TECHNOLOGICAL PROBLEMS OF HYDROGEN PEROXIDE PRODUCTION BY THE ANTHRAQUINONE PROCESS: A SYSTEMATIC REVIEW, CAUSAL ANALYSIS AND A DATA-COLLECTION PROTOCOL

Almost all hydrogen peroxide manufactured worldwide comes from the anthraquinone (AO) process, yet the technological problems of that process are studied as separate, largely disconnected topics. This review classifies those problems along the process chain and maps the causal links between them, using a corpus of 46 sources whose digital object identifiers were individually verified against the Crossref and OpenAlex registries. The problems of the hydrogenation, oxidation, extraction and distillation stages prove not to be independent: they converge at system level on four outcome indicators reagent and adsorbent consumption, explosion and fire hazard, energy intensity, and wastes and emissions. Eight causal links documented in the reviewed literature are distinguished from three that remain hypotheses and require measurement: the effect of residual hydrogen peroxide returned from extraction on catalyst activity, the effect of water accumulating in the working solution on carrier degradation, and the feedback of degradation products on the attainable hydrogenation degree. Together these three close a self-reinforcing loop which, if confirmed, would explain a slow but progressive deterioration of plant performance. Six topics yielded no source meeting the inclusion criteria: solvent losses and volatile organic compound emissions from the solvent-recovery vent; water and residual hydrogen peroxide treated as independent variables; leaching of palladium into the working solution; inert accumulation and purge policy in the hydrogen loop; aqueous effluent streams of an AO plant; and membrane reactors for hydrogenation. To make the hypotheses and gaps testable, a fifteen-item data-collection protocol is proposed that specifies for each item the measured quantity, analytical method, unit, target precision, sampling point, frequency and the specific link it tests, allowing the review to be connected directly to laboratory or plant data obtained later.