Porosity CH 5
Porosity CH 5
Porosity CH 5
5
CONTENTS
5 EFFECT OF HYDROCARBONS ON
POROSITY DETERMINATION
5.1 Acoustic Log
5.2 Density Log
5.3 Neutron Log
LEARNING OBJECTIVES
Having worked through this chapter the Student will be able to:
General:
1. Determine porosity from an acoustic, density and neutron log.
Acoustic Log
2. Describe the Wyllie time-average equation for determination of porosity and
identify the variables in the equation.
3. Describe the conditions in which this equation is valid, the terms in the
equation, and the sources of error in the use of the equation.
4. Describe the empirical equation for determination of porosity and identify the
variables in the equation.
5. Describe the conditions in which this equation is valid, the terms in the equation
and the sources of error in the use of the equation.
6. Describe which of the above is the most appropriate equation to use in a given
situation.
7. Describe the effects of hydrocarbon, shales, carbonates and uncompacted sands
on the acoustic log.
Density Log
8. Describe the equation for determination of porosity and identify the variables
in the equation.
9. Describe the conditions in which this equation is valid, the terms in the equation
and the sources of error in the use of the equation.
10. List the density of some common minerals, water, oil and gas.
11. List some of the minerals whose density cannot be read directly from the density
log.
12. Describe the effect of temperature, pressure and salinity on the density of
water.
13. Describe the effects of hydrocarbon, shales and pressure on the density log.
Neutron Log
14. List the conditions under which the neutron logs are calibrated.
15. List the sources of error and therefore corrections to the neutron log readings
to derive porosity.
16. Describe the effects of hydrocarbon and shales on the neutron log.
2
Determination of Porosity
5
1 INTRODUCTION
• Acoustic log,
• Density log and/or
• Neutron log.
The measurements of the neutron, density, and sonic logs depend not only on porosity
(φ) but also on the formation lithology, on the fluid in the pores, and, in some instances,
on the geometry of the pore structure. When the lithology and, therefore, the matrix
parameters (tma, rma, φma) are known, correct porosity values can be derived from any
one of these logs, appropriately corrected for environmental effects, in clean water-
filled formations.
This chapter will describe the way in which the porosity of a rock can be determined
from the:
• Acoustic log,
• Density log and/or
• Neutron log.
The most simple environment in which the porosity of a formation can be determined
is a clean water bearing formation. The first part of this chapter will therefore describe
how the porosity in a clean water bearing formation can be determined from an
Acoustic, Density and Neutron log.
• Hydrocarbons
• Shales
• Carbonates
1
and therefore their impact on determination of porosity when these situations exist
will be described at the end of the chapter.
The values listed are for nonporous substances i.e. for purely matrix material. Porosity
decreases the velocity of sound through the rock material and, correspondingly,
increases the interval transit time.
µs/ft)
∆ tma(µ
V ma(ft/se c) µs/ft)
∆ tma(µ (commonly
use d)
t LOG = φ t f + (1 - φ ) tma
or
Equation 1
t - tma Porosity using the Time-
φ = LOG
t f - tma Average Equation.
where
t Log is the reading on the sonic log in µs/ft,
t ma is the transit time of the matrix material,
t f is the transit time of the saturating fluid (about 189 µs/ft for freshwater mud
systems).
4
Determination of Porosity
5
Generally, consolidated and compacted sandstones have porosities of between 15 and
25%. In such formations, the response of the sonic log seems to be relatively
independent of the exact contents of the pores: water, oil, gas, or even disseminated
shale.
This empirical transform can be approximated over the range of normally encountered
porosities by the following equation:
Equation 2 φ SV = C
(tLOG − tma )
Porosity using the t LOG
Empirical Equation
The value of the constant C has a range of 0.625 to 0.7 depending upon the
investigator. 0.7 was originally proposed for C and this value is often used in chart
books (e.g. Chart Por-3). However, more recent transit time-to porosity comparisons
indicate 0.67 is more appropriate.
For the case of a gas-saturated reservoir rock, C becomes 0.6. It should be used when
the rock investigated by the sonic tool contains an appreciable amount of hydrocarbon
in the gassy (vapour) phase. Because of the very shallow depth of investigation, this
condition normally exists only in higher porosity sandstones (greater than 30%).
The empirical transform exhibits several salient features. First, it appears that all pure
quartz sandstones may be characterised by a unique matrix velocity, slightly less than
18,000 ft/sec. A value of 17,850 ft/sec (or tma = 56µs/ft) is suggested. Limestone and
dolomite also seem to exhibit unique matrix velocities: 20,500 ft/sec (or tma = 49µs/
ft) for limestone, and 22,750 ft/sec (or tma = 44µs/ft) for dolomite.
It can be seen from Chart Por-3 that in sandstone, the transform yields slightly greater
porosity values over the low to-medium-porosity range (i.e., the 5 to 25% range) than
does the time-average equation using an 18,000 ft/sec matrix velocity. In fact, at 15%
porosity the transform indicates a porosity similar to that given by the time-average
equation using a matrix velocity of 19,500 ft/sec. Thus, it appears the higher matrix
velocities used in sonic interpretation in the past have been selected to force the time-
average equation to yield a truer porosity over the low to medium range; this is true
for both carbonates and sandstones.
For moderately high porosity sands (21-24%), the proposed empirical transform
generally corresponds to the time-average equation using vma = 18,000 ft/sec. Above
30% porosity, however, sonic transit time increases much more rapidly than porosity,
and its response quickly departs from that predicted by the time-average equation.
This is the region in which the time-average equation would require a “lack of
compaction” correction. The transform eliminates the need for the correction factor
and yields porosity directly.
1
The density log responds to the electron density of the formations. For a substance
consisting of a single element, the electron density index, ρe, is related to the bulk
density by,
Equation 3
ρe = ρb
2Z Electron Density in a
A substance consisting of a
single element
where,
ρb is the actual bulk density,
Z is the atomic number (number of electrons per atom), and
A is the atomic weight (ρb / A is proportional to the number of atoms per cubic
centimetre of the substance).
Element A Z 2Z / A
H 1.008 1 1.9841
C 12.011 6 0.9991
O 16.000 8 1.0000
Na 22.990 11 0.9569
Mg 24.320 12 0.9868
Al 26.980 13 0.9637
Si 28.090 14 0.9968
S 32.070 16 0.9978
Cl 35.460 17 0.9588 Table 2
Atomic number and atomic
K 39.100 19 0.9719
weight for formation
Ca 40.080 20 0.9980
elements
For a molecular substance, the electron density index is related to the bulk density by,
Equation 4
∑ Z ′s
ρ e = ρ b 2 ,
Electron Density in a
Mol. Wt. molecular substance
where ΣZ ‘s is the sum of the atomic numbers of atoms making up the molecule (equal
to the number of electrons per molecule) and, Mol. Wt. is the Molecular weight.
For most formation substances, the bracketed quantities in Equations 3 and 4 are very
close to unity (Column 4 of Tables 2 and 3). When the density tool is calibrated in
freshwater-filled limestone formations, the apparent bulk density, ρa, as read by the
tool is related to the electron density index, ρe, by:
6
Determination of Porosity
5
ρa = 1.0704 ρe − 0.1883.
Table 3
Properties of various
formation substances
For liquid-filled sandstones, limestones, and dolomites the tool reading, ρa is practically
identical to actual bulk density, ρb. For a few substances, such as sylvite, rock salt,
gypsum, anhydrite, coal, and gas-bearing formations, the corrections shown in Figure 1
are needed to obtain bulk density values from the density log readings.
Figure 1
Correction needed to
generate true bulk density
from log density
ρb = φ ρ f + (1 − φ )ρma .
1
For usual pore fluids (excepting gas and light hydrocarbons) and for common
reservoir matrix minerals, the difference between the apparent density, ρa, read by the
density log, and the bulk density, ρb, is so trivial that it is disregarded. Solving for φ,
ρma − ρb Equation 5
φ = ,
ρma − ρ f Porosity from the Density
Log
where,
ρma is the matrix density (see Table 3).
ρf is the fluid density
ρb is the bulk density read from the log (=ρa)
The fluid in the pores of the permeable formations, within the relatively shallow zone
investigated by the tool (about 6 in.), is usually mostly mud filtrate. This mud filtrate
may have a density ranging from slightly less than 1 to more than 1.1 depending upon
its salinity, temperature, and pressure. Figure 2 shows the densities of water and NaCl
solutions at various temperatures, pressures, and salinities. At 750 F and atmospheric
pressure, the relation between NaCl water salinity and density may be approximated
by:
ρw = 1 + 0.73 P,
Chart Por-5 shows density log (FDC log) porosities versus ρb readings for various
matrices and ρf values of 1 through 1.2.
8
Determination of Porosity
5
Figure 2
Density of water at various
temperatures, pressures and
salinities
The responses of the neutron tools primarily reflect the amount of hydrogen in the
formation. Since oil and water contain practically the same amount of hydrogen per
unit volume, the responses reflect the liquid-filled porosity in clean formations.
However, the tools respond to all the hydrogen atoms in the formation, including those
chemically combined in formation matrix minerals. Thus, the neutron reading
depends mostly on the hydrogen index of the formation. The hydrogen index is
proportional to the quantity of hydrogen per unit volume, with the hydrogen index of
fresh water at surface conditions taken as unity.
The standard conditions for CNL tool and Dual Porosity tool calibration are:
If there are departures from these conditions, the logs will require corrections. The
combined correction for all factors, usually small, yields a value of corrected neutron
porosity index. Chart Por - 14c provides the corrections to the CNL and Dual Porosity
thermal neutron measurements for borehole size, mudcake thickness, borehole and
formation-water salinities, mud weight, pressure, and temperature.
10
Determination of Porosity
5
5.2 Density Log
If residual hydrocarbons exist in the region investigated by the FDC tool, their
presence may affect the log readings. The effect of oil may not be noticeable because
the average fluid density, ρf (from ρo and ρmf ) will probably still be close to unity.
If there is appreciable residual gas saturation, its effect will be to lower the ρa. Figure
3 shows the corrections that must be added to the recorded ρa values to obtain true ρb
values when low-pressure gas or air (ρg approx. = 0) occupies the pores.
Figure 3
Correction needed to
generate true bulk density
from log density
The apparent density of gas, as seen by the density log, can be computed if the
composition and density of the gas are known. Figure 4 is a chart showing, for a gas
of specified composition, the values of ρgas (actual density) and ρagas, the apparent gas
density seen by the density tool (based on electron density) as a function of pressure
and temperature. If formations saturated with gas are in the vicinity of the borehole,
use ρga instead of ρf in Equation 5.
1
Figure 4
Gas density and hydrogen
index as functions of
pressure and temperature
for a gas mixture slightly
heavier than methane
The quantitative response of the neutron tool to gas or light hydrocarbon depends
primarily on hydrogen index and another factor - the “excavation effect.” The
hydrogen index can be estimated from the composition and density of the hydrocarbon.
For light hydrocarbons (gases), Figure 4 provides an estimate of its hydrogen index,
Hh . The hydrogen index of heavier hydrocarbons (oils) can be approximated by the
equation:
Equation 6
H0 = 1.28 ρ0 . Hydrogen Index of Oils
This equation assumes the chemical composition of the oil is n CH2. Ho is derived
from the comparison of the hydrogen density and molecular weight of water to those
of oil.
12
Determination of Porosity
5
Another set of equations can be used to estimate the hydrogen index of hydrocarbon
fluids:
Equation 9
4 - 2.5 ρh
General Hydrogen Index of Hh = 9 ρh .
Oils 16 - 2.5 ρh .
Mathematical investigations indicate that the effect of gas in the formation near the
borehole is greater than would be expected by taking into account only its smaller
hydrogen density. Previous calculations had been made as if the gas-filled portion of
the porosity were replaced by rock matrix. The new calculations show that when this
additional rock matrix is “excavated” and replaced with gas, the formation has a
smaller neutron-slowing characteristic. The calculated difference in the neutron log
readings has been termed the excavation effect. If this effect is ignored, too-high
values of flushed-zone gas saturation and too-low values of porosity are given.
Figure 5 shows the corrections needed for excavation effect. The values of porosity
for sandstone, limestone, and dolomite lithologies are plotted. Intermediate porosity
values can be interpolated.
The ordinate scale is used to correct neutron log porosities. An additional ordinate
scale is provided for correcting porosities derived from a neutron-density cross plot
that does not contain the excavation effect correction. Excavation effect corrections
have already been incorporated into Chart CP-5.
The corrections for excavation effect given by Figure 5 can be approximated by the
formula
Equation 10
Corrections for excavation [ ]
∆φ Nex = K 2 φ 2 SwH + 0.04 φ (1 - SwH )
effect
where ∆φNex, φ, and SwH are in fractional units. For sandstone the coefficient, K, is 1;
for limestone it is about 1.046, and for dolomite it is about 1.173. Note that the second
term of this equation is rather small and can often be disregarded.
1
Figure 5
Correction for excavation
effect as a function of Sxo
for three values of porosity
and for Hg. Effects of
limestone, sandstone and
dolomite included within
the shaded bands.
14
Determination of Porosity
5
6.2 Density Log
Interpretation of the density log can be affected by shale or clay in the formations.
Although the properties of shales vary with the formation and locality, typical
densities for shale beds and laminar shale streaks are of the order of 2.2 to 2.65 g/cm3.
Shale densities tend to be lower at shallow depths where compacting forces are not as
great. Dispersed clay or shale disseminated in the pore spaces may have a somewhat
lower density than the interbedded shales
Also, the neutron tool measures water of crystallisation. For example, nonporous
gypsum (CaSO4 + 2H2O) has a large apparent porosity because of its significant
hydrogen content.
In vuggy formations, the velocity of sound seems to depend mostly on the primary
intergranular porosity, and the porosity derived from the sonic reading through the
time-average formula (φSV) will tend to be too low by an amount approaching the
secondary porosity. Thus, if the total porosity (φt) of a formation exhibiting primary
and secondary porosity (φ2) is available (from a neutron and/or density log, for
example, the amount of secondary porosity can be estimated:
Equation 11
Secondary Porosity in φ 2 = φt - φ SV
vuggy formations
When the formations are not sufficiently compacted, the observed t values are greater
than those that correspond to the porosity according to the time-average formula, but
the φ versus t relationship is still approximately linear. In these cases, an empirical
correction factor, Cp, (Equation 12) gives a corrected porosity, φSVcor :
Equation 12
t - tma 1
φ SVcor = Porosity in uncompacted
t f − tma Cp
sands
The value of Cp is given approximately by dividing the sonic velocity in nearby shale
beds by 100. However, the compaction correction factor is best determined by
comparing φSV, as obtained from Equation 1, with the true porosity obtained from
another source. Several approaches are possible.
The R0 method: Compare the sonic and induction or laterolog values in a clean water
sand. The value of R0 found from the resistivity is divided by Rw to obtain F. Then
φ is found from F (Chart Por-1) and compared with φSV from Equation 1 (sonic
porosity without compaction correction). The value of Cp is equal to φSV/φ. This value
of Cp can then be used to analyse nearby potential hydrocarbon-bearing sands.
Density-sonic cross plot method: When sonic and density logs are available, ρb (in
ordinate) and t (in abscissae) values are cross plotted over several sands in the interval
of interest. If the sands contain no gas and some of them are clean, a line drawn from
the matrix point through the points lying toward the upper right will be the clean sand
line (Figure 6). For any given porosity value on this clean sand line, there will be a
value of t. Enter this t in Chart Por-3 and go vertically to the φ value. The intersection
will give the value for Cp. If a sand is known to be clean and liquid filled, then Cp =
φSV/φD.
Neutron method: The previous two methods require a clean sand. If the sands are
shaly, neither method can be safely used. If an SNP or CNL* neutron log is available,
φN may be compared with φSV (or t ) using Chart Por-3. Differences between φN and
φSV in water-filled sands are due to lack of compaction. For such sands, Cp = φSV/φN.
16
Determination of Porosity
5
Figure 6
Density-sonic crossplot as
used for compaction-factor
determination
The sonic travel time in shales normally decreases with increasing burial depth. A plot
of this trend, tsh versus depth, defines the normal compaction. Departures from this
trend toward higher values suggest an abnormal, over-pressured section (Figure 7).
With experience in the area, the magnitude of the over-pressure can often be related
to the difference between the actual shale transit time and that expected from the
normal compaction trend line.
1
Figure 7
Detecting Overpressured
zone using the sonic log
18
Determination of Porosity
5
Figure 8
The Density log in
overpressured shales
1
The readings of all neutron logs are affected to some extent by the lithology of the
matrix rock. SNP (Sidewall Neutron Porosity) and CNL (Compensated Neutron
Log) logs are usually scaled for a limestone matrix. Porosities for other lithologies are
obtained from Charts Por-13a and Por-13b or from scales on the log headings. The
SNP corrections apply only to logs run in liquid-filled holes. When the hole is gas
filled, the lithology effect is reduced to a negligible level, and porosity may be read
directly subject to limitations.
Shear-Wave Interpretation
All the preceding discussion has concerned compressional transit time interpretation.
With the Array-Sonic tool and full-waveform recording, it is now possible to obtain
shear-wave transit time measurements on a more routine basis. Application of the
shear wave in formation evaluation is only now beginning to be explored. It is obvious
that shear-wave velocity data will be useful in calculating rock elastic or inelastic
properties and as an adjunct to shear seismic data.
Shear-wave transit time data are also useful in identifying matrix minerals and pore
fluids. For example, a cross plot of compressional transit time, tc, and shear transit
time, ts, can be used to identify the mineral content of the various rocks traversed by
the wellbore. The technique is similar to other porosity log cross plotting techniques
(e.g., density-neutron, sonic-density, sonic-neutron).
There is evidence that the shear-wave transit time may be useful for fluid identifica-
tion. Laboratory observations suggest that light hydrocarbon saturation decreases the
velocity of the compressional wave (relative to brine saturation) through the porous
rock and increases the velocity of the shear wave.
A relationship between porosity and shear velocity (or interval transit time) has also
been noted. Indeed, the time-average relationship (Equation 1) and the empirical
relationship (Equation 2) that relate compressional transit time to porosity appear to
apply to shear transit time as well. Of course, appropriate matrix and fluid parameters
must be used. For shear-wave propagation, the parameters are approximately:
20
Determination of Porosity
5
Sandstone, tma = 86 µs/ft
Limestone, tma = 90 µs/ft
Dolomite, tm = 76 µs/ft
Anhydrite, tm = 100 µs/ft
Water, tm = 350 µs/ft
These values are tentative. Further experience with the shear transit time may lead to
some refinement. Also, the listing of a shear transit time value for water is somewhat
imaginary since water does not support shear-wave propagation. However, the use
of the listed value for water in the time-average equation does seem to yield acceptable
porosity values.
1
Chart POR-1
Chart POR-3
Chart POR-5
Chart POR-13a
Chart POR-13b
Chart POR-14c
Chart POR-15a
Chart CP-5
22