Transverse Dispersion in Slug-Mode Chemical EOR Processes in Stratified Porous Media
- Michael R. Wheat (Imperial C.) | Richard A. Dawe (Imperial C.)
- Document ID
- Society of Petroleum Engineers
- SPE Reservoir Engineering
- Publication Date
- May 1988
- Document Type
- Journal Paper
- 466 - 478
- 1988. Society of Petroleum Engineers
- 5.3.2 Multiphase Flow, 4.1.2 Separation and Treating, 4.1.5 Processing Equipment, 5.6.5 Tracers, 1.6.9 Coring, Fishing, 4.3.4 Scale
- 0 in the last 30 days
- 152 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
Summary- Transverse dispersion in multilayered media of moderate permeability contrast has been examined experimentally and theoretically. permeability contrast has been examined experimentally and theoretically. Continuous and slug-mode displacements at unit and nonunit mobility ratios have been conducted in a layered beadpack model with a permeability contrast of approximately 3: 1. Layer-width and residence-time effects have been quantified and scaled to core and reservoir dimensions. Transverse dispersion can be significant for reservoirs with layers less than 100 cm [40 in.] thick and core material with layers less than 1 cm [0.4 in.] thick. The data can be used for validating simulation packages written for chemical dispersion.
All reservoirs and the majority of core material are heterogeneous, with stratification (layering) being the most common feature. In EOR chemical processes, the chemicals are often injected as a fraction of the PV (a slug of fluid) followed by chase fluid. As the slug progresses through the porous matrix, it will move faster through the more permeable portions, and transverse dispersion-movement of chemical from areas of high to low concentration (or more correctly, chemical potential) normal to the flow-will occur across the concentration discontinuities created. This dispersion will dilute the concentration of chemical in the slug and may reduce its ability to mobilize residual oil.
As flow progresses farther, the layering can cause the slug integrity to be totally lost (Fig. 1). This slug breakdown accelerates the chemical dilution process because of the large areas now available for transverse dispersion. If the chemical concentration is reduced to below the critical value needed to ensure that oil remobilization occurs, the effectiveness of the process will be destroyed. (This value is obtained from slim-tube or long-core tests.)
In multiwell tracer tests, which have been proposed to determine interwell matrix properties before the application of EOR processes, transverse dispersion will cause a redistribution of fluid processes, transverse dispersion will cause a redistribution of fluid components between the layers. The interpretation of these tests has been limited by a lack of both an understanding of the mechanisms involved and a mathematical model with which to deconvolve the resultant effluent profiles. This work will help clarify these mechanisms and thus, it is hoped, aid in the development of an effective mathematical model.
The effects of viscous crossflow in layered media on continuousand slug-mode processes in the absence of dispersion have been discussed previously. This paper examines the effect transverse dispersion has on linear, miscible slug-mode displacements in multilayered media. Each layer in this work is essentially homogeneous and isotropic, with the heterogeneity of the system a result of the differing permeability of the layers. The unit-mobility-ratio case is considered first both experimentally and theoretically, and quantitative conclusions pertinent to field and long-core test experimes are made. The major interest has been in reservoir layers 15 to 100 cm [6 to 40 in.] thick and chemical residence times of 1 to 5 years and in core material with layers up to I cm [0.4 in.] thick. This range of layer widths and time scales will cause the most disruptive dispersive mixing. The study therefore extends the work on slug displacement by Koonce and Blackwell to include long-time effects shown to be important in continuous displacements by Lake and Hirasaki and Marle et al. The complications introduced by nonunit mobility ratios are then discussed and illustrated by experimental results.
Slug-Mode Displacements. Fig. 1 shows a slug-mode process at various stages of displacement when there is no fluid mixing. The system considered contains two layers, with the conductance (per-meability/porosity) of Layer A greater than that of Layer B. A consequence of this conductance contrast, more slug chemical is injected into Layer A than Layer B (Fig. la), and as the displacement progresses, the slug portion in Layer A will move faster than that in Layer B (Fig. 1b-d).
If the slug fluid contains chemicals with concentrations not equal to those in the resident or chase fluid, then mass transfer wig occur across each of the exposed fluid boundaries shown in Fig. Id as a result of microscopic dispersion (a combination of molecular diffusion and microscopic convective dispersion). Microscopic dispersion perpendicular to the direction of flow (transverse dispersion) will have a greater effect on the chemical distribution than that parallel to the direction of flow (longitudinal or axial dispersion) parallel to the direction of flow (longitudinal or axial dispersion) because of the respective areas over which they act. Clearly, the maximum dispersion effect will occur when the two slug portions have separated and no longer support each other laterally (Fig. 1d). Here, the maximum interfacial area for mass transfer by transverse dispersion is exposed. This dispersion may reduce the chemical concentration to below that required to remobilize residual oil. This aspect will be discussed later.
One method to prevent the process from being adversely affected by transverse dispersion is to design the slug volume, Vs, so that the two slug portions support each other for the duration of the displacement. The minimum slug volume required to achieve this is obtained when the overtaking point, X,,, (defined as the distance when the leading front in Layer B, XBI, has traveled the same distance as the trailing front in Layer A, XAI, Fig. 1c), occurs at the end of the system (i.e., X., = 1.0). This slug volume can be shown to be
where Ab is the fraction of the system width that is high-conductance media and CA and CB are the layer conductances. Fig. 2 shows the value of V, required for a range of conductance contrasts and layer widths. Clearly, even for modest conductance contrasts of about 10, large volumes of slug chemical would be needed. Therefore, for practical slug-mode EOR processes, the effects of transverse dispersion must be considered. It should he emphasized, however, that the process will not become ineffective at the overtaking point, but rather that this is the point where the portions of the slug become separated and the maximum area for transverse dispersion appears.
|File Size||954 KB||Number of Pages||13|