Rojas.-thermodynamic Characterization of a PVT of Foamy Oil

SPE 69724Thermodynamic Characterization of a PVT of Foamy Oil Douglas J. Romero and Belkis Fernandez, SPE, Schlumberger, and Gonzalo Rojas,SPE, UDO Copyright 2001, Society of Petroleum Engineers Inc. This paper was prepared for presentation at the 2001 SPE International Thermal Operations and Heavy Oil Symposium held in Porlamar, Margarita Island, Venezuela, 12-14 March 2001. 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 Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of 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. Abstract This work is based on theoretical studies and/or experimental observations carried out diverse authors whom have investigated on the " foamy oil phenomenon " developing a methodology to characterize these crude oils thermodynamically, taking as bases nonconventional PVT analysis and using as research tool the application of equations of state and methods known for the determination of the equilibrium constants liquid-gas. It is presented two new correlations developed in this work for the calculation of the viscosity of heavy crude oils, which are based on values of molar fractions of liquid and gas in equilibrium. The proposed methodology was validated using conventional and nonconventional PVT data of wells located in the Orinoco Belt, Jobo Field, Morichal Area obtaining excellent results and showing that the proposed methodology is applicable to conventional and nonconventional heavy crude oils in all the possible scenes, that is to say, whether necessary information is available. Introduction When we analyse foamy crude oils under the optics of primary production mechanisms known traditionally, have not been able to explain with exactitude the production behavior in these reservoirs1; this has given basis to many people to do research with the objective of explainning the origin of this atypical behavior. To such extreme to establishment the theory of "foamy oil phenomenon ", which considers a transient state or supersaturation condition, in which take place the characteristics named " atypical " that identify to this type of heavy crude oils. It’s of extreme importance for the petroleum industry to model the thermodynamic behavior of foamy oils reservoirs, since a good characterization fluid increases the probabilities to obtain better numerical reservoir simulations of and thus to be able to consider with most exactitude the total recovery. Characterization of Foamy Oils The foamy oil phenomenon has appeared only in heavy and extra-heavy crude oils, since in these crudes the viscous forces surpass to the gravitational forces on the productive life of reservoirs, reason why this phenomenon goberns the production behavior of these reservoirs. This phenomenon to appear after that reservoir pressure reaches the bubble point pressure, from this pressure the petroleum production increases, the gas bubbles expand to displace petroleum towards wells quickly. Depending of pressure and extraction rate of wells placed in the reservoir is possible that gas bubbles be produced with the oil. With the purpose to characterize the ability that have some heavy crudes to show the foamy oil behavior and entrap gas that is released by each decrease of pressure. Nonconventional PVT analysis were developed which defer from the method used traditionally. The conceptual difference between this analysis and the method used conventionally, is that the flash and differential liberation tests are carried out without agitation of the cell2. Generally, the results obtained by means of conventional and nonconventional PVT analysis differ remarkably mainly in the values of bubble point pressure obtained by both methods for foamy oils. Frequently values of bubble point pressure smaller are obtained in nonconventional PVT analysis because occurs entrapping of gas within the oleic phase of the crude reason why it’s deduced that supersaturation phenomenon occurs, the gas bubbles released to pressure far below the bubble point pressure are dissolved and/or dispersed within the phase of petroleum in a perfect hydrodynamic equilibrium. It later on that such hydrodynamic equilibrium should be broken to obtain free gas. It had been determinate by laboratory experiences that viscosity of foamy crude oils obtained using capillary viscometers are more suitable to make reservoir numerical simulations than those obtained with rotational viscometers, in this case, strench caused by shaking the crude sample breaks the dispersion gas/oil and release bubble gas to build a free gas cap separated of the crude oil. 000147*(Rs*(γg/γo)0. ρo. The compressibility was caculated according to Sheng and Maini5. To determine formation volumetric factor of oil (FVFo) at pressures below bubble point pressure. The criteria used for the selection was the range or conditions for which these correlation’s can be used. The methodology proposed (Fig. FERNANDEZ AND G. Option 1 From data PVT (conventional or nonconventional) the equilibrium constants condition by the method proposed by Zhou3 are determined. Bg. shows two options to use depending of the case is analysed. two new correlations were developed for the calculation of the viscosity of heavy crude oils and foamy oils. with the purpose of choosing those that displayed minor percent error with respect to PVT conventional and nonconventional data. The compressibility and viscosity. ROJAS Methodology A set of correlations were selected to be used in the determination of thermodynamic properties of heavy and extra heavy crude oils. a modification of SPE 69724 Standing4 correlation is recommended: Bo=0. One to be used at pressures below or equal to bubble point pressure and the another one at pressures above bubble point pressure. both options use the same correlations. The PVT analyses used in the validation of the model correspond to following wells: • • • Well 1 (conv. whether necessary data is available from a conventional or no-conventional PVT test. B.and nonconv. These correlations are the following: At pressures below or equal bubble point pressure: µ = 0. finaly the thermodynamic properties and viscosity of the crude oils are caculated. and nonconv. Co. 2). Bare Field – Hamaca Area Well 2 (conv.2 D.25*T)1. ROMERO.5+1. Mathematical and statistical algorithms and sensitivities analyses were applied to obtain a mathematical expression based on molar fractions of gas and liquid in equilibrium and viscosities of liquid and gas values respectively. Then the crude thermodynamic properties are determined according to the following expressions: X=1/K a= R * T * ρo * 5. which are based on values of molar fractions of liquid and gas in equilibrium.).55* X*(P – Pb ) + µ ob (6) µ liq values used are reported in experimental tests as average values for each field in study. for both. Viscosity Given the situation that not exist reliable mathematical correlation for calculation viscosity of foamy crudes oil.5*(µ liqY+µgX)+0. whereas for viscosity no one of the correlations found in the bibliography yields good results since the error with respect to experimental values were superior to 98%.etc. GOR’s are calculated to pressure and temperature required using Millán6 equation developed specifically for heavy oils. which are modified to supersaturation condition (foamy oil) by Sheng and Maini method5 to determine gas and liquid molar fractions at different pressure and temperature conditions. afterwards to determine gas and liquid molar fractions in equilibrium condition applying method proposed by Zhou3. two very important properties to study the behavior of foamy oil under pressure and temperature variations. also were considered diverse methods to determine the equilibrium constants of hydrocarbons systems and several correlations published to calculate the viscosity of those crudes.6146 MWo * P Rs = X *a 1− X (1) (2) (3) Option 2 With data obtained from separator test.972+0. above and below of bubble point pressure.2). Bare Field – Hamaca Area Well 3 (conv. basically the difference between both options is in the calculation of solution GOR. Afterwards of selecting the correlations to be used. In the determination of other thermodynamic properties of fluids as Bo.) Cerro Negro Field – Morichal .175 (4) the oil density at different pressures was determined using Millán6 correlation by heavy oils. a methodology to calculate fluid thermodynamics properties was development which may be applicable in all the possible scenes. The complet set correlations used in this work are showed in Appendix A. differential and viscosity test.). this to be modifycated to foamy oil condition according to Sheng and Maini5 method . the cut parameter was the value of gravity API of the crude. Validation of Mathematic Model The validation of the selected mathematical model was made from a group of 8 conventional and nonconventional analyses PVT pertaining to reservoirs of the Orinoco Oil Belt and to the Jobo Field of the Morichal area. throw an analysis of correlations presented in texts of reservoirs engineering4. Worksheets were created to study the correlations. The criteria raised by Smith1 was considered widely for development of these new correlations. depending on the available information (Fig. These new equations reproduce with greater exactitude the viscosity behavior with respect to pressure variation.25((Y*µ liq + X*µ g)+(µ liqY*µgX)) (5) At pressures above bubble point pressure: µ = 0. that is to say. both. In validation of correlations developed and presented in this paper (4 & 5). Zhou. : “Understanding Foamy Oil Mechanisms for Heavy Oil Reservoirs During Primary Production”.4 %. As it’s displayed in Figures 8 and 9. M. Oklahoma. considering that viscosity is one of the most difficult property to predict. C.SPE 69724 • • THERMODINAMIC CHARACTERIZATION A PVT OF FOAMY OIL Well 4 (conv. Universidad de Oriente. and nonconv. Kv4 and Kv5: Constant values for each hydrocarbons system K’(p): Equilibrium constants modified for foamy oils. Mirabal. Millán. bbl/STB. (1996). y Rojas.. Rico. in comparison with those results obtained using others correlations.) Jobo Field – Morichal Area Figure 3 shows results obtained in GOR calculation for the conventional tests. SPE 36749. The correlations developed in this work for calculation viscosity of heavy oils provide good results for any heavy crude type.E. E: “Correlaciones para Crudos Pesados Venezolanos”. was obtained compressibility of foamy oil much greater than conventional heavy oils. nevertheless this hypothesis has not been verified experimentally. adim. K = Equilibrium constant. for his assistance as well as numerous practicing engineers who provided us data for the study. these results obtained are very reliable because are very close to experimental values.: “Computin and Selecting Parameters in numeriacal Modelling of Oil Field Thermal Recovery”. X = Molar fraction of gas phase in equilibrium.6-9 October 1996. only three were made using capillary viscometer and the other were made with rotational viscometer. Subscripts o = oil g = gas ob= bubble point oil b = bubble point liq = liquid Acknowledgments The authors wish to thank Schlumberger and Universidad de Oriente for permission to publish this work. The average error percentage is located between 10 and 20 %. I. Thesis. Whereas for nonconventional tests. Rs = Solution gas-oil ratio. Annual Technical Conference and Exhibitio. graphically can be observed in Figure 4. W: “The Properties of Petroleum Fluids”. 2. a much better match with experimental values of all PVT analyses used was obtained applying correlation’s developed in this job (4 & 5). Denver. FVF = Formation volume factor. .8%. psia. Kv3. all results show that these correlations have a good confiability grade. USA (1973) 5. Maini. for no-conventional PVT test of Well 1 (Fig 6). adim. :“Fluid Flow and Sand Production in Heavy Oil Reservoirs Under Solution Gas-Drive”. adim. B. avoiding the accomplishment of PVT tests. Viscosity tests in PVT analysis used for validation are different between themselves.. the average error percent was 6. FVFo results obtained. The compressibility of foamy oil crudes presumes that it’s between 5 to 10 times greater than conventional heavy oil crude. a comparison between this and other recommended correlations by others authors to calculate viscosity of foamy oil and heavy oil crudes was made. Kv2. GOR. X. in comparison with obtained by other correlations. 6.Colorado. 3. Nomenclature µ = Viscosity. The developed methodology reproduces in very acceptable degree the behavior of the fluids with respect to the variations of pressure and temperature and is applicable for conventional or foamy heavy crude oils in spite of the little existing information. Kv1. considering that viscosity is a property very difficult to determinate in heavy oils and foamy oils. adim.USA. applying the 3 developed methodology. The average relative error for the developed correlations (4 & 5) was 10. Conclusions 1.. adim. cps Y = Molar fraction of liquid phase in equilibrium. PennWell Books. scf/STB. Otero. A. G. Jimenez. J.. γ = Specific gravity. K(p): Equilibrium Constant for convencional crude oils. as it’s showed in Figure 10. Tulsa.E. (1995) 4. which reduces the total costs of a developing project.7 psia. F. References 1. Smith.5 %. in these cases the error percentage not surpass 1%. is observed a good match with respect to the experimental values. Values obtained through Option 2 of proposed methodology are named synthetic PVT. (1986) 2. Figure 5 shows compare results obtained for GOR calculation in all the possible scenes for well 1. y Hayes.. A reliable synthetic PVT can be made. M. SPE 36750. 3. McCain. We also with to thank engineers Juan Cova. 14. The calculations of isothermal compressibility was made by Sheng and Maini5 method proposed for foamy crude oils. like for all possible scenes in which the methodology proposed can be applied (Fig 7).: “ A Dynamic Model to Simulate Foamy Oil Flow in Porous Media”. R. conventional or foamy oil. 2 (3) 15-22. average error percent was 4. T= temperature. which were superior to 98%.adim. Cesar Diez and Marcelo Laprea. psc: Normal pressure. Sheng. P = Pressure. SPE 15094. G. introducing values of liquid and gas molar fractions in equilibrium determined according to proposed methodology..) Arecuna Field – Hamaca Well 5 (conv. in three analyzed cases. Special Oil & Gas Reservoirs . Huerta. SPE 69724 Modification of equilibrium constants at foamy oil state:Sheng and Maini method Solution GOR ( Rs ).0125 ∗ API − 0. 2 ∗ Ty − 1180 ∗ γ g ∗ 12 .175 γg F = Rs ∗ ( )^ 0 . ROJAS • Appendix A – Correlation Set.(psc> p> psb) (A-11) .psc) psb− psc (A-10) . 61 ∗ API P * 10 ^ 5 .2 (A-2) • OFVF (Bo): Standing & Beggs correlation at pressures below bubblepoint pressure Bo = 0. 50074 * ( Bo / Bob ) (A-5) Above bubblepoint pressure ρ o = ρ ob * e ^ ( Co * ( P − Pb )) (A-6) • Isothermal compressibility Vasquez & Beggs correlation Co = • − 1433 + 5 ∗ Rsb + 17 .031 (p > pb) (A-1) K’(p) = K(pb).4 ) ∗ 10^ (0.00091 ∗ Ty ))^1.972 + 0.2048 18.K(psb) * (p .000147 ∗ F ^1.00075 * T ) (A-4) Below bubblepoint pressure:Millán correlation ρo = ρ ob * 1 . This calculation is made to determinate equilibrium constants coeficients (Kvalues) for obtain equilibrium constants (K). FERNANDEZ AND G. B. When PVT analyses report don’t displayed results of separators test.25 ∗ T γo (A-3) • Density extra heavy oils: At bubbepoint pressure: Millán correlation ρ ob = (1 .83 Rs = 1. 6499 e ^ ( 0 . Standing correlation are following: Rs = γg ∗ (( P + 1.2353 * Pb ^ − 0 . ROMERO. Rsb ∗ ( P / Pb)^0.02483 ) e ^ ( 0 . • Extra heavy oils: Millán correlation K’(p) = K(p).5 + 1 .4 D. Standing correlation is recomended to determinate GOR at temperature differents. (A-7) Equilibrium constants: Sheng & Zhou method − Kv4  Kv 2  e    K =  Kv1 +  *  T + Kv5 P   (A-9) (A-8) Ê' (p) = K(p)+ (psb> p> pb) K(pb). Well 2 Exp. GOR results obtained for nonconventional tests 900 1000 . scf/stb GOR. 60 40 Well 1 Calc. Well 2 Calc. 60 40 Well 4 Calc Well 1 Calc Well 3 Calc Well 4 Exp. Well 4 Exp.GOR results obtained from conventional PVT tests. psia Figure 4. 100 200 300 400 500 700 Pressure . Well 3 Exp. Bo. Well 5 Calc. Well 3 Calc.value ) Define Pressure Stage of LD. Well 5 Exp. psia Figure 3. Methodology proposed diagram Figure 1. scf/stb Determination of Equilibrium Constants ( K ) in Foamy Oil State to Viscosity Test Yes Determination of Equilibrium Constants ( K ) at Pressure of LD test andTyac. CERRO NEGRO MACHETE ZUATA HAMACA Thermodynamics Properties Crude Determination of CCE Test (Co. Well 1 Exp. and Viscosity Tests Thermodynamics Properties Crude Determination of LD ρo. CCE. V. Well 1 Exp Well 3 Exp 600 800 20 20 0 0 0 0 200 400 600 800 1000 1200 1400 1600 Pressure.etc) Test (GOR. rel) To Consider This Crude in Foamy Oil State No Determination of Equilibrium Constants ( K ) at Pressures of Viscosity Test Calculate Viscosity. Show Results Figure 2.SPE 69724 THERMODINAMIC CHARACTERIZATION A PVT OF FOAMY OIL 5 Sucre Anzoátegui Monagas Guárico Define Input Data Delta Amacuro ELT OCO B ORIN N Option 2: Input Data To Use Correlations Option 1: Input Data from Laboratory and define analysis type GOR Calculation by Correlations Calculate Equilibrium Constants Coefficients( K . Validation of methodology. Orinoco Belt localization 120 120 100 100 80 80 GOR . Well 4 Calc. 95 0 200 400 600 800 1000 Pressure .PVT No-Conv 0 Calc.PVT No-Conv 1 20 Exp. cps 1500 Pressure ( psia ) 1200 3000 2500 2000 1500 1000 0.15 4000 Smith. 1. psia Figure 5.0008 0.0004 0. Well 3 Exp. GOR results for well 1 (all possible scenes) Figure 6. Well 4 Calc. Exp. bbl / STB Visc.0012 Well 1 Conv Well 1 Foamy Well 4 Conv Well 4 Foamy Well 3 Conv Well 3 Foamy 0.0016 5000 4000 Well 1 Exp. Smith.0002 500 0 0 0 0 500 1000 1500 2000 2500 3000 3500 4000 Pressure . psia Figure 9.0006 0.(2) Maini Correl.05 Exp. Results obtained for all PVT analises used in validation of new viscosity correlations 200 400 600 800 1000 4500 Pressure . Well 2 Exp. Well 5 Exp.PVT Conv 40 Well 1 Calc.PVT Sintetic No-Conv 0 0.scf/stb 80 60 Calc.PVT Conv Calc. Well 3 Calc.05 Calc. bbl / STB GOR . ROMERO.(1) 3500 3000 1. Well 1 Calc. psia 500 600 700 800 900 1000 Pressure. Well 3 Exp. Developed 1. Isothermal compressibility results obtained for nonconventional tests 1200 . ROJAS SPE 69724 1. Oil formation volume factor results obtained for nonconventional tests 1.2 120 100 1. 1 / psia 4500 3500 Viscosity . psia 2000 2500 Figure 8. 0. Well 4 Exp. Comparison between correlations developed in this job and others correlations recomended by others authors to calculate heavy oils viscosity Figure 7.1 1. Well 1 Exp. Well 5 Calc. B.PVT No-Conv Exp.001 0.95 0 200 400 600 800 1000 1200 0 100 200 300 400 Pressure . Well 2 Calc.0014 Isothermal Compressibility (Co) . psia Figure 10.PVT Sintetic Conv Calc.15 OFVF (Bo).PVT Conv Calc. Well 4 Exp. Well 4 Calc.6 D.PVT Conv 2000 1500 1000 Exp.PVT Sintetic No-Conv 0 500 1000 0.1 2500 Viscosity ( cps ) OFVF (Bo) . FERNANDEZ AND G.PVT No-Conv Calc.PVT Sintetic Conv 1 500 Calc. Well 3 Calc. FVFo results for well 1 (all possible scenes) 0.