The mechanism of stabilization of Ribonuclease A (RNase A) in the freeze-dried state by the inclusion of carbohydrates, polyols, polymers or amino acids was investigated. Two proposed mechanisms of stabilization were tested: (1) the solution state theory of preferential exclusion and hydration and the subsequent vitrification potential of stabilizing excipients; and (2) water replacement by excipients and requisite hydrogen bonding between the protein and stabilizing excipient. The objective of this work was to characterize protein-induced changes in the surface characteristics and bulk properties of the stabilizing excipient. Specific test methodologies were used to characterize the excipient-protein interaction potential, including immersional calorimetry, dynamic sorption analysis, differential scanning calorimetry (DSC), and FTIR. Testing was done on the freeze-dried excipient, RNase A and excipient/RNase A formulations. Moisture content of the formulations was determined using KF and TGA and found to be consistently less than 2% for the formulations. Concentration of the protein was determined using fluorescence spectroscopy and correlated with a coupled enzyme assay to determine specific activity. Protein instability, as exemplified by a reduction in specific activity, was evident in the sucrose and glycine formulations. Depression of the glass transition temperature (T{dollar}\sb{lcub}\rm g{rcub}{dollar}) of freeze-dried sucrose was attributed to complexation between the sucrose and RNase A, an increase in protein mobility due to a redistribution of moisture in the freeze-dried material, or to RNase A acting as a plasticizer. Comparison of heats of solution ({dollar}\rm\Delta H\sb{lcub}SOLN{rcub}{dollar}) for glycine formulations suggest that glycine binds to the protein surface. Comparison of monolayer values (W{dollar}\sb{lcub}\rm m{rcub}{dollar}) suggest that high molecular weight polymers, such as PVP and Ficoll 70, do not associate with the protein surface due to steric effects. Examination of spectral shifts in the Amide I and II bands of RNase A was inconclusive. It can be concluded from these results that for RNase A, excipients which do not associate with the protein in the freeze-dried state, yet remain amorphous following freeze-drying, act as stabilizers. It is anticipated that this work will contribute to the understanding of protein stabilization in the freeze-dried state and promote a more rational approach to the formulation development of biopolymers.