Article pubs.acs.org/EF
Asphaltene Stability Prediction Based on Dead Oil Properties: Experimental Evaluation Doris L. Gonzalez,*,† Francisco M. Vargas,∥ Elham Mahmoodaghdam,† Frank Lim,‡ and Nikhil Joshi§ †
Schlumberger Reservoir Sampling and Analysis, 16115 Park Row, Suite 150, Houston, Texas 77084, United States The Petroleum Institute, Post Office Box 2533, Abu Dhabi, United Arab Emirates ‡ Anadarko Petroleum Corporation, 1201 Lake Robbins Drive, The Woodlands, Texas 77380, United States § Moulinex Business Services LLC, 157 Silverwood Ranch Drive, Shenandoah, Texas 77384, United States ∥
ABSTRACT: The asphaltene content effect on crude oil properties was investigated for a series of deepwater Gulf of Mexico (GOM) fluids with asphaltene contents varying from 4 to 15 wt %. The objective of the study was to conduct flow assurance screening tests on GOM samples collected from different sands and determine properties of the dead oil and the asphaltene stability. Densities, refractive indices, and viscosities were measured at different temperatures in dead oils with three different asphaltene contents. The properties showed defined tendencies with the asphaltene content and with the temperature. The application of the one-third rule in the calculation of properties, such as solubility parameter and viscosity of dead oil systems, was evaluated. This approach also provides an alternative to calculate the refractive index based on densities obtained from an equation of state. The analysis also shows the important role that the asphaltene content plays in determining the viscosity of crude oil and evaluates the possibility of predicting viscosity from refractive index, as proposed by Vargas et al. Another important aspect to evaluate is the prediction of the asphaltene stability in the crude oil by measuring basic dead oil properties, such as density and refractive index. The asphaltene instability trend (ASIST) method was used to predict the asphaltene precipitation onset at reservoir conditions. In this analysis, the asphaltene stability was studied on the heaviest and lightest samples (high and low asphaltene content) by determining the minimum quantity of precipitant required to initiate asphaltene flocculation, followed by measurement of the refractive index of the mixture at the onset conditions. The asphaltene precipitation kinetic effect was also considered in this study.
■
INTRODUCTION The objective of this study was to conduct flow assurance screening studies on Gulf of Mexico (GOM) samples from different sands and determine properties of the dead oil and the asphaltene stability. Asphaltenes are the heaviest components, with high molecular weight, density, and aromaticity, in oil samples. In this paper, the dead oil analysis was conducted on samples with different asphaltene contents of 4, 7, and 15 wt %. Asphaltenes play an important role in the rheological behavior of reservoir fluids.5 Previous studies indicate that the viscosity of dead oils is sensitive to its asphaltene content.6 In 2001, Wang et al. developed the asphaltene instability trend (ASIST) method4 for predicting asphaltene precipitation onset using refractive index (RI) measurements.3 In this work, the asphaltene stability was studied on samples with the lowest and highest asphaltene content by determining the minimum quantity of n-alkane precipitant required to initiate asphaltene flocculation, followed by measurement of the RI of the mixture at the onset conditions. Samples with 4 and 15.5 wt % asphaltene content were titrated with n-heptane (n-C7), n-undecane (n-C11), and n-pentadecane (n-C15), and the RI of each of the mixtures at their onset condition (initiation of asphaltene flocculation) was measured at temperatures above the measured wax appearance temperature (WAT) of the stock tank oils (STOs). The ASIST method was used in this study to predict the asphaltene precipitation onset pressure (AOP) at © 2012 American Chemical Society
reservoir conditions. The asphaltene precipitation kinetic effect was also considered.
■
EXPERIMENTAL SECTION
The following steps were followed during experimental work: (1) The dead oil samples were pretreated by heating and agitating in an ultrasonic bath at 70 °C to homogenize. (2) The SARA analysis, which includes saturates, aromatics, resins, and asphaltenes content, was measured using a modified version of the American Society for Testing and Materials (ASTM) D4124 method, as implemented by Schlumberger Reservoir Sampling and Analysis. n-Heptane was used for asphaltene precipitation. (3) The WAT of the conditioned STO samples was measured using a cross polar microscope (CPM). (4) Density, viscosity, and RI of dead oils with different asphaltene contents were measured at temperatures ranging from 40 to 208 °F (4.4 to 97.8 °C) using an Anton Parr densitometer, Index Instruments refractometer, and the capillary and rheometer (cone−plate) techniques. Asphaltene stability study: (5) The refractive indices of the precipitants n-C7, n-C11, and n-C15 [high-performance liquid chromatography (HPLC)grade quality] were measured at the chosen temperature of 45 °C (113 °F) and 60 °C (140 °F). (6) The precipitants, n-C7, n-C11, and n-C15, were added to the STOs to obtain mixtures with various ratios of precipitant in the mixture. (7) The mixtures of the precipitant and oils were allowed to settle for different times: 20 min, 5 h, 24 h, and 4 days (kinetic study). (8) Subsequently, a sub-sample from each ratio was observed under a high-pressure microscope (HPM) to identify Received: May 15, 2012 Revised: August 22, 2012 Published: August 28, 2012 6218
dx.doi.org/10.1021/ef300837y | Energy Fuels 2012, 26, 6218−6227
Energy & Fuels
Article
Table 1. SARA Analysis
Table 4. RI for Dead Oils
sample
saturates (wt %)
aromatics (wt %)
resins (wt %)
asphaltenes (wt %)
1 2 3 4
39.23 51.56 56.0 57.52
35.9 31.94 30.7 30.4
9.01 8.78 8.5 8.27
15.52 7.6 4.7 3.7
temperature 68 104 140 174.2
°F °F °F °F
(20 (40 (60 (79
°C) °C) °C) °C)
sample 1
sample 2
sample 3
sample 4
1.5272 1.5168 1.5068 1.4977
1.5120 1.5035 1.4944 1.4858
1.5086 1.4992 1.4894 1.4807
N/A 1.4931 1.4829 N/A
Table 2. Wax Properties sample
wax content (wt %)
1 2 3 4
0.9 N/A 0.62 1.07
WAT 85.3 104 86 93.4
°F °F °F °F
(29.6 °C) (40 °C) (30 °C) (34.1 °C)
Table 3. Composition of the Hydrocarbon Fluids component
MW
propane butane pentane C6 C-pentane benzene cyclohexane C7 C-hexane toluene C8 E-benzene xylene C9 C10 C11 C12+ MW calculated MW measured °API − STO density (g/cm3)
44.10 58.12 72.15 86.20 84.16 78.11 84.16 100.20 98.19 92.14 107.00 106.17 106.17 121.00 134.00 147.00 373.8
sample 1 (wt %)
sample 2 (wt %)
sample 3 (wt %)
sample 4 (wt %)
0.48 1.12 2.12 4.49 0.71 0.12 0.45 4.61 1.05 0.44 5.84 0.25 0.95 5.36 6.15 5.23 60.65 270.02 289.99 22.5 0.919
0.53 1.18 2.22 4.14 0.90 0.13 0.68 4.69 1.55 0.47 5.96 1.15 0.27 5.50 6.38 5.22 59.03 265.59 275.18 26.1 0.898
0.56 1.34 2.91 4.94 1.08 0.04 0.80 5.08 1.60 0.55 6.62 0.45 0.71 5.71 6.54 5.33 55.75 256.38 268.7 26.95 0.893
0.59 1.35 2.49 3.37 0.90 0.15 0.77 5.70 1.69 0.61 7.34 0.46 1.42 5.44 6.59 5.44 55.69 256.63 266.63 28.4 0.885
Figure 1. Effect of the temperature on RI for GOM dead oils.
asphaltene flocculation. With this instrument, it is possible to observe particles with a diameter of 1.0 μm and larger. (9) Finer volume ratios were prepared using a separate sample, and steps 6−8 were repeated to identify the minimum volume of precipitant required for asphaltene flocculation. The minimum precipitant volume requirement was termed as the onset of the asphaltene flocculation point at the chosen temperature. (10) The ASIST method was used to predict the asphaltene precipitation onset at reservoir conditions.
Figure 2. Effect of the asphaltene content and temperature on RI for GOM dead oils.
Table 5. Density for Dead Oils
■
temperature 40 60 100 150 208
RESULTS AND DISCUSSION The SARA Analysis. The SARA analysis was performed on dead oil samples with different asphaltene contents (Table 1). Wax Properties. The wax content and the WAT of STO sample were measured to design the tests in such a way that wax would not be precipitated during the asphaltene instability measurement. The WAT of the oils was determined as shown in Table 2. The compositional analysis of the dead oil samples is summarized in Table 3. RI. RI values of the dead oils were measured as a function of the temperature using an automatic refractometer (Table 4).
°F °F °F °F °F
(4.4 °C) (15.6 °C) (37.8 °C) (65.6 °C) (97.8 °C)
sample 1 (g/cm3)
sample 2 (g/cm3)
sample 3 (g/cm3)
0.930 0.915 0.900 0.874 0.865
0.916 0.908 0.892 0.873 0.853
0.904 0.896 0.880 0.861 0.844
The accuracy of the equipment is ±0.0001. The temperature of the samples was controlled to ±0.1 °F with a circulating water bath. Figures 1 and 2 show the effect of the temperature and asphaltene content on RI. According to Figures 1 and 2, the RI is higher for dead oils with greater asphaltene content and lower temperatures. RI shows a 6219
dx.doi.org/10.1021/ef300837y | Energy Fuels 2012, 26, 6218−6227
Energy & Fuels
Article
Table 7. RI Prediction Using the One-Third Rulea asphaltene T density FRI = RI RI (wt %) (°F) experimental density/3 calculated experimental difference
15.5
7.6
4.7
68 104 140 174 68 104 140 174 68 104 140 174
0.914 0.900 0.886 0.872 0.905 0.891 0.877 0.863 0.892 0.878 0.863 0.849
0.3048 0.3000 0.2952 0.2906 0.3018 0.2970 0.2922 0.2876 0.2973 0.2925 0.2877 0.2832
1.5216 1.5119 1.5022 1.4930 1.5154 1.5057 1.4961 1.4870 1.5065 1.4968 1.4872 1.4782
1.5276 1.5168 1.5068 1.4977 1.5120 1.5035 1.4944 1.4858 1.5086 1.4992 1.4894 1.4807
0.0060 0.0050 0.0046 0.0047 −0.0034 −0.0023 −0.0017 −0.0012 0.0021 0.0024 0.0022 0.0025
a
Equipment accuracy = 0.0001 RI units. Experimental error bar = ±0.002 RI units.
Figure 3. Effect of the temperature on the density of the dead oils with different asphaltene content.
Table 8. Viscosity for Dead Oils (cP)
Table 6. FRI/Density for Dead Oils with Different Asphaltene Contents at Different Temperatures
temperature
FRI/density T (°F)
15.5 %
7.4 %
4.7 %
68 104 140 174.2
0.3365 0.3361 0.3359 0.3360
0.3315 0.3321 0.3323 0.3326
0.3345 0.3347 0.3346 0.3348
40 60 100 150 208
°F °F °F °F °F
(4.4 °C) (15.6 °C) (37.8 °C) (65.6 °C) (97.8 °C)
sample 1 15.5 wt % asphaltene
sample 2 7.6 wt % asphaltene
sample 3 4.7 wt % asphaltene
16455 2993 207.6 41.8 15.1
248.8 116.4 41.8 17.7 10.5
105.1 55.9 21.4 10.5 6.3
Figure 5. Viscosity behavior of the dead oils as a function of the temperature and asphaltene content.
model, where the molar refractivity (Rm) is a function of the RI, the molecular weight (MW), and the mass density (ρ).
Figure 4. One-third rule applied to pure components and dead oils with different asphaltene contents.
⎛ RI2 − 1 ⎞ MW MW = FRI Rm = ⎜ 2 ⎟ ρ ⎝ RI + 2 ⎠ ρ
linear decrease with an increasing temperature in the measured range (Figures 1 and 2); they are nearly parallel to each another. Dead Oil Density. The density of the dead oil samples was measured at different temperatures (Table 5 and Figure 3). Density is higher for fluids with higher asphaltene content (Figure 3) because asphaltenes are polynuclear aromatics (PNAs) and have the highest densities. Dead Oil Density versus RI. The “one-third rule” described in the reference by Vargas et al.,1 states that a relationship between density and RI is obtained from the Lorentz−Lorenz
On the basis of measurements for pure hydrocarbon components, the ratio between Rm and MW is a constant with a value around 1/3. F Rm 1 = RI ∼ ∼ 0.333 MW ρ 3
The crude oil samples evaluated in this study show that the relationship between their RI and density is around 1/3 (Table 6). 6220
dx.doi.org/10.1021/ef300837y | Energy Fuels 2012, 26, 6218−6227
Energy & Fuels
Article
The viscosity of the STO is proportional to the asphaltene content of the sample. At low temperatures (
