Investigation of Unexpectedly Low Field-Observed Fluid Mobilities During Some CO2 Tertiary Floods
- P.D. Patel (Shell Development Co.) | P.G. Christman (Shell Development Co.) | J.W. Gardner (Shell Development Co.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- November 1987
- Document Type
- Journal Paper
- 507 - 513
- 1987. Society of Petroleum Engineers
- 4.3.4 Scale, 5.8.7 Carbonate Reservoir, 4.1.9 Tanks and storage systems, 5.2 Reservoir Fluid Dynamics, 4.3.3 Aspaltenes, 5.4.2 Gas Injection Methods, 5.3.2 Multiphase Flow, 5.4.7 Chemical Flooding Methods (e.g., Polymer, Solvent, Nitrogen, Immiscible CO2, Surfactant, Vapex), 4.6 Natural Gas, 5.2.1 Phase Behavior and PVT Measurements, 5.4 Enhanced Recovery, 1.8 Formation Damage, 5.4.1 Waterflooding, 1.6.9 Coring, Fishing
- 0 in the last 30 days
- 411 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Summary. Phase behavior, inorganic precipitation, and wettability are investigated as possible reasons for the unexpectedly low field-observed mobilities during some CO2 floods, in particular, the Denver Unit, Wasson field CO2 pilot. The observed mobility was not a near-wellbore effect and probably played a major role in reservoir sweep: the low effective probably played a major role in reservoir sweep: the low effective permeability offset the detrimentally low CO2 viscosity. Experimental and permeability offset the detrimentally low CO2 viscosity. Experimental and simulation studies, supplemented by literature data, lead to the conclusion that rock wettability could be the root cause of these low fluid mobilities. Phase behavior effects, though they may play a role, are not necessary to explain the injectivity behavior, and inorganic precipitates probably have little effect under the conditions investigated here. Thus, probably have little effect under the conditions investigated here. Thus, current simulator modeling of low fluid mobilities, which are based on arbitrary permeability reduction factors allegedly caused by phase behavior, appears unjustifiable even though overall simulator results may be acceptable.
Under Denver Unit, Wasson field conditions (2,000 psi and 105 deg. F [13 790 kPa and 40.5 deg. C]), CO2 is about 10 times less viscous than brine. Thus, during a field test of CO2 flooding, injectivity might be expected to increase sharply when CO2 injection follows a brine preflood. Instead, in the pilot, injectivity dropped abruptly upon preflood. Instead, in the pilot, injectivity dropped abruptly upon CO2 injection and remained less than that of preflood brine during most of the CO2 injection period. Fig. 1 illustrates this behavior. This paper addresses reasons for such an injectivity trend.
A literature survey shows that declines in injectivity during field solvent floods are not uncommon. Two other west Texas tertiary pilots, at Slaughter and Levelland, exhibited some reduced pilots, at Slaughter and Levelland, exhibited some reduced injectivity during the CO2 injection phase. Rich- and lean-gas injection, alternated with water, produced reduced solvent injectivities in Amoco's Project "A" Exxon's Seeligson, and Sonatrach's Hassi Messaoud projects. Table 1 summarizes the characteristics of fields exhibiting reduced solvent injectivities. No single charac-teristic stands out as the cause for such injectivities.imulator modeling of these low fluid mobilities has used arbitrary permeability reduction factors attributed mainly to pore-plugging "asphaltic" precipitation on CO2/crude interaction. In addition to this phase-behavior effect, inorganic precipitates resulting from rock/CO2/brine interaction and/or wettability considerations are candidate causes of low solvent injectivities (Fig. 2). Here we investigate the importance of each mechanism.
Effect of Phase Behavior
Literature Data. The interaction of CO2 with crude oil may result in asphaltic precipitation and formation of multiple hydrocarbon phases. Both situations could lead to lower total fluid mobility during reservoir floods. Instances of asphaltic precipitation have been noted by numerous authors. The effect such precipitation has on flow through porous media, however, has received less attention because of, no doubt, the experimental difficulties involved. Patton et al. performed sandpack displacements using CO2 and Patton et al. performed sandpack displacements using CO2 and a crude oil for which organic precipitates were observed in visual tests. Substantial permeability damage did not result. Monger reported unusual injectivity reduction and eventual core plugging during a CO2 displacement of highly asphaltic crude in Berea rock. A second experiment done at slightly higher pressure, however, apparently did not plug up. Campbell and Orr described CO2 flooding experiments in etched glass plates that approximate flow in reservoir rock. In displacements of the same crude with which Patton et al. reported substantial organic precipitation on Patton et al. reported substantial organic precipitation on contact with CO2, Campbell and Orr saw "no evidence of any restriction to flow caused by the precipitate."
Besides the asphaltic phase sometimes observed, multiple hydrocarbon liquid/vapor phases always occur during CO2 flooding of oil reservoirs. Shelton and Yarborough speculated that such multiphase flow, perhaps in conjunction with asphaltic precipitates, can "significantly affect the injectivity of a field precipitates, can "significantly affect the injectivity of a field project." Henry and Metcalfe monitored the pressure across cores in project." Henry and Metcalfe monitored the pressure across cores in which the effluent hydrocarbon multiphases of a CO2/oil sandpack displacement flowed. A slight increase in pressure drop was reported when three, rather than two, hydrocarbon phases flowed in the core.
In summary, there has been no clear experimental evidence that phase-behavior effects (both multiphase flow and organic precipitation) phase-behavior effects (both multiphase flow and organic precipitation) result in the field-observed low fluid mobilities previously mentioned. On the other hand, there is no clear evidence to discount these causes. Systematic studies of the effects of asphaltic precipitation on flow behavior are planned by some precipitation on flow behavior are planned by some investigators. A coupling of flooding experiments with models predicting such precipitation may be necessary. predicting such precipitation may be necessary. Experiments. Realizing the difficulties of directly assessing the impact of phase behavior on fluid mobilities during CO2 floods, we wanted to determine whether phase behavior is a necessary reason for low fluid mobilities during some CO2 floods.
Fig. 3 shows injectivity results during CO2 displacements of waterflood residual oil in 9-in. [23-cm] -long Denver Unit, Wasson field cores. The coreflooding flow system has been described by Lease and Van Egmond. Pressure was monitored across the entire core in these experiments. Injectivity was normalized with that at the conclusion of the waterflood:
where q is the volumetric flow rate and delta p is the pressure drop. Thus, normalized injectivity at CO2 flooding inception was one. The movable PV for an experiment was simply the total PV minus the nearly constant irreducible brine volume left after many PV's of CO2 flooding.
|File Size||683 KB||Number of Pages||7|