SPE 145703

Fluid Composition Equilibrium; a Proxy for Reservoir Connectivity

1,2 2 1 1 2
Thomas Pfeiffer , Zulfiquar Reza , David S. Schechter , William D. McCain , Oliver C. Mullins
1)
Texas A&M University
2)
Schlumberger
Copyright 2011, Society of Petroleum Engineers

This paper was prepared for presentation at the SPE Offshore Europe Oil and Gas Conference and Exhibition held in Aberdeen, UK, 6–8 September 2011.

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 th e paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessar ily 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 prohi bited. 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

Assessing reservoir connectivity during the earliest stages of reservoir evaluation is highly desirable for
successful field development. Nevertheless, it has long been problematic to assess reservoir connectivity prior to
production. Recently, downhole fluid analysis has enabled facile assessment of fluid compositional gradients
vertically and latterally. Using equations of state, the extent of fluid compositional equilibrium can be established.
Only a process that stretches across the entire age of the reservoir is likely to capture geologic events that cause
compartmentalization. Fluid composition equilibration requires mixing of the entire content of the reservoir which
occurs only on the geologic time scale. Restrictive flow barriers are not compatible with thorough mixing of fluids
throughout the reservoir. Fluid composition equilibration provides a tight constraint to test connectivity.
In this paper, the time constants for fluid composition equilibration are evaluated in numerical simulations.
Equilibration processes are simulated in a simplified model over geologic timescales at isothermal conditions
where diffusion and gravity are the active mechanisms. A variety of initial conditions and reservoir fluid types are
considered. The effect of barriers on the equilibration time is investigated for single and multiple barriers. The
results are compared with analytical calculations. Longer equilibration times correspond to tighter constraints on
connectivity.
This work shows the progression of compositional gradients over geologic time until all components have reached
zero mass flux. It investigates the foundation of connectivity studies that rely on fluid composition.
Determination of fluid equilibrium should become part of the standard procedure for reservoir connectivity
evaluation.

INTRODUCTION

A compartment is a part of a reservoir that needs to be penetrated by a well to be drained. Undetected flow
barriers may lead to disappointing production results. Establishing reservoir connectivity prior to production is
highly desirable.

Identifying compartments is not very difficult. The use of wireline formation testers to acquire static formation
pressure surveys is common practice. The word static refers to pressure gradient analysis in single or in multiple
wells prior to production at virgin conditions. Even subtle discontinuities and differences in pressure gradients can
indicate reservoir compartmentalization (Canas et al. 2008; Elshahawi et al. 2005; Jackson et al. 2007).

Identifying reservoir connectivity is more difficult, especially prior to production. Static pressure surveys fail at
identifying reservoir connectivity. Care must be taken when the survey indicates a continuous gradient between
2 SPE 145703

different parts of the reservoir. Equilibrated pressure over geologic time does not guarantee flow communication
during the production time.

There are two contributing factors that need to be considered: Mass flow and Time. The amount of mass flow that
is necessary to equilibrate a certain amount of differential pressure depends on compressibility. Compared to the
entire volume of the reservoir this amount is small. The amount of time it takes to equilibrate pressure depends on
permeability, viscosity, compressibility and the geometry of the system. It is very short compared to age of the
reservoir. Example calculations and numerical reservoir simulations clearly support these statements and are
detailed later in this paper.

A small amount of fluid will take a short amount of time to equilibrate pressure across a barrier. It is therefore
likely that a flow barrier does not to cause a discontinuous static pressure profile. Consequently, to identify such a
barrier, a different approach than static formation pressure surveys is needed. A process that involves moving the
entire content of the reservoir and that takes a very long time to do so ist fluid compositional equilibration. To
reach fluid compositional equilibrium the entire content of the reservoir needs to mix. The dominant mechanism of
reservoir mixing is molecular diffusion, a very slow process.

The understanding of the properties of the hydrocarbon column is vital to the optimal development plan of a field.
The assumption that any period of geologic time assures fluid equilibrium or even uniform fluid columns is grossly
inaccurate (Mullins 2008).

A technology that reveals fluid property variations in the reservoir as part of wireline formation testing is DFA. The
fluid properties become available as soon as the data is acquired. Fluid composition, inferred from the data, is
used for EOS modeling under the assumption that the fluid column is compositionally equilibrated (Xian et al.
2010). If the EOS-predicted composition does not match a subsequent measurement, the assumption that the
fluid column is equilibrated may not hold. A strong indication of connectivity is given if all measurements match
the EOS prediction, as mass flux from all parts of the reservoir is necessary to reach compositional equilibrium.

For a variety of reasons, it is preferable to assess fluid equilibration using downhole fluid analysis (DFA) and
using the laboratory analysis as complementary measurements. With DFA, fluid complexities are recognized in
real time and a more comprehensive data suite can be obtained with the wireline formation tester. Laboratory
analysis does not afford this capability. In addition, with DFA, systematic errors often cancel, especially with the
same tool, same well, same operator, and same calibration. Laboratory analyses have absolute error bars on
each analysis. Moreover, the DFA measurement of relative asphaltene content is based on optical absorption
measurements and is far superior to physical separation methods used in the laboratory (Mullins 2008). Finally,
the new Flory-Huggins-Zuo equation of state for asphaltenes in reservoirs has been developed with a focus on
DFA data (Zuo et al. 2008).

STATIC FORMATION PRESSURE SURVEYS AND DOWNHOLE FLUID ANALYSIS

Naturally, there are statistical errors and limitations attached to the pressure measurement. For formation
fluids of low densities, like gases and gas condensates, the limitations of the sensor technology may become an
issue. Kabir and Pop (2007) discuss this subject and present case studies and outline methods to quantify the
accuracy of the pressure survey. Hashem et al. (2004) present further techniques to improve the data quality of
formation testing.

Pressure equilibration occurs relatively quickly in comparison to the age of the reservoir and can happen across
flow barriers (Muggeridge and Smalley 2008; Mullins 2008). If the gradient suggests a continuous pressure
profile, compartmentalization cannot be ruled out. The advantage of integration of DFA and pressure gradient
analysis has been indicated by numerous authors (Dubost et al. 2007; Elshahawi et al. 2005; Montel et al. 2003;
O'Keefe et al. 2007; Venkataramanan et al. 2006; Xian et al. 2010). Pressure analysis is still the first step in
SPE 145703 3

identifying compartments. After this first step, DFA is applied to characterize the fluid content in the zones that
were previously identified by the pressure analysis. Some parts of the reservoir may exhibit a continuous pressure
profile. DFA helps to establish if they are in flow communication. Two case studies proved the DFA prediction of
connectivity (based on evaluating asphlatene content and GOR distributions) were correct with production data
(Mullins et al. 2010).

RESERVOIR CHARGING AND MIXING

Reservoirs fluids are not compositionally equilibrated at the time of charging. Stainforth (2004) tests two end-
member hypotheses for trap filling. One assumes perfect mixing of the fluid in the trap and one assumes no
mixing at all. In the majority of cases, the latter assumption matches reality. Layers of fluids stack into the
reservoir from the top down in the order they arrive from the source rock. Figure 1 visualizes this process
schematically.

Heavier components are liberated first from the source rock, and then lighter components follow. These light
components do not mix well with the contents that are already present in the reservoir; rather, they get to the top
of the reservoir via a high permeability path.

Figure 1 Charge history of a reservoir. The fluids layer into the trap in the order they arrive from the source rock
(Stainforth 2004). Graphic from Mullins (2008).

Gas may be present throughout the entire trap-filling process. A chemical potential is created by the difference in
concentration of each component throughout the communicating fluid column. Molecular diffusion acts to
eliminate the chemical potential. Gravity pulls the heavier components towards the bottom of the reservoir. When
the sum of all forces on each molecule adds up to zero, the component mass flux stops and compositional
equilibrium is reached. At that point the entirer content of the reservoir had to be moved to reach this state.
Convective forces are only relevant in settings of extreme temperature gradients. The critical Raleigh number, a
measure of the ratio between the viscous and the buoyant forces, is rarely exceeded (England et al. 1995).

DIFFUSION COEFFICIENTS

In the absence of convection, diffusion is the prevailing mechanism to reach fluid compositional equilibrium.
2
Albert Einstein (1905) related the mean square displacement R of a particle under Brownian motion to the
product of its molecular diffusion constant D and the diffusion time t:
4 SPE 145703

A representative choice of diffusion coefficients is not trivial. The Einstein-Stokes-Equation relates the diffusion
coefficient D inversely to the viscosity of the fluid μ – with the Bolzman constant kB and absolute temperature T.

The diffusion coefficient of a spherical particle is also a function of its radius r. Thus, as a frist approximation,
when comparing the diffusion coefficients of hydrocarbons the difference should be in the order of the cubic root
of the chain length.

Freed (2007) found that the self-diffusion coefficient of each component in a petroleum fluid mixture scales with its
chain length in form of a power law. Freed (2009) presents a method to apply a correction to the diffusion
coefficients for use at elevated temperatures and pressures. Freed’s method takes the mean chain length of the
entire mixture into account. Thus it can be applied to any composition of petroleum reservoir fluid—gas or liquid.
Based on the review of the current literature, the diffusion coefficients of the pseudo components used in this
work are best approximated using Freed’s power law.

The choice of diffusion coefficients is the result of an extensive literature review. For a summary of this review the
reader is refered to the publication by Pfeiffer et al. (2011).

SIMULATION MODEL

Two possible reservoir settings are compared in a theoretical model: a laterally extended reservoir and a
thick, block-type reservoir. The distance over which diffusion takes place varies by an order of magnitude
between the two settings. Two imaginary wells in Figure 2 and Figure 3 visualize possible access points to
evaluate the fluid composition. The same depth difference is linked to a different diffusion length in each model.

Figure 2 Visualization of a laterally extended reservoir. The reservoir is found 200m deeper in well 2 than in well
1. The fluid composition must equilibrate across a distance of 2000m. A baffle to the flow (reduction in
permeability and porosity) is introduced halfway between the wells.

A simple reservoir model is constructed to numerically simulate pressure and fluid compositional equilibration in a
static, closed system under isothermal conditions. Various initial conditions are simulated, in some cases for up to
150 million years. A certain level of simplicity of the model is necessary to keep the run times of the simulation
reasonable.

Figure 3 Visualization of a thick, block-type reservoir. Regardless of the lateral distance between the wells, the
fluid composition equilibrates over a constant distance of 200m. A baffle is introduced in the middle of the sand
similar to the baffle in Figure 2
SPE 145703 5

A vertical stack of 100 grid blocks (200m×500m×2m) is set up in two different ways to represent the two reservoir
settings. In the thick, block-type setting the grid blocks are vertically stacked directly on top of each other (Figure
4). In the laterally extended setting, corner point geometry shears the blocks like a deck of cards over a distance
of 2000m (Figure 5).

Both models are initialized with the following parameters: temperature 150°C and 100% hydrocarbon saturation in
single phase. Capillary pressure effects are not considered in this work. Porosity, permeability and initial pressure
distribution are summarized in Table 1.

Figure 4 Simulation model of a thick, block-type reservoir. 100 grid blocks (500m×200m×2m) stacked to form a
one-dimensional model. Grid blocks 49 and 50 have reduced permeability and porosity. The top half of the model
is initialized with uniform pressure and fluid composition that differ from the initial condition in the bottom half.

Figure 5 Laterally extended reservoir. 100 grid blocks (500m×200m×2m) sheared sideways over a distance of
2000m. Grid blocks 49 and 50 have reduced permeability and porosity.

Table 1 Initial conditions in different sections of the reservoir simulation model

Grid block Porosity Permeability Pressure

% mD bar
1-48 20 100 390
49 0.1 0.5 390
50 0.1 0.5 370
51-100 20 100 370

Four typical laboratory reports build the basis to fill the model with hydrocarbon fluids. Each of them is
representative of a type of reservoir fluid: black oil, volatile oil, gas condensate, and gas. The composition of all
example fluids is extended to C45+ and then grouped into pseudo components following the methodology of Al-
Meshari and McCain (2005). The mole fraction of the pseudo components is then slightly varied up or down to
create a heavier and a lighter version of the original fluid. Both models are then initialized with the heavy version
of the fluid in the bottom half of the model and the lighter fluid version in the top part.

Figure 6 through Figure 9 show the mole fractions of the pseudo components of the original four reservoir fluids
and their heavier and lighter versions.
6 SPE 145703

Figure 6 Composition of black oil (mole fractions). A heavier and a lighter version are created by slightly varying
the composition of the pseudo components while still staying within the classification of a black oil.

Figure 7 Composition of volatile oil (mole fractions). A heavier and a lighter version are created by slightly varying
the composition of the pseudo components while still staying within the classification of a volatile oil.

Figure 8 Composition of gas condensate (mole fractions). A heavier and a lighter version are created by slightly
varying the composition of the pseudo components while still staying within the classification of a gas condensate.

Figure 9 Composition of gas (mole fractions). A heavier and a lighter version are created by slightly varying the
composition of the pseudo components while still staying within the classification of a gas.
SPE 145703 7

The compositional variation creates not only a density difference that will evoke gravitational segregation, but also
a difference in chemical potential for every component in the mixture.

Both models (representing the two reservoir settings) are simulated for each of the four reservoir fluid types. The
time it takes for the pressure to reach a constant gradient is compared to the time it takes until the composition of
all components in all grid blocks remains constant over time. Figure 10 and Figure 11 compare the times to
equilibrate pressure and fluid composition for both reservoir settings.

Figure 10 Equilibration times comparison for four different reservoir fluids in thick, block-type reservoir. Pressure
equilibrates in a matter of days; fluid compositional equilibration takes more than seven orders of magnitude
longer.

Figure 11 Equilibration times comparison for four different reservoir fluids in laterally extended reservoir. The
longer distances result in overall longer equilibration times. The gap between pressure equilibration and fluid
compositional equilibration still remains more than seven orders of magnitude.

Various baffle properties result in different equilibration times. The following behavior of the reservoir simulation
software can be observed: a reduction in permeability influences the pressure equilibration time, independent
from a change in porosity. A change in porosity influences the time for fluid compositional equilibration,
independent from a change in permeability. Both properties were reduced at the baffle by the same factor to
influence both equilibration processes equally.
8 SPE 145703

Even when the baffle is completely eliminated, the gap between pressure and fluid compositional equilibration
remains. Figure 12 and Figure 13 show the comparison of equilibration times for the two models with uniform
porosity and permeability throughout the entire model.

Figure 12 Comparison of equilibration times in the absence of a baffle for a thick, block-type reservoir.
Equilibration occurs one order of magnitude faster compared to the case with a baffle. The difference between
pressure and fluid compositional equilibration remains more than seven orders of magnitude.

Figure 13 Comparison of equilibration times in the absence of a baffle for a laterally extended reservoir.
Equilibration occurs one order of magnitude faster compared to the case with a baffle. The difference between
pressure and fluid compositional equilibration remains more than seven orders of magnitude.
SPE 145703 9

The 200 fold reduction in pororsity and permeability at a single barrier is compared to a set of 5 barriers. Each of
these five barriers has a reduction in permeability and porosity of a factor of 40. They are spaced at equal
distance from each other throughout the reservoir model. Figure 14 visualizes the spacing and location of the
barriers in the laterally extended reservoir model. The barriers form six compartments in the reservoir model.

Figure 14 Model with multiple barriers. Five barriers are spaced at equal distance to each other. Each of these
five barriers has a reduction in permeability and porosity of a factor of 40.

The top three compartments are initialized at 20bar higher ressure as the bottom three compartments. They
contain initially the lighter version of the reservoir fluid. The bottom three compartments contain the heavy version
of the reservoir fluid. The pressure profile is ploted versus depth for various time steps in Figure 15. In this
example the reservoir fluid in the model was black oil. None of the six compartments are visible in the pressure
profile after only 3 month. The pressure profile remains constant after that for the rest of the simulated 150 Million
years.

Figure 15 Pressure vs Depth for various time steps. The reservoir fluid here is black oil. After only 3 month, none
of the six compartments are visible in the pressure profile.

The molfraction of the C36+ component group is plotted versus depth for comparison in Figure 16. The
compartments remain visible as stair step profile until the fluid column is compositionally equilibrated after 250
million years.
10 SPE 145703

Figure 16 C36+ mole fraction versus depth. The compartments remain visible in the distribution of the fluid
composition until the reservoir is compositionally equilibrated after 250 million years.

The influence of five barriers with a 40 fold reduction in permeability and porosity on the equilibration time is found
to be the same a one barrier with a 200 fold reduction. The overall equilibration times are found to be equal to the
ones shown in Figure 10 and Figure 11. The notable difference is the shape of the profile as the number of
stairsteps represents the number of compartments.

ANALYTICAL COMPARISON

The mass flow required for pressure equilibration depends on the compressibility of the system. Under the
-5
assumption of a total compressibility of 10 1/psi it would take only 1‰ of the reservoir fluid content to equilibrate
a 100psi pressure difference.

Muggeridge and Smalley (2008) present an equation to estimate the time scale of pressure equilibration:

where t is time; µ, fluid viscosity; Φ, matrix porosity; ce, effective compressibility; L, length over which equilibration
occurs; HB, the thickness of the baffle; Δp, the initial pressure difference across the baffle; δp, gauge resolution or
maximum allowed pressure difference when called equilibrated; and k B, baffle permeability. The use of SI units for
all variables returns the time in seconds.

Muggeridge and Smalley assume one side of the baffle as infinite acting. In our closed system this assumption
does not hold. Nevertheless, this equation offers an easy-to-use back-of-the-envelope comparison that is
independent from the work presented in this paper.

For the case of black oil in the laterally extended reservoir model with a 200-fold reduction of permeability at the
baffle, the equation estimates pressure equilibration within 0.3 years. The simulation estimates 2 years. Table 2
summarizes the values used in the calculation prior to SI unit conversion.
SPE 145703 11

Table 2 Input variables for analytical estimation of pressure equilibration

µ 0.522 cP
Φ 0.2 fraction
-5
ce 1×10 1/psi
L 2000 m
HB 4 m
Δp 20 bar
δp 0.01 bar
kB 0.5 mD

To estimate the time for fluid compositional equilibration, Einstein’s equations of mean square displacement of a
particle can be modified for the use in a porous medium. The diffusion length is multiplied by a tortuosity factor:

where t is time; D, the diffusion coefficient; L, the diffusion length; and τ, the tortuosity factor. The effect of
tortuosity on the actual diffusion length needs to be considered. Tortuosity accounts for the increased length
through which matter must diffuse in a porous medium as opposed to a straight line. England et al. (1987)
suggest a value of sqrt(3) for rock matrices. They also point out that, additionally, large-scale tortuosity effects
may be present. Dykhuizen and Casey (1989) also suggest a value of sqrt(3), but restrict its applicability to
homogeneous and isotropic rocks. In nature, where pores vary in cross section and coordination number,
tortuosity may well deviate from this value.

-4 2
For a diffusion length of 2000m and a diffusion coefficient for methane of 8.854 × 10 m /day, the above equation
returns a time of 18.5 million years. The model predicts 16 million years to equilibrate gas (80% methane) in the
laterally extended reservoir.
12 SPE 145703

CONCLUSIONS

A series of numerical simulations of one-dimensional reservoir models show that the time required for fluid
compositional equilibration exceeds the time required for pressure equilibration by seven or more orders of
magnitude. The results are in reasonable agreement with simple analytical calculations.

Baffles to the fluid flow in the reservoir become invisible in static pressure gradients after a very small fraction of
the age of the reservoir. Compartmentalization is substantially more likely to show in the distribution of fluid
composition than in a discontinuity in the pressure gradient. Connectivity is far more robustly analyzed by
determination of fluid compositional equilibrium than by static pressure communication.

Downhole fluid analysis evaluates reservoir fluid composition and assesses reservoir connectivity where static
pressure gradients fail to reveal highly valuable information.
SPE 145703 13

REFERENCES

Al-Meshari, A.A., and McCain, W.D. 2005. New Strategic Method to Tune Equation-of-State for Compositional
Simulation. Paper SPE-106332 presented at the SPE Technical Symposium of Saudi Arabia Section, Dhahran,
Saudi Arabia, 14–16 May. DOI: 10.2118/106332-MS.

Canas, J.A., Pop, J.J., Dubost, F.X. et al. 2008. Advanced Pressure and Fluid Gradient Analysis of a Deep
Offshore Reservoir. Paper SPE 115429 presented at the SPE Annual Technical Conference and Exhibition,
Denver, Colorado, 21–24 September. DOI: 10.2118/115429-MS.

Curtiss, C.F. and Bird, R.B. 1999. Multicomponent Diffusion. Industrial & Engineering Chemistry Research 38 (7):
2515-2522.

Darken, L.S., 1948. Diffusion, Mobility and Their Interrelation through Free Energy in Binary Metallic Systems.
Transactions of The American Institute of Mining And Metallurgical Engineers 175: 184-201.

Dubost, F.X., Carnegie, A.J.G., Mullins, O.C. et al. 2007. Integration of In-Situ Fluid Measurements for Pressure
Gradients Calculations. Paper SPE 108494 presented at the International Oil Conference and Exhibition,
Veracruz, Mexico 27-30 June. DOI: 10.2118/108494-MS.

Dykhuizen, R.C. and Casey, W.H. 1989. An Analysis of Solute Diffusion in Rocks. Geochimica et Cosmochimica
Acta 53 (11): 2797-2805.

Einstein, A., 1905. On the movement of small particles suspended in a stationary liquid demanded by the
molecular-kinetic theory of heat. Annalen der Physik 17: 549- 560.

Elshahawi, H., Hashem, M.N., Mullins, O.C. et al. 2005. The Missing Link—Identification of Reservoir
Compartmentalization through Downhole Fluid Analysis. Paper SPE 94709 presented at the SPE Annual
Technical Conference and Exhibition, Dallas, Texas 9-12 October. DOI: 10.2118/94709-MS.

Elshahawi, H., Venkataramanan, L., McKinney, D. et al. 2006. Combining Continuous Fluid Typing, Wireline
Formation Testers, and Geochemical Measurements for an Improved Understanding of Reservoir Architecture.
Paper SPE 100740 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, 24–
27 September. DOI: 10.2118/100740-PA.

England, W.A., Mackenzie, A.S., Mann, D.M. et al. 1987. The Movement and Entrapment of Petroleum Fluids in
the Subsurface. Journal of the Geological Society 144 (2): 327-347.

England, W.A., Muggeridge, A.H., Clifford, P.J. et al. 1995. Modelling Density-Driven Mixing Rates in Petroleum
Reservoirs on Geological Time-Scales, with Application to the Detection of Barriers in the Forties Field (Ukcs).
Geological Society, London, Special Publications 86 (1): 185-201.

Freed, D.E., 2007. Dependence on Chain Length of NMR Relaxation Times in Mixtures of Alkanes. The Journal
of Chemical Physics 126 (17): 174502.

Freed, D.E., 2009. Temperature and Pressure Dependence of the Diffusion Coefficients and NMR Relaxation
Times of Mixtures of Alkanes. The Journal of Physical Chemistry B 113 (13): 4293-4302.

Grathwohl, P., 1998. Diffusion in Natural Porous Media: Contaminant Transport, Sorption/Desorption and
Dissolution Kinetics. Topics in Environmental Fluid Mechanics. Boston: Kluwer Academic Publishers.

Grogan, A.T., Pinczewski, V.W., Ruskauff, G.J. et al. 1988. Diffusion of CO2 at Reservoir Conditions: Models and
14 SPE 145703

Measurements. SPE Reservoir Engineering 3 (1): 93-102.

Hashem, M., Elshahawi, H., and Ugueto, G. 2004. A Decade of Formation Testing—Do's and Don'ts and Tricks of
the Trade. Paper presented at the SPWLA 45th Annual Logging Symposium, Noordwijk, The Netherlands, June
6-9.

Jamialahmadi, M., Emadi, M., and Müller-Steinhagen, H. 2006. Diffusion Coefficients of Methane in Liquid
Hydrocarbons at High Pressure and Temperature. Journal of Petroleum Science and Engineering 53 (1-2): 47-60.

Jackson, R.R., Carnegie, A., and Dubost, F.X. 2007. Pressure Measurement and Pressure Gradient Analysis:
How Reliable for Determining Fluid Density and Compositional Gradients? Paper SPE 111911 presented at the
Nigeria Annual International Conference and Exhibition, Abuja, Nigeria, 6–8 August. DOI: 10.2118/111911-MS.

Kabir, C.S. and Pop, J.J., 2007. How Reliable Is Fluid Gradient in Gas/Condensate Reservoirs? SPE Reservoir
Evaluation & Engineering 10 (6): 644-656.

Liu, H., and Macedo, E.A. 1995. Accurate Correlations for the Self-Diffusion Coefficients of CO2, CH4, C2H4,
H2O, and D2O over Wide Ranges of Temperature and Pressure. The Journal of Supercritical Fluids 8 (4): 310-
317.

Montel, F., Bickert, J., Hy-Billiot, J. et al. 2003. Pressure and Compositional Gradients in Reservoirs. Paper SPE
85668 presented at the Nigeria Annual International Conference and Exhibition, Abuja, Nigeria, 4–6 August. DOI:
10.2118/85668-MS.

Muggeridge, A.H. and Smalley, P.C. 2008. A Diagnostic Toolkit to Detect Compartmentalization Using Time-
Scales for Reservoir Mixing. Paper SPE 118323 presented at the Abu Dhabi International Petroleum Exhibition
and Conference, Abu Dhabi, U.A.E., 3–6 November. DOI: 10.2118/118323-MS.

Mullins, O.C., 2008. The Physics of Reservoir Fluids; Discovery through Downhole Fluid Analysis. Houston,
Texas: Schlumberger.

Mullins, O.C., Zuo, J.Y., Freed, D.E., et al. 2010. Downhole Fluid Analysis Coupled with Novel Asphaltene
Science for Reservoir Evaluation. Paper RRR presented at the SPWLA 51th Annual Logging Symposium, Perth,
Australia, June 19–23.

O'Keefe, M.D., Godefroy, S., Vasques, R.R. et al. 2007. In-Situ Density and Viscosity of Reservoir Fluids
Measured by Wireline Formation Testers. Paper SPE 110364 presented at the Asia Pacific Oil and Gas
Conference and Exhibition, Jakarta, Indonesia, 30 October–1 November. DOI: 10.2118/110364-MS.

Pfeiffer, T., Reza, Z., McCain, W.D., Schechter, D. S., Mullins, O.C. Determination of Fluid Composition
Equilibrium – a Substantially Superior Way to Assess Reservoir Connectivity than Formation Pressure Surveys.
Paper presented at the SPWLA 52th Annual Logging Symposium, Colorado Springs, CO, USA, May 14-18.

Riazi, M.R., 1996. A New Method for Experimental Measurement of Diffusion Coefficients in Reservoir Fluids.
Journal of Petroleum Science and Engineering 14 (3-4): 235-250.

Riazi, M.R. and Whitson, C.H. 1993. Estimating Diffusion Coefficients of Dense Fluids. Industrial & Engineering
Chemistry Research 32 (12): 3081-3088.

Stainforth, J.G., 2004. New Insights into Reservoir Filling and Mixing Processes. In Understanding Petroleum
Reservoirs: Towards an Integrated Reservoir Engineering and Geochemical Approach, ed. J.M. Cubitt, W.A.
England, and S.R. Larter. SP 235: 115-132. London: Geological Society.
SPE 145703 15

Taylor, R. and Krishna, R. 1993. Multicomponent Mass Transfer. Wiley Series in Chemical Engineering. New
York: Wiley.

Venkataramanan, L., Gustavson, G., and Weinheber, P. et al. 2006. Pressure Gradients and Fluid Analysis as an
Aid to Determining Reservoir Compartmentalization. Paper presented at the SPWLA 47th Annual Logging
Symposium, Veracruz, Mexico, June 4–7.

Xian, C., Zuo, J.Y., Haddad, S.S. et al. 2010. Reservoir Architecture Characterization: EOS Based Downhole
Fluid Analysis, Openhole Logs and Pressures Integration. Paper SPE 130900 presented at the International Oil
and Gas Conference and Exhibition in Beijing, China, 8–10 June DOI: 10.2118/130900-MS.

Zuo, J.Y., Zhang, D., Dubost, F. et al. 2008. EOS-Based Downhole Fluid Characterization. Paper SPE 114702
presented at the 2008 Asia Pacific Oil & Gas Conference and Exhibition Held in Perth, Australia, October 20–22.