Bioresource Technology 167 (2014) 484–489
Contents lists available at ScienceDirect
Bioresource Technology journal homepage: www.elsevier.com/locate/biortech
Hydrolysis of ionic cellulose to glucose Huyen Thanh Vo a,b, Vania Tanda Widyaya a, Jungho Jae a,b, Hoon Sik Kim c, Hyunjoo Lee a,b,⇑ a
Clean Energy Research Center, Korea Institute of Science and Technology, Seoul 136-791, South Korea University of Science and Technology, Deajeon 305-355, South Korea c Department of Chemistry, Kyung Hee University, Seoul 130-701, South Korea b
h i g h l i g h t s
g r a p h i c a l a b s t r a c t
Water-soluble ionic cellulose was
synthesized from cellulose and ionic liquid. Hydrolysis of water-soluble cellulose generated glucose in aqueous media. Sulfonated active carbon showed high catalytic activity and good reusability.
a r t i c l e
i n f o
Article history: Received 24 March 2014 Received in revised form 5 June 2014 Accepted 8 June 2014 Available online 24 June 2014 Keywords: Cellulose Water-soluble Hydrolysis Ionic liquid Glucose
a b s t r a c t Hydrolysis of ionic cellulose (IC), 1,3-dimethylimidazolium cellulose phosphite, which could be synthesized from cellulose and dimethylimidazolium methylphosphite ([Dmim][(OCH3)(H)PO2]) ionic liquid, was conducted for the synthesis of glucose. The reaction without catalysts at 150 °C for 12 h produced glucose with 14.6% yield. To increase the hydrolysis yield, various acid catalysts were used, in which the sulfonated active carbon (AC-SO3H) performed the best catalytic activity in the IC hydrolysis. In the presence of AC-SO3H, the yields of glucose reached 42.4% and 53.9% at the reaction condition of 150 °C for 12 h and 180 °C for 1.5 h, respectively; however the yield decreased with longer reaction time due to the degradation of glucose. Consecutive catalyst reuse experiments on the IC hydrolysis demonstrated the catalytic activity of AC-SO3H persisted at least through four successive uses. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Cellulose, the most abundant biopolymer in nature, attracts much interest as a resource for biofuels like bioethanol and biobutanol as well as for platform chemicals such as 5-hydroxymethylfurfural (HMF) and levulinic acid (LA). In the conversion of cellulose to fuels or chemicals, the ﬁrst step is depolymerization of cellulose to its monomeric compound, glucose, via hydrolysis. ⇑ Corresponding author at: Clean Energy Research Center, Korea Institute of Science and Technology, Seoul 136-791, South Korea. Tel.: +82 2958 5868; fax: +82 2958 5809. E-mail address: [email protected]
(H. Lee). http://dx.doi.org/10.1016/j.biortech.2014.06.025 0960-8524/Ó 2014 Elsevier Ltd. All rights reserved.
The major obstacles for the hydrolysis of cellulose in mild conditions are the high crystallinity of cellulose resulting from strong inter- and intra-molecular hydrogen bonds and the low solubility of cellulose in water and common organic solvents. Homogeneous catalysts such as mineral acids (Camacho et al., 1996) and enzymes (Katz and Reese, 1968) demonstrate superior hydrolysis activity to heterogeneous catalysts due to their easier accessibility to the reaction center. However, homogeneous catalysts have many disadvantages such as reaction system corrosion, waste recycle expense for mineral acids and cause of the undesired glucose degradation products like HMF in enzymatic processes. Heterogeneous catalysts have been widely studied for cellulose hydrolysis due to their advantages in separation and catalytic
H.T. Vo et al. / Bioresource Technology 167 (2014) 484–489
activity. These catalysts include sulfonated active carbon (Suganuma et al., 2008; Toda et al., 2005), polymer based acids (Rinaldi et al., 2010), and layer-transition metal oxides (Takagaki et al., 2008; Zhang and Fang, 2012). Among these materials, sulfonated active carbon showed superior catalytic activity for cellulose hydrolysis. In a recent report, the hydrolysis of microcrystalline cellulose (MCC) with sulfonated activated carbon achieved 64% of total reducing sugars (TRS) yield, with 4% of glucose yield at 100 °C for 6 h (Suganuma et al., 2008). Layer-transition metal oxides have also been used as acid catalysts for the hydrolysis of biopolymers. Proton-containing transition metal oxide, HNbMoO6, exhibited a remarkable performance for the hydrolysis of cellobiose and starch (Takagaki et al., 2008). This catalyst achieved 20% glucose yield in the hydrolysis of starch at 100 °C for 15 h. However, for the hydrolysis of cellulose, the catalytic activity of HNbMoO6 was meager due to the low density of Bronsted acid site as well as low surface area. In heterogeneous reaction, due to the water insoluble characteristic of cellulose, only the exterior acid sites of the catalysts can be accessible to cellulose and this leads to low hydrolysis efﬁciency. To enhance the accessibility of water molecules or catalyst sites on the hydrolysis centers of cellulose chains, various kinds of pretreatment have been employed, for example, ball milling (Onda et al., 2008) and ionic liquid (Li and Zhao, 2007) treatment. Ionic liquids (IL) can be used either as a pretreatment reagent that destroys the cellulose crystalline structure before hydrolysis or as a reaction solvent in which dissolution and hydrolysis of cellulose occur simultaneously. For instance, by using ILs (chloride anionic ILs such as 1-butyl-3-methylimidazolium chloride ([Bmim][Cl]))and acid catalyst, the conversion of cellulose to sugars could be increased even under mild reaction conditions. Li et al. obtained about 77% of TRS yield and maximum 43% glucose yield by using [Bmim][Cl] and mineral acids (H2SO4 and HCl) at 100 °C for 9 h (Li and Zhao, 2007). At the catalytic system of [Bmim][Cl] and solid acid catalyst, about 28% of TRS yield was achieved at 100 °C for 5 h with Amberlyst-15 DRY resin (Rinaldi et al., 2008) and 69% of TRS yield was produced at 130 °C for 3 h with sulfonated active carbon (Guo et al., 2012). Although the conversion of cellulose could be increased much by using ionic liquids as a solvent for the hydrolysis, the use of corrosive mineral acid and/or low glucose yields is the limitation of these methods. Recently, we reported that the reaction of cellulose with dimethylimidazolium methylphosphite ([Dmim][(OCH3)(H)PO2]) formed ionic cellulose, 1,3-dimethylimidazolium cellulose phosphite (Scheme S1, Supplementary data), which can be dissolved in water due to the ionic structure (Vo et al., 2012). In this paper, we studied the hydrolysis of this ionic cellulose dissolved in water to obtain glucose using various kinds of acid catalyst (Scheme S2, Supplementary data). We also used other water-soluble cellulose derivatives for hydrolysis and compared the TRS yields with our ionic cellulose.
(Kim et al., 2010). Carboxymethylcellulose (CMC, H-form) was synthesized by the neutralization of CMC-Nawithphosphoric acid. This process was described in Supplementary data (Fig. S1, Supplementary data). Catalysts for hydrolysis like phosphorous acid, sulfuric acid, dry Amberlyst-15 (4.8 mmol/g acidity) and Naﬁon-NR50 (0.9 mmol/g acidity) were obtained from the Sigma–Aldrich Chemical Co. Activated carbon was received from Strem Chemicals Inc. 2.2. Preparation of ionic cellulose Microcrystalline cellulose (0.5 g) and [Dmim][(OCH3)(H)PO2] (5 g) were loaded into 25 mL one-necked round bottom ﬂask. The mixture was then heated to 120 °C in an oil bath and allowed to react for 1 h. After the reaction, the mixture was cooled to room temperature and diluted with 10 mL of water. To this diluted solution was dropped with 20 mL of acetonitrile to produce the precipitates. The resulting precipitates were ﬁltered and washed with acetonitrile at least three times, and ﬁnally, vacuum dried at room temperature overnight to give the desired ionic cellulose. The degree of substitution of phosphorous in ionic cellulose was calculated based on the phosphorous content as described in previous literature (Vo et al., 2012). Ionic cellulose obtained has the degree of substitution of phosphorous of 0.36 which corresponded to the molecular weight of ionic glucose unit of 219 g/mol. 2.3. Hydrolysis of ionic cellulose Typically, 0.1 g of ionic cellulose dissolved in 5 mL of distilled water and catalyst were introduced into a sealed pressure glass tube (Ace, 15 mL, pressure limit is 20 bar). The glass tube was placed in an oil bath which was maintained at 150 °C for the desired reaction time. After the reaction, sample was cooled to room temperature and centrifuged at a rate of 10,000 rpm for 10 min to separate the solid and liquid products. For the product analyses, solid product was washed with water and dried under vacuum for 12 h, and was then characterized by FT-IR. Aqueous solution was analyzed using HPLC (Younglin 9100) equipped with RI detector (YL9170) and column (Shodex SUGAR-KS802, 8.0 300 (mm) ID) which was maintained at 80 °C. The mobile phase was deionized water at a ﬂow rate of 0.6 mL/min. The total reducing sugar (TRS) in the liquor samples were analyzed by DNS method (Supplementary data) (Li and Zhao, 2007; Miller, 1959).The concentration of cellobiose, glucose, fructose, levulinic acid and HMF were calculated based on the standard curve obtained with known concentrations of the substances. The yields of each component were calculated based on the total glucose unit in the ionic cellulose used as follows. Yield of glucose ð%Þ ¼ 100 ½glucose produced ðmolÞ=IC used ðgÞ=219ðmol=gÞ
2.4. Preparation and characterization of sulfonated carbon material 2. Experimental 2.1. Materials Microcrystalline cellulose (MCC) and other cellulose derivatives, such as cellulose acetate, ethyl cellulose, methyl cellulose, sodiumcarboxymethyl cellulose (CMC-Na) and hydroxyethyl cellulose, were purchased from Sigma–Aldrich Chemical Co. 1,3dimethylimidazolium methylphosphite ([Dmim][(OCH3)(H)PO2]) was prepared according to our recent report (Vo et al., 2012). Solvents were purchased from J.T. Baker and used as received. Decrystalized cellulose (DC) was synthesized by dissolving MCC in [Bmim]Cl (10 wt%) at 130 °C for 2 h as described in literature
Sulfonated active carbon was prepared as described in the reference (Suganuma et al., 2008). The activated carbon powder (1 g) was stirred with concentrated sulfuric acid (96%, 20 mL) at 150 °C for 24 h under N2 gas with ﬂow rate of 50 mL/min. After cooling to the room temperature, the black solid was repeatedly washed with distilled-water (3 L). The solid was then hydrothermally treated in a bomb-reactor at 150 °C for 3 h for the complete removal of H2SO4 physically adsorbed on the carbon material. The solid was washed again with water until pH of solution was about 6–7. Finally, the sulfonated active carbon was dried under the vacuum at 70 °C for 12 h. The FT-IR spectrum of synthesized carbon appeared the vibration bands at 1372 cm1 (S=O stretching) and
H.T. Vo et al. / Bioresource Technology 167 (2014) 484–489
1025 cm1 (SO3-stretching) giving the evidence of the presence of –SO3H groups in the resulting material (Fig. S2, Supplementary data). Sulfur content of the catalyst was determined to be 0.80 wt% by CHNOS elemental analysis, corresponding to 0.25 mmol of SO3H/g of carbon. The BET surface area of catalyst was 1175 m2/g. 2.5. Instrument The catalyst was characterized by FT-IR, elemental analysis, and BET techniques. Infrared spectra were obtained by using Nicolet FT-IR spectrometer (iS10, USA) equipped with a SMART MIRACLE accessory. C, H, N, O, and S contents of the catalyst were characterized by CHNOS elemental analyzer (Model: Fisons EA 1108). The Brunauer–Emmet–Teller (BET) surface area was determined by a Belsorp-mini II instrument (BEL Inc., Japan). Products were analyzed by 1H-NMR recorded using Bruke Avance 400. 3. Results and discussions 3.1. Hydrolysis of ionic cellulose The hydrolysis of ionic cellulose (IC) was conducted in the presence of various kinds of catalyst as shown in Table 1. In a typical reaction, IC (100 mg, 0.45 mmol of glucose unit) dissolved in water (5 ml) and the catalyst (0.025 mmol of H+) was added in a sealed pressure glass tube and heated at 150 °C for 12 h. After the reaction, the product solution was analyzed by HPLC. The results revealed that not only glucose but also cellobiose, 5-HMF and small amounts of levoglucosan and levulinic acid were formed together. 3.1.1. Hydrolysis in the absence of catalyst In the absence of catalyst, the hydrolysis of ionic cellulose produced cellobiose and glucose with yields of 8.4% and 14.6%, respectively (Table 1, Entry 1), indicating that the hydrolysis of IC in water happened substantially even without additional catalysts. The formation of glucose from the hydrolysis of IC without catalysts could be ascribed to the acidic ionic compound [Dmim][(OH)(H)PO2] (pKa = 2.9), which was separated from the IC, as shown in Scheme 1. At 150 °C, hydrolysis of the C–O–P bond in IC could also happen to generate [Dmim][(OH)(H)PO2]. 1H-NMR spectrum of hydrolysis product in Fig. S3 (Supplementary data) veriﬁed the existence of [Dmim][(OH)(H)PO2] along with the produced saccharides. In fact, the addition of [Dmim][(OH)(H)PO2] increased the yield of glucose up to 23.5% (Table 1, Entry 2), however, compared to other sulfonic acid-based catalysts, the catalytic activity was not high due to its relatively weak acidity. Besides [Dmim][(OH)(H)PO2], we also observed some white precipitates after the reaction. IR and XRD analyses revealed that the precipitates were decrystallized cellulose (DC) (Figs. S4 and
Table 1 Ionic cellulose hydrolysis using different catalysts.a Entry
1 2 3 4 5 6 7
– [Dmim][(OH) (H)PO2] H3PO3 H2SO4 AC-SO3H Amberlyst-15 Naﬁon-NR50
Yield (%) H+/GUb
0 4.0 0.05 0.05 0.05 0.05 0.05
24.6 30.6 32.0 36.8 52.0 42.5 30.3
8.4 5.6 1.5 1.6 5.6 4.0 0.3
14.6 23.5 28.9 30.4 42.5 31.8 27.8
1.7 5.4 2.6 1.8 0.5 2.4 1.6
a Reaction condition: cellulose derivative (100 mg), water (5 mL) and acid catalyst (0.025 mmol), 150 °C, 12 h. b H+/GU is the molar ratio of H+ (acid catalyst) to glucose unit in IC.
S5 in Supplementary data). Furthermore, no glucose or cellobiose having phosphite groups were detected after the reaction. These results suggest, in the absence of extra catalyst, the hydrolysis of C–O–P bond occurred faster than that of glucosidic bond in the hydrolysis of IC. 3.1.2. Effect of acid catalysts Various homogeneous and solid acid catalysts were tested for the reaction. The molar ratio of the H+ in acid catalyst to the glucose unit (GU) in IC was set to 0.05. Polyprotic acids such as H3PO3 and H2SO4 were regarded as monoprotic acids due to the decreased acidity from the second proton. As shown in Table 1, the H3PO3 (pKa1 = 2.0) and H2SO4 (pKa1 = 3.0) produced glucose with yields of 28.9% and 30.4%, respectively, which, as expected, were two times higher than that produced from the reaction conducted in the absence of catalysts (Table 1, Entry 3 and 4). However, interestingly, in the case of heterogeneous AC-SO3Hcatalyzed reaction, the TRS and glucose yields reached 52.0% and 42.5% when the H+/GU ratio was 0.05 (Table 1, Entry 5), which were higher than those at H2SO4 system. The higher glucose and TRS yields of AC-SO3H catalyst could be ascribed to the slower C–O–P hydrolysis rate than that of homogeneous H2SO4. To verify this possibility, [Dmim][(OCH3)(H)PO2] was hydrolyzed to [Dmim][(OH)(H)PO2] and methanol using AC-SO3H and H2SO4 at 100 °C (Scheme S3, Supplementary data). Fig. S6 (in Supplementary data) reveals the formation of methanol in H2SO4 catalyzed reaction was faster than that at AC-SO3H system. Therefore, it could be concluded that, in the IC hydrolysis catalyzed by H2SO4, dephosphorylation proceeded faster than glycosidic bond hydrolysis to generate insoluble cellulose which could not be hydrolyzed anymore. In fact, the hydrolysis of decrystallized cellulose (DC), which was obtained from the solution of cellulose dissolved in 1butyl-3-methylimidazolium chloride ([Bmim]Cl) (Kim et al., 2010), revealed that the catalytic activities of both H2SO4 and AC-SO3H for this reaction were very poor (Table 2, Entry 1). Table 1 also shows that other heterogeneous catalysts, Amberlyst-15 and Naﬁon-NR50 obtained 31.8% and 27.8% of glucose yields which were comparable to those of homogeneous acid catalysts, but lower than that of AC-SO3H. Although these three solid acid catalysts have SO3H functional group, they are different in their acid strength. The Hammett’s acidity values (H0) of ACSO3H, Amberlyst-15, and Naﬁon-NR50 are 11, 2.2, and 13, respectively (Rys and Steinegger, 1979; Moa et al., 2008; Suganuma et al., 2008). However, the differences in AC-SO3H, Amberlyst-15, and Naﬁon-NR50 performances seem most likely to come from the surface area not from the acidity. The surface area of AC-SO3H is 1,177 m2/g while those of Amberlyst-15 and Naﬁon-NR50 are only 45 and