Performance structure and mechanism of Pd Ag alloy catalyst for selective

Journal of Catalysis 300 (2013) 205–216 Contents lists available at SciVerse ScienceDirect Journal of Catalysis journal homepage: www.elsevier.com/locate/jcat Performance, structure and mechanism of Pd–Ag alloy catalyst for selective oxidation of glycerol to dihydroxyacetone Shota Hirasawa a, Hideo Watanabe b, Tokushi Kizuka b, Yoshinao Nakagawa a,⇑, Keiichi Tomishige a,⇑ a b Department of Applied Chemistry, School of Engineering, Tohoku University, 6-6-07 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan Faculty of Pure and Applied Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8573, Japan a r t i c l e i n f o Article history: Received 29 November 2012 Revised 15 January 2013 Accepted 17 January 2013 Available online 23 January 2013 Keywords: Biomass Palladium Silver Oxidation Glycerol a b s t r a c t The mechanism of the oxidation of glycerol to dihydroxyacetone over Pd–Ag catalysts was discussed Characterization results suggest that the metal composition of the surface of the crystalline Pd–Ag alloy particles is almost the same as the bulk composition The synergetic effects of Pd and Ag appear on the oxidation of the secondary OH group of vic-diols The reaction order with respect to glycerol concentration over Pd–Ag/C was zero, suggesting the strong interaction between glycerol and the catalyst surface The reaction order with respect to O2 pressure was 0.4, suggesting that the rate-determining step is the reaction involving oxygen species These reaction trends and characterization results support the mechanism where the terminal OH group of glycerol is adsorbed on the Ag site and the neighboring secondary OH group (CH–OH) is attacked by the oxygen species dissociatively adsorbed on the Pd site Ó 2013 Elsevier Inc All rights reserved Introduction Conventional resources, mainly fossil fuels, are becoming limited because of the rapid increase in energy demand Biomass is one of the renewable resources and can be used to produce various chemicals and fuels like petroleum In contrast to petrorefinery, biorefinery is a facility that integrates biomass conversion processes and equipment to produce fuels, power, and chemicals By utilizing new chemical, biological and mechanical technologies, the biorefinery will replace the present petroleum-based energy systems with greener and more sustainable ones [1,2] Glycerol is a particularly important building block for biorefinery because the expanding biodiesel production gives large amounts of inexpensive glycerol as a by-product [3] Therefore, new efficient procedure including oxidation, hydrogenolysis, dehydration, esterification, transesterification, and polymerization for the transformation of glycerol into valuable chemicals are highly desired [4–6] In particular, oxidation with molecular oxygen is an attractive way of conducting the transformation since molecular oxygen is inexpensive and readily available and ultimately co-produces only a benign by-product (H2O) The oxidation of glycerol leads to a large number of products such as dihydroxyacetone (DHA), glyceric acid (GLYAC), glyceraldehyde (GLYALD), hydroxypyruvic acid (HYPAC), and glycolic acid ⇑ Corresponding authors Fax: +81 22 795 7214 E-mail addresses: yoshinao@erec.che.tohoku.ac.jp (Y Nakagawa), tomi@erec.che.tohoku.ac.jp (K Tomishige) 0021-9517/$ - see front matter Ó 2013 Elsevier Inc All rights reserved http://dx.doi.org/10.1016/j.jcat.2013.01.014 (GLYCAC) Therefore, control of reaction selectivity is necessary to obtain the desired compound GLYAC is an important intermediate for more deeply oxidized products such as tartronic acid and mesoxalic acid The production of GLYAC by catalytic aerobic oxidation of glycerol has been intensively investigated using monometallic or bimetallic catalysts such as Au, Pt, and Pd in a basic medium [7–17] DHA is another desirable product because of its application in cosmetics and in the fine chemicals industry [18] DHA is currently produced from the microbial fermentation of glycerol by Gluconobacter suboxydans However, the productivity is low because of the low substrate concentration and the long fermentation time [19] The direct catalytic oxidation of glycerol to DHA has been investigated for a few decades, and Pt–Bi catalysts have been mentioned as selective ones for DHA formation (Table 1) [7,20–24] In the recent publications, the initial selectivity is about 80% and the DHA yield is about 50% (at 80% glycerol conversion) over Pt– Bi/C using batch reactor Nie et al succeeded in the synthesis of DHA using Pt–Sb catalysts in 2012, while the DHA yield was comparable to those using Pt–Bi catalysts [25] Recently, we reported that Pd–Ag alloy catalyst is very effective for selective oxidation of glycerol to DHA under neutral conditions [26] The physical mixture of Pd/C and Ag/C showed almost no enhancing effect on the DHA selectivity and activity, showing the synergetic effects in Pd–Ag alloy catalysts [26] Over the Pd–Ag/C catalyst, the high selectivity to DHA (85%) was maintained in the longer reaction time, which is in contrast to the case of the Pt–Bi system where DHA selectivity decreased at longer reaction time One problem in the Pd–Ag/C system is that the reaction stops before the total 206 S Hirasawa et al / Journal of Catalysis 300 (2013) 205–216 Table Some typical results of catalytic oxidation of glycerol to dihydroxyacetone investigated by several groups Entry Catalyst Glycerol/(Pt or Pd) molar ratio Solvent Oxidant Temp (K) Time (h) Conversion (%) DHA selectivity (%) Ref 2%Pd–2%Ag/C 5%Pt–1%Bi/C 7.4%Pt–2.9%Bi/C 5%Pt–5.4%Bi/C 3%Pt–0.6%Bi/C 5%Pt–5%Bi/C 3.8%Pt–3.0%Sb/MWCNT 115 843 – 1755 117 424 558 H2O H2O H2O H2O H2O H2O H2O O2 (0.3 MPa) air (0.1 MPa) air (0.1 MPa) O2 (0.1 MPa) O2 (0.3 MPa) O2 (150 ml/min) O2 (150 ml/min) 353 323 333 393 353 333 333 24 – 52 30 75 20 80 92 90 85 66 49 51 60 49 51 [26] [20] [7] [22] [23] [24] [25] conversion of glycerol probably because of the poisoning by the produced acid For the improvement of DHA yield, the elucidation of reaction mechanism and mechanism of inhibition by by-products is necessary In this article, we discussed the reaction mechanism over Pd–Ag catalysts based on the performance and structure of catalysts with various types of supports and different amount of Ag as well as reaction kinetics and the reactivity trends of various alcohols In addition, the deactivation of Pd–Ag/C catalyst was investigated in terms of the types of poisonous compounds and the adsorption site Experimental 2.1 Catalyst preparation The carbon black (Vulcan-XC72, Brunauer–Emmett–Teller (BET) surface area 254 m2/g) supplied by Cabot Corporation Ltd., the SiO2 (G-6, BET surface area 535 m2/g) supplied by Fuji Silysia Chemical Ltd., the TiO2 (P-25, BET surface area 50 m2/g) supplied by Nippon Aerosil Co., Ltd., the a-Al2O3 prepared by the calcination of c-Al2O3 (KHO-24, BET surface area 133 m2/g) supplied by Sumitomo Chemical Co., Ltd., the ZrO2 (BET surface area 15 m2/g) supplied by Soekawa Chemical Co., Ltd and the CeO2 (HS, BET surface area 77 m2/g) supplied by Daiichi Kigenso Kagaku Kogyo Co., Ltd were used as supports of the catalysts Pd/C and Ag/C catalysts were prepared by impregnating the carbon with an aqueous solution of Pd(NO3)2 (N.E Chemcat Co., Ltd.) and AgNO3 (Wako Pure Chemical Industries, Ltd.), respectively The Pd–M/C (M = Ti, Mn, Ni, Ag, Re, Ir, Au, and Bi) were prepared by co-impregnation method using mixed aqueous solutions The precursors used were Pd(NO3)2, TiCl3 (Wako Pure Chemical Industries, Ltd.), Mn(NO3)2Á6H2O (Wako Pure Chemical Industries, Ltd.), Ni(NO3)2Á6H2O (Wako Pure Chemical Industries, Ltd.), AgNO3, NH4ReO4 (Soekawa Chemical Co., Ltd.), H2IrCl6 (Furuya Metals Co., Ltd.), HAuCl4Á4H2O (Soekawa Chemical Co., Ltd.), and Bi(NO3)3Á5H2O (Wako Pure Chemical Industries, Ltd.) The loading amount of Pd was wt.% and that of secondary metal was represented by the molar ratio of the additive to Pd (typically 1) After evaporating the solvent and drying at 383 K for 12 h, Pd–Ag/SiO2, Pd–Ag/TiO2, Pd–Ag/Al2O3, Pd–Ag/ZrO2, and Pd–Ag/CeO2 were calcined in air at 573 K for h and were reduced under H2 flowing at 473 K for h, followed by passivation with O2/ He (2%) at room temperature Pd–Ag/C was directly reduced under H2 flowing at 473 K for h, followed by passivation with O2/He (2%) at room temperature All catalysts were in powdery form, with less than 0.1 mm Conversion ð%Þ ¼ P 2.2 Activity tests The pretreatment of C-supported catalyst was performed in a glass test tube (70 ml) equipped with magnetic stirrer and balloon filled with oxygen [26] An aqueous solution of glycerol (5 wt.%, 20 ml) and the catalyst (0.050 g) were put into the reactor The reactor was depressurized and filled with oxygen The reactor was then heated to 353 K During the experiment, the stirring rate was fixed at 500 rpm (magnetic stirring) After h, the reactor was cooled down The catalyst was collected by centrifugation The wet catalyst was used immediately after the pretreatment Activity test was performed in a 190-ml stainless steel autoclave with an inserted glass vessel The catalyst with or without pretreatment was put into an autoclave together with a spinner and a 20 ml aqueous solution of glycerol After sealing, the reactor was filled with 0.3 MPa oxygen The autoclave was then heated to 353 K, and the temperature was monitored using a thermocouple inserted in the autoclave During the experiment, the stirring rate was fixed at 500 rpm (magnetic stirring) After an appropriate reaction time, the reactor was cooled down The autoclave contents were transferred to a vial, and the catalyst was separated by filtration The standard conditions for the reaction were as follows: 353 K reaction temperature, 0.3 MPa initial oxygen pressure, h reaction time, g glycerol, 19 g water, and 50 mg supported metal catalyst The parameters were changed appropriately in order to investigate the effect of reaction conditions Details of the reaction conditions are described in each result The products were analyzed using high-performance liquid chromatography (Shimadzu Prominence) equipped with UV–VIS detectors and refractive index detector An Aminex HPX-87H column (Bio-Rad) was used for separation Diluted H2SO4 (0.01 M) was employed as eluent For the analysis, the samples were diluted with the eluent A measuring time of 25 min, a column temperature of 333 K, and a flow of 0.6 ml minÀ1 were applied The chemical standards we have used for the identification of the possible products were dihydroxyacetone (90–100% purity, MP Biomedicals), DL-glyceraldehyde (97.0+% purity, Wako Pure Chemical Industries, Ltd.), DL-glyceric acid (40% in water, Tokyo Chemical Industry Co Ltd.), b-hydroxypyruvic acid (95.0+% purity, Sigma Aldrich), tartronic acid (98% purity, Alfa Aesar), calcium mesoxalate trihydrate (98+% purity, Tokyo Chemical Industry Co Ltd.), glycolic acid (97.0+% purity, Wako Pure Chemical Industries, Ltd.), glyoxylic acid monohydrate (95.0+% purity, Wako Pure Chemical Industries, Ltd.), oxalic acid (98.0+% purity, Wako Pure Chemical Industries, Ltd.), and formic acid (98.0+% purity, Wako Pure Chemical Industries, Ltd.) The HPLC charts of the standards are shown in Supplementary Material Fig S1 The conversion and the selectivity were defined on the carbon basis and defined as follows: P fðmol of productÞðnumber of carbon atoms in the product moleculeÞg  100 fðmol of product of glycerolÞðnumber of carbon atoms in a product or glycerol moleculeÞg fðmol of productÞðnumber of carbon atoms in a product moleculeÞg Selectivity ð%Þ ¼ P  100 fðmol of product Þðnumber carbon atoms in a product moleculeÞg 207 S Hirasawa et al / Journal of Catalysis 300 (2013) 205–216 Turnover number (TON) and turnover frequency (TOF) were calculated based on the number of surface noble metal atoms determined by CO adsorption: TON = {sum of product amount (mol)}/{adsorption of CO (mol)} TOF (hÀ1) = {sum of product amount (mol)}/[{adsorption of CO (mol)}{reaction time (h)}] The mass balance was also confirmed in each result, and the difference in mass balance was always in the range of the experimental error The agreement in terms of the mass balance indicated that polymeric by-products were not formed (±10%) The used catalyst was collected by centrifugation A slight loss ( SiO2 > Al2O3 > ZrO2 > TiO2 > CeO2 In terms of TON, carbon, SiO2, and Al2O3 supports gave high activity However, the reaction rate over Pd–Ag/ Al2O3 rapidly decreased Therefore, we focused on Pd–Ag/SiO2 as well as Pd–Ag/C in the following studies The effect of Ag loading on the catalytic activity was examined on the Pd–Ag/SiO2 The amount of Ag was varied from 0.5 to in the molar ratio of Ag to Pd (entries 3–5) The catalysts with more Ag tend to be more selective for DHA formation, and this trend was also observed for Pd–Ag/C [26] On the other hand, the catalysts with more Pd tend to be more active for glycerol oxidation, and this monotonous trend was different from that of Pd–Ag/C where the catalyst with Ag/Pd = was most active The reaction time dependences of the aerobic glycerol oxidation over Pd–Ag/SiO2 and pretreated Pd–Ag/C (Ag/Pd = 1) were compared (Fig 1) Both SiO2- and C-supported catalysts showed high selectivity to DHA The initial reaction rate over Pd–Ag/SiO2 was comparable to that over Pd–Ag/C Unlike in the case of Pd–Ag/C which has an induction period, it is speculated that the active structure of Pd–Ag/SiO2 is easily Table Comparison of the catalytic performances in the glycerol oxidation over modified Pd/C (2 wt.% Pd, secondary metal/Pd = 1) Entry 10 Catalyst Pd/C Pd–Ti/C Pd–Mn/C Pd–Ni/C Pd–Ag/C Pd–Re/C Pd–Ir/C Pd–Au/C Pd–Bi/C Blank CO uptake amount (lmol/g) 50.1 60.7 51.3 –a 16.2 94.3 70.5 77.8 2.3 Conversion (%) 2.8 2.1 3.2 6.3 9.5 7.4 4.3 6.1 1.8 0.0 Selectivity (%) TON DHA GLYALD GLYAC HYPAC GLYCAC 66.1 44.8 64.9 45.6 81.9 47.1 52.3 47.8 54.1 22.5 32.2 22.6 23.5 8.9 21.8 24.2 20.9 33.1 10.9 22.1 12.0 26.6 5.6 25.5 21.5 25.5 10.1