Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride

ARTICLE IN PRESS

JID: JECHEM

[m5G;April 28, 2017;11:18]

Journal of Energy Chemistry xxx (2017) xxx–xxx

Contents lists available at ScienceDirect

Journal of Energy Chemistry journal homepage: www.elsevier.com/locate/jechem

http://www.journals.elsevier.com/ journal-of-energy-chemistry/

Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride✩ Jiayi Chen a,1, Yao Ge a,b,1, Yuanyuan Guo a, Jinzhu Chen a,c,∗

Q1

a b c

Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, Guangdong, China College of Chemistry and Chemical Engineering, Xinjiang University, Urumqi 830046, Xinjiang, China College of Chemistry and Materials Science, Jinan University, Guangzhou 510632, Guangdong, China

a r t i c l e

i n f o

Article history: Received 4 November 2016 Revised 18 April 2017 Accepted 18 April 2017 Available online xxx Keywords: Biomass Carbon nitride 5-Hydroxymethylfurfural Hydrogenation

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

a b s t r a c t Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural (HMF) to 2,5-dihydroxymethyltetrahydrofuran (DHMTHF) with 96% selectivity and a complete HMF conversion is obtained over palladium catalyst supported on mesoporous graphitic carbon nitride (Pd/mpg-C3 N4 ) under pressured hydrogen atmosphere in aqueous media. The excellent catalytic performance of Pd/mpg-C3 N4 is attributed to hydrogen bonding-related competitive interactions between reactant HMF and “intermediate” 2,5dihydroxymethylfuran (DHMF) with the support mpg-C3 N4 , which leads to a deep hydrogenation of DHMF to DHMTHF. © 2017 Published by Elsevier B.V. and Science Press.

1. Introduction The production of liquid fuels and organic chemicals from plentiful and renewable biomass is a matter of growing interest to reduce dependence on fossil fuel sources [1–4]. Biomass is one of the most abundant carbon-containing renewable resources on earth for the production of chemicals and fuels [5,6]. A recent review critically discussed nano-scale metal-catalysts for valorization of biomass-related cellulose, chitin, lignin and lipids [7]. The platform molecule 5-hydroxymethylfurfural (HMF), a typical representative of the furan family, is viewed as one of the key precursors for the production of chemicals and liquid transportation fuels [8–10]. It can be converted into 2,5-dihydroxymethylfuran (DHMF) and 2,5-dihydroxymethyl-tetrahydrofuran (DHMTHF) with further hydrogenation (Fig. 1) [11,12]. Typically, DHMTHF has been used as a high value added chemical with important applications as a solvent as well as polymer monomer [13,14]. In addition,

✩ This work was supported by the National Natural Science Foundation of China (21472189), Natural Science Foundation of Guangdong Province, China (2015A030312007), Science and Technology Planning Project of Guangzhou City, China (201707010238), and Jinan Double Hundred Talents Plan. ∗ Corresponding author at: College of Chemistry and Materials Science, Jinan University, Guangzhou 510632, Guangdong, China. E-mail address: [email protected] (J. Chen). 1 These authors contributed equally to this work.

DHMTHF can be converted to 1,6-hexanediol which is an important monomer for special applications [11,15–20]. The reaction of HMF to DHMTHF has been studied extensively over various metal catalysts. Previously, Nakagawa and Tomishige investigated selective reduction HMF into DHMTHF by using Ni– Pd bimetallic catalysts [11]. Dumesic and co-workers developed monophasic and biphasic reactor systems for selective transformation of HMF into DHMTHF over ruthenium, palladium, and platinum catalysts [12]. Xu and co-workers achieved the conversion of fructose to DHMTHF by a combination of fructose dehydration and HMF hydrogenation [21]. Very recently, the conversion of HMF to DHMTHF was also investigated over palladium immobilized on metal-organic frameworks [22]. Recently, selective hydrogenations of phenol [23–26], nitrophenol [29], quinolone [30], nitroarenes [31], phenylacetylene [32], benzoic acid [33], styrene [34], and 2-methylfuran [34] were reported by using mesoporous graphitic carbon nitride (mpg-C3 N4 ) supported transition metal nanoparticles owing to the unique properties of mpg-C3 N4 such as a large surface area and accessible crystalline pore walls. Recently, Yan and co-workers reported mesoporous, nitrogen-containing carbon materials from chitin carbonization for heavy-metal (CrVI , HgII , PdII ) removal and styrene epoxidation [35]. In addition, the presence of electron-rich nitrogen atoms in the supports can enhance π -binding ability, improve basicity and hydrophilicity, and modify interactions between metal sites, reactant, intermediate, product, and support [23–34,36,37].

http://dx.doi.org/10.1016/j.jechem.2017.04.017 2095-4956/© 2017 Published by Elsevier B.V. and Science Press.

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

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

ARTICLE IN PRESS

JID: JECHEM 2

[m5G;April 28, 2017;11:18]

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

Fig. 1. Molecular interactions in the hydrogenation of HMF over Pd/mpg-C3 N4 .

the reactor, which was sealed, purged with H2 three times, and pressurized with H2 (1.0 MPa) at room temperature, and then heated to the desired reaction temperature. The reactor was maintained at the desired temperature for 4 h. After the reaction, the reactor was cooled to room temperature. Products were separated by filtration and water-soluble products were analyzed by HPLC and liquid chromatography-mass spectrometer (LC-MS).

87

51

In this research, mpg-C3 N4 supported palladium (Pd/mpg-C3 N4 ) was investigated as bifunctional catalyst for selective hydrogenation of HMF to DHMTHF under pressured hydrogen atmosphere yielding DHMTHF selectivity of 96% with a complete HMF conversion (Fig. 1). Our results further indicated that the observed excellent catalytic performance of Pd/mpg-C3 N4 is presumably related to hydrogen bonding-related competitive interactions between reactant HMF and “intermediate” DHMF with the support mpg-C3 N4 , which leads to a deeper DHMF hydrogenation to DHMTHF (Fig. 1).

2.4. Reusability of the catalyst

94

52

2. Experimental

95

53

2.1. Catalyst preparation

After reaction, the catalyst Pd/mpg-C3 N4 was separated from the reaction mixture, and washed with de-ionized water and ethanol, dried at 60 °C over night in a vacuum oven. Then, the recovered catalyst was used for the next cycle under the same conditions.

43 44 45 46 47 48 49 50

54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82

mpg-C3 N4 : mpg-C3 N4 was prepared according to the literature method [37]. Pd/mpg-C3 N4 : K2 PdCl4 (63 mg) in H2 O (40 mL) was treated with activated mpg-C3 N4 (500 mg) and the mixture was stirred for 8 h at 80 °C. Then NaBH4 (40 mg) solution was added to this suspension at 0 °C and stirred for 4 h to synthesize Pd/mpg-C3 N4 . Finally, Pd/mpg-C3 N4 (Pd 4.1 wt% based on ICP–AES) was separated by filtration, washed with distilled water and dried at 100 °C overnight under vacuum. 2.2. Characterization of catalyst The Brunauer–Emmett–Teller (BET) surface area measurements were performed with N2 adsorption–desorption isotherms at 77 K (SI-MP-10/PoreMaster 33, Quantachrome), after degassed under vacuum at 423 K for 12 h. The specific BET surface areas were evaluated using the method in the p/p0 range from 0.05 to 0.3. X-ray diffraction (XRD) patterns were obtained with PANalytical X’pert Pro multipurpose diffractometer at 40 KV and 40 mA, using Ni-filtered Cu-Kα radiation (λ = 0.15418 nm). High-resolution transmission electron microscopy (HRTEM) was recorded using a JEM2100HR instrument. Samples for HRTEM analysis were prepared by placing a drop of the suspension sample in ethanol onto carboncoated Cu grids, followed by evaporating the solvent. X-ray photoelectron spectroscopy (XPS) data was obtained with Kratos Axis Ultra (DLD) photoelectron spectrometer operated at 15 kV and 5 mA at a pressure of about 5 × 10−9 torr using Al Kα as the exciting source. The binding energies were referenced to the C 1s photoelectron peak (284.6 eV). Inductively coupled plasma–optical emission spectroscopy (ICP–OES) analysis was performed with PerkinElmer Optima 80 0 0 instrument.

83

2.3. Selective hydrogenation of HMF to DHMTHF

84

The hydrogenation of HMF under H2 was conducted in a 60 mL stainless steel autoclave reactor. HMF (63 mg, 0.5 mmol), Pd/mpgC3 N4 (30 mg, Pd 4.1 wt%) and water (3 mL) were introduced into

85 86

2.5. Analysis of product HMF and the hydrogenation products were determined by HPLC (Shimadu LC-20AT) equipped with UV and refractive index (RI) detectors. The analysis was conducted with a Shodex Sugar SH-1011 column (300 × 8 mm) using HPLC grade H2 SO4 (0.005 M) water solution as the eluent and a flow rate of 0.5 mL/min at a column temperature of 50 °C. The conversion of HMF was determined by a UV detector (284 nm); whereas, the concentrations of products such as DHMF, DHMTHF and hexanetriol (HT) were monitored by an RI indicator.

88 89 90 91 92 93

96 97 98 99 100 101 102 103 104 105 106 107 108 109

3. Results and discussion

110

3.1. Catalyst characterization

111

As shown from N2 adsorption/desorption isotherm (Fig. 2), the Brunauer–Emmett–Teller (BET) surface area of mpg-C3 N4 and Pd/mpg-C3 N4 are 145 m2 /g and 119 m2 /g, respectively. Moreover, the BET surface area of Pd/mpg-C3 N4 decreases from 119 m2 /g for the fresh one to 98 m2 /g for the recovered one after five-time recycling, indicating the adsorption of humins formed on the surface of Pd/mpg-C3 N4 through the reaction [12]. X-ray powder diffraction (XRD) patterns of mpg-C3 N4 and Pd/mpg-C3 N4 are compared in Fig. 3. For mpg-C3 N4 , the peak at 27.4° was a characteristic peak (002) corresponding to the stacking of conjugated inter-layers; whereas, the peak at 13.1° was indexed as (100) showing in-plane ordering of tri-s-triazine units (JCPDS 87-1526) [38–40]. Notably, the diffraction peaks of palladium species in Pd/mpg-C3 N4 are unobserved (Fig. 3), which is presumably due to the small nanoparticles size and high dispersion of noble nanoparticles on the support surface. In addition, the XRD patterns of the recovered Pd/mpg-C3 N4 are almost the same as those of fresh one, indicating no apparent loss of crystallinity for the recovered catalyst (Fig. 3).

112

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130

ARTICLE IN PRESS

JID: JECHEM

[m5G;April 28, 2017;11:18] 3

Intensity

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

mpg-C3N4 Pd/mpg-C3N4 recovered Pd/mpg-C3N4

Fig. 2. N2 adsorption–desorption isotherms of mpg-C3 N4 , Pd/mpg-C3 N4 (Pd 4.1 wt%) and recovered Pd/mpg-C3 N4 (Pd 4.1 wt%) after a five-cycle experiment with the reaction conditions described in Fig. 7.

131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

The transmission electron microscopy (TEM) analysis of mpgC3 N4 and Pd/mpg-C3 N4 revealed that both the structure and mesoporosity of mpg-C3 N4 are well maintained after the loading of Pd. As shown in Fig. 4(b) and (d), the well-dispersed Pd nanoparticles can clearly be observed over mpg-C3 N4 and the average size is about 3.6 nm. For recovered Pd/mpg-C3 N4 , the TEM image (Fig. 4c) shows spherical palladium nanoparticles with average size of 5.4 nm, which is a little bigger than that of the fresh one (Fig. 4e). Therefore, the slightly reduced selectivity toward DHMTHF for Pd/mpg-C3 N4 recycling in HMF hydrogenation can reasonably be related to the adsorption and accumulation of oligomeric products on the Pd/mpg-C3 N4 surface and/or the increase in average particle size of recovered Pd nanoparticles on the mpg-C3 N4 support. Fig. 5(a) shows the X-ray photoelectron spectroscopy (XPS) of mpg-C3 N4 and Pd/mpg-C3 N4 . Peaks corresponding to carbon and

(a)

10

30

40

50

60

70

80

2 theta / degree Fig. 3. XRD patterns of mpg-C3 N4 , Pd/mpg-C3 N4 (Pd 4.1 wt%) and recovered Pd/mpg-C3 N4 (Pd 4.1 wt%) after a five-cycle experiment with the reaction conditions described in Fig. 7.

nitrogen for mpg-C3 N4 and palladium for Pd/mpg-C3 N4 were presented in the survey scan. The high resolution N 1 s XPS spectra of Pd/mpg-C3 N4 show almost the same peaks as those of mpg-C3 N4 (Fig. 5b). The predominant peak in N 1 s spectra of Pd/mpg-C3 N4 at 398.6 eV was assigned to the sp2 -hybridized nitrogen atoms in C–N=C, demonstrating the presence of triazine rings. The peak at about 399.6 eV was indexed as the sp3 -hybridized tertiary nitrogen N–(C)3 ; whereas, the peak at 401.3 eV was attributed to the amino group functionalized with a hydrogen (C–N–H), suggesting the existence of the defects [40–44]. Notably, the XPS analysis of fresh and spent Pd/mpg-C3 N4 did not show any evident changes of nitrogen content and nitrogen species.

(b)

(d)

20

(c)

(e)

Fig. 4. TEM images of (a) mpg-C3 N4 , (b) fresh and (c) recovered Pd/mpg-C3 N4 (Pd 4.1 wt%); particle-size distribution of (d) fresh and (e) recovered Pd/mpg-C3 N4 (Pd 4.1 wt%) after a five-cycle experiment with the reaction conditions described in Fig. 7.

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

147 148 149 150 151 152 153 154 155 156 157 158

ARTICLE IN PRESS

JID: JECHEM 4

[m5G;April 28, 2017;11:18]

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

(a)

(b)

(c)

Fig. 5. (a) XPS scan survey and (b) N 1s XPS spectra of mpg-C3 N4 and Pd/mpg-C3 N4 (4.1 wt%), (c) Pd 3d XPS spectra of Pd/mpg-C3 N4 .

164

The Pd 3d XPS of the Pd/mpg-C3 N4 catalyst can be deconvolved into two main characteristic peaks (Fig. 5c). The doublet with binding energies at 335.5 and 340.8 eV, assigned to Pd 3d5/2 and Pd 3d3/2 , respectively, can be ascribed to the Pd° state. The other peaks at 337.2 and 342.6 eV are assigned to the PdII state [24,25,27,28].

165

3.2. HMF conversion into DHMTHF

166

To probe the reduction network of HMF, HMF selective reduction to DHMTHF was further performed with pressured hydrogen atmosphere in water with Pd/mpg-C3 N4 as the catalyst. Hydrogenation of HMF results in the reduction of aldehyde group of HMF first to give DHMF; subsequently, a deep DHMF hydrogenation leads to saturation of DHMF furan ring to form DHMTHF (Fig. 1). In addition to DHMF and DHMTHF, hexanetriol (HT) was detected as side-product, owing to hydration/dehydration of DHMF, followed by reduction of ring-opening intermediate, and a further hydrogenolysis [12,45]. An initial study was conducted to explore the influence of metal sites on the selectivity to the HMF hydrogenation. Pt, Ru and Pd were loaded on the mpg-C3 N4 as the catalysts, respectively, for aqueous-phase hydrogenation of HMF. As shown in Table 1 (Run

159 160 161 162 163

167 168 169 170 171 172 173 174 175 176 177 178 179

1), Pt/mpg-C3 N4 was not active for HMF hydrogenation in the hydrogenation system, and the HMF conversion was extremely low under the investigated conditions. In the case of Ru/mpg-C3 N4 , the HMF conversion of 90% was obtained with the DHMTHF and HT selectivities of 60% and 40%, respectively (Table 1, Run 2). The DHMF was, however, unobserved under the investigated conditions indicating that the ruthenium catalyst was more active both in terms of DHMF hydrogenation and DHMTHF hydrogenolysis (Table 1, Run 2). In contrast, Pd is found to show high yield and selectivity than Pt and Ru as shown in Table 1 (Runs 1–3). In addition to the influence of metal site, variable amounts of Pd were loaded on the mpg-C3 N4 support, and the influence of the palladium loading level in Pd/mpg-C3 N4 on the HMF hydrogenation was investigated (Table 1, Runs 3–6). Generally, the conversions of HMF were >99% in all investigated conditions. Moreover, the DHMF selectivity smoothly decreased with Pd loading amount; whereas, the DHMTHF selectivity gradually increased with the Pd content to the highest of 96% at 4.1 wt% Pd in Pd/mpgC3 N4 . The above results thus suggested that Pd/mpg-C3 N4 was effective catalyst for HMF hydrogenation even under relatively lower Pd loading level of 1.2 wt%. However, the low loading levels of Pd in Pd/mpg-C3 N4 (Pd 1.2–2.5 wt%) is insufficient to completely hydrogenate the in situ obtained DHMF into DHMTHF under the in-

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202

ARTICLE IN PRESS

JID: JECHEM

[m5G;April 28, 2017;11:18]

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

5

Table 1. Selective hydrogenation of HMF over various catalysts. Run

Catalyst M/mpg-C3 N4 (M wt%)

HMF conversion (%)

DHMF selectivity (%)

DHMTHF selectivity (%)

HT selectivity (%)

1 2 3 4 5 6 7a 8b 9 10

Pt/mpg-C3 N4 (5.0 wt%) Ru/mpg-C3 N4 (4.8 wt%) Pd/mpg-C3 N4 (4.1 wt%) Pd/mpg-C3 N4 (1.2 wt%) Pd/mpg-C3 N4 (2.5 wt%) Pd/mpg-C3 N4 (8.4 wt%) Pd/mpg-C3 N4 (4.1 wt%) Pd/mpg-C3 N4 (4.1 wt%) Pd/C (5.0 wt%) Pd/CNT (4.5 wt%)

9 90 >99 >99 >99 >99 >99 >99 >99 >99

8 0 0 30 12 0 14 0 6 7

0 60 96 68 84 93 82 95 82 67

0 40 4 2 4 6 4 4 10 25

Reaction conditions: HMF (63 mg, 0.5 mmol), catalyst (30 mg), water (3 mL), reaction time (4.0 h), 60 °C, H2 (1.0 MPa). a 15 mg catalyst was used, b 45 mg catalyst was used.

Table 2. Selective hydrogenation of HMF to DHMTHF over various catalysts.

203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244

Run

Catalyst

Temperature (°C)

H2 pressure (MPa]

HMF Conversion (%)

DHMTHF Yield (%]

Ref.

1 2 3 4 5

Ni-Pd/SiO2 Ru/CeO2 Pd/MIL-101(Al)-NH2 Pd-Ir/SiO2 Pd/mpg-C3 N4

40 130 30 275 60

8 1.8 1.0 8 1.0

99 >99 >99 >99 >99

95 91 96 95 96

[11] [12] [22] [46] This work

vestigated condition (Table 1, Runs 4–5). Therefore, the hydrogenation of HMF over Pd/mpg-C3 N4 proceeds the reduction of aldehyde group first to form DHMF, followed by the saturation of furan ring to yield DHMTHF. Moreover, DHMF was observed as the “intermediate” for HMF-to-DHMTHF transformation (Fig. 1). Therefore, the yield of DHMTHF reaching 96% with the HMF conversion >99% under 1.0 MPa H2 was achieved over Pd/mpg-C3 N4 , this result is comparable to the reported catalysts (Table 2). In addition to the influence of Pd loading level in Pd/mpg-C3 N4 , the influence of Pd/mpg-C3 N4 loading amounts on the HMF hydrogenation was investigated by varying the amount of Pd/mpg-C3 N4 used in the hydrogenation system (Table 1, Runs 3, 7–8). As expected, a complete HMF conversion was observed during the reaction process, further indicating a high activity of Pd/mpg-C3 N4 for HMF hydrogenation. In addition, the more Pd/mpg-C3 N4 there is, the higher selectivity of DHMTHF is. Moreover, a decreased selectivity of DHMF is also observed with an increasing Pd/mpg-C3 N4 loading amounts (Table 1, Runs 3, 7–8). These results suggest that a further DHMF conversion into DHMTHF is promoted by Pd/mpgC3 N4 dosage. The influence of support on HMF conversion was investigated. Recently, Dumesic and co-workers investigated Ru-promoted selective hydrogenation of HMF, and suggested that the support materials of the Ru catalysts with high isoelectric points produced DHMTHF with high selectivities [12]. In our case, in addition to Pd/mpg-C3 N4 , carbon materials such as activated carbon and multi-walled carbon nanotubes (CNTs) supported palladium catalysts (Pd/C and Pd/CNT) were prepared and tested for HMF hydrogenation. Full conversions of HMF were observed over Pd/C and Pd/CNT with DHMTHF selectivities of 82% and 67%, respectively (Table 1, Runs 9–10). Obviously, Pd/mpg-C3 N4 is found to have more activity and selectivity than Pd/C and Pd/CNT as shown in Table 1 (Runs 3, 9–10). The catalytic performance of Pd/mpg-C3 N4 , Pd/C and Pd/CNT was further quantitatively compared in terms of turnover frequency (TOF). Herein, the TOF for the HMF hydrogenation was measured under a low HMF conversion level around 10%, given as the amount of consumed HMF per amount of surface Pd atoms per hour for the investigated Pd catalyst. The dispersibilities for Pd/mpg-C3 N4 , Pd/C and Pd/CNT were 1.3%, 0.9%, 2.3%, respectively, based on CO-based pulse chemisorption experiments. The resulting TOFs for HMF hydrogenation were 2195 molHMF /molsurface Pd h for Pd/mpg-C3 N4 , 2536

molHMF /molsurface Pd h for Pd/C and 532 molHMF /molsurface Pd h for Pd/CNT. Recently, Wang and our group reported the synthesis of cyclohexanone from phenol by using palladium supported on mpgC3 N4 and polyaniline wrapped carbon-nanotubes, respectively [24,25,27,28]. Owing to the presence of nitrogen in the support mpg-C3 N4 and PANI/CNT, the formation of hydrogen bonding between phenol and support seem to play an important role on the high selectivity to cyclohexanone production. In our case, the obtained high yield of DHMTHF is presumably attributed to a special function of support mpg-C3 N4 . Herein, Fig. 1 demonstrates the stepwise sequence for the HMF hydrogenation. There is a hydrogen bond between HMF and mpg-C3 N4 , which can facilitate the anchorage of HMF molecule on mpg-C3 N4 . However, once the HMF is hydrogenated to DHMF, there is a stronger hydrogen bonding interaction between DHMF and mpg-C3 N4 than that of mpg-C3 N4 and HMF, due to the fact that DHMF shows two hydroxyl groups. Thus, the stronger interaction between the in situ obtained DHMF and mpg-C3 N4 leads to a deep DHMF conversion to DHMTHF (Fig. 1). Adsorption experiments of HMF and DHMF on mpg-C3 N4 were examined and the HPLC analyses show that the HMF and DHMF adsorptions on mpg-C3 N4 are around 0.03 mmolHMF /gmpg-C3N4 and 0.22 mmolDHMF /gmpg-C3N4 , respectively. mpg-C3 N4 shows enhanced DHMF adsorption than that of HMF, presumably owing to the fact that DHMF is much more hydrophilic and can strongly interact with mpg-C3 N4 support. The significant difference of the adsorption between HMF and DHMF on mpg-C3 N4 can possibly conduce to the high selectivity of DHMTHF on catalyst Pd/mpg-C3 N4 . As depicted in Fig. 6(a), the impact of hydrogenation temperature of HMF on the selectivities to DHMF and DHMTHF using Pd/mpg-C3 N4 as the catalyst in aqueous media was investigated. In general, a full conversion of HMF was found over Pd/mpg-C3 N4 with DHMTHF as the primary product in 4 h at all investigated temperatures. At 40 °C, the DHMF and DHMTHF selectivities were 38% and 61%, respectively, suggesting that low reaction temperature is insufficient to fully hydrogenate the in situ obtained DHMF into DHMTHF. When the reaction temperature was elevated to 60 °C, DHMF was completely converted into DHMTHF with the selectivity of 96%. Hexanetriol (HT) with the selectivity of 4% was identified as the by-product. However, a further increased reaction temperature to 120 °C led to a slightly reduced DHMTHF selectivity to 90%, suggesting that the hydrogenation of HMF under moderate

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286

ARTICLE IN PRESS

JID: JECHEM 6

[m5G;April 28, 2017;11:18]

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

(a)

(b)

(c)

Fig. 6. Selective hydrogenation of HMF to DHMTHF over Pd/mpg-C3 N4 as a function of (a) reaction temperature, (b) reaction time and (c) H2 pressure. Reaction conditions: HMF (63 mg, 0.5 mmol), Pd/mpg-C3 N4 (30 mg, Pd 4.1 wt%), water (3 mL) and (reaction time (4.0 h, H2 (1.0 MPa) for (a); 60 °C, H2 (1.0 MPa) for (b); reaction time (4.0 h), 60 °C for (c)).

287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309

temperature promotes the formation of DHMTHF. In contrast, high temperature for HMF hydrogenation favors both the DHMF conversion into DHMTHF and the further DHMTHF hydrogenolysis to HT. Fig. 6(b) shows the influence of reaction time on the conversion of HMF to DHMTHF on Pd/mpg-C3 N4 . Full conversion of HMF was achieved at 30 min with the primary DHMTHF product. The selectivity of DHMF reaches a maximum of 77% at 15 min. The selectivity of DHMF decreased with reaction time ranging from 15 min to 4 h, and the selectivity of DHMTHF gradually increased, suggesting that prolonging the hydrogenation reaction time favors a further conversion of DHMF. After 4 h, full DHMF-to-DHMTHF hydrogenation was achieved, leading to an overall selectivity of 96% to DHMTHF. And the HT selectivity increased ranging from 0 to 7%. These results suggested that the conversion of HMF into DHMTHF was stepwise and DHMF was the “intermediate” (Fig. 6b). The effects of hydrogen pressure on HMF reduction was studied in Fig. 6(c). The HMF conversion significantly increased from 8% to 99% with the H2 pressure ranging from 0.1 to 0.5 MPa. Complete HMF conversion was observed at 0.5 MPa with the selectivities of DHMF and DHMTHF 23% and 71%, respectively, under the investigated conditions (Fig. 6c). The highest DHMF selectivity of 23% at 0.5 MPa was observed and later full consumed with reaction pressure raised to 1.0 MPa; whereas, the highest DHMTHF selectivity of

96% was obtained at 1.0 MPa. The HT selectivity slowly enhanced from trace at 0.1 MPa to 8% at 2.0 MPa. The reusability of the Pd/mpg-C3 N4 catalyst was also investigated for HMF hydrogenation by conducting five consecutive experiments. The Pd/mpg-C3 N4 was washed completely with methanol–water and then re-examined for the next cycles. The DHMTHF selectivity slightly dropped from 96% to 89% with complete HMF conversion after a five-cycle experiment (Fig. 7). Accordingly, the DHMF selectivity raised from 3% to 9%. Inductively coupled plasma–atomic emission spectroscopy (ICP–AES) analysis of the aqueous solution after each cycle did not show even traces of leached Pd, suggesting an efficient interaction between mpg-C3 N4 support and palladium nanoparticles.

322

4. Conclusions

323

In summary, Pd/mpg-C3 N4 shows excellent catalytic performance for the selective hydrogenation of HMF into DHMTHF in aqueous medium under pressured hydrogen atmosphere. The excellent catalytic performance of Pd/mpg-C3 N4 is attributed to hydrogen bonding-related competitive interactions between reactant HMF and “indermediate” DHMF with the support mpg-C3 N4 , which leads to a further hydrogenation of DHMF to DHMTHF.

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

310 311 312 313 314 315 316 317 318 319 320 321

324 325 326 327 328 329 330

ARTICLE IN PRESS

JID: JECHEM

[m5G;April 28, 2017;11:18]

J. Chen et al. / Journal of Energy Chemistry xxx (2017) xxx–xxx

DHMF;

Product selectivity / %

100

DHMTHF;

HT

3

4

8

9

96

95

93

89

89

1

2

3

4

5

80

60

40

20

0

Run Fig. 7. The reusability of Pd/mpg-C3 N4 . Selectivity to DHMF (white), DHMTHF (grey), and HT (black) as a function of recycling times for the hydrogenation of HMF over Pd/mpg-C3 N4 . Reaction conditions: HMF (63 mg, 0.5 mmol), Pd/mpg-C3 N4 (30 mg, Pd 4.1 wt%), water (3 mL), reaction time (4.0 h, H2 (1.0 MPa)).

331

Acknowledgments

332

337

The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China (21472189), Natural Science Foundation of Guangdong Province, China (2015A030312007), Science and Technology Planning Project of Guangzhou City, China (201707010238), and Jinan Double Hundred Talents Plan.

338

Supplementary materials

339 340

Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.jechem.2017.04.017.

341

References

333 334 335 336

342 343 344 345 346 347 348 349 350 351 352

[1] G.H. Wang, J. Hilgert, F.H. Richter, F. Wang, H.J. Bongard, B. Spliethoff, C. Weidenthaler, F. Schüth, Nat. Mater. 13 (2014) 293–300. [2] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558. [3] R.J. van Putten, J.C. van der Waal, J.G. de Vries, C.B. Rasrendra, H.J. Heeres, J.G. de Vries, Chem. Rev. 113 (2013) 1499–1597. [4] T.F. Wang, M.W. Nolte, B.H. Shanks, Green Chem. 16 (2014) 548–572. [5] G.J.S. Dawes, E.L. Scott, J. Le Notre, J.P.M. Sanders, J.H. Bitter, Green Chem. 17 (2015) 3231–3250. [6] J. Reichert, B. Brunner, A. Jess, P. Wasserscheid, J. Albert, Energy Environ. Sci. 8 (2015) 2985–2990. [7] Y.Z. Wang, S. De, N. Yan, Chem. Commun. 52 (2016) 6210–6224.

7

[8] Y. Román-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, Nature 447 (2007) 982– 985. [9] M. Dashtban, A. Gilbert, P. Fatehi, RSC Adv. 4 (2014) 2037–2050. [10] S.P. Teong, G.S. Yi, Y.G. Zhang, Green Chem. 16 (2014) 2015–2026. [11] Y. Nakagawa, K. Tomishige, Catal. Commun. 12 (2010) 154–156. [12] R. Alamillo, M. Tucker, M. Chia, Y. Pagán-Torres, J. Dumesic, Green Chem. 14 (2012) 1413–1419. [13] M.H. Tucker, R. Alamillo, A.J.G. Crisci, M. Gonzalez, S.L. Scott, J.A. Dumesic, ACS Sustainable Chem. Eng. 1 (2013) 554–560. [14] C. Moreau, M.N. Belgacem, A. Gandini, Top. Catal. 27 (2004) 11–30. [15] T. Buntara, S. Noel, P.H. Phua, I. Melian-Cabrera, J.G. de Vries, H.J. Heeres, Angew. Chem., Int. Ed. 50 (2011) 7083–7087. [16] S. Koso, N. Ueda, Y. Shinmi, K. Okumura, T. Kizuka, K. Tomishige, J. Catal. 267 (2009) 89–92. [17] M. Chia, Y.J. Pagan-Torres, D. Hibbitts, Q. Tan, H.N. Pham, A.K. Datye, M. Neurock, R.J. Davis, J.A. Dumesic, J. Am. Chem. Soc. 133 (2011) 12675–12689. [18] T. Buntara, S. Noel, P.H. Phua, I. Melián-Cabrera, J.G. de Vries, H.J. Heeres, Top. Catal. 55 (2012) 612–619. [19] K. Chen, K. Mori, H. Watanabe, Y. Nakagawa, K. Tomishige, J. Catal. 294 (2012) 171–183. [20] S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa, K. Tomishige, Appl. Catal. B 111-112 (2012) 27–37. [21] Y.L. Yang, Z.T. Du, J.P. Ma, F. Lu, J.J. Zhang, J. Xu, ChemSusChem. 7 (2014) 1352– 1356. [22] J.Z. Chen, R.L. Liu, Y.Y. Guo, L.M. Chen, H. Gao, ACS Catal. 5 (2015) 722–733. [23] J.W. Zhong, J.Z. Chen, L.M. Chen, Catal. Sci. Technol. 4 (2014) 3555–3569. [24] Y. Wang, J. Yao, H.R. Li, D.S. Su, M. Antonietti, J. Am. Chem. Soc. 133 (2011) 2362–2365. [25] X. Xu, H.R. Li, Y. Wang, ChemCatChem. 6 (2014) 3328–3332. [26] Z.L. Li, J.H. Liu, C.G. Xia, F.W. Li, ACS Catal. 3 (2013) 2440–2448. [27] J.Z. Chen, W. Zhang, L.M. Chen, L.L. Ma, H. Gao, T.J. Wang, ChemPlusChem. 78 (2013) 142–148. [28] A.B. Chen, Y.L. Li, J.Z. Chen, G.Y. Zhao, L.L. Ma, Y.F. Yu, ChemPlusChem. 78 (2013) 1370–1378. [29] X.-H. Li, X.C. Wang, M. Antonietti, Chem. Sci. 3 (2012) 2170–2174. [30] Y.T. Gong, P.F. Zhang, X. Xu, Y. Li, H.R. Li, Y. Wang, J. Catal. 297 (2013) 272–280. [31] Z.L. Li, J.L. Li, J.H. Liu, Z.L. Zhao, C.G. Xia, F.W. Li, ChemCatChem 6 (2014) 1333– 1339. [32] D.S. Deng, Y. Yang, Y.T. Gong, Y. Li, X. Xu, Y. Wang, Green Chem. 15 (2013) 2525–2531. [33] X. Xu, M.H. Tang, M.M. Li, H.R. Li, Y. Wang, ACS Catal. 4 (2014) 3132–3135. [34] L.-H. Gong, Y.-Y. Cai, X.-H. Li, Y.-N. Zhang, J. Su, J.-S. Chen, Green Chem. 16 (2014) 3746–3751. [35] Y.J. Gao, X. Chen, J.G. Zhang, N. Yan, ChemPlusChem 80 (2015) 1556–1564. [36] Y.J. Zhang, A. Thomas, M. Antonietti, X.C. Wang, J. Am. Chem. Soc. 131 (2009) 50–51. [37] F. Su, M. Antonietti, X.C. Wang, Catal. Sci. Technol. 2 (2012) 1005–1009. [38] X.C. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J.M. Carlsson, K. Domen, M. Antonietti, Nat. Mater. 8 (2009) 76–80. [39] H.F. Shi, G.Q. Chen, C.L. Zhang, Z.G. Zou, ACS Catal. 4 (2014) 3637–3643. [40] F. Goettmann, A. Fischer, M. Antonietti, A. Thomas, Angew. Chem., Int. Ed. 45 (2006) 4467–4471. [41] A. Thomas, A. Fischer, F. Goettmann, M. Antonietti, J.-O. Müller, R. Schlögl, J.M. Carlsson, J. Mater. Chem. 18 (2008) 4893–4908. [42] E. Raymundo-Piñero, D. Cazorla-Amorós, A. Linares-Solano, J. Find, U. Wild, R. Schlögl, Carbon 40 (2002) 597–608. [43] C. Chang, Y. Fu, M. Hu, C.Y. Wang, G.Q. Shan, L.Y. Zhu, Appl. Catal. B 142-143 (2013) 553–560. [44] F. He, G. Chen, Y.G. Yu, S. Hao, Y.S. Zhou, Y. Zheng, ACS Appl. Mater. Interfaces 6 (2014) 7171–7179. [45] J. Horvat, B. Klaic, B. Metelko, V. Sunjic, Tetrahedron Lett. 26 (1985) 2111–2114. [46] Y. Nakagawa, K. Takada, M. Tamura, K. Tomishige, ACS Catal. 4 (2014) 2718– 2726.

Please cite this article as: J. Chen et al., Selective hydrogenation of biomass-derived 5-hydroxymethylfurfural using palladium catalyst supported on mesoporous graphitic carbon nitride, Journal of Energy Chemistry (2017), http://dx.doi.org/10.1016/j.jechem.2017.04.017

353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415