Hydrogen peroxide is an important industrial commodity. For large scale industrial use, the production via the established anthraquinone process is still economical most favourable. However, smaller applications would require substantial logistic efforts for the distribution. For these applications the local production could be more suitable. Electrochemical methods using renewable energy would further support the environmental friendliness. A simple electrochemical way to produce hydrogen peroxide is the partial reduction of oxygen from air according to equation [1].

A limiting factor is the undesired further reduction of hydrogen peroxide to water at the same electrode according to equation [2]. As the reversible potential for this reaction is much higher than that for the oxygen reduction, the reaction is kinetically favoured. Some efforts were made to find catalyst with a low activity for this undesired reaction. One candidate material for an acidic environment are gold catalyst with a small amount of added palladium (1, 2). In own test this was confirmed in RRDE tests (3).

In this contribution the transfer into a full cell will be described. As cell a Micro Flow Cell with 10 cm² electrode area and n Electro MP Cell with 100 cm² electrode area were used, both ElectroCell A/S, Denmark. Both cells were modified to accommodate house made gas diffusion electrodes. On the cathode side commercial oxygen evolution electrodes by ElectroCell A/S were employed. The gas diffusion electrodes were made from fuel cell type GDLs H23C8 by Freudenberg, Germany. Catalyst layer was either applied by sputter coating or by airbrushing a catalyst ink of carbon supported catalyst. 2 M sulphuric acid was used as electrolyte on both sides. The electrolyte was flown through gaps between the electrode and the separation membrane made from FumaSep F-101020 PK by Fumatech, Germany (cf. Fig. 1). The gap width in 10 cm² cell was 4 mm on both sides. In the 100 cm² cell the anodic gap width was increased to 8 mm on the anode side in order to decrease the gap on the cathode side to 1.5 mm.

Hydrogen peroxide concentrations were measured using a commercial electrochemical sensor by AMT, Germany downstream of the cell.

Tests were either performed by passing a small electrolyte flow through the cathode side once a sampling the product (single pass) or by cycling the electrolyte at a high flow rate till a stable concentration was reached (cycled batch).

For the 10 cm² cell tests with electrodes sputtered with a 50 nm platinum layer were compared to testes with electrodes coated with 20 wt% PdAu/C catalyst with 0.15 mg cm^-2 metal loading.

In single pass tests with the platinum electrode the highest observed hydrogen peroxide concentration of 0.3 wt% was obtained at an electrolyte flow rate of 0.1 ml min^-1 at a current of 250 mA corresponding to a current efficiency of ca. 11% (c.f. Fig. 2). Increasing the flow rate to 0.5 ml min^-1 decreased the H₂O₂ concentration to 0.2% wt% which, however, corresponds to a current efficiency of 37.8%. This can be seen as indication for hydrogen peroxide reduction at longer residence time. Accordingly, no built up in concentration of H₂O₂ was observed in cycled batch operation. Surprisingly, for the electrode coated with PdAu/C catalyst a much lower hydrogen peroxide concentration of 0.1 wt% was obtained at 0.1 ml min^-1 flow rate in spite of a higher current of 550 mA (c.f. Fig. 3), so that the current efficiency was less than 2%. This can only be explained by a long hold back time of hydrogen peroxide in the porous structure of the supported electrode.

The strong influence of the local concentration is further supported by the finding that during first tests in the larger cell with smaller gap and thus stronger flow it was possible to obtain hydrogen peroxide concentrations of 0.5 wt% with the sputtered platinum electrode. Here, also an accumulation of hydrogen peroxide during cycled batch operation was possible so that hydrogen peroxide concentrations up to 0.6 wt% were accomplished.

In the contribution more details on the tests will be discussed supporting the importance of the cell design for the reaction selectivity.

1. J. S. Jirkovsky, I. Panas, E. Ahlberg, M. Halasa, S. Romani and D. J. Schiffrin, Journal of the American Chemical Society, 133, 19432 (2011).
2. J. S. Jirkovsky, I. Panas, S. Romani, E. Ahlberg and D. J. Schiffrin, Journal of Physical Chemistry Letters, 3, 315 (2012).
3. C. Cremers, B. Kintzel, K. P. Duraisamy, K. Pinkwart and J. Tübke, Meeting Abstracts, MA2016-02, 1652 (2016).