Prediction of Optimal Salinities for Surfactant Formulations Using a Quantitative Structure–Property Relationships Approach

Abstract

Each oil reservoir could be characterized by a set of parameters such as temperature, pressure, oil composition, and brine salinity, etc. In the context of the chemical enhanced oil recovery (EOR), the selection of high performance surfactants is a challenging and time-consuming task since this strongly depends on the reservoir’s conditions. The situation becomes even more complicated if the surfactant formulation is a blend of two or more surfactants. In the present work, we report quantitative structure–property relationships (QSPR) correlating surfactants’ structures and their composition in a mixture with optimal salinity (S opt), corresponding to minimal interfacial tension in the reference brine/surfactants/n-dodecane system, at T = 313 K and P = 0.1 MPa. Particular attention was paid to selected families of surfactants: α-olefin sulfonate (AOS), internal olefin sulfonate (IOS), alkyl ether sulfate (AES), and alkyl glyceryl ether sulfonate (AGES). The models were built and validated on the database containing S opt values for 75 surfactants’ formulations. Molecular structures of amphiphilic molecules were encoded by functional group count descriptors (FGCD), ISIDA substructural molecular fragment (SMF) descriptors, and CODESSA molecular descriptors (CMD). For mixtures, descriptors were calculated as linear combinations of descriptors of individual compounds weighted by their mass fractions in mixtures. Different machine-learning methodssupport vector machine (SVM), partial least-squares (PLS) regression, and random subspace (RS)have been used for the modeling. Both global (on the entire database) and local (on individual families) models have been built. Models display reasonable accuracy (about 0.2 log S opt units) which is comparable with the experimental error of measured S opt. Our results show that the suggested approach can be successfully used to build predictive models for relatively small data sets of mixtures of chemical compounds.

Chemoinformatics at IFP Energies Nouvelles: Applications in the Fields of Energy, Transport, and Environment

“…The used database included S* values for 75 surfactant mixtures belonging to AOS, IOS, AES and AGES families . Three types of descriptors (FGCD, SMF, and MD3) were considered in order to extract surfactant molecular features, and three machine learning methods (SVM, PLS, and RS) were applied to derive models.…”

“…Models were trained on the entire database and on two subsets containing families of structurally similar surfactants, AOS/IOS and AES/AGES; this led to the development of 27 models. From comparisons drawn, SVM−SMF trained on the entire database was the most interesting combination, both from the accuracy of its predictions and its applicability domain . Using the SVM−SMF model, 64 % of surfactant mixture S* values are predicted within 10 % deviation with respect to reference experimental values; and 91 % are predicted with less than 20 % error.…”

“…[50] In the frame of research performed by "The EOR Alliance", [83] we proposed the development of QSPRs to predict the optimal salinity for some surfactant families, and a temperature of 40 8C, a brine (H 2 O + NaCl), and an oil (n-dodecane). [82], [84] In the objective to speed up the ASP formulation procedure, we also proposed models based on thermodynamics. [81,85] The used database included S* values for 75 surfactant mixtures belonging to AOS, IOS, AES and AGES families.…”

“…[81,85] The used database included S* values for 75 surfactant mixtures belonging to AOS, IOS, AES and AGES families. [84] Three types of descriptors (FGCD, SMF, and MD3) were considered in order to extract surfactant molecular features, and three machine learning methods (SVM, PLS, and RS) were applied to derive models. Models were trained on the entire database and on two subsets containing families of structurally similar surfactants, AOS/IOS and AES/AGES; this led to the development of 27 models.…”

Equivalent Alkane Carbon Number of Live Crude Oil: A Predictive Model Based on Thermodynamics

“…While advanced methods have been recently proposed to speed up ASP formulation procedure [4,5], the HydrophilicLipophilic Deviation (HLD) concept, as proposed by Salager et al [6], is still the keystone of numerous studies [1,7]. At HLD = 0, the Salager relation (Eq.…”