OTC-27226-MS

Prediction of Asphaltene Onset Pressure from Dead Oil Stability

Mehdi Haghshenas, Soban Balashanmugam, Doris Gonzalez, and Marney Pietrobon, BP Plc

Copyright 2016, Offshore Technology Conference

This paper was prepared for presentation at the Offshore Technology Conference held in Houston, Texas, USA, 25 May 2016.
This paper was selected for presentation by an OTC 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 Offshore Technology Conference and are subject to correction by the author(s). The material does not necessarily reflect
any position of the Offshore Technology Conference, its officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the
written consent of the Offshore Technology Conference 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 OTC copyright.

Abstract
Asphaltene deposition is a major flow assurance issue in some of the deepwater reservoirs in the Gulf of
Mexico. Prediction of asphaltene onset pressure (AOP) is essential for the management of the asphaltene
threat. Generally, determination of AOP requires pressurized down-hole samples with very low or no
oil-based mud contamination. The opportunity to acquire such high quality samples is not always
available. This paper presents a novel approach for the prediction of asphaltene onset pressure from dead
oil samples based on the asphaltene instability trend method.
The asphaltene instability trend method compares the estimated live oil refractive index and the onset
refractive index extrapolated to high pressure/high temperature conditions to infer asphaltene instability.
The main laboratory measurements that are used for AOP prediction are refractive indices of oilprecipitant mixtures at the onset of asphaltene precipitation. The asphaltene instability prediction is based
on the assumption of a linear relationship between onset refractive index and temperature and also the
molar volume of precipitant. In this work, we incorporated the use of equation of state and high pressure
/ high temperature stock tank oil density measurements to estimate the refractive index of live oil. Also,
the molar volume of solution gas was estimated by an equation of state and used to extrapolate the onset
refractive index. These modifications minimized the sensitivity of calculations to the lab data and
provided more accurate AOP predictions compared to the original method.
The modified approach was applied to several oil samples from the Gulf of Mexico. The estimated
AOPs were within 20% of measured AOPs from live oil samples. The new method provides an alternative
for asphaltene stability prediction when a high-quality bottom-hole sample for direct measurement is not
available or the lab data is inconclusive.

Introduction
The Flory-Huggins model originally used for polymer molecules interactions relates the excess energy
of inter-molecular interactions to the difference between solubility parameter of solute and solvent.
Buckley et al. (1998) assumed that dominant intermolecular interaction energy governing asphaltene
precipitation is the London dispersion interaction. With this assumption, they showed that interaction
energy is expressed as a function of the differences between the squares of the refractive indices rather
than the differences between the solubility parameters of asphaltene and solvent. Therefore, solubility
parameter could be estimated from refractive index which is measured more easily than solubility

OTC-27226-MS

parameter. Furthermore, the instability of asphaltene could be related to the refractive index of asphalteneprecipitant mixture at the onset of asphaltene precipitation. This forms the basis for the asphaltene
instability trend (ASIST) which compares the live oil refractive index and the onset refractive index at
high pressure/high temperature conditions to infer asphaltene instability as developed by Jianxin Wang as
his PhD thesis (Wang 2000). In 2004, Wang et al. published the application of dead oil titration data in
AOP prediction. Later in 2006, they presented a one-point asphaltene instability trend method as a
screening tool for asphaltene instability prediction. In 2009, Creek et al. used asphaltene instability
induced by low molecular weight alkanes to predict live oil asphaltene stability. A 2011 publication by
Montesi et al. in the Offshore Technology Conference notes that asphaltene instability trend provides the
most accurate representation of asphaltene precipitation for the Blind Faith field and its predictions are
used to manage asphaltenes. Gonzalez et al. (2012) presented application of normal-alkane titration in
predicting AOP for samples with high asphaltene content (4 and 15.5 weight percent). However,
application of the published method to fluids from a Gulf of Mexico reservoir did not provide accurate
AOP predictions; hence, the analysis method was revisited and the modifications were made to develop
a new approach as explained below.

Materials and Methods

The dead oil samples tested in this study are obtained from flash down of bottom-hole samples. The
pertinent bottom-hole samples were collected from five Gulf of Mexico wells using pressure-compensated
sampling bottles with low oil-based mud and chemical contamination. This study focused on the samples
that could be tested for live AOP measurement to use the measured data for method validation.
The basic lab data for the Wang et al. (2004) method involves titration of a dead oil sample with three
different normal paraffin precipitants (usually n-heptane, n-undecane and n-pentadecane) at three different
temperatures (usually 40, 50 and 60C). The titrated aliquots are viewed under an optical microscope. A
complete test on single oil provides the following data:

Density of oil post centrifuging (usually carried out to clean the oil of any sediments, free water
etc.)
Refractive index of the centrifuged, interference free oil (usually at 40, 50 and 60C)
Onset volume percent and refractive index for nC7, nC11 and nC15 (i.e. the volume percent of
precipitant in the oil and the refractive index of the mixture at which the asphaltenes in the oil are
only just stable, with each titrant and at each temperature)
Change in onset volume percent and refractive index (with each precipitant) with temperature.
Previous studies note the effect of kinetic time on the onset refractive index (Wang et al., 2004 and
Gonzalez et al., 2012). Kinetic time refers to the time spent between mixing the oil with precipitant and
assessment of the mixture for asphaltene flocculation; with longer kinetic times, smaller amounts of
precipitant are required to precipitant asphaltenes. Variable kinetic times add another degree of uncertainty to the analysis; therefore, in this study a constant kinetic time of 30 minutes is used in all the tests.
The purpose was to keep this time as short as possible while ensuring it is long enough to achieve
repeatability of test results.

Prediction of Asphaltene Onset Pressure

Wang et al. (2000) observed that onset refractive index is linearly correlated to the square root of molar
volume of precipitant. This linear trend is then used to find the onset refractive index (RI) of solution gas
assuming that solution gas acts as the asphaltene precipitant in depressurization of live oil. Our onset RI
data for three precipitants (nC7, nC11, nC15) showed that a plot of onset RI vs precipitant molar volume
). Therefore, for the
(Vm) exhibits a more linear trend than onset RI vs square root of molar volume (

OTC-27226-MS

fluids in this study we used molar volume instead of its square root for extrapolation to find the onset RI
for solution gas. The molar volume of solution gas is estimated using Hall-Yarborough compressibility
factor (Hall and Yarborough, 1982).
A major assumption of this method is the linear trend of onset RI with temperature. This trend is used
to extrapolate onset RI from test temperatures to reservoir temperatures (Wang et al., 2004). We examine
the linear relationship between onset RI and temperature performing the tests at 40, 50, and 60C. The test
temperatures should be above wax appearance temperature to avoid interference of wax particles with
asphaltene flocculation. However at temperatures higher than 60C, the evaporation of n-heptane and light
end of crude results in higher, non-repeatable RI measurements.
The onset RI should be compared to live oil RI to infer asphaltene stability. Because it is not feasible
to measure refractive index of oil at live conditions, the live oil RI is estimated from dead oil RI and
solution gas refractivity at surface condition. Synovec and Yeung (1983) showed that the overall refractive
index of an ideal mixture can be calculated based on the volume fractions, as follows:
(1)

where ni and i are the refractive index and volume fraction of each component in the mixture,
respectively. Although solution gas and stock tank oil (STO) do not form an ideal mixture, this method
is the only available way to estimate RI of live oil at reservoir conditions.
(2)

where
Vliquid
n
refractive index (RI)
M
molecular weight
R
refractivity

density
The STO densities in the above equation come from high-pressure/high-temperature measurements.
Gas densities could come from lab measurements or an equation of state. In the Vliquid equation above,
is calculated from gas oil ratio (GOR) and surface gas and oil densities available in the PVT
reports. Gas refractivity (Rgas) is calculated from molar fraction and refractivity of gas components up to
hexane (C6):
(3)

The pure component refractivities are listed in Table 1 (Hadrich, 1975).

OTC-27226-MS

Table 1Pure component molar refractivities.

Component
N2
CO2
C1
C2
C3
i-C4
n-C4
i-C5
n-C5
C6

Name

Molar refractivity (cc/mole)

Nitrogen
Carbon dioxide
Methane
Ethane
Propane
i-Butane
n-Butane
i-Pentane
n-Pentane
Hexane

4.752
6.835
6.70
11.35
16.00
20.64
20.64
25.29
25.29
29.94

STO refractive index is calculated from STO density at high pressure and high temperature using
Lorenz-Lorentz equation (Wang, 2000):
(4)
where R is molar refraction, M is molecular weight, and is density. Although this equation is
developed and tested for pure compounds, it can be used to find STO RI at high pressures and high
temperatures,
(5)
is then plugged into Equation (2) to find live oil refractive index. Vargas et al. (2009 and
2010) showed that for a variety of crude oils FRI/ is always approximately equal to 1/3. The so-called
one-third rule could be used as a check for Equation (4).

Results and Discussion

The analysis results for 5 dead oil samples from various reservoirs in Gulf of Mexico are discussed in this
paper. Table 2 shows the bulk properties of the dead oil samples and associated live oil properties. The
asphaltene onset pressure is measured on the live oil samples using a solid detection system with infrared
signal and high pressure microscope. The flashed down dead oil is then used for onset measurements as
described above. Table 3 shows the onset refractive index of each sample with three precipitants at three
test temperatures. The onset volume and PVT properties of individual samples are used to construct the
stability envelope shown in Figure 1. The crossing of live oil refractive index and onset refractive index
indicates the asphaltene onset pressure. Table 4 shows the predicted AOP of individual samples compared
to the measured AOP using solid detection system. The predicted AOP for samples 1, 3, and 4 is very
close to the measured AOP (less than 10% error). Sample 2 showed a very low measured AOP (5,200
psia) which is expected from low GOR and high reservoir temperature for this sample. The predicted AOP
for this sample is 25% higher than the measured value which is the largest error amongst the samples
tested. Sample 5 did not show any AOP in the live measurement indicating a very stable oil with respect
to asphaltene precipitation. Dead oil tests confirmed this observation where 83% of n-heptane was
required to precipitate asphaltenes. This is much higher than the onset precipitant volume for the other
samples (30 to 60 volume percent). With such high volume percent, hence low onset RI, no AOP was
predicted for Sample 5 using the modified method (Figure 5). To compare the results of modified
approach to the published method of Wang et al. (2004), Sample 1 was tested at various kinetic times to
generate the asphaltene stability envelope using the original method as shown in Figure 6. The effect of

OTC-27226-MS

kinetic time makes it difficult to assess asphaltene stability and quantify the asphaltene onset pressure. The
measured asphaltene onset pressure (using high pressure microscope) is above 8,000 psi which is
confirmed with field observations. Similar comparison for Sample 3 shows that the original method does
not predict any AOP for this sample by using 2 and 4 hour kinetic times (Figure 7); whereas the modified
approach shows a very high AOP of 12,000 psi (Table 4) which is confirmed by lab measurement and
field observations.

Table 2Live and dead oil properties of the samples.

Sample No.
1
2
3
4
5

Density at 20C (g/cc)

RI at 20C

API Density

GOR (scf/stb)

Saturation Pressure (psia)

Measured AOP (psia)

0.8712
0.8784
0.8930
0.8702
0.8974

1.4885
1.4951
1.5038
1.4904
1.5047

32.1
30.7
25.3
31.1
25.6

910
717
1351
1015
908

3606
3013
7668
4582
5964

8500
5200
12000
6400
No AOP

Table 3Onset refractive index for each precipitant and temperature.

Sample 1

Precipitant

Sample 2

C7
C11
C15
Precipitant

Sample 3*

C7
C11
C15
Precipitant

Sample 4

C7
C11
C15
Precipitant

Sample 5**

C7
C11
C15
Precipitant
C7

Onset
40
1.4359
1.4448
1.4563
Onset
40
1.4469
1.4564
1.4656
Onset
22
1.4286
1.4402
1.4523
Onset
40
1.4218
1.4353
1.4477
Onset
40
1.3996

Temperature(C)
50
60
1.4307
1.4274
1.4413
1.4373
1.4529
1.4504
Temperature(C)
50
60
1.4424
1.4373
1.4520
1.4475
1.4614
1.4570
Temperature(C)
40
60
1.4238
1.4184
1.4364
1.4314
1.4482
1.4450
Temperature(C)
50
60
1.4169
1.412
1.4309
1.4265
1.4435
1.4393
Temperature(C)
50
60
1.3933
1.3881

*Sample 3 tests were run at different temperatures than other samples.

**Sample 5 showed a high level of stability with nC7; hence, nC11 and nC15 tests were not necessary.

OTC-27226-MS

Figure 1Live oil refractive index and onset refractive index for Sample 1.
Table 4 Predicted AOPs in comparison with live AOP measurements.
Sample No.
1
2
3
4
5

Measured AOP (psia)

Predicted AOP (psi)

Error

8500
5200
12000
6400
No AOP

8500
6500
11000
6500
No AOP

0%
25%
-8%
2%
NA

Figure 2Live oil refractive index and onset refractive index for Sample 2.

OTC-27226-MS

Figure 3Live oil refractive index and onset refractive index for Sample 3.

Figure 4 Live oil refractive index and onset refractive index for Sample 4.

OTC-27226-MS

Figure 5Live oil refractive index and onset refractive index for Sample 5.

Figure 6 Live oil refractive index and onset refractive index for Sample 1 using the Wang et al. (2004) method with 3 kinetic times (10
minutes, 5 hours, and 24 hours).

OTC-27226-MS

Figure 7Live oil refractive index and onset refractive index for Sample 3 using the Wang et al. (2004) method with 2 kinetic times (2
hours and 4 hours).

Conclusions
The method presented in this paper provides an alternative to live AOP measurement by using dead oil
samples to predict asphaltene stability at live conditions. This method uses the concept offered by Wang
et al. (2004) along with modifications that provide accurate AOP predictions where the original method
predictions were not consistent with live measurements and field observations. The main enhancements
are the use of equation of state to predict gas molar volumes for extrapolation of onset refractive index
and removing the effect of experiment kinetic time from AOP estimation. The predicted AOPs for the
tested samples are accurate for practical design and mitigation purposes and eradicate the need for high
quality live oil samples which are expensive and difficult to acquire. Further testing on fluids from Gulf
of Mexico and other regions with known asphaltene instability is needed to extend the application of this
approach.

Nomenclature
AOP
ASIST
FRI
GOR
HP/HT
HPM
M
n
PVT
STO
R

asphaltene onset pressure

asphaltene instability trend
refractive index function
gas oil ratio
high pressure high temperature
high pressure microscope
molecular weight
refractive index (RI)
pressure volume temperature
stock tank oil
refractivity

10

OTC-27226-MS

RI
Vm
x

refractive index
molar volume
mole fraction
density
volume fraction

References
Buckley, J.S., Hirasaki, G.J., Liu, Y., Von Drasek, S., Wang, J.X., and Gill, B.S. Asphaltene Precipitation and Solvent
Properties of Crude Oils. Petroleum Science and Technology, 1998, 16(3&4), 251285.
Creek, J. L., Wang, J., & Buckley, J. S. Verification of Asphaltene-Instability-Trend (ASIST) Predictions for LowMolecular-Weight Alkanes. SPE Production and Operations. May 2009.
Gonzalez, D. L., Vargas, F. M., Mahmoodaghdam, E., Lim, F., Joshi, N. Asphaltene Stability Prediction Based on Dead
Oil Properties: Experimental Evaluation. Energy & Fuels 2012, 26, 6218 6227.
Hadrich, J. The Lorentz-Lorenz Function of Five Gaseous and Liquid Saturated Hydrocarbons. Applied physics, 1975,
7(3), 209 213.
Hall, K.R. and Yarborough, L. 1973. A New Equation of State for Z-Factor Calculations. Oil & Gas Journal. 1982. 71
(25), 8292:
Montesi, A., Pinnick, R. A., Subramanian, S., Wang, J., & Creek, J. L.. Asphaltene Management in GOM DW Subsea
Development. Offshore Technology Conference. January 2011.
Synovec, R.E. & Yeung, E.S. Quantitative Analysis without Analyte Identification by Refractive Index Detection.
Analytical Chemistry. 1983, 55, 1599 1603.
Vargas, F.M., Gonzalez, D.L., Creek, J.L., Wang, J., Buckley, J., Hirasaki, G.J., & Chapman, W.G. Development of a
General Method for Modeling Asphaltene Stability. Energy & Fuels, 2009, 23 (3), 11471154.
Vargas, F. M., Chapman, W. G., Application of the One-Third rule in hydrocarbon and crude oil systems. Fluid Phase
Equilibria 2010, 290, 103108.
Wang, J. Predicting Asphaltene Flocculation in Crude Oils. PhD Thesis, New Mexico Institute of Mining & Technology.
April 2000.
Wang, J. X., Buckley, J. S., Burke, N. E., & Creek, J. L. A Practical Method for Anticipating Asphaltene Problems. SPE
Production & Facilities. August 2004.
Wang, J., Creek, J. L., & Buckley, J. S. Screening for Potential Asphaltene Problems. SPE Annual Technical Conference
and Exhibition, San Antonio, Texas, USA. September 2006.