Theoretical and experimental study of glycolaldehyde reaction pathways on Ni/Pt(111) and Ni/WC surfaces
Conference Paper
Overview
Identity
View All
Overview
abstract
Global energydemand growth and green house effect are two of the main drivers for thedevelopment of replacing fossil fuels. Biomass-derived molecules are apromising class of alternative resources because they offer the advantages ofbeing widely available, renewable, and potentially carbon-neutral. Suchmolecules generally contain more oxygen atoms than are found in petroleum-basedfeedstocks. Previously, small alcohols and polyols were used asrepresentatives of oxygenates to be studied with gradually increasing thecomplexity of the molecular structure. The result revealed the possibility ofoxygenates reforming to H 2 and CO (syngas) by selectivelycontrolling the C-H, C-C, C-O, and O-H bond scissions. In this work,glycolaldehyde (HOCH 2CH=O), which contains both OH and C=Ofunctionalities similar to many biomass derived molecules, was studied as theprobe molecule for biomass conversion to syngas. The current study of glycolaldehyde activity utilized density functional theory (DFT) prediction, aswell as experimental verification using temperature programmed desorption (TPD)and high resolution electron energy loss spectroscopy (HREELS). Bimetallic catalysts are known to often exhibit uniqueproperties different from either of the parent metals, and show potential forbiomass conversion. The Ni/Pt(111) bimetallic system has been extensivelyinvestigated and was therefore studied for glycolaldehyde reactions. As established in previous studies ofalcohols and polyols, enhanced catalytic conversion of these molecules onthe bimetallic surfaces could be correlated to the binding energies and thesurface d-band center with respect to the Fermi level. The binding energy ofglycolaldehyde was found to increase as the surface d-band center approachedthe Fermi level, with the NiPtPt(111) configuration exhibiting the highestbinding energy and thus predicted to present the highest activity. In order to verify the DFT prediction, glycolaldehyde TPD experiments were performed on Pt(111), NiPtPt(111), PtNiPt(111) surfacesand a thick Ni(111) film. H 2 and CO were observed as the reaction products, which confirmed the prediction to produce syngas from glycolaldehyde. The activity of glycolaldehyde on each surface was quantified from the TPD peak areas. Among the four surfaces, the NiPtPt(111) surface showed the highest activity, consistent with the DFT prediction. HREELS experiments ofglycolaldehyde were employed to show the intermediates in the reactions. At 300K, the C-C peak disappeared and a CO peak was observed, demonstrating the C-Cbond cleavage to produce CO, which was consistent with the TPD result. However, the favorable NiPtPt(111) bimetallic structure is notstable at high temperature; the top monolayer Ni atoms tends to diffuseinto the Pt bulk and stay beneath of the Pt surface. The resulting PtNiPt(111)subsurface configuration showed significantly lower reforming activity. Since tungstenmonocarbide (WC) has been shown to possess similar electronic propertiesto Pt(111), Ni-modified WC surfaces were proposed to replace NiPtPt(111) in theglycolaldehyde study. Parallel DFT glycolaldehyde binding energy on theone-monolayer Ni-modified WC (1ML NiWC) surface was calculated and found to besimilar to that on NiPtPt(111). TPD experiments of glycolaldehyde were performedon WC and Ni-modified WC surfaces. On clean WC surface, ethylene was observedas the product, which was consistent with the HREELS result. On Ni-modified WCsurface, H 2 and CO were produced, exhibiting the same chemistry with NiPtPt(111) configuration. A similar glycolaldehyde reforming activity was alsofound on the 1ML NiWC surface after the quantification of reaction activity. Theseresults suggested that Ni monolayer catalysts supported on WC may be preferableto Ni/Pt bimetallics as active and selective catalysts for biomass reformingwith higher stability and lower cost.
