SPE 78538 Characterization of Tar From a Carbonate Reservoir in Saudi Arabia: Physical Aspects

Harouaka A. S, B. Mtawaa and W. A. Nofal, SPE, KFUPM/RI

Copyright 2002, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 10th Abu Dhabi International Petroleum Exhibition and Conference. 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, as presented, have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material, as presented, does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or members. Papers presented at SPE meetings are subject to publication review by Editorial Committees of the Society of Petroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the S ociety 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 where and by whom the paper was presented. Write Librarian, SPE, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.

Introduction The present paper discusses detailed physical characterization of several extracted and RFT bottom hole tar samples obtained from a carbonate reservoir in Saudi Arabia. The chemical aspect has already been presented elsewhere [1]. Tar is defined as extra heavy oil with a gravity ranging between 29 and 9API albeit the distinction between heavy oil and tar is rather shady. Tar mat is present in abundance in the Middle East, Africa and elsewhere. In recent publications Kaufman et al. [2] mentionned the presence of a tar zone at the water/oil contact in Burgan field. This has been known for a long time. However, to ascertain lateral and vertical delineation of the tar zones appear to be still unresolved. Thick tar zones identified through visual observation and Latroscan analyses of weathered core samples were reported in the Raudhatain field [3]. A tar mat is generally a thick column laying between an aquifer underneath and a much lighter oil reservoir above. This peculiar location poses a host of challenging problems for an efficient management of the lighter oil and ultimately for a proper management of the tar zone itself [4]. In this study, density of tar was measured with a digital Anton Parr densiometer having a maximum temperature range of 300 F and a pressure limit of 6000 psi. Based on these measurements, specific gravity and API gravity were determined. A rolling ball viscometer was used to measure tars viscosity at elevated pressures and various temperatures. The effect of visbreaking or permanent viscosity reduction due to thermal alteration has also been examined.

Abstract

A detailed physical characterization of tar from a carbonate reservoir in Saudi Arabia was made to evaluate its mobility and ways of establishing contact between the lighter oil and its aquifer. Density and viscosity measurements were carried out on several tar samples, under simulated reservoir conditions of pressure and temperature. Other physical parameters such as simulated distillation, pour point and penetration index were also experimentally determined. Tar physical properties were found to vary with depth and area within the same field. The obtained experimental results showed a gradual increase in density and viscosity from the tar/oil contact towards the tar/water contact. This increase was much more pronounced in the neighborhood of the tar/water contact. Density and viscosity of tar diluted with toluene were in excellent agreement with those of pure tar. The density of non preserved tar varied between 0.956 g/cc at 200F and 1.008 g/cc at 76F while that of preserved tar varied between 0.944 g/cc at 200F and 0. 991 g/cc at 76F. The tar samples analyzed appear to behave as Newtonian fluids.

[Harouaka A. S, B. Mtawaa and W. A. Nofal]

The pour point, which measures the lowest temperature at which tar is observed to flow when cooled and examined under prescribed conditions, was determined following ASTM standard procedure: ASTM D97-66. Tar does not change from its semisolid state to the liquid state at any definite temperature, but it gradually becomes softer as the temperature rises. For this reason, the determination of the softening point has been carried out following the standard procedure described in IP 58/83 [5]. The penetration index and flash point have also been determined by standard procedures. Distillation of tar has been conducted using the procedure described in ASTM D402-73. The sample is distilled at a controlled rate to a temperature of 680 F, and the volumes of distillate obtained at the specified temperatures are measured. The residue and the distillates can then be tested as required. Experimental procedures and background Modifications of existing oil characterization methods were necessary to facilitate experimental determination of the physical properties of tar. One typical modification is to dilute tar with a suitable organic solvent prior to testing. Measurements are then made using different dilution ratios. The Physical properties of pure tar are determined by extrapolation to zero dilution ratio. Chirinos M. L. et al. [6] studied the effect of temperature (ambiant to about 194F) and percentage of added diluent on the density and viscosity of nearly 100 crude oils obtained from the Venezuelan Orinico Oil Belt. An Atomix mixer, a Fisher specific gravity hydrometer and a Haak Rotovisco RV3 viscometer were used to perform this study. These crudes and their mixtures with diluents behave as Newtonian but not as plastic fluids. Expressions to estimate the density and kinematic viscosity of these oils and their mixtures with diluents at any temperature were also presented. Tobey et al. [7] reported that when core plugs obtained from tar zones near the oil-water contact in the Arab-D formation in Saudi Arabia were extracted with a series of solvents (naphtha, toluene, and methylene chloride or trichloroethane) for at least 72 hours; their porosity improved and their permeability showed only a

small change after the initial cleaning with naphtha. It was hypothesized that macro pores were easily accessed by the solvents and hence cleaned right after the naphtha and toluene extractions. The micro pores and the smaller pore-throats remained completely or partially plugged by tar that differ from one location to another within the same reservoir. Puttagunta V. R. et al. [8] developed a generalized viscosity correlation to predict, with good accuracy, the viscosity of any Alberta bitumen's or heavy oils, which are known to vary widely from deposit to deposit, over a wide range of pressure and temperature, usually encountered in in-situ recovery methods. This correlation requires only one viscosity measurement at 86F and atmospheric pressure. Erno B. P. et al. [9] measured the viscosity of oil samples extracted at 194F using a Beckman 28/M heated centrifuge from preserved cores obtained from various depths and different wells in the Lower Cretaceous Clear water "B" heavy oil reservoir in the Caribou lake area. They observed that the viscosity increases consistently with depth and correlates well with structural elevation. Simulated distillation data showed that the viscosity variations are due to compositional differences. Samples from other heavy oil reservoirs such as McMurray, the Wabiskaw and the Waseca sand at Pikes Peak in west central Saskatchewan were found to exhibit similar trends. Svrcek and Mehrotra [10] measured the viscosity, density and solubility of CO2 , CH4 and N2 in bitumen samples obtained by toluene extraction from the Athabasca tar-sands. These measurements were carriedout over a temperature range of 77 to 212F and a pressure up to 1450 psi. They observed that CO2 has the highest solubility in Bitumen and reduces its viscosity drastically. On the other hand N2 has quite low solubility in the bitumen and less effect on its viscosity. They also reported that dead oil viscosity decreases drastically with the increase of temperature and is less affected by the pressure while that of live oil decreases with pressure and temperature. Following is a brief description of the experimental procedures and conditions under which the physical characterization of tar samples retrieved from the tar mat zone, in a carbonate reservoir from Saudi Arabia, was determined.

[Paper Number]

[Characterization of Tar From a Carbonate reservoir in Saudi Arabia: Physical Aspects ]

Density Density measurements were carried-out using a digital Anton Paar densiometer (DMA-60). The measuring principle involves the period of oscillation of a vibrating U-shaped sample tube filled with the sample or through which the sample flows continuously. It was equipped with a Ruska mercury injection pump to charge the tar sample into the sample tube and pressurize it to the desired pressure, and a Heto constant temperature oil bath to maintain the sample tube at the desired temperature. Before being used, the densiometer was calibrated using vacuum and a standard oil sample with a density close to that of the tar samples to be analyzed. The density is calculated once the sample is introduced into the sample tube and its period of oscillation is stabilized at the desired pressure and temperature. 1. Extracted tar To measure the density of extracted tar, it must be injected into the densiometer's sample tube as explained earlier. However, at atmospheric pressure and different temperatures, it does not have the necessary mobility to be injected. This is why it has been diluted in toluene. The density of extracted tar-toluene homogeneous mixtures was then measured at different toluene concentrations (weight %). The density of the extracted tar is simply taken as the value extrapolated to zero toluene concentration. 2. Dead RFT tar A tar sample taken out of the RFT bottom hole sample transfer cell and saved in a container was left open to atmosphere for few days until the associated gas was liberated. Its density was then measured following the same procedure used for the extracted tar. 3. Live RFT tar The cylinder containing the live RFT tar was connected to the inlet port of the densiometer sample tube while another cylinder was connected to its outlet port to let the tar sample flow continuously from the source cell to the second cylinder via the sample tube. The sample cell was then heated to the desired temperature using a temperature regulator and display along with heating tape.

The sample was subsequently pressurized to the desired pressure using a Ruska mercury injection pump. A second pump was used to maintain flow through the densiometer sample tube. The density of the undisturbed RFT tar sample was then determined at different pressures and temperatures. Viscosity Viscosity measurements of extracted and dead RFT tar were carried out at ambient pressure and different temperatures using a highly sensitive torque measuring system made of a plate-cone type Contraves Low Shear-30 viscometer and a Haak-M regulated constant temperature oil bath. Tar samples (0.50 cc) were heated to the desired temperature and poured into the measuring cup of the viscometer. Similarly, the viscosity of tar/toluene homogeneous mixtures were measured at various toluene concentrations (volume %). Results were extrapolated to zero toluene concentration to find the pure tar viscosity. Viscosity measurements of extracted and live RFT tar were also obtained at various high pressures and temperatures using an ROP rolling-ball viscometer DT14001, along with a Haak-M constant-temperature oil bath and a Ruska mercury injection pump. An important feature of this viscometer is the utilization of a magnetic field for detecting the rolling ball. Accurate ball location can be detected even when the ball is completely coated with the high viscosity tar. The viscometer was calibrated using a standard oil sample with a viscosity of 27 cp at 210F and 477 cp at 100F. Viscosity values were calculated using the following relation: = A *(?b - ?t ) * t + B Where: is the tar viscosity (cp), A is the slope of the calibration curve at the desired angle, ?b and ?t are the densities of the steel ball and tar respectively, t is the measured time (sec.) and B is the intercept of the calibration curve at the same angle.

[Harouaka A. S, B. Mtawaa and W. A. Nofal]

Other Physical Parameters

Simulated distillation There are several distillation methods covering the boiling range distribution of petroleum products. The extended ASTM D2887-73 (standard test methods for boiling range distribution of petroleum fractions by gas chromatography) was followed for the distillation of extracted and RFT tars. This method is mainly applicable to petroleum products and fractions with a final boiling point of 1000 F or less but extended to cover a higher boiling point range. The apparatus used included a Hewlet-Packard level 4,5880A Gas Chromatograph, a flame ionization detector (FID), a liquid nitrogen cooling system, and a cartridge tape unit. The column used was a 1/8 x 20 inch stainless steel, 10% UC-W982 on 80/100 mesh chromosorb P-AW. The results obtained are shown in Table 1 below. The initial boiling point (IBP) of extracted tar (552 F) was almost twice that of RFT tar (228F). However, at the end, their yields (around 1000F) are almost equal 10-11% of the total sample. This indicates that their compositions are mainly the same. The difference at low temperature may be due to the fact that the extracted tar has lost its lighter components keeping in mind the cores were exposed to atmosphere for a long period of time. Table 1. Distillation data of extracted and RFT tars.
Extracted tar % OFF IBP* 1 2 3 4 5 6 7 8 9 10 11 RFT tar Temperature F 552 228 590 297 639 377 675 444 702 518 723 588 752 662 790 736 828 811 871 892 916 999 975
equal

1. Penetration index The penetration index is the consistency of a bituminous material expressed as the distance in tenths of a millimeter a standard needle vertically penetrates a sample of the material under known conditions of loading, time, and temperature. Penetration tests were carried out for the extracted and RFT tar samples at 77oF according to ASTM D5-83 using the penetration apparatus (Penetrometro-697) and its accessories (sample containers, timer, a water bath, needles, and weights). It was found that penetration index of the extracted and RFT tar samples were 67 and 196 units, respectively as shown in Table 2 below. Table 2. Penetration index of extracted and RFT tar.

Tar sample

Penetration*

Extracted Tar RFT tar

67 196

*One penetration unit = 0.1 mm, Time = 5 sec, needle weight with spindle = 50 gms.

2. Flash point The flash point is defined as the lowest temperature, corrected to a barometric pressure of 760 mmHg, at which application of a test flame causes the vapor of the sample to ignite under specified testing conditions. A Gallenkamp-Autoflash apparatus, a form of the Pensky-Martens closed tester, was utilized to determine the flash point of the extracted and RFT tar samples according to ASTM D93-80. 3. Pour point The pour point is the lowest temperature, expressed as a multiple of 5F at which the oil/tar is observed to flow when cooled and examined under prescribed conditions. The pour point tests were conducted according to ASTM D 97 using a test apparatus assembled in house (Brookfield Ex 200 thermostatic bath, a test jar, and two certified thermometers). The obtained pour points were 163 oF for the extracted tar and 89 oF for the RFT tar.

*Initial boiling point: The point at which a cumulative area count

to 0.5% of the total area under the chromatogram is achieved.

[Paper Number]

[Characterization of Tar From a Carbonate reservoir in Saudi Arabia: Physical Aspects ]

Results and discussions This section summarizes the main objective of this study, the physical properties determination of several tar samples including extracted and RFT bottomhole samples. Extracted Tars: Various extracted tar/toluene mixtures (20, 40, 60, 80 and 100% by weight) were prepared manually. Once homogenization was completed, densities were measured at atmospheric pressure and temperatures ranging from ambient to 210oF to include the reservoir temperature, estimated at 190 oF. At each temperature, the density of the mixtures behaved as expected, giving a linear variation with solvent concentration as shown in Figure 1. Each straight line was extrapolated to zero toluene concentration to determine the density of extracted tar at that particular temperature. It has been determined that within the range of temperatures considered in these experiments, the densities of tar/toluene mixtures follow an equation of the form: m = Wd *( d - t ) + t Where, m is the density of the mixture in g/cc, d is the density of the diluent (toluene) in g/cc, t is the density of tar in g/cc and Wd is the weight fraction of diluent. Experiments to evaluate the effect of pressure on the density of extracted tar were performed at three different temperatures (220, 230, and 240oF). They were generated maintaining a constant temperature while decreasing the pressure from 3500 to 1000 psig. The results obtained are shown in Figure 2. As in the case of extracted tar, densities of RFT tar/toluene mixtures were measured at atmospheric pressure and various temperatures (102, 120, 160, 180, 190 and 200o F). Here also there is a linear relationship between densities and solvent concentrations as indicated by the straight lines shown in Figure 3. Extrapolation to zero toluene concentration yields the density of RFT tar at the desired temperature. When these densities are plotted against

temperature along with the densities of extracted tar as shown in Figure 4 one can see, rather cle arly, that at temperatures less than 180o F the density of the RFT tar is lower than that of the extracted tar. However, beyond 180o F the two curves coincide. This is expected since at higher temperatures RFT tar loses its lighter components and has the tendency to become similar in composition to the extracted tar. Experiments to determine the effect of pressure on the density of RFT tar were also performed at three temperatures (160, 190, and 220o F). The data were taken while decreasing the pressure from 3500 to 1000 psig and maintaining the temperature constant. The density increases linearly, at constant temperature, with increasing pressure within the specified pressure range as shown in Figure 5. Viscosity Measurements Extracted Tars: The viscosity of extracted tar was measured at different temperatures ranging from 160 to 270 oF and atmospheric pressure. The results obtained are shown in Figure 6. The log-log plot of shear rate vs. shear stress depicted in Figure 7 shows a straight line with a slope of one indicating the extracted tar behaves as a Newtonian fluid with some plastic behavior as the yield (intercept) is nonzero. As in the case of extracted tar the viscosity of RFT tar was also measured at atmospheric pressure and the same temperatures as the extracted tar. The results obtained are shown in Figure 8. A comparison between Figures 6 and 8 indicates clearly that the RFT tar has a lower viscosity than the extracted tar. The RFT tar is also believed to behave like a Newtonian fluid as indicated in the log-log plot of shear rate vs. shear stress (Figure 9). The slope shown in Figure 9 is slightly less than one indicating a pseudoplastic behavior. The yield on the other hand is essentially zero. One should keep in mind that the extracted tar is implicitly filtered during the extraction process as most solid particles are separated from the recovered tar. The RFT tar, on the other hand, has not been filtered. Based on this analysis, it is believed that the RFT tar behaves like a Newtonian fluid under reservoir conditions. The standard way to plot viscosity vs. temperature is on ASTM viscosity vs. temperature charts for liquid petroleum products, which for Newtonian

[Harouaka A. S, B. Mtawaa and W. A. Nofal]

fluids generally yield a straight-line relationship between kinematic viscosity and temperature. The straight line is based on a mathematical expression that allows the calculation of viscosity, in centistokes (cSt) at any temperature T in degree Rankin, as follows: Log log (? + 0.7) = A - B log T. Where, ? is the kinematic viscosity in centistokes, A is the intercept and B the slope. Various tar toluene mixtures, ranging from nine to 32 % by weight of toluene, were prepared manually for both extracted and RFT samples. Once homogenization was achieved, viscosity measurements of tar/toluene samples were performed at 180 and 190oF and atmospheric pressure. The results are shown in Figures 10 and 11 respectively. At each temperature the viscosity of the tar/toluene mixtures exhibited a linear variation with solvent concentration as shown in Figures 10 and 11. Each straight line was then extrapolated to zero toluene concentration to find the viscosity of extracted and RFT tar samples at that particular temperature. At any temperature the viscosity of tar/toluene mixtures appear to follow an equation of the form: Y = A-BC. Where, Y = log log (? + 0.7), A is the straight-line intercept, B is the slope of the straight line and C is the solvent concentration in volume %. The straight-line equation, at extracted tar/toluene mixture viscosity is: Y = 0.70651-0.015519*C. The straight-line equations for RFT tar/toluene mixture viscosity at 180 and 190o F are respectively: Y = 0.62468 - 0.01474 * C. Y = 0.59805 - 0.015742 * C. Figure 12 shows a plot of kinematic viscosity vs. solvent concentration for both extracted and RFT tar/toluene mixtures at 180 and 190o F, and atmospheric pressure. One may observe from Figure 12 that an 180oF, for

extrapolation to zero solvent concentration gives kinematic viscosities comparable to those for pure tar at a given temperature for both extracted and RFT tars. The viscosity of RFT tar was also measured at different temperatures (160 up to 230oF) and pressures ranging from 1000 to 3500 psig. The results are shown in Figures 13. It can be seen from Figure 13 that the viscosity of the RFT tar decreases with increasing temperature and decreasing pressure, unlike that of the dead RFT, which decreases significantly with temperature and remains more or less unchanged with pressure. CONCLUSIONS The following conclusions may be drawn from the detailed physical characterization of several extracted and live RFT bottom hole tar samples obtained from a carbonate reservoir in Saudi Arabia: 1. The physical properties of the analyzed tar samples were found to vary with depth and area within the same field. 2. A gradual increase in density and viscosity from the tar/oil contact towards the tar/water contact was observed. This increase was much more pronounced in the neighborhood of the tar/water contact. 3. At each temperature, the density of tar/toluene mixtures behaved as expected, giving a linear variation with solvent concentration. 4. Density and viscosity of tar diluted with toluene were in excellent agreement with those of pure tar. 5. The viscosity of the RFT tar sample decreases with increasing temperature and decreasing pressure. 6. The viscosity of dead RFT decreases significantly with temperature and remains more or less unchanged with pressure. 7. The extracted tar is believed to behave as a Newtonian fluid with some plastic behavior 8. The RFT tar is also believed to behave as a Newtonian fluid with a pseudoplastic behavior. ACKNOWLEDGEMENTS The authors which to acknowledge the support of the Research Institute of King Fahd University of

[Paper Number]

[Characterization of Tar From a Carbonate reservoir in Saudi Arabia: Physical Aspects ]

Petroleum and Minerals (KFUPM/RI). They also thank KFUPM/RI for their permission to publish this paper. The authors are grateful to A. A. Habelreeh for his valuable contribution to the final manuscript preparation. REFERENCES 1. Harouaka A.S., H.K. Asar, A.A. Al-Arfaj, A.H. AlHusaini and W.A. Nofal (1991). Characterization of tar from a carbonate reservoir in Saudi Arabia: Part 1-Chemical aspect Paper SPE 21004 presented at the SPE International symposium on oilfield chemistry held in Anaheim, California, Febr. 20-22. 2. Kafman R.L., H. Dashti, C.S. Kabir, J.M. Pederson, M.S. Moon, R. Quttainah and H. Al-Wael (1997). Characterizing the Greater Burgan Field: Use of Geochemistry and Oil Fingerprinting Paper SPE 27803 presented at the 10th SPE Middle East Oil Technical Conference and Exhibition, Bahrain, March 15-18. 3. Nemcsok S., N.H. Morisson, A. Carruthers and S.H. Abdullah (1998). Sedimentary Interpretation of a Multilayered clastic Oil Reservoir: Impact on Development Plans For the Zubair Reservoir, Raudhatain field Paper SPE 48972 presented at the Annual Technical Conference and Exhibition, held in New Orleans, LA, Sept. 27-30. 4. Harouaka A.S. and H.K. Asar (1990). Tar mats evaluation- a resource and a nuisance First Saudi symposium on energy utilization and conservation held in Jeddah, Saudi Arabia, March 4-7. 5. Haines W.E. (1976). Extension of the US Bureau of Mines- API Scheme For the Characterization of Heavy Oils Erdoel Kohle -Ergas-Petrochem V. 29, No 9, August. 6. Chirinos, M.L., J. Gonzalez and I. Layrisse (1983). "Rheological Properties of Crude Oils from the Orinoco Oil Belt and their Mixtures with Diluents" Rev. Tech. Interp, July 1983, 103-115. 7. Tobey M. H., H.I. Halpern, G.A. Cole, J.D. Lynn, J.M. Al-Dubaisi and P.C. Sese (1993). Geochemical Study of Tar in the Uthmaniyah Reservoir Paper SPE 25609 presented at the SPE Middle East Oil Technical Conference and Exhibition, Bahrain, April 3-6. 8. Puttagunta V. R., B. Singh, E. Cooper and B. Thunder (1988). A Generalized Viscosity Correlation for Alberta Heavy Oils and Bitumens" Preprint Paper 158 presented at the fourth UNITAR/UNDP conference on heavy crude and tar sands, Edmonton, Alberta, Canada August 7-12.

9. Erno B. P., J.R. Chriest and R.C. Wilson (1991). Depth-Related Oil Viscosity Variation in Canadian Heavy Oil Reservoirs JCPT, Vo. 30, No. 3, MayJune, 36-41. 10. Svrcek W.Y. and A.K. Mehrotra (1982). Gas Solubility, Viscosity and Density Measurements for Athabasca Bitumen JCPT, July-August, 31-38.