Abstract
Mesoporous heteropolyacid (HPA) nanorods having a composition of [C16H33N(CH3)3]xH3-xPW12O40 ((CTA)xH3-xPW, x = 1, 2, and 3) were synthesized by surfactant encapsulation and were evaluated for their catalytic activity in cellulose hydrolysis. The (CTA)H2PW nanorods were found to be most active with 57.2% yield of 5-hydroxymethylfurfural (5-HMF) at ~100% conversion in water/methyl isobutyl ketone (MIBK) biphase, which was higher than (CTA)H2PW nanosphere at 140 °C for 11 h. The yields of 5-HMF and glucose were obtained as 4.5% and 54.3% at 160 °C for 8 h in water system, respectively. (CTA)H2PW nanorods showed higher tolerance to such feedstocks as lignocellulose, i.e. corn straw with 19.8% and 8.3% yields for glucose and xylose at 35.4% conversion in water. Moreover, (CTA)H2PW nanorods showed higher stability and long duration with ten times reuse. (CTA)H2PW nanorods presented higher efficiency and reusability in conversion of cellulosic biomass.
Download PDF
Full Article
Mesoporous Heteropolyacid Nanorods for Heterogeneous Catalysis in Polysaccharide Conversion
Jiaying Song, Yiming Li, Xueyan Zhang, Dan Zhang, Zijiang Jiang,* and Xiaohong Wang *
Mesoporous heteropolyacid (HPA) nanorods having a composition of [C16H33N(CH3)3]xH3-xPW12O40 ((CTA)xH3-xPW, x = 1, 2, and 3) were synthesized by surfactant encapsulation and were evaluated for their catalytic activity in cellulose hydrolysis. The (CTA)H2PW nanorods were found to be most active with 57.2% yield of 5-hydroxymethylfurfural (5-HMF) at ~100% conversion in water/methyl isobutyl ketone (MIBK) biphase, which was higher than (CTA)H2PW nanosphere at 140 °C for 11 h. The yields of 5-HMF and glucose were obtained as 4.5% and 54.3% at 160 °C for 8 h in water system, respectively. (CTA)H2PW nanorods showed higher tolerance to such feedstocks as lignocellulose, i.e. corn straw with 19.8% and 8.3% yields for glucose and xylose at 35.4% conversion in water. Moreover, (CTA)H2PW nanorods showed higher stability and long duration with ten times reuse. (CTA)H2PW nanorods presented higher efficiency and reusability in conversion of cellulosic biomass.
Keywords: Heteropolyacid; Mesoporous; Nanorod; Hydrolysis; Cellulose; 5-HMF
Contact information: Key Lab of Polyoxometalate Science of Ministry of Education, Faculty of Chemistry, Northeast Normal University, Changchun 130024, P. R. China;
* Corresponding author: wangxh665@nenu.edu.cn
INTRODUCTION
5-Hydroxymethylfurfural (5-HMF) is an important bioplatform chemical that has been produced from the conversion of monosaccharides or polysaccharides (Delidovich et al. 2014). Directly transforming cellulose to 5-HMF might be more economical and energy efficient without competing with the edible sugars. The process contains three cascade reactions: hydrolysis to glucose, isomerization to fructose, and final dehydration to 5-HMF in the presence of acidic catalysts (Li et al. 2018). Compared with homogeneous acids (HCl or H2SO4), heterogeneous catalysts might be more environmentally friendly with easy separation, recyclability, and no corrosion (Jing et al. 2018). The difficulty in mass-transfer between solid catalysts and solid cellulose might hinder their application in 5-HMF production.
Heteropolyacids (HPAs) are acidic catalysts with strong Brønsted acidity; they have been widely used in cellulose conversion (Geboers et al. 2010; Fan et al. 2011; Chambon et al. 2011; Palkovits et al. 2011; Sun et al. 2012; Qu et al. 2012; Deng et al. 2012; Tsubaki et al. 2013; Xie et al. 2014; Zhao et al. 2014; Zhao et al. 2015; Sun et al. 2015a; Kumar et al. 2016; Zhao et al. 2018a; Zhang et al. 2018; Almohalla et al. 2018; Zhang et al. 2019). To overcome the mass-transfer difficulty, surfactant-type HPAs were developed and have shown benefits including concentrating substrates and providing special surroundings by their hydrophobic tail and hydrophilic head (Li et al. 2004; Tang et al. 2012; Lü et al. 2010; Zhang et al. 2010; Cheng et al. 2011; Zhao et al. 2011; Zhang et al. 2011; Zheng et al. 2013; Lu et al. 2015; Sun et al. 2015b; Sun et al. 2016; Zhang et al. 2016a). The compounds [C16H33N(CH3)3]H2PW12O40 ((CTA)H2PW), Cr[(DS)H2PW12O40]3 (DS is the abbreviation of dodecyl sulfate), (CTA)H3PW11CrO39, (CTA)H4PW11TiO40, [HOCH2CH2N(CH3)3]H2PW12O40 and [HOCH2CH2N(CH3)3] H4AlW12O40 all showed good activity in cellulose transformation to glucose, 5-HMF, or levulinic acid (LA), which the best yields of 5-HMF, LA and glucose were obtained as 75.0%, 74.8%, and 75.9% upon [HOCH2CH2N(CH3)3]H2PW12O40 and [HOCH2CH2N(CH3)3]H4AlW12O40, respectively (Zhang et al. 2016a; Sun et al. 2016). However, these surfactant-type HPAs did not exhibit porous characteristics, which could provide more reacting room for transformations. Various solid hybrids with porous structure have been found to be active for cellulose to 5-HMF (Zhang et al. 2016b; Zhang et al. 2017; Zhao et al. 2018b). In this concept, surfactant-type HPAs with porous structure would be designable.
Surfactant-type HPAs had been developed to fabricate different architectures with tunable morphologies including one-dimensional wires (Kang et al. 2004) and fibers (Carraro et al. 2008), two-dimensional thin-films (Liu et al. 2002; Bao et al. 2009) and disks (Nisar et al. 2009a), and three-dimensional vesicles (Zhang et al. 2008; Bu et al. 2009), spheres (Li et al. 2007; Nisar et al. 2009b), tubes (Ritchie et al. 2009), flowers (Nisar et al. 2009c), and cone (Nisar et al. 2011). A surfactant-type HPA with nanorod morphology and porous properties was designable due to the well-matched linear shape and porous cavity for interactions between active sites and cellulose (Zhang et al. 2016c).
Based on the above, surfactant-type HPAs (CTA)xH3-xPW nanorod with mesoporous structure had been synthesized through simply controlling their initial usages. Such nanorod showed enhanced activity in cellulose hydrolysis either in water or in H2O/MIBK biphase. Higher efficiency can be attributed to their one-dimensional morphology, which is a good match to linear cellulose, porous structure favoring for the reaction. Moreover, (CTA)xH3-xPW nanorod showed easily separation compared to their spherical species. The cellulose or even lignocellulose was hydrolyzed into 5-HMF or glucose in H2O/MIBK biphase or in water systems upon (CTA)H2PW due to its stronger Brønsted acidity and special nanorod morphology, which also showed higher stability and longer duration.
EXPERIMENTAL
Materials
Microcrystalline cellulose (white, average particle size 50 μm) was obtained from Beijing InnoChem Science & Technology Co., Ltd. All other reagents were of AR grade and used without further purification. The 3,5-dinitrosalicylic acid (DNS) reagent was prepared according to ref (Cowan et al. 2001). H3PW12O40 was prepared based on Duan et al. (2013).
Characterization
The elemental analysis was obtained using a Leeman Plasma Spec (I) ICP-ES and a P-E 2400 CHN elemental analyzer. FTIR spectra were recorded on a Nicolet Magna 560 IR spectrometer (KBr discs) in the 4000 to 400 cm-1 region. X-ray diffraction (XRD) patterns of the catalysts were carried out using a Japan Rigaku Dmax 2000 X-ray diffractometer with Cu Kα radiation (λ = 0.154178 nm) (Rigaku Corporation, Japan). UV-vis spectra (200-600 nm) were recorded on a Cary 500 UV-vis-NIR spectrophotometer. DR-UV-vis spectra (200-600 nm) were obtained on a UV-2600 UV-vis spectrophotometer (Shimadzu). The 31P NMR spectra of the catalysts were achieved with a Bruker AM 400 spectrometer at 161.9 MHz. SEM images were determined by a SU8010 scanning electron microscope. The energy dispersive X-ray analysis (EDX) was performed to calculate the contents of P, W, O, N, and C elements. TEM micrographs were recorded on a Hitachi H-600 transmission electron microscope. Brunauer-Emmett-Teller (BET) specific surface area, total pore volume and average pore width of the catalysts were determined by a N2 physisorption experiment using an automatic gas adsorption system (ASAP 2020, Micromeritics Instrument Corp, USA) at -196 °C. The sample was degassed at 90 °C for 8 h before the physical adsorption of N2. The electro-potential variation was measured with an instrument of ZDJ-4B automatic potentiometric titration, using a BestLab Non-aqueous pH Titration electrode (Shanghai, China).
Chemical Tests
The acidic strength was measured by the titration according to the previous literature (Pizzio and Blanco 2007). The concentrations of 5-HMF and LA were determined on Agilent Technologies 7820A GC system fitted with an Agilent J&W Advanced capillary GC column (Shanghai, China). The concentration of glucose was measured in the aqueous phase by HPLC equipped with a refractive index detector (Shimadzu LC-10A, HPX-87H column) column at 35 °C. The mobile phase was H2O with a flow rate of 0.5 mL/min at 75 °C. The concentration of xylose was measured by HPLC equipped with a refractive index detector using a UltimateXB-NH2 (4.6 mm×150 mm, 5 μm) column at 35 °C. The mobile phase was H2O/acetonitrile (1/4 v/v) with a flow rate of 1.0 mL/min. The error bars were obtained as the standard deviation of three measurements, which were calculated based on the following:
.
Methods
50 mL of hexadecyltrimethylammonium bromide (CTAB) aqueous solution (20 mM) was added into 50 mL of H3PW12O40 (HPW, 20 mM) with stirring. The white precipitate formed immediately. Then the mixture was kept still for 24 h. The white precipitate was separated and dried under vacuum at 50 C. Then it was calcined at 200 C for 3 h, 72.3% yield. The other hybrids (CTA)xH3-xPW with different x values were synthesized using the same procedure except using MCTAB: MHPW = 2: 1 and 3: 1, instead.
Catalytic Procedure
For the hydrolysis of cellulose into the 5-HMF in the double solvent MIBK/H2O system, the mixture cellulose (0.1 g), and the catalyst (0.09 mmol) were added to H2O (0.5 mL) and MIBK (5 mL). Then they were placed in a steel autoclave for 11 h at 140 °C under magnetic stirring (300 rpm). The reaction was stopped by rapidly cooling the reactor in an ice bath at 0 C for 30 min. After the temperature reached room temperature, the reaction solution formed three layers: the first layer was the organic phase which contained the desired product 5-HMF. The second layer was the aqueous phase that contained fructose and glucose. The third layer was solid which contained unreacted cellulose and the catalyst. The mixture was centrifuged, then washed three times with water and ethanol, and dried under vacuum at 60 °C for 24 h. Repeated experiments only increased some fresh cellulose to 0.1 g as the initial amount due to the inability to separate the catalyst in the mixture from the unreacted cellulose and by-products humin by centrifugation. The total amount of the catalyst leaching after ten cycles of experiments was detected by UV-vis spectroscopy. Cellulose conversions (wt %) were determined by the difference between the cellulose weight before and after the reaction. The quantitative analyse of 5-HMF was determined by Gas Chromatograph (GC).
For the hydrolysis of cellulose in water, a mixture of cellulose (0.1 g) and catalyst (0.09 mmol) was added into water (7 mL). Then the mixture was heated at 160 °C in a steel autoclave lined with Teflon for 8 h with stirring (300 rpm). The reaction was stopped by rapidly cooling the reactor in an ice bath at 0 °C. After the temperature reached room temperature, the mixture was centrifuged to separate the catalyst and unreacted cellulose. Cellulose conversions (wt%) were determined by the change in cellulose weight before and after the reaction. 5-HMF was determined by HPLC (high-performance liquid chromatography). The concentration of glucose was measured in the aqueous phase by HPLC equipped with a refractive index detector (Shimadzu LC-10A, HPX-87H column) column at 35 °C. In both two systems, the conversion of sucrose, cellobiose, starch and corn straw was done in the same procedure as cellulose conversion.
Total Reducing Sugars (TRS) Analysis
A mixture that contained 2 mL of the DNS reagent and 1 mL of the reaction sample was heated for 2 min in a boiling water bath, then cooled to room temperature with flowing water, and mixed with deionized water to a volume of 25 mL. The color intensity of the mixture was measured in a UV757CRT model spectrophotometer at 540 nm. The concentration of total reducing sugars was calculated based on a standard curve obtained with glucose. (Cowan et al. 2001)
Adsorption Experiment
In order to confirm the adsorption capacity of the catalyst to cellulose, the adsorption experiment was conducted. During the adsorption experiment, a mixture of cellulose (0.1 g) and the catalyst (0.09 mmol) were mixed together in H2O (5.5 mL) in a steel autoclave lined with Teflon at a 140 °C for 1 h. The powder solids were removed and dried in a vacuum of 60 C. Then it was used to determine the adsorption of cellulose upon (CTA)H2PW nanorod by IR spectroscopy.
RESULTS AND DISCUSSION
Characterization of the Catalysts
From the results of the elemental analysis (Table 1), the molar ratio of W and P in catalysts was P: W = 1: 12, corresponding to that with Keggin structure of PW12O403-. The contents of C, H and N were inherent with the calculated values indicating the formula as (CTA)xH3-xPW12O40 (x = 1-3).
Figure 1 shows the IR spectra of the (CTA)xH3-xPW nanorods, giving the four characteristic peaks similar to their parent H3PW12O40 (1075, 976, 899, and 796 cm-1) (Deltcheff et al. 1983), indicating that the catalysts maintained the original heteropolyacid structure during reaction with CTAB. In addition, peaks of C-H stretching vibration at 2916 and 2845 cm-1, and C-N at 1467 cm-1 confirmed the presence of CTA+ in the hybrids. The characteristic peaks belonging to PW12O403- anion shifted depending on the CTA+ contents, showing the existence of anion-cation interaction between the polyoxometalate anion and CTA+. The Keggin structure of (CTA)xH3-xPW (x = 1-3) nanorod was further determined by DR-UV-vis spectroscopy, which presented one peak at 267 nm assigned to O → W change transition (Fig. 2).
Fig. 1. IR spectra of (a) H3PW12O40, (b) (CTA)H2PW nanorod, (c) (CTA)2HPW nanorod, (d) (CTA)3PW nanorod, and (e) (CTA)H2PW nanorod after the reaction
Table 1. Elemental Analysis, Total Acidity and Hydrolysis Activity a of H3PW12O40 and (CTA)xH3-xPW Nanorod
The XRD patterns of H3PW12O40 and (CTA)xH3-xPW showed that all exhibited typical X-ray diffractograms of Keggin anion at 10.2°, 20.6°, 25.1°, 34.5°, and 53.7°, indicating the retain of the structure after reacting with CTAB (Fig. 3). And slight shifts for XRD of (CTA)xH3-xPW determined the existence of (CTA)xH3-xPW and no physical mixture of PW12O403- anion and CTAB (Wang et al. 2016).
The 31P MAS NMR spectrum of the (CTA)H2PW nanorod gave one peak at -17.99 ppm (Fig. 4a) with some shift compared to its parent H3PW12O40 (-15.6 ppm). This result confirmed the formation of (CTA)H2PW nanorod and no physical mixture of CTA+ and H3PW12O40.
Fig. 2. DR-UV-vis spectra of the catalysts. (a) H3PW12O40, (b) (CTA)H2PW nanorod, (c) (CTA)2HPW nanorod, and (d) (CTA)3PW nanorod
Fig. 3. XRD patterns of (CTA)xH3-xPW nanorod. (a) H3PW12O40, (b) (CTA)H2PW nanorod, (c) (CTA)2HPW nanorod, (d) (CTA)3PW nanorod, and (e) (CTA)H2PW nanorod after the reaction
.
Fig. 4. 31P MAS NMR spectrum of the (CTA)H2PW nanorod (a) before the reaction, and (b) after the reaction
The SEM images of (CTA)H2PW with different concentrations of CTAB and H3PW12O40 determined the synthetic procedure of nanorods (Fig. 5). At lower concentrations, CTAB firstly reacted with H3PW12O40 through ion exchanging reaction to form micellar sphere with size of 100 nm (Fig. 5a). Increasing their concentrations both to 15 mM, the micellar sphere grew bigger and then self-assembled at 20 mM to nanorods with diameters from 300 nm to 400 nm (Fig. 5b, 5c). When concentrations were further increased to 25 mM, the regular nanorod of (CTA)H2PW was the main product in SEM with diameter of 500 nm (Fig. 5d). Furthermore, the ordered mesoporous structure of (CTA)H2PW nanorod was determined by TEM (Fig. 6a). Combined with SEM and TEM, the self-assembly performance of CTAB and H3PW12O40 was as follows: (1) CTA+ reacted with H3PW12O40 through ion-exchanging to (CTA)H2PW single molecule; (2) (CTA)H2PW self-assembled to form micellar sphere at lower concentrations; and (3) The obtained micelles were further stacked to form cylinders that were attached together as nanorod. The nanorod presented an ordered mesoporous structure (Scheme 1). From the IR and XRD of samples with different initial concentrations (Fig. 7), it could be concluded that such samples all showed the Keggin structure. This indicated that CTA+ reacted with H3PW12O40 firstly in a stoichiometric manner to (CTA)H2PW, then it self-assembled into nanosphere or nanorod depending to the initial usages of CTAB and H3PW12O40. The SEM images of (CTA)2HPW nanorod and (CTA)3PW nanorod are given in Fig. 8.
Fig. 5. SEM images of CTAB reacting with H3PW12O40 with different concentrations of (a) 10 mM, (b) 15 mM, (c) 20 mM, and (d) 25 mM
Scheme 1. The synthetic procedures for (CTA)xH3-xPW nanorod
Fig. 6. The TEM image of (CTA)H2PW nanorod at concentrations of 20 mM (a), EDX pattern (b), and compositional EDX mapping of the (CTA)H2PW nanorod (c)
Fig. 7. IR spectra of and XRD patterns of CTAB reacting with H3PW12O40 with different concentrations of (a) 10 mM, (b) 15 mM, (c) 20 mM, and (d) 25 mM
Fig. 8. The SEM image of (CTA)2HPW nanorod (a) and EDX pattern (b), and the SEM image of (CTA)3PW nanorod (c) and EDX pattern (d)
The energy dispersive X-ray spectroscopy (EDX) measurement (Fig. 6b) showed that the (CTA)H2PW nanorod had no other impurities from the W, C, O, N, and P elements. The molar ratio of P and W obtained from the Fig. 6b was about 1:12, being consistent with the ICP elemental analysis, indicating that the (CTA)H2PW nanorod maintained the Keggin structure. And the molar ratio of W to C was 12:19, indicating the formula of nanorod as (CTA)H2PW, which is coherent with elementary analysis. The energy dispersive spectroscopy (EDX) mapping (Fig. 6c) revealed a uniform dispersion of the component elements W, C, O, N, and P, demonstrating the strong interaction between CTA+ and H3PW12O40.
The N2 adsorption-desorption isotherm and the pore size distribution of Barrett-Joyner-Halenda (BJH) for (CTA)H2PW nanorod are given in Fig. 9. The sample exhibited a typical type IV isotherm and the relative pressure P/P0 had a very pronounced H1 type hysteresis loop at 0.4 to 0.8, indicating that the prepared nanorod had a mesoporous structure. The pore size distribution of Barrett-Joyner-Halenda (BJH) indicated that the pore size of the catalyst was mainly distributed at 2.5 to 3 nm. The surface area of (CTA)H2PW nanorod was 80.3 m2/g.
Fig. 9. (a) Nitrogen adsorption-desorption isotherm of (CTA)H2PW nanorod, and (b) its BJH pore size distribution
Effect of Different Catalysts on Cellulose Hydrolysis
Firstly, the different catalysts were subjected to screening tests for cellulose hydrolysis in aqueous medium (Fig. 10). It could be clearly seen that without any catalyst, cellulose was hardly converted under the reaction conditions employed. In a previous report, micellar (CTA)H2PW showed certain activity in cellulose hydrolysis, which presented 39.3% yield of glucose at 44.1% conversion in water at 170 °C for 8 h (Zhao 2011). It could be seen that the conversion of cellulose was improved to 64.8% in the presence of (CTA)H2PW nanorods, while the yield of glucose also was increased to 54.3%. This improvement was attributed to its wire-like morphology (Zhang et al. 2016c) and higher surface area. The conversion of cellulose and yield of glucose also depended on the composition of the catalysts as (CTA)H2PW > (CTA)2HPW > (CTA)3PW (Table 1), which was coherent with their acidic contents as (CTA)H2PW (1.41 mmol/g) > (CTA)2HPW (0.72 mmol/g) > (CTA)3PW (0.04 mmol/g).
Fig. 10. Different catalysts on cellulose hydrolysis in water. Reaction conditions as 0.1 g of cellulose, catalyst (0.09 mmol), water (7 ml), 160 °C for 8 h
Due to its higher efficiency (CTA)H2PW was used as the main catalyst in cellulose hydrolysis in water. The reaction conditions were optimized as usage of catalyst, volume of water, reaction temperature, reaction time, and different usage of cellulose (Fig. 11). It could be seen that the amount of catalyst increased from 0.05 mmol to 0.09 mmol (Fig. 11a). The conversion of cellulose and the yield of glucose were increased from 42.1% and 34.1% to 64.8% and 54.3%, respectively. This increase could be attributed to an increase in the active sites of the catalyst. However, when the amount of catalyst was increased to 0.1 mmol, the yield of glucose decreased to 47.2%, while conversion was not increased significantly. This decrease was attributable to the higher usage of catalyst, leading to further decomposition of glucose. Water was an essential reactant for the hydrolysis of cellulose. The yield of TRS and glucose increased with added volume of water from 4 mL to 7 mL (Fig. 11b). However, further increasing usage of water decreased the efficiency, which might be due to a reduction in acidic sites. Increasing the reaction time could increase the conversion rate, the yield of glucose increasing first and then decreasing, and the glucose yield reached the maximum at 8 h (Fig. 11c). As the temperature was increased from 140 to 160 °C, the conversion rate and glucose yield increased from 41.6% and 30.7% to 64.8% and 54.3%, respectively. The cellulose usages was increased from 0.1 g to 0.3 g with lowing yields, the usage of 0.1 g was chosen to be the optimum usage (Fig. 11d). In summary, the reaction conditions were optimized to be in 7 mL water at 160 °C for 8 h upon 0.09 mmol (CTA)H2PW nanorod and 0.1 g of cellulose to give 54.3% yield of glucose at 64.8% cellulose conversion (Fig. 11).