Gas Condensate Properties 1770s

SPE-170601-MS Correlations to Estimate Key Gas Condensate Properties through Field Measurement of Gas Condensate Ratio J.E. Paredes, R. Perez, and L.M. Perera, PEMEX; C.J. Larez, Schlumberger

Copyright 2014, Society of Petroleum Engineers This paper was prepared for presentation at the SPE Annual Technical Conference and Exhibition held in Amsterdam, The Netherlands, 27–29 October 2014. This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its officers, or . Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.

Abstract When a hydrocarbon reservoir is discovered it is important to know the type of fluids that are present as well as their main physicochemical characteristics, this is normally obtained by performing a PVT analysis to a representative fluid sample of the reservoir. In most cases, having a PVT analysis can take several months, which limits the number and type of reservoir studies that can be carried out during this period. The only parameter that can be measured almost immediately after a well is completed, are conventional production measurements. In some cases this production measurement can be obtained before completing the well by using special testing or measuring equipment such as a DST (Drill Stem Test, acronym in English). It is important to obtain preliminary values of properties such as: molar percentage of heptane and heavier components (% mole of C7⫹), molecular weight of the original fluid (MW), maximum retrograde condensation (MRC) and dew point pressure (Pd). Most of these properties are very important for exploitation of gas condensate reservoir and their early early availability will allow engineers to carry out reservoir studies that will ensure an efficient exploitation and maximize the final recovery of the liquids present in the reservoir. This work proposes a series of correlations, obtained from analyses of several PVT studies collected mainly from reservoirs of Mexico and other regions of Latin America. These correlations can be used to estimate reservoir properties for gas condensate such as: % mole of C7⫹, MW, MRC, Pd and Z-factor at Pd. The only parameter needed to use these correlations is the value of the gas condensate ratio (GCR) of the fluid during the early stage of production. These empirical equations should be valid for any gas condensate reservoir worldwide; although a range of usability is proposed for a better performance of the correlations.

Introduction Fluids present in an oil reservoir are the result of a series of thermodynamic changes of pressure and temperature suffered by the original mixture of hydrocarbons over time and during its migration from the

2

SPE-170601-MS

Figure 1—Hypothetical p-T diagram for gas condesate13.

container rock to the trap. Reservoir pressure and temperature increase with depth and their relative relationship will influence the behavior of the light and heavy components that the fluid could contain. In general, the content of light components in a mixture of hydrocarbons increases with temperature and depth, which may result in reservoirs near the critical point; gas condensate reservoirs are included in this kind of fluids. Looking at a phase envelope it can be appreciated that the temperature in this type of reservoir is located between the cricondentherm and the critical temperature (Figure 1). The critical point is to the left side of the cricondenbar and the lines of quality are predominantly pushing towards the bubble point line. Production wells from gas condensate fields can suffer strong changes in pressure and temperature through the production cycle and can quickly enter the two phase region (Figure 1) in any point. These sudden changes can for the high variability found in measured values of gas-condensate ratio between 600 and 10,000 m3/m3 approximately1. Recoverable liquid is in general light colored and presents densities ranging from 0.75 to 0.80 gr/cm3. Due to the importance of knowing early key parameters in gas condensate reservoir such as dew point pressure in other to optimize production scheme early on, it is necessary to have as soon as possible a PVT analysis. Production goals and operational conditions in most cases prevent to carry out this type of analysis and it is therefore necessary to use empirical correlations to estimate some reservoir properties necessary to perform any reservoir analysis that could lead to optimize the exploitation of the reservoir. PVT correlations are important tools when there is little information available from fluids samples. PVT correlations may come in the form of empirical equations, graphics or in some cases as both forms solutions. First correlations were published by Standing2 in 1947, using known values of temperature and GOR to obtain bubble point pressure and the oil formation volume factor. After Standing correlations,

SPE-170601-MS

3

Table 1—Descriptive statistics of properties analyzed in the study. Variable

Average

Minimum

Maximum

Standard Dev.

Reservoir T (°C) Dew point (Kg/cm2) MW (g/gmol) GCR (m3/m3) C7⫹ (% mol) Z-Factor (Adim) MRC (%)

131.6 361.95 32.28 2, 619 6.52 1.049 18.81

87.0 202.5 20.23 598 1.17 0.814 1.07

186.9 537.8 45.09 12, 199 12.98 1.322 38.50

29.9 70.87 7.10 2, 759 3.27 0.113 10.80

several authors (Glaso3 1980, Sutton and Farshad4 in 1984) developed a set of equations that make it easier to have critical parameters needed to perform reservoir analysis such as material balance, nodal analysis and numerical simulation. Hydrocarbon reservoirs can be classified into black oil, volatile oil, gas condensate, wet gas and dry gas and each of these fluids type has unique features and properties, governed by different thermodynamic phenomenon, which hinders the use of one common equation to reproduce their thermodynamic behavior. In the case of this work, it is focused on obtaining a set of equations to describe the behavior of some properties for gas condensate reservoirs. This paper presents a set of empirical correlations obtained from the analysis of different PVT analysis from fluid samples of gas condensate reservoirs mainly from Mexico and other parts of Latin America. These correlations can be applied to any gas condensate reservoir in the world whose characteristics are within or similar to the range of the data used in this study. Likewise a comparative analysis was made between the results obtained with the proposed correlations proposals and others similar correlations found in the literature.

Available data There were more than 100 PVT analyses available during the development of this work. These PVT analyses correspond to 41 gas condensate reservoirs. Each of this PVT analysis was subjected to a prior process of validation using molar balance techniques proposed by Bashbush5, Whitson6 and Hoffmann Crump7 to ensure the consistency of its composition and discard those with unreliable information. The ranges for each of the variables select from these PVT analyses, as well as its statistical descriptions are presented in the Table 1. Most of the statistical models used for the development of the correlations are non-linear. The software MatLab™ was used for the construction of these solutions. The ranges shown in Table 1 can be used to ensure the reliability and applicability of the proposed equations before using them in any reservoir, especially for reservoir temperature and the gas condensate ratio.

Development of correlations After each PVT validation was done, a database with key parameters and properties for gas condensate reservoir was built. Previously published works8, 9 explain the relationship that may exist between the maximum retrograde condensation, average molecular weight, molar percentage of C7⫹ and the gas condensate ratio. Based on this, a statistic and numerical analysis were made to find a possible correlation between these parameters. After the analysis was done, three correlations were found. Correlation to calculate the % mole of C7ⴙ The following empirical equation is proposed to calculate the % mole of C7⫹ of a fluid sample, knowing field measurements of GCR (m3/m3):

4

SPE-170601-MS

Figure 2—% mole of C7ⴙ. Experimental data and fitted curve.

(1) where: A ⫽ 2, 258.31297446475 B ⫽ ⫺0.816117090683706

Figure 2 shows the experimental data and curve represented by the proposed equation. It’s recommended to use the correlation for values of GCR between 600 and 8,000 m3/m3. Correlation to calculate MW (g/gmol) The following empirical equation can be used to estimate the MW using measurement of the GCR (m3/m3): (2) where: A ⫽ 687.1830018707650 B ⫽ ⫺0.429370789691255

Figure 3 shows the experimental data and curve represented by the proposed equation. Similarly to previous correlation, this equation it’s reliable and recommended to be use within the same range for GCR. Correlation to calculate the MRC (%) The data analysis showed that it was impossible to obtain an empirical equation that adequately correlates the MRC with the GCR as with previous equations. One possible explanation for this it’s that for values of GCR bellow 1, 200 m3/m3 and reservoirs near the critical region (critical point), the fluid thermodynamic behavior is governed by the molar composition of heavy fractions9. Based on this assumption, the

SPE-170601-MS

5

Figure 3—MW. Experimental data and fitted curve.

focus of the analysis was change to find a correlation between MRC and the mole fraction of C7⫹. An empirical equation is proposed to estimate the MRC according to the % mole of C7⫹: (3) where: A ⫽ 2.921696607928510 B ⫽ 2.640466985198960

Figure 4 shows the experimental data and curve represented by the proposed equation. The range of applicability is the same as previous equations and the value of the % mole of C7⫹ can be obtained from Equation 1. Nomograph complementary to the three empirical equations presented in this work, a graphical solution or nomograph was built; this nomograph relates equations 1 to 3, see Figure 5. This graphics solution allows to quick and accurately estimate % mole of C7⫹, MW and MRC, without the need for complex calculations, which is very useful for field engineers. Another correlation was developed to estimate dew point pressure. In the literature there are several published correlations but internal testing showed that theses correlation have strong variations between measured and calculated data and were only reliable for a very specific range of reservoir temperature and specific gravity; these limitations can be associated with the limited number of fluid sample used for the development of those correlations. In the case of the correlation proposed in this paper, several parameters were analyzed. The result showed that the dew point pressure have a strong correlation with the gas Z-factor measured at the dew point pressure. Correlation to calculate the Gas Z-Factor at dew point pressure (Adim) To estimate the value of the gas Z-Factor at dew point pressure, the following empirical correlation, based on the % mole of C7⫹ and MW is proposed:

6

SPE-170601-MS

Figure 4 —MRC, experimental data and fitted curve.

Figure 5—Nomograph for calculation of MRC, MW and % mole of C7ⴙ using GCR measurements.

(4) where: A ⫽ ⫺89.543548

D ⫽ ⫺16.777186

B ⫽ 34.620795

E ⫽ 7.2393271

C ⫽ 29.186098

SPE-170601-MS

7

Figure 6 —Fitted surface for gas compressibility factor (Z-Factor) at dew point pressure.

Figure 6 shows the experimental data and curve represented by the proposed equation. The values of the % mole of C7 ⫹ and MW can be estimated using equations 1 and 2 respectively. Correlation to calculate dew point pressure (kg/cm2) The following empirical equation is proposed to calculate dew point pressure for a fluid sample of a gas condensate reservoir, knowing the Z-Factor at the dew point pressure, which can be calculated using the equation 4. The proposed correlation is: (5) where: A ⫽ ⫺0.27522065593 B ⫽ 2.121956466266 C ⫽ 2.129009818

In the Figure 7, experimental data and curve represented by proposed equation are presented. The final step was to analyze the robustness of the correlation for dew point pressure. A data set was used to calculated dew point pressure using other previously published correlations and the proposed in this paper. The correlations used were proposed by Marrufo10 and Ovalle11. As you can see in Figure 8, the results showed that the correlation proposed in this work has less than 10% error in 90% of the data to which it was applied; unlike the correlations of control with errors greater than 20% in approximately 50% of the same set of data.

Range of applicability All proposed correlations can be used to estimate key properties by knowing measurement of GCR from any exploratory or development well of the reservoir. A range of applicability was established, mainly to GCR values of the well that is going to be used. The range for GCR is 600 to 8000 m3/m3. This range also goes according to the classification of fluid type1, since if a fluid has GCR value lower than 600 m3/m3, it can be considered a volatile oil instead of a gas condensate reservoir and with values above 8000 m3/m3 the fluid in the reservoir begins to behave like a wet gas reservoir.

8

SPE-170601-MS

Figure 7—Natural logarithm of dew point pressure vs. eZ and fitted curve.

Figure 8 —Results obtained using several correlations to estimate dew point pressure and error comparative.

Another important point to note is when it’s the best time to measure GCR values on the field, the answer is during the initial stage of the reservoir exploitation and before the dew point pressure is reached. This is why it is recommended to use the GCR obtained from a stabilized production well test, at the beginning of its production if possible.

Practical application The proposed correlations were used to estimate and compare results on two recently drilled exploratory wells. Both wells have a full feature PVT analysis report. The information from these wells weren’t used

SPE-170601-MS

9

Table 2—Comparative table for results obtained on control wells. Exploratory well #1

% Mole C7ⴙ (%)

MW (g/gmol)

MRC (%)

Z @ Pd

Pd (kg/cm2)

Calculated Experimental data measurement (PVT report) % Error

10.54 10.31 2.3

40.81 39.08 4.4

28.16 31.03 9.2

1.043 1.005 3.8

367.9 397.0 7.3

Exploratory well #2

% Mole C7ⴙ (%)

MW (g/gmol)

MRC (%)

Z @ Pd

Pd (kg/cm2)

Calculated Experimental data measurement (PVT report) % Error

9.05 8.92 1.4

37.66 36.25 3.9

23.80 22.95 3.7

1.125 1.049 7.3

414.9 396.0 4.8

for the development of the correlations and therefore they were considered control wells. From field measurements, the value of GCR was 718 m3/m3 for Exploratory well #1 and 866 m3/m3 for Exploratory well #2. Table 2 presents the results obtained. The accuracy of the proposed correlations was measured by calculating the % deviation from the experimental measurements. The results obtained by the correlations are very similar and that there is less than 10% error in all cases analyzed.

Conclusions A set of correlations to estimate key properties of gas condensate reservoirs, using gas-condensate ratio (GCR), was obtained from the analysis of more than 100 PVT studies mainly from Mexico and other regions of Latin America. These correlations should be valid for any gas condensate reservoir worldwide; although a range of usability is proposed for a better performance of the correlations. The properties that can be estimated are: % mole of C7⫹, MW of the original fluid, maximum retrograde condensation (MRC), dew point pressure and Z-Factor at dew point pressure. This validation methodology indicates that the percentage of error is lower than 10% in the proposed correlation to calculate Pd while previously published correlations may have error greater than 20% and have a limited range of applicability. The correlations proposed in this paper were successfully applied to a couple of exploratory wells, obtaining an error lower than 10% when compared to lab measurements. Nomenclature PVT Pressure Volume Temperature Analysis DST Drill Stem Test Pd Dew Point Pressure Heptane Plus and Heavier Fractions C7⫹ MW Molecular Weight of The Reservoir Fluid MRC Maximum Retrograde Condensation Z-Factor Gas Compressibility Factor GCR Gas-Condensate Ratio GOR Gas-Oil ratio A Coefficient A B Coefficient B C Coefficient C

10

SPE-170601-MS

D E

Coefficient D Coefficient E

References 1. Méndez L.T., Teyssier S.J., “Caracterización de Fluidos de Yacimientos Petroleros”, Revista del Instituto Mexicano del Petróleo”, Vol XI No 4, October 1979. 2. Standing, M.B.: “A Pressure-Volume-Temperature Correlation for Mixture of California Oils and Gases” Drill. And Prod. Prac., API (1947), 275. 3. Glas␾, ⌽: “Oil-System Correlations” Petroleum Production Handbook, T.C. Frick (ed.), SPE, Richardson, TX (1962), Chap 19. 4. Sutton, R.P and Farshad, F.F.: “Evaluation of Empirically Derived PVT Properties for Gulf of Mexico Crude Oils” Paper SPE 13172, Houston, Texas September 1984. 5. Bashbush J. L., “A Method to Determine K-Values From Laboratory Data and Its Applications” SPE 10127, Dallas, Texas, October 1981. 6. Whitson C. H., “Evaluating Constant Volume Depletion Data” SPE 10067, Dallas, Texas, October 1981. 7. Hoffmann A.E., Crump J.S. and Hocott C.R., “Equilibrium Constants for a gas condensate system” SPE 219, Petroleum Transactions, AIME 1953. 8. Paredes J.E., Perez R., Larez C.J., “Correlation to Predict and Validate the Maximum Retrograde Condensation in Gas Condensate Reservoirs” SPE 158494 San Antonio, USA, October 2012. 9. Cho S.J., Civan F., Starling K.E., “A Correlation To Predict Maximum Condensation for Retrograde Condensation Fluids and Its Use in Pressure-Depletion Calculations” SPE 14268, Las Vegas, Nevada, September 1985. 10. Marrufo I., Maita J., Him J., Rojas., “Correlations To Determine Retrograde Dew Pressure and C7⫹ Percentage of Gas Condensate Reservoirs on Basis of Production Test Data of Eastern Venezuelan Fields” SPE 75686, Calgary, Alberta, May 2002. 11. Ovalle A.P., Lenn C.P., McCain W.D., “Tools To Manage Gas/Condensate Reservoirs; Novel Fluids-Property Correlations on the Basis of Commonly Avaliable Field Data” SPE 112977, Doha, Qatar, November 2005. 12. McCain W.D.: “The Properties of Petroleum Fluids”, PennWell Publishing Company, TulsaOklahoma, 1990. 13. Whitson C.H., Michael R. Brulé: “Phase Behavior”, SPE Monograph Series Volume 20, Richardson-Texas, 2000. CONVERSION FACTORS

1 1 1 1

m3 ⫽ 6.28981 stb m3 ⫽ 35.314667 scf m3/m3 ⫽ 5.614584 scf/stb kg/cm2 ⫽ 14.223296 psi

STATISTICAL PARAMETERS

Coefficient of determination (R2). Defined by the ratio of the sum of squares over the total sum of squares: (A) Absolute error calculations. Defined as the sum of all the relative errors divided by the number of values used in the calculation of the average error. (B)