(235f) Reactor Design Optimization for Direct Synthesis of Hydrogen Peroxide

Inoue, T., National Institute of Advanced Industrial Science and Technology (AIST)
Kikutani, Y., Kanagawa Academy of Science and Technology
Hamakawa, S., National Institute of Advanced Industrial Science and Technology (AIST)
Mizukami, F., National Institute of Advanced Industrial Science and Technology (AIST)
Kitamori, T., The University of Tokyo

Reactor design was investigated for the direct reaction between hydrogen and oxygen to produce hydrogen peroxide. Hydrogen peroxide is primarily produced by the anthraquinone process, in which 2-alkyl-9, 10-anthraquinones are hydrogenated with palladium catalyst in the first step, and the products are subsequently oxidized by air to produce hydrogen peroxide. This process is safe and easily scaled-up, but anthraquinone decomposes gradually during the redox cycle. That results in trace amounts of contamination, and further purification step is required for obtaining products.

Compared with the current anthraquinone process, direct production from hydrogen and oxygen can enjoy less energy consumption and environmental benignancy. This reaction is conducted as multiphase reaction among hydrogen - oxygen mixture (gas phase), aqueous medium (storing hydrogen peroxide produced, liquid phase) and supported palladium catalyst (solid). In spite of the advantage, widespread application of the process has been limited by the risks inherent in handling the explosive mixture. Moreover, high pressure from 2 MPa to 10 MPa are typically needed to promote mass transfer from gas to liquid phase.

Recently, microfabricated reactor technology has been successfully applied to selective hydrogen peroxide production by handling explosive mixture of hydrogen and oxygen, as shown by one of the authors. 1 Still, there is more room for optimizing the microfluidic design to promote the productivity of hydrogen peroxide. Scaling out technology should be also implemented so that the productivity meets the industrial requirement. As the reaction is conducted as three phase reaction, reactor design is crucial for the productivity.

We conducted the reaction in a glass fabricated microreactor. As it is transparent, flow regime inside the reactor can be easily monitored during the reaction, which is crucial for the reactor performance. 1 Also glass as the reactor material has been proved to be suitable for fine chemical processes, for example in Lonza's c-SSP process. 2 Microchannels were fabricated on a glass wafer (30mm width X 70mm length X 0.7mm depth) by the combination of chemical etching (Institute for Micro Technology (IMT), Japan) and mechanical machining. Chemical etching was suitable for fabrication of a microchannel with less than 100 micrometer width, while mechanical machining is suitable for that of larger channels with more than 200 micrometer width. Once microchannels as well as inlet/outlet holes were fabricated on a glass wafer with the same dimensions, the two wafers were thermally bonded (IMT). Catalyst was loaded into the channel (600 micrometer width X 300 micrometer depth) as slurry.

We found that the gap between the size of microchannels for gas and liquid flows is crucial for the reactor design, especially for validating proper gas - liquid distribution into the reactor. When gas channel was with more than 100 micrometer width, the flow regime was unstable because of the crosstalk between gas and liquid dispensing channels, causing pulsing flow into the catalyst bed. The crosstalk between the gas - liquid dispensing channel was significantly suppressed when the width of gas inlet channel was reduced to 50 micrometer (depth of the channel became 25 micrometer because of the geometrical characteristics of wet etching). As both of the gas and liquid flows were driven by , we regard that the difference of pressure drop by gas and liquid is responsible for this crosstalk issue. 3 Gas channels should be narrow enough in order to compensate the pressure drop difference between gas and liquid channel.

Proper gas - liquid distribution in the reactor is crucial for the reactor performance, which will be presented at the conference. Also another effort on scaling out is in progress, and reactor performance will be presented at the conference.

(1) Inoue, T.; Schmidt, M. A.; Jensen, K. F. Microfabricated Multiphase Reactors for the Direct Synthesis of Hydrogen Peroxide from Hydrogen and Oxygen. Ind. Eng. Chem. Res. 2007, 46, 1153.

(2) Roberge, D. M.; Bieler, N.; Thalmann, M. Microreactor technology and continuous processes in the fine chemical and pharmaceutical industries. PharmaChem. 2006, June, 15.

(3) Wada, Y.; Schmidt, M. A.; Jensen, K. F. Flow Distribution and Ozonolysis in Gas - Liquid Multichannel Microreactors. Ind. Eng. Chem. Res. 2006, 45, 8036.