OTC-26771-MS

Sand-Silt-Clay Evaluation Models: Which One to Use - A Case Study in the

Malay Basin
Gamal Ragab Gaafar, Michael Mehmet Eltunbay, and Shaharudin B A Aziz, PETRONAS Carigali SDN BHD;
Ehab Najm, Halliburton

Copyright 2016, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference Asia held in Kuala Lumpur, Malaysia, 22–25 March 2016.

This paper was selected for presentation by an OTC 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 Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the
written consent of the Offshore Technology Conference 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 OTC copyright.

Abstract
Clastic-reservoir characterisation of the Malay basin poses numerous challenges resulting from the silty
and clayey nature of its reservoirs. To date, different practitioners use shale volume (Vsh) and clay volume
(Vcl) in an exchangeable fashion, where the former is rock and the latter is defined as either size or
mineral content. Inaccurate quantification of Vcl magnifies the errors inbound in the calculation of
porosity and saturation because of the impact of clay density and conductivity in relevant equations.
Different models were developed to evaluate sand, silt, and clay reservoirs, which may be applicable
in some local areas and not applicable in others. One of the more common caveats in some of the available
models is the mix between the silt volume, Vsilt, and the volume of dispersed clay, Vcl_disp, which has
an impact on the effective porosity and water saturation calculations, Sw, especially when evaluating
lowresistivity/low-contrast (LRLC) zones using the Thomas-Stieber model.
In the presented study, the same data set was treated with three different deterministic models to solve
for sand, silt, and clay volumes. These models include: 1) the sand-silt-clay-water model, also known as
the Malay or SSC model; 2) a model that uses maximum porosity to calculate the silt volume; and 3) a
technique that uses nuclear magnetic resonance (NMR) clay-bound and capillary-bound volumes from the
NMR porosity model.
To select the most accurate and reliable petrophysical approach, the results were compared with X-ray
diffraction (XRD) analysis results in the area of the study, and a comparison of grain-size distributions of
actual data with grain-size distributions obtained from NMR T2 measurements by a recently developed
model was made.
The proposed technique helps in selecting maximum sand and shale porosities as one of the essential
parameters in the Thomas-Stieber model for typifying and quantifying the shale and for deciding whether
the laminated-dispersed or laminated-structural shale model should be used.
The refined sand, silt, and clay volumes and porosities along with tensor resistivity data were then input
into the tensor model for a petrophysical evaluation across anisotropic sand intervals.
The proposed model will help to minimise the uncertainty as a fit-for-purpose approach by improving
the accuracy in the calculated mineralogy, porosity, saturation, and consequently, reserve estimations.
2 OTC-26771-MS

Introduction
There are many publications that discuss the numerous clastic reservoir classifications; one of the most
recent publications is by Lazar et al. (2015), which concluded that fine-grained rocks can be represented
within the full spectrum of clastic sedimentary rocks on a ternary diagram with percent of sand, coarse
mud, and fine mud as end members (Fig. 1). The grain-size boundaries in this ternary space is defined as
follows: fine mud (clay and very fine silt) is less than 8 ␮m, medium mud (fine and medium silt) ranges
from 8 to 32 ␮m, coarse mud (coarse silt) ranges from 32 to 62.5 ␮m, and sand ranges from 62.5 to 2000
␮m. The benefits of the proposed grain-size boundaries are very well explained by Lazar et al. (2015).
Siltstone and claystone are terms well entrenched in the literature of fine-grained sedimentary rocks but,
have somewhat variable definitions and their use has generated confusion in the past. Following the
proposed grain-size boundaries (Fig. 1) and making a distinct separation of textural and compositional
meaning, one could use, however, the siltstone, mudstone, and claystone terms as alternatives to the
coarse, medium, and fine mudstone terms. In this alternative scheme, a claystone, for example, is a rock
composed of more than two-thirds fine mud-size grains, less than a third coarse mud-size grains, and less
than a quarter sand-size grains. Similarly, siltstone is a rock composed of more than two-thirds coarse
mud-size grains, less than a third fine mud-size grains and less than a quarter sand-size grains.

Figure 1—Nomenclature guidelines for fine-grained sedimentary rocks: texture (grain size) (left) and classification (Blatt 1983) based
on size in ␮m (right).

Although there are many books and papers that talk about shale and clay, they do not differentiate or
explain them clearly or correctly. To date, both the term ⬙volume of shale⬙ (Vsh) and term ⬙volume of
clay⬙ (Vcl), are used interchangeably. Moreover, there is not enough in the literature that addresses the
volume of silt (Vsilt) nor sand-silt-clay. According to Spooner (2014), shale is a rock and is typically
defined as an indurated, finely laminated, sedimentary rock, composed primarily of clay, mud, and silt.
The important feature to note is that this definition does not describe the mineralogy but rather the grain
size. In this definition, clay refers to clay-sized particles (i.e., ⬍1/256 mm). Whilst clay can refer to grain
size, it can also refer to clay minerals, and it is the dual meaning of the word clay that is at the heart of
the confusion in the industry. When computing porosity, one should account for the clay minerals (for
example, in determining when their density is different from the matrix density). When computing water
OTC-26771-MS 3

saturation, one should account for the excess conductivity resulting from the clay minerals. In both cases
it is the volume of clay minerals to correct for, not the volume of shale.
According to Barlai (1971), the presence of silt 1) decreases the electrical resistivity of productive
zones, 2) increases natural gamma-ray activity of the rocks, 3) increases neutron porosity, and 4) in
contrast, does not decrease spontaneous potential (SP) deflections. These four effects of silt can result in
misinterpretation in identifying the reservoir or productive zones, volume of clay, porosity, and fluids
saturation.
Application of a binary sand-shale litho-porosity model (Fig. 2) suggested that effective porosity, ⌽e,
should approach zero in the area of the apex of the boomerang distribution on a typical neutron-density
crossplot (Heavysege 2002). Core plug samples from this same facies (in the Malay basin) have measured
porosities in the range of 10% to 15 %. Since core porosity was assumed to represent effective porosity,
⌽e, then it was assumed that the binary model was underestimating effective porosity, ⌽e. At the same
time, sieve data revealed the presence of a significant amount of silt-sized grains, typically in the order
of 30% to 60%. It was concluded that a binary sand-shale litho-porosity model was inappropriate for these
silty reservoirs, and that more effective porosity, ⌽e, needed to be accommodated. The development of
silty reservoir models had been spawned. In the Malay Basin model, the silt was accommodated at the
expense of clay minerals, or shale. As a consequence, the locally used silty reservoir models have
underestimated the amount of clay minerals associated with the silt-sized grains in the low-energy facies
at the apex of the boomerang. Recent core studies have included petrographic descriptions (TS, SEM), as
well as XRD, XRF, wet chemistry and NMR (Fig. 3). All have defined the presence of clay minerals (20%
to 40%) at the silt point.

Figure 2—Neutron-Density Cross-Plot Representing Sand-Silt-Clay Boomerang Shape and XRF/XRD for Core Samples in the Malay
Basin (Heavysege 2002).
4 OTC-26771-MS

Figure 3—The litho-porosity model based on the neutron-density crossplot.

This paper focuses on the volumetric sand-silt-clay calculation technique; details on porosity and water
saturation will be included in a later publication.
Methodology

There are different models developed to evaluate sand-silt-clay reservoirs, among which, three
different techniques were used and are discussed in this paper. The first two techniques are based on a
densityneutron crossplot while the third uses NMR porosities.
Phi_max Technique
Barlai (1971) shows that clay gradually fills the maximum sand porosity, ⌽sd,Max. Thus, in a shaly
sandstone not containing silt, the value of the effective porosity for a given clay content Vsh, will be
(⌽sd,Max - Vsh). If in such a rock type the gradually increasing silt fraction Vsilt, is substituted for the sand
fraction Vsand, it can be assumed that with increasing volume of silt the macro-porosity ⌽sd is decreasing
linearly with Vsilt per the following function:
(1)

Rearranging this equation, one can solve for volume of silt, Vsilt. Some other commercial software
uses the following equation:
(2)

Where,
Vcl ⫽ Volume of clay
Vsh ⫽ Volume of shale
Vsilt ⫽ Volume of silt
⌽e ⫽ Effective porosity
⌽sd ⫽ Macro pores
⌽max ⫽ Maximum reservoir porosity
OTC-26771-MS 5

Malay Model
The litho-porosity model based on the density-neutron crossplot is illustrated in Fig. 3 (Kuttan et al.
1980). The basic framework of this model is defined in terms of quartz (Q), water (W), dry clay (M), and
⬙mineral⬙ silt (X). The data points, which have boomerang type distribution, can be related to these end
members. Thus, the point K represents a clean, wet, quartz sandstone, point L represents a wet, pure clay,
and P represents a wet, pure siltstone. These end members are referred to as sand, clay, and silt
respectively. Thus, any rock (matrix and porosity components) can be described in terms of these end
members.
The position of the dry silt point (X) is defined by the extension of the water-silt line to a grain density
of 2.68 g/cc. The dry clay point (M) is obtained by extending the sand-silt trend (KP) to intersect the
water-clay trend (WL). Although this method of obtaining M is strictly a constructional approach, the
density and neutron responses of this point fall within the range of accepted values. The actual location
of the point M is not critical to the interpretation.
In this model, based on the end members defined above, sand and clay are considered to be mutually
exclusive. Thus the KP trend is essentially clay free, and the PL trend is essentially sand free.
In Fig. 3 the water-wet silt line (WP) is projected through X to intersect the quartz-dry clay line QM
at N. A matrix ratio (MRx) is determined for this point where
(3)

Similarly, any data point (i.e. point A in Fig. 2) can be projected to intersect QM at Y, and the matrix
ratio (MR) of the data point can be determined where
(4)

The value of MR can be compared with MRx to determine whether the data point lies on the sand-silt
or the silt-clay trend. For the sand-silt trend the line QX represents the zero line of total porosity, and
isoporosity lines are constructed parallel to it. From Eqs. 2 and 3, total porosity equals effective porosity.
Within this trend, sand and silt are the only components considered to be present and hence,
(5)

where Vsd and Vsi are the bulk volume fractions of dry sand and dry silt respectively. Vsi may be
expressed in terms of the matrix ratios MRx and MR:
(6)

For the silt-clay trend, the line XM represents the line of zero total porosity, and iso-porosity lines are
drawn parallel to it. Within this trend, silt and clay are the only components considered to be present and
hence,
(7)

where Vsi and Vc1 are the bulk volume fraction of dry silt and dry clay respectively. Vsi can be
expressed in terms of the matrix ratios (MR and MRx) and total porosity as follows:
(8)

Please refer to the referenced paper for the porosity and water saturation calculation.
NMR Technique
NMR total porosity, ⌽TNMR, CBW, and BVI obtained from either T1 or T2 relaxation can be used
to calculate effective porosity, ⌽E, and volumes of sand, silt, and clay in the reservoir, provided that NMR
porosity is corrected for HI, cutoffs are obtained from the core, and there is a relation between T1 or T2
6 OTC-26771-MS

distribution and pore size distribution. The total bound fluid volume BFV is split between clay-bound
(⌽clay) and capillary-bound (⌽silt) volumes with a clay cutoff (Claverie et al. 2007). The volumes of clay
and silt, and the effective porosity can then be calculated with the following equations:
(9)

(10)

(11)

(12)

This method can provide a robust and fast volumetric analysis. Results from the listed equations can
be compared with the sand-silt-clay model or any other volumetric analysis model used to evaluate the
reservoir for verification (Fig. 3).
An illustrative workflow for the three different techniques is shown in (Fig. 4). It is clear that the third
technique is the simplest in terms of number of steps.

Figure 4 —Workflow for the three different Vsilt calculation techniques.

Results Validation
As a normal practice, petrophysical analysis validation is done using results obtained from routine or
special core analysis. For volumetric analysis in particular, data from XRD and/or core grain-size analysis
can be used. However, if either data is unavailable, an alternative can be used. There are two different
techniques, recently developed, that use an NMR T2 relaxation spectrum that transfers it to grain-size
distribution. These two techniques are 1) micro structure rock modeling (MSRM) and 2) cutoff technique.
OTC-26771-MS 7

MSRM
Theoretically, NMR T2 relaxation can respond to the pores that are filled with wetting phase (e.g. water
in water wet), but it does not necessarily respond to the pores filled with non-wetting fluids. To derive
grain-size distribution from NMR, relaxation time measurement and the MSRM model are used to
construct the pore/grain size distribution and to simulate NMR response by considering mineralogy,
partial water saturation, surface relaxivity, and the surface roughness factor of the grains (Chen et al.,
2010).
Cutoff Technique
This technique is very simple and applies different T2 cutoffs (clay bound and capillary bound) on the
NMR T2 relaxation spectrum. Each cutoff represents a specific pore size, which is then translated to grain
size. For sand pore size, the large pores containing free fluid will always have some capillary-bound fluid
associated with them, even if it is a very small amount (Lowden, 2003). These cutoffs can be constant or
variable according to the nature of the reservoir. To find the volume of this bound fluid requires an
estimate of the free fluid/bound fluid ratio. If, for example, the large pores are associated with sands, the
free fluid/bound fluid ratio is determined by calculating the ratio everywhere the purest sands are
encountered. How the cleanest sands are identified depends on the preferred method for identifying
lithology from the log suite. The free fluid /bound fluid ratio must then be extended away from the
reference points (e.g. by interpolation or by dividing the reservoir into zones and computing the average
value; see Fig. 3). Having selected the method to generate a continuous curve of free fluid/bound fluid,
the values are used to calculate the porosity of the sands, which is then subtracted from the capillary-
bound fluid to obtain the porosity of the silts:
Bound fluid associated with sands ⫽ measured free fluid/(free fluid/bound fluid_sand)
Sand porosity ⫽ sand bound fluid ⫹ free fluid
Silt porosity ⫽ capillary-bound fluid - sand bound fluid
At this stage, each T2 distribution in the log is divided into three components that are broadly related
to lithology. Another technique for calculating the sand and silt porosity is by using capillary-bound BVI
calculated using spectral technique, SBVI, where the technique takes into consideration the contribution
of each pore size to bulk volume irreducible, BVI, using the following equation:
(13)

where the subscript i represents the bin number, the SBVI coefficient ci represents the irreducible water
saturation associated with each T2 bin, and Pi is the porosity associated with each bin. This parameter ci
must be obtained from NMR core analysis.
According to Lowden (2003), for Silt pore size, the SDR equation requires a T2 value that represents
the average pore size of the silts. In most cases, the bound fluid part of the T2 distribution has a uni-modal
T2 shape, where the representative T2 value is the mode. The modal T2 is found by searching for the
maximum amplitude in the T2 distribution between the cutoffs for clay-bound water and bound fluid.
There may be circumstances where a representative T2 value cannot be found (e.g., the T2 distribution
has a poly-modal or flat shape). Where this situation occurs, the T2 value for the specific pore size
population must be extended away from the reference points using a method similar to that used for the
free fluid/bound fluid ratio of the larger pores, although the method may differ if the depositional controls
on lithology associated with different pore size populations are independent. In this work, the following
assumptions are made:
8 OTC-26771-MS

● The reservoir is water wet.

● NMR responds to the pores filled with wetting phase (i.e., clay-bound and capillary-bound can be
trusted to represent clay and silt pore sizes respectively.
● A volume of pores higher than capillary-bound T2 cutoff can be used to represent pore sizes larger
than silt size without a need to break it into different sizes.
That said, the NMR relaxation spectrum was divided using clayand capillary-bound T2 cutoff into three
main pore sizes (clay, silt and sand respectively), and porosity was translated to volumes of different sizes.
This data was used to validate the processing results from the three techniques.
Case Study
Kuttan et al. (1980) describes the Malay Basin as a large, asymmetrical, Tertiary basin located off the east
coast of West Malaysia within the South China Sea. It extends approximately 400 miles along its
northwest-southeast axis and is about 100 miles wide. The Malay basin reservoirs generally consist of
very fine-grained sediments with fresh to brackish formation waters. Grain-size analysis of the reservoirs
shows that the clay content is generally low although the silt content may be significant. Core data and
log responses indicate that the porosity distribution and solids conductivity are controlled largely by the
silt-size particles. Under the conditions that exist in the Malay Basin, these silt-size particles appear to
behave as a specific mineral type.
Extensive core data in the field indicates that the reservoirs are very fine grained. Sieve analysis, thin
section, scanning electron microscope and X-ray diffraction studies show these rocks to have a high-silt
and low-clay content. Although the reservoirs are silty, their producibility is generally good.
The triple combo acquired while drilling (displayed in the composite Fig. 5) indicates that the reservoir
is sand-shale sequence. An excavation effect on the density neutron suggests it is a gas-bearing reservoir.
Also, there are some coal beds throughout the logged section.

Figure 5—Study well location and geologic column.

OTC-26771-MS 9

Figure 6 —Composite of triple combo acquired while drilling.

The gamma ray is relatively high, which is interpreted as resulting from the presence of feldspars. On
the neutron-density crossplot, the dry and wet end points for sand-silt-clay are identified by analysis. The
gas-bearing sand is clearly clustered above the sand-water line.
10 OTC-26771-MS

Figure 7—Raw neutron-density crossplot.

OTC-26771-MS 11

Figure 8 —Sand-silt-clay calculated using Phi-max technique.

12 OTC-26771-MS

Figure 9 —Sand-silt-clay model results.

OTC-26771-MS 13

Discussion
The data was processed using the three different techniques discussed earlier in this paper. The following
are the results from the three techniques:
● The Phi_max model overestimates the volume of clay, Vclay, in reservoir section the matter that
drops effective porosity and hydrocarbon volume.
● The Sand-Silt-Clay model seems to overestimate the volume of silt, Vsilt, on the account of
volume of clay, Vclay. This can be controlled by zoning the logged section and changing the
dry/wet endpoints for silt and clay. This step requires a good knowledge about the different
volumes and availability of XRD and any other volumetric calibration means. Also, the model is
very much dependent on density neutron data and its integrity. Moreover, if the density of
hydrocarbon is unknown or is input incorrectly, it will impact the estimated volumetrics.
● The NMR model is relatively simple and robust; however, it is very much dependent on the
porosity of clay bound in wet clay. Two other major considerations are 1) a good knowledge of
T2 cutoffs for clay and capillary bound is necessary, otherwise cutoffs can be selected visually if
they can be easily picked on the T2 relaxation spectrum; 2) the total porosity must be corrected
for HI in case of gas or light hydrocarbon. Finally, though it is not the case in this study, the
technique will be limited in heavy oil or when diffusion brings the fluids signal to the expected
capillary-bound water signal.

Results Validation
Due to the lack of core and/or XRD analysis results during the time of studying this well, NMR was used
to validate results by transferring pore-size distribution to grain size distribution. As discussed earlier, for
simplicity, this research uses the cutoff technique to divide the reservoir into clay, silt and sand sizes. Figs.
11 and 12 show cumulative pore-size distribution at two different depths of the reservoir (denoted by red
dots in Fig. 10). The plots suggest that the volume of clay, Vclay, is very low (5% to 20%) compared to
the volume of silt, Vsilt, (10% to 50%), which is supported by the basin description by Kuttan et al. (1980)
and is indicated in the sieve analysis displayed in the ternary plot (Fig. 13). When comparing results from
the three different models with NMR estimated grain-size distribution and core analysis in the basin, one
can easily conclude that the NMR model gives the best answer.
14 OTC-26771-MS

Figure 10 —A comparison between conventional, sand-silt-clay and NMR volumetric Models.

Figure 11—NMR cumulative T2 pore size distribution.

OTC-26771-MS 15

Figure 12—Cumulative T2 pore size distribution (percentile).

Figure 13—Malay Basin core sieve analysis displayed in ternary diagram.

16 OTC-26771-MS

Applications in Anisotropic Reservoirs, LRLC

NMR can be used to minimise the uncertainty in anisotropic sand/shale laminated reservoirs. While it can
play a key role in lithology independent porosity and hydrocarbon typing and quantification and
permeability, it can also provide volumes of sand-silt-clay provided that porosity is corrected for hydrogen
index and polarization effect. According to Minh and Sundararaman (2006), the NMR response is
independent of the lamination distribution and geometry (i.e. NMR cumulates sand/shale thicknesses that
are below the tool vertical resolution). In anisotropic reservoir analysis, whether using Tensor or Thomas
Stieber models, NMR can provide sand porosity (⌽sand), volume fraction of sand (Fsand), and/or
net/gross ratio using the following equations:
(14)

(15)

Using ⌽Sand100% & ⌽Clay100% gives

(16)

(17)

Details on the role of NMR in anisotropic reservoir evaluation to provide the volume fraction of sand
or shale needed to balance the horizontal and vertical resistivity equations in tensor model are explained
in Fanini et al. 2001 and Shray and Borbas 2001.

Conclusion
● Similar to clays, the presence of silt in the reservoir adds to the reservoir characterisation
challenges.
● There are different techniques for volume of silt quantifications.
● Validation with core is necessary to prove the convenient technique to be used.
● In the absence of core, NMR can play a rule in validating sand-silt-clay calculated volumes by
applying the grain-size distribution obtained from the T2 relaxation spectrum. However, some
work needs to be done to help ensure the effectiveness of the technique.

Acknowledgement
The authors wish to thank the management of PETRONAS, PETRONAS Carigali Sdn. Bhd, and
Halliburton for granting permission to develop the workflow and to publish this paper. We also express
our appreciation to the respective technical review groups whose input has helped to produce the final
document.

References
Barlai, Z. 1971. Some Principal Questions of Well Logging Evaluation of Hydrocarbon-Bearing Sandstones with a High
Silt and Clay Content: Experience Acquired by the Field Application of a New Method. The Log Analyst 12 (3).
SPWLA-1971-vXIIn3a2.
Blatt, H. 1983. Sedimentary Petrology. Geological Magazine 120 (05): 516 –517. http://dx.doi.org/10.1017/S0016756800027576.
Chen, J., Chen, S., Altunbay, M. et al 2010. A New Method of Grain Size Determination for Sand-Control Completion
Applications. Presented at the SPE International Symposium and Exhibiton on Formation Damage Control, Lafayette,
Louisiana, 10-12 February. SPE-128011-MS. http://dx.doi.org/10.2118/128011-MS.
Claverie, M., Hansen, S.M., Daungkaew, S. et al 2007. Application of NMR Logs and Borehole Images to the Evaluation
of Laminated Deepwater Reservoirs. Presented at the Asia Pacific Oil and Gas Conference and Exhibition, Jakarta,
Indonesia, 30 Oct – 1 November. SPE-110223-MS. http://dx.doi.org/10.2118/110223-MS.
OTC-26771-MS 17

Fanini, O.N., Kriegshauser, B.F., Mollison, R.A. et al 2001. Enhanced, Low-Resistivity Pay, Reservoir Exploration and
Delineation with the Latest Multicomponent Induction Technology Integrated with NMR, Nuclear, and Borehole
Image Measurements. Presented at the SPE Latin American and Caribbean Engineering Conference, Buenos Aires,
Argentina, 25-28 March. SPE-69447-MS. http://dx.doi.org/10.2118/69447-MS.
Heavysege, R.G. 2002. Formation Evaluation Of Fresh Water Shaly Sands of the Malay Basin, Offshore Malaysia.
Presented at the SPWLA 43rd Annual Logging Symposium, Oiso, Japan, 2-5 June. SPWLA-2002-SS.
Kuttan, K., Stockbridge, C.P., Crocker, H. et al 1980. Log Interpretation in the Malay Basin. Paper Presented at the
SPWLA 21st Annual Logging Symposium, Lafayette, Louisiana, 8-11 July. SPWLA-1980-II.
Lazar, O.R., Bohacs, K.M., Macquaker, J. et al 2015. Capturing Key Attributes of Fine-Grained Sedimentary Rocks in
Outcrops, Cores, and Thin Sections: Nomenclature And Description Guidelines. Journal of Sedimentary Research 85
(3): 230 –264.
Lowden, B. 2003. A New Method For Separating Lithologies and Estimating Thickness-Weighted Permeability Using
Nmr Logs. Presented at the SPWLA 44th Annual Logging Symposium, Galveston, Texas, 22-25 June. SPWLA-
2003-1414P.
Minh, C.C. and Sundararaman, P. 2006. Nuclear Magnetic Resonance Petrophysics in Thin Sand/Shale Laminations.
Presented at SPE Annual Technical Conference and Exhibition, San Antonio, Texas, USA, 24-27 September.
SPE-102435-MS. http://dx.doi.org/10.2118/102435-MS.
Shray, F. and Borbas, T. 2001. Evaluation of Laminated Formations Using Nuclear Magnetic Resonance and Resistivity
Anisotropy Measurements. Presented at the SPE Eastern Regional Meeting, Canton, Ohio, USA, 17-19 October.
SPE-72370-MS. http://dx.doi.org/10.2118/72370-MS.
Spooner, P. 2014. Lifting the Fog of Confusion Surrounding Clay and Shale in Petrophysics. Presented at the SPWLA 55th
Annual Logging Symposium, Abu Dhabi, United Arab Emirates, 18-22 May. SPWLA-2014-VV.