(173g) Investigating the Effect of Acids and Halides on Direct Synthesis of Hydrogen Peroxide

H₂O₂ is an environmentally benign and selective oxidizer which can potentially replace chlorine and chlorinated oxidizers in many industrial applications. It is currently produced on large scale by the energy- and cost-intensive anthraquinone auto-oxidation (AO) process which makes it more expensive than chlorine. Direct synthesis (DS) of H₂O₂ is a greener alternative to AO with less energy and cost requirements, but DS suffers from low H₂O₂ selectivities on Pd (the most commonly used catalyst for DS) in the absence of caustic additives like acids and halides (> 90% by AO vs. < 60% using DS).¹ Although acids and halides increase selectivity, they also lead to dissolution of Pd from the surface which complexes with the anionic species in solution to form homogeneous Pd²⁺ complexes (e.g., PdCl₄^2-) that may catalyze H₂O₂ formation or decomposition, and contribute to the reported changes in selectivity.^2,3 Consequently, it is unclear if the active sites for direct synthesis are homogeneous Pd complexes, halide-modified Pd nanoparticles, or if these species are somehow linked within a single catalytic cycle. Kinetic measurements need to be combined with spectroscopic tools to probe the relationships between rates, selectivities, and the populations of distinct homogeneous and heterogeneous Pd species to elucidate the true nature of active catalytic species in direct synthesis.

Here, we combine transient and steady-state measurements of H₂O₂ and H₂O formation rates with in situ UV-Vis spectroscopy to probe the identity of active catalytic species that form H₂O₂. Pd atoms exist in equilibrium between a heterogeneous colloidal state (Pd⁰) and a collection of homogeneous Pd²⁺ complexes due to the presence of both reducing (H₂) and oxidizing (e.g., O₂, Cl^-) environment in the solvent. Reactions are conducted in a semi-batch reactor by introducing Pd salts (e.g., 0.01- 0.05 mmol of PdCl₂, PdSO₄, Pd(NO₃)₂) into methanol (140 mL) saturated with H₂ and O₂ reactants (4.8 kPa H₂, 4.8 kPa O₂, 278 K). In situ UV-Vis spectra show that at steady state approximately 10 - 20% of the total Pd exist in the Pd²⁺ form while the remaining Pd atoms exist in the Pd⁰ state at these conditions. After the reaction attains steady state, an excess of HCl (10 mmoles) is injected into the solvent which increases the concentration of Pd²⁺ species to 25 – 50 %. We also observe an increase in the selectivity to H₂O₂ accompanied by a decrease in the rate of formation and rate of hydrogenation of H₂O₂. A second HCl injection increases the population of the homogeneous Pd²⁺ complexes to ~ 40-80% but the rates of formation and hydrogenation as well as selectivity are unaffected on subsequent HCl injection. The rate constants for H₂O₂ formation (k_formation) and H₂O₂ hydrogenation (k_{hydrogenation}) are calculated by fitting the rate equation for net H₂O₂ formation to the H₂O₂ concentration as a function of time.⁴ The ratios of k_formation and k_{hydrogenation} increase two- to six- fold after the first HCl injection for each salt tested, but remain constant after the second injection. H₂ conversions decrease after each HCl addition which alludes to a loss of active sites on introduction of HCl. These observations are incompatible with soluble Pd²⁺ species acting as the active catalytic species for the formation of H₂O₂. Rather, the colloidal Pd nanoparticles produce H₂O₂. The Cl^- ions modify the surface of the colloidal Pd⁰ species and these chloride-modified nanoparticles inhibit the H₂O₂ hydrogenation after the first HCl injection which increases the k_formation/k_{hydrogenation} ratio and selectivity towards H₂O₂ after the first HCl spike. The decrease in the H₂O₂ formation rates occurs due to the reduction in number of active colloidal Pd sites due to partial conversion to Pd²⁺. Adding additional Cl^- ions after the first injection does not have any effect other than changing the distribution of Pd²⁺ and Pd⁰ species because the surface of the colloids are already saturated with Cl^- following the first HCl injection. The k_formation/k_{hydrogenation} ratios are also different for each Pd salt prior to adding HCl, which shows that the corresponding anions (Cl^-, SO₄^2-, NO₃^-) also modify the surface of active Pd colloids altering the activity and selectivity of the process.

Having established the true nature of active catalytic species in direct synthesis, we will attempt to understand how these acid and halide promoters are affecting the selectivity of direct synthesis by conducting activation enthalpy measurements using commonly used additives like HCl, HBr and H₂SO₄ on supported Pd nanoparticles. These promoters possibly change the enthalpies of formation of H₂O₂ and H₂O and the binding energy of oxygen to the nanoparticles due to which the H₂O₂ selectivity changes. Comparison of activation enthalpies and selectivities will help us understand the role played by these commonly used additives in improving the H₂O₂ selectivity.

(1) Wilson, N. M.; Bregante, D. T.; Priyadarshini, P.; Flaherty, D. W. In Catalysis; Spivey, J., Han, Y.-F., Eds. 2017; Vol. 29, p 122.

(2) Dissanayake, D. P.; Lunsford, J. H Journal of Catalysis 2003, 214, 113.

(3) Dissanayake, D. P.; Lunsford, J. H. Journal of Catalysis 2002, 206, 173.

(4) Wilson, N. M.; Priyadarshini, P.; Kunz, S.; Flaherty, D. W. Journal of Catalysis 2018, 357, 163.