Local solvation structures govern the mixing thermodynamics of glycerol-water solutions

Abstract

Glycerol is a major cryoprotective agent and is widely used to promote protein stabilization. By a combined experimental and theoretical study, we show that global thermodynamic mixing properties of glycerol and water are dictated by local solvation motifs. We identify three hydration water populations, i.e., bulk water, bound water (water hydrogen bonded to the hydrophilic groups of glycerol) and cavity wrap water (water hydrating the hydrophobic moieties). Here, we show that for glycerol experimental observables in the THz regime allow quantification of the abundance of bound water and its partial contribution to the mixing thermodynamics. Specifically, we uncover a 1 : 1 connection between the population of bound waters and the mixing enthalpy, which is further corroborated by the simulation results. Therefore, the changes in global thermodynamic quantity - mixing enthalpy - are rationalized at the molecular level in terms of changes in the local hydrophilic hydration population as a function of glycerol mole fraction in the full miscibility range. This offers opportunities to rationally design polyol water, as well as other aqueous mixtures to optimize technological applications by tuning mixing enthalpy and entropy based on spectroscopic screening.

Fig. 1. Schematic representation of the hydration shell of a glycerol molecule with bound (blue colored) and wrap (red colored) water molecules. The non-shell (bulk-like) water molecules are indicated with gray color.

Fig. 2. The frequency dependent (a) absorption coefficient and (b) average molar extinction coefficient of glycerol–water binary mixtures with various mole fraction of glycerol (X_gly). Arrows indicate increasing X_gly.

Fig. 3. (a) Molar effective extinction coefficient (ε_hydration) at 20 °C as calculated by subtracting the bulk water and bulk glycerol spectra from the glycerol–water mixtures spectra (eqn (6)). The yellow dotted line represents pure water and pure glycerol. ε_hydration increases with frequency for all the solutions in the 350–450 cm⁻¹ spectral region (gray shaded area). The dotted black line is the linear fit in that region for glycerol mole fraction; X_gly = 0.8, as an example. The slope obtained from such fit is then used for interpreting the spectral trends. (b) The slopes derived from experimental ε_hydration spectra are plotted for all X_gly at three different temperatures (5 °C, 20 °C and 40 °C, respectively). The lines are guided to the eye and the error bars are indicated by the shaded regions. (c) The populations of various types of water in glycerol–water mixtures as predicted by MD simulations.

Fig. 4. The three-body angle distributions for the (a) bound and (b) wrap waters exhibit a monotonic increase in the population of tetrahedrally-coordinated waters (θ = 109.5°) and a monotonic decrease in the population of icosahedrally-coordinated waters (θ = 64°) relative to pure water at 18 °C. (a) At low concentrations, the bound waters are more icosahedral and less tetrahedral than pure water. (b) Wrap waters show higher tetrahedral populations and lower icosahedral populations than pure water for the entire concentration range. (c) Here, we depict the population of tetrahedrally-coordinated waters p_tet for wrap (red diamonds), sample-averaged (black squares), and bound (blue circles) as a function of glycerol mole faction. For reference, we indicate the pure water p_tet with a horizontal dashed line.

Fig. 5. (a) Linear correlation between ΔH_mixing (from Ref a, as an example) and the slope in the libration range of the ε_hydration spectra (Δε_hydration/Δν), associated to the bound water population. The blue line is a linear fit. (b) The trend of the experimentally determined slope (Δε_hydration/Δν) as a function of glycerol content is compared with that of mixing enthalpy values (ΔH_mixing). ΔH_mixing values are taken from four literature data sets: Ref a; calculated, Ref b; extrapolated, Ref c, Ref d, as well as from the present MD simulations. All data sets are scaled (from 0 to −1) for better comparison.