E

SPE 36740
Giilfi
6-’ m
.

ScttSetyd Pdoleum
. .

Ettgltwem

Pressure and Composition Effect on Wax Precipitation: Experimental Data and Model
Results
Huanauan Pan, Abbas Firoozabadi, SPE, Reservoir Engineering Research Institute; Per Fotland, Norsk Hydro
Production, A.S.

Copyright 199S, Society oi Petrc4eum Engineem, Inc

alkane mixtures using calorimetry and microscopy.’ He found
Thii papw was prc$amct la presentation M tha 193s SPE Annual Technbal calerence and that the precipitation is unstable and segregates into two solid
ExhbMon hekl in OenWr, Co&ado, IJS.A,. H October 199S.
phases when the chain-length difference between the two
Thrn papa was sebdd lo! ww4ntd0n ky m SPE Prqpm Committee Iollowiig rawiaw cd
information contained in m mtmtrad stied & the authof( G). Contents d tho F+OI, M alkanes exeeeds a f~ed value. He also found the segregated
presented, km not teen mviawd by tha Sadoty ot Petrobum Engrneem and am sutajd t.
correction ty tha author(s). The nwteriat, as presented, doos nci -.sNY retkt any
phases consist predominantly of pure components. Reeently,
pdii 01 the %ckty cd Petmbum Enginaem, ~s offkem, or rmmbers. Pvm pmsentod at Snyder et al.2’34studied the kinetics of the segregation using
SPE meetings am subject to publication revi.w ty Edtwial Cwnmiltees oi the S06ety C4
Petrobum Engkeem. Permission to m?py is reslricied to an abstract ot IW4 nmre than !3KI spectroscopy, ealorirnetry and electron diffraction, They
words. Illustrations may not be cap+od, The abet ract ohould omtain COIWPIIWOIJS
duw+dodgnwt C4whom md by +om the paper was pmmmt.d. Wrhe Lbrwkn, SPE, P.O. observed that the rate of the segregation is very sensitive to
%x B3383S, Richardson, TX 7=33S3S, U.S.A., k 01.214-S52-9435. the chain-length difference. Hansen et al.s observed phase
transitions of the precipitated wax from the North Sea erudes.
Abstract Based on these observations, Lira-Galean~ Firoozabadi, and
Wax precipitation is often studied using the stock tank oil. Prausnitz6developed a thermodynamic multisolid wax model.
However, precipitation may be very different in well tubing The calculated results from this model were in agreement
and production facilities due to the effects of pressure and with data of stock tank fluids from the North Sea given by
composition. As an example, the cloudpoint temperature may W.B. Pedersen et al? The model of Lira-Galeana, et ai.6
decrease as much as 15 K from atmospheric pressure to the lumps all the various chemicaJ species for a given carbon
saturation pressure of 100 bar mostly due to the dissolution of number and avoids the use of paraffins (P), naphthenes (w
light gases into the oil (i.e. due to composition changes ). It is and aromatics (A) by assigning average properties such as the
also often assumed that the addition of solvents such as C5 melting point temperature, heat of fusion, and critical
and C6decreases the cloudpoint temperature. @ the contrary, properties. However, K.S. Pedersen, et al.* and Ronningsen et
from our modeling results, we have found that the mixing of al.g have shown that the wax precipitated from a petroleum
a erode with a solvent increases the cloudpoint temperature fluid consists primarily of normal paraffins, iso-paraffms, and
(i.e., enhances the wax precipitation). naphthenes. The aromatics do not precipitate as the wax. The
In this study, the cloundpoint temperature at live oil use of PNA analysis avoids assigning average properties to
conditions and the amount of the precipitated wax at stock the carbon numbers. We have recently observed that certain
tank oil conditions are provided for three crudes. A modifkxl light oils and condensates with an API gravity of 45 may have
multisolid wax precipitation model is used to study the effects a cloudpoint temperature as high as 333 K. Such a high
of pressure and composition on wax precipitation. The cloudpoint temperature might be attributed to a very high
modeling results reveal that an increase in methane and C02 concentration of paraffins and naphthenes. Accordingly, a
concentration decreases the cloudpoint temperature while an predictive model should consider the PNA analysis to account
increase in C5 concentration increases the cloud point for physical properties of various hydrocarbon spezies.
temperature. Most of the literature data on wax precipitation are
measured at stock tank oil (dead oil) conditions. Since the
Introduction concentration of light components at high pressures has a
Wax formation is a major problem in well tubing and subsea significant effect on cloudpoint temperature, a proper
pipelines. The current methods which include the interruption predictive model and experimental data for reseIvoir fluids
of production to scrub the deposited solid are very costly. A will be very useful in investigating the wax formation
thermodynamic predictive model is important in solving the problems at well tubing, production facility, and pipeline
problem. conditions. The experiments by Weingarten and Euchner’O,
Dorest studied the solid precipitated from binary normal Colgate and Sivarnamanl’, and Brown, et al.’z reveal that the

579
2 H, PAN, A. FIROOZABADI, P. FOTIAND SPE 3S740

cloudpoint temperature of a live oil decreases as the fraction of a crude can be divided into paraffins (P),
saturation pressure increases. Brown et SI.12also state that the naphthenes (N), and aromatics (A). PNA distribution is a key
cloudpoint temperature increases linearly with pressure for to wax precipitation modeling and average properties of a
any crude, Brown, et al.]z provide the detailed compositional fraction, which have proved very successful in vapor-liquid
data for one crude, but other authors do not provide the calculations of crude systems, may not be appropriate for wax
compositional data which is the required input to an equation precipitation prediction. In this study, we dhide each
of state (EOS)-baswl thermodynamic model undefined component of crude into three speciex normal
The effects of pressure and composition on wax paraffins (P), naphthenes and iso-paraffins (N) and aromatics
precipitation currently are not clear. The common belief is (A). We also develop the correlations for estimating fusion
that the addition of intermediate components such as C4and properties for these species.
C5 may prevent wax precipitation. In some cases, even the
mixing of a crude with condensate liquids has been tried. Fusion Properties. The melting-point temperature, ~’, the
Reality may contradict this belief.
The purpose of this work is to study the effects of enthalpy of fusion, Ah/ , and the heat capacity of fusion,
composition and pressure on both the cloudpoint temperature ACti, are evaluated for P, N, A species according to the
and the amount of precipitated wax. For this purpose, we fmt following correlations:
investigate wax composition. We then modify the multisolid We use the correlations proposed by Won]4 to estimate
model of Lira-Gakana, et SL6Next we present experimental T’ and Ah/ for normal alkanes:
results on wax precipitation for three reservoir fluids at stock
tank and live conditions. The calculated results for these three ~f = 374.5+0.02617Mi _20172/~i ...................(1)
oils using the modit%xlmuhisolid model are presented and
compared with data. We then discuss the effects of pressure Ah{ =0.1426A4Wi~f .................................................(2)
and composition and offer simple theoretical interpretations,
followed by conclusions and recommendations The correlations proposed by Lira Galeana et al! are
used to calculate ~.f for naphthenes, iso-paraftlns and
Wax Composition
Wax precipitation has often been described as paraftln aromatics species, and Ah/ for naphthenes, and iso-
precipitation in the literature. It has also been established paraffins:
experimentally that in the precipitated wax, only
hydrocarbons heavier than, say, C15to Cm are present13. In ~f = 333.46-419.01 exp(-0.(K18547MWi) ................ (3)
order to ascertain the wax composition, we have considered a
model synthetic oil with the composition shown in Table 1. A~r = 0.05276~i~f ..........................................(4)
Figure 1 illustrates the wax precipitation process as the
Figure 2 (data source: Ref. 15) shows that the entropy of
temperature demases at 1 atm. The model used to perform
the calculations will be presented later in this paper. fusion for aromatics, Al/ / ~f, is atmost independent of
Examination of the results presented in Fig. 1 reveals that (to molecular weight. We propose the following correlation to
a temperature of 235 K): estimate the enthalpy of fusion for aromatic species
1. The precipitated wax does not contain aromatics. Ah\ = 11.2~f .............................................!...o..........(5)
2. The normal paraffins with the same carbon number as
The correlation proposed by K.S. Pedersen et at.]s is
the naphthenes precipitate fiit.
tiopted to estimate ACPi for dl P, N, ad A species:
3. At high temperatures, n-paraffins essentially constitute
the wax phw, however, at lower temperatures, naphthenes
contribution becomes important. ACfi = 0.3033~i –4.635x 104~iT ..................... (6)
4. The lightest component that can be found in the wax
phase is P-CMwhich precipitates around 237 K. Critical Properties and Acentric Factor. The correlations
All of the above results are in line with the experimental for critical properties of Cavett”, and Lee and Kesler’* are
observationsag and are predicted by our modification of the based on the properties of light hydrocarbons (< Cm ) and
multisolid wax model. their application to heavy hydrocarbons is questionable. The
recent work of Riazi and A1-Sabhaflgis more suitable for our
Modiflcatlon of Multkolid Wax Model
PLUPOSC. These authors provide a general correlation of
The multisolid model of Lira-Galeana, et SI.s is based on the critical properties and acentric factors for P, N, A species.
precipitation of certain heavy components of the crude with The heavy hydrocarbon data can be obtained by extrapolating
average properties, such as enthalpy of fusion, assigned to their correlation. We found the extrapolation gives values that
each fraction. According to the molecular structures, each are too low for the critical pressure, Pc, of heavy

580
SPE 3S740 PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 3

hydrocartmns. The following correlation is then proposed: expansivity will increase with decreasing temperature.
Therefore the cloudpoint temperature can be obtained by
P=i= A – Be=mi ..................................................... (7) beating the minimum on the expansivity-temperature plot.
to estimate Pc of a hydrocarbon with the mohndar weight of In the measuring process, the cell containing the sample is
more than 300 glmol. Coefficients A, B, and C in ~. (7) for placed in a system of five heating ovens which are all
P, N, and A speciesare listed in Table 2. The correlation of connected to a personal computer in order to stabilize the
Riazi and A1-Sahhaf is adopted for the critical temperature, temperature of the sample to within 0.005 K. The temperature
Tc, and critical volume, Vc, in this work. is controlled stepwise. The sample is considered to be at
The acentric factor of aromatics from Riazi and At- equilibrium when no change in temperature or pressure
Sahhafs correlation is also found to be too high for heavy occurs over a given period. The pressure is controlled to
components. We propose to estimate the acentric factor of within 0.01 bar and the volume is recorded from the position
aromatics from of the piston (resolution 0.0001 cc). ‘f’hewhole experimental
procedure is automated, The method has been verified by
ln(oi = –36,1544 – 30.94MW~0XX1( MWi s 800 ) ... (8) comparing data on stabilized erodes and by high pressure
measurements on a transparent mixture in visual cells.
When MWi 2 800, we set @i= 2,0. Riazi and A1-Sahhafl
Details on the method with results both on binary, ternary and
expression for the acentric factor of paraffins and naphthenes crude systems will be published later. The measured
is used in the study. cloudpoint temperatures using this procedure are listed in
Table 7 for the two crudes.
Measurements A temperature controlled ultracentrifuge was used to
Three crudes were used in the experimental study of wax measure the precipitated amount of wax for the stcck tank oil.
precipitation. These erudes are denoted Oil #1, Oil W and A given amount of the crude is centrifuged at a high speed for
Oil #3 in this paper. The measurements and results are 24 hours at a given temperature (in the range of 279 to 323
deserked as follows: K). The supematant is carefully removed and the precipitated
material is weighed and analyzed by a high temperature gas
Oil #1. The cloudpoint temperature measurements for Oil #1 chromatography(GC), which can analyze normal alkrmes to
were made by viscomeuy. Four cloudpoint temperatures were C*O.As there will be some co-precipitated oil, it is necessary
measured: three for the crude saturated with a separator gas at to perform a correction according to the GC analysis. The oil
saturation pressures of 34, 68 and 102 bar at 311 K and one part of the chromatogram is seen as peaks with decreasing
for the stock tank oil. The compositions of the crude for the intensity as a function of elution time. The wax components
saturated conditions and for the stocktank are listedin Table are normally observed as ‘lump’of peaks at later elution times
3. The measured data are listed in Table 7. than the oil peaks. Upon integration, the oil part can be
subtracted from the chromatogram and what is left is
Oil #2 and Oil #3. The compositions for these two North Sea regarded as the ‘pure’ wax. The measured amount of the
crudes both at stock tank and at reservoir conditions are precipitated wax for the two crudes at different temperatures
providtxt in Tables 4 and 6, respectively. The detailed are listed in Table 7.
analysis of C1O+for Oil #2 is listed in Tabte 5 where the
amount of normal paraffins was obtained by high temperature Modeling Results and Comparison with Dsta
gas chromatography. The TBP analysis of Clw for Oil #3 is We used the modified multisolid model deseribed aime to
listed in Table 6. calculate the wax precipitation of Oil #1, Oil #2 and Oil #3.
A new teehnique was applied to detect the cloudpoint The Peng-Robinson equation of state (PR-EOS)m is used in
temperature at elevated pressures for Oil #2 and Oil #3. The this study. The binary interaction coefficients, kg between
method is based on measuring the volume as a function of hydrocarbons rue calculated from:z’
temperature at constant pressure. From the recorded vatues of
*vytlvve *
volume and temperature it has been found that the thermat
expansivity is a constant for a stock tank oil. Below the
cloudpoint temperature, the expansivity increases with
kf=l.O-
[)
%
v:. + V*
..................................................(9)

deereasing temperatur~ from the changes in curvature of the where VCis critical volume estimated using the correlation in
graph one can obtain the cloudpoint temperature. When the Ref. 19. The exponential factor 0 in Eq. (9) is determined by
crude contains gas, the thermal expansivity is no longer a matching the bubble point pressure of reservoir fluids; 9
constant, it is a function of temperature. We have observed, values are 1.0, 0.7 and 0.9 for Oil #1, Oil #2 and Oil #3,
for the bottom hole samples of different reservoir fluids above respectively. The ki value between C02 and Nz is set to -0.02.
the cloudpoint temperature, the expansivity will increase as Table 8 lists the ky values between COZ and hydrocarbons,
temperature increases. Below the cloudpoint temperature, the and between N2and hydrocarbons. The ktivalue between COZ

581
4 H. PAN, A. FIROOZABADI, P. FOTIAND SPE 36740

and hydrocarbons heavier than Cj is set to O.1S0 and the cloudpoint temperature of 011 W. However, the decrease of
value between N2 and hydrocarbons heavier than cc is set to the cloudpoint tempemture with the saturation pressure in Oil
0.120. We use these values of the interaction coefficients both #3 is less than that in Oil #1 and Oil #2.
in vapor-liquid equilibrium (V-L-E) and liquid-solid Figures 4 and 5 show the amount of the precipitated
equilibrium (L-S-E) calculations. Brown et al.*2 report that phase for Oil W and Oil #3, respectively. There is good
for the Soave-Redlich-Kwong (SRK)Z EOS, they had to use agreement between model results and measurements at
different interaction coefficients for V-L-E and L-S-E. Due to atmospheric pressure for both oils. These two figures also
this reason, they used the Simplified Perturbed Hard Chain show the calculated results at live oil conditions for Oil #2
Theory (SPHCT)23EOS. For our model with the PR-EOS, the and Oil #3; the amount of the precipitated wax decreases with
same interaction coefficients are used for V-L-E and L-S-E. the increasing saturation pressure.
The calculated results and comparison with the data for the
three crudes are described as follows: Effact of Pressure and Composition
We first study the effect of pressure and the addition of light
Characterization. The two parameter Gamma distribution components on the cloudpoint temperature of the model
function” was first used to generate the molar disrnbution of synthetic oil, which was described above; the composition is
the plus fraction of the crude, The distribution parameters are listed Table 1. Figure 6 shows the effect of pressure on the
a =1.68, 2.25, 3.00 and ~ =147.024, 72.03 and 28.43, for cloudpoint temperature of the model synthetic oil (the solid
the plus fractions of Oil #1, Oil #2 and Oil #3, respectively. line) which has no methane, and its n-pentane content is 40
Then the distribution was lumped into two or three percent (sex Table 1). Note that the symbols from figures 6
pseudocomponents of nearly equal weight. Hydrmarbons CIO to 12 are not experimental data, they are the calculated values
to the plus fraction were rdso lumped in groups of five and must lie on the curves. The cloudpoint temperature for
consecutive carbons. The PNA distribution for Oil #1 and Oil the original oil at atmospheric pressure is 331.3 K but as the
#3, and the NA distribution for Oil #2, are not available. pressure rises, it increases to 333.4 K at 500 bar. This result
Therefore, we have to estimate the distributions. Erickson, et and other calculations to be discussed, all indicate that the
aL25have shown that the amount of normal paraffins follows pressure increase causes the cloudpoint temperature to
an exponential decay with carbon number using high increase for a liquid mixture of fixed composition. The effect
temperature gas chromatography; the amount of normal of composition is observed by studying the cloudpoint
rdkanes becomes negligible at very high molecular weights. temperature of the model synthetic oil also shown in Fig. 6
We used an exponential decay approximation to estimate the when the oil is mixed with various hydrocarbons. The
distribution of normal alkanes of all pseudocomponents for introduction of Cl decreases the cloudpoint temperature.
Oil #1 and Oil #3. The aromatic distribution of the Unexpectedly, the increase in the amount of solvents such as
pseudocomponents for the three crudes is approximated CS increases the cloudpoint temperature. Figure 7 shows the
according to the experimental data by Jaubert, et al,2GWe also precipitated amount at different methane concentrations at
assume that the fraction is resin-like if its molecular weight is 110 bar. The dissolution of methane not only decreases the
between 800 to 950 g/mol, and is asphaltene-like if its cloudpoint temperature, but also decreases the relative
molecular weight is more than 950 g/mol. Therefore, the amount of solid phase. Figure 8 depicts the precipitated solid
amount of paraffins and rtaphthenes is set to zero for the amount and the precipitated solid phases when the model
heaviest pseudocomponents with molecular weight more than synthetic oil is diluted by C5; the cloudpoint temperature
800 g/mol. The results of characterization for the three crudes increases about 5 K for 909i0of C5 in the model synthetic oil.
are shown in Table 9. The identity of precipitated solids also changes and solid
naphthenes appear at higher temperatures for the diluted
Results. Figure 3 shows the calculated results and the mixture, compared with the originat oil. Figures 6 and 8
experimental data of the cloudpoind temperature vs. imply that the addition of solvents such as CS is not beneficial
saturation pressure for the three oils. There seems to be a in reducing the cloudpoint temperature. However, the amount
substantial effect of pressure and composition on the of the precipitated wax may decrease with increasing solvent
cloudpoint temperatttm of Oil #1. The calculated cloudpoint concentration. An explanation will be provided later.
temperature at 100 bar is 11 K less than it is at 1 bar and the Figure 9 shows the effect of pressure on the cloudpoint
measured temperatures reveal even larger disparity (15 K). temperature of Oil #1, Oil #2 and Oil #3 for the stock tank
This difference is due to the combined effect of pressure and composition. As the pressure increases, the cloudpoint
composition, Later we will show the effect of pressure and temperatures of Oil #1 and Oil #2 increase by about 1.5 K per
composition separately. 100 atrn, pressure has less effect on the increase of the
Figure 3 shows that the agreement b?tween the data and cloudpoint temperature for Oil #3 than it does for Oil #1 and
calculated results for Oil #2 and Oil #3 is good. There is a Oil #2. Figure 9 also shows that the cloudpoint temperature
significant effect of pressure and composition (about 12 K) on increases more quickly at low pressures than at high

582
SPE 36740 PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 5

pressures. These results are different from the work of Brown from their model is not, therefore, verified.
et al.12 Next we examine the effeet of diluting a reservoir fluid
Figure 10 illustrates the effect of composition on the with a gas such as Cl, and a solvent such as C5, on the
cloudpoint temperature of Oil #1. The mixing of the stock oil cloudpoint temperature at constant pressure, The liquid
with pure hydrocarbons results in an increase in the fugaeity of the first precipitating component, ~i~, is equal to
concentration of the added hydrocdmn. Similar to Fig. 6, the fugacity of its pure solid at the cloudpoint temperature,
methane lowers the cloudpoint temperature. The addition of
30 mole prcent methane lowers the cloudpoint temperature J?= &.(~,~) .....................................................(n)
from 294 K to 285 K when the pressure is about 100 bar.
Pressure increase may partly compensate for methane Let’s denote the liquid fugacity of the fmt precipitating
concentration. When 10 percent methane is dissolved in the component by f;’when the crude is diluted at constant
crude, the cloudpoint temperature decreases by 3 K, however, temperature and pressure. The fugaeity of the solid does not
as the pressure increases atx)ve the saturation pressure, the change with the dilution beeause pressure and temperature
cloudpoint temperature may exceed that of the swk tank oil. are kept constant. If f~f is less than ~~, then f~’is also less
F~. 10 shows that the increase in the amount of C3 has a
small effect on cloudpoint temperature. For hydrocarbons than fi~We(T, P), from Eq.(1 1). In this case, the volubility of
such as Cs and Cg, there is no benefit in mixing them with the the precipitating component in solution increases.
crude and in fact they increase the cloudpoint temperature. Consequently, the cloudpoint temperature decreases. On the
Figure 11 depicts the effect of Cl, Nz and COZamount on other hand, if ~V’ is larger than f~, the sohrbility of the
the cloudpoint temperature for Oil #l. It is interesting to note
precipitating component deereases in the crude, i.e., the
that the dissolution of C02 can significantly lower the
cloudpoint temperature increases.
cloudpoint temperature due to its high volubility at low
The expression of fugacity of the fust precipitating
pressures. Nitrogen can also lower the cloudpoint
component in liquid can be written as:
temperature.
Figure 12 shows the effect of Cl, Nz and C02 composition
on the amount of the praipitated wax for Oil #1 at 220 bar. Al = ~i#):p ..............................................................(l2)
This figure reveals that dissolving methane, carbon dioxide
and nitrogen all decrease the amount of the precipitated wax. After dilution, the fugacity expression becomes:
Next, we provide thcoretictd analysis of the pressure and ~“ = x;~:p ............................................................ (13)
composition effects observed above.
Combination of Eq.(12) and Eq.(13) yields
Theoretical Interpretations
We fwst examine the effeet of pressure on the cloudpoint
temperature. The volubility of the first precipitating
~’ x; $;
—.—— ........................................................... (14)
component which is expected to lx a heavy normal paraffin, ~’ Xi ~~
at the cloudpoint temperature can be expressed as:z’

..................................(10)precipitating
Due to dilution, x; < xi. Since the interaction between the
‘i=
F(T)exp[-p(v:~uJ)l heavy component and a light component such as
methane (both in the liquid phase) is very strong, the fugacity
where xi is the mole fraction (volubility) in the liquid and
coefficient @:1 often decreases due to the dissolution of
F(T) is only a function of temperature. The liquid molar
volume, Vi’, is a weak function of pressure and the change of methane, i.e., $ ~ e $ ~. Therefore, x~$~ < ~i$ ~, or ~~1e ~’ .
Consequently, dissolving methane or other light component
solid molar volume, Vi’ , with pressure is negligible. Pressure
in a crude results in the decrease of cloudpoint temperature,
appears explicitly in Eq. (1O), and its increase leads to the Now let’s examine the change of the fugaeity coefficient of
decrease of volubility of the precipitating component. In other the precipitating heavy component when a solvent such as Cs
words, the cloudpoint temperature will increase as a result of or C/J is added to a crude, The fugacity coefficient often
pressure increase. The liquid molar volume often varies more
increases by diluting a crude with C5 i.e., $:1 > t$~. This
quickly at low pressures than itdoes at high pressures. We
can deduce from Eq.( 10) that the decrease of cloudpoint increase in $: is such that xj*+’/ > xj$~ , i.e. fi’1> j’.
temperature is more pronounced at low pressures than it is at Therefore, dilution of a crude with Cs causes the cloudpoint
high pressures. Figs. 6 and 9 confii this deduction. The temperature to increase. This deduction contradicts the
Iinew relation between cloudpoint temperature and pressure common belief that the addition of a solvent will result in
for any stock tank oil, which was presented by Brown et al. 12

583
6 H. PAN, A. FIRCXXABADI, P. FOTIAND SPE S6740

decrease of the cloudpoint temperature. 1 = liquid phase

Tlte practical consequence of the above analysis is that the s = solid phase
addition of solvents, and mixing of cmdes with condensates
may result in wax precipitation enhancement and should be Subscripts
avoided. A = aromatic
c = critical property
Conclusions and Recommendations i =component i index
The major conclusions drawn from this work are: j = component j in&x
1. Results based on our thermodynamic model N = naphthene and iso-para#in
demonstrate that dissolving components such as methane and P = normal parafin
CQ in a crude decreases the cloud point temperature
significantly. On the other hand, the cloudpoint temperature Acknowledgments
increases when the crude is diluted with solvents such as C5 We thank Norsk Hydro and other member companies of the
and Ct. Due to the SbOng CffCCt of composition on the wax Reservoir Engineering Research Institute for the financial
precipitation, results from stock tank measurements can not support. Amwo Production Company provided data for one
be applied to high pressure live oil conditions. of the three samples. Dr. Michael Riley’s help in the
2. Our modeling results also show that normal paraffins preparation of the paper and his comments are appreciated,
readily precipitate, followed by naphthenes and iso-paraffms
while aromatics stay in the liquid phase and are not present in References
the wax. The cloudpoint temperature is strongly influenced by 1. Dorset, D.L.:’’Chain Length and the Cosolubility of n-Paraffii
heavy normal paraffins while the amount of the precipitated in the Solid State~ Macromolecules (1990) 23,623.
wax depends on the concentration of n-paraffins as well as 2. Snyder, R.G., Gob, M. C., Snvastsavoy, V.J. P., Strauss, H. L.,
and Dorsct, D.L.: “Measurement of the Growth Kinetics of
iso-paraffins and naphthenes.
Microdomains in Binary n-Alkane Solid Solutions by Infrared
In view of the significant difference in the physical
Spcctmscopy~ J. Phys. Chem. (1992) %, 10008.
properties of the paraffins, naphthenes, and aromatics of a 3. Snyder, R.G, Conti, G., Strauss, H. L., and Doraet, D.L.:
given carbon number in regard to wax precipitation, we ‘Thertnally-ftrduced Mixing in Partially Microphage Segregated
recommend PNA analysis be provided for wax precipitation Binary n-Alkane Crystals,” J. Phys. Chem. (1993) 97,7342.
studies, Since hydrocarbons with a carbon number of 15 and 4. Snyder, R.G., Snvataavoy, V.J.P., Cates, D.A., Strauss, H.L.,
higher appear in the wax, the analysis to C@ or Cm is also White, J.W., and Dorset, D.L., “Hydrogen/Deuteriurn Isotope
recommended. The n-paraffin analysis to Cw or beyond, can Effects on Microphage Separation in Unstable Crystalline
be currently done13. Aromatic analysis to C4Shas also been Mixtures of Binary n-Alkanes,” J. Phys. Chem.(1994) 98,674.
reported recently by Jaubert, et al. using NMR.~ 5. Hansen, A.B., Larsen, E., Pederscn, W,B., Nielsen, A.B., anc
R@mingsen, H.P.: “Wax Precipitation from North Sea Crud
Oils. 3. Precipitation and Dissolution of Wax Studied by
Nomenclature
Differential Scanning Calorimetry,” Energy & Fuels (1991) 5,
f = figacity, bar 914.
ku = interaction coefj%ient between components i and 6. Lira-Galesna, C., Firoozabadi, A., and Prattanitz, J.M.:
j “Thcrrnodynarnics of Wsx Precipitation in Petroleum
A-W = molecular weight, glmoi Mixtures,” AIChE J. (1996) 42,239.
P = pressure, bar, atm. 7. Pedersen, W.B., Hansen, A.B., Larsen, E., N]elsen, A.B., and
T = temperature, K R@nningsen, H.P.: “Wax Precipitation from North Sea Crude
v=molar volume, Cclmcd Oils. 2. Solid-Phase Content as Function of Temperature
Determined by Pulsed NMR;’ Energy & Fuels (1991)5, 908.
x = mole fraction
8, Pedersen, K.S,, Fredenslund, Aa., and Thomassen, P,:
properties of OIIS and Natural gases,” Gulf publishing Co.,
Greek Letters Houston, TX(1989), Chap. 14.
A CPi = heat-capacity offusion of component i, cal/K- 9. R@tstingsen. H. P., Bj@mdal, B., Hansen, A.B., and Pederaen,
mol W.B.: “Wax Precipitation from North Sea Crude Oils. 1.
Ah{ = enthalpy offision of component i, caltmol Crystallization and Dksolution Temperatures, and Newtonian
and Non-Newtcmian Flow Properties,” Energy & Fuels (1991)
9 = exponential factor in Eq. (9)
5, t195.
4 = fugacity coefficient 10. Weingarten,J. S., and Euchner, J.A.: “Methods for Predicting
to = acentric factor Wax Precipitation and Deposition,” SPE Production
Engineering (Feb. 1988) 121.
11. Colgate, S.0., and Sivaraman, A.: “Acoustic Resonance
Superscripts
Determination of the Effect of Light Hydrocarbons on Wsx
f = melting point
Appearance Points in a Njord Well Fluid,” Paper presented at

584
SPE 36740 PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 7

the Twelfth Symqmsiurn of ‘Ilmrnophysicsd Pmperdes,

Boulder, Colorado, Jun. 19-24,1994. TABLE 1- COMPOSITION AND MOLECULAR
12. Brown, T. S., Niesen, V.G., and Erickson, D.D.: ‘The Effect of WEIGHT OF THE MODEL SYNTHETIC OIL
Light Ends and High pressure on Paraffin Formation,” paper
SPE 28505 presented at the 1994 SPE hnual Technical
Conference and Exhibition, New Orleans, Sept. 25-28. GQtnQ. MY
“P-C5 40 72
13. Philp, R.P.: “High Temperature Gas Chromatography for the “P-clo 5 144
Analysis of Fossil Fuels: A review,” J. High Resolution “N-C1O 5 142
Chromatography (1994) 17,398. “A-C1O 5 136
14. Won, K.W., “’lkrrnodynarnica for Solid Solution-Liquid-Vapor P-c 15 5 212
E@ilibrin Wax Phase Formation !Yom Heavy Hydrocarbon N-C15 5 210
A-C 15 5 204
Mixtures,” Fluid Phase Equilibria (1986) 30,265.
P-C20 2 264
15. American Petroleum Institute, Selected Values for Physical and N-C20 3 282
lltetmcsdynamic Properties of Hydroadwrts and Related A-C20 2 276
Compounds, Rcs. Proj. 44, Texas A & M. University, College P-C25 1 352
Station, TX, Section m (1964). N-C25 3.5 w
16. Pedersen, K.S., Skovborg, P., and R@mingsen, H.P.: “Wax A-C25 2.5 344
P-C30 1 424
Recipitation from North Sea Crude Oils. 4. Thermodynamic
N-C30 2 422
Modelirrg,” Energy& Fuels (1991)5, 924. A-C30 2 416
17. Cavett, R.H.: “Physical Data fm Distillation Calcrdations- P-C35 1 492
Vapor-Liquid Equilibrium,” Proc., 27th. API Meeting, San N-C35 2.5 490
Francisco, CA (1964). A-C35 1.5 464
18. Lee, B,I., Kesler, M. G.: “A Generalized Thermodynamic P-C40 0.5 562
Correlation Based on Three Parameter Correspondirtg States,” N-C40 1 560
A-C40 1.5 554
AIChE J. (1975) 21,510.
P-C45 0.5 632
19. Riazi, M. R., A1-Sahhaf, T.A.: “Physical Properties of n- N-C45 1 m
Alkanes and n-Alkyfhydmcmbmw Application to Petioleum A-C45 1.5 624
Mixtures,” Ind. Ertg. Chem. Res. (1995) 34,4145.
20. Robinson, D.B., Peng, D.Y. and Chung, S.Y.-K.: “The
Development of the Peng-Robinsorr Equation and Its ‘P-nonnat
prllmm
Application to Phase Equilibrium in a System Containing ‘N.iso-para!Tm
and naphthcne
‘A-aromatic
Methanol,” Fluid Phase Equilibria (1985) 24,25.
21. Chueh, P.L. and Prausnitz, J.M.: “Vapor-Liquid Equilibria at
High Pressures: Calculation of Partial Molar Volume in Non-
Polar Liquid Mixtures;’ AIChE J. (1967) 13, 1099. TABLE 2- COEFFICIENTS OF EQ, (7)
22. Soave, G.: “’Equilibrium Constants fmm a Modified Redlich-
Kwong Equation of State: Chem. Eng. Sci. (1972) 27,1197. Eaaffio Nyttl&B eyurdc~
23. Kim, C. P., Vimalchand, P., Donohue, M.D. snd Sandier, S.1.,: A 0.679091
B -22,1795 -27.6292 -42.9311
‘local Compositims Model for Chairdike Molecules: A New
c 0.00264174 o.oo4495tx 0.00561927
Simplified Version of the Perturbed Hard Chain Theory,”
AIChE J. (1986) 32, 1726.
24. Whitson, C. H., “Characterizing Hydnxarbon Pfus Fractions,”
SPEJ (Aug. 1983) 683.
25. Erickson, D. D., Niesen, V.G., and Brown, T.S.:
‘Twrrrtcdynarnic Measurement and Prediction of Paraffin
Precipitation in Crude Oil,” paper SPE 26604 presented at the
1993 SPE Annual Technical Conference and Exhibition,
Houston, TX, Oct. 3-6.
26. Jaubert, J.-N., Neau, E., Peneloux, A,, Fressingne C., and
Fuchs, A.: “Phase Equilibrium Calculations on an Indonesian
Crude Oil Using Detailed NMR Analysis or a Predictive
Method to Assess the Properties of the Heavy Fractions,” Jnd.
Eng. Chem. Res. (1995) 34,640.
27. Rausnitz, J.M., Lichtenthaler, R.N., and Azevedo E.G.:
Molecular Thermodynamics of Fluid Phase Equilibria,
Prentice-Hall Inc., Errglewood Cliffs, NJ (1986), Chap. 9.

585
8 H. PAN, A. FIROOZABADI, P. FOTIAND SPE 36740

TABLE 3- OIL #1 DATA

P(bar) TABLE 5- NORMAL PARAFFIN DISTRIBUTIONS IN
1.0 “ C1O+ OF 01L#2
LUQIQ%%Qb%a%E?%%%
w’%? - 0.2457 13.1157 24.5103 33.7031 Nmnd Parafrina Non-normalPamffii
C2 30:1 0.3433 3.6278 3.8029 3.894 cornp. mQkf&ME/CQolp!- MN!
C3 44.1 1.2781 4.5489 4,2133 3.658 N-C1O 1.10513 142 ●R-C1O 5.172’20 134
IC4 58.1 0.6048 1.5747 13965 1.2487 N-Cl 1 0.97961 156 R-Cl 1 4.44901 147
NC4 58.1 2.3328 3.7094 3.2575 2.6859 N-C12 0.87894 170 R-C12 3.88277 161
IC5 72.2 0.6242 1S304 2.0215 1.7486 1S332 N-C13 0.77614 184 R-C13 4.25163 175
NC5 72,2 0.6306 2,2744 2.%55 1.9978 1.7364 N-C14 0.72284 198 R-C14 3.85340 190
C6s 66.2 0.6627 4,3528 3.6846 3.1681 2.7556 N-C15 0.73708 212 R-C15 3.19127 2(M
C7S 97,5 0.7232 70409 5S66 4.75al 4.1317 N-C16 0.62821 226 R-C16 2.63989 222
C6s 111,1 0.7416 8.5073 6.5015 5.5512 4.8149 N-C17 0.59000 240 R-C17 2.33663 237
C9S 124.7 0.7572 6.2413 4.7139 4.0230 3.4876 N-C18 0.50846 254 R-C18 2.49054 251
Clos 138.7 0.7560 5.8502 4.4761 3.8184 3.3104 N-C19 0.44588 268 R.C19 2.15084 263
Clls 153.0 0,7763 4.7261 3.5504 3.0293 2.6253 N-C20 0.39453 282 R-C20 1.68018 275
C12S 167.2 0.7635 4.1475 3.1143 2.6571 2.2Q27 N-C21 0.33451 296 R-C21 1.56445 291
C13S 178.3 0.&180 4.276 3,2103 2.7391 2.3736 N-(L?2 0.30743 310 R-c22 1.40910 305
C14S 193.6 0,8124 3.653S 2,7430 2.3403 2.0261 N-C23 0.27001 324 R-C23 1.37820 318
c15s 207.9 0,8156 4.0267 3,0229 2.5791 2.2350 NC24 0.23816 338 R-C24 1.13932 331
C16S 222.1 0,8195 3.0815 2.3208 1.8601 1.7159 N-(25 0.22039 352 R-(25 0.99953 345
cl 7s 236.2 0.8230 2.7844 2.0877 1,7686 1.5510 N-(%26 0.18290 366 R-C26 1.02207 359
Cl 6S 250.3 0.8263 2.895 2.1733 1.8542 1.6066 N-C27 0.14406 380 R-C27 0.86298 374
C18S 264.5 0.82?)5 2.6891 2.0187 1,7224 1.4926 N-C28 0.13047 394 R-(28 0.89314 388
C20S 278.6 0.8325 2.219 1.6668 1,4212 1,2316 N-C’29 0.10775 408 R-(Y29 0.84513 402
C21S 292.7 0.6354 2.@159 1.5359 1.3104 1,1356 N-C30 0.08772 422 R-C30 0.75544 416
C22S 306.8 0.6362 1.8446 1.4598 1.2456 1,0794 N.(I31 0.06888 436 R-(T31 0.61233 430
C23S 320.8 0.6410 1.7396 1.- 1.1143 0.885s N-c32 0.05482 450 R-C32 0.53563 444
C24S 334.9 0.6436 1,6283 1.2223 1.0428 0.9037 N-C33 0.06876 464 R-C33 0.53780 458
C25S 346.9 0.6466 1A660 1.1005 0.9390 0.8137 N-C34 0.07264 478 R-C34 0.42828 472
C28S 362.9 0.8492 12924 0.9702 0.6278 0.7173 N-C35 0.05311 492 R-C35 0.37749 486
C27S 376,9 0.8518 1.1907 0,8939 0.7627 0.2608 N-C36 0.03050 506 R-C36 0.34654 500
C28S 380,9 0.6544 1.1033 0.6262 0,7C66 0.6124 N-C37 0.02318 520 R-c37 0.29744 514
C28S 404,9 03569 0.8835 0.7456 0.6364 0.5515 N-C38 0,01875 534 R-C38 0.28634 528
N-C39 0.01218 548 R-C39 0.27894 542
C30S 418,9 0,8594 0.W89 0.6623 0.5622 0.5045
N-C40 0.01150 562 R-C40 0.26581 556
C31 S 432.9 0.6619 0.6768 0S061 0.4335 0.3756
N-C41 0.00970 576 R-C41 0.25928 570
C32S 446.9 0.W44 0.7532 0.5654 0.4624 0.4180
N-C42 0.00859 590 R-C42 0.20070 584
C33S 460.9 0.W68 0.6694 0.5175 0,4416 0.3627
N-C43 0.00621 604 R-C43 0.19316 598
C34S 474.9 0.@692 0.629 0.4722 0.4029 0.3491
N-C44 0.00513 618 R-C44 0.16376 612
C35S 468.9 0.6716 0.5756 0.4321 03667 0.3195
N-C45 0.00571 632 R-C45 0,16282 626
C36S 502.9 0.8740 0.5294 0.3974 0.3391 0.2K139
N-C46 0.00362 646 R-C46+ 3.70820 806
C37S 516,9 0.8763 0.4829 0.3625 0.3083 0.2660 N-C47 0.00376 660
C36S 530.9 0.8786 0,4435 0.3328 0.2841 0.2462 N-C48 0.M1278 674
C39S 6449 0.6808 0.4036 0.W32 0.2567 0.2242 N-C49 0.00364 688
Wos+ 8000 4,7618 3.5697 3.0627 2.6541 N-C50 0.IXI181 702
W - molecular weight, g/rode N-C51 0.00151 716
- density al 289 K N-C52 0.00129 730
saturation pressure at311 K N-C53 0.00109 744
N-C54 0.0CG94 758
N-C55 0.00078 772
TABLE 4- OIL #2 DATA N-CM 0JXK172 786
stock Reservoir Densily Mdacular N-C57 0.00078 800
Tank Oil fluid Weight N-C58 0.0C090 814
S&ulEflQl!3% m&k&9bm N-C59 0.00043 828
N-C60 0.00074 842
C02 o 1,365 N-C61 0.00056 856
c1 o 51.283 N-C62 0.00029 870
C2 0,067 7,344 N-C63 0.00024 884
C3 0.536 4.430
IC4 0.301 0.755 lon-normalpaftins for rhe same cartrwr number
NC4 1.240 2,037
IC5 1.008 0.772 0.6246
NC5 1.730 1.287 0.6308
C6 3.701 1.967 0.6636 86.03
C7 8.213 2.524 0.7366 91.94
C8 10.623 3.456 0.7543 105.45
C9 6.672 2.101 0.7640 120.23
Clo+ 65.666 20.644 0.6621 282.15

!586
SPE 3674o PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 9

TABLE 6- OIL #3 DATA TABLE 8- Kij FOR COZ.HYDROCARBONS AND

Nz- HYDROCARBONS
stock Reservoir Dansity MdecYJlar c, c, c, c, c, C6

p
Tank 011 Fluid
Lm210%nlQ19%
O.000 0.266
!@
Weight
Lxnm.lQ
C02
Nz
0.103 0.130
0.031
0.135 0.130
0.042
0.125 0.150
0.091 0.095 0.095 0.120
I
C02 0.028
cl 0.128 26.517
0.240 4,254
TABLE 9- CHARACTERIZATION OF OIL #1, OIL #2,
C2
C3 1.186 5208 AND OIL #3
IC4 1.235
NC4 3.782 mQIQ%MM QQmEr0QlQ%J!4kY
IC5 1.586 0.6246 “P-c\o+ 34130 1636 P-C1O+ 44627 167.0
NC5 2.567 0.6308 N-C1O+ 9.6705 163,6 N-C1O+ 6.4827 158.6
C6 5.235 3.577 0.6701 64.40 A-Cl O+ 9.6705 163.6 A-C1O+ 15.1263 159.6
C7 8.152 5.133 0.7260 93,58 P-cl 5+ 0.9288 233.6 P-C15+ 2.8096 236.6
C8 9.418 5.682 0.7346 109.21 N-C15+ 7.2836 233.6 N-C15+ 3.U28 233.3
C9 7.024 4.178 0.7621 124.00 A-C15+ 7.2836 233.6 A-C15+ 8.W&4 233.3
Clo+ 60.730 35.9n 0.8987 345.63 P-C20+ 0,4214 304.6 P-C20+ 1.5446 306.6
N-C20+ 3.0217 304.6 N-C20+ 2.1514 301.6
TBP Data of C,* A-C20+ 6,1349 304.6 A-C20+ 5.0188 301.6
Er.a@Qo MY P-C25+ 0.2539 374.3 P-C25+ 0.7856 375,1
CP 1 3.4053 140 N-C25+ 1.9114 3743 N-C25+ 1.3880 372,2
CP2 2.9271 147 A-C25+ 3.8807 374,3 A-C25+ 3.2408 372.2
CP3 6.8023 170 P-C30+ 0.1097 444.8 P-C30+ 0.3528 448.8
CP4 5.9862 180 N-C30+ 1.1706 444.8 N-C30+ 1.4348 440,4
CP5 2.1881 200 A-C30+ 2.3767 444.8 A-C30+ 1.4348 440,4
CP6 3.7005 220 P-C35+ 0.0244 514.4 P-C35+ 0.1377 510.5
CP7 2.6450 232 N-C35+ 07955 514.4 N-C35+ 1.5694 511.7
CP8 2.6111 2!33 A-C35+ 1.6151 514.4 A-C35+ 0.0174 511.7
CP9 1.6962 258 P-CP1 o.Ml13 643.6 P-C40+ 0.0468 590.3
CP1O 2.3105 273 N-CP1 0.2198 643.6 N-C40+ 1.1864 586.5
CP1l 20807 280 A-CP1 1.9269 643.6 A-C40+ 00491 586.5
CP12 1.2881 303 P-CP2 0.OOOO 807,2 P-C46+ 0.0259 713,4
CP13 1.8528 318 N-CP2 0.0000 807.2 N-CP1 0.1143 724,3
CP14 3.2358 374 A-CP2 1,6303 807.2 A-CP1 2.0266 724.3
CP15 8.0726 563 P-CP3 0.0000 1083,4 P-CP2 0.0000 734.4
CP16 2.7177 654 N-CP3 0.0000 1083,4 N-CP2 0.0157 901.3
CP17 7.1790 708 A-CP3 1.0033 1083.4 A-CP2 1.5486 801.3
eservoir Tmmeratme: 352.85K QiL#3
~turationpresh at reservoir temperature: 107.0har =1 —M!&
3.6579 157.1
N-CP1 4.7384 157,1
A-CP1 4.7384 157,1
TABLE 7- MEASURED CLOUDPOINT POINT P-CP2 2.8452 201,2
TEMPERATURE AND THE AMOUNT OF THE N-CP2 4.5198 201.2
PRECIPITATED WAX FOR OIL #1, OIL #2 AND A-CP2 4.5198 201.2
P-CP3 1.8518 252,1
OIL #3 N-CP3 2.9644 252.1
W?l A-CP3 4.4466 252.1
l(bar) 1 34 68 102 P-CP4 1.2662 330.2
VAT(K) 28593 294.26 284,82 260,93 N-CP4 2.8849 33o,2
A-CP4 4,3273 330,2
)iLlt2 P-CP5 0,2077 563.0
‘(bar) 1.0 50.0 120,0 N-CP5 2.7995 563.0
VAT(K) 313.15 309.15 301.15 A-CP5 5.0654 563.0
“(K) 313,15 303,15 293,15 283.15 278,15 P-CP6 0.0487 6540
VT%(WAX)O,l 0,5 1,81 5.59 8,2 N-CP6 0.5292 6540
A-CP6 2.1398 654.o
P-CP7 0.0380 666.0
)iL#3 N-CP7 0.7083 666.0
y)ar) 15 40 70 100 30 A-CP7 3.2007 666.0
IVAT 318,7 315.7 314.2 314.2 313.7 P-CP8 0.0028 743.7
r(K) 313,1 3031 283.1 283.1 278.1 N-CP8 0.2600 743.7
h/T%(WAX) 0,161 0.70 2.19 7.03 12.48 A-CP8 2.S680 743.7
“P- C,& indimtes five consecutive @rbon numbers, i.e. C, O-Cl,

587
10
H, PAN, A. FIROOZABADI, P. FOTLANIY
SPE 3674o

230 250 270 290 3T0 330

Temperature, K

Fig. l-Calculated wax amount and precipitating species

for the model synthetic oil (P=l atm).

40

0 Aromatics
Naphthenes
z lso-Paraffins
~~ Correlation for Naphthenes and lso-Paraffins
. I = = = - Correlation for Aromatics

mm mm--m-m-mm mm

o , * I I , I , , I I , 1

0 50 100 150 200 250 300 350 400

Molecular Weight, gjmol
Fig. 2-Melting entropy vs. molecular weight.

588
SPE 36740 PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 11

1 1 I I I I t 1 +

o
N
m

o
O&

:-u L C&

i
8
● ✛ ■ ;4 I
8
1

0 o
CJ
co
 ... ..””
334 ~ Pressurized stock tank oil
... ..””
... ..”” ...... 0..... 20% c1
t . ..” ..- 0 --’ 30%c~

333
.,””
ti .. >..
. .....
. ...
s ..

... 1:
1:

.- t:

-“ 1.

.“ 1;
.- t.
.’ b:
,
.“
1 #**
‘$:...
.;
●
A=”* 0’ U,p
i ●’* / 0/ \ ...
\ ..
*\ .“ t.
,.. / 0
i, ~’” 00 ...
#’ 8..
/
s.

4
‘\* L Original oil
329 0
\ ~ -“-”-”*.-”.”” 50% C5
. / .--&-- 20%C1
w
—+- 30% cl
I —.. + - Saturation point
328[, tii, iiii Ii,,,, tt,, i,tt,,,,,, I,, t,,,,,, i,,,,,,,,,
100 200 3(MI 400 50(-) 240 260 280 300 320
Pressure, Bar Temperature, K
Fig. 6-Calculated cloudpoint temperature Fig. 7-Calculated wax amount at
vs. pressure for the model synthetic oil. different Cl concentrations for
the model synthetic oil (P=l 10 bar).
SPE 3674o PRESSURE AND COMPOSITION EFFECT ON WAX PRECIPITATION: EXPERIMENTAL DATA AND MODEL RESULTS 13

r5 I .,* ‘-%

,, -R-. ‘.
,, [,\–-._~
(,
., ---- -> N

235 255 275 295 315 335 355

Temperature, K

Fig.8-Calculated wax amount and precipitating species at two

different C5 concentrations the model synthetic oil (P=l atm).

310
I

305

100 200 300 400 500

Pressure, bar
Fig. 9-Calculated cloud oint temperature
vs. pressure for the stoc ! tank oils.

591
14 H. PAN, A. FIROOZABADI, P. FOTLAND SPE 36740

592