Oil recovery from tank sludge

Thorough tank cleaning means removing all of the sludge and restoring the
vessel’s full storage capacity

Marcello Ferrara and Cristina Ferrara ITW

A
lthough somehow neglected by the indus- 10%. Thereafter the tank is ventilated until the
try, tanks are vital for production: indeed, LEL falls to a lower value.
no production can occur without proper The tank is cleaned by manual removal of sludge
storage. Tank cleaning normally occurs for main- and, when applicable, high pressure water jetting,
tenance and inspection purposes, and less fre- with the following results:
quently for recovering storage capacity. However, • The value of the oil is lost together with the
any sludge at the bottom of the tank equals vol- sludge
ume lost in the tank’s useful storage capacity. • High amounts of waste (water, sludge, oil) are
When selecting the proper tank cleaning tech- generated, which need to be disposed of
nology, a number of evaluations should be per- • There is significant use of equipment and
formed: for instance, manpower requirements, manpower
the effectiveness of the method, the equipment • HSE concerns apply during job execution
used, the downtime involved, the waste produced, • Downtime is huge.
environmental and safety concerns, and utili- Last but not least, the manual cleaning job is
ties consumed. In general, all of the costs (direct very hard and manpower cleaning efficiency is
and indirect) involved in the cleaning have to be very low. Workers are acting in a confined space,
evaluated. wearing heavy personal protective equipment with
In an effort to reduce operating costs, many oil breathing apparatus, and can hardly move inside
companies aim for the lowest price without taking the tank as they are walking on a very viscous and
into consideration the full picture, just the clean- sticky medium. In the case of high ambient tem-
ing cost. This is particularly true for tank clean- peratures, working becomes possible for some
ing, wherein many conventional manual cleaning minutes only. Workers are therefore allowed a long
operations are available, most of which offer the resting time within their working shift after a cer-
same technology. tain time inside the tank.
It has to be mentioned that nowadays it is very
Conventional tank cleaning methods difficult to find skilled workers for this job which
Manual cleaning is the oldest and today’s most is surely the most arduous in the industry. If
widely used method of tank cleaning. Often, this this method is chosen for its ‘economy’, the
is also thought of as the only effective, easy and above items must be kept in consideration along
cheap technology. with the ‘simple’ cleaning cost. The cost impact
But what really happens during manual clean- might be a multiple of the mechanical cleaning
ing? The first step is emptying the tank to the low- cost and so will affect the economy of the entire
est possible level; thereafter, only sludge, deposits project.
and a certain amount of oil will be left in the tank. To overcome the drawbacks of manual clean-
The manways need to be opened for man entry, ing, some other methods have been developed (‘no
so breathing apparatus is mandatory. The tank man entry’ techniques) but all have their related
is not safe for unprotected entry and a num- problems. For instance:
ber of additional requirements need to be com- • Cleaning with robots improves safety and some-
plied with. Among them, any job must be stopped times the duration of operations but it does not
whenever the lower explosive limit (LEL) exceeds have any impact on sludge recovery.

www.digitalrefining.com/article/1001431 PTQ Q4 2017 1

 Heavy
hydrocarbons

ITW chemical
action

Sediment

Figure 1 Asphaltene stabilsation and sediment release Figure 2 Solid heavy paraffin sludge

• Physical separation methods involve the use of able liquid product. It is chemically stabilised,
multiple equipment like centrifuges and decant- there is no danger of subsequent precipitation,
ers to enhance sludge recovery. This method and therefore the sludge can be pumped from the
effectively separates sediments only from the tank as a stable liquid and fully reprocessed.
sludge. No real oil recovery occurs, only the cre- Sediments are left at the bottom of the tank
ation of an unstable oil phase which will precipi- or separated during circulation. Release of sedi-
tate again once pumped into another tank. ments from precipitated asphaltene is not pos-
• In crude oil washing, crude oil is used as a sible unless the asphaltene structure becomes
‘solvent’ for sludge. Here is a simple question: disjointed. The concept is illustrated in Figure 1.
if crude is a solvent for the sludge, why has the According to the method, after the first step of
sludge precipitated? Crude is not a solvent for the dissolving and stabilising the sludge, a second
sludge and this method simply moves the sludge step of tank decontamination follows in order to
from one tank to another (a dispersion method). allow entry under safe conditions.
Integration with a physical separation method This second step normally reduces the LEL to
adds only the removal of sediments but, again, 0%, as well as H2S and benzene levels to 0 ppm.
does not impact real oil recovery. No work interruption is needed as LEL safe val-
• Conventional chemical cleaning uses a disper- ues are not exceeded.
sant to solve the problem. In reality, the problem As the developer of the technology, ITW can
is transferred to the receiving tank and the chem- tailor-make chemicals to target the many types
ical does not impact real oil recovery. of sludge in the oil industry. The following case
To solve all of these problems, ITW has pat- histories illustrate some applications of the
ented a technology for tank cleaning. technology.

ITW tank cleaning technology Case history: cleaning paraffinic sludge

ITW has a novel technology for asphaltene stabi- Eagle Ford is well known as an opportunity
lisation. According to this approach, the sludge is crude. Its processing, however, poses a num-
removed by the addition of a proprietary chemi- ber of problems from storage until the actual
cal, without the need to open manways. CDU run (fouling of preheat train and furnace).
The patented chemical contains proprietary ITW developed a chemistry capable of dissolv-
asphaltene stabilisers, paraffin ing the associated heavy par-
solvents and fluidising agents. affinic sludge. Figure 2 illustrates
A different proprietary chemi- the solid heavy paraffin iso-
cal is used to dissolve polymers lated from a paraffinic sludge,
when cleaning tanks in the pet- while Figure 3 illustrates the
rochemical industry. same solid heavy paraffin dis-
The chemical is circulated solved by ITW proprietary
inside the tank and, while cir- chemistry.
culating, it will dissolve and The same technology can be
stabilise the sludge. With the used for cleaning preheat trains
technology, the sludge is trans- Figure 3 Solid heavy paraffin sludge or pipelines fouled with par-
formed into a reusable, pump- dissolved by ITW affinic sludge.

2 PTQ Q4 2017 www.digitalrefining.com/article/1001431

 Figure 6 Collapsed roof and extra
Figure 4 Solid sludge before cleaning Figure 5 Tank internals after cleaning supports

Case history: collapsed roof tank cleaning ogy’s ability to solve the problem.
Besides the drawbacks mentioned earlier, man- Before giving the go-ahead, the ship’s owners
ual cleaning is not always possible. This is the performed a thorough washing of some tanks with
case whenever safety conditions do not allow per- hot light cycle oil, which proved ineffective against
sonnel entry, for instance when the sludge level is this type of sludge.
above manways, or when the tank’s roof has col- ITW injected its patented asphaltene stabi-
lapsed or heavily corroded. liser directly into the vessel’s tanks. Immediately
In the case of a 150 000 m3 crude oil tank, after the injection, the cargo pumps were able to
the roof had collapsed and personnel entry was start running so that circulation could take place.
almost impossible. The refiner chose ITW tech- Circulation lasted about 24 hours once operating
nology as the best available approach to cleaning. conditions had been achieved.
This involved dissolving and pumping out about At the end of the cleaning, ITW applied its
10 000 m3 of sludge in about 10 days. The stabi- improved degassing/decontamination technology,
lised fluid was fully reprocessed in a CDU without to reduce the time for safe entry. At the manways
any operating problem. opening, the following parameters were recorded:
ITW also took care of the engineering part of LEL: 0%; hydrocarbons: 0 ppm; H2S: 0 ppm; ben-
the job, namely calculating the static condition zene: 0 ppm; ammonia: 0 ppm; pyrophorics: nil.
of the roof and the number and position of addi- This was achieved after only 12 hours’ circula-
tional supports needed for safe entry of personnel tion. Ventilation time was reduced from about 48
for final cleaning. hours to 12 hours.
The job was completed when a gas-free certif- Visual inspection of the tanks confirmed the
icate could be issued to perform maintenance results of ITW Online Cleaning. The tanks were
activity. Figures 4, 5 and 6 illustrate some events clean to the bottom level and the tank walls were
during the cleaning operation. clean.
No waste was generated by the process. All of
Case history: oil tanker cleaning the sludge was converted into oil and then recov-
A product tanker experienced hard sludge precip- ered as slop oil for subsequent reuse. Degassing/
itation due to incompatible blending of heavy fuel Decontamination technology does not form an
oil and paraffinic crude oil. The blending occurred emulsion and uses biodegradable chemistry,
inside the vessel’s tanks and created a very hard, therefore the washing fluids could be discharged
coke-like layer in all of the vessel tanks. directly to a treatment plant for oil/water separa-
Cargo pumps were blocked by hard sludge and tion. Figures 7, 8 and 9 illustrate some results of
the vessel could not operate. The tank walls were the cleaning.
also covered with deposits.
The ship’s owners wanted to avoid mechan- Case history: Orimulsion tank cleaning
ical cleaning of the vessel tanks, which would Orimulsion is a bitumen emulsion in water,
have led to at least 3-4 months’ dry-docking, with obtained from the world’s largest deposit in the
related costs and downtime. Orinoco basin in Venezuela.
ITW made a study of a deposit sample to verify Raw bitumen has an extremely high viscos-
the performance of its Online Cleaning technol- ity and specific gravity between 8° and 10° API
ogy in this case. Lab tests confirmed the technol- gravity at ambient temperatures and is unsuita-

www.digitalrefining.com/article/1001431 PTQ Q4 2017 3

 Figure 7 Jackhammer sludge before Figure 8 Tank bottom visible after ITW Figure 9 Cargo pump freed from
cleaning cleaning sludge

ble for direct use in conventional power stations. er’s construction (roughly 30 years ago). Manual
Orimulsion is made by mixing bitumen with cleaning was not the solution however, as boiler
about 30% fresh water and a small amount of turnaround was scheduled for only 20 days and an
surfactant. The resulting emulsion behaves simi- important revamp had to be implemented.
larly to fuel oil. Because manual tank cleaning would be a dirty,
A 50 000 m3 Orimulsion tank contained about time consuming and risky operation, the manage-
1500 m3 of sludge, which the operator rated ‘mis- ment decided to test ITW technology. The pur-
sion impossible’ to remove. The operator wanted pose of the job was to have indications of cleaning
to convert the tank to store virgin naphtha to be during continuous injection of a proprietary ITW
fed to an ethylene plant and therefore the cleaning chemical: timely monitoring had to be applied to
requirements were very strict; even minimal resi- confirm sludge dissolution. Digging before start-
dues of black oil could not be tolerated. ing the job showed a 50 cm layer of coke-like solid
Additionally, no modification of the tank struc- material.
ture was allowed (for instance, no removal of any To give more added value to the application,
piece of the roof or walls) because time was not ITW formulated a tailor-made chemical contain-
available to carry out repairs. ITW completed the ing both asphaltene stabilisers and combustion
job in less than 30 days so that the operator could catalysts. The chemical was injected upstream of
immediately fill the tank with a virgin naphtha the service storage tank; the additivated fuel oil
cargo. entered the tank from the bottom.
After about one and a half months of treat-
Case history: onstream tank cleaning ment, digging in the tank revealed a 20 cm layer
A power station needed to clean a fuel oil service of solids, 30 cm of sludge with no measurable
tank which had not been cleaned since the boil- viscosity at 100°C, and 150 cm of more viscous
fuel oil.
These results were interpreted
Day 48 Day 75
in the sense that ITW additive,
Starting treatment Boiler by entering the tank from the
with MEG F 223
Day 0 bottom, was solubilising the
Analysis deposit sludge so that this solubilised
on filter
Day 43
sludge rendered the lower por-
Ash = 17.6% tion of the oil more viscous.
150 cm oil with
Asphaltenes = 8.8% higher viscosity After still another month of
than the upper layer Oil near sludge treatment, digging was per-
Analytical data:
30 cm sludge
Viscosity @ 100ºC formed in the tank again and
Analytical data:
Viscosity @ 100ºC
= 32.6 cSt the results were surprising: the
Ash = 0.01%
not measurable solid layer had disappeared, as
Ash = 7.4% 50 cm sludge had the very viscous sludge. In
20 cm solid layer Analytical data:
Fuel oil Viscosity @ 100ºC their place, a single non-viscous
MEG F 223 = 132 cSt sludge layer (viscosity 132 cSt at
Ash = 3.7%
100°C) was found. The results
Figure 10 Timing of onstream tank cleaning are summarised in Figure 10.

4 PTQ Q4 2017 www.digitalrefining.com/article/1001431

 The results of tank cleaning
Fuel oil characteristics during ITW injection
were also confirmed by those
in the preflame and post-flame
Parameter/analyses #1 #2 #3 #4 #5 #6
zones of the boiler. Since the Density @ 15°C, g/cc 0.9648 0.9695 0.9781 0.9688 0.9781 0.9695
additive contains both asphal- Viscosity @ 50°C, °E 48.1 32.2 45.4 23.6 44.4 46.4
Sulphur, wt% 2.3 2.14 2.09 2.5 1.73 1.93
tene stabilisers and combustion Water, vol% 0.3 0.1 0.3 0.1 0.2 0.5
catalysts, it continued its action Sediments for extrac., wt% 0.04 - - - - -
Asphaltenes, wt% 5.4 7.9 10.93 5.0 10.0 9.0
downstream of the tank. Both SHF, wt% 0.33 0.14 0.02 0.06 0.08 0.10
pre-flame and post-flame zones CCR, wt% 11.1 11.2 13.5 5.7 14.1 14.1
were also cleaner than before. Ash, wt% 0.011 0.012 0.02 0.01 0.01 0.01
Sodium, ppm 8 6 10 3 20 37
In particular, the hot filters’ ΔP Vanadium, ppm 41 66 55 12 40 37
was almost nil after cleaning, on Nickel, ppm 34 32 31 11 18 21
a normalised basis.
The fact that the filters did not Table 1
increase the ΔP further confirms
that the heavy sludge had been
fully converted into a stable liq-
uid. As additional confirmation,
fouling factor monitoring in the
preheaters revealed no increase
in fouling.
The most significant improve-
ments in boiler operating param-
eters can be summarised as
follows: Figure 11 SBR tank deposit Figure 12 SBR tank deposit
• Combustion chamber pressure dissolved by chemical treatment
decreased from about 260 mm
H2O to about 230 mm H2O Summary of tank cleaning costs
• Flue gas pressure drop in the
Item Conventional cleaning ITW cleaning
Ljungstroem decreased from Time to transform sludge to reusable product Not applicable 3-10 days1
about 100 mm H2O to an average Time to achieve safe entry conditions 10-20 days 1-3 days
of about 70 mm H2O Time to manually remove sludge 60-180 days1 6-20 days1
Total manpower >20 workers 1
6-8 workers
• Air temperature at the Ljung- Man-hours working in confined space 4400-43,200 h 440-4320 h
stroem outlet increased from Hazardous waste 4500 tons 450 tons
Cost to dispose and transport hazardous waste ($700/ton) 3 150 000 315 000
395°C to about 403°C Value of sludge when converted to a reusable product
• Normalised burner pressure (in this case, 26 500 bbl x $50/bbl) Not applicable 1 325 000
was almost constant. 1. Depending on tank size and amount of sludge
The improved cleanliness
of the boiler translated into Table 2
improved combustion efficiency:
specific steam production per Kcal of incom- A styrene butadiene rubber tank was scheduled
ing fuel improved from about 0.0566 tonnes to be cleaned for maintenance. The deposit was a
steam/fuel Kcal to about 0.0574 tons steam/fuel very hard solid which a mechanical cleaning con-
Kcal, an improvement of about 1%. tractor had failed to remove. ITW took a sam-
Table 1 summarises the inlet fuel oil character- ple of the solid and showed that the same could
istics during the period of continuous chemical have been dissolved by the company’s chemistry.
injection. The results of dissolution are shown in Figures 11
and 12.
Case history: SBR tank cleaning
In the petrochemical industry, ITW’s approach is Full value of cleaning
the same, but a proprietary polymer dissolution The above examples show that the mere ‘clean-
chemistry is used to perform the job. ing’ cost is only a portion (sometimes very small)

www.digitalrefining.com/article/1001431 PTQ Q4 2017 5

 of the real cost of the project. Tank cleaning is precipitation occurs following reprocessing as all of
not a standard job or simply a mechanical or the precipitated compounds are transformed into
chemical cleaning task. a liquid and stabilised. By utilising the technology
Each tank to be cleaned is normally a spe- for tank cleaning, it is therefore possible to achieve
cific case. Perhaps nobody knows the history of these results:
the tank, or the tank has collected sludge from • Real and effective recovery of hydrocarbons
cleaning activities in other tanks. Therefore the • Reduction of cleaning time
real value of cleaning will depend on the spe- • Reduction of waste disposal costs
cific application and, most importantly, will vary • Safe and environmentally friendly operations.
according to local environmental legislation,
waste disposal and transportation costs, and so Marcello Ferrara is the Chairman of ITW. He has 31 years’
on. experience in the petroleum business, holds international patents
To give some indication of the order of magni- for new processes and additive compositions for environmental
control and for improving petroleum/petrochemical processes,
tude of the task, Table 2 summarises some of the
and holds a PhD in industrial chemistry.
items which should be considered when evaluat- Email: [email protected]
ing the cleaning costs of a tank. Cristina Ferrara is a Process Engineer with ITW. She has five
years’ experience in the oil industry and in the design of cleaning
Conclusion programmes and holds a chemical engineering degree from
Palermo University, Italy.
Tanks are a valuable asset for any production site.
Email: [email protected]
No production can occur without a proper storage
facility. But cleaning can became a complex and
time consuming item if a proper strategy is not LINKS
chosen. The evaluation of tank cleaning technol-
ogy must follow a thorough evaluation of all of the More articles from: ITW Technologies
direct and indirect costs of cleaning.
More articles from the following categories:
ITW tank cleaning technology enables tank
Corrosion and Fouling Control
cleaning under no-entry conditions, thus avoiding Crude and Vacuum Units
environmental and safety concerns, but also stabi- Heavy and Sour Feedstocks
lises the sludge and transforms it into a fully reus- Storage Tanks and Terminals
able and reprocessable liquid product. No sludge

6 PTQ Q4 2017 www.digitalrefining.com/article/1001431