SPE-171940-MS

Microporosity: Characterization, Distribution, and Influence on Oil

Recovery
Shawn M Fullmer, Sean A Guidry, Jonas Gournay, and Emily Bowlin, Exxonmobil; Gary Ottinger and
Abdulla Al Neyadi, ZADCO; Gaurav Gupta and Bo Gao, Exxonmobil; Ewart Edwards, ZADCO

Copyright 2014, Society of Petroleum Engineers

This paper was prepared for presentation at the Abu Dhabi International Petroleum Exhibition and Conference held in Abu Dhabi, UAE, 10 –13 November 2014.

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
Microporosity is very common in limestone reservoirs globally and is especially significant in many large
Mesozoic reservoirs in the Middle East. Despite its common occurrence there is:
1. Wide variation in its definition,
2. Uncertainty around characterization, genetic controls, and distribution
3. A rudimentary understanding of its influence on reservoir performance and hydrocarbon recovery.
The results of this study, based on a global survey of microporosity and specific Middle Eastern case
studies, provide clarity on each of these topics.
One volumetrically significant type of microporosity occurs between micron size subhedral crystals of
low magnesium calcite in matrix and within grains. This micro-pore system is very homogenous in terms
of pore size distribution with 90% of pores between 1 and 3 microns in diameter. Pore throat radii range
between 0.1 and 1.5 microns. Porosity, permeability, and capillarity relationships reflect this homogeneity
for rocks dominated by microporosity. Rocks with less than approximately 80% microporosity exhibit a
marked increase in pore system heterogeneity.
A pore geometry characterization approach incorporating digital image analyses of petrographic
thin-sections was developed and provides a very effective means of rapidly characterizing and quantifying
the total pore system, including microporosity.
The lateral and stratigraphic distribution of microporosity is systematically related to the distribution
of depositional facies and the regional extent of burial diagenetic processes. Factors that inhibit burial
diagenesis, such as hydrocarbon charge, also have a strong influence on the nature and distribution of
microporosity.
Remaining oil saturation in microporous limestone, as measured from centrifuge capillary pressure and
steady state (SS) core flood experiments, is negatively correlated with the percent fraction of micropo-
rosity. Due to the homogenous nature of the micro-pore system, rocks dominated by microporosity have
more favorable oil recovery than rocks with mixed pore systems. In the specific cases studied here, water
provides more favorable recovery than gas. These results have implications for resource assessment, field
2 SPE-171940-MS

development planning and optimization of ultimate recovery in limestone reservoirs with significant
microporosity.

Introduction
The heterogeneity typical of carbonate pore systems poses a significant challenge for the characterization
and development of carbonate reservoirs. Micro-pores are a common feature within carbonate pore
systems and numerous studies have investigated the nature and origin of carbonate microporosity.
However, no consensus exists on how and where micro-pores form, the most effective ways to charac-
terize the micro-pore system, controls on its distribution, or what influence microporosity has on oil
recovery. Wide variation in the use of the term microporosity further complicates issues involving
micro-pore space. The results of this study, based on a global survey of microporosity and specific Middle
Eastern case studies, provide clarity on each of these topics. This work is focused on microporosity and
micro-pores hosted in limestone but also includes an evaluation of the influence of microporosity in mixed
pore systems. Pore space associated with dolomitization or structural deformation is outside the scope of
this paper.
This work is focused on four primary objectives:
1. Synthesize global observations regarding the nature of limestone microporosity and provide a
universal genetic classification of the same.
2. Develop a pore geometry characterization approach incorporating digital image analyses of
petrographic thin-sections to rapidly and systematically characterize and quantify the total pore
system, including microporosity.
3. Apply lessons learned about the origins and evolution of microporosity to better understand factors
controlling its distribution.
4. Investigate the influence of microporosity on oil recovery for rocks containing a range of
microporosity abundance.
The intent of this paper is to provide a summary of results for each of these objectives rather than a
thorough review of each. The integration of results to improve understanding of reservoir performance is
emphasized. The data and results section of the paper is organized in the order in which the objectives are
presented above.
Observations regarding the nature of the micro-pore system were synthesized from a global study of
limestone microporosity carried out over the last several years at Exxonmobil URC. The data sets included
in this global study span geologic age, depositional and diagenetic setting, burial history, geographic
location, and burial depth. Data and analyses were synthesized from over twenty different oil and gas
reservoirs as well as outcrop and research core sample sets. Data constraining the distribution of
microporosity and its influence on oil recovery in this paper are from a case study evaluating these topics
in a large Cretaceous offshore oil reservoir located in the Southern Arabian Gulf.
The findings from this study have significant implications for reservoir characterization and field
development activities as well as important insights into asset valuation and maximizing ultimate oil
recovery in limestone reservoirs with significant microporosity. Integration and synthesis across the
geology and engineering disciplines, as demonstrated here, facilitate new understanding of field perfor-
mance and illuminate new opportunities for optimizing value.

Methods

Industry standard analytical techniques and characterization methods used in this study include: thin
section and scanning electron microscope (SEM) petrography, conventional core analysis (CCAL), and
mercury injection capillary pressure analysis (MICP). For the sake of brevity a detailed description of
SPE-171940-MS 3

Figure 1—Thin section and SEM photomicrographs depicting the nature of limestone micro-pores in both grain-dominated rocks (AC) and
mud-dominated rocks (D-F).

these standard methods and techniques is not included here. A new approach to pore system character-
ization is presented in the paper and as such a brief description of this methodology is provided here. Thin
sections are scanned at a resolution of 0.64 microns/pixel and image segmentation is performed on the
entire thin section scan to isolate pore space (sensu Anselmetti et al. 1998). Image analysis on ion milled
bulk rock and pore cast SEM images captured at 5000 times magnification is performed to quantify pore
metrics for pore space below the resolution of a standard thin section. Polygons generated around
individual pores are evaluated for size, shape, and other metrics which are used to classify pore type and
generate size distributions. Pore type distributions are plotted in a 4 axis tetrahedron with interparticle
pores, separate vugs (Svugs), touching vugs (Tvugs) (sensu Lucia, 1995), and micro-pores at each apex
of the tetrahedron. The percent fraction microporosity is calculated by subtracting TPS calculated
macroporosity from CCAL derived plug porosity.
Results from special core analysis (SCAL) experiments are synthesized in this study. Remaining oil
saturations are derived from primary imbibition measurements via water-oil centrifuge capillary pressure
(Pc) experiments and water-oil and gas-oil steady state core flood experiments. Saturation values reported
are minimum oil saturations at maximum negative Pc for the centrifuge derived data and remaining oil
saturation after water flood (Sor(wf)) or gas flood (Sor(gf)) normalized to an equivalent fractional flow
of water or gas for the steady state core flood experiments. All steady state core flood experiments were
performed at reservoir conditions with live reservoir fluids on preserved core samples. Centrifuge
capillary pressure experiments were conducted at elevated temperature with dead oil.

Data and Results

Nature of Limestone Microporosity
The most volumetrically significant type of microporosity in limestone rocks is pore space between
micron-sized crystals of low magnesium calcite (micrite) that occur in matrix material and within grains
(Figure 1).
4 SPE-171940-MS

Figure 2—Pore size distribution for limestone micro-pores.

Figure 3—Crystal textures, pore throat size, and porosity-permeability relationships for Type I, Type II, and Type III micro-pores.

The micro-pore system is very homogenous in terms of pore size distribution with 90% of pores
between 1 and 3 microns in diameter (Figure 2). Three sub-types of micro-pores are differentiated by the
micro-texture of the micrite the pore space is associated with and/or a distinct clustering in pore throat size
distributions (Figure 3). Porosity – permeability relationships correlate well with micro-texture and pore
throat size groupings.
Type 1 micro-pores are associated with a granular subhedral micrite texture characterized by a loose
framework of subhedral crystals that are randomly oriented. Crystal boundaries are characterized by
punctic (point) and serrate (flat) contacts. Type I micro-pores have a pore throat radii distribution that
ranges between 0.1 and 1.5 microns and average pore throat radius of 0.7 microns. These larger pore throat
SPE-171940-MS 5

sizes result in higher porosity and permeability values relative to Type II and Type III micro-pores. Type
II micro-pores are associated with either a granular euhedral or clustered micrite micro-texture. Clustered
textures are characterized by a loose framework of agglutinated crystals of random orientation. Individual
crystals are distinguishable but crystals are overgrown and are commonly vertically stacked. Type II
micro-pores display tall and narrow pore throat size distributions centered at approximately 0.2 microns
and have intermediate values of porosity and permeability relative to the other types. Type III micro-pores
are associated with fitted micrite micro-textures with crystal morphologies characterized by a dense
mosaic of micro-crystals that are partially to fully cemented together. Crystal faces are scalloped and
interfacial boundaries typically form 120 degree triple junctions. Type III micro-pores display lower and
broader MICP profiles centered at 0.06 microns and have correspondingly lower relative values of
porosity and permeability.
Figure 3 provides an illustration of the relationship between micrite micro-texture, pore throat size, and
porosity – permeability for rocks dominated by microporosity. This relationship is largely a function of
the narrow pore and pore throat size distribution across micro-pore type and the inter-crystalline
connectivity of the micro-pore system. The micrite microtextures observed represent the diagenetic
evolution, to be described in greater detail below, of microporous limestone in the burial environment.
Because microporosity is ambiguously defined within the industry it is important to select a definition that
is consistent with rock based observations and has utility for understanding and predicting microporosity
distribution and explaining reservoir properties and performance. The classification and characterization
of micro-pores described here, first presented in Kaczmarek et al, 2011, provides a genetic and universal
definition of limestone microporosity. This definition is quantitative and tied to rock properties making
it a useful tool for integrating pore system and rock property data as well as understanding the link
between the pore system and reservoir performance.

Total Pore System Characterization

The total pore system (TPS) characterization approach is designed around the concept that pore size and
pore type distribution are key variables controlling rock performance (Lucia, 2007). In order to link the
pore system and performance data these key variables must be quantified in a systematic and repeatable
way. For the characterization system to be both cost effective and broadly applicable it must be based on
evaluation of standard data types, such as thin sections, rather than costly and potentially inaccessible data
types. The TPS characterization approach was developed with these requirements in mind. TPS consists
of analyzing pore geometry and pore type distributions from image and pore metric analyses on high
resolution thin section and/or SEM images (see methods for further discussion). Pore space is isolated via
image segmentation and pore morphometric analyses are performed on the resultant pore polygons (figure
4). Similar image based techniques are described by Anselmetti et al. 1998. The end product of the TPS
workflow is a distribution of pore size comparable to pore throat size distributions from MICP or pore size
distributions from NMR and the percent fraction of each pore type (interparticle pores (IP), separate vugs
(Svugs) sensu Lucia 1995, touching vugs (Tvugs) sensu Lucia 1995, and micro-pores as defined above).
The pore type distributions are plotted in a 4 axis tetrahedron with the end member pore types listed above
at each apex of the tetrahedron (figure 4). In cases where no Tvugs are present a ternary plot is used
instead. These results are validated against MICP, nuclear magnetic resonance (NMR), Focused Ion beam
SEM (FIBSEM) and micro computed tomography (Micro-CT) analyses on a representative subset of
samples to ensure fidelity of statistics across 2dimensional and 3-dimensional data types. The TPS
approach provides a quantitative basis for investigating the influence of varying proportions of micro-
porosity on petrophysical properties and integrating pore geometry characterization with SCAL and
dynamic data.
In the previous section a robust relationship between porosity and permeability was illustrated for rocks
that are dominated by microporosity. The driver behind the relationship is thought to be associated with
6 SPE-171940-MS

Figure 4 —Pore space tetrahedron that underpins the TPS characterization approach and an illustration of the image analysis workflow used to
analyze pore geometry.

the homogenous size distribution characteristic of the micro-pore system. In contrast, carbonate rocks
commonly exhibit very complex porosity – permeability relationships due to complex and heterogeneous
pore systems. A critical question that follows is “How much microporosity is required for a rock to exhibit
the micro-pore dominated behavior shown in figure 3?” To illustrate the utility of the TPS approach and
address this important question the following example comparing variably cemented grainstones with
intragrain microporosity is presented.
Figure 5 compares three different grainstone rock types (RRT’s), RRT 37, RRT 27, and RRT 17. The
photomicrographs shown are representative of the depositional texture, grain size and sorting, pore types,
and diagenetic modification typical of rocks belonging to the respective RRT designated. RRT 37 consists
of grainstones with intragrain microporosity and minor pore occluding calcite cementation (highest
relative reservoir quality (RQ)). RRT 27 consists of grainstones with intragrain microporosity and
common pore occluding calcite cementation (intermediate relative RQ). RRT 17 consists of grainstones
with intragrain microporosity and pervasive pore occluding calcite cementation (lowest RQ). TPS
characterization was performed on approximately 50 samples with roughly equal representation from each
RRT. Figure 6 shows the resulting pore type distribution color-coded by RRT. The samples consist mostly
of interparticle pores and micro-pores and the majority of the samples (with the exception of one) contain
greater than 50% microporosity. The RRTs are distributed regularly along the IP%-Micro% axis from
better quality (RRT 37) with more interparticle porosity to the poorest quality (RRT 17) clustering at over
90% microporosity.
A positive correlation can be observed between porosity and permeability, again roughly distributed by
RRT but with some scatter especially apparent for RRT 27 (figure 7). Two poro-perm trends are evident,
one with relatively higher permeability for a given porosity and the other with relatively lower permea-
bility for a given porosity. RRT 37 and RRT 17 are easily differentiated by both porosity and permeability
SPE-171940-MS 7

and lie along distinctly different trends. In contrast,

RRT 27 appears to span both trends with a range of
permeability corresponding to a given porosity.
Based on TPS characterization the pore system for
RRT 17 is predominantly microporous (figure 6).
When the porosity and permeability data for micro-
pore dominated samples (from figure 3) are plotted
together with the grainstone data it is clear that all of
RRT 17 and some of RRT 27 can be classified as
micro-pore dominated rocks (figure 8). Grainstones
designated as RRT 27 have undergone variable cal-
cite cementation which has resulted in the occlusion
of some interparticle space. It is apparent that some
RRT 27’s, those that plot along the micro-pore
dominated trend, have undergone sufficient cemen-
tation such that the intragrain micro-pore system is
the effective pore system and that the macro-pore
space is not contributing to permeability in a sig-
nificant way. It is also apparent that it is difficult to
separate micro-pore dominated RRT27’s from those
that plot along trend with the better quality RRT
37’s by the method that was employed to classify
these rocks into RRT’s originally.
Plotting the percent fraction of microporosity
against plug permeability (Figure 9) shows a nega-
tive correlation between the two variables. This
relationship is consistent with expectations due to
the interdependence of permeability and interparti-
cle porosity. In this example higher percent fraction
microporosity corresponds with increased cementa-
tion of interparticle pore space and a concomitant
decrease in permeability until some threshold where
presumably the micro-pore system is the only re-
maining connected pore volume available to con- Figure 5—Grainstone rock types with intragrain microporosity and
tribute to permeability. Comparing the data points variable proportions of macro-pore occluding calcite cement. See text
for discussion.
from RRT27 that fall along the micro-pore domi-
nated trend in figure 8 (blue oval in figure 10) with
the position of data points from the same samples along the % Microporosity axis in figure 9 (blue box
in figure 10) constrains the position of this threshold at approximately 80% microporosity (fig 10). The
samples with less than 80% microporosity plot along trend with the higher permeability RRT 37’s with
characteristically heterogeneous mixed pore systems.
A comparison of pore size distribution, facilitated by TPS characterization, with pore throat size
distribution further substantiates the validity of the 80% threshold and provides additional insight into the
sensitivity of permeability to cementation of interparticle pore space. The pore size distribution for an
RRT 27 with 83% microporosity is more similar to the pore size distribution of an RRT 37 with only 73%
microporosity than an RRT 17 with 98% microporosity (figure 11A). In contrast the pore throat size
distribution of the same RRT 27 is more similar to the pore throat size distribution of the RRT 17 than
that of the RRT 37 (figure 11B). The implication of this observation is that while significant interparticle
8 SPE-171940-MS

Figure 6 —Pore type distribution generated by TPS approach for variably cemented grainstone rock types with intragrain microporosity.

Figure 7—Porosity – permeability relationships for variably cemented grainstone rock types with intragrain microporosity.
SPE-171940-MS 9

Figure 8 —Comparison of porosity – permeability relationships for variably cemented grainstone rock types with intragrain microporosity and rocks
dominated by microporosity (data set from Figure 3).

Figure 9 —Percent fraction microporosity calculated using TPS approach plotted against permeability for variably cemented grainstone rock types
with intragrain microporosity.

space remains uncemented in the RRT 27, illustrated by pore sizes more similar to the RRT 37, sufficient
closure of the pore throats connecting the interparticle space has occurred such that the connected pore
throat size is more similar to the exclusively microporous RRT17. In other words, the extent of
cementation that occludes just enough interparticle pore space to push the percent fraction of micropo-
rosity above 80% (83% in this case) is not enough to significantly diminish the macropore size distribution
but is sufficient to close off the pore throats connecting the macro-pores.
10 SPE-171940-MS

Figure 10 —Comparison of Porosity-Permeability and percent fraction microporosity-permeability. Blue oval in poro-perm plot and blue box in
percent fraction microporosity plot identify the position of data points from the same sample subset in each respective plot. A threshold of 80%
separates RRT 27’s that plot along trend with micro-pore dominated samples and those that plot on trend with the more heterogeneous RRT37’s.

Controls on Micro-pore Distribution

In order to gain understanding about the controls on the distribution of microporosity it is necessary to
revisit the origin of the actual pore volume that microporosity is associated with and the diagenetic
processes that act to modify the pore space that the pore volume is arranged within. Commonly, limestone
micro-pores are a product of calcite burial stabilization of preexisting carbonate material (aragonite,
high-Mg calcite, or low Mg calcite) that was deposited as grains or matrix and had significant initial
porosity (Loucks and Lucia, 2013). Some grain types are more susceptible to becoming microporous than
others. Susceptibility appears to be mainly a function of original porosity and how the original porosity
is organized. This variation in susceptibility manifests itself as a major control on the distribution of
microporosity at the field to regional scale. The precursor constituents, extrapolated from modern
SPE-171940-MS 11

Figure 11—Pore size distribution from TPS for three variably cemented grainstone samples that span the 80% microporosity threshold (A). Pore
throat size distribution for the same three samples (B).

equivalents, of a microporous limestone were deposited with an initial pore structure at the scale of a few
nanometers. If the depositied constituents are not dissolved or extensively altered by near surface
diagenesis then stabilization or dissolution-reprecipitation in the shallow burial environment transforms
the pore system from a highly porous network arranged at the nanometer scale to a less porous, though
significant porosity can be preserved, network arranged at the micrometer scale (Loucks and Lucia, 2013).
The granular micrite fabrics discussed in section 1 are interpreted to have developed in the manner
described. Continued burial diagenesis drives the evolution of the system from the granular microtexture
to a fitted microtexture as compaction-related, low-Mg calcite cements overgrow the micrite grains and
occlude micro-pore space. The granular micro-texture is preserved at depth where diagenesis is incom-
plete due to cement inhibition by hydrocarbon charge (Kaczmarek, et al, 2011).
Questions concerning the distribution of microporosity can be separated into two general categories:
1) controls on the distribution of the occurrence and relative abundance of microporosity encountered
vertically or laterally and 2) controls on the distribution of the type of microporosity encountered
12 SPE-171940-MS

Figure 12—Simple comparison of the depositional influence on percent fraction microporosity in a given rock. Rudist floatstone in A has lower
percent fraction microporosity relative to B because of the presence on non-microporous rudist grains and higher proportion of inter-granular pore
space.

vertically or laterally. The factors controlling the distribution of occurrence and relative abundance of
microporosity are closely related to the depositional origins and diagenetic evolution of microporosity.
Microporosity occurs in limestone materials that contain significant depositional porosity that is arranged
in a micro- to nano-pore structure. These are the carbonate materials that are systematically predisposed
to preserve microporosity at depth. The relative proportion of material that is predisposed to preserve
microporosity that is deposited in a given location will control the relative abundance of microporosity in
that location. In other words, the relative proportion of microporosity in a given location or depositional
facies is a function of the ratio of depositional components that are micro-pore prone to those that are not.
The standard suite of controls on sedimentation in a carbonate depositional environment, such as
hydrodynamic energy, light penetration, salinity, etc., will dictate that ratio. An example illustrating this
concept is provided in figure 12. The photomicrograph at the top of figure 12 is a rudist floatstone with
abundant primary interparticle pore space. The photomicrograph at the bottom of figure 12 is a peloid
SPE-171940-MS 13

Figure 13—Micrite textural evolution with successive burial and longer residence time in the water-leg.

packstone with large benthic foraminifera. The rudist fragment in the upper image is not micro-pore prone
while most of the other material, grains in this example, is microporous. The intergranular space remains
open, does not contain mud, and is therefore not considered microporous either. It follows then that the
relative contribution to total porosity from microporosity in this facies is limited to microporosity in the
peloids, algal grains, and forams. Contrast that with the facies represented in the lower photomicrograph.
In this facies all of the grains and the mud in between the grains is microporous resulting in a higher
abundance of microporosity relative to the rudist floatstone. In this way the distribution of the relative
abundance of microporosity is a function of the distribution of depositional facies. The rules that are
employed to explain and predict the lateral and vertical distribution of depositional facies can also be
employed to predict the lateral and vertical distribution of microporosity.
Diagenetic controls related to burial processes also play a key role in the distribution of the relative
abundance of microporosity and have a profound influence on the type of microporosity encountered. As
a unit of limestone is buried, the evolution of crystal microtexture and the micro-pore space associated
with them will progress from granular microtexture with significant microporosity to fitted microtexture
with little microporosity at a regional scale unless there are significant variations in local subsidence. It
follows then that the extent to which burial diagenesis has progressed for a given sample will exert a
control on the abundance and type of microporosity remaining in that sample. Factors that inhibit
diagenesis, such as hydrocarbon charge, prevent continued progression of microtextural evolution and in
effect lock-in the micro-pore type at the time of hydrocarbon charge. This imposes a second diagenetic
control on the distribution of the type of microporosity as rocks with longer residence time in the water
leg will have more diagenetically mature types of microporosity (Type III). Figure 13 illustrates this
concept. The upper SEM image shows an example of Type I microporosity and is from a sample located
in a crestal position of an oil reservoir. The assumption is that this sample, relative to the flanks of the
field, has been in the oil leg the longest. The SEM image in the middle shows an example of a micro-pore
system that is transitory between Type I and Type III. This sample is from a position located within the
flank of the reservoir in the transition zone. In this case, longer residence time in the water-leg, relative
to the first sample, has resulted in increased cementation which is reflected in the partial fitting of slightly
larger micrite crystals. The lower SEM image shows a very mature fitted microtexture and is from a
sample located in the aquifer below the reservoir. It is likely that this sample never experienced
hydrocarbon charge and as a result underwent complete burial stabilization and cementation.
Influence of Microporosity on Oil Recovery
The final question posed in this study concerns the influence of microporosity on oil recovery. TPS
characterization results were integrated with remaining oil saturation values obtained from special core
14 SPE-171940-MS

Figure 14 —Plot of minimum oil saturation vs. permeability (A) and minimum oil saturation vs. percent fraction microporosity (B).

analysis experiments including water-oil (W-O) centrifuge capillary pressure and both water-oil and
gas-oil steady state core flood experiments. Steady state core flood tests were performed on preserved core
samples at reservoir conditions with reservoir fluids. Centrifuge capillary pressure experiments were
conducted at elevated temperature with dead oil. Samples included in the analysis range from mixed pore
system rock types with as little as 50% microporosity to micro-pore dominated rock types with greater
than 95% microporosity. The results reported here are limited to rocks with Type I (granular Subhedral)
micro-pores. The influence of other micro-pore types on oil recovery will be the topic of future work.
Twenty six companion plugs were selected from the W-O centrifuge Pc data set for detailed TPS
characterization and integration work. Minimum oil saturation measured at maximum negative Pc was
used as a representation of displacement efficiency for the centrifuge Pc data. This end point value was
plotted against permeability and percent fraction microporosity for the 26 samples characterized. Mini-
mum oil saturation is positively correlated with permeability and negatively correlated with percent
fraction microporosity (figure 14). In other words, the most permeable rocks in the study have the highest
minimum oil saturations and least favorable displacement while the lowest permeability rocks have the
lowest minimum oil saturations and most favorable displacement. Likewise, rocks with the least percent
fraction microporosity had the least favorable displacement while rocks with the greatest percent fraction
microporosity had the most favorable displacement.
The results from seven steady state core flood experiments were used to investigate the relationship
between microporosity and remaining oil saturation after both waterflood and gasflood. TPS character-
ization was carried out on multiple companion plugs to the native state plugs used to assemble the steady
state composites. Remaining oil saturation after waterflood is referred to as Sor(wf) and remaining oil
saturation after gasflood is Sor(gf). Sor(wf) and Sor(gf) values are reported at a fractional flow of water
or gas respectively that approaches 0.99 to provide a consistent basis for comparison. Sor(wf) and Sor(gf)
are plotted against percent fraction microporosity and compared in figure 15. Sor(wf) for rocks with less
than 80% microporosity is approximately 45% and decreases systematically as percent fraction micro-
porosity increases from 80% to near 100%.Rocks with the highest percent fraction microporosity have the
most favorable displacement efficiency with Sor(wf) of 29%. Sor(gf) is higher in general compared to the
waterflood results. Composites with less than 80% microporosity had Sor(gf) hovering around 60%.
Sor(gf) also decreases systematically with increasing percent fraction microporosity above the 80%
threshold to a low of 48% Sor(gf) at 98% microporosity.
Results from both the centrifuge Pc experiments and steady state core flood experiments indicate that
rock types with higher percent fraction microporosity have more favorable recovery than those with more
SPE-171940-MS 15

Figure 15—Plot of Sor(wf) and Sor(gf) vs. percent fraction microporosity. Remaining oil saturation values are reported at a fractional flow of water
or gas respectively that approaches 0.99 to provide a consistent basis for comparison.

equable proportions of microporosity to macroporosity. The 80% threshold observed to differentiate

between rocks wherein the pore and pore throat size distribution of the micro-pore system controls
permeability and those that have a mixed pore system with contribution to permeability from connected
macro-pores is persistent for differentiating more favorable and less favorable recovery in these same
rocks. The primary distinguishing characteristic separating these two rock types from a flow/recovery
standpoint is the incredibly homogenous pore system in micro-pore dominated rocks versus the hetero-
geneous pore system in rocks with a mix of pore types. Figure 16 illustrates this concept by comparing
a grainstone that has connected inter-granular macroporosity as well as intra-granular microporosity with
a mud-dominated packstone that has a pore system consisting almost exclusively of micropores. While the
SEM images show the consistent nature of microporosity between the two samples, the photomicrographs
and pore throat size distributions show significant heterogeneity in the grainstone example and the
homogenous nature of the mud-dominated packstone example. We hypothesize that it is this difference
in the relative heterogeneity of the pore system that is driving the differences in remaining oil saturation
observed in the SCAL results. The heterogeneous pore system in rocks with less than 80% microporosity
has less favorable displacement due to significant pore scale permeability contrasts and more tortuous
flow paths. Rocks with greater than 80% microporosity have more favorable displacement due to simple
flow paths and very consistent pore scale permeability.
16 SPE-171940-MS

Figure 16 —Comparison of a grainstone with a heterogeneous mixed pore system (A) and a mud-dominated packstone with a homogenous micro-pore
system (B) to illustrate the sensitivity of oil recovery to pore system heterogeneity.

Conclusions
The most volumetrically significant type of microporosity in limestone occurs between micron size
crystals of low magnesium calcite in matrix and within grains. This micro-pore system is very homog-
enous in terms of pore size distribution with 90% of pores between 1 and 3 microns in diameter. Pore
throat radii range between 0.1 and 1.5 microns. Pore throat size is related to micrite micro-textures.
Together these define three distinct micro-pore type classes. Porosity, permeability, and capillarity
relationships reflect the homogeneous nature of rocks dominated by microporosity and correlate well with
micropore type. Rocks with less than approximately 80% microporosity exhibit a marked increase in pore
system heterogeneity. The classification and characterization of micro-pores described in this study
provides a genetic and universal definition for limestone microporosity. This definition is quantitative and
tied to rock properties making it a useful tool for integrating pore system and rock property data as well
as understanding the link between the pore system and reservoir performance.
A pore geometry characterization approach incorporating digital image analyses of petrographic
thin-sections was developed and provides a very effective means of rapidly characterizing and quantifying
the total pore system, including microporosity.
The TPS approach provides a quantitative basis for investigating the influence of varying proportions
of microporosity on petrophysical properties and integrating pore geometry characterization with SCAL
and dynamic data.
The lateral and stratigraphic distribution of microporosity is systematically related to the distribution
of depositional facies and the regional extent of burial diagenetic processes. Factors that inhibit burial
diagenesis, such as hydrocarbon charge, also have a strong influence on the nature and distribution of
microporosity.
Paramount among the findings from this study is that pore system heterogeneity is a primary control
in the relationship between microporosity and oil recovery efficiency. In general, remaining oil saturation
SPE-171940-MS 17

is positively correlated with permeability and negatively correlated with percent fraction of microporosity
due to the inherently homogenous pore and pore throat size distributions for limestone micro-pores. The
lowest remaining oil saturations measured were in rocks with the highest percent fraction microporosity.
This translates to an expectation of more favorable recovery but lower unstimulated rates for these rock
types. Rocks with higher permeability and lower percent fraction microporosity had the highest remaining
oil saturations. The expected performance from these rock types is less favorable recovery but higher
rates. The 80% threshold previously discussed distinguishes rock types with more favorable versus less
favorable recovery. Water provides more favorable recovery than gas across all combinations of pore type.
The findings from this study have significant implications for reservoir characterization and field
development activities as well as important insights into asset valuation and maximizing ultimate oil
recovery in limestone reservoirs with significant microporosity. Integration and synthesis across the
geology and engineering disciplines, as demonstrated here, facilitate new understanding of field perfor-
mance and illuminate new opportunities for optimizing value.

Acknowledgements
The authors would like to thank ADNOC, ZADCO and its Shareholders for their review and permission
to present this paper and our Project Team members who have contributed to it.

References
Anselmetti, F.S., Luthi, S., Eberli, G.P., 1998. Quantitative Characterization of Carbonate Pore
Systems by Digital Image Analysis. AAPG Bulletin 82(10): 1815–1836.
Kaczmarek, S.E., Fullmer, S.M., Hasiuk, F.J., Mitchell, J.C. 2011. Characterization and Genesis of
Limestone Microporosity: A Global Study. Paper presented at 14th Bathurst Meeting of Carbonate
Sedimentologists, Bristol, United Kingdom, 12-14 July
Loucks, R.G., Lucia, F.J., Waite, L.E., 2013. Origin and Description of the Micropore Network Within
the Lower Cretaceous Stuart City Trend Tight-Gas Limestone Reservoir in Pawnee Field in South Texas.
GCAGS Journal 2: 29 –41
Lucia, F.J., 2007. Carbonate Reservoir Characterization: An Integrated Approach, second edition,
348. Springer Berlin Heidelberg New York