Re-injection of produced water is of increasing importance as water cuts continue to increase worldwide. It provides an environmentally acceptable solution to the disposal of produced water, and contributes to pressure maintenance when injection takes place into the reservoir itself. Injection can take place under matrix injection or fracturing conditions. In both cases, the performance of the injection well and the distribution of the injected water are strongly influenced by the build-up of formation impairment around the wellbore or the fracture face. Solid particles and small oil droplets dispersed in the injection water are deposited in the formation by a process of filtration, and therefore will cause this impairment.
As part of a study on formation damage, different concentrations of oil droplets and hematite particles suspended in water were injected into sandstone core samples. A particle analyzer was used to measures the inlet and outlet concentrations of each particle type, concentration and size. The analysis confirms the high impact of oil concentration with the presence of solid particle on permeability damage. Preparation of oily-water emulsion is very challenging to achieve without chemical agents or mechanical devices. The presence of solid particles present in oily-water emulsion act as emulsion stability agents. Hematite particles being oil-wet, improved emulsion stability. Permeability impairment due to oily-water alone at residual oil formation is insignificant, and the size of oil droplets does not play a big role in injecivtivity decline. However, oily-water containing solid particles causes severe permeability damage. Furthermore, if the deposition of solids increases to a critical concentration, it could create a high differential pressure mobilizing some of the trapped residual oil.
Production of hydrocarbons is usually accomplished by the production of aquifer water. This produced water consists of formation water and/or water that has previously been injected into the formation. As more oil is produced, the amount of produced water increases. Consequently, this requires water management. One technique is reinjection for disposal, pressure maintenance, or enhanced oil recovery. An important and difficult task in the reinjection process is the ability to predict the impact of water quality on well injectivity. This is mainly due to the poor knowledge of deposition mechanism in the formation of suspended solids and oil droplets in the produced water.
A successful injection project can be executed if two key questions are answered: is the water going where it is supposed to; and can the desired injection rate be met? There are several factors that can influence produced water reinjection (PWRI) performance. Most studies indicate that water quality is a very important parameter for reservoir impairment processes, such as fine migration of the same formation during the injection operation (Civan 1989, 1991, Ohen 1991, 1993, 1996), clay swelling (Barkman 1975), and biological effects. Recent researchers have shown that completion fluids fines, an order of magnitude smaller than a given pore size, can also cause considerable damage (Gruesbeck 1982, Wojtanawicz 1987, Bennion 1994, 1995). In addition, larger quantities of suspended oil may impair injectivity by reducing the permeability of the formation to water, especially in injection zones that initially have less than the mobile oil saturation. Oil tends to foul deep-bed filtration and enormous effort is made to remove the majority of the oil content before reinjection operations.
Produced water containing oil in the produced water in the range of 500mg/l to 5,000mg/l or higher must be treated before reinjection of water for two reasons. First of all, the oil in the injection water may cause damage to the formation. Hence, the oil content of the injection fluid must be reduced to a suitable level for use. Secondary, the oil that is recovered from the produced water is routed to the oil sales meter to generate cash for the operation (Thro 1994, Van der Zande 2000, Janssen 2000).