An Elucidation of Substrate Effects in Graphene-Based Sensing Characteristics─Interfaces between Organic Solvent and Graphene

Most graphene-sensor researches have focused on direct graphene modifications to enhance performance. However, supporting-substrate effects on graphene sensing mechanisms remain underexplored. Because of graphene 2D architecture, substrates affect its surface potential, wettability, and molecular adsorption. These effects intensify in the presence of polar molecules, e.g., water molecules, further complicating the sensing characteristics. To explore these effects, this study investigates the influence of substrate on the sensing capabilities and mechanisms of graphene field-effect transistors (GFETs) in organic solvents through electrical-transport measurements. Specifically, we compare partially suspended graphene FETs (PS-GFETs) and oxide-supported graphene FETs (OS-GFETs) in response to dimethyl sulfoxide (DMSO), ethanol, and isopropanol (IPA) at different concentrations. By quantifying Dirac-point hysteresis, we experimentally show that the hysteresis correlates with molecular polarity, following the trend DMSO < ethanol < IPA. Moreover, OS-GFETs exhibit a 1.5-fold sensitivity enhancement compared to PS-GFETs when detecting organic solution concentrations. Employing the two-dimensional hydrogen bond network (2D-HBNS) model, we theoretically illustrate that hydrophobic PS-GFET surfaces maintain equilibrium through hydration shell and 2D-HBNS formation. In contrast, hydrophilic OS-GFET surfaces disrupt this balance, enhancing van der Waals interactions and attracting organic molecules. This leads to superior sensitivity in OS-GFETs. To further validate this hypothesis, we introduced poly(methyl methacrylate) (PMMA) and polytetrafluoroethylene (PTFE) layers on the SiO₂ substrate. The experiments show it changes graphene-surface hydrophilicity and graphene-sensor sensitivity. These findings establish a theoretical and experimental framework for optimizing graphene-based sensors. This framework elucidates a solute–solvent interfacial interaction model for polar liquids, aiming to improve the sensing characteristics of 2D materials.