Isotope Effects in Methanol Synthesis and the Reactivity of Copper Formates on a Cu/SiO2 Catalyst

Abstract

Here we investigate isotope effects on the catalytic methanol synthesis reaction and the reactivity of copper-bound formate species in CO 2 -H 2 atmospheres on Cu/SiO 2 catalysts by simultaneous IR and MS measurements, both steady-state and transient. Studies of isotopic variants (H/D, 12 C/ 13 C) reveal that bidentate formate dominates the copper surface at steady state. The steadystate formate coverages of HCOO (in 6 bar 3:1 H 2 :CO 2 ) and DCOO (in D 2 :CO 2 ) are similar and the steady-state formate coverages in both systems decrease by *80% from 350 K to 550 K. Over the temperature range 413 K-553 K, the steady-state methanol synthesis rate shows a weak H/D isotope effect (1.05 ± 0.05) with somewhat higher activation energies in H 2 :CO 2 (79 kJ/mole) than D 2 :CO 2 (71 kJ/mole) over the range 473 K-553 K. The reverse water gas shift (RWGS) rates are higher than methanol synthesis and also shows a weak positive H/D isotope effect with higher activation energy for H 2 /CO 2 than D 2 /CO 2 (108 vs. and 102 kJ/mole) The reactivity of the resulting formate species in 6 bar H 2 , 6 bar D 2 and 6 bar Ar is strongly dominated by decomposition back to CO 2 and H 2 . H 2 and D 2 exposure compared to Ar do not enhance the formate decomposition rate. The decomposition profiles on the supported catalyst deviate from first order decay, indicating distributed surface reactivity. The average decomposition rates are similar to values previously reported on single crystals. The average activation energies for formate decomposition are 90 ± 17 kJ/mole for HCOO and 119 ± 11 kJ/mole for DCOO. By contrast to the catalytic reaction rates, the formate decomposition rate shows a strong H/D kinetic isotope effect (H/D *8 at 413 K), similar to previously observed values on Cu(110).

Simultaneous MS-IR Studies of Surface Formate Reactivity Under Methanol Synthesis Conditions on Cu/SiO2

“…One significant difference is the previous report of a hydrogen effect on formate decomposition on Cu/SiO 2 [20], a result not observed in this study. We have previously reported an H/D isotope effect on Cu/SiO 2 [25]. The rates of DCOO decomposition in this study fall on the low side of the band because of this kinetic isotope effect.…”

“…More specifically, any surface precursors are either (1) exchanged very rapidly with the gas phase or with formate itself or (2) present in such small quantities that their effects on the surface residence time of (the carbon in the) formate is inconsequential [32]. A minor deviation from first order decay is seen in the Ar:D 2 transient, similar to that discussed in reference [25], where the change in formate reactivity during the decay was suggested to be due to the surface heterogeneity of the Cu sites on supported catalyst. Changes in composition of the coadsorbate population during the decay can also contribute to this aspect.…”

“…D 2 rather than H 2 was used to avoid the mass spectral fragment overlaps in the H 2 system. As previously described [25], the only identifiable IR features which appear under reaction conditions (other than gas phase features of carbon monoxide, carbon dioxide and water) are assigned to copper-bound bidentate formate. In this study we used the narrow features at 1321 and 1307 cm After steady state catalytic behavior is reached, an isotopic transient (SSITKA) experiment is performed by switching the gas to D 2 ?…”

“…These studies indicate a saturation formate coverage of 0.25 on Cu(110) [30], Cu(111) [31] and Cu/SiO 2 [20]. Our measured IR signals, which decrease with temperature under reaction conditions [25] were used to scale the coverage. Figure 1 shows selected IR spectra of the -OC(D)Ostretch region at various times during the experimental sequence described above.…”

“…Chorkendorff and co-workers documented the removal of formate from Cu(100) in 5.8 bar hydrogen [5] with relatively high rates. Our previous work showed that over the temperature range 100-240°C, H 2 at pressures up to 6 bar has little or no effect on the lifetime of formate on a Cu/SiO 2 catalyst [25]. We also showed that little or no methanol is produced during the decomposition of formate under inert gas and/or hydrogen containing atmospheres [26].…”