The ability of solid peroxides to provide sustained release of both oxygen and hydrogen peroxide makes them potentially suitable for oxygen release or antibacterial applications. Most recent reports using solid peroxides to augment oxygen levels do so by compounding solid peroxide powders in polymers to retard the aqueous decomposition. Compounds with peroxidase activity may be added to reduce hydrogen peroxide toxicity. Peroxides are rarely pure and are mixed with oxide and themselves decompose to form hydroxides in water. Therefore, even if buffering strategies are used, locally the pH at the surface of aqueously immersed peroxide particles is inevitably alkaline. Since pH affects the decomposition of peroxides and hydrogen peroxide stability, this study compared for the first-time the aqueous decomposition products of hydrogen and inorganic peroxides that are in use or have been used for medical applications of have been evaluated preclinically; calcium peroxide (CaO₂), magnesium peroxide (MgO₂), zinc peroxide (ZnO₂), sodium percarbonate (Na₂CO₃.1.5H₂O₂) and hydrogen peroxide (H₂O₂). Since plasma can be approximated to be carbonate buffered phosphate solution, we maintained pH using carbonate and phosphate buffers and compared results with citrate buffers. For a given peroxide compound, we identified not only a strong effect of pH but also of buffer composition on the extent to which oxygen and hydrogen peroxide formation occurred. The influence of buffer composition was not previously appreciated, thereby establishing in vitro parameters for better design of intentional release of specific decomposition species.

Statement of significance

This paper compares for the first time the aqueous decomposition products oxygen and hydrogen peroxide of solid peroxy compounds of metal cations, (calcium, magnesium, sodium and zinc) across a pH range that could feasibly be found in the body, (pH 5,7, 9) either physiologically or pathologically. We find that in addition to pH, buffer composition is also a critically important factor, making translation from in vitro models challenging. Cytotoxicity was related to hydrogen peroxide release, alkalinity and in the case of zinc peroxide to the cation itself. In vitro and preclinical studies generally report release data from polymer-peroxide composites and rarely compare peroxides with one another. Together our data provide guidance for oxygen and ROS delivery from these inorganic materials.