Welcome on the ECCE-6 CDROM.

Conference logo

European Congress of Chemical Engineering - 6
Copenhagen 16-21 September 2007

Abstract 702 - Process intensification of electro-organic synthesis by microstructuration

Process intensification of electro-organic synthesis by microstructuration

Special Symposium - EPIC-1: European Process Intensification Conference - 1

EPIC-1: Intensified Hydrodynamics & Structured Environments (IHSE-2)

Mr ANIS ATTOUR
Laboratoire des Sciences du Génie Chimique (LSGC-ENSIC)
LSGC-groupe GPM
1, rue Grandville
BP 451
54001 Nancy Cedex
France
France

Keywords: Microreactor, Electro-organic synthesis, Process intensification, reactor model

Process intensification of electro-organic synthesis by microstructuration

A.Attour1*, S.Rode1, F.Lapicque1, A.Ziogas2 and M.Matlosz1
1 : Laboratoire des Sciences de Génie Chimique, CNRS-ENSIC, BP-451, 1 rue Grandville, F–54001 Nancy, (*auteur correspondant) attour@ensic.inpl-nancy.fr
2 : Institut für Mikrotechnik Mainz (IMM) GmbH, Carl-Zeiss-Strasse 18-20 D-55129 Mainz

Electroorganic synthesis is a domain in which microstructured devices are especially promising, with potential improvement of both production rates and selectivity. This study deals with design, build-up and validation tests of a new microreactor for organic electrosynthesis. The miniaturised cell is characterized by a 100 µm electrode gap leading to a reduced ohmic drop even with low-conducting organic electrolytic media, and to high mass transfer coefficients. The anode is composed by 10 glassy carbon elements, each with an area of 1x1 cm2. The cathode facing the segmented anode is a 10x1 cm2 stainless steel plate. The reactor was designed in order to obtain high conversion of the substrate in continuous operation without recirculation of the liquid. The reactor was constructed by Institute of Microtechnics of Mainz (IMM). The model reaction was the anodic oxidation of the 4-methoxytoluene to 4-methoxy-benzaldehyde-dimethylacetal in methanol: this reaction is carried out at production scale in capillary-gap cells operated at low conversion per pass and integrated in a recycle loop, and exhibits selectivity problems because of side-reactions. The reduction of the methanol solvent at the cathode results in formation of hydrogen.
We first used the cell without segmentation of the anode, each element being connected to the same potential: optimal conversion and selectivity were searched by selecting the nature of supporting electrolyte and its concentration as well as flow rate and the applied current. Moreover, in order to increase the cell productivity, with conditions closer to industrial practice, the reagent concentration was increased to 0.1 M – higher concentrations couldn’t be used because of significant overheating of the cell due essentially to the high ohmic drop leading to heat dissipation by Joule effect. The concentration of the supporting electrolyte, KF, exerts an influence on the reaction selectivity: decreasing the KF concentration to 0.01 M increases noticeably the acetal production. For optimal conditions (0.2 ml/min, 0.1 M of reagent and 0.01 M of KF), selectivity up to 86 % were attained with the single-pass microreactor. This result is higher than that obtained in industrial practice (68%).
A reactor model was established, based on the kinetics of the different electrochemical reactions and on mass balances on the different species. The reactor was considered to be in plug-flow with the mass transfer to the electrodes described by Newton-type laws. The electrochemical kinetics were described by Tafel-type laws. Following this model, the reactor performance depends on three non dimensional numbers related to operating parameters: a non dimensional current, comparing the applied current to the current necessary for complete reagent conversion, a number of transfer units, comparing the mass transfer velocity to the electrolyte flow velocity and a Wagner number, comparing the polarisation resistance to the ohmic resistance. It is shown that the current distribution is uniform in the cell for low Wagner numbers and non uniform for high Wagner numbers. Compared experimental production rates obtained with and without local control of current density confirmed the predictions of the model.
Acknowledgement These investigations are carried out within the IMPULSE project founded by EU (contract n°011816-2)

Presented Wednesday 19, 15:20 to 15:40, in session EPIC-1: Intensified Hydrodynamics & Structured Environments (IHSE-2).

Conference logo