The problem tested in the experiment is whether the protein-osmolyte-water solution stabilizes or destabilizes the unfolded state of the protein. This problem was used to test the mechanism by which osmolytes, small organic compounds, interact with the protein to affect its stability. The proposed hypothesis indicated that the free energy transfer of the protein backbone from water to a water/osmolyte solution, Δgtr, is negatively correlated with the fractional polar surface area of ​​an osmolyte. Δgtr is the unit used to measure the degree to which an osmolyte stabilizes the protein given that Δgtr>0 (stabilized) and Δgtr<0 (destabilized). The independent variable was fractional polar area (SA) and the dependent variable was energy change or stability (Δgtr). The model included two other adjustable parameters or independent variables: polar and nonpolar SAs and the interaction between the solvent and the protein backbone. The Δgtr was calculated using all adjustable parameters. The first experiment involved the quantitative solvation model in which the interaction energy of the solvent is a function of the polarity of the interactant. The other experiment implied that the number of energetically equivalent ways to achieve a given interaction was a function of the surface area of ​​the interactant. To carry out the experiment, the researchers used a 1 M osmolyte solution. Using the X-ray structure of eight stabilizing osmolyte calculations were carried out. Although comparisons between osmolyte structure and Δgtr values ​​indicated no obvious correlation, there was a clear correlation between Δgtr and fractional polar area (R = 0.88). The second experiment used the quantitative solvent model (water and osmolyte) to test interactions with backbone polar groups. This indicated that the skeleton/osmolyte interactions became increasingly favorable as the osmolyte became increasingly polar. The result indicated for an osmolyte concentration of 1 M, the calculated and measured Δgtr values ​​are in good agreement. In particular, as the fractional polar surface area increases, the interaction of osmolytes with the protein backbone becomes increasingly favorable, i.e. their Δgtr values ​​decrease. The change in free energy for folding/unfolding will depend linearly on the osmolyte concentration. The statistical mechanics model was used to calculate the average protein backbone energy in osmolytic solutions. About 90% of the calculated Δgtr have a correlation greater than 0.80. The success of the second experiment was consistent with their hypothesis. Interactions between a single amide nitrogen and two carbonyl oxygen backbones were used for solvent interactions. Each of these sites had a positive, neutral or negative ether charge presented by the solvent (water or osmolyte).