Petrophysics/Petrofísica/Petrofísica
Comparative analysis for resulting petrophysical property ranges using different mercury injection capillary pressure methods
Análisis comparativo para los rangos de propiedades petrofísicas resultantes usando diferentes métodos de presión capilar de inyección de mercurio
Análise comparativa para as faixas de propriedade petrofísica resultante usando diferentes métodos de pressão capilar de injeção de mercúrio
Alfonso Quaglia
Ing°Geó°. Inter-Rock USA, LLC. (International Rock Analysts). Correo-e: quagliaa@inter-rock-ca.com
Antonio Montilva
Ing°Geó°. Consultor G&G Inter-Rock. Correo-e: antonio.montilvalarre@gmail.com
Juan Carlos Porras
Ing°Geó°. MSc. Inter-Rock Panamerican. Correo-e: porrasjc@inter-rock-ca.com
Rafael Panesso
Ing°Geó°, Esp. Inter-Rock Panamerican-Colombia. Correo-e: panessor@inter-rock-ca.com
Recibido: 29-6-20; Aprobado: 31-7-20
Abstract
Mercury injection capillary pressure is one of the most important methods to determine certain petrophysical properties including reservoir quality, storage & flow capacity, etc. At the same time, it is possible that uncertainties may be present depending upon the applied measurement method, apparatus calibration, sample size & shape, etc. Nevertheless, the used methodology in this investigation took care of certain conditions to minimize the mentioned uncertainties. A set of consolidated sandstone samples, covering varieties of reservoir quality rocks, were used for lab measurements. Proper applied methodology, maximizing consistency and homogeneity for individual core plugs, made possible analyzing variations that occur in the estimation of certain petrophysical properties such as: porosity, water saturation and Hydrocarbon column from diverse mercury injection capillary pressure tests, either manual or automatic. Additionally, rock types were characterized using the Winland R35 empirical model. Routine and special core analyses were performed for this purpose. Results of this investigation indicate that, as the quality of the rock decreases, the greater the uncertainty in the calculations of the petrophysical properties, so it is recommended to take into account the percentage of variation obtained for each petrophysical property determined on each rock type, based on both: similar rock quality and different lab measurements systems.
Resumen
La presión capilar de inyección de mercurio es uno de los métodos más importantes para determinar ciertas propiedades petrofísicas, incluida la calidad del depósito, la capacidad de almacenamiento y flujo, etc. Al mismo tiempo, es posible que existan incertidumbres dependiendo del método de medición aplicado, la calibración del aparato, tamaño y forma de la muestra, etc. Sin embargo, la metodología utilizada en esta investigación se ocupó de ciertas condiciones para minimizar las incertidumbres mencionadas. Se utilizó un conjunto de muestras consolidadas de arenisca, que abarcaban variedades de rocas de calidad de reservorio, para las mediciones de laboratorio. La metodología aplicada adecuada, maximizando la consistencia y la homogeneidad de los tapones de núcleos individuales, permitió analizar las variaciones que ocurren en la estimación de ciertas propiedades petrofísicas como: porosidad, saturación de agua y columna de hidrocarburos de diversas pruebas de presión capilar de inyección de mercurio, ya sea manual o automática. Además, los tipos de rocas se caracterizaron utilizando el modelo empírico Winland R35. Se realizaron análisis básicos de rutina y especiales para este propósito. Los resultados de esta investigación indican que, a medida que disminuye la calidad de la roca, mayor es la incertidumbre en los cálculos de las propiedades petrofísicas, por lo que se recomienda tener en cuenta el porcentaje de variación obtenido para cada propiedad petrofísica determinada en cada tipo de roca, basado en ambos: calidad de roca similar y diferentes sistemas de medición de laboratorio.
Resumo
A pressão capilar por injeção de mercúrio é um dos métodos mais importantes para determinar certas propriedades petrofísicas, incluindo qualidade do tanque, capacidade de armazenamento e fluxo, etc. Ao mesmo tempo, pode haver incertezas dependendo do método de medição aplicado, da calibração do dispositivo, tamanho e forma da amostra, etc. No entanto, a metodologia utilizada nesta pesquisa tratou de certas condições para minimizar as incertezas mencionadas. Um conjunto de amostras consolidadas de arenito, abrangendo variedades rochosas de qualidade de reservatório, foi utilizado para medições laboratoriais. A metodologia aplicada adequada, maximizando a consistência e homogeneidade dos plugues centrais individuais, permitiu analisar as variações que ocorrem na estimativa de certas propriedades petrofísicas, tais como: porosidade, saturação da água e coluna de hidrocarbonetos de vários testes de pressão capilar de injeção de mercúrio, manual ou automático. Além disso, os tipos de rocha foram caracterizados usando o modelo empírico Winland R35. Foram realizadas rotinas básicas e análises especiais para este fim. Os resultados desta pesquisa indicam que, à medida que a qualidade das rochas diminui, a maior incerteza nos cálculos das propriedades petrofísicas, recomenda-se levar em conta o percentual de variação obtida para cada propriedade petrofísica determinada em cada tipo de rocha, com base em ambos: qualidade de rocha semelhante e diferentes sistemas de medição laboratorial.
Palabras clave/Keywords/Palavras-chave:
Capillary pressure of mercury injection, columna de hidrocarburos, coluna de hidrocarbonetos, hydrocarbon column, porosidad, porosidade, porosity, presión capilar de inyección de mercurio, pressão capilar de injeção de mercúrio, saturación de agua, saturação de água, water saturation, Winland R35.
Citar así/Cite like this/Citação assim: Qujaglia et al.(2020) o (Quaglia et al, 2020)
Referenciar así/Reference like this/Referência como esta:
Quaglia, A., Montilva, A., Porras, J. C., Panesso, R. (2020, agosto). Comparative analysis for resulting petrophysical property ranges using different mercury injection capillary pressure methods. Geominas 48(82). 69-87.
Introduction
The quality of reservoir rocks is determined, among other factors, by the pore volume, the pore throat size distribution and the permeability. (Hartmann & Beaumont, 1999).
Winland's equations relate the porosity and permeability of a rock with various parameters obtained from the capillary pressure curves through mercury injection method used. More specifically, they relate to porosity and uncorrected air permeability with the Pore throat radius (Hartmann & Beaumont, 1999). Using the porosity and permeability data, it's possible to determine the Pore throat radius that dominates the flow through the Winland equation also known as Winland R35.
On the other hand, Pore throat size distribution of a rock can be obtained from the capillary pressure curves. Among some methods through which capillary pressure can be measured in the laboratory are: the centrifuge method, the porous plate method and the mercury injection method; the latter being the method to be used in this study, since it is the most used in the industry, for pore type determination purposes.
This study complied with quality standards in sample preparation. Among these standards, a computerized axial tomography or better known as a CT Scan was performed, to guarantee homogeneity of the samples. The permeabilities of the rocks were calculated applying Darcy's Law and using the gas permeameter. Different capillary pressure tests were performed, using either digital mercury porosimeter and manual mercury pump. Likewise, to fulfill the fundamental purpose of this research work, it was divided into several stages, going from justification, introduction to the topic, background and objectives; followed by overview, sampling, measurements, generated workflow and results discussion.
Background
In 1842 the possibility of forcing mercury into wood was already mentioned to obtain its porous structure. In 1921 Washburn suggested obtaining the pore size distribution from the pressure-volume data obtained by mercury intrusion. By 1940, this technique began to be applied, and in 1945 Rittler and Drake were the first to publish a large number of experimental data, including the description of the equipment and the operation mode. Nowadays, with the commercial development of equipment, it's the increasingly used test (Van Brakel J, Modry S and Svata M, 1981).
As well known, Rocks are minerals aggregates with different physical, chemical and geometric properties. Furthermore, rocks have complex internal geometries, known as microstructures, that exhibit a great variety of heterogeneities, such as the disorder in mineral arrangements, the variability in their mineralogical composition, the degree of fracturing, the grain size, the number of pores and their size, among other things, which depend on the measurement scale. With a sufficiently large measurement scale all parts of a rock would present similar physical properties, but with a small measurement scale (such as the grain size scale) this same rock would exhibit particular heterogeneities (Guéguen & Palciauskas, 1994).
Due to the formation process of the sedimentary rocks and the distribution of local stresses, the individual grains are not in continuous contact with each other, leaving significant empty spaces (pores) within them. This porous space in terrigenous rocks is a complex irregular system, which is sometimes interconnected but sometimes not, and whose sizes vary from micrometers to tens of Virtually, all the macroscopic physical properties of rocks are influenced by the microstructure of the pores. However, this detailed microscopic information cannot be measured, hence the importance of certain measurable macroscopic parameters, such as porosity, permeability, and capillary pressure, among others, that provide essential information on the porous structure (Guéguen & Palciauskas, 1994).
H.D. Amoco's Winland in late 1970s showed a statistical correlation between the optimal flow through rocks and the pore throat radii, when 35% of a rock's pore space is saturated with a phase that does not wet during a capillary pressure test. He called this pore throat size r35 or dominant pore size. (See table 1).
Table 1. Pore types and size ranges. (Hartmann & Beaumont, 1999).
As a result of his work with 322 samples, of which only 82 (56 sandstones and 26 carbonates) had already corrected low permeabilities, Winland developed an empirical relationship between porosity, air permeability (uncorrected), and Pore throat radius (Equation 1).
Log (r35) = 0.732 -0.588Log (k) -0.864Log (φ) (1)
Where:
r35 = Pore throat radius (µm)
k = Uncorrected Air Permeability (mD)
φ = Porosity (%).
Methodology
In this experimental investigation, a group of samples will be subjected to certain conditions to observe the effects that are generated as dependent variables. The overall methodology consisted in five stages:
1) Background review: geological information, Handbook procedures, Core inventory, sample classification according to reservoir quality, etc.
2) Core Plug Sampling and preparation: CT Scan, Plugging, shaping, Cleaning & Drying.
3) Routine core analysis (RCAL): Porosity, Permeability, Grain Density & Sample description.
4) Special core analysis (SCAL): Capillary Pressure Tests using three different methods for mercury injection.
5) Comparative analysis and discussion of results.
1.-Background review:
In order to complement the knowledge and information of the study, previous studies were reviewed, as well as laboratory handbooks, research related to the subject, such as specialized bibliography and technical reports. Once necessary information was selected, an inventory of available samples was done (Figure 1.1), and those resulted as most suitable for analysis were chosen in order to achieve results with the greatest certainty possible. Previous analyses performed on available core were used as reference to try to obtain a set of samples that will reflect various rock types and reservoir qualities.
Figure 1.1. Available core for the study.
After performing CT Scanning on previously selected core samples, those with a high degree of heterogeneity (bioturbations, fractures, laminations, etc.) were discarded. In this case, one of the samples was discarded for presenting a fracture that could be detected thanks to the help of a 3D computed tomography. (Figure 2.2).
Figure 2.2. 3D section showing discarded sample with undetectable fracture with naked eye.
2.2.-Plugging & Sample Preparation:
For plugging the five core samples, a diamond drill steel tube drill was used (Figure 2.4), which has a water-based cooling system. When cutting each of the 5 core samples, two cylindrical plugs were obtained from each one (10 plugs). It should be noted that due to limitations in the volume of available rock, plugs were cut to a diameter of one inch (2.54 cm). The length of the plugs were variable but made sure keeping the same length for each pair of plugs of the same rock quality.
2.-Core plugs Sampling and Preparation:
2.1.-Sample QC & Selection
Computerized axial tomography (CT Scan) (Figure 2.1) was performed on six pre-selected core samples in previous stage in order to avoid certain undesired features and fractures, which were not detectable by simple observation, and thus optimize the samples to be studied. Siemens CT Scanner, Somaton Spirit model was used to scan available core.
Figure 2.1. CT Scanning on Available Cores.
Best efforts were made to honor all reservoir qualities present and selecting the most homogeneous samples in order to compare the different measurements as accurately as possible and thus discard features that could introduce bias in the study results. Figure 2.3.
Figure 2.3. Electronic Transversal cut from CT Scan to optimize selection.
Figure 2.4. Diamond bit core drill.
Each pair of plugs was matched to the same length. For this, it was necessary to use a diamond edge disc cutter (Figure 2.5), then the surfaces of the plugs were shaped to eliminate any type of irregularity. This was done for two main purposes: a) to shape them into regular cylinders and b) to match the length of the plugs in pairs.
Once these activities were completed, 10 plugs were obtained (5 pairs), where the length of each pair of plugs were the same (2 inches minimum); Enough length to perform porosity, permeability and mercury injection capillary pressure tests. (Figure 2.6).
2.3.-Plugs Cleaning & Drying:
For plugs cleaning, separation method was used by distillation, using the Dean-Stark apparatus (Figure 2.7). This method consists of separating the components of the mixtures based on the differences in the boiling points of components. It should be noted that each pair of plugs was placed in a different Dean-Stark apparatus and that the cleaning process took around three weeks due to the saturation of some plugs. This process was performed in order to extract possible remains of hydrocarbons, water, and mineral salts, etc..., potentially affecting porosity, permeability, and capillary pressure tests.
Figure 2.5. Diamond edge disc cutter.
Figure 2.6. Set of study shaped plugs.
Figure 2.7. Dean-Stark apparatus.
Subsequently, the plugs were placed in a drying oven (Figure 2.8) at 80°C of temperature for approximately 72 hours until a constant weight was obtained; this was done in order to eliminate any trace of moisture and solvents that may have remained in the plugs from cleaning process. After the necessary time had elapsed, plugs were removed from the oven and were provisionally placed in a container with a lid (desiccator), until they cooled down (Figure 2.9) and thus protect them from moisture.
Figure 2.8. Used conventional oven.
Figure 2.9. Desiccator used to protect the plugs from moisture.
3.-Routine core analysis (RCAL).
3.1.-Porosity & Primary Grain Density:
Obtaining absolute porosity through helium porosimeter consists of a digital pressure gauge, as well as 2 manometers, a digital temperature gauge, a reference volume selector, a vacuum extractor valve, a sample holder where the grain density is measured, a helium regulating valve, and a panel where the valves that allow the passage of helium through the equipment are controlled. In addition, it is connected to a helium cylinder that has its respective manometer, stopcocks, and safety (Figure 3.1).
To obtain the absolute porosity for each plug, the gas expansion method was performed through the following activities:
Figure 3.1. Helium porosimeter.
Sample Volume: because the plugs have the shape of regular cylinders, their volume could be easily calculated through the cylinder volume formula (Equation 2).
V = π.r2.L (2)
Where:
V=Volume(cm )
r=Radius(cm)
L=Length(cm)
To obtain the lengths and radii of the plugs, an electronic Vernier (Figure 3.2) was used, which has an appreciation of 0.001 cm. To obtain the weight of each sample, a digital scale was used that has an appreciation of 0.01 g.
Figure 3.2. Electronic Vernier used to determine sample dimensions.
Grain volume and grain density were determined through gas expansion method at constant temperature, for which the samples were placed in a core cell with known volume. Helium was then allowed to pass into cell A and kept trapped at a pressure of 100 psi, then helium was released from cell A into the cell B by means of a duct which has a sufficiently small diameter, so that the volume of gas that could be found in that duct was neglected, therefore, the only volume taken into account was that of cell A. Then, the first volume of gas is equal to the volume of cell A, the second volume of gas is equal to the volume of cell A plus the volume of cell B minus the grain volume of the sample. The grain density was then calculated using the grain volume and the weight of dry sample. (Equation 3)
GD=Ws/Vg (3)
Where:
GD = Grain density (g/cc)
Ws = Sample Weight (g)
Vg = Grain volume (cc)
Then helium passed through the porosimeter connected to a hydrostatic core holder (Figure 3.3) capable of withstanding overload pressures, then an initial pressure of 100 psi was injected into the samples, then by pressure difference, using Boyle's Law the pore volume was calculated. Once this value was found, equation 4 was used to obtain the absolute porosity.
Ø =(Vt-Vg)/Vt×100 (4)
Where:
ø= Porosity (%)
Vt = Total Sample Volume (cc)
Vg = Grain volume (cc)
Figure 3.3. Core holder for overload porosity.
3.2.-Samples Description:
See table 2.
Table 2. Grain density & Sample description.
3.3.- Permeability:
Permeability determination was based on Darcy's law, using the gas permeameter (Figure 3.4). It consists of digital temperature and pressure indicators, a flow rate meter, a shut-off valve that connects the system to a specific rotameter (20 cc, 200 cc, 2,000 cc) as well as various valves that regulate the amount of gas and its path through the equipment. In addition, it should be noted that it is connected to a hydrostatic core holder capable of withstanding overload pressures and to a nitrogen cylinder that has its respective manometer, stopcocks, and safety.
Gas permeability (Nitrogen) was determined using the principle of fluid compressibility based on Darcy's law. The permeability measurement was performed by placing the clean samples in a special Hassler (core holder) and circulating a flow of nitrogen (N2) through the sample to a constant rate. The selector valve connects the system to a specific rotameter that depends on the gas flow with which it works. It is necessary to start with the
Figure 3.4. Gas permeameter.
highest flow rotameter “2,000 cc”, but if the reading is too low, the rotameter selector valve should be changed to the medium flow valve “200 cc”, and in case the flow rate readings are still low, the rotameter selector valve should be changed back to the “20 cc” minor flow position.
Based on the parameters recorded by the equipment: flow rate (Q), differential pressure across the sample (ΔP) and temperature (T) together with the gas viscosity (µs) calculated from temperature, atmospheric pressure ( Pa), area section (Am) and the length of the sample or plug, the gas permeability (Ka) was calculated (Equation 5).
(5)
Where:
Pa = atmospheric pressure (psi)
µs = Viscosity of the gas (cp)
Q = Flow rate (cc/s)
SL = Sample length (cm)
P1 -P2 = ΔP = Differential pressure (psi)
Am =Area section (cm2)
4.-Customized Mercury Injection Capillary Pressure Procedure.
4.1.-Sample Preparation:
This workflow was based on a “customized” procedure to generate the mercury injection capillary pressure curves. A series of activities were performed in a laboratory with necessary equipment and conditions for this purpose:
To perform capillary pressure tests by mercury injection, each of the ten plugs was divided into three equal parts. This, in order to use three different capillary pressure methods, so that these measurements could be directly compared. For this, a diamond edge disc cutter was used, thus obtaining thirty samples in total in the form of discs (Later, one of the plugs (three Subsamples) was discarded as a result of handling, so finally 27 samples were analyzed), with a diameter of 1 inch (2,54 cm) and 1 cm long. (Figure 4.1).
Figure 4.1. Sub-samples obtained for capillary pressure analysis by mercury injection.
Digital scale (0.0001 g appreciation) was used to weigh each dry sample. Subsequently, grain volume and density were determined using the gas expansion method at constant temperature for each Sub-sample (1-inch x 1 cm). It can be said that it is quite similar to that previously performed on the original plugs. In this case, a modified Frank Jones porosimeter with digital pressure meters was used. The equipment was connected to a computer which takes the information directly from the porosimeter and processes it. This analysis also helped to validate the porosity measurements previously done.
Once total volume and grain volume of each sample was determined, the pore volume was calculated
(Equation 6), this was necessary at the time to determine the type of penetrometer to use for each sample, since each one requires a different mercury injection volume.
Vg = Vtotal -Vp (6)
Where:
Vg = Grain volume (cc)
Vtotal = Total volume (cc)
Vp = Pore volume (cc)
4.2.-Capillary pressure Methods:
Once the Sub-samples were prepared, the mercury capillary pressure measurements were made, using the following three methods:
a) Mercury manual pump up to 2,000 psi.
b) Digital mercury porosimeter up to 2,000 psi.
c) High pressure mercury digital porosimeter up to 60,000 psi.
a) Mercury manual pump up to 2000 psi.
To obtain the capillary pressure curves by manual mercury injection pump at a maximum pressure of 2,000 PSI (Figure 4.2), it is necessary to previously weigh, clean and dry the samples before placing them in the sample holder. After that, vacuum was made for approximately thirty minutes. The nitrogen cylinder was opened and the pressures were increased in intervals of 3 psi until reaching 30 psi, after that the pressure in-crease was in of 20 psi intervals until reaching 100 psi, in the 100 psi the increase in pressures was at intervals of 100 psi until reaching 500 psi, and at 500 psi the increase in pressure was again varied to 250 psi until reaching 2,000 psi which is the maximum pressure required for this analysis. Finally, the in-crease in the pore volume injected with mercury was plotted as a function of the injection pressure to obtain the capillary pressure curve.
Figure 4.2. Manual mercury pump used for capillary pressure up to 2,000 psi.
Figure 4.3. High Pressure Mercury injection equipment.
b) Digital mercury porosimeter up to 2,000 psi.
Using the digital mercury porosimeter and at a maximum pressure of 2,000 PSI, consists on a similar procedure used to perform the tests at 60,000 psi, with the difference that the pressure intervals increase is modified and the equipment was set to reach a maximum injection of 2,000 psi as the highest pressure port. This test and the one performed with the mercury manual pump were done at a maximum pressure of 2,000 psi as this is the maximum pressure range at which most of the mercury injection capillary pressure tests are commonly found.
c) Digital High-pressure mercury porosimeter up to 60,000 psi.
Capillary pressure was placed inside which is a glass or capillary tube through which mercury flows to the sample. It should be noted that for this measurement the Lab team made pressure increases at intervals of sixty seconds, until reaching the maximum pressure possible for these ports, which is 32 psi, once the run at the low-pressure port was completed, the penetrometer was extracted.
Once the weight of the penetrometer with the sample and mercury is known, the values are entered in the equipment, and the penetrometer is assembled in the high pressure port to continue with the second part of the analysis, in this port the pressures go from 28 psi up to 60,000 psi, then the equipment performs the same procedure as in the previous stage, making vacuum and increasing the pressure at intervals of sixty seconds until reaching 60,000 psi. Figure 4.3.
5.-Comparative Analysis. Porosity, Permeability, Rock Typing, Water Saturation & Hydrocarbon Column.
5.1.-Porosity.
Primarily, Porosity was determined using the constant temperature gas expansion method. As seen, porosities vary between 3.8% and 15%, being very similar in rocks of the same type. In decreasing order we find samples (Plugs) E2 and E1 with an average porosity of 14.8%, samples (Plugs) D2 and D1 with an average porosity of 10%, samples
tests by mercury injection using the digital mercury porosimeter at a maximum pressure of 60,000 PSI were based on the volume of grains obtained for each sample where it was necessary to select the appropriate penetrometer for each one. Once the penetrometer was selected, the sample (Plugs) C2 and C1 with an average porosity of 7.85%, samples (Plugs) B1 and A2 with an average porosity of 5.5% and finally, the sample with the lowest porosity is sample (Plug) A1 with 3.8%. Figure 5.1. Much more detail is shown in figure 5.2, Where Sub-samples show consistent difference regarding the mercury injection method employed. As seen, highest porosity values were reported mostly when High Pressure Mecury Injection system was used.
Figure 5.1. Porosity values for each sample (Plug). Boyle Gas Porosimeter. Plugs.
Figure 5.2. Porosity values for each Sub-sample (Discs) by Mercury Injection Method.
5.2.-Permeability.
Permeability values were determined using the Darcy's law principle of fluid compressibility. As seen, the permeabilities vary between 0.004 mD and 119 mD. In decreasing order we find samples (Plugs) E2 and E1 with an average permeability of 116.5 mD, samples (Plugs) D2 and D1 with an average permeability of 27.3 mD, samples (Plugs) C1 and C2 with an average permeability of 1.56 mD, samples (Plugs) A2 and B1 with an average permeability of 0.026 and finally, sample (Plug) A1 with the lowest permeability of 0.004 mD. Figure 5.3.
Figure 5.3. Gas Permeability values for each sample. Plugs.
5.3.-Rock typing.
With the information obtained from porosity to Helium and gas permeability tests, the first approximation for rock typing was determined based on pore throat size, using the Winland Plot and equation 1.
The first approach to determine the existing rock types was made using the Winland R35 plot, as shown in figure 5.4. It shows the distribution of porosity and permeability of samples, separated by Iso-lines which represent equal values of pore throat radius (R35) and rocks of similar quality (equal pore throat radius).
In order to validate the dominant pore throat radius that best fits the porosity, permeability and capillary pressure data, several plots were prepared. Using this type of plots, the correlation between the pore throat radius calculated from the capillary pressure curves (Equation 7) and those calculated from the Winland R35 (Equation 1) from routine porosity and permeability data analysis, can be observed.
(7)
Where:
r= pore throat radius expressed in microns (µ)
C= constant
γ = interfacial tension (Dyne/cm)
θ = contact angle wetting phase and rock surface (º)
PClab= Capillary pressure from laboratory (psi)
The distribution of the porosity and permeability data generated in conventional analyses performed on selected plugs from this investigation gives an idea of existing Rock Types (Petrofacies). Diagonal lines represent equal values of pore throat radius (R35), data plotted along these lines represent rocks of similar quality, which are grouped into four families representing different rock types (Petrofacies). The rock types were determined according to the Winland plot and are delimited by pre-established ranges of pore throat radius, which in this case corresponds to a mercury saturation of 35% (R35) (rock type mega> 10 microns , macro type between 2 and 10 microns, meso rock type between 0.5 and 2 microns, micro rock type between 0.5 and 0.1 and nano rock type <0.1 microns).
To determine dominant pore throat radius, 27 samples were tested through different mercury injection capillary pressure methods. These samples were obtained from subdividing the nine available plugs in 3 equal parts (as mentioned in sample preparation previous section) and separated into three groups (a, b & c); the first group underwent mercury injection capillary pressure at a maximum pressure of 2,000 psi using manual mercury pump, the second group underwent the same tests at a maximum pressure of 2,000 psi, but using digital mercury apparatus and finally the third group underwent high pressure mercury injection (60,000 psi) using the same digital equipment as for the second group. Figures 5.5, 5.6, 5.7 and 5.8 show the results of these tests.
Figure 5.4. Petrofacies. Pore Throat Radius Lines & K/PHI iso-lines using Winland R35.
Figure 5.5. Capillary Pressure Curves @ 2,000 psi, using a manual mercury pump.
Figure 5.6. Capillary Pressure Curves @ 2,000 psi, using digital equipment.
Figure 5.8. All Capillary Pressure Curves: @ 60,000 psi Digital, @ 2,000 psi Digital, @ 2000 PSI Manual.
Figure 5.7. Capillary Pressure Curves @ 60,000 psi, using digital mercury porosimeter.
Another way to Validate the Rock Type model to determine pore throat radius that best fits the capillary pressure, porosity and permeability data, would be using the following “One to one Plots”. These correlation charts were constructed by plotting the Pore throat radii data obtained from the laboratory capillary pressure methods and those calculated from Winland equation as shown in figure 5.9.As seen, correlations between the pore throat radius calculated from each capillary pressure test, and the ones calculated from Winland R35 equation, showed the best fit on the 45° line. Based on these results, it was determined that dominant pore throat size corresponds to 35% mercury saturation.
Figure 5.9. Pore Throat Radius Correlation between Capillary Pressures and those from Winland equation.
It is important to highlight that the capillary pressure tests performed by mercury injection on the 27 Subsamples allowed defining the Rock Type Model, based on the mercury saturation of 35% (R35), which in turn, was compared with the rock types defined from poro-perm Winland Chart. (Figure 5.4)
Regarding Entry Pressures it was observed from laboratory tests that Sub-samples E1a/E1b/E1c and E2a/E2b/E2c correspond to the upper zone of macroporous rock type, close to the limit between the macro and mega zone, with inlet pressures going from 5 to 10 psi, representing the samples with the best flow and storage capacity from available sample set. Sub-sample sets labeled as D1a/D1b/D1c and D2a/D2b/D2c also correspond to a macro rock type category, but with a lower flow and storage capacity than the previous sample set, with an average inlet pressure range of approximately 15 to 20 psi. Subsample sets C1a/C1b/C1c and C2a/C2b/C2c correspond to meso rock types, with inlet pressures ranging from 40 to 80 psi. Sub-sample sets B1a/B1b/B1c andA2a/A2b/A2c correspond to micro rock type category, with a wide range of inlet pressures of approximately 200 and 400 psi. Finally, Sub-samples A1a/A1b and A1c represent a nanoporous rock, with inlet pressures that, depending on the used method, could range from 300 psi to 700 psi, but due to its lower porosity, it represents the sample with the worst storage capacity of available Sub-samples. Additionally, incremental mercury saturation vs pore throat radius plot was built as an accurate option to determine the dominant pore throat size for each rock type. Figure 5.10.
Figure 5.10. Incremental mercury saturation plot from the three Capillary Pressure methods. Showing acceptable consistency in Rock Typing identification.
Figure 5.11. Capillary Pressure Curves @ 60,000 psi Digital Vs. @ 2,000 psi Digital Vs. @ 2,000 psi Manual; showing that the highest difference between pore throat sizes of same Rock Types (Black arrows) is not greater than 10%.
Figure 5.12. Variation of irreducible water saturation (Swirr) by rock type, from different measurements.
5.4.-Hydrocarbon Column Height & Vertical Distance to Oil Water Contact from Capillary Pressure tests.
Height of the hydrocarbon column is a function of the rock type and the buoyancy pressure, which is related to the difference in density (ΔP) between the hydrocarbon (po, pg) and
Finally, it was observed that, regarding Rock Type identification, there was certain consistency from methods used, always being the high-pressure mercury injection method (HPMI), the most optimistic one in terms of dominant pore throat size. The highest difference of pore throat size range resulted from the three methods per rock type was between 1% and 10%. Figure 5.11.
5.3 Irreducible Water Saturation Analysis from Capillary Pressure tests.
After the three methods of mercury injection capillary pressure were performed, it was observed some variation of irreducible water saturation (Swirr) by rock type when using the manual pump & the digital apparatus with both settings: 2,000 psi and 60,000 psi. Figure 5.12.
Much more detail can be seen in figure 5.13 regarding how variation in irreducible water saturation in-creases as rock quality decreases. Additionally, as seen in figure 5.14, macro and meso rock types show the minimum Swirr variation, going from 3.4% to 4.5%, while in micro and nano rock types the maximum variation of Swirr was higher, showing values between 8.7% and 11.6%. Nano rock Type was not considered as reservoir rock in this study as it presented high Swirr values > 80%. It is important to know that a capillary pressure value of 1,000 psi is taken as the maximum equivalent height of the hydrocarbon column to determine experimental Swirr.
Figure 5.13. Variation of irreducible water saturation (Swirr) by rock type, observed in Sub-samples.
Figure 5.14. Variations of Swirr for each sample from different Sub-sample measurements.
the water (pw) . This force is used to overcome the capillary barrier (capillary pressure) that opposes the penetration of a non-wetting fluid. At the top of a hydrocarbon column of height h, buoyancy induces a pressure difference from water (ΔP=PO-PW) that is easy to calculate using the fluid balance formula (equation 8).
ΔP= Δp•g•h (8)
Where:
ΔP = Hydrocarbon pressure difference with respect to water (psi)
Δp= density difference of hydrocarbon and water (g/cm3)
g= Gravitational acceleration (cm2/s)
h= Hydrocarbon Column Height (feet)
Continuing with the procedure to determine hydrocarbon column height by rock type, air-mercury laboratory system was used, assuming water-oil reservoir system, as the water-wetting phase, and similar reservoir oil of intermediate characteristics with a density of 0.8 g/cm using the following equation (9):
(9)
Where:
h = height of the hydrocarbon column (feet)
Pc field = Capillary pressure at reservoir conditions (psi)
Dens wetting phase = Density of the wetting fluid phase (g/cc)
Dens Non-wetting phase = Density of the Nonwetting fluid phase (g/cc)
The density of water (wetting phase) and oil (nonwetting phase) are known parameters, so the only unknown variable was the capillary pressure under reservoir conditions, which was calculated using the following equation (10):
(10)
Where:
Pc lab = Capillary pressure taken as a reference to estimate Swirr (psi)
IFT w-o = Interfacial tension of the system in the reservoir (dyne/cm)
IFT lab = Interfacial tension of the system used in the laboratory expressed (dyne/cm)
(θ) w-o = Water-oil contact angle at reservoir conditions (°)
(θ) lab = Contact angle of the system used in the laboratory (°)
Hydrocarbon column height was determined using (equation 9) and considering, as explained in Swirr section, a value of 1,000 psi for equivalent height over Free Water Level (FWL) which resulted in about 700 feet. It should be noted that a contact angle of 50° was used from existing wettability tests, where it was concluded from previous core analyses that these rocks were “water wet”. Moreover, average hydrocarbon density about 0.8 g/cc from Laboratory reports was used. Nano-porous rock type was discarded, as it was considered Non-reservoir rock. At the same time, it was possible to determine the “Vertical Distance to Oil Water Contact” by using the “Flat part” of capillary pressure curves as “oil water contact” with the help of the “Displacement Pressure Value”. This vertical distance, in some sense, represents a sort of a hydrocarbon column as well, but measured from the highest reference position to the hydrocarbon-water contact. As mentioned before, this case reproduces an oil-water system, assuming the same reservoir characteristics of those from where core was taken. Figure 5.15 shows how vertical distance to Oil-Water Contact decreases as the quality of the rock decreases. Besides, when quality of the rock decreases, variation of the vertical distance to O-W contact is greater in each Rock Subsample from the same original plug.
Figure 5.15. Variation in vertical distance to Oil-Water Contact by Sub-sample in feet (Ft.).
Discussion
• Core samples correspond to consolidated sandstones of medium to fine grain, with a range of grain densities between 2.62 and 2.68 g/cm3.
• Overall, Porosity ranges between 3.8% and 15% when all samples were measured with Boyle gas porosimeter; while using mercury injection methods the Porosity range was from 2.8% to 14.8% and Porosity difference for same Rock Type Sub-samples ranges between 0.5 to 2 %.
• The permeability of the samples was determined using Darcy's law and the range of permeabilities obtained for all available samples varies between 0.004 and 119 mD.
• From Core Analyses, four petrofacies or rock quality categories were distinguished: (macro-meso-micro & nano) with a range of pore throat radii that varies between 0.06 and 8,35 microns. These Categories were determined using incremental mercury saturation charts, which had the highest certainty and accuracy when compared with the "One to One" plots.
• Winland R35 was the most suitable model to determine Rock Types.
• Swirr variations were found in measurements made with both digital devices and the manual mercury pump, ranging from 3.4% to 4.5% for macro and meso porous rocks and approximately from 8.7% to 11.6%. in micro porous rocks.
• In the same way, variations of "Vertical Distance to Oil-Water Contact" were observed in the measurements made with both digital devices and the manual mercury pump, which ranged from 0.2% to 0.6% for macro rocks; 2.2% to 5.8% for meso-porous rocks and approximately 5.8% to 15.4% in micro-porous rock types.
• In general, capillary pressure measurements on both, the digital equipment and the manual mercury pump, tend to generate greater variations in lower quality rock property measurements.
• It would be ideal to extend this type of studies to different reservoir nature and thus take advantage of this experience to be able to determine sensitivities in different geological environments from that of this study; where it would be possible to work with a larger number of samples, complementing additional information such as: thin sections, scanning electron microscopy, mineralogical tests, etc. in order to reduce uncertainty.
References
Barenblatt, G., Entov, V., Ryzhik, V. (1990). Theory of fluid flows through natural rocks, Kluwer Academies Publishers, Dordrecht, The Netherlands, vol. 3, pp 171-290.
Borger, H. D. (1952). Case history of Quiriquire field, Venezuela, AAPG., Bull, pp 2291-2330.
Djebbar Tiab & Erle Donaldson. (2003). Petrophysics Theory and practice of measuring reservoir rock and fluid transport properties, Second edition Gulf Professional Publishing, Oxford, pp 87 – 329.
Guéguen, Y., Palciauskas, V. (1994). Introduction to the physics of rocks, Press. Princeton, NJ, pp 17-126.
Figure 5.16. Shows the difference of Vertical Distance to O-W contact in percent from different Sub-sample measurements corresponding to each original plug, where high macro rocks have a vertical distance to OW contact variation between 0.01 and 0.1%, the low macro rock types a variation between 0.2 and 0.6%, meso rock types a variation between 2.2 and 5.8% and finally micro rock types a variation of between 5.8 and 15.4%.
Figure 5.16 Difference in percent of Vertical Distance to O-W Contact for each sample when the different Sub-sample measurements were compared.
Figure 5.17. Workflow for comparative analysis of different mercury injection capillary pressure methods.
Hartmann & Beaumont (1999). Chapter 9, predicting reservoir system quality and performance in exploring for oil and gas traps, E. Beaumont & N. Foster, eds., Petroleum Geology, Fredericksburg, Texas
Hurtado, J. (2000). Metodología de la investigación holística. Sypal, Caracas, Venezuela.
Mavko, G., Mukerji, T., Dvorkin, J. (2003). The rock physics handbook. Tools for seismic analysis of porous media, Cambridge University Press, United Kingdom.
Mendoza, E. (1998). Introducción a la física de rocas, USB Editorial Equinoccio, Caracas, Venezuela, pp 37-64.
Panesso, R. (1999). Caracterización petrofísica integrada yacimientos LCV-9, SM-601, arenas L2M, R3M campos La Ceibita – Soto Norte, PDVSA, Anaco, Venezuela, pp. 87-88.
Pittman, E. (1992). Relationship of porosity and permeability to various parameters derived from mercury injection capillary pressure curves for sandstone, AAPG Bulletin
Pirson, S. (1965). Ingeniería de yacimientos petrolíferos, Ediciones Omega S.A., Barcelona, pp 63 – 87.
Rzhevsky, V. & Novik G. (1971). The physics of rocks, Mir Publishers, Moscow, pp 465–489.
Van Brakel, J. Modry, S., Svata, M. (1981). Mercury porosimetry: state of art, Powder Technology, pp.1-12.
The author(s) declare(s) that she/he/they has/have no conflict of interest related to hers/his/their publication(s), furthermore, the research reported in the article was carried out following ethical standards, likewise, the data used in the studies can be requested from the author(s), in the same way, all authors have contributed equally to this work, finally, we have read and understood the Declaration of Ethics and Malpractices.