evaluation of air injection eor to apply a light oil...

32nd Annual Symposium & Workshop IEA Collaborative Project on Enhanced Oil Recovery

- 1 -

Evaluation of Air Injection EOR to apply a Light Oil Reservoir with Low Oil saturation after water flooding

Katsumo Takabayashi, Haruo Maeda, Hiroshi Uematsu INPEX CORPORATION

Akasaka Biz Tower, 5-3-1 Akasaka, Minato-ku, Tokyo 107-6332 Japan [email protected]

Abstract

Air injection EOR operation for light oil reservoirs has been paid an attention as a low cost enhanced oil recovery process for more than 10years. But, the actual application of the process is limited because of its complicated mechanism.

Firstly, Combustion Tube (CT) test which is one of the popular tests for the evaluation of in-situ combustion was conducted at two different initial oil saturations (40% and 17%) with the oil taken from our candidate domestic oil field. Results indicated higher oil recoveries (71% and 76%, respectively) even in the low oil saturation. Based on these results, we considered air injection would be a promising approach for our domestic oil field, in which most part of the oil reservoir has been already swept by water due to a strong aquifer support and especially some area is believed to be reached to irreducible oil saturation (13%). Following PVT experiments, Accelerating Rate Calorimetric (ARC) tests and CT tests, a thermodynamic model was constructed. Based on results of air injection simulation studies for the CT tests, with a newly constructed thermodynamic model, we concluded that the distillation process at high temperature (combustion) front would be a predominant mechanism of such higher oil recovery of the CT test. Sensitivity studies were also conducted with various number of pseudo components using constructed thermodynamic model. These studies indicated number of pseudo components drastically affects to the distillation process of lighter hydrocarbons. Air injection simulation was also carried out for the field-wide with a simplified sector model incorporated with the constructed thermodynamic model. The results indicated that oil bank can be formed by air injection but the injected air overrides the oil bank and causes early oxygen breakthrough in the case of low oil saturation reservoir, resulting in a low incremental oil recovery. On the other hand, in the case of higher oil saturation reservoir, a larger oil bank is formed so that the timing of oxygen breakthrough becomes more moderate and a higher incremental oil recovery is expected. For example, more than 10% of incremental oil recovery will be expected if residual oil saturation is as high as 30%.


- 2 -

Finally, we concluded that air injection would be a valuable EOR even in the water-flooded reservoir, if residual oil saturation is still higher, but our candidate oil field has been entirely water-flooded and oil saturation is insufficient to apply the air injection.

Introduction

The air injection process is a unique enhanced oil recovery process which is applicable to light oil reservoirs by injecting inexpensive and ubiquitous air. The injection scheme is very similar to a fire flood which is applicable to heavy oil reservoir. In-situ reaction of oxidation or combustion between oil and injected air becomes a dominant role to enhance the oil recovery. However, due to the differences of oil properties between light oil and heavy oil, the mechanisms of oil recovery are also different [1]. For example, reduction in oil viscosity by thermal effects is a critical factor to enhance the recovery of heavy oil but such oil viscosity reduction is less important in the air injection due to comparatively low viscosity of the targeted oil.

The fire flood was patented in 1920 and the field trial started in the US. Many field

projects were reported in 1960’s but then its number declined year by year. The air injection projects at light oil reservoirs increased in 1990’s, however the current commercial projects are operated only in Williston Basin in the US [2-5]

Many technical papers[6-8] have described oil recovery mechanisms of the air injection process. The followings are principal factors occurring in-situ under the air injection;

(1) Field re-pressurization by injecting air, (2) Thermal effects by the oxidation or combustion, (3) Oil swelling mainly by the in-situ-generated CO2, (4) Oil viscosity reduction by CO2 dissolution, (5) Vaporizing oil by thermal effects, (6) Oil sweeping by the flue gas, (7) Super-extraction by water and (8) Near-miscibility/miscibility of the generated flue gases and the oil

One of the most important factors is replacement of the oil by the flue gas so that the air injection is called sometimes as the flue gas injection. When the air injection is applied to the water flooded oil reservoir that is under the irreducible oil saturation, the flue gas drive will not be effective in incremental oil recovery. But Moore et al. [1,9] mentioned that a high temperature reaction zone (combustion zone) may play an important role, that is vaporization of the oil by thermal effects. The vaporized oil is driven by the flue gas and moves forward. When such vaporized oil reaches in the low temperature zone, it starts condensing and forms the oil bank. If such distillation process works effectively in the reservoir, we will be able to expect a successful application of the air injection to the oil reservoirs under the irreducible oil saturation.


- 3 -

Laboratory tests

For the purpose of collecting fluid properties of the oil taken from our candidate oil field, PVT tests, Accelerating Rate Calorimeter (ARC) tests, Thermo Gravimetric Pressurized Differential (TG/PDSC) tests and High pressure Ramped Temperature Oxidization (HP-RTO) tests were carried out. .All thermal experiments were conducted with the stock tank oil and the major thermal properties are tabulated in Table 1.

The Combustion Tube tests (CT test) were conducted at lower oil saturation case

(So=17%) and relatively higher oil saturation case (So=40%). The testing procedure is as below;

1 Pack with the crushed core 2 Saturate with the brine (measure the porosity) 3 Inject the brine at a set rate to determine the permeability 4 Inject the stock tank oil until its breakthrough 5 (In lower oil saturation case) Inject water until its breakthrough to simulate

waterflooded condition 6 Inject the stock tank oil to pressurize to 3,600psig and preheated to 103degC 7 Inject N2 to establish air permeability while heating the inlet up to 200degC 8 Start air injection

The test results are shown in Figure 1 A stable combustion front was formed and an

incremental oil recovery was observed in both tests.

Simulation Study

Parameters setting for simulation studies

Firstly, we defined five fluid models as shown in Table 2 and EOS parameters of each pseudo-component were determined by matching with the results of PVT tests. The stoichiometry and kinetic parameters of oxidation/combustion process were tuned through history matching with the results of CT tests by using a thermal simulator “STARS” developed by Computer Modeling Group Ltd., Calgary. Table 3 shows the one-dimensional model used for the history matching and Figures 2-4 show the results of the history matching with five fluid models (FM1 - FM5), which are adjusted to represent the composition of the stock tank oil. In our simulation study, the pseudo-components of heavier than C4 were confirmed to be consumed by oxidation/combustion process, so that they were defined as a fuel in the thermodynamic model.

Secondly, we constructed three-dimensional model representing a quarter of the 5-spot pattern model as shown in Figure 5 and Table 4. and conducted simulation study to evaluate the effects of number of pseudo-components and air injection rate on production performance by using this three dimensional model. Figures 6-9 shows the production performances with various air injection rate of 100 - 0.1 Sm3/m2-hr. These results show that the oil production performances at a higher air injection rate are not so much affected by the number of pseudo-components, however the effect of the number of pseudo-components becomes larger at a lower air injection rate. In our calculation, six(6) pseudo-components model (FM1) showed


- 4 -

the most intermediate performances between them in terms of the time of oil production start and the amount of cumulative oil production in volume. These results suggested that the six(6) pseudo-components model (FM1) would be the most adequate fluid model in this study.

Figure 10 shows the oil recovery in weight at the air injection rate of 0.1 Sm3/m2-hr. The

oil recovery (wt%) became the lowest at the six(6) pseudo-components model (FM1) . We can explain these phenomenons as follows. Figures 11 and 12 show the compositional distribution simulated by the one dimensional model with the fluid model FM1, which represents the composition of the live oil composition as shown in Table 2, at 3.7PV of the air injected with air injection rate of 1.0 Sm3/m2-hr and 0.1 Sm3/m2-hr, respectively. The solid line in each figure indicates temperature profile and the point of the peak temperature shows the location of the combustion zone. At the high air flux case, temperature in front of the combustion zone shapely drops and reached to the reservoir temperature. In contrast, at the low air flux case, a decrease of temperature in front of the combustion zone was more gradual. Such differences of the temperature profile in front of the combustion zone are caused by the heat transfer due to the differences of propagation velocity of combustion zone. Furthermore, at the low air flux case, noticeable heterogeneous compositional distribution was observed. Such behavior is caused by the in-situ distillation process. Light components vaporized by the proximity of the combustion zone move forward as gas phase and condense at low temperature zone. By such continuous distillation process in front of the combustion zone, the light components of the oil are stripped and the oil near the combustion zone becomes gradually heavier. Such remaining oil will be consumed as fuel by propagation of the combustion zone and light oil shall be produced preferentially.

Application of air injection to the candidate oil field

We decided to adopt the six(6) pseudo-components model as mentioned in the previous section. Our candidate field located in Japan and commenced oil production in the early 1964s. The reservoir is a stratigraphic trap type and a primary gas cap had been existed. The reservoir dip is steep in the southeastern structure. The oil reservoir is sandstone and the reservoir pressure is supported by a strong edge water drive. Almost entire reservoir area was swept by invasion of the edge water and oil saturation in the targeted area is around 13 %, which is closed to the irreducible oil saturation (Sor).

As shown in Figure 13, the reservoir model represents a half of the 5-spot pattern model

(335m x 167.5m) with 10 layers. The grid size is 4m x 4m x 1m. Reservoir properties in the model are assumed as homogeneous and the reservoir dip is assumed to be 13 or 30 degree angle. Table 5 shows the details of the reservoir model. As oil reservoir has been completely swept by the edge water, we assumed that the current oil saturation (13%) is the irreducible oil saturation (Sor) in this study to evaluate the possibility to apply air injection.

Firstly, we evaluated the effects of reservoir dip angle, ratio of vertical and horizontal

permeability (Kv/Kh) and air injection rate on the oil recovery at 13% of initial oil saturation which is setup as Sor. As shown in Figure 14, the effect of reservoir dip angle is insensitive. At the Kv/Kh=0.1 case, higher air injection rate is more preferable but even in such a case,


- 5 -

we cannot expect the incremental oil recovery more than 0.3%. Figure 15 shows the typical oil production behavior at the simulation run with reservoir dip angle of 13 degree, Kv/Kh=0.1 and air injection rate of 9MMscf/d. No oil produces until 1,500 days after staring air injection and oxygen breakthrough occurs immediately after oil production starts. When oxygen concentration in the effluents at the producer reaches 3 mol%, oil production is automatically shut down due to a safety reason.

Figure 16 shows the temperature and oil saturation profile at the cross-section between

injector and producer at the moment when oxygen breakthrough occurred. We can see the oil bank was formed by propagation of the combustion zone, however such oil bank has not reached at the producer due to the producer shut-in because the air including a high oxygen concentration overrode such oil and reached at the producer.

If the residual oil saturation (Sor) is larger than 13%, can we expect a higher oil recovery?

We conducted other simulation runs with Sor = 20% and 30% cases. Oil recovery is improved by an increase of the oil saturation (=Sor) as shown in Figure 17. The result suggests the current oil saturation at the candidate field is much closer to the lowest limit to apply the air injection. Figure 18 shows oil saturation and temperature profile of various residual oil saturation cases at 1,300 days after starting air injection. In case of Sor=30%, large oil bank was formed even though the reservoir is under the irreducible oil saturation. In addition, such oil bank itself prevents early oxygen breakthrough so that a high incremental oil recovery more than 10% was achieved.

We confirmed that the vaporization process and the flue gas drive by thermal effects

would play a dominant role when the air injection is applied to the oil reservoir under irreducible oil saturation. However, we have to conclude our candidate field is not suitable for application of the air injection due to mainly low oil saturation. If we could identify the area in which a higher remaining oil saturation is expected, we might justify our proposal to conduct a pilot test. But, we will need to investigate carefully the fact that some portion of oil is consumed as a fuel to maintain the combustion zone and its effect on the oil recovery.

Conclusions

We would like to conclude our study as follows:

1. Vaporization process and the flue gas drive by thermal effects would play a dominant role when the air injection is applied to the oil reservoir under irreducible oil saturation.

2. To represent the vaporization process in the reservoir simulation, the adequate number of pseudo-components has to be carefully evaluated. Otherwise such behavior will not be represented in the simulation study. In our oil system, 6-pseudo components model is selected to represent the vaporization process to be took place near the combustion zone.

3. Our simulation results indicated that oil bank can be formed even under the condition of the irreducible oil saturation. However, in the case of low oil saturation, the injected air easily overrides the oil bank and causes early oxygen breakthrough and a low incremental oil recovery.

4. In the reservoir with high oil saturation, we can expect forming a large oil bank and early


- 6 -

oxygen breakthrough will be prevented and a high incremental oil recovery will be expected.

Acknowledgement

The authors wish to thank INPEX Co., Ltd. and Japan Oil, Gas and Metals National Corporation for the permission to publish this paper. Thanks are also due to R.G. Moore, S.A. Mehta, M.G. Ursenbach and all members of the In Situ Combustion Research Group at the University of Calgary for their technical contributions to laboratory studies. The technical advices in numerical simulation provided by Dubert Gutierrez of Computer Modelling Group Ltd. are also acknowledged.

References

1. Juan, E.S., Sanchez, A, del Monte, A, Moore, R.G., Mehta, S.A., and Ursenbach, M.G., 2003, Laboratory Screening for Air Injection-Based IOR in Two Waterflooded Light Oil Reservoirs, Canadian International Petroleum Conference, 2003-215

2. Fassihi, M.R., Yannimaras, D.V., and Kumer, V.K, 1997, Estimation of Recovery Factor in Light-Oil Air Injection Projects, SPERE August 1997, pp173-178.

3. Kumer, V.K., Fassihi, M.R., and Yannimaras, D.V., 1995, Case History And Appraisal of The Medicine Pole Hills Unit Air-Injection Project, SPERE August 1995, pp198-202.

4. Clara, C., Zelenko, V, Schirmer, P, and Wolter, T., 1998, Appraisal of The Horse Creek Air Injection Project Performance, presented at the 8th Abu Dhabi International Petroleum Conference & Exhibition held in Abu Dhabi, United Arab Emirates, 11-14 October, p573-583, SPE49519

5. Gillham, T.H., Cerveny, B.W., Fornea, M.A., and Bassiouni, Z, 1998, Low Cost IOR: An Update on The W. Hackberry Air Injection Project, presented at the 1998 SPE/DOE Improved Oil Recovery Symposium held in Tulsa, Oklahoma, 19-22 April 1998, p401- 406, SPE39462

6. Clara C., Durandeau M., Quenault G., and Nguyen T.H, 2000, Laboratory Studies for Light-Oil Air Injection Projects: Potential Application in Handil Field, SPE Asia Pacific Oil and Gas Conference and Exhibition held in Jakarta, Indonesia, 20–22 April 1999., SPE 54377.

7. D. V. Yannimaras and D. L. Tiffin, 1994, Screening of Oils for In-situ Combustion at Reservoir Conditions via Accelerating Rate Calorimetry, presented at the SPE/DOE Ninth Symposium on Improved Oil Recovery held in Tulsa Oklahoma U.S.A. 17-20 April 1994.

8. O. S. Shokoya, S. A. Metha, R. G. Moore, B. B. Maini, and M. Pooladi-Darvish, 2001, Does Miscibility of In Situ Generated Flue Gases With Light Crude Oils Contribute to Oil Recovery Under High Pressure Air Injection?, presented at the Petroleum Society’s Canadian international Petroleum Conference 2001, Calgary, Alberta, Canada, June 12-14, 2001.

9. Moore R.G., Mehta S.A., and Ursenbach M.G., 2002, A Guide to High Pressure Air Injection (HAPI) Based Oil Recovery, presented at SPE/DOE Improved Oil Recovery


- 7 -

Symposium in Tulsa, Oklahoma, 13-17 April 2002, SPE 75207


- 8 -

Figure 3 Oxygen concentration in the effluent

gas at the CT test @So=40% Figure 2 Oil and water production at the

CT test @So=40%

Figure 1 Results of the CT test (So=40%(Left) & 17%(Right))


- 9 -

Cum Oil & Water Cut

Time (day)

Cu

mu

lati

ve O

il S

C (

m3)

Wat

er C

ut

SC

0 10 20 30 400.0

5.0

10.0

15.0

20.0

25.0

0.0

0.2

0.4

0.6

0.8

1.0

Cum Oil: FM1 (6comp)Cum Oil: FM2 (4comp)Cum Oil: FM3 (3comp)Cum Oil: FM4 (3comp)Cum Oil: FM5 (2comp)WC: FM1WC: FM2WC: FM3WC: FM4WC: FM5

Cum Oil & Water Cut

Time (day)

Cu

mu

lati

ve O

il S

C (

m3)

Wat

er C

ut

SC

0 100 200 3000.0

5.0

10.0

15.0

20.0

25.0

0.0

0.2

0.4

0.6

0.8

1.0


Cum Oil & Water Cut

Time (day)

Cu

mu

lati

ve O

il S

C (

m3)

Wat

er C

ut

SC

0 500 1,000 1,500 2,000 2,5000.0

5.0

10.0

15.0

20.0

25.0

0.0

0.2

0.4

0.6

0.8

1.0


Cum Oil & Water Cut

Time (day)

Cu

mu

lati

ve O

il S

C (

m3)

Wat

er C

ut

SC

0 10,000 20,000 30,0000.0

5.0

10.0

15.0

20.0

25.0

0.0

0.2

0.4

0.6

0.8

1.0


Figure 4 Calculated position of 175 deg.C front at the CT test @So=40%

Figure 5 Schematic of three-dimensional reservoir model

Figure 6 Calculated oil production and water cut at the air flux of 100 Sm3/m2/hr

Figure 7 Calculated oil production and water cut at the air flux of 10 Sm3/m2/hr

Figure 8 Calculated oil production and water cut at the air flux of 1.0 Sm3/m2/hr

Figure 9 Calculated oil production and water cut at the air flux of 0.1 Sm3/m2/hr


- 10 -

Figure 10 Calculated oil recoveries at three dimensional reservoir model

Figure 11 Compositional distribution at the one –dimensional model at air injection rate of 1.0 Sm3/m2/hr

Figure 12 Compositional distribution at the one –dimensional model at air injection rate of 0.1 Sm3/m2/hr

Figure 13 Three-dimensional reservoir model


- 11 -

Figure 14 Sensitivity study at the three-dimensional model

Figure 15 Oil production performance at reservoir dip angle 13 degree, Kv/Kh=0.1 and air injection rate 9MMscf/d


- 12 -

Figure 16 Temperature and oil saturation profiles in the cross-section of three-dimensional model

Figure 17 Effects of residual oil saturation on oil revovery

Figure 18 Oil saturation and temperature profiles at Sor= 13%, 20% and 30%


- 13 -

Table 1 ARC tests of the oil

Ignition

Temp.

Activation

Energy

Sample

Test Type

℃℃℃℃ Kcal/gmol

Oil only Closed 163 68

Oil+Core Closed 158 79

Oil+Core Closed (Isoage)

163 100

Oil+Core Flowing 158 100

Table 2 Fluid model

Fluid Composition (mol%) Fluid

Model

Number of

Pseudo components

Frequency Factor C1 C2-

C3 C4-C6 (Fuel)

C7-C11 (Fuel)

C12-C17 (Fuel)

C18+ (Fuel)

FM I 6 3E+8 35 10 8 24 13 10

FM 2 4 3E+8 35 10 32 23

FM 3 3 3E+8 35 18 47

FM 4 3 3E+8 45 32 23

FM 5 2 5E+8 45 55


- 14 -

Table 3 Parameters of the one-dimensional model

Grid Size 1.0 m X 100

Reservoir Pressure（（（（MPa) 18, 25

Initial Oil Saturation (%) 13.3

Porosity (%) 20

Permeability (md) 10,000

Well Spacing (m) 100

Air Flux (Sm3/m2-hr) 0.1-100

Table 4 Parameters of the three-dimensional quarter model

Porosity 21.5 %

Permeability (Kh) 100 mD

Kv/Kh 1.0

Reservoir Pressure 18 Mpa

Oil Saturation 13.3% (=Sor)

Over/Under-burden

Heatloss

K = 24 Btu/ft-day-degF

C= 35 Btu/ft3-degF

Well Spacing 170 m

Air Injection Rate 3 MMscf/d

number of grid 30 x 30 x 10


- 15 -

Grid Size 4m x 4m x 1 m

Table 5 Parameters of the three-dimensional half model

Porosity 20 %

Permeability (Kh) 200 mD

Kv/Kh 0.1 or 1.0

Reservoir Pressure 18 Mpa

Oil Saturation 13% (=Sor)

Over/Under-burden

Heatloss

K = 24 Btu/ft-day-degF

C= 35 Btu/ft3-degF

Well Spacing 240 m

Air Injection Rate 9 MMscf/d or 1MMscf/d

Reservoir dip 13 degree or 30 degree

number of grid 34 x 67 x 10

Grid Size 4m x 4m x 1 m

evaluation of air injection eor to apply a light oil...

Documents