Table 4. Comparative Analysis of Wettability Measurement Techniques Using MDS, CFD, and ML Approaches.

Type	Objective	Algorithm	Key finding	Ref
MDS	Studying nanoscale wettability of different rock surfaces for ore flotation and oil recovery	Condensed-phase Optimized Molecular Potentials for Atomistic Simulation Studies (COMPASS) force-field, Andersen thermostat, and Ewald summation methods	Rock wettability ranks as gypsum > calcite > halite > silica > graphite	(58)
			Silica’s water contact angles change with oil polarity, ranging from 58° in hexane to 118° in toluene
	To utilize MDS for simulating some phenomena at a molecular level (oil/brine/rock system) and develop a model that correlates free energies with contact angles	Forcite module in Materials Studio, incorporating the COMPASS force field and Nosé–Hoover thermostat	The model accurately predicts experimental results (R2 ∼ 0.98)	(228)
			MDS effectively and economically assesses sandstone wettability changes
	Analyzing shale organic matter’s wettability under reservoir conditions	COMPASS force field	Water becomes more wettable with rising temperature under certain pressures	(203)
			Water wettability drops with higher salinity; Mg2+ and Ca2+ ions affect wetting more than Na+
	Studying how interface wettability affects Lennard-Jones fluid flow in nanochannels via molecular dynamic simulations	Non-Equilibrium Molecular Dynamics (NEMD) simulation with a modified Lennard-Jones potential	Wettability greatly affects interfacial hydrodynamic resistance	(229)
			Near solid walls, temperature and pressure profiles vary due to wettability, with higher interfacial friction increasing temperatures
	Assessing wetting properties and adsorption capacities of surfactants (ALES, SLES, TD, SDS) on coal dust, by using MD simulations and quantum calculations	Amorphous cell module	Wetting performance is mainly influenced by surfactants’ molecular structure, EO groups, and hydrolytic cations	(230)
			NH4+ in ALES and the combined hydrophilicity of EO groups and SO4– in ALES and SLES greatly improve their wetting and adsorption on coal dust
	To simulate nano water droplets’ wetting behavior on flat and pillar surfaces using molecular dynamics	Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) with Lennard-Jones potential and SPC/E water model	Higher surface energy on flat surfaces reduces the contact angle, indicating better wettability	(231)
			Nanostructured pillar surfaces are more hydrophobic than flat ones, with sparser pillars increasing the contact angle
	To assess how reservoir temperature and kerogen structure affect shale kerogen wettability	LAMMPS with Lennard-Jones potential and SPC/E water model	Kerogen contact angles vary by type and maturity, with Type II being more water-wet and Type III the least	(232)
			Rising reservoir temperatures decrease contactangles by 62%, increasing water-wettability
CFD	To study the dynamic wetting and wrapping process of coal dust by spray droplets	A numerical simulation study focusing on the dynamic wetting and wrapping process of coal dust particles	The study found that a particle size ratio (θ) over 2 between droplets and dust allows droplets to fully cover and wet coal dust, while a ratio under 1 results in ineffective wetting	(233)
	To determine the optimal conditions for effectively wetting and settling respirable dust in mining environments		Droplet velocity is key to wetting efficiency; lower velocities cause poor coverage, but at around 20 m/s, wetting becomes faster and more effective
	To evaluate how surfactant-based nanofluids change wettability and their impact on enhancing oil recovery	multicomponent multiphase, compositional model	Surfactant-enhanced nanofluids can shift rock wettability in tight oil reservoirs from oil-wet to water-wet; this change improves capillary pressure and permeability, aiding in increased oil extraction	(234)
	To understand how self-propelled droplets on superhydrophobic surfaces with varying wettability control their jumping direction	The interFoam solver in OpenFOAM which is based on the VOF model	Surface energy gradients, influenced by wettability, direct droplet movement toward smaller contact angles	(235)
	To examine how reservoir wettability, pore geometry, and fluid viscosity ratios impact relative permeability	LBM for simulating two-fluid flow in porous media	Increasing reservoir rock wettability from neutral to strong reduces the wetting fluid’s relative permeability	(236)
			This effect on the wetting fluid’s permeability diminishes at higher viscosity ratios, while the nonwetting fluid’s permeability is significantly higher in strongly wet conditions compared to neutral wettability
	To assess how silicon oxide nanoparticles, affect wettability in glass micromodels to boost oil recovery	CFD (with method performed using Ansys Fluent software)	Initial wettability significantly affects oil saturation and recovery	(197)
			Nanofluid-induced wettability alteration is time-dependent
			Silica nanoparticles form a nanotexture coating on solid surfaces, changing the wettability from oil-wet to intermediate-wet, enhancing oil recovery
	To explore microtexture lubrication mechanisms through the lens of interface wettability	CFD	Isotropic textures with low skewness and high kurtosis improve wettability and spreading	(237)
			Groove-shaped textures cause anisotropic spreading through continuous triple contact lines, improving wettability
	To evaluate the wettability of silicon microstructures made	CFD with VOF model	The VOF model accurately simulates macroscopic contact-angle characteristics on complex surfaces	(238)
	To analyze and understand discrepancies between experimental and theoretical contact angles of these surfaces complex surfaces		A 2D CFD model with an equivalent triangular waveform precisely predicts contact angles
	To examine how surface wettability, inclination, liquid properties, and impact velocity affect drop impact and spreading dynamics	VOF model	The Static Contact Angle (SCA) model is precise for less wettable surfaces (SCA > 90°), matching experimental results	(239)
			For more wettable surfaces (SCA < 90°), the dynamic contact angle (DCA) is higher initially, requiring the DCA model for accurate predictions
ML	To employ ML tools to efficiently determine the contact angle, is crucial for assessing shale wettability in CO2-enhanced oil recovery, CO2 sequestration in saline aquifers, and hydraulic fracturing	LR and RF	The operating pressure, temperature, total organic content (TOC), and mineral matter significantly affect shale wettability	(240)
			The RF model proved reliable for predicting contact angles, offering an alternative to complex experimental methods in CO2 sequestration applications
	To improve carbon capture, utilization, and storage by leveraging ANN and ANFIS for estimating the contact angle in coal-water-CO2 systems.	ANN and ANFIS	The ANN and ANFIS models demonstrated high accuracy in predicting contact angles in coal formations	(222)
			A new empirical equation derived from the ANN model showed high accuracy, confirming the reliability of machine learning in accurately predicting contact angles
	To estimate shale wettability, a key factor in CO2-EOR and CO2 storage in shale formations	Decision tree, RF, function networks, and gradient boosting regressor	The operating pressure, temperature, rock mineralogy, and TOC as significant factors affecting shale wettability	(241)
			Machine learning models, especially the gradient boosting regressor, were effective in predicting contact angles
	To explore how rock mineral composition and properties of oil-based drilling fluids affect the wettability of tight gas sandstone formations	General Regression Neural Network (GRNN)	The mineral compositions of quartz, feldspar, carbonate, and clay significantly affect the wettability of tight sandstone	(242)
			The GRNN model accurately predicted wettability
	To evaluate shale wettability based on formation pressure, temperature, salinity, TOC, and theta zero	Multilayer perceptron (MLP) and radial basis function neural networks (RBFNN)	The Radial Basis Function Neural Network-Many-Objective Optimization (RBFNN-MVO) emerged as the most accurate	(243)
			Sensitivity analysis identified theta zero, TOC, pressure, temperature, and salinity as key factors influencing shale wettability
	To assess shale wettability by analyzing critical factors such as formation pressure, temperature, salinity, TOC, and theta zero	MLP and RBFNN	The RBFNN-MVO model was the most accurate among tested ML models for predicting shale wettability	(225)
			The RBFNN-MVO model was validated as an efficient and cost-effective tool for shale wettability prediction
	To assess the wettability of CO2/brine/rock minerals (quartz and mica) in geological formations	Fully connected feedforward neural networks, extreme gradient boosting, k-nearest neighbors, decision trees, adaptive boosting, and RF	FCFNN model was more effective than other machine learning techniques in predicting the wettability of the mineral/CO2/brine system, achieving an R2 above 0.98 and an error below 3%	(224)