Evaluation of Gas Saturation During Water Imbibition Experiments

ABSTRACT

Knowledge of gas saturation history is very important in determining gas recovery from gas reservoirs with water influx. Water imbibition is known to control gas recovery. Spontaneous imbibition experiments have been traditionally employed to determine gas saturation.

On-line NMR relaxometry is introduced as a method for monitoring co-current imbibition. A group of plugs from a Western Canadian sandstone reservoir were selected and a series of imbibition tests were run. NMR was used to measure the amount of water imbibed in the cores and the gas saturation during each experiment was, in turn, measured. The values of final residual gas saturation and the production profiles were compared to the results from corresponding counter-current imbibition tests. The correlations of residual gas saturation with initial imbibition rate and other operating parameters were investigated. Through interpretation of the NMR spectra, bound water T2cutoff values were obtained. Water distribution in different pore sizes during the experiments was also calculated. The initial imbibition rate in different pore sizes was measured.

Empirical equations from the literature, which were used to describe the behaviour of co-current imbibition tests, were applied to the experimental data. The proposed methodology can be used to evaluate the mechanisms of water imbibition in gas reservoirs.

Introduction
The understanding of the mechanisms that govern spontaneous water imbibition in gas-water systems is important to the development of natural gas reservoirs. Many researchers indicate that residual gas saturation is affected by factors such as wettability, imbibition rate, initial water saturation, and also the experiment methods (1-3). In this work, a group of sandstone plugs underwent co-current imbibition tests. The fluid saturations were measured using Nuclear Magnetic Resonance (NMR) relaxometry. The results were compared with those coming from counter-current imbibition tests from previous work (4, 5). Some results from the literature ( 6, 7) were also tested against our experimental results. In 1960, Handy (6) developed an equation to describe the behaviour of co-current imbibition tests. He postulated that the weight or volume of imbibed water is proportional to the square root of imbibition time:

Equation (Available In Full Paper)

In this equation, A and N wt are the cross-section area of the core and the volume of water imbibed into the core, respectively. S wf is the water saturation behind the imbibition front. K w and P c are the effective water permeability and capillary pressure at S wf, respectively. Finally, Φ?is the porosity and µ w is the viscosity of the imbibing brine.

Li and Horne (7) point out there are some disadvantages to Handy’s equation (6) since effective water saturation behind the front and capillary pressure cannot be calculated separately, and the relationship between the square of weight gain and time is not a straight line during the later period of water imbibition. They developed an equation to describe the relationship between the imbibition rate and gas recovery based on the assumption of a piston-like imbibition flow. In their equation, the imbibition rate should be in a linear relation with the reciprocal of gas recovery.

A full version of this paper is available on OnePetro Online.