Reactions at the water/solid interface are central to develop more sustainable processes, from biomass upgrading to the use of unconventional activation as in photocatalysis and electrocatalysis. To gain atomistic insight on those reactions, modelling approaches were constantly improved in the past decade. Several approaches are available nowadays in the literature, from continuum models to a full explicit description of the liquid water.[1] Moving to a full description of reactivity the water/metal interface is still beyond a full complete DFT approach provided the minimal size of the periodic cell that is necessary and the minimal sampling required. Moving to classic molecular mechanics (MM) is then a natural choice to avoid the large computational burden associated with ab initio molecular dynamics. But pair-wise force fields fail to properly describe chemisorption. This is why we proposed a QM/MM approach (MMSolv) to keep a high-level description of the surface/adsorbate interaction.[2] MMSolv combines ab initio adsorption energies with solvation energies computed at the MM level using alchemical transformation. Then, the quality of the prediction relies strongly on the quality of the force field. Typically, most of the water/metal force fields are not describing well the near chemisorption character of this interaction.[3] Our GAL force field was specifically designed to describe the strength of this interaction.[3][4] This strength is strong enough to challenge the alchemical transformation. We circumvented this issue passing through a well-behaved physisorption potential for Pt/water as implemented in our SolvHYBRID package.[5] Then, we validated MMSolv against experimental data on the adsorption of aromatics at the Pt/water interface.[5] MMSolv performs noticeably better than PCM thanks to a proper inclusion of the surface solvation. It can be easily used with oxides surfaces, given that the chemisorption of water is well described.[6] It was then compared with other approaches, from continuum models to enhanced ab initio molecular dynamics (thermodynamic integration, metadynamics) on the adsorption of ethanol at the alumina/water interface.[7] Ultimately, combining MMSolv with DockOnSurf,[8] our new tool to screen thousands of adsorption modes for flexible molecules, we identified the best adsorption modes of polyols such as xylitol at the alumina/water interface. This demonstrates the ability of MMsolv to gain atomistic insight into adsorption at the solid/liquid interface, including the (de)-solvation of the surface and polyfunctional molecules. References 1.M. Saleheen and A. Heyden, ACS Catal. 8 (2018), 2188-2194 2. S. N. Steinmann, P. Sautet, and C. Michel, Phys Chem Chem Phys, 18 (2016), 31850–31861 3. S.N. Steinmann, R. Ferreira De Morais, A. W. Götz, P. Fleurat-Lessard, M. Iannuzzi, P. Sautet and C. Michel, J Comp Theo Chem, 14 (2018), 3238-3251 4. P. Clabaut, P. Fleurat-Lessard, C. Michel, S.N. Steinmann, J Chem Theo Comput, 16, (2020) 4565-4578 5. P. Clabaut, B. Schweitzer, A.W Goetz, C. Michel, S.N. Steinmann, J Chem Theo Comput, 16 (2020) 6539 - 6549 6. J. Rey, S. Blanck, P. Clabaut, S. Loehlé, S. N. Steinmann, C. Michel, J Phys Chem B 125 (2021) 10843