186 Energy & Fuels 2009, 23, 186–192

Comparison of Conventional and Catalytic in Situ Combustion

Processes for Oil Recovery
Musa B. Abuhesa and R. Hughes*
Cockcroft Building, Institute for Materials Research, UniVersity of Salford, M5 4WT Manchester, U.K.

ReceiVed September 23, 2008. ReVised Manuscript ReceiVed NoVember 4, 2008

Laboratory in situ combustion experiments were performed to test the concept of using in situ combustion
to catalytically upgrade a medium-heavy Clair crude oil (19.8° API), using a low-pressure combustion cell.
Tests were performed for both noncatalytic and catalytic procedures and for both dry and wet forward combustion
runs. The results showed that the presence of catalyst promotes the combustion reactions and the resultant oil
was upgraded by up to 11 points. High oil recoveries were achieved for all runs and ranged from ∼53 to 74%
original oil-in-place. The highest oil recovery was achieved during wet noncatalytic tests, while the lowest
was achieved during the dry catalytic tests.

Introduction
In situ combustion (ISC) is a thermal recovery process in
which heat is generated within the reservoir by igniting a
part of the oil-in-place to improve the flow of the unburnt
region. The combustion is supported by continuous injection
of air or oxygen-enriched air or steam/air injection at one or
more wells. In the field, the ignition is started either by
electrical means or by a gas burner, and in some cases
autoignition can be achieved.1
There has been interest in the application of catalysts to the
ISC process in order to promote further upgrading of the
produced oil. Fassihi et al.,2 Greaves et al.,3 and Rashidi and
Bagci4 studied the effect of the clay content and metallic
derivatives on the oxidation reaction. Lower activation energies
were observed for the various reactions involved in the
combustion process. They also reported that increasing the clay
content increased the air requirement because of increased fuel
availability, which is indicative of the effect of the clay and
metallic additives. The addition of a nickel/alumina catalyst to
the sandpack in a combustion tube experiment, using a Middle
East crude of 18° API gravity, has been reported to give an
8-point gravity increase in addition to a 50% sulfur removal in
the produced oil.5 Similar upgrading of the produced oil has
* To whom correspondence should be addressed. Telephone: +0044-
161-2955081. E-mail: r.hughes@salford.ac.uk.
(1) Tadema, H. J.; Weijdema, J. Spontaneous ignition of oil sands. J.
Oil Gas 1970, 68 (50, Dec), 77–80.
(2) Fassihi, M. R.; Brigham, W. E.; Ramey, H. J. Reaction kinetics of
in-situ combustion: part 1- observations. Soc. Pet. Eng. J. 1984, 24 (Aug),
399–407.
(3) Greaves, M.; Field, R. W.; Al-Shalabe, M. I. In-Situ combustion
studies for North Sea Forties and Maya crude oils. J. Chem. Eng. Res. Des.
1987, 65, 29–40.
(4) Rashidi, F.; Bagci, A. S. Effect of pressure and clay content on
combustion kinetics of heavy oil in limestone medium. 5th UNITAR
Conference, Caracas, Venezuela, Apr 4-9, 1991; Elsevier: New York, 1991;
Vol. 32, pp 3-327.
(5) Moore, R. G.; Laureshen, C. J.; Mehta, S. A.; Ursenbach, M. G.;
Belgrave, J. D.; Weissman, J. G.; Kessler, R. V. A down-hole catalytic
upgrading process for heavy oil using in-situ combustion. J. Can. Pet.
Technol. 1999, 38 (13), 96–72.
been reported by Weissman et al.6 and for light crude oil by
Kisler and Shallcross.7 The effect of rocks of varying miner-
alogical composition and of varying clay content has also been
studied by Ranjbar8 and Vossoughi et al.9 The former suggested
that the controlling mechanism for fuel formation is the
conversion of crude oil components by radical polymerization.
Although combustion and oxidation processes for oil recovery
have traditionally been associated with heavy oil, it has been
demonstrated recently that air injection into light oil reservoirs
can provide good recoveries,10,11 while Al -Saffar et al.12 have
shown that the separation of a light oil fractions into individual
SARA fractions followed by combustion can provide useful
information on this light oil oxidation process.
Most laboratory studies of ISC have used combustion tubes
or thermal analysis techniques. Experiments with catalyst
addition have employed both techniques, and in addition to the
combustion studies of Moore et al.5 and Weissman et al.,6 there
has been preliminary work by Bahri.13
The application of horizontal wells to thermal recovery via
steam-assisted gravity drainage (SAGD) and displacement
(6) Weissman, J. G.; Kessler, R. V.; Sawicki, R. A.; Belgrave, J. B. D.;
Laureshen, C. J.; Mehta, S. A.; Moore, R. G.; Ursenbach, M. G. Down-
hole catalytic upgrading of heavy crude oil. Energy Fuels 1996, 10, 883–
889.
(7) Kisler, J. P.; Shallcross, D. C. The effects of metallic catalysts on
light crude oil oxidation. In situ 1996, 20 (2), 137–160.
(8) Ranjbar, M. Improvement of medium and light oil recovery with
thermo-catalytic in-situ combustion. J. Can. Pet. Technol. 1995, 34 (8, Oct),
25–30.
(9) Vossoughi, S.; Willhite, G. P.; El-Shoubary, Y.; Bartlett, G. W. Study
of clay effect on crude oil combustion by thermo-gravimetry and differential
scanning calorimetry. J. Therm. Anal. 1983, 27, 17–36.
(10) Yannimaras, D. V.; Sufi, A. H.; Fassihi, M. R. The case for air
injection into deep light oil reservoir. 6th European IOR Symposium,
Stavanger, Norway, May 23-29, 1991.
(11) Yannimaras, D. V.; Taffin, D. L. Screening oils for in-situ
combustion at reservoir conditions by accelerating-rate calorimetry. SPE
ReserVoir Eng. 1995, 10 (1), 36–39.
(12) Al-Saffar, H. B.; Hasanin, H.; Price, D.; Hughes, R. Oxidation
reactions of a light crude oil and its SARA fractions in consolidated cores.
Energy Fuels 2002, 15 (1), 182–188.
(13) Bahri, S. In-situ combustion for upgrading of heavy oil. Ph.D.
Thesis, Chemical Engineering Unit, University of Salford, Salford, U.K.,
2002.

10.1021/ef800804a CCC: $40.75  2009 American Chemical Society

Published on Web 12/08/2008
In Situ Combustion Processes for Oil RecoVery
Figure 1. Schematic flow diagram of the experimental apparatus.
Figure 2. Cross-sectional view of the combustion cell (modes 1 and 2).
processes is documented by Aguilera et al.14 The combination
of horizontal well technology coupled with catalyst addition
surrounding a horizontal well in the form of a perforated tube
has been studied extensively by Greaves and co-workers over
the past few years,15-18 and the improved recovery and
upgrading of the produced oil have been demonstrated. In the
present paper, we report on experimental studies of ISC of a
medium-gravity crude in a combustion tube at relatively low
pressure, comparing the results of (a) normal combustion, (b)
combustion with the addition of catalyst, and (c) combustion
with the addition of catalyst and with a perforated tube acting
as a horizontal production well.
Experimental Section
Apparatus. The major components of the experimental apparatus
used in the experimental setup are shown schematically in Figure
1. The system consists of the following interrelated parts: combus-
tion cell, flow metering and pressure regulation units, temperature
measurement and control, fluid production unit, and gas analysis
(14) Aguilera, R.; Artindale, J. S.; Cordell, G. M.; Ng, M. C.; Nicholl,
G. W.; Runions, G. A. Horizontal Wells: formation eValuation, drilling
and production, including heaVy oil recoVery; Gulf Publishing Company:
Houston, TX, 1991.
(15) Greaves, M.; Tuwil, A. A.; Bagci, A. S. Horizontal producer wells
in in-situ combustion (ISC) processes. J. Can. Pet. Technol. 1993, 32 (4),
58–67.
(16) Greaves, M.; Al-Shamali, O. In-situ combustion (ISC) process using
horizontal wells. J. Can. Pet. Technol. 1996, 35 (4), 49–55.
(17) Xia, T. X.; Greaves, M. 3-D physical model studies of down-hole
catalytic upgrading of Wolf Lake heavy oil using THAI. Canadian
International Petroleum Conference, Calgary, Canada, June 12-14, 2001;
CIPC: Calgary, Canada, 2001; Paper 2001-17.
(18) Xia, T. X.; Greaves, M. Down-Hole upgrading of Athabasca Tar
Sand Bitumen using THAI-SARA analysis. Presented at SPE International
Thermal Operations and Heavy Oil Symposium, Margarita Island, Ven-
ezuela, Mar 12, 2001; SPE 69693.
Energy & Fuels, Vol. 23, 2009 187
unit. The combustion cells used (Figures 2 and 3) comprise a
cylindrical tube constructed from 316 stainless steel with dimensions
530 mm in length, 53 mm inside diameter, and 3 mm wall thickness.
Gas flow through the system was controlled by means of mass flow
controllers and measured with a rotameter flowmeter. The longi-
tudinal temperature of the tube wall was maintained within (10
°C of the set temperature by means of three separately controlled
heater windings and the maximum temperature set at 400 °C with
a ramp rate of 5 °C/min. Three thermocouples were positioned at
100 mm intervals on the outside wall of the tube to measure its
temperature. Another moveable thermocouple was inserted inside
the central thermowell to monitor the sandpack temperature. The
produced liquids were separated by centrifugation, after collection
in special bottles, prior to measurement of the oil and water
recovered and the produced oil’s API gravity. A Varian 3300 gas
chromatograph equipped with a “CTR I” column and thermal
conductivity detector head was used for analysis of the produced
gas composition. This column allows analysis of the fixed gases,
nitrogen (N2), oxygen (O2), methane (CH4), carbon dioxide (CO),
and carbon monoxide (CO2), at room temperature. The carrier gas
used was helium.
Procedure
Mixtures of washed Buckland silica sand (∼355 g), Clair oil,
and water (in some experiments) were used in this study; in
addition, catalyst was added if it was to be used in the catalytic
runs. Oil saturation of 20% or 40% and water saturation of 20%
(based on the sand pore volume) were used in this work. Table
1 shows the common operating conditions.
The sample bed was combusted in an air feed stream
containing 21% oxygen, with an operating pressure of 3 bar
and a temperature ramp of 5 °C/min, over the range from room
temperature to 400 °C. Gas analysis was carried out intermit-
tently at 15 min intervals during each experiment.
188 Energy & Fuels, Vol. 23, 2009 Abuhesa and Hughes

Figure 3. Cross-sectional view of the combustion cell (mode 3).

Table 1. Common Operating Conditions

parameter value
sand type coarse silica
sand weight, g 355
sand porosity, % 36.53
crude oil Clair
density, kg/m3 935
API gravity, °API 19.8
temperature, °C 400
heating rate, °C/min 5
pressure, bar 3
air flow rate, mL/min 250
air flux, sm3/(m2 h) 6.80

Three experimental modes were studied. First (R-1-R-4

without catalyst) and second (R-5-R-12 with catalyst) mode
experiments were performed by packing the sample bed (100
mm length) at the middle of the combustion tube and packing
the catalyst at the end of the sample bed in the catalytic runs.
The rest of the tube contained porcelain beads, for preheating
the inlet oxidant air and maintaining the temperature of the fluid
produced from the combustion to the bed exit. CoMo catalyst
was used at either 1 or 5 g loading in experiments R-5-R-8.
The catalyst used was a standard HDS catalyst of 1/16 in. CoMo
extrudates on an alumina support, presulfided and supplied by
Akzo Nobel BV, Amsterdam, The Netherlands. To test for any
decline in the catalyst activity following the catalyst experiments,
the same catalyst (5 g of CoMo) was used in a series of four
experiments (R-9-R-12), and this was collected and weighed
after each experimental run and reused in the next experiment
under the same conditions.
Third mode experiments (R-13-R-16) were performed using
a small perforated tube, which serves as a horizontal well, fixed
in the lower part of an outlet fitting of the combustion cell
(Figure 3). For catalytic runs using this mode, the catalyst was
packed around the perforated tube. The perforated tube was
made of 316 stainless steel of 9.1 mm o.d. with five perforations
of 1.5 mm diameter at 10 mm intervals arranged in a single
row of phase angle 360°. The perforated section was covered
with 100-gauge wire mesh to prevent the sand from entering
this production line.
Results and Discussion
The combustion processes were very similar because the
sample beds in all experiments were identical and the same
processing parameters were used. Because of this, the analysis
of the results was based on the stabilized period of the
combustion run (HTO region).
Noncatalytic ISC Experiments (First Mode). The effluent
gases from the combustion cell are the result of different
Figure 4. Temperature and produced gas composition vs time: R-1,
without catalyst, So ) 20%, Sw ) 0%.
chemical reactions occurring in the ISC process. Carbon oxides
(CO2 and CO) are the major uncondensed products that are
produced by these reactions; together with any unreacted
oxygen, they can provide a good indication of the performance
of the ISC process. Generally, the outlet gases (CO2, CO, and
O2) in the individual composition profiles tend to exhibit similar
trends.
Figures 4 and 5 illustrate the composition of the outlet gases
CO2, CO, and O2 with combustion temperature as a function
of time for R-1 and R-2. They represent the general behavior
of the combustion for all four runs. Table 2 summaries the
results during the stabilized period of the combustion run for
each experiment.
R-1 was performed as a dry forward combustion run, with
an oil saturation of 20%. From Figure 4, the carbon oxides (CO2
and CO) started production at about 1 h from the beginning of
the experiment, while at the same time, most of the oxygen
supplied has been consumed. The CO2 and CO concentrations
during the stabilized period of combustion averaged 15.02 and
3.87%, respectively, while the O2 concentration was stable at
about 0.93%, which indicates that most of the oxygen has been
consumed by combustion reactions.
R-2 was carried out as the same conditions as R-1 but with
a higher oil saturation of 40%. Figure 5 shows that the values
of the CO2 and CO concentrations at the stabilized period of
combustion averaged 15.59 and 4.47%, respectively, and an O2
level at about 1.20%. Compared with R-1, the duration of the
reaction time for the stabilized period of carbon oxides and
In Situ Combustion Processes for Oil RecoVery Energy & Fuels, Vol. 23, 2009 189

Table 3. Combustion Performance: Catalyst Addition

Experiment
experiment
R-5 R-6 R-7 R-8
oil saturation So, % 20 40 40 40
H2O saturation Sw, % 0 0 0 20
catalyst weight, g 1 1 5 5
combustion time, h 4.25 5 5.75 5.75
air requirement, (sm3/m3) 226.7 272 309 283.3
fuel burnt, % OOIP 16.9 11.1 17.4 16.1
oxygen utilization, % 94.86 94.81 94.71 93.9
fuel consumption, kg/m3 11.45 15.11 23.57 21.8
oil recovery, % OOIP 53.1 65.3 61.3 67.2
API gravity, °API 23.8 23.4 27.1 29.2

respectively, than those in R-2. Another effect observed was a

decrease in the peak temperature to 447 °C, compared with 464
°C in R-2. The presence of water results in an increase in steam
production. The decrease in the peak temperature in this
combustion experiment is probably due to the combined effect
Figure 5. Temperature and produced gas composition vs time: R-2, of external heat losses and increases in the steam zone size.
without catalyst, So ) 40%, Sw ) 0%.
All experiments in this set were carried out with an air flow
Table 2. Combustion Performance: First Mode (Noncatalyst rate of 250 mL/min, except R-4, which was conducted, with a
Experiments) higher air flow rate of 375 mL/min to investigate the effect of
experiment an increase in the air flow rate on the oxidation of the crude
oil. A comparison with those conducted under the same
R-1 R-2 R-3 R-4
conditions except for the air flow rate showed that the produced
oil saturation So, % 20 40 40 40
gas profiles had a similar trend. At this higher air flow rate, the
H2O saturation Sw, % 0 0 20 0
combustion time, h 4.25 5 5.25 3.5 reaction starts earlier and the duration of the reaction time was
fuel burnt, % OOIP 16.04 12.73 12.30 13.37 shorter as expected and is attributed to the higher combustion
oxygen utilization, % 95.6 94.3 94.3 94.2 front velocity (0.033 m/h) compared with 0.027 m/h for previous
air requirement, sm3/m3 206 247 226 309 experiments. Increasing the air flux from 250 mL/min (R-2) to
oil recovery, % OOIP 56.3 67 70.50 69.2
API gravity, °API 21.9 22.9 24.8 24.5 375 mL/min (R-4) increased the fuel consumption because of
fuel consumption, kg/m3 10.86 17.24 16.79 18.1 higher levels of thermal cracking at the high temperatures
achieved. Also, levels of carbon oxides production were higher.
The average oxygen utilization at the stabilized period was
very high for all experiments and ranged between 94 and 96%.
This is an indication that most of the oxygen was consumed in
the high temperature zone to produce CO and CO2.
In all of the experiments, a substantial oil upgrading was
obtained as evidenced by an increase in the oil’s API gravity
by ∼2 to 5° API, compared with that of original Clair crude oil
(19.8° API).
Table 2 gives the total oil recovery as a percentage of the
original volume of original oil-in-place (OOIP) and the produced
oil’s API gravity for each experiment with other combustion
parameters.
The presence of water gave a higher yield and a lighter oil
production. The oil recovered in R-3 (wet) was higher than that
of R-2 (dry), at 70.5 and 67% OOIP, respectively. Also the
produced oil in R-3 had a slightly higher API gravity than that
in R-2. This is probably due to the increased hydrogen
generation due to coke gasification as the water vaporizes and
contacts the residual coke ahead of the combustion front.
Figure 6. Temperature and produced gas composition vs time: R-3,
without catalyst, So ) 40%, Sw ) 20%. Catalytic ISC Experiments (Second Mode). The first four
runs (R-5-R-8) in this mode were performed as both dry and
oxygen production was slightly longer in R-2 than in R-1; also wet ISC experiments with catalyst packed at the outlet part of
the levels of carbon oxides production were slightly higher. This the sample bed. The heated oil and hot gases, generated by
is due to a higher oil content used in R-2. combustion reactions, are passed over the catalyst bed, and
In order to investigate the effect of water on the oxidation of hydrogenation takes place to produce upgraded oil. The other
the Clair crude oil, R-3 was performed at the same conditions four runs (R-9-R-12) were conducted to investigate the catalyst
as R-2 but with the addition of 20% water saturation (based on life with time. All runs follow the same operating conditions
the sand pore volume). Behavior similar to that of previous runs of the noncatalytic runs given in Table 1.
was observed (Figure 6). The results were compared with those The combustion performances of the four experiments in this
of R-2. CO2 and CO produced levels averaged at 15.91% and set are summarized in Table 3, with the measured produced
3.91%, respectively, which were slightly higher and lower, gases and maximum combustion temperatures illustrated in
190 Energy & Fuels, Vol. 23, 2009
Figure 7. Temperature and produced gas composition vs time: R-7,
with catalyst (5 g), So ) 40%, Sw ) 0%.
Figure 8. Temperature and produced gas composition: R-8, with
catalyst (5 g), So ) 40%, Sw ) 20%.
Figures 7 and 8 for experiments R-7 and R-8, respectively. The
plots are typical for all experiments in this series.
Generally, the carbon oxides started production at about 1 h
from the beginning of the experiment, while at the same time,
most of the oxygen supplied had been consumed. From a close
inspection of the figures and a comparison with the noncatalytic
experiments, the main finding in this set was that the CO
declines more rapidly with time compared with runs R-1-R-4.
Also, the CO2 level is significantly higher, mirrored by a lower
level of CO. This reduction in the CO level suggests that some
conversion of CO to CO2 has occurred, giving rise to a higher
value of CO2 than that produced in noncatalytic runs. Similar
results have been reported by Moore et al.,5 who attributed this
to the effectiveness of the heated catalyst zone in converting
CO and H2O into CO2 and H2 via the water-gas shift reaction
(WGS).
R-7 and R-8 were performed as dry and wet catalytic
combustion runs, respectively, with the same conditions as those
of R-6 but with 5 g of catalyst instead of 1 g; in addition, 20%
water saturation was used in R-8. Figures 7 and 8 (R-7 and
R-8) have almost the same period of combustion time, which
means that the presence of water has no significant effect on
the combustion time. The most interesting observation in
Abuhesa and Hughes
comparison between noncatalytic and catalytic experiments was
that in the catalytic runs the CO2 production level was higher
and was mirrored by a reduction in the CO level especially for
the heavier catalyst weight (5 g) runs. Also, the combustion
duration was almost the same when 1 g of catalyst was used,
but the duration was longer when more catalyst (5 g) was loaded.
This indicates that both the presence and loading of the catalyst
promote the combustion reactions.
From Table 3, the lowest value of the air requirement was in
R-5 (226.7 m3/m3), which was performed as a dry run with
lowest oil saturation (20%). The highest value was in R-7 (309
m3/m3), which was conducted with higher oil content (40%)
and higher catalyst weight. The air requirement in R-6 was 272
m3/m3, which is lower than that of R-7. This is due to the heavier
catalyst weight used in R-7 (5 g) compared with R-6 (1 g).
This indicates that more air is required as more catalyst is loaded
because more oxygen is needed. In R-8 (wet), the air require-
ment value is lower than that of R-7; this indicates that the
presence of water reduces the air requirement. This is in
agreement with the results reported by Bagci19 and Al-Saghr,20
who confirmed that wet experiments consumed less fuel and
required less air because of the high displacement efficiency
provided by steam generated during the wet experiments. Also
Greaves et al.3 stated that a considerable reduction in the air
requirement occurs with wet combustion because less fuel is
consumed. If the air requirement values of this catalytic set are
compared with those of a noncatalytic set, from Tables 2 and
3, it can be seen that the air requirement for catalytic runs is
observed to be higher than that of noncatalytic runs.
Produced Oil. The produced oil from ISC experiments is
the result of several processes such as pyrolysis, distillation,
and cracking, which are governed by temperature. Weissman
et al.6 passed the oil over a hydroprocessing catalyst bed located
in the heated zone, which resulted in a product that was
significantly upgraded as compared with either the original crude
oil or thermally processed oil (noncatalytic). In the present set
of experiments, the catalyst bed was packed at the outlet end
of the sample bed, in order for the produced oil to flow over it
at high temperature, facilitating the catalytic reaction to take
place; as a result, upgraded oil is produced.
From Table 3, the produced oil had a gravity increase that
varied upon the experimental conditions. For all of the four runs
(R-5-R-8), the gravity increased by ∼4 to 10°API, as compared
with the original Clair crude oil of 19.8°API. This was due the
combined effect of thermal and catalytic reactions. On the other
hand, the oil recovery as a percent of the original volume of
OOIP varies from about 53.1 to 67.2 volume % of the OOIP.
From this, it is seen that the oil production was affected by
experimental conditions. The API gravity of the produced oil
using 1 g of catalyst appeared to be independent of the oil
saturation at approximately 23.9° API, giving an increase of
∼4° API. When the catalyst weight was increased to 5 g, in
R-7 (dry) and R-8 (wet), the produced oil was more highly
upgraded to 27.1 and 29.2° API and about 7 and 9° API higher,
respectively, as compared with the original Clair oil. This
indicates that the presence of water promotes the catalytic
upgrading. Also, it is clear that the catalyst loading promotes
the crude oil combustion performance by producing higher oil
(19) Bagci, A. S. The application of dry and wet combustion on
limestone containing heavy oils with the analysis of combustion reactions
kinetics. Ph.D. Thesis, Middle East Technical University, Ankara, Turkey,
1986.
(20) Al-Saghr, A. M. Three-dimensional physical model studies of air
injection: in-situ combustion and downhole catalytic upgrading using
horizontal wells. Ph.D. Thesis, University of Bath, Bath, U.K., 1998.
In Situ Combustion Processes for Oil RecoVery
API gravity. This finding is in agreement with the results
reported by Xia and Greaves,17 who stated that when the total
catalyst loading was reduced, the level of oil upgrading
decreased.
Oil recovery increased from 53.1 to 65.3 volume % of OOIP
as the oil saturation increased from 20% (R-5) to 40% (R-6).
These two runs were performed with a catalyst weight of 1 g.
Using 5 g of the catalyst instead of 1 g in R-7 reduced the oil
recovery to 61.3 volume % of OOIP compared with the run at
the same conditions (R-6) because more oil burned when more
catalyst was loaded. More oil was produced in R-8 when 20%
of water saturation was used, compared with the dry run, R-7,
probably because of the high displacement efficiency provided
by the steam generated.
The API gravity of the produced oil and oil recovery were
compared for catalytic with noncatalytic runs. As can be seen
in Tables 2 and 3, the oil produced by noncatalytic ISC was
upgraded by about 2 to 5° API depending on the experimental
conditions, compared with the original crude oil, while in
catalytic ISC runs, the oil produced was upgraded by about 4
to 10° API, depending on the experimental conditions and
especially on the catalyst weight used. Greaves and Xia21
reported similar findings. Also, the presence of water gave a
further slight increase in the API gravity for both catalytic and
noncatalytic ISC tests. This is probably due to the increased
generation of hydrogen by the WGS with the catalyst used. Also,
the oil recovered as a percentage of the volume of OOIP from
catalytic runs was less than that from noncatalytic runs. This is
due to more oil being burnt during the catalyst runs.
The above results are consistent with those reported by Al-
Saghr,20 who concluded that the oil produced by noncatalytic
ISC experiments was upgraded by 2-3 points, while that
produced by catalytic ISC was upgraded by 7-10 points for
both dry and wet ISC experiments, respectively, compared with
the original crude oil. It was also stated that less oil was
recovered from the catalytic runs compared with noncatalyst
runs. He also reported that water injection was beneficial in
improving the quality of the produced oil by promoting
hydrogen generation by WGS and the possibility of restoration
of the catalyst activity by the presence of steam.
Catalytic Life Experiments (Second Mode). All four runs
were conducted under the same experimental conditions as those
in previous runs, with the catalyst placed at the tube exit, but
the same catalyst (5 g) was used in all four runs. After each
experimental run, the catalyst was weighed; it was then replaced
and reused in the next run. All runs were conducted under dry
conditions with an oil saturation of 40%.
The discussion of this set (R-9-R-12) is focused on the coke
deposition on the catalyst and the degree of produced oil
upgrading. Generally, the produced gas composition profiles tend
to exhibit similar trends. The duration of the combustion time
was almost the same in all runs because of the same common
combustion parameters. Also, the CO level was slightly higher,
while that of CO2 was slightly lower in R-11 and R-12 compared
with R-9 and R-10. This might be due to a decrease of the
catalyst activity with time, as can be seen in Figure 9. This
figure shows the results of a coke deposition weight increase,
based on the 5 g of fresh catalyst used for the first time in R-9
(reused in the next runs as collected), along with the produced
oil’s API gravity, both for each run. It is clear that the coke
deposition increased during the first two runs (R-9 and R-10)
(21) Greaves, M.; Xia, T. CAPRI down-hole catalytic process for
upgrading heavy oil: produced oil properties and composition. Canadian
International Petroleum Conference, Calgary, Canada, June 12-14, 2001;
CIPC: Calgary, Canada, 2001; Paper 2001-23.
Energy & Fuels, Vol. 23, 2009 191
Figure 9. API gravity and % of catalyst weight changes for catalyst
life tests. At 0 h, original oil and fresh catalyst used.
as expected at 16 and 20% by weight of the catalyst, after
combustion times of 4.5 and 9 h, respectively. This coke
deposition was mirrored by a significant increase in the produced
oil’s API gravity as compared with the original Clair crude oil,
which averaged ∼8° API. This is a clear indication that the
catalyst still retains a good activity after a total of 9 h of
combustion time. After 13.5 h (R-11), a slight decline of
both the coke deposition and the produced oil’s API gravity
occurred. The same behavior was noted after 17.5 h (R-12) with
almost the same degree of decrease as that in R-11. This is
probably due to coke accumulated on the catalyst surface, from
the first two runs (R-9 and R-10), which may limit the active
sites of the catalyst, when used in the second two runs (R-11
and R-12). From these results, it is clear that the catalyst activity
decreased with time, as can be observed in the slight reduction
of the API gravity for both of the last runs. Another factor is
that the catalyst used was initially presulfided, which is a
requirement for hydrogenation activity. In the absence of
sufficient sulfur in the feed stream, it is possible that with time
the carbon oxides produced and any residual oxygen present
could convert the sulfide form to oxide and thus reduce the
catalyst activity. Also the weight of the coke deposit decreased,
which may be attributed to the change in its composition when
used for longer periods or to its light fractions being volatilized
at high temperature. Xia and Greaves18 reported that coking of
the catalyst could reduce its activity.
Perforated Tube Experiments (Third Mode). Four experi-
ments were carried out in this series, two using no catalyst at
0% and 20% water saturation (R-13 and R-15) and two with
catalyst (R-14 and R-16) with again 0% and 20% water
saturation. Temperature and gas production plots for these
experiments were generally similar to those of the previous
experiments, and the main results are given in Table 4. In
general, the behavior during these experiments was similar, with
some differences in produced gas levels, depending on the
experimental conditions. As in the previous experiments, the
presence of water promotes the combustion reaction, as
represented by the slightly higher CO2 and lower CO levels
obtained. Also, the presence of catalyst around the perforated
tube (horizontal well) promotes oil upgrading. Comparing the
dry and wet catalytic runs (R-14 and R-16) with those of the
noncatalytic runs (R-13 and R-15), the level of CO2 is slightly
192 Energy & Fuels, Vol. 23, 2009 Abuhesa and Hughes

Table 4. Results of Third Mode Runs crude oil (19.8° API), using a low-pressure combustion cell.
experiment The produced oil from the noncatalytic experiments was
R-13 R-14 R-15 R-16 thermally upgraded by ∼2-5 points depending on the experi-
oil saturation So, % 40 40 40 40 mental conditions, whereas that produced oil from the catalytic
H2O saturation Sw, % 0 0 20 20 experiments was much more upgraded, by ∼5-10 points
catalyst no yes no yes because of the combined effects of thermal and catalytic
combustion time, h 5.5 5.75 5. 5 5.75
upgrading. The higher upgrading in the catalytic tests was
stabilized period, h 2.25 2.5 2.25 2.5
fuel burnt, % OOIP 10.9 12.0 10.5 11.7 achieved with a larger catalyst loading of 5 g, which confirms
oxygen utilization, % 93.4 94.6 89.2 90.8 that the presence and loading of catalysts are beneficial in
fuel consumption, kg/m3 14.8 16.3 14.2 15.8 improving the quality of the produced oil. In contrast, the
O2 requirement, sm3/m3 79.3 89.3 71.4 75.2
oil recovery, % OOIP 71.3 64.5 74.3 69.7 recovered oil from the catalytic experiments was lower than
API gravity, °API 23.1 27.9 25.3 30.7 that recovered from the noncatalytic experiments. The carbon
CO2, mol % 14.5 16.1 15.0 17.3 oxides molar ratio was lower in the catalytic experiments
CO, mol % 4.45 2.6 3.8 2.2
because of a lower level of CO, which was mirrored by a higher
Table 5. Comparison between Perforated Tube (Third Mode) level of CO2 compared with the noncatalytic runs, confirming
Experiments and First and Second Mode Experiments that the presence of catalyst promotes the WGS by converting
first and second modes third mode CO into CO2. During wet combustion tests, for both catalytic
R-2 R-3 R-7 R-8 R-13 R-14 R-15 R-16 and noncatalytic tests, a reduction in the peak temperature was
oil saturation So, % 40 40 40 40 40 40 40 40 noticed, and it was observed that the fuel consumption and air
H2O saturation Sw, % 0 20 0 20 0 0 20 20 requirement were slightly lower, compared with dry forward
catalyst no no yes yes no yes no yes combustion tests. However, they were slightly higher in the
fuel consumption, 17.24 16.79 23.57 21.8 14.8 16.3 14.2 8.81 15.8
kg/m3 catalytic tests compared with the noncatalytic cases. Also, the
oxygen utilization, % 94.3 94.3 94.71 93.8 93.4 94.6 89.2 90.8 recovered oil as a percentage of OOIP and the quality of
oil recovery, % OOIP 67 70.50 61.32 67.1 71.3 64.5 74.3 69.6
API gravity, °API 22.87 24.80 27.13 29.2 23.1 27.9 25.3 30.7 the produced oil as measured by the API gravity were slightly
higher, while that of CO is lower in the catalytic runs, as can higher than those of the produced oil during dry combustion
be seen in Table 4. tests. It is clear that the presence of water improves the
These higher values of CO2 suggest that some conversion of combustion processes. From the results obtained from the
CO to CO2 may have occurred, giving rise to higher values of catalytic life tests, the catalyst activity was observed to decrease
CO2 than those produced in noncatalytic runs. Al-Saghr19 has with time because of coke deposition on the catalyst surface.
reported similar results. Also, Moore et al.5 have reported similar The results obtained from the perforated tube (horizontal well)
trends of gas composition. They attributed this to the effective- experiments followed the same behavior as the second mode,
ness of the heated catalyst zone in converting CO and H2O into except that the oil recovered and the degree of produced oil
CO2 and H2 via WGS. upgrading were slightly higher. This could be caused by the
Comparison between Perforated Tube Experiments location of the sample bed, which was now placed close to
with First and Second Mode Experiments. Generally, in a
the outlet side of the combustion cell compared with that in
comparison of the runs of the third mode with those of the first
mode 1, when the sample bed was positioned in the center of
and second modes under the same conditions, for both catalytic
the combustion cell.
and noncatalytic experiments and dry and wet conditions, similar
behavior was observed. In the third mode (perforated tube)
experiments, slightly higher oil was recovered from these Acknowledgment. The authors acknowledge National Oil
experiments compared with the other modes. This slight increase Corporation and Zueitina Oil Company, Libya, for financial support
in recovered oil might be due to the placement of the sample of this research.
bed near the outlet of the tube compared with that in the first
mode, which was placed at the middle part of the combustion Nomenclature
tube. Also it confirms that more oil was produced using HTO ) high-temperature oxidation
horizontal wells. On the other hand, almost the same degree of ISC ) in situ combustion
upgrading of the produced oil from both modes was observed, R-1, R-2, ... ) run number
as can be seen in Table 5. The oxygen utilization was high in OOIP ) original oil-in-place
both modes and averaged ∼89-95%, but it was slightly lower WGS ) water-gas shift reaction
in the second mode.
Supporting Information Available: Additional information in
Conclusions the form of experimental plots of evolved gas analysis. This material
is available free of charge via the Internet at http://pubs.acs.org.
ISC experiments have been performed to investigate the
criteria of thermocatalytic upgrading of medium-heavy Clair EF800804A