Solvent Additive-Induced Deactivation of the Cu-ZnO(Al2O3)-Catalyzed γ-Butyrolactone Hydrogenolysis: A Rare Deactivation Process

Abstract

This work reports initial results on the effect of low concentrations (ppm level) of a stabilizing agent (2,6-di-tert-butyl-4-methylphenol, BHT) present in an off-the-shelf solvent on the catalyst performance for the hydrogenolysis of γ-butyrolactone over Cu-ZnO-based catalysts. Tetrahydrofuran (THF) was employed as an alternative solvent in the hydrogenolysis of γ-butyrolactone. It was found that the Cu-ZnO catalyst performance using a reference solvent (1,4-dioxane) was good, meaning that the equilibrium conversion was achieved in 240 min, while a zero conversion was found when employing tetrahydrofuran. The deactivation was studied in more detail, arriving at the preliminary conclusion that one phenomenon seems to play a role: the poisoning effect of a solvent additive present at the ppm level (BHT) that appears to inhibit the reaction completely over a Cu-ZnO catalyst. The BHT effect was also visible over a commercial Cu-ZnO-MgO-Al₂O₃ catalyst but less severe than that over the Cu-ZnO catalyst. Hence, the commercial catalyst is more tolerant to the solvent additive, probably due to the higher surface area. The study illustrates the importance of solvent choice and purification for applications such as three-phase-catalyzed reactions to achieve optimal performance.

Figure 1

Characterization of the calcined Cu₇₀Zn₃₀ oxide catalyst precursor. (A) Nitrogen adsorption isotherms (−196 °C). (B) BJH pore size distribution derived from the N₂ sorption data. (C) XRD diffraction pattern. (D) Temperature-programmed reduction.

Figure 2

Binary Cu–ZnO catalyst performance. Hydrogenation of γ-butyrolactone over Cu–ZnO. (Left) Concentration profiles of the reactants and products as a function of time: γ-butyrolactone (filled circles), 1,4-butanediol (empty circles), and carbon balance (filled squares). Reaction conditions are given in Table S4. Solvents: (A) dioxane and (B) THF (type A). (Right) Appearance of the catalyst after reaction. The physical appearance of the catalyst under dioxane looked normal, a black-colored material. For the THF’s catalyst residue (THF, type A), the color indicated that the Cu catalyst might have sintered. The green-colored residue was unfamiliar to us. In panel (B, right), the solvent was removed and the catalyst was dried to better observe its appearance after reaction. GBL, γ-butyrolactone; BDO, 1,4-butanediol.

Figure 3

Hydrogen solubility in different solvents (THF, tetrahydrofuran; Dio, dioxane) as a function of pressure at 180 °C, calculated by the UNIQUAC contribution thermodynamic gas/liquid binary equilibrium method (Aspen Plus, Aspen Technology, Inc.).

Figure 4

Proposed mechanically induced sintering mechanism under turbulent stirring conditions for a Cu–ZnO catalyst during the liquid-phase hydrogenolysis of GBL under THF as solvent. GBL, γ-butyrolactone; THF, tetrahydrofuran.

Figure 5

Binary Cu–ZnO catalyst performance. Hydrogenation of γ-butyrolactone over Cu–ZnO. Concentration profiles of the reactants and products as a function of time: γ-butyrolactone (filled circles), 1,4-butanediol (empty circles), and carbon balance (filled squares). Reaction conditions are given in Table S4. Solvents: (A) BHT-containing THF (type A) and (B) BHT-free THF (type B). GBL, γ-butyrolactone; BDO, 1,4-butanediol; and BHT, 2,6-di-tert-butyl-4-methylphenol.

Figure 6

Active site model and reaction mechanism for the ring-opening hydrogenolysis of γ-butyrolactone to 1,4-butanediol over Cu–ZnO. Adapted with permission from Hamminga et al.

Figure 7

XRD diffraction patterns for the commercial Cu(O)–ZnO–MgO–Al₂O₃ catalysts, including the as-received oxidic material (fresh) and the after-reaction catalysts. The patterns of two CuO–ZnO reference compounds were included to help in the identification since the commercial catalyst contains broad reflections: (α) Cu₅Zn₉₅ and (β) Cu₅₀Zn₅₀, where the subscripts represent the relative mole composition. Both were prepared using the method described in Section 2.2.

Figure 8

Performance of a commercial Cu–ZnO–MgO–Al₂O₃ catalyst in the hydrogenation of γ-butyrolactone using (A) THF as a solvent, with and without BHT as an additive, and (B) using 1,4-dioxane as a solvent, with and without BHT as additive. For THF, BHT comes as an additive in one of the commercial grades, whereas for 1,4-dioxane, the BHT was added by us to reach the same concentration, 250 ppm. For the experiments containing BHT, the catalyst was reused in a second cycle using fresh reagents after in situ reduction. The concentration profiles can be found in Figure S7. GBL, γ-butyrolactone; THF, tetrahydrofuran; BHT, 2,6-di-tert-butyl-4-methylphenol.

Figure 9

XPS spectra of the (1) Cu 2p_3/2, (2) Zn 2p_3/2, and (3) Al 2s core level regions for the various Cu(O)–ZnO–MgO–Al₂O₃ materials: (A) fresh commercial unreduced catalyst; (B) after-reaction, pure THF; (C) after-reaction, pure dioxane; (D) after-reaction, BHT-containing THF, and (E) after-reaction, BHT-containing dioxane. THF, tetrahydrofuran; BHT, 2,6-di-tert-butyl-4-methylphenol.