Asphaltene Stability of Some Iraqi Dead Crude oils

Asphaltene is one of the fractions of the crude oil which is soluble in aromatics such as benzene or toluene and insoluble in alkane such as n-heptane, n-pentane or petroleum ether (mixture of alkane compounds). Asphaltene precipitation is one of the most common problems that sometimes occurs in both oil recovery and refinery processes as a result of changing in pressure, oil composition, or temperature. Therefore the stability of asphaltene in the crude oil must be studied to show the tendency of it for precipitating asphaltene to prevent it (Asphaltene precipitation and deposition problem) and eliminate the burden of high treatment costs. In the present study, saturate, aromatic, resin and asphaltene (SARA) analysis of the six dead crude oil samples from different Iraqi oil fields was conducted by using open column liquid chromatography after separating the asphaltene from them through filtration process. The asphaltene stability of dead crude oil samples was studied depending on changing the composition of them by adding the petroleum ether as an alkane and using colloidal instability index (CII) to determine the tendency of these crude oil samples to precipitate asphaltene depending on the SARA analysis results of these dead crude oils. All of dead crude oil samples showed the instability of asphaltene depending on this index and this means that all of them might precipitate asphaltene if the composition of these crude oil samples changed due to existing with the alkane in the live case in wells (Live oil is oil containing gas phase at reservoir conditions) such as injection of gas which has high ratio of alkane or the expanding the gas in the oil when the pressure decreases until reaches bubble point pressure. The refractive index of the dead crude oil samples was measured experimentally and calculated by two correlations which were Fan et al. correlation and Chamkalani correlation. The last one showed the best match between the experimental and calculated values of the refractive index of the dead crude oil samples.

There are different models of asphaltene precipitation.One of them is colloidal model which based on that asphaltenes form centers of the micelles and bordered and spread in the oil by resins.Precipitation will take place when the resins are separated from the colloid causing aggregation and phase separation, Zendehboudi, et al., 2014.Colloid stability is a balance between attractive and repulsive forces: when particles collide, if the attractive forces are stronger, they aggregate and dispersion may destabilize.When repulsive forces dominate, the system will remain in a dispersed state, Verdier, 2006.By the addition of normal alkane liquids such as n-pentane or n-heptane or adding petroleum ether to crude oil, the resin molecules tends to desorb from the surface of asphaltene in order to re-establish the thermodynamic equilibrium.The desorption of peptizing resins induces the asphaltene micelles to agglomerate in order to reduce the overall surface free energy and with the sufficient titration of n-alkanes, the asphaltene molecule aggregate to such an extent that it precipitates by overcoming the Brownian forces of suspension, Sulaimon, and Govindasamy, 2015.Asphaltene stability has been studied by many researchers as mentioned below: Yen, et al., 2001, used SARA analysis by liquid chromatography and applying CII (Oil with CII > 0.9 is unstable and oil with CII < 0.7 is stable), Oliensis spot test (Oil with spot test numbers of 9 or less has been known to unstable), and Asphaltene Precipitation Detection Apparatus APDU (A very unstable oil usually has an APDU value between 0 and 1 at inflection) as an indexes of aspaltene stability.By these three indexes, they found the Alaskan oil was unstable., et al., 2002, correlated between RI and SARA fractions (HPLC technique) and depended on ΔRI (The difference between RI of oil and RI of oil and alkane mixture at asphaltene onset point) to show the stability of the crude oil.They showed that oil with ΔRI < 0.04 or 0.05 had asphaltene precipitation problem while oil with ΔRI > 0.06 had no asphaltene precipitation problem.

Fan
Chamkalani, 2011, developed a relation for SARA and RI and compared it to Fan et al correlation, by utilizing RI to investigate the stability of asphaltene.Sulaimon, and Govindasamy, 2015, showed that Fan, et al., 2002, correlation given is reliable in estimating the refractive index of crude oil with the maximum absolute deviation of 2.0%.
Rogel, et al., 1999, studied the composition of crude oils and its effect on asphaltene stability and also chemical and structural characterization of asphaltenes and resins and their relation to asphaltene stability.They found the relationships between flocculation onsets of the crude oils and structural parameters of the asphaltenes.
In the present work, SARA analysis of the dead crude oil samples was made by open column liquid chromatography to find the weight percent of each SARA fractions (Saturate, aromatic, resin, and asphaltene fractions) of the dead crude oil samples then applied the index of colloidal instability index (CII) to determine the asphaltene stability.It also measured the refractive index (RI) of these samples by the refractometer instrument (RFM960) and also calculated RI of oil by two correlations.

EXPERIMENTAL WORK 2.1 Materials
To conduct SARA analysis, the following materials required:

1-Dead crude oil samples
The experimental tests were conducted with six dead crude oil samples from different Iraqi oil fields as depicted in Table 1.

Instruments and Devices
In this study the following devices and instruments were used respectively: 1-The pycnometer for measuring the density and then calculating the API gravity of the dead crude oil samples.2-The BROOKFIELD DV-II+Pro Viscometer (HB) for measuring the dynamic viscosity (µ) of the dead crude oil samples.3-The simple distillation device for heating the dead crude oil samples to extract the volatile compounds from them before conducting SARA analysis because they causes weight loss during the experiment tests of SARA.4-The filtration system to extract asphaltene fraction of the dead crude oil.5-The burette and silica gel for designing open column chromatography to separate SAR fractions (Maltenes).6-The refractometer (RFM960) for measuring the refractive index (RI) of the dead crude oil.

Experimental Work Procedure
The following flowchart shows the procedure of the experimental work that have been applied to achieve the practical part of this study: As in the above flowchart, the procedures of the experimental work were as follows: 1-Measuring API gravity of the dead crude oil samples using a pycnometer.2-Measuring the dynamic viscosity (µ) of the dead crude oil samples using BROOKFIELD DV-II+Pro Viscometer (HB).3-After the two previous steps, the experimental work divided into: a-Determining the stability of asphaltene depending on colloidal instability index (CII) which required simple distillating the dead crude oil samples (Approximately 210 ºC) and then conducting SARA analysis on them.b-Selecting the best correlation for calculating the refractive index of the dead crude oil from SARA analysis results by comparing the results of these correlations with the measurement values of RI.

SARA Analysis
Saturate, aromatic, resin and asphaltene weight percent of the dead crude oil samples were found by conducting SARA analysis after simple distillation (Getting toped oil by simple distillation at a temperature of 210 ºC).This analysis was shown in the following flowchart depending on ASTM D2007: The SARA analysis of the toped oil after simple distillation was conducted as shown in Fig. 3 depending on ASTM D2007.The weight of the taken toped oil was 0.5 g which mixed with 20 ml of petroleum ether and left for one day to precipitate asphaltene.Then the mixture of the toped oil and petroleum ether were filtrated through whatman No.42, 110 mm diameter filter paper to get the weight percent of the asphaltene in the oil which precipitate on filter paper by Eq. (1).After designing the column of separation, the maltenes poured above the saturated column with petroleum ether.The first fraction of the maltenes which dropped through the column was saturated (Yellow color material) and this separated by adding the petroleum ether above maltenes because of the polarity between it and saturate fraction and collected in a beaker.When the saturated fraction separation completed, the second solvent used was benzene which separated the aromatic fraction (Brown color material) because of high polarity between them and collected in a beaker.Benzene addition was stopped when there was no aromatic in the column.Finally, the mixture of chloroform and ethanol was poured through the column to displace the resins (Black color material) and then collected in a beaker.Then each one of them (SAR fractions) was dried by oven to get the net weight and the weight percent of the SAR fractions was calculated by Eq. ( 2).

Refractive Index Measurement
The refractive index of the dead crude oil was measured using refractometer (RFM960) as shown in

RESULTS AND DISCUSSION
The API gravity, dynamic viscosity (µ), and SARA analysis results of dead crude oil samples are illustrated in Table 3.These dead crude oils ranged from light to heavy oils depending on their API and the dynamic viscosity values in comparison with Tharanivasan, 2012 study.The sample Kz was the more light and the sample Qy was the heavy one depending on their API and dynamic viscosity values.The relation between saturate weight percent versus API and resin weight percent versus API as shown in Fig. 6 and Fig. 7.As shown in Fig. 6 and Fig. 7 an increase in the saturate weight percent with API (Light crude oil had high saturate weight percent) and a decrease in resin weight percent with API (Heavy crude oil had high resin weight percent) and this matched with, Subramanian, et al., 2016, study where they found out that heavy crudes usually have a lower tendency to give asphaltene deposition problems in spite of their higher asphaltene content because of its high resin ratio that surrounds the asphaltene and prevent it to precipitate.
To show the asphaltene stability, the colloidal instability index (CII) of the dead crude oil samples was calculated by the Eq.(3) as introduced by, Yen, et al., 2001 and the results were as illustrated in Table 4.

R e s i n w t % V S A P I
All of the dead crude oil samples illustrated in Table 4 had CII greater than 0.9 which means that these crude oil samples were unstable depending on the CII technique which introduced by, Yen, et al., 2001.
The refractive index (RI) of the dead crude oil samples was measured by the refractometer (RFM960) and also calculated by Eq. ( 4) which introduced by the, Fan, et al., 2002 and the comparison between them as shown in Table 5.
RI= (S%*1.4452+ A%*1.4982 + (R + As) %*1.6624) /100 (4)   6 and Table  7.The relationships between RI and each fraction of the SARA were plotted as shown in Fig. 8 and Fig. 9.       show an increase in the refractive index values (RI) of the dead crude oil samples with increasing resin and asphaltene weight percent (R% and As%) and a decrease in RI with increasing saturate and aromatic weight percent (S% and R%) and this accepted with the Chamkalani correlation which showed the same relationship.This also proved the less absolute deviation of the calculated refractive index (RI calculated) values of the dead oil samples by Chamkalani correlation than Fan, et al., correlation as shown in Table 5 and Table 6.This conclusion is also confirmed by Fig. 12 and Fig. 13. because the first one gave the best relationship between the measured and calculated refractive index and the coefficient of determination (R 2 ) was greater than Fan, et al., 2002, correlation.Due to the previous relationships between measured refractive index (RI) values and SARA fractions ratio of the dead crude oil samples as Fig. 8 and Fig. 9 showed the compatibility with the Chamkalani, 2011, equation, where there is shown an increase in the refractive index (RI) with an increase in asphaltene and resin percent and a decrease in it with an increase in saturate and aromatic percent in the oil.While in the Fan, et al., 2002, correlation, there is an increase in RI with an increase in the all SARA fractions percent.

CONCLUSIONS
1-All of dead crude oil samples were unstable according to colloidal instability index (CII) and this means that if any rich alkane gas exists with the oil will precipitate asphaltene (All dead crude oil samples have asphaltene precipitation problem).
2-If the gas which is rich of alkane compounds such as methane exists with crude oil in the live case (Live oil which contain gas) by different process such as gas injection for EOR processes, decreasing the viscosity of the heavy crude oil by gas injection, decreasing the pressure due to production until reaching bubble point pressure and stimulation methods, it will cause asphaltene precipitation because of asphaltene instability.
3-Light crude oil sample (Khabbaz crude oil) was also unstable although the oil had a low asphaltene weight percent this means that the asphaltene precipitation do not depend on asphaltene weight percent in the crude oil.
4-The two correlations (Fan et al. correlation and Chamkalani correlation) of calculating refractive index of the dead crude oil samples have been gave the acceptable match with the measured one but Chamkalani correlation is more acceptable.
It was first defined by Boussignault in 1837 to represent the material that precipitates out of petroleum due to adding of petroleum ether, Panoganti, 2013.It is defined as a solubility class of crude oil fractions that can precipitate with n-alkanes, such as n-pentane and n-heptane, or petroleum ether but remains soluble in aromatic solvents, such as toluene and benzene, Powers, 2014.Asphaltene is also defined as association of aggregates with 2-6molecules per aggregate.The aggregates are either colloidal particles or macromolecules, Zendehboudi, et al., 2014.The collection of the aggregates forms clusters and the summation of the clusters lead to deposition after precipitation as shown in Fig. 1, Hoepfner, 2013.

Figure 2 .
Figure 2. Flowchart of the experimental work.
. %: Asphaltene weight percent.Asp.Weight is the weight of the asphaltene which results from subtracting the weight of filter paper from the total weight of asphaltene and filter paper.Maltenes (SAR fractions of the oil) after that have been entered to open column chromatography which consisted of burette and silica gel as in the Fig. 4.
This correlation gave the more acceptable results of the refractive indexes of the dead crude oil samples than , Fan, et al., 2002 when they compared with measured values of the refractive indexes because of less absolute deviations.This result supports the two introduced correlations and also the accuracy of the SARA analysis which conducted by designing open column chromatography and then the method of determining the stability of asphaltene.The refractive index (RI) values and SARA fraction weight percent are shown in Table

Figure 8 .
Figure 8. RI and S% relationship of dead oils.Figure 9. RI and A % relationship of dead oils.

Figure 9 .
Figure 8. RI and S% relationship of dead oils.Figure 9. RI and A % relationship of dead oils.

Figure 10 .
Figure 10.RI and R% relationship of dead oils.Figure11.RI and as % relationship of dead oils.

Figure 11 .
Figure 10.RI and R% relationship of dead oils.Figure11.RI and as % relationship of dead oils.

Figure 12 .
Figure 12.Measured RI and calculated RI by Figure 13.Measured RI and calculated RI by Chamkalani correlation.Fan, et al., 2002, correlation.

Table 1 .
Dead crude oil samples.
2-Precipitant, extraction and resolution materials of SARA fractionsThese materials which were used in the experiments, their advantage, and description of each one are illustrated in Table2.

Table 2 .
Used materials with their descriptions and advantages.

Table 3 .
API gravity, dynamic viscosity (µ), and SARA analysis results of dead crude oil samples.

Table 4 .
Colloidal instability index (CII) values of dead crude oil samples.

Table 5 .
Measured RI and calculated RI by Fan et al correlation.

Table 6 .
Measured RI and calculated RI by Chamkalani correlation.