SPE Member
Abstract Foams have been proposed for use as mobility control agents and to improve conformance in various EOR processes. Multiple correlation analysis has been used to generate equations relating foam data to the chemical structure of alcohol ethoxylates and alcohol ethoxylate derivatives being considered for use as CO2, "foaming" agents. Important surfactant structure parameters determining foaming performance include molecular weight, mole percent ethylene oxide, and hydrophobe carbon number and chemical structure.
Steam foaming agents have also been studied using multiple correlation analysis. Mobility reduction factors (MRF) observed when an alpha-olefin sulfonate (AOS) and 200 deg. C steam were injected into sand packs could be correlated with AOS concentration and the gas:liquid volume ratio, Branching at the 2-position of AOS surfactants reduced the MRF substantially. This reduction appeared to be related primarily to inductive effects of the side chain.
Other surfactant properties of importance in enhanced oil recovery include critical micelle concentration, cloud point, interfacial tension, and adsorption. These properties have been related to surfactant structure using multiple correlation analysis.
Introduction Various surfactants are under investigation in our laboratory for use as mobility control agents in steam and CO, enhanced oil recovery processes. Literature reports have described the laboratory evaluation of large numbers of surfactants for these applications. Many of the tested surfactants were close chemical structure analogs of each other. If a means could be found to correlate surfactant chemical structure with performance parameters such as mobility reduction factor, foam stability, critical micelle concentration, and cloud point, surfactant performance could be predicted. The objective of the present work was to use the technique of multiple correlation analysis (see Appendix 1) to do this. In addition, the terms of the correlation equation define the surfactant chemical structure parameters that determine the magnitude of the performance property being studied. The value of the coefficients of the various terms of the correlation equation establish the relative importance of each term in determining the magnitude of the performance parameter.
Sometimes chemical structure parameters alone are insufficient to construct a correlation equation having an acceptable (is greater than 0.90) correlation coefficient, r. This situation is particularly likely to occur when studying commercial surfactants that are actually blends of several components. In these cases, inclusion of terms representing easily measured surfactant physical properties such as surface tension and cloud point can be used as independent variables in the correlation equation. In addition, the nature of the performance tests, one atmosphere Ross-Miles foam tests in particular, may be such that chemical structure parameters alone are insufficient to generate correlation equations having r is greater than 0.90.
This study was restricted to linear correlation equations. While non-linear equations that have higher correlation coefficients and greater quantitative predictive power could be generated, theoretical insights from such non-linear equations are often obscured by the complexity of the various terms.
The results of one atmosphere foaming experiments of the types used here are subject to may test design variables. Bikerman has emphasized that foam heights or volumes "are specific for the shaking procedure selected and possess no general validity."
P. 395^