Electrochemical Multisensor System for Monitoring the Hydrogen Peroxide Direct Synthesis in Microreactors ^†

Abstract

We present an electrochemical sensor system for the detection of hydrogen peroxide inside a direct synthesis microreactor. The setup allows the online, in situ measurement of high reactant concentrations by amperometric detection across the micro channel width and length. The robust integration of the electrochemical cell in the microreactor was demonstrated. Hydrogen peroxide was detected under reaction conditions (pH 3–4, presence of bromide) showing linear behaviour up to 2 mM with high sensitivity and excellent stability. The linear range was increased up to 10 mM by applying a diffusion limiting pHEMA layer to the electrode surface.

1. Introduction

Hydrogen peroxide (H₂O₂) with its use as ecofriendly, strong oxidizing agent is one of the key parts of a more sustainable chemical industry. The heterogeneous catalytic direct synthesis of H₂O₂ from the elements H₂ and O₂ provides an attractive alternative for the decentralized production in comparison to the established anthraquinone method [1]. Conventional reactor designs are not fit to yield the full potential of the direct synthesis process, as the mass transport resistances in the liquid reaction medium are production limiting factors. A way to overcome these challenges is the use of microreactors [2]. However, defining and optimizing the process parameters for these microreactors is challenging as commercially available analytic systems only allow the concentration measurement of the reactants H₂, O₂ and H₂O₂ at the outlets of the microreactor. To measure the concentration distribution of the reactants inside the microchannels with a high spatial and temporal resolution, for the first time we developed a measurement setup of miniaturized electrochemical sensors.

2. Materials and Methods

The sensor setup, which is based on platinum electrochemistry, allows the amperometric measurement of H₂O₂ through direct oxidation on the electrode surface [3]. By changing the applied electrode potential it is possible to also selectively measure the concentrations of the dissolved process gases O₂ [3] and H₂. For the robust integration of the sensors and to achieve a packaging compatible with the microreactor layout and its harsh environmental conditions, platinum electrodes with a diameter of 300 μm were encapsulated into an Stycast epoxy housing, including an electrodeposited Ag/AgBr reference electrode. As shown in Figure 1a, each prepared sensor plug contains two platinum electrodes as working (WE)/counter electrode (CE) or alternatively the reference electrode (RE). Depending on the chosen setup, several of these plugs can be mounted in the microreactor as schematically shown in Figure 1b, allowing the determination of the reactants gradient over the channel length. All amperometric measurements were performed in a 3-electrode-setup, inside a test microreactor with an electrolyte containing 0.15 mM H₂SO₄ and 4 mM NaBr at a flowrate of the electrolyte of 1.5 mL min⁻¹ if not stated otherwise.

3. Results and Discussion

One of the main challenges for the sensor setup to overcome is the sulfuric acid electrolyte (PH 3–4) also containing a NaBr as catalytic promoter. As shown in Figure 2a, the bromide ions inhibit oxidation of H₂O₂ in the potential range of 400 to 700 mV vs. Ag/AgBr, which makes it necessary to shift to higher detection potentials for the selective amperometric detection of hydrogen peroxide.

Furthermore, by increasing the measurement potential, an increase in background signal in the absence of hydrogen peroxide is visible. This effect is caused by the oxidation of the bromide ions to bromine at these potentials. For the selective detection of hydrogen peroxide it is therefore necessary to make a trade-off between very high detection and low background signals as shown in Figure 2b. For the following experiments a measurement potential of 850 mV vs. Ag/AgBr was chosen. By applying a measurement potential of 850 mV vs. Ag/AgBr we were able to detect H₂O₂ concentration linearly up to 2 mM inside the reactor (R² = 0.997) with an excellent sensitivity of 727 μA cm⁻² mM⁻¹. As shown in Figure 3a even after constantly measuring for 19 h the sensitivity for the detection concentrations up to 1 mM of hydrogen peroxide showed no loss in sensitivity (727 μA cm⁻² mM⁻¹ vs. 715 μA cm⁻² mM⁻¹; both with R² = 0.999). Due to the nature of the direct synthesis process of H₂O₂, the expected concentrations which can be obtained inside the microreactor setup exceed these concentrations more than tenfold. Therefore, the linear range for the amperometric detection of hydrogen peroxide was significantly increased up to a concentration of 10 mM (R² = 0.999) by applying a 20 μm thick layer of pHEMA hydrogel on the platinum electrode (Figure 3b) [3]. However, the sensitivity of the measurement was decreased due to the mass transport limiting effect of the hydrogel to a value of 24 μA cm⁻² mM⁻¹. A further increase in linear response might be obtainable by increasing the size of the pHEMA hydrogel layer. However, that will also decrease the sensitivity for the H₂O₂ detection further.

4. Conclusions

In this work, we present the first results for the electrochemical measurements of H₂O₂ with the described microreactor setup. The influence of the presence of bromide ions in the electrolyte on the detection process was examined. Furthermore, the overall sensitivity, linearity and stability of the amperometric H₂O₂ detection was evaluated. By applying a pHEMA-based hydrogel to the surface of the electrode a tenfold increase in linear range was achieved. Hereby we successfully showed the measurement of H₂O₂ under reaction conditions inside a microreactor.