SPE-175557-MS

A Comparative Study of Various Laboratory Methods for Capillary

Pressures Determination in Conventional Oil Fields of Kazakhstan.
Marlen Tleuberdinov, Nurlan Zhulomanov, and Alimzhan Adilbayev, Weatherford Laboratories

Copyright 2015, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Reservoir Characterisation and Simulation Conference and Exhibition held in Abu Dhabi, UAE, 14 16 September
2015.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents
of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Society of Petroleum Engineers, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written
consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may
not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract
Since relative permeability and capillary pressure studies are of great importance and considered by
majority of exploration companies as studies of key parameters used in reservoir engineering, we have
decided to conduct research studies on capillary pressure curves. Unfortunately, some oil companies in
Kazakhstan are not familiar with SCAL studies; therefore, we provide guidelines to create SCAL
programs based on both technical requirements and financial resources. Results of our laboratory studies
of various fields of Kazakhstan are presented in this paper comparing different methods to determine
capillary pressure for conventional oil sandstones. Our studies included comparison between three
fundamental techniques for capillary pressure determination: mercury injection, centrifuge and porous
plate methods. Theoretical approach suggests that porous plate method is the most accurate since
irreducible water saturation is achieved more uniformly. However, each method has its own advantages
and disadvantages. Those include such factors as: time, presence of a wetting phase, accurate capillary
pressures within both high and low saturation zones. We evaluated 25 samples from 2 fields in
Kazakhstan. Analyzing capillary pressure curves helped us to distinguish various effects in conventional
reservoirs and uncertainties associated with them.
We also compared different models of J-function. It is defined that in some cases Pc curves based on
Leverett J-function fail to display realistic values against wetting-phase saturations. Ideally, this model
works in Berea sandstone reservoirs. However, many factors such as lithological characteristics, clay and
minerals content, distribution of fluids affect Pc models. We were able to identify limiting range of
applicability and further we provided our recommendations on which Pc models should be used for
different types of reservoirs. Summarizing aforementioned, comprehensive recommendations with regard
to type of the reservoir rock were designed which is a discussion point whilst negotiating with customers.

Introduction
Capillary pressure, Pc, is the difference in pressure across the interface between two phases. It depends
upon interfacial tensions, wetting angles, porosity, permeability and fluid saturations. It is the most
fundamental rock-fluid property in multi-phase flows which describes flow processes in the reservoir
(Green 2008). The petroleum industry has been using three methods for determining capillary pressure

SPE-175557-MS

curves. Those are porous plate, centrifugation and mercury injection. Each method has its own advantages
and limitations. In this paper we have evaluated seven different fields from two distinct basins in
Kazakhstan. Based on laboratory experiments on these samples we have noticed both similarities and
significant differences in results. This paper discusses causes and effects of difference in capillary pressure
as well as recommendations to get the most accurate data for a particular scenario. Since SCAL services
are not cheap, it is essential to design optimized SCAL programs in order to follow cost-saving policy
which is of current interest since oil industry is currentrly struggling.

Procedure
Cleaning
In order to perform analyses on core samples, the core has to be properly cleaned and dried. For the
cleaning procedure use of methanol/toluene/isopropanol/chloroform is required. Each company has its
own procedure for mixing the before mentioned chemicals. The cleaned core has to be dried in the oven
at the specified temperature until complete dryness of a core sample is achieved. The drying has to be done
very cautiously, since there might be clay content in some sandstones which could be damaged and
restoration of the core sample becomes no longer possible. This could greatly affect the results resulting
in a wrong interpretation. Hands on core description by a trained geologist before cleaning of samples may
be useful so that the proper technique is applied to clean and dry samples.
Mercury Injection
Samples are weighed and loaded into the glass cells. Mercury is forced into the evacuated pores of
samples until desired 60000 psi pressure is achieved. The injected volume of mercury is recorded and used
to determine vacuum/mercury capillary pressure curves.
The mercury acts as a non-wetting fluid. The major assumption for this method is the wettability of the
rock, since here it is assumed that the rock is strongly water wet. The main advantage of this method is
to be able to get a good data on pore geometry of the rock. Also, the method is quick and can be
accomplished within several hours.
Effect of Heterogeneity: Before putting the sample into the apparatus, the sample has to be carefully
observed for presence of any inclusions, fractures etc. There might be significant differences in results
comparing to other two methods.
Effect of Clay content: can be misleading. MICP results can greatly vary from those of the porous plate
method. Since some clay minerals such as smectite tend to swell and block the pore throats, irreducible
water saturation can be higher for a porous plate method, whereas the mercury does not react with smectite
and hence no blockage of pore throats resulting in low values for water saturation (Newsham 2004).
Porous Plate
The porous plate method is considered to be the most accurate for determining capillary pressure. The core
samples are 100% saturated with brine, and then placed in core holders where confining pressure is
constant. The capillary pressure system is between air and brine. The pressure incrementally increased as
water gets squeezed out of the sample. Before each increment the water production has to completely
stabilized i.e until no water is expelled from the core. The pressure is increased up to 200 psi. The water
gets squeezed uniformly which results in a uniform capillary pressure distribution. As opposed to the
mercury method, where wettability of the rock is strong this method uses actual wettability which in turn
provides different results. Comparing to the faster centrifuge technique, porous plate method is direct
procedure which requires less data processing (Raghuraman 1998).
Centrifuge
The method of centrifuge was introduced by Hassler and Brunner (Green 2007). Samples are spun at
different rotational speeds and the amount of expelled water is measured. Therefore, production stabilization is achieved to obtain capillary pressure curves.

SPE-175557-MS

Discussion
In this work, we compared three fundamental techniques that are used to measure capillary pressure in
laboratory conditions. Each technique has its own advantages and disadvantages.
Mercury injection is quick and relatively cheap. There is a good representation of pore geometry
achieved using this method. However, there are some limitations: sample is destructed after the test;
results are a few percent higher for residual water saturation (very optimistic). Also, accuracy of Pc curves
for high clay content samples is questionable (Sabatier 1994).
Centrifuge is a valuable tool for measuring different rock parameters such as capillary pressure,
wettability and relative permeability. Centrifuge technique is relatively fast, allows getting full capillary
pressure curves in several days to several weeks, depending on sample properties. As opposed to porous
plate method, centrifugal forces can create non-uniform fluid distribution along core length. This makes
data interpretation difficult; however, there are several methods (reference to Forbes and Hassler-Brunner)
that are proved to give good results. In these calculations end-face saturations are used to plot Pc curve,
which refer to saturation at the inlet of the core (Christiansen 1992). In addition centrifuge tests could be
run under reservoir confining stress and temperature, which allows accounting for wettability changes.
However, in comparison with porous plate method, centrifuge is expected to be more influenced by
capillary end-effect (Sylta 2010).
The comparison of J-function and capillary pressure.
In the first field, J function distribution from MICP and PcRI showed 2 distinct lithotypes i.e four silty
dolomites and one coarse-grained siltstone. For centrifuge similar rocks were evaluated and different
J-function distribution was observed. The capillary pressure via mercury injection gave very optimistic
irreducible water saturation; whereas porous plate showed higher value for water saturation. This could
be attributed to wettability of the rock, which was taken in account with a proper brine composition in the
porous plate method. On the contrary, mercury injection method correlates fairly well with strongly
water-wet rocks. Centrifuge results fell into a good range with other methods. Some variations that can
be observed on the graphs are might be due to the interpretation method. There are a few industry
recognized methods such as Forbes, Hassler-Brunner, Rajan, etc. For the purpose of this test, results were
interpreted using Forbes method. Results may greatly vary depending on an interpreter. Also, unlike the
other two methods, centrifuge method is indirect thus, it could add to the uncertainty in the capillary
pressure results.
In the second field, silty sandstones were evaluated. The J-function distribution showed one large
lithotype, whereas capillary pressure was not quite consistent. The main reason for significant difference
for capillary pressure was high heterogeneity of core samples, which had high clay inclusions. The trims
selected for mercury injection could be unrepresentative due to high heterogeneity and great variation of
clay content across the core.

Conclusions
We have conducted a laboratory study comparing different capillary measurement techniques. We would
like to summarize our results and offer the following conslusions:
1. In air-brine system, MICP does not account for a content of clays in samples. In some cases
mercury injection method demonstrated underestimation of water saturation obtained by the
porous plate method (Greder 1997).
2. We believe that the tests that use true wetting and non-wetting phases (air-brine system) represent
more realistic curves since mercury injection does not include any true wetting phase. All MICP
data is converted from vacuum-mercury to other fluid systems numerically which is questionable
since we create a pseudo air-water system (Newsham 2004). Meanwhile, mercury injection
could be used to quantify the degree of heterogeneity of rock samples (Greder 1997).

SPE-175557-MS

3. Centrifuge method is a relatively fast and reliable technique to obtain realistic Pc curves. However,
it is of fundamental importance to account for capillary end-effect. Partially this can be done by
rotating samples long enough to reach equilibrium saturation state.
4. Porous plate method is the most accurate method which involves using reservoir conditions,
wettability effects.
Overall, this model is supposed to help exploration companies to improve understanding of SCAL
techniques, to deisng more efficient and less expensive SCAL programs and to lessen uncertainties in
reservoir models.

Table 1Basic petrophysical properties

Sample
7x
7
10
21ds
24ds
1
2
3
4
5

Field

Porosity

Gas permeability

Grain density

Lithology

1
1
1
1
1
2
2
2
2
2

0,270
0,192
0,258
0,234
0,244
0,213
0,182
0,214
0,190
0,199

133,5
6,08
105,9
103,6
84,5
3,37
2,57
2,75
2,58
2,85

2,67
2,81
2,82
2,81
2,80
2,65
2,65
2,65
2,65
2,65

coarse-grained siltstone
silty dolomite
silty dolomite
silty dolomite
silty dolomite
silty sandstone
silty sandstone
silty sandstone
silty sandstone
silty sandstone

Figure 1Capillary pressure sample 1

SPE-175557-MS

Figure 2Capillary pressure sample 2

Figure 3Capillary pressure sample 3

SPE-175557-MS

Figure 4 Capillary pressure sample 4

Figure 5Capillary pressure sample 5

SPE-175557-MS

Figure 6 Capillary pressure sample 7x

Figure 7Capillary pressure sample 7

SPE-175557-MS

Figure 8 Capillary pressure sample 10

Figure 9 Capillary pressure sample 21ds

SPE-175557-MS

Figure 10 Capillary pressure sample 24ds

Figure 11J-function

10

SPE-175557-MS

Figure 12J-function

Figure 13J-function

Figure 14 J-function

SPE-175557-MS

11

Figure 15J-function

Figure 16 J-function

Nomenclature
Pc:
MICP:
SCAL:
Swi:

capillary pressure
Mercury Injection Capillary Pressure
Special Core Analysis
initial water saturation

References
1. Green, D., Gardner, J. S., Balcom, B. J., McAloon, M., & Cano-Barrita, J. 2008. Comparison
Study of Capillary Pressure Curves Obtained Using Traditional Centrifuge and Magnetic Resonance Imaging Techniques. Presented at SPE Symposium on Improved Oil Recovery, 20-23
April, Tulsa, Oklahoma, USA. SPE-110518-MS. http://dx.doi.org/10.2118/110518-MS
2. Greder, H. N., Gallato, V., Cordelier, Ph., Laran, V., Munoz, V., dAbrigeon, O. 1997. FORTY
COMPARISONS OF MERCURY INJECTION DATA WITH OIL/WATER CAPILLARY
PRESSURE MEASUREMENTS BY THE POROUS PLATE TECHNIQUE. Presented at the
International Symposium of the Society of Core Analysts held in Calgary, AB, Canada.

12

SPE-175557-MS

3. Newsham, K. E., Rushing, J. A., Lasswell, P. M., Cox, J. C., & Blasingame, T. A. 2004. A
Comparative Study of Laboratory Techniques for Measuring Capillary Pressures in Tight Gas
Sands. Presented at SPE Annual Technical Conference and Exhibition, 26-29 September, Houston, Texas. SPE-89866-MS. http://dx.doi.org/10.2118/89866-MS
4. Sylta, K-E.H. 2010. Primary Drainage Capillary Pressure Curves in Heterogeneous Carbonates
with Ultracentrifuge and NMR. Bergen, Norway: Centre for Integrated Petroleum Research.
Department of Physics and Technology, University of Bergen.
5. Sabatier, L. 1994. COMPARATIVE STUDY OF DRAINAGE CAPILLARY PRESSURE MEASUREMENTS USING DIFFERENT TECHNIQUES AND FOR DIFFERENT FLUID SYSTEMS. Presented at the International Symposium of the Society of Core Analysts held in
Stavanger, Norway
6. Christiansen, R. L., & Cerise, K. S. 1992. Geometric Concerns for Accurate Measurement of
Capillary Pressure Relationships With Centrifuge Methods (includes associated paper 27739).
SPE Formation Evaluation 704: 311314. SPE-19026-PA. http://dx.doi.org/10.2118/19026-PA
7. Raghuraman, B., Ramakrishnan, T.S., & Supp, M. 1998. CAPILLARY PRESSURE AND
RESISTIVITY MEASUREMENTS USING THE CENTRIFUGE: A NEW INTERPRETATION
METHOD. Paper SCA-9811 in the Proceedings of the International Symposium of the Society of
Core Analysts held in Hague, Netherlands
8. Green, D.P., Dick, J.R., Gardner, J., Balcom, B.J., Zhou, B. 2007. COMPARISON STUDY OF
CAPILLARY PRESSURE CURVES OBTAINED USING TRADITIONAL CENTRIFUGE
AND MAGNETIC RESONANCE IMAGING TECHNIQUES. Presented at the International
Symposium of the Society of Core Analysts held in Calgary, Canada.