Abstract

In this paper, a semi-empirical method for evaluating the injectivity decline in a produced water reinjection operation is proposed. The characteristics of injectivity decline during the pumping of produced water are represented as two-stage behavior - an early exponential decay and a late linear declining injectivity. The analytical modeling of coupled flow-transport phenomena for a two-layered medium reveals that the particle deposition during produced water re-injection constitutes a critical component of two stage decay-fading responses. The method was validated through matches between the modeling results and field measurements. The investigated concepts and developed technologies detailed in this paper can be applied directly to the produced water re-injection operations in order to predict possible injectivity decline and assist in understanding the mechanisms behind the decline phenomenon.

Introduction

Produced water can come from several sources:

  • a) formation water,

  • b) injected water for waterflooding and

  • c) condensed water from gas production, etc.

Water cuts can be quite high (some figures suggest a worldwide average of up to 75%)1. Traces of oil (OIW) and solids (TSS) may stay with the produced water after separation. Solids can also be accumulated from turbulars during injection. The composition of produced water is highly variable. In some cases, produced water may be treated to remove some of OIW and TSS before injection. Produced water disposal by re-injection is regulated and controlled by local agencies. Sometimes, produced water is used for pressure maintenance.

Even with some treatment (and treatment may not be economically or technically desirable), produced water can contain insensible amounts of residual hydrocarbons, heavy metals, radionuclides, inorganic species, suspended solids and chemicals. These can result in injectivity decline (near-well) and formation damage (anywhere) even if injection is above fracturing pressure. The present study examines permeability reduction and injectivity impairment by particulates formation or hydraulic fracture plugging.

Maintaining injectivity becomes problematic when solid particles and entrained oil either reside in porosity or plug pore throats, particularly if injection protocols do not allow injection above fracturing pressure. Matrix injection is often susceptible to rapid injectivity degradation. For any level of reliability of matrix injection, a forecast of injectivity decline is needed in order to identify methods for retaining acceptable rates. There are some predictive methods. Among them, WID (Water Injectivity Decline 2), jointly developed by the University of Texas at Austin and the Norwegian University of Science and Technology (NUST), is based on analytical evaluations of changes in injection rate using relationships between particle deposition, permeability reduction, and normal flow resistance 2.,3. Other approaches associate the injectivity decline with the permeability reduction due to the building up near-well filter cakes 2,4–8.

A simple alternative for predicting injectivity decline is introduced in this paper. The method is an empirical one, but appears to be supported by related theoretical studies, by physical intuition and field phenomena. When critical parameters are sensibly defined, the method can approximate injectivity if previously measured data are not available.

This content is only available via PDF.
You can access this article if you purchase or spend a download.