The neutral form of glucosamine, C6H13NO5, one the most effective and most widely used over-the-counter health supplements for the relief of osteoarthritis, is very unstable in air. It is marketed as chloride and sulfate salts. Unlike the stable glucosamine chloride, direct use in pharmaceutical formulations of the sulfate, ostensibly the physiologically more active form, is hindered by its strong hygroscopicity. Copious patent literature exists describing methods for stabilization of the sulfate by converting it into double and/or mixed salts, usually with alkaline or earth alkaline sulfates and chlorides. Aiming to unravel the structures of the alleged double/mixed salts, we attempted synthesis of the stabilized forms of the sulfate following literature procedures. Our repeated attempts did not yield true glucosamine sulfate or any real (in the chemical sense) double or mixed salts. Instead, Fourier transform infrared spectroscopy, powder X-ray diffraction, thermogravimetric analysis, and elemental analyses consistently showed that physical mixtures of the stable glucosamine chloride, which has a strong propensity to crystallize out from solutions, and the respective alkaline salts are obtained in all cases. Expectedly, these mixtures were non-hygroscopic. The analysis of the commercially available sample of "glucosamine sulfate" showed that it is a mixture of glucosamine chloride and K2SO4, in accordance with the above conclusions. By using a simple ion exchange in glucosamine chloride, we devised a simple method to generate glucosamine sulfate. As anticipated, the latter is a very hygroscopic powder in the solid state and is chemically moderately unstable in solution. Along with the conclusions based on the products obtained following published procedures, reaction of this compound with alkali chlorides readily affords the (non-hygroscopic) glucosamine chloride in a mixture with the respective alkali sulfates. We are tempted to conclude that the alleged "stabilization" of glucosamine sulfate by formation of double/mixed salts is (in the chemical sense) misleading. We believe that these compounds have probably never been obtained, and the related published synthetic procedures should be reinvestigated. The conclusions of this study could have important implications on the effective amount of the active ingredient required to achieve physiological activity, because such "stabilized" mixtures contain less than the optimal amount of the physiologically active ingredient, which could also have some commercial implications.
