Quantitative assessment of surface elemental and mineralogical effects on rock wettability: evidence from several rock types

Predicting rock wettability remains a fundamental challenge in petroleum reservoir characterization, directly impacting enhanced oil recovery strategies and remaining oil-in-place estimates. While mineralogy and surface chemistry are recognized as key controlling factors, their relative contributions and the mechanisms governing their influence remain poorly quantified, particularly in heterogeneous multi-mineral systems. This study systematically investigates the relationships between rock compositions, surface elemental chemistries, and wetting behaviors across diverse lithologies to address this knowledge gap. Using a combination of micro-X-ray fluorescence (µ-XRF) spectroscopy, X-ray diffraction (XRD) analysis, and contact angle measurements on 126 rock samples representing seven major lithological units, multivariate statistical analysis was conducted to quantify and compare the predictive power of elemental composition versus bulk mineralogy. The analysis reveals highly significant associations between elemental compositions and wetting tendencies: increased water-wetting behavior correlates with enrichments in silicon (Si), potassium (K), calcium (Ca), titanium (Ti), cobalt (Co), rubidium (Rb), and phosphorus (P), while oil-wetting tendencies correlate with iron (Fe), arsenic (As), and bromine (Br). The logistic regression model based on elemental composition explains 79% of wettability variance (Tjur’s R² = 0.79), compared to only 23% for mineralogical composition (Tjur’s R² = 0.23). Multivariate analysis of variance (MANOVA) confirms a highly significant effect for elemental profiles (F(2, 230) = 33.84, p < 0.001) versus non-significant effect for mineral profiles (F(2, 230) = 1.73, p = 0.179). Although calcite demonstrates a tendency toward oil-wetting and quartz toward water-wetting, these mineralogical relationships lack the statistical rigor observed in elemental correlations. The findings indicate that surface elemental composition shows a substantially stronger statistical association with wettability than bulk mineralogy under the conditions tested here (Tjur’s R² = 0.79 vs. 0.23) under ambient conditions. The novelty of this work lies in its comprehensive quantitative comparison across diverse lithologies, providing the systematic evidence that elemental composition significantly outperforms mineralogy in predicting wettability, with implications for reservoir characterization and enhanced oil recovery strategies.