Handbook Vapor
Handbook Vapor
NIPER--580
DE92 001070
BY
Partha S. Sarathi
David IC Olsen
October 1992
Prepared for
U.S. Department of Energy
Assistant Secretary for Fossil Energy
P. 0.Box 1398
Bartlesvik OK 74005
This document io
PUBLICLY RELEASABLE
A
Authorizing official
Date:
-9 -0d
Prepared by
IIT Research Institute
National Institute for Petroleum and Energy Research
P. 0. Box 2128
Bartlesville, OK 74005
DISCLAIMER
This report was prepared as an account of work sponsored by an
agency of the United States Government. Neither the United States
Government nor any agency Thereof, nor any of their employees,
makes any warranty, express or implied, or assumes any legal
liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, product, or process
disclosed, or represents that its use would not infringe privately
owned rights. Reference herein to any specific commercial product,
process, or service by trade name, trademark, manufacturer, or
otherwise does not necessarily constitute or imply its endorsement,
recommendation, or favoring by the United States Government or any
agency thereof. The views and opinions of authors expressed herein
do not necessarily state or reflect those of the United States
Government or any agency thereof.
DISCLAIMER
Portions of this document may be illegible in
electronic image products. Images are produced
from the best available original document.
TABLE OF CONTENTS
Abstract .......................................................................................................
Acknowledgments...........................................................................................
Page
1
2
TABLES
4
1.1 Properties of saturated steam ........................................................................
Heat
content
of
wet
steam.
Btu/lb
..................................................................
6
1.2
1.3 Approximate contribution of various steamflood mechanisms to overall recovery
of 13' to 15' API gravity oil .......................................................................
10
ILLUSTRATIONS
6
1.1 Heat content of wet steam. Btu/lb ..................................................................
of
steamflood
process
...................................................................
7
Schematic
1.2
8
1.3 Schematic of steamflood temperature profile ......................................................
Schematic
of
steamflood
saturation
profile
........................................................
9
1.4
12
1.5 Effect of temperature on oil viscosity ............................................................
Schematic
of
cyclic
steam
process
................................................................
14
1.6
15
1.7 Schematic of push-pull cyclic steam process ....................................................
CHAPTER 2 . Criteria for Evaluating Steam Injection Prospects .....: ..........................
18
Introduction ............................................................ . !..............................
18
Rock and Fluid Properties ...........................................
i ........ ................. 18
Rock Type .................................................................................. 18
Pay Zone Thickness ...........................................................................
18
......... 18
Depth and Reservoir Pressure ......................................................
iii
TABLE OF CONTENTS-Continued
Page
TABLES
2.1 Cyclic steam screening criteria ....................................................................
2.2 Steamflooding screening criteria ..................................................................
19
20
ILLUSTRATIONS
2.1 Estimated heat loss as function of depth ......................................................... 22
2.2 Typical pilot pattern configuration for an anisotropic reservoir ............................... 23
2.3 Unconfined steamflood pattern with peripheral injectors ......................................
27
iv
..
.
. _ _ _ _.
TABLE OF CONTENTS-Continued
TABLES
3.1 List of major initial expenses to be included in the steam injection project economic
evaluation ............................................................................................
32
3.2 Steps for steam injection prospect evaluation process selection. and field
implementation ...................................................................................... 32
3.3 Initial investigation .................................................................................
33
3.4 Laboratory analysis and data processing .........................................................
33
34
3.5 Comprehensive pilot feasibility study and design ..............................................
3.6 Field pilot implementation .........................................................................35
3.7 Expanded field operation .......................................................................... 35
CHAPTER 4 . Economics of Steam Injection ......................................................
38
Introduction .........................................................................................
38
38
Economic Factors in Steam Injection Operation ............................................
Estimation of Economically Recoverable Oil ................................................................ 43
. .
Steam Injection Project Costs .................................................................
44
Cost Estimate ...................................................................................
45
Development Costs ............................................................................ 45
Well Costs .................................................................................. 45
45
Water Treatment Facilities .....................................................................
Steam Generation ..............................................................................
48
Steam Distribution System .................................................................... 49
Other Surface Facilities ........................................................................
50
Free Water Knockout Unit .................................................................... 50
Heater Treater ................................................................................... 50
Well-Testing Unit .............................................................................. 50
LACTUnit ...................................................................................... 50
TankBattery .................................................................................... 50
Operating Costs ................................................................................ 50
Raw Water Production and Transportation Costs ..........................................
50
Water Treatment ................................................................................ 50
Steam Generation Operation Costs ...........................................................
51
Power ............................................................................................
51
Well Pulling and Workover Costs ........................................................... 51
Other Operating Costs .........................................................................
51
Financial Costs .................................................................................
51
Summary ............................................................................................
52
References ...........................................................................................
55
Appendix 4-A-Review of Department of Energy Steam Drive Predictive Models ........ 57
Background ..................................................................................... 57
Review of Available Steamflood Models from BPO .................................. 57
Model Recommendation .................................................................. 58
References for Appendix 4-A .................................................................
58
Appendix 4-B--Calculation of Capital Charge Rate ........................................... 61.
TABLE OF CONTENTS-Continued
TABLES
Page
4.1 Kern County. CA. new steam injection well costs. excluding well head injection
equipment ............................................................................................
40
4.2 Kern County. CA. new production well costs. excluding downhole pump ................. 46
4.3 Cost to drill and equip a 1.000 ft steam injection well .........................................
46
47
4.4 Cost to drill and equip a 1.000 foot production well ...........................................
4.5 Index of drilling and equipping wells unadjusted for depth ................................... 48
49
4.6 Characteristics of conventional lease crude fired steam generator ............................
4.7 Conventional oil fired surface steam generator costs ........................................... 49
4.8 Comparisons of economic factors affecting oil production from selected oil
production states .................................................................................... 53
4.9 Capital Charge Rates ............................................................................... 54
54
4.10 Suggested average life tables for oilfield related equipment ...................................
CHAPTER 5 . Water Treatment for Steam Genelation ............................................ 62
Introduction ......................................................................................... 62
Water for Steam Generation ....................................................................... 62
Sources of Water for Steam Generators .....................................................
62
Water Treating Consideration ................................................................. 63
Total Hardness ............................................................................. 65
Alkalinity ................................................................................... 65
Oxygen ...................................................................................... 65
Sulfides .....................................................................................
65
Dissolved Solids.................... !...................................................... 65
Suspended Solids ..........................................................................
65
Iron .......................................................................................... 65
Oil ........................................................................................... 66
Silica ........................................................................................ 66
66
pH ...........................................................................................
Biological Growths ........................................................................ 66
66
Feedwater Treating Equipment ....................................................................
Ion Exchange Units ............................................................................ 66
Sodium Zeolite Softener ....................................................................... 68
Principles of Operation ........................................................................
68
Salt Requirements ..............................................................................
69
Softener Equipment and Operation ...........................................................
71
Brining System ................................................................................. 74
Softener Operation ............................................................................. 75
Backwash ...................................................................................
76
Brining ...................................................................................... 76
Slow Rinse ................................................................................. 77
Fast Rinse .................................................................................. 77
Limitations of Strong Acid Resins ........................................................... 77
Weak Acid Resins ..............................................................................
79
Regeneration ...............................................................................
79
Neutralization ..............................................................................
79
Weak Acid Cation Exchange Resin Equipment and Operation ...........................
80
Limitation of Weak Acid Cation Exchange Resins ......................................... 80
Ion Exchange Unit Operation Problems .....................................................
80
Resin Stability ..................................................................................
81
vi
. -
...............
TABLE OF CONTENTS-Continued
Varying Water Quality .........................................................................
Poor Operational Practices ....................................................................
Mechanical Problems ..........................................................................
Fouled Resins ..................................................................................
Hardness Leakage ..............................................................................
Water Treatment for Steam Injection Projects ..............................................
Steam Injection Water Softening System ....................................................
Selection of Ion Exchange Resins ............................................................
Ion Exchange Calculations ....................................................................
Use of Chelating Agents ......................................................................
Hardness Monitors .............................................................................
Overview of Other Water Treatment Equipment ............................................
Oil and Solid Removals .......................................................................
Gravity Separation .............................................................................
Flotation .........................................................................................
Induced Gas Flotation .........................................................................
Operation Principle .............................................................................
Filtration .........................................................................................
Choice of Filters ................................................................................
Pressure Filters .................................................................................
Diatomaceous Earth (DE) Filters .............................................................
Selection of Diatomite ........................................................................
Diatomite Filter Problems ....................................................................
Deaeration ..........................................................................................
Oxygen Exclusion .............................................................................
Oxygen Removal ..............................................................................
Mechanical Deaeration ........................................................................
Vacuum Deaeration ...........................................................................
Steam Deaeration ..............................................................................
Gas Stripping ..................................................................................
Chemical Deaeration ..........................................................................
Oxygen Scavenging by Sulfur Oxide .......................................................
Reclamation of Produced Water For Steam Generation ..................................
Silica Scaling ..................................................................................
Disposal of Excess Produced Water ........................................................
Water Treating Costs .........................................................................
Guidelines to Water Treatment System Design and Operation ...........................
Design Concept ................................................................................
Source Water Properties and Problems.....................................................
Equipment Consideration ....................................................................
Operation of a'Water System ................................................................
Summary ............................................................................................
References ..........................................................................................
General References ................................................................................
Appendix 5-A.............................................................................................
Water Chemistry ...................................................................................
Definition .......................................................................................
Ions, Valence and Radicals ..................................................................
pH ...............................................................................................
Water Analysis ................................................................................
vii
Page
8c
82
82
83
83
85
85
86
88
88
91
91
92
93
93
95
95
98
98
98
99
101
101
102
102
103
103
104
105
107
109
110
113
114
116
117
118
120
121
122
122
123
124
126
129
129
129
130
131
131
TABLE OF CONTENTS-Continued
Page
TABLES
5.1 Typical analysis of raw waters used in the San Jaoquin Valley. California. steam
injection projects ....................................................................................
64
5.2 Approximate total dissolved solid (TDS) content and hardness of California oilfield
produced water ......................................................................................
64
65
5.3 Steam generator feedwater requirements .........................................................
67
5.4 Methods for removal of common raw water impurities........................................
5.5 Effects of salt level on the peiformance of a typical synthetic sodium zeolite exchanger . . 7 1
72
5.6 Zeolite softener seiection chart ....................................................................
5.7 Rules-of-Thumb for selection of ion exchange resins for oilfield steam generator
feedwater softener ..................................................................................
88
5.8 Ion exchange softening system calculations..................................................... 89
5.9 Concentration of EDTA and its sodium salts necessary to complex 1 ppm of hardness
ion .................................................................................................... 90
5.10 Induced gas flotation cell selection chart .........................................................
96
5.11 Feedwater oil reduction achievable using a depurator .......................................... 97
5.12 Summary of advantages and disadvantages of water deoxygenation systems .............104
5.13 Oxygen towers selection chart ................................................................... 109
5.14 Typical SO2 scrubber waste analysis ............................................................ 113
5.15 Typical water analysis for water treatment cost estimation ................................... 118
5.16 Fresh water treatment costs .......................................................................
119
5.17 Produced water treatment costs ..................................................................
119
..
ILLUSTRATIONS
5.1 Effect of feedwater hardness and salt dosage on resin capacity ...............................
5.2 Effect of brine hardness and salt dosage on water softening costs ...........................
5.3 Schematic of a typical ion exchange water softener ............................................
5.4 Schematic of the internals of a typical zeolite softener unit ....................................
70
72
73
74
viii
...
ILLUSTRATIONS-Continued
Page
Schematic
of
a
bulk
salt
storage
facility
..........................................................
75
5.5
Effect
of
water
hardness
and
salt
dosage
on
hardness
leakage
from
primary
beds
.........
78
5.6
Resin
exchange
capacity
as
a
function
of
water
hardness
and
salt
dosage
...................
78
5.7
Schematic
of
steam
injection
water
softener
piping
arrangement
.............................
86
5.8
Flow
schematics
of
steamflood
ion
exchange
softener
regeneration
cycle
...................
87
5.9
5.10 Schematic of an automatic hardness monitor ....................................................
92
94
5.1 1 Schematic of a typical dissolved gas flotation unit ..............................................
96
5.12 Schematic of an induced gas flotation cell .......................................................
5.13 Schematic of a typical up-flow sand filter ......................................................
100
5.14 Schematic of a vacuum deaerator ................................................................ 106
5.15 Schematic diagram of a spray type steam deaerator ........................................... 106
5.16 Schematic of a counter current gas stripping unit .............................................. 108
5.17 Schematic of sulfur burner process for oxygen deaeration ................................... 111
5.18 Flow diagram for Kern River. CA. steamflood produced water reclamation plant ........ 116
CHAPTER 6 . Steam Generation ...................................................................
Introduction ........................................................................................
General Features of Oil Field Steam Generators ..........................................
Generator Selection ...........................................................................
Design Requirements For Oilfield Steam Generators.....................................
Steam Generator Components ..........................................................
1 . Feedwater System ....................................................................
2 . Feedwater Preheater .................................................................
3 . Fuel System ...........................................................................
4 . Combustion Air System .............................................................
5 . Convection Section ..................................................................
6 . Radiant Section .......................................................................
Steam Generator Controls ....................................................................
Process Description ...........................................................................
Fuel System ....................................................................................
Combustion Air System ......................................................................
Steam Generator Specifications .............................................................
Fuels For Oilfield Steam Generators ........................................................
ix
145
145
145
147
152
152
152
152
153
153
153
154
155
155
157
158
159
161
162
162
162
163
163
164
165
165
166
167
167
.169.
173
173
175
TABLE OF CONTENTS-Continued
Bage
ILLUSTRATIONS
6.1 Iosmetric view of a skid-mounted oilfield steam generator ................................... 146
6.2 Schematic of oilfield steam generator ........................................................... 147
6.3 Approximate steam generator output (lbs/hr/bhp)at various pressures. steam quality
and feedwater temperature ........................................................................ 149
6.4 Daily fuel (lease crude) costs for operating various capacity steam generators.............150
6.5. Daily fuel (natural gas) costs for operating various capacity steam generators ............. 150
6.6 Tubing arrangements in the steam generator flue gas convection section ...................154
156
6.7 Layout of a typical oilfield steam generator burning lease crude .............................
6.8 Approximate flue gas dewpoints for heavy oils ................................................ 171
171
6.9 Thermal efficiency versus flue gas temuerature for 10" API crude ..........................
172
6.10 Thermal efficien6 versus flue ias temperature for 15" API crude ..........................
6.1 1 Thermal efficiency versus flue gas temperature for natural gas (1.000 Btu/SCF) ......... 172
.....
..
..
~-.
TABLE OF CONTENTS-Continued
6.12
6.13
6.14
6.15
6.16
6.17
6.18
6.19
6.20
6.2 1
Page
207
CHAPTER 7 . Steam Distribution ..................................................................
207
Introduction ........................................................................................
Components of a Steam Distribution Network ............................................ 207
General System Pressure and Temperature Design Requirements ......................
209
Main Steam Headers..........................................................................
212
Lateral Steam Lines ...........................................................................
213
213
Steam Flow Rates and Pressure Drop ..................................................
213
Steam Shut-off Valves ...................................................................
Expansion of Steam Lines ...................................................................
214
Expansion of the Steam Headers ............................................................ 217
Wellhead Connection ......................................................................... 217
Support of Steam Lines ......................................................................
217
Insulation and Heat Loss.....................................................................
220
Effects of Branching on Steam Quality ..................................................... 223
Wellhead Equipment.......................................................................... 224
229
Wellhead Steam Rate Controt Devices ......................................................
229
Static Chokes in Critical Flow ..........................................................
232
Adjustable Chokes Operated in the SubcriticalFlow Regime .......................
The Effect of Steam Quality on Injection Rate Control............................... 234
Control of Individual Well Steam Quality .................................................. 234
Steam Distribution System Maintenance and Monitoring ................................
235
Summary ...........................................................................................
235
236
References..........................................................................................
237
Appendix 7-A . Steam distribution network components suppliers .........................
TABLES
7.1
7.2
7.3
7.4
7.5
ILLUSTRATIONS
7.1 Schematic of a typical steam distribution system layout ...................................... 208
7.2 Typical expansion loops used in steam injection operations to relieve steam flow line
thermal stresses ....................................................................................
215
xi
TABLE OF CONTENTS-Continued
Page
Schematic of different expansion methods for steam injection lines ......................... 215
Z-bend expansion loop sizing chart ............................................................. 216
Schematic of a typical swivel joint used at the steam injection wellhead .................... 218
Schematic of expansion manifold connection to steam injection well Christmas tree .....218
Steam injection line heat loss versus cash loss .................................................221
Schematic of insulated pipe for heat loss transfer estimation ................................. 222
Schematic of pipe tee installations in steam injection lines ....................................
224
225
Schematic of typical steam injection wellhead arrangement ..................................
Schematic of steam injection wellhead details .................................................. 226
226
Schematic of a typical steam injection well casing stuffing box ..............................
Schematic of a typical steam injection well tubing stuffing box .............................. 227
Schematic of steam injection wellhead assembly showing casing and tubing stuffing
box arrangement ...................................................................................
227
7.15 Schematic of steamflood production wellhead arrangement showing the facilities
for pump hookup .................................................................................. 228
7.16 Schematic of acceptable and nonacceptable choke inserts for steam injection rate
control choke bean ................................................................................. 230
7.3
7.4
7.5
7.6
7.7
7.8
7.9
7.10
7.11
7.12
7.13
7.14
CHAPTER
Introduction ........................................................................................
238
238
TABLES
8.1 Steam injection wellhead steam quality .........................................................
247
ILLUSTRATIONS
8.1 Schematic of equipment arrangement for steam quality measurement by the vapor-liquid
separation technique ............................................................................... 240
8.2 Schematic of a vibration densitometer for steam quality determination ..................... 248
8.3 Schematic of gamma ray steam quality measurement technique ............................. 249
xii
TABLE OF CONTENTS-Continued
Page
TABLES
9.1 Recommended operating temperature range for thermal well tubular goods ...............258
9.2 Tensile Properties of various grades of seamless casing and tubing ........................ 259
9.3 Typical casing completions in California thermal wells ....................................... 262
ILLUSTRATIONS
9.1
9.2
9.3
9.4
9.5
9.6
xiii
258
260
260
264
265
265
TABLE OF CONTENTS-Continued
Page
9.7 Schematic showing retrievable and permanent packer position in steam injection well ...277
.9.8
Schematic detailing packer installation for multizone completion ............................ 277
9.9 High-pressure steam injection well packer details .............................................
278
:.....284
CHAFER 10. Thermal Well Cementing Practices ..........................................
Introduction ........................................................................................
284
Steam Injection Well Cementing Requirements ...........................................
284
. -ThermalCement Chemistry ......................................................................
284
Portland Cement ............................................................................... 284
High Alumina Cement .......................................................................... 288
. Steamflood Cementing Practices ................................................................. 288
Steam Well Problems .......................................................................... 288
Cementing Practices .......................................................................... 292
Summary ...........................................................................................
293
...................293
References......................................................................
Appendix 10-A. Thermal well cementing service vendors................................... 295
ILLUSTRATIONS
10.1
10.2
10.3
10.4
10.5
10.6
10.7
10.8
296
296
296
296
297
298
298
299
299
300
301
301
302
302
304
304
305
306
309
xiv
. .
-~
...................
TABLE OF CONTENTS-Continued
ILLUSTRATIONS
Paee
11.1 Schematic of a diluent pump used in the Cat Canyon field of California to produce the
6" M I oil ........................................................................................... 303
C
310
310
310
311
311
314
316
318
318
318
320
321
TABLES
12.1 Cost effectiveness of insulated pipe .............................................................
319
ILLUSTRATIONS
12.1 Heat loss in the wellbore and casing temperature during steam injection ................... 312
12.2 Variation of overall heat transfer coefficient (U) with tubing temperature for
different tubing surfaces and annulus pressure ................................................ 313
12.3 Estimated heat loss as function of depth ........................................................ 313
12.4 Schematic of basic insulated tubular design ....................................................
315
12.5 Typical insulated tubing construction ...........................................................
315
12.6 Exploded view of a field installed insulated tubular ...........................................
317
CHAPTER 13 . Sand Control Problems .............................................................
Introduction .........................................................................................
Occurrence and Identification of Problems-Causing Sands ..............................
Prevention of Problems in Producing Wells ...............................................
Use of Open-Hole or Gravel Pack Completion ............................................
Screens .........................................................................................
Gravel Packing ...................................................................................
Design of Gravel Packs for Thenha1 Wells.................................................
Chemical Methods of Sand Control .........................................................
Summary ...........................................................................................
References ...........................................................................................
Appendix 13-A. Sand control and gravel pack system vendors .............................
.
322
322
322
323
324
325
326
329
333
333
334
337
'TABLES
13.1
13.2
13.3
13.4
325
326
328
329
TABLE OF CONTENTS-Continued
ILLUSTRATIONS
13.1 Types-of screens ...................................................................................
13.2 Gravel size selection ...............................................................................
13.3.Schematic of a thermal well gravel pack completion ..........................................
13.4 High-temperature screen ..........................................................................
Paee
327
328
330
331
References...........................................................................................
Appendix .14-A. Steam injection profie control service vendors ...........................
354
354
357
TABLES
14.1 Procedure for determining steam injection profile using a spinner .......................... 341
14.2 Properties of tracers ............................................................................... 344
14.3 Procedure for running radioactive tracer survey with sodium iodide........................ 344
14.4 Procedure for running radioactive tracer survey using inert gas tracers..................... 345
14.5 Calculation of the critical pressure ratio ......................................................... 351
ILLUSTRATIONS
14.1
14.2
14.3
14.4
14.5
14.6
14.7
14.8
14.9
14.10
xvi
. .
. . . . .
TABLE OF CONTENTS-Continued
CHAFTER 15. Emulsion Problems in Steam Injection Projects .................................
Introduction ........................................................................................
Definition of an Emulsion ....................................................................
Emulsion Formation and Stability...........................................................
Methods Used to Break Emulsions .........................................................
Emulsion Breaking ............................................................................
A . Oil-in-Water Emulsions .............................................................
B. Water-in-Oil Emulsions .............................................................
Summary ...........................................................................................
General References ................................................................................
Appendix 15-A. Emulsion treatment chemical suppliers .....................................
Paee
358
358
358
359
360
361
361
363
364
364
365
TABLES
16.1 Specifications for horizontal freewater knockouts .............................................
16.2 Specifications for vertical freewater knockouts ................................................
16.3 Computerized automation of oilfield productions-costs and benefits......................
375
375
388
ILLUSTRATIONS
16.'
16.2
16.3
16.4
16.5
16.6
16.7
16.8
xvii
367
368
371
374
374
376
379
380
TABLE OF CONTENTS-Continued
16.9
16.10
16.11
16.12
16.13
Page
381
382
382
384
385
xviii
TABLE OF CONTENTS-Continued
Page
TABLES
18.1
18.2
18.3
18.4
18.5
18.6
18.7
xix
BY
Partha S. Sarathi and David K. Olsen
ABSTRACT
More than 80% of the total steam injection process operating costs are for the production of
steam and the operation of surface and subsurface equipment. The proper design and operation of
the surface equipment is of critical importance to the success of any steam injection operation.
However, the published monographs on thermal recovery have attached very little importance to
this aspect of thermal oil recovery; hence, a definite need exists for a comprehensive manual that
places emphasis on steam injection field practices and problems. This handbook is an attempt to
fulfill this need.
This handbook explores the concept behind steam injection processes and discusses the
information required to evaluate, design, and implement these processes in the field. The emphasis
is on operational aspects and those factors that affect the technology and economics of oil recovery
by steam. The intended purpose of this handbook is twofold: (a) to provide operators with a
ready reference, a starting point when searching for information to engineer a steam injection
project; and (b) to provide a bibliographic source on the subjects discussed in the handbook. It
should, however, be emphasized that this handbook is not a detailed "trouble shooter" manual on
field problems. Steam injection process problems and solutions are field specific, and only
The handbook is organized in the above manner so that the reader will have a good
understanding of the engineering and financial requirements of a steam injection project before he
reads the steam injection field practices, problems and solutions.
ACKNOWLEDGMENTS
This work was sponsored by the U.S. Department of Energy (DOE) under Cooperative
Agreement DE-FC22-83FE60149 and performed within project BEllB as outlined in the FY91
Annual Research Plan (DOE Report NIPER-465). The work of this nature could not have been
completed without the help and cooperation of various personnel. Many people have made
significant contributions to the preparation of this report. In particular, the authors wish to offer
their thanks to John Ball, consultant, for his assistance in the preparation of chapters 12, 13, and
14; Bruce Ramzel, and Summer DOE Interns, Heather Horstman, Yesh Tyagi, William Lucas,
Johnathan Grigsby and Robert Pendergrass 11, all of NIPER, for their help in preparing the
figures; and Rex Thomas, Consultant, for his assistance in gathering materials for the report. We
offer our thanks to Michael Madden for his assistance in preparing chapter 18 and Arden Strycker
and Min Tham, all of NIPER, and Thomas B. Reid of the U.S.Department of Energy, Bartlesville
Project Office, for their critical reviews of the manuscript. We also extend our appreciation to Bill
Linville and Ron Kendall of NIPER for their editorial review. We also thank the American
Petroleum Institute for permission to reproduce portions of RP11T report and various technical
journals and equipment manufacturers for permission to use their illustrations, tables, and charts.
CHAPTER 1
FUNDAMENTALS OF STEAM INJECTION PROCESSES
INTRODUCTION
Steam injection is a thermal drive process that adds heat to the reservoir to expand the oil-inplace, reduce its viscosity, provide drive energy and thereby improve the displacement efficiency
of injected fluid. For more than 100 years, steam has been used in attempts to solve the problems
of low recovery from heavy oil reservoirs. Its full potential was not realized until the early 50s
when field testing of the steam injection process began. In the United States in 1990,
steamflooding recovered 520,000 barrels of oil per day (BOPD) or 73% of all enhanced oil
recovery. 1
This manual explores the concepts behind steam injection processes and discusses the
information required to evaluate, design, and implement these processes in the field. The emphasis
is on the practical aspects of steam injection. Details of equations and calculation methods for
estimating steam injection performance are not discussed in this manual. Other publications24
cover these subjects in detail. There is a heavy emphasis on surface and subsurface facilities, field
practices, and operational problems which are not discussed in sufficient detail in other
publications. The material presented is directed toward engineers and independent operators who
have become familiar with routine waterflood operations but have had no exposure to thermal
operations.
This chapter reviews the steam injection process and recovery mechanisms. In subsequent
chapters, various aspects of steam injection are discussed in detail.
Basics of Steam Injection
Although in situ reservoir energy is responsible for the displacement of oil through porous
rock into wellbores, it is not always sufficient for oil displacement, especially in shallow heavy oil
reservoirs. Hence, energy needs to be imparted to the reservoir to allow the oil to flow toward a
wellbore. This usually takes the form of mechanical displacement such as pushing the oil with
water or gas. In steam injection, steam is used as the displacing fluid. Steam augments the
reservoir energy by parting its heat content as well as mechanical energy because of pressure
differential.
Steam is an ideal fluid for adding energy to a reservoir because of its high heat content per
pound. For example, water<at400" F contains 375 Btdlb, but saturated steam at 400" F has 1,201
Btu/lb, or over three times the heat content of water.
To fully realize why steam is effective in producing oil, one needs to understand the
properties of steam as well as what happens in a reservoir when steam is injected. Some properties
of saturated steam for a few selected temperatures and pressures are listed in table 1.1.
Absolute
pressure,
pia
Saturation
temperature.
"F
Heat of
liquid
Btunb
Latent
heat of
vaporization,
Btwlb
Heat
content
of steam,
Weight
oflcuft
of steam,
Volume
of 1 Ib
of steam,
Vdum
0f1ib
of liquid.
Btdb
Ib
Cuft
aft
50
28 1.O
250.2
923.9
1,174.1
0.1175
8.514
0.01727
100
327.8
298.5
888.6
1,187.2
0.2257
4.431
0.01774
150
358.4
330.6
863.4
1,194.1
0.3318
3.014
0.01809
200
381.8
355.5
842.8
1,198.3
0.4372
2.287
0.01839
250
401.0
376.1
825.0
1,201.1
0.5425
1.843
0.01 865
300
417.4
394.0
808.9
1.202.9
0.6482
1.543
0.01889
400
444.6
424.2
780.4
1.204.6
0.8614
1.161
0.01934
500
467.0
449.5
755.1
1,204.7
1.0781
0.9276
0.01975
600
486.2
471.7
732.0
1,203.7
1.2991
0.7698
0.02013
700
503.0
491.6
710.2
1,201.8
1.5254
0.6556
0.0205
800
518.2
509.8
689.6
1,199.4
1.7576
0.567
0.02087
900
532.0
526.7
669.7
1,1%.4
1.9964
0.501
0.02123
1,ooo
544.6
542.6
650.4
1.192.9
2.2424
0.446
0.02159
1.100
556.3
557.5
631.5
1.189.1
2.4964
0.401
0.02195
1.200
567.2
571.9
613.0
1.184.8
2.759
0.3624
0.02232
1,300
577.4
585.6
594.6
1.180.2
3.0311
0.33
0.02269
1.400
587.1
598.8
576.5
1,175.3
3.3137
0.3018
0.02307
1,500
596.2
611.7
558.4
1,170.1
3.6076
0.2771
0.02346
2,000
635.8
672.1
466.2
1,138.3
5.3104
0.1883
0.02565
2,500
668.1
731.7
361.6
1.093.3
7.6523
0.1307
0.02859
3.000
695.3
801.8
218.4
1,020.3
0.085
0.03428
11.765
Note that the heat content of steam is considerably higher than the heat content of liquid
water. This difference is the latent heat and represents the amount of additional energy required to
convert boiling water at a given pressure into steam at the same pressure and temperature. Latent
heat is large at lower pressure and decreases with increase in pressure and becomes zero at
705.47' F temperature and 3,208.2 psia pressure. This temperature and pressure is known as the
critical point of water. Above this point, water exists only as a single phase. Also, note that steam
saturation temperature increases with pressure and the rate of increase is greatest at lower
pressures. This rate of change in temperature becomes smaller at higher pressure. For example,
when the steam pressure increases from 100 to 200 psia, the saturation temperature changes from
328" to 381" F or an increase of 53" F, On the other hand, when the pressure of steam increases
from 1,000 to 1,100 psia, the corresponding increase in saturation temperature is only about 12" F.
The other important characteristic of steam is the change in volume. The volume of 1 lb. of
saturated steam at various pressures and temperatures is depicted in Table 1.1. At 400 psia and
444' F, 1 lb of water occupies 0.0193 cu ft, but 1 lb of saturated steam at the Same conditions has a
volume of 1.161 cu ft. Therefore, the saturated vapor occupies about 60 times the volume of
water. Naturally, this ratio decreases with an increase in steam pressure but still is significant at
the pressure range where most thermal projects operate. For example, at 1,500 psia, the upper
limit for many steamflood projects, the ratio is 12. This is one of the reasons why steam heating
has had such success in the thermal process. A given amount of reservoir heating can be
accomplished with a much smaller unit weight of steam than hot water.
Steam Quality
Another term frequently encountered in the steam injection processes is steam quality. This
term refers to the degree of dryness of steam. For example, an 80%quality steam refers to a steam
water mixture containing 80% steam and 20% water by weight. A 100% quality steam refers to
steam containing no water and is known as dry saturated steam. The heat content of a wet steam
(mixture of steam and water) is always lower than that of a dry saturated steam. For example, a
70% quality steam at 200 psia contains about 946 Btuflb, and dry steam at the same pressure
contains 1,200 Btu/lb, or about 1.3 times as much heat as the wet steam. At higher pressures, this
ratio becomes smaller since the effect of steam quality becomes less predominant. This is because
the enthalpy of water increases with pressure and the latent heat of vaporization decreases with
pressure.
In table 1.2, the heat content of several qualities of steam is presented. The heat content of
various quality steam is also shown graphically in figure 1.1. Most oilfield steam generators are
designed to give 80%quality steam. Nevertheless, in the 100 to 1,500 psia pressure range, where
most steam injection projects operate, wet steam carries more heat than hot water. For example, 1
lb of an 80%quality steam at 200 psia carries 674 Btu more heat than 1 lb of boiling water at the
same temperature. This allows an operator to introduce more heat in a reservoir per pound of
injected fluid.
Steam Injection Process
What happens in a reservoir when steam is injected? Obviously, steam moves through the
reservoir and in the process it heats the oil and displaces it towards a producing well. Figure 1.2
depicts schematically a steamflood process.
Temperature.
T
100
90
80
70
1,174.1
1,187.2
1.194.1
1.198.3
1.201.1
11204.6
1,204.6
1.204.7
1.203.7
1,201.8
1.199.4
11196.4
1,192.9
1.189.1
1,184.8
1.180.2
1.175.3
.._ ..
1,170.1
1,138.3
1.093.3
1,020.3
1.081.7
1.098.2
1.107.7
1i114.0
1.118.6
11122.0
1,126.6
1.129.1
1.130.5
1.130.8
1.130.4
11129.4
1,128.0
1.125.9
1,123.6
1.120.7
1.117.7
..._
1.114.3
1.091.7
1,057.1
998.4
989.3
1.009.4
1.021.3
1,029.7
1.036.1
11041.1
1.048.5
1.053.6
1.057.3
1.159.8
1.061.5
11062.5
1,062.9
1,062.7
1.062.3
1.061.3
1.060.0
1;058.4
1,045.1
1,002.1
976.5
8%.9
920.5
935.0
945.6
953.6
960.5
970.5
978.1
984.1
988.7
Psi
281.0
327.8
358.4
381.8
401.o
417.4
444.6
467.0
486.2
503.0
518.2
532.0
544.6
556.3
567.2
577.4
587.1
596.2
635.8
668.1
695.3
50
100
150
200
250
300
400
500
600
700
800
900
1.m
1,100
1.200
1.300
1,400
1,500
2.000
2.500
3,OOo
992.5
995.5
997.9
999.6
1,001.0
1.001.8
1.002.4
liG.6
998.4
984.8
954.7
60
50
40
20
804.5
712.2
742.8
762.3
776.9
788.6
798.5
814.4
827.1
837.7
846.7
854.6
861.6
867.8
873.3
878.4
882.9
887.1
619.8
653.9
676.0
692.6
706.1
717.6
736.4
751.5
764.5
775.7
785.6
794.6
435.0
476.2
503.3
524.1
541.1
250.2
298.5
330.6
355.5
376.1
394.0
424.2
449.5
486.2
491.6
831.7
848.6
861.2
871.1
879.3
892.4
902.6
910.9
917.7
923.6
928.5
932.8
936.4
939.7
942.4
944.7
946.7
951.8
948.7
932.8
555.8
580.3
600.5
618.1
633.6
647.7
660.7
672.7
683.8
694.5
704.5
714.1
723.4
765.3
802.8
810.1
817.1
809.7
829.4
835.1
858.6
876.3
889.2
890.9
905.2
912.5
911.0
804.0
845.4
1100
300
200
100
10
20
30
40
50
60
70
80
90
100
509.8
526.7
542.6
557.5
571.9
585.6
598.8
611.7
672.1
731.7
801.8
TANK B A l l E R I
IMPERMEABL
As steam moves through a reservoir between injector and producer, it creates several
temperature fluid-flow regions. These regions are designated as the steam zone, hot condensate
Ts
Ts = STEAM TEMPERATURE
T, = ORIGINAL RESERVOIR TEMPERATURE
1
2
3
4
= STEAM ZONE
= HOT CONDENSATE ZONE
= OIL/COLD WATER ZONE
= INITIAL ZONE
Bt-
Tr
DISTANCE
80
1 =
2 =
3 =
4 =
STEAM ZONE
HOT CONDENSATE ZONE
OIL / COLD WATER ZONE
INITIAL ZONE
70
60
50
40
30
20
10
0
DISTANCE
200
10-12
1-2
Temperature, O F
300
400
20-24
2-4
10-14
1-2
1-2
24-26
2-4
12-15
1-2
1-2
Steam distillation, the principal oil recovery mechanism in the steam zone, recovers reservoir
fluids in the following manner. When steam contacts the crude, a portion of steam condenses and
provides heat to the crude. The mixture of heated crude and hot condensate begins to boil when
the vapor pressure of the mixture (equal to the sum of the partial pressure of water. and crude) is
equal to or exceeds the system pressure. The mixtixe will always boil at a temperature lower than
either of the constituents boiling point at the system pressure. The degree of boiling point
lowering depends on vapor phase composition, since this composition influences total system
pressure.
Steam distillation results in the rapid vaporization of oil and water. This rapid boiling
disturbs and partially redistributes oil from a dead end pore to a connecting pore. This phenomena,
known as the chipping efect,6 results in a more efficient displacement of the crude oil by the
injected steam.
The chipping effect is more pronounced in a heavy oil reservoir than in a light oil reservoir
because the low volatility of the heavy oil inhibits effective separation of the lighter components
from the crude. As the volatility of a crude increases, the chipping effects diminishes and the
importance of distillation increases. Depending upon the volatility of crudes, up to 70% of oil-inplace has been recovered in laboratory 2-D steamfloods.9 However, in the field the contribution of
steam distillation to overall recovery is expected to be lower.
Gas Stripping
Along with steam distillation, gas stripping also occurs in the steam zone.6 Gas stripping
occurs because in the steam zone, steam selectively desorbs light fractions from the crude.
However, this process is less efficient than steam distillation.
In Situ Solvent Drive
A substantial portion of the light oil fraction and the carrier steam condenses in the cooler
region ahead of the steam zone and forms the hot condensate zone.
10
The condensed steam (hot water) being more viscous than steam, reduces steam fingering.
The degree of reduction of steam fingering depends on the oil-water mobility ratio. The steam
condensate flows along with the oil to form a hot water drive.
In the hot condensate zone, the lighter fractions of the oil mix with the in situ oil and dilute it.
This dilution reduces the overall density and viscosity of the oil. As the steam front pushes further
towards the producers, the light ends accumulate and extract additional light fractions from the oil
with which it comes in contact. Thus, the volume of the light oil solvent bank grows as it moves
through the reservoir. The result is a miscible displacement of the oil phase. The amount of
additional oil recovery due to this miscible displacement has been estimated at 3 to 5 8 of the initial
oil-in-place for some heavy 0ils.~0
The net effects of dilution and extraction of oil in the hot water flood region are (a) decrease
in residual oil saturation and (b) reduction in the amount of heavy fractions at the trailing edge of
the hot water flood region. Since the oil left behind in the hot water flood region is likely to be
steam distilled by the advancing steamfront, only a small amount of the heavy oil fraction will be
left behind by the steamflood.
Viscosity Reduction
The most obvious effect of heating a heavy oil reservoir is reduction of oil viscosity.
Figure 1.5 shows this pronounced change in viscosity. This plot shows the effect of temperature
on the viscosity of a 14 API gravity oil. It is evident that the Viscosity improvement is greatest at
lower temperatures and tends to be marginal after reaching a certain temperature. Greater viscosity
reductions are experienced in the more viscous low API gravity crudes than in higher API gravity
crudes. Note that the viscosity of a typical 14 API crude at 80 F is reduced from 1,445 to about
47 CP at 175 F; more than thirty-fold. In the case of a low viscosity crude, viscosity reduction
with temperature is not as dramatic as with the more viscous crude. The plot clearly indicates the
advantage of temperature in making viscous crude more flowable in the reservoir rock. Thus,
the net result of increasing temperature is to improve the mobility ratio. With the oil viscosity
lowered, the displacement efficiency and sweep efficiency are improved.
The change in oil viscosity with temperature is reversible, i.e. the oil viscosity returns to its
original value when the temperature decreases to the initial value. This reversibility of the change
in oil viscosity with temperature explains the formation of an oil bank.
When a steamfront moves through a reservoir, the temperature immediately ahead of the front
increases, thereby decreasing the oil viscosity. Oil is readily displaced from this high temperature
region to a region where the temperature may be considerably lower. In this low temperature
region, the oil regains its viscosity; thus, retarding the oil flow. Consequently, a large amount of
11
1500
Po
1000
5
8
500
0
0
50
100
150
250
200
TEMPERATURE,
300
350
oil accumulates as an oil bank. This bank, often observed in steamflooding heavy oils, is
responsible for high oil production rates and low water-oil ratios.
Thermat Expansion
Crude oil, like most liquids, increases in volume when heated. This change in volume,
increases its saturation and fluidity. The amount of swelling of an oil because of temperature rise
depends on the composition of the oil. Light oils expand more than heavy oils; thus, thermal
expansion is more effective in recovering light oils than heavy oils. Thermal swelling is an
important oil recovery mechanism in a hot water drive. Depending upon the type of oil and initial
saturation, as much as 10% of initial oil-in-place can be recovered by thermal expansion.
Solution Gas Drive
As the temperature ahead of a steamfront increases, the heated crude expels the dissolved
gases. These liberated gases expand, push the oil, and aid in oil recovery.
Gravity Segregation
Gravity segregation occurs because steam is lighter than oil or water. This difference in
density causes steam to rise to the top of a sand and spread out areally. The oil heated by steam
12
expands and becomes lighter and less viscous, permitting the steam to move quickly in the upper
part of a producing zone. Thus, the reservoir becomes divided into two layers, a steam-invaded
override zone at the top and a noninvaded zone at the bottom (see figure 1.2).
At first, the overriding steam spreads areally, but as steam injection continues, the steam zone
grows downward forcing the hot water in front of it. Thus, the oil at the interface between the
steam and hot water can be stripped from the reservoir rock and transported towards producing
wells along with the hot water condensing from the steam zone. With time, and at the expense of
recycling steam, an entire reservoir could be heated this way. However, with very viscous oil, this
process would not be economical because of inordinate amounts of time needed to achieve good
areal coverage.
Cyclic steam injection, also known as huff-n-puff, or steam soak, involves the injection of
steam into a producing well for a short time. The well is then shut in for several days to permit the
soaking of the reservoir by steam, and then placed on production (see figure 1.6).
The cyclic steam-process is similar to hydraulic fracturing. However, instead of increasing
the flow capacity of the reservoir, the viscosity of the oil is reduced. The result is the same. The
producing capacity of the well is increased whether the permeability is increased or the viscosity
reduced. Although the terms are sometimes used interchangeably, steamflooding has the same
relationship to the cyclic steam process as waterflooding does to hydraulic fracturing with water.
There is one difference, in that the heat injected is not produced except as sensible heat in the
produced fluids. .The injected heat is expended in heating and decreases the viscosity of the oil in
the reservoir. After repeated cyclic steam treatments, it is possible to displace some of the oil
between wells.
13
a. Steam Injection
14
STEAM
PACKER
The well is then shut in for several days to permit the soaking of the reservoir by steam.
During this soaking period, injected steam condenses as it distributes heat to a larger volume of the
reservoir. After the steam is soaked for several days, the well is allowed to flow under primary
production and is later pumped.
Cyclic steam process performance depends on existing reservoir pressure. Heat is needed to
make the heavy crude mobile, but energy is required for it to flow into a wellbore. Where some
reservoir pressure exists, injected steam can reinforce natural reservoir energy, increase the
pressure differential, and allow oil to flow naturally at economic rates and volumes.
Both injection and soak times generally increase as the total volume of steam injected
increases. During the production period, which may range from 1 to 7 months, the well pressure
decreases and some of the steam condensate vaporizes, driving heated oil toward the producer.
The well is produced until the decline in production warrants another treatment with steam.
When economical amounts of oil are no longer being produced, the whole cycle is repeated.
Usually three complete cycles are nsed in a single well. Response to cyclic steam injection
depends on .formation thickness, oil-in-place, volume of steam injected, and the number of
preceding cycles. Regardless of reservoir type, cyclic injection becomes less efficient as the
15
number of cycles increases, With each succeeding cycle, oil production declines, water cut
increases, and the cycle becomes longer.
After many cycles, large fractions of injected steam are produced as water, making water
handling important. From an economic standpoint, a good indicator of performance is the
produced oil to injected water ratio. When this ratio falls below 1, the project becomes marginal
and the cyclic steam process is frequently converted to steamflood.
REFERENCES
1. Pautz, J. F. and R. Thomas. Applications of EOR Technology in Field'Projects-1990
update. DOE Report, NIPER-513, November 1990.
2. Prats, M. The,mal Recovery. SPE Monograph Series No. 7, SPE of ATME, Dallas,
TX, 1982.
3. Burger, J., P. Sourieau and M. Cornbarnous. Thermal Methods of Oil Recovery. Gulf
Publishing Co., Houston, TX, 1985.
4. Boberg, T. C. Thermal Methods of Oil Recovery, John Wiley 6r. Sons, New York,
1988.
17
CHAPTER 2
CRITERIA FOR EVALUATING STEAM INJECTION PROSPECTS
INTRODUCTION
Several factors must be taken into consideration when evaluating candidate reservoirs for
steam injection operation. These include reservoir rock and fluid properties, crude oil
characteristics, field history and current well status. Tables 2.1 and 2.2 summarize the criteria that
must be considered in evaluating a prospect for cyclic steam and steamflood. Note that these
recommendations are general guidelines with which to identify target reservoirs for further study,
and the use of engineering judgment in the application of these criteria is advocated.
This chapter discusses each of these criteria and their relative importance to a reservoir
engineer evaluating the properties for a particular steam injection project. Although the screening
criteria for cyclic steam treatment and steamflooding are slightly different, the following discussion
is applicable to both processes.
Rock and Fluid Properties
Rock Type
Any formation that permits the injection of steam at an acceptable rate should be considered
for additional study. Since sandstone generally has high permeability and limestone and dolomite
have low permeability, steamflooding has mainly been applied to unconsolidated sandstone
reservoirs. However, steam has been injected successfully into a fractured consolidated sandstone
reservoir in Wyoming' and in a fractured dolomite reservoir in Southwest France.%
18
Water
Reservoir connate water properties are not critical. Water for stem generator should be chemically
treated so it is, soft, slightly alkaline, and free of oxygen, solids, oil, H2S (hydrogen sulfide), and
dissolved iron.
Reservoir
Thickness > 20 ft
Depth < 3,000 ft
Porosity > 25%
Oil-in-place > 1,000 bbllacre-ft
Permeability > 250 mD (millidmy)
Favorable Factors
1.
2.
3.
1.
2.
3.
4.
2.
3.
4.
19
Viscosity (Po)is not critical, but less than 10,000 CP(centipoise) oil is preferred.
Gravity is not critical by itself, but 10 to 20' API oil is preferred.
water
Formation water properties are not critical.
Water for steam should be chemically treated so that it is slightly alkaline and free from factors
such as hardness, oxygen, solids, oil, H2S (hydrogen sulfide), and dissolved iron.
Lithologv
Low clay content
Reservoil:
Thickness (h) > 15 ft
Depth < 4,500 ft
Porosity ($)> 25%
Oil Saturation (So) > 0.4
Oil-in-place > 600 bbl/acre-ft ($SO ) 0.08)
Permeability (k) is not critical by itself, but a value > 300 mD (millidarcy) is preferred.
Transmissibility,
>50mD-ft
CP
&tors Which I
1.
2.
3.
High porosity
High net to gross pay
High well density
4. High quality water
5 . Low fuel costs
6 . Usable existing wells
5 . Homogeneous formation
1.
2.
3.
4.
5.
6.
Steam is normally injected into shallow reservoirs. With increasing depth, heat losses in the
wellbore as well as to the overburden and underburden of the formation increases. However, the
incorporation of insulated downhole steam injection tubulars in planned project operations will
reduce the thermal losses to the overburden soil and therefore improve the process efficiency.
Normally, depths less than 4,500 ft are preferred for steamflooding. The deepest steam
injection to date has been in an 8,500 ft well in the Boscan reservoir in the Lake Maracaibo Basin
of VenezuelaP The use of insulated tubing at this field maintained heat losses at the design value
of 5% of the injected heat. Figure 2.1 shows the wellbore heat loss accompanying steam injection
through a 2 7/8-in. uninsulated tubing and an insulated tubing in a 7-in. casing. The well depth
was 3,000 ft, and the steam injection rate was 950 bbl/d of cold water equivalent. The steam
injection pressure was chosen as 1,800 psig. The overall heat losses were estimated using
Rameys method5 for both insulated and uninsulated tubing. From this plot, it is clear that there is
no technical justification to use depth as a criterion of ineligibility for steam injection. Rather, it
may be that the injection of steam into a deeper formation will violate an economic constraint.
Economic computations which consider the extra cost of insulated tubulars versus the savings in
fuel will have to be carried out.
The high pressures associated with deep reservoirs will also prohibit the use of steam. With
increasing depth, steam injection pressure generally increases with a corresponding increase in
steam temperature. If the reservoir pressure is higher than 3,200 psig, the critical pressure of
steam, steam injection is nearly impossible. Even at reservoir pressures close to 2,000 psig, the
operating pressure would be close to 3,000 psig. Additionally, with increase in pressure the total
heat of steam (sum of latent heat and sensible heat of steam) decreases.
Further, a high reservoir pressure may bring about a low steam injection rate. This will
affect the economics of the process because of increased wellbore heat loss, lower production
volumes, longer project life, and resulting increased heat losses to adjacent strata. Ideally,
reservoir pressure should be less than 500 psig.
21
4200
2400
1800
1200
600
0.5
1.5
2.5
Ti =-ki hi
PI
where ki is the permeability to the fluid phase i and Pi, is the fluid viscosity, and hi is the pay
thickness. The ratio ki@i is called the fluid mobility.
The fluid mobility is a function of the fluid, fluid saturation, and the displacement
temperature. Adverse transmissibilities of the injected and reservoir fluid render the steamflood
process ineficient. A low reservoir permeability and/or a very high oil viscosity results in low oil
mobility. The effect of low permeability on the process efficiency was discussed in the previous
paragraph. Scaled physical model studies of steamflooding for different oil viscosities indicate a
strong negative influence of oil viscosity on the efficiency of the steam drive process.
The problem of low oil mobility and reservoir transmissibility has been addressed by some
operators by injecting steam above the formation pressure and allowing the fractures to provide a
path between injector and producer. An example of this is the Saner Ranch fractures assisted
steamflood process.6 In summary, adverse fluid mobility and reservoir transmissibilities can be
22
modified by tailoring the process to the reservoir. As long as these modifications are economically
sound, there can be no restrictions on these parameters.
Stratification
Steamflooding generally works best in a massive sand with no stratification. However,
reservoirs with shale stringers are frequently encountered.
If the shale stringers are very thin (e2-ft thick) and continuity of the sand can be traced from
injector to producer, the formation can still be used for steam injection. If the shale breaks are
thick, the pay sands should be flooded separately. In such cases, a packer is used to separate the
steam injection for the upper and lower sands, thus avoiding injection into the shale break. If the
shale break is greater than 40-ft thick, flood the two zones separately.
A n isotropy
An anisotropic reservoir is one in which the reservoir properties vary areally. The most
common anisotropic effect is preferential permeability, which causes fluid to flow nonradially from
the injection well. As long as the injection and producing well are in communication, anisotropy
will not hinder steam injection.
If the preferential permeabilities have been found, the project can be designed to account for
this effect. For example, an inverted seven-spot pattern can be rotated and stretched (see
figure 2.2) along the major high permeability axis of an anisotropic reservoir to achieve uniform
steam breakthrough at the producing wells.
23
Oil Saturation
A minimuni oil content (the product of oil saturation and porosity) is necessary in order to
offset the energy requirements of a steamflood process. A rule-of-thumb in the oil industry says
that the product of oil saturation and porosity (9 X SOi) must be at least 0.13 or 1,000 bbl/ac-ft, for
steamflooding. Thus, if porosity is 0.2, the oil saturation should be at least 0.65. For light oils
that can be flooded to a lower residual saturation, the oil content must be greater than 0.08 or 600
bbl/ac-ft.
24
1
I
This combination of porosity and oil saturation implies that the reservoir should have enough
recoverable oil to cover the energy requirements of the process, and to supply additional
production to make the process economically attractive. Dugdale and Belgrave9 performed a
detailed energy analysis and concluded that it is possible to perform steamflood at oil content
values below 0.1 in heavy oil reservoirs. Thus, no general guidelines on the minimum oil content
requirement for a feasible steamflood project can be given. Individual reservoirs must be analyzed
independently. The only valid guideline for oil content is that it should be high enough to furnish
the energy needs of the process and supply sufficient additional oil production to make the process
economical.
Clay Content
Clay content of a reservoir is not a restriction in selecting a steam injection candidate.
Reservoirs with or without nonswelling clays can be used for steam injection projects. Some
reservoirs contain water sensitive clays, such as montmorillonite clays which swell when contacted
with injected steam or water. The swollen clays greatly reduce the formation permeability.
Nevertheless, this type of reservoir can also be used for steam injection if the clays are properly
stabilized. For example, in a California reservoir containing swellable clays, swelling was
prevented by injecting a saturated potassium chloride (KC1) slug near the wellbore. In addition, a
0.5% (wt) KCI solution was continuously injected into the generator feedwater downstream of the
water softener. This approach was effective in treating swellable clays so that steam injection rate
could be maintained. If the reservoir contains swelling clays which cannot be controlled and the
effective permeability would be reduced to less than 100 md, this reservoir should be excluded
from steam injection.
Crude Oil Characteristics
Gravitv -Crude oils with gravities from 6" to 50" API are amenable to steam injection.
Steamflooding has usually been successful in heavy oils of 8' to 25" API gravity. In the range
between 26"to 50" API, steam distillation is the major recovery mechanism. Most fields producing
oil with gravity higher4than 40' API are deeper than 5,000 feet, a depth that was once impractical
for injecting steam. However, with the development of insulated tubing, it is now possible to
recover light oils from these very deep formations.
Viscosity - Since a minimum oil mobility is required at reservoir temperature for a
displacement process to operate, an upper limit is usually imposed on oil viscosity for the steam
injection process. These values are in the range of 15,000 CPfor steamflood. For steamflooding,
Yan et al.10 reported a decrease in recovery from 32 to 29% when viscosity was increased from
500 CP to 4,000 CPfor a 15-ft thick reservoir. The corresponding steam oil ratios were 6.6 and
25
8.8, respectively. Doscher,11 on the basis of scaled physical model studies, concluded that very
viscous crudes could not be recovered economically. However, steam injection has been
successfully carried out in Canadian reservoirs containing highly viscous crudes. These include
Cold Lake, Primrose, Peace River, etc.
Pattern Configuration
While a producing well in a confined pattern can capture all movable fluid inside a pattern,
such patterns have an unfavorable producing to injection well ratio. On the other hand, an
unconfined pattern has a favorable producing to injection well ratio, but it loses a large portion of
movable fluids outside the pattern. Therefore, if possible, multiple patterns with more than one
injector and producer (such as the pattern shown in fig. 2.3) are preferred.
26
3.75 ACRES
equipment necessary for the treatment will be described in detail in a subsequent chapter.
Enough fuel must be available to fire the steam generator so that high quality steam can be
generated continuously. Fuel such as natural gas or diesel oil can be used. After oil production
reaches a steady level, part of the produced crude could be used as the generator fuel.
Water D isposa 1
As mentioned previously, produced water can sometimes be used as generator feedwater
after treatment. "he rest of the produced water, after the removal of the oil particles must be
disposed of properly. The produced water after treatment, for example, can be disposed by
injecting into a disposal well completed in a different formation. If water disposal is a problem in
the area, the field should not be considered for steam injection.
27
SUMMARY
Guidelines for selecting candidate reservoirs for steam injection are presented. These are
general guidelines that reflect the current technology and economic climate. The criteria presented
should not be regarded as sacred, Le., that all criteria must be met before the process can be
considered for a particular reservoir. Each reservoir shouid be examined closely on an individual
basis and engineering judgment applied before a decision can be made to pilot testa reservoir.
REFERENCES
1 . Chappelle, H. H., G. P. Emsurak and S. L. Obernyer. Screening and Evaluation of
Enhanced Oil Recovery at Teapot Dome in the Shannon Sandstone: A Shallow, Heterogeneous
Light Oil Reservoir. Pres. at the SOC.of Pet. EngJDOE Symposium on Enhanced Oil Recovery,
Tulsa, OK. Apr. 20-23, 1986. SPUDOE paper 14919.
- Lacq Superior Field. J. Pet. Tech., v. 34, No. 4, pp. 873-80, April 1982.
6 . Stang, H. R. and Y. Soni. The Saner Ranch Pilot Test of Fractured-Assisted Steamflood
Technology. Pres. at the SOC.of Pet. Eng. Annual Tech. Conf. and Exhib., Houston, TX,Sept.
16-19, 1984. SPE paper 13036,
28
CHAPTER 3
PROJECT PLANNING
INTRODUCTION
Steam injection is typically a high cost, low profit operation-hence proper planning is
important to achieve maximum economic benefits. The success depends on proper selection of
reservoirs for injection, sound program planning, consideration of capital expenditures and high
operating costs involved, and an awareness of the degree of success which can reasonably be
expected.
The objective of this chapter is to appraise operators of the various steps which must be
considered when planning a steam injection pilot.
Reservoir Selection
In selecting a reservoir for the steam injection process, all readily available information must
be taken into consideration. These factors must then be carefully appraised and the prospect
selected. Criteria for evaluating steam injection prospects are discussed in detail in Chapter 2.
Some of the parameters involved include depth, oil-in-place, porosity, inhomogeneities, sand
thickness, oil mobility, and special conditions that may influence steam injection mechanisms such
as gas cap and bottom water.
Depth
While depth is a controlling factor costwise, there is no trend or concentration of projects in
any particular depth range. Projects have been successfully implemented at depths from a few
hundred to over 4,500 ft. A deeper depth requires a higher injection pressure, which means that
high-pressure steam-injection equipment may be required. Also, since the temperature is
determined by steam pressure, higher steam pressure translates to higher temperature.
Disadvantages of high steam temperature include greater heat losses, lower recovery, casing or
tubing failure in older wells, and accelerated corrosion of well equipment.
On the favorable side, increased depth means that a greater pressure drawdown can be
applied to producing wells. This will yield higher producing rates-for a given crude in a specific
formation.
Oil-In-Place
Oil-in-place at the time of steam injection initiation is another important consideration. One of
the most frequently &ked qu ons is: How much oil must be in place to support a steam injection
operation? There' is no sim
swer to this question. The only thing that can be said about oilin-place is that it should be high enough to sustain a cash flow sufficient to pay for high operating
costs and provide an acceptable return on the capital invested. Chances for an economically
successful steam injection operation increase as oil-in-place increases above 600 bbl/acre-ft.
29
unsuitable for steam injection, the cost of drilling and completion of wells for steamflood use
should be estimated. At this stage, maximum use of applicable correlations and rules-of-thumb
should be made.
30
A complete economic study should consider both capital investment and operating costs,
equipment and completion problems, and the cost of complying with regulatory requirements.
Factors that must be considered in the planning stage include: type and quality of available fuel;
water availability; required water treating and sizing of water handling equipment; necessary steam,
injection and production facilities to handle anticipated volumes of injection and production fluids;
condition of existing wells and equipment; additional well equipment; drilling and remedial costs;
safety precautions; availability of market outlets for the produced fluid; and waste disposal costs.
These and other equally important points must be carefully evaluated to arrive at an economic
justification for starting a project. Table 3.1 lists major capital and operating costs that must be
considered in the preliminary economic evaluation. Economic analysis should also include the cost
of regulatory compliance expenses.
Loss of revenue due to operational problems should be included in the projected annual cash
flow. The likely operational problems include: mechanical failure of steam generation equipment,
sand production and attendant pumping problems, emulsion problems, casing failure in old wells,
surface equipment malfunction, etc. Finally, in making economic evaluation, the operator should
consider special tax breaks given for enhanced oil recovery projects. If the economic analysis is
improperly done, the profitability of the project is jeopardized and the possibility of obtaining
adequate financing is materially reduced.
31
Eouioment
1.
Steam generators
3.
4.
1.
2.
3.
4.
1.
2.
3.
4.
5.
Injection wells
Producing wells
Water disposal wells
Water supply wells
Water treatment costs (include production or purchase price of raw water, transportation cost, chemicals
necessary for treatment to generator quality, and labor to accomplish all of this)
Steam production costs (include fuel cost, power for generators and fuel oil heating, and labor to operate
equipment and system)
Well preparation cost (include cost of pulling or downhole workover)
Water disposal well (include waste water treatment cost)
Emulsion treatment cost
3.
4.
5.
6.
32
Geological Information
Reservoir depth and thickness
structure
Reservoir Data
- Rock and fluid properties
- Pressure, temperature, and saturations
- Fluid withdrawals
- Recovery mechanisms
Feasibility Study (Idealized Conditions)
Computer analysis based on existing or readily-available information
Determine expected producing rates and fuel consumption versus time
Economic Factors
- Preliminary evaluation based on expected producing rates, crude oil selling price, capital and operating costs,
and taxes.
*
*
Drill core holes to get samples of reservoir rock and fluids. and obtain logs.
Laboratory measurement of rock and fluid properties
Porosity, permeability, and compressibility
Water-oil and gas-oil relative permeabilities at elevated temperatures
Rock composition (swelling or dispersible clay material)
Effect of salinity on permeability
API gravity of crude oil
- Oil viscosity versus temperature
Distillation characteristics (light oils)
- Chemical composition of oil (sulfur content)
- Analysis of produced water and other source waters
33
Site selection(s)
Initiate environmental studies
Design facilities
Steam generators
Fuel
Pollution control
- Steam distribution system
Surface lines
Insulation
- Oil gathering system
Flow lines
Production tanks
De-emulsification equipment
Upgrading if necessary
- Water treating system
Collection tanks
Flotation cells
Filtration equipment
Softening equipment
Pumps
Monitoring system
Process Monitoring
Environmental Monitoring
Economic analysis
- Drilling and completion costs:
(New vs existing wells)
- Initial investment:
Steam generators (lease or buy)
Facilities
Pumping equipment
Wellhead and downhole equipment
- Operating costs:
Fuel for generators
Water plant operation
Electrical power
Wages
- Repair and maintenance costs:
Sand production workovers
Equipment repair
System leak repair
Taxes
Estimated gross revenue
34
*
*
Evaluation
- Economic appraisal of pilot
Pilot installation
- Drilling and completion
- Injection system
- Production gathering and treating system
- Metering and monitoring system
- Waste disposal system
Pollution control system
*
*
*
*
*
these more or less conventional practices and the enhanced recovery processes being developed
today are as follows:
1. More detailed and exacting laboratory screening tests to determine process applicability.
2. More specialized and costly equipment, some of which may have to be designed. New
well completion procedures and materials may be required.
3. Proper project evaluation requires the' taking of more detailed and complex data,
necessitating the installation of-specialized and more costly monitoring equipment.
4. All of the above requires more and better trained field personnel in order to evaluate and
cope with operating problems and environmental considerations.
These differences, along with many others such as special chemicals and storage facilities,
etc., dictate the need for a carefully controlled and monitored field pilot test.
One of the most difficult phases in developing any new oil recovery process is the design,
implementation, and interpretation of field pilot tests. They are expensive, yet represent only one
of many possible sets of operating conditions. One must take every possible precaution to choose
35
an optimum set of conditions and to design a pilot that provides maximum opportunities for
interpretation.* To summarize:
Pilot Test
SUMMARY
Steam injection is a practical, tried and proven method of increasing both rate of recovery and
ultimate recovery from certain types of reservoirs under particular conditions. A successful
application of the steam injection process requires a systematic investigation starting with
preliminary screening and culminating with full-scale field development. Economics must be
considered in each step. Rewards from a successful steam injection project will be substantial.
Unsuccessful projects can be avoided if proper consideration is given to all aspects of the
operation.
A good understanding of the reservoir and recovery process is essential for successful
implementation of a steam injection project. Before starting a steam injection project, an operator
should have a good estimate of the capital money required, have access to it, and be willing to
spend it. The operator should keep in mind that all investments involve a certain degree of risk,
and steam injection projects are no exception. The risk of failure of a steam injection project is
minimized if highly reliable and unbiased data are used in preparation and evaluation. In the final
analysis, experience, judgement, and knowledge of the reliability of the input data should be used
as a guide in deciding whether the proposed economic gains from a project justify the expected
risk.
REFERENCE
1 . Pursley, S. A., R. H. Healy and E. J. Sandvik. A Field Test of Surfactant Flooding,
Loudon, Illinois, J. Pet. Tech., v. 25, No. 7, July 1973, pp. 793-802.
37
CHAPTER 4
ECONOMICS OF STEAM IN ECTI,
INTRODUCTION
The basic advantage of steam injection over that of other recovery methods is increased
ultimate recovery of lower gravity viscous oil in a shorter period of time. Steamed wells are
usually allowed to produce to their maximum ability; hence, rapid payouts are possible. In
addition, the process can be contained within a small area, thus making unitization unnecessary.
Steam injection has evolved into a mature process over the past 30 years. This advancement in
technology has removed many of the unknown factors associated with the process and has
improved production forecasting. From the lending institution prospective, this has lowered the
risks associated with the process and has permitted better evaluation of steam injection projects for
loan purposes.
In spite of the technological advancements, steam injection projects are still considered as
high-cost, low-profit operations. The production of heavy oil by steam injection is a more
complex and expensive undertaking than conventional oil recovery practices. Steam injection
requires high capital investments, and operating costs are high. Maintenance costs are generally
high since reservoirs that make good steam injection candidates are usually very unconsolidated
and lend themselves to sand production and require above normal workovers. Furthermore, the
market price realized for the produced crude oil needs to be reduced for the lower quality product
and any costs incurred for delivery to refineries. The very best steam recovery fields in California
are barely profitable operations at current heavy oil prices. Hence, a careful analysis of the
economics of the proposed project must be performed to determine feasibility. Assuming that the
feasibility of a project had been established and that a flood has been started, many other factors
will constantly arise that must be balanced against each other. Thus, a steam injection project
requires ;I constant study of engineering and economics, and the task is not completed until the
project is ready to be terminated.
The purpose of this chapter is to present an overview of the economics of a steam injection
project. Some of the factors which govern the economics of a steam injection operation are
outlined. Costs of steam injection operations are presented.
Economic Factors in Steam Injection Operation
An economic analysis of any injection operation whether it is waterflood or steam injection is
largely a balancing of the costs for such an operation against the value of the additional oil
recovery. The factors that govern the economics of individual steam injection operations are often
numerous and varied. Many of these factors are peculiar to individual areas or operators, and it is
not the purpose of this discussion to consider such factors and how they might affect economic
38
conditions of steam injection. Likewise, there are factors that more or less frequently impact the
economics of individual steam injection operations, and these will be presented briefly. These
factors include (a) cost of the prospect, (b) reservoir depth, (c) oil content of the reservoir, (d) net
production pay, (e) demand and price for the crude, (f) abandonment, (8) availability and quality of
raw water, (h) availability and cost of fuel and power, (i) oil transportation facilities,
(j)location, (k) size, (1) existing lease terms, (m) water disposal facilities, (n) local environmental
regulations, and (0)availability and cost for labor and field service.
The order of importance of these factors is not necessarily the order of their listing. To
illustrate how these factors influence the economics of an individual operation, a brief discussion
of the related factors is in order.
The value of a thermal recovery prospect often is difficult to determine and may vary greatly.
The effect of this factor is often disregarded in considering the economics of a steam injection
operation. In areas where the possibilities of thermal recovery are not well established, stripper
well properties can be acquired at little more than salvage value. Under such conditions, the cost
of acquisition is of no significant importance, even though the true value of the prospect may be
great. Conversely, in an area where steam injection operations are well established, and where the
cost of such prospects more nearly reflect their true economic value, the cost of acquiring desirable
steamflooding acreage is likely to be very high, almost prohibitive, to the average operator
interested only in the production of oil. The thermal acreage of San Joaquin Valley is a good
example of this later condition. Thus, the economic values of thermal recovery prospects may vary
from salvage values to values representing a significant portion of the profit that is to be realized
from the development of the prospect. This value should always be included in the economics of
any steam injection operation.
In the development of a steam injection operation, perhaps the most important of all factors
that affect the economics are: (1) the oil content of the reservoir-measured in terms of barrels of
oil per acre-foot of reservoir; (2) the depth to the producing formation; and (3) the demand and
wellhead price for the oil. The economic effects of these factors are interrelated, and one can
hardly be considered without the others.
The oil content at the beginning of a steam injection operation is a key indicator of the
likelihood of the economic success of the project. Steam injection recovery operations are energy
intensive, and proximately one-third of the produced oil is used to generate steam. Total oil
recovery must be sufficient to pay all costs of recovering that oil, at the prevailing oil price, plus
provide an acceptable rate of return on the investment. Hence, minimum amounts are needed to
permit significant incremental recovery and to exceed fuel requirements. Chances for an
economically successful steamflood increase as oil-in-place increases above 600 barrels per acrefoot.
39
r
The cost and efficiency of steam injection processes are strongly depth dependent. The
average 1990 Kern County, California, steam injection project well completion costs are shown as
a function of depth in table 4.1. From this table, it is clear that well costs are largely controlled by
depth. Deeper depths require higher injection pressures which means that high-pressure steaminjection equipment may be required. Also, the higher pressures required for deeper reservoirs
may lead to higher steam temperatures and, therefore, higher reservoir heat losses. Furthermore,
the latent heat content of high-pressure steam is low. Other disadvantages of high-pressure, hightemperature steam include casing or tubing failure in older wells and accelerated corrosion of well
equipment. These factors increase operating and maintenance costs. Inasmuch as development,
operation, and maintenance costs for intensive steam injection operations increase with depth, there
is a minimum recovery per acre of oil for a given price that can be considered profitable. As the
depth increases or the price of oil decreases, higher recoveries per acre are necessary to offset
increased costs. In the same fashion, as the depth decreases or the price of oil increases, the
minimum recovery per acre to obtain profit decreases. These are general relationships, the other
factors being equal. In any proposed steam injection operation, however, these factors must be
considered and carefully weighed to deternine the probable economics of the operation.
Other factors encountered in the proposed development of a steam injection operation can be
as important to the economics of the operation as the factors just described. In many old
properties, particularly extremely old properties, many of the wells have been improperly plugged
and long abandoned. The condition of these abandoned wells can be of major importance to a
thermal operation. The old Midcontinent fields are good examples of this later condition. Geysers
of steam, hot water and hot oil were formed at the surface after implementation of a steam
TABLE 4.1.
Well depth,
ft
0-250
25 1-750
751-1,250
1,251- 1,750
1,751-2.250
2.25 1-2,750
2,751-3.250
Eastern
Kern County
inj. well Cost. $
Western
Kern County
inj. well cost, S
37,000
73,000
89,000
121.000
4
4
.
m
160.OOO
2 0 0 . ~
240,000
79,000
94,000
141,000
189.000
236,000
283,000
process at shallow depths in Missouri and Oklahoma as a result of communication to the surface
via improperly plugged and abandoned wells. These unfortunate Occurrences can be remediated at
added cost by drilling new wells and changing injection patterns. Poorly plugged and
40
undocumented abandoned wells are sources of tremendous loss of valuable oil and steam. Such
losses reduce expected recovery and add materially to costs. When an operator is planning steam
injection process implementation in old fields, a thorough investigation is necessary, with
documentation, to try to locate all old abandoned wells for the prevention of surface contamination
and loss of recovery. With the increasingly stringent environmental regulations, such surface
contamination materially affects the economics of an operation due to cleanup costs and penalties.
While far from being universally practiced, it is recommended that any and all wells drilled within
the area to be flooded, be reopened and replugged even the so called dry holes, in order to
prevent escape of oil andor injected steam to the surface. Such a strategy in the long run will be
less costly than cleanup and surface restoration costs. In view of the added costs of locating and
replugging abandoned wells in older fields, it is obvious that the oil recovery per acre from such
fields must be higher than that of younger properties for the process to be economical.
The availability of an adequate supply of good quality water at a reasonable cost for steam
generation is important to steam injection operations. Large-scale steamflood operations require
large volumes of good quality water, and usually a shortage of water seriously affects the rate of
oil recovery and the economics of a operation. The cost of furnishing suitable water for steam
generation varies considerably, depending upon the source of water and the amount of water
treatment required. This varies not only from field to field but also from day to day, particularly if
produced water is recycled. When sufficient volumes of water can be acquired at reasonable cost,
the total cost of furnishing feed water for steam generation is usually small compared to steam
generator fuel cost and has minimal effect on the economics of steam injection operations. In areas
where water is scarce and extremely valuable, its cost could be prohibitive.
Steam injection operations are energy intensive, and fuel costs greatly affect the economics of
an operation. Natural gas and lease crude are the most common types of fuels used to fire thermal
EOR steam generators. In a typical California steam injection operation, fuel costs account for
more than 50% of operation and maintenance costs. Gas-fired steam generators are less expensive
to install and operate than oil-fxed steam generators. In many areas, gas is not readily available,
and lease crude is burned to generate steam. This reduces the volume of fuel available for sale.
Since improvement in steam generator thermal efficiency reduces fuel requirements and reduces
cooling loads of flue gas scrubbers, generator efficiency has an appreciable effect on the economics
of an operation. For example, the fuel requirements of an oil-fmd 50 MM Btu/hr steam generator
(the standard oil field unit) can be reduced roughly by 5,300 barrels of oil per year by increasing
the thermal efficiency by about 5%. Thus, even a marginal improvement in thermal efficiency can
materially affect the economics of an operation and may result in prolonging its economic life.
The location and size of a steam injection operation also can affect overall economics. There
are many factois inherent to the location of the property which affect development and operating
41
costs. It is more expensive to implement a steamflood project in urban areas such as Los Angeles,
California, than in rural area such as rural Kern County, California. Availability and cost of labor
and oilfield services often depend on location. Oil pipeline outlets or other means of oil
transportation often have an influence on the price received for oil at the wellhead. The size of the
operation alone may have considerable effect. Usually, a large operation can be developed and
operated for lower unit costs than a smaller operation. The shape and size of the area often affect
development and operating costs. A square area usually is less expensive to develop than a long
narrow area, such as that of the shoestring sands of eastern Kansas. Well spacing also affects
development costs. Spacing controls the number of injection and production wells to be drilled
and this, in turn, largely controls the development and operation cost per acre. In thermal
operations, wells are closely spaced to minimize heat loss between injectors and producers and to
improve areal and vertical sweep efficiency. In selecting the well spacing, the improvement in
recovery must be balanced against development and operating costs.
Existing wells may or may not be used. One should not hesitate to use an old well as a
producer, provided it is located not more than 10% off the desired geometrical location and
provided, of course, that it is in good physical condition and can handle thermal stresses resulting
from exposure to hot produced fluid. In most instances, old wells are not in good physical shape
and require either cleaning out, replacement of tubing and casing, or some other major remediation.
Since steam injection wells are subjected to high thermal stresses due to exposure to live steam, old
wells should not be used as injection wells, unless they are worked over. The cost of this work
seldom justifies the expense involved, even if it is very close to a proposed pattern input. Usually,
it is less costly to drill and complete a new injection well than to convert an old exiting well to an
injector.
Existing primary lease terms and obligation to non-operating interests are also important in
thermal operations. Obviously, if royalty interests are high or nonoperating interests do not
participate in the cost of the operation, the profit is materially reduced. Geology of the reservoir or
field also significantly impact the economics of the thermal operation. For example, the shoestring
reservoirs of southeast Kansas, though very high in oil saturation, unfortunately are highly
compartmentalized and discontinuously bedded, allowing very little movement of steam through
the producing formation. Such floods will seldom be a financial success. Steamflood is seldom
economically viable in formations with a gas cap or in reservoirs with an underlying aquifer.
Steam injected near the gas cap almost invariably moves preferentially through the gas cap,
allowing little movement of steam through the zone of oil saturation. Similarly, in oil reservoirs
with an under-lying aquifer-if steam enters the aquifer-all injected heat will be dissipakd and the
steam process may not be economical. Integrated engineering and geologic analysis prior to
process implementation will improve the chances for success.
42
Federal and local environmental regulations and requirements also significantly impact the
economics of thermal operations. For example, in California the air emission standards are so
stringent that the cost of compliance with the state air quality standards is becoming increasingly
prohibitive for many steam injection operators. Obtaining necessary permits to operate the steam
generators is the single most difficult and costly aspect of thermal recovery operations in
California. Disposal of project-generated wastes such as scrubber liquor, scrubber sludge, water
softener regeneration brine, etc. calls for special measures and can materially impact the economics
of an operation. The cost of compliance with environmental regulations must be taken into
consideration in any analysis of the economics of steam injection operations.
The brief factors which have been presented, not only control the economics of steam
injection operations in general but also offer a sound basis for determining the profitability of the
proposed operations. A thorough analysis of these and other factors, individually and collectively,
not only determines the merit of thermal EOR in general but also the extent to which these
operations are justified. Today, thermal recovery of oil is based on sound engineering and proven
practices. It is usually possible to determine approximately how much oil can be recovered by a
thermal method from a given reservoir. With this information on hand, thermal recovery is largely
a problem of economics, i.e., balancing the total cost of recovery against the estimated value of the
oil to be recovered. However, predicting the economics of thermal recovery is quite complex and
requires extensive knowledge and experience. Although the literature on the theoretical and
reservoir aspects of steam injection operation is voluminous, steam injection recovery costs are not
as well documented. The published economic studies?-5 of necessity, generally present overall
costs rather than specific costs. In this report, an attempt is made to present a more detailed
breakdown of equipment costs.
Estimation of Economically Recoverable Oil
It is obvious that the economics of any steam injection project are based on the simple
formula:
Gross Profit = Gross Revenue - All Costs
Estimation of gross revenue requires a reliable estimate of production. Several steamflood
recovery models have been developed for use in estimating incremental ultimate oil recovery and
potential producing rates under various operating conditions. The U.S. Department of Energy
(DOE),through the DOE Bartlesville Project Office (BPO), has made available to interested
operators and others four steam-drive enhanced oil recovery models that may be acquired free of
charge.* The four models include algorithms for calculating oil and water production rates,
optimum steam injection rates, surface and wellbore heat losses, and a basic economic analysis
package. A brief description of the features of these models and their limitations is presented in
appendix 4-A.
44
Cost Estimate
To determine whether a project will be economically successful, the costs specific to steam
injection operation need to be analyzed in detail. These costs are grouped into development costs,
operational costs, and acquisitiodlease costs. Administrative costs are excluded. Steam injection
project costs are detailed in the following. All costs are for a typical 1,000-ft, shallow, heavy oil
well in California. Unless otherwise noted, all costs are in 1991 dollars.
Development Costs
Well Costs
The cost estimates for injection and production wells are shown in Tables 4.3 and 4.4.
These cost estimates are based on the Kern County, California, assessors1 estimation of the cost of
drilling and equipping new steam injection and production wells (see Tables 4.1 and 4.2), and the
Independent Petroleum Association of America (IPAA) Cost Study Committee1 Index of Drilling
and Equipping Wells.6 The IPAA Drilling Index for the years of 1983 through 1989 is shown in
Table 4.5.
Other capital costs associated with a steam injection project include: (1) source water wells
and distribution lines; (2) water treatment facilities; (3) fuel supply facilities; (4) steam generator;
(5) flue gas scrubber system; (5) steam distribution system; (6) wellhead injection facilities; and
(7)surface facilities. The capital costs presented in the following section were gathered from
various sources including equipment vendors and California steamflood operators.
45
TABLE 4.2. - Kern County, California, New Production Well Costs, Excluding
Downhole Pump'
Well Depth,
ft
Western
Kern County
prod. well cost. $
Eastern
Kern County
prod well cost. 0
TABLE 4.3.- Cost to Drill and Equip a 1,000-ft Steam Injection Well (in 1991 Dollars)
0 19,500
6,500
2,600
700
Fuel
5,300
700
35300
Total drilling a s l s
LonPinn:
3.500
w:
1,000
600
2,200
1,100
Supervision
1.100
Overhead
46
2,800
2,800
7,300
900
600
12.300
4,000
9500
$85,000
25,000
$1l0,Oao
TABLE 4.4. - Cost to Drill and Equip a 1,000-ft Production Well (in 1991 Dollars)
m:
Payment to drilling contractor
site preparation
Transportation and setup of rig
Fuel
700
5,300
700
$35.300
a:
and wireline evaluation services
$3,500
&r,&y
Side wall sampling, base charge
Sampling, Wsample
for 50 A zone,samplel5 ft. 10 samples
Transport
supervision
Overhead
:
$2,000
2,800
7,300
900
supervisim
Overhead
Casing ?'trea+d and mu led, S12.3ift 1,ooOA
Production tubing 2-3/8", &.OB 1,OOO A
600
12,300
4,000
$82.800
Subsurface rod pump assembly with gas ancha, 3/4" API class C sucker rod
228,000 in Ibtorque API 228-213-86 pumping unit
Total cust of production well
$11,000
32UQQ
$116,800
47
TABLE 4.5. - Index of Drilling and Equipping Wells Unadjusted for Depth6 (1984 = 100)
Weight
Payments to drilling contractors
pnrrbmed Items:
Road site preparation
Transportation
Fuel
Drilling mud and additives
services
Directional tilling services
Perforate
Famation testing
Cement and cementing services
Casing and tubing
Casing hardware
Special tool rentals
Drill bits and reamers
Wellhead equipment
Mix. equipment and supplies
1983
1984
1985
1986
1987
1988
1989
29.8
114.3
100.0
101.0
85.9
64.4
77.3
73.5
6.3
2.9
1.2
5.8
96.7
100.0
101.8
107.5
100.0
100.0
100.0
100.0
101.7
100.0
95.3
100.6
103.6
100.0
58.4
100.6
106.2
100.0
65.4
100.6
108.5
100.0
58.8
100.6
110.1
100.0
68.8
100.6
100.0
0.5
100.0
100.0
100.0
100.0
100.0
100.0
87.4
100.0
88.9
133.2
104.6
138.8
164.1
4.1
0.8
2.0
4.2
4.3
16.8
95.6
101.6
100.0
105.9
107.4
105.2
100.0
100.0
100.0
100.0
100.0
100.0
67.4
98.8
100.0
89.3
86.5
92.0
920
106.0
98.8
103.9
99.1
97.4
108.1
96.8
73.0
98.8
100.0
94.0
90.8
98.8
98.8
100.9
101.2
101.8
99.8
100.4
106.5
99.3
79.5
104.0
151.7
95.4
97.5
120.0
162.0
116.5
105.5
103.6
99.6
109.3
102.9
15.7
115.4
182.3
101.2
102.3
129.9
129.9
102.0
118.8
106.6
106.0
103.5
115.2
105.3
94.4
97.5
106.4
111.8
0.7
_.
105.2
10.0
_.
.
.
100.0
100.0
100.0
100.0
100.0
100.0
100.0
89.9
98.8
100.0
94.0
92.7
98.8
98.8
100.0
100.4
103.8
100.8
98.1
103.3
100.4
100.0
98.7
Plugging
3.1
1.6
2.0
2.5
0.7
6.1
103.3
99.9
107.2
100.9
98.5
100.3
97.9
Sub-TotalPurchased Items
70.2
102.4
3.0
120.0
$20,000. A 1,000-bbl soft water storage tank with gas blanket will cost $30,000. Produced water
treatment and disposal cost will total $80,000.
Steam Generation
The operating characteristics of a steam generator are given in Table 4.6. Steam generator
and accessories costs, excluding that of pollution control equipment, are given in Table 4.7. The
steam generation operation cost, including water and fuel costs, are also shown in Table 4.7.
Labor costs are taken from reference 7. A typical 50 MM Btdhr steam generator burns
approximately 7 gpm (gallons per minute) of crude oil and will require approximately 17 scf/min
(standard cubic feet per minute) of air (52,280 lbm/hr air) for complete combustion. A flue gas
scrubber (pollution control equipment) to handle approximately 25 scfm flue gas costs $360,000.
The operations and maintenance cost of the scrubber is approximately $200,000 per year ($0.2 per
barrel of produced team).^
48
Heat input
Heat output
Design capacity
Average daily output
Annual output
Steam quality
steam Lwnditioos
Water requirement
Electricity requirement
Fuel requirement
6.2MM BtulbM.
TABLE 4.7. - Conventional Oil-Fired Surface Steam Generator Costs4 (1991 Dollars)
rn
Steam Generator (50 MM Btu/hr unit)
Steam piping. valves, insulation
$ 500,000
IE240.000
Total
Power (7gkwhr)
Maintenance (4% of cepital investment)
Operating labor (1.5 aperatdshift 0 $17/hr)
Overhead
Water (including purchase and treatment cost) 0 ebb1
Fuel (74,000bbl/yr @ $14/bbl)
Total steam generator -rating cost
Steam generation cost = $1.49 per bbl-steam
740,000
$ 22,400
$ 29,600
$ 184,000
$
$
184,000
62,000
51.036.000
$1.518,000
49
50
removal, pH control chemicals, and other chemicals as required. In Kern County, California, it
costs approximately 4 cents to treat 1 bbl of raw water containing 1,000 ppm (parts per million)
total dissolved solids. Details of water treatment costs can be found in Chapter 5.
Financial Costs
Royalties and taxes (federal and state income taxes) make up a substantial portion of the cost
of thermally produced oil. Hence, they must be properly accounted for in economic analysis.
Income and severance taxes vary from state to state. The current federal tax code must be
consulted for federal income taxes, investment tax credits, intangible drilling costs (IDC), and
depreciation on capital equipment.
Though royalties may range from 10 to 25%, most steam injection projects are on old leases
where a royalty rate of 12.5% on gross revenue usually applies. As previously mentioned, the
profit potential of a steam injection project is severely impacted by a high royalty rate.
51
The various oil-production-specificstate taxes are shown in Table 4.8. These can be used to
arrive at oil production economics on a state-specific basis. The federal corporate income tax for
the year 1990-91 is 34%. The current federal tax law permits domestic independent producers to
fully deduct intangible drilling costs (IDC).
An IDC is any cost that is necessary for the drilling and completion of a well but which in
itself has no salvage value (e.g., wages, fuel, repairs, supplies, transportation, etc.)
Other charges that must be included in the financial analysis of a steam injection project
include (a) cost of borrowing the necessary capital and (b) depreciation of capital equipment.
Approximately one-third of the wellhead costs goes toward the financial costs of the operation.
The suggested average life for steam injection operation related equipment is given in Table 4.10.
Capital charge rates, as arrived at by Rand Corporation4 for steamflood projects, are given in
Table 4.9. The annual capital charge rate accounts for (1) the cost of capital, (2) depreciation, and
(3) federal and state income taxes including investment tax credits. This table is based on 1982 tax
rates and presented here for illustration purposes only. The details of its calculation are presented
in Appendix 4-B. This appendix should be consulted for estimating capital charges for new
projects.
Table 4.9 is based on the following assumptions: (a) combined federal and state income tax
assumed to be 50%, (b) production and local tax rate assumed to be 10% of gross revenue. All
investments are assumed to use 100% equity financing with 10% minimum acceptable real rate of
return. Equipment life (book life) is assumed to be equal to project life. The tax life is assumed to
be 5 years with straight line depreciation. A 10% investment tax credit is assumed in the
calculation.
SUMMARY
Thermal recovery operations require a sizeable capital investment, and most components
require much more maintenance than normal oilfield equipment. The success of an individual
steam injection operation is affected by several factors. However, the principal factors are the
oiVsteam ratio (barrel of oil produced per barrel of feedwater injected as steam), the depth to the
producing formation (deeper formations have larger heat losses and, at the same time, less latent
heat content in the injected steam), and the price and demand for the oil. Other factors may become
extremely important and affect the economics of the operation. As examples, in certain areas the
quality of feedwater is so poor that the cost of water treatment may be excessive; or the cost of
furnishing fuel for steam generation tends to make total operating costs higher. Hence, an
evaluation of the different factors which could influence costs must be made before any steam
injection operation is undertaken. A complete engineering and economic study of a prospect will,
52
TABLE 4.8. - Comparisons of Economic Factors Affecting Oil Production From Selected
Oil Producing States
Missouri2
New
Mexico'
12.5
Site
specific
12.5
Site
specific
12.5
Site
specific
12.5
Site
specific
7.04
None
3.754
3.153
None7
None
Kansas'
Oklahoma'
12.5
Site
specific
4.333
Cdo'
North
Dakota'
12.5
Site
specific
12.5
Site
specific
12.5
Site
specific
None
4.64
2-55
53.6
None
None
Stripper
wells
1. Variable
stripper
Illinois1
Texas'
1. Variable
stripper
None
Vintage
2. New oil
& g=
None
None
None
2. New oil
Other
3. Tertiary oil
Incremental
50% for
None
3. Workovers
prod.
EOR
&
Deep &
tight gas
Secondary
tertiary
only
Ad Valorum Tax
Yes8
None
None
Yes*
Yes
Yes
Yes
None
yes9
Yes8
Yes
yes9
Yes
None
Yes
Yes
Yes11
Yes12
None
Yes''
Yes
Yes12
None
Rat
015Olyr
9.7
7.4
Variable
8.9
1.3
8.4
6.4
10.2
99
83
99
79
98
62
76
92
64
20
89
18
24
'
99
32
Kansas Inc., Strategic Analysis of the Oil and Gas Industry in Kansas. Arthur D. Little. Inc., Cambridge, Mass.,April 1990.
Personal commuaication with K.Deason. Missouri Dept. of Natural Resources, and S. Even, Missouri Dept. of Revenue, July 1990.
Gross Lease Revenue (N.B.- Does nc4 discount transportation and marketing costs).
Gross Lease Revenue less Marketing and Transport Costs.
Less than $25,000 at 2%. $25,000-$100.000 at 346, $100.000-$300.000 at 4% $3Oa,oOO and over at 5% on corporate/individualoiVgas revenues.
An extraction tax is assessed at the rate of 6.5% for old wells and 4% for new wells.
Each state is attempting to mitigate declining oil production and declining revenues to the state and have or are considering economic
incentives for enhanced oil recovery.
Ad valorum tax levied on the economic value of each producing unit. Appraisal value calculated by applying present worth factor to
future revenue to derive a net worth for each lease.
Tax h s i s derived from apportioned revenue derived within state as determined by three factor formula that is equally weighted. A twofactor formula is available for qualifying companies. Rates are $0 $25,000 at 4.5%. > $25,000 at 6.75%.
lo Separate accounting for oil and gas income on all taxable income.
Of shareholder equity 0.1%. minimum of $20 and maximum of $2,500.
l2 Of business and investment capital 0.125%. minimum of $10 and maximum of $20,000.
'
53
Project life,
rate
0.334
0.304
0.2m
0.243
0.230
0.218
0.208
0.199
0.173
0.163
4.0
4.5
5.5
6.0
6.5
7.0
7.5
8.0
10.0
11.0
TABLE 4.10. - Suggested Average Life Tables for Oilfield Related Equipment1
Life
XaKS
computers
Injeaion System
Collection System
Scrubber
Water Softener
Filters
Preheaters
Steam Generators
5
10
15
15
15
20
Compressas
Automatic Well Testing (AWT) Unit
Oil Analyzer
Vapor Recovery
Gauging Equipment
Flow splitter
B.S. & W. Monitor
Air Exchanger
Electronic Panels
Waste Water Plant
Boiler Plant
Giw Plant
Cogeneration Plant
20
20
20
40
10
10
10
10
10
10
15
15
15
15
15
15
20
L.A.C.T.Units
20
20
20
20
20
20
20
The economic feasibility of thermal operations has been proved over and over again in
California and other parts of the world for many years. In spite of higher capital and operating
costs, thermal operations have proved profitable in these areas and have recovered large volumes
of heavy oil which otherwise might have been abandoned. Successful steamflooding under a
54
favorable geological setting is entirely dependent upon the three basic factors: engineering,
economics, and experience.
REFERENCES
1. Maples, J. W. Kern County Assessor-Oil and Gas Properties Appraisal Parameters,
1991-92-Assessor, County of Kern Oil and Gas Division, Bakersfield, CA, March 1991.
2. Lewin and Associates-Economics
Report No. DOEJETI12072-2, May 1981.
8 . Bartlesville Project Office, U.S. Department of Energy, P.O. Box 1398, Bartlesville, OK
74005 (Phone: 9 18-337-4293).
55
APPENDIX 4-A
REVIEW OF DEPARTMENT OF ENERGY
STEAM DRIVE PREDICTIVE MODELS
BACKGROUND
Estimating the amounts of economically recoverable heavy oil requires the use of a sound,
yet simple, thermal EOR model. For this, the Department of Energy (DOE) Bartlesville Project
Office (BPO) currently maintains four steamdrive enhanced oil recovery models and one in situ
combustion oil recovery model
This appendix reviews the publicly available thermal recovery DOE models and recommends
the model most suitable for estimating oil production by steam injection.
Review of Available Steamflood Models from BPO
The four steamdrive thermal recovery models available through the BPO contain several
common features. These features include algorithms for the determination of surface and wellbore
heat losses, engineering costing algorithms, and a basic economic analysis package. The common
technical features include a correlation for estimating residual oil saturation to steam, an optimum
steam-injection rate calculation, and a method for calculating oil and water production rates, which
includes fluid shrinkage and the presence of free gas. The models also use common routines for
handling data input, data verification, and data output.
Each of the BPO steam drive models is discussed in more detail as follows:
1. sUPRI Model. The SUPRI model, also called the Williams et al. model,l-* utilizes the
Marx and Langenheim3 method to predict the growth and volume of a steam zone created by
continuous steam injection. These calculations assume that the growth of a steam zone and the
volume of oil displaced by steam are controlled by heat losses to the adjacent overburden and
underburden.
2. Jones Model. The Jones model4 is a steamdrive performance predictive model based on
work by van Lookerens and by Myhill and Stegemeier.6 This model uses a steam-injection-rate
optimization function based on a vertical conformance factor and a steamdrive algorithm based on
the Mam and Langenheim3 equations as modified by Mandl and Volek.7 The model delays the
arrival of the oil bank.at the producing well and maintains ultimate oil recovery to less than the
mobile oil in-place.
3. Gomaa Model. The Gomaa model8 predicts steamflood performance by utilizing a set of
empirical correlations. These correlations are based on observed field performance data from the
California Steamflood (notably Kern River) projects. The key correlation variables include vertical
heat losses, vertical sweep, and fractional oil recovery as functions of reservoir thickness, heat
injection rate pzr acre-foot of reservoir, bottomhole steam quality, and mobile oil saturation. The
57
correlation curves are developed by regressing the results of a finite-difference steam drive
reservoir simulator. A simple method of estimating water production rates has been added to the
original Gomaa model by BPO.
4. Intercomp Model. The Intercomp model provides a more complex and thus more realistic
description of steamdrive performance. The model utilizes the reservoir and injected steam
properties to predict steamflood performance. The model calculations allow for the development
and eventual production of three fluid banks created during steamflooding: ( i ) the cold liquid
bank, (2) the hot oil, condensate, and water banks, and (3) the steam zone.
.
, - The cold liquid bank, created by the displacement of unheated oil and water by the steam and
hot liquid banks, represents the initial response of a reservoir to steamdrive. The hot liquid bank
contains heated oil and water and exists just ahead of the advancing steam zone front. The steam
bank follows and generally overrides the, hot liquid fluid bank. The model varies the densities,
viscosities, residual saturations, and relative permeabilities of these banks with temperature. Heat
losses from the reservoir to the overburden and underburden are calculated from simple heat
balances, and heat is produced when the heated zone reaches the producing well. Initially the
reservoir rock" is assumed to be homogeneous, isotropic, and incompressible, in thermal
equilibrium with the adjacent strata, and with uniform oil, water and gas saturations. A detailed
description of this model is presented in references 9 and 10.
Model Recommendation
A comparative assessment of these models by Dowd et al. indicated that all of the models
display reasonable and expected sensitivity to the steam injection rate, steam quality, and grosslnet
pay ratio. However, the simple Gomaa model exhibits no sensitivity to reservoir permeability or
crude oil viscosity. Based on the model's ability to match simulation results and history match of
field data, Dowd et al. recommend that the Intercomp model be used to predict steamflood
recovery. However, it must be pointed out that the algorithm used in the Intercomp model to
" after steamflood often leads to an overprediction of oil recovery.
calculate residual oil saturation
Since this significantly affects the project economics, model results should be used with caution.
58
5. Van Lookeren, J. Calculation Methods for Linear and Radial Steam Drive in Reservoirs.
Pres. at the 52nd Annual Fall Meeting of the SOC.of Pet. Eng., Denver, Colorado, 1977. SPE
paper 6788.
6 . Myhill, N. A. and G. L. Stegemeier. Steam Drive Correlation and Prediction. J. Pet.
Tech., v. 30, No. 2, February 1978, pp. 173-182.
7 . Mandl, G. and C. W. Volek. Heat and Mass Transport in Steam Drive Processes. SOC.
Pet. Eng. J., March 1969, pp. 59-79.
59
APPENDIX 4-B
CALCULATION OF CAPITAL CHARGE RATE4
The methodology used to calculate the capital charge rate is as follows:
CRFr,B =
r)book life
(1 r) book
+
life
-1
(1 + r >
x CRFr,B
(1 - tax rate)
1 - tax rate
fb
rs
fs
61
( 1 - tax rate)
CHAPTER 5
WATER TREATMENT FOR STEAM GENERATION
INTRODUCTION
Successful operation of steam generation equipment depends primarily upon a good source
of feedwater combined with an effective water treating system. Quality of steam generator
feedwater is of critical importance in steam injection projects. Field experience indicates that most
steamer downtime is caused by water treating problems. Although a once-through steam generator
can tolerate relatively high amounts of total dissolved solids (TDS), strict adherence to feedwater
quality requirements is essential. Operating outside a design quality range can result in scale
deposition in the water sides of steam generator tubes. Deposition of scale reduces heat transfer,
reduces flow though the tubes, and results in tube failure through formation of localized hot spots
on the tube walls. Corrosive components of the water also may cause tube failure. Therefore, it is
imperative that the feedwater be properly conditioned to minimize downtime and costly
maintenance.
Those embarking on steam injection projects should have an understanding of water quality
requirements and the reason these requirements must be met for optimum operation of the steam
equipment. Furthermore, since water provides a medium for trouble causing chemical reactions
such as scaling and corrosion, it is essential that the operator at least have a rudimentary knowledge
of water chemistry. In Appendix 5-A, oilfield water chemistry is briefly discussed.
In .this chapter, the steam injection process water quality requirements, problems, and
remedies are presented. Since the availability of fresh water for steam generation is becoming
increasingly scarce in many locations, treatment and use of produced water as steam generator
feedwater should be considered. A discussion on the treatment of produced water is also
presented.
A list of feedwater treatment equipment and chemical vendors is included in Appendix 5-C.
This list is included to serve as a reference and is not intended to be the recommended list of
suppliers.
The main point to consider before undertaking a steam injection pilot is that the available
water supply is adequate to sustain a full-scale steamflood operation for a considerable length of
time. It would be expensive to undertake a pilot program based on a limited supply of fresh water
because if the pilot proved successful (and it should be favorable on paper before it starts) the
ultimate need may be 20 to 30 times the rate required by the pilot for several years to come. The
pilot design should be based on the use of inferior quality water to assure the availability of an
adequate supply of water for future expansion.
Naturally occurring water supplies contain considerable quantities of impurities. Water is
such a superb solvent, that it has the power to dissolve virtually all inorganic substances to some
extent. The impurities found in the water result from dissolving the various gases and mineral with
which it comes in contact. Some of the typical impurities found in water sources are related to the
origin of these sources. Moving supplies such as rivers and creeks generally contain mud, silt and
other suspended matter, dissolved gases and minerals as well as bacteria, algae, and other organic
matter. Because of natural settling, static sources are usually low in suspended matter. However,
shallow lakes are easily disturbed by storms and are likely to contain higher amounts of suspended
solids than deeper lakes.
Well waters, in general, are richer in dissolved mineral content, but poorer in suspended and
organic matter and algae due to the filtration effect of the earth. As a rule, the mineral content of
well water increases with depth. Typical analyses of subsurface waters used in California
steamflood operations are shown in Table 5.1. Finally, produced water is usually contaminated
with oil and contains extremely high amounts of dissolved solids and organic matter. The
dissolved mineral content of these waters often exceeds 20,000 ppm (see Table 5.2).
Since the type and amount of impurities found in ground and surface water supplies are
considerably different, methods to produce the same end purity will vary with the source of water.
Further, it is highly unlikely that water from any two surface sources in the same area would
contain the same kind or amount of impurities. Therefore, it is essential that a specific water
analysis be conducted before developing any water treatment plans. Additionally, source water
should be monitored periodically to assure that changes which occur in a water supply due to usage
are compensated by a water treating plant.
63
(9) silica, (10) pH, (1 1) carbon dioxide, and ( 1 2) biological growth. The following will discuss
briefly how these impurities can affect steamflood equipment performance. For a more complete
discussion, please refer to Appendix 5-B.
TABLE 5.1. - Typical Analysis of Raw Waters Used in the San Jaoquin Valley, California, Steam
Injection projects4
Water analysis
-
50
45
460
60
40
IO
15
<1.0
<I
240
20
< I .o
.o
Coalinga
480
320
35
< I .o
40
200
20
350
45
I IO
50
360
120
2.000
80
400
200
C1.0
<I
to
None
3.0
200
600
.o
None
7.8
<o.I
None
None
7.9
2.0
MclGttrick
Kern
River
Midway
Sunset
Impurities
in water
<I
.o
None
7.7
3.0
to
None
5.0
3,000
<I .o
None
7.5
<O.l
None
Note: All values are in parts per million as CaC03 except those indicated with an asterisk.
TABLE 5.2. - Approximate Total Dissolved Solid (TDS) Content and Hardness of California
Oilfield Produced Water10
Hardness
rZPm
(as CaCo3)
Area-Field
2,000
3.000
1.600
800
I00
TDS
PPm
25.000
30,000
28.000
10,oOo
1.oOo
6.000
1,500
25,000
300
500
64
5.000
8 ,ooO
Hardness
Total dissolved solids (TDS)
AIblinity
Silica
Dissolved oxygen
Chlorine
PH
Iron
Hydrogen sulfide
Turbidity
Oil
e1 ppm
<4.000 ppm
~ 1 . 5 0 0ppm
4 0 ppm
~0.01
ppm
e1 ppm
9-1 1
<0.05ppm
0
0
0
Total Hardness: The hardness is the measure of the amount of calcium and magnesium ions
contained in water. These ions are responsible for the steam generator scaling and their
concentration in the boiler feedwater should be held below 1 part per million (ppm) to prevent
scaling. This is normally accomplished with the aid of an ion exchange resin bed.
Alkalinitv: Natural water contains carbonate and bicarbonate ions which, under the influence
of heat, breakdown into hydroxide (OH-) and carbon dioxide causing the water to become alkaline.
Although excess hydroxide alkalinity can result i n caustic embrittlement, a moderate alkaline
environment helps to reduce corrosion and keeps silica in solution. Therefore, alkalinity levels of
less than 1,500 ppm need not be treated.
Qxvren: Dissolved oxygen is the primary contributor to corrosion and should be excluded
from feedwater. API recommends that the maximum amount of dissolved oxygen in the feedwater
be kept below 0.01 ppm and preferably at 0.0 ppm.
S m :
Q11: The presence of oil in the feedwater contributes to film formation and coking in the
generator tube and results in their eventual failure. Oil also fouls the water softener resins.
Therefore, the feedwater should be free of oil.
Silica: Silica is troublesome in oilfield steam generators because of its scaling tendency.
However, the presence of hardness ions such as calcium and magnesium is necessary for the silica
to form scale. Since these ions are usually removed by ion exchange process, silica removal is not
necessary. Satisfactory operations with silica contents of 150 ppm are possible in the absence of
hardness ions.
The pH of the feedwater should be maintained between 9 and 11 to keep silica in
solution and to avoid corrosion of the steel parts of the system.
BioloPical Growths: If the total dissolved solids content of the raw water is high, bacteria
may grow in the feedwater. Since bacteria and algae will foul the ion exchange system, their
growth must be controlled. The most common method of controlling bacterial growth in oilfield
water is by means of chlorine or other biocide agents. Chlorine will attack only the exposed
organisms and will not penetrate slime. Biocides are effective in eliminating slime forming
bacteria. Biocides should be chosen with the help of a vendor representative, sincz the type needed
will depend on the organism, contact time, and temperature of the water to be treated, etc.
Table 5.4 summarizes the various methods of treating common feedwater impurities.
m:
66
H2S Gas
co2 Gas
0 2 Gas
Sediment, Turbidity
Bacteria
Oil
Hardness
Chlorine
Iron
ions without reducing the total solids content. Inexpensive to purchase, operate, and maintain, the
system consists of either a cation exchange resin which removes cations (such as calcium,
magnesium, iron, and manganese) or an anion exchange resin which removes anions (such as
carbonate, bicarbonate, chloride, and silica). For steam operations, most commonly used ion
exchange system for softening water is the synthetic sodium zeolite cation exchange unit. The
synthetic sodium zeolite resins are derivatives of sulfonic acids and are commonly referred to as
strong acid resins. Strong acid resin softeners efficiently reduce the calcium and magnesium
content of a raw water, but they do not affect the alkalinity, silica, or the total dissolved solids
content of the water.
Many different types of ion exchange materials are used in water softening operations. The
ion exchange material and method used to soften water will depend largely on the quality of raw
water and the quality of water required for a particular use. For example, the water quality
required for steam generation calls for almost zero hardness, and a sodium zeolite softener alone is
not satisfactory if the TDS content of raw water is greater than 2,000 ppm.10 Depending on the
TDS content of raw water, other types of resins are used alone or in conjunction with sodium
zeolite resins to obtain water of the required quality. The raw water quality and economics will
dictate the resin choice. Weak acid resins are also used in steam injection operations. These resins
exhibit a much greater capacity than conventional zeolite resins, are highly selective, and offer
more efficient regeneration. Weak acid resins are discussed more fully in a later section.
Before discussing the details of ion exchange equipment, four terms describing resin
efficiency are very important and should be remembered.
1.
of an exchanger is defined as the amount of a specified ion that a given volume
of resin can remove from a raw water at stated conditions. Exchanger capacities are used
in characterizing ion exchange materials and in numerical calculations pertaining to ion
exchange operations. The capacity of an ion exchange resin is expressed as grains per
cubic foot (1 grain CaC03ku ft = 2.29 ppm; see Appendix 5-A for definitions). For
67
example, a resin is said to have a capacity of 2,500 grains per cubic foot, if 1 cubic foot
of this resin upon saturation contains 2,500 grains (5,725 ppm) of cations expressed as
calcium carbonate. The capacity of ion exchange resins varies greatly. The capacities of
naturally occurring zeolites vary between 2,500 and 5,000 grains per cubic foot. The
capacities for synthetic zeolites range between 15,000 and 35,000 grains per cubic foot.
In comparison, weak acid resins typically have a capacity as high as 70,000 grains per
cubic foot, but the operating capacity is usually in the range of 20,000 to 40,000 grains
per cubic foot.10
2. Leak= is the quantity of unwanted ions present in the treated water, after its passage
through the resin bed. Leakage is a function of raw water composition, resin capacity,
regenerant concentration, and the condition of the lower levels of the exchanger bed.
Leakage, as applied to an oilfield water, is discussed in a later section.
3. Flow Rate is expressed as the volume of water passing per square foot of cross-sectional
area of resin bed per minute. Flow rate must be regulated so as to allow proper contact
time between the water and resin. Since the flow rate can significantly effect the
operating capacity of the resin beds, it should be limited to about six to eight gallons per
square foot of resin beds.
4. Rate of ExchanE is affected by resin type, flow rate, temperature, and resin particle size.
For a given resin, a decrease in particle size results in increased rate of exchange and
increased leakage. Higher temperatures increase the rate of exchange. For a given
particle size, different resins exhibit different rates of exchange.
Principles of operation4
In the sodium zeolite water softening process, the hard water is passed through a bed of
active sodium cation exchange resins. As the water flows through the resin bed, the resins replace
the objectionable calcium and magnesium ions in the water with the nonobjectionable sodium ions.
Using the symbol 2 for the zeolite radical, the softening process can be summarized as follows:
68
CI
The resin now contains calcium and magnesium ions and a few residual sodium ions.
When the ability of the zeolite bed to remove hardness has been exhausted, the softener is
temporarily taken out of service and backwashed for regeneration. The regeneration consists of
passing sodium chloride brine through the units, replacing the calcium and magnesium ions with
sodium ions. The regeneration step may be represented as follows:
Ca (or Mg)Z + 2Na C1
(2)
The regenerating solution, along with water used to rinse the softener after regeneration, is
discharged to waste. The process of regeneration of spent resins with sodium chloride is called
brining.
It should be mentioned at this point, that no chemical reaction is involved in the ion exchange
process. The softening or regeneration proceeds only because there is great excess of one ion in
the water compared to the resin. Although sodium zeolite treated water is nearly free of calcium
and magnesium ions, some hardness leakage is inevitable. The term leakage implies a slipping
of some of the unwanted ions into the effluent. Leakage occurs because the ion exchange process
is in dynamic equilibrium, i.e., reactions 1 or 2 are reversible. The sodium ion in the raw water,
plus the sodium ion released from the ion exchange bed by exchange of hardness ion, has a
regenerating effect on the resin as it is removing hardness. This results i n hardness leakage
through the bed and into softened water. Hardness leakage increases with increasing TDS in the
raw water.
The frequency of regeneration needed depends on the flow rate, the hardness content of the
raw water, the exchange capacity of the resin, volume of resin in the bed, and amount of salt used
per regeneration. Of these, the operator can exercise control only on the quantity of salt used per
regeneration. The other parameters are fixed by the system design and raw water hardness.
Salt Requirements1 1
Zeolites have a greater affinity for divalent cations such as Ca++ and Mg* than monovalent
cations such as Na. The divalent cation affinity increases with. an increase in atomic weight.
Thus, during the softening cycle, Ca++ ions are-more readily removed from raw water than
Mg.
Therefore, magnesium hardness predominates in the softened water than calcium
hard ness.
69
Since the zeolite more readily releases sodium in exchange for calcium and magnesium, an
excess sodium chloride must be used for regeneration. This means the amount of sodium in the
brine must exceed the total amount of equivalent calcium and magnesium in the exhausted resins.
The amount of salt needed for regeneration is determined by the acceptable effluent quality
limit and plant capacity. By using greater quantities of salt for regeneration, the capacity of the
softener can be increased. In figure 5.1, the effect of raw water hardness and salt dosage on resin
capacity is depicted. However, capacity increase is not proportional to regenerant increase. The
effect of salt level on the performance of a typical zeolite softener is shown in Table 5.5.
Capacities in this table are stated in grains of CaC03 per cubic foot of zeolite and salt consumptions
in pounds per cubic foot of resin and also in pounds per kilograin of hardness. From this table, it
is apparent that an increase in salt usage is out of proportion to the increase in softener yield. For
example, increasing the salt dosage from six pounds per cubic foot of resin to 15 I b k u ft of resin (a
160%increase in salt usage) will result in about only a 50% increase in softener capacity. Thus,
the higher the salt level, the less efficient is the exchange process and greater is the operating and
waste handling costs. The effect of feedwater hardness and salt dosage on water softening costs is
shown in figure 5.2.
30
28
Y
24
16
0
1000
2000
3000
4000
5000
FIGURE 5.1. - Effect of feedwater hardness and salt dosage on resin capacity."
70
TABLE 5.5. - Effects of salt level on the peifoimance of a typical synthetic sodium
zeolite exchanger12
Pounds per cubic foot
of resin
Ut Reauiremeni
Pounds
per kilograin
Resin caoacity
(Grains of CaC03 per
cubic foot of resin)
0.33
18.000
0.4
20,000
10
0.42
24,000
15
0.5
30.000
Table 5.6 is a zeolite water softener sizing chart provided by one manufacturer. This chart
provides a perspective on the relative size of the equipment and choice between salt dosage and
capacity.
71
0.4
15
10
0.3
0.2
0.1
0
0
1000
3000
2000
5000
4000
FIGURE 5.2. - Effect of brine hardness and salt dosage on water softening costs.
Polisher
tank
dimension
Dia x side,
inches
dimension
Dia x side.
inches
PlillWy
30x72
48 x 12
60x72
72 x 72
90x96
120 x 9 6
24x60
42 x 60
48x60
60x60
78x72
96 x 72
Total
resin
capacity.'
kilograins
Service
rate,
GPM
flow
CaC03
Resin
volume,
cubic feet
Primary
Polisher
@ 30. bed,
Ib
depth
55
120
480
1.200
1,800
2.550
4,800
7,200
190
280
440
7 80
72
Salt
dosage.*
16
40
60
85
8
24
32
50
240
600
900
90
1.200
2,400
140
3,600
H W RATE
c- RAW WATER INLET
P R W Y BAFFLE
RlEEBoARD
EGEtmwl
INLET
STEaSHELL
I \
**-
ON EXCHANSE RESIN
TREATED
WATER
UNDER DRAN
SYSTBV!
Cull5
The resin bed contains a strong acid cation resin. The bed is leveled and hydraulically graded
so that the coarse particles are in the bottom of the bed and the finest at the top. This arrangement
permits even flow of water through the bed. The quantity of resin used will depend upon the
exchange capacity of the resin, the hardness of the water being treated, and the amount of water to
be softened between regenerations. The bed must also be of sufficient size to allow proper contact
time between the water and the resin. A minimum depth of 30 in. is recommended for all systems.
The softened water is collected by the underdrain system, located at the bottom of the
softener. This system also collects the waste brine and rinse water and distributes the backwash
water during the backwashing operation. A well-designed underdrain system permits the even
collection of softened, rinse, and salt waters from all portions of the bed and distributes the
backwash water so that it flows evenly upward through all portions of the bed. An uneven
distribution will lead to channeling and hardness leakage and cause loss of capacity. Uneven
backwash can also result in loss of resin through carryover.
Although the underdrain system design varies with the vendor, most vendors use either a
deflector plate type or a header-lateral type design. In the deflector plate design, a specially
designed deflector plate housed inside a false bottom permits the even collection or distribution of
water.
73
Either a rnultiport valve or a valve nest is used to direct the flow of water and brine in and out
of the softener. The valve nest is comprised of six valves: raw water inlet and outlet valves, brine
inlet valve, rinse water outlet valve, and the backwash water inlet and outlet valves. In most
installations, air, water, or motor-operated automatic valves are used to control the flow. Very few
installations utilize manually operated gate valves. In large installations, a single multiport valve is
used instead of a valve nest to control the flow of water through the softeners during the various
cycles of operation. As the name implies, the multiport valve is comprised of several ports, and
they are positioned precisely at the desired location by manipulating the valve motion. The
positioned port then directs the flow of water in the same manner as the opening and closing of six
separate valves. Automation is usually used in the operation of a multiport valve. Although more
expensive, multiport valves are preferred over valve nests due to the elimination of operational
errors caused by opening or closing the wrong valves. The internal details of a typical ion
exchange water softener are shown in figure 5.4.
Brining System
The brining system usually is comprised of an epoxy-lined salt storage tank in which
saturated brine is formed. The saturated brine is then transferred to a small-diameter brine
LOCATION OF
DISTRIBUTOR
l.
FERRULED
OPENING
74
measuring tank where the brine is diluted to the desired concentration. The small-diameter
measuring tank allows a deeper drawdown and gives a more accurate measurement.
In large steamflood operations, where considerable amounts of salt are being used to
regenerate the spent resin, it is more cost-effective to purchase salt in truck loads than purchasing it
bagged. A reduction in handling costs is another economic advantage of bulk salt usage. A typical
design of a bulk salt storage facility for steamflood operation is shown in figure 5.5. It consists of
a large subsurface concrete brine pit, usually located adjacent to a railroad track or roadway, where
salt can be brought to the plant in bulk and simply dumped into a salt storage and brine preparation
tank. An auger is usually available to unload the shipment of salt from the truck into a pit. This
tank usually has a float-operated valve to control the addition of water into the tank. The saturated
brine is pumped at intervals into a brine measuring tank from which it is pumped to the softener
after dilution. Plastic pipes are usually used to transfer brine from the tank to softener.
Softener Operation
A sodium zeolite softener operates through two cycles: softening and regeneration. During
the softening cycle, the raw water enters the softener through the inlet distributor and percolates
FLOAT
VALVE
BEAM
GRAVEL
75
through the zeolite bed. As it flows through the resin bed, the hardness-causing ions-calcium
and magnesium-are taken up by the resin, and an equivalent amount of sodium is given up by the
resin to the water. The softened water is collected by the underdrain system and transferred to the
point of use.
When a softener is exhausted, it is regenerated. Several different approaches are used to
signal the end of a softening cycle. Some plants regenerate after a fixed number of hours in
operation. Others monitor the hardness of the sample-softened water and initiate regeneration
when the hardness exceeds the desired level. Another common technique used to establish the
need of regeneration is to monitor the quantity of water softened between regeneration. A metering
device is used to initiate regeneration automatically when the preset numbers of gallons have been
softened.
The regeneration cycle involves the following four steps: (1) backwash, (2) brining,
(3) brine displacement and slow rinse, and (4) fast or final rinse.
Backwash-The purposepf backwash is to cleanse the resin bed of filtered particulates and
resin fines and regrade the bed. Backwashing is accomplished by sending a strong flow of water
through the underdrain system into the resin bed. As the water flows upward through the bed, it
expands, cleanses, and hydraulically regrades the bed. The waste water is discharged to the waste
through a raw water distributor.
The backwash regrades the bed by bringing the coarsest particles to the bottom of the bed and
the finest to the top. This assures proper water distribution through the bed. Expansion fluffs the
resin bed and improves the brine resin contact efficiency. Channeling and high pressure drop
through the resin bed are eliminated through the removal of fines. Backwash must be carried out
for a minimum of 15 minutes or until the backwash water effluent is clear.
The backwash flow rates depend on the water temperature, resin type, and height of the free
board space. The rate should not be excessive to cause resin loss, yet should be sufficient to
produce a minimum of 50% bed expansion. Backwash rates usually vary from 6 to 8 gpm at
ambient temperature; however, manufacturers recommendations should be carefuIly followed. A
clear, suspended, matter-free water should be used for backwashing.
Brining-The purpose of the brining operation is to regenerate the spent resin. A strong
salt solution enters the softener through the resin bed. As the brine trickles down through the bed,
it removes the calcium and magnesium ions from the resin and simultaneously replaces them with
an equivalent amount of sodium ions.
The brine flow rate must be slow enough to provide sufficient brine-resin contact time, yet
fast enough to prevent channeling of the brine. To ensure optimum contact time and regeneration
efficiency, one vendor recommended that a 10% brine be added at a rate of 1 gpm per cubic foot of
resin in the softener.
76
Slow Rinse-The
77
1000
2000
4000
3000
5000
FIGURE 5.6. - Effect of water hardness and salt dosage on hardness leakage from primary
beds. 1
30
25
0
5z
20
c5
15
10
2Q:
zcn
w
0.1
0.2
0.3
0.4
0.5
0.6
FIGURE 5.7. - Resin exchange capacity as a function of water hardness and salt d0sage.I'
78
expensive than the conventional sti-ong acid resins and can be regenerated only with a strong acid
such as hydrochloric or sulfuric acids. However, weak acid resins can be operated near 100%
efficiency, and this will lower the regenerant requirements which generally results in a savings of
regenerant and waste-handling costs. Since the goal is to obtain the most complete softening at
minimum cost, raw water and softened water quality requirements usually dictate the resin choice.
Regenera tion
(RC00)2Ca + 2HC1 e 2RCOOH + CaC12
Neutralization
RCOOH + NaOH
-+ RCOONa + H 2 0
Because of its strong preference for hydrogen ions, weak acid resins can be regenerated even with
a very low acid concentration.
Hydrochloric acid is preferable to sulfuric acid as the regenerant. When sulfuric acid is used
to regenerate the resin bed, calcium sulfate will precipitate. Because of the extreme insolubility of
calcium sulfate, the cleaning will take considerable time. Calcium sulfate precipitation can be
avoided by using a very dilute acid (below 1%) and maintaining a veiy high flow rate. Regenerant
efficiency, however, is rapidly lost at low concentration. Further, a high flow rate results in higher
pressure drop and considerable force is exerted on the resin. The resin must have physical strength
to withstand this force. Since most metal chlorides are water soluble, a high concentration of
hydrochloric acid can be used in regenerating the bed. This not only eliminates the problems
associated with sulfuric acid, but also lowers the regenerant storage and handling costs.
80
all indicative of ion exchange system operations problems. There are many possible causes for
these problems, and corrective actions must be taken to alleviate them. In this section, ion
exchange system-related operational problems and soh tions are briefly reviewed.
The ion exchange system-related problems can be broadly grouped into the following classes:
(a) ion exchange resin stability; (b) problems arising from changes in raw water quality;
(c) problems associated with poor operational practices such as those resulting from the use of
offspec regenerant and improper regeneration schedule; (d) mechanical problems; and (e) resin
capacity failure due to fouling. Although hardness leakage is not an operational problem, it is
discussed in this section because many of the operational problems result in hardness leakage.
Resin Stability
Cation exchange resins are highly stable and experience very little capacity loss during the life
of an exchanger. Their useful life varies from 5 to 15 million gallons treated per cubic foot of
resin. The resin degrades either due to physical breakdown by attrition or oxidation. Resin
degradation will result in poor quality, treated water. In such cases, the resins must be replaced.
The life of the resins can be prolonged by minimizing or eliminating the oxidative conditions.
The most frequently encountered oxidant in raw water is chlorine. When lake or river water is
used as the feed, chlorine is usually added to the water to control slime and bacterial growth.
Excess chlorine in raw water will oxidize the cation resin, causing the resin to become jelly-like.
While oxidation does not affect the exchange sites, the mushy resins result in severe channeling
and increased pressure drop. Excessive pressure drop through the bed or flow channeling results
in significant reduction in ion exchange capacity and increased operational costs. Free chlorine in
supply water should be measured and controlled to prolong resin life. Free chlorine concentration
must be reduced to less than 1 ppm by deaeration or chemical means. Resin breakdown can be
minimized by avoiding excessive backwash. It is a good practice to replace annually about 10% of
increased to 5,000 ppm, this would result in 10% loss in resin capacity. Otherwise, the hardness
leakage through the bed would be increased by IO%, and an expensive aftertreatment would be
required to avoid generator tube scaling. Alteimtively, by reducing the throughput in the bed or by
increasing the salt dosage, the lost capacity can be restored. Hence, a periodic chemical analysis of
the incoming water to the resin bed should be made for these difficulties to be anticipated.
82
compaction and more resin breakage and loss of capacity. A broken underdrain nozzle will result
in excessive backwash flow and loss of resin.
The exchanger unit must be periodically inspected for mechanical deficiencies and corrective
measures taken if needed. Since steamflood operation water treatment plant shutdown can be very
costly, periodic checkouts are an insurance against trouble.
Fouled Resins
Ion exchange materials, at times, lose their effectiveness because of the accumulation of
materials on their surfaces and in their pores. This accumulation of material, or fouling, is a result
of an adsorption of contaminants present in the raw water, which is not completely removed during
the normal backwash and regeneration steps. The substances that frequently foul ion exchange
resins in oilfield operations include iron, oil, bacterial growth, and oxidants. The effects of these
fouling agents on exchanger performance have been discussed in detail in a previous section.
Fouling problems are best handled by good preventive measures.
Oxidant fouling agents such as free oxygen and chlorine can be eliminated at the source by
adding a chemical scavenger such as sodium sulfite as far upstream as possible. Bacterial slime
and algae can be controlled by closing the system to air and sunlight and adding a bactericide to the
raw water. By eliminating oxygen pickup at the source, the iron pickup can be minimized.
Injection of sodium sulfite will keep the iron in reduced form and eliminate iron fouling. Water
containing excessive amounts of suspended matters must be filtered before softening. This is
because the softening system has a limited ability to filter suspended matters and may be difficult to
backwash. Fouled softener must be taken out of service periodically and backwashed several
hours at the maximum permissible rate to dislodge and remove the adherent fouling agents. Iron
fouled resins can be cleaned by washing the fouled resins with dilute hydrochloric acid or sodium
hydrosulfite. Proper prewater treatment is the best insurance against fouling agent troubles.
Hardness Leakage517
In the operation of a water softener, one of the first things that must be dealt with is the
leakage of hardness ions through the ion exchange bed. As previously defined, the term hardness
leakage implies a slipping of some of the hardness ions present in the feedwater into the softened
water. Because of the nature of the sodium zeolite process, a certain amount of hardness leakage is
inevitable; however, the leakage must be kept to a minimum. A small amount of hardness ion in
boiler feedwater is acceptable because in the temperature range at which a steam generator operates,
these ions remain in solution in the liquid phase of the wet steam. However, the degree of
solubility of hardness ion varies according to the specific feedwater, and no generalization can be
made. In some water, only 0.2 ppm hardness leakage can be tolerated, while in others a 3 ppm
hardness can be tolerated. Reasons for these variations are unclear. Each feedwater must be
83
checked to determine what hardness is tolerable at a given steam quality. A safe bet is to use zero
hardness water, but complete elimination of hardness may not be economical in some installations.
Several factors are responsible for hardness leakage through a resin bed. Because of the
need for the use of uneconomical excess regenerating chemical (salt) to completely regenerate the
resin bed, the bed is never completely regenerated. Therefore, there is always some hardness ions
present in the bed. These ions probably will leak during the subsequent softening cycle. Further,
if the feedwater contains considerable sodium ion, the sodium will displace some of the calcium
(hardness) ion previously removed from the feedwater. Consequently, the displaced calcium will
be picked up by the softened water and hardness leakage occurs.
Sloppy operation and equipment malfunctioning also contribute to hardness leakage. Lessthan-satisfactory rinsing after regeneration will leave substantial amounts of hardness ions in the
bed. These will show up in the effluent for a short time after regeneration. Additional rinsing is
one way to solve this problem. A malfunctioning brine pump or meter can cause insufficient brine
to flow through the bed during-the brining operation and result in incomplete regeneration and
hardness leak. A fouled resin bed will reduce the resin capacity and contribute to leakage. Poor
backwash leads to bed compaction and channeling and results in hardness leakage.
Leakage is inevitable in the ion exchange system. Leakagc, however, can be minimized by
following prudent operating practices and improved design. By automating the zeolite water
softener operation, operator errors can be eliminated, and the danger of hard water getting into the
soft water lines can be avoided. Automation also permits each cycle to perform exactly the same
way each time and eliminate insufficient backwashing or rinsing by maintaining proper flow rate
and quantity. The system can also be made to shut down in the event of an equipment
malfunction. In most large steam injection projects, the water softening plant is fully automated
and requires very little attention.
Leakage can also be minimized substantially and effluent quality improved by utilizing multistage units and by using strong acid-weak acid exchangers in tandem. Further, the resin bed
should never be exhausted completely. The bed capacity begins to drop off and hardness leakage
begins to increase as a greater fraction of the theoretical capacity of the bed is exhausted.
Consequently, insofar as efficiency of regeneration is concerned, the bed should be operated only
at a fraction of its designed capacity, While this will reduce hardness leakage, it also will increase
thc regeneration frequency and hence the operating cost. Therefore, a compromise must be
reached between operating costs and the amount of hardness leakage that can be tolerated. It is
recommended that the bed be exhausted to about 60% of its total capacity. This not only improves
the performance and lowers regeneration chemical requirement but also prolongs the resin life.
Also, by operating the exchanger below its rated capacity, the system will be able to accommodate
84
a wide variation in the raw water quality. Pretreatment of raw water to remove suspended matter
and to reduce soluble foulants will also improve resin capacity and reduce leakage.
Water Treatment f o r Steutn Itijection Projects
Source water determines the complexity of steam iiijection project water-treatment systems.
Depending on the raw water source, systems range from a simple sodium zeolite ion exchange unit
to a facility that includes clarifiers, coagulators, filters, skimmers, oil separators, coalescers,
depurators, softeners, and deaerators. As a general rule, poorer water quality requires more
elaborate treatments and higher treating costs.
The water sources and the water quality requirements for steam injection operations are
discussed in previous sections. The general principles, operational techniques, and problems of
ion exchange softeners, are outlined. The principal objectives of this section are (1) to explain the
operational differences between the conventional cation exchange system and the one used in steam
injection operations and (2) to summarize features of some other common water treating equipment
and to offer some guidelines to their operation. Other topics discussed include: description of
processes for the reclamation of produced water for steam generations, water treatment monitoring
equipment, and water treatment costs.
Steam Injection Water Softening System 1 2
The oilfield cation exchange water softening system differs from the previously described
conventional cation exchange system in two ways. Because of the stringent water quality
requirements in steam injection operations, oilfield water-softening systems utilize a primarysecondary softener setup and a special upflow-downflow countercurrent regeneration technique to
produce water with less than 0.25 ppm hardness for steam generation. The steamflood water
softening system consists of a primary exchanger which removes the bulk of the hardness from the
water, followed by a polisher (secondary) unit which scavenges the last traces of hardness.
Depending on the water.requirements, two or more such trains (primary and polisher) are used to
control water hardness. Oilfield water-softener systems are offered as fully self-contained, skidmounted units complete with two trains of softener vessels (2 primary and 2 polishers), brine
meter, regeneration meter, valves and piping, control panel, and pumps.
The two trains are installed in parallel so that a continuous supply of softened water for steam
generation is assured. Raw water passes through one of the primary exchangers and one of the
polishing units in series to some point short of exhaustion of the resin in the primary bed. Then
the used pair is regenerated.automatically as the feedwater is diverted through the other pair. This
operation is automated either by hardness-sensing devices or through an automatic time cycle. A
schematic of the softeners and piping arrangement is shown in figure 5.8. The primary softeners
85
RAW WATER
A
I
POLISHING
m
HEATERS
produce into a common manifold. The polishing softeners receive the water from this manifold.
Each primary and polisher train is regenerated independent of other trains. Flow schematics of the
softening regeneration cycle is shown in figure 5.9. Note that soft water is used for backwashing
and brining of polisher to minimize hardness leakage. Note also that countercurrent regeneration is
used in the polisher to ensure complete regeneration ut the bottom of the bed. Since portability is
an important design factor, the softener train is usually oversized to cope with unexpected raw
water quality changes. The softened water is usually stored in a galvanized or plastic-coated tank
under a blanket of nitrogen to keep oxygen out. Water is usually gravity fed from this tank to a
generator feed pump.
86
RAW WATER
RA
iER
DILUTION WATER
BRINE
PRIM
SOFIWATER
SOWATER
SOFTENING
BACKWASHING
RAW WATER
BRINE DISPLACEMENT
AND SLOW RINSE
PRIMARY RINSE
BRINING
RAW WATER
POLISHER RINSE
Experimental studies indicate that the strong acid cation exchangers are not satisfactory to soften
waters containing greater than 5,000 ppm
S. Weak acid cation exchange resins have much
higher capacity and can be used to soften water containing up to 20,000 ppm TDS. Water
containing up to 10,000 ppm TDS can be softened using a strong.acid cation resin in the primary
unit, followed by a weak acid cation resin in the polisher. Because of the effectiveness of the weak
acid polisher, a higher than normal hardness leakage from the primary unit can be tolerated without
sacrificing the quality of the final product from the.polisher. The amount of salt, hydrochloric acid
and sodium hydroxide (regenerant chemicals) required will depend on the extent of hardness
breakthrough from the primary. Table 5.7 shows a general rule-of-thumb for selecting ion
exchange resins for thermal EOR operations. This table can be used as a guideline in selecting a
system .
r-
TABLE 5.7. - Rules-of-Thumb for Selection of Ion Exchange Resins for Oilfield Steam
Generator Feed water Softener10
Hardness
Resin
<2,000
<700
700-5,000
<700
5,000-10,000
400
5,000-10,000
500-1,500
50- 15,000
1,500-3,000
88
Na/H = 2.17
Na, m q n
Hardness, mg/L as CaCO3
=
=
=
where
FWV
TH
LH
and
3. Regeneration cost (RC) Ghbl for both 100% strong acid resin systems and
100%weak acid resin systems.
RC
ww
and
RC=0.00244 CR (TH-LH)
EC
CR
= cost of regenerants, 4/113
TH
= total hardness in feedwakr, mg/L as CaC0-j
= leakage hardness in softened water, mg/L as CaCO3
LH
EC
= exchange capacity, kgr/ft3
0.00244 is a conversion factor
4. Regeneration Cost (RCC), ebb1 for combined strong acid resiidweak acid resin
systems.
RCC, ebb1
Cost for Strong Acid Resin + Cost for We'ak Acid Resin
= 0.00244 CRS (TH-LH)
ECS
where
and
ami
where
EC
TH
EC
CTH-J-H)
89
acetic acid (EDTA). EDTA does not eliminate hardness, but only prevents it from forming scales.
EDTA is preferred to other chelating agents because of its theimal stability.
The amount of EDTA necessary to complex 1 ppm of calcium ion depends on which sodium
salt of EDTA is used. The amounts of various EDTA salts necessary to complex 1 ppm of
hardness ion and the solubility of the EDTA salts in water are shown in Table 5.9. Even though
the amount of tetra sodium EDTA necessary to complex 1 ppm of the cation is higher than the other
varieties, it is used almost exclusively in the steam injection applications because of its higher
solubility.
A 38% solution of these salts is most commonly used in the steam injection application. To
ensure complete chelation of the hardness ions, an excess (10 to 15 ppm) of the chelating agents is
carried in the process water. The EDTA is diluted and stored in a 55-gallon polyethylene drum. A
reciprocating metering pump is used to inject the EDTA directly into the waterline immediately
upstream of the steam generator feedwater pump.
One important consideration in EDTA use is its effect on corrosion. EDTA, at generator pH
and reducing conditions, will chelate iron and corrode boiler tubes. Hence, addition of EDTA
should be closely controlled to prevent excess. It should be added to the feedwater line so that it is
well diluted before reaching the generator.
California steam injection field experience indicates that complete hardness removal is not
necessary, and a small amount of hardness in the feedwater is acceptable. Unfortunately, the
maximum acceptable level of hardness in the feedwater without scaling the generator varies with
the feedwater and must be established for each feedwater. Chelating agent such as the EDTA
should be used only if the acceptable level of hardness cannot be achieved economically with the
ion exchange process.
TABLE 5.9. - Concentration of EDTA and its Sodium Salts Necessary to Complex
Chelating agent
Solubility
in water
@ 80' F
gmn/loO cm3 H 2 0
0.02
Mg++
7.4
12.0
11.1
9.5
15.4
57.0
9.6
15.6
103 .O
10.4
16.9
90
Hardness Monitors 17
Monitoring hardness is essential for successful treatment. Tests for hardness should be run
at least once each shift. Ideally, hardness should be monitored continuously, in addition to
periodic manual tests.
A number of different approaches are used to monitor and control hardness. The most basic
and inexpensive is the analytical approach. The simplest hardness monitoring tool samples the
softened water at frequent intervals (10 minutes or so) and indicates whether the measured
hardness is above or below the setpoint. In most models, 1, 3, 5 and 7 ppm are used as the
setpoints.
Since this instrument is basically a titration apparatus, practically no maintenance is needed.
However, technician must periodically check the instrument to assure its proper operation. The
capacity of the reagent reservoir i s sufficient to pelform about 1,500 analyses and can be refilled in
a matter of minutes.
A more useful hardness monitoring instrument is the one that not only monitors hardness,
but also takes appropriate action to control it. A number of manufacturers market hardness
monitors that are designed to regulate hardness. These instruments are basically automatic titrators
fitted with an actuating mechanism designed to take appropriate action in the event the hardness
exceeds the preset value. Depending on the model, the actuator will sound an alarm and/or actuate
automatic regeneration equipment for zeolite softener. Some models will also shut down the steam
generator. Units are usually calibrated for setpoints from 0.25 to 3.0 ppm hardness.
A more sophisticated hardness regulator uses colorimetric principles to control hardness.
These are simple single reagent colorimetric analyzers designed to measure the intensity of the
color developed by the addition of reagent to the sample. The intensity is proportional to the
hardness and is indicated on a direct reading meter as pprn hardness. A schematic of one such
colorimetric analyzer is shown in figure 5.10. These -instruments usually analyze and record
hardness every 2 minutes. Most models are also equipped with a signal generator (pneumatic or
electrical) to start automatic zeolite regeneration or to shut down the generator.
91
u DRAIN
92
Like the conventional waterflood operation, the produced fluids from the wellhead in the
steam injection operations is sent to the oil dehydration plant for separation into oil and water. The
separated water is sent to the water treatment plant for further processing. This discussion begins
at this point, as this section is concerned with the treatment of produced water, not oil dehydration.
Oil dehydration operations are discussed elsewhere in this report.
The degree to which the oil and solids must be removed from produced water depends on the
end use water quality requirements. If the water is to be used as steam generator feed, total
removal of oil and solid is a must. If the waters are to be disposed of or used as irrigation water,
complete oil and solid removal is not necessq. Depending on the local regulations, the oil content
of e a t e d water is permitted to vay from a low of 10 ppm to as high as 40 ppm.
There are three practical methods of removing oil and solids from water. Each has its distinct
place in water treatment. The three methods are gravity separation, flotation, and filtration.
Gravity Separation
When a mixture of oil and water is allowed to stand, it will separate into two distinct phases
because of the differences in density between oil and water. This process is highly effective with
low-density (high API gravity) oils but becomes less effective with low API gravity oils. This is
because the density difference between water and oil determines the driving force causing
separation of oil and water into two phases.
Oil and water are separated in settling basins. Settling basins were used in large California
steam injection projects in the early 1970s to clarify produced water from dehydration plants. The
introduction of highly effective induced gas flotation cells (depurators) in 1970 for the removal of
suspended oil from produced water practically eliminated the need for settling basins in steamflood
operations. Unlike the settling basins, flotation cells require only a minimum of space for largecapacity throughput and can handle emulsions. Settling basins are rarely used in present day steam
injection operations. However, it is dvisable to use a small skid-mounted API type oil-water
separator or skim tanks upstream of the depurator to handle any unexpected oil load in feedwater
because of dehydration plant upsets.
The subject of gravity sGparation of oil and water has been covered in detail in an API
publication.20 This publication sets forth clearly the design and operational requirements of
settling basins and should be consulted for details..
Flotation
Flotation is a process in which a gas is used in a mechanical system- to give a lift to
suspended oil or solids. Two types of flotation machines are used in the oilfield: (a) dissolved gas
(or air) flotation (DGF) machine and (b) induced gas (or air) flotation (IGF) machine.
93
The dissolved gas tlottltion (DGF) process has been used in the
Dissolved Gas
oilfield for a number of years in the removal of oil and suspended solids from produced water.
The process involves contacting the water in a pressurized chamber with air or nitrogen. The
volume of gas dissolved in the water is directly proportional to the applied pressure. Chemicals are
usually added to the water to facilitate higher degrees of removal. Adequate retention time must be
provided in the dissolving tank for the gas to dissolve in water. The operating pressure usually
varies from 40 to 90 psi, depending on the application. Schematics of a typical DGF unit are
shown in figure 5.1 1.
After water has been saturated with gas, the gas-water mixture is transferred to the flotation
chamber through a pneumatic pressure relieving valve. When the pressure is relieved, gas comes
out of solution, forming minute bubbles. These bubbles then attach themselves to the suspended
oiI or solid particles. Adhesion of gas bubbles to the suspended particles results in the reduction of
their specific gravity. This reduction of the effective specific gravity causes the particles to rise to
the surface of the flotation cell. The accumulated surface mass is removed by rotating scraper arms
into a launder for removal. The clarified water is discharged into an effluent tank. The efficiency
of a DGF machine is of the order of 75 to 80%.
Though DGF machines are very efficient in removing suspended oil from water, their
efficiency suffers when the oil load exceeds 100 ppm and emulsion exists. As the incoming oil
load increases, increasing amounts of oil are left in the water. The disposal of sludges created by
the DGF machine is a problem in itself. Since the maximum flow capacity of these types of
machine is usually limited to about 2.0 gpm/ft2 of surface area, they require large real estate.
VARIABLE
-@bRECIRCULATION
METER
BOTTOM DRAIN
HOPPER
Once popular in the oil patch, DGF machines are increasingly being replaced by induced gas
flotation (IGF) machines, which are much more efficient than DGF machines and can handle much
higher contaminant levels. The IGF machines do not create sludge and require much smaller space
compared to DGF machines because of larger throughput.
Induced Gas F L ~ t a t i o n ~ ~ - ~ ~
The shortcomings of the dissolved gas flotation cell, necessitated the need for a process that
clarifies an emulsion-containing oilfield water efficiently and economically. This led to the
development of the induced gas flotation process. An IGF cell is basically a froth flotation cell
(used in the benefication of ore in the mining industry) modified for oilfield application. The
induced gas flotation machine is a natural aspirating shear force flotation cell designed to lower the
oil content of water efficiently. Since its introduction in late 1969 in a California steam injection
operation, the IGF machine became the primary treatment unit for the clarification of produced
water in steam injection operations. The induced gas flotation cells are commonly referred to as
depurator in the oil patch. The depurators are vely effective in the removal of emulsified oil and
suspended solids from produced water. They also lower the oxygen content of the water, if gas
other than air is used in their operation. The depurator, together with a chemical flotation reagent,
can remove oil and suspended solids from the water quickly and economically. Present day
depurators can process in excess of 170,000 barrels of water per day containing up to 1,500 ppm
oil.
Operat ion ~ r i cip~e24
n
The depurator typically consists of four flotation cells connected in series, with each cell
having 1 minute retention time; 4 minute retention time for the entire machine. A schematic of a
machine developed by WEMCO Corporation (the most popular IGF machine in the United States)
is shown in figure 5.12. Table 5.10 shows the specifications and dimensions of these machines.
These are included here to illustrate the range in capacities and overall dimensions of these units,
but their inclusion does not necessarily imply endorsement.
Each cell is equipped with a motor-driven, self-aerating rotor mechanism. As the motordriven rotor spins rapidly, it acts as a pumpand draws in gas from the gas blanket above the water,
and the water from below and forces them through a dispenser at high velocity. This creates a
shearing force and uses the gas to be broken up into a myriad of minute bubbles. As these
bubbles rise to the
ce, oil particles and solids attach to the interface of the gas bubbles. The
oil and suspended solidsgather in a dense froth at the surface and are removed from the cell by
skimmer paddles into a float flume for collection and treatment.
BEARING HOUSING
8t SHAFT ASSEMBLY
MIXTURE FLOW
LIQUID VORTEX
PATHS
96
A polymer or other flocculation enhancer is often added upstream of the depurator to improve
its efficiency. These compounds also break oil-in-water emulsions, gather suspended solids, and
stabilize the gas bubbles to promote froth flotation. In thermal operations, nitrogen or natural gas
instead of air are used to form the bubbles. A pressure of 0.5 to 1.0 ounce maintains a gas blanket
between water level and the air-tight cover. The absence of oxygen prevents the bacterial growth
and retards corrosion. The gas is recycled, requiring only about 0.1 to 0.2 cu ft/bbl makeup gas.
The depurators are highly efficient in cleaning the oily water. The efficiency of an IGF
machine is of the order of 95% and stays at that level even at high contamination levels. The oil
concentration of the inlet water is usually lowered from about 400 ppm to less than 20 ppm.
Table 5.1 1 gives the oil reduction achieved in selected California steam injection operations using
the depurator. These machines, however, do lose efficiency as the oil concentration in the coater
exceeds 1,500 ppm and may fail completely if considerable amounts of oil (greater than 5.000
ppm) come through, This is because the free oil inhibits the formation of foam which, of course,
destroys the flotation action. If the produced water contains dissolved iron, it can be removed in
the flotation process by using air in the first cell and nitrogen in subsequent cells. Methane is
usually used to purge the air and maintain a positive gas pressure in the freeboard area of
subsequent cells. The air will oxidize the dissolved iron and remove it as ferric hydroxide.
Since the IGF machines are very efficient in removing the oil from the produced water, they
should be considered by operators contemplating the use of produced water for steam generation,
or for surface discharge. Depurators offer maximum cleaning ability with minimum space
requirement. However, before a depurator can be recommended for a specific application,
bblld
Feedwater
flow rate.
bblld
Inlet water
PPrn
Exit water
PPrn
Amount
of
c hernical
added,
PPm
Ventura, CA
3 00
9.000
75-100
10-15
12
Wilmington, CA
750
25,000
81-175
6-10
Kern River, CA
450
13,000
60-350
2-15
77,175
70,000
100-400
5-10
15
Midway Sunset, CA
97
carefully monitored laboratory tests should be carried out to arrive at the proper size of machine
and preferred flotation chemical: There are laboratory-size flotation cells available for bench
testing, and often vendors of the depurators can conduct such tests and make recommendations.
Filtration
Filters are commonly employed in those steam injection operations that utilize surface or
produced water for steam generation. Surface waters generally carry considerable amounts of
suspended solids and must be removed before the water is softened. In such cases, primary
filtration is necessary. Produced waters usually contain oil and no filterable solids. Oil and solids
from such water can be removed more economically using flotation cells. In such cases, filtration
should be viewed as a cleanup or polishing operation.
Choice of Filters
The choice of a filter should be based on a number of factors, which must be evaluated in
each case. These include: (1) quality of effluent desired, (2) amount of suspended solids,
(3) nature of the suspended solids, (4)capital cost vs. operational cost, ( 5 ) space available,
(6)flexibility of operation, (7) variability in quality of incoming water, and (8) degree of reliability
of operation required.
A wide choice of filters is available for the removal of suspended solids and other impurities
from the water. These include: (1) slow sand filters. (2) gravity filters, (3) pressure filters, and
(4) diatomaceous earth (DE) filters. Each has various characteristics limiting them to specific
applications; therefore, the selection should be based on specific project requirements. In steam
injection, both conventional pressure filters and precoat (diatomaceous earth) filters are used.
Pressure filters are installed in operations handling waters containing large amounts of suspended
solids. Precoat filters are very effective in removing suspended solids and oils. They are widely
used in operations that recycle produced water for steam generation. They are usually installed
downstream of depurators to remove the last traces of oil and solids from produced water prior to
softening.
Pressure ~ i l t e r s 2 5
These iilters use sand, antharcite or garnet, or a combination of the above as the filter media.
Pressure filters are normally operated by forcing water down through a column of filter media such
as sand. As the dirt builds up in the filter sand, the flow rate through the filter will decrease (or the
pressure drop will increase). At some point, the tlow is reversed and the accumulated solids are
flushed to waste. As the filter is backwashed, the tinest sand tends to accumulate at the top of the
filter.
During filtration, only the top inch or so of the sand acts as a filter. The remaining medium
acts only as a support. Most modern filters operate at higher flow rates and pressures to drive the
dirt down into the bed and make more use of the bed depth for filtration.
To improve the filtration efficiency, multimedia filters are increasingly being used in some
steam injection operations. In these filters a combination of sand, anthracite, and garnet is used.
The filter media is graded, so that the lightest and coarsest particles are at the top, and the fmest and
heaviest medium is at the bottom. With this type of arrangement, the coarsest particles are retained
at the top of the filter, and the finer particles are driven deeper into the filter bed to be collected on
the finer and more dense filter media. Thus, the whole filter bed takes part in the filtration
operation and results in higher throughput and longer filtration cycles.
Both upflow and downflow arrangements are used in the operation of the multimedia filters.
In the upflow arrangement, the flow is from bottom to top. The rated capacity of the upflow filter
is 6 to 8 USGPM/ftZ and the backwash rates vary from 15 to 20 USGPWftZ for a period of about
10 minutes. Figure 5.13 shows an upflow filter arrangement. Upflow filters can remove particle
sizes up to 10 microns without the use of a coagulant aid. Particle up to 2 microns in size can be
removed with the help of a coagulant aid such as alum. Though upflow filters can be used to filter
particles up to 2 microns, they are inherently unstable. A change in flow rate, pressure fluctuation,
or vibration can result in channeling and poor filtration.
In the downflow filters, the raw water enters at the top and flows downward. The capacity
of downflow filters are much higher than in upflow filters and are much more stable operational
wise. Downflow filters are usually rated at 15 to 20 USGPM/ft2 and are backwashed at a rate of
15 USGPM/ft2 for 10 minutes. Particles size up to 1 micron can be effectively filtered out with the
help of a filter aid. Because of its inherent stability and higher throughput, downflow
arrangements are more popular in the oil patch.
RELIEF VALVE
MANWAY
SIGHT GLASS
OUTLETS - FILTERED
& FLUSH WATER
COARSE GRAVEL
DlSTAlEUTlON PLATE
INLET NOZZLES
INLETS - MAIN
INLET MANIFOLD
CHEM. & DIP CONN.
100
essential to maintain a loose pack in the filter cake and to prevent premature plugging of the filter
cake. Without the body feed, the filtration efficiency will suffer. The higher the body feed rate,
the greater is the filtration rate and higher will be the operating costs. Hence, an optimum body
feed rate that results in highest filtration rate at the lowest cost should be used. In a typical thermal
EOR operation, the filtration rate is maintained at about 2 gaYmin/ft2 of filtration area. The body
feed concentration is maintained at about 0.7 p u t diatomite per part of suspended solid and oil.
Usually, a material handling system consists of storage silos, shakers, and screw feeders are used
to handle the volume of diatomite used in treating the raw water.
When the filter cycle is completed, the sluicing or washing cycle is initiated. The diatomite is
cleaned by the use of high-velocity water jets to remove the cakes. After sluicing, the filter casing
is opened, inspected and manually cleaned if necessary. The amount of backwash water needed to
clean the DE filter is much less than the pressure filters.
Selection of Diatomite
Several grades of diatomite are available for use as ;Ifilter aid. The choice of grade depends
entirely on the size and characteristics of the impurities to be removed and the effluent clarity
requirements. The finest diatomite should be used for removing tight emulsions, ultra fine
colloids, etc. The finest diatomite, however, has the lowest flow rate. The coarsest grade of
diatomite yields the highest flow rate and should be used to filter water containing coarse particles.
The final selection of filter aid (diatomite) should be made on the basis of field trials to avoid costly
misapplications. Filter companies and diatomite suppliers can conduct such tests at small cost to
prospective users and make recommendations.
Diatomite Filter Problems
Diatomite earth filters are very efficient in removing oil from water and have the advantage of
high flow rate and low space requirements. They are capable of reducing the oil content of water
from 20 to 0 ppm. Unfortunately, an improperly operated filter can be the biggest source of
trouble. Diatomaceous earth filters require more closely monitored operators attention than other
types of filters. This is because an improper or inattentive operation will result in a poorly coated
screen that results in holes, causing impurities and slurry feed to go through the filter and plug the
ion exchange resin beds. Past experience indicates that most operational problems can be traced to
poor operational practices. Some of the causes of DE filter failure in the past include the
following:26~~
1 . Operator leaves backwash valve open partially or completely after backwashing filter,
resulting in partial or complete bypassing of the filter. This is worse than no filter at all, because in
addition to bypassing the filter, diatomite is continuously fed to the water.
2. Malfunctioning of the body feed devices will result in inadequate precoating of the
screens and loss of filtration efficiency.
3. Temporary shutdown on a momentary pressure surge may cause part of the precoat to
drop off the screen. Subsequent operation of filter will result in all slurry feed and all suspended
solids to go through the filters and plug the lines.
4. Failure to flush the lines containing slurry (body feed) during backwash will cause lines
to plug and filter to fail.
Diatomaceous earth filters are very efficient in removing the oil from water and have the
advantages of high flow rate and low space requirements. However, if the filter cake becomes oil
saturated, some oil will leak through the filter. Hence to prevent the leakage the filtration is never
carried to completion. This increases the cost of a filter run. It is sometimes necessary to install
strainers or in-line filters downstream of a DE filter to strain out any solids going through the filter
because of malfunction.
A properly designed and operated DE filter will produce high quality effluent consistently.
Several filter companies offer a highly automated and packaged DE filter system that eliminates
most of the operator related problems.41 Such a system, however, is not cheap; and the savings
resulting from the reduction in labor, and cost of continuous attention, must be compared in a
decision made against initial capital expenditure.
Diatomaceous earth filters are not recommended for use in small steam injection operations
because of the logistics involved in the supply of diatomite and high cost of disposal of used
diatomite and the associated filtered solids.
DEAERATION
Deaeration of raw feedwater is critical to the control of corrosion of oilfield steam generators,
piping, and equipment. The primary source of corrosion in steam injection operation is probably
dissolved oxygen in raw water. Oxygen removal before water is heated is very important; its
importance is second only to the removal of hardness.
Oxygen Exclusion
The first step in controlling oxygen is to eliminate it at the source, if possible. Subsurface
source water does not contain oxygen; however, it is aerated inadvertently through improper
handling. By taking the necessary steps, oxygen pickup from the surface can be eliminated.
These steps include the use of a sealed type wellhead (with gas blanket) and use of packers to seal
water wells. Since raw water is usually stored piior to treatment, storage tanks should be roofed
and gas blanketed. The inlet to a tank must be submerged. An inlet (above water level) without a
gas blanket can pick up 5 ppm or more of oxygen. Submerging entrance piping to a storage tank
without gas blanket, while lowering the oxygen pickup, can still result in an oxygen pickup of
102
103
System
Disadvantages
Advantiiges
~
B. Vacuum
~-
flared generally.
2. Requires "sweet"gas free of
sulfur compounds.
3. Amount of tolerable C02 in
the gas is limited by pH,
scaling and associated problems.
1. Mech,mically simple.
2. Capable of reducing residual oxygen to
essentially zero.
3. Minimum initial capital cost.
C. Chemical Scavenging
lowering total pressure of the system. In the countercurrent gas stripping technique, the
concentration of dissolved oxygen in the surrounding atmosphere is reduced by dilution with the
scrubbing gas. In steam deaeration, the solubility is decreased by raising the water temperature.
Vacuum ~eaeration29
In the vacuum deaeration process, the partial pressure of oxygen, and hence its solubility in
the water, is lowered by decreasing the system's total pressure.
It is well known that water boils at different temperatures, depending on the pressure which
is maintained in the vapor space above the water. For example, water will boil at 60" F, if the
pressure reading is 0.5214 inches of mercury (0.256 psia) which corresponds to a vacuum of
104
about 29.4 inches of mercury. It is usually assumed that the solubility of any gas is zero at the
boiling point of the liquid. Therefore, by maintaining a vacuum of 29.5 inches of mercury,
oxygen can be completely removed from the water. Since it is not economical to maintain water at
boiling condition, complete removal of oxygen by vacuum deaeration is not possible. To maintain
an oxygen content of about 1 ppm in water at 60' F, approximately 28 inches of vacuum must be
maintained in the system.
For deaeration to be effective, two conditions must be met: (1) water must be kept in an
agitated state and (2) a very large surface area should be available. In the vacuum deaeration
operation, the water is sprayed ontcl a packed column maintained at about 28 inches of vacuum.
As the water droplets cascade through the packing, it forms a thin film over the packing. It
requires only a short distance for oxygen to travel for release in a thin film.
A properly designed and operated vacuum tower can reduce the oxygen content of water
down to about 0.8 ppm. A schematic of a vacuum deaerator setup used in a steam injection
operation is shown in figure 5.14. The dimensions of the tower will vary, depending on the
volume of water to be deaerated and the flow rate desired. For example, the dimension of a typical
tower used in one oilfield deaeration operation is 4 ft diameter by 55 ft high and is rated for 30,000
barrels of water per day.
Vacuum deaeration is seldom used in steam injection operations because of its high operating
expenses and complexity of operation. Further, the deaerated water must be chemically treated to
remove the residual oxygen.
Steam Deaeration30
Steam can be used to remove oxygen from source water. The principle of removal is to
(1) increase the temperature of the water (decreasing the oxygen solubility) and (2) put water vapor
(steam) into the gas space over the water (which decreases the partial pressure of oxygen). The
advantages of steam are as follows:
(1) It is readily available;
(2) it heats the water and reduces the oxygen solubility;
(3) does not contaminate the water; and
(4) most of the steam used to scrub the water is recovered as condensate and only a small
portion of the steam utilized to remove the noncondensible gases is vented to the
atmosphere.
Steam deaerators are widely used in power plants to deaerate boiler feedwater. The two major
types of steam deaerators are the tray-type and spray-type. Only spray-type deaerators are used in
the oilfield. Figure 5.15 is a schematic of a typical oilfield spray-type deaerator.
105
i-flRASCHlG
RING SECTIONS
<
i i
&*-I
Tlup
DIRECT CONTACT
VENT CONDENSER
OVERFLOW
CONNECTION
106
The major components of a spray-type steam deaerator include a carbon steel shell, springloaded inlet spray valve, direct contact vent condenser and a steam scrubber for final deaeration.
The inlet spray valve and direct contact vent condenser section are fabricated out of stainless steel.
The incoming feedwater is sprayed in a finely divided state into an atmosphere of steam in the
preheating section. The mist of water, upon coming into contact with the steam, is heated to near
steam temperature. Most of the noncondensible gases are released to the steam and the heated
water falls into a reservoir and drains to the lowest section of the steam scrubber. The heated water
is then directed into the path of the incoming steam.
Steam enters the deaerator through the steam inlet nozzle and is directed to the scrubbing
section via baffles. Here the large volume of steam scrubs and heats the water from the preheat
section to the saturation temperature. The intimate contact between steam and water in the scrubber
section results in the release of the remaining noncondensible gases. The deaerated water drops
into the storage section below for pick up by the feedwater pump.
The steam, after flowing through the scrubber, passes into the preheating section and heats
the incoming spray of water. Most of the steam condenses in the preheating section and becomes
part of the deaerated water output. Finally, the noncondensible gases are expelled through a vent
in the top of the deaerator, along with small amounts of steam into the atmosphere.
Though steam deaerators are capable of reducing the oxygen content of water to about 0.01
ppm, they are not widely used in steam injection operations because of capital equipment needs and
operational and maintenance problems. Some of the operational problems include clogging of
spray valves and the inlet distributors and scaling problems. Further, these units require constant
operators attention to ensure that the noncondensible gases are properly vented to keep the partial
pressure of oxygen to a minimum. Use of large amounts of steam to deaerate the water increases
the process steam demand and decreases the net steam available for injection because of fixed
steam generation capacity. Further, the size and cost of the apparatus increases sharply with the
volume of water to be deaerated. Since it is not physically possible to completely deaerate the
water using a steam deaerator, the need for a chemical oxygen scavenging system is not eliminated;
however, the scavenger requirements will be considerably less otherwise.
Gas Stripping3 I
Use of natural gas to strip oxygen from water is practiced in large waterflood operations, but
no steamflood operation using this technique has been reported. The underlying principle behind
this deaeration technique is the reduction of the concentration of the dissolved oxygen in water by
diluting it with the stripping gas. This reduces the partial pressure of oxygen in the gas mixture
and causes oxygen to come out of the water.
107
A schematic of a countercurrent gas stripping unit is shown in figure 5.16. Gas stripping is
normally performed in bubble tray column. Bubble tray column promotes intimate contact between
the water and gas by providing a large surface area.
m t i o n : The water enters near the top of the column and flows down through the trays
passing across each tray and then down the staggered downcomers to the next tray. The stripping
gas enters the column near the bottom and passes upward through the bubble caps on the trays. As
it flows up the column, the gas bubbles go up through the downflowing water and remove a good
portion of the oxygen. The gas leaves the top of the vessel through a stainless steel wire mesh mist
extractor, through a gas outIet line, and a gas backpressure valve. The deaerated water collects in
the bottom of the column. The discharge is controlled by a liquid level control and a large capacity
diaphragm dump valve.
These columns are usualIy designed to operate at 50 psig or less. With higher operating
pressure, more trays are required, and the cost will increase rapidly. Typical specifications for
oxygen desorption towers are shown in Table 5.13. Gas stripping operations are not as efficient
as vacuum deaeration and can lower the oxygen content of water to about 0.9 ppm. Gas stripping
units, however, are less expensive to operate than a vacuum tower. Chemical scavenging must be
used to remove the residual oxygen.
OXYGEN
DESORPTION
TOWER
30
36
42
48
54
60
66
72
Tower
height,'
ft
30
30
30
30
30
30
30
30
Working
pressure,
psi
50
50
50
50
50
50
50
50
No. and
type of
trays
10 single
10 single
10 single
10 single
10 single
10 single
10 single
10 single
pass
pass
pass
pass
pass
pass
pass
pass
4,400
6,340
8,680
11,350
14,460
27,690
2 1,350
25,390
6.600
9.5 10
13,020
17,030
2 1.70
26,540
32,030
38,080
8,800
12,680
17,360
22,710
28,930
35.390
42,710
50,780
Gas
con~umption,~
MSCFIday
44.0
63.4
86.8
113.6
114.7
117.0
213.6
253.9
1 B S on
~ using 10 trays on 24' tray spacing.
2 Capacities based on 30% downcomer area, 80% of flood, and 4 secoiid rctcnhon time in downcorners.
3 Gas consumption based on 5 cu fthbl of water and maximum water capacity.
Chemical Deaeration6
Removal of oxygen from water by chemical means is the most popular and widely practiced
technique in steam injection operations. Several methods of chemical scavenging exist. These
include catalyzed sodium sulfite treatment, hydrazine treatment, and sulfur dioxide method.
The use of catalyzed sodium sulfite is cheaper than other chemical treatment methods and is
the chemical of choice with steam injection operators. It is inexpensive and is very effective in
removing oxygen from water. Theoretically, 8 ppm of sodium sulfite is needed to remove each
1 ppm of dissolved oxygen. The oxygen scavenging reaction is given by
The above reaction is very slow at ambient temperature. For rapid removal of oxygen at ambient
temperature, a small amount of catalyst (about 10 parts per billion, ppb, in the water) such as
cobalt sulfate must be added to the sulfite solution. Catalyzed sodium sulfite will remove the
oxygen in a matter of seconds as compared to minutes with the uncatalyzed compound. To be
certain that all of the oxygen is removed, an excess sulfite is added to the water so that a sulfite
residual of 20 to 30 ppm is carried into the process water. It is recommended 10 Ib of catalyzed
sodium sulfite be used per pound of oxygen.6
The sodium sulfite is usually added to the water at a point downstream of water softener.
Some operators inject the sodium sulfite solution upstream of the softener to provide adequate time
109
for the chemical to react. The advantage of injecting ahead of the softeners is that it will protect the
steel parts of the softeners and the steel piping against oxygen corrosion. However, addition of
sodium sulfite ahead of the softener will increase the load on the softeners from the added sodium
and will cause extra hardness leakage. Further, the excess may not be available to scavenge any
oxygen that may enter the system due to leaky valve, pump failure, etc.
Both catalyzed and uncatalyzed sodium sulfite are readily available from water treatment
chemical companies. Since uncatalyzed sodium sulfite is less expensive and the amount of catalyst
needed (0.001 ppm of cobaltous ion in water) is very small, it is recommended that uncatalyzed
sodium sulfite be purchased in bulk from chemical supply companies and the catalyst be added in
the field. The catalyst, cobaltous hexahydrate, can be purchased and fed with sodium sulfite as a
catalyst without materially increasing the cost of sulfite. The equipment needed for sodium sulfite
treatment includes storage vessels, mixing tanks, and chemical feed pumps.
Although sulfite is inexpensive and generally satisfactory for removing oxygen from the
feedwater, there are times when hydrazine is used in preference to sulfite. If the oxygen content of
the source water is consistently low, hydrazine may be economical to use because little excess is
required for complete oxygen removal at high temperature. Only 3 pprn of hydrazine is needed to
remove I. ppm oxygen from the water. Further, at temperatures above 500' F, the sulfite may
decompose to give sulfur dioxide which is corrosive. Hydrazine is stable at these temperatures.
Also, hydrazine does not require mixing or large storage vessels and is easy to feed.
Hydrazine is considerably more expensive and the reaction rate is very slow at room
temperature but can be increased somewhat by using a catalyst. Even with the use of a catalyst, the
rate of hydrazine-oxygen reaction is not fast enough to make use of hydrazine practical in many
steam injection applications. Also, hydrazine is a hazardous chemical, and certain precautions in
handling should be observed. In most thermal operations, the cost of hydrazine prevents it from
being used where high concentrations of oxygen are present in the feedwater.
110
A large California steam injection operator utilizes this method to scavenge oxygen from
feedwater. Theoretically, 4 lb of'sulfur dioxide is required to scavenge 1 lb of oxygen from water.
A sulfur burner is used to generate sulfur dioxide for scavenging oxygen. The sulfuric acid and
the excess sulfurous acid is neutralized by the natural bicarbonate alkalinity of the water as follows:
H2SO4 + 2NaHC03
-+Na2S04 + 2H2CO3
The use of a sulfur burner to generate sulfur dioxide for scavenging oxygen from boiler
feedwater is well known and practiced since the early 1940s by the chemical process ind~stries.3~
However, this technique was not adopted by the oil industry until the early 1960s, when the first
such unit was installed in a West Texas waterflood project to deoxygenate the process water.32 A
schematic of the sulfur burner process used in the West Texas field is shown in figure 5.17.
WATER
SOURCE
111
The sulfur burning method of scavenging oxygen from feedwater was adopted by California
steam injection operators in the early 1980s. The overall process scheme used in steam injection
operations is similar to those used in waterflood operations, but differs in the specifics because of
the special needs of the steam injection operation. In this process, the sulfur is supplied to a sulfur
burner as required to produce the sulfur dioxide necessary to scavenge oxygen in the incoming
water. The hot sulfur dioxide is cooled to about 150" F by passing it through a water-cooled heat
exchanger. The cooled sulfur dioxide is then introduced to the bottom of an absorption tower
composed of ceramic packings. A side stream of source water taken from the source water storage
tank is introduced into the top of the tower at the desired rate. The process water is sprayed onto
the packing through the spray nozzles. The water reacts with the upflowing sulfur dioxide and
converts it into diluted sulfurous acid which falls to the bottom of the tower. This solution is then
pumped into the main stream of water. A dilute solution of cobalt chloride is added to the water to
speed the reaction of sulfurous acid with the dissolved oxygen.
The absorption tower design, operation, and problems are similar to those of the generator
flue gas scrubber. According to the operator, the installation costs for such a system were slightly
over $175,000 (1991 dollars), and the operating costs were about 1 @/barrelof water treated.
Another inexpensive source of sodium sulfite is the generator flue gas scrubber liquor.33
Depending on the sulfur content of the fuel oil burned in the generator, the scrubber liquor contains
between 3% and 5% sodium sulfite which can be used to scavenge oxygen from the feedwater.
The use of scrubber liquor to scavenge oxygen not only eliminates chemical requirements, but will
also lower the spent liquor disposal cost. According to one California steam injection operator
who utilizes scrubber wastes to scavenge oxygen, the total capital investments needed to implement
such a system is slightly over $35,000 (1991 dollars) and includes a 200-bbl fiberglass holding
tank for scrubber liquor; a chemical feed pump; two 450-gpm cartridge filters; and associated
pipes, valves, and fittings.
The requisite volume of scrubber liquor is transferred from the scrubber waste tank into the
fiberglass holding tank. This allows the heavy metal precipitates and other debris to fall to the
bottom of the holding tank. The clear supernatant liquor then flows through the cartridge filter and
introduced into the upstream of the ion exchanger units. According to the operator, between 1 and
2 bbl of scrubber effluent (containing 5% sodium sulfite) was required to remove 4 ppm oxygen
from 1,000 bbl of feedwater. The net savings in the operating and chemical costs were such that
the operator was able to recover his investment in less than 6 weeks.
It should be realized, however, the benefits of using scrubber waste to scavenge oxygen may
not be realized in all operations. This is because the scrubber waste contains high amounts of
sodium (as high as 105,000 ppm in some liquor) which makes an additional sodium load on the
ion exchange resin and may cause increased hardness leakage. The scrubber liquor is also high in
112
iron, which is harmful to the resin. Also, if the amount of oxygen in the steam generators is
restricted in the burn process, sulfur, sodium sulfide, thiosulfate and hydrogen sulfide may also be
generated.l Sulfur, sodium sulfide, and thiosulfate are known to inhibit the ability of sodium
sulfite to remove oxygen from the generator feedwater.6 Hence efforts should be taken to ensure
that generators operate with sufficient oxygen to minimize the harmful effects of oxygen
deficiency. Also, a complete analysis of the scrubber waste must be performed and its oxygen
scavenging potential established prior to its use as an oxygen scavenger. A typical SO2 scrubber
waste analysis is shown in Table 5.14.
Constituent
pp m
105,000
less than 1
35
2,400
0
44,000
53,000
84,000
33,000
160
Sodium
Calcium
Iron
Chloride
Carbonate
Bicarbonate
Sulfate
Sulfite
Bisulfite
Silica
Total dissolved solids
225.000
5,600
PH
6.7
113
However, before the produced water can be used for feedwater, the residual oil, suspended
solids, and hardness must be removed. The required waste water treatment operations are more
complex than conventional fresh water treatment techniques. The complexity and corresponding
higher costs of produced water treatment have made reclamation uneconomical in many small
operations. However, in larger operations, recycling is a viable alternative to conventional water
handling practices due to high fresh water requirements and high disposal costs.
Before designing the treatment facility, the geochemical analysis of the produced water and
its scaling tendencies must be investigated. Oilfield produced waters show a wide variance in the
TDS and hardness content and contain significant amounts of oil and suspended matter.
Experience indicates that conventional sodium zeolite softeners are not effective at a TDS level
above 20,000 ppm, and weak acid softening resins must be utilized to remove the hardness. The
TDS of oilfield produced water from thermal EOR fields in California varies from 1,500 ppm in
the valley to more than 30,000 ppm near the coasts. The TDS and hardness of the produced water
can usually be lowered, and the softening costs reduced, by blending it with municipal or city
water. The favorable economics depend on the ready availability of large volumes of city water at
cheaper costs. Most major operators in California blend the produced water with city water before
softening to reduce costs. Because of their high TDS content, produced waters are more likely to
cause plugging problems in the generator tubes, if the steam quality is allowed to exceed the
solubility limits of sodium salt. The use of produced water does require monitoring steam quality
more closely.
Silica Scaling
Another major area of concern regarding produced water use has been the possibility of silica
scaling. The produced water from a steamflood usually contains between 100 and 400 ppm silica
as Si02. This silica is dissolved from the quartz present in the formation by the unvaporized and
condensed water that has a pH >9. Silica forms scales at high pressure in the presence of metal
ions such as iron, aluminum, calcium, and magnesium. Although API recommendations permit up
to 150 ppm silica in generator feedwater, fields studied indicated that up to 9.9% of inlet dissolved
silica will precipitate in one form or another in the convection and radiant sections of generator^.^^
Even a trace amount of metal ion will result in silicate scale formation in boiler tubes. Factors that
contribute to silicate precipitation include not only the silica level but also salinity, alkalinity
(HCO3-), pH, and trace amounts of metal ions.
Depending on the vapor velocity, entrained liquid droplet, and water film thickness, silicate
scale can be mobilized and cause erosion and eventual failure of the tubes.30 Field studies
indicated that irrespective of the steam quality, silica precipitation will occur, and the metal loss is
higher at higher quality steam due to greater rate of transport of the deposit present. Further,
114
neither the acceptable level of silica concentration in the feedwater required to produce 75% or
higher quality steam, nor a .water treatment process to achieve this concentration, is known.
Since the solubility of silica increases rapidly above pH 10, it is suggested that silica
deposition can be prevented by pH modification; however, adding caustic to buffered waters is
difficult to c ~ n t r o l .If~ soda ash is used to increase pH of the water, it may decompose in the
generator and release carbon dioxide to accelerate corrosion.
To mitigate silicate scale problems, high silica waters should generally be avoided in steam
generation processes. If the use of high silica water cannot be avoided, silica must be removed by
precipitation or by total demineralization (anion exchangers) to avoid scaling problems. Since the
formation of silica scales requires the presence of metal ions such as iron, calcium, and
magnesium, complete removal of these ions is technically and economically more attractive than
silica removal.
Reclamation of produced water for steam generation involves the following steps:
(a) Flotation Process: The induced gas flotation process further reduces the oil and
suspended solids to about 20 ppm and 80 ppm, respectively.
(b) Filtration: DE filters reduce the oil and suspended solids in the produced water to 0
PPm.
(c) Oxygen Scavenging: Chemical scavengers are used to eliminate the dissolved oxygen,
(d) Ion Exchange Softener: Softening is done in two stages-primary and polisher-to
reduce the hardness to 0 pprn.
Figure 5.18 summarizes the sequence in which the unwanted constituents of the produced
water are removed. Operational details of the equipment used in these steps were discussed in a
previous section.
Daily monitoring should form an integral part of produced water reclamation plant operation.
Monitoring provides adequate warning of possible upset conditions. The following components of
the produced water must be monitored constantly.
Oil concentration in incoming water
Filter pressures Dissolved oxygen and iron
Water hardness
Plant inflow and outflow volumes.
The oil concentration is probably the most critical factor since high concentrations will increase
filter differential pressure and reduce filter cycle length. A significant reduction in filter cycle
would curtail generator feed supply and may result in generator shutdown.
**
115
CHEMICAL
TREATMENT
vT
PRODUCEDWATER
OXYGEN
SCAVENGER
IN1 FT
1
I
SEDIMENTATION
BASIN
DEPURATOR
UNIT
NITROGEN
DIATOMCEOUS
FEEDWATER TO
STEAM GENERATORSf7
CAKE
riLitn
L A I IUN
EXCHANGE
SOFTENER
FLOTATION
CELL
4
DISCHARGETO
JRREGATJON CANAL
DEPURATOR
UNIT
FIGURE 5.18. - Flow diagram for a Kern River, CA, steamflood produced water reclamation
plant.18
Higher filter pressure is an indication not only of high oil concentration but may also indicate
bacterial growth in the settling tanks. High oil concentrations are controlled by ensuring proper
operation of the upstream dehydration plant. Bacterial growth is controlled by periodic biocide
treatment of the holding tanks.
Dissolved oxygen and iron are monitored as corrosion control indicators.
Monitors are utilized for identifying hardness leakage and to indicate when corrective action
is warranted.
Plant inflow and outflow volume monitoring is needed to ensure efficient plant operation.
116
Underground injection wells associated with the disposal of oilfield produced waters are classified
by the U.S. Environmental Protection Agency (EPA) as Class I1 wells.
Produced water must be treated before its disposal. For example, current federal regulations
require that the oil content of produced water not exceed 48 ppm. The methods of treatment and
equipment used for produced water disposal are similar to those used in the reclamation of
produced water for steam generation. Treatment of produced water need not be elaborate. The
operators objective should be to provide only sufficient treatment to meet the regulatory
requirements and render the water suitable for injection into formations without decreasing the
injective capacity of disposal wells. It is essential that thorough laboratory studies be conducted on
the proposed injection water, thereby determining the minimum water treatment requirements;
scrubber wastes must be neutralized and filtered before injection. If the produced water is to be
discharged to a surface facility such as a canal, a more thorough treatment may be in order.
The treatment and disposal of steamflood produced water in a pollution-free manner is an
essential part of operational expenses and must be so recognized. With proper planning, design,
and operation, produced water may be disposed of at a minimum cost. Produced water disposal
system design, operation, and economics are beyond the scope of this report. Steam injection
produced water treatments are covered in a previous section of this report. Disposal of the
produced water by underground injection is similar to that for oilfield produced brines and is
covered thoroughly in an API p ~ b l i c a t i o nwhich
~ ~ should be consulted.
117
TABLE 5.15. - Typical Water Analysis for Water Treatment Cost Estimation18
~-
Impurities
~~
Sodium
Iron
PH
Oil,ppm
640
260
570
90
15
45
95
70
60
60
7.5
---
5,000
960
270
75
20
20
270
170
2,000
1,540
7.4
20
Fresh and produced water treatment costs are based on the following chemical prices:
Diatomaceous earth, ton
Chlorine gas, ton
Oxygen scavenger, lb
(25%sodium sulfite solution)
Sodium chloride (salt), ton
Hydrochloric acid (HCI), ton
Sodium hydroxide (NaOH), ton
$300
$270
$0.75
$39
$92.5*
$325
118
Chemical
Chemical usage
rate
Chemical
requirement
per bbl water
w r bbl water
Bacteria control
Oxygen removal
Strong acid
Chloriile
Sodium sulfite
0.1 gm
10 gm
3x
0.016
softening &
regeneration
Sodium chloride
0.7 Ib
a914
Process
Chemical
costs, $
0.020
Power
Operational (labor)
Repair and maintenance
0.003
0.015
iu-m
$0.04
Process
Filtration
Oxygen
Chemical
Chemical usage
rate
Chemical
requirement
per bbl water
Chemical
costs, $
per bbl water
Fiber and DE
0.7 part DE
sodium sulfite
9 mghpm 0 2
0.03 Ib
10 gm
$0.02
NaCI
0.3 Ib
0.005
HCI
0.16 Ib
0.02
NaOH
0.17 Ib
removal
Strong acid
softening &
regeneration
Weak acid
Softening &
regeneration
0.01
0.08
Power
Labor
Repair & maintenance
0.01
0.06
nsl
TotaVbbl
$0.165
119
success of a steam injection project can be just as dependent on water handling practices as on such
items as steam generation, reservoir performance, and sweep and displacement efficiency. As
previously mentioned, a problem-plagued water treatment system will result in costly steamer
downtime and loss of production. A properly engineered and operated water treatment system can
result in a more profitable operation than otherwise. The purpose of this section is to outline what
needs to be considered in the design and operation of water treatment system to ensure maximum
benefits from the installed equipment.
Design Concept
Design of a water treatment system for steam injection requires careful planning and
engineering. It may require more technical effort initially, but will assure relatively trouble-free
operation over the life of the project. There is no such thing as a standard water system design.l
Each project is unique and has its own water processing requirements. A treatment facility must be
tailored realistically to meet anticipated needs, with provision for addition or expansion when
required. In designing the process, the objective should be to keep it simple. The simpler the
system, the better chance it will be operated successfully. During the planning and design stage,
major potential problems should be identified and taken into account. Since everything cannot be
foreseen, no attempt should be made to anticipate all problems and provide for all contingencies.
The approach to the planning and designing of a water treatment system for steam generation
can be broken into three steps, and certain minimum information is required for each step. In
general terms, the timing of items that need to be considered are as follows:
(a) Feasibility Stud ies: The planning of the water treatment system should coincide with
other early work on the project. At this stage, identify all water sources capable of meeting the
projects water requirements. More often than not there is only one adequately sized water source
existing at a given site. In cases where multiple water sources exist, it is necessary to consider
them separately to determine the procurement (either drilled, purchased, or piped in) and
processing costs for each source. The main objective at this step is to select the most likely water
source and rough-in expected treatment equipment requirements and order-of-magnitude costs.
The quality of feedwater will determine the equipment requirements and operating costs. The
analyses of several typical source waters used in the California steam injection projects are shown
in Table 5.2. Each type of water presents a different set of problems.
In this step, water quality data are refined, and comparative
(b)
economics of alternatives are made using firmer numbers. More exact determination of equipment
requirements, materials, and operating costs should be considered.
w:
(c) Final Desicn: This step must cover such items as exact equipment sizing, location, and
brand names. Installation drawings should be reviewed to locate sampling and coupon monitoring
120
points, drain and flushing locations, etc. While these details may seem insignificant, they are
extremely important to the profitable operation of the project.
The complexity of the water treatment system depends on the quality of the source water.
The better the quality of raw water the less complex the system will be. Subsurface sources of
water usually require less equipment to treat and are preferred over other sources. Steam injection
demands the handling of large volumes of water. For various reasons, a single source may not be
able to meet a projects water demand, and it is a poor practice to design a water handling system
based on a single source. Overall plans should include the handling of water from different
sources. Flexibility should be built into the initial design to provide for the handling of less than
desirable water. As a generality, the water treatment system should be designed to treat the readily
accessible and inexpensive water that is available in sufficient quantities, with provisions for
addition of supplementary equipment if and when necessary. Most water can be treated with a
reasonably simple system, if sufficient emphasis is placed on careful planning during design.
121
Eq u ipm en t Consideration
At each step prior to the final plans of a steam injection project, the type of equipment
believed necessary to produce water of acceptable water quality should be reviewed. Since it is not
possible to predict with certainty all of the possible problems which can arise in a
15- to 20-year steam injection project, provisions should be made for possible future additions of
water-handling equipment. Only that equipment needed to start plant operation should be installed
initially. If additional problems arise regarding water quality after a plant is in operation, a detailed
study must be conducted to identify the causes and possible solutions. If the study indicates the
need for new equipment and/or chemical treatment to achieve the desired boiler feedwater quality,
then the system should be modified accordingly. Installation of each piece of new equipment must
be justified economically.
In the design of pipelines, consideration should be given to the possibility of hydraulic shock
and surge which may damage pipes of low tensile strength. Provisions should be made for the use
of cleanout pigs for removal of scale and other fouling materials. Air pocket vents should be
included. Leak-testing equipment should be available. If separators are to be used, they should be
selected based on the volume of suspended solids and type of oil that must be removed, the
reactivity of the water, and the value of the land on which they would be located. Separators,
however, should be used only if the amount of free oil is expected to exceed the handling capacity
of the depurators. This can occur when the dehydration equipment malfunctions. Efforts should
be made to prevent upsets of dehydration equipment. Installation and operation of deaeration
equipment is very expensive, Before considering deaeration equipment, other methods of control
of oxygen should be thoroughly evaluated. In the case of dissolved oxygen, elimination of the
source of oxygen can be the most profitable approach. Even the shallow water sands do not
contain dissolved oxygen, and the oxygen found in these waters is from the surface. Oxygen
pickup from the surface can be eliminated by proper design and gas blanketing of the supply well.
All chemicals, which may include bactericide, oxygen scavengers, de-emulsifiers etc.,
should have a mix hopper provided with a transfer arrangement to a feed tank. In the case of
sodium sulfite, the tanks should be closed and blanketed to prevent oxygen contamination. All
chemicals should be fed into the system through a variable metering type injection pump. Materials
should be selected for corrosion resistance to chemicals in the concentrated and dilute states.
122
Frequently, once a system is installed, put on stream, and debugged, the operation and
maintenance of the water-handling system is relegated to the backburner until trouble developssoftener resin deterioration, equipment scales, corrosion failures etc. Considerable amounts of
money are usually spent on remedial work that often destroys a projects profit picture.
As with other preventive maintenance, water-treatment system monitoring can go a long way
towards reducing such problems. A few thousand dollars spent on monitoring has a large potential
payout in decreased operating costs. A systematic review and evaluation of water quality will help
detect changes and conditions before serious problems develop. Automatic and manual hardness
and oxygen monitoring should be made to periodically determine and follow water treatment plant
performance. Oxygen content at the discharge of each piece of equipment should be checked
periodically. Performance of the ion exchange bed should be monitored carefully. Excessive
pressure-drop through the bed and poor quality treated water are indications of resin degradation.
As a rule, about 10% of the resin bed should be replaced annually. Using such a systematic
approach, the efficiency of water treatment equipment and its operation may be monitored. Such
information, in conjunction with the steam generator performance and the injection well
performance data, can be utilized to evaluate the flood performance.
Apparently, insignificant items can greatly influence water quality. In a well designed and
operated water-handling system, most of these can be classed as good housekeeping items. Yet
the importance of good maintenance programs cannot be over-emphasized. Routine maintenance
schedules and reporting procedures should be set up and periodically reviewed. Last, but certainly
not least, is the importance of operators training. Many operating problems can be avoided or
minimized by personnel who understand the what, why, and how of good maintenance
management.
Finally, it should be kept in mind that the steam injection process is, above all, an oilfield
operation. Each project is unique and has its own problems. Circumstances, when fully
understood, will dictate the best course of action.
SUMMARY
Quality of boiler feedwater is of critical importance in steam injection projects. Past field
experience indicates that the majority of steamer downtime is caused by water treating problems.
Therefore, it is imperative that feedwater problems be examined with care. This chapter discusses
in detail the feedwater treating requirements for a single-pass steam generator and the effect of
specific impurities on steamflood equipment performance. Guidelines to water treatment system
design are also provided. Water treatment softening-unit operations and problems are detailed.
Process requirements for reclaiming produced water for steam generation are also detailed.
123
REFERENCES
V.
1. Burns, W. C. Water Treatment for Once Through Steam Generators. J. Pet. Tech.,
17, NO. 4, Aphil 1965, pp. 417-421.
2. Hagist, F. C. and D. R. Fincher. What You Ought to Know About Water Treatment for
Thermal Recovery. Oil & Gas J., v. 63, No. 46, Jan. 11, 1965, pp. 64-70.
8 . Roberts, J. C. and J. W. Williams. Steam Injection, Oil Recovery Techniques and Water
Quality. J. Water Pollution Control Federation, v. 42, No. 8, Part 1, August 1970, pp. 14371445.
9. Haseltine, N. G. and C. M. Beeson. Steam Injection Systems and Their Corrosion
Problems. Materials Protection, v. 4, No. 10, October 1965, pp. 57-61.
10. Reyes, R. B. Softening of Oilfield Produced Water by Ion Exchange for Alkaline
Flooding and Steamflooding. Pres. at the California Regional meeting of the SOC.of Pet. Eng.,
Ventura, CA, Mar.23-25, 1983. SPE paper 11706.
11. Bradley, B. W. Influence of Salt Dosage and Hardness on Series Softener Performance.
Pres. at the 42nd Ann. Fall Tech. meeting of the SOC.of Pet. Eng., Houston, TX, Oct. 1-4, 1967.
SPE paper 1951.
12. Serck Baker-Water Softener Systems-Oil Field Steam Generation and Other Enhanced
Recovery Techniques. Tech. Bulletin, Serck Baker Inc., Huntington Beach, CA, 1982.
13. Owens, M. E. and B. G. Bramley. Performance of Equipment Used in High Pressure
Steam Floods. J. Pet. Tech., v. 18, No. 12, December 1966, pp. 1525-1531.
14. Brusset, M. J., A. N. Edgington and R. F. Gilmour. Equipment Performance in an
Alberta High Pressure Steam Injection Project. Paper pres. at the 18th Ann. Tech. meeting of the
Pet. SOC.of CIM, Banff, Alberta, Canada, May 24-26, 1967.
15. Bealessio, M. Sources of Fresh Water for Steam Generators. Producers Monthly,
v. 32, No. 9, September 1968, pp. 8-10.
124
16. Bradley, B. W. and L. K. Gatzke. Steamflood Heater Scale and Corrosion. J. Pet.
Tech., v. 27, No. 2, February 1975, pp. 171-178.
17. Welder, B. Q. Automation Does Well in Steam Injection: The Oil & Gas J., v. 63,
NO. 46, NOV.15, 1965, pp. 164-173.
18. Carrell, M. A. Reclaiming Produced Water for Steam Generation in the Kern River
Field. Pres. at the 54th Ann. Fall Tech. Conf. of the Soc. of Pet. Eng., Las Vegas, NV, Sept. 2326, 1979. SPE paper 841 1.
25. Sivalls, R. C. Water Treating Equipment Design Manual. Sivalls Inc., Tech. Bulletin,
Sivalls Inc., Odessa, TX, 1969.
26. Olmsted, B. C. and G . R. BelI. Diatomite Filtration of Water for Injection. Paper
presented at the Third Biennial Symp. on Microbiology, sponsored by API Pacific Coast District
Study Committee on Treatment and Control of Injection Waters, Nov. 29-30, 1964, Anaheim,
CA, pp. 201-219.
27. Smith, M. L. Waste Water Reclamation for Steam Generator Feed, Kern River Field,
California. Pres. at the 42nd Ann. California Regional meeting of the SOC.of Pet. Eng., Los
Angeles, CA, Nov. 4-5, 1971. SPE paper 3689.
125
3 1. Weeter, R. F. Desorption of Oxygen From Water Using Natural Gas for Countercurrent
Stripping, J. Pet. Tech., v. 17, No. 5 , May 1965, pp. 515-520.
32. Perry, L. N. and W. J. Frank. Sulfur Burning Method of Scavenging Oxygen From
Water. Paper 906-11-E, pres. at the API Southwestern District Spring meeting, Division of
Production, Hobbs, NM, March 1966. API Drilling and Production Practices, 1966, pp. 70-73.
34. American Petroleum Institute. Subsurface Salt Water Disposal. Book 3 of the
Vocational Training Series. American Petroleum Institute, Division of Production, Dallas, TX,
1960.
35. Beckett, J. K. Low-Cost Water Treatment Solves Disposal Problems. World Oil,
v. 173, No. 6, November 1971, pp. 64-67.
36. Burton, R. Waste Water Recycling in Steamflood Operations. Pres. at the California
Regional meeting of the SOC. of Pet. Eng., Ventura, CA, Mar.23-25, 1983. SPE paper 117 10.
37. Khatib, 2. J., E. E. Olson and M. C . Place Effect of High Silica Content on Scale
Deposition and Pipe Wall Loss in Oilfield Steam Generators. Pres. at the 64th Ann. Tech. Conf.
and Exhib, of the SOC.of Pet. Eng., $an Antonio, TX, Oct. 8-1 1, 1989. SPE paper 19760.
38. BETZ Handbook of Industrial Water Conditions, 8th Ed., Chapter 8: Ion Exchange,
p. 54. BETZ Laboratories Inc., Trevoise, PA, 1980.
39. Kemmer, F. N.: The NALCO Water Handbook-Chapter 12: Ion Exchange, pp. 11,
McGraw Hill Book Company, New York, 1979.
40. Battle, J. L.: Salt Water Disposal. Pres. at the 15th Ann. Southwestern Petroleum Short
Course, Lubbock, TX, Apr. 18-19, 1968, pp. 193-206.
41. Maloney, G . F.: Diatomite Filters-Methods of Automation. Paper pres. at the Third
Biennial Symp. on Microbiology, sponsored by API Pacific Coast District Study Committee on
Treatment and Control of Injection Waters, Nov. 29-30, 1964, Anaheim, CA, pp. 170-186.
42. Ostroff, A. G.: Introduction to Oilfield Water Technology, pp. 88, Prentice-Hall Inc.,
Englewood Cliffs, NJ, 1965.
GENERAL REFERENCES
BETZ Laboratories. BETZ Handbook of Industiial Water Conditioning, 8th Ed., Chapters 6, 8
and 9. BETZ Laboratories, Inc., Trevoise, PA, 1980.
Crawford, P. B. Water Processing For Thermal Oil Recovery Programs. Producers Monthly, v.
29, No. 4, April 1965, p. 18.
Crawford, P. 3. Water Treatments For Steamfloods-Part 11-Hardness
Exchangers. Producers Monthly, v. 29, No. 11, November 1965, p. 28.
126
Removal by Ion
127
APPENDIX 5-A
WATER CHEMISTRY
D ef i n i t i o n
The purpose of this appendix is to define terms frequently encountered in oilfield water
chemistry. The treatment by necessity is very simplistic, but sufficiently accurate for our
purposes.
Elemen& are chemical substances which cannot be chemically decomposed to give two or
more simpler substances. Examples of elements of importance in oilfield water chemistry (and
their symbols) include: hydrogen (H), sodium (Nn), calcium (Ca), magnesium (Mg), oxygen (0),
and carbon (C).
Elements are made up of matter called ptomg. An atom is the smallest particle of an element
that can enter into a chemical reaction.
An atom is made up of electrons, protons and neutrons. An electron carries a negative
electrical charge and has negligible mass. The proton carries a positive electrical charge and has a
unit mass. The neutron carries no charge and also has unit mass.
The nucleus of an atom is made up of protons and neutrons. Thus, the nucleus is positively
charged. In an atom the electrons move about the nucleus. The positive charges in the nucleus are
balanced by the negatively charged electrons orbiting it. Thus, an atom is electrically neutral.
Since the mass of an electron is negligibly small, for all practical purposes, the mass of an atom is
considered to be the mass of its nucleus. The mass of an atom is equal to the sum of the number of
protons and neutrons in the nucleus.
The number of electrons in an atom is always equal to the number of protons in the atom, but
the number of neutrons in the nucleus of an atom may or may not be the same as the number of
protons in it. When the number of protons and neutrons in an atom are different it is called an
isotope of an element.
For the purpose of this report, the number of protons and neutrons in a nucleus are assumed
to be the same. An exception to this is the nucleus of the hydrogen atom. The nucleus of the
element hydrogen contains one proton and no neutron. Only the electrons in an atom take part in a
chemical reaction.
Atoms are identified by name, atomic number and atomic mass. The atomic number is the
number of electrons moving around the nucleus of an atom. It is also the number of protons in the
nucleus. The mass of an atom is the sum of protons and neutrons in the nucleus. The atom
hydrogen has one electron orbiting about a nucleus made up of one proton and no neutron. Thus,
its atomic number is 1 and its atomic mass is also 1. The nucleus of a calcium atom is composed
of 20 protons and 20 neutrons. Thus, its atomic number is 20 and its atomic mass is 40.
129
The atomic weiPht of an element is'the ratio of the mass of an element to that of an arbitrary
standard element, the carbon atom. The atomic mass of a carbon atom is 12. Therefore, when we
say that the atomic weight of hydrogen is 1, it actually means that a hydrogen atom weighs 1/12 as
much as an atom of carbon.
:Elements combine to form molecules . For example, an oxygen
molecule is comprised of two oxygen atoms. Molecules made up of a number of different atoms
are called compounds. A compound is formed by chemically combining two or more elements in
definite proportion by weight. It is not possible to identify individual elements in a compound,
unless the compound is chemically separated.
Molecular w e i a is the weight of a single molecule and is equal to the sum of the weights of
its constituent atoms. Molecular weights, like atomic weights are also relative weights. When we
say that the molecular weight of water is 18, we mean that a molecule of water weights 18/12 times
the weight of a carbon atom.
When the molecular weight of a substance is expressed in grams, it is termed as gram
molecular weight or simply gram-molg or &.
130
The elements and radicals present in water combine with each other on an equivalent basis to
form a salt. For example, one equivalent weight of Na+ combines with one equivalent weight of
SO4- to form the salt sodium sulfate Na2S04.
PH
A pH of solvent such as water is the measure of its acidity or alkalinity. A pure water
molecule dissociates itself into hydrogen ions (H+) and hydroxyl ions (OH-) according to the
equation
H20 w H+ + OHThe extent to which water dissociates is given by the dissociation constant k. The dissociation
constant is defined as the product of the number of hydrogen and hydroxyl ions present in one liter
of water. Thus,
Thus, there are 10-7 moles of H+ ions present in 1 liter of water. Since this is such a small
number, it is more convenient to express the hydrogen ion concentration in terms of a logarithmic
function. This expression is called a pH and defined as
[;+I
Water Analysis
Water Analvsis is merely a listing of the concentration of various chemical species present in
the water. The results of the water analysis are expressed in several ways. Some of the more
common ways of reporting water analysis include: parts per million (ppm); milligram per liter
(mg/L), milliequivalent per liter (meq/L), and ppm as CaC03. These at-e defined as follows:
Parts -Der million (ppm) means one part in a million part, for example, 1 gram in a million
grams of water, or 1 pound in a million pounds of water.
Millipram per liter (mg/L) expresses the weight of an ion or radical in 1 liter of water. For
fairly fresh water, 1 mg/L is approximately equal to 1 ppm. In water analysis, the units m g L and
ppm are often used interchangeably.
131
Millieauivalents Der liter (meq/L) - Some laboratories report water analysis results in
milliequivalent per liter:
meqL of an ion = mg/L + equivalent weight of the ion
PPM as CaC03
Steamflood water analysis reports are frequently expressed in this unit. When the
concentration of a species is expressed in ppm as CaCO3, it means that the concentration of the ion
(or radical) is given as though it were calcium carbonate (CaC03). For example, ppm Mg++
expressed as CaCO3 is given by
ppm Mg++ as CaCO3 = ppm Mg++ X
132
APPENDIX 5-B
OILFIELD STEAM GENERATOR WATER QUALITY REQUIREMENTS
INTRODUCTION
Oilfield steam generator water quality requirements are much more stringent than those
required for the conventional power boilers. Poor quality feedwater can result in steam generator
tube failure. The following discussion is directed toward how specific impurities can affect steam
generator performance.
Total Hardness16-l 7
Hardness is a measure of the amount of calcium and magnesium salts contained in water.
The salts of calcium and magnesium are the most common source of scale in steam generator tubes
because the solubility of many calcium and magnesium salts in water decreases with increases in
temperature. Probably the most common source of scale is a breakdown of calcium bicarbonate
with heat to form calcium carbonate, illustrated as:
calcium bicarbonate + Heat + calcium carbonate + water + carbon dioxide
Other calcium salts that may deposit are calcium sulfate and calcium silicate. The most common
magnesium salts expected to scale are magnesium hydroxide and magnesium silicate.
Since the solubilities of these scale-forming salts are so low at tube wall temperatures, they
tend to precipitate at tube walls and plug the tubes. This results in reduced heat transfer and flow
through the tube causing hot spots to develop. Hot spots result in localized overheating and
subsequent tube rupture. The breakdown of bicarbonates also releases carbon dioxide which can
react with steam condensate to form carbonic acid and cause metal corrosion and tube failure.
The most trouble-free steam generator operation is achieved by maintaining boiler feedwater
hardness to near zero level. This is critical because, at typical steamflood operations temperatures
of 400" to 600" F, the solubility of calcium carbonate in distilled water is from 4 to 7 ppm. At 70%
steam quality, the hardness would be concentrated three and one-third times. At 80% steam
quality, there will be a fivefold concentration, and at 90% a tenfold concentration. Thus,
theoretically the solubility of calcium carbonate in 80% steam ranges between 0.8 and 1.4 ppm,
and at 90% the solubility reduces to- 0.4 to .0.7 ppm. Obviously, the higher the steam quality
desired, the more carefully must feedwater hardness be controlled. Operating a steam generator
with hardness values exceeding these limits for as little as 5 hours can result in tube scaling.
Therefore, it is important to monitor the feedwater hardness level continuously to detect any
hardness leakage. Since the solubility limits given above vary according to the chemical
composition of the feedwater, each feedwater must be analyzed to determine what hardness level is
tolerable for a given steam quality.
133
By far, the most common method of removing hardness for steam generation is through
sodium cation exchange. In the ion exchange method, all of the objectionable calcium and
magnesium ions in the feedwater are replaced by nonobjectionable sodium ions. The sodium salts
are highly soluble in water and contribute to the total dissolved solids (TDS) content of the
feedwater.
C02
carbon
dioxide
H20
water
+ Heat +
20H- + C02
hydroxide
The carbonates and hydroxides will combine with hardness and other divalent ions to form scale
deposits. However, since the ion exchange system removes all hardness, scale formation will not
be a problem.
With pressures above 600 psi, a feedwater containing 2,000 ppm of sodium bicarbonate
could theoretically form nearly 5,000 ppm of sodium hydroxide in the water phase of 80% quality
COf-
H20
steam, and corrosion or caustic embrittlement of tubes would be likely. However, for this to
occur, the following three conditions must exist at the same time.*
134
Suspended Solids
Common suspended solids found in an oilfield water include mud, silts, and corrosive
products. The presence of these materials can cause plugging and fouling of steam generator
tubes, water treating plants and injection wells. In most field operations, filtration is all that is
required to control suspended solid buildup. Suspended solid levels in the feedwater should be
maintained below 5 ppm and preferably below 1 ppm.2
Oxygen
Dissolved oxygen can exist in surface waters as a result of contact with the atmosphere and in
subsurface waters as a result of equipment suction leaks as well as pulling a vacuum on a well
because of falling water levels. Dissolved oxygen can be extremely corrosive in steam generators.
Field experience shows that even a small amount of oxygen (less than 1 ppm) can cause
nonuniform pitting3 Therefore, as for hardness, the dissolved oxygen content of the feedwater
must be reduced to m.
Since improper handling of an initially oxygen-free water source can result in aerated water,
monitors should be utilized to detect oxygen leakage. To minimize the treatment necessary to
remove dissolved oxygen, the-operator should take steps *toeliminate unnecessary oxygen pickup.
Unnecessary oxygen,pickup can be avoided 6y using packers to,seal water wells, gas blanketing
storage tanks as well as water wells,- submerging entrance piping to storage tanks and by
maintaining adequate pressure on the suction side of the pumps.3 Oxygen from feedwater can be
removed either by a deaerator and/or by chemical means.
135
Sulfides
Sulfide i s unacceptable in steam generator feedwater because of the high rate and severe
nature of the corrosion it produces. Sulfides also cause plugging problems upstream of a
generator. The presence of hydrogen sulfide (H2S) in the steam phase could cause corrosion and
perhaps plugging of injection wells because of precipitation. For these reasons, as well as the
hazardous nature of H2S, sulfides should be reduced to nearly zero (less than 0.1 ppm). The
presence of bacteria in many freshwater sources is the cause of H2S in these waters. If the source
of sulfide is indigenous to the water source and another source cannot be located, treatment could
be very expensive and the process quite complex. Hydrogen sulfide is generally reduced with
deaeration, but chemical means such as chlorination have to be used for complete removal. If
chlorination is used, it must be carefully controlled to avoid added corrosion potential and to avoid
oxidizing effects of chlorine on ion exchange resin.
Iron5
The natural iron content-of feedwater is quite low. Its presence is usually indicative of
corrosion. Feedwater picks up iron either from a gathering system as a result of corrosion of steel
piping or can acquire it through contact with formation minerals. Irrespective of its origin, the
presence of iron in feedwater is troublesome and should be kept below 0.05 ppm.
Iron may exist in solution as a ferric (Fe+++) or ferrous (Fe") ion. Depending upon the
form in which it exists, iron in feedwater can cause difficulties in the ion exchange process by
fouling the resin bed or can form scale deposits in the generator tubes. To prevent scale deposits in
steam generator tubes, the iron content of feedwater should be kept below 0.05 ppm. The ion
exchange systems are extremely efficient in loweiing the feedwater iron content to this level. Both
the Fe++ and Fe+++ ions are removed by the resin in a manner analogous to calcium and
magnesium ions. The ferrous ions can be removed from the resin bed by backwash. However,
the ferric ion fouls the resin bed by forming a gelatinous precipitate on or in the resin. This results
in capacity loss and makes the unit difficult to backwash. Special backwash and regeneration
techniques must be employed to restore the exchange capacity of the bed. These techniques are
expensive and increase water treatment costs.2
The ferric ion is formed by the oxidation of ferrous ion. Hence, iron should be kept in the
reduced form (ferrous form) as water is being softened. Oxidation of ferrous ion can be avoided
by removing dissolved oxygen with a reducing agent (such as an excess amount of sodium sulfite)
upstream of the resin bed. This will act to reduce the ferric ion to ferrous ion and, thus, keep the
resin unfouled. The reducing agent should be added to the water before caustic is added for pH
control. After the iron has been removed, it is highly desirable to avoid iron pickup in long
pipelines between a softener and a steam generator. One way of accomplishing this would be to
136
keep residual sodium sulfite in the water to avoid oxygen contamination and iron pickup. Another
method of avoiding iron pickup is to use plastic pipes or internally plastic coated pipes.
oil4
The presence of free oil in feedwater can result in the formation of hard asphaltic scales on
steam generator tube walls. Deposition of scale reduces heat transfer and causes tubes to fail
through formation of localized hot spots on tube walls. Oil is especially troublesome in the fouling
of ion exchange resins. From the viewpoint of the resin bed, the oil content should be reduced to
zero ppm. Coarse media filtration can remove oil up to 50 ppm. Above this limit, pretreatment by
induced clarifkation, air flotation, and diatomaceous earth filtration may be required.
PH5
The pH of feedwater plays a very important role in thermal recovery operations. The
solubility of many common ions varies greatly with the pH of water. The concentration of calcium
ion in water goes through a minimum at a pH of 9 to 10. Above a pH of 10, the concentration of
calcium ion in water increases rapidly due to the formation of soluble calcium hydroxide. The
concentration of magnesium ion in water decreases rapidly with increase in pH and approaches
zero at a pH of around 10.6. The solubility of iron in water also decreases with increase in pH and
approaches zero at a pH of 9. Thus, at a pH of approximately 10, both magnesium and iron ions
would precipitate out. Since calcium, magnesium, and iron ions are the most troublesome
impurities in feedwater, problems resulting from these ions can be minimized by assuring that the
steam generator feedwater is slightly alkaline.
Field experience indicates satisfactory operations can be obtained by maintaining the pH of
feedwater between 9 and 11. This alkaline environment also keeps silica dissolved and reduces the
corrosion of steel. However, higher operation pressure will lower the optimum operating range to
between 10 to 10.5. The pH control is usually done by adding sodium hydroxide (NaOH) into the
water systems at some point between the softener and feedwater injection pump. This is done to
prevent the neutralizing of oxygen scavengers which may have also been added.
Silica6
Silica is present in most water supplies and is a much greater problem in power boilers than
in oilfield steam generation systems. Silica is troublesome in oilfield steam generators because of
its scaling tendency. Silica can-produce scaling, both directly and as a constituent of complex
mineral scales. However, formation of complex scales requires the presence of metal ions such as
iron, calcium, and magnesium. Elimination of these ions will effectively prevent the formation of
complex silicate scales.
137
Control of silica problems in steam generator systems consists primarily of maintaining silica
solubility. Silica solubility is a function of temperature, alkalinity, and concentration of other ions
with which silica forms complex scales. By maintaining the hardness of the feedwater near zero
level and pH around 11, silica can be kept in solution. Satisfactory operations have been
maintained with silica content of as much as 150 ppm. Steam generated by the system must
contain enough liquid to prevent silica from precipitating. Tolerable limits of silica in feedwater
vary from 100 ppm at pressures to 50 psi to 50 ppm at pressures up to 1,500 psi. It is
recommended that the silica concentration in feedwater be maintained around 40 ppm.
Suspended or colloidal silica can be effectively eliminated by a coagulation process. Soluble
silica can be removed by a strong base anion exchange to almost any level. Since silica content of
fresh water is much less than the tolerable limit, no silica removal is required. However, it is
suggested that before any equipment expenditures are made for silica removal, a very critical look
at all operating conditions and water analysis data be made to decide if such expenses are justified.
Biological Growths7-8
Feedwater used for steaming operations are usually rich in total dissolved solid content and
are conducive to growth of organic matter. If left untreated, these organic growths can foul and
plug an ion exchange resin bed.
Some of the common bacteria found in oilfield waters include sulfate-reducing bacteria, iron
bacteria, and slime-forming bacteria.
Sulfate-reducingbacteria are anaerobic bacteria. However, they are quite capable of thriving
in oxygenated systems, provided they find some scale to congregate under. These bacteria reduce
sulfate ions in water to sulfide ions and produce H2S as a byproduct. They cause corrosion, and
the iron sulfide produced as a byproduct of corrosion reaction is an excellent plugging material.
Iron bacteria are aerobic bacteria and grow well even with only trace amounts of oxygen.
Iron bacteria precipitate a sufficient quantity of fenic hydroxide to cause severe plugging problems.
As discussed previously, ferric ions foul ion exchange resins readily and are expensive to remove.
Slime-forming bacteria are aerobic bacteria and produce dense masses of slime on solid
surfaces. They are magnificent pluggers and contribute to corrosion.
One method of controlling aerobic bacteria is to seal and gas blanket water supply wells. The
other and the most common method of controlling bacterial growth i n oilfields is by chemical
means. Chlorine is the most widely used inorganic chemical to kill bacteria. Chlorine, however,
attacks only exposed organisms and will not penetrate slime or scale masses which often hide
sulfite-reducing bacteria. There are other biocides (a chemical which kills other forms of life, in
addition to bacteria) available which will control sulfate-reducing bacteria.
138
Since chlorine is a very strong oxidizing agent, it will oxidize iron and hydrogen sulfide.
Once it reacts, it is no longer available to kill bacteria. Hence, to determine the total amount of
chlorine needed it is necessary to establish how much chlorine will be used by reaction with other
materials. The amount used by the system is called the chlorine demand. The amount of chlorine
in excess of that required to control bacteria and meet chlorine demand is the excess chlorine. The
excess chlorine concentration in the feedwater must be less than 1 ppm because chlorine can
oxidize ion exchange resins and reduce its effectiveness. Excess chlorine concentration in
feedwater can be reduced to less than 1 ppm by chemical deaeration upstream of an ion exchange
unit.
Other biocides should be chosen with the aid of vendor representatives, since the type needed
will depend on contact time, pH, and temperature of the water treated. These biocides should be of
the nonionic type as others will clog filters and softeners.
Turbidity
The turbidity of water being fed into steam generators should be zero. Softening systems
have a limited ability to filter out turbidity. If the turbidity is high, a separate means must be used
for its reduction. A combination of coagulation, flocculation, and sludge removal is generally
required.
Carbon Dioxide9
Although fresh sources of water contain carbon dioxide (COz), its major corrosive effect, as
far as steam generation is concerned, comes from the decomposition of sodium bicarbonate. In
theory, the bicarbonates break down at high temperatures and release carbon dioxide. The C02
then reacts with steam condensate to form carbonic acid and cause of the metal condensate system
to corrode. In a typical oilfield steam generator, the tubes carry both steam and water not flashed
to steam. The water phase is usually highly alkaline due to sodium carbonate and hydroxide in
solution. On the other hand, the steam phase carries free C02 which redissolves in the condensate
forming carbonic acid. If the velocities in the boiler tubes are high and if there is a lot of turbulence
in the flow stream, then the two phases are intimately mixed, the acid is neutralized by the
alkalinity, and minimal corrosion occurs. However, if the velocity is low, the vapor can condense
without intimate contact with the alkaline water and cause corrosion.
There is much disagreement among operators as to which situation actually occurs. This led
to two different approaches to handle the C02 problem. One group feels that treatment is not
necessary because C02 will eventually be neutralized by the alkaline water. This group, however,
carefully excludes oxygen (which can accelerate C02 corrosion) from the system and closely
monitors corrosion rates. The other group feels that treatment for C02 corrosion is necessary.
One method of treatment is to use sodium hydroxide or volatile film ng amines (such as
139
140
APPENDIX 5-C
FEEDWATER TREATMENT EOUIPMENT AND CHEMICAL VENDORS
A.
B.
3. Saltech Corporation
1035 Diesel Drive
P.O. Box 26872
El Paso, TX 79907
Telephone: (915) 592-7982
4. Permutit Company
30 Technology Drive
Warren, NJ 07059
Telephone: (908) 668- 1700
1-800-631-0878
5. Graver Water
2720 U.S. Highway No. 22
Union, NJ 07083
Telephone: (908) 964-2400
Rockford, IL 6 1105
Telephone: (815) 877-3041
Columbia, MO 65202
Telephone: (314) 449-00 11
1/800-325-0104
8 . NATCO
Division of National Tank Company
Bakersfield, CA 93308
Telephone: (805) 325-4146
141
C.
3. Saltech Corporation
1035 Diesel Drive
El Paso, TX 79907
Telephone: (9 15) 592-7982
4. WEMCO
1796 Tribute Road
P.O. Box 15619
Sacramento, CA 95852
Telephone: (916) 929-9363
2. Petrolite Corporation
OXield Chemical Group
16010 Barker's Point Lane
Suite 600
Houston, TX 77079
Telephone: (7 13) 558-5200
5. Graver Water
2720 U.S. Highway No. 22
Union, NJ 07083
Telephone: (908) 668-1700
7. Permutit Company
30 Technology Drive
Warren, NJ 07059-0920
Telephone: (908) 668- 1700
1-800-631-0878
F.
2. Celite Corporation
P.O. Box 519
Lompoc, CA 93438
Telephone: 1-800-654-3103
3, Liquid-Solid Corporation
185 Le Grand Avenue
P.O. Box 9
North Vale, NJ 07647
Telephone: (201) 784-1570
Q
Pg
G.
1 . Saltech Corporation
1035 Diesel Drive
El Paso, TX 79907
Telephone: (915) 592-7982
2. Permutitcompany
.
30 Technology Drive
Warren,NJ 07059
Telephone: (908) 668- 1700
1-800-631-0878
2. ACS Environmental
303 Silver Spring Road
Conroe, TX 77303
Telephone: (409) 856-4515
3. AFL Industries Inc.
3661-B West Blue Heron Blvd.
Riviera Beach, FL 33404
Telephone: (407) 844-5200
3. Graver Water
2720 U S . Highway No. 22
Union, NJ 07083
Telephone: (908) 964-2400
J.
1. WEMCO
P.O. Box 15619
1796 Tribute Road
Sacramento, CA 95852
Telephone: (916) 929-9363
H . Clarifier. Deaerators
(Mechanical, Vacuum, Steam)
1. Permutit Company
30 Technology Drive
Warren, NJ 07059
Telephone: (908) 668- 1700
1-800-631-0878
2. Graver Water
2720 U.S. Highway No. 22
Union, NJ 07083
Telephone: (908) 964-2400
P.O.2792
Odessa, TX 79760
Telephone: (915) 337-3571
4. Permutit Company
30-T Technology Drive
Warren,NJ 07059
Telephone: (908) 668-1700
1-800-631-0878
K . Water Treatment Plant Contro1s
(Multi Port Valves, Hardness
Moni tors,Oxygen Monitor, etc.)
1. Saltech Corporation
1035 Diesel Drive
El Paso, TX 79907
Telephone: (915) 592-7982
Cells
011Wate
- r Separator. Skimmers,
Hydrocyloncs, etc,
1. Modulur Production Equipment Inc.
P.O. Box 690965
Houston, TX 77269-0965
Telephone: (7 13) 320-8796
143
M.
Renfro Road
Bakersfield, CA 933 12
Telephone: (805) 589-2597
144
CHAPTER 6
STEAM GENERATION
INTRODUCTION
Steam for thermal recovery is usually generated in direct-fired, forced-circulation, oncethrough generators. Since a typical steam injection operation requires large quantities of steam for
continuous injection, the cost of treating the feedwater for steam genemtion must be kept low. The
once-through design permits the use of low quality feedwater that is essentially free of hardness
and suspended solids, but of relatively high level of total dissolved solids (TDS). Oilfield steam
generators are usually rated in millions of BTUs per hour of heat output. Some manufacturers also
rate them in terms of pounds per hour of steam output, or in terms of boiler horsepower.
Typical oilfield steam generators range in size from 10 to 180 million BTU/hr (MM BTU/hr)
heat output. The smaller size generators are generally used in pilot projects and in steam soak
applications, whereas the larger units are used in continuous steam injection operations. In fact,
the 50 MM BTU/hr unit has become the industry standard for steamflood applications. While all
oilfield steam generators are offered as skid-mounted units. smaller units (up to 40 MM BTU/hr)
are also offered as fully self-contained trailer-mounted units. The size of a fully-assembled skidmounted 50 MM BTU/hr unit is about 11 ft by 80 f t and its weight ranges from 170,000 to
230,000 p ~ u n d s . ~ *The
~ g 50 MM BTU/hr unit is the largest skid-mounted unit that can be
transported easily by rail to an oil lease. A list of steam generator vendors is included in Appendix
6-F.
General Features of Oil Field Steam Generators
The oilfield steam generator, also known as the wet steam generator, thermal recovery heater,
oil field heater, etc., differs from conventional power boilers, in that is specifically designed to
produce low quality steam from saline feedwater with minimum treatment. Oilfield steam
generators can handle zero hardness feedwater containing up to 6.000 ppm TDS an3 generate up to
80% quality steam. The cap on the quality is set to prevent the precipitation and deposition of
dissolved solids on the boiler tubes.
Basically, in a generator of this type, water enters one end of a heated tube and leaves the
other end as wet steam. This type of construction permits a great deal of flexibility in operation,
allows quick response to load changes, and does not require constant supervision. Further, since
the water storage capacity of the generator is small. only a small amount of steam-water mixture is
released if the heating coil fails. Thus, these units are safe to operate under continuous operating
conditions. However, in order to respond to quickly to changing load conditions, it is important to
control precisely the rate of flow of water through the tube, the fuel injection rate, and the air flow.
Figure 6.1 is an isometric view of a skid-mounted oilfield steam generator showing the major
145
CONVECTION SECTION
RADIANT SECTION
FEED WATER
HEAT EXCHANGER
components on the supporting skid.1 The generutor is fully self-contained and equipped with
necessary controls and instrumentation to monitor operating pressures, flows, temperatures, etc.
The standard oilfield steam generator flow chart is shown in figure 6.2.
A self-contained unit is ready for operation, and the only field work required to prepare a
generator for operation after transportation consists of connecting unit's discharge pipe to the well,
connecting feedwater to pump, connecting electrical power, and connecting fuel supply to the
generator.
146
WATER1
4
SEPARATOR
FEEDWATER
FUEL
Generator Selection
Generator capacity must be carefully chosen for each application to avoid idle steam capacity.
Oilfield steam generators are available with pressure ratings of 1,000, 1,500 and 2,500 psig, based
on coil pressure capability. They come in a range of standard sizes from 10 to 150 MM BTU/hr
heat output (approximately equivalent to 650 to 9,864 bbl of water (as steam) per day) and are
designed to burn both gas and oil.
Steam rates in the United States are normally reported in barrels of steam per day (BSPD)
regardless of steam pressure or quality. This is also the cold water equivalent (CWE) rate. The
CWE rate is much smaller than the actual volumetric rate for wet steam and is equivalent to the
mass flow rate. The term BSPD originates from the early day steam injection practice when the
generator feedwater meter indicator was calibrated in barrels and it was the most convenient place
to measure the amount of steam injection.
Table 6.1 is a generator selection chart provided by one manufacture.2 Table 6.2 presents the
capacities and fuel requirements for these units. The electrical requirements for selected units are
depicted in table 6.3. The capacity and fuel requirements in these tables are based on feedwater
temperature of 100 F and 80% quality steam at 1,000 psig. The approximate steam generator
147
per hour
boiler, hp.
314
553
657
747
896
1.195
1,494
2,100
3.000
4,500
= 100'
steam Pressure = 1,OOO psig.
Generator Efficiency = 90%.
a Feedwater Temperature
I'ound\
per hour
BTU x IO6
unit
Barrels
per day
lhrrcl\
pcr hour
10
9.600
27.4
657
18
17.280
40.3
1.I 83
22
21.120
60.3
1,447
25
24.000
68.5
1.644
30
28,800
82.2
1.972
40
38.400
109.6
2,630
50
48.000
137.0
3.288
70
67.200
10 1.8
4.603
100.
96.000
274.0
6,576
150
144.000
61 1 .O
9.864
F.
boiler. ho
Fuel o
Ba+: 130
Gallons
ner hour
G a!
) .C
D BTLIIGAL
Barrels
&is:
MSCFhour
1.M
3TUISCF
MMSCFIday
ner day
79
44.8
11.36
0.2727
314
141
80.6
20.45
0.4909
553
172
98.5
25.00
0.66
657
196
112.0
28.41
0.6818
747
235
134.4
34.10
0.8182
896
314
179.1
45.45
1.09
1.195
392
223.9
56.82
1.3636
1,494
549
313.5
79.55
1.9091
2.100
784
447.8
1 13.64
2.7273
3,000
1,176
67 1.8
170.46
4.0909
4,500
314
657
1.195
896
1.494
1SO0
2.500
1.Ooo
1,500
2.500
41
47
60
19
25
36
60
72
95
34
50
80
89
100
137
65
76
114
116
134
172
110
146
160
179
216
1 54
191
91
134
output in pounds of water evaporated per hour for other operating pressures, steam quality, and
feedwater temperatures is shown in figure 6.3. The approximate fuel cost for operating various
capacity generators by burning 10' to 20" API California crudes is shown in figure 6.4. This
figure is based on Table 6.2 and can be used to estimate approximate fuel cost for the generation of
steam. In figure 6.5, the cost of steam generation by burning natural gas (1,000 Btdscf) is
depicted.
26
30
62
66
1200
1100
I;
000
I-
2 900
800
W
700
600
2400
2000
1600
1200
800
OPERATING PRESSURE-PSIG
400
FIGURE 6.3. - Approximate steam generator output (lb/hr/bhp) at various pressures, steam
quality and feedwater temperature.
149
14000
12000
tt)
$1 0000
8000
-I
UI
6000
4000
18 mm BTUlHR
>
n
2000
0
20
10
30
40
FIGURE 6.4. - Daily fuel (lease crude) cost for operating vaiious capactiy generators.
12000
70 mrn BTUlHR
0
I--
v)
10000
40 mm BTU/HR
00
2a
30 mm BTUlHR
8000
25 mm BTUlHR
6000
18 mm BTU/HR
4000
2000
0
0
FIGURE 6.5. - Daily fuel (natural gas) costs for operating various capacity steam generators.
150
Figure 6.3 assumes steam generator efficiency as 90%. The following procedure is
employed to graphically determine the steam generator capacity from Fig. 6.3. To determine the
capacity of a 50 MM Btu/hr generator delivering 80% quality steam at 1,500 psig (feedwater
temperature assumed to be 100" F), enters the chart at 1,500 psi and proceeds vertically to 80%
quality line. Then proceeds horizontally to locate the 100" F feedwater temperature line and then
goes vertically to read off the evaporation capacity as 33 Ib/hr/bhp. From Table 6.1, the boiler
horsepower corresponding to a 50 MM Btu/hr generator is 1,494 and hence generator output in
lbdhr is approximately 49,300 lb (= 33 x 1,494) per hour.
Since the steam requirements for thermal operation are not uniform and vary with time, it is
important to size steam generators properly, especially for small projects. It is important to bear in
mind that in any operation there will be downtime for maintenance, and it is unlikely that all
injectors will be in service at any given time. Hence, to avoid idle steam capacity, selection should
be based on maximum expected steam injection rate per injector and the number of injectors
expected to be in service at any given time. A minimal size generator that fulfills this requirement
with some spare capacity to meet unexpected demands should be specified. Selection based on the
steam requirements to start an entire project at one time should be avoided. Further, in small
projects where steam requirements are small, consideration should always be given to having two
or more smaller units instead of one large unit, This permits more flexibility in operation as steam
demands change by bringing in or removing from service one or more units. This strategy also
ensures that at least some steam is available all the time, in the event of failure of one or more units.
In steam stimulation projects, excess steam capacity can be avoided by steaming the wells in
rotation. For example, consider a steam soak operation involving 20 wells to be steamed at the rate
of 3,200 barrels of equivalent steam per cycle, two cycles per year. Also, assume that the duration
of each injection cycle is 3 weeks. If all of the wells are to be steamed simultaneously, this
operation will need about 3,050 barrels of steam per day. From table 6.1, to satisfy this
requirement, the operator must specify a 50 MM Btu/hr unit. On the other hand, if only 5 wells are
to be steamed at a time, the maximum steam requirement reduces to 763 barrels of steam per day,
and,smaller 18 MM Btu/hr unit would be sufficient to meet f i e peak demand with ample spare .
capacity. Such a generator is also small enough to be trailer-mounted and can be moved from well
to well; thus, minimizing heat losses in distribution lines. Another advantage .of adopting a
staggering steaming policy in steam soak operations is that the production decline from the wells
steamed first is offset by the production rise from wells steamed later.
In large steamflood operations, due to continuous steam injection requirements, steam
demand is large and several steam generators are employed. Since steam requirements decrease
with time as injected steam reaches producers, steam demand fluctuates, and the operator must bear
this in mind while sizing steam generating capacity. Choice of steam generator size should be
151
based primarily on the overall expected oil-steam ratio.3 As noted previously, generators should
not be sized based on total project steam requirement, but should be based on the desired steam
injection rate per injector and the number of injectors to be used at any one time. For this reason, it
is preferable in steamflood operations to undersize steam generation capacity relative to total project
requirement. Also, it is preferable to have a few standby portable units to take care of any
unexpected surge in demand or to provide steam to peripheral wells that are too far away from the
main unit.
152
predicted flue gas dewpoint temperature, thus eliminating the condensation of acid gas
components of the flue gas.
3. Fuel System
Modem oilfield steam generators are designed to burn different fuels including natural
gas, fuel oil, lease crude, compressed natural gas (CNG), etc. The lease crude (fuel oil)
fuel system normally includes fuel oil strainer, steam-fuel oil heat exchanger, electric oil
heater, fuel oil temperature controller, pressure regulators. bypass system, air and steam
atomizing systems, burner assembly, and safety shut-off valves.
153
FLUE GAS TO
ATMOSPHERE
REFRACTORY
BAFFLES TO
MINIMIZE
FLUE GAS
BY PASSING
REFRACTORY
LINING
MOWS
FINNED
TUBES
STEEL
CASllNG
I
FLUE GAS FROM
RADIANT SECTION
3-ROWS OF BARE
SHOCK TUBES
FIGURE 6.6. - Tubing arrangements in the steam generator flue gas convection section.2
6.Radiant Section
The radiation section is comprised of a cylindrically shaped combustion chamber housing the
radiant section tubes, tube hanger support systems and the burner assembly. The dimensions of a
50 M M B T U h unit radiant section as provided by one manufacturer is 11 ft in diameter and 40 ft
long. The tubes within the radiant section are 3 in. in diameter and are held in place by primary
hangers in the endplate and by three or four rows of secondary hangers attached to the curved wall.
Depending on the outlet pressure ratings, schedule 40,80, or 160 pipes are used. These tubes are
arranged in rows and travel the length of the generator. In most cases, they penetrate the two ends
of the radiant section. The burner is located in the center of one flat end. On the other end, the
transition section connects the radiant section to the convection section. The radiant section shell
wall is lined with light-weight refractory bricks to protect it from the high-temperature corrosive
154
combustion gases. The type of refractory used depends upon the nature of fuel burned and the
environment.
155
FIGURE 6.7. - Layout of ii typical oilfield steam generator burning lease crude.l9
The preheated feedwater enters the top of the convection section and flows downward and
countercurrently to the hot flue gases from the radiant section. The 1,600 F hot flue gases from
the radiant section enter the convection section at the bottom and contact the lower, most-bare tubes
of the convection section. These bare tubes are chosen to withstand the hot flue gas temperatures
and are located 4.5 in. apart. The flue gas, after losing part of its heat to the bare tube, then
contacts a series of finned tubes at the upper section at a reduced temperature. The flue gas then
exits the top of the convection section at a temperature of 300" to 600" F.
The finned tubes, which increase the heat transfer rate between the feedwater and flue gas,
cannot be used in the bottom section because of increased corrosion at the high temperature. The
transition between finned tubes and bare tubes is carefully chosen to ensure that the maximum
permitted fin tip temperatures are not exceeded. Since burning of lease crude results in an
accumulation of soot in the convection section heat transfer suifaces, operation of the generator at
156
optimum performance will require regular cleaning of the convection section tube surfaces. The
cleaning cycle varies from 3 weeks to 3 months, depending on the fuel oil.
After supplying the heat requirements of the feedwater preheater, and usually the heat needed
to preheat the fuel oil, the effluent from the convection section enters the radiant section of the
steam generator. The tubes within the radiant section are heated by direct radiation from the
combustion gases and radiation from the refractory lining behind the tubes. The heat transfer
process at work in the radiant section is not simple. To guarantee that the correct heat flux is
delivered to the tubes, manufacturers use sophisticated computer programs to calculate tube
spacing and tube placement from the refractory wall. Heat flux rates of 20,000 BTU/hr-ft* are
typical in oilfield steam generators. Failure to model correctly the heat flux rates can result in poor
generator efficiency and premature tube failures.
Design and construction practices, which are common for oilfield steam generators, include
closely spaced tubes to minimize heat absorbed by the refractory, and reducing tube size to
minimize the heat release in the event of a tube failure. The feedwater leaves the radiant section as
80% quality steam.
Since some formations are susceptible to swelling, it is not possible to inject wet steam into
such reservoirs. For these, the effluent steam and water from the generator must be separated and
only dry steam injected. The separated water can be utilized to preheat feedwater.
Fuel System
Although only the fuel oil system is shown in figure 6.7, oilfield steam generators are
designed to bum different fuels, including lease crude, natural gas, liquefied petroleum gas (LPG),
etc. To provide efficient combustion of fuel oil, they must be pumped into the burner, divided into
small droplets and intimately mixed with air before burning. The process of dividing the fuel oil
into small droplets is called atomization.
Steam or compressed air is generally used as the atomizing agent. The atomization takes
place in the burner nozzle. The atomization is accomplished by forcing the crude through the fuel
oil nozzle at sufficient pressure to guarantee a well-dispersed pattern. An atomizing air compressor
and/or an atomizing steam system is used to atomize the fuel. To assure the formation of proper
droplet size and spray pattern shape at the nozzle exit, the oil must.enter the nozzle at correct
viscosity. Most nozzles are designed to function correctly at an oil viscosity of 100 to 150 Saybolt
Second Unit (SSU).
A typical 14' API Kern River, CA, crude must be preheated to about 250" F to reach the
proper nozzle viscosity. Hence, oilfields are equipped with il fuel oil preheater to preheat the fuel
oil. Electricity, hot water, or steam is used to preheat fuel oil. I n figure 6.7, both electricity and
steam are used to preheat fuel oil. In many leases, produced oil is hot enough to fire a generator.
157
Close control of the temperature of the oil pumped to the burner is critical. Close temperature
control is needed to provide oil at constant pressure and viscosity. If the oil were to overheat and
become less viscous, the burner could overfire. If the oil were too cool and become more viscous,
it would be difficult to handle and impossible to atomize and burn. The oil pump and piping
between oil storage tank and steam generator must also be insulated to maintain correct oil
temperature. For proper nozzle operation, the fuel oil must be delivered to the nozzle at the correct
pressure. For a 50 MM BTU/hr generator, the fuel oil must be delivered to the nozzle at 100 psi at
the rate of about 6.5 gpm.
The basic components of a generator fuel system burning lease crude include fuel pressuring
pump, fuel oil heater, atomizing equipment, and heated fuel storage. A fuel oil system supplied by
the generator manufacturer usually includes a bypass system to bypass any fuel in excess of that
required by the nozzle. Gear pumps are often used for fuel oil service. However, the lubricity of
the fuel oifshould be considered before specifying the fuel pumping equipment.
When natural gas is used as the fuel in an oilfield steam generator, a separate fuel system is
used. Instead of a fuel nozzle, a pipe ring is used at the transition from the burner throat to the
radiant section. Natural gas is supplied to this ring at approximately 10 inches of water column.
The natural gas is released into the combustion chamber through small nozzles (holes) in the ring,
where they mix with combustion air and burn. The flame pattern of the gas flame is more stable
than the oil flame and hence are more easily controllable. Typical gas fuel requirements for a 50
MM BTU/hr generator are about 1,000 standard cubic feet per minute (SCFM) of 1,000 BtdSCF
gas.
Cornbustion A ir System
The air needed for the combustion of the fuel is provided by a high-pressure air blower. A
typical 50 MM BTU/hr generator burning 6.5 gpm of crude will require approximately 17 SCFM
or 52,280 lb/hr of air. This air is not filtered unless the atmosphere is dust laden. Filtering will
entail additional blower horsepower.
The combustion air system includes a forced-draft type air blower and associated controls to
regulate the discharge of the blower. The air blower system must be designed for local operating
conditions. The BTU output of a steam generator at sea level is higher than that of higher altitudes.
Also, if the generator is installed inside an ill-ventilated building (to protect it from harsh winter),
the blower is likely to suck hot air from inside the building, resulting in efficiency degradation.
Hence, the blower must be located in a well-ventilated area.
The blower delivers air to the radiant section of the generator through the burner nozzle
throat. The air and the atomized fuel are turbulently mixed at the transition from the throat to the
radiant section, and combustion is initiated.
158
The rate of air delivery to the combustion process is determined by the shutters placed at the
blower discharge. The amount of excess oxygen in the flue gas, as it leaves the radiant section,
controls the position of these shutters. Excess oxygen concentration for normal operations is
between 1.5% and 3%. Excess oxygen outside this range causes the shutter controller to take
appropriate action.
To assure complete combustion, steam generation equipment has to be operated with about
3% excess air. Any excess air above this amount will lower the steam generator efficiency,
because the additional excess air robs the heat from the combustion process that otherwise could be
used to generate steam. Otherwise, the use of less than 1.5% excess air will probably cause
incomplete combustion.
As part of the efficient operation of the unit, the stack gases released by the generator should
be monitored to assure that complete combustion is occurring and no undue air is being used. The
quantitative determination of total air (100% + excess air) admitted to an actual combustion process
requires a complete flue gas analysis for C02, 0 2 , CO and N2. However, N2 is usually
determined by difference. No oxygen in the flue gas indicates incomplete combustion with
resultant fuel wastage. Too much excess oxygen indicates needlessly high stack losses. By
performing these analyses and adjusting generator operation when necessary, the operator can save
fuel.
Steam Generutor S p e c i f i c ~ t i o n s ~
Since steam injection operations are site specific, oilfield steam generators must be custom
designed to meet the requirements of a specific site. Operators must provide certain minimum
information to steam generator vendors, to assure that minimum design, safety and environment
considerations are met. The American Petroleum Institute (APJ) recommends that steam injection
operators include the following minimum information in steam generator specification sheets:
(a) Fuel oil analysis (see appendix 6-A for fuel oil analysis specification sheet)
(b) Fuel gas analysis (see appendix 6-B for fuel gas analysis specification sheet)
(c) Other fuels analysis (if any)
(d) Feedwater analysis (see appendix 6-C for feedwater analysis specification sheet)
(e) Utilities specification
(i) Electric power: -V -Ph -HZ
(e.g.,. 440 V, 3 Eh, 60 Hz)
F dewpoint
(ii) Compressed air: -SCFM __ Psig
(f)
159
-Max
Min -Design
160
0')
161
Natural Gus
If natural gas is readily available, it is the most convenient fuel because it can simply be
metered to the burner at a pressure that assures the desired volume feed rate. Other advantages of
gas are cleanliness of operation, relative simplicity of heat input control and fewer operating
problems. However, if the gas is allowed to accumulate in faulty valves or fittings which are not
readily detectable, explosions can occur. The explosion hazard can be minimized with well-trained
operators. Some of the problems associated with the use of natural gas as generator fuel are:
water in gas supply lines, inadequate gas pressure, and trow rate as well as excessive pressure
drop under maximum demand operations. Typical gas consumption rates for different sizes of
steam generators are shown in table 6.2,
Liquefied Petroleum Gus (LPG)
Light hydrocarbons such as propane and butane are distributed and stored under pressure as
liquefied gas. When the pressure is released, the liquid boils producing gaseous fuels. Use of
LPG has most of the advantages of natural gas, but does require storage capacity and vaporizers to
convert LPG into gaseous form for combustion. Liquefied petroleum gases are more expensive
than natural gas. Typical consumption rates for propane range from about 1,500 gallons per day
(2,520 BTU/SCF) for 5 MM BTU/hr heat release rate to 15,000 gallons per day for 50 MM
BTU/hr units.
Lease Crude
Lease crude is the most commonly used steam generator fuel i n the oil patch. However, the
operating difficulties and hence the costs associated with burning lease crude are significantly
higher than those associated with burning natural gas. Lease crudes contain water which is
corrosive and susceptible to sludge formation upon storage. Additives must be added to the crude
to minimize these problems.
The following requirements must be met before crude oil can be used as generator fuel.5
1. Oil properties such as the API gravity at 60" F, viscosity from 100" to 300" F, percentage
of basic sediment and water (BS & W), and tlash point must be known to successfully
control burning. The flash point is defined as the temperature at which the oil begins to
vaporize. These vapors will tlash when ignited.
2. The burner control system must be evaluated to assure that i t is compatible with the oil
pump.
3. The atomizing medium must be able to vapoiize the oil.
162
... .
163
Tube failure can also occur as' a result of improper burner operation. In oil-fired generators
burning lease crude, the fuel oil must be preheated to the right temperature before burning. Too
high or too low a preheating temperature can result in erractic burner operation and poor flame
pattern. Hampton6 has pointed out that the following problems can result from too high or too low
a preheating temperature.
1 . Poor atomization will result in poor combustion.
2. Too high a preheat temperature will result in preignition and erratical firing of burner.
3. Too low a preheat temperature makes ignition of fuel almost impossible, especially on a
cold start.
4. Improper preheating of fuel oil will result in soot and carbon formation on burner
throats and in the combustion area.
5 . Improper preheating also will result in oil pump cavitation.
6 . Fuel oil pump cavitation will result in reduced fuel input to burners and poor burner
pressure regulation.
I
Modern oilfield steam generators are highly instrumented to ensure safe operations and to
shut down if an unsafe condition exists. Because of this high degree of automation, steam
generators experience frequent shutdowns and reactivation. Such frequent shutdowns and restacts
result in cyclic thermal stresses on the refractory material and tubes.' As a result, the refractory
lining of the inner surface of a generator will decay and fail. The loss of the insulating properties
of the refractory will permit the coil-osive combustion gases to come in contact with the outer shell
of a generator and can result in skin failure due to corrosion. Further, the expansion and
contraction of radiant section tubes will cause tube sealing packings in the tube hanger to fail. In
addition, the tubes within the radiant section of a generator will suffer material loss from their outer
surface because of the erosive action of the hot combustion gases. Tubes within the convection
section will deteriorate with time due to aging and coil-osive attack by the products of combustion.
Hence, steam generators must be rebuilt periodically.
California field experience has shown that on an average a 50 MM BTU/hr steam generator
burning lease crude must be rebuilt once every 7 years. In general, the first rebuilt requires the
replacement of all the refractory and a portion of the tubes. The second rebuilt requires the
replacement of refractory and all tubes.
Other mechanical components of the steam generation system will also fail, mostly as a result
of poor maintenance. The rotating parts of the blower will fail prematurely, if it is not aligned
properly. Accumulation of soots on generator tubes will cause blower motors to overload and
eventually fail. Presence of excessive water in the lease crude will result in burner failure. The
I64
fuel oil nozzle is another troublesome area requiring frequent cleaning. The seals of the fuel and
feedwater pumps also fail frequently and require repair. Field experience will better define other
areas of frequent repair. Most of the steam generation system problems, however, can be kept to a
minimum with good and well thought out maintenance.
(ix) Since conditions such as too cold a fuel oil temperature, too wide a flame angle, too
much excess air, or too much fuel supply will all result in excessive flame
impingement on tubes and rapid coke formation, they should be avoided.
In addition, the operator must monitor and record daily those parameters needed for
calculations and eftlcient operation of the steamflood. These include: water quality, steam quality,
lubricating oil levels of the feedwater pump and air compressors, and filter and strainer conditions.
The API recommended steam generator daily log is shown in table 6.4. Also, since the single
most expensive activity in steamflood operation is the production of steam by burning fuel, a daily
record of fuel use should be part of every steamflood operations. This will be discussed in some
detail in a subsequent chapter on tools for monitoring steamflood performance.
Water
Inlet eedwater
temp. flow rate
Date
or
time
Feedwater
pump disch.
pressure
Steam
outlet press.
Steam
Stack
outlet ~ e m p (emp.
Radiant
tube
sk~ntemp.
Fuel
nozzle
press.
Radiant
Section
pressure
Fuel
rate
Excess
02
(v) The condition of the burner tip, burner throat, fuel nozzle. flame diffuser, and air
blower vanes should be inspected and cleaned as required.
166
(vi) It is also recommended that the convection section heat transfer surface (inside and
outside) be cleaned regularly to reduce stack temperature and improve generator
efficiency. Soots should be removed from the outside surface and any scale buildup
inside the tubes must be removed by circulating a weak solution of hydrochloric acid
through the tubes at room temperature.
The API recommended monthly steam generator inspection checklist is shown in table 6.5.
Annuul Inspection
In addition to the monthly inspection, all components of steam generation systems (water,
fuel, instrument, air, safety devices, etc.) should be thoroughly inspected once a year and
permanent records should be kept. External inspection should include calibration and replacement
of pressure and temperature instruments; inspection of safety relief valves for the accumulation of
rust, dirt, or foreign matter; and inspection of electrical equipment, damaged insulation, broken
wires, and corrosion. Internal inspections should include inspection of all tube surfaces for
erosion, corrosion, deformation, bulging, cracks and sagging, and inspection of tube hangers,
yokes and hanger bolts for cracks and stress deformation and inspection of refractory material for
cracks. The API recommended annual steam generator inspection list is included in appendix 6-E.
The above recommended practices should be considered only as the starting point for a good
and thorough maintenance schedule. Assistance of the manufacturer of steam generation
equipment must be enlisted in developing a good flexible and optimum maintenance schedule,
Operutional P r o b l e m Because of Poor Muintenunce
Failure to follow manufacturers recommended maintenance schedules may cause several
operational problems and result in costly downtime. Some of these problems are detailed in the
following paragraphs.
A trouble-free steam injection operation begins with the use of good quality feedwater for
steam generation. Ideally, the feedwater must have less than 1 ppm hardness, less than 0.05 ppm
dissolved oxygen, a pH value between 9 and 1 1 , and less than 4.000 ppm TDS. Failure to use a
good quality feedwater will result in formation of scales inside the tubes and cause tubes to
corrode. Scale formed inside the tubes can cause a variety of problems in addition to lowering the
overall heat transfer rate and formation of hot spots. Feedwater should be free of oil. Oil
contamination of the feedwater may lead to asphaltene deposition inside the tubes. Oil and other
suspended matter are removed from the feedwater by flowing it through a mechanical filtering
system. Such filters should be backwashed every 24 hours to minimize cake buildup and prolong
their life.
167
TABLE 6.5. - API Recommended Steam Generator Monthly Maintenance and Inspection Check
List4
'
Inspected
OK
Description of item
Faulty
Replaced
In spec tor
Commenl
Feedwater pump
crankcase-drain and
refill as specified by
manufacturer.
Feedwater pump motor'
lubricate as specified by
manufacturer.
Convection coil2
Radiant coil
Fuel oil filters and
strainers3
Water filters
Air filters
Burner blower motor'
Miscellaneous motors
Burner stabilizer/diffuser
plate
1 Keep all motors clean and ventilation openings clear of dust, dirt and other debris. Do not over grease. WARNING:
Disconnect all power sources to the unit and discharge all parts which may retain an electrical charge before attempting
any maintenance or repair. Screens and covers must be maintained i n place when unit is in operation.
Some small motors have sealed-for-life type bearings which require n o relubrication.
Motors that do require lubrication, can be regreased by stopping the motor. rcmovlng thc drain plug and pumping new
grease into fillhole. Run the motor with the drain plug removed, for a short period, to discharge excess grease. Replace
the drain plug.
Motors that operate at speeds greater than 1,800 RPM should be lubricated on a more frequent maintenance schedule
depending on duty cycle.
2. On occasion it may be necessary to remove deposits from between the fins on the tubes in the convection section. The
frequency of cleaning the fin tubes will be determined by the type of fuel oil being used.
The convection section fin tubes may require cleaning when increase in backpressure nf approximately 2 in w.c., above
new and clean condition, is indicated on the radiant section manometer.
3. The frequency of service required for the filters and strainers is determined by operating time using fuel oil and the
quality of the fuel oil being used. One indication of a dirty element is a drop in oil pressure to the burner (as indicated on
pressure gauge).
Every effort should be made to burn only clean fuel in the generator. If field gas is used as
fuel, it should be oil-free. Oily gas will cause a gas pilot solenoid valve to become sticky and
result in solenoid pilot valve failure.
168
..
When firing lease crude, the oil should be filtered to remove sand, sludge, and other solid
material. Dirty fuel causes nozzle plugging and contributes to erratic flame pattern and eventual
tube failure. Presence of sand in the fuel will lead to excessive wear of the pump, pressure
regulators, and nozzle. Heavy sludges give rise to erratic burning and eventual flame failure.
Sludge problems can be alleviated by adding additives to the crude. Lease crude should also be
free of excessive free water. Up to 3% water, as long as it is dispersed in the oil, is not detrimental
to combustion. Excessive water can lead to burner failure and refractory deterioration.
Thus, good maintenance practices should be an integral part of a steam injection operation not
only to lower operating costs and increase thermal efficiency, but also to ensure the safety of field
personnel.
169
Since the flue gases heat content accounts for the major portion of generator heat losses,
steam generator efficiency can be improved by reducing the stack heat losses. The amount of heat
lost via flue gases is proportional to the temperature of the flue gases and to the mass flow rate of
the vented gases. To the extent possible, flue gas heat is recovered in the convection section and
indirectly in the feedwater preheater.
Since steam generators are mostly fired with sulfur bearing fuels, such as the California
crude oils, the flue gases contain corrosive sulfur oxides. The amount of heat recoverable from the
flue gases is limited by the dewpoint temperature of these corrosive acid gases. If cooled below
the dewpoint, these gases condense and form highly corrosive sulfuric acids. Thus, the exhaust
gases cannot be cooled below the acid dewpoint which is about 400' F under typical operating
conditions. Therefore, steam generator thermal efficiencies are limited by stack gas dewpoint
temperatures.
The approximate flue gas dewpoints for heavy oils are shown in figure 6.8 as a function of
percent sulfur in the fuel oil and percent excess air used for combustion. As can be seen for a
given amount of sulfur in the fuel, the stack gas dewpoint increases with percent of excess air.
Further, the greater the excess air used in the combustion process, the lower will be the thermal
efficiency because the additional air traveling through the generator robs heat from the combustion
process that otherwise could be converted to steam enei*gy. Typical net thermal efficiencies (as
defined by equation 1) for three different fuels are shown in figures 6.9 through 6.11, as a
function of the flue gas temperature and percent excess air.2 These charts assume a 2% radiation
heat loss. As can be seen from these plots, the thermal efficiency increases as stack gas
temperature and percent excess air decrease.
A knowledge of thermal efficiency is essential for the proper operation of a steam generator.
A drop in thermal efficiency is usually an indication that either the tubes are fouled andor too much
excess air is being used for combustion. A stack gas analysis for CO;! will give an indication of
excess air being used. Steam generation equipment is not normally operated with excess air above
7%. Any amount greater than that will result in greater heat losses and lower thermal efficiency.
Otherwise, if too little excess air is used for combustion. it will result in incomplete combustion
and smoking and soot deposition on the surface of the tubes. In gas-fired steam generators, where
soot should not be a problem, loss in thermal efficiency generally indicates that scale has formed
inside the tubes. The presence of scale is indicated by an increase in flue gas temperature. Scale
buildup is also indicated by an increase in the feedwater pump outlet pressure. Accumulation of
soot on the external surfaces of the tube also reduces the heat transfer rate to the feedwater and
increases the flue gas temperature.
170
152
0
O,
148
144
5n
140
136
w
?i
,I- ,,
132
128
B
a
124
120
95
ll
......................
generatgrsett;"g.ratfiation.l~ss~~
.....................
:......................
;.....
70
300
400
500
600
700
800
900
1000
FUEL GASTEMPERATURE. OF
FIGURE 6.9. - Thermal efficiency versus flue gas temperature for IO" APT crude.*
171
100
150
200
250
300
350
400
450
94
90
86
82
78
FIGURE 6.10, - Thermal efficiency versus flue gas temperature for 15" API crude.2
$?
W'
94
Ps
90
a
>
w
I
5
Gz
86
LL
82
LL
-I
3w
f:
78
100
160
220
280
340
400
FUELGAS TEMPERATURE, OC
FIGURE 6.1 1. -.Thermal efficiency versus flue gas temperature for natural gas (1,000 BtdSCF).2
172
Any increase in the steam generator thermal efficiency reduces the fuel requirements of the
steam generator and sharply reduces the cooling load on the sulfur dioxide scrubbers (where they
are used). For example, the fuel requirements of an oil-fired 50 MM BTUihr steam generator can
be reduced roughly by about 5,300 bbl of oil per year by increasing the thermal efficiency 5%.
For this reason, one manufacturer offers steam generators equipped with special add-on convection
sections. Corrosion-resistance, plastic-coated heat exchange tubes are used in this section to
capture heat from the corrosive flue gases. This arrangement pennits the cooling of sulfur oxide
containing flue gases to about 150' F (well below their dewpoint temperature) and increases the
thermal efficiency by about 8%.
173
RECIRCULATION
RECIRCULATION
BLOWDOWN
PUMP
BLOWDOWN
TANK
174
and scale downstream equipment, but also result in particulate pollutants such as sodium salts left
after evaporation of the scrubbing liquor mist. The two most common mist eliminators are the wire
mesh pad, and the chevron mist eliminator. The chevron mist eliminator is a set of slats set in such
a way to impart a zigzag flow to the gas over a distance ranging from a few inches to few feet.
175
MM BTU/hr trailer-mounted unit. Since the unit is modular in nature, a number of such units can
be manifolded together to achieve the desired output capacity.
These units are light-weight, portable, inexpensive, and need no water treatment. According
to the manufacturer, a 10 MM BTU/hr unit costs less than $40,000.
Fluidized Bed Combustion (FBC) Steam Generutor9
Conventional oilfield steam generators use oil or gas as fuel. In fields where a coal source is
readily accessible, it may be more economical to use inexpensive and more plentiful coal as fuel.
One vendor offers a steam generator whose design features a fluidized bed combustion with an inbed steam generating coil and economizer section. These generators permit the burning of high
sulfur coal and various solid fuels to generate steam in an oilfield environment. The fluidized bed
design has the potential for providing uniform and low heat tluxes, minimum heat retention and
quick response to load changes and low fouling tendencies. This type of steam generator was field
tested in a South Texas steamflood project where steam was injected into a tar sand to recover a
heavy -2" API gravity tar.10 This type of generator was a natural fit for this project due to the
existence of an abundance source of low BTU coal and limestone in the project vicinity.
Operation Principle: Figure 6.13 is a schematic of the FBC steam generator. The crushed
solid fuel (coal, lignite, petroleum coke, waste wood, etc.) is mixed with limestone and fed into the
combustion chamber, where air is circulated at high velocity to agitate and suspend the burning
solids. The limestone reacts with the sulfur in the coal to limit the production of sulfur dioxide.
Feedwater is pumped through the heat exchanger coils in the combustion chamber where it is
heated and converted to steam. The operating conditions of the combustion are such that
production of nitrogen oxides and carbon monoxide are also minimized. Emissions from the FBCtype steam generator are kept within stringent environmental standards without using external
cleanup equipment.
The economics of the FBC steam generator depends upon the cost of solid fuel and the
capacity of the generator. It is unlikely that the generator will be economical to operate at capacities
less than 50 MM BTU/hr. This system should be considered only if a ready and reliable source of
inexpensive solid fuel is assured and stringent air quality standards have to be met.
The Wet Air Oxidation Boiler
wet 8aXidatk9n CQnCeQtl
The term wet oxidation as used here refers to a flameless combustion process capable of
releasing enormous amounts of heat under certain conditions. Around the turn of the century, it
was discovered that dissolved oxygen will attack water-wetted fuels directly from the liquid phase
176
1. Water Storage
2. Preheater
3. Economizer
4. Cyclone
5. Bag House
6. Fly Ash Collector
7. Dust Collector
8 Ash Collector
9. Coke
10. Limestone
11. Fuel Injection System
12. Forced Draft Fan
13. External Boiler
14. Forced Draft Fan
15. Injection Well
16. Spent Solids
if the materials are pressurized to 300 to 3,000 psi. Under these conditions the flameless oxidation
process generates temperatures of 300" to 700" F as compared with conventional combustion which
produces flame of 1,800" F at atmosphere pressure. However, the total number of BTUs released
by wet oxidation is the same as that released by conventional combustion. The principal reaction
products are also the same: carbon dioxide and water.
WAO Boiler12
Although wet air oxidation reactors generate enormous amounts of heat and are used
extensively in the waste disposal industry for the treatment of aqueous wastes and sludges, they
have not been used in the oil patch. Their inherent characteristics. however, make them ideal for
thermal EOR applications.
Figure 6.14 shows the schematic of one WAO arrangement for steam injection application.
The feed to the WAO boiler consists of fuel slurry, water, and compressed air (or oxygen). It
oxidizes the fuel and produces steam, C02. and N2 which can then be injected into the reservoir.
The rational behind the suggestion to use WAO process in steam injection application is that steam
contributes thermal energy to the reservoir, while the gas phase (C02 and N2) contributes both
thermal energy and gas drive/solubility effects to the process. Since by its inherent nature, the
process does not produce any S 0 2 , NO,, or particulate, and no emission control is necessary, and
the process can be used at locations where stringent air quality standards have to be met.
177
REACTOR
FUEL
OIL
PREPARATION
WATER
BLOWDOWN
4
I
A'R
COMPRESSED
AIROR02
GENERATION
/--
u
RESERVOIR
FIGURE 6.14. - Schematic of a wet air oxidation boiler arrangement for steamflood application.12
Other advantages of a WAO boiler for steam injection application are as follows:
(a) Inexpensive fuels such as high sulfur petroleum coke, high sulfur coal, lignite, high
(b)
sulfur crude-water emulsion, wood wastes, or any other solid waste product can be
used as feed to the reactor.
Feedwater requires no treatment. Brackish or oiltield brine or produced water can be
(c)
used as the source of water. Dissolved feedwater solids are concentrated and removed
in the blowdown.
The steam produced i n the WAO unit is essentially 100% quality; hence, the water
requirement per unit heat injected into the reservoir is signiticantly lower.
The major disadvantages of the WAO boiler are that it requires the use of an air compressor or a
compressed oxygen source to generate s t e m . and the process is corrosion prone. Hence, it is
more expensive to generate steam using the WAO boiler than a conventional boiler. Another
disadvantage of the WAO boiler is the problem of the disposal of reactor blowdown sludge.
The WAO process is a well-developed technology. However, the present economic situation
and the absence of field performance data are apparently the compelling reasons for the lack of
interest in WAO boilers. Because of increasing costs of meeting stringent air and water quality
178
regulation requirements, WAO boilers with their emission and water usage advantages may prove
to be cost-effective in steam injection applications.
179
I
I
I
WATER
WATER SYSTEM
I
1
I
I
I
I
I
IIUECT014 LIIiE
7 0 WELLS
FUEL SYSTEM
FUEL PUMP
ATMOSPHERIC
AIR IN
AND DRIVERS
CHEMICAL SYSTEM
CHELUCAL SUPPLY
TANKt
;a
CODE:
CONTROL VALVES
-D+
BLOCK VALVES
RELIEF V A L E S
STRAINERS
FILTERS
PUMPS
CHECKVALES
180
Disadvantages
1.
1.
3.
4.
5.
BTU of
heat injected.
I.
181
FEED
(SOFTENING)
RAW WATER STORAGE
PNA
There are nearly a dozen different DSG designs in existence. Much research has gone into
the development of DSG without significant success. Existing DSG can be broadly classified into
two groups: (1) low-pressure generators or indirectly fired DSG and (2) high-pressure generators
or directly fired DSGs. Here the terms high and low refer to the combustion pressure, not the
injection pressure. The two designs differ in the method of transfering heat from hot combustion
gases to produce steam.
Low-Pressure DSG
The key features of this generator are as follows:
(a) Energy is transferred to the water through a heat exchanger, thus, enabling the
combustion process to be conducted at a pressure less than the injection pressure.
(b) Smallest well in which the generator can be placed is 10 in. in diameter.
(c) The combustion products are returned to the surface where they must be cleaned up.
182
FUEL VALVE
COMBUSTION
CARBON STEEL
WATER
High Pr ess u re DS G
The distinguishing features of this generator are as follows:
(a) Steam is produced by direct contact with high-pressure combustion gases, whic..
eliminates the need for any heat exchangers.
(b)
The overall size of the generator is considerably smaller than a low-pressure version
and this permits the installation of the generator in existing injection wells.
(c)
The hot combustion products are injected into the formation and repressurize the
reservoir. This repressuiing can enhance recoveiy by improving flow characteristics.
Figure 6.18 is a schematic of a high-pressure DSG developed by Sandia National
Lab0ratory.1~The typical operating parameters for this system are shown in table 6.8. In the
Sandia design, fuel and air are injected and mixed in the upper portion of the combustion chamber.
Since the combustion operates at the injection pressure, a large air compressor is required at the
surface to provide high-pressure air for the combustion. The fuel and air are thoroughly mixed at
high pressure in the combustion chamber prior to ignition. Once sustained combustion is
achieved, high-pressure feedwater is sprayed directly into the flame through spray nozzles. The
water is flashed into steam, and the mixture of steam and combustion products is injected into the
formation.
The overall thermal efficiency of a high-pressure DSG is about 80%. Even though the
efficiency is not as high as that for the low-pressure DSG, it is considerably higher than that for a
conventional surface generator.
Results of the field tests conducted to date in evaluating the reliability of high-pressure DSG
under oilfield conditions have indicated a number of technical problems. Corrosion of burner
components seems to be the most severe and coininon problem among all DSCs. Other major
problems encountered include difficulties in controlling combustion at high pressure and frequent
packer failures resulting from the harsh DSG operating environment.
184
WATER
AIR ASSISTED
ATOMIZING NOZZLE
12"
-4STEAM
10-20
1,000-3.000
2,500-5.000
No. 2 diesel
80
Very low
Emissions
Combustion pressure, psia
Minimum casing size, in.
Generator length, ft
1,000-3.000
7
Even though DSGs have been field tested successfully, these tests are of short duration-the
longest being 6 months. The DSG needs to be field tested over an extended period of time (2 to 3
years) to establish its reliability and identify operational problems.
The economic advantage of DSG over surface generation is debatable. Even though DSG
has higher thermal efficiency, this efficiency does not always translate into dollars and cents
because the cost of fuel used to drive compressors often exceeds the savings in heat. Further,
since most heavy oils of interest in the U.S. are found in shallow reservoirs (<4,000 ft), a surface
generator with insulating tubing may be more cost-effective in delivering unit heat input to the
reservoir. The advantage of DSG over suiface generators in reducing the cost of compliance with
environmental regulations is also debatable because the injected com bustion gases more than likely
will break through to the producing wells and thereby reduce the economical advantage of DSG.
185
Based on a heat-balance study of a conventional surface steam generation system and the
downhole steam generation system, Sandia National Laboratory concluded that DSG has a lower
appeal to low-injectivity shallow reservoirs and high-injectivity deeper reservoirs.15 Figure 6.19
depicts the steam generator system selection chart recommended by Sandia. This graph, however,
excludes any consideration of investment and operating costs. The costs of energy injected into a
reservoir (1981 dollars) as a function of depth for surface and downhole steam generators are
shown in figure 6.20.
A fair comparison of the economics of the DSG versus surface steam generation is difficult
because such studies involve the comparison of an unproven technology with that of an established
technology. Published comparative economic evaluations are all preliminary and generic in nature
and reflect author's bias. To arrive at an honest conclusion, a site-specific process economic study
of the options is necessary. In table 6.9, the published equipment cost data are shown. Most of
the data were published in 1980-82. These are updated to 1991 dollars using the U.S. Producer
Price Index. Note that the equipment costs vary with improvement in technology, effect of
inflation, and the general state of the economy.
7500
S
6000
DOWNHOLE
OF
SELECT
DOWNHOLE
SYSTEM
.I/
4500
L
EFFICIENCY
SELECT
CONVENTIONAL
SYSTEM
1500
OL""""'
0
"
'
12
16
x io3
186
20
6'
0
1500
3000
4500
6000
7500
DEPTH, ft
FIGURE 6.20. - Cost of energy injected into reservoir versus reservoir depth.15
Author
and
Reference
Had5
Year
19822
Generator costs,
1991 $ per
MM Btu/hr
15,900
Water
treatment
plant costs,
1991 $ per
Well drilling
and
completlon
costs,
MM Btu/hr
1991 $
5,600
Generator
costs,
1991 .% per
MM H t u h
$81/ft
'38,200
26,360
+ $8.000
32,650
Water
lreatmen t
plant costs,
1991 $ per
M M Btu/hr
Well drilling
and
completlon
costs,
1991 $
6,900
4 $ 1 per
~ ft
+ $12,600
per ft
+ $10,600
CORC02:
198 1
10,600
8,200
'74,000
Eson21
1982
14,575
4.000
'
LOW
187
59,200
1,200
At the present time, because of the current economic situations and the unproven nature of the
technology, there is lack of interest in the oil patch for DSG. It is unlikely that the DSG will be
used in the shallow California reservoirs. DSG applicatjons may be limited to special situations
such as steamflooding of heavy oil reservoirs in Alaska where the fragile nature of the environment
and arctic climate preclude the use of surface steam generation facilities. It may also find a
practical application offshore where its compact size would be an asset.
Cogeneration
Another method of steam generation currently in use, principally in California, involves the
use of a heat-recovery steam generator (HRSG) as an integral component of a process known as
cogeneration. In the cogeneration process, the fuel (such as natural gas) is burned in a gas turbine
to generate electricity and the heat from the hot exhaust gas is recovered in a HRSG to generate
:
HEAT RECOVERY STEAM
J
0-1
FUEL
FLUE GAS
__I_)c.r3
STEAM TO
PRODUCTION
CrrY
WATER
I
b
OTHERUSES
188
plants, offshore oil platforms, and crude oil pipeline applications. Steam injection operators did
not utilize cogeneration technology until the late 1970s. principally due to economic reasons. In
the past, the low oil piice and the larger capital expenditure for cogeneration equipment combined
with the absence of a ready market for the excess generated electricity made cogeneration
economically unattractive for steam injection operators. However, the passage of the Public
Utility Regulatory Policy Act (PURPA) i n 1978 and the rising prices of crude in the 1970s
improved the attractiveness of cogeneration for steam injection operations. One of the provisions
of PURPA requires that state utility agencies must mandate local public utility companies to
purchase any excess power that a cogenerator has to offer. The contract with the electric utility
company is a key factor in ensuring the economic success of a large cogeneration system. In
California, the public utility commission has arriinged for standard offers. Cogenerators can take
advantage of them or can negotiate a special contract with the utility, if they so choose.
In California, most cogeneration systems are owned by the operators or by companies
specially formed for this purpose. These cogeneration plants are large units with a power
generating capacity of 80 MW or more. In table 6.10. the capital cost itemization of an 80 MW
California cogeneration project is shown. l * These data are included here to indicate the capital
investment requirements for large size cogeneration plants. Small steam injection operators can
merely purchase steam from these companies. The constraints to such an arrangement are heat loss
from the steam piped long distances prior to injection and long-term purchase commitments.
Long-term purchase commitments, while they assure a steady supply of steam at a negotiated price
for the duration of a project, also require operators to pay for steam not used due to project
shutdown. For this reason, most small steam injection operators i n California prefer to generate
the steam themselves.
M$
Power generation
Steam generator
General facilities
Switchyard & transmission
Emission offsets
Startup
Project engineering & management
Construction interest
9 . Offsite pipelines
10. Total
1.
2.
3.
4.
5.
6.
7.
8.
2.5.637
8.900
7,578
1,385
1,170
2,051
2.160
565
1.320
50.766
% of
Total cost
50.5
17.5
14.9
2.7
2.3
4.0
4.3
1.2
2.6
Small operators who do not wish to enter a contract with large cogenerators yet wish to avail
themselves to the benefits of cogeneration can do so by installing small custom-designed skidmounted cogeneration units. These are designed for unattended operation in a dusty outdoor
environment and generate approximately 2.5 MW (8.53 MM BTUlhr) of electricity and 40 MM
BTU/hr (2,600 bbVd) of 80% quality steam at 1,500 psig. One vendor's economic summary of a
packaged cogeneration unit (updated to 1091 dollars using the U.S. CPI) is shown in table 6.1 1.l6
These cogeneration economic data are included here for illustration only, and the actual vendor
quote may differ from these 'off-the-cuff' estimates. The net cost shown in the table is the cost of
the system minus the cost of the steam generator that would have been required if the cogeneration
system had not been installed.
According to one steam injection operator who installed such a system (in the mid-80s) on
his lease, the generated power cost from Cogeneration units were about 25% of purchased power
costs and the payout was about 2.4 years.17 His annual maintenance expense averaged about 5%
of the initial turbine investment.
With the exception of the turbine and the electric generator, a cogeneration facility is designed
and operated to produce steam similarly to a standard oil field steam generator. Both require
essentially the same feedwater quality (zero hardness and low TDS) and operate within the same
pressure ranges. Since a cogeneration steam generator is designed to recover the waste heat from
the hot exhaust gases of a turbine, only a convective heat transfer section is used in a cogeneration
facility. Since cogeneration units are designed to burn sweet natural pas, no SO2 is formed. The
NO, emissions are usually controlled by iiijecting demineralized water into the turbine combustion
chamber.
The operating problems of cogeneration steam generators are similar to those of standard oil
field steam generators. The discussion of the operation and problems of a turbine generator set is
beyond the scope of this report.
SUMMARY
Steam generators used in oilfields differ from conventional power boilers. Oilfield steam
generation equipment and techniques have evolved over the past 30 years to a highly specialized
state. This chapter reviewed the once-through wet steam generators that are used almost
exclusively in oilfields. Essential operatins and maintenance techniques are discussed. The
importance of good operation practices and mechanical maintenance schedules that reduce the
overall maintenance and operating costs are emphasized. Salient features of other specialized
oilfield steam generation equipment are detailed.
$3,000,000
$200.000
$3,200,000
Operating costs2
Turbine fuel, 37 MM BTUhr
Supplemental fuel, 36 MM BTU/hr
Demineralized feedwater
Treatment and disposal of waste water
Operation and maintenance
Total operating costs
yabuhlu2
= $126,000/month
= $100.000/mon th
= $226,000/month
$
97,000/month
$3,200,000
$700.000
Net cost
$2.500.000
38.000
2,500
7 00
1.2
0.07
0.8
I10
0.31
x fuel cost
$/MM BTU
REFERENCES
1 . Warren, K. W. Advances in Oil Field Steam Generation. Meyer, R. F., J. C. Wynn
and J. C. Olson, eds. The Future of Heavy Crude and Tar Sands. Second International UNITAR
Conference in Caracas, Venezuela, February 1082, McGraw-Hill, New York, pp. 685-693.
3. Peachey, B. R. Design Consideration for Very Large Oilfield Stearn Generators. Pres.
at the 35th Ann. Tech. Meeting of the Pet. SOC.of ClM, Calgary, Canada, June 10-13, 1984.
Paper No. 84-35-8 1.
4. American Petroleum Institute. API Recommended Practice for Installation and Operation
of Wet Steam Generators. API Publication API RP 11T, 1st Ed., March 1983. American
Petroleum Institute, Production Dept., Dallas, TX, 1983.
5. Gjerde, E. Raw crude: A Heat Source for Steam Flooding. Pet. Engr. v. 36, No. 13,
December 1964, pp. 92-94.
6. Hampton, L. A, How various Fuels Affect the Design and Operating Costs of Steam
Generators. Pres. at the 19th Ann. Tech. Meeting of the Pet. SOC.of CIM, Calgary, Canada, May
7-10, 1968. Paper No. 6812.
7. Prats, M. Operational Aspects of Steam Injection Processes. In: Donaldson, E. C., G.
V. Chilingarian and T. F. Yen, eds. Enhanced Oil Recovery 11-Processes and Operations,
Elsevier Science Publishing Co., New York, 1989, p. 359.
8. Modular Steam System: Vapor Tech Steam Generator-A Lightweight Steam Generator
for In-field Steam Flooding-Modular Steam System. Technical Bulletin, Earth Resources
Technology Services, Phoenix, AZ.
9 . Martin, W. L., M. W. Britton and R. A. Harmon. Conocos South Texas Tar Sands
Project. In: Meyer, R. F., J. C. Wynn and J. C. Olson, eds. The Future of Heavy Crude and
Tar Sands. Second International UNITAR Conf., Caracas, Venezuela, February 1982. McGrawHill, New York, pp. 987-997.
10. Jones, 0. and J. S. Davis. Solid Fuels Proven for Oil Field Steam Generation. Oil &
Gas J., v. 81, No. 8, Feb. 27, 1983, pp. 189-192.
11. Clark, S. W. Wet Oxidation Downhole Steam Generator for Recovery of Deep Heavy
Oil. Pres. at the Third International UNITAR Conf. on Heavy Crude and Tar Sands, Long Beach,
CA, July 22-31, 1985. Proceedings v. 3, pp. 1,116-1,119.
12. Balog, S. E., R. K. Kerr and L. A. Pradt. The Wet Air Oxidation Boiler for Enhanced
Oil Recovery. J. Can. Pet. Tech., v. 21, No. 5 , September-October 1982, pp. 73-79.
13. Sperry, J. S. Development and Field Testing of the Vapor Them Process. Pres. at the
Third ERDA Symp. on Enhanced Oil & Gas Recovery and Improved Drilling, Tulsa, OK, Aug.
30-Sept. 1, 1977. Proceedings, v. 1, pp. D-2/1-D-2/14.
14. Bader, B. E., R. L. Fox, D. R. Johnson, A. B. Donaldson and D. A. Krueger. Deep
Steam Project. Quarterly Report Oct. 1-Dec. 3 1, 1978. Sandia Laboratories Report SAND-790562, 1979.
15. Hart, C. M. Comparative Analysis of Steam Delivery Cost for Surface and Downhole
Stem Drive Technologies. Sandia National Laboratories Report No. SAND-8 1-0758, 1981.
16. Solt, J. C. and I. R. McNeill. Cogeneration in the Oil and Gas Industry Solar Turbines
Technical Publication No. I T S 43, Solar Turbines Inc., Div. of Caterpillar, Oak Brook, IL, 1986.
192
17. Livesay, J. D. Long-Term Performance of Small Cogeneration Units in Oil Field and
Gas Plant Operation. Pres. at the Pet. SOC.of CIM and the Soc. of Pet. Engineers International
Technical Meeting, Calgary, Canada, June 10-13, 1990. Paper CIM/SPE 90-14.
18. Harbor Cogeneration Project Fact Sheet - Union Pacific Resources, Wellmington, CA,
1989.
19. NATCO: "Therma Drive" Steam Generator. Technical Bulletin No. 1701, National
Tank Company, Tulsa, OK, 1990.
20. Nguyen, D., S. Singh and S. Wong. Technical and Economic Criteiia for the Selection
of Downhole Steam Generators in Alberta. Pres. at the 38th Ann. Tech. Meeting off the Pet. SOC.
of CIM, Calgary, Canada, June 7-10, 1987. Paper No. 87-38-04.
21. Eson, R. L. Direct Fired Downhole Steam Generator-Field Tests. Meyer, R. F., J. C.
Wynn and J. C. Olson, eds. The Future of Heavy Crude and Tar Sands. Second International
UNITAR Conf., Caracas, Venezuela, February 1982, McGraw-Hill, New York, pp. 737-743.
22. Schiimer, R. M. and R. L. Eson: A Direct-Fired Downhole Steam Generator-From
Design to Field Test. J. Pet. Tech., v. 37, No. IO, October 1985, pp. 1903-1908.
Bibliography
193
194
Uneration
Anon. Cogeneration Unit to Assist Heavy Oil Recovery. Oil & Gas
NOV.7, 1983, pp. 68-69.
Western, E. R. and D. W. Nass. Cogeneration Improves Thermal EOR Efficiency. Oil &
Gas J., v. 88, No. 40, Oct. 15, 1990, pp. 41-43.
Davis, J. S., W. W. Young and C. J. Lyns. Use of Solid Fuels Possible for Field Steam
Generation. Oil & Gas J., v. 79, No. 22, June 8, 1981, pp. 129-134.
Sperry, J. S., R. S . Poston and F. S. Young. Development and Field Testing of the Vapor
Therm Process in the Carlyle Pool-Allen County, Kansas. Pres. at the Fourth Ann. DOE Symp.
on Enhanced Oil & Gas Recovery & Improved Drilling Methods, Tulsa, OK, Aug. 29-31, 1978.
Proceeding, v. 2, No. lB, pp. D-2/1-D-2/29.
Sperry, J. S., F. S. Young and R. S. Poston. Debelopment and Field Testing of a Process
for Recovering Heavy Crude Oil in the Carlyle Pool-Allen County, Kansas Using the Vapor
Therm Generator. U.S. Dept. of Energy Report No. DbEIBETC-2880- 1, September 1980.
Young, F. S. and R. W. Krajicek. The Vapor Therm Process for Oil Recovery of Viscous
Crude Oil. In: Meyer, R. F., J. C. Wynn and J. C. Olson, eds. The Future of Heavy Crude and
Tar Sands. Second International UNITAR Conf., Caracas, Venezuela, February 1982, McGrawHill, New York, pp. 466-468.
105
APPENDIX 6-A
BTU/lb.
BTU/l b.
Lb/ft3
ssu
ssu
ssu
ssu
% Wt
% Wt
% Wt
% Wt
% Wt
% Wt
% Wt
% Wt
mg/l
mg/l
mg/l
mg/l
Lb/100 bbl
197
APPENDIX 6-B
APP RECOMMENDED FUEL GAS ANALYSIS SPECIFICATION SHEET4
BTU/SCF
BTU/SCF
Ft3/lb.
O F
Components (Mol %)
Carbon Dioxide ( C a )
Hydrogen Sulfide (H2S)
Nitrogen (N2)
w g e n (02)
Methane (C&)
Ethane (c2H6)
Propane (C3H8)
Butanes + (C4+)
%
%
APPENDIX 6-C
API RECOMMENDED WATER ANALYSIS SPECIFICATION SHEET4
Source:
Pressure:
Temperature:
Components
*Calcium, (Ca)
*Magnesium, (Mg)
*Oxygen, ( 0 2 )
Barium, (Ba)
*Iron, (Fe)
*Sodium, (Na)
Copper, (CUI
Potassium, (K)
Ammonium,
*Chloride, (Cl)
Iodide, (I)
Bicarbonate, (HCO3)
Carbonate, (CO3)
Silica, (Si02)
Sulfate, (SO4
Hydroxide, (OH)
Hydrogen Sulfide (H2S)
Organic Acids
*Total Alkalinity
*Total Hardness, (CaC03)
*Total Dissolved Solids
*Total Suspended Solids
(m)
*PH
Color
*Oil
*Minimum Data Required
mgn
APPENDIX 6-D
API RECOMMENDED MAINTENANCE PRACTICE
FOR OILFIELD STEAM GENERATOR4
6 . 3 . 4 If the Wet Steam Generator is shut down the
condition of the convection section and radiant
section tubes should be noted. Fin loss
deterioration o r fouling should be noted and
recorded.
200
Unit No.
Inspection Date
Location
Inspectors Signature
Visual check of external wiring, conduit, piping, tubing and fittings for obvious leaks or defects.
All gauges must be checked for proper calibration.
* Consult state and local codes for test procedures.
Electrician
Date
Mechanic
Date
~~
Operator
Date
Foreman
Date
20 1
b.
202
DescriDtion of item
OK
Radiant section
Blowdown valve
ANNUNCIATOR PANEL
Flame monitor
High steam kmp.
High tube temp.
High stack temp.
High burner temp.
Low oil temp.
High steam p r e x
High atom. press.
Low atom. press.
Low gas press.
Low oil press.
Low instrument air press.
Low cornhushon air press.
203
ReDlaced
Comments
e w o f l t e m
L
la
c
OK
PNEUMATIC SYSTEM
Air compressof
Air receiver tank
HigMow press. switch
Air relief valve
Pressure gauge
Gauge valve
Instrument air filter
Air pressure regulator
Air pressure regulator
Pneumatic transmitter
Transmitter valve manifold
Water flow rate gauge
Burner control signal
Flow purge solenoid
High purge solenoid
Low fire solenoid
High signal selector
Manual loader (water)
Bias regulator (burner)
Burner operator w/positioner
Pressure controller
MAIN GAS SYSTEM
Gas inlet
Manual gas stop valve
Safety shut off valves
Vent valve
Gas pressure regulator
Gas Orifice Flanges
PILOT GAS SYSTEM
Manual pilot gas stop valve
Pilot regulator
Pilot solenoid valve
204
Renlac;eP.
Comments
205
Renlaced Comments
APPENDIX 6-F
2. N A T O
Division of National Tank Company
4550 Pierce Road
Bakersfield, CA 93308
Telephone: (805) 325-4146
Contact: Bert Miles
1. NATCO
Division of National Tank Company
4550 Pierce Road
Bakersfield, CA 93308
Telephone: (805) 325-4146
206
CHAPTER 7
STEAM DISTRIBUTION
INTRODUCTION
Steam leaving generators is directed to injection wells through a network of pipelines called
the steam distribution system. The high temperature and pressure of the steam to be injected
influences the size and installation of lines transmitting it. In the past, a steam distribution system
was considered merely a pipeline between steam generators and injection wells. However,
experience of the past decade indicated that the network can behave as an unpredictable phase
separator delivering near 100% quality steam to one well and at the same time deliver hot water to
another well on the same distribution line. Hence, the steam distribution network must be
designed not only for appropriate pressures and temperatures, but also to provide desired steam
quality at a desired flow rate to each injection well in the network. Consideration must be given to
maintaining optimum line velocities and provide for the measurement of steam quality at points
other than the generator outlet.
The objective of this chapter is to present an overview of the features and requirements of a
steam distribution network in a modern steam injection project. The basic function of the steam
distribution network in a steam injection project is to deliver steam reliably and safely to injection
wells at a desired rate and quality.
Components of a Steam Distribution Network
A steam distribution network consists of the following major components: (a) main headers;
(b) lateral steam lines; (c) wellhead steam rate controls, and (d) wellhead assembly. The design of
a steam distribution system is based on the following factors:
1. The expected rate of delivery of steam to each injection well.
2. The expected system operating pressure (This will be principally a function of the
expected reservoir injection pressure).
3. The desired steam quality at each injection well.
4. Minimize the effects of phase-splitting.
The principles of-operation of a steam distributionsystem are simple and straightforward.
Figure 7.1 shows a schematic of a typical steam distribution system employed in a steam injection
project. The steam generator(s) deliver steam to the main header usually at a steam quality of 70 to
80%. These steam headers range in size from 12 to 24 inches, depending on the steam generation
capacity of the project and project steam requirements.
Steam is transported from a steam header to individual injector well lines by a steam lateral
system. The lines making u p this system typically range between 3 and 10 inches i n nominal
diameter.
207
STEAM HEADER
R //
The rate of delivery of steam to each individual well is controlled by the installation of some
form of rate control, in general at the wellhead itself. These control devices are currently of two
forms. The first is merely a static choke driven by system pressure to critical (or sonic) flow. The
second device widely used is a manually adjustable choke or control valve operated in the
subcritical flow regime. Both forms of control are discussed in detail later in this chapter.
Finally, the steam injector downhole tubing can be expected to expand and contract as its
temperature rises and falls. Since it hasibecome common practice to install a downhole tubing
packer, the movement of the expanding tubing downhole can become somewhat restricted. The
movement of the tubing through the wellhead can become considerable. It is the purpose of the
wellhead assembly not only to deliver the'steam from the distribution network to the injection well
tubing, but also to provide for the expected tubing movement at the wellhead.
The discussion within this chapter is:organized into the following sections:
(a) General system pressure and temperature design requirements;
(b) Main steam headers;
(c) Lateral steam lines;
208
TABLE 7.1.- Design Properties and Allowable Working Pressures (psig) for Piping'
ASTM A106, grade B seamless pipe - petrolcum refinery p,iping code
for pressure piping ANSI B31.3-1966 - Coiwsioti allowatlce = 0.05
Nom
pjpe
I D,
thk..
(d)
-2010
100
200
300
400
500
600
700
800
0.109
0.622
2.260
2.160
2.050
1.050
1.850
1.750
1,630
1,220
1.050
0.113
0.154
0.824
0.742
I *940
3.460
1.850
1.760
3.140
I .670
2.980
1,590
2.830
1.500
3,300
3.680
1.390
2.480
1,050
1.870
1.679
2.172
2.844
3.659
,315
0.133
0.179
0.250
0.358
1.049
0.957
0.815
0.599
2.1 IO
3.470
5.720
9,540
2.010
3.320
5.470
9.1 10
1.910
5.200
1.820
3.OW
4.940
8.230
1,730
2.840
4.680
7.800
1.640
2.690
4.440
7.390
1.510
2.490
4.1 10
6,850
1,140
1.880
3.090
5,150
.900
0.145
0.200
0.343
0.400
1.610
1.520
2.530
4.100
6.9 IO
4.180
1.100
I .680
2.780
4.600
7.230
1.600
2.660
xx
2.718
3.632
7.445
6.409
I .450
2.400
3.9 70
6.240
1.370
2.280
3.770
5.910
1,300
2. I60
3,570
5.610
1.200
2.000
3.310
5.190
910
1.510
2,490
3,910
S40
X80
160
3.653
5.022
6.445
2.375
0.154
0.2 18
0.343
0.436
2.067
1.939
1.687
1.503
1.170
2.490
4,600
6.290
2.380
4.400
6,010
1.410
1,340
2.260
5.710
1.210
2.010
3,770
5.140
1.140
1.930
3,570
4.880
1.060
4.180
1.270
2.150
3.970
5.420
1,790
3.310
4,510
800
1,350
2.490
3.400
0.216
0.300
0.438
0.600
3.068
2.900
2.624
2.300
1.640
2.560
4. I30
6.090
I .570
2.110
3.940
5.820
1.490
2.320
3,750
5.530
1.120
2.210
3,560
5.260
1.350
2.090
3.370
4,980
1.280
1.980
3.200
4.720
1.180
1.840
2.960
4,370
890
1.380
2,230
3.290
0.237
0.337
0.531
0.674
4.026
3.826
3.438
3.152
1.4-w
2,280
3.980
5.3 IO
2.180
I .380
1,250
1.970
3.440
1.580
1.180
1.870
3.260
4.340
1.120
1.770
3,090
4.120
1.040
1.640
5.070
1,310
2,070
3.620
1.820
2.860
3.810
780
1,230
2.150
2.870
0.280
0.432
0.7 18
0.864
6.065
5.761
5.187
1.897
1.210
1.160
2.070
3.7ho
4.6W
1.970
1.100
I .8xo
1,050
I .7XO
4.460
3.410
4.230
3.240
4.020
990
1.690
3.070
3.810
940
I .600
2.9 I O
3.620
870
1.480
2,700
3.350
1.120
2.030
2,520
0.322
7.981
7.625
5.187
6.813
1.100
1.870
3.760
3.700
I .050
I .ow
900
1.530
3.070
3,030
860
I .450
2.910
2.870
790
1.340
2,700
2.660
2.030
2.ooo
1.125
10.020
9.750
8.500
1.490
3.730
1,420
3.570
0.375
0.500
1.312
12.000
11.750
10.126
890
1.250
3.70
850
1.190
3.540
3.360
3.200
0.250
0.375
0.500
13.500
15.250
13.000
486
810
1.I40
465
7XO
442
740
420
700
1.090
1.030
0.250
0.375
0.500
15.500
15.250
!5.000
426
710
407
0.250
0.375
0.500
17.500
17.250
17.000
Sch.
No.
Wt of
pipe.
1b/ft
0 D,
in.
If2
S40
0.851
0.840
314
S40
X80
1.131
1.474
S40
X80
size.
in.
160
xx
1-1R
S40
X80
160
xx
S40
X80
160
7.58
10.25
14.33
18.58
3.500
S40
X80
160
10.79
14.99
22.51
27.54
4.500
S40
X80
160
18.98
28.58
45.30
53.17
6.625
28.56
43.4
45.30
14.7
8.625
xx
xx
xx
S40
X80
I60
xx
10
12
14
16
18
9.030
40.5
54.7
115.7
10.750
S40
X80
12.750
160
49.6
65.4
160.3
S40
X80
160
36.7
62.6
72.1
14.000
S40
X80
160
42.1
62.6
82.8
16.000
S40
47.4
70.6
93.5
18.Ooo
160
0.500
0.718
0.906
S40
X80
160
X80
in.
0.365
0.500
in.
1.500
1.687
1.030
3.800
3.590
1.790
3.590
3.540
Y 80
3.150
8.660
6.560
950
600
1.010
1.700
1.610
3.410
3.360
3.240
3.200
930
1.350
890
840
I .280
1,220
3.060
800
1.160
2.900
740
1.070
2.690
560
810
2,020
810
770
1.140
1,080
730
I .020
3.030
690
970
2,870
640
900
2.660
480
680
2.000
398
980
670
930
377
630
880
349
580
820
263
437
620
331
550
770
306
510
710
231
382
540
293
486
272
450
630
205
339
474
3.100
3.210
387
650
368
610
349
580
990
680
950
900
860
810
378
630
361
344
600
880
830
327
540
7ho
3 10
520
720
210
660
570
800
680
Nom
pye
size,
in.
20
24
Sch.
No.
OD,
pipe,
1bfit
thk,
in.
in.
0.250
0.375
S40
X80
160
52.7
78.6
104.1
20.000
S40
X80
160
63.4
94.6
125.5
24.000
I D.
Wall
Wt of
(d)
-2010
In.
100
200
0.500
19.500
19.25
19.000
340
570
790
325
540
760
0.250
0.375
23.500
23.250
282
468
0.500
23.000
660
270
441
630
300
400
500
60
520
720
291
485
680
278
460
650
436
620
256
244
231
383
540
219
363
309
425
600
401
570
700
244
404
264
510
570
184
304
426
203
336
470
153
253
354
150-lb
300-lb
4001b
9Wlh
600-lh
1.500-lh
2.500-lb
100
200
300
215
400
240
210
180
500
150
600
130
700
110
720
700
680
665
625
555
470
960
930
910
890
835
740
635
1,440
1.400
1,365
1.330
1.250
1.110
940
2.160
2.100
2,050
2.000
1.875
1.660
1.410
3.600
3.500
3,415
3,330
3.125
2.770
2.350
10
20
40
60
80
100
120
I40
160
I80
200
220
240
260
280
300
320
800
Expansion,
in.1100 ft
0.145
0.293
0.430
0.593
0.725
0.898
Change in
lemperalure.
Expansion,
i d 1 0 0 ft
340
2.862
3.029
3.21 1
3.375
3.566
3.740
3.929
4.100
4.296
4.487
4.670
4.860
5.051
5.247
5.437
5.627
5.831
360
380
400
420
440
1.055
460
480
I .209
1.368
1.528
1.69I
1.851
2.020
2.183
2.350
2.519
2.690
SO0
520
5 40
560
580
600
620
640
660
21 1
6,000
5,830
5,690
5.550
5.210
4,620
3,920
The purpose of the main steam headers is to collect steam produced by one or more steam
generators at one or more sites and deliver this steam to various steam laterals for distribution. The
steam collected by the main headers will vary in steam quality from generator to generator. In
addition, the quality of the steam delivered by a generator during its initial start-up may approach
zero.
Depending on the volume of steam it is expected to carry, the main header size may vary
from 12 to 24 in. Steam headers usually have extra strong wall thickness-more than that required
by the code-to allow for extra margin of safety. The size of the steam pipe is usually determined
using empirical steam flow equations such as the Unwin formula (Eq. 7.1) or Fritzsche's formula
(Eq. 7.2)6
The Unwin formula for steam flow is
AP =
0.002705 (1 +
3$r) VLW2
74,500 d5
(7.1)
where
d = internal diameter of the pipe, in.
L = length of pipe, in.
P = pressure in lb per sq in. gauge
W = rate of flow of steam in Ib per hr
V = specific volume of steam, cu ft per lb
The Fritzsche formula is believed to give more accurate results for pipes IO-in. and larger, carrying
steam at high velocities.
The headers are usuaily sized based on the minimum acceptable pressure drop in the header.
If the available steam pressure is lower than the acceptable pressure drop, the header should have
been sized larger than normal to minimize pressure losses. Since the headers carry large volumes
of steam and are located adjacent to the generators where there is greater exposure to personnel,
they should be designed and fabricated using the most stringent standards. Construction practices
should be based on accepted standards, and pipe welds should be subjected to more frequent X-ray
inspections.
2 12
A shut-off valve must be provided at each steam lateral connection to the header. This will
facilitate the isolation of any lateral for repair or maintenance. A bypass valve should also be
installed in parallel with the shut-off valve to allow for slow preheating of the lateral.
Because a steam generator can introduce significant amounts of cold water to the steam
header during the startup and shutdown phases, i t is a common field practice to install a
"blowdown" line. This line permits the operator to divert the discharge from the generator to
atmosphere and to a sump during the startup or shutdown phases. By using this line, the operator
can minimize the stresses placed on the steam header system.
shock, but liquids left in the line as a result of construction and testing can be delivered to the
injection wells at low and therefore safe velocities. During the steam bypass operation, injection
wells may be opened and closed in sequence beginning with the wells nearest the beginning of the
branch. The presence of flowing liquid in the lines may be detected by the audible differences at
the well rate control device (the steam choke or the control valve).
In laying out the steam lateral system, the designer should avoid the use of circular paths or
steam flow loops, for example, terminating one lateral branch into another. Although this might be
attractive in minimizing pressure losses, it can lead to hazardous situations when a portion of the
system must be isolated and depressurized for repair or maintenance.
214
TO INJECTIONWELLS
FIGURE 7.2. - Typical expansion loops used in steam injection operations to relieve
steam flow line theimal stresses.3
EXPANSON
il
EXPANSION
EXPANSION -9
FIGURE 7.3. - Schematics of different expansion methods for steam injection lines.2
215
Dead anchors must be installed between expansion devices to force the pipe movements to
the expansion device. The location of the dead anchors is largely determined by the piping layout.
The expansion devices are usually installed approximately midway between two anchors. Table
7.4 can be used to select the loop size and anchor spacing for vertical loops. Z-bend type
expansion joints can be sized using figure 7.4.
TABLE 7.4.- Loop Size vs. Spacing
Loop Size. ft
pipe. Size,in.
Spacing. ft
lleight
150
200
250
150
I
8
2" sch.40
3" Sch. 40
8
9
8
9
IO
250
200
12
9
..
11
II
rn
LJ U
10
12
I4
15
I6
,
I
200
250
6'Sch. 40
LW
I
,.A
4- sch. 40
Width
.-
17
40
35
30
a=
25
i 20
15
10
5
Fin
-_
100
150
200
250
300
350
400
216
I 3
16
15
Expansion loops must be equipped with bleed valves at the upstream side5 so that it is
possible to do the following:
(a)
Bleed off the line when any shutdown occurs. This is very important in cold climate,
where the lines may otherwise freeze during winter.
(b)
(c)
the line and the wellhead. Figure 7.5 shows a typical swivel joint used at the injection wellhead
which permits vertical movement of tubing and horizontal movement of steam line. Figure 7.6
shows an expansion manifold used to connect the Christmas tree and flow line of a steam injection
well. The dotted part of this figure indicates movement of tree and manifold due to thermal
expansion.
Support of Steani Lines
An expansion is not effective unless both lines to the loop are supported. Otherwise, the line
may creep toward one loop and not retui-n to its original position upon cooling. The determination
of the spacing between these supports should be based on basic statics and the assumption that the
longitudinal pressure stress i n a pipe is no more than one-half the allowable stress. This
217
s
T
FIGURE 7.5. - Schematic of a typical swivel joint used at the steam injection wellhead.3
assumption would allow the bending stresses due to weight loading to be as much as one-half the
allowable stress. Since the effects of thermal expansion on bending stress are rarely completely
known, a span length that produces bending stress no more than one-quarter the allowable stress is
preferred. The formulas for support spacing N (ft) are:
For a continuous beam:
N = I'ZSI~W~
For a free span:
N = VZSI6wf
Allowable
spacing
of supports, ft
where N =
Section modulus of pipe, inches3
2m
- 0.0982 (D4-d4)I D
I
=
Moment of inertia, inches4
(7.3)
17.4)
z =
0.0491 (D4-d4)
Cross-sectional metal area, inches2
I7 (D2-d2)/ 4
Radius of gyration, inches
-14
rg
D
d
Arg2
=
=
219
cold water to the steam header. If this should occur, "water hammering" of the header can be
expected. This will result in very high instantaneous pressures within the header. These pressures
will act to place the header in a stressed condition similar to that o f a loaded structural column. The
Euler equation, which predicts the failure load for structural columns, suggests that the strength of
the column can be increased by "pinning" the column at points along its length. Properly designed
guided supports can serve to "pin" the header. The equations described under the laterals section
should be used as a guide for selecting header pipe support spacing, but in no case should the
spans calculated using these equations be exceeded.
220
Temperature difference, O
Nominal
pipe size
2
3
4
3
4
300
400
500
578
935
1,177
1,701
1.040
1.496
1.894
2,740
1.552
2.25 1
2,845
4,127
2,198
3.196
4,043
5,878
107
158
203
299
143
21 1
27 I
399
178
263
338
498
214
316
406
598
98
130
139
185
246
297
410
236
314
380
522
250
488
70 1
886
1,279
Buried pipe: (24 in. cover-average California)
89
132
169
249
Insulated pipe: (1-1/2 in. 85% Magnesia)
2
3
4
6
77
103
125
171
3000
2500
2000
1500
1000
184
223
307
158
216
600
10
20
30
40
FIGURE 7.7. - Steam injection line heat loss versus cash loss.
22 1
50
Depending on the expected operating pressure of the steam distribution system, the internal
operating temperature of the system can be expected to range between 350" and 600" F. At these
temperatures, the decision on the type and thickness of the pipe insulation becomes a question of
economics. Two types of pipe insulation are commonly used in steamflood operations:
1 . Fiber glass wool wrapped with a thin skin of sheet metal (usually aluminum) - the
thermal conductivity of the fiber glass wool is 0.022 Btu/hr-ft-OF.
2. Calcium silicate wrapped with an impervious barrier followed by a thin skin of sheet
metal (usually aluminum) - the thermal conductivity of the calcium silicate is 0.017
Btu/hr-ft-"F.
The convection coefficient for heat transfer from the outside skin of the insulation surface is
generally taken to be in the range of 2 to 5 Btu/hr-ft2-"F.
The rate of heat transfer from a unit length of pipe may be calculated by referring to
figure 7.8 and the following radial heat transfer equation.
q/L = Uo (a
AT)
= 27c (Ti - To) /((ln(r/ri) /k)
where
+ lhr)
q/L
UO
A
AT
Ti
air,O F
Pipe wall temperature-assumed
TO
r
ri
FIGUKE 7.8. - Schematic of insulated pipe for heat loss transfer estimation.
222
(7.5)
Until the late seventies, it was common practice to assume that steam delivered to a steam
lateral system at 80% quality would, after traveling through the various branching points of the
steam distribution network, aii-ive at each wellhead at 80% quality or less after any adjustments for
heat losses in the lines. In 1978, Chevron published a research paper which demonstrated that this
was not the case. They demonstrated that in laboratory tests steam quality was not being evenly
distributed between two branches of a pipe tee but rather, depending on flow rate (and some other
parameters as well), the liquid phase would favor flowing through either branch and was rarely
split evenly between the two.
Chevron went on to demonstrate that the usual practice of installing the pipe tee in a "straightthrough" fashion (refer to Figure 7.9) further provoked the problem of unequal phase splitting. As
a result of further research into the problem of phase splitting at pipe tees, the installation of pipe
223
tees in the "dead-end" fashion of Figure 7.9 is preferred. I n addition, an even split into the two
branches is preferred with a 30-70 split considered the maximum acceptable uneven split. It is also
considered good practice to install the pipe tee as level as possible.
The problem of unequal phase splitting at steam distribution system branches is unfortunately
very complex. No less than six flow regimes have been identified for the combined flow of liquid
and gases (vapors), and some authors have identified seven or more. The flow regime of a vaporliquid mixture can be very unstable and therefore very difficult to predict. As a result, computer
pipe network models which properly account for the phase splitting phenomenon are currently
being developed. Their further development will require modeling of existing steam distribution
networks and comparison of the modeling results with actual field measurements. To further
complicate the problem, steam distribution systems currently in operation seem to demonstrate a
tendency towards less than stable delivery of constant steam qualities. That is, at times a given
system may deliver a relatively high quality of steam to a particular well and a short time later the
same well may be delivered a low quality. For the present, the designer will be left with only the
guidelines of good engineering practice as previously described (dead-end tees with relatively even
flow splits between branches).
Wellhead Equipment
Wellhead design for a steam injection well depends on depth, temperature, and pressure of
the steam and casing and tubing sizes. The wellhead pressure and therefore the system pressure
depend on the maximum steam injection rate that the well can sustain.
Straight-Through Tee
Dead-End Tee
224
Figure 7.10 illustrates a typical wellhead assembly used in a steam injection project. Figure
7.11 shows the wellhead details for a typical steam injection well.
As previously discussed, all tubular goods in an injection well will expand as temperature
increases. But the amount of expansion will not be the same for the tubing. casing, and surface
pipe. The reason for this uneven expansion is that these strings are of different lengths, and at
different temperatures. Hence, provisions for this differential expansion must be made at the
wellhead, unless a packer or an expansion joint is able to accommodate the tubing movement.
Downward expansion of tubing and casings are discussed in a later chapter. Surface expansion of
casing and tubings are usually handled by wellhead stuffing boxes. Figure 7.12 shows a typical
thermal wellhead stuffing box that allows casing to expand and contract as required when a well is
steamed. A tubing stuffing box for steam injection well is shown in figure 7.13. A tubing stuffing
box is similar in design to a casing stuffing box except that it has provisions to allow for
installation of a packer.
A variety of stuffing box designs are used in steam injection applications, depending on the
m o u n t of expansion anticipated, depth, temperature, and pressure. Two such designs are shown
in figure 7.14. The wellhead shown in figure 7.14a provides for upward expansion of casing and
downward expansion of tubing with the rise in temperature. The wellhead shown in figure 7.14b
provides upward expansion of both casing and tubing. This wellhead design contains two separate
-?GATE
VALE
MASTERGATE VKM
ISERIES 400
WEMHW)
TOP FLANGE FOR
5 WWffi
225
TE
ANNULUS
VALVE
FIGURE 7.12. - Schematic of a typical steam injection well casing stuffing box.4
226
FIG IRE 7.13. - Schematic of a typical steam injection well tubing stuffing box.4
TUBING HANGER
STUFFING BOX
(a)
(b)
FIGURE 7.14. - Schematic of steam injection wellhead assembly showing casing and tubing
stuffing box arrangement.3
227
stuffing boxes, an upper tubing stuffing box and a lower casing stuffing box for the tubing and
casing expansion. Since this arrangement shows all the tubing weight is set on the packer, this
design is not recommended for use in steam injection wells deeper than 1,000 ft. Use of such a
design in deeper wells may cause the tube to buckle because of its own weight and result in
mechanical failure.
The wellheads used in cyclic steam wells should be equipped to handle fluid production in
addition to the free tubing and casing expansion facility. One such wellhead design is shown in
figure 7.15. This wellhead is equipped with a combination stuffing box-packoff unit that not only
allows free movement of tubing and casing through the wellhead, but also allows steam to be
injected without pulling the rods. To steam the well, the pump is lifted off' i t s seat, and the packoff
unit is tightened to provide a seal around the polished rod.5 At the end of the steaming period, the
packoff unit is loosened, the pump is seated, the stuffing box is checked for leakage, and the well
is put on production. A three-way, two-position valve is usually used in huff 'n puff operations to
permit steam injection and oil production through the same wellhead outlet. At the end of the
steaming cycle, the valve is simply turned to the production line as soon as the packoff unit is
loosened and the pump is seated.
FROM GENERATOR
PACK-OFF UNIT
228
As pointed out in a previous section, wellhead assembly must be equipped with a moveable
connection to allow for the free horizontal expansion of the tlow line and vertical expansion of
tubing and casing. This is accomplished through the use of three swivel joints (see figure 7.10).
Grease packed swivel joints such as those offered by Bwco are the most widely used swing joints
in thermal EOR fields. These joints are designed to accommodate all expansion and temperature
ranges.
Liberal use of hammer unions should be included in the wellhead assembly design to provide
for not only well service and workover, but also for future maintenance and replacement of swivel
joints and the flow rate control device.
where
W
d
229
RADIUS
(ROUND)
EDGE
bu/
CARBURIZED &
HARDENED ALLOY
DESIRED
HEX NUT
UtSiHtU
CARBIDE
INSERT
u tsintu
FLOW
DESIRED
FLOW
230
L
Pi
P2
Vi
The specific volume of steam is given by the general equation of steam properties ax
v1 = x v, + (1-x) Vf
where
v,
vf = Specific volume of the liquid phase @ Pi, ft3/lbm
Notice that as long as the upstream pressure is more than 1.72 times greater than the
downstream pressure both of the above equations apply and that neither is a function of any of the
downstream conditions (in the case of steam injection operations, the wellhead pressure).
It is particularly important when using the static choke method of rate control to make certain
that the chokes themselves are installed with the rounded radius of the choke insert facing the
incoming steam and also that the inside surface of the choke is smooth over its entire length
(usually 6 in.). Figure 7.16 illustrates several types of choke inserts available and it further
indicates those choke inserts which are considered acceptable. During the life of a steam injection
operation, it may be expected that these choke inserts will be changed several times. The inserts
should be carefully inspected with each change, and as new shipments are received, they also
should be carefully inspected to guarantee that they possess the desired round entrance and smooth
interior.
General field practice is to measure the upstream and downstream pressures across each
choke at 2-week intervals. This allows a technician to determine that the choke is or is not
operating at critical flow (subcritical flow is an indication that well work is required because the
perforations have become plugged or locked, usually by sand fill). The pressure readings also
allow a technician to calculate the current rate of delivery of steam to the well and make a choke
hsert size change to either decrease or increase the rate. This allows adjustment or trimming of the
steam system to provide the desired rate of flow to all wells i n the system. In large steamflood
projects, the injection well pressure is monitored and controlle'd with the help of computers.8
The use of static flow chokes under critical tlow is attractive for steam injection service
because of their simplicity and ease of maintenance and also because they are relatively immune to
the condition of injection wells. If injection pressures should become higher than about 250 psi, a
significant amount of power is consumed to raise generator feedwater pressure to provide the
pressure upstream of the static choke to guarantee critical flow. As ;Iresult, adjustable chokes or
23 1
control valves operating in the subcritical flow regime are often wed where injection pressures are
expected to be high.
Adjustable Chokes Operated in the Subcritical Flow Regime
In steam injection projects where the steam injection pressures are expected to exceed 250
psi, the adjustable choke, or control valve, is selected as the means of controlling the rate of steam
delivery to the injector. The principle of operation is the same as that of any control valve. The
adjustable choke is placed in the path of flow, and the variable orifice within the choke is adjusted
to sufficiently restrict flow to achieve the desired rate. The rate of flow across the choke or control
valve can then be determined using the standard control valve equations for gas or steam. The
following equation can be used to size a control valve for steam flow.
Q = 1.06 m
where
fi)
pressure upstream of the valve. But, although this equation is very useful in initially sizing the
control valve, it is nearly impossible to make use of this equation to estimate the current rate of
delivery across the valve. This difficulty arises because the exact position of the valve stem must
be known in order to determine both C, and C1. As the valve stem position is changed in an effort
to adjust steam rate, the values of both C, and C1 must both be re-evaluated.
To solve the difficulties associated with the use of the above equation, an orifice plate is
generally installed with the adjustable choke. The static and differential pressures across the static
choke are then measured, an estimate of the stearn quality is made, and the orifice plate flow
equation is used to calculate the rate of steam delivery across the orifice plate. By monitoring the
differential and static pressures across the orifice as the valve stem position of the choke is
232
adjusted, the desired rate of steam flow may be obtained. The orifice plate flow equation is
repeated as follows:
Q, = C F, F, fi fl
Rate of delivery of s t e m across oiitice plate, Ibmhr
The orifice coefficient
Correction factor based on wet steam, specific volumes
fqx
Vf
1.062 W7-F
w =
233
pressures or the cost of power necessary to raise feedwater pressure to the required system
pressure becomes unacceptable.
The Effect of Steam Quality on Injection Rate Control
To make use of either steam rate control system properly, the steam quality at the control
device should be accurately known. This results from the square root of the density term which is
present in both equations describing the rate of flow across either the choke or the control valve.
The accurate estimation of steam quality delivered to the wellhead and therefore across the control
device is by no means a simple task. The problem is made more difficult because no reliable
procedure or equipment has yet been developed to accui'ately measure the steam quality at the
wellhead. Current steam quality measurement techniques and practices are discussed in detail in
chapter 8.
In the past, field practice has been to calculate steam quality at the wellhead as the steam
quality at the steam generator site less the effects of hear loss in the distribution system. While
this makes the solution of the steam rate equations possible, it does not accurately reflect the true
quality delivered to the wellhead, and the mte of delivery of steam across the control device cannot
be accurately calculated. Current research and field testing is directed at the following:
1. A better understanding of how the steam distribution system delivering steam to the
wellhead operates in regard to steam quality.
2. Techniques to measure the steam quality at the wellhead.
Recent laboratory tests have shown that it is possible to design and operate the steam
distribution system to provide more reliable steam quality values to each wellhead.
234
quality) with steam to produce desired quality. The problem remains as to the most desirable
location for the mixing of steam and hot water.
Mixing hot water with steam results in a two-phase mixture. Phase separation will result if
this mixture is transported over long horizontal distances. Further, since steam velocities in the
distribution system usually range from 30 to 70 fdsec, the presence of liquid slug in steam can
result in line failure at tees and ells. Hence, it is preferable to mix steam and hot water at the
wellhead to avoid the liquid slug problems.
The mixing chamber is usually located at the downstream side of the steam rate control device
(usually a critical flow choke). Steam and water are introduced into the chamber through metering
devices and intimately mixed before being injected into the well. A turbine meter installed at the
water line measures the volume of water introduced into the chamber. Steam rate is measured
using the static choke.
Though this device is satisfactory for formulating steam of desired quality, the inlet steam
quality must be known accurately if it is to be effective. Since the quality of the steam entering the
mixing chambers is usually not precisely known, this limits the usefulness of this device. To
overcome this limitation, the steam is first passed through a separator located upstream of the
mixing chamber. The steam leaving the separator is then metered through a choke into the mixing
chamber and mixed with a known quantity of water. Since the steam leaving the separator would
be of 100% quality, the quality of steam entering the well can be closely controlled.
SUMMARY
A properly designed and installed steam distribution system is relatively trouble free to
operate and requires very little maintenance. The steam distribution system must be designed, not
only for the appropriate pressures and temperatures, but must also provide the desired steam
quality and flow rate to each injection well in the system. The major component of a steam
235
distribution network includes main steam headers, lateral steam lines, wellhead assembly, and
wellhead steam rate controls.
Steamheader and lateral steam lines must be designed and constructed in accordance with
recognized standards such as ANSI Standard 31.4. Expansion loops must be incorporated in
steam lines to provide for the expansion and contraction of these lines due to temperature changes.
Since temperature changes also affect the wellhead connections, they must be equipped with
stuffing boxes to allow for the free movement of tubing and casing and tubing setting with steam
temperatures. The rate of delivery of steam to each individual well is controlled using a static or
adjustable choke. The performance of the controlled devices must be checked at least once a week
to assure proper functioning. In addition, a check should be made whenever a change in the steam
distribution system pressure occurs. The integrity of the steam line insulation must be periodically
evaluated and the insulation replaced as warranted.
REFERENCES
1. Natural Gas Processors Suppliers Association. Engineering Data Book, 9th Edition.
Natural Gas Processors Association, Tulsa, OK, Section 10, 1972.
6 . Crocker, S. Piping Handbook, 4th Ed. Chapter 2, Flow of Steam in Pipes, pp. 246253. McGraw Hill Book Company, New York, 1945.
7. Chien, S. F. Critical Flow of Wet Steam Through Chokes. J. Pet. Tech., v. 42, No. 3,
March 1990, pp. 363-370.
8. Schoen, D. M. and T. L. Frazier. Operational Review of a Large-Scale Saturated Steam
Delivery System. Paper presented at the SOC.of Pet. Eng, California Regional Meeting, Ventura,
CA, Apr. 4-6, 1990. Paper SPE 20026.
236
APPENDIX 4 - A
STEAM DISTRIBUTION NETWORK COMPONENTS SU PPLIERS
A.
C.
4. Progressive Fabricators
6882 N. Broadway
St. Louis, MO 63147
Telephone: (314) 385-5477
D.
Piping Insulation
5. Hydril Company
P.O.Box 60458
3300 N. Beltway 8
Houston, TX 77032-341 1
Telephone: (7 13) 449-2000
6 . Trico Industries Inc.
B.
CHAPTER 8
STEAM QUALITY
INTRODUCTION
The term steam quality is commonly used to describe the vapor content of a two-phase steamwater mixture. It is a dimensionless quantity and defined as the inass ratio of vapor to the total
mass of liquid and vapor.1 Thus, an oilfield steam generator is said t o produce 80% quality steam,
if the produced steam contains 80% by weight steam and 20% by weight water.
A knowledge of steam quality is essential for the efficient operation of a steam injection
operation. Steam quality is one of the operator-controlled parameters that can impact the
economics of a steam injection project. Steam quality enters into the calculation of steam generator
efficiency and steamflood performance.
Steam generator output quality is required for several reasons. The output quality affects the
steam generator energy output. It is desirable to know the energy output of a generator, so that it
can be compared with the total energy injected at wells. If a reasonable energy balance cannot be
obtained, this is an indication that one or more of the meteling devices is not functioning properly
and maintenance is required. Second, steam quality from each generator is required by plant
operators to ensure that generator tiring rates are properly set and stay in adjustment. Third, it is
desirable to compare energy output from a generator with fuel consumption on a regular basis to
detect generator efficiency. Any such change is usually the result of a change in operating
conditions, which can be corrected to provide good penemtor efficiency.
The performance of a steam injection operation is commonly measured in terms of oil-steam
ratio, i.e., barrels of oil recovered per barrel of steam iiljected. The oil-steam ratio generally
improves, if the heat content of the injected steam is high. The enthalpy or total heat of steam is
strongly dependent on the steam quality, especially at lower pressures. At a given pressure a
higher quality steam carries more heat than a lower quality steam and, other things being equal, a
higher quality steam can recover more oil than a lower quality steam. Steam quality must also be
known to calculate accurately the wellhead injection rate using devices such as the critical choke.
Ideally, it would also be desirable to have a knowledge of steam quality at the bottom of the
injection string, as it enters the sandface. This would give the exact amount of energy entering the
formation and, in conjunction with the surface steam quality, would also allow an evaluation of the
effectiveness of the tubing insulation, if present.
The steam quality can be measured using a number of techniques. However, these
techniques are only approximate, since in general the steam quality varies considerably with the
operating conditions. In this chapter, some of the various methods currently available for the
measurement of steam quality have been evaluated. Only techniques of interest to steam injection
operation are reviewed. Some of the more recent techniques developed to measure steam quality at
238
wellheads in multiple well distribution systems are discussed. Limitations of each of the methods
reviewed are also given.
230
ORIFICE
ADJUSTABLE
COMPUTER
STEAM FROM
FLOW CHOKE
FIGURE 8.1. - Schematic of equipment arrangement for steam quality measurement by the vaporliquid separation technique.
To obtain reliable results, the separator chamber must be thoroughly warmed to avoid any
steam condensation. If not, dilution of the liquid phase would occur and give a lower apparent
quality. In addition, insulation should be provided generously throughout the flowing chambers to
approach an adiabatic condition (no net gain or loss of heat to the surrounding area) as much as
possible. Quality can be calculated by dividing the mass rate of vapor flow by the sum of the mass
rates of vapor and liquid streams.
Tests performed by Hodgkinson2 indicated that a separator system could provide accurate
measurements of steam quality in the range of 40 to 80%. The measured values deviate from
standard chemical titration tests (discussed later) by only about 5 to 7%.
Mahon3 has evaluated the separator sampling techniques. He recommends samples to be
taken at distances of several pipe diameters (50- 100) downstream of any elbow, tees, valves, or
other flow constrictions. It is also essential that pressure drops through the separator be kept to a
minimum to avoid flashing of the water phase. Before taking samples, the separator must be
allowed to warm up and stabilize for a period of at least 15 minutes.
240
measurement of steam quality, has been investigated by several author^.^-^ This technique is
preferred by many TEOR operators, since it does not require sampling or phase separators. The
standard equation for single-phase flow through sharp-edge orifice plates or flow nozzles is
where
Wh
Apw
V
= specific volume, cu ft/lb
C
= meter factor
The meter factor, C , is an empirically determined flow constant which depends on the size
and geometry of the orifice plate and the flow conditions. It is nonnally considered to be constant
for a range of fluid properties and flow rates. This is u reasonable assumption provided the
Reynold's number stays above 105 for all flow conditions.
For metering flow of a wet steam through the orifice, the specific volume in Eq. 1 must be
adjusted to account for the degree of wetness. The specific volume of a two-phase steam-water
mixture (assuming the mixture to be homogeneous) is given by
= (x) v,
+ (1 - x) Vf
where
= steam quality, fraction of mass in vapor phase
V,
= specific volume of saturated steam, ft3/lb
Vf
= specific volume of saturated liquid, ft3/lb
Although this correction is appropriate for very high qualities, it is known to introduce
significant errors for sharp-edged orifices as the flowing stream becomes wetter. Wilson4
X
determined that errors in mass throughput in the 1,500 to 2.000 psia range were about 5% at 80%
quality, 10% at 60% quality, and 15 to 20% at 40% quality.
James10 determined experimentally that Eq. 1 could be used directly if the specific volume
term is calculated from Eq. 2 using an apparent or pseudo-quality that is related to the true quality
by
1.5
Xapparent - Xylle
(3)
24 1
(4)
242
provides a convenient tool for this check. By chemical titration, the concentration of C1- in the
liquid part of the steam-water mixture is compared with the concentration of the same ion in the
feedwater, and, assuming good quality water samples, the results obtained can be very accurate.
This method is, at present, the most widely used technique for obtaining steam quality in the field.
The sampling technique for obtaining a specimen of the water fraction from the two-phase
flowing mixture is one of the most critical steps in the procedure. This liquid fraction sample is
obtained from separators or traps in the output line of the generator and must be drawn through a
cooler to avoid flashing. If the sample is drawn off too quickly, vapor may also be withdrawn and
condensed, with a subsequent dilution of the liquid phase, therefore implying less water
evaporation in the generator and a lower output quality. On the other hand, if liquid is withdrawn
too slowly, the liquid obtained will be some composite of the effluent liquid previously left in the
separator, also giving incorrect results.
Jon concentrations are found by adding a dilute acid to the water samples. A phenolphthalein
color indicator is used to determine when sufficient acid solution has been added. More acid
solution must be added to the outlet sample than to the feedwater sample, and the ratio of the acid
added to each sample determines the quality of the outlet sample. For example, if the effluent
sample requires an acid volume five times greater than the feedwater sample, then the final
dissolved solids concentration is five times the original, or what is the same, four-fifths of the
feedwater has been vaporized, and the steam quality is 80%.
Chemical titration techniques suffer the same limitations as all other techniques requiring
samples to be taken from the flowing two-phase pipeline; that is, the need to take a sample from a
representative cross section of the pipe and in a segment of the line located far away from any
discontinuities. Errors will be introduced into the measurement if a sample is taken very close to a
tee, elbow, or any pipe fitting.
generator. This, however, is not always the case. The effluent liquid phase contains carbonate
and hydroxyl ions produced by the thermal decomposition of bicarbonate ions in the feedwater.
The hydroxyl ions in the effluent liquid provide a conductivity out of proportion to the concentrated
total dissolved solids and when this conductivity is used directly, the calculated quality is too high.
To avoid this situation, both input and effluent liquid samples must be neutl-alized to a pH of 7 with
a weak acid. Over- or underneutralization of either sample can lead to erroneously calculated steam
qualities. Secondly, because conductivity varies substantially as a function of temperature, it is
necessary to correct conductivities of feedwater and effluent samples to a common temperature
base if they have not been measured at the same temperature originally.
flowing mixture. However, the quality obtained from this method is only applicable to the outlet
of the steam generator. It cannot be used at either the wellhead or downhole locations.
Even when intended for measurement of steam quality at the generator outlet, this method is
not in standard use because it is error prone, more time-consuming, just as involved as the
chemical titration method, and yields less accurate values.
245
sample to mix with the absorbing medium. Then, the temperature rise of the absorbing medium is
measured, and from this measurement the enthalpy of the steam-water sample is calculated.
Knowing the enthalpy of the steam-water mixture, its quality can be obtained by comparison with
the values for saturated liquid and vapor at the tlowline pressure. The absorbing medium with
known thermal properties must be selected and its flow rate measured accurately. The rate of flow
of the steam-water mixture must also be measured and this can be done by noting the amount of
condensate if the indirect technique (with a heating coil) is used.
Although available, this technique is not in common use because calibration of the enthalpy
transfer chamber requires very precise measurement, and the criteria assuring simple design and
ease of operation have not been met.
246
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
26
27
28
29
30
31
Flow rate,
bbl/day
425.3
453.9
19.1
600.7
727.4
301.4
398.6
247.2
283.5
365.1
530.1
188.3
379.4
100.3
293.8
25 1.3
507.5
486.3
278.7
113.0
478.8
276.0
256.1
226.7
432.2
333.5
395.9
308.9
25 1.3
397.9
Steam quality,
%
62.1
78.1
100.0
4.2
6.3
52.7
53.4
100.0
81.6
88.3
43.2
66.5
100.0
95.1
93.4
68.3
48.5
85.7
89.9
95.1
64.6
72.7
89.0
93.3
56.2
73.7
63.8
91.3
95.6
95.8
Vibrating Densitometer1
The vibrating densitometer operates on the principle that the natural or resonant frequency of
a vibrating tube containing a two-phase mixture depends on the bulk density of the two-phase
mixture. Hence, by measuring the resonant frequency, the bulk density of the two-phase mixture
can be calculated. The bulk density, along with the pure vapor and liquid density at the operating
temperature, can be utilized to calculate the steam quality.
Figure 8.2 shows a schematic of a vibrating densitometer which consists of a pair of sensor
tubes made of magnetic material and are vibrated by an electromagnet. When the frequency of
vibration of the tubes corresponds to its natural or resonant frequency, a large amplitude
247
UPPER
MANIFO
R.T.D.
LJ
TUBE
FLOW
p =(A/f2)+B
whereP
-1
x = -1-Vf /(Vg-Vf)
[P
(9)
where
x
= Steamquality
p
= Bulk density of two-phase mixture (Ibm/ft3)
vf = Specific volume of liquid (ft3/lbm)
= Specific volume of vapor (ft3/lbm)
Vibrating densitometer technique is still being evaluated. Its accuracy, reliability and ease of
operation are not known at this time.
SOURCE
TRANSMITTER
I
STANDARD
TWISTED PAIR
249
The gamma steam transmitter calculates steam quality from data supplied by the density
sensor and by a downstream pressure transmitter and outputs to an indicator.
To obtain reliable steam quality, the system must be calibrated over a wide range of
conditions and the calibrations must be checked periodically. The system is drift prone and hence
accuracy suffers.
The void fraction, a,is the volume occupied by the gas-phase in a wet steam mixture. The
steam quality, x, which is defined as the ratio of the mass tlow rate of the gas-phase steam to the
mass flow rate of the wet steam, is related to ct by
a=
s(1-x)
1 +XP1
where s is the slip ratio. The slip ratio is defined as the ratio of average gas-phase velocity to
average liquid-phase velocity. Hence, if the slip ratio and pressure are known, the steam quality
can be determined from Eqs. 10 and 11. However. the slip ratio, s, depends on pressure, mass
flux, steam quality, flow pattern, and orientation of flow. These relationships are not well
established and, therefore, calibration of the system over a wide range of conditions must be
undertaken. The principle of operation of a neutron densitometer is shown schematically in
figure 8.4.
The neutron densitometer steam quality measurement system is shown schematically in
figure 8.5. To measure steam quality accurately, the meter must be calibrated. The system is
usually calibrated with the steam pipe empty and with the pipe full of water. Measurement of
250
FAST
NEUTRON
SOURCE
STEAM-W AT ER
MIXTURE
THERMAL
NEUTRONS
TRANSMITTED
NEUTRONS
---+,I,l
NEUTRON DETECTOR
MODERATOR
SHIELD
PIPE
utron densitometer. 15
PROCESSOR
WELLHEAD
FIGURE 8.5 - Arrangements for steam quality measurement using a neutron densitometer.16
25 1
steam quality also requires knowledge of steam pressure and an approximate mass flow rate. The
mass flow rate can be determined by measuring the pressure drop across a flow nozzle. The
flowing steam pressure is determined using a pressure transducer. The steam quality is determined
using the information provided by the neutron densitometer and flow nozzles. A data acquisition
system interfaced with an on-line computer is utilized to calculate and display steam quality, mass
flow rate, and enthalpy delivery rate to the well. Further details on the system can be found in
references 15 and 16.
The system is still in developmental stage and has been field tested to a limited extent in
Canada. No units have been installed to date in the United States.
SUMMARY
Accurate measurement of steam quality in an oilfield environment is both cumbersome and
difficult because steam distribution through manifolds, laterals, and tees results in uneven splitting,
and no easy way exists to determine steam quality with any degree of confidence. Some of the
techniques currently used in oilfields to measure surface steam quality include the orifice plate
method, dissolved solids technique, liquid-vapor separator method, enthalpy determination
method, and stack gas analysis technique. All of these techniques have limitations and do not lend
themselves to continuous, real time measurement of steam quality. Newer techniques, such as
those based on the vibration densitometer and theimal neutron densitometer are promising but are
expensive and require extensive calibration. No technique is available to measure steam quality at
the sandface. At the present time, downhole steam quality is infeired from wellhead steam quality
measurements. A list of vendors that can assist in steam quality measurement are shown in
Appendix 8-A.
REFERENCES
1 . Strome, R. K. and D. M. McStravick. Continuous Steam Quality Measurement in a
Steam Distribution System Quality Measurement. Presented at the SOC.of Pet. Eng., California
Regional Meeting, Ventura, California, Apr. 8- 10, 1987. SPE paper 16348.
252
5. Wilson, T. M. Steam Quality and Metering. The Journal of Canadian Pet. Tech., v. 15,
No. 2, April-June, 1976, pp. 33-38.
6. Mattar, L. Nicholson, M., K. Aziz, and E. A. Gregory. Orifice Metering of Two-Phase
flows. Journal of Pet. Tech., v. 31, No. 8, August 1979, pp. 955-961.
7. Bennett, M. D. and K. Martin. Calibration of a Small Ventuiimeter for Wet Steam at
Low Reynolds Number. Proc. Instn. Mech. Eng., London, v. 184, Pt. 3C, 1970, pp. 151-153.
13. Jones, J. and R. L. Williams. A Two-Phase Flow-Splitting Device That Works. Pres.
at the SOC.of Pet. Eng., International Thermal Operations Symposium, Bakersfield, California,
Feb. 7-8, 1991, SPE paper 21532.
14. Kay-Ray Inc. Steam-Quality Measurement System. Tech. Bulletin KR484-789, 1986.
Kay-Ray Inc. Division of Rosemont, Mt. Prospect, Illinois 60056.
15. Woiceshyn, G. E., P. S. Yuen, H. John and J. J. Manzano-Ruis. Measurement of
Steam Quality, Mass Flow Rate and Enthalpy Delivery Rate Using Combined Neutron
Densitometer and Nozzle. Pres. at the Fifth SPE/DOE Symp. on Enhanced Oil Recovery of the
SOC.of Pet. Eng., Tulsa, Oklahoma, Apr. 20-23, 10x6. SPE/DOE paper 14907.
16. Wan, P. T. Measurement of Steam Quality Using a Neutron Densitometer. Pres. at the
Pet. SOC.of CIM and Soc. of Pet. Eng. Joint International Tech. Meeting, Calgary, Canada, June
10-13, 1990. Paper CIM/SPE 90-88.
253
Sandia National
254
APPENDIX 8-A
STEAM OUALITY MEASUREMENT SYSTEM VENQORS
A.
0-1
'fice Plate Vendors
1. Lambda Square Inc.
P.O. Box 1119
Bay Shore, NY 11706
Telephone: (516) 587-1000
Houston, TX 77255
Telephone: (7 13) 686-3444
255
CHAPTER 9
THERMAL WELL COMPLETION PRACTICES
INTRODUCTION
Thermal well completions, while basically the same as those for conventional wells, differ in
mechanical aspects. The main difference between completing a steam injection well and a
conventional well is the handling of the problems associated with elevated temperatures. High
temperatures cause tubular goods in an injection well to expand. If the injection or production of
hot fluids is interrupted for some reason such as mechanical failure of a steam generator or a well
workover to correct an unexpected downhole problem, the well bore temperature will fall causing
the tubing to contract. The heating and cooling of the tubular goods lead to severe thermal
stresses. If these stresses are allowed to exceed the design stress of the tubular goods, casing and
tubing failure may result. In addition, the effect of temperature on downhole completion
equipment must be considered in equipment selection.
Movement of sand particles into a production wellbore from semiconsolidated or
unconsolidated sands, as fluid is produced, can be a major production problem in thermal wells.
The sand production mechanism can be extremely complex and is influenced by all completion
operations ranging from first bit penetration to the start of production. The problems of sand
control in thermal recovery wells are especially troublesome because of high production rates of
reservoir fluids that result from the high volume of fluid injected in these processes. The greatest
single problem in steam injection operation is that of making lasting, high productivity well
completions resistant to erosion and corrosion at elevated temperatures.
Thus, thermal well completions must be designed to withstand elevated temperatures without
damaging to wellbore equipment. In brief, the severe service seen by thermal wells calls for
specific thermal completion designs. This chapter describes and recommends certain well
completion practices currently used in the industiy .
Provided in Appendix 9-A is a list of vendors that deal in thermal well completion tools and
services. This list is not intended to be the recommended vendors; it is included here so that a
prospective operator can contact these or other vendors for additional information.
CASING DESIGN
256
250 to 3,200 psi. Most of these failures have occurred at the couplings. Examination of failed
casings show extensive compression damage to the joints in the form of telescoping or buckling to
the point of collapse or breaking.1
It is important that the temperatures effect on the yield strength, ultimate tensile strength, and
modulm of elasticity of casing materials be considered piior to preparing specifications for casing.
Quantitatively, casing will expand or contract with temperature changes in an amount determined
by the following expression,
(1)
AL = U T L
a
AL
AT
= change in temperatuiE, O F
L
= length of the casing, ft.
An excellent rule of thumb is that the unstressed casing will elongate 0.83 inch per 100 ft of length
per 100" F increase in temperature. Thus, a 2.000-ft tubing string subjected to a 500' F
temperature will elongate about 83 in. or almost 7 ft. The casing will also contract during a cooling
period.
If the shoe of the casing is cemented and the top is free to expand, the casing will expand out
of the ground, thus relieving casing stress. If the casing is not free and the top is securely fastened
at the bradenhead, changes in temperature will set up thermal stress in the casing. This stress will
result in the slackening of the tension generally transmitted and concentrated in the lowermost free
section of the string in the form of compression buckling.* A good rule of thumb for estimating
thermal compressive forces in tubular goods is that the compression changes by about 207 psi per
O F change in temperature.3
Thermal stress, if it exceeds the elastic limit of the tubular goods, will cause both tubing and
casing to buckle helically, like a coil of spring. The magnitude of the stress depends largely on the
moment of inertia of the pipe and the radial clearance between pipe wall and any straightening
support.3 The support structure for casing is the open hole; for tubing, the supporting structure is
the casing. In all but very shallow wells, hole deviations and formation sloughing will cause
compressive stresses to build up at intermittent joints and may cause heated casing and tubing to
fail.
An obvious way to avoid compressive casing failure is to ensure that the compressive
stresses do not exceed the yield strength of the casing material. In new wells, this can be achieved
by employing an appropriate grade of casing; by cementing the casing under tension; and by
controlling casing temperature through the use of therinal packers.
257
160
140
120
100
80
400
200
600
800
1000
TEMPERATURE. '' f
Willhite propose that casing failure can be avoided by selecting a casing strong enough to avoid
yielding in compression at the maximum expected temperature. Holliday6 noting that the tensile
strength of steel is at a maximum of 500' F (see figure 9.1) proposes that one need not be unduly
concerned about casing stress exceeding its yield point at steam temperature. However, casing
strength must be high enough to avoid tension fiilure upon subsequent cooling. A review of
reported casing failure indicates that casings generally fail i n tension, than in compre~sion.~
Hence, compressive casing stresses are normally tolerable. Holliday's criteria permit much higher
allowable casing temperatures and steam pressures than those of Willhite or Gates. The
recommended maximum casing temperature for various grades of casing reported by these authors
are shown in table 9.1
TABLE 9.1 - Recommended operating temperature range for thermal well tubular goods
Pcrinissible s u r f x e stem temperature.
Casing
(MI Grade)
Willhite
Gates
H-40 270-300
J-55
N-80 350-400
270-330
240-215
470-580
350
350-410
595
620-730
P-1lo
--
258
O F
Holliday
450
750
Table 9.2 lists physical properties of four grades of seamless tubing and casing at various
temperature^.^ The ultimate tensile strength of these steels decreases slightly up to 300" F,
increases to a maximum at 500' F, then rapidly decreases with temperature as shown in figure 9.1.
The yield strengths and elastic limits are shown as function of temperature in figures 9.2 and 9.3,
respectively.
In some shallow reservoirs, steam is injected directly down the casing. These wells are
usually less than 600 ft deep, so the steam temperatures are low and casing failure is unlikely. In
deep wells, steam is rarely injected down the casing. The probability of casing failure in deep
wells is much higher due to higher steam temperatures and pressures. In deep steam injection
TABLE 9.2. - Tensile properties of various grades of seamless casing and tubing4
~
Tensile Properties1
Grade
Condition
Outside
diameter,
in.
Wall,
thickness
In.
Tensile
test
temp .
OF
Y leld
strength,
Teiisile
strength,
Elastic
limit,
PSI
PSI
PSI
Modulus
of
elasticity
1,000
PSI
~~
J55
Hot
rolled
13-518
0.313
RT
3 00
5 00
7 00
9 00
59.320
55.820
57 $870
53,130
44,44 5
107,950
100,040
122.125
98.490
68,030
52,350
50,750
42,050
37,925
28,800
25.750
24.650
23,800
21,550
24,000
N80
Normalized
13-318
0.500
RT
300
5 00
7 00
83,750
80.1 5 0
83.970
74.415
124,050
120.675
133.500
116,775
69,550
65.9850
59.850
51.900
27,900
25,400
21,600
16.300
5) 00
62.715
85,000
40.850
15.700
RT
3 00
5 00
7 00
900
106.800
102.100
104.850
136.275
131.600
139,100
78.050
73,500
70,950
27,600
26,500
26,200
95,900
86,640
129,100
107,375
678,500
59,500
25,200
18,700
RT
128,825
114,850
110,025
100,310
78.545
145,500
139,100
146.050
124,650
92.145
120,200
95,000
86,750
76,800
51.400
27,900
27,300
25,500
24,100
19.200
P 105 Normalized
10-7/8
0.375
& tempered
P 1 10 Quenched
10-314
& tempered
0.547
3 00
5 00
700
900
tests.
259
140
120
100
80
60
40
-" I
311
200
400
600
800
TEMPERATURE.
1000
140
'g
120
Q
i
0
100
u)
n
E!
>
80
60
40
0
J55
v
r\
"
200
400
600
800
1000
TEMPERATURE. O F
260
wells, steam is rarely, if ever, injected down the casing for the fear of risking casing failure due to
thermal stresses. In deep wells, casing temperatures can be maintained below the injected steam
temperature by injecting the steam down the tubing and by preventing the steam from entering the
tubing-casing annulus through the use of thermal packers.8
26 1
Field Operator
ft
Depth,
Psig
Max
Pressure,
in.
Diameter
Grade
Type
type
Coupling
~~
Kern River
Chevron
7 00
350
7.0
K-55
STC
Midway-Sunset
Chevron
1,200
450
7.0
K-55
STC
Mt. Poso
Shell
1 .goo
600
8 SI8
K-55
BUTT
San Ardo
Texaco
2.350
800
8 SIX
K-55
STC
Guadalupe
Union
3,000
1,800
7 .O
N-80
BUll-
completed using K-55 grade casings with buttress threads and landed in tension. Canadian
operators generally use N-80 grade casing because of the higher injection pressures and
temperatures. Casing grades should be selected based on expected operating conditions and
anticipated thermal stresses. Since most field experience shows that casing failures occur at a
coupling, use of buttress or other premium couplings is recommended. However, this will
increase the initial costs of completion. In shallow wells, where the operating pressures and
temperatures are lower, the standard short threads and couplings (STC) may be used without the
risk of failure. Many operators in the shallow Kern River (CA) fields, still uses the less expensive
short threads and couplings. Hence, the steam injection completion practice for a given field
should be a compromise between the initial completion costs and the cost of risking failure of the
well. Engineering judgement plays a large role in arriving at a safe completion design. Since
failures usually occur at the joint, the engineer should rate the joint first. apply a safety factor, and
then select the casing based on anticipated maximum casing temperature.
Casings are usually landed in tension after prestress to protect the casing from elevated
temperature. Studies indicate that a prestressed casing is less subjected to buckling than a
conventionally run string. By placing high tensile stress initially, it is possible to absorb a higher
temperature increase and still remain below the yield point in compression.10 For example, if the
safe operating temperature limit for a stress free N-80 casing were set at 500' F, this limit can be
increased by another 100' F, by stressing the casing. The higher operating temperature allows for
higher injection pressure, since the saturated steam pressure increases from 68 1 psig at 500" F to
1,543 psia at 600' F.
Casings are normally prestressed while cementing. The casing is run and cemented through
a shoe with a small volume of quick setting thermal cement. The shoe is anchored to about 200 ft
262
to the bottom to facilitate the stretching operation. lo The rig is moved off and casing jacks stress
the casing while the remainder of the cement sets up. In this way, as the casing heats up, it merely
relieves the tensile forces already in the structure, and thermal elongation is eliminated.
The amount of prestress is a function of steel grade, coupling strength, mud weight, and
collapse resistance. To prevent crushing when the load is distributed over the slip area, the
prestressed casing must be landed on a reinforced cellar designed for the load. In the past, a
two-stage cement job was used to stress the casing. Only a few operators currently use this
technique. Though the two-stage technique is less expensive (about $5,000 per job), the operator
risks casing failure when the overburden sloughs in and holds the pipe or when the tail section
does not hold.11
263
diameter with 7 in. common in deeper wells. California regulation requires a packer in continuous
steam injection wells to protect the casing. Expansion joints are needed to accommodate the
differential expansion and contraction of the tubing when heated or cooled. Expansion joints are
usually placed at the packer. Thermal packers are discussed more fully in a later section. This
section describes certain thermal well completion practices used in the industry. It should be
noted, however, that well completion practices for steam injection can and do vary from field to
field and operator to operator in specifics, depending on the circumstances, and no generalization is
possible. Examples of typical steam injection, production, and observation well completion
features are shown in figures 9.4, 9.5 and 9.6, respectively. The discussions in the following
section, however, are generic in nature and should be of assistance to the engineer designing a
thermal well completion program.
Drilline Fluids
Both productivity and injectivity can be adversely affected by washouts through the
production interval and by reduction in permeability due to fresh water invasion into the formation
and resultant swelling of water sensitive clays. Drilling fluids, therefore, have to be inhibited
against clay hydration, generally with 1 to 10% brine or KC1 solutions andor other clay stabilizing
compounds, such as hydroxy-aluminum.
CHECK VALVE
GATE VALVE
GATE VALVE
STEAM FROM
STEAM GENERATOR
~1 I,
fl I
r HCASING
~EMENTTOSURFACE
)J.
264
CONTROL VALVE
ORIFICE WELL TESTOR
COOLING WATER
200 +I-
TOPOFSAND
7 N-80 CASING
CEMENT TO SURFACE
WELASS G CEMENT
TREATED WITH 40%
SILICA FLOUR 8 2%
CALCIUM CHLORIDE
3/4'SUCKER RODS
GRAVEL PACKED
W/ 3/16' (APP 4-6 MESH)
D.H. PUMP
5' +/- WIO
CONTACT
I O L C CONDUCTOR
7' K 55 CISING
e7m
no=
2 3'8'
TUBING
CLASS%' W S F 39.CaClz
1 5% R X l (RETARDER)
2'. 321 STAlNtESS STEEL
12 THERMOCOUPLE PAIRS
Cementing
Poor cementing techniques could cause channeling and may contribute to excessive water or
gas production in certain wells. Formation damage from cement particle invasion into the
production zone could be a severe problem in certain enhanced recovery wells. For this reason,
open-hole completions or a cement "puddling" technique, designed to minimize column height or
head while cementing casing through the production zone, is sometimes recommended. In most
areas, casing is set through and cemented with high silica flour cement. For an open-hole
completion operation, the well is drilled to gauge to the top of the producing formation and then
cemented in accordance with the best available cementing practices: Le., the cement should possess
low-fluid loss, and the casing string should be centralized, equipped with scratchers, and rotated or
reciprocated during the cementing operation. A high-silica tlour cement is recommended for all
zones experiencing high temperature.
During the cement "puddling" operation, cement with proper water loss and setting time
additives is spotted in open hole through the production interval while pumping through the drill
string equipped with scratchers. The drill string is usually rotated and withdrawn slowly through
the production zone during the cementing operation to remove, or scratch, mud cake from the side
of the wellbore and to prevent the formation of pockets of mud in the cement. After the cement has
been "puddled in," the drill string is removed from the well; centralized casing is run to the top of
the cement pool and then slowly through the cement to the bottom of the well. The cement is then
permitted to set prior to perforating. Thermal well cementing practices are discussed in more detail
in chapter 10.
Perforating
Jet perforating is a piercing and pulverizing process that creates a tlow path through the
casing, cement sheath, and into the formation. However, cement, copper, lead and carbon from
the jet charge can penetrate and mix with crushed formation material. sometimes causing significant
perforation damage and restrictions on productivity and injectivity.
Perforations should be gravel packed with properly sized sand to prevent the "tunnel" from
being filled with formation fines when the well is placed on production. If gravel is not sized
properly, sand will enter the perforated area, become trapped in the gravel, and cause a
prohibitively high pressure drop across the peiforations.
Completion fluids flowing across the interval at the time of casing perforation should be free
of solids because they can be driven into the peiforation by the force of the charge. Perforation
debris removal methods include underbalance, backflow, perforation washing, acid stimulation,
and swabbing. Underbalance, backflow, and perforation washing, for the most part, are effective
remedial measures, but are not always 100% successful. Acid stimulation under certain conditions
could be responsible for reduced permeability and compressive strength of the formation,
266
Screens
Slotted liners, wire wrapped screens and pre-packs, have been used effectively in certain
areas to control sand production. However, it is not recommended that the well be permitted to
create its own pack around a screen placed across the producing interval. This procedure will
generally result in a disturbance of the formation, possibly causing excessive fine sand production
267
or plugging before the screen or pack stabilizes. All screens, therefore, should be used in
conjunction with gravel packs to retain gravel packing material which, in turn, supports the
formation sand.
The all-welded, wrapped-on-pipe screen has the highest flow capacity. Wire and pipe are
normally 304 stainless steel, but for extremely corrosive atmospheres special alloys such as
Inconel 625 and Incoloy 825 are available. Some laboratory corrosion tests have shown that
Inconel 625 has a better high-temperature corrosion and oxidation resistance than Incoloy 825.
A slot or wire width opening of two times the 10 percentile diameter of the formation sand
has been proved effective in areas containing sand with broad size distribution. In areas with
relatively uniform sand size, slots equal to the 10-15 percentile formation sand diameter have been
used satisfactorily. The self cleaning, keystone type slot or wire opening is recommended over
that of parallel sided slots in which particles 1/2 the slot width tend to wedge and plug.
Pre-Packs
Pre-packs are designed to bridge fine formation sands without the need for gravel packing.
They are fabricated with an outer cylinder of either unconsolidated or resin-consolidated sand
which is supported by a slotted mandrel. Since pre-packs necessarily must be somewhat smaller
than the wellbore into which they are positioned, the formation may subsequently shift and cause
productivity impairment. Pre-packs may plug when (1) the formation contains large quantities of
clay, (2) asphaltenes or paraffins are precipitated by viscous, low gravity crude oils, (3) emulsions
are formed, and (4) carbonate or sulfate scale is deposited. They offer the following advantages:
(1) large surface area for filtering, (2) compatible with most well fluids, (3) high permeability (50
to 100 darcies), (4) high compressive strength (5,000 psi at 150" F), (5) can be run on a wire line
or tubing, and ( 6 )withstand temperatures to 500" F.
To reduce the chances of formation movement around the pre-pack and the risk of plugging
with formation fines or clay, this device should be used in conjunction with underreaming and
conventional or consolidated sandpacks.
Open-Mole Gravel Packing
Open-hole gravel packs should provide higher productivity than inside gravel packs because
restrictive casing perforations are eliminated, and the underreamed wellbore improves radial flow
into the well. Open-hole gravel packing are preferred over other completion techniques in stratas
of clay and shale since the gravel should prevent migration of the shale or clay to the screen and
sealing it off.
Gravel packing consists of underreaming the wellbore 6 to 12 in. larger than the casing
diameter; running and hanging the wire wrapped liners; packing with gravel using a viscous, highgravel concentrated (15 lb/gal) slurry at low circulation rates to prevent intermixing of the sandpack
with formation sand; washing to remove gravel bridges; and then repacking, if necessary. During
268
the foregoing operations, a nondamaging completion tluid should be used as suggested earlier.
Also, preflushes of clay stabilizing material should be used prior to both underreaming and gravel
packing as will be discussed later. In addition, the hole should be enlarged slowly to prevent
corkscrewing; a caliper log should always be run in the underreamed hole to determine if sloughing
has occurred; and debris should be removed from the well prior to running the liner.
The ratio of gravel size to formation sand size should be estimated from Karpoff s rule which
states that gravel size should range from 5 to 10 times the 50 percentile or median formation sand
diameter. Sand movement should be properly restrained under these conditions since the gravel
pore openings, theoretically, are properly sized relative to the sand particle diameter.
Effectiveness of gravel packs can sometimes be determined by radioactive tracer methods.
Gravel packs in open hole injection wells sometimes tend to slump soon after the start of injection,
and logging may indicate that repacking is required.
Consolidated Pack
The consolidated pack is a slurry consisting of a carrier fluid, resin (or other type of bonding
agent), coupling agent, sand, and catalyst, which is blended at the surface and pumped behind a
liner or in open hole where the resin or bonding agent is permitted to cure. This process is
designed to achieve gravel packing and sand consolidation in a single injection step. If a liner is
used, any consolidated material remaining in the wellbore is simply drilled out. In most open hole
completions, no liner is used and the consolidated pack is redrilled through the entire producing
zone with a regular bit. All the same rules and precautions associated with conventional open-hole
gravel packing and drilling also apply to this particular technique.
The main advantage of the consolidated pack treatment is that consolidation of the sand is
assured in all areas where the sand laden slurry can be pressure packed. The pack possesses
excellent strength (5,000 psi) and permeability (60 darcies) and is stable in most well environments
up to 500' F. The process works well with clay bearing formations, and excess resin can be used
to consolidate a portion of the formation behind the pack to stabilize both sand and clay when a
special clay stabilized compound, such as Halliburton's Claylok or Claysta, is injected ahead of the
resin.
Cased -Hole Com d eti on
Cased-hole completions in steam injection operations generally are recommended only for
injection wells although they may be used for producers under certain circumstances. If it is found
that open-hole completion cannot be satisfactorily performed, or is not desired for some reason,
then setting the casing with the best cementing technology should be implemented along with the
best available well completion, peiforating, gruvel packing. and/or chemical sand consolidation
methods.
269
Clap Stabilization
Serious reductions in permeability of the formation can occur when clay minerals restrict
flow by either expanding to fill the pore volume or dispersing and lodging in restrictions. The
expanding force acting on clays is caused by the affinity of exchangeable interlayer cations and
surfaces for fresh water. The inherent negative charge on almost all clay minerals is neutralized by
adsorption of cations, producing a positive suiface charge with a negative inner charge. These clay
particles with an electric double layer repel each other and, thus, disperse.
Chemicals are now available which stabilize many water-sensitive formations. One such
compound, a positively charged hydroxy-aluminum ion which is prepared in the field by reacting
aluminum chloride and sodium hydroxide in a high shear device, is absorbed and polymerized on
the negatively charged clay mineral surfuce to form a stabilizing network that prevents further
migration. This chemical, Claylok, a trademark of Chevron Research Company, has been
successfully used to improve permeability following acid treatments, stabilize formation fines,
desensitize clay bearing formations prior to steam injection, and reduce formation damage during
perforating and gravel packing. Most field brines and fresh water will not remove the treatment,
but mixtures having pH values below 4 or above 9 will damage its effectiveness.
Halliburton has a process that uses an organic, oil-soluble chemical, trade name SFD-3, to
shrink water-swollen clays and then surround them with a hard furan-type resin which increases
permeability and stabilizes clay in some dirty formations. Another Halliburton product, Claysta, is
organic in nature and stabilizes clay in a manner similar to Claylok. but has the added advantage
that it is stable to subsequent acid or caustic treatments.
It appears that these clay stabilizing compounds have possible application in open hole as
well as in cased hold completions. Injection of these chemicals prior to underreaming and gravel
packing may eliminate or significantly reduce formation damage caused by clay swelling.
Incorporating these inhibitive chemicals into the gravel pack slurry, drilling fluid, and/or other
completion fluids may also prove to be advantageous. All possibilities should be evaluated in the
laboratory before implementation.
270
*
**
***
27 1
B. Case 2
Assumption
1. Well has not been previously steamed.
2. Sand production problems.
3. Formation sensitive to fresh water.
4. Well needs to be gravel packed or re-gravel packed.
General Procedure
1. Bail out, clean up well, and acidize, if necessary (Steps 1-4 in Case 1).
2. If well is perforated, shoot additional perforations (8 shots per ft) under 2% KCl
solution.
3. Install gravel pack in hydroxy aluminum while keeping well full with hydroxy
aluminum. Pack fluid should incorporate hydroxy aluminum, or 2% KCI solution
when possible.
4. If open hole, remove old liner, install new liner, and gravel pack in hydroxy
aluminum.
c. casd
Assumptions
1 . Well has been previously steamed.
2. Sand production problems
3. Clay and sand migration are problems.
4. Formation sensitive to fresh water.
General Procedure
1. Bail out and clean up well.
2. Cool well with filtered (1 micron) 2% KC1 solution until temperature of formation
around wellbore is less than 200" F.* Twenty-five barrels of cooling water per
perforation is usually a sufficient quantity for cooling.
*Temperatures greater than 200 F will cause aluminum chloride to precipitate in formation.
Screening
Certain preliminary tests should be run in the laboratory to determine which completion
technique is most likely to be successful. The following is an outline of proposed tests which
should be run for the purpose of designing the best or optimum well completion:
1. Permeability, porosity, and water sensitivity.
2. Clay content and types of clays present, if any,
272
3 . Grain size distribution - to be used for the proper selection of slotted liner size and/or
conventional or consolidated grain size.
4. Effect of steam condensate - permeabilities before and after injection.
5 . Fluid penetration studies - determine depth of fluid penetration in core and resultant
damage from conventional mud, extremely low water loss mud, invert emulsion, chalk
emulsion and fluids containing bridging material of calcium carbonate, and oil soluble
resins.
6 . Degree of clay or fine sand migration in core samples duiing fluid or steam injection.
7 . Effectiveness of clay stabilizing compounds, such as Claylok, Claysta, and SFD-3-furan
resin combination, in native cores.
8. Compatibility of clay stabilizing compounds i n presence of and as a part of well
completion fluids.
9 . Chemical consolidation and compressive strength of consolidated native cores and
compare with clean sand chemical consolidation.
Procedures
In the following section, the drilling and completion procedure followed by a California
steam injection operator is outlined.14 These procedures detail the steps involved in the drilling
and completion of steam injection and production wells. These procedures are meant only to be a
guide for the engineer in designing such well completions.
1. Steam Injection Well
a. Drill the well with "best" low-water loss drilling mud using current drilling practices
for that area.
b. Run 2-4 joints (depending on thickness of reservoir) of casing and set the casing with
a high-silica flour cement, using centralizers and scratchers, while rotating or
reciprocating the casing to remove mud cake and pockets o f mud from annulus during
the cementing operation. A synthetic lubricant should always be used on threads
while making up casing.
c. Plug back well with 20-40 mesh sand and cement plug or standard bridge plug to
depth 1-2 ft below intended perforation.
d . Install BOP. Pressure test casing to 3.000 psi with fresh water.
e. Run neutron, gamma ray, CCL, and other desired well surveys.
f. Run open-ended 2-7/8 in. O.D. EUE tubing,* equipped with casing scraper, to plugback and reverse-circulate well with 10 bbl diesel fuel followed by 2% caustic and
1% detergent solution until returns are clean. Displace cleaning solution with fresh
27 3
j.
k.
1.
m.
0.
p. Steam injection will continue over the life of the prc?ject at viirious increased rates,
depending on the size of the steamtlood.
*
**
2. Production Well
a. For liner type completions, drill to the top of the producing formation with standard
drilling mud using current drilling practices for that area. For "set through"
completions, skip steps a, c, and e.
b. Set the casing with standard low-water loss cement, using centralizers and scratchers,
while rotating or reciprocating the casing to remove mud cake and pockets of mud
from annulus during the cementing operation.
274
C.
d.
e.
f.
g.
h.
1.
j.
k.
1.
m.
n.
0.
P.
9.
r.
Drill through the production sand with the "best" completion fluid as determined from
laboratory screening tests. Clay stabilizing agents should be incorporated into the
completion fluid, if needed.
If required, inject clay stabilizer through entire production interval.
Underream formation 6-12 in. using "best" completion fluid to circulate out cuttings.
Run caliper log and other desired logs with "best" completion fluid.
Run properly designed prefaced liner or wire wrapped perforated pipe and perform
tailored gravel pack or consolidation pack with clay stabilized completion fluid, or
Perform properly tailored open hole consolidation pack with clay stabilized
completion fluid. Permit bonding agent to cure sufficiently, and drill through with
standard bit, using best completion fluid. A slotted liner used only to properly guide
the bit could be incorporated into this process.
Lower tubing to bottom and reverse circulate, and clean with 4% KCI solution.
Set retrievable bridge plug, install steam quality expansion type casing head, install
BOP, and recover bridge plug. In place of expansion type casing head, a telescoping
union or expansion joint could be run above packer if packer is used. If no packer is
used, no expansion joints are required; however, downhole cooling may be required
during production of hot well fluids.
Run 2-7/8-in. O.D. tubing, packer (if required), 2 joints special alioy tubing,
expansion joint (if required), and seating nipple.
If packer is used, space out to set packer inside special alloy casing at least 20 ft
above pay zone, gravel pack, or consolidated pack. If no packer is used, position
bottom of tubing at top of pay. Tubing should be lined with straight phenolic
coating. All materials below packer and including packer should be fabricated from
special alloy material. If no packer is used, the bottom 8 joints of tubing should be
special alloy.
If packer is used, displace annulus fluid with salt water containing 80-100 ppm cobalt
catalyzed sodium sulfide oxygen scavenger and Tretolite KW 12 or Visco 938 film
coating chemical at 0.5% by volume.
Set packer; remove BOP and install wellhead.
Run rods and pump with coupon rack on-bottom.
Test well. When itis evident that well will flow, recover rods, pump, and coupon
rack.
Set coupon rack in seating nipple and swab well if necessary.
After well has cleaned up and stabilizes, conduct static BHP and BHT surveys.
27 5
DOWNHOLE E Q U I P M E N T
After successful drilling and completion of the wells, the next important step is to install a set
of downhole equipment to minimize thermal stresses on casing and improve steamflood
(stimulation) efficiency. Downhole equipment includes thermal packers and expansion joint and
steam deflectors.
Thermul Puckers
The injection tubing string should be equipped with a thermal packer. The thermal packer
provides a seal between the injection tubing and the inside of the casing, closing the annulus to
steam injection. In the case of an open hole completion, packers seal against the formation. Use
of packer with an expansion joint will: (1) reduce thermal stress, (2) reduce primary cement
failures, (3) reduce wellbore heat loss to the casing and formation, and (4) prevent the casing from
bursting if injection pressures become too high. A thermal packer will normally reduce the average
temperature of the casing and cement sheath by approximately 120' F and reduce thermal stress by
approximately 20,000 psi. To obtain the full benefits of the packer, the annulus must be vented
early in the heating period to remove any steam formed by evaporating water in the annulus. Not
all steam injection wells are completed with thermal packers. Many cyclic steam injection wells
and a few steamflood injectors in California are completed without packer to reduce well servicing
and workover costs. These are discussed in a later section.
Packer selection for thermal recovery wells is primarily a problem of selecting retrievable or
permanent equipment. This decision is influenced by such considerations as pilot or large scale
flood, soak, or drive etc. Whenever long periods of uninterrupted injection are planned,
permanent completion is preferred. Retrievable packers are the preferred choice if long
interruptions of the injection well are anticipated, so that the packer is available for use in another
well.
Retrievable thermal packers come in either single or double giip models. Single grip models
are used for steam injection below the packer and are usually tension set. This way pressure from
below tends to further set the packer. Double grip packers are used for pressure control above and
below the packer. Many thermal well retrievable packers come with an integral expansion joint.
Thermal packers come in a variety of sizes to fit casing from 4- 1/2 to 13-3/8 in. OD, and most are
designed with an operating limit of 2,100 psia and 640' F.
Packers should be located as high as possible in the weil to permit downward tubing
expansion, but close enough to the formation to confine the steam coming from the tubing.
Typical retrievable and permanent packer locations for steam injection wells are shown in figure
9.7.13 Packer installation for a multiple zone steam injection well is shown in figure 9.8.13
27 6
EXTENSION
JOINT
RETRIEVABLE
PACKER
SEAL
ASSEMBLY
DRILLABLE
PACKER
FIGURE 9.7. - Schematic showing retrievable and permanent packer position in steam injection
well. 13
4
-
ac v
-
THERMAL TRIMMED
PERMANENT PACKER
LOWER ZONE PERFORATIONS
277
The performance of a packer tubing system plays a major role in the success of any steam
injection project. Continued high injection rates without excessive heat losses are necessary for
efficient utilization of injected heat and sustained oil production. In this respect, packer plays a
crucial role in the economic success of the project. This is because: (a) The cost of replacing failed
wells, packers and tubing is high in terms of capital and lost production; (b) The price of a packertubing system is large and it must provide sufficient increased earnings or savings to justify their
installation; and (c) The selected packer should be able to lower the wellbore heat losses (and they
usually do). The cost of heat loss can be high in terms of operating costs, if steam costs (fuel,
water, labor) are high, or oil production, if steam generating capacity is restricted.
Packers do fail-packer failures in high pressure injection wells (pressure >700 psi,
temperature >500" F) are generally caused by expansion joint assembly failure (seal failures).
Details of a packer system used by a California thermal EOR operator to alleviate the casing
stress problems associated with high pressure steam injection ( I ,400 psi and 588" F) in shallow
wells (less than 3,000 ft) are shown in figure 9.9.14 The major components of the system are a
BALL JOIN7
XPANSON JOINT
EXPANSIONXWNT
278
locking type compression set packer, insulated tubing, and a surface expansion joint. Use of
custom packers such as this and those offered by major packer vendors (Guiberson-Dresser and
Otis) have practically eliminated the well failures.
The most common operational problem associated with the packers include leakage or
sticking of packers and thermal expansion joints. These problems can be minimized by carefully
running and setting the packers. Sufficient slip area must be provided; Otherwise, danger of
ruining the casing exists when high pressures are applied
Packer Installation
The reliability of a thermal packer can be enhanced and costly workovers avoided, if certain
precautions are followed prior to the packer installation. Gates4 advocates the following:
1 . Scrape the casing to remove any cement or scale to provide a clean, smooth surface on
which to set the packer.
2. Displace fluid in the hole below the packer setting depth by bailing or by dumping water
down the annulus. To obtain the full benefits of the packer, the annulus must be vented
during the early heating so as to remove any steam which occurs by evaporation of water
in the annulus.
3. Provisions must be made for the expansion and contraction of the tubing.
Seal Systems
One of the keys to successful steam injection operation is the ability to seal against high
pressure, high temperature steam. Downhole sealing points include ( 1 ) casing to open hole;
(2) tubing to casing annulus; (3) packer to casing, and (4) tubing to packer.
Casing to open hole and tubing to casing annulus seals involve thread connections. By
design, threaded joints on oil country tubular goods will not be pressure tight because of the root-
crest clearance, unless coated with a suitable thread .compound. Thread seal reliability depends
heavily on the mechanical properties of the fillers used in these compounds. In steam injection
applications, special high temperature sealing compounds with nonmetallic fillers must be used to
prevent leakage. Present day thermal well thread sealing compounds, though expensive (about
$8.00 per pound), are highly reliable and contain fillers that retain their physical properties even
after long periods of exposure to the steam injection environment. These sealants are rated to
1,200' F and employ a blend of copper, molybdenum disulfide, and teflon as the filler material.
Packer to casing and tubing to packer sealing depends on the reliability of elastomers used in
the thermal packers. Present day packer sealing elements are fabricated out of EPDM elastomers
with carbon fiber reinforcements. Though packer vendors rate these seals to about 600" F and
3,000 psi, field experience indicates these seals usually fdil at temperatures above 525" F. Hence,
before selecting a particular packer, its temperature and pressure limitations must be investigated.
27 9
SUMMARY
Steamflood injection and production well completion practices, while basically the same as
those used for conventional waterflood injection and production wells, differ in mechanical aspects
related to the heat factor. Thermal we11 completions are more expensive but are necessary to avoid
failures due to the stress caused by steam injection temperatures. A carefully engineered well
completion program for steam injection may make the difference between economic success and
failure of a seemingly attractive project. Downhole mechanical problems and excessive heat loss
can be minimized if the well completion is given as much technical consideration as the surface
equipment.
Thermal production well completion practices are similar to those of injection wells from a
tubing and casing consideration. The primary factor in selecting casing size is the size of
production equipment to be placed inside the casing. In the United States, most production wells
are completed open-hole with a slotted liner through the production interval. Careful consideration
should be given in selecting thermal packers, and temperature and pressure limitations of the
packer should be investigated.
REFERENCES
1. Nelson, C. G. Program is Designed to Analyze Casing Buckling in Thermal Recovery.
Oil & Gas J., v. 73, No. 49, Dec. 8, 1975, pp. 79-82.
2. Holmquist, D. E. Prestressed Casing Tolerates High BHT's. World Oil, v. 173, No. 6,
November 1971, pp. 80-82.
280
28 1
APPENDIX 9-A
B.
Casings and T u b i n s
5 . Kilsby-Roberts
P.O. Box 9500
Brea, CA 92622
Telephone: (7 14) 579-8823
3. Hydril Company
P.O. Box 60458
Houston, TX 77205-0458
Telephone: (7 13) 449-2000
4. Halliburton Services
10 15 Bois d' Arc, Drawer 1431
Duncan, OK 73536
Telephone: (405) 251-3760
1. Hydril Company
P.O. Box 60458
Houston, TX 77205-0458
Telephone: (7 13) 449-2000
5 . SchlumbergerWell Services
2x3
CHAPTER 1 0
THERMAL WELL CEMENTING PRACTICES
INTRODUCTION
Cement is almost exclusively used to fill the annular space between casing and open hole in a
well completion operation. The primary functions of any cementing job are (1) to restrict fluid
movement between formations and (2) to support the casing.' Good cement jobs are essential for
the successful operation of a well. In thermal wells it is even more ciitical. Steam injection wells
present special challenges when one designs a cement job for such wells. In most cases, the
cement slurry is subjected to relatively low temperatures during the cement job and early curing.
However, after the cement sets, it must withstand the thermal shock associated with the initiation
of steam injection. Another complicating factor is the problematic weak and unconsolidated
formation often encountered with thermal wells. This chapter presents an overview of current
steam injection well cementing practices. Vendors such as those listed in Appendix 10-A should
be consulted for additional information.
Steam Injection Well Cemenling Requirements
Although cements used in thermal wells are subject to thermal shock at the initiation of steam
injection, they must retain adequate compressive strength and low permeability during the life of
the well. Thermal stress on the cement and casing requires uniform high quality bonding
throughout the well. It is also advantageous to have low theimal conductivity cement such that the
injected heat goes predominantly to the oil bearing sand rather to the overburden.
Good cementing practices, such as the use of spacers, centralizers, and mud conditioners are
also very important. However, such practices are wasted unless the cement is properly designed
for long-term stability and adequate performance characteristics. Two kinds of cement, with
additives, are routinely used in the completion of steam injection wells: Portland cement (API
Class A, G or H cement) and high alumina cement. In general, Portland cement is useful in
applications up to about 650" F; however, beyond 600" F, use of high alumina cement is
284
Ordinary Portland cement is essentially a calcium silicate material. the primary components
being dicalcium silicate and tricalcium silicate. When water is added to the cement, it hydrates to
form a calcium silicate hydrate gel. It is this gel that is responsible for the strength and integrity of
the set cement at ordinary temperatures. In addition, a substantial amount of calcium hydroxide is
also liberated.
The calcium silicate hydrate gel provides good binding at lower temperatures, but at about
230" F, this gel undergoes a phase transition and converts to a phase called "alpha dicalcium silicate
hydrate' (a-gel). The a-gel is, unfortunately, a much denser material that is prone to shrinkage.
The shrinkage is accompanied by a loss of compressive strength and an increase in permeability.
Though the amount of strength reduction is severe (from about 4,000 psi to 2,800 psi in one
week), it is still sufficient to support the casing in a well (the minimum API acceptable compressive
strength is about 500 psi). However, the real problem is the drastic changes in permeability. The
changes in permeability may create greater problems. Cements exhibiting 0.01 mD permeability
after one day at 320' F can exhibit 1 mD permeability after one month. Noimally, this permeability
increase is of little importance in the high permeability reservoirs associated with the steamflood
operations. However, this permeability increase will cause steam to migrate to zones behind the
casing and weaken the casing-cement bond. This weakened bond is easily broken by the thermal
expansion of the casing. Thermal well cement studies indicate. that for the prevention of interzonal
communication, the water permeability must not exceed 0.1 mD at steam injection temperature^.^
In figures 10.1 and 10.2, the compressive strength and permeability performance of neat Portland
cements at 400" F are shown, respectively.7
Portland cement slurries are usually made suitable for steam injection applications by
stabilizing them with the addition of silica flour. When silica flour is added to the cement, a
portion of the additive reacts with the calcium hydroxide to form alpha dicalcium silicate hydrate
(a-gel). The remaining silica reacts with the a-gel to form what is known as the tobermorite
phase.5 The tobermorite phase is a better cementing binder than a-gel and retains the high
compressive strength and low-permeability charactelistics of the PortIand cement. In figures 10.3
and 10.4, the compressive strength and permeability performance of silica flour augmented
Portland cements are shown, respectively.7
For steam injection well cementing applications, normally 30 to 60% (by weight of dry
cement), silica flour is added to the Portland cement. The average being about 40%. The
minimum temperature at which silica flour should be used is 230" F. Recent research indicates that
regular silica flour is the best additive for temperatures ranging from 230' to 600' F.6 Above
650" F, these mineral phases begin to dehydrate to form calcium oxide. In steam injection
operations, the dehydrated cement will begin to rehydrate in the presence of steam and begin to
disintegrate due to volume increase.
285
5000
4000
3000
2000
1000
0.000
0
12
15
18
21
24
10
I '
1
0.
0.01
0.001
0
12
15
18
21
24
286
8000
'E
7500
E
p
7000
K
I-
n
w
6500
L
(0
$a
6000
5500
d
5000
0
12
15
18
21
24
'0
0.1
=!
5
5
B
I-
0.01
0.001
0
12
15
18
21
24
287
Alumina silicate cements such as the 'lPozzolan" cements, which contain fly ash, can
chemically bind the free calcium hydroxide in Portland cements5 but are often unreliable at the high
temperatures associated with a steam well, and their use is not recommended.3 Thus, for all
practical purposes, the Portland cements are satisfactory for steam injection well completion up to
600" F, if sufficient silica flour is added to the cement. Since very few steam injection operators
inject steam above 600" F, silica flour augmented Portland cement is satisfactory for most steam
injection applications. Silica flour is also inexpensive and is compatible with all commonly used
retarders, friction reducers, low water loss additives, and lost circulation materials.
288
primary cementing. After the cement has set, however, a steam injection well must withstand
temperatures up to 650" F, but not until steam injection begins.
The formations associated with steam injection wells are usually unconsolidated, weak and
have low fracture gradients. Hence, lost circulation of cement and fluid loss problems are often
present. To avoid fracturing the weak formations with cement and the associated cement loss, low
density cement slurries are necessary. Another problem associated with the thermal well is the
thermal expansion of the casings and wellhead growth. This calls for a cement with good
expandable properties. To minimize heat loss to the over- and underburdens, the cement should
have good insulating properties.
These unique characteristics of thermal wells and the associated performance requirements of
cementing materials are often at cross purposes. For example, as mentioned previously, thermal
well cement must retain the strength and low-petmeability characteristics of ordinary cement at
elevated temperatures. This requirement can be met through the use of high density cement
slurries. Unfortunately, since the formations associated with steam wells are weak, high density
slurries are generally not suitable because of the loss circulation problem. Further, thermal well
completion calls for cements with good insulating properties (i.e., low thermal conductivity
characteristics), but high density cements are poor insulators. Therefore, considerable research has
been performed to formulate cements with the above desired properties.
Until a few years ago, low density Portland cement slurries were prepared by adding light
weight filler materials (such as perlite, gilsonite, powdered coal and diatomaceous earth) or water
absorbing thickeners such as bentonite and sodium silicate. The performance of selected low
density slurries are shown in figures 10.5 through
These figures indicate that high
compressive strength and low water permeability are not linked. Further. field experience indicates
that conventional, filler cements with densities less than 12.5 pounds per gallon (ppg) perform
poorly in steam injection wells due to permeability increase.8
Recently, two new methods of preparing low density slurries have been developed, These
methods not only reduce the slurry density to a very low level, they were also able to retain the
compressive strength and low-permeability characteristics of the cement at elevated temperatures.
The method includes (1) addition of hollow glass or ceramic micro spheres9 and (2) foamed
cernent.l* Of these two techniques, foam cements are preferred for thermal applications due to
their superior performance.7
With the addition of proper surfactants and stabilizers, it is now possible to prepare low
density slurries with high compressive strength and extremely low-permeabilities. Foamed cement
made from a base slurry stabilized for high temperatures exhibits excellent performance
289
8000
7000
A'
6000
5000
4000
3000
2000
1000
12
15
18
21
24
100
10
iz
W
Q
0.1
a
W
0.01
0.001
12
290
5000 .
3000
2000
1000
12
loo
i4z
3I
0.1
0.01
I-
0.001
'
12
291
characteristics at steam injection temperatures. Although foamed cement with densities as low as 9
ppg show excellent performance characteristics in terms of compressive strength, their water
permeabilities are somewhat higher than microsphere ~ e m e n t . ~
A typical cement formulation recommended for most steam injection wells could consist of
50% Class G cement, 35% -200 mesh silica flour and 15% microspheres.
Cementing Practices
Given a good composition of thermally stable, bondable cement, it is essential that the cement
be applied properly to assure adequate zonal isolation and uniform bond quality. Good cementing
practices is even more critical in thermal wells to ensure that the integrity of the bond is maintained
under temperature induced stress conditions. Cement should be returned to the surface on every
string of casing to assure complete fill-up of cement in the casing hole annulus and to anchor the
casing firmly to the ground.
Prior to the cementing operation, the drilling mud must be removed to assure good stable
bonding. Mud removal can be enhanced through the use of spacers and washes.9 Most operators
use chemical preflushes ahead of the cement job to help remove the mud in the casing and
formation wall. The chemical preflush, which is normally water-based, should contain a good
mud thinner and a surfactant to leave the pipe and formation surface in a water-wet condition.
Care should be exercised in selecting the mud thinners, since some of them (such as quebracho,
lignosulfonates) added to water may retard cement setting.1 Dilute mix of Portland cement put in
turbulant flow is an excellent preflush and aid in erosion of gelled mud and filter cake. Because of
great variation in mud system, the chemical wash should be selected for the particular mud system
in use.
Effective centralization is a critical factor in obtaining a good primary cement job. Enough
centralizers must be used to produce a uniform sheath of cement. In straight hole, one centralizer
per joint is recommended. Centralizers should be placed in gauge sections of the hole as
determined by caliper or other logs. In crooked hole, centralizer placement depends on hole
deviation.
Floating and guiding equipment should be used to assure good cement job. Float collars
should be located one or two joints above the float shoe to prevent mud contaminated cement from
being placed outside the bottom casing j0ining.l Use of scratchers should be given serious
consideration, since they aid in the removal of mud cake and break the mud gel. They also help to
mix up the mud with cement, thus reducing channeling. Hole washout must be taken into
consideration in estimating the cement volume required to achieve the desired fill-up. The actual
cement volume to be used should be at least 15% more than the estimated value.
292
The success of the cement job depends on the quality of cement mixing. Weight of cement
slurry should be monitored to ensure that the correct amount of water is used to prepare the slurry.
Wiper plugs should be used to wipe mud off casing ahead of cement as well as to separate mud
and cement. Consideration should be given as to whether or not to reciprocate and/or rotate the
casing while cementing to assure successful mud displacement and enhance the formation of good
uniform cement sheath. Cement should be circulated near turbulence condition to aid in the
removal of mud and to reduce the possibility of mud channels remaining in the productive zone.
Use of friction reducers with the slurries promote turbulence flow and should be considered.
Good operator and service company supervision during cementing operations is critical to
minimize cement job failures. The slurry should be placed at the highest practical rate without
exceeding the fracturing pressure of the formation. Use of sonic or ultrasonic logs are
recommended to monitor the presence of cement and the quality of casing-cement bond. Proper
interpretation of the log is critical and the service of experienced engineers should be enlisted to
interpret the logs.
SUMMARY
This chapter presents a brief overview of thermal well cementing practices. Neat Portland
cements (API Class A, G or H) are not appropriate for completing steam injection wells. These
cements lose their compressive strength and show permeability increases at temperatures above
230" F. However, Portland cements with appropriate additives can be used to cement steam
injection wells, provided well temperatures do not exceed 600' F. Above this temperature, high
alumina cements are recommended. Fine mesh silica flour 30% to 60% by weight of dry cement
should be mixed with Portland cement and used in thermal applications.
Since the formation associated with the steam injection wells are usually unconsolidated and
weak, high density cements should not be used. Ceramic microspheres can be used to prepare
ultra low density cements with acceptable peiformance for steam injection wells.
critical in steam injection well completion. Cement returns to
Good cementing practices
the surface is a must. The hole should be well circulated prior to cementing. Pipe should be
suitably equipped with floating and guiding equipment. Adequate number of centralizers must be
used to assure good cement sheath. During cementing operations, mud removal can be greatly
enhanced through the proper application of spacers and washers. Use of bottom plugs, chemical
washes, excess cement, and good supervision will assure good cementing job.
REFERENCES
1. Allen, T. 0. and A. P. Roberts. Production Operations, v. 1 ; Well Completions,
Workover and Stimulation, chapter 4. Oil and Gas Consultants International, lnc., Tulsa,
Oklahoma, 1978.
293
3. Nelson, E. B. Improved Cement Slurry Designed For Thermal EOR Wells. Oil & Gas
J., v. 84, No. 48, Dec. 1, 1986, pp. 39-44.
4. Eilers, L. H. and R. L. Root. Long-Term Effects of High Temperature on Strength
Retrogression of Cements. Pres. at the 46th Annual California Regional Meeting of the SOC. of
Pet. Eng., Long Beach, California, Apr. 8-9, 1976. SPE paper 5871.
Institute of Mining, Calgary, Alberta, Canada, June 10-13, 1984, Paper No. 84-35-1 15.
9. Griffin, T. J. and W. Chimilowski. Cement Spacers and Washes Can Improve
Stimulation and Production Characteristics of Oil and Gas. Pres. at the 29th Annual Tech. Meeting
of the Pet. SOC.of the Canadian Institute of Mining, Calgary, Alberta, Canada, June 13-16, 1978.
Paper No. 78-29-12.
10. Olanson, M. T. Application of Foam Cements in Alberta. Pres. at the 35th Annual
Tech. Meeting of the Pet. SOC.of the Canadian Institute of Mining, Calgary, Alberta, Canada,
June 10-13, 1984. Paper No. 84-35-72.
11. Cain, J. E. and S . H. Shryock. Cementing Steam Injection Wells in California. J. Pet.
Tech., v. 18, No. 4, April 1966, pp. 431-436,
294
APPENDIX 10-A
THERMAL WELL CEMENTING SERVICE VENDORS
A.
205
CHAPTER 1 1
THERMAE WELL LIFTING EQUIPMENT
INTRODUCTION
The production of oil from thermal wells by artificial lifts is more difficult and expensive than
producing from conventional wells for several reasons. Lifting conditions can vary widely during
steam stimulation cycles. Temperatures may range from 75' to 600' F, viscosities from 10,000 CP
to 1 cP, water-oil ratios from 0.1 to 20, liquid production rates from 10 to 1,000 bbllday, gas-oil
ratios from 50 to 1,000 SCF/STB and sand cuts from 0 to 18%. These conditions impose severe
strain on the lifting systems. Produced vapor, though mostly steam, may also contain acid gases.
The presence of high temperature, high water cut and acid gases can result in downhole corrosion.
Although different types of artificial lift equipment has been utilized to lift oil from steamed well,
the most prevalent form of lift for fluids in thermal operations is the conventional rod pump.
Corrosion and abrasion (due to sand production) in combination with the mechanical stresses of a
reciprocating pump jack takes a heavy toll on the downhole pump. Problems associated with the
lifting of heavy oil in thermal well by pump jacks has resulted in the emergence of newer breeds of
pump jacks offering the control and flexibility required for optimizing production in these wells.
This chapter presents an overview of the pumping problems associated with the lifting of
heavy oil from thermal wells. The problems are discussed primarily as they affect rod pumps.
THERMAL WELL PUMPllNG PROBLEMS
Thermal wells are generally associated with low gravity, high viscosity crudes, produced
from unconsolidated formation. Some of the pumping problem5 associated with such wells
include rod floating, sand production, steam flashing, varying produced gravity, varying inflow
rates, and system operational problems.
Rod Full und Rod Drag
In heavy oil wells, the rate of movement of the sucker rod on the downstroke is greatly
influenced by oil viscosity. In steam operations, prior to steam breakthrough, the downhole
temperatures are low enough that the oil tends to stay viscous. The viscous oil can cause
considerable viscous drag in the production equipment. This viscous drag or frictional force on the
rod can be easily overcome during the upstroke of the pump jack. However, during the
downstroke only gravitational forces aid the rod movement and the frictional forces can
considerably slow down the rod movement.
The rate at which the rods move on the downstroke is called the rod fall or rod float rate. It is
typically in the range of 10 to 60 ft per minute (0.5 to 3 strokes per minute). Thus, the rod float
rate constrains the beam pumping unit operation. Attempting to run the pump jack at a speed faster
206
than the rate at which the rod can sink creates a condition known as rod float. Rod float will result
in the polished rod clamp and carrier bar to separate on pump downstroke.
Since pump jacks are normally a single speed device. the carrier bar tmvels at the same speed
on the upstroke as on the downstroke. The resulting collision between the downward moving
clamp and rod string and upward moving carrier bar can cause major damage to both surface and
downhole equipment.
To accommodate extra rod stresses resulting from viscous forces. oversized rod strings are
usually used. Since the oversized rod reduces the annular clearance between the rod coupling and
the tubing inside diameter, it aggravates the rod float problems. Most operators attempt to soften
the rod float effect by providing maximum annular clearance with the use of larger diameter tubes.
To date, many operators avoid rod float conditions by lowering the pump speed downstroke
through the use of motor sheaves.2 Since the fluid production rate is limited not by pump size, but
rather by pumping speed, rod float restricts well productivity. Production restrictions can result in
high fluid levels in the well, which increases the backpressure on the well. Increased backpressure
reduces inflow rates and will result in the loss of average production.
Many operators use a longer stroke surface unit and downhole pump to increase production.
However, this is a costly and only partially effective solution. A larger subsuiface pump diameter
could be installed to increase production, but pump diameters are limited by sand settling
problems, pump compression ratios and protection from high temperatures because of steam.
Thus, productivity of a well is severely impaired by rod float. On some wells, increased
backpressures cause sand to settle.
Further, the high viscosity of heavy oil causes conventional subsurface pumps to operate less
efficiently because of slow closure of traveling and standing valves.3 The viscous friction of fluid
flowing through the traveling valve on the downstroke further reduces the rod fall rate. To avoid
rod float, the upstroke travel rate must match the downstroke rate. This imposed slow upstroke
results in sluggish closure of the traveling valve. Thus, a percentage of production is lost on every
stroke.3 In many wells, restricted flow because of rod flo{t will keep a well from being pumped
off.
I
Sanding
the world's heavy oil is found in unconsolidated formation with high sand content.
n from unconsolidated heavy oil reservoirs is one of the major problems affecting
heavy oil production from thermal wells. Production of sand particles with the fluid causes severe
I
e equipment and ac
f abrasive effects of sand on do
operating problems bec
I
of sand across the pe
ns. Experience with thermal
ry projects in unc
reservoirs indicates that despite the use of sand exclusion measures, some sand production is
297
unavoidable. Furthermore, fluctuating wellbore conditions due to well stimulation and varying
production rates may also cause an increase in sand production. The presence of sand in the
production fluids damage the sucker rod pump. Damages to the pump include erosion of the ball
and seat, and seizure of the p l ~ n g e r . ~
Unlike conventional operations where a well is routinely pumped off and the pump jack shut
down to reduce the effects of fluid pound, pumps in thermal wells are kept running to avoid
complications imposed by a well sand flow. When fluid stops moving, the sand held in
suspension in the fluid settles out on the pump. This condition often necessitates costly workover
and frequent servicing of the pumping equipment. In extreme cases, sand problems dictate a
complete recompletion of the well. In wells where sanding is severe, operators have reported
pulling sucker rod pumps only after three weeks of service because of wear and plunger seizureP
Steam Breakthrough
When thermal wells are pumped too fast, hot fluid and in some cases steam itself is brought
back to the surface. This can.result in damage to the suifixe equipment and cause production rates
to drop. Steam interference can significantly intluence pump performance and adversely affect
pump efficiencies. Field investigations in Canada revealed that many cyclic wells where steam is
produced along with oil, steam significantly affected the pump performance. The barrel fillage
ranged from 60% to 80% and the corresponding pump efficiencies were 30% to 68%.5
Production Conditions Variations
In steam operations, the composition. gravity and viscosity of the produced oil can and does
change daily. This is because of the steam distillation and solvent drive mechanism associated with
such processes. Further, oil and stimulation fluids (steam, hot water, and noncondensible gas) do
not flow into the wellbore at a constant ratio. This is because of the gravity override and fingering
of the stimulation fluids as well as uneven heating of the reservoir sand.
Varying inflow rates and the fluid properties causes rod Fail rate to change. As discussed
previously, fluctuation in the rod fall rate seiiously impairs the pump performance. The traditional
approach to this problem is to monitor the operation of the beam unit closely and implement
appropriate corrective action as needed. If the operator foresees a change in rod float rate or a
pump off condition becomes imminent, he alters the pump speed rate to correct the problem. This
is usually done by changing the pump drive belt sheaves. This approach is not only labor intensive
and inefficient but also very expensive. Further, the daily changes in fluid gravity causes the beam
pump to remain out of balance most of the time.
298
Traditional approaches used in the United States thermal operations to improve the pump
performance include: (1) setting the pump speed to achieve an estimated average production;
(2) alter pump speeds mechanically by changing sheave size; (3) use of variable speed motors to
change pump speed; and {4) use of pu
ch of these approaches have
tion, the man-hours needed to
,-pump efficiency is reduced
r steam breakthrough. Further,
uction rate, this can result in
high fluid level
and unnecessary maintenance. 'The limitation of changing motor sheaves to alter pump speed were
299
already discussed. The use of a variable speed motor to change pump speed, though attractive
from the operational prospective, has its limitations. Use of variable speed motors require
someone to watch production rates or fluid pound indications to determine when to change the
pump speed. Over and under counterbalancing can be severe when variable speed motors are
used.3 The required gear box torque will be increased considerably if the unit is slowed on the
downstroke, because the counterbalance inertia cannot be used to assist the motor.
The inadequacies of the conventional beam pumps to meet the service demands of heavy oil
production have led to the development of alternate pumping techniques. These include
modification of conventional pump jacks to improve its flexibility and novel pumps such as the
diluent pumps, cavity pumps and chamber lift pumps. None of these pumps are widely accepted
and they are used principally in the heavy oilfields of Canada and Venezuela, where the oil is more
viscous than those found in the United States.
300
and are field tested in Canada. The authors are not aware of any of these units being used in the
United States.
H E P Pump7
This is a hydraulic, variable speed, surface pumping unit that utilizes a hydraulic system to
transmit energy from the prime mover to the sucker rod string. The pump was developed to
eliminate the gear box in the conventional beam unit in fdvor of a high torque, low speed motor.
The system uses Llydraulics to transmit the power from the prime mover to the sucker rod
strings, Electronics to control the rod motion, and a Pneumatic system to provide counterbalance
effects. Hence, the name HEP.
The system utilized a double-acting piston and cylinder hydraulic motor, mounted on the
wellhead to actuate the rod string. The piston rod is connected to the polished rod by means of
flexible wireline, carrier bar and rod clamp. A variable displacement pump, which is capable of a
wide range of discharge rates, is used to hydraulically transmit input power from the prime mover
to the hydraulic piston. The displacement pump discharge rate and direction of motion are
controlled electronically. In this way, the motion of the polished rod is controlled automatically.
Stroke speed and the velocity profile are varied by changing the control panel settings. This feature
allows upstroke and downstroke rod speeds to vay independent of each other. Counterbalance for
the pumping system is provided by a nitrogen filled hydraulic system. The unit offers 3 different
stroke lengths and the desired stroke length can be selected by a switch inside the electronic control
panel.
The HEP has exhibited its superiority over conventional beam pumps in the field tests.
However, the system has two major drawbacks that prevent its widespread application. First, the
reliability of hydraulic cylinders (the system that transmits power from the prime mover to the rod
strings) in relation to long-term continuous operation is suspect. Secondly, difficulties have been
encountered in effectively counterbalancing the system. There are two concerns with the nitrogen
filled hydraulic counterbalance system: leakage and actual counterbalance effect.* As a gas, the
nitrogen pressure is dependent on ambient temperature and wide vaiiations in temperatures affect
the counterbalance capability. lso, since the pressure varies inversely as volume, large volumes
are needed to minimize pressure fluctuations.
PC 3000 System3
The PC 3000 system operates like an air-balanced beam pump unit and utilizes a low speed,
high torque hydraulic motor and a nitrogen system to provide the needed counterbalance. A
hydraulic system is utilized to lift the beam, rod string:, and tluid load on the upstroke. A
hydraulically driven winch is used to pull the beam i n the downstroke. This system also allows the
30 I
operator to vary the upstroke and downstroke speed independently to accommodate any pumping
condition. The unit is still in the developmental stage and waiting field tiial.
302
TYPE X
CONVENTIONAL
DILUENT PUMP
TYPE Y
DILUENT INJECTION
DILUENT INJECTION
DILUENT INJECTION
PRODUCTION
LRODUCTION
DISCONNECT
PUMP
BARREL
SPRAY NOZZLE
"EXPLANATION
BLENDED CRUDE
FIGURE 11.1. -- Schematic of a diluent pump used in the Cat Canyon field of
California to produce the 6" API oi1.9
1. Fluid level has to always be a minimum of 300 ft above the pump to avoid starter damage
due to friction heat.
2. In wells with high sand production, a prepack filter must be used below the pump to
avoid starter damage.
3. Wells with high GOR will require gas anchors and pumps with sufficiently large
capacities to handle gas and liquid.
303
Despi :its disadvantages, rod pumps are used almost exclusi rely i n U.S. thermal oil
production. This is because the alternative lifting methods such as the gas lift are more problem
prone and economically unattractive to operate. Both insert pumps and tubing pumps are used in
the U.S. steamflood operation. Insert pumps are preferred over tubing pumps due to the
undesirability of pelforming a tubing job to replace a worn pump barrel in a potentially hot well.
Most operators use the heavy-wall, full-barrel, top hold down pumps with stainless steel friction
ring assemblies. Many operators employ slow speed, longer stroke pumps to handle large
volumes of fluid produced after steam breakthrough. Operators' experience indicate that it is not
possible to obtain accurate liquid levels in wells producing large volumes of steam. Field
experience in such wells, indicate basing pumps off condition on fluid pounding alone is not sound
and can often lead to steam lock in the pump. l4
controllers are increasingly being used in California thermal operations to improve pumping
efficiency.13-14 One large California cyclic steam operator set the initial pumping unit speed to
satisfy the peak production rate for the well. As the production rates declined and a pumped-off
condition was approached he was able to detect the condition with the help of a pump-off controller
and shut down the unit for a preset time to allow entry of fluid to the well. This solved the
problems of pumping unit and downhole damages associated with tluid pound.l3 This approach
also has the advantages of power savings and reduced maintenance cost while providing an
efficient way to produce steam stimulated wells. The savings realized by this particular operator
included:14 31% reduction in power consumption; 23% reduction in well work expenditure and a
substantial increase in per well oil production. In spite of its demonstrated advantages, pump-off
controllers are not yet used widely because of its high initial cost (about $4,50O/well installed) and
availability of cheaper alternatives such as variable speed motors.
SUMMARY
The production of oil from thermal wells by artificial lifts is more difficult and expensive than
producing conventional wells. Pumping problems encountered in the production of heavy oil from
thermally stimulated wells include: high sand production, varying produced gravity, varying
inflow rates, varying viscosity, vapor interference and system problems. Because of its versatility,
reliability and low cost, rod pumps are the pump of choice in theimal operations. However, severe
problems are encountered in the use of rod pumps as the lifting device in thermally stimulated
heavy oil wells. These include rod floating and overloading due to high fluid viscosity, pump
sticking and erosion with high sand production, heavy crude plugging, and gas locking under
steamflood conditions. The inadequacies of the conventional beam pumps to meet the service
demands of heavy oil production led to the development of alternate pumping techniques. These
include modification of conventional pump jacks to alleviate some of the aforementioned problems,
and novel pumps such as the diluent pumps, cavity pumps and chamber lift pumps.
These new breeds of pumps are still in the prototype stage and used principally in the heavy
oil fields of Canada and Venezuela. In the United States, despite its drawbacks, rod pumps are
used almost exclusively in the thermal operation. Insert type rod pumps are used in the United
States thermal operation. Use of pump-off controllers and variable speed drives are used to extend
the operating range of a given rod pump.
A list of thermal well lifting equipment vendors is included in Appendix 11-A. It is
recommended that the operators follow the pump vendor's recommendation in selecting the
appropriate lifting equipment for the production well.
305
REFERENCES
1. The University of Texas-Lessons In Well Servicing and Workover-Artificial Lift
Methods, Lesson 5-A home study course. Petroleum Extension Service, The University of
Texas, Austin, Texas, 197 1.
2. Beyea, D. F. and M. J. McKinnon. Optimizing Pump Speed Throughout Production
Cycles. Pres. at the Fourth Annual Canadian Heavy Oil and Tar Sands Technical Symposium,
Calgary, Alberta, Canada, Feb. 18, 1987.
3. Cockshutt, D. and W. Thomas. New Developments in Beam Pumping Technology: PC
3000. Pres. at the Canadian Heavy oil Association Thermal Well Completion Seminar, Edmonton,
Alberta, Canada. Aug. 11, 1988. Proceedings of Thermal Well Completions Seminar, D. J.
Theriault, Editor.
4. Sudol, T., R. Ridley and D. Nguyen. New Sucker Rod Pump Design for Sand
Producing Wells. Pres. at the 5th UNITAR International Conference on Heavy Crudes and Tar
Sands, Aug. 4-9, 1991, Caracas, Venezuela-I?. F. Meyer, Editor. Heavy Crude and Tar
Sands-Hydrocarbons for the 21st Century. v. 2, published by Petroleos de Venezuela, S.A.,
Caracas Venezuela, October 1991.
Norcen-Bodo Steam Pilot. Pres. at the Canadian Heavy Oil Association Thermal Well Completion
To Work. Pres. at the 5th UNITAR International Conference on Heavy Crudes and Tar Sands,
Aug. 4-9, 1991, Caracas, Venezuela. Conference Proceedings, v. 2, R. F. Meyer, ed. Published
by Petroleos de Venezuela, S.A., Caracas, Venezuela, October 1991.
11. Uzcategui, E. Application of Progressive Cavity Pump in Maraven's Heavy Oil Fields.
Pres. at the 5th UNITAR International Conference on Heavy Crudes and Tar Sands, Aug. 4-9,
1991, Caracas, Venezuela. Conference Proceedings, v. 2, R. F. Meyer, ed. Published by
Petroleos de Venezuela, S.A., Caracas, Venezuela, October 199 1 .
306
307
APPENDIX 1 I - A
THERMAL WELL LlFTHNG EQUIPMENT SUPPLIERS
R o d h mD- Vendors
30')
CHAPTER 12
USE OF INSULATED TUBULARS IN THERMAL PROJECTS
INTRODUCTION
Heat losses in the steam injection well can be very detrimental to a steamflood project.
Unless efforts are made to minimize losses, the economics of the project can be severely impacted
by increased energy costs and, in the case of deep reservoirs, the steam quality can be reduced to
an unsatisfactory level. High heat losses can also raise casing temperatures to a level that risks
failure under the stress of thermal expansion.
Insulated tubing is frequently used to minimize heat loss in the injection well, and maximize
the downhole steam quality. This will also lower the longitudinal expansion and thermal stress in
the casing.
The object of this chapter is to discuss briefly the wellbore heat losses and how they can be
minimized using insulated tubing. The topics discussed include ( 1 ) wellbore heat losses;
(2) advantages of using insulated tubing; (3) insulated tubing design requirements; (4) insulated
tubing construction and selection criteria; (5) insdated tabing installation procedure; (6) insulated
tubing performance; and (7) insulated tubing cost effectiveness.
Wellbore Heat Losses
Generally, in thermal recovery of petroleum, low quality steam is used for injection so that
low quality feedwater, free of hardness, may be used to generate steam. Because the saturated
steam carries less heat than superheated steam, it is important that the steam reach the formation
with as little heat loss as possible. As deeper formations are considered, the heat losses in the
wellbore become even more important.
Crawfordl has shown that for 2.5 in. tubing in a 7-3B-in. casing, the heat losses in Btu'dd
may be approximated from a chart. His example was developed for a well depth of 1,000 ft in
which 250 bbVd of 400' F steam was injected. The total heat loss could range from 1 million
Btu'dd up to more than 16,000,000 Btu'dd. This depends on the overall heat transfer coefficient
between the tubing and the formation. This coefticient ranges from 5 for the lower figure to 240
for the higher loss. For uninsulated tubing, a reasonable value for the coefficient is between 80
and 160, giving a total heat loss between 10,000,000 and 15,000,000 Btu'dd, or approximately
15,000 pounds of steam condensed per day. This is equivalent to 45 barrels of water. For deeper
wells, the loss becomes even greater; at 2.000 ft, 100 barrels of condensation, and at 5,000 to
6,000 ft, 200 barrels.
The important variables in calculating the heat loss are the diameters of the tubing and the
casing, the temperature and pressure of the injected steam, and the depth of the well. Equations
have been developed by Ramey2 and Satter3 but they lire complicated, and tedious calculations are
310
necessary to apply them. However, charts have been prepared by Huygen and H ~ i t tand
, ~ they are
much easier to apply. One of these charts is shown in figure 12.1. This chart is for 2-1/2-in.
tubing in 5-1/2-in. casing, and gives heat losses both for injection into the casing and injection
through the tubing. For 400" F steam injected through tubing, the heat losses for the first day
would be 600 Btdhr-ft. For the 1,000 ft well, the daily loss would be 14.4 million Btu. After a
year, this would decrease to 9.6 million. Using a chart for 7-in. casing, these numbers do not
change appreciably. They check reasonably well with the previous calculation. It is of some
interest that injection through the casing would increase these values for heat loss to 36 million Btu
and 14.9 million Btu.
These losses have caused a number of studies to be made for methods of improving the
efficiency of steam delivery. One showed that a packer at the bottom of the well would prevent
steam from coming back up the annulus between the tubing and the casing. An attempt to use an
insulating fluid in the annulus gave problems largely due to decomposition of the fluid.5 Painting
the outside of the tubing with aluminum paint gave some success but is difficult to apply. Willhiteg
has shown how these procedures affect the heat transfer coefticient in figure 12.2.
31 I
4000
1
_.____......
_._...._.....
.I
3000
2000
MONTH
8
1 YEAR
1000
0 0
200
400
600
800
CASING TEMPERATURE - O F
FIGURE 12.1. - Heat loss in the wellbore and casing temperature during steam injection (2-1/2-in. tubing-5-1/2-in.
casing).4
300
400
500
700
600
TUBING TEMPEFMTURE.
3600
3Ooo
2400
1800
1200
600
0 '
0
0.5
1.5
2.5
313
approaches than insulated surface lines. The in-well tubulars must be of restricted diameters to
conform to well-casing size. Sufficient casing clearance and size tolerance must be provided to
accommodate couplings and the effects of well deviation from vertical. The insulated tubular must
be able to sustain the various loads associated with vertical running and pulling as well as the need
to support its own weight when suspended from the wellhead.
The environment of the well is such that corrosion resistance to well fluids and sour gas is
also a problem not faced by the surface-line insulation. To satisfy these special needs, the design
trend for insulated tubulars has been to use an inner API tube to convey the steam and an outer API
tube to carry the loads and protect the insulation in the sealed annulus between the tubes. In these
dual tube applications, with an inner tube diameter of 2 in. or greater and an outer tube of 4.5 to
6 in. diameter, the limit of net insulation thickness is approximately 1 in. This requires highly
effective insulation material and a relatively large differential expansion across a small gap due to
thermal differences from the inner to the outer steel tubes.
3 14
COLLAR
TUBING
CASING
6.DOLISTANDARD BUTTRESS
THREAD COUPLING,
INNER- TUBING
BFI I OWS
DIFFUSER SLEEVE
BELLOWS
WELD R ~ N G S
M U T U Y E A ( W J ) NSUA
SYSTEM PLUS VACUUM
AM ORAOE K-56
*AM
AM 5CTWlTPESS CCUPLNG
MODIREDFOR SEM RiNG
N CouPLlNo M U A T O R
CVTAWAY s13wNQ
KAWAYXI COUPLNGCoNffiUUATIc
315
The individual assemblies are joined by conventional threaded couplings secured at the time of
deployment into the well. The newest and most effective of the ThermocaserMseries is the 750
model which has the insulation annulus at a high vacuum supported by gathering agents to
maintain the vacuum. An exploded view of this insulated tubular is depicted in figure 12.6.
BUTTRESS
INNER
TUBE
OUITER
CASTING
BUITRESS
COUPLING
COUPLING
INSULATOR
THERMOCASE
JOINT
317
SUMMARY
Heat losses in the wellbore we quite appreciable and become greater as the depth of the well
increases. Consequently, much consideration has been given to methods for minimizing these
losses. Isolation of the tubing from the casing by a packer gives some improvements as does
painting the tubing with aluminum paint. However. the most effective method is to insulate the
tubing. This has been shown to be cost effective with a payout of less than 2 years. Vendors such
as the one listed in Appendix 12-A can help in the design and selection of insulated tubulars for
steam injection wells.
318
L = 3,000 ft
P = 1,500 psia
T = 596' F
Heat in (Btu/hr)
(Ibhr)
Above ground heat loss
(Btuhr)
Below ground
A. Heat loss (Btuhr)
B. Condensation loss
(Btuhr)
(Ib/hr)
Heat out ( B t u h )
Casing temperature, O F
Heat loss, %
Steam quality at sandface, %
1.
2.
3.
4.
5.
6.
7.
I2,500.000
( 1 0.704)
254.000
12,500,000
(10,704)
254.000
2 .700.000
8 10,000
2,781,500
(5,OO)
6.764.500
490
45.9
53
764.913
(1,375)
10,671.087
224
14.6
87
Payout
AfkusX
1. 80% s t u t y (5.1 ton s t e a m / t o o e d
10,704
hr
Yr
oill;
1 ton
Steam x 24 hr x 365 day x ___
day
Yr 2,000 lb
4. -A
V'
1,379 m
Y'
x 2,000-!-!Lx
$40/ft x 3,000 ft x
ton
bbl
x=
= $122,58O/yr
350 lb x 0.9 bbl
yr
$122,580
Yr
= 11.8 mos.
319
3 in. Kawasaki
Thermocase"
5 in. x 3-112 in.
REFERENCES
1. Crawford, Paul B. Heat Loss When Injecting Steam Down Tubing in Thermal Recovery
Programs. Producers Monthly, v. 30, No. 11, November 1966, pp. 16-17.
2. Ramey, H. J., Jr. Wellbore Heat Transmission. J. Pet. Tech., v. 14, No. 4, April
1962, p. 427.
V.
3. Satter, Abdus. Heat Losses During Flow of Steam Down a Wellbore. J. Pet. Tech.,
17, NO. 7, July 1965, pp. 845-851.
4. Huygen, Hans H. A. and J. L. Huitt. Wellbore Heat Losses and Casing Temperatures
During Steam Injection. Producers Monthly, v. 30, No. 8, August 1966, pp. 2-8.
5 . Willhite, G. P., J. H. Wilson and W. L. Martin. Use of an Insulating Fluid for Casing
Protection During Steam Injection. J. Pet. Tech., v. 19. No. 1 1 , November 1967,
pp. 1,453- 1,456.
6. Davis, J. S. and J. P. Fanaritis. The Application and Economics of the Use of Insulated
Injection Tubing With Oil Field Steam Generators. Chapter 70 in The Future of Heavy Oil and Tar
Sands, edited by R. F. Meyer, J. C. Wynn, J. C. Olson. Second International UNITAR Conf.,
Caracas, Venezuela, February 1982, McGraw-Hill, New York, pp. 675-684.
320
APPENDIX 12-A
32 1
CHAPTER 13
SAND CONTROL PROBLEMS
INTRODUCTION
Some oil-bearing formations are so weakly held together that sand grains are produced with
the oil. This presents several difficulties. If the sand is lifted to the surface, separation and
disposal are required. It poses abrasion problems both to the tubing and to the pumps. It may clog
flow patterns and reduce the production of oil. Clogging may occur in the formation itself or in the
slots, perforations, or screens of the producing equipment.
While sanding may be a problem with primary production wells, the problem becomes much
greater when steam is injected into the formation. Some of these extra difficulties may be
mentioned. Steam dissolves the gravel packing at the bottom of the well as well as the formation
sand. It pushes the gravel away from the screen and into the formation sand. The dissolved silica
precipitates as the steam cools. The condensed steam swells clays and precipitates scales.
Because of the non-uniformity of the Occurrence of the unconsolidated sands, it is difficult to
give a general treatment of the problems. Each well must be evaluated on its own to give the most
advantageous solution to the problem. However, there is a great deal of experience which can be
brought to bear on the design.
This chapter will discuss some of the parameters used in deciding how to design the well
completion. The subjects to be discussed are: (1) Occurrence and identification of problem
causing sands, (2) Prevention of problems in producing wells, (3) Use of open-hole or gravelpack completion, (4) Screens, ( 5 ) Gravel packing (6) Design of gravel packs for steam injection
processes, and (7) Chemical methods of sand control.
Occurrence and Identification of Problem-Causing Sands
Sand control must be considered if the well produces or is expected to produce sand grains
with the oil. This is the result of producing from an unconsolidated sand formation.
Unconsolidated sands range from the so-called "quicksand," in which the binding forces of the
formation are very minor through the partially consolidated sand which releases some sand during
oil production. Cores are difficult to obtain from"quicksand." Cores from the partially
consolidated sand cores appear to be solid but crumble when handled.
Unconsolidated sands most frequently occur in the shallow tertiary formations which are the
primary target of steam stimulation operations. Sanding is a problem in California, in Alberta,
Canada, in Venezuela, and other locations, but does not occur with sufficient regularitv to be
predicted from location alone. However, the occurrence of sand in nearby wells should alert the
operator to be prepared for sand.
322
Many times a steam stimulation project follows a period of production without stimulation,
and the history of the primary production should indicate whether sanding is likely to be a
problem. However, in wildcat or frontier situations, it may be worthwhile to conduct a test on the
well before deciding on the completion method.
Testing usually involves producing the well at gradually increasing rates until either sand is
produced, or the anticipated rate of production is reached. The test may be made either on the well
or on a particular interval of the formation. If sand is produced, then sand control methods should
be considered.
There are several ways in which logs can be used to predict the tendency of a well to produce
sand. One such method is the combined modules technique, developed by Stein and Hiltiel and is
based on data from acoustic and density logs. The combined modules is compared with results
from a well test; and, assuming that all wells in the field are completed similarly, their values for
the combined modules would show their tendency to produce sand. The lower the value, the less
likely is the well to produce sand. A later report indicated a method for applying the combined
modules to wells of different completion geometries. This uses a comparison of well parameters
(production rate, formation volume factor, permeability,viscosity of reservoir fluid, number of
perforations) to make the comparison with the test well. Another method is the mechanical
properties log developed by Schlumberger which is based on resistivity, density, acoustic, and
neutron logs.
Some wells produce sand free until water cuts increase significantly or until production
problems occur. Consequently, it is necessary to be alert for signs of sanding after production
starts. Some of the production problems that may be observed are: sucker rod pump failure by
sand cutting or seizing; sand fill up in settling tanks or oil-water separators; and erosion of surface
chokes and elbow and tee fittings by flowing sand. A common problem that starts sand flow in a
producing well results when of slots in a slotted liner erode to the extent they no longer prevent
sand entry.
323
cases, adaptation of open hole gravel packing techniques has resulted in the deposition of dirt and
fluid loss additives on the formation and plugging of screens with solids. Some of this plugging is
removed by the first steam that is injected, but this produces thief zones. This is due to both the
solubility of silica in steam and the physical force of steam injection.
Surveys2-S of steam stimulation projects have revealed a number of problems relating to
sanding. These are shown in Table 13.1. To make an intelligent choice for a production program
it is necessary to obtain as much information about the character of the formation as possible. For
example, in some wells the production of sand leads to cavities which may cause the casing to
deform; in other wells the sand body may be fluid enough to reconform and fill the space.
324
Problem
Solution
Sanding;
Reference
(2 1
Replaced adapter
(2 )
Tatum Field
Oklahoma
Sanding
California Fields
Sanding
Sodium aluminate
consolidation treatment
Gravel-packing inside
open hole (no screens)
(5 1
Gravel-Packing
~
~~
Dome Petroleum, on the other hand, investigated4 sand control methods including open hole,
gravel pack, and cased completions with screens. Dome decided that cased hole would work the
best in Lloydminster fields. Conclusions on the use of screens are shown in table 13.2. Dome
concluded that each project had different constraints on the producing wells, which were not
compatible with screen usage (e.g., high gas rates or tough emulsions), but that wire-wrapped
screens inside casing resulted in an effective completion and improved a specific well's cash flow.
They recommended research be performed to find a better method of pumping fluid containing
high sand cuts.
Screens
The simplest, most consistently reliable approach to sand control is the application of
mechanical sand retention devices. Screens, slotted liners, and gravel are used. In this aspect of
sand control technology, the most important single design consideration is the proper sizing of
linear openings or gravel pore space relative to producing formation particles.
Formation grain size is required to properly size mechanical devices for bridging or absolute
stoppage, without undue limitation of flow capacity. This information is obtained through sieve
analyses of representative samples of formation material. Formation samples may be obtained
from sidewall samples or cores prior to completion. Cores provide the best source because
sidewall, bailed, or flowed samples are subject to contamination.
When slotted liners-orwirezwrapped screens a!re used for sand control without a gravel pack,
slot width must be properly sized relative to the forination sand to be excluded. Coberly showed6
that where sands have a broad size distribution, as in California, a slot width of twice the
325
Experience
Screens were not effective on cold production but were successful
with hot production. Saud production and periods of high servicing
OcCutTed occasionally and good production runs were obtained both with and
without a screen in the well. Screens were. not worth the extra cost.
Prairie Lake
Lindbergh
to make the well economic, if the sand problems are not severe. In wells
The use of screens was unsuccessful but a shut-in revived the wells.
Morgan
Note: Both the Lindbergh and the Morgan fields were combination
thermal drives utilizing combustion with added water.
10 percentile diameter of the formation would facilitate bridging and retain most of the sand. The
10 percentile diameter is the theoretical sieve size that will pass 90% of the total formation sample.
Where sands are more uniform and rounded, the recommendation may be to make the slot width
equal to the 10 percentile diameter so that more of the sand will be retained. Slots may be tapered,
with the small dimension on the outside of the liner or screen, so that it will be self cleaning.
Slotted liners7 may be obtained with slots ranging from 0.012-0.50 inches. While they have
a relative low initial cost, there are disadvantages to their use. The smallest slot widths may be too
large for some producing formations. Compared to wire-wrapped screens they have a relatively
small inlet area and, because they are made from low carbon steel, are somewhat subject to
corrosion or erosion.
Available wire-wrapped screen types include: wrapped-on-pipe, grooved type, ribbed type,
and all-welded (Fig. 13.1). The wire is normally stainless steel wrapped on J or K grade pipe.
Slot openings as narrow as 0.003 inch can be obtained. Wire-wrapped screens are less subject to
corrosiodemsion than slotted pipe, and they have greater flow capacity. The wrapped-on-pipe
screen has the least flow capacity, and the all-welded screen has the highest flow capacity.
Gravel Packing
The use of a gravel pack is the most common procedure in the control of sand. The term
"gravel" is a misnomer, as a fine sand is the usual medium. The size of the gravel is selected
similarly to the selection of slot size for screens. The sieve analyses of the formation sand are used
to determine the size of the gravel which will satisfactorily retain the formation sand when the well
is producing oil.
326
.~
IMPRESSED
RIBBED TYPE
FIGURE 13.1. -Types of Screens.7
The median size of the formation sand grains is obtained from the sieve size, which will pass
50% by weight of the sand. The diameter of the sand grain corresponding to that sieve size is
obtained8 from table 13.3. According to work by Saucier,g gravel six times that size will be
most satisfactory (Fig. 13.2) with larger gravel allowing too much formation sand to pass and
smaller gravel reducing the production rate. For example if the formation sand median is 140
mesh, the sand grain size is 0.0049 in. and six times that would be 0.0294 inch. The latter
corresponds to 25 mesh, and referring to Table 13.4,lO the commercial size sand would be 20-40
U.S. mesh.
There are three variations on gravel packing. The first involves packing the open hole, and
this provides the highest production rate if the formation sand is retained sufficiently. The second
327
MESH
u.s.2
Series
2 1R
3
Tyler
Series
Inches
2 It2
3 10
4
10
12
10
14
16
l2
20
u.s.2
Millimeters
Tyler
Senes
20
Millimeters
100
115
150
170
200
0.0280
0.0276
0.0232
0.0197
0.0195
0.0165
0.0164
0.0138
0.0117
0.0116
0.0098
0.0097
0.0083
0.0082
0.0070
0.0069
0.0059
0.0058
0.0049
0.0041
0.0035
0.0029
0.71
0.701
0.589
0.50
0.495
0.42
0.417
0.351
0.297
0.295
0.250
0.246
0.210
0.208
0.177
0.175
0.149
0.147
0.124
0.1104
0.062
8.00
7.925
6.73
6.68
6.73
5.613
4.76
4.699
4.00
3.%2
3.36
3.327
2.83
2.794
2.38
2.363
2.00
1.981
1.68
1.651
1.41
1.397
1.19
230
250
0.0024
0.046
1.168
1.00
0.991
0.84
0.833
270
325
400
270
325
400
0.002 1
0.0017
0.0015
0.0394
0.0390
0.0331
0.0328
16
25
Inches
0.315
0.312
0.265
0.263
0.223
0.221
0.187
0.185
0.157
0.1%
0.132
0.131
0.111
0.110
0.0937
0.093
0.0787
0.078
0.0661
0.065
0.0555
0.055
0.0449
14
18
SIEVE OPENING
series
3 1n
MESH
SIEVE OPENING
24
28
30
35
32
40
35
42
45
50
48
60
60
70
65
80
80
100
120
140
170
200
1.2
ez
.
t
o.8
0.6
BW
0.4
isnU
RECOMMENDED
o.2
0
0
10
12
14
16
18
328
20
0.088
0.074
0.053
0.044
0.037
TABLE 13.4. - Most Commonly Utilized Gravel Sizes for Sand Control
Gravel size,
U.S.Mesh
8-16
12-20
16-30
20-40
40-60
Gravel size,
Screen Gauge,
in.
in.
0.094-0.041
0.066-0.033
0.041-0.023
0.033-0.0165
0.0165-0.0098
0.028
0.02
0.016
0.012
0.008
involves packing the annulus between the liner and the casing. The third squeezes the gravel
through the casing and the formation.
The procedure for gravel packing as outlined by Suman11 for the third variation consists of
six steps:
1. Run a wash tool with cup packers to clean the perforations and circulate out a cavity
behind the cement. Backsurging may be used in low pressure wells that could be damaged by
washing.
2. Prepack perforations by pumping properly sized gravel into the cavity outside the cement
using a viscous fluid and high gravel concentration up to 15 pounds sand per gallon fluid (1,800
kilograms per cubic meter).
3. Apply squeeze pressure and drive fluid out of gravel to compact the perforation pack.
Reverse out excess gravel.
4. Run screen and circulate gravel into place using viscous or non-viscous fluid.
5 . Pull work string with screen wash pipe.
6 . Run production tubing, latch into top of screen and set a production packer to keep screen
pack in place.
Although laboratory tests11,12 indicate that the severe conditions in the well might be
expected to destroy the gravel pack, most installations seem to hold up fairly well. Vigilance is
necessary to be sure that the pack is doing its work, and replacement may be called for if the sand
production increases.
329
--S1
--
THERMAL PACKER
--
SHUIROUTSAFETYJOINT
--
BUNKPIPE
HCTEMP SCREEN
BULL PLUG
The most important element in a successful thermal gravel pack installation is the ability to
retain sand both from the gravel pack or the formation without restricting production. The
retention of sand is just as important as the inability to produce or inject because of plugging.
High Temperature Screens. Wire-wrapped screens and slotted liners are most
commonly used to hold the gravel-pack material in place. A typical wire-wrapped screen for high
temperature sand control application is shown in figure 13.4. Slot plugging has been shown to be
a primary disadvantage of slotted liners and not a consequence of a wire-wrapped screen. This is
330
PIPE BASE
33 1
largely due to the dot opening being paralleled to the direction of expansion. As the slotted liner
expands due to the temperature change, the gravel pack grains are "rolled" or forced into the slot
opening, since the moving liner is in direct contact with the grains.
Slotted liners are generally designed to provide only 2.5% to 3% of their surface area open to
flow. This is required to maintain the tensile strength. With such a small amount of area open to
inflow, even a minor amount of plugging will restrict production. Slotted liners, with slots cut
perpendicular to the axis pose significantly reduced tensile strength and are unsuitable for most
applications.
Wire-wrapped screens offer a greater amount of open surface area, some 10 to 20 times that
of an equal size slotted Liner. Successful screen designs must address one of the two
consequences of thermal stimulation, i.e., resistance to thermally generated forces or allowance for
expansion. Screen designs should permit a large ID screen which will not restrict production or
injectivity, but the OD should be sized to provide a minimal radial clearance of 0.75 to 1.0 in. with
the casing ID. This will facilitate gravel pack removal.
Expansion Joints. Expansion joints are run in the thermal gravel pack system to protect
the screen assembly, gravel-pack extensions, and gravel-pack packer from forces caused by the
screen base elongation and contraction. It is designed to prevent buckling of the blank pipe and
screen assemblies by relieving the thermally generated forces. The system must be designed to
allow for the expansion of the entire assembly. The expansion joint should be rotationally locked
to facilitate removal of the gravel-pack assembly if necessary. l4
To eliminate space-out difficulty, and to ensure that the assembly is positioned on bottom, the
expansion joint should be shear-pinned in the fully expanded position while running the assembly
into the well. The packer is set and the gravel pack is performed with the expansion joint in
position. When the forces caused by thermal expansion exceed the shear pin ratings, the
expansion joint is activated and collapses to relieve the expansion forces.
The shear value of the pins should be adequate to prevent premature shear while landing the
screen on the bottom, but must also be weak enough to shear before any damage occurs due to
screen expansion. Normally a 14,000 lb force shear rating is acceptable.
The high-temperature seal is the same as the high-temperature screen seal. Neither of these
seals are exposed to large pressure differentials and primarily serve as barriers to movement of
solids. Since the function is not that of a dynamic seal, premature wear and failure is not a
concern.
High Temperature Packer: The primary objective of the gravel-pack packer is to create a
seal that will perform satisfactorily for the life of the completion. The high-temperature packer is a
retrievable, seal-bore packer which can be used both as a gravel-pack packer and an
332
injectiodproduction packer. The primary design consideration given to this piece of equipment is
the selection of a packing element system.
Through the use of insulated tubing, heat losses can be minimized and a higher temperature
steam can be injected into the formation. With the ability to reach a 550' F (290' C) downhole
temperature and 2,000 psi (13,780 kPa) sandface pressures, conventional oilfield packing systems
are not adequate. Packing elements that do not fail at high temperature must be used in the packing
system to maintain sealing integrity. Carbon fibers, graphite; and, until recently, asbestos are used
as packing elements in high temperature applications.
The bulk-material-packing-element systems typically utilize multiple packing elements and
low-alloy-steel backup rings with wire mesh extrusion rings. These rings help support the bulk
materials which are not resilient and therefore lack compression-set resistance. Compression-set
resistance, a feature of elastomeric packing elements, is the elastic characteristic that maintains
sealing integrity.
Some packers intended for thermal applications employ a metal-to-metal packing element
consisting of soft alloys such as brass, lead, or aluminum. Such packing systems are very good
gravel migration preventers but are generally capable of withstanding only a 1,500 psi (10,300
kPa) pressure differential. Critical internal packer connections are typically sealed through the use
of threaded metal-to-metal seals. With careful design and manufacture, these seals are effective in
providing proper pressure containment at elevated temperatures.
SUMMARY
Sand control is an important consideration in steam stimulation of heavy oil because much of
the heavy oil occurs in formations with unconsolidated sands. However, because its occurrence is
not easily predictable, each well must be designed based on its specific characteristics.
In some instances, the preferable procedure is to produce the sand with the oil and perform
the separation above ground. However, it is often more desirable to leave the sand in the
formation.
333
Although both mechanical and chemical methods are available for sand control, the
mechanical techniques are usually the choice for thermd wells. These include slotted liners,
screens, gravel packs, and various cmbjnations of these. Special equipment for thermal wells
include temperature screens, expansion joints, and high temperature packers. Vendors such as
those listed in Appendix 13-A are well equipped to design a sand control program for thermal
production wells and should be consulted.
REFERENCES
1. Stein, N. and D.W. Hiltie. Estimating the Maximum Production Rate Possible from
Friable Sandstones Without Using Sand Control. J. Pet. Tech., v. 24, No. 9, September 1972,
pp. 1157-1160.
2. Chu, C. State-of-the Art Review of Steamflood Field Projects. J. Pet. Tech., v. 37,
No. 10, October 1985, pp. 1887-1902.
3. Lea, J.F., and D.W. Middleton. Development of Sand Control Techniques for Steam
Stimulated Wells in the Lindbergh Field. Energy ProcessinglCanada (May-June 1986) 14-17.
4. Matjemson, D.M., and J.A. Sayre. Dome Petroleum's Experience with Sand Control in
Lloydminster Heavy Oil Fields. Pres. at the 38th Annual Technical Meeting of the Pet. SOC.of
CIM, Calgary, Alberta, Canada, June 7-10, 1987. Paper No. 87-38-05.
5. Farouk, Ali, S.M., and .R.F. Meldau. Current Steamflood Technology. J. Pet. Tech.
v. 31, No. 10, October 1979, pp. 1332-1342.
6. Coberly, C.J. Selection of Screen Openings for Unconsolidated Sands. API Drilling
and Production Practices, 1937, pp. 189-201.
7. Suman, G. 0. Jr. World Oil's Sand Control Handbook. Gulf Publishing Co.,
Houston, TX, 1975.
8. Ayres, H. J., and J. Ramos. Guidelines to Sand Control-Part
v. 44,No. 9, September 1972, pp. 82-98.
2. Petroleum Engineer,
9. Saucier; R.J. Considerations in Gravel Pack Design. J. Pet. Tech., v. 26, No. 2,
February 1974, pp. 205-212.
10. Halliburton Services. Sand Control Brochure, 1974.
11. Suman, G. 0. Jr, and R. E. Snyder. Sand Control Design Criteria and Gravel Packing
Concepts. Pres. at the SOC.Pet. Eng. International Tech. Symposium, Beijing China, Mar. 1826, 1982. Paper SPE 10031.
12. Underdown, D.R. and Kamalendu Das. Stability of Gravel Packing Materials for
Thermal Wells. Presented at the SOC. of Pet. Eng. International Symposium on Oilfield and
Geothermal Chemistry at Denver Co., June 1-3,1983. Paper SPE 11793.
334
13. Sacuta, A., D.M. Nguyen, and G. A. Kissel. Stability of Nickel-coated Sands as a
Gravel Pack Material for Thermal Wells. Presented at the SOC.Pet. Eng. California Regional
Meeting, Oakland, CA, Apr. 2-4, 1986. Paper SPE 15058.
14. Weirich, J.B. and T.E. Zaleski. Thermal EOR requires special design for gravel packs.
Oil Gas J., v. 84, No. 46, November 17, 1986, pp. 45-50.
15. Weirich, J. B. and E. T.Zaleski. Gravel Pack Equipment Requires Special Design for
Thermal Recovery. Presented at the Canadian Heavy Oil Association Thermal Well Completion
Seminar, Edmonton, Alberta, Canada, Aug. 11, 1988. Paper No. 3
335
APPENDIX 13-A
SAND CONTROL AND GRAVEL PACK SYSTEM VENDORS
1. Baker Sand Control System
1010 Rankin Road
P.O. Box 61486
Houston, TX 77208- 1486
Telephone: (7 13) 443-7590
2. Solum Oil Tool Corporation
2750 Rose Avenue
Signal Hill, CA 90806-1928
Telephone: (213) 427-5504
3. Halliburton Services
Drawer 1431
Duncan, OK 73536
Telephone: (405) 25 1-3760
4. Guiberson Division
Dresser Industries
P.O. Box 6504
Houston, TX 77265
Telephone: (7 13) 750-230 1
337
CHAPTER 14
STEAM INJECTION PROFILE CONTROL
INTRODUCTION
When steam is injected into a nonuniform formation, the steam enters the formation with a
flow inversely proportional to the resistance encountered. Thus, the greater amount of steam will
enter the more permeable (less resistant) sections of the formation while lesser amounts of steam
enter the less permeable (more resistant) sections. If the formation consists of several sand layers
separated by shale stringers, the steam will seek the most permeable layer and as that zone is
depleted the resistance becomes less until a thief zone may develop where the steam is diverted
from the oil and has little effect on its recovery.
In early completions where an open hole gravel pack was used with the tubing ending at the
top of the formation, the amount of steam entering each layer of the sand was determined by the
permeability of the layer. In addition, water would separate from the steam and migrate to the
bottom because of higher density, and the bottom layer would only be exposed to water instead of
steam. The first improvement was to extend the tubing to the bottom sand layer so that the mixed
waterhteam would contact the lowest layer but also would be forced up the annulus to contact the
other layers.
It is, of course, desirable to know how effectively the steam is contacting the oil, and
considerable effort has gone into measurement of the amounts of steam going into the various parts
of the formation. Originally, the spinners that were used to measure water injection were applied
to steam measurement. A temperature survey was also used. However, more sophisticated
methods using radioactive tracers have become common.
Various methods have been used to improve the injection profile. These depend in part on
the completion method that was chosen for the well. It is also desirable to know the profile at
some distance from the injection, and monitoring wells. These methods also reveal the effects of
steam override in which the steam has risen to the top of the formation because of gravity, leaving
the bottom part of the formation untouched.
This chapter will discuss (1) distribution of steam liquid and vapor in the wellbore,
(2) measurement of steam injection profile including spinners, steam surveys, and radioactivity
methods, (3) methods of improving steam injection profiles, and (4) monitoring wells.
Vendors such as those listed in Appendix 14-A should be consulted to develop a steam
injection profile control program.
Distribution of SteamIWater in Tubing
Steam of less than 100% quality has the possibility of separating into two phases. The
extremes range form mist flow to annular fl0w.l These two conditions are shown in figure 14.1.
338
.......
.... .-.; .
.......
: -.0.Y..
.,
.
e
*
\*
1:.
'I
. . .........
. .. . ::
..
.
:
-........
...,..-.-.La.. -.c I..:::
,***..':.'
'::.
I
e
..:'.
.I
1.
..i.. ':
..
...... ..-.
>) .
.
..........
...... ...
:,:>*' -.:., . .. ,
- . ....
:'*.
:e:.
e
.:.
I.
:
.....
- .;. :.::.: :.
' .* :.:
.... ..........
. .. ..:.?,:.I
*..
.
.......
.....
-(;-.I'. . .:.
:. . . .
*
*:e,.
e
;.
a.
**
\*
. . . . . . ....
1.:~.
MIST FLOW
ANNULAR FLOW
FIGURE 14. 1. - Mist and annular flow regimes for two-phase fl0w.l
For the mist flow, the velocity of the two phases are essentially equal, while for the annular flow
the velocities may be quite different. By the use of radioactive tracers, a determination of the
regime existing in the process may be determined.
The effect of the phase separation is important to the determination of the injection profile
because the water brings much less heat to the formation than the steam. Thus, if a lower zone is
receiving only water, it will be stimulated much less than the upper zones which receive steam.
For this reason, completion practices have changed fro he early open hole completion with the
a completion where the tubing goes to
tubing terminating ahead of the form-ationto be stimula
the bottom of the well and introduces the steam there to rise up the annulus.
339
Spinners
The spinner is a direct approach to measuring the velocity of the steam at different positions
in the we1L2 The tool is basically a free rotating impeller positioned in the pipe by a centralizer.
The steam passing the tool causes the impeller to rotate at a rate proportional to the velocity of the
steam. The rate of rotation is translated into an electrical signal which is transmitted to the surface
through the logging cable.
The configuration of the spinner is shown in figure 14.2. In many respects it is similar to the
spinners used to measure the injection of water into a waterflood injection well, but because of the
higher temperature of the steam environment some changes are necessary. Conventional electrical
insulation and packing materiaIs are not satisfactory and teflon has been used. The high velocity of
the steam results in high rotation rates that require strengthening the impeller and its support.
340
The survey procedure is likewise similar to that for the water injection. However, because of
the higher velocity of the steam, it may be necessary to bypass the steam while the spinner is
lowered through the tubing. A general procedure for a spinner survey is given in table 14.1.
TABLE 14.1. - Procedure for Determining Steam Injection Profile Using a Spinner2
1.
Record impeller RPM at the following stations: 30 to 50 ft below the bottom of the tubing, 5 to 10 ft above the liner
top, 5 to 10 ft below the liner top, and just above the perforations.
Make a down survey at 40 to 60 fpm and stop at the depth where the impeller stops rotating.
3.
4.
Make a down survey at 60 to 80 fpm to pickup. This run will locate the liquid level, if any. If the surveys check, the
tool is withdrawn.
5.
The recordings at the surface are a plot of spinner RPM versus depth. The plot is equated to percent steam entering the
formation by assuming the highest value of RPM is equal to 100% of the steam and the zero RPM represents 0% of the
steam.
Examples of two runs are shown in figure 14.3. Figure 14.3a shows a well where the steam
is being distributed over the face of the formation. Figure 14.3b shows a well that illustrates the
need for profile improvement. Although the perforation interval is about 500 ft, practically all of
the steam is entering a 15 ft section. This also shows the liquid level as some 150 ft above the
bottom of the perforations.
Temperature Surveys
A temperature survey2 during steam injection shows only the lowest point of steam injection
which is indicated by a rapid cooling to the normal well temperature. This start of rapid cooling is
also the top of the liquid level and should check with the value determined by the spinner survey.
The temperature survey is primarily a supplement to the spinner survey in determining the injection
profile.
The temperature measuring tool used in steam wells is the normal instrument adapted for
work in the higher temperature. It has a thermistor with a much wider range of temperatures. An
improved instrument using a platinum resistance thermometer was recently described.3
Used as a supplement to the spinner survey, the temperature survey is shown in figure 14.3.
In figure 14.3a, the temperature is shown to slowly cool off but with no liquid level to cause the
rapid cooling. In figure 14.3b, at the point of injection, the temperature drops but then becomes
constant down to the liquid level where it starts to cool down to the formation temperature.
34 1
DEPTH
700
290
O F
300
310
ATURE
DEPTH
TEMPERATURE
lo00
1100
LIQUIDLEVEL
./
1200
1300
270
280
290
300 310
320
TEMP O F
(b)
342
When the temperature survey is used by itself to determine the profile, it is usually used
during the soak period. It is assumed that the hottest sections received the most steam. Thus
qualitatively, a profile can be obtained by equating injection to sections of elevated temperatures.
343
Xn-133
1-131
100%
Charge
-1
Half-life
10.7 yrs.
5.2 days
8.0 days
Energy level
517 kev
81 kev
364 kev
0.5
99.0
100
% Gamma Emissions
The procedure for the survey using iodide-131 is given in table 14.3. The most common
survey is run with the sodium iodide dissolved in water. If the injected steam is in one phase or in
the mist regime, this is satisfactory. The sodium iodide may also be dissolved in methyl alcohol.
This has been shown8 to tag both the gas phase and the liquid phase. To use this system, it must
be recognized that the velocity of the gas phase is faster than that of the water phase. This results
in two spikes on the record, and they must be interpreted to show how much steam and water are
entering a formation. Methyl or ethyl iodide has been used in an attempt to place the iodide in the
steam phase, but it has been shown5 that each decomposes in the hot steam and forms hydrogen
iodide which is soluble in water.
TABLE 14.3. - Procedure for Running Radioactive Tracer Survey With Sodium Iodide2
1.
Run a background gamma log from the surface to pickup depth. As in the case of the temperature tool, it is usually not
necessary to bypass the steam while running into the well.
2.
Bring the gamma logging tool to a station above the liner top and record the radiation intensity continuously.
3.
Inject the tracer slug. As the slug passes the gamma tool the increased intensity will be recorded.
4.
As soon as the slug is past the logging tool, run a gamma log down to pickup depth.
5.
Immediately make a log up to the surface to detect any possible casing leaks.
6 . The steam injection profile is constructed using the gamma logs. The increase in intensity over background for any
interval is attributable to the tracer being carried to that area by the injected steam. The area under the enhanced gamma
curve less the area under the background log is equated to the total amount of steam being injected. By dividing the
interval into the zones of particular interest and measuring each area, the percentage of the steam going into any zone
can be calculated. A cumulative curve can then be drawn by adding up the percentage and plotting percent of steam left
against depth.
344
The procedure for using the inert gases is shown in table 14.4. Although this is a more
expensive test, it adequately measures the steam going into the various zones. The vapor phase of
the steam is the most important phase from a heat content and volume standpoint. Both krypton
and xenon are thermally stable gases and have no charge so that they do not associate with the
formation. The plating out technique is therefore not applicable. The use of this method on three
types of completions is shown in figure 14.4 where the tubing ends (a) above the perforations,
(b) below the perforations, and (c) within the perforations. For both (a) and (b), measurements are
taken at four positions while for (c) only three positions are necessary.
The velocity can also be used to measure the quality of the steam at the injection area.4 The
wellhead steam flow rate, the downhole pressure, and the vapor velocity are used to develop this
equation:
Steam Quality (in percentage) =
247 x Density of vapor x Distance between detectors x cross-sectional area
Steam flow rate (BPD water equivalent) x Transit time in sec.
In the above equation:
Density is in l b d c u ft;
Distance between Gamma ray detectors is in ft; and
Cross-sectional area is in sq ft.
TABLE 14.4. - Procedure for Running Radioactive Tracer Survey Using Inert Gas Tracers
1.
The dual gamma ray detector is lowered into the well and held stationary at a specified depth.
2.
A slug of radioactive inert gas tracer is injected into the surface steam line.
3 . The gamma ray counts from the surface and downhole detectors are displayed on the chart recorder and stored in the
computer.
4.
The tool is then moved to another depth of interest and another slug of tracer is injected.
5.
The vapor phase injection profile is calculated directly from the tracer transit times across the dual gamma tool. In
some cases, vapor velocity may be too high for.accurate determination of transit time across the tool. In these cases
transit times from surface to tool measured at two different depths are substituted. These transit times are converted to
an equivalent transit-time based on the distance between detectors on the logging tool.
6. The vapor phase injection profile is determined by material balance between total flow and flow at different depths. At
a given depth, the fraction of the total flow by the tool is given by:
Fraction of total flow =
345
ISTOP.
Ie
0
I
FIGURE 14.4.
An example of the combination method4 is a steam injector completed with four 0.25 in. jet
holes per ft over three perforated intervals (Fig. 14.4). The tubing ends within the bottom set of
perforations. The steam packer is sei 5 ft above the top perforations. Both a vapor phase survey
using krypton and a conventional liquid phase survey using sodium iodide were run. Surface to
tool transit times were determined from stops 1 and 3 to determine the 100% flow transit time in
the annulus. It was 0.83 seconds. Stops 2 and 3 showed annular transit times of 1.7 and 1.1
seconds, respectively. Using the equation in table 14.4, the percentage of total vapor flow going
by stops 2 and 3 is 49% and 75%. Thus, the flow is 49% out of the top perforations, 26% out of
the middle set, and 25% out of the bottom set. These results and the results from the sodium
iodide survey are shown in figure 14.5. The liquid phase survey shows almost all of the liquid
exiting the bottom set of perforations. Calculating the steam quality by the above equation shows
that between stops 1 and 3, it was 37%. This can then be used to calculate the heat distribution for
the well, and this is shown in figure 14.5 to be 33%, 20%, and 47% for the top middle, and
bottom perforations respectively,
346
il
STOP 1
782
796
812
81 7
830
853
STOP 2
STOP 3
49%
I
126%
@3%
133%
r
J
20%
1-
TT]25%
47%
347
The unconsolidated nature of the California sands have caused several schemes of profile
improvement to be developed. Notable among them are the limited access method based on
perforation designs and the concentric tube and the parallel tube methods.
EXPANSION JOINT
OUTER TUBING
UPPER
THERMAL PACKER
INNER TUBING
CROSS-OVER JOINT
LOWER
THERMAL
PACKER
EXPANSIONJOINT
349
600'
700'
800'
350
.-
For critical flow to be maintained (at a steam quality of GO%), the ratio of reservoir pressure
(or discharge pressure) to downhole injection pressure must be less than 0.578 (Table 14.5). For
a bottomhole injection pressure of 450 psia, the reservoir pressure must be less than 260 psia. As
formations considered for steam injection are generally low pressure reservoirs, this does not pose
a severe restriction.
To help achieve uniform injection profiles, it is important that the diameter and shape of each
perforation be as uniform as possible. Therefore, perforating guns and charge designs are
carefully selected to provide the most accurate hole size possible.
To assure proper sizing of the perforations, some designs for limited entry injection place the
perforations in the tubing instead of the casing. As these can be prepared on the surface, they can
be more accurately sized.11-12 In addition, limited entry outlets may be designed and constructed
in various ways as long as critical flow may be achieved and maintained. For example, critical
flow nozzles, chokes, or steam deflectors can be used.
TABLE 14.5. - Calculation of the Critical Pressure Ratiol3
When the discharge (downstream) pressure of a perforation is reduced below
a certain critical pressure the mass flow of steam does not increase for any further
decrease in the discharge pressure. The critical pressure ratio (xc), expressed as a
ratio of the critical discharge pressure (P,) to the bottomhole injection pressure (P),
can be calculated as follows.
Critical discharge ratio:
(14.1)
(14.2)
where
P,
C,
C,
fs
w, s
=
=
--1 .oo
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1 .o
1.02
1.04
1.06
1.08
1.10
1.13
1.16
1.20
1.24
1.29
35 1
XC
---0.606
0.603
0.598
0.594
0.589
0.584
0.578
0.572
0.564
0.556
0.547
Critical flow nozzles consist essentially of a short cylinder with a flared approach section. In
general, the length of the straight section tube is about 1/2 times the throat diameter but can be
longer. In addition, a tapered divergent section can be added at the exit of the throat to provide
some degree of pressure recovery of the exiting steam.
Steam chokes are quite similar to nozzles but have no convergent entry or divergent exit
sections. Chokes are commonly used in surface lines to limit steam flow. However, they may be
used downh01e.l~
An example of a steam deflector is shown in figure 14.9. A portion of the steam flow exits
the tubing string through single or multiple slots located concentrically around the tubing. The total
area of the milled slots is equal to the cross-sectional area needed to achieve and maintain critical
flow.
WELDS
MANDREL
MILLED SLOTS
OUTER SKIRT
352
An example of limited entry injection using critical flow chokes is shown in figure 14.10. A
thermal packer is placed between the two target oil zones to keep the injected steam separated once
it has passed through the limited entry outlets. The pipe above the top packer may be bare or
insulated. However, the section of the pipe exposed to low pressure steam in the casing between
the two packers should be insulated to prevent excessive heat loss from the steam inside the tubing.
EXPANSION JOINT
353
SUMMARY
Steam injection profiles are a measure of the effectiveness with which the steam contacts the
oil-bearing formation. They can be measured by spinner surveys, by temperature surveys, and by
radioactive tracer techniques. When the profile is found to be undesirable, it can be changed by
diverting the steam. This can be done by plugging a thief zone or by directing the steam to a
particular zone. In the parallel tube method, tubes lead the steam to particular sections of the
formation. Limited entry techniques may also be used in which perforations are placed opposite to
sections according to the amount of steam required by that section. Monitoring wells are used to
check the temperature profile out in the formation.
REFERENCES
1. Elson, T. D. and K. C. Hong. Distribution of Steam Liquid and Vapor in an Injection
Wellbore. R. F. Meyer, J. C. Wynn, J. C. Olson, eds. The Future of Heavy Oil and Tar Sands,
Second International UNITAR Conf., Caracas Venezuela, February 1982, McGraw-Hill, New
York, pp. 662-665.
2. Bookout, Donald E., J. J. Glenn, Jr., and Herman E. Schaller. Injection Profiles During
Steam Injection. Producers Monthly, v. 31, No. 8, August 1967, pp. 2-8.
5. Nguyen, T. V. and C. E. Stevens. The Use of Inert Gas Radioactive Tracers for Steam
Injection Profiling. Pres. at the SPE California Regional Meeting, Long Beach, CA, Mar. 23-25,
1988. Paper SPE 20035.
6. Nguyen, T. V., B. Davenport, C. E. Stevens, J. C. Reis, and R. S. Millhone. steam
Injection Profiling. U.S. Patent 4,817,713, Apr. 7, 1989.
7. Nguyen, T. V. and B. Davenport, Steam Injection Prbfiling. U.S. Patent 4,793,414,
Dec. 27, 1988.
8 . Crowe, T, L. A Radioactive Steam Comparison Between Krypton-85, Xenon-133, NaI131 (Methyl Alcohol Base) and NaI-131 (Water Base). Pres. at the SOC.Pet. Eng. 60th California
Regional Meeting, Ventura, CA, Apr. 4-6,1990. Paper SPE 20035
9. Burkill, Guy C. C. How Steam is Selectively Injected in Open Hole Gravel Packs.
World Oil, v. 194, No. 1, January 1982, pp. 127-136.
354
11. Hong, K. C. and S. Griston. New Methods for Controlled Injection of Steam Into
Multiple Sands. Pres. at the 61st Ann. Tech. Conf. and Exhib. of the SOC.Pet. Eng., New
Orleans, LA, Oct. 5-8, 1986. Paper SPE 15472.
12. Gates, C. F. and S. W. Brewer. Steam Injection Into the D and E Zones, Tulare
Formation, South Belridge Field, Kern County, CA. J. Pet. Tech., v. 14, No. 21, March 1975,
pp. 343-348.
13. Small, G. P. Steam-Injection Profile Control Using Limited-Entry Perforations. Pres. at
the 1985 California Regional Meeting of the SOC. Pet. Eng., Bakersfield, CA, Mar. 27-29, 1985.
Paper SPE 13607.
14. Hong, K. C., S. Griston and J. W. Ault. Limited Entry Method for Multiple Zone,
Compressible Fluid Injection. U.S. Patent 4,640,355, Feb. 3, 1987.
15. Webb, C. H. Downhole Fixed Choke For Steam Injection. U.S. Patent 4,770,244,
Sept. 13, 1988.
16. Widmyer, R. H. Use of Monitor Observation Wells in the Monitoring and Evaluation of
Oil Recovery Projects. Pres. at SPEIDOE Fifth Symp. on Enhanced Oil Recovery, Tulsa, OK,
Apr. 20-23, 1986. Paper SPE/DOE 14956.
17. Griston, S. Fluid Effects in Temperature Observation Wells. Pres. at the 64th Annual
Tech. Conf. and Exhib. of the SOC.of Pet. Eng., San Antonio, TX, Oct. 8-11, 1989. Paper SPE
19740.
18. Patzek, T. W. and M. T. Koinis. Kern River Steam Foam Pilots. Pres. at the SPWDOE
Enhanced Oil Recovery Symp., Tulsa, OK, Apr. 17-20,1988. Paper SPE/DOE 17380.
355
APPENDIX 14-A
B.
Survevs
1. Halliburton Logging Services
550 Ming Avenue, Suite 190
Bakersfield, CA 93309-840 1
Telephone: (805) 833- 1952
357
CHAPTER 15
EMULSION PROBLEMS IN STEAM INJECTION PROJECTS
INTRODUCTION
A large portion of the oil produced by steam injection process is accompanied by water in the
form of an emulsion. The emulsion problem is much more severe in thermal operations than in
waterflood. The stability of the emulsion varies widely with the crude oil properties and producing
formation. These emulsions are difficult to break and the separation of the two phases consumes
considerable time, money, and efforts and requires specialized equipment and techniques. To
minimize the transportation and process costs, the crude oil purchasers limit the water content of
the oil they purchase to less than 3%. The cost of processing the produced fluids to meet the
purchaser's specification is second only to the fuel cost, and any savings in the treatment cost
significantly impacts the economics of the operation and improves the profits. The treatment costs
can be lowered by studying the treatment problem and selection and use of appropriate treating
methods, equipment, and procedures. This requires a basic understanding of the emulsions and
treating methods. The purpose of this chapter is to appraise the reader about how the crude oil
emulsions are formed and to discuss briefly the methods used in treating and breaking the
emulsion. The equipment used to dehydrate the crude oil is discussed in detail in chapter 16.
Definition of an Emulsion
An emulsion is a stable mixture of two immiscible liquids with one liquid intimately
dispersed in the second liquid in the form of fine droplets. The dispersed fluid is called the internal
or discontinuous phase while the continuous fluid is called the external phase. For an emulsion to
remain stable, the presence of a third component called the emulsifier is necessary. The emulsifier
inhibits the coalescence of the droplets. A common example of an emulsion is mayonnaise which
is a suspension of olive oil in vinegar with egg as the emulsifying agent.
An oilfield emulsion is an intimate mixture of oil and water. When water is dispersed in oil,
it is called a water-in-oil (w/o) or a "normal" or "regular" emulsion. It is the most common form of
oilfield emulsion. When oil is dispersed in water, it is called an oil-in-water (o/w) or a"reverse"
emulsion. Reverse emulsions are more common in steam injection operations due to large
production of water. A w/o emulsion is usually very thick and viscous. A drop of this emulsion
does not disperse when added to water. This is the kind of emulsion that is treated in wash tanks
and heater treaters. The o/w emulsion is usually less viscous than the oil itself and is usually
brownish in color. A drop of this kind of emulsion will readily disperse when added to water.
The following discussion is concerned with w/o emulsions because most oilfield emulsions are of
this type.
358
359
with 50% oil and 50% water will separate more readily than an emulsion wherein the
percentage of the dispersed phase is considerably less than the external phase. This
phenomena has to do with the distance between water particles which directly
influences the force of attraction between droplets. The greater the distance between
particles, the weaker the force of attraction, and the less the probability of coalescence.
IV. Age 0f Emulsion: A fresh emulsion is usually less difficult to treat than an aged
emulsion which has had a chance to set and settle out most of its free water. What
remains, after the free water has settled out, is an emulsion with fewer droplets of
dispersed water; therefore, there is less chance of colliding and coalescing, and
consequently the emulsion is more stable than when it is first produced.
V. w s : Solids such as sand, silt, clays which are often produced with oil and water
tend to concentrate at the oil-water interface and contribute to emulsion stability.
Heat also melts and solubilizes solids such as paraffms which may be acting to stabilize
the emulsion, thereby removing them from the interface.
Heat may also cause the fluid droplets to expand to a point that they will rupture their
captive films, especially in the presence of chemical deemulsifiers.
Chemicd: Dehydration in most steam injection operations is accomplished by the use of chemical
emulsion breakers. This method is popular because the chemicals are easily applied to the
emulsion at reasonable cost, and usually minimizes the amount of heat and settling time required.
(d)
360
The chemical deemulsifiers are surface active agents that migrate to the oil-water interface,
lower the tension between oil and water and thus rupture or weaken the captive film sufficiently to
disperse the emulsifier back into the oil. For the chemical to be effective it must have (a) a strong
attraction to the oivwater interface; (b) an attraction for water droplets with similar charge and bring
them together; (c) the ability to neutralize the emulsifier and promote the rupture of the interfacial
film; and (d) the ability to make solid particles such as clays to water wet and cause them to leave
the interface and be diffused into the water droplets.
It must be remembered that a given chemical emulsion breaker may be excellent and efficient
for one emulsion and entirely unsatisfactory for another. It is important for effective and
economical treatment that the operator enlist the service of the chemical supplier such as those listed
in Appendix 15-A to analyze his specific problems and recommend a suitable chemical
deemulsifier.
Mechanical: If the intensity of the collision between the dispersed droplets is sufficiently high, the
interfacial film can be ruptured faster than the droplets can be reformed, and the emulsions breaks.
This is the principle behind mechanical methods of treatment of emulsion. Mechanical agitation,
ultrasonic vibrations, and centrifuging are some of the ways emulsions are resolved. Agitation
techniques are not widely employed in the oilfield because agitating most oilfield emulsions has the
tendency to make the emulsion worse.
Electrical: The small water droplets dispersed in the crude oil can be coalesced by subjecting the
water in oil emulsion to a high voltage electric field. Under the influence of the electric field, water
droplets become mutually attractive due to induced electric charges on the drops. The effect of
these charges and the resultant forcible collision of drops overcome the stabilizing films, and a
rapid separation of oil and water follows.
Emulsion Breaking
A. Oil-in-Water Emulsions
These will be discussed first because they are more prevalent in steam injection operations.
A stable oil-in-water emulsion is a colloidal system of electrically charged oil droplets surrounded
by an ionic atmosphere. Steamflooding encourages the formation of these emulsions because of
the high water to oil ratio, shearing in the formation, and the presence of emulsifying agents such
as silt and clays in the formation. There is nothing that can be done to prevent the formation of oilin-water emulsions.
Some emulsification is also caused in surface equipment by pumping wells off, gases coming
out of solution, and centrifugal pumps in the gathering system. In the steam injection operation,
the pumps are often pounded in order to keep the wells pumped off and to produce the maximum
amount of oil. The emulsions, however, are made worse as a result. The principal gas that comes
out of solution in a steam injection operation is carbon dioxide. Since the release of carbon dioxide
causes water to become acidic, it encourages the formation of oil-in-water emulsions. Therefore,
the only thing that can be done to reduce the formation of oil in water emulsions in the wells and
production gathering system is to design the system such that the fluids are in laminar flow and
avoid centrifugal pumps wherever possible.
Oil-in-water emulsions can be broken by chemical and/or physical methods. Chemicals are
commonly used to treat these emulsions and to enhance mechanical treatment. In breaking
emulsions, the stabilizing factors must be neutralized to allow the emulsified droplets to coalesce.
The accumulated electrical charges on the emulsified droplet are neutralized by introducing charges
opposite to that of the droplet. Chemical emulsion breakers provide this opposite charge. The
dielectric constants of water and oil cause emulsified oil droplets to carry negative charges.
Therefore, to destabilize an oil-in-water emulsion, a cationic (positive charge) emulsion breaker
should be used.
The treatment of oil in water emulsions is normally divided into two steps:
1. Coagulation. This is destruction of the emulsifying properties of the surface-active agent
or neutralization of the charged oil droplet.
2. Flocculation. This is agglomeration of the neutralized droplets into large separable
globules.
Both inorganic and organic oil in water emulsion breakers are available from chemical service
companies. Organic emulsion breakers are the preferred emulsion breakers in Kern River Field
because part of the produced water is usually discharged into an agricultural canal, and there are
limitations on the amount of inorganic materials in that water. "Inorganic" emulsion breakers
contain inorganic salts such as zinc or chromium or iron which increase the specific gravity of the
water and encourage separation of water and oil.
The selection of a chemical for breaking oil-in-water emulsions must be done with care to
make sure that this chemical is compatible with the oil treating chemical and does not have any
undesirable side effects. One good rule-of-thumb is to use products from the same chemical
company for both treatments. The chemical companies are very aware of incompatibility problems
and can supply products that work together effectively.
The preliminary testing for oil-in-water emulsion breakers is the bottle test in which the
service representative places samples of the emulsion in glass bottles, adds various doses of his
chemical, shakes the bottles and observes the effect of the chemical on water clarity. The
following pitfalls are present in this kind of bottle testing:
1. The sample should be fresh and tested immediately at the site. Do not believe any results
on a sample that has been shipped. Several days of agitation on a truck or train serve to
break most emulsions.
362
2. The chemical should be used neat (100% straight out of the bottle) even though only very
small amounts are being used, and measurement is difficult. Dilution of oil-in-water
emulsion breaker chemicals affects their pH and decreases their effectiveness.
3. The amount of agitation used in the testing should simulate the amount of agitation
actually found in the system. Emulsion breakers which work only after prolonged hard
shaking will not usually work in a system that only has s short run of pipe that is in
laminar flow.
One final precaution: The effectiveness of the chemical program is generally composed of 30
to 40% the effectiveness of the chemical and 60 to 70% the effectiveness of the chemical company
service representative. Chemical companies all have effective chemicals, however, there will be
large variations in the abilities of the representative who provides the service.
B. Water-in-Oil Emulsions
Water-in-oil emulsions are viscous, concentrated emulsions which are formed when oil
comes into contact with water and solids. It has been observed in many steam injection operations
that breaking the obvious oil-in-water emulsion that is produced as a byproduct of steamflooding
reveals a "hidden" water-in-oil emulsion which then must be treated in order to be able to sell the
oil. Water-in-oil emulsions are stabilized by formation fines and asphaltenes.
Water-in-oil emulsions are also broken by a combination of time, temperature, and chemical.
The chemical treatment of this kind of emulsion is directed toward destabilizing the dispersed water
droplets and solids or destroying the emulsifying agents. The mechanical facilities which are
usually wash tanks or heater treaters are critical here. The facilities and the chemical must work
together for cost-effective separation. There is frequently a close balance between heat and
chemical. The field engineer needs to study carefully the cost of heat vs. the cost of the chemical in
his own system.
It should be emphasized that no two oilfield emulsions are alike. The chemicals and
procedures used to treat the emulsion produced from one field may not work on an emulsion from
a different field. Within the same field, emulsions can vary from well to well and from year to year
during the life of the field.
The selection of a chemical to break water-in-oil emulsions must be done with the same care
as with oil-in-water emulsions. Bottle testing is again used as a guide to selecting the chemical.
Here the oil phase emulsion is poured into bottles, allowed to come to the temperature of the
system, chemical added and the water drop recorded over a period of a few hours. At the end of
the test a sample of oil from near the oil-water interface should be tested to determine how much
water is remaining, that is, whether the sample is "dry" and the oil can be shipped.
The following precautions apply to this kind of testing:
1 . Fresh samples should be used within a few hours of sampling to get reliable results.
363
SUMMARY
There are two kinds of oilfield emulsions: oil-in-water emulsions, in which water is the
continuous phase and water-in-oil emulsions, in which oil is the continuous phase. Both kinds of
emulsions are broken by a combination of time, temperature, agitation, and chemical.
The selection of emulsion breaking chemicals should be accompanied by bottle testing which
carefully simulates system conditions. The emulsion breaker chemical is only as good as the
chemical service company representative who administers the chemical program. If the oilfield is
remote, and no service will be provided, the company should make provision for training of its
own personnel in chemistry and emulsion breaking.
GENERAL REFERENCES
1. Bansbach, P. L. and D. V. Bessler. Cold Treating of Oilfield Emulsions. Pres. at the
22nd Annual Southwestern Petroleum Short Course, Lubbock, TX, April 1975,.pp. 241-249.
2. Smith, H. V. and K. E. Arnold. Crude Oil Emulsions in Petroleum Engineering
Handbook, edited by H. B. Bradley. Chapter 19, SOC.of Pet. Eng., Richardson, TX,1987, pp.
19-1 to 9-15.
APPENDIX 15-A
EMULSION TREATMENT CHEMICAL SUPPLIERS
1. Nalco Chemical Company, VISCO Division
4851 Stine Road
Bakersfield, CA 933 13
Telephone: (805) 834-6590
2. Tretolite Chemicals
Petrolite Oil Field Group
333 Palmer Drive
Bakersfield, CA 93309
Telephone: (805) 397-5105
3. Baker Performance Chemicals Inc.
3920 Essex Lane
Houston, TX 77027
Telephone: (7 13) 599-7400
4. Eisenman Chemical
3900 Essex Lane
Houston, TX 77027
Telephone: (7 13) 439-8764
365
CHAPTER 16
SURFACE PRODUCTION FACILITIES
INTRODUCTION
Aside from water treatment and steam generation systems, a steam injection project contains
several types of hardware for the collection, treatment, and disposal of the produced fluids.
Collectively, this hardware is called the surface production facility (see Fig. 16.1) and includes a
production gathering system and the oil dehydration and storage system. A casing gas vapor
recovery system is used to recover and process the produced vapor from the well casing and is also
found in many large steam injection projects. The pipings, vessels, and other items associated
with the vapor recovery system are not part of the production facilities.
In steam injection operations, more than a third of the nonfuel operation costs result from the
operation of the production facilities. History of steam injection operations reveal many projects
have failed as a direct result of an improperly designed and operated surface facility. The proper
design and operation of the surface equipment is of critical importance to the success of any steam
injection operation. The objectives of this chapter are to present an overview of the design and
operation features of various production equipment and their operational problems. A list of
production equipment vendors is presented in Appendix 16-A. The service of these or others
should be enlisted in the selection of equipment.
PRODUCTION GATHERING SYSTEM
The production gathering system is nothing more than a multitude of pipelines and associated
pumps used to transport the produced fluids to the treatment plant. The automatic well test system
(AWT) is an integral part of the gathering system. The system may contain one or more lines that
branch off to each well or it may consist of a separate line to each well. as distribution dictates. The
complexity of the production gathering system network depends on the size of the project.
A schematic of a typical production gathering system network is shown in figure 16.2. It
consists of (1) a production manifold that receives the fluids from the production well and
transports them to the oil dehydrations facility directly or through a main gathering manifold; (2) a
series of lines from individual wells, which route the production either to the production manifold
or to a common test line or to a purge line through an automatic three-way diverter valve; (3) a
common test line that gathers the flow from individual wells and transports it to a test tank for the
purpose of gauging and testing the oil and water production; and (4) a purge line that also serves as
an observation line. In a cyclic operation, the production manifold also serves as the steam
injection manifold for a certain period of time when the wells are being steamed. Similarly, the line
to each well from the manifold is a dual purpose line used for steaming the well for a certain period
366
L9
PRODUCTION
AWT
OILDEHY~RATIONPLANT
368
at the lease line and metering of the oil. If such is the case, then separate facilities must be
constructed at the lease line to separate oil and water. The lease facilities usually consist of a
primary and secondary wash tank and one or more shipping tanks. The oil is skimmed from the
primary to secondary and then to the shipping tank. When the shipping tank is full, it is manually
gauged, prior to shipping to the oil dehydration plant. A positive displacement pump is utilized to
pump the oil back into the production gathering system. The water is waterlegged into the drain
tank. Any oil collected at the top of the drain tank is skimmed and pumped back into the
production gathering system using positive displacement pumps. Depending on the size of the
project, several drain tanks (each with a capacity of 500 to 1,000 bbl) are utilized to hold the water
from the lease tanks. The water is also discharged into the production gathering lines using
centrifugal pumps.
Since the produced fluids are in emulsified form, an emulsion breaker (a chemical that aids
the breakup of the emulsion) is injected into the gathering line to resolve the emulsion. High
temperature, longer residence time, and agitation all aid in the breakup of the emulsions. Since all
of the parameters favorable to the breakup of the emulsion are present in the gathering system,
usually only a small amount of chemical is added to the system to initiate the breakup of emulsion.
The emulsion breakers feed points are usually located downstream of the AWTs. Only a minimum
amount of chemical needed to initiate the breakup of the oil-in-water emulsion must be added and
full advantage of the residence time in the line should be taken. Careful attention must be paid to
the amount of treatment chemicals used because overtreating would inhibit the resolution of waterin-oil emulsion formed in the oil dehydration plant.
Use of single reverse emulsion breaker throughout the gathering system is recommended to
permit the ease of administration and to avoid potential incompatibility problems which could exist
if two or more chemicals were used. Emulsions and emulsion breaking are discussed more fully in
chapter 15.
369
In steam injection operations, data obtained from AWT units are utilized as the number one
diagnostic tool to determine the changes in reservoir characteristics and to establish well workover
needs. These data are also utilized in many instances to gauge the success (both technically and
economically) of the steam injection process.1 Hence, it is important that data gathered from AWT
units be accurate and reliable. To illustrate how vital these data are, consider a typical waterflooded
reservoir in the United States.
Many U.S. waterflooded reservoirs are currently producing at high watercuts, and most of
them contain wells that are capable of producing in excess of 500 barrels of fluid per day (bfpd).
For such wells, a small error in the watercut will result in a large percentage error in the oil
production rate. For example, a well producing 600 bfpd at a 95% watercut is producing 30
barrels oil per day (bopd). If the watercut is measured at 94% or 1% less than actual value, the
calculated oil rate would be 36 bopd. The 6 bopd difference represents a 20% error in oil
production rate. Not only is the oil production rate of this well incorrect, other wells in the system
will have to make up the difference if an allocation method is used. Since this is usually the case,
this may result in the continued production of uneconomic wells. Unfortunately, it is sometimes
very difficult or uneconomical to obtain accurate production data because of the time consuming
nature of the tests and difficulties in the calibration and operation of metering devices and sampling
equipment. The problem is even more acute in steam injection operations. The hawking of a
variety of well test hardware designs, as well as a wealth of recent AWT related papers, is a
testimony to the difficulties in obtaining accurate and reliable production data.
370
COMMON LINE
TO GATHERING
TEST LINE
OBSERVATION LIN
CAPACITANCE
PROBE
FLOWMETER
37 1
system, a number of factors should be considered in determining the most desirable method of
treating the crude oil to meet sales requirements. Some of these factors are (1) tightness of the
emulsion, (2) specific gravity of the oil and produced water, (3) corrosiveness of the crude,
produced water and casing head gas, (4) scaling tendencies of the produced water, ( 5 ) quantity of
fluid to be treated and percent of water in the fluid, and (6) equipment operating pressures.
In steam injection operations, chemicals and heat are used to destabilize the emulsion. The
addition of heat has a dual effect, Le., the specific gravity and viscosity of the oil both decrease as
the temperature is increased. This combination of decreased specific gravity and viscosity of the
oil amplify the differences in densities between oil and water. This difference in densities aids in
the separation of water from oil.
A variety of equipment is used in steam injection operations to dehydrate the oil and to
remove sand. These include: freewater hockout vessels, wash tanks, heater treaters, and storage
tanks. The sediments from the tank bottoms are cleaned using desanders. In addition, accessory
equipment that is not basically necessary to the operation of the dehydration plant is also included
as part of the dehydration plant. These include the lease automatic custody transfer (LACT) units
and cooling towers to cool the produced waters for treatment in the depurators.
Not all of these units are found in all operations. The amount and type of equipment needed
depend on the volume of the fluids to be handled, severity of emulsion and sanding problems,
waste water specification, etc. Several different types of equipment or systems may satisfactorily
resolve an emulsion, but particular types of equipment or systems may be superior to others
because of basic considerations in design, operation, initial cost, maintenance, and operating costs,
and performance.5 Effort should be made to select the minimum number of pieces of equipment
for each treating system to lower initial and operating costs.
Basic descriptions of various pieces of equipment used in a steamflood dehydration plant are
as follows.
372
Depending on the volume of the produced fluids, one or more FWKO vessels are used in the
process. The vessel must be large enough to permit the separation of water from oil in reasonable
time. Both vertical and horizontal configurations are utilized. Schematics of horizontal and vertical
FWKOs are shown in figures 16.4 and 16.5, respectively. In tables 16.1 and 16.2, the size and
capacities of the horizontal and vertical vessels are givenP
The residence time in the FWKO ranges from 3 to 6 hours. The free water is removed from
the vessel through a water dump valve (usually controlled by a capacitance type interface
controller). This prevents the dumping of oil out of the water dump valve by keeping the
emulsion-freewater interface level at a predetermined set point. Normally, there is no oil discharge
valve and the oil, emulsion, and gas is flowed through a connection in the top of the vessel on to
the treating system.
Variations of freewater knockouts are often used to accomplish specific treating purposes.
Flow splitters are one variation of the freewater knockout designed for specific applications. The
flow splitter vessels have two or more separate compartments to collect the oil before it is
discharged from the freewater knockout. The oil is discharged in multiple streams, depending on
the number of separate oil compartments in the vessel. The free water is discharged from the
vessel as it would be in a standard horizontal freewater knockout using an interface level control
and water discharge valve. The oil flows over adjustable weirs into separate compartments and
each is equipped with an oil level control and separate oil discharge valve. Any gas separated is
discharged out the top of the vessel through a separate gas line. The purpose of these vessels is to
ratio the oil out of the freewater-knockout to different pieces of treating equipment. This type of
vessel would have application where the oil volume, produced and passed through the freewater
knockout, was so large that it could not be handled by a single piece of treating equipment.
Therefore, the flow splitter is used to ratio the oil flow to several treaters. Or, the operator may
already have several items of treating equipment on hand and wants to use them in a large central
treating installation where the flow has been increased due to a steamflood.
Many steamflood operators use a cone-bottom FWKO. This design is used if sand
production is anticipated to be a major problem. Normally, 45" and 60" cones are used as the
bottom head of the vessel. Water jets are usually used to dislodge and flush the sand from the
vessel.
Emulsion breakers and antifoaming agents are usually added upstream of the FWKO to
partially break the emulsions. The gross emulsions leaving the FWKO usually consist of 40%
water and 60% oil and, depending on the operation, the water content of emulsion may range from
10% to 50%. The water leaving the FWKO vessel usually contains about 100 to 400 ppm of oil
and this oil is usually recovered in a depurator.
373
is" MANWAY
.....-....-....- .....
BAFFLE
VALVE
SAFETYVALVE
INTERFACE FLOAT
16" MANWAY
374
~~
~~~~
~~
Volume of
fluid in vessels
Specifications
Size
Dia. x
length, ft
3 x 10
4 x 10
6 x 10
6 x 15
8 x 15
10 x 15
10 x 20
10 x 30
10 x 40
10 x 50
12 x 40
12 x 50
W.P.
psi
Inlet &
Outlet
Conn., in.
50
50
50
50
50
50
50
50
50
50
50
50
3
4
4
4
4
6
6
6
6
8 In - 6 Out
8
8
Standard
water
valve,
in.
Drain
Conn.,
in.
2
3
3
3
3
3
4
4
4
4
6
6
6
6
6
6
6
4
4
4
4
4
4
4
VOl.
Water,
bbl
Vol.
Oil,
bbl
Total
Vol.,
bbl
6.9
12.5
29.1
41.7
76.7
124.0
159.0
228.8
296.8
366.7
43 1.8
532.5
5.4
9.3
23.0
32.8
60.5
102.9
131.2
188.1
243.7
300.5
371.0
471.7
12.3
21.8
52.1
74.5
137.2
226.9
290.2
416.9
540.5
667.2
802.8
989.7
Volume of
fluid in vessels
Specifications
Size
Dia. x
length, ft
4x
6x
8x
10 x
Standard
water
valve,
in.
Approx.
wt. Ib.
W.P.
psi
Inlet &
Outlet
Conn., in.
10
10
50
50
4
4
3
4
3
4
2,000
3,500
10
50
6,300
50
4
4
10
8,700
Drain
Conn..
in.
VOl.
Water,
bbl
10.2
24.1
45.4
75.0
Vol.
Oil,
bbl
11.2
25.2
Total
Vol.,
bbl
21.4
49.3
44.8
90.2
69.9
144.9
The FWKO is relatively maintenance free. The main problem which occurs is the buildup of
solids on the bottom of the tanks. The tanks must be cleaned periodically, depending on the
severity of solid buildup.
Settling Tanks
Some steam injection operators use settling tanks instead of heater treaters to dehydrate oil.
Various names are given to these settling tanks and some of the most common are gun barrels,
wash tanks and dehydration tanks. Designs for these tanks differ in detail from vendor to vendor
and field to field. The essential components of a settling tank include: a gas separation chamber or
gas boot, a downcomer, water wash section, spreader or deflector plates, and oil collectors. The
375
internal design of the wash tanks vary from operator to operator and depends on the overall
process selected for the facility, emulsion properties, flow rates, and desired effluent qualities.
Many settling or wash tanks employ heat to aid in the treatment process. Heat can be added to the
liquid by an indirect heater, a direct heater, or any type of heat exchanger.
A direct fired heater, also called a "jug heater," is one in which the fluid to be heated comes in
direct contact with the immersion type heating element or heating tube. The units are normally
constructed so that the heating element can be removed for cleaning, repair, or replacement. Direct
heaters are usually used to heat low pressure noncorrosive liquids.
An indirect fired heater is one in which the fluid passes through tubes immersed in a bath of
water or other heat transfer medium. The heat transfer medium, in turn, is heated by an immersion
type heating element similar to the one used in a direct fired heater. Indirect fired heaters are more
expensive than direct fired heaters and cost more to maintain. Indirect fired heaters are generally
used to heat corrosive or high pressure fluids.
Schematics of a wash tank, used by one large California steam injection operator to dehydrate
the oil, is shown in figure 16.6. These are 10,000 barrel welded mild steel tanks and have
internals consisting of two stacks of heating coils and a spreader bar. The tanks are maintained at
190" F by circulating steam through the heating coils.
GAS LOOP
m-l
GAS BOOT
E' {
I 1
SKIMMER TROUGH
OIL OUTLET
I
,THERMAL
PLATES
(TYPICAL)
'
PERFORATED
DEFLECTOR PLATES
TO DEPURATOR
376
The function of the gas boot is to separate gas from the incoming fluid stream. The emulsion
enters the gas boot, where a momentum change causes gas to separate. The gas boot is usually a
simple piece of pipe as shown in figure 16.6. In some units, they are more elaborate and contain
nozzles or baffles to help separate gas. The oil and water then passes through the gas boot to enter
the bottom of the tank through a spreader device. An oil-water interface is maintained in the tank
by means of an outside water leg on the tank or through the use of an electronic probe. Electronic
probes are most often used in heavy oils to sense the interface and automatically trigger a water
dump valve.
A downcomer directs the emulsion below the oil-water interface to the water wash section.
In large tanks, such as the one shown in figure 16.6, a spreader is used to distribute the flow over
the entire cross-section of the tank. The function of the spreader or the deflector plate, which
contains small holes or slots, is to disperse the emulsion into very small droplets which are
subjected to washing action as the droplets float to the top of the tank. This permits the water
released from the emulsion to drop out. If the droplet size is too big, there will not be sufficient
"washing" action and the oil will take longer to dehydrate.
The whole key to efficient wash tank operation is maintaining a stable oil-water interface
which is well above the accumulated sediment. The plant operator should check the location and
appearance of oil pad every hour and adjust the pad as needed to maintain a good interface.
However, all changes should be made slowly.
Maintenance of wash tanks consists of keeping the heating coils in good condition and
cleaning the accumulated tank bottoms every 3 months. Since heat aids in the breakup of the
emulsion, it is important that heating coil surfaces be film free to promote better heat transfer. Oil
cuts in the wash tanks must be monitored every 2 hours or so with the objective being to have a
water and sediment free oil (less than 3% basic sediment and water [BS&W]) under normal
operating conditions. In large operations, the oil from the first wash tank is skimmed to a second
wash tank and then to the shipping tank. This is to accommodate any upsets and to ensure that the
dehydrated oil meets the pipeline specification.
Heater Treater
Most steam injection operators use heater treaters to break emulsions. A heater treater is a
pressure vessel that operates on the same principle as the wash tank. A heater treater can heat the
emulsion to temperatures above the boiling point of water, without boiling the water.
The heater treater (also called an emulsion treater) combines all of the various pieces of
equipment used to treat an emulsion in one vessel. The heater treater is designed to include in one
unit any or all of the following elements: oil and gas separator, freewater knockout, heater, water
washing section, filter section, stabilizing section, heat exchanger and electrostatic field.
377
Although the heater treaters can be operated at atmospheric pressure, they are often operated
under low working pressure (typically from 10 to 50 psig), depending on the construction of the
vessel and the type of controls used. Heater treaters are available in both vertical and horizontal
configurations. The emulsion, along with the treating chemicals, enters the vessel near the top and
flows downward.
Vertical Treaters
One of the most commonly used crude oil emulsion treating system is the vertical heater
treater. A variety of designs are available. Three such designs are shown in figures 16.7, 16.8
and 16.9.
In the design shown in figures 16.7 and 16.8, the emulsion enters near the top of the treater
into a gas separation section. In figure 16.9, the emulsion enters near the bottom and flows
upward into a gas separation. This section must have adequate space to separate gas from the
liquid. The gas separation section usually has an inlet diverter and a mist extractor.
The emulsion flows through a downcomer to the bottom portion of the treater, which serves
as a FWKO and water wash section. This section should be sized for sufficient retention time to
allow the free water to settle out. This will minimize the amount of fuel needed to heat the liquid
rising through the heating section.
The oil and emulsion flows upward through the water which serves as a washing medium.
The water is heated by a fire tube projecting into this compartment. After leaving the heated water
wash, the emulsion rises into a settling space where water is broken out of the emulsion, settles
out, and falls back into the water wash. The clean oil rises and passes through the oil outlet. The
water in the bottom of the unit passes out a water outlet to the disposal system. The oil-water
inteiface level is controlled by an interface controller or an adjustable external water leg. The
operating pressure is kept well above the saturated steam pressure at the operating temperature to
prevent the water from boiling.
Horizontal Heater Treater
The external and internals of a typical horizontal heater treaters are shown in figures 16.10
and 16.1 1. The horizontal treaters operate much like vertical treaters. The incoming emulsion,
with chemical added, enters the front section of the treater where gas is flashed. The liquid flows
downward to near the oil-water interface where the emulsion is water washed and the free water is
separated. The oil and emulsion rises past the fire tubes and flows into a second section of the
vessel where water is broken out of the emulsion and falls back into the bottom. The clean oil is
removed through the outlet. The oil-water interface in the inlet section of the vessel is controlled
by an interface controller, which operates a dump valve for the free water.
378
GAS OUTLET,
RELIEF VALVE
SAFETY HEAD
(SELECTED ASME CODE
SIZES ONLY)
MIST EXTRACTOR
GAS EQUALIZER
EQUALIZING LINE
GAS SEPARATING
EMU1SlON CONDUCTOR
\
FLAT BAFFLE &
FILTER SUPPORT
CONICAL BAFFLE &
FILTER SUPPORT
FILTER
SECTIONS
DOWNCOMER
HOOD
NORMAL INTERFACE
FR EEWATER
KNOCKOUT BY-PASS 1
OIL VALVE
OIL
OUTLET
Bk
WATER VALVE
II
379
WATER
GAS OUTLET
SAFETY HEAD
(SELECTED ASME CODE
SIZES ONLY)
GAS EQUALIZER
INLET+
EQUALIZING LINE
REMOVABLE ADJUSTABLE
GAS SEPARATIN&
SECTION
WATER
OIL
-------
LEVEL
It
FIREBOX
HOOD
EMULSION
CONDUCTOR
NORMAL INTERFACE
- ----- LEVEL
EMULSION
WATER VALVE
CONNECTION
380
RELIEF
SAFETY HEAD
(SELECTED ASME CODE
SIZES ONLY)
vALvE!L.A
7
GAS SEPARATING
SECTION
ADJUSTABLE
/I
38 1
eater.^
RELIEF
SAFETY
l-7
J
I
INSTR.
NELGAS8
SUPPLY
GAS OUTLET/
TO SALES
THERMOSTAT \
/ \\
GAS OUTET
TO GAS PIT
WATER
OUTLET
INLET
ouin
GAS
OUTLET
REUEF
382
Generally, horizontal treaters have a larger settling or coalescing section than vertical treaters
and, therefore, are often used to treat heavier oils. The larger sizes of the horizontal treaters aIlow
for larger f i e boxes and therefore the capability to handle larger volumes of fluids. Also, the oilwater interface area in a horizontal treater is larger, and this permits more rapid settling of oil and
water particles within the vessel. One possible disadvantage of horizontal treaters is that they
require more floor space than vertical designs.
The dimensions of a typical horizontal heater treater used in a steamflood operation are 12 ft
in diameter by 30 ft in length and can process 540 barrels of fluid. These units are usually
operated at 60 psig and 260' F.
Electrostatic Treaters
Electrostatic treaters (see figure 16.12), often called chemelectric or electrochemical treaters,
are similar to horizontal heater treaters except that high voltage, alternating current electric grids are
added. Electricity is often an effective means of breaking emulsions. As the heated emulsion rises
through the electric field, the water droplets are given a charge. When charged, the droplets move
about rapidly, colliding with each other with enough force to coalesce into larger and larger drops
until they settle out. The clean oil continues to rise to the top of the vessel where it is collected and
removed.
The electrical system consists of a transformer and electrodes which are suspended, one
above the other, in one section of the unit. The electrodes are usually arranged so their crosssectional area is perpendicular to flow of the fluids.
With electrostatic treating units, it is possible to use lower heating temperatures than are used
with the other units. It is also sometimes possible to treat the emulsion at the temperature at which
it is produced with no additional heat. Not only do lower temperatures require less fuel to fire the
heater, they also lessen problems with scale and corrosion formation and, most importantly, reduce
the chance of gravity and volume losses that occur at higher temperatures. One disadvantage of an
electrostatic treater is that the grid is prone to plugging by solids. This would lead to a particularly
difficult situation with oil from unconsolidated formations such as those found in California.
Practically all oil requiring treatment contains a certain amount of entrained sludge and solids,
and a bottom drain is provided on most heater treaters to permit removal of these materials. The
presence of a large quantities of solids can be a problem in any heater treater, but particularly for
those electrostatic units where plugging is frequently observed.
Emulsion Breaker Treatment
A good emulsion breaker treatment is vital to the success of most heater treater operations,
because the emulsions are generally severe enough that heat alone will not resolve the emulsion in a
383
GRID TRANSFORMER
EMULSION
INLET
EMULSION SPREADER
GAS
OUTLET
reasonable amount of time. Careful bottle testing (see chapter 15), will identify chemicals which
will provide fast water drop and dry oil at the interface. It must be remembered, however, that the
results of bottle testing cannot always be taken too literally. In the cases of unconsolidated
formations, such as Kern River, CAYcontrol of solids and a good interface are vital to the success
of the separation process, be it wash tanks or heater treaters. A small amount of emulsion
remaining at the interface after the first day of operation will build up on the subsequent days and
eventually upset the treater entirely.
As with all oil and water separation processes, careful monitoring is the key to success. Oil
cuts (BS&W) and oil counts should be monitored every 2 to 4 hours. In addition, pressures,
temperatures, chemical usage, and fuel consumption should be monitored daily for cost-effective
operation.
384
HYDRATION FACILITIES
TRANSRR PUMP
A
A
large operations, the tank bottoms are usually cleaned once every 3 months. The tank bottoms
usually contain considerable amounts of oil and can be recovered by cleaning the sediments. The
profits realized from the sale of this oil usually pays for the cleaning and disposal costs of the tankbottom materials.
The oily bottom sediment is first hosed into a concrete pit. A slurry pump, located at the
bottom of the pit, is utilized to transfer the sediment to a process tank. The contents of the process
tanks are thoroughly mixed and transferred to a sediment tank through a set of hydrocyclones. The
solids, which drop out in the hydrocyclone, are further treated. The liquid portion is returned to
the process tank. The liquid portion is recycled several times through the hydrocyclone until the
fluid returning to the process tank is nothing but water and oil. The most common type of
hydrocyclone used to separate the fluid from sand is the cyclone desander.
385
Cyclone Desanders
Cyclone desanders (hydrocyclones) are commonly used in steam injection operations to
separate sand particles from the liquid. Cyclone desanders are conical-shaped devices that make
use of centrifugal force to separate the solid from the liquid.
The fluid enters through a tangential opening in the large end of the cone-shaped housing.
This results in a whirling motion or "cyclone" action of the fluid. The fluid and solid particles
move downward in a spiral pattern forcing the heavy particles to move toward the outer perimeter
of the cone. Gravity forces these particles to slide downward and force out the apex. The
remainder of the fluids move toward the vacuum created at the center of the cone, and are drawn
off at the top as overflow. The overflow contains particles less than 30 microns in size and sends
them to a sediment tank.
The sediment accumulated in the sediment tank is pumped to a decanting centrifuge using a
cavity pump. A light solvent is usually injected into the pump suction to aid in the recovery of oil.
The decanting centrifuge is a rotating cone-shaped drum, which turns at high speed and thus
effectively separates the solids. An augur within the rotating drum scrapes the particles off the side
of the drum and forces them out of the discharge port. The fluid in the center of the drum is
discharged through the liquid discharge port. The cleaned solids are hauled to a disposal site.
Operation, Maintenance and Monitoring Considerations
Neither the equipment nor the operation of a steam injection oil dehydration plant are
particularly complicated. However, careful monitoring of fluid and equipment is critical for proper
plant operation. In large steam injection operations, where the dehydration plant spreads over a
large area, the operation is manpower intensive. Some of the operating and maintenance problems
associated with dehydration plant include: malfunction of burners in the heater treater; deposition of
soot on the walk of the fm tube; deposition of scales and solids on the heating tubes and nearby
surfaces; sand buildup; erosion of pumps and valves, due to the abrasive actions of sand;
malfunction of water dump valve; malfunction of instruments and controls; and corrosion. Smiths
has discussed, in considerable detail, these and other operational problems associated with
emulsion treating equipment. He also discussed the economics of crude oil emulsion treatment.
Interested readers are directed to reference 5 for more details.
Automation of Dehydration Plant
Because of the large physical size of a dehydration plant, and the equipment used, operations
tend to become highly labor intensive in large steam injection operations. For example, the oil
dehydration plant in one large California steamflood operation covers 11 acres and consists of
7 freewater knockouts; 17 heater treaters; 21 storage tanks; and 5 LACT units; and numerous
pumps, valves, and other miscellaneous equipment.6 The operation and maintenance of such a
386
plant is highly labor intensive. If an upset occurs, it may take from 20 minutes to one hour to
locate the trouble spot. Operations such as these would highly benefit from automation. The
current trend in large steam injection operations is to automate production operations to reduce
operational costs. The oilfield automation goes by various names such as "Supervisory Control
and Data Acquisition (SCADA)" and "Sample Control and Alarm Networks (SCAN)." These are
discussed more thoroughly in references 6 and 7. Depending on the complexity of operation, the
installation costs will range from 1 to 3 million dollars. Maintenance costs include those associated
with the daily upkeep of the system, monthly computer software maintenance, monthly hardware
maintenance, and skilled labor. In Table 16.3, the cost and benefits of oilfield-production
automation is depicted.* Automation is expensive, and an economic analysis must be performed to
justify the investment and maintenance costs of the system.
SUMMARY
As a result of severity of emulsion problems and sand production in steam injection
operations, the cost of processing the produced fluids to meet the pipeline specification is
significant and second only to the fuel cost. By judicious selection of equipment, and adopting
prudent operation practices, treatment costs can be lowered. A variety of equipment, along with
chemical deemulsifiers, are utilized to treat the thermally produced oil. These include gun barrels,
heated wash tanks, heater treaters, electric dehydrators, and hydrocyclones. Not all the equipment
is used in all the steam injection operation. The nature and severity of produced emulsions dictate
the type of equipment most appropriate for a particular job. The type, configuration, size and
usage varies from field to field.
This chapter has briefly described the facilities and equipment commonly used in steam
injection operations to process the produced fluid. The overall treatment costs can be lowered by
selecting the equipment and facilities based on ease of operation, initial cost, maintenance cost, and
performance. Efforts should be made to select the minimum number of pieces of equipment and
simplest design to optimize initial and operating costs. The system should be designed to
accommodate the maximum anticipated throughput. The amount of excess capacity to be built into
the treating system should be based on an assessment of the cost of the extra capacity balanced
against the risk of not being able to treat the peak throughput. Only the equipment needed to
accomplish the task should be purchased and installed at,the start of the project. Additional
features or equipment can be added later in the life of the project, if necessary.
387
GlwilIa
m S C 0 i l M m
a
b
c
d
e
f
g
h
i
j
k
I
Telemetry equpmcni
huechon well end dences
Cabk tnslallallon
a Tank levels
b lnp3mn volumes
c Roductlm volumes
d chcck hnc and vesvl pressure
SofIwm hceme
-W
b Loadingalam
c Ability 10 track pump from well to well m m quickly
d Ability to dctect scak buildup in (be pump and tbg.
mae quickly
e Ability lo delcct luting travel mom quickly
f Ability 10 deled stickhg pump more quickly
g Ability lo detect pump lagging bottom more quickly
h Ability lo detect all overloads prior to failure more
quickly
a
b
c
d
5-
3a Repairs a replacement
b Ahgnment
4-
Software unpkmnetlon
10
Labor
a Replaa csbk
I
2
Power Cpppymplipp
a0
2Loadccllandcable
P
a Manlor cathodtc protectlon
b Corurol i n ~ u o nvolumes
c Momtor flowhoe preuure
-1
@e11er data)
3 kyE&irs
or rephamen1
..
a Replacemat
7 Reld cable N U
a Cable replacement
lprcuon W&
1
Rate
a Turhnc meter rates and p L u p
b Frequency to DC converter
d Calibiatlon
c Cable
3
4
Pressure
a Transmiter
b Cahle
c Cshbrauon
Field cabk cub
RTUboardS
DMIsccllaocovr
I
Cleanrng fluids
E
F
REFERENCES
1. Flahive, R.M. and P. J. Hickey. Production Well Testing Utilizing a Mass Flow Meter.
Paper pres. at the 62nd Annual Fall Meeting of the SOC.of Pet. Eng., Dallas, TX, Sept. 27-30,
1987. SPE paper 16882.
2. Hohn, P. P. and F. R. Morton. Capacitance Water-Cut Probe Utilization in the Kern
River Field. Paper pres. at the SOC.of Pet. Eng., California Regional Meeting, Bakersfield, CA,
Apr. 5-7, 1989. SPE paper 18780.
3. Christianson, B. A. and E. L. Berger. San Ardo Field Production Testing System
upgrade. Paper pres. at the SOC.Bet Eng. International Thermal Operations Symp., Bakersfield,
CA, Feb. 7-8, 1991. SPE paper 21533.
4. Sivalls, C. R. Crude Oil Treating Systems Design Manual, Technical Bulletin No. 166,
Rev. 2. Sivalls Inc., Odessa, TX, September 1986.
5. Smith, H. V. and K. E. Arnold. Crude Oil Emulsions, Chapter 19 in Petroleum
Engineering Handbook. Edited by H. B. Bradley, SOC.of Pet. Eng., Richardson, TX,1987, pp.
19-1 to 19-34.
6. Strome, R. K. and K. P. Beal. Heavy Oil Dehydration Facility Retrofitted With
Computer. Paper pres. at the SOC.of Pet. Eng. International Thermal Operations Symp.,
Bakersfield, CA, Feb. 7-8, 1991. SPE paper 21535.
7. Shore, R. A. The Kern River SCAN Automation System-Sample Control and Alarm
Network. Paper pres. at the SOC.of Pet. Eng., California Regional Meeting, Bakersfield, CA,
Nov. 8-10, 1972. SPE paper 4173.
8. Jentsch, W. A. and R. D. Marrs. Computerized Automation of Oilfield Production
Operations: An Extensive Five Year Study Into the Costs and Benefits. Paper pres. at the Soc. of
Pet. Eng., 61st Annual Fall Meeting, New Orleans, LA, Oct. 5-8, 1986. SPE paper 15392.
389
APPENDIX 16-A
STEAM IN.1ECTION PRODUCTION FACILITIES VENDORS
A.
B.
CHAFTER 17
STEAM INJECTION PROJECT PERFORMANCE MONITORING TOOLS
INTRODUCTION
The cost-effective operation of a steam injection project will require diligent monitoring. The
performance of an individual well and surface equipment, especially the steam generator, must be
constantly monitored and appropriate action taken when problems arise in order to reduce operation
and maintenance costs.
Fuel, labor, and supplies make up more than 50% of the steam injection project's operating
expenses. Hence, constant monitoring and control of these costs are crucial to the economics of
the operation. In a typical steam injection project, approximately 30% of the produced oil is
consumed to generate steam. Any improvement in the steam generator thermal efficiency would
reduce total fuel use and increase the operating revenues. For example, the fuel requirements of an
oil-fired 50 MM Btu/hr steam generator can be reduced by roughly 5,300 bbVyr by a 5% increase
in thermal efficiency. This would result in an increase of over $63,00O/yr in operating revenues at
an oil price of $12.00/bbl. Thermal efficiency of the steam generator can be improved by
monitoring and controlling the excess oxygen used in the combustion process and the flue gas exit
temperature. In addition, fuel requirements can be reduced by optimizing the steam injection rate to
an individual well pattern. When carefully monitoring the steam injection profile and the casing
gas production, the operator can optimize the injection rate.
Apart from the water treatment and steam generation costs, a steam injection project will incur
numerous other expenses such as those associated with the operation and maintenance of surface
production facilities, servicing of the wells, disposal of produced waters, and steamflood
monitoring. These expenses are by no means minor and must be carefully monitored and
controlled to improve the project's operating margins. The operational costs can be lowered by
fine tuning the performance of all field facilities.
Monitoring tools and procedures provide the data necessary for optimizing project
performance. These include the acquisition of physical data such as pressure, temperature, flow
rate, and fluid volumes as well as economic data such as fuel usage rate and itemized costs of
operating the surface facilities. The performance data must be made available to the field personnel
on a predetermined time schedule so that corrective actions can be taken to improve equipment
performance. The economic data must also be made available to the engineering staff and
management so that informed decisions can be made.
Several steam injection operation monitoring tools have evolved over the years and can be
broadly grouped into two categories: (1) physical data monitoring tools and (2) economic data
monitoring tools. Examples of physical data monitoring tools include an automatic well testing
392
system, steam generator fuel usage report, water treatment plant throughput, and electrical power
consumption. Examples of economic data monitoring tools include surface facility operational cost
reports and capital equipment replacement cost reports. Following is a brief discussion of various
steam injection operation monitoring tools.
operational problems reported on a daily basis. A daily report of plant throughput volumes should
be prepared with appropriate comments regarding the operation of the plant during the preceding
24-hrperiod. Also, as in the case of fuel usage report, it should be prepared and distributed to the
operating personnel on time.
394
forecast the effect of such a change, but also enables them to pinpoint the success or failure of this
change in operations. The operating expense report also allows the performance of 'what-if
calculations and forecasts the effect of physical and operating parameter changes on operating
expenses under various scenarios.
395
CHAPTER 18
ENVIRONMENTAL ISSUES
INTRODUCTION
In recent years, steam injection operators have been faced with increasingly stringent
environmental regulations which are oriented towards preventing or ameliorating significant
degradation of air and water quality and land use. These environmental laws and regulations
significantly impact the design and operation of steam injection processes. Federal, state, and local
agencies are all involved in the enactment and implementation of various environmental laws and
regulations. Although the laws and government agencies that administer the environmental
programs differ from state to state, the overall pattern of administering these laws is uniform
throughout the country.
Prospective steam injection operators are required to obtain necessary environmental permits
before an operation can begin. These permits require operators to monitor the environmental
conditions of their operations, keep certain records, and report periodically to the enforcing
agency. Violation of environmental regulations will result in severe penalties or criminal
convictions. Hence, it is essential that the operators be aware of the essence of these regulations,
so that inadvertent noncompliance with applicable environmental rules and regulations can be
avoided.
The purpose of this chapter is to outline the steps involved in obtaining an environmental
permit and to briefly summarize the environmental laws of relevance to steam injection operations.
Only federal laws are summarized. Since the environmental regulatory and compliance
requirements of the states vary from state to state, they are not discussed. Recently, the U.S.
Department of Energy published an EOR environmental regulation handbook1 detailing state
environmental laws. It is recommended that the prospective operator consult this or other
publications2-5 for the requirements of the state where they plan to conduct an operation.
OBTAINING AN ENVIRONMENTAL PERMIT
Before a steam injection operation can begin, an operator must obtain necessary
environmental permits from the federal, state and local agencies to conduct the Operation. This
section, adapted from reference 1, describes the general procedures for obtaining environmental
permits and offers some general pointers about post-construction compliance and enforcement.
1.1 Obtaining an Environmental Permit
Step 1. Define Operabon
'
396
The specific type of operation planned will determine which permits will be needed, and how
much time should be allowed from project conception to actual operation. For example, a chemical
recovery process may involve handling, use, and disposal of chemicals that are very strictly
regulated. A thermal recovery project, on the other hand, will have little or no solid waste to
dispose of, but will generate significant amounts of air pollution. The more precisely one can
define the project at the start (e.g., which chemicals will be used; what size compressor will be
needed) the better one can focus on the relevant laws and permits.
Step2. Idenb'fv the Facility Si&
Because environmental laws are designed to protect the quality of the environment and
because environmental conditions vary from place to place, the permits needed and the conditions
for obtaining them will depend on the site chosen. If several alternative sites, or a general area has
been defined, environmental conditions and permit requirements should be used as criteria for
selecting a specific site. An operation that is unacceptable at one site may be acceptable at another
site only a short distance away. Air quality, for example, can be extremely variable.
Step3. SurvevEnvironmental Conditions at the Cand idate Site(sl
A quick survey of environmental conditions should identify critical problem areas before any
substantial investments are made. The types of information to look for at this time include general
descriptions of air and water quality, hydrological factors affecting injection for reservoir pressure
maintenance and/or waste disposal, special topographic features (e.g., wetlands, floodplains),
special land areas in the vicinity (e.g., natural preserves, federally-owned lands), and similar
characteristics that might entail extraordinary environmental standards or permit procedures.
Step 4. Identify the Required Permits
Steps 1-3 may not actually occur in sequence. Rather, they are three of the elements in the
environmental evaluation of site and operational feasibility. Once a site or small number of sites
have been chosen and the operating mode determined, the required permits can be identified. This
chapter will be useful as a starting point for this step, but it will be important to verify current
requirements with the regulatory agencies before proceeding further. The following considerations
should be kept in mind when identifying the required permits.
a. Review potential requirements for every aspect of the operation that "pollutes" or
modifies the natural environment: air pollution (including on-site storage of oil or
volatile chemicals), wastewater streams, anything injected into a well for either
production or waste disposal, surface and subsurface construction, and surface
transport ( h c k s and roads, pipelines). Some details (e.g., roads) may not be covered
by state or federal law but will probably be subject to municipal or county permits.
b. Review both federal and state requirements. An operator may even need permits from
both the federal and state governments for the same discharge. If the facility site is on
397
federal lands, the permit requirements established for that area by the agency in charge
should be identified, in addition to the general environmental laws described in this
handbook.
step 5. s
i Di
n
rmi '
cies
Early contact with the agencies responsible for granting permits has several benefits:
1. The latest requirements and any pending regulatory changes are immediately available.
2. Clarification of confusing requirements and direct verification as to whether or not a
permit is needed.
3. Immediate help is available for determining requirements of specific application forms.
4. "Good faith" can be established in complying with agency rules.
5 . Personal contacts can help smooth out any administrative problems.
6 . Avoids surprises of requirements not clearly spelled out in a specific law.
7 . A facility can be described in favorable terms before opposing entities contact the
agency.
8 . Plans can be changed early on if modifications are necessary to avoid unattainable
requirements or if newly promulgated regulations have to be met.
Step. 6. Identify the Environmental. Engineering. or Other Needed Studies
Very close attention should be given to the information required to obtain each permit. A
common frustration experienced by many permit applicants is to be told by an agency, at the last
minute, that they need to undertake an environmental study to answer a particular question.
Careful reading of permit applications and discussions with agency staff should enable
environmental requirements to be identified early in the process so the project will not be delayed.
Step 7. Initiate the Necessary Stud'ies
For some facilities, the environmental studies will take one or two years to complete. A case
in point for steam injection facilities is ambient air quality monitoring. The U.S. EPA requires a
full year of monitoring data for a pre-construction permit application, which means that planning
for monitoring must begin about two years before the expected date for starting construction.
- plications
Step 8. Complete the Permit AD
Once all required information has been collected, the application forms can be completed.
Usually, there is no advantage in filing applications far in advance of the deadline; in fact, some
permits become invalid if they are not "used" within a certain time. On the other hand, if the
expected review time is, for example, 6 months, submitting the application 9 to 12 months before
construction will provide a cushion for bureaucratic delay and negotiation of disputed terms or
appeals from adverse agency decisions.
398
399
400
1.2.2 Enforcement
Monitoring. Recordkeeping, and ReportinP
More often than not, a permit will require monitoring of environmental conditions, the
keeping of certain records, and periodic reporting to the agency. The U.S. EPA and most state
agencies take these requirements seriousIy, because they rely on self-monitoring and self-reporting
to identify trouble spots. The few criminal convictions that EPA has obtained for violation of
environmental laws have involved failures to report activities, falsification of monitoring data, and
similar infractions. Close adherence to these requirements will stand an operator in good stead if
involved in more substantial problems with the agency.
Reportinp Spills
A special reporting requirement in the Clean Water Act applies to spills of oil or hazardous
substances. Most states have a similar law. Under federal law, any spill of oil (or any spill above
specified amounts of over 300 listed substances) that reaches or may reach surface water must be
reported immediately after it is discovered to the EPA or to the Coast Guard. The source of the
spill will be liable for a moderate penalty ($500 - $5,000) in addition to all costs of clean-up or
containment.
InsDections
Virtually every environmental law authorizes state or federal officials to inspect a facility and
owners operation records during business hours. (They are also authorized to enter a facility at
any time to take emergency measures to protect the environment or the public health.) Although
the Supreme Court has ruled that OSHA inspections cannot be conducted against the owners will
without a warrant, the decision may not apply to environmental laws. In any case, such warrants
will be routinely issued because of the broad authority granted to these agencies.
1
Enforcemen
*Pnl
The environmental laws usually allow an agency to choose between suing a violator
immediately in court and issuing some type of administrative order to the violator to correct a
violation. In practice, agencies almost always use the administrative order approach, if only
because litigation is time-consuming and expensive. In fact, many agencies prefer to use informal
negotiations rather than formal orders to obtain compliance. This is especially true at the state
level; as a rule, EPA prefers the formal approach.
For minor violations, the agencies usually seek simply to have the problem corrected. More
significant violations, or repeated violations, will probably also result in a civil penalty. A civiI
penalty, which the agency can collect just by proving the violation, is designed to compensate for
40 1
the environmental damage that occurred. It carries no implication that the source acted criminally;
criminal penalties are separately authorized and can be imposed only after a court conviction.
The statues often authorize enormous civil penalties, such as $25,000 per day. However,
.
penalties of that size are reserved for the largest corporations and the worst violations. More
routinely, penalties or small facilities will vary from about $500 to almost $10,000 for the entire
violation, depending on how long it lasted, how serious the environmental harm was, and how
much money the violator saved by allowing the violation to occur. The compliance record of the
violator, his efforts to correct the violation, and his general attitude toward the agency will also
influence the amount of the penalty. If one feels strongly that the amount of the penalty that the
agency is asking for is too high, the whole case can be brought to court where the judge sets the
penalty.
Final authority for all enforcement matters lies with the courts. If in disagreement with m
agency about whether a violation occurred, about the corrective measures to prevent further
violations, or about the penalties or other sanctions they seek, the matter may be brought to court to
be decided by a judge or ajury, as the case may be. The agency usually has the choice of going
directly to the court without providing advanced notification to the violator.
In order to answer correctly the questions in the permit application, a prospective thermal
EOR operator should be aware of the essence of the environmental rules and regulations of
relevance to his project. The following sections briefly summarize the major federal environmental
laws and regulations of relevance to thermal operations.
Maior Laws
The environmental laws that have the greatest impact on steam injection processes pertain to
air quality, water quality, hazardous wastes and environmental quality. Following is a list of major
federal laws that have relevance to steam injection processes.
Air Quality -Clean Air Act and Amendments
Water Quality -Clean Water Act and Safe Drinking Water Act
Hazardous Wastes -Resource Conservation and Recovery Act (RCRA)
Environmental Quality -National Environmental Policy Act (NEPA)
402
et seq.). Regulations promulgated by federal agencies are initially published in the Federal
Register (FR). The final versions of the regulations are then incorporated into the Code of Federal
Regulations (CFR). The CFR is updated quarterly and published in its entirety annually. Since
regulations take effect as soon as the final version appears in the FR, a person cannot be sure of
full regulatory compliance by reviewing the current edition of the CFR. Regulations pertaining to
the environmental area can be found in the Title 40 of CFR.
Enforcement Agencies
The environmental regulations are enforced by a number of agencies. At the federal level, the
U.S. Environmental Protection Agency (EPA) is entrusted with enforcing laws dealing with air
quality, water quality, hazardous wastes, and hazardous materials. At the state level, many
different agencies are involved with implementing and enforcing laws pertaining to different
environmental areas. The State EOR environmental regulatory agencies are shown in Table 18.1.
The agency jurisdiction may sometimes overlap. As an illustration, a summary of different
agencies and their principal area of authority in California are outlined.
Air Resources Board (ARB)
The ARB is a state agency whose primary responsibility is to coordinate the state-wide air
quality programs. It is also responsible for implementing programs designed to attain the state and
national ambient air quality standards. In addition, this agency supervises the overall scheme to
control toxic air pollutants and conducts research on various air quality matters.
Air Pollution Control and Air Oualitv M a n w m e n t Districts (APCWAQMD)
These are county or multi-county agencies responsible for implementing the state air quality
laws pertaining to stationary sources of air pollution. There are 34 single county APCDs, five
multi-county APCDs, and three AQMDs.
State Water Resources Control Board (SWRCB)
Like ARB, SWRCB is a state agency whose primary function is to coordinate the states
water quality program. The SWRCB has the authority to regulate the discharge of wastes to land
and injection of toxic waste to wells.
Regional Water Quality Control Boards (RWOCB)
RWQCB is a regional agency responsible for administering the states water quality program
within the region. It also has the authority to develop water quality control plans, issue of waste
discharge permits, and regulate waste disposal sites. The state is divided into nine RWQCBs.
Deaartment of Health Services t DHS)
DHS is a state agency entrusted with implementing the states hazardous waste laws and
regulations including hazardous waste facility permitting.
403
State
Air Quality
Water Quality
(NPDES & Class I Wells)
Underground
Injection
(Class I1 Wells)
Hazardous
Waste
Alabama
Air Division
Dept. of Envirn. Mgt.
1751 Congressman
WL Dickenson Dr.
Montgomery, AL 36130
2051271-7700
Water Division
Dept. of Environ. Mgt.
1751 Congressman
WL Dickenson Dr.
Montgomery, AL 36130
205127 1-7700
Land Division
Dept. of Envirn. Mgt.
175 1 Congressman
WL Dickenson Dr.
Montgomery, AL 36130
205127 117700
California
Toxic Substances
Control Division
Dept. of Health Services
400 P street
Sacramento, CA 95814
9 161324-7193
Air Pollution
Control Division
Colorado Dept. of Health
4210 E. 11th Ave.
Denver, CO 80220
303133 1-8500
Waste Management
Cdorado Dept. of Health
4210 E. 11th Ave.
Denver, CO 80203
303133 1-4830
1102 Q Street
P.O. Box 2815
Sacramento, CA 95812
9161322-2990
Colorado
Illinois
Kansas
Conservation Division
Kansas Corp. Commission
200 Colorado Bldg.
202 w. 1st street
Wichita, KS 67202-1286
3161263-3238
Conservation Division
Kansas Corp. Commission
200 Colorado Bldg.
202 W. 1st Street
Wichita, KS 67202-1286
3161263-3238
Conservation Division
Kansas Corp. Commission
200 Colorado Bldg.
202 w. 1st street
Wichita. KS 67202-1286
3 161263 -3 23 8
State
Air Quality
Water Quality
(NPDES & Class I Wells)
Underground
Injection
(Class I1 Wells)
Hazardous
Waste
Louisiana
Office of Conservation
Dept. of Natural Resources
P.O. Box 94275
Baton Rouge, LA 70804-9725
5041342-55 15
Office of Conservation
Dept. of Natural Resources
P.O. Box 94275
Baton Rouge, LA 70804-9725
5041342-5515
Michigan
Permit Section
Air Quality Division
Dept. of Natural Resources
P.O. Box 30028
Lansing, MI 48909
5171373-7023
Mississippi
Montana
Envir. Services
Cogswell Building
1400 Broadway
Helena, MT 59620
4061444-3454
State
New Mexico
Air Quality
Environmental Dept.
1190 St. Francis Drive
P.O. Box 26110
Santa Fe, NM 87502
5051827-2850
5051827-5000
North Dakota
7011224-2969
Water Quality
(NPDES & Class I Wells)
Underground
Injection
(Class I1 Wells)
5051827-5000
5051827-5000
7011221-5188
7011224-2969
7011221 -521 0
4051271-5220
4051231-2541
4051521-2500
Hazardous
Waste
Environment Dept.
1190 St. Francis Drive
P.O. Box 26110
Sank Fe, NM 87502
5051827-2929
701/221-5188
ch
Oklahoma
4051521-2302
Hazardous Waste
Management Services
Oklahoma State Dept.
of Health
lo00 NE 10th Street
Box 53551
Oklahoma City, OK 73152
4051271-7047
Texas
5 12/451-571 1
5121463-6893
5121463-6790
5121463-6790
State
Utah
Wyoming
g
4
Air Quality
Water Quality
(NPDES & Class I Wells)
Underground
Injection
(Class XI Wells)
Hazardous
Waste
Normally, the lowest level of government agency is given the authority to implement the
various laws and regulations. For example, the APCD is responsible for enforcing air quality laws
and regulations including issuance of permits for new air pollution source within the district.
However, some districts are permitted to regulate only portions of laws. In such cases, the agency
which retains portions of authority may also be involved, and agency overlaps can occur.
The major federal environmental regulations that impact steam injection operations are briefly
described in the following pages. Other applicable federal regulations of relevance to steam
injection operations are summarized elsewhere.1-2 The abbreviations used in this chapter are listed
in Table 18.2.
Air Oualitv Regulations ImDacting Steam Injection Operations
The federal government plays a leading role in developing a regulatory framework within
which all thermal enhanced oil recovery operators must operate. Federal environmental regulations
and standards supersede all less stringent state and local regulations. State and local
governments, however, do have the authority to promulgate and enforce environmental standards
more stringent than their federal counterparts. For example, State of Californias air quality
requirements are more stringent than those stipulated under federal statute. Kern County, CA, has
adopted even more stringent measures to attain and maintain state and national ambient air quality
standards. This section briefly describes, in general terms, the federal air quality regulations to
which the steam injection operators must comply.
The U.S. Environmental Protection Agency (EPA) is the primary agency entrusted with the
responsibility of implementing the Federal Clean Air Act (CAA). The purpose of CAA is to protect
and enhance the air quality and to attain the National Ambient Air Quality Standards (NAAQS).
Under the CAA, each state has the primary responsibility to prepare and submit to the EPA, a State
Implementation Plan (SIP) which details the method by which NAAQS are attained and
maintained. EPA regulations pertaining to air quality can be found at 40 CFR Subchapter C - Air
Programs.
The provisions of CAA that impact steam injection operators are as follows:
National Ambient A i r Qualitv Standards,
National Ambient Air Quality Standards (NAAQS) are the primary guidelines used to
measure the air quality of a region. NAAQS sets the ceilings for which each pollutant may not
exceed. EPA set the NAAQS according to established criteria. These criteria are reviewed at least
once in every 5 years by an independent scientific committee. The air quality standards are set on
the basis of scientific data and analysis only. EPA is not required to consider economic or
technical feasibility in setting air quality standards.
408
DOG
EPA
LAEX
NPDES
NSPS
PL
PSD
RCRA
RWQCB
SDWA
SIP
SWRCB
USC
UIC
There are two types of air quality standards: primary and secondary. The object of the
primary standards is to protect the public health, while allowing for ample margin of safety. The
aim of the secondary standard is to specify a level of air quality considered necessary to protect the
public welfare from known or anticipated adverse effects, including the effects on economic values
and personal comfort.
CAA has designated six pollutants as harmful, and standards have been established. These
pollutants include: sulfur dioxide (S02),nitrogen oxides (NO,), particulates, carbon monoxide
(CO), ozone, and lead. At one time; EPA had adopted NAAQS for hydrocarbons, but these
standards were later rescinded. These standards are implemented through SIPS (CAA, Section
110). Most steam injection operations employ lease crude fired steam generators. These crude oil
fired steam generators emit S-02,NO,, and particulates. In addition, oilfield wellheads can emit
hydrocarbons, depending on the degree of control employed.
EPA has designated 247 Air Quality Control Regions (AQCR). Although NAAQS should
not be exceeded in any of the regions, they are not directly enforceable. EPA has set emission
limitations for each pollutant to attain NAAQS. If the actual air pollutant concentration in an air
district exceeds NAAQS, more stringent pollution control devices on the emission source are
imposed to lower the concentration to an appropriate level. EPA has rated the AQCR as either
409
clean or nmattainment for each criteria of pollutant. Nonattainment areas are those areas in
which state and federal ambient air quality standards have not been met for the specified criteria
pollutant. Procedures to bring the nonattainment areas to attainment are implemented by SIPS.
ional primary and secondary ambient air quality standards are shown in Table 18.3.
TABLE 18.3. - Current National primary and Secondary Ambient Air Quality Standards
Air contaminant
Averaging time
Nitrogen dioxide1
Annual average
Sulfur dioxide
Annual average
Federal primary
standard
100 pg/m3
(0.05 P P ~ )
Federal secondary
standard
100 pg/m3
(0.05 P P ~ )
24 Hour
3 Hour
1 Hour
Annual geometric
Mean
50 pg/m3
50 pg/m3
24 Hour
150 pg/m3
150 pg/m3
Hydrocarbons
(corrected for
methane)
3 Hour
6-9 AM
160 pg/m3
2(0.24 ppm)
Ozone
1 Hour
Suspended
particulates
235 pg/m3
0.12 ppm)
8 hour
Carbon monoxide
8 Hour
1 Hour
Lead.
90 Day
30 Day
Source:
ppm = parts per million
Pg/m3= micrograms per cubic meter
mg/m3 = milligrams per cubic meter
lNitmgen dioxide is the only one of the nitrogen oxides considered in the ambient standards.
*Maximum 3-hour concentration between 6 and 9 A. M.
Note:
410
State Imp1ementa t ion Plans (SIPI (42 USC 7410. Sec. 110)
To implement the CAA, states are required to adopt and submit to EPA, a State
Implementation Plan (SIP) detailing with the basic strategies for implementation, maintenance, and
enforcement of NAAQS within the state. SIP forms the blueprint for achieving air quality goals
within a state. The SIP provides the emission limitations, schedules, and time tables for
compliance by stationary sources of air pollution such as oilfield steam generators. EPA must
approve the SIP for each state. SIP should contain measures necessary for achieving the primary
and secondary standards. In setting these measures, the state does not need to consider their
technical or economical feasibility.
SIP must assure attainment of NAAQS by prescribed dates. SIP must meet federal
requirements, but each state may choose its own mix of emissions for stationary and mobile
sources to meet NAAQS. CAA prohibits the adaptation of any SIP without a public hearing, and
the sources affected by SIP are expected to participate. Requirements of SIP are contained in 42
USC, Sections 7407(a) and 7410(a)(l).
Emission Limitations
CAA has developed minimum technological standards for various sources of air pollution.
These sources are divided into two groups: stationary and mobile sources. Since stationary
sources are the originator of pollutant in steam injection operations, only stationary source
standards are outlined here.
In establishing the framework for controlling emissions from stationary sources, CAA makes
a distinction between new and existing sources. Since it is typically less expensive for new
sources to incorporate state-of-the-art control technologies, the regulatory burdens fall more
heavily on the new sources. Existing sources are regulated through SIP. Each state must develop
a standard of performance for existing sources using EPA guidelines. If a state fails to establish
standards or if the state standards are not acceptable to EPA, EPA may promulgate its own
standards.
To control the pollutants from new stationary sources, CAA directs EPA to establish New
Source Performance Standards (NSPS). NSPS reflect the degree of pollution control achievable
through the best available and adequately demonstrated pollution control technique. In selecting
the best available control technology for each pollutant, the EPA must take into account the cost of
achieving such emission reduction and energy requirements.
EPA has delegated the authority to implement and enforce NSPS to various state and local
agencies; although it always retains independent enforcement authority.
41 1
412
Averaging
Pollutant
so,
Time
Class I
Class I1
1 Ye=
24 hours
3 months
5
10
19
37
None
1Y
None
2
3 hours
1 year
Ozone
75
None
40
182
700
None
None
None
None
91
512
None
None
None
None
None
None
None
None
8 hours
1 hour
1 hour
III
37
20
5
25
24 hours
Class
so21
Emissions
Ilblbblfuel)
N%
Particulate
Emissions
HC
HC
16
365
7.2
2.5
0.7
0.1
21
10.3
2.9
0.43
10
50
29 2
7.2
2.5
0.7
0.1
52.6
18.3
5.1
0.73
20
115
292
7.2
2.5
0.7
0.1
120.9 4 2
11.8
1.60
220
29 2
7.2
2.5
0.7
0.1
220
21.5
3.07
50
76.7
~~
d. Obtain emission offsets to achieve a net reduction in emissions. An offset refers to the
amount by which other sources must reduce their emissions to allow for the amount to be produced
source; thus, making room for development.
e m i t future new source construction in NAA, EPA allows existing source to bank
banked offset can be sold or traded to other sources.
The BACT for the major steam injection operation related pollutants are as follows:
(a)
The EPA approved BACT for the control of SO2 from oil fired steam generator is fuel oil
containing less than 0.05% by weight sulfur, or the equivalent rate achieved by scrubbing.
BACI for NOy
(b)
For NOx, the EPA is requiring a case-by-case analysis. It maintains that BACT, at a
minimum, shall be a modification of burner that will result in the least amount of oxygen to be
consumed for combustion. California steam injection operators frequently utilize low NOx burners
to control NOx emissions.
(c) BACT for Particulate Matter
BACT for particulate matter has not been defined. EPA maintains that S@ scrubbers should
satisfy the BACT requirements for particulates. This is because the SO2 scrubbers, although
primarily designed for So2, removal, also remove a small percentage of particulate matter.
Particulate matter emissions from oilfield steam generators are small, and it would take an
average of nine or more 50 MM BTUhr units to produce more than 250 tondyr.
Lowest Achievable Emission Rate (LAER)
Steam injection operators wishing to locate in nonattainment areas must comply with lowest
achievable emission rate (LAER). LAER for a particular pollutant is the most stringent emission
limit for the pollutant contained in the SIPS of the U.S. In no instance can the LAER be permitted
to be less stringent than the amount allowed under an applicable new source performance standard.
414
Unlike BACT, the LAER does not permit cost effectiveness to be taken into consideration. Thus,
LAER is more stringent than BACT.
The LAER for an oil fired steam generator is considered to be 0.05 wt % sulfur in fuel or the
equivalent rate achieved by scrubbing. This is based on technology achievable in practice for these
types of sources. LAER, however, is a changing thing and must reflect the advances in
technology and SIPS requirement. As SIP requirements change and become more stringent,
LAER must reflect such changes. Similarly, as scrubber technology advances, LAER must also
reflect such changes.
415
Steam injection operation effluent guidelines are enforced through the National Pollution
Discharge Elimination System (NPDES). To receive a discharge permit, a stea
must comply with all applicable effluent limitations for the particular dis
operations. NPDES
ffective for terms not exceeding 5 years.
EPA or by a state having a certified CWA program.
I
416
417
418
TABLE 18.6. - EPA List of Exempt and Nonexempt Oilfield Wastes from RCRA
Hazardous Management Requirements4
Nonexempt wastes
Exempt Wastes
ProdUcedWater
Drilling fluids
Drilling cuttings
Rigwash
Drilling fluids and cuttings from offshore opfatbns
disposed of onshore
Well completion, treatment, and stimulation fluids
Basic sediment and water and other tank bottoms from
storage facilities that hold product and exempt waste
Accumulated materials such as hydrocarbons, solids,
sand, and emulsion from production impoundments
Pit sludges and contaminated bottoms from storage or
disposal of exempt wastes
Workover wastes
Gas plant dehydration wastes, including glycol-based
compounds, glycol filters, filter media, backwash
and molecular sieves
Gas plant sweetening wastes for sulfur removal,
including amine, amine filters, amine filter media,
backwash, precipitated amine sludge, iron sponge,
and hydrogen sulfide scrubber liquid and sludge
Packing fluids
Producedsand
Pipe scale, hydrocarbon solids, hydrates, and other
deposits removed from piping and equipment prior
to transportation
Hydrocabon-bearing soil
Pigging wastes from gathering lines
Wastes from subsurface gas storage and retrieval,
except for the listed nonexempt wastes
Constituents removed from produced water before
419
420
TABLE 18.7. - Estimated Air Emissions From a Typical Steam Injection Project During Site Preparation Activities
Activity
Equipment
Operating time
Working over
existing injection
and production
wells
350 hp diesel
engine to
operate rig
100 hrs/well
Drilling
observation wells
250 hp diesel
engine to
opixate auger
drill
3 hrs/well
Dozers (3)
5 0 0 hrs./dozer
Emission factors/gm/bp-hr
(Diesel engines)
Emission factors l b h r
(Heavy duty equipment)
Total time,
hours
CO
85,400
3.03
1.12
9.74
36.87
HC
NO,
SO2
TSP
14.0
0.93
1.0
460.88
30.62
32.92
14.0
0.93
1.0
0.38
0.41
CO
HC
0.14
0.24
5.05
0.35
0.16
0.56
0.18
3.80 0.26
0.12
1.56
0.19
2.40 0.18
0.17
1.56
0.19
2.40
0.18
0.17
2.15
0.15
0.99 0.09
0.14
-1.72
0.12
0.80 0.07
0.11
1.80
0.30
2.96
13.50
2.25
22.20
0.41
0.16
2.3
0.14
0.14
0.82
0.32
4.60
0.28
0.28
NO,
SO2
TSP
~~
1,500
3.03
1.12
1.25
0.46
5.14
Site preparation
Loadera (4)
Tractors (4)
Transportation
of material
and equipment
General
Heavy duty
mucks (10)
Miscellanexjus
500 hrdoader
400 hrdtractor
1500 hrs/truck
400 bours
1,500
2,000
1,600
15,000
4,000
Local groundwater quality has the potential to be affected by steam injection waste water
effluent disposal activities. Some of these include:
Potential of seepage of injected or reinjected materials through the formation into
underground aquifers;
Potential of leaching from unlined waste disposal pits;
Potential of accidental surface spills during storage and transport of waste effluents; and
Potential of secondary fractures, which may connect the formation with aquifers.
0
Total protection of groundwater from waste is impossible because the control of all flow
paths from waste to groundwater aquifers, except casing leaks, is beyond reliable engineering
control.
Waste Management
A typical steam injection process generates wastes both during site preparation phase and
field testing phase. The wastes produced from site preparation activities include:
Clay, silt, and dust particles;
Organic and inorganic matter from decayed vegetation;
Drilling wastes such as mud, brines, etc; and
Oil and grease from construction machinery.
The first two types of wastes are disposed of by landfill and are of no environmental concern.
Drilling wastes and oilfield brines, because of their potential damage to local water and eco system,
require careful handling and must be disposed of in state-approved sites.
The wastes generated from field operations activities include:
Sludge produced from scrubbers;
Spent liquids such as scrubber fluids;
oilywastes;
Hazardous (toxic) wastes
The handling and disposal of these wastes call for special measures and are outlined as follows:
The scrubber liquids must be treated onsite prior to their disposal into local surface
waters;
Scrubber sludges are difficult and costly to dispose of and are a potential source of
contamination of local surface and subsurface water at the disposal site;
The oily wastes are collected separately and are disposed of according to local regulatory
requirements; and
The hazardous wastes are handled in accordance with the state regulatory requirements.
422
SUMMARY
The environmental requirements for candidate domestic steam injection oil recovery projects
were identified and evaluated from federal, state, and local regulatory view points. From these, the
following conclusions have been derived.
Environmental regulations pertaining to ambient air quality and groundwater quality
greatly impact the growth of the steam injection processes;
Current regulations are sufficient for controlling noise pollution; and
Environmental issues such as solid- and liquid-waste handling procedures require sitespecific details that must be evaluated against the applicable governmental regulatory
requirements.
REFERENCES
1. Madden, M. P., R. P. Blatchford and R. B. Spears. Environmental Regulations
Handbook for Enhanced Oil Recovery. U.S. Dept. of Energy Report No. NIPER-546, December
1991.
2. Sarathi, P. Environmental Aspects of Heavy Oil Recovery by Thermal EOR Processes.
Pres. at the SOC.of Pet. Eng. Western Regional Meeting, Long Beach, CA, Mar. 20-22, 1991.
SPE paper 21768.
Solid
423
APPENDIX 18-A
THERMAL EO R ENVIRONMENTAL CONSULTANTS
A.
Consultants
1. WZIInc.
4800 Easton Drive, Suite 114
Bakersfield, CA 93309
Telephone: (805) 326- 1112
425