ASPHALTENE PRECIPITATION GAS-OIL RATIO WELLS

IN

HIGH

Sunil Kokal

Abdullah Al-Ghamdi

Dimitrios Krinis

Kokal is a reservoir fluid property specialist in Saudi Aramcos Research and Development Center in Dhahran. He is an expert in hydrocarbon phase behavior, crude oil emulsions and asphaltenes. Kokal has a BS degree from the Indian Institute of Technology, New Delhi, and a PhD degree from the University of Calgary, Canada, both in chemical engineering. He has written more than 70 technical papers and has authored the chapters on Crude Oil Emulsions and Reservoir Fluid Sampling, for the new revised edition of the Society of Petroleum Engineers (SPE), SPE Petroleum Engineers Handbook (due in 2004). He is a registered professional engineer in Alberta, Canada and is a member of SPE. He has served on local and international levels on many SPE committees and is now a technical editor for SPEs Reservoir Evaluation and Engineering publication. His e-mail address is Sunil.Kokal@aramco.com. Al-Ghamdi is a laboratory scientist working in Saudi Aramcos Research and Development Center in Dhahran. His research activities focus on emulsion studies and demulsifier characterizations. Al-Ghamdi also investigates asphaltenes precipitation from crude oil and the mixing of hydrocarbons from deeper reservoirs. In addition, he studies reservoir fluids. He holds a BS degree in industrial chemistry from King Fahad University of Petroleum and Minerals in Dhahran. Al-Ghamdi has authored and coauthored several technical papers. He is a member of the American Chemical Society and SPE. His e-mail address is ghamam2l@aramco.com.sa.

38

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

Krinis is a senior reservoir engineer working in a reservoir management position with Saudi Aramco. He has participated in several oil and gas condensate field development projects undergoing depletion, water injection or gas recycling in the North Sea, the Far East and the Middle East with Marathon Oil and Shell. Krinis has a BS in mining and metallurgical engineering (1984) from the National Technical University of Athens, Greece, and a PhD in petroleum engineering (1990) from Heriot-Watt University in Scotland. He is interested in hydrocarbon fluid phase behavior, transient well testing, reservoir simulation and horizontal/multilateral well completion and performance. His e-mail address is Dimitrios.Krinis@aramco.com.

ABSTRACT
Asphaltene deposits have been observed in a number of high gas-oil ratio (GOR) wells in North Ghawar. Even though the oil reservoir is undersaturated, two small gascaps are present as a result of gas injection during the 1960s and 1970s. New development wells drilled recently to produce oil and gas from the gas-cap areas have experienced asphaltene deposition. The cause of precipitation is the stripping of the asphaltenes from the crude by the gas. This paper describes the results of an investigative study initiated to determine the precipitation mechanism and ways to alleviate the deposition problem. Asphaltene precipitation experiments were conducted at reservoir conditions in a special pressure-volume-temperature (PVT) apparatus. The effect of GOR on asphaltene precipitation was determined by titrating the reservoir oil with gas-cap gas. Bulk deposition tests were also performed at different GORs with reservoir fluids. The results demonstrate that the onset of asphaltene precipitation occurs at relatively low GOR values. However, the amount of asphaltene precipitated at the onset is negligible. Asphaltene precipitation and deposition increase with increasing GORs. Asphaltene deposition envelopes are provided for the reservoir oil as a function of pressure and temperature. Guidelines are provided to alleviate the problem by controlling the GORs. Recipes for solvent treatment, including asphaltene dispersants, are also described in the paper.

Fig. 1. Wells with solid asphaltenes observed.

INTRODUCTION
Ghawar Field is one of the major oil fields in Saudi Arabia. In the northern part of the field, some wells have experienced solid built-up in the wellbore. Analyses of solid samples from several wells have shown the presence of asphaltenes that may have precipitated during crude production and have started to deposit on the wellbore. The solid

Fig. 2. Solids from a well.

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 39

Fig. 3. Gas-oil ratio of wells with asphaltene deposits.

Fig. 4. Tag depth for a well.

deposition has been observed in high GOR wells. A location map of the wells is shown in fig. 1; photographs of the solid deposit from one well are shown in fig. 2. The Arab-D Reservoir of Ghawar Field contains an undersaturated light oil. The bubble point pressure is ~1,900 psi at the reservoir temperature of 101.6C (215F), and the average GOR is ~570 scf/stb. The reservoir pressure at present is over 3,000 psi. In the 1960s and 1970s, the associated gases from part of the field were injected back into the reservoir at two locations due to the unavailability of gas processing facilities and to avoid excessive flaring. The injected gases have formed two separate gas caps in the field (north and south gas-caps, fig. 1). In recent years, oil production has started from these gascap regions. Due to the presence of the gas cap, some of the free gas flows into the oil production wells, increasing their total GORs. The coning or cresting of gas into the oil has caused limited plugging in a few wells in the north and south gas-cap areas. The gas strips the oil of asphaltenes, which precipitate and deposit in the wellbores. Plugging of the wellbore by asphaltenes or organic deposits has the potential to reduce productivity and cause production impairment. Furthermore, several more gas-cap wells are to

be drilled in the area, and their productivity may be impacted by the deposition tendency. Fig. 3 shows the GOR for 11 wells in which asphaltene deposits have been observed. The solid line shows the average GOR for the entire field (~570 scf/stb). Except for one well, the GOR for all wells is higher (in some cases substantially higher) than the average field GOR. The high GOR is a consequence of gas coning/cresting in the wells. The free gas strips the asphaltenes from the crude that deposits in the wellbore. One well was tagged over a period of time to ascertain the buildup of asphaltenes. Fig. 4 shows the tag depths and indicates a loss of wellbore accessibility of ~61 meters (~200 ft), over 18 months. Recent results show that the build up has stabilized, and the asphaltenes may be dragged with the oil to the gas-oil separating plant (GOSP). Other wells are also being monitored and have shown some buildup activity. Asphaltenes comprise the heaviest and most polar fraction of crude oils (1 & 2). Asphaltenes exist in the form of colloidal dispersions and are stabilized in solution by resins and aromatics that act as peptizing agents. Asphaltene precipitation and deposition may occur deep inside the reservoir, near the wellbore and/or in processing facilities (3). Asphaltene precipitation is a function of pressure, temperature and live crude oil composition. Asphaltenes have a tendency to precipitate as the pressure is reduced, especially near the bubble point (however, precipitation can occur even at pressures higher than the bubble point, depending on the crude). Another important reason for precipitation is the stripping of crude oil by gas. When gas is added to the crude (as may be happening during the production from the gas-cap wells), the composition of the crude changes and may lead to precipitation. This is the same mechanism during deasphalting of crude in a refinery where propane and butane are used for stripping the asphaltenes. The precipitated asphaltenes then deposit near, or in, the wellbore. This may lead to an increase in formation damage (skin) and subsequently more precipitation. This study was initiated to investigate the causes of asphaltene precipitation and deposition in the high GOR wells. If the stripping mechanism can be confirmed, then appropriate stimulation programs can be designed to remove the solid deposits. Moreover, guidelines can be provided to produce the oil wells at a certain maximum GOR to keep the deposition under control. This paper presents the extensive experimental work conducted and preliminary field data to address the issue. The study includes a set of phase behavior experiments to determine the conditions for the onset of asphaltene precipitation

40

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

from live crude at different GORs. The extent (amount) of asphaltene precipitation was also determined. From these data, asphaltene solubility in crude oil at different conditions (including different GORs) was determined. Solubility tests with solvents were also conducted to determine the best solvents for stimulation. Several commercial asphaltene dispersants were also evaluated, first with the dead crude and with live oil, for their ability to disperse asphaltenes (i.e., reduce/eliminate their precipitation). These results are presented in the paper.

E X P E R I M E N TA L I N V E S T I G AT I O N S
ASPHALTENE PRECIPITATION Onset
Fig. 5. Cross-section of the pressure-volume-temperature cell with laser light source and detector.

A state-of-the-art solids detection system (SDS) was used to measure the onset of asphaltene precipitation from live crude oils (4). It is a high-pressure, high-temperature apparatus and consists of a variable volume and a visual PVT cell retrofitted with fiber optic light transmission probes to measure the onset of organic solids precipitation concurrently with fluid volumetric data. A schematic of the core is shown in fig. 5, and the principle of operation is depicted in fig. 6. The SDS apparatus consists of a laser power source, a fiber optic bundle that carries the laser light into the PVT cell, the actual PVT cell containing the crude at pressure and temperature, another fiber optic bundle that carries the received laser light from the PVT cell, and a power meter that measures the amount of light received (fig. 5). The PVT cell is mounted inside a temperature-controlled oven and has an effective volume of 110 cc. The PVT cell is a windowed cell that permits visual observation of the oil inside the cell. A variable-volume displacement pump controls the volume and pressure of the fluid inside the cell. The PVT cell also has a specially designed magnetically-coupled impeller mixer that provides powerful mixing and maintains equilibrium in the fluid system. The PVT cell is designed for operation at pressures of up to 10,000 psi and temperatures up to 182C (360F).
Crude Oil and Gas Properties

Fig. 6. Schematic of the principle of operation of the asphaltene precipitation cell.

The crude oil was sampled using a conventional bottomhole sampler. The crude oil fluid composition is shown in table 1. It has a bubble point pressure of ~1,900 psia at a reservoir temperature of 101.6C (215F) and a GOR of 580 scf/stb. The crude oil properties do not vary significantly across the area of interest. The dead crude has an asphaltene content of ~3.0 wt%. The composition of the injected gas (that was injected in the 1960s and 1970s) is presented in table 1. This was the

associated gas from the crude after processing at the gas-oil separating plant (GOSP). The gas used in the experiments was prepared from the high-pressure production trap (HPPT) gas after flashing it at 1,300 psia and 23.8C (75F). The composition of the flashed gas (used in all the experiments) is shown in table 1. The flashed gas matches the injected gas composition closely.
Procedures

The procedures for detecting the onset of asphaltene precipitation from crude oils is described elsewhere (4). Example plots of precipitating and non-precipitating fluids are shown in fig. 6. For the non-precipitating fluid, the laser power (as measured by the power meter) increases monotonically

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 41

Mole % Component N2 CO2 H2S C1 C2 C3 C4 C5 C6 C7+ C7+MW C7+SG BPP (psia) GOR (scf/stb)
Table 1. Crude Oil and Gas Properties

Reservoir Fluid 0.14 5.89 1.82 24.01 9.79 7.49 4.92 3.95 3.14 38.85 240 0.8652 ~1900 @ 220F

Injects Gas (Actual) 0.41 12.30 1.91 56.00 17.45 8.20 2.64 0.84 0.25 0.00

Injected Gas (Experiment) 0.34 12.62 2.49 56.00 16.23 8.39 2.86 0.83 0.25 0.01
Fig. 7. Onset of asphaltene precipitation with crude oil from a well without gas.

(almost linearly) as the pressure drops to the bubble point. At this point, the evolving gas bubbles scatter light, and the transmittance drops sharply. For the precipitating fluid the laser power drops before the bubble point. The point at which the curve deviates from the straight line corresponds to the onset of asphaltene precipitation. This drop in laser power is related to the precipitation of solids that scatter light and cause the power of the transmitted light to deviate from the expected linear curve. The point determines the onset pressure for solids precipitation. To determine the effect of the GOR on asphaltene precipitation from live crude oil, pressurized gas was injected into the PVT cell containing reservoir fluid at single-phase conditions. Approximately 30 cc of reservoir fluid were taken in the PVT cell, and a simple depressurization test (described above) was conducted at the reservoir temperature of 101.6C (215F). Following the depressurization test, the sample was pressurized to 3,000 psi at 215F and equilibrated at these conditions for 24 hours. Gas injection was started at a rate of ~1 cc/hr with continuous stirring. The results are described below. Fig. 7 shows the results of depressurization tests for one well. No onset of asphaltene precipitation was observed during depressurization. Several experiments were conducted with samples from other wells, and none of them

Fig. 8. Onset of asphaltene precipitation with crude from a well with gas.

showed any asphaltene precipitation during depressurization. Some of the samples were taken through repeated depressurization and pressurization steps without noticing any onset of precipitation. The results indicate that the crude does not have a natural tendency to precipitate when pressure is reduced. The results of gas injection are presented in fig. 8. The laser light transmittance increases initially as the gas dilutes the crude and then starts to decline after ~1.6 cc of gas injection. This is the point of asphaltene onset and is shown clearly in the inset chart in fig. 8. As more gas is added, the amount of asphaltene precipitation increases, and the light transmittance declines monotonically. After 4 cc of gas are added, the amount of light passing through the cell is not measurable. As more gas is added, the light transmittance starts to increase (~6 cc), indicating a dilution effect or a redissolution effect. After ~7 cc of gas are added, the transmittance declines again as the bubble point is reached. The gas addition was stopped after 9 cc. The results of gas injection are re-plotted in terms of gas-oil ratio in fig. 9. The results indicate that the onset of asphaltene precipitation will occur at a GOR value of ~625 scf/stb. This suggests

42

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

Fig. 9. Onset of asphaltene precipitation with crude from a well with gas.

Fig. 10. Onset of asphaltene precipitation with gas added.

that the crude is nearly saturated with asphaltenes and a small amount of gas is enough to start the precipitation. This is in agreement with data from a nearby field (5), where precipitation occurs during depressurization. Similar behavior was observed (not reported here) when the fluid from a nearby well was tested with the gas (similar procedures). A depressurization test was conducted in another experiment in which 7 cc of gas were added to 30 cc of oil. The results are shown in fig. 10. The crude oil and gas mixture was first pressurized to more than 6,000 psi. During depressurization, the onset of asphaltene precipitation started at ~4,500 psia. The bubble point (bubbles observed through the PVT cell) was ~3,000 psia. The results indicate that some of the precipitated asphaltenes (during gas addition) redissolved when the mixture was pressurized to 6,000 psia.

arate PVT cell was installed in the apparatus. It is essentially a high-pressure filtration apparatus and is referred to as the bulk deposition apparatus. It consists of two floating piston pumps as shown in fig. 11 (schematic) and fig. 12 (photograph). One of the cylinders is connected to a pump. The sample is taken in one of the cylinders and equilibrated for ~48 hours. The sample is then transferred into the second empty cylinder through a filter assembly. The experiment is conducted at constant pressure and temperature, representing reservoir conditions. The deposited asphaltenes are trapped on the filter assembly that is removed at the end of the experiment and weighed. The amount of asphaltenes precipitated is calculated in ppm, or as a percent of the total oil charged. Several bulk deposition experiments were conducted on well samples at 3,000 psi and 101.6C (215F). The pressure was selected to keep the samples in single phase. Sixty cc of reservoir fluid were charged into one of the cylinders. A known amount of gas was then added to the sample. The apparatus was rocked for 24 hours at these conditions. The crude oil was then filtered by transferring it into a second cylinder. After all the sample had been transferred, the filter assembly was opened, and the filtered solids were weighed. A photograph of the filtered solids is shown in fig. 13. From the weight of the filtered solids (mostly asphaltenes), the amount of precipitated asphaltenes was calculated. Fig. 14 shows the amount precipitated in ppm as a function of gas added (or GOR). Table 2 presents the data used in the plots. The results indicate that the amount of precipitate increases with an increasing GOR. While the amount of asphaltenes precipitated is not significant per unit volume of oil, the total amount can be significant over a period of time. Another point to note is that not all the precipitated asphaltenes will deposit. Some of the precipitated asphaltenes will be carried with the oil to the GOSP. Moreover, some of the asphaltenes will re-dissolve as the pressure is reduced below the bubble point in the production strings. Complex hydrodynamics and deposition kinetics will determine how much asphaltene will deposit and

A S P H A LT E N E P R E C I P I TAT I O N : BULK DEPOSITION

The experiments described previously show the onset of asphaltene precipitation; they do not quantify the amount of asphaltenes precipitated. To quantify precipitation, a sep-

Fig. 11. Schematic of a bulk asphaltene precipitation apparatus.

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 43

Fig. 14. Bulk asphaltene precipitation at reservoir conditions.

GOR (scf/stb)
Fig. 12. Photograph of bulk asphaltene precipitation apparatus.

550 60

597 60

643 60

736 60

1250 1950 60 60

Amount oil charged (cc) Amount gas charged (cc) Pressure (psi) Temperature (F) Amount precipitated (mg) Precipitated asphaltenes (ppm)

30

60

3000 3000 3000 3000 3000 3000 215 215 215 215 215 215

17.9

33

58

62

81.1

132

403

739

1304 1396 1827 2973

Table 2. Bulk Deposition Test Results

Fig. 13. Bulk asphaltene precipitation showing precipitated asphaltenes.

plug the wellbores. Some of this is being observed in the existing wells.

A S P H A LT E N E S O L U B I L I T Y T E S T S
Several tests were conducted to determine the solubility of asphaltenes in solvents. Two solvents were selected for the tests: diesel and xylene. Diesel is readily available

and cheap. The solubility tests were conducted to select appropriate solvents for cleaning the plugged asphaltenes in the wellbores. A simple procedure was used to determine the solubility of asphaltenes in these two solvents. The precipitated solids (fig. 2) were ground, and a fixed amount (5 g) was placed in a teabag. The teabag was then suspended in a solvent (100 mL) from a hook attached to a balance. The solvent was not stirred. The weight of the teabag (with the asphaltenes) was monitored. Using Archimedes principle, the loss in the weight of asphaltenes was calculated. The results are presented in fig. 15 for xylene, diesel, and mixtures of xylene and diesel. The results show that diesel is not a very good solvent, and less than 10 percent of the asphaltenes dissolved after soaking for three hours. On the other hand, complete solubility was attained with xylene in

44

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

Fig. 15. Solubility of asphaltenes in different solvents. Fig. 16. Asphaltene dispersant tests with dispersant C-2.

under three hours. For cleaning purposes, xylene is a better solvent and a 50:50 (diesel:xylene) mixture was recommended for cleaning the wells.

A S P H A LT E N E D I S P E R S A N T T E S T S : DEAD CRUDE
One way to inhibit the asphaltenes from precipitating is by using asphaltene dispersants or inhibitors (6, 7, & 8). Asphaltene inhibitors are specialized polymeric chemicals that stabilize asphaltenes in the crude and prevent them from precipitating. In their role, they act in the same manner as resins and peptize asphaltenes. However, it is believed that they have a much stronger association with the asphaltenes, and hence, have a stronger peptizing effect than natural resins. Consequently, the inhibitors stabilize and disperse asphaltenes and prevent them from precipitating and flocculating. Several commercial asphaltene dispersants were evaluated for their tendency to disperse asphaltenes in the crude. The screening test used for evaluation was the heptane precipitation test (6, 7, & 8). Six different commercial asphaltene inhibitors were selected for testing. Different amounts (in ppm) of these inhibitors were first added to 50 mL of nheptane and shaken. Five hundred microliters of the dead crude oil was then injected into these tubes, shaken and left to stand for a period of 24 hours. The amount of asphaltenes precipitated at the bottom of the tubes was measured and related to the base case where no inhibitor was added. The results for one of the inhibitors are shown in fig. 16, and the results for all the inhibitors are presented in fig. 17 as a function of inhibitor concentration. The results show that dispersant C-2 was one of the best and was further used for live oil tests described below.

Fig. 17. Asphaltene dispersant tests.

A S P H A LT E N E D I S P E R S A N T T E S T S : LIVE CRUDE
While the tests with the dead crude are good for screening inhibitors, they do not provide an estimate of how effective the inhibitors will be at reservoir conditions. Based on the dispersant selection process described above, the best dispersant C-2 was selected for testing with live crude oil. Four different concentrations were evaluated using the bulk deposition apparatus to estimate the effectiveness of the dispersant at 3,000 psi and 215F. Sixty cc of gas were added to 60 cc of crude oil (in each case). The concentrations used were 0, 10, 100 and 1,000 ppm. A special technique was used to mix the dispersant with the crude oil. The dispersants were diluted in xylene (a 10 percent solution), and aliquots of the required dosages (microliter volumes using appropriate syringes) were added to the empty, clean cylinder (figs. 11 and 12) before crude charging. The crude was then charged into the cylinder and equilibrated. The cylinder assembly was rocked for 48 hours at the set temperature and pressure before filtering (described earlier). After filtering (transfer from one cylinder to the other), the cylinders were cooled down to room temperature, and the fil-

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 45

Fig. 18. Effect of dispersant on bulk deposition tests at reservoir conditions.

Fig. 19. Density of wellhead oil samples.

ter assembly was opened, and the filtered solids were weighed. From these measurements, the amounts of precipitated asphaltenes were calculated in ppm. The results are presented in fig. 18. The results indicate that the dispersant was effective, especially at the higher concentrations, in reducing the amount of precipitation.

and 4. In 11 wells, black, solid asphaltenes were obtained. In several wells, only water and inorganic solids (from the formation) were obtained. The black solids were generally found in the highest GOR wells. The results clearly indicate that asphaltenes are precipitating in the wells.

MODELING
A commercial PVT simulator (PVTsim by Calsep) was also used to simulate the asphaltene phase behavior. The required input for the simulator were the fluid compositions (table 1), the amount of asphaltenes in the crude oil and any minor adjustment or tuning to match the bubble point pressure. The results are presented in fig. 20. The bubble point pressures are well-matched. The simulator was then used to predict the phase boundaries for asphaltene precipitation with injected gas. A few minor adjustments had to be made to the amount of asphaltenes in the crude, but a fairly good trend analysis was obtained with the PVT software. The use here highlights that, with a few experimental points and minor tuning, such a package can be used to determine the phase envelope for asphaltene precipitation. Fig. 21 shows the results for the bulk deposition tests. The results are presented in terms of asphaltene wt% as a percent of the original asphaltenes in the crude. PVTsim predicted that the amount of asphaltenes will decline beyond a GOR of 1,000. Experimental results showed an opposite effect; asphaltene precipitation increased with GOR (fig. 14).

W E L L H E A D S A M P L E A N A LY S E S
As part of the investigative study, several wellhead samples (depressurized) were obtained from selected wells across the field. These samples were analyzed for basic sediment and water (BS&W), emulsions and saturates, resins, aromatics and asphaltenes (SARA) analysis and density of the oil samples. No free water was observed in any of the wellhead samples. If water was present, it was in the form of a tight emulsion. The oil was separated using minimal demulsifier, and its density was measured. The results are shown in fig. 19 and fall within a tight band of 0.875 g/cc at room temperature. A SARA analysis was also conducted on the separated oil and the asphaltene values are plotted in fig. 20. There was significant variation in the asphaltene values due to contamination of the oil with emulsion (not complete water separation). The average value is ~3 wt% asphaltene in the dead crude. These tests were conducted to identify any area anomalies.

BAILER AND SAND-BAILER S A M P L E A N A LY S E S

Bailer and sand-bailer samples were also obtained from selected wells including all high GOR wells. A bailer sample is collected from a specified depth, and the sand-bailer sample is obtained from the bottom of the well. The bailer sampler is an open-cup sampler. BS&W, SARA (for separated oil) and solid analyses were performed on the bailer and sand-bailer samples. The results are presented in tables 3

DISCUSSION
The results of this study clearly indicate that asphaltene precipitation is a function of gas injection and increases with increasing GOR. Asphaltene precipitation onset starts at relatively low GOR (~625 scf/bbl). The amount of asphaltene precipitated is relatively small. Moreover, there is evidence that some of the asphaltenes dissolve as the pressure is

46

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

Well No GOSP 1 2 56 218 235 270 425 438 452 20 429 474 484 391 1 1 1 1 1 1 1 1 1 2 2 2 4 6

Sampling Date 24-Feb-01 26-Feb-01 24-Feb-01 21-Feb-01 21-Feb-01 25-Feb-01 25-Feb-01 26-Feb-01 21-Feb-01 26-Feb-01 27-Feb-01 25-Feb-01 3-Mar-01 10-Mar-01

BS&W (%) 97.0 0.0 52.0 0.0 0.0 6.3 0.0 97.5 0.0 80.6 98.4 98.7 99.9 92.8

Density (g/cc) 0.8785 0.8826

SARA S A R A

Sample Type sand-bailer sand-bailer

Depth (ft) 6892 6815 6915 7050 7010 6810 7565 7114 6472 7080 6990 7000 7025 7750

Remarks 10% solids, calcite/dolomite not enough sample small amount sample 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 83% solids, mostly calcite

44.14

40.13

12.79

2.94 ~98 ~98

sand-bailer sand-bailer sand-bailer sand-bailer

~98

sand-bailer sand-bailer sand-bailer sand-bailer sand-bailer sand-bailer sand-bailer sand-bailer

29% solids, calcite/dolomite 65% solids, calcite/dolomite 90% solids, calcite/dolomite 64% solids, calcite/dolomite

Table 3. Sand-Bailer Sample Data

Well No GOSP 439 454 472 472 476 397 469 469 485 487 257 477 488 1 1 1 1 2 4 4 4 4 4 6

Sampling Date 20-Feb-01 20-Feb-01 11-Sep-00 11-Sep-00 20-Feb-01 21-Jul-01 25-Jul-01 20-Feb-01 24-Jul-01 29-Jul-01 21-Feb-01 09-Jul-01 17-Jul-01

BS&W (%) 0.0 99.0 0.0 0.0 89.0 0.0 0.0 0.0 0.0 0.0 87.0 0.0 0.0

Density (g/cc)

SARA S A R A

Sample Type bailer bailer

Depth (ft) 6880 7078 6772 6772 6398 6980 6878 6825 6930 6841 7020 NA 6847

Remarks Empty bottle 2% solids, calcite/dolomite 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 51% solids, calcite/dolomite 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes 44% solids, calcite/dolomite 100% black solids, mostly asphaltenes 100% black solids, mostly asphaltenes

~98 ~99 ~98 ~98 ~99 ~98 ~98 ~98 0.8833 ~98 ~98

bailer bailer bailer bailer bailer bailer bailer bailer bailer bailer bailer

Table 4. Bailer Sample Data

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 47

Other wells in the area are being monitored for solid buildup to evaluate whether or not the deposition is building. Recent data indicate that the buildup in several wells has stabilized. This could be due to the high flow rates in the well and the complex hydrodynamics of fluid flow and depositional environment. Furthermore, no dramatic reduction in oil productivities has been observed in the wells being monitored.

CONCLUSIONS
The main conclusions of this study are: Asphaltene deposition has been observed in several high GOR wells. The precipitation has been linked to gas coning/cresting in these wells. No significant productivity decline has been observed in the wells. However, solid buildup in some wells has resulted in loss of wellbore accessibility of more than 30 meters (100 ft). The gas from the gas-cap titrates the produced oil and causes precipitation of asphaltenes that deposit in the wellbores. A commercially available PVT apparatus was used successfully to determine both the onset and amount of asphaltene precipitation with gas titration. The data indicate that the precipitation onset occurs at relatively low GORs. The amount of precipitation is very small at onset conditions. The amount of asphaltene precipitation increases with increasing gas volumes or GORs. A PVT simulator was partially successful in providing precipitation trends. It can be a useful tool to estimate precipitation envelopes and trends. However, it needs to be tuned with some experimental data. Simple tests are described to select a suitable solvent for asphaltenes. Several commercial asphaltene dispersants were also evaluated for their effectiveness in dispersing asphaltenes. One of the dispersants was used with live fluids, and the results are positive in terms of a reduction in asphaltene precipitation at reservoir conditions. The presence of asphaltenes also makes the produced emulsions tight at the wellhead. This may lead to emulsion problems at the wet crude handling facilities.

Fig. 20. Asphaltene content of wellhead oil samples.

Fig. 21. Asphaltene precipitation boundaries for a well fluid with injected gas.

reduced below the bubble point. These effects have resulted in preventing severe asphaltene buildups in the wells under consideration. Another factor that has kept the deposition under control is the high flow rates in these wells that prevent the asphaltene flocculation process and carry the asphaltenes to downstream processing facilities. Some deposition is taking place in the wells as observed in the bailer samples.

PROPOSED FIELD TRIALS

The recommendation being made to production engineering is to select a well and clean it with a solvent. Based on experiments conducted, a 50:50 xylene and diesel mixture is recommended for field trials. The solvent mixture will be allowed to soak for a minimum of 24 hours, and the wells will be flowed back. Samples from the flow-back will be analyzed to determine the effectiveness of the solvent cleanout procedures, and recipes will be modified for future cleanouts. While there are no plans at this stage to use dispersants (for their cost), their use will be evaluated after a few wells have been cleaned and their performance measured. The field trials will be described in a future paper.

R E C O M M E N D AT I O N S
The main recommendations of this study are: The oil wells should produce at as low a GOR as possible. This will reduce the amount of asphaltene precipitation and subsequent deposition. Continuous monitoring of asphaltene buildup in the

48

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005

wellbores should be maintained. Clean wells with a solvent. Xylene is preferable. A 50:50 mixture of xylene and diesel is adequate provided enough soaking time is allowed. Monitor the cleanout procedures to improve recipes for future cleanouts Investigate the use of asphaltene dispersants in severe deposition cases. Suspended asphaltenes in the crude may lead to emulsion and other production problems at the wet crude handling facilities and may necessitate increased demulsifier dosages.

REFERENCES
(1) Kokal, S.L. and S.G. Sayegh, (1995) Asphaltenes: The Cholesterol of Petroleum, SPE 29787 paper presented at the 6th Middle East Oil Show, Manama, Bahrain, March. (2) Michell, D.L. and J.G. Speight, (1973). The Solubility of Asphaltenes in Hydrocarbon Solvents, Fuel, vol 53, pp 149-152. (3) Leontaritis, K.J. and G.A. Mansoori, (1989). Asphaltene Deposition: A Comprehensive Description of Problem Manifestations and Modeling Approaches, SPE paper 18892 presented at the SPE Prod. Operations Symposium held in Oklahoma City, Okla., March 13-14. (4) Hammami, A., D. Chang-Yen, J.A. Nighswander and E. Strange, (1995). An Experimental Study of the Effect of Paraffinic Solvents on the Onset and Bulk Precipitation of Asphaltenes, Fuel Science & Tech. Int., Vol 13(9), pp 1167-1184. (5) Kokal, S.L. et al., (2002). Productivity Decline in Oil Wells Related to Asphaltene Precipitation and Emulsion Blocks, SPE paper 77767 presented at the SPE Annual Technical Conference and Exhibition, San Antonio, Texas, October. (6) Stephenson, W.K., (1990). Producing Asphaltenic Crude Oils: Problems and Solutions, Petroleum Engineer Int., pp 24-32. (7) Alleson, S.J., A. Hammami, H. Maeda and K. Ohno, (1996). Control of Asphaltene Deposition Laboratory Screening and Field Evaluation of Asphaltene Inhibitors. (8) Yen, A., Y.R. Yin and S. Asomaning, (2001). Evaluating Asphaltene Inhibitors: Laboratory Tests and Field Studies, SPE paper 65376, Intl. Symposium on Oilfield Chemistry, Houston, Texas.

ACKNOWLEDGMENTS
The authors would like to acknowledge Saleh Al-Khudair, Nashi Al-Otaibi from Production Engineering and Hisham Nasr-El-Din for solvent cleaning and stimulation design. The onset and bulk deposition tests were performed by Mohammad Al-Dokhi and Hussain Al-Hakeem. The bailer samples were analyzed by N.S. Meeranpillai and Mohammad Al-Awaisi.

N O M E N C L AT U R E
Gas-oil ratio Gas oil separating plant High-pressure production trap Pressure-volume-temperature Solids detection system GOR GOSP HPPT PVT SDS

SAUDI ARAMCO JOURNAL OF TECHNOLOGY SPRING 2005 49