In-sItu CombustIon PIlot basIC DesIgn anD laboratory exPerIments

When it must be decided to develop a field with an enhanced oil recovery method, first it is needed to have a reservoir characterization model of high quality. Then the choice of the best suited method has to be carried out. For any method, a preliminary study has to be performed in order to help to decide. In the case of an in-situ combustion field development, various patterns are considered; at the same time, duration for the combustion front to move from the injector to a producer is analyzed. Field examples of various patterns are presented. The amount of air to inject in case of dry combustion and of air and water in case of wet combustion has to be determined in order to design air compressors and water pumps. The amount of air is a function of the volumetric sweep efficiency and of the oil and the matrix from the reservoir. Lab experiments must be performed in the reservoir matrix with the reservoir oil to determine the air requirement, which is the amount of air needed to burn a unit volume of reservoir. The amount of water is also determined by lab tests. Then the flows of air and of water are determined, which allows the design of compressors and pumps. The amount of oil produced is calculated taking into account the sweep efficiency in the different zones in front of the combustion. Production of oil, water and gas and their compositions obtained at the lab scale are presented. A scheme of the production, treatment and storage for a pilot field test is shown. In conclusion, a diagram shows the general guidelines for the preparation and implementation of field experiments using in-situ combustion.


Introduction
Due to the increase in the world population and to the increase in its welfare, more and more energy is needed.Renewable energies are not sufficient to answer the demand and fossil energies will remain for a long time the main resources.New fields are more and more difficult to find and more and more expensive to develop.The mean recovery factor of the developed fields is around 35 % of the oil in place.New methods -Enhanced Oil Recovery (EOR) -are the solution for increasing this recovery factor.
Whatever the method, the first step to produce a reservoir is to perform its characterization in order to know the quantities of hydrocarbons in place and the complexity of the structure for the flow of fluids (Fig. 1) (Burger et al., 1985).
Thermal methods are the best EOR methods for heavy, extra-heavy oils and bitumen.In steam injection, heat is produced on the surface and injected into the reservoir through the vaporized water.In in-situ combustion, the heat is produced in the reservoir itself, meaning better energy efficiency than in the case of steam.
The principle of in-situ combustion is to burn with air or oxygen, some of the oil in place.The heat generated by the combustion is used to decrease the viscosity of the remaining oil and hence facilitates its production.Combustion is started by an artificial mean or by self-ignition, increasing the temperature up to a value where a combustion front is initiated and moves from the injection well to production wells.Only an oxidizing gas -dry combustion -is injected or air and water -wet combustion -are injected simultaneously or alternately.The aim of the water injection is to recover the heat stored in the reservoir behind the combustion front.In both cases, there is formation of a coke-like material, which will act as  the fuel for the combustion.This fuel is produced by cracking and pyrolysis of the oil in front of the combustion front.The amount of fuel is lower in the case of wet combustion due to the heat transfer from behind the front, allowing a better sweep efficiency in front of the combustion front.
Before implementing a field test for in-situ combustion, it is needed to check if the reservoir fulfills the screening criteria already known, to define the pattern on which ISC will be tested, to estimate the performances which can be reached, to determine through lab tests the main parameters governing the ISC process and to define the main equipment which will be used for performing the pilot test.Utilization of thermal numerical models like Stars or Eclipse will be very useful for optimization of the parameters involved in the combustion process, but analytical models like the one from Nelson and McNiel (Nelson, McNiel, 1961) give a better understanding of the ISC.

screening criteria
Screening criteria (Gadelle, Clause, 2013) have been developed to select the best method of EOR for a given reservoir (Fig. 2).For an in-situ combustion field experiment, it is recommended that the reservoir depth is enough high for not having an overburden fracture when injecting air, the maximum depth is limited by the economics of the process (cost of air injection at high pressure).Sand or sandstone are better matrix than carbonates which being very often fissured or fractured, are not recommended for a gas injection such as air injection.Viscosity or specific gravity must be high.

Pattern size
Different patterns (Renard, 2011) with varying spacing between wells and numbers of wells (5, 7, 9 spots) can be used at the pilot level (Fig. 3).The case of the THAI process (combustion with horizontal and vertical wells is not considered here).The inverted 7-spot is the most regular with the same distance between injector and producers; the inverted 5-spot is also very regular but having fewer wells, its sweep efficiency will be lower.Generally the field is then developed with patterns identical to the one used for the pilot; in case of a dipping reservoir, line drive is preferred to usual patterns (Fig. 3).In the Suplacu de Barcau field in Romania (Gadelle et al., 1981), the combustion was ignited in a 5-spot pattern, which, after the good results obtained, was extended to a 9-spot pattern.Then the combustion was pursued in several patterns.
After some years, instead of continuing to produce the field with patterns, it was decided to start a line drive for the combustion in order to benefit from gravity displacement due to the dip angle of the reservoir.The pattern size in case of 5-spot is described on Fig. 4. The rate of the combustion front is generally comprised between 5 and 10 cm/day; here a mean value of 7.5 cm/day for the rate of the combustion front is used; with such a value, the duration to sweep 1 ha (100 m*100 m) is 2.6 years.In the case of the Chichimene field in Colombia (Fig. 5), the first design was a distance between the injector and the producers of 450 m; duration for the front to arrive to a producer was 12 to 25 years with rate of the front of 10 and 5 cm/day.This delay was too long and it was decided to modify the pattern in order to have only 110 m between the injector and a producer, meaning a time of 3 to 6 years with the previous rates.

amount of air and water to be injected amount of air
The amount of air to be injected to sweep the pattern is depending of the size of the pattern and of the thickness of the layer, of the areal and vertical sweep efficiencies and of the air requirement to burn a unit volume of reservoir which is the ratio between the air flux and the combustion front rate.

Sweep efficiency
The volumetric sweep efficiency by the air injected depends on the pattern used for the pilot test.For an inverted 5-spot, the areal sweep efficiency lies around 65 %.The vertical air sweep efficiency is depending on the thickness of the layer; due to segregation effects, it decreases with the

air requirement
The air requirement which is the ratio of the air flux to the rate of the combustion front represents the amount of air needed to burn a unit volume of porous medium.This value does not depend of the flow rate and the pressure; it is a characteristic of the reservoir and its oil.It cannot be extrapolated from other tests performed with different oils or matrix.This is shown in Fig. 6; the 2 oils considered have quite similar properties in terms of sp.gr., viscosity, asphaltene content and the matrix have amount of metals susceptible to act as catalyst for the combustion, of the same order of magnitude.
The air requirement is obtained through laboratory tests performed in combustion tubes.Figure 7 shows the equipment for combustion tube tests manufactured by Xytel Inc. (Gadelle, Clause, 2012).It can work up to pressure of 350 bar and temperature of 800 °C.It consists of a thin inner tube of 10 cm diameter and 200 cm length which can sustain a pressure of 10 bar and an outer shell which withstands very high pressure.Around the inner tube 33 heaters are used to prevent heat losses; temperatures are measured in the axis of the inner tube and at its wall at the level of each heater; wall temperature at each heating collar is maintained around 10 °C lower than the axis temperature in order to be sure to not inject heat.In the annulus between the inner tube and the outer shell, insulating material is placed for decreasing the heat losses, and it is under a nitrogen pressure.Air and water in case of wet combustion are injected through a specific manifold.At the outlet, the gases are measured and analyzed; the oil and water produced are also measured and analyzed.
Figure 8 shows the position of the combustion fronts vs. time in an experiment (Gadelle et al., 1981) started in dry combustion and followed by wet combustion.The slope of the curve represents the rate of the combustion front.Knowing the air flow rate and the diameter of the combustion tube, it is easy to calculate the air flux.From the air flux and the rate of the combustion front, the air requirement is calculated.
As shown in Figure 9, it is needed to carry out several tests to determine the best conditions for propagating the combustion in dry or wet mode.
The experiments carried out in a combustion tube must be realized with the oil and the matrix from the reservoir as mentioned earlier.In many cases the availability of reservoir matrix is difficult and when it is consolidated, this material must be crushed and sieved before using.If there is not enough reservoir matrix, it could be necessary to find a sand and additives (such as clay) having the same properties regarding ISC as the initial matrix.Then the artificial sand will be used for quite all the experiments.Ramped Temperature Oxidation (RTO) can be used to find a substitute to the original matrix.
The equipment for RTO, manufactured by Xytel Inc., is shown in Fig. 10 (Gadelle, Clause, 2012).A sample containing oil and sand (or crushed core material) is heated from room temperature up to 500-600 °C.An oxygen containing gas or nitrogen is circulated through the sample.Thermal effects are observed with thermocouples and the reactions are followed through the oxygen consumption and the CO 2 and CO formation (Fig. 11).Two reactions appear, one at low temperature which is the fixation of oxygen on hydrocarbon molecules (practically no formation of carbon oxides), the second at high temperature where all the oxygen consumed is transformed into carbon oxides.The first reaction is an oxidation reaction whereas the second is a combustion reaction.In some cases a third reaction peak is seen at very high temperature.It is attributed to insoluble organic material linked to the matrix (kerogen).
The second peak, corresponding to a combustion reaction, gives an order of magnitude of the amount of fuel which will be burned during the in-situ combustion.The first peak, corresponding to hydrocarbon oxidation, normally does not exist during in-situ combustion, because the operator wants that all the oxygen is consumed at the front and that there is no residual oxygen in front of the combustion front.The virgin oil is heated by a gas containing no oxygen or a very small amount, and is submitted to pyrolysis and cracking in an inert atmosphere.Results are shown on Figure 11.
This equipment is very useful not only for choosing a substitute to the original matrix or to compare the chemistry of different oils in different matrix, but also to check the influence of catalysts which can be injected during the ISC process.

amount of water to be injected
The combustion tube allows defining the best ratio between the flow rates of air and water in wet combustion.The amount of water to be injected during the operation is then calculated from the amount of air.The water can be injected simultaneously or sequentially with the air.Generally it is preferred to inject alternately the air and the water for minimizing the corrosion effect.

Air and water flow rates
The maximal injected air flow rate is generally reached at the end of the pilot test when the combustion is the largest; this value is depending on the vertical and horizontal efficiencies and on the area of the pattern and thickness of the layer.It is assumed that the rate of the combustion front is at its minimum value which is in the range of 4 cm/day.
The air injection schedule can be to use immediately the maximal flow rate of air or to increase the flow of air step by step until it reaches the maximal air flow rate.The second possibility is probably better because during the combustion displacement, it minimizes the air going through the front and so the possible detrimental low temperature oxidation.
If the water is injected slug by slug, it is necessary to take into account during the calculations the ratio of water/oil and the duration of the slugs of air and water.

air and Water Facilities
From the amount of air and water to be injected and their respective flow rates, it is easy to determine the maximum injection pressure.Then the design of the compressors and pumps can be carried out.Fig. 12 shows such air and water facilities.

Prediction of performances amount of oil
The volume of oil produced during the pilot test can be evaluated by considering the displacement phenomena in the various zones (Fig. 13) (Burger et al., 1985): Zone 1: zone effectively burned from which the oil has been burned or displaced; Zone 2: steam zone downstream the burned zone, where the oil saturation is quite the residual oil saturation after a steam displacement; Zone 3: hot water zone downstream from the steam zone, where the oil saturation is the residual oil saturation after a hot water displacement; Zone 4: hot water zone below the zones 1 and 2 with the same residual oil as in zone 3; Zone 5: zone not affected by the combustion.
From the values of air injected and oil produced, it is possible to calculate the Air/Oil Ratio (AOR), which is a measure of the interest of the project.
Correlation between the volume burned and the volume of oil displaced in a specific field have also been proposed and can also be used to determine the amount of oil recovered (Gates, Ramey, 1980).

amount of water
The water produced is the connate water and the water of combustion which is calculated from the stoechiometric equation of burning the coke material; in case of wet combustion, part of the water injected with air is produced.

amount of gas
The gas produced is the nitrogen injected with the oxygen for the combustion, and the carbon oxides.The amount is not much different of the amount of air injected.When the oil contains some sulphur, hydrogen sulphide will be produced, and safety measures have to be taken.

Production facilities
Figure 14 shows facilities for a pilot test.First there is a gas -liquid separator, followed by separation between oil and water with the use of heater -treater for braking emulsions.All the gas is sent to a stack or today it is treated to eliminate through oxidation all the residual hydrocarbons, carbon monoxide.In case a specific treatment is designed for H 2 S. The water after treatment is disposed of.The oil goes to storage.

technology
Technology is a very important point when implementing a pilot.On this side, it is necessary to define the completion of the injection well and of the production wells taking in account the high temperature encountered during the combustion (high quality stainless steel for tubing and casing, thermal cement, thermal packers).Cooling of the wells by internal tubing is a requisite.All this part must be carried out by specialists of the different domains.
Only one point which is very important has to be considered here: the ignition.Ignition is the first operation to be realized when starting the pilot.A good ignition will generally allow a successful combustion, but a bad ignition means a failure.The different ways for starting the combustion are self ignition or artificial ignition.

self ignition
If the oil in place is sufficiently oxidizable under bottom hole conditions, spontaneous ignition may occur near the

Figure 1 .
Figure 1.Diagram of the Study of Reservoir Data Prior to an EOR Test

Figure 2 .Figure 5 .
Figure 2. Screening Criteria for Enhanced Oil Recovery Process

Figure 6 .Figure 8 .
Figure 6.Influence of Oil Characteristics on Coke Deposit

Figure 13 .
Figure 13.Sweep Efficiencies for In-situ Combustion