Journal of Petroleum Science and Engineering 42 (2004) 157 – 170

www.elsevier.com/locate/petrol

Asphaltene precipitation and deposition in the near wellbore

region: a modeling approach
Reyadh A. Almehaideb
Chemical and Petroleum Engineering Department, College of Engineering, UAE University, P.O. Box 17555, Al-Ain, United Arab Emirates

Abstract

In this work, a new single-well model to simulate asphaltene precipitation, deposition, and plugging of oil wells during
primary production is developed. A four-component (asphaltene – oil – gas – water), four-phase (solid – oil – water – gas) limited
compositional formulation is described to account for asphaltene flow with crude oil and its precipitation, deposition, and
plugging near the well, with the resulting effect on well productivity. The model accounts for the asphaltene precipitation
phenomena directly through the use of laboratory values of asphaltene solubility versus pressure data. Four to six primary
variables are solved simultaneously in the model. These include the pressure, three of the four phase saturations, the gas
saturation (bubble point) pressure, and the asphaltene saturation pressure. The model is implemented in cylindrical (r – z)
coordinates to match the flow direction around the well.
This model is applied to a typical well configuration using a recombined asphaltic UAE crude oil sample. The asphaltene
solubility versus pressure relationship is first evaluated experimentally using the gravimetric method. Then, the asphaltene
precipitation and deposition around a typical production well is then simulated. Results of the model demonstrate how the
asphaltene precipitation zone develops around production wells as the pressure at the well declines, thus precipitating
asphaltenes as the pressure drops below the upper asphaltene precipitation pressure. Also, they show the growth of deposited
asphaltene near the well with time, as the plugging problem takes a few hundred days to develop after the start of production.
Application of this model will help operators to better forecast the onset of asphaltene precipitation as it relates to flow rate
history of the well and the pace of plugging induced by asphaltenes, and thus better plan remedial measures.
D 2004 Elsevier B.V. All rights reserved.

Keywords: Asphaltene; Plugging; Deposition; Modelling; Wellbore flow

1. Introduction tigators also define it as the pentane insoluble fraction

of crude oil, even though amounts and character of
1.1. Definition of asphaltenes asphaltenes precipitating are different (Kabir and
Jamaluddin, 1999). Their presence and stability in
Asphaltenes are defined as ‘‘the wax-free fraction crude oil is closely related to the intermediate molec-
of crude oil that is insoluble in n-heptane but soluble ular weight (MW) compounds called resins (Hirsch-
in hot toluene/benzene’’ (ASTM 6560). Some inves- berg et al., 1982; Burke et al., 1988). Resins are
defined as the fraction of crude oil insoluble in
propane but soluble in n-heptane. This is because
E-mail address: reyadh@uaeu.ac.ae (R.A. Almehaideb). both resins and asphaltenes are insoluble in propane,

0920-4105/$ - see front matter D 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.petrol.2003.12.008
158 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

C3. Both asphaltenes and resins are heavy aromatics value in terms of wt.% of asphaltene precipitate at the
and form the most polar parts of crude oil. The MW of bubble point. Afterwards, asphaltene may start to
asphaltene is generally higher than MW of resins dissolve back in oil as dissolved gas evolves below
(Mohammed et al., 1996). The fraction of crude oil the bubble point, changing the composition of the
precipitated by propane (C3) may be liquid especially liquid. Thus, to describe the asphaltene precipitation
at high reservoir temperatures but the fraction precip- phenomena, one needs to know the asphaltene pre-
itated by n-heptane (n-C7) is solid (Jamaluddin et al., cipitation envelope (APE), shown schematically in
2001). Together, asphaltenes and resins are sometimes Fig. 1, and the amounts precipitated in wt.% at each
called Asphalt. Asphaltene occurs naturally in crude pressure step, shown schematically in Fig. 2.
oil in the range of 0– 60% with most crude less than
10% (Mohammed et al., 1996). The ratio of asphal- 1.2.1. Experimental determination of asphaltene
tenes to resins is around 1:1 – 1:20 for stable crude, precipitation
while for unstable crudes it is usually less than 1:1. It Asphaltene precipitation is evaluated experimen-
is also important to distinguish asphaltenes and resins tally using a variety of methods. The most widely
from wax, a solid that may also sometimes deposit used methods are the gravimetric (static) method
near the wellbore and is composed of high molecular and the filtration (dynamic) method. The gravimetric
weight aliphatic hydrocarbons. Field experience of method, which was used in this work, is based on
operators, however, is that waxes precipitate in much evaluating asphaltene precipitation from asphaltene
smaller quantities than asphaltenes (Leontaritis, solubility versus pressure performed on a visual
1998). PVT cell. Jamaluddin et al. (2001) presented a
recent comparison of these methods and a descrip-
1.2. Asphaltene precipitation tion of experimental methods for evaluating asphal-
tene precipitation.
For asphaltene to deposit in porous media, it has to
precipitate first from crude oil, as only suspended 1.2.2. Thermodynamic modelling of asphaltene pre-
(free) asphaltene may deposit and plug the porous cipitation
media. Phase behavior studies have shown that Thermodynamic models of asphaltene precipitation
asphaltene precipitation starts to occur at a pressure are mathematical equations that can be used to predict
higher than the bubble point, called the upper asphal- asphaltene precipitation and the amounts precipitated.
tene precipitation pressure, and reaches a maximum There are two views with regard to presence of asphal-

Fig. 1. Schematic of the asphaltene precipitation envelope.

 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 159

Fig. 2. Schematic of the asphaltene solubility versus pressure, after Kabir and Jamaluddin (1999).

tenes in crude. One view believes that asphaltenes 1.3. Asphaltene deposition models
precipitation, just like other components, is reversible.
Therefore, thermodynamic equilibrium models that There are the two recognized modes of deposition
give precipitation as a function of the composition, for asphaltene on porous media, adsorption and me-
pressure, and temperature, are justified. Thermody- chanical entrapment.
namic equilibrium models such as the thermodynamic
Micellization model, and the pure solid model are used 1.3.1. Adsorption
for this purpose. The other view holds that asphaltenes Asphaltene particles physically adsorb on the rock
precipitation is irreversible since asphaltenes are dis- surface due to polarity, affinity, or other attractive
persed in suspension and stabilized in crude through force. This process is reversible with asphaltene
their interaction with resins to form colloids, so that if desorbing from porous media when the concentration
this suspension is destabilized, for example by a of suspended asphaltene in crude oil decreases. Stud-
reduced concentration of resins due to gas injection, ies on asphaltene adsorption on rock materials by a
asphaltenes precipitate irreversibly. number of investigators (Collins and Melrose, 1983;
Due to these different views, a number of ap- Dubey and Waxman, 1991; Gonzalez and Travalloni-
proaches to thermodynamically model the asphaltene Louvisse, 1993) have shown that asphaltene adsorp-
precipitation and thus define the asphaltene precipita- tion can be modelled using a Langmuir isotherm
tion envelope (APE) shown schematically in Fig. 1 equation of the form:
have emerged. They include the Liquid-Solubility Wsa;max Ka Csf
Model (Hirschberg et al., 1982; Pan and Firoozabadi, Wsa ¼ ð1Þ
Ka Csf þ 1
2000; Wang and Civan, 2001), The Thermodynamic
Colloidal Model (Leontaritis and Mansoori, 1987; where Csf is the mass of suspended asphaltenes per
Correra and Donnaio, 2000), Thermodynamic Micel- mass of the oil phase, wsa is the mass of adsorbed
lization Model (Pan and Firoozabadi, 1998, 2000), asphaltene per mass of rock, wsa, max is the maximum
and the Pure Solid Asphaltene Model (Nghiem and adsorbed mass fraction, and Ka is the ratio of adsorp-
Coombe, 1996; Nghiem et al., 1998, 2000; Kohse et tion/desorption rate constants.
al., 2000). More recent experimental evidence has The contribution of adsorption to total amounts
leaned more towards the reversibility of asphaltene deposited is typically small (Leontaritis, 1998). How-
precipitation, i.e. the thermodynamic Micellization ever, adsorption may play a significant role in mod-
model, and the pure solid models. This last model ifying the wettability of the rock and thus the relative
will be taken as basis for the proposed single well permeabilities in the zones affected by asphaltene
model. deposition.
160 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

1.3.2. Mechanical entrapment deposition as a fraction of bulk volume in the pluggable

Several empirical and physically based equations and unpluggable pores, respectively. Ali and Islam
have been proposed to model physical blocking of (1998) used only the pore-throat plugging term for
porous media by asphaltene. They are generally based pluggable pores and the surface deposition and entrain-
on the theory for fines migration and deposition in ment terms for non-pluggable pores. Nghiem et al.
porous media by Gruesbeck and Collins (1982). In (2000) used only the pore-throat plugging term, to
this model, there are two parallel flow branches, one model mechanical entrapment of asphaltene.
that has relatively small pores so that fines depositing Leontaritis (1998) presented another model for
may plug the channel, and the other where surface near wellbore formation damage due to asphaltene
deposition occurs but the pore throats are large entrapment in pores. His model considered only pore-
enough so that the channels do not plug. The distri- throat plugging and is tied to a special (Colloidal)
bution of pluggable and non-pluggable channels thermodynamic model that gives the asphaltene par-
depends on the size distribution of fines (in this case ticle size distribution, and tying it to pore throat sizes
asphaltene) and pore throats. There are three factors that would entrap a certain fraction of asphaltene
contributing to entrapment, surface deposition, en- concentration.
trainment, and pore-throat plugging. Ali and Islam
(1998) and Wang and Civan (2001) used all three 1.4. Asphaltene plugging models
parameters in modelling the asphaltene deposition
process. Their equation is of the form: The plugging model refers to the effect of asphal-
tene deposition on porosity and permeability. Various
yðEA Þ=yt ¼ aCsf ;v /  bEA ðvo  vcr;o Þ investigators have indicated that porosity is reduced
þ ci ð1 þ rEA Þuo Csf ;v ð2Þ by the amount of fractional pore volume occupied by
asphaltene deposition, i.e.:
The first term represents surface deposition, the
second entrainment of deposits, and the third the pore / ¼ /i  EA  EAA ð4Þ
throat plugging rate. In Eq. (2), EA is the volume
where /, /i are the current and initial porosities, and
fraction of deposited asphaltene in the bulk volume, a
EAA is fractional pore volume of asphaltene adsorbed,
is the surface deposition rate constant, Csfv is the
which it typically small.
volume ratio of suspended asphaltene in the oil phase,
The effect of deposition on permeability is mod-
uo is oil phase Darcy velocity, vo and vo,cr are
eled in a variety of ways. In Ali and Islam (1998)
interstitial velocity of oil phase, and the critical
model, the pluggable and unpluggable pathways have
interstitial velocity needed for entrainment, so that
different initial permeabilities, and their permeabilities
b, the entrainment rate constant, is zero when the
change in a different manner due to plugging. For
interstitial velocity is less than this critical velocity.
pluggable pores, the instantaneous permeability due to
The pore-throat plugging coefficient ci is zero when
plugging, kp, is related to the initial permeability of
the pore-throat diameter is larger than a critical
the pluggable pathways, kpi, through the following
diameter (i.e. the channel is unpluggable), and r is
empirical equation:
the snowball rate constant. When there are both
pluggable and unpluggable pores in the porous media, 4

the total plugging is the sum of plugging in the kp ¼ kpi :eaEp ð5Þ
pluggable and unpluggable pores multiplied by their
while for unpluggable pores, the instantaneous per-
fraction: meabilities with asphaltene deposition, knp, is related
to the initial permeability of the unpluggable path-
EA ¼ fEA;p þ ð1  f ÞEA;np ð3Þ ways, knpi, through the following equation:

where f is the fraction of pore space containing plug- knpi

knp ¼ ð6Þ
gable pathways, and EA,p and EA,np are the asphaltene 1 þ bEnp
 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 161

where kpi, knpi, a, and b are constants for the rock precipitation versus pressure are used directly in the
media of interest, and the overall permeability, at any fluid flow model, which treats dissolved and precip-
time, is described by: itated asphaltenes in an analogous manner to the
treatment of free and dissolved gas in black oil
k ¼ fkp þ ð1  f Þknp ð7Þ simulators. Furthermore, the current model is imple-
mented in radial geometry to accurately represent
Nghiem et al. (1998) and Nghiem et al. (2000) near-wellbore velocities, which are critical for accu-
used two correlations for the effect on permeability, rate determination of entrainment and pore-throat
the first is: plugging effects, as indicated by Eq. (2).
1=Rf ¼ ð/=/i Þa ½ð1  /Þ=ð1  /i Þb ð8Þ

where Rf is the permeability reduction factor and a 2. Single-well model for asphaltene precipitation,
and b are parameters whose values are typically 60 deposition, and plugging
and 90. The second is a linear permeability reduction
model described by either: 2.1. Mathematical formulation

Rf ¼ 1 þ ðRf ;max  1Þðwsa =wsd;max Þ The proposed single well model is an extension of
when wsd < wsd;max ð9Þ the reservoir part of the traditional black oil (oil,
water, gas) equations described by Almehaideb et al.
or, (1989). A fourth material balance equation is included
Rf ¼ Rf ;max when wsd > wsd;max ð10Þ for the asphaltene component. The model equations in
cylindrical (r– z) coordinates are as follow:
where wsd is the wt.% asphaltene deposited through Oil:
both adsorption and mechanical entrapment. 

Wang et al. (1999) used Civan et al. (1989) model 1 B Bpo BD B
rko  co þ
for permeability reduction, namely: r Br Br Br Bz



k ¼ ki ð/=/i Þ3 ð11Þ Bpo BD B /So
 ko  co ¼ þ qo ð13Þ
Bz Bz Bt Bo
where ki is the initial permeability. Wang and Civan
(2001) used a modified version of the above equation Water:
by including the valve or gate effect, fp, as a coeffi-
cient for permeability modification: 

1 B Bpw BD B
rkw  cw þ
k ¼ ki fp ð/=/i Þ3 ð12Þ r Br Br Br Bz



Bpw BD B /Sw
Currently published papers that describe asphaltene  kw  cw ¼ þ qw
precipitation, deposition, and plugging around pro- Bz Bz Bt Bw
duction wells, combine two models, a thermodynamic ð14Þ
model that describes asphaltene precipitation from
crude oil due to pressure drawdown near the well, Gas:
and a fluid flow model that also accounts for asphal- 


tene deposition, and plugging. For example, Wang 1 B Bpg BD Bpo BD
rkg  cg þ rRs ko  co
and Civan (2001) model estimates asphaltene precip- r Br Br Br Br Br
itation through the liquid Solubility Model, Nghiem et 


B Bpg BD Bpo BD
al. (2000) use the Pure Solid Asphaltene phase be- þ kg  cg þ Rs ko  co
Bz Bz Bz Bz Bz
havior model to estimate precipitation, and Leontaritis

(1998) used the Colloidal model. In the current work, B Rs /So /Sg
¼ þ þ qg þ Rs qo ð15Þ
the experimentally determined values of asphaltene Bt Bo Bg
162 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

Asphaltene: voir volume, D is the depth, positive downward, B is


 the formation volume factor, / is the porosity, and r
1 B Bpo BD B and z are the radial and vertical dimensions, respec-
rðkA þ RA ko Þ  co þ
r Br Br Br Bz tively. For the asphaltene component, SA is the sus-


Bpo BD pended asphaltene saturation (volume of suspended
 ðkA þ RA ko Þ  co
Bz Bz asphaltene particles per pore volume of the reservoir),


 RA is the dissolved asphaltenes oil ratio [std. vol./std.
B /SA RA /So EA SA
¼ þ þ þ þ RA qo vol.], and BA is the asphaltene formation volume
Bt BA Bo BA So
factor [res. vol./std. vol.].
ð16Þ The model assumes that the oil component exists
only in the oil phase and the water component exists
The saturations of the four phases are related only in the water phase, while the gas component can
through: exist either as free gas or dissolved in the oil phase.
So þ Sw þ Sg þ SA ¼ 1:0 ð17Þ Similarly, the asphaltene component exists either as
suspended solid particles in the oil phase, deposited in
and the phase pressures are related through the cap- the porous media, or as dissolved in the oil phase. The
illary pressures, which are functions of saturations model also assumes that suspended (and dissolved)
using either of Stone’s models: asphaltene flow with the oil phase at the same velocity.

Pcow ¼ Po  Pw ; Pcgo ¼ Pg  Po ð18Þ 2.2. Implementation of the model

In the above equations, S and P are the phase The above model is discretized using central in
saturations and pressures respectively for the oil (o), space and backward in time discretizations scheme
water (w), gas (g), and suspended asphaltene (A) yielding a fully implicit model, and solved using
phases; k is the phase mobility, Newton’s method. The inner boundary condition is
defined by a constant total oil flow rate for the well.
k:krl
k1 ¼ ; ð19Þ Based on this set value, the oil flow rate at each layer
l1 is defined through the ‘‘transmissivity-allocation
where l is the phase index; generally for o, w, g, and scheme’’, namely:
A. However, the mobility definition for oil and
suspended (free) asphaltene is modified here for the TWok
Qok ¼ P Qo;tot ð22Þ
assumption that both flow at the same velocity. TWok
Therefore, where TWok is the transmissivity of oil at layer k
defined by the product of the mobility and a geometric



k:kro So factor, TC, that also includes the skin effect as:
ko ¼ ;
lo So þ SA TWok ¼ TCk  kok ð23Þ



k:kro SA
kA ¼ ; ð20Þ where the geometric factor for each layer k is:
lo So þ SA
C1  k  D:z
c is the gravitational coefficient for each phase, TC ¼ ð24Þ
ðlnðr1 =rw Þ þ sÞ
g and C1 is a conversion factor for units other than S.I.
c1 ¼ q1;std  sinðhÞ; ð21Þ
gc The flow rates of the other components (w, g, and A)
is then defined as follows:
where ql, std is the phase density at standard condi-
tions, / is the declination angle. q is the source/sink TWgk
Qgk ¼ Qok þ Rsk  Qok ð25Þ
(production) rate in standard volume/time per reser- TWok
 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 163

Table 1
TWwk Summary of asphaltene solubility test
Qwk ¼ Qok ð26Þ
TWok Pressure, psia Asphaltene, mass%
5500 0.25
5000 0.23
TWAk 4500 0.17
QAk ¼ Qok þ RAk  Qok ð27Þ 4000 0.165
TWok
3500 0.16
3000 0.03
The transmissivities at the well for free gas, water, 2800 0.09
and suspended asphaltene are defined in an analogous 2300 0.1
1800 0.12
manner to the definition of oil phase transmissivity at
1200 0.63
the well. The outer boundary condition is a no-flow 500 1.18
boundary. 14.7 2.99
The above equations were solved simultaneously
using the Newton’s method on a cylindrical (r –z) grid
with a time-step control routine. The grid system ing is employed, in which the primary variable for gas
employed is depicted in Fig. 3. The system of equa- switches from Psat above the bubble point to Sg below
tions (one for each component and cell/block) gener- the bubble point pressure. Similarly, the primary
ated by above equations is solved using a standard variable for asphaltene switches from PA,sat above
band-solver. The primary variables chosen are water- the upper asphaltene precipitation pressure ( Puap) to
phase pressure ( pw) the water saturation (Sw), and SA below Puap. The system of equations is solved
either the saturation pressure or saturations for the gas iteratively at each time step for these variables and
( Psat/Sg) and asphaltene ( PA,sat/SA). Variable switch- other reservoir parameters are then derived from these

Fig. 3. Schematic of grid blocks in region near the wellbore showing the affected zone.
164 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

Fig. 4. Experimental asphaltene solubility versus pressure.

primary variables. The equations solved, as per New- The deposited asphaltene volume as a fraction of
ton’s method, in matrix notation is: bulk volume (EA), which shows on the RHS of the
asphaltene material balance equation, Eq. (16), accu-
AP¼F ð28Þ mulates over time as long as suspended asphaltene is
present in oil, and is described by:
where A is the coefficient matrix, P is the vector of
primary variables—four for each block, and F is the EA ¼ EA þ ðdEA =dtÞDt ð29Þ
forcing vector. The coefficient matrix is made up of
The deposition rate (dEA/dt) is evaluated using Eq.
the derivatives of all four material balance equations
(2) with variables evaluated at the old time step. This
for each block with respect to the primary variables,
‘‘decoupling’’ of evaluation of asphaltene suspended
while the forcing vector is the negative of the residual
in oil and asphaltene deposited in porous media is
of each material balance equation. Convergence at
done to save an additional inside iteration level and
each time step is achieved when this residual is lower
reduce the simulator CPU time with very little impact
than a set minimum convergence criteria. Other con-
on the results due to the relatively small maximum
vergence criteria are also set on the changes in
time steps allowed in the solution routine (less than 1
pressures and saturations over iterations. After con-
day) for a process that takes several hundred days.
vergence, the solution is then advanced in time and a
Nghiem et al. (2000) also used a similar decoupling in
new set of primary variables is solved for at the new
their solution scheme, while Wang and Civan (2001)
time.
did include an additional internal iteration scheme in
Table 2 each time step and evaluated the deposition rates
Basic data for run #1 using the variables evaluated at the new time step.
Top of reservoir 8400 ft The three different models for asphaltene deposi-
Base of reservoir 8550 ft tions and plugging, namely Ali and Islam (1998),
Well radius, rw 2 in.
External radius, re 10,000 ft
Initial pressure 4000 psia Table 3
Upper asphaltene precipitation pressure 3600 psia Asphaltene solubility ratio, vol/vol, used in the simulation model
Bubble point pressure 3000 psia
Initial porosity 0.12 Pressure, psia Asphaltene solubility, ft3/ft3
Initial horizontal permeability 40 md 5500 0.002229
Initial vertical permeability 12 md 5000 0.002051
API of oil 32.8 4500 0.001516
Asphaltene density 60 lbm/ft3 4000 0.001338
Reservoir rock density 165 lbm/ft3 3500 0.001427
Connate water saturation 0.22 3000 0.000268
 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 165

Fig. 5. Sand-face pressure at middle of perforations versus time for runs #1 and #2.

Nghiem et al. (1998, 2000), and Wang and Civan a- Extractionof around 3 g of asphaltene from stock
(2001) were implemented in the model, with the tank oil using the ASTM 6560 method as the n-
choice to run either available. This last model was heptane insoluble solid residue.
used in the asphaltene deposition and plugging case b- Mixing around 80 cc of stock-tank crude with the
discussed below. extracted asphaltene and placing the sample in the
visual PVT cell, and adding pressurized natural
gas to the stock-tank oil until a live –oil mixture
3. Results and discussion with a bubble point of around 3000 psi was
synthesized.
3.1. Experimental results c- Pressurizing the sample to 6000 psi and then
reducing the pressure in increments. At each
A complete set of experiments to evaluate asphal- increment, 2 –3 cc of the sample were withdrawn
tene saturation in live crude versus pressures was in a veil and analysed for asphaltene content using
conducted using the gravimetric method. The set the ASTM D-3279 method. The summary of these
consisted of the following. results is listed in Table 1 and plotted in Fig. 4.

Fig. 6. Productivity Index versus time for runs #1 and #2.

166 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

Fig. 7. Suspended asphaltene saturation versus time for near wellbore blocks, run #1.

It is interesting to note the similarity in character are listed in Table 3. Fig. 5 shows the flowing bottom-
between these results and the results reported by hole pressure versus time at the middle of the perfo-
Jamaluddin et al. (2001) shown schematically in rations where the flow behavior changes from tran-
Fig. 2, showing a region of almost flat solubility at sient to pseudo-steady state at around 200 days, and
high pressure followed by a reduction in solubility to Fig. 6 shows the changes in productivity index (P.I.)
a minimum value near the bubble point and then an for this well during the transient flow region until it
increase in solubility below the bubble point as gas is stabilizes at around 1.7 (STB/d)/psi under pseudo-
evolved and the composition of the crude is changed. steady state condition. This P.I. provides the basis for
The results of this test are used as input into the comparison with the plugging model in run #2. Fig. 7
asphaltene single well model developed in this work. shows the growth of the asphaltene precipitation zone
around the well versus time as pressure drop below
3.2. Numerical results the upper asphaltene precipitation pressure of 3600
psi. It shows that this zone grows continuously
The asphaltene solubility test results were used as outward from the well and the concentrations of
input for the single well asphaltene precipitation, suspended asphaltene go higher with time as the
deposition, and plugging model developed in this pressure drops near the well. The growth of the
work. Two test cases are discussed below. The base asphaltene precipitation zone gives an indication of
case involves asphaltene precipitation but without the zones that will be affected by plugging.
deposition and plugging. The second is made with
deposition and plugging using Wang and Civan de-
position and plugging models. This last model is
chosen because it is more general and was developed Table 4
Parameters for adsorption and deposition used in run #2
based on experimental data.
Maximum asphaltene adsorption, wsa,max 0.1 mg/g
Ratio of adsorption/desorption reaction, Ka 2500
3.2.1. Run #1: asphaltene precipitation (base case) Surface deposition coefficient, a 0.085/day
A 2000-day base run was performed using the data Entrainment coefficient, b 0.0
listed in Table 2. The asphaltene solubility data Critical interstitial oil velocity, vcr,o 100 ft/day
discussed in Section 3.1 were converted to a [vol/ Pore throat plugging coefficient, cI 0.1/ft
vol] basis and also used as input in the model. These Snowball coefficient, r 0.1/ft
 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 167

Fig. 8. Accumulation of deposited asphaltene around the wellbore for run #2.

3.2.2. Run #2: asphaltene precipitation, deposition, after about 3 years. The porosity and permeability in
and plugging the near wellbore region are shown in Figs. 9 and 10.
The same data used in the base case are run again Again, one can notice that asphaltene deposition
with Wang and Civan model for deposition and severely reduced the porosities and permeability near
plugging included. The parameters used for adsorp- the well. The flowing bottom hole pressure and the
tion and plugging are listed in Table 4. The precipi- productivity index versus time for this run are dis-
tated asphaltene quickly adsorbed on the rock surface played in Figs. 5 and 6 for comparison. The flowing
up to the maximum absorbed value used in the run bottom hole pressure was reduced due to plugging
(0.1 mg/g), but this adsorption had very little effect on from 3267 psi to 3038 psi after 1750 days, equivalent
porosity and permeability. Fig. 8 shows the accumu- to a skin of 4.04. The productivity index dropped from
lation of deposited asphaltene in the near-wellbore a stabilized value of 1.7 STB/D/psi to 1.18 STB/D/psi
region. Severe plugging has developed near the well after 1750 days, a drop of 30.6%. Beyond 1750 days,

Fig. 9. Change in porosity versus time for near wellbore blocks, run #2.
168 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

Fig. 10. Change in horizontal permeability versus time for near wellbore blocks, run #2.

more severe plugging occurred, further reducing the 3. The choice of cylindrical geometry, which
well productivity index. accurately predicts pressure drop around the well,
allowed for accurate determination of the asphal-
tene precipitation zone. The size of this region
4. Conclusions increased progressively with time as the pressure
dropped below the upper asphaltene deposition
Based on this study, the following conclusions can pressure in a larger portion of near wellbore
be stated: region.
4. Different models for deposition and plugging can
1. A four-component single well extended black oil be used in the model, provided the parameters of
model with a variable asphaltene saturation these models are representative for the well to
pressure routine has shown that it can handle the be studied. The Wang and Civan model was
phase behavior aspect of asphaltene precipitation used to demonstrate the plugging effect for the
by modelling it as the fourth component and test well. For this well, severe plugging
treating it in a similar fashion to the treatment of developed after about 4.8 years of production
free and dissolved gas in variable bubble point with over 30% drop in the productivity index
routines. This treatment allowed experimental and the development of an equivalent skin of
solubility data to be inputted directly into the about 4.04.
model. Also, the transport of dissolved and
suspended asphaltene within the oil phase is
handled consistently. Nomenclature
2. Laboratory analysis methods, using the gravimetric A Coefficient matrix
method, were used to define the upper asphaltene B Formation volume factor, ft3/ft3
precipitation pressure and asphaltene solubility in C Concentration, ft3/ft3.
oil versus pressure data, which are needed as input C1 conversion factor for other than S.I. units
for the model. These experimental measurements D Depth, ft.
are comparable to gas solubility versus pressure E Volume fraction in bulk volume,
data, derived from PVT reports, that are used in dimensionless
black-oil simulators. F Forcing vector
 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170 169

f Fraction of pores pluggable by asphaltene through the UAE University Center for Externally
deposition Funded Research (eFORS). Special thanks are due to
Ka Ratio of adsorption/desorption rate con- Samir El-Hardelo, Khalid Abdul-Fattah, and Hasan
stants, dimensionless Kamal for performing the live crude oil recombination
k Permeability, md and the gravimetric analysis to determine asphaltene
P Vector of primary variables—four for each solubility versus pressure.
block
p Pressure, psia
References
Q Flow rate, STB/D
RA Dissolved asphaltene oil ratio, ft3/ft3
Ali, M.A., Islam, M.R., 1998. The effect of asphaltene precipitation
Rs Solution gas – oil ratio, SCF/STB on carbonate-rock permeability: an experimental and numerical
Rf Permeability reduction factor approach. SPE Production and Facilities, August. Society of
S Saturation, fraction of void volume Petroleum Engineers, Dallas, Texas, pp. 178 – 183.
TC Constant part of transmissivity Almehaideb, R., Aziz, K., Pedrosa, O., 1989. A Reservoir/wellbore
model for multiphase injection and pressure transient analysis,
TW Transmissivity at the well block
paper 17941. SPE Middle East Oil Show, Manama, Bahrain,
t Time, days 11 – 14 March, 1989.
V Volume, ft3 Burke, N., Hobbs, R., Kashou, S., 1988. Measurement and model-
vcr,o Critical interstitial oil velocity for entrainment ing of asphaltene precipitation from live reservoir fluid systems,
w Mass per mass of rock paper 18273. SPE Annual Technical Conference and Exhibition,
Houston, TX, October 2 – 5, 1988.
Civan, F., Knapp, R., Ohen, H., 1989. Alteration of permeability by
fine particle processes. Journal of Petroleum Science and Engi-
Greek neering 3 (1/2), 65 – 79.
a Surface deposition coefficient, 1/day Collins, S.H., Melrose, J.C., 1983. Adsorption of asphaltenes and
b Entrainment coefficient, 1/ft water on reservoir rock minerals. International Symposium on
ci Pore throat plugging coefficient, 1/ft Oilfield and Geothermal Chemistry, Denver, 1 – 3 June 1983.
Paper SPE 11800.
r Snowball coefficient, dimensionless Correra, S., Donnaio, F., 2000. OCCAM, onset-constrained colloidal
k mobility asphaltene model, paper 58724. SPE International Symposium on
/ porosity, fraction of pore space Formation Damage, Lafayette, LA, 23 – 24 February 2000.
c gravity coefficient, psi/ft Dubey, S.T., Waxman, M.H., 1991. Asphaltene adsorption and de-
sorption from mineral surfaces. SPE Reservoir Evaluation and
D Difference
Engineering, August 1991. Society of Petroleum Engineers,
Dallas, Texas, pp. 389 – 395.
Gonzalez, G., Travalloni-Louvisse, A.M., 1993. Adsorption of as-
Subscripts and superscripts phaltene and its effect on oil production. SPE Production and
A Asphaltene Facilities, May 1993. Society of Petroleum Engineers, Dallas,
i Radial layer index Texas, pp. 91 – 96.
Gruesbeck, C., Collins, R.E., 1982. Entrainment and deposition of
k Vertical layer index fine particles in porous media. SPEJ, December 1982. Society of
l Phase index Petroleum Engineers, Dallas, Texas, pp. 847 – 856.
o, g, w Oil, gas, water, respectively Hirschberg, A., Jong, L., Schipper, B., Meyers, J., 1982. Influence
sa Adsorbed asphaltene of temperature and pressure on asphaltene flocculation paper,
sf Suspended asphaltene 11202. SPE Annual Technical Meeting and Exhibition, New
Orleans, LA, September 26-29, 1982.
uop Upper onset pressure for asphaltene Jamaluddin, A., Creek, J., Kabir, C., McFadden, J., D’Cruz, D.,
precipitation Joshi, J., 2001. A comparison of various laboratory techniques
to measure thermodynamic asphaltene instability, paper 72154.
SPE Asia Pacific Improved Oil Recovery Conference, Kuala
Acknowledgements Lumpur, Malaysia, 8 – 9 October 2001.
Kabir, C., Jamaluddin, A., 1999. Asphaltene characterization and
mitigation in south kuwait’s marrat reservoir, paper 53155. SPE
This work was part of a research project funded by Middle East Oil Show, Manama, Bahrain, 20 – 23 February
the Abu-Dhabi National Oil Company (ADNOC) 1999.
170 R.A. Almehaideb / Journal of Petroleum Science and Engineering 42 (2004) 157–170

Kohse, B., Nghiem, L., Teikoku, H., Ohno, K., 2000. Modelling SPE Annual Technical Conference and Exhibition, New Or-
phase behavior including the effect of pressure and temperature leans, LA, 27 – 30 September 1998.
on asphaltene precipitation, paper 64465. SPE Asia Pacific Oil Nghiem, L., Kohse, B.F., Farouq Ali, S.M., Doan, Q., 2000. As-
and Gas Conference and Exhibition, Brisbane, Australia, 16 – 18 phaltene precipitation: phase behaviour modelling and composi-
October 2000. tional simulation, paper 59432. SPE Asia Pacific Conference on
Leontaritis, K.J., 1998. Asphaltene near-wellbore formation damage Integrated Modelling for Asset Management, Yokohama, Japan,
modeling, Paper 39446. Lafayette, LA. SPE Formation Damage 25 – 28 April 2000.
Control, February 18 – 19, 1998. Pan, H., Firoozabadi, A., May, 1998. Thermodynamic micellization
Leontaritis, K.J., Mansoori, G.A., 1987. Asphaltene deposition dur- model for asphaltene aggregation and precipitation in petroleum
ing oil production and processing: a thermodynamic colloidal fluids. SPE Prod. and Facilities. Society of Petroleum Engineers,
model, paper 16258. SPE International Symposium on Oilfield Dallas, Texas, p. 118.
Chemistry, February 4 – 6, 1987, San Antonio, TX. Pan, H., Firoozabadi, A., February 2000. Thermodynamic micelli-
Mohammed, S., Arisaka, K., Kumazaki, Y., 1996. Integrated anal- zation model for asphaltene precipitation from reservoir crudes
ysis of asphaltene deposition from field production data and at high pressures and temperatures. SPE Production & Facilities
laboratory experiments, paper 49501. SPE Abu Dhabi Interna- 15 (1), 58 – 65.
tional Petroleum Exhibition and Conference, Abu Dhabi, Wang, S., Civan, F., 2001. Productivity decline of vertical and
U.A.E., 11 – 14 October 1996. horizontal wells by asphaltene deposition in petroleum reser-
Nghiem, L., Coombe, D., 1996. Modelling asphaltene precipitation voirs, paper 64991. SPE International Symposium on Oilfield
during primary depletion, paper 36106. SPE Latin American/ Chemistry, February 13 – 16, 2001, Houston, TX.
Caribbean Petroleum Engineering Conference, Trinidad and To- Wang, S., Civan, F., Strycker, A.R., 1999. Simulation of paraffin
bago, April 23 – 26, 1996. and asphaltene deposition in porous media, paper 50746. SPE
Nghiem, L., Coombe, D.A., Farouq, A., 1998. Compositional sim- International Symposium on Oilfield Chemistry, February 16 –
ulation of asphaltene deposition and plugging, paper 48996. 19, 2001, Houston, TX.