Spe 139715 Ms

SPE 139715
Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Experimental Evaluation of the Flue-Gas Injection of Barrancabermeja
Refinery as EOR Method
J.E. Rivera De La Ossa, Ecopetrol; A. Bejarano Wallens, SPS Oil; A. Florez Anaya, Pacific Rubiales;
N.S. Santos, Universidad Industrial de Santander
Copyright 2010, Society of Petroleum Engineers
This paper was prepared for presentation at the SPE International Conference on CO2 Capture, Storage, and Utilization held in New Orleans, Louisiana, USA, 10–12 November 2010.
This paper was selected for presentation by an SPE program committee following review of information contained in an abstract submitted by the author(s). Contents of the paper have not been
reviewed by the Society of Petroleum Engineers and are subject to correction by the author(s). The material does not necessarily reflect any position of the Society of Petroleum Engineers, its
officers, or members. Electronic reproduction, distribution, or storage of any part of this paper without the written consent of the Society of Petroleum Engineers is prohibited. Permission to
reproduce in print is restricted to an abstract of not more than 300 words; illustrations may not be copied. The abstract must contain conspicuous acknowledgment of SPE copyright.
Abstract
Because of its petrochemical processes Barrancabermeja refinery produces a considerable amount of flue gas daily. The
Llanito oil field has an original oil volume of 314 million barrels (MBls) and it has produced 37.9 MBls up to December
2009, which corresponds to a recovery factor of only 12%, being really low for 50 years of exploitation.
The application of an enhanced oil recovery process represents an alternative for improving the final recovery factor. The
availability of sources (Water, gas, chemicals, steam, etc) and the reservoir characterization dictate the feasibility of the
project according to a study carried out. An environmentally undesirable by-product resulting from refining processes such
as flue gas (Nitrogen + CO2) becomes important when used as raw material in the secondary or tertiary recovery of oil which
affects the increase in production and the final recovery of hydrocarbons.
This study evaluated experimentally the effect of the injection of the Barrancabermeja refinery´s flue gas as an enhanced oil
recovery method applied to B Sandstone of the Llanito oil field. The volume of the refinery´s flue gas was characterizated
and calculated; a study of fluid-fluid interaction was carried out between the flue gas and petroleum at reservoir conditions.
The interaction between reservoir fluid and the porous media was evaluated and the best flue gas injection mechanism was
determined in order to increase the recovery factor for the B sands of Llanito oil field. In addition to increase the final
recovery factor, the study aims to reducing flue gas emissions and consequent environmental benefit.
Introduction1
Globally there is a concern to reduce the adverse effects of industrial flue gases on the environment. As a consequence, it has
been a lot of emphasis to inject the flue gases that contain high CO2 concentration and catch them in adequate geological
formations. A feasible way is inject it in oil reservoirs, by means of which it can be additionally achieved the enhanced
recovery of hydrocarbons, with the consequent increase in the final recovery factor in the reservoir.
The flue gas is relatively a cheap gas and widely available to improve the displacement efficiency of depleted reservoirs or
where the oil has not been reached, especially at the current levels of oil’s supply and demand worldwide. The profitability
2 SPE 139715
potential of the recovered oil through the injection of the flue gas from producer reservoirs as well as depleted and mature
reservoirs is more important, especially for reservoirs with little or null water production and for those with low porosity and
permeability where the waterflooding as a secondary recovery method is not possible.
The goal of this work was to evaluate experimentally the effect of the flue gas injection of the Barrancabermeja refinery as a

recovery method for the B sands of the Llanito field, and in order to accomplish this target it was done these evaluations and
studies with fluids and porous media representatives of the reservoir.
Fluid sampling was performed at reservoir and surface conditions, making “capture” a crude oil bottom sample from a zone
that it has not been drained in the B sands of the Llanito field and this was done in a newly drilling well, in other words, that
sample had the original conditions of the reservoir when the sampling was done. Subsequently, it was characterized the
reservoirs fluids and it was done the characterization study of the refinery emissions to verify that those streams have the
average and characteristical compositions of a flue gas. This was confirmed by reviewing studies that were found in the
technical literature. Other important target was quantifying the emissions and verifying the volume available of flue gas at
reservoir conditions.
A study of fluid-fluid interaction that include a PVT conventional analysis and special PVT analysis of bottom oil sample and
flue gas mixture, at different saturation pressures of this mixture was carried out.; this was the first PVT analysis of a fluid
sampled at reservoir conditions in the B sands of the Llanito field. Similarly, it was done the Displacement Efficiencies
evaluation for different processes of flue gas injection and its possible combinations of the processes with the continuous
waterflooding, obtaining the development of the following tests at porous media level: continuous waterflooding at pressure
and flow constant, injection of a flue gas Slug followed by the continuous waterflooding and the alternated gas-water
injection.
The purpose of these evaluations was to optimize the flue gas to be injected and to obtain the maximum recovery for each
process applied at reservoir conditions, obtaining some favorable Displacement Efficiencies for Slugs processes plus
continuous waterflooding and also for alternated gas-water injection by using less volume of flue gas. It was showed that the
injection of the flue gas is technically feasible and that the application in the B sands of the Llanito field is possible, which
will allow the dispose of some emissions of a pollutant gas and in this way reduce the effect of this gas on the environment.
2
Llanito field generalities
Llanito field was discovered in the 60’s by the Colombian Petroleum Company ECOPETROL, this has been
identified as a production mechanism: the gas in solution, partial water drive and it has been implemented artificial lifting
systems by mechanical and progressive cavities pumping. The field has been exploited for 50 years and the quality of the
crude oil is 21 ºAPI. The formation water has a high salinity average of 28.000 ppm of chlorides. The original oil reported is
314 MBOOIP, with a cumulative production to December 2009 of 37, 9 Mbls of oil and 19.000 MPC of gas.
The tertiary Mugrosa formation B and C zones are the main sands producers, with an original reservoir pressure estimated in
2400 psi to -5.200 feet under the sea level for the B zone and 2.800 psi to -6.200 feet under the sea level for the C zone. The
Llanito field is located in the Valle Medio del Magdalena basin in the department of Santander, at the north of the “De
mares” concession. It limits at the north with the San Silvestre swamp and at the west with the El Llanito swamp. The field is
located at 12 km at the north of the Barrancabermeja refinery, see Figure 1.
SPE 139715 3
The field area has an approximate extension of 70 km2 where it has been drilled about 223 wells, which 106 belong to the
Llanito field, being 63 producer wells by January 2010. Nowadays, it counts with a crude oil production of 4500 BAPD and
a gas production of 3 MPCD.
Figure 1. Location of Llanito field.

4 SPE 139715
Reservoir properties
The area is complex in the Llanito field, basically by the high statigraphy control observed in the electric logs.
Producer crude oil zones are observed with resistivity of 2-5 ohm-m and with initial water saturations that reached 60%.
Studies attributed the low resistivity to the content and existing type of clay.
In the Llanito field the 56% of the production comes from the B zone in the Mugrosa formation. The Table 1 presents the

petrophysics properties of the Llanito field and the Table 2 the fluids properties.
Table 1. Petrophysics properties of Llanito field.
Parameter Llanito Field
Zone A Zone B Zone C Zone D
Average Depth (ft) 4000 5200 6200 7000
Average Thickness (ft) 26 50 40 18
Average Porosity (%) 24 21 20 18
Average Permeability
480 340 240 160
(mD)
Swi (%) 42 51 46 47
Gas Drive + Water Drive

Production Mechanism
Gas Drive
Crude type Naphthenic
Table 2. Fluids properties of Llanito field.
Parameter Llanito Field
Zone A Zone B Zone C Zone D
API Gravity 18.4 20 22 24
Viscosity (cP) 50 19.4 12.1 5
Volume Factor 1.058 1.094 1.105 1.2
Volume Factor @ Pb 1.061 1.094 1.09 1.206
GOR (SCF/STB) 150 500 400 600
Bubble point (psi) 1600 2175 2240 2900
Characterization and quantification of flue gas of the Barrancabermeja Refinery1,3

One of the main goals of this project was to characterize and quantify the emitted flue gas of the refinery based on
technical criteria and worldwide experience. Taking into account the above, was confirmed if the available volume of gas is
enough to be injected at reservoir conditions and in this case in particular in the B sand of the Llanito field, which is show
follow:
SPE 139715 5
The flue gas characterization of the Barrancabermeja refinery, a complete analysis was performed monthly by plants and
global monthly, which was definite the stream of the flue gas. As a result was obtained characteristic values of molar
composition of Nitrogen, Carbon Dioxide, Carbon Monoxide, Methane and Oxygen stream present in the gas that is emitted
on environment. In 2009 the Ecopetrol-ICP run 115 monitoring with the device MADUR GA 21 and 52 gas chromatography
with the following average composition of flue gas:

CO2= 14,98 %
N2= 80,10 %
O2= 3,27 %
CH4= 0,06 %
CO= 1,68 %
Volume quantification of flue gas available at reservoir conditions

The main sources of flue gases generation of the Barrancabermeja refinery are ovens, boilers, catalyst regenerators
of the catalytic rupture of these units, the by-products of the hydrogen generation in the paraffin plant and UNIBON among
other processes. As it is presented in Table 3, it can be done a preliminary approximate of the flue gas volume with values
corresponding to the year 2009, without taking into account some processes that also emitting flue gas emissions. To create
this preliminary quantification it was taken into consideration these emissions streams:
Table 3. Emission volume of the flue gases in the Barrancabermeja refinery, average year 2009
3 3
Smokestack m /s ft /D
Distral Boiler* 37 112.894.992
Foster wheeler* 32 97.638.912
Balance unity Boiler* 20 61.024.320
North central 30 91.536.480
IV Model 19 57.973.104
Orthoflow 10 30.512.160
UOP I 31 94.587.696
UOP II 50 152.560.800
Sulphuric acid* 2,5 7.628.040
Total 231,5 706.356.504
Conversion factor 1 m3/s = 3051216 ft3/D
According to Table 3, nowadays 706 MMPCD of flue gases are produced in the refinery at normal conditions,
however, this quantity can be higher or lower depending on all the emissions can be collected or that they can go in and out
of service in the different petrochemicals processes. Therefore, it was done a sensitivity analysis of the available flue gas
volume in the surface and its corresponding equivalent flue gas volume at reservoir conditions. The values of the available
gas in the surface that were taken into account were: 300, 700, 1000 and 1500 MMPCD and the potential reservoir pressures
of 500, 1000, 1500, 2000, 2500, 3000, 3500, 4000 and 4500 psi at 144,5 F. To perform the sensitivity analysis of the flue gas
volume available at reservoir conditions, was used the PVT-PRO software. The volume at bottom conditions is presented in
Table 4.
6 SPE 139715
Table 4. Flue gas volume in surface and its equivalent at reservoirs’ conditions. (Sensitivity according to
the bottom pressure)
T
Volume measured at the surface to 14.7 psi, 60 ° F
Bottom pressure
500 1000 1500 2000 2500 3000 3500 4000 4500
(psi)

0.999
Z 0.9927 0.9921 0.9974 1.008 1.0231 1.042 1.0641 1.0888 1.1156
1
Volume at
surface Volume at bottom hole conditions y 144,5 °F, (MMPCD)
MMPCD
300 9.90 5.02 3.38 2.57 2.09 1.77 1.55 1.39 1.27
700 23.09 11.71 7.88 5.99 4.87 4.14 3.63 3.25 2.96
1000 32.99 16.72 11.26 8.56 6.96 5.91 5.18 4.64 4.23
1500 49.49 25.09 16.90 12.84 10.44 8.87 7.77 6.96 6.34
1,4,5
Evaluation of fluid – fluid compatibility
In order to analyze the behavior of flue gas from the refinery and live oil from sands B of Llanito field, the next evaluations
was performed:
- Sampling in the bottom of the B sands of the well Llanito 127 and quality control of the sample.
- Conventional analysis PVT.
- Study of interaction crude live - flue gas.
- Study of asphaltenes due to the gas injection.
Sampling in sand bottom and quality control of the sample of the well Llanito 127
First, it made the data adquisition of pressure and temperature values from the base of the interesting range in the B
sands of the well Llanito 127 just drilled and completed to surface, with those values it can obtain the gradient static of the
fluids and identify the contacts water - oil and gas presents in the well, the identify was done with a pressure sensor with
electronic memory PPS25 and a "Slick-Line" unit. After obtained the static gradient, it took a bottom simple to 6761 feet @
2350 psi and 144.5 ° F, the sampling was performed whit a Single Phase Sampler (SPS), which captures the sample to
reservoir conditions and keeping the sample at higher pressure than that obtained in reservoir conditions to avoid the
formation of two phases when get the sampling to the surface. The transfer pressure to the free piston cylinders was to 5500
psi.
Before perform the PVT analysis of fluids recovered, it was observed in the laboratory the opening pressure of the
sample chamber (SPS). In order to determine the gas/oil ratio and the API gravity of the residual oil it performed a
instantaneous release of gas, the results were density 0.8643 g/cc at 5000 psi and 144.5 °F, contains 307.3 cubic feet of gas at
14.7 psia and 60 ° F per barrel of residual oil, and the residual oil has an API gravity of 22.7. Furthermore, it performed a test
of pressure-volume ratio which established the bubble pressure of 2260 psia at temperature of 144.5 ° F. The above analysis
was done as a quality control of the sample before the conventional analysis PVT.
SPE 139715 7
Conventional PVT analysis

After a period of restoration at 5000 psi and 144.5 °F during 5 days of the bottom sample, it obtained the following
results: It has a density of 0.8674 g/cc to 5000 psi and 144.5 °F; from the test of instantaneous release of gas allowed to
establish that the fluid contains 288.9 cubic feet of gas at 14.7 psia and 60 °F per barrel of residual oil, see Table 5.

Table 5. Summary of test of Flash Liberation @ Reservoir Conditions
G.O.R. 288,9 SCF/STB
Bo 1,0904 Bl Res/STB
Oil density(Sat) 0,8674 g/cc
Dead oil density 0,9170 g/cc
Gas specific gravity 0,7079
Molecular Weight Used C30+

580 gr/mol
The molecular weight of the fluid was 194.51 lb/Lbmol, density of the residual oil at 60 °F was 0.917 g/cc and the
residual oil API gravity was 22.65. The fluid is subjected to expansion isobaric to 5000 psi from room temperature (68 °F) to
the reservoir temperature (144.5 ºF), obtaining a thermal expansion of 1.037 and a coefficient of Isobaric thermal expansion
4870 e-04 (1/°F). Through a study of Pressure - Volume ratio conducted at reservoir temperature (144.5 º F) it was found that
the fluid has a bubble pressure of 2175 psia and a density at saturation pressure of 0.8561 g / cc, View 6 table.
Table 6. Summary Pressure – Volume Ratio Test

Bubble Pressure (Pb) 2175 psia
Thermal Expansion @ 5000 Psi 1,037

Vol. @ 144.5 °F
Vol. @ 68 °F
Coef. Isobaric thermal expansion 4,87E-04 1/°F

5000 psig (68°F-144.5°F)
Fluid density @ reservoir

P.b psia y 144.5°F 0,8561 g/cc
Fluid compressibility @ Ty
( From P yac. at Pb ) 4,41E-06 1/psi
During the differential liberation of six stages at reservoir temperature was found that the fluid contains 268.74
cubic feet of gas at 14.7 psia and 60 °F per barrel of residual oil and 1.1590 barrels of saturated-fluid at reservoir temperature
per barrel of residual oil at 60 °F.
The density and viscosity of the oil were measured over a large range of reservoir pressures at reservoir temperature
using an electromagnetic viscometer and a high pressure and temperature densimeter. The viscosity at the bubble pressure is
8 SPE 139715
10.160 cP and density 0.85 g/cc, Table 7 shows the behavior of the oil viscosity under different conditions of pressure @
144.5 ° F.
Table 7. Viscosity and Density of Oil Saturated @ 144.5 ° F.

PRESSURE OIL DENSITY OIL VISCOSITY
(psia) (g/cc) (cP)

5011 0,8603 13,5400
4510 0,8584 12,8260
4010 0,8565 12,3140
3510 0,8539 11,6510
3010 0,8519 10,9590
2513 0,8496 10,4860
2311 0,8488 10,3470
Pb = 2175 0,8500 10,1600
2009 0,8521 10,1960
1509 0,8604 11,6930
1009 0,8692 13,3960
509 0,8790 15,1910
108 0,8878 16,9590
Study of interaction live oil - flue gas

The purpose of this study is to analyze the behavior of live oil sample obtained in the bottom sampling with the flue
gas to possible injection pressures at reservoir conditions, for which it prepared a synthetic gas with composition similar to
the average was obtained in the characterization of flue gas emissions from the refinery in Barrancabermeja.
The tests were performed by injection of known volumes of flue gas with the following molar ratios with respect to
the original fluid (Live Oil): 4.59%, 8.19%, 12.2%, 16.41%, which led to increases in the bubble pressure of the fluid, bubble
pressures are shown in Table 8, these bubble pressures were determined experimentally by testing of expansion at constant
composition made to the original fluid and the fluids obtained at each stage of gas injection.
Table 8. Saturation pressures for each mole fraction of gas injected.
Psat. (psia)
% molar injection
0,00 2160
4,59 2606
8,19 2961
12,2 3415
16,41 3915
After each injection of flue gas, fluid-gas system was left stirring for two days at pressure of 5000 psig and reservoir
temperature (144.5 ° F) to obtain a single phase and restore the new condition of the mixture. Table 9 shows the swelling
factors of the fluid obtained from the results of the tests. The results indicate that the increase of this property not present a
typical behavior, which could indicate the possible formation of a phase of asphaltenes.
SPE 139715 9
Table 9. Swelling Factor for each mole fraction of flue gas

Psat. (psia) Swelling Factor
1,0000
2160
2606 1,0207

2961 1,0291
3415 1,0332
3915 1,0436
For each injection were taken from the cell two portions of different fluid for determinate experimentally the
viscosity and density at each saturation pressure. The behavior of these two parameters can be seen in Table 10 and Figure 2
and Table 11 and Figure 3.
Table 10. Fluid density @ Psaturation for each flue gas mixture
Psat. (psia) Density(g/cc)
2160 0,8528
2606 0,8516
2961 0,8560
3415 0,8670
3915 0,8547
Figure 2. Fluid density at saturation pressure

10 SPE 139715
Table 11. Fluid viscosity at saturation pressure for each gas mixed with flue gas.
Psat. (psia) Viscosity (cP)
2160 10,16
2606 9,02
2961 7,97

3415 9,65
3915 8,22
Figure 3. Fluid viscosity at saturation pressure
The experimental behavior of these two properties does not match the result expected, because in the process of gas
injection is expected to decline further clarified in the density and viscosity of oil contacted by the gas. Additionally the
results indicate that the fluid is not kept uniform during the process, which is typical of asphaltenes precipitation processes.
Determination of onset precipitation asphaltene by injection of flue gas
The objectives were to evaluate the phenomena of precipitation and reversibility of the asphaltenes in a crude
sample of the well Llanito 127 at reservoir conditions, determine the onset of asphaltenes precipitation in a crude sample of
crude Llanito 127 mixed with a molar fraction of flue gas at temperature and pressure conditions and the test of reversibility
of asphaltenes. The results of these studies shown that if the asphaltenes present in crude precipitate when subjected to
thermodynamic changes and determine the thermodynamic condition to which these precipitation phenomena occur or if on
the contrary, their effect can be discarded as a potential risk or as a cause of damage. The applied technique corresponds to
"solid Detection by Near Infrared."
Restoration Stage: the bottom sample is subject to restoration processes that involves a thermal expansion, and then
keep stirring at temperature and pressure conditions for an approximate time from one to two weeks to restore
thermodynamic equilibrium of fluid in the reservoir. A portion of it was taken to the reservoir pressure equals to 2350 psi and
145 ° F to begin measuring the onset of precipitation.
SPE 139715 11
In the case of the sample which flue gas was injected, the restoration period was 8 days, at conditions of 5000 psi and 145 °F;
time which is expected to complete solubilization of the gas. After this time, the sample was ready to determinate the onset of
precipitation.
Reversibility test for asphaltenes: This test determines if the application of thermodynamic changes of pressure

flocculate or not asphaltenes and if the flocculation is reversible. This test only takes into account macroscopic changes of the
behavior of the asphaltenes phase in the sample. An aliquot was taken of the restored crude sample (60 cm3) and transferred
to a cell at 2350 psi and 145 º F; the sample was kept in balance for a period of 12 hours. The cell prepared for this test is the
equipment Bulk Volume, which has a filter with a membrane of 0.45 microns.
After reaching the equilibrium conditions of 2350 psi and 145 ° F. A crude oil sample was taken to measure the
content of asphaltenes, this sample did not pass through the filter of 0.45 microns (M1), then at 2350 psi and 145 °F is taken a
second sample of oil but now passing through the filter of 0.45 micron and identified asphaltenes present (M2). Similar data
of % of asphaltenes correspond to an optimal stage of restoration of the sample.
An isothermal expansion was done up to 2280 psi, value slightly higher than the oil bubble pressure, corresponding
to the maximum precipitation of asphaltenes. These conditions were taken another sample of crude to measure the content of
asphaltenes (M3). A decrease in the content of asphaltenes with respect to the sample M2 indicates that asphaltenes flocculate
when the pressure decreases. Finally, return to the original conditions by isothermal compression and take a fourth sample
(M4). The comparison between the M3 and M4 sample to able to determine the reversibility of the phenomenon. Table 12
lists the samples taken to reversibility test of crude without flue gas.
Table 12. Reversibility Test (Without Flue Gas)

Pass throw filter 0.45
Identification Conditions % asphaltenes weight
micrs
M1 P=2350 psi / T= 145°F No 3.96
M2 P=2350 psi / T= 145°F Si 3.98
M3 P= 2280 psi /T =145°F Si 2.64
M4 P=2350 psi / T= 145°F Si 2.84
Asphaltenes precipitation test
To determine the asphaltenes precipitation onset in the oil sample, we used a spectrophotometric technique. The
equipment used was the Near Infrared to have a system for measuring light transmission in the infrared region near to detect
optical changes in the oil and measure the asphaltenes precipitation onset. The test is qualitative and determines a condition
of temperature and pressure where asphaltenes are separated from the solution and show a tendency to join.
The transmittance of light is changing steadily during the experiment progresses and in the precipitation points the
transmittance changes abruptly whenever the asphaltenes are out of solution and block the passage of light between fiber
optic probes. To establish the asphaltenes precipitation onset is carried out an isothermal expansion by stages until
12 SPE 139715
corresponding bubble pressure and continuously monitors the transmitted light reading, the changes in reading are associated
with the asphaltenes precipitation threshold. The procedures described above are prerequisites for the development of EDA
(Envelope Asphaltenes Deposits), which corresponds to the region of thermodynamic space P-T, into which the asphaltenes
lose its stability and begin their flocculation and subsequent precipitation. The variation of the readings of % of transmitted
light is recorded by a photometer and stored for realization of the corresponding curve.

SARA analysis and Colloidal Stability Index (CSI): Table 13 lists the results of the SARA analysis performed to the
bottom sample SARA crude. With these we determined the Colloidal Stability Index (CSI) which allows knowing
(qualitatively) the trend of crude oil, which can be: a precipitate or not to precipitate asphaltenes. Given the Colloidal
Stability Index as:
Table 13. SARA analysis for crude oil of well Llanito 127.

Fraction Result %w
Saturated 47.529
Aromatics 36.737
Resins 11.466
Asphaltenes 4.269
S+A > 0.9, Crude oil with trend to asphaltenes precipitations.

AR+R
Where: S: saturated A: asphaltenes, Ar: Aromatics y R: resins.
For crude oil simple at bottom conditions of well Llanito 127.
S+A = 1.074; Results was a crude oil with trend to precipitation.
AR+R
Asphaltenes precipitation test in crude oil at bottom conditions: Pressure at which asphaltenes precipitation begins in a
sample at bottom hole of crude at Llanito 127 is 3100 psi @ 145 ª F. On this point, the asphaltenes have lost their connection
with the resins (It is they who keep them in solution) and have started to agglomerate to the size necessary to obstruct the
passage of light. The figure 4 presents the EDA obtained by PVT+ Simulator where it can appreciate the envelope or critical
region in which the crude sample of the well 127 Llanito present higher possibility to precipitate the asphaltenes. The critical
region of precipitation begins below the superior apex of the envelope (green curve) to the saturation curve (purple curve),
becoming more severe it precipitation at points near the saturation line.
SPE 139715 13

Figure 4. EDA for a bottom hole sample in well Llanito 127.
Asphaltenes precipitation test in a sample of crude oil and flue gas mixture: The pressure at which precipitation starts
asphaltenes in a sample of bottom hole crude and flue gas mixture of Llanito 127 is 4800 psi at 145 º F. In Figure 5 we can
see the onset of precipitation of asphaltenes in a sample of bottom hole Llanito mixed with flue gas. In this graph we can see
that the onset of precipitation occurs at 4800 psi where it was observed that the intensity of light transmitted through the
sample decreases and this point is considered as the beginning of the precipitation of asphaltenes.
Figure 5. Asphaltenes precipitation onset of crude oil at bottom plus flue gas at 13.3% molar for Llanito 127.
14 SPE 139715
1
Evaluation of behavior of flue gas injected with the porous media
The best method to evaluate miscible and immiscible coreflood experiment is the representative cores samples because it
provides a more accurate idea of the increase in the recovery using these mechanisms of displacement. By this means it is
possible to determine the optimal volume of flue gas to be injected to produce the best oil recovery. For our particular case
the method of recovery to be applied to fluids and sands B of Llanito field is immiscible for the characteristics of the oil at

reservoir conditions and fluid to be injected.
Given the above, tests of Displacement Efficiency were performed for defining the optimal scheme of flue gas injection in
the B sands of Llanito field. The various processes that were analyzed are: continuous injection of flue gas, the injection of a
"Slug" of flue gas followed by continuous injection of water and alternating gas \ water injection (WAG), all tests were
compared with continuous Water injection or waterflooding.
Determination of the optimal scheme for flue gas injection
For this case, the different processes of flue gas injection were compared with continuous waterflooding (WF) that was
previously analyzed in core samples of the B sands. The main target was to determine the feasibility of water injection in the
Llanito field and we could concluded that there is no problem about the critical flow rate neither sensitivity to water injection
and the best injection flow rate was 0.48 cm3/min equivalent to 2 BAPD / ft in field. The following displacement tests were
carried out in order to choose the best scheme or process of flue gas injection, Table 14:
Table 14. Summary of displacement tests performed

Test Nº Composite Nº Test type
1 2 Continuous water injection at Q = 0.48 cm3/min
2 1 Continuous flue gas injection Q = 1 cm3/min
3 2 Continuous flue gas injection at P = 4700 psi

Injection of flue gas “Slug” 0.3 VP followed by
5 2
continuous water injection (WF) at P = 4700 psi
6 2
Injection Alternating of flue gas / water ratio (1:3) at P =
7 2
4350 psi
Injection Alternating of flue gas / water ratio (1:1) P =
8 2
4350 psi
PV: Porous volume

Continuous flue gas injection: For this type of injection process were performed three coreflood experiments (Tests # 2, 3 y
4): Continuous flue gas injection at constant flow rate Q = 1cm3/min, Continuous flue gas injection at P = 4700 psi and
Continuous flue gas injection at P = 3600 psi. For these three injections of the flue gas is evident in general: poor control and
poor mobility Displacement Efficiency. Although, there was a good mobility up to 0.3 Porous Volumes Injected (PVI)
compared with WF. In general, after the gas breakthrough (BT) provides a very poor sweep. There was a very young
fingering gas possibly caused by the viscous interdigitation.
SPE 139715 15
Because of the low sweep in the tests at 3600 and 4700 psi at continuous flue gas injection, it was decided to inject a "Slug"
gas PV of 0.3 followed by Waterflooding (Tests # 5 y 6), as well as, two tests of WAG at different ratios (Tests # 7 y 8).
above, in order to improve mobility control, and to see if there is improvement in the ultimate recovery of oil. Next Table
presents the summary of the oil recories for these tests:

Tabla 15. Summary of the obtained oil recovery.
Final oil
Oil recovery during
Test recovery
breakthrough [%]
[%]
Waterflooding 37.5 41.2
3
Continuous gas injection @ Q= 1 cm /min 55.15 61.95
Continuous gas injection @ P= 4700 psi 18.7 23.64
Continuous gas injection @ P= 3600 psi 18.07 32.82
Slug gas injection 0.3 VP + waterflooding at 4700 psi 38.75 39.7
Slug gas injection 0.3 VP + waerflooding at 3600 psi 35.42 40.05
Injection Alternating of flue gas / water ratio (1:3) to compelte 0.3 PV
31.8 34.66
followed waterflooding at P = 4350 psi
Injection Alternating of flue gas / water ratio (1:1) to compelte 0.3 PV
35.44 38.56
followed waterflooding at P = 4350 psi
The oil recovery for continuous gas injection at constant rate is the higher, but, is not a practical process because it needs 2,8
PV of flue gas. At field level is not profitable. See the next Figure.
Figure 6. Continuous flue gas injection @ Q cte= 1 cm3/min vs Continuous waterflooding @ Q=0.48 cm3/min.
SPE 139715
Figure 8. Summary of a slug flue gas injection (0.3PV) + waterflooding vs waterflooding

Figure 7. Summary of the continuous flue gas injection vs waterflooding
16
SPE 139715 17

Figure 9. Summary of a water alternating gas vs continuous waterflooding.
Selection of the best scheme of flue gas injection in the B sands of Llanito field
With the purpose of selecting the best scheme of flue gas injection, these processes were evaluated:
- Continuous flue gas injection at constant pressure.
- Injection of a flue gas Slug = 0.3 VP followed by continuous waterflooding injection at constant pressure.
- Alternated flue gas / water, 1:1 ratio, until completing 0.3 PV of gas, followed by continuous gas injection at
constant pressure.
From the evaluation that was done, the best recoveries were obtained, compare with the continuous waterflooding, those are
shown in table 16 and the recovery tendencies can be observed in figure 10.
Table 16. Summary of the crude oil recoveries obtained with the best methods
Oil recovery during Final oil
Final Sor
Test breakthrough recovery
[fraction]
[%] [%]
Waterflooding continuous 37.5 41.2 0.26

Gas injection slug 0.3 VP + Waterflooding at
4700 psi 38.75 39.7 0.28
Gas injection slug 0.3 VP + Waterflooding at
3600 psi 35.42 40.05 0.27
Water alternating gas injection, 1:1 ratio, to complete
0.3 VP follow by WF at P= 4350 psi 35.44 38.56 0.29
18 SPE 139715

Figure 10. Best gas injection processes vs. Continuous Waterflooding (WF)
Analyzing both results, we obtain the following conclusions:

1. The process of alternated injection: flue gas /water with a 1:1 ratio until completing 0.3 PV of gas and followed by WF at
P=4350 psi was the best injection scheme using flue gas.
2. The processes that involved the injection of a flue gas Slug and the alternated gas/water injection allow the optimization
of the gas to be injected, and getting better displacement efficiency.
Table 17. Summary of displacement tests performed

Test Nº Composite Nº Test type
1 2 Continuous water injection at Q = 0.48 cm3/min
2 1 Continuous flue gas injection Q = 1 cm3/min

5 2
6 2
Injection Alternating of flue gas / water ratio (1:3) at P =
7 2
4350 psi
Injection Alternating of flue gas / water ratio (1:1) P =
8 2
4350 psi
PV: Porous volume

SPE 139715 19
Conclusions
1. According to the average of the obtained compositions through chromatography, the range of composition for the
majority components (CO2 and N2) in the analyzed gases from the principal emissions sources of the refinery are 11,6 and
16% for the CO2 and 79,5 and 80,6 for the N2. According to the bibliographic review the flue gas is composed of 85-88% of
nitrogen and 12-15% of CO2.

2. In accordance with the emissions’ volumes that were reported, it is count with a diary volume of flue gas of 706
MM ft3/D, which is a respectable value for any project of gas injection. This value could be higher since some flows are not
reported as units of Topping and soon begin to operate two large refining plants.
3. During the fluid-fluid interaction tests it was observed that the experimental behavior of the density, viscosity and
°API of the crude oil and flue gas mixture it does not correspond with the expected, due to that in the process of gas injection
there were expected a decrease on the density and viscosity of the fluid. Additionally, the results will indicate that the fluid
did not maintain homogeneous during the process, which is typical in the asphaltenes precipitation processes.
4. The analyzed bottom sample corresponding to the 127 Llanito well begins the asphaltenes precipitation at 3100 psi
and 145°F conditions, making it this point in the superior apex of the thermodynamics region, ADE (asphaltenes deposition
envelope).
5. According to the obtained results in the onset determination of asphaltenes precipitation, at original reservoir
conditions, pressure equal to 2350 psi and 145°F, we found asphaltenes in the formation or the reservoir.
6. The crude oil sample at bottom conditions corresponding to the well Llanito 127 with a mixture of flue gas at 13.3%
begins the asphaltenes precipitation at 4800 psi and 145°F conditions.
7. The injection of a flue gas slug of 0.3 PV of gas followed by continuous waterflooding at constant pressure of 4700
psi showed a favorable behavior in the first 0.4 PV injected, which is equivalent to 63% of the total recovery, but, after the
breakthrough, the sweep is not efficient. Compared with the waterflooding at 0.4 PV injected it is only get the 48.8% of the
total recovery, which shows that in the first 0.4 PV injected is more efficient the Slug of flue gas followed by waterflooding
at constant pressure.
8. The process of alternated injection of flue gas / water, 1:1 ratio, until to complete 0.3 PV of gas, followed by the
continuous waterflooding at constant pressure of 4350 psi presented a promising behavior until the first 0.6 PV injected
(gas/water cycle). In this point the recovery was 37% equivalent to 96% of the total recovery for this process, while for the
waterflooding at 0.6% PV it was obtained a recovery of 34,5%, equivalent to 83,7% of the total recovery. This shows that the
WAG (water alternating gas) 1:1 ratio was more efficient than the waterflooding until the first 0.7 PV injected and until
completing 1 PV is virtually similar.
9. The process of alternated injection: flue gas/ water with a 1:1 ratio until completing 0.3 PV of gas and followed by
waterflooding at 4350 psi was the best injection scheme using flue gas.
10. The processes that involved the injection of a Slug of flue gas and the alternated gas/water injection allowed the
optimization of the gas to be injected and that a better sweep is achieved, improving the mobility ratio.
20 SPE 139715
Recomendations
1. It is recommended to monitor refinery plants which are not measured the flow of emissions, for example Topping
units in order to have a total value or overall emissions of flue gas at the refinery.
2. From the onset study of asphaltenes precipitation it is advisable to carry out a measurement stage at the reservoir
level, consistent on doing a screening of dissolvent products and asphaltenes inhibitors.

3. From the onset study of asphaltenes precipitation it is advisable to evaluate the sample to other conditions of
temperature (at least 3) with the purpose of being able to obtain the entire EDA thermodynamical region.
4. It is advisable to do an analysis of the economical variables with regards of the best injection processes, where it is
taken into account the risk and uncertainty of applying them in a field.
Acknowledgements
This paper is based on industry data obtained by ECOPETROL S.A and ECOPETROL-ICP and the authors give thank to this
company. Special thanks also goes to the engineers Policarpo Romero J. and Luis Carlos Niño for their support in preparing
this paper.
References
1. RIVERA, J.E. Evaluación de la Inyección de Gas de Combustión de la Refinería de Barrancabermeja como Método
de Recobro en el Campo Llanito.
2. Informe, Ecopetrol S.A. Producción incremental Area Llanito. Año 2000.
3. SACHICA, J.A., NIÑO, Y.A., “Factibilidad técnico económica del uso del CO2 de la gerencia complejo de
Barrancabermeja en el recobro terciario de petróleo de los campos Yariguí Cantagallo, y San Silvestre, Gala y
Llanito”, Bogotá DC, 2005
4. Ecopetrol S.A., Instituto Colombiano del Petróleo-ICP, Reporte PVT-ID0157-122.
5. Ecopetrol S.A. – ICP, Determinación del onset de precipitación de asfáltenos en una muestra de fondo del pozo
Llanito 127. Laboratorio de Daños a la Formación. 2009.

Spe 139715 Ms

Uploaded by

Copyright:

Available Formats

Spe 139715 Ms

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Spe 139715 Ms

Uploaded by

Copyright:

Available Formats

SPE 139715

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Copyright 2010, Society of Petroleum Engineers

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Figure 1. Location of Llanito field.

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Table 1. Petrophysics properties of Llanito field.

Parameter Llanito Field

Zone A Zone B Zone C Zone D

Average Depth (ft) 4000 5200 6200 7000

Average Thickness (ft) 26 50 40 18

Average Porosity (%) 24 21 20 18

Gas Drive + Water Drive

Table 2. Fluids properties of Llanito field.

Parameter Llanito Field

Zone A Zone B Zone C Zone D

API Gravity 18.4 20 22 24

Viscosity (cP) 50 19.4 12.1 5

Volume Factor 1.058 1.094 1.105 1.2

Volume Factor @ Pb 1.061 1.094 1.09 1.206

GOR (SCF/STB) 150 500 400 600

Bubble point (psi) 1600 2175 2240 2900

Characterization and quantification of flue gas of the Barrancabermeja Refinery1,3

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Volume quantification of flue gas available at reservoir conditions

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Conventional PVT analysis

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Oil density(Sat) 0,8674 g/cc

Dead oil density 0,9170 g/cc

Gas specific gravity 0,7079

Molecular Weight Used C30+

Table 6. Summary Pressure – Volume Ratio Test

Thermal Expansion @ 5000 Psi 1,037

Coef. Isobaric thermal expansion 4,87E-04 1/°F

Fluid density @ reservoir

Table 7. Viscosity and Density of Oil Saturated @ 144.5 ° F.

(psia) (g/cc) (cP)

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Study of interaction live oil - flue gas

Table 8. Saturation pressures for each mole fraction of gas injected.

Table 9. Swelling Factor for each mole fraction of flue gas

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Psat. (psia) Density(g/cc)

Figure 2. Fluid density at saturation pressure

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Figure 3. Fluid viscosity at saturation pressure

Determination of onset precipitation asphaltene by injection of flue gas

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Table 12. Reversibility Test (Without Flue Gas)

M1 P=2350 psi / T= 145°F No 3.96

M2 P=2350 psi / T= 145°F Si 3.98

M3 P= 2280 psi /T =145°F Si 2.64

M4 P=2350 psi / T= 145°F Si 2.84

Asphaltenes precipitation test

Downloaded from http://onepetro.org/SPECO2/proceedings-pdf/10CO2/All-10CO2/SPE-139715-MS/1756226/spe-139715-ms.pdf/1 by Universidad de America Bogota user on 17 October 2022

Table 13. SARA analysis for crude oil of well Llanito 127.

S+A > 0.9, Crude oil with trend to asphaltenes precipitations.