Impact of Dispersion on Gel Placement for Profile Control

Abstract

Summary A key issue in gel technology is how to place gels in thief zones withoutdamaging oil-productive zones. This study explores the influence of diffusion, dispersion, and viscous fingering during placement of gels to modify injectionprofiles. These phenomena usually will not eliminate the need for zoneisolation during gel placement in unfractured injection wells. During gelplacement in parallel laboratory corefloods, diffusion and dispersion can causeone to conclude erroneously that zone isolation is not needed in fieldapplications. Gel treatments are more likely to improve sweep efficiency inwells where fractures are the source of the channeling problem. Introduction At the peak of activity, 35% of the EOR projects in the U.S. were polymerprojects. About 60% of these polymer projects were gel polymer projects. About60% of these polymer projects were gel treatments rather than traditionalpolymer floods. The objective of gel treatments is to block fractures orwatered-out, high-permeability zones so that subsequently injected fluids aremore likely to enter and to displace oil from other strata. Many gel projectshave been very successful, but unfortunately, others have been technicalfailures. One study revealed that less than 45% of near-wellbore gel treatmentswere successful. The sporadic success rate for gel treatments may be partly aresult of the way the gels were placed in the reservoir. In most cases whengelling agents were injected, zones were not isolated, so the chemicals hadaccess to all open intervals. Much of the gel formulation entered fracturesand/or high-permeability streaks. However, some of this fluid penetrated intostrata that one does not want to plug. Therefore, a key issue in gel technologyis how to place gels in fractures or thief zones without damagingoil-productive zones. Two recent studies examined how injection-flow profiles are modified byunrestricted injection of Newtonian and non-Newtonian gelling agents. Thesestudies found the following.Zone isolation is much more likely to be needed during placement of gelsin unfractured wells than in fractured wells.Productive zones in unfractured wells can be seriously damaged if zonesare not isolated during gel placement.If zones are not isolated during gel placement, the minimum penetrationinto unfractured, low-permeability zones can be achieved by use of a water-likegelling agent (having a resistance factor of unity).Compared with water-like gelling agents, the non-Newtonian rheology ofexisting polymeric gelling agents will not reduce the need for zone isolationduring gel placement. This study explores the influence of diffusion, dispersion, and viscousinstabilities during placement of gels to modify injection profiles. Inparticular, these phenomena are examined to determine profiles. In particular, these phenomena are examined to determine whether they can be exploited tooptimize gel placement. Several terms should he defined for the reader's benefit. The terms" gelant" and "gelling agent" refer to the liquid formulationbefore gelation. Resistance factor, F, is defined as water mobility divided bygelant mobility, and is equivalent to the effective viscosity of the gelant inporous media relative to that of water. Residual resistance factor, F, isdefined as water mobility in the absence of gel divided by water mobility inthe presence of gel. Residual resistance factor is a measure of thepermeability reduction caused by gel. Gelant Penetration In Oil-Productive Strata A common misconception in the application of gel treatments is that injectedgelling agents will exclusively enter high-permeability, watered-out channelswithout penetrating to any significant extent into less-permeable, oil-bearingstrata. Straightforward application of the Darcy equation reveals that gellingagents can penetrate to a significant degree into all open intervals. Forexample, if a gelant penetrates 50 ft [15.2 m] radially from an injection wellinto the most-permeable layer of a multilayer reservoir, then the gelant canpropagate at least 5 ft [1.5 m] radially into a zone that is 100 times lesspermeable, as Fig. 1 shows. Fig. 1 plots the depth of penetration (final radiusminus wellbore radius) of gelant into a penetration (final radius minuswellbore radius) of gelant into a less-permeable zone (Layer 2, k2) when thegelant reaches 50 ft [15.2 m] into the most-permeable zone (Layer 1, k). (Thewellbore radius is 0.5 ft [0.15 m], and all layers have the same porosity.)This information is shown for two Newtonian fluids (F =1 and F =100) and twonon-Newtonian fluids. The non-Newtonian fluids included a xanthan solution anda partially hydrolyzed polyacrylamide (HPAM) solution. (Flow properties of thenon-Newtonian polyacrylamide (HPAM) solution. (Flow properties of thenon-Newtonian fluids are described in Ref. 5.) Note that for a givenpermeability ratio, the three viscous fluids penetrate to a greater depth inthe less-permeable layer than does the water-like fluid (F =1). For the calculations represented in Fig. 1, no crossflow occurs betweenlayers. If crossflow can occur between layers or flow paths in a reservoir, viscous gelants will penetrate into low-permeability layers to a greaterextent. In fact, under some circumstances (if the gelant/water mobility ratiois less than the permeability contrast between adjacent layers), the depth ofpenetration of gelant in a low-permeability layer can be almost the same asthat in an adjacent high-permeability layer. Thus, if crossflow can occur, viscous gelants can damage oil-productive zones to a greater extent than theycan if crossflow is not possible. In preparing Fig. 1, diffusion, dispersion, chemical retention, andinaccessible PV effects were neglected. The impact of chemical retention andinaccessible PV on these calculations has been described previously. The roleof diffusion and dispersion is discussed in this paper.