Water-Tolerant Mesoporous-Carbon-Supported Ruthenium Catalysts for the Hydrolysis of Cellulose to Glucose



original image

Best Supporting Role: Ru/CMK-3 is a water-tolerant and reusable catalyst for the hydrolysis of cellulose, and exhibits high glucose yields and turnover numbers due to a synergistic effect between CMK-3 and Ru. CMK-3 contributes to the conversion of cellulose into oligosaccharides, while Ru promotes the hydrolysis of oligosaccharides to glucose.

The conversion of renewable biomass into useful chemicals is one of the most important subjects in green and sustainable chemistry.1 To avoid conflicts with food production, as seen in the last few years, nonfood biomass should be used as a resource for the production of chemicals. Cellulose is the most abundant nonfood biomass resource produced via photosynthesis,2 and therefore the conversion of cellulose has attracted significant attention as a key issue in the utilization of biomass.3 Cellulose is a water-insoluble polymer composed of glucose units linked by β-1,4-glycosidic bonds.2, 4 The hydrolysis of cellulose produces glucose, which is a versatile precursor to fuels, plastics, pharmaceuticals, and other value-added chemicals. However, the efficient hydrolysis of cellulose to glucose with low environmental impact remains a challenge (Scheme 1).

Scheme 1.

Conversion of cellulose into glucose via soluble oligomers.

Among the known methods for the hydrolysis of cellulose, sulfuric acid can be used as an active homogeneous catalyst;5 however, liquid sulfuric acid is highly corrosive and neutralization is required to dispose of the acid after reaction. Cellulase enzymes can be used to selectively hydrolyze cellulose to glucose;6 however, the drawbacks of this enzymatic method are a low reaction rate, high costs, and the difficulty of recovering the enzyme from the reaction mixtures. Sub- or supercritical water has also been applied to the hydrolysis of cellulose, but the selectivity of this method for glucose is low due to the thermal instability of glucose at high temperature.7 For these reasons, heterogeneous solid catalysts would be favorable choices for cellulose hydrolysis, because solid catalysts can be applied under a wide range of conditions, easily separated from the reaction mixture, and reused in repeated reactions.5

Since we reported the conversion of cellulose into sorbitol by supported Pt and Ru catalysts under H2 pressure,8 other groups have also reported the degradation of cellulose to sorbitol or glycols using various supported metal catalysts.9 Sulfonated carbons or resins have also been used for cellulose hydrolysis;10 however, the turnover numbers (TONs) for glucose formation were as low as 1.6–5.1 and sulfonic groups might hydrolyze and leach into the water solvent by elevation of the temperature to enhance catalytic activity.

In our research on the synthesis of sorbitol from cellulose, the formation of glucose as an intermediate was observed, and this motivated this study of cellulose hydrolysis by supported metal catalysts without the use of sulfonic acid groups. Mesoporous carbon materials (CMKs),11 with large surface areas and highly oxygen-functionalized surfaces, are chosen as a support, and Ru/CMK-3 is found to be a water-tolerant and reusable catalyst for hydrolysis of cellulose to glucose with high yield.

The N2 adsorption results for CMK-3 and Ru/CMK-3 are presented in Table 1. The Brunauer–Emmett–Teller (BET) surface area of Ru/CMK-3 was 1100 m2 g−1 and the pore diameter, calculated with the Barrett–Joyner–Halenda (BJH) method, was 3.8 nm, which were almost the same as those obtained for CMK-3 (1120 m2 g−1, 3.8 nm, respectively). The small-angle X-ray scattering (SAXS) patterns of Ru/CMK-3 and CMK-3 showed similar scattering peaks for (100) at 2θ=1.0° (Figure 1), which indicated that the structure of CMK-3 was retained during the preparation of Ru/CMK-3. The XRD patterns of Ru/CMK-3 and CMK-3 displayed broad signals of turbostratic carbon at 24° and 43° (Supporting Information, Figure S1),12 so that peaks from Ru crystals could not be detected. A transmission electron microscopy (TEM) image of Ru/CMK-3, shown in Figure 2 confirmed the ordered mesoporous structure of CMK-3, although Ru particles were not observed. From these characterization data, it is suggested that Ru is not fully reduced to form zero-valent nanoparticles but highly dispersed on the surface of CMK-3, without plugging of the CMK-3 pores.

Table 1. Textural properties of CMK-3 and Ru/CMK-3, by N2 adsorption at 77 K.
EntryCatalystBET surface area [m2 g−1]BJH pore diameter [nm]Pore volume [cm3 g−1]
  1. [a] After hydrolysis of cellulose.

22 wt % Ru/CMK-311003.81.37
32 wt % Ru/CMK-3[a]11003.81.40
Figure 1.

SAXS patterns for CMK-3 and 2 wt % Ru/CMK-3.

Figure 2.

TEM image of 2 wt % Ru/CMK-3.

Table 2 summarizes the results of cellulose hydrolysis by using carbon-supported Ru catalysts. To avoid the further degradation of glucose formed at elevated temperatures, the reactor was rapidly cooled from 503 K to 298 K (Figure S2).13 The selectivity for glucose decreased and those for byproducts increased when the reaction temperature was kept at 503 K for more than 1 min. Ru/CMK-1, Ru/XC-72 (carbon black), Ru/AC (activated carbon), and Ru/C60 were also used, to allow a comparison of the catalytic activities (Table 2, entries 3, 6–10). Among these catalysts, Ru/CMK-3 had the highest cellulose conversion and glucose yield. In entry 3, the major products were glucose (yield 28 %) and soluble oligosaccharides (total 15 %: dimer 7.7 %, trimer 4.4 %, tetramer 2.1 %, pentamer and hexamer <0.5 %). The TON based on bulk Ru for glucose formation was 52, which is one order of magnitude higher than those of previously reported solid sulfonated acid catalysts (TON based on SO3H: 1.6–5.1 for 24–27 h).10 Other minor products were fructose (2.0 %), mannose (1.3 %), levoglucosan (1.6 %), 5-hydroxymethylfurfural (5-HMF; 3.0 %), and furfural (0.5 %). The conversion of cellulose was 59 %, which was determined from the weight loss of the solid residue after reaction. Accordingly, the selectivity for glucose and oligosaccharides was 72 %, based on the cellulose conversion. Ru/CMK-1 also showed a good yield of glucose (27 %), whereas Ru/XC-72 (16 %), Ru/AC(W) (11 %), Ru/AC(N) (14 %), and Ru/C60 (5.0 %) were less active than the Ru/CMK catalysts. Likewise, CMK-1-supported Pt and Pd catalysts were also used for the reaction, but the yields of glucose (13 %) were lower than that for the Ru/CMK-1 catalyst. In addition, a small amount of glucose (4.6 %) was formed without catalysts, which is due to the formation of H3O+ in the hot compressed water.

Table 2. Hydrolysis of cellulose by carbon-supported Ru catalysts.[a]
EntryCatalystYield based on carbon [%]Cellulose conv.[b] [%]
  1. [a] Cellulose, 324 mg; catalyst, 50 mg (metal, 2 wt %); water, 40 mL, 503 K (Figure S2). Selectivity based on the cellulose conversion is shown in parentheses. [b] Cellulose conversion is calculated from the weight loss of the solid residue after reaction. [c] Dimer–octamer. [d] 5-Hydroxymethylfurfural. [e] Dimer, 3.0 %; trimer, 3.4 %; tetramer, 3.2 %; pentamer, 1.7 %; hexamer, 1.6 %; heptamer, <1 %. [f] Dimer, 8.4 %; trimer, 6.9 %; tetramer, 4.2 %; pentamer, 2.0 %; hexamer, <1 %. [g] Dimer, 7.7 %; trimer, 4.4 %; tetramer, 2.1 %; pentamer and hexamer, <0.5 %. [h] 5 wt % Ru. [i] 10 wt % Ru. [j] Carbon black Vulcan XC-72 (Cabot). [k] Activated carbon (Wako). [l] Activated carbon SX Ultra (Norit).

1None4.6 (19)14.1[e] (58)
2CMK-320.5 (38)22.1[f] (41)
3Ru/CMK-327.6 (47)14.8[g] (25)
4Ru/CMK-3[h]29.0 (47)10.3 (17)
5Ru/CMK-3[i]34.2 (51)5.1 (8)
6Ru/CMK-126.8 (52)10.6 (20)
7Ru/XC-72[j]15.6 (33)21.8 (46)
8Ru/AC(W)[k]11.3 (29)18.8 (49)
9Ru/AC(N)[l]13.7 (34)19.9 (50)
10Ru/C605.0 (18)15.9 (58)
11Pt/CMK-113.3 (35)21.1 (55)
12Pd/CMK-113.2 (31)16.5 (38)

Re-use experiments of 2 wt % Ru/CMK-3 were performed. In these experiments, the catalyst was separated by centrifugation of the reaction mixture and re-used three times without any treatment. The yields of glucose, which were calculated based on the added fresh cellulose in each run as the residual cellulose in the last run was less reactive, were 25–27 % in these experiments (Table S1), and the total TON based on bulk Ru in the experiments was 145. In addition, no change was observed for the N2 adsorption analysis of Ru/CMK-3 before and after the catalytic reaction, as shown in Table 1. The XRD pattern of the catalyst also remained the same after the reaction (Figure S1). These results demonstrate that Ru/CMK-3 is a reusable catalyst for the hydrolysis of cellulose in water. Oxide-supported Ru catalysts were also tested for this reaction (Table S2). Although a Ru/Al2O3 catalyst showed a good yield of glucose (25 %), the catalyst deactivated in the re-use experiments due to a structural transformation of Al2O3 into boehmite [AlO(OH); Figure S3].

In order to obtain information regarding the role of the support and metal, CMK-3 and 2–10 wt % Ru/CMK-3 catalysts were employed for the hydrolysis of cellulose (Table 2, entries 2–5). When CMK-3 was used as a catalyst, the yield of glucose was 21 % and that of oligosaccharides was 22 % (total 43 %). It was noted that CMK-3 itself catalyzed the hydrolysis of cellulose in water at 503 K. In the presence of Ru on CMK-3, the yield of glucose increased from 21 % to 28 % (Ru 2 wt %), 29 % (Ru 5 wt %), and 34 % (Ru 10 wt %), whereas the yield of oligosaccharides decreased from 22 % to 5 % by increasing the Ru loading. Accordingly, the total yield of sugars (glucose plus oligosaccharides) was ca. 40 %, independent of the Ru loading. The conversion of cellulose was slightly improved by increasing the content of Ru. Although both CMK-3 and Ru are active for the hydrolysis reaction, the main role of CMK-3 is to convert cellulose into oligosaccharides, and that of Ru is to transform oligosaccharides into glucose (Scheme 2); therefore, the synergistic effect of CMK-3 and Ru achieves a good yield of glucose and high TONs.

Scheme 2.

Reaction scheme for the hydrolysis of cellulose to glucose.

To clarify the role of Ru, the hydrolysis of cellobiose was performed as a test reaction using CMK-3 and 5 wt % Ru/CMK-3 as catalysts. The yield of glucose without a catalyst was 8 % and CMK-3 did not improve the glucose yield (7 %), as shown in Table 3. It is noticeable that the hydrolysis of cellulose is accelerated by CMK-3, which would be due to the higher interaction of CMK-3 with cellulose than that with cellobiose.10c In contrast, Ru/CMK-3 gave a markedly higher glucose yield (25 %) than CMK-3, which indicates that Ru catalyzes the hydrolysis of β-1,4-glycosidic bonds of soluble cellobiose. This result suggests that Ru facilitates the acid-catalyzed hydrolysis of cellulose in water under the reaction conditions.

Table 3. Hydrolysis of cellobiose by the Ru/CMK-3 catalyst.[a]
EntryCatalystYield based on carbon [%]
  1. [a] Cellobiose, 342 mg (2 mmol); catalyst, 50 mg; water, 40 mL; 393 K, 24 h.

35 wt % Ru/CMK-324.90.70.9026.5

In conclusion, Ru/CMK-3 is a water-tolerant and reusable catalyst for the hydrolysis of cellulose that provides high glucose yields and TONs. A synergistic effect between CMK-3 and Ru is observed, in which the main role of CMK-3 is the hydrolysis of cellulose to oligosaccharides while Ru promotes the conversion of oligosaccharides into glucose.

Experimental Section

CMK-1 and −3 were synthesized according to literature procedures.11 Activated carbons were purchased from Wako [activated charcoal, denoted as AC(W)] and Aldrich [Norit SX Ultra, denoted as AC(N)]. Carbon black was supplied by Cabot (XC-72) and C60 (fullerene) by Hoechst.

Ru/CMK-3 catalysts were prepared by a conventional impregnation method as follows: RuCl3 aq. (0.202 mmol in 5 mL water; Ru metal loading 2 wt %) was added dropwise to a mixture of CMK-3 (1.00 g) and water (20 mL), and the mixture was stirred for 16 h. After drying in vacuo, the solid was reduced in a fixed-bed flow reactor with H2 (30 mL min−1) at 673 K for 2 h. No methanation of the carbon support was observed during the H2 reduction.

Hydrolysis of cellulose was carried out in a hastelloy high-pressure reactor (OM Lab-Tech MMJ-100, 100 mL). Cellulose (Merck, Avicel) was milled using ZrO2 balls at 60 rpm for 4 days. The milled cellulose (324 mg, 1.86 mmol glucose-units, containing 7.0 wt % physisorbed water), Ru/CMK-3 (50 mg), and water (40 mL) were charged in the reactor (glucose-unit/Ru=186), and the mixture was heated from 298 to 503 K in 15 min with stirring at 600 rpm, and then rapidly cooled to 298 K (Figure S1). The products were separated by centrifugation and decantation, and water-soluble products were analyzed by using high-performance liquid chromatography (HPLC; Shimadzu LC10-ATVP or Hitachi LaChrom Elite, refractive index detector). The columns used in this work were a Phenomenex Rezex RPM-Monosaccharide Pb++ column (ø 7.8×300 mm, mobile phase: water 0.6 mL min−1, 353 K), a Shodex Sugar SH-1011 column (ø 8×300 mm, mobile phase: water 0.5 mL min−1, 323 K), and a TSKgel Amide-80 (ø 4.6×250 mm, mobile phase: acetonitrile and water (60:40) 0.8 mL min−1, 313 K).

N2 adsorption measurements (Belsorp-mini II) were performed at 77 K. X-ray diffraction (XRD; Rigaku Miniflex) and small-angle X-ray scattering (SAXS; Rigaku RINT 2000) measurements were conducted using Cu Kα radiation. Transmission electron microscopy (TEM; JEOL JEM-2000ES) images were obtained at an accelerating voltage of 200 kV.


This work was supported by a Grant-in-Aid (KAKENHI, 20226016) from the Japan Society for the Promotion of Science (JSPS). The authors would like to thank Prof. W. Ueda for the SAXS measurements, Dr. M. Lin for helpful advice on the synthesis of CMK, and Mr. K. Kasai for performing preliminary experiments.