Conversion of Levulinic Acid to Î³-Valerolactone over Few-Layer

Subscriber access provided by KUNGL TEKNISKA HOGSKOLAN

Article

Conversion of levulinic acid to gamma-valerolactone over few-layer graphene-supported ruthenium catalysts Chaoxian Xiao, Tian-Wei Goh, Zhiyuan Qi, Shannon Goes, Kyle Brashler, Christopher Perez, and Wenyu Huang ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b02673 • Publication Date (Web): 09 Dec 2015 Downloaded from http://pubs.acs.org on December 10, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Conversion of levulinic acid to gammavalerolactone over few-layer graphene-supported ruthenium catalysts Chaoxian Xiao,1 Tian-Wei Goh,1 Zhiyuan Qi,1 Shannon Goes,1 Kyle Brashler,1 Christopher Perez,2 and Wenyu Huang1,2,* 1

Department of Chemistry, Iowa State University, Ames, Iowa 50011, United States 2

Ames Laboratory, US Department of Energy, Ames, IA 50011, United States

ABSTRACT: Few-layer graphene (FLG) supported ruthenium nanoparticle catalysts were synthesized and used for the hydrogenation of levulinic acid (LA), one of the “Top 10” biomass platfrom molecules derived from carbohydrates. FLG supported ruthenium catalyst shows 99.7% conversion and 100% selectivity towards gamma-valerolactone (GVL) at room temperature in a batch reactor under high-pressure hydrogen. This catalyst showed four times higher activity and exceptional stability in comparison with traditional activated carbon supported ruthenium catalysts (Ru/C). X-ray photoelectron spectroscopy (XPS) and Fourier transform infrared spectroscopy (FTIR) studies suggest that the superior catalytic properties of Ru nanoparticles supported on FLG in LA hydrogenation could be attributed to more metallic Ru content present in the Ru/FLG than that in Ru/C.

ACS Paragon Plus Environment

1

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

KEYWORDS:

Biomass

conversion;

cellulose;

Page 2 of 23

levulinic

acid;

gamma-valerolactone;

hydrocarbon fuel; hydrogenation; graphene. 1. Introduction Depleting fossil resources and increasing concerns over greenhouse gas emissions have led to global interests in producing fuels and value-added chemicals from renewable biomass feedstocks. Cellulose is the dominant component in various biomass sources, such as municipal solid waste, paper pulp sludge, sewage waste, animal manure, agricultural biomass, and forest waste. Therefore, the conversion of cellulose and its derivatives has drawn intense studies in biomass conversion. Levulinic acid (LA), recognized as one of the “Top 10” most promising platform molecules derived from biomass by US Department of Energy, is considered to be a very promising candidate because of its easy production.1,2 LA can be produced in high yield (up to ~80%) from cellulosic feedstocks using 15wt.% sulfuric acid as the catalyst.3 The commercialization of producing LA has been announced by several plants located at USA (Segetis) and Italy (GFBiochemicals). However, the needs to use lime to neutralize acid in the product solution and the energy intensive distillation for the separation of LA increased production costs. Recently, several groups made significant progress to address this issue by replacing homogeneous mineral acid with solid acid catalysts.4-8 Additionally, Gurbuz et al. reported that LA can be extracted directly from aqueous solution using alkylphenol solvents to form a biphasic system.9 LA can also be separated by forming ethyl levulinate via esterification with ethanol in the presence of acid catalysts.10 The hydrogenation and subsequent dehydration of LA produces gamma-valerolactone (GVL) as the major product. GVL is a versatile chemical, which has been used as a high-quality solvent, food and fuel additive.11-13 Notably, it can be used as the feedstock to produce hydrocarbon


2

Page 3 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

fuels.13 Bond and co-workers had developed an integrated process combining the conversion of GVL to butene via decarboxylation over SiO2/Al2O3 catalyst with the subsequent oligomerization of butene over HZSM-5 or Amberlyst 70 catalysts. The final products consist of mainly C8, C12 and C16 olefins, which are good candidates for gasoline and jet fuels.13 SerranoRuiz et al. studied the conversion of GVL over water-stable Pd/Nb2O5 bifunctional catalyst. The products contain pentanoic acid and 5-nonanone, which can serve as a source of chemicals or be further upgraded to hydrocarbon fuels via hydrodeoxygenation.14 The catalysts for the hydrogenation and dehydration of LA mainly include heterogeneous catalysts,15 such as Raney Ni,16 Cu-Al2O3,17 Cu-ZrO2,17 CuO-Cr2O3,15,18 Pt/C,19 PtO2,16 Pd/C,19 Ru/C,20,21 Ru/SiO2,22 Ru/TiO2,23 RuSn/C,24 and Mo2C.25 In 1930, Schuette and Thomas reported the production of GVL by reducing LA with hydrogen at room temperature using PtO 2 as the catalyst.16 The best yield of GVL could reach 87%. Among all the catalysts studied since then, Ru/C was found to be the most effective for this reaction. Al-Shaal et al. reported that 100% conversion and 97.5% selectivity to GVL could be achieved over traditional activated carbon supported ruthenium catalyst (Ru/C) at room temperature, representing the best result obtained so far.26 However, they also found that the Ru/C catalyst suffered from continuous deactivation during recycling, with the yield decreasing from 30% in the 1st cycle to approximately 10% at 4th cycle. Therefore, designing catalysts with enhanced stability during recycling is highly demanded. Our strategy is to employ a new type of carbon material, few-layer graphene (FLG), as the support material instead of activated carbon to synthesize Ru catalysts. Graphene is a single layer of sp2-bonded carbon atoms arranged in a honeycomb-like lattice. It could be prepared by various methods such as exfoliating graphite with physical or chemical means,27 chemical vapor


3

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 23

deposition (CVD) of hydrocarbons,28 epitaxial growth on suitable substrates,29 or carbonization of carbonaceous compounds.30 Because single layer graphene can re-stack together to form more layers, FLG is commonly obtained after synthesis.27 Graphene has received considerable interests as a promising catalyst support owing to its low cost, high surface area, excellent thermal and chemical stability, and rich surface chemistry.31 It has been reported that graphene or graphene supported metal/metal oxide composites could be used as excellent catalysts for a number of catalytic applications including hydrolysis,32 photocatalysis,33 electrocatalysis,34 Fischer-Tropsch synthesis,35 Suzuki coupling reaction,36 etc. Herein, we utilized a polyol approach to synthesize Ru nanoparticles, and then loaded them onto FLG to make supported Ru catalysts. The FLG supported Ru catalysts showed high activity and selectivity in the hydrogenation and dehydration of LA in the presence of molecular hydrogen. Since the concentration of LA produced from raw biomass is only 5wt.%, we performed most of the LA hydrogenation reaction in an aqueous solution only containing 5wt.% LA to demonstrate that no pre-concentration is necessary before the LA to GVL conversion.

2. Results and discussion 2.1 Synthesis and characterizations of the catalysts FLG-supported Ru nanoparticle catalysts (Ru/FLG) were prepared with a bottom-up approach (Experimental details in Supporting Information). Ru nanoparticles were synthesized by reducing RuCl3 precursor in ethylene glycol at 160C in the presence of sodium hydroxide and polyvinylpyrrolidone (PVP). As-synthesized Ru nanoparticles were isolated from ethylene glycol after adding excess acetone, purified by washing with ethanol/hexanes, and then re-dispersed in ethanol to form a stable solution. Subsequently, FLG was added in the above solution to allow


4

Page 5 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Ru nanoparticles to be adsorbed onto graphene surface. Finally, Ru/FLG catalysts were obtained after removing the solvents. The TEM images of Ru/FLG are shown in Figure 1ab. Ru nanoparticles were dispersed uniformly on graphene surface without any aggregation. The average particle size was found to be 1.1  0.2 nm (Figure 1c). The actual loading of Ru in graphene supported Ru nanoparticle catalyst was 2.0% by weight, determined by inductively coupled plasma optical emission spectroscopy (ICP- OES) analysis.

Figure 1. ab) TEM image of as-synthesized 2.0% Ru/FLG, and c) size distribution of Ru nanoparticles. The size distribution was obtained by counting more than 200 nanoparticles.

Ru/FLG

FLG

GO

Graphite

5

15

25

35

45

55

65

2 


5

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 23

Figure 2. XRD patterns of pristine graphite, graphite oxide (GO), FLG, and Ru/FLG.

FLG was obtained with thermal exfoliation of graphite oxide at 700°C under 10% H2/Ar flow (Experimental details in Supporting Information). Figure 2 shows the X-ray diffraction (XRD) patterns of pristine graphite, as-synthesized graphite oxide (GO), FLG, and Ru/FLG. Pristine graphite shows characteristic (002) peak at 26.6. After oxidation, the (002) peak of graphite oxide downshifted to 11.3, indicating an interlayer separation of about 7.8 Å. The dominant interlayer distance suggests graphite oxide layer was mostly intercalated.37 After thermal exfoliation under 700C in flowing 10% H2/Ar, the sharp peak at 11.3 disappeared, suggesting the reduction of GO and exfoliation of GO layers.38 However, the broad peak at 26.2 for FLG may indicate some amount of reclustering.39 Therefore, the graphene support used in this study mainly contains multi-layered graphene/graphite. After loading Ru nanoparticles on FLG, the XRD pattern did not shown significant changes, mainly because very small particles do not have long-range ordering to facilitate visible XRD peaks and the loading of Ru is low. Graphene samples were also characterized with N2 physisorption (Figure S1 and Table S1). The adsorption-desorption of FLG and Ru/FLG showed type IV characteristics with a hysteresis loop in the range of 0.49-0.99, suggesting the mesoporous structure of these samples (Figure S1). Brunauer-Emmett-Teller (BET) surface area of FLG was determined to be 420 m2/g. The specific surface area values are lower than the theoretical value for graphene monolayer (2630 m2/g),40 which also indicated the existence of multilayer graphene support used in this study. However, after loading ruthenium nanoparticles, surface area dramatically declined (74 m2/g), mainly owing to the re-stacking of graphene layers during the preparation of Ru catalysts. Meanwhile, the pore volume also dropped from 1.15 to 0.20 cm3/g.


6

Page 7 of 23

It is of interests to study the chemical states of ruthenium in graphene-supported catalysts. However, it is difficult to use XPS to characterize the oxidation state of Ru due to the overlap of Ru 3d and C 1s bands in the range of 280-290 eV (Figure S2). X-ray absorption near edge spectroscopy (XANES) can overcome this drawback and is able to measure the sample without exposure to air, which is important to measure air-sensitive metals such as Ru. We measured the oxidation state of Ru in Ru/FLG after redution in situ with 3.5% H2/He at 150C for 1h. As shown in Figure 3, the white line of Ru K edge XANES of Ru/FLG sample is located at 22137 eV, which is between those of Ru foil (22132 eV) and RuO2 (22139 eV) standard samples. Based on the linear fitting41 of Ru/FLG sample using the two standard samples as the references, Ru component in Ru/FLG catalyst is found to be mainly composed of 46.2% Ru0 and 53.8% Ru4+. The significant amount of oxidized Ru could be due to the small particle size, which is hard to be reduced.42-44

Normalized absorption

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Ru/FLG Ru foil RuO2

1.0

0.5

-0.1 22050

22100

22150

22200

Energy (eV)

Figure 3. XANES of Ru K edge of Ru/FLG catalyst and the comparison of standard samples (Ru foil and RuO2 powder). Ru/FLG sample was reduced under 3.5% H2/He at 150C for 1 h to


7

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 23

remove surface adsorbed oxygen. Subsequently, XANES measurement was conducted at room temperature under the protection of 3.5% H2/He. 2.2 Catalytic performance of Ru catalysts in the presence of molecular hydrogen The hydrogenation and dehydration of LA (Scheme 1) was carried out in aqueous solutions containing 5 or 10wt.% LA (Table 1). When Ru/FLG catalyst was employed (substate to metal tatio = 4400), 99.7% conversion of LA and 100% selectivity to GVL could be obtained at room temperature within 12 hours (Entry 1, Table 1). Even when 5wt.% LA aqueous solution was used (substrate to metal ratio = 1460), we can still achieve 99.3% conversion of LA and 97.7% selectivity to GVL within 8 hours (Entry 2, Table 1). Since 5wt.% LA could be readily obtained from the hydrolysis of carbohydrates in the presence of 1-5% mineral acid,3,45 our results indicate that the solution from LA production process doesn’t need any pre-concentration process (e.g. extraction and distillation). The LA hydrogenation product, GVL, can be separated from the aqueous solution at lower costs due to its lower boiling point and more hydrophobic nature than LA.46 After neutralization of the crude LA solution, Ru/FLG catalyst could be directly applied to hydrogenize LA for the production of GVL, which will decrease the cost in GVL production.

Scheme 1. Hydrogenation and dehydration of LA to produce GVL.


8

Page 9 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

We also prepared several other ruthenium catalysts as the control samples and tested them under the same reaction conditions (Entries 36, Table 1) (Experimental details in Supporting Information). When FLG was replaced by other support materials, such as activated carbon (AC), silica gel, or Vulcan XC72R carbon, lower conversions (5363%) and selectivities to GVL (6471%) were observed (Entries 46). Furthermore, similar conversion (60.4%) and selectivity (53.8%) were observed when no support was used (PVP-stabilized Ru nanoparticles, Entry 12). The average production rates of GVL were 65-84 molGVL gRu–1 h–1, only about a half of that obtained using Ru/FLG (178 molGVL gRu–1 h–1). When a commercial catalyst Ru/C was used (Entry 7), the average production rate of GVL (63 molGVL gRu–1 h–1) was at the same level as the other control catalysts. As a comparison, Entries 10&11 showed the results obtained from the literature using Ru/C and PtO2, which showed moderate activity in this reaction under ambient temperature. However, the average production rate of GVL was only about 6.56.8 molGVL gRu–1 h–1 in each case. In general, we found that graphene support enhanced the activity of the catalysts by 24 folds compared with other traditional supports. To better understand the function of graphene, we also employed wetness-impregnation (WI) method to prepare FLG-supported ruthenium catalyst (2.8% Ru/FLG-WI). Slightly lower conversion (95.5%) and selectivity to GVL (88.5%) could be achieved (Entry 3). The production rate of GVL is 155 molGVL gRu–1 h–1, comparable to that obtained over Ru/FLG catalyst. We also prepared a control catalystan oxidized Ru catalyst supported on graphene (denoted as “Ru/FLG, oxidized”) by treating Ru/FLG sample in 10 mL 30% hydrogen peroxide. The oxidized sample was separated and washed with deionized water, and subsequently used for LA


9

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 23

conversion reaction under the same reaction conditions. Please note that no reduction procedure was carried out at 80C for oxidized Ru sample, which was adopted by all other Ru samples in typical operation procedure. As the result shown by Entry 13 in Table 1, this catalyst showed significantly lower conversion (84.2%) and selectivity to GVL (85.0%), suggesting that reduced Ru (Ru(0)) shall be the major active site of the reaction. Ru/FLG catalyst was further tested under solvent-free conditions (Entries 89) to show its versatility in LA conversion under various reaction conditions. Solvent-free condition is preferred if LA was separated in advance, because it could facilitate the later isolation and processing of GVL derivatives.26 63.9% conversion and 89.1% selectivity to GVL could be achieved over this catalyst after 32 h. The average production rate of GVL is slightly lower (130 molGVL gRu–1 h–1) compared with solution condition, but is much higher than the reported highest value.26 Complete conversion and 93.5% selectivity to GVL were obtained after 58 h, suggesting that Ru/FLG is also an excellent catalyst for this reaction under solvent-free conditions.

Table 1. Catalytic performance of ruthenium catalysts in the hydrogenation of LAa Entry

Catalyst

LA S/Mb Catalyst t Conv. conc. Amount (%) (h) (wt.%) (mg)

Selectivity (%) GVL

Othersc

Average production rate of GVLd (molGVL gRu–1 h– 1 )

1e

2.0% Ru/FLG

10

4400

10

12 99.7

100

0

365

2

2.0% Ru/FLG

5

1460

15

8

99.3

97.7

2.3

178

3

2.8% Ru/FLG-WI

5

1460

10.6

8

95.5

88.5

11.5

155

4

2.1% Ru/AC

5

1460

14.2

8

59.1

64.5

35.5

70


10

Page 11 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

5

0.7% Ru/SiO2

5

1460

43.7

8

53.0

66.7

33.3

65

6

2.0% Ru/Vulcan XC72R

5

1460

15

8

63.4

71.6

28.4

84

7f

3.6% Ru/C

10

1740

10

12 90.9

68.6

31.4

63

8g

2.0% Ru/FLG

100

7330

30

32 63.9

89.1

10.9

130

9g

2.0% Ru/FLG

100

7330

30

58 100

93.5

6.5

118

10h

5% Ru/C

100

348

250

50 100

97.5

2.5

6.8

11i

PtO2



286

400

44

87





6.5

12

Ru-PVP

5

1460



8

60.4

53.8

46.2

45

13

2.0% Ru/FLG, oxidized

5

1460

15

8

84.2

85.0

15.0

131

a

Reaction conditions: 10 mL H2O, 0.5 g LA, 40 bar H2, 20C. b S/M: the molar ratio of substrate (LA) to active metal (Ru). Products are identified by GC and GC-MS. c Byproducts mainly include angelica lactone and other unknown products. d Average production rate of GVL = moles of produced GVL /(mass of Ru  reaction time), molGVL gRu–1 h–1. e 1.0 g LA. f Use commercial carbon supported ruthenium catalyst (Ru/C). The loading amount was determined with ICP-MS. g Solvent-free reaction conditions. The reaction was conducted at 37C due to the higher melting point of LA (3335C). h From literature:26 5.0 g LA, 0.25 g 5% Ru/C, 25C, 12 bar H2, 50 h. i From literature:16 0.5 mole of LA, 150 mL of ethyl acetate, 0.4 g PtO2, r.t., 2.33.0 bar H2, 44 h, mechanical stirring.

A kinetic study was carried out to investigate the reaction mechanism of LA hydrogenation (Figure 4). We found that gamma-hydroxyvaleric acid (GHA) is the main intermediate at 2h by using 1H NMR (Figure S3), suggesting that GHA is the major reaction intermediate in this reaction, as shown in Scheme 1. Graphene supported Ru catalysts could be separated from aqueous solution after reaction with filtration, and reused for multiple times. As shown in Figure 5, both conversion (99.30.7%) and selectivity (95.02.5%) were maintained for at least five runs. Since the deactivation of catalyst could be masked at high conversion, the stability was further confirmed in an additional


11

ACS Catalysis

recycling experiment conducted at low conversion with a shorter reaction time of 2 h (Figure S4). Similarly, the conversion and selectivity did not change significantly within five runs.

Conversion & yield (%)

100 80 60 Conversion

40

yield of GVL

20

0 0

2

4 6 Reaction time (h)

8

Figure 4. Time dependent conversion of LA and yield of GVL obtained using 2.0% Ru/FLG catalyst in the hydrogenation and dehydration of LA. Reaction conditions: 10 mL H2O, 0.5 g LA, 15 mg catalyst (0.297 mg Ru), S/M = 1460, 20C, 40 bar H2.

120

Conversion & selectivity (%)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 23

Conversion Selectivity

100 80

60 40 20 0 1

2

3 Run

4

5

Figure 5. Recycling experiment of 2.0% Ru/FLG catalyst in the hydrogenation and dehydration of LA. Reaction conditions: 10 mL H2O, 0.5 g LA, 15 mg catalyst (0.297 mg Ru), 20C, 40 bar H2, 8 h (the same condition as entry 2 in Table 1).


12

Page 13 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Since traditional Ru/C shows continuous deactivation in this reaction, 26 the exceptional stability of graphene supported Ru catalyst is very interesting. In order to study the influence of the support, we also tested the stability of activated carbon supported ruthenium (Ru/AC) during recycling experiment. We found that Ru/AC catalyst showed a low conversion (59%) for the first run, and then deactivated continuously within subsequent cycles (Figure S5). The conversion dropped significantly to 43% in the 4th run, similar to the behavior of traditional Ru/C catalyst observed by other researchers.26 These results suggest that the graphene support stabilized the catalyst from deactivation during the reaction. Furthermore, we studied the influence of the preparation method on catalyst stability by comparing Ru/FLG and Ru/FLG-WI, which were prepared with polyol and wetness impregnation methods, respectively. We found that Ru/FLGWI catalyst also showed excellent stability throughout the recycling experiment (Figure S6), 84.44.8% conversion and 88.52.8% selectivity could be steadily obtained during 4 runs. This result indicated that the stability of graphene supported ruthenium catalysts were less influenced by the preparation method. Graphene support could be the key factor responsible for exceptional stability. Leaching and aggregation are common reasons of catalyst deactivation. Hence, we analyzed the concentration of leached ruthenium in the aqueous solution after separating the catalyst with inductively coupled plasma mass spectrometer (ICP-MS). The ratio of leaching amount of ruthenium in aqueous solution relative to total ruthenium in the catalyst was determined to be 0.92% and 0.64% for Ru/AC and Ru/FLG catalysts, respectively. The low level of leached ruthenium suggests that leaching shall not be the main reason of continuous deactivation of Ru/AC catalyst.


13

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 23

Hence, we also collected the TEM images of Ru/FLG and Ru/AC after reaction (Figure S7 and S8). The average particle size of Ru/FLG catalyst did not change (1.1  0.2 nm) after reaction. However, we found that average particle size of Ru/AC catalyst slightly increased from 1.1  0.2 nm to 1.4  0.3 nm. The size distribution histogram showed that more particles with the size of 1.42.2 nm formed after reaction. Less surface ruthenium atoms in larger particles could lead to lower activity. Thus we conclude that the aggregation of Ru nanoparticles during the reaction could be the main reason of deactivation of the Ru/AC catalysts. 2.3 The interaction of Ru nanoparticles with carbon support The enhanced activity and stability of graphene supported Ru catalysts may be attributed to the interaction between graphene and Ru nanoparticles. Several characterization techniques, including XPS, FTIR, and Raman spectroscopy, were used to study this interaction. The results were discussed below. As discussed in Section 2.1, Ru 3d signal is interfered with C1s, and thus is difficult to be resolved. Instead, we measured Ru 3p to study the oxidation states of Ru. We measured Ru samples prepared with either wetness impregnation or polyol reduction (Figure 6ab). As shown in Figure 6a, the Ru 3p 3/2 binding energy of FLG supported Ru (Ru/FLG-WI, prepared with wetness impregnation method) is 462.6 eV. The peak deconvolution suggests that the surface is composed of 52.3% Ru (IV) and 47.7% Ru(0). Activated carbon supported Ru showed higher binding energy at 462.9 eV, which is closer to the RuO2 standard (463.1 eV). This sample contains more Ru(IV) (66.1%) and less Ru(0) (33.9%) compared with graphene supported catalyst (Ru/FLG-WI). These results suggest that graphene supported Ru processes more Ru (0) on the surface. We also observed the similar trend on the samples prepared with polyol reduction approach (Figure 6b). Graphene supported Ru (Ru/FLG) showed a lower binding energy (462.6


14

Page 15 of 23

eV) and higher concentration of Ru(0) (36.8%) compared with activated carbon (Ru/AC) (462.8 eV and 29.4%, respectively). Because TEM already showed that the particle size of Ru are similar, XPS results suggest that more Ru(0) present in graphene samples could be responsible to the improved activity, selectivity and stability during recycling. a)

Ru(IV)

b)

462.6 Ru(0)

462.6 Ru(0)

Ru 3p3/2

Ru/FLG

Intensity (a.u.)

462.9 Ru/AC-WI

463.1 RuO2

470

Ru(IV)

Ru 3p3/2

Ru/FLG-WI

Intensity (a.u.)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

465

460

455

462.8 Ru/AC

463.1 RuO2

470

Binding Energy (eV)

465

460

455

Binding Energy (eV)

Figure 6. Ru 3p3/2 XPS spectra of graphene supported Ru (Ru/FLG), activated carbon supported Ru (Ru/AC), and RuO2 standard sample. a) Samples prepared with wetness impregnation method; b) samples prepared with polyol reduction method. We performed XPS peak fitting using CasaXPS program. All XPS spectra are calibrated by C1s peak at 284.6 eV. The peak width of Ru(IV) and Ru(0) were restricted to be the same during fitting. And the positions of Ru(IV) and Ru(0) were restricted to 463.0463.2 and 461.6461.7 eV, respectively. These fitting parameters are adopted from a reported work.47 Asymmetric and symmetric peak shapes were applied to Ru(IV) and Ru(0), respectively.

We also performed FTIR studies of CO adsorption on Ru catalysts using Diffuse Reflectance Infrared Fourier Transform spectroscopy (DRIFTs). We studied all the Ru samples, but due to the strong infrared absorption of carbon we only found that Ru samples prepared with wetness


15

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 23

impregnation gave clear CO signal. As shown in Figure S9, when Ru was supported on graphene (Ru/FLG), we found a single CO adsorption peak at 1977 cm1. When activated carbon was used (Ru/C; commercial catalyst), two peaks are resolved at 1977 and 2038 cm1, respectively. The peak at 2038 cm1 could be assigned to linearly adsorbed CO on Ru(0) atom.48,49 However, the peak at 1977cm1 was located between the range of linearly adsorbed CO on high defect sites and bridge CO adsorbed on Ru surface. Given that CO adsorption on Ru catalysts typically show high intensity peak of linear CO adsorption, and graphene supported Ru nanoparticle is extremely small (1.1 nm), we assign the peak at 1977 cm1 to linear CO adsorbed on high defect sites on Ru nanoparticle surface. The disappearance of the CO adsorption peak at 2038 cm1 on graphene supported Ru sample suggests that this sample only possesses electronrich Ru(0) atoms at defect sites that can donate more d-band electrons to * orbital of adsorbed CO molecules. These electron-rich Ru(0) atoms at defect sites could be the reason for the superior catalytic properties of Ru/FLG in LA to GVL conversion. These FTIR results are consistent with the XPS characterization because more electron-rich Ru sites can result in XPS peak shift to lower binding energies. When Ru nanoparticles were supported on graphene, the defect sites of graphene could be beneficial to the stabilization of Ru nanoparticles. During the preparation process, pristine graphite was oxidized by potassium permanganate, giving abundant functional oxygenated groups on the surface of graphite oxide. After exfoliation and reduction in hydrogen flow, defective sites were generated in graphene. The defects of graphene samples can be measured with the ratio of the intensities of D and G bands from Raman spectroscopy (Figure S10).40 The high D/G ratio (ID/IG = 0.82) of FLG sample suggests a large amount of defects exist in graphene sample after thermal reduction process. These defective sites could serve as excellent anchor


16

Page 17 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

points to stabilize Ru nanoparticles because the adsorption energy of Ru on single vacancy site on graphene (7.41 eV) is much larger than that on pristine graphene (3.26 eV).50 After loading Ru nanoparticles, we found that D/G ratio slightly decreased (ID/IG = 0.79) (Figure S10), which indicates a larger sp2 domain formed in Ru/FLG sample.51 No obvious peak shifts was detected, however, probably because the change in electronic structure of  and * bands that defective sites bonding is too little.52 It has been proposed that the origin of enhanced adsorption of Ru nanoparticles on graphene is mainly due to the hybridization between the dsp states of the Ru particles with the sp2 dangling bonds at the defect sites.50 The interaction of graphene and Ru nanoparticles can also upshift the d-band center of Ru, thus facilitates the activation of hydrogen molecule, providing enhanced hydrogenation activity. The superior stability of graphene supported Ru catalysts over Ru/C and Ru/AC could also be explained by the strong interaction between Ru nanoparticles and graphene defect sites that will prevent the leaching or migration of Ru nanoparticles.

3. Conclusion Few-layer graphene supported Ru catalysts were synthesized by loading pre-synthesized Ru nanoparticles onto the surface of graphene. At room temperature, the catalyst showed 99.7% LA conversion and 100% GVL selectivity in the hydrogenation and dehydration of LA. The graphene supported Ru catalysts are 24 times more active than ruthenium loaded on other traditional support materials. We found that FLG supported Ru catalysts showed exceptional stability during the recycling experiment. The conversion (~99%) and selectivity (9399%) could be maintained for at least five runs. TEM characterizations indicate no obvious aggregation of Ru nanoparticles after the


17

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 23

reaction. On the contrary, activated carbon supported Ru catalysts showed continuous deactivation. We found that aggregation of Ru nanoparticles could be responsible for the deactivation of activated carbon supported ruthenium catalysts based on ICP-MS and TEM studies. XPS and FTIR results suggest that graphene supported Ru catalysts possess more Ru(0) atoms on the surface compared with activated carbon supported ones, which could be related to improved activity, selectivity and stability of Ru/FLG in LA conversion to GVL. Graphene with large amount of defects could effectively prevent the migration and aggregation of supported ruthenium nanoparticles, owing to the strong interaction between the dsp states of the Ru nanoparticles with the sp2 dangling bonds at the defect sites of graphene.

ASSOCIATED CONTENT Supporting Information. Experimental details, N2 adsorption, XPS, recycling experiments, and TEM images. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *Phone (515)294-7084; Email: [email protected] Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes


18

Page 19 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

The authors declare no competing financial interest. ACKNOWLEDGMENT This work is supported through the funding from Iowa Energy Center. We thank Iowa State University for startup funds. We also thank Gordon J. Miller for use of his XRD, and Igor I. Slowing for use of his ICP-AES. ABBREVIATIONS LA, levulinic acid; GVL, gamma-valerolactone; AL, Anjelica lactone; ICP-OES, inductively coupled plasma optical emission spectroscopy; ICP-MS, inductively coupled plasma mass spectrometer; XRD, X-ray diffraction; XANES, X-ray absorption near edge spectroscopy; XPS, X-ray photoelectron spectroscopy; AC, activated carbon; G, exfoliated graphene; GO, graphite oxide; FLG, few-layer graphene; HMF, hydroxymethylfuran; PVP, polyvinylpyrrolidone; HPLC, High-performance liquid chromatography.


19

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 23

REFERENCES (1) Bozell, J. J.; Petersen, G. R. Green Chem. 2010, 12, 539-554. (2) Climent, M. J.; Corma, A.; Iborra, S. Green Chem. 2014, 16, 516-546. (3) Bozell, J. J.; Moens, L.; Elliott, D. C.; Wang, Y.; Neuenscwander, G. G.; Fitzpatrick, S. W.; Bilski, R. J.; Jarnefeld, J. L. Resour., Conserv. Recycl. 2000, 28, 227-239. (4) Potvin, J.; Sorlien, E.; Hegner, J.; DeBoef, B.; Lucht, B. L. Tetrahedron Lett. 2011, 52, 5891-5893. (5) Weingarten, R.; Conner, W. C.; Huber, G. W. Energy Environ. Sci. 2012, 5, 7559-7574. (6) Wettstein, S. G.; Alonso, D. M.; Chong, Y.; Dumesic, J. A. Energy Environ. Sci. 2012, 5, 8199-8203. (7) Alonso, D. M.; Wettstein, S. G.; Mellmer, M. A.; Gurbuz, E. I.; Dumesic, J. A. Energy Environ. Sci. 2013, 6, 76-80. (8) Alonso, D. M.; Gallo, J. M. R.; Mellmer, M. A.; Wettstein, S. G.; Dumesic, J. A. Catal. Sci. Technol. 2013, 3, 927-931. (9) Gurbuz, E. I.; Wettstein, S. G.; Dumesic, J. A. ChemSusChem 2012, 5, 383-387. (10) Nandiwale, K. Y.; Niphadkar, P. S.; Deshpande, S. S.; Bokade, V. V. J. Chem. Technol. Biotechnol. 2014, 89, 1507-1515. (11) Heeres, H.; Handana, R.; Chunai, D.; Borromeus Rasrendra, C.; Girisuta, B.; Jan Heeres, H. Green Chem. 2009, 11, 1247-1255. (12) Bond, J. Q.; Wang, D.; Alonso, D. M.; Dumesic, J. A. J. Catal. 2011, 281, 290-299. (13) Bond, J. Q.; Alonso, D. M.; Wang, D.; West, R. M.; Dumesic, J. A. Science 2010, 327, 1110-1114. (14) Serrano-Ruiz, J. C.; Wang, D.; Dumesic, J. A. Green Chem. 2010, 12, 574-577. (15) Wright, W. R.; Palkovits, R. ChemSusChem 2012, 5, 1657-1667. (16) Schuette, H. A.; Thomas, R. W. J. Am. Chem. Soc. 1930, 52, 3010-3012. (17) Hengne, A. M.; Rode, C. V. Green Chem. 2012, 14, 1064-1072. (18) Yan, K.; Liao, J.; Wu, X.; Xie, X. RSC Adv. 2013, 3, 3853-3856. (19) Upare, P. P.; Lee, J.-M.; Hwang, D. W.; Halligudi, S. B.; Hwang, Y. K.; Chang, J.-S. J. Ind. Eng. Chem. (Amsterdam, Neth.) 2011, 17, 287-292. (20) Ding, D.; Wang, J.; Xi, J.; Liu, X.; Lu, G.; Wang, Y. Green Chem. 2014, 16, 3846-3853. (21) Galletti, A. M. R.; Antonetti, C.; De Luise, V.; Martinelli, M. Green Chem. 2012, 14, 688694. (22) Bourne, R. A.; Stevens, J. G.; Ke, J.; Poliakoff, M. Chem. Commun. 2007, 4632-4634. (23) Luo, W.; Deka, U.; Beale, A. M.; van Eck, E. R. H.; Bruijnincx, P. C. A.; Weckhuysen, B. M. J. Catal. 2013, 301, 175-186.


20

Page 21 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

(24) Wettstein, S. G.; Bond, J. Q.; Alonso, D. M.; Pham, H. N.; Datye, A. K.; Dumesic, J. A. Appl. Catal., B 2012, 117-118, 321-329. (25) Mai, E. F.; Machado, M. A.; Davies, T. E.; Lopez-Sanchez, J. A.; Teixeira da Silva, V. Green Chem. 2014, 16, 4092-4097. (26) Al-Shaal, M. G.; Wright, W. R. H.; Palkovits, R. Green Chem. 2012, 14, 1260-1263. (27) Paton, K. R.; Varrla, E.; Backes, C.; Smith, R. J.; Khan, U.; O'Neill, A.; Boland, C.; Lotya, M.; Istrate, O. M.; King, P.; Higgins, T.; Barwich, S.; May, P.; Puczkarski, P.; Ahmed, I.; Moebius, M.; Pettersson, H.; Long, E.; Coelho, J.; O'Brien, S. E.; McGuire, E. K.; Sanchez, B. M.; Duesberg, G. S.; McEvoy, N.; Pennycook, T. J.; Downing, C.; Crossley, A.; Nicolosi, V.; Coleman, J. N. Nat Mater 2014, 13, 624-630. (28) Kim, K. S.; Zhao, Y.; Jang, H.; Lee, S. Y.; Kim, J. M.; Kim, K. S.; Ahn, J.-H.; Kim, P.; Choi, J.-Y.; Hong, B. H. Nature 2009, 457, 706-710. (29) Berger, C.; Song, Z.; Li, X.; Wu, X.; Brown, N.; Naud, C.; Mayou, D.; Li, T.; Hass, J.; Marchenkov, A. N.; Conrad, E. H.; First, P. N.; de Heer, W. A. Science 2006, 312, 11911196. (30) Zhang, B.; Song, J.; Yang, G.; Han, B. Chem. Sci. 2014, 5, 4656-4660. (31) Georgakilas, V.; Otyepka, M.; Bourlinos, A. B.; Chandra, V.; Kim, N.; Kemp, K. C.; Hobza, P.; Zboril, R.; Kim, K. S. Chem. Rev. 2012, 112, 6156-6214. (32) Upare, P. P.; Yoon, J.-W.; Kim, M. Y.; Kang, H.-Y.; Hwang, D. W.; Hwang, Y. K.; Kung, H. H.; Chang, J.-S. Green Chem. 2013, 15, 2935. (33) Wang, P.-Q.; Bai, Y.; Luo, P.-Y.; Liu, J.-Y. Catal. Commun. 2013, 38, 82-85. (34) Yan, Y.; Xia, B.; Qi, X.; Wang, H.; Xu, R.; Wang, J. Y.; Zhang, H.; Wang, X. Chem Commun (Camb) 2013, 49, 4884-4886. (35) Sun, B.; Jiang, Z.; Fang, D.; Xu, K.; Pei, Y.; Yan, S.; Qiao, M.; Fan, K.; Zong, B. ChemCatChem 2013, 5, 714-719. (36) Gao, Y.; Chen, X.; Zhang, J.; Asakura, H.; Tanaka, T.; Teramura, K.; Ma, D.; Yan, N. Adv. Mater. 2015, n/a-n/a. (37) Jeong, H.-K.; Lee, Y. P.; Lahaye, R. J. W. E.; Park, M.-H.; An, K. H.; Kim, I. J.; Yang, C.W.; Park, C. Y.; Ruoff, R. S.; Lee, Y. H. J. Am. Chem. Soc. 2008, 130, 1362-1366. (38) Li, J.; Lin, H.; Yang, Z.; Li, J. Carbon 2011, 49, 3024-3030. (39) Guo, Y.; Bao, C.; Song, L.; Yuan, B.; Hu, Y. Ind. Eng. Chem. Res. 2011, 50, 7772-7783. (40) Zhu, Y. W.; Murali, S.; Cai, W. W.; Li, X. S.; Suk, J. W.; Potts, J. R.; Ruoff, R. S. Adv. Mater. 2010, 22, 3906-3924. (41) Xiao, C. X.; Goh, T. W.; Brashler, K.; Pei, Y. C.; Guo, Z. Y.; Huang, W. Y. J. Phys. Chem. B 2014, 118, 14168-14176. (42) Joo, S. H.; Park, J. Y.; Renzas, J. R.; Butcher, D. R.; Huang, W. Y.; Somorjai, G. A. Nano Lett. 2010, 10, 2709-2713.


21

ACS Catalysis

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 23

(43) Grass, M. E.; Zhang, Y. W.; Butcher, D. R.; Park, J. Y.; Li, Y. M.; Bluhm, H.; Bratlie, K. M.; Zhang, T. F.; Somorjai, G. A. Angew. Chem., Int. Ed. 2008, 47, 8893-8896. (44) Huang, W.; Kuhn, J. N.; Tsung, C. K.; Zhang, Y.; Habas, S. E.; Yang, P.; Somorjai, G. A. Nano Lett. 2008, 8, 2027-2034. (45) Yuan, J.; Li, S.-S.; Yu, L.; Liu, Y.-M.; Cao, Y.; He, H.-Y.; Fan, K.-N. Energy Environ. Sci. 2013, 6, 3308-3313. (46) Sen, S. M.; Alonso, D. M.; Wettstein, S. G.; Gurbuz, E. I.; Henao, C. A.; Dumesic, J. A.; Maravelias, C. T. Energy Environ. Sci. 2012, 5, 9690-9697. (47) Morgan, D. J. Surf. Interface Anal. 2015, 47, 1072-1079. (48) Kantcheva, M.; Sayan, S. Catal. Lett. 1999, 60, 27-38. (49) Chin, S. Y.; Williams, C. T.; Amiridis, M. D. J. Phys. Chem. B 2006, 110, 871-882. (50) Liu, X.; Yao, K. X.; Meng, C.; Han, Y. Dalton Trans. 2012, 41, 1289-1296. (51) Stankovich, S.; Dikin, D. A.; Piner, R. D.; Kohlhaas, K. A.; Kleinhammes, A.; Jia, Y.; Wu, Y.; Nguyen, S. T.; Ruoff, R. S. Carbon 2007, 45, 1558-1565. (52) Diez-Betriu, X.; Alvarez-Garcia, S.; Botas, C.; Alvarez, P.; Sanchez-Marcos, J.; Prieto, C.; Menendez, R.; de Andres, A. J. Mater. Chem. C 2013, 1, 6905-6912.


22

Page 23 of 23

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Catalysis

Table of Contents Graphic


23

Conversion of Levulinic Acid to Î³-Valerolactone over Few-Layer

Recommend Documents