- Email: [email protected]
Isomerization of glucose at hydrothermal condition with TiO2, ZrO2, CaO-doped ZrO2 or TiO2-doped ZrO2
G Model
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 Haruyuki Kitajima a , Yoshimasa Higashino a , Shiho Matsuda a , Heng Zhong b , Masaru Watanabe a,b,∗ , Taku M. Aida a , Richard Lee Smith Jr. a,b a b
Graduate School of Environmental Studies, Tohoku University, 6-6-11 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan Research Center of Supercritical Fluid Technology, Tohoku University, 6-6 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan
a r t i c l e
i n f o
Article history: Received 27 October 2015 Received in revised form 5 January 2016 Accepted 19 January 2016 Available online xxx Keywords: Glucose Fructose Hydrothermal Isomerization ZrO2 solid solution
a b s t r a c t Catalytic activity of TiO2 , ZrO2 and ZrO2 solid solutions (with CaO or TiO2 ) for the isomerization of glucose into fructose under hydrothermal conditions (120–180 ◦ C for 5–15 min reaction time) was evaluated by experimental kinetic studies and surface acidity–basicity measurements. Kinetic studies were conducted with batch reactors heated by microwave. Under hydrothermal conditions regardless of the temperature, fructose yield was always below 10% and fructose selectivity was rapidly decreased with increasing glucose conversion. By adding 10 mM NaOH, fructose yield reached 20% at 160 ◦ C for 5 min. In the presence of TiO2 , fructose yield and selectivity were similar (160 ◦ C, 47% of glucose conversion and 14% fructose yield). ZrO2 showed higher catalytic activity (160 ◦ C, 63% glucose conversion and 21% fructose yield) compared with TiO2 . The catalytic activity of TiO2 doped ZrO2 was between TiO2 and ZrO2 . To increase the basicity of ZrO2 , CaO was doped into the ZrO2 matrix. For experiments with the 24 wt% CaO doped ZrO2 , fructose selectivity was higher than 70% even at 30% glucose conversion at 160 ◦ C for 15 min. For systematical understanding catalytic activity of the metal oxides used, acidity and basicity of the catalysts were measured by temperature programmed desorption (TPD) using either CO2 or NH3 . It was found that basicity on the surface increased with increasing the amount of CaO in the ZrO2 solid solution while acidity increased with amounts of TiO2 . For a simple network model of glucose reactions, rate constants were fitted to the data assuming a simple network model and they were completed with the acid-base properties of TiO2 , ZrO2 and CaO doped and TiO2 doped ZrO2 . As a result, the reactivity was found to be correlated with the ratio of the base to acid sites on the surface and fructose formation was linearly proportional to the base/acid mole ratio on the surface of the catalysts. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Glucose is widely found in natural resources such as carbohydrates, saccharides and cellulosic materials and it is an important chemical for food and biomass chemistry. Transformation of glucose into its isomer, namely fructose, is important as a primary step for modifying of sugars for dietary purposes or for as chemical use. It is well known that the isomerization of glucose into fructose is catalyzed by alkali (or solid base) catalysts through proton transfer or Lewis acid catalysts through hydride transfer. Liu et al. [1] found high catalytic activity of organic amines on glucose isomerization in hot water at 100 ◦ C for 30 min reaction time, where glucose conversion was 51% and fructose yield was 32% (fructose selectivity:
∗ Corresponding author. E-mail address: [email protected] (M. Watanabe).
63%). However, homogeneous catalysts have disadvantages in their reuse and recycle due to the difficulty of their separation from the reaction products. To overcome the difficulty of reuse and recycle homogenous catalysts, solid catalysts are preferable. Tin coated zeolite, which is a solid Lewis acid catalyst, has been demonstrated to have high catalytic activity for glucose isomerization in water at 110–140 ◦ C, giving around 30% of fructose yield at 50% of glucose conversion [2]. Tin-coated zeolite, on the other hand, requires a complicated and long-period production process (hydrothermal treatment at 180 ◦ C for up to 40 d). Alkali-metal ion exchange zeolite has also been demonstrated to have activity for glucose isomerization [3–5] but metal cations tend to leach from the surface of the zeolites under hydrothermal conditions (90–95 ◦ C). Talcites have been tested for glucose isomerization in hot water [3,5,6] and leaching of magnesium ion in the talcite was observed [6].
http://dx.doi.org/10.1016/j.cattod.2016.01.049 0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
2
As a water-tolerate solid base catalyst, metallosilicates [7], zirconia promoted with Cs [8], zironosilicate [9] and zirconium carbonate [10] have been tested to isomerize glucose into fructose under hydrothermal conditions at 80–150 ◦ C. Zirconiumsilicate and zirconium carbonate have good stability under hydrothermal conditions and their catalytic activity is also high (for zirconiumsilicate, glucose conversion was 46% and fructose selectivity was 52% at 80 ◦ C for 1 h; for ziroconium carbonate, glucose conversion was 45% and fructose selectivity was 76% at 120 ◦ C for 20 min). Surface basicity of these ziroconium-based catalysts was evaluated by CO2 -TPD [9] or titration method [10] and these basic properties were considered to be related to the catalytic activity, but the correlation between surface properties and catalytic activity was not clarified. Previously, anatase-TiO2 and tetragonal ZrO2 , which are stable under hydrothermal conditions at 200 ◦ C, were used to study the of isomerization glucose [11,12] and it was found that zirconium could promote the reaction. Surface basicity and acidity of the zirconium was evaluated by CO2 - and NH3 -TPD and it was considered that higher basicity of the zirconia was closely related to its higher catalytic activity than anatase-TiO2 , of which its basicity was weaker [12]. For the design of an appropriate solid base catalyst for glucose isomerization having water-tolerance, zirconium-based catalyst seems to be preferable. According to Moreau et al. [5], the rate of glucose isomerization depends on pH (NaOH amount) of water media in hot water (90 ◦ C). Lee et al. [13] clearly showed that the acidity–basicity of zirconium can be controlled by doped-metal cation and their amount. In this work, catalytic activity of CaO- or TiO2 -solid solutions with zirconium for glucose isomerization in hydrothermal water at 120–180 ◦ C was evaluated with experimental reaction studies and correlation of the rates with the catalyst surface acidity–basicity. For comparison, anatase-TiO2 and tetragonal-ZrO2 were tested as catalysts for glucose isomerization along with glucose reaction with and without NaOH. To determine the acidity–basicity of the metal oxides used in this study, temperature programmed desorption method (TPD: NH3 -TPD for acidity and CO2 -TPD for basicity) was used. Based on the kinetic data, a simple network kinetic model was developed and the rate constant of each reaction step was determined by fitting a model to the experimental data. The correlation between the rate constant and acidity-basicity of the metal oxides was elucidated.
2. Experimental 2.1. Materials Anatase TiO2 was purchased from Wako Co., Ltd., and used as it is. Tetragonal ZrO2 was prepared from Zr(OH)4 obtained from Nakarai Tesque (Kyoto, Japan) by calcination at 400 ◦ C. The crystal structure of the calcined ZrO2 was found to be mixture of monoclinic and tetragonal from determination by XRD analysis [12]. ZrO2 solid solutions (CaO or TiO2 were doped into the ZrO2 matrix) was provided by Daiichi Kigenso Co., Ltd. (Osaka, Japan). The doped amounts of CaO were 4.3, 14, and 24 mol% denoted as Ca-4.3, Ca14, and Ca-24, respectively and those of TiO2 were 7.5 and 15 mol% denoted as Ti-7.5 and Ti-15, respectively. These CaO- and TiO2 doped ZrO2 were calcined at 700 ◦ C for 1 h. The structure of all the doped ZrO2 was cubic (from XRD analysis, not shown here) regardless of the kind and amount of doped metal oxide. Table 1 shows some physical properties of the metal oxides that are used in this study. The amounts of acid and base sites were measured by a temperature programmed desorption (TPD) method. Probe substrate for acid and base sites were NH3 and CO2 respectively. The procedures and the conditions of TPD measurement were described below. In summary, ZrO2 has the similar amount of acid and base
sites. The amount of base sites on CaO-doped ZrO2 was higher than that of acid sites and the basicity increased with increasing the doped amount of CaO. On the TiO2 -doped ZrO2 and the anataseTiO2 , the amount of acid sites was higher than that of base sites. Glucose and fructose were purchased from Wako-Pure Chemicals, Co., Ltd. (Osaka, Japan) and used without further purification. Pure water was obtained by ion exchange and distillation equipment (Yamato). 2.2. Apparatus and procedures A microwave irradiation apparatus containing high-pressure glass reactor was employed to perform the glucose isomerization experiments. The detail of the apparatus is given in previous reports [14,15]. Briefly, multi-mode microwave oven (m-reactor, Shikoku Keisoku, Co., Ltd.) was used to heat a high-pressure glass reactor (inner volume: 10 mL, maximum pressure: 10 MPa, Hiper glasstor, Takatsu Techne Co., Ltd.) The glass reactor was covered with polycarbonate outer tube and PEEK caps. A vacuum was used between the outer cover and the glass reactor to reduce heat losses by convection. Water was fed into the evacuated space to assist in the termination of the reaction after the reaction completed. An aqueous solution (5 g) of glucose (2 wt%) and catalyst (0.1 g) were loaded into the glass reactor. After sealing the glass reactor, 1.3 MPa of nitrogen gas was introduced to inhibit solution boiling. For comparison, glucose reaction experiments without catalyst and with 10 mM of NaOH instead of pure water were conducted. The loaded reactor was placed in the microwave oven set at 120–180 ◦ C and temperature was maintained at the settings for 3–30 min. After the reaction, the product solution consisting of the reactants, products, water and catalyst was recovered with rinsing the reactor by water after cooling and depressurization. Solid catalysts and water soluble products were separated by vacuum filtration. 2.3. Analysis and definition Liquid products were analyzed by HPLC (column: Shodex SH1011, detector: RI, effluent: 1 mL/min of 5 mM H2 SO4 sulfuric acid aqueous solution, column oven: 60 ◦ C) and yields were defined on a molar basis. The detected components were glucose, fructose, 5-hydroxymethyl-furfural (5-HMF), glyceraldehyde, erythrose, glycoaldehyde, lactic acid, formic acid, acetic acid, and levulinic acid. In this study, the focus of the research was on glucose isomerization thus glucose yield, fructose yield and fructose selectivity (fructose yield/glucose conversion) were mainly evaluated (Fig. 1). 3. Results and discussion 3.1. Experimental data Time profiles of glucose reactions are shown in Fig. S1 (Supplementary material). Glucose was relatively stable at 140–180 ◦ C for 30 min and the maximum conversion was 23% at 180 ◦ C for 30 min and then fructose yield was 7%. Fig. S2 shows the glucose reaction with 10 mM NaOH at 140 ◦ C and 160 ◦ C for 3–15 min. The maximum conversion was 30% and then fructose yield was about 20%. For both cases (without catalyst and with NaOH), glucose conversion and fructose yield reached a plateau that resembled chemical equilibrium. Fig. 2 shows the fructose selectivity against glucose conversion with and without NaOH. Without catalyst, fructose selectivity drastically dropped and the formation into the other products was formed with progressing glucose reaction. In the presence of NaOH, selectivity of glucose isomerization into fructose was enhanced, but the selectivity was still low with fructose
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
3
Table 1 Physical properties of TiO2 , ZrO2 and doped ZrO2 solid solutions used in this study. Catalyst
Crystal structure
BET (m2 /g)
Amount of acid site ( mol/g)
Amount of base site ( mol/g)
Density of acid site ( mol/m2 )
Density of base site ( mol/m2 )
ZrO2 TiO2 Ca-4.3(700) Ca-14(700) Ca-24(700) Ti-7.5(700) Ti-15(700)
Monoclinic/tetragonal Anatase Monoclinic Monoclinic Monoclinic Monoclinic/tetragonal Monoclinic/tetragonal
110 4.7 46.8 54.7 79.2 55.4 197.1
670 79 410 490 480 560 400
550 42 530 570 510 240 230
6.1 16.8 8.8 9.0 6.1 10.1 2.0
5.0 8.9 11.3 10.4 6.4 4.3 1.2
All calcination times were 1 h. Hyphenated values refer to the amount of dopant in weight percent doped into ZrO2 .
Fig. 1. Microwave irradiation apparatus with high-pressure glass reactor. (1): microwave oven, (2): polycarbonate outer tube, (3): thick-walled glass reactor, (4): thermocouple, (5): PEEK cap, (6): Ar gas, (7): valve, (8): valve for gas replacement, (9): cooling water, (10): vacuum pump, (11): pressure indicater.
Fig. 3. Fructose selectivity against glucose conversion: (a) with tetragonal/monoclinic ZrO2 catalyst, (b) with anatase TiO2 catalyst. Symbols : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
Fig. 2. Fructose selectivity against glucose conversion with and without 10 mM NaOH. Symbols: : no catalyst 140 ◦ C, 䊐: no catalyst 160 ◦ C,: no catalyst 180 ◦ C, 䊏: NaOH 140 ◦ C, : NaOH 160 ◦ C.
selectivity being about 70% at 30% of glucose conversion, 160 ◦ C for 30 min reaction time. Figs. S3 and S4 show the fructose yield versus reaction time for the glucose isomerization reaction with ZrO2 or TiO2 catalysts at
120–160 ◦ C under hydrothermal conditions for 3–30 min reaction time. Figs. S3 and S4, 5-HMF yields are also shown because 4–7% yields of 5-HMF were formed while 5-HMF formation was negligible with and without NaOH. The formation of 5-HMF is evidence that some amounts of acid sites exist in ZrO2 and TiO2 catalysts and that the acidity of TiO2 was most likely higher than that of ZrO2 . Catalytic activity of glucose isomerization of ZrO2 was higher (maximum fructose yield and glucose conversion were 21 and 63%, respectively) than that of TiO2 (maximum fructose yield and glucose conversion were 14% and 47%, respectively). Fig. 3 shows the fructose selectivity against glucose conversion in the presence of ZrO2 and TiO2 at 120–160 ◦ C. Except for the reaction at 120 ◦ C in the presence of ZrO2 , fructose selectivity decreased with increasing glucose conversion. The maximum selectivity of fructose was 71%, which was achieved at the reaction with TiO2 , but the fructose selectivity drastically decreased with the increase of glucose conversion. The reaction behavior at 120 ◦ C in the presence of ZrO2 implies that an intermediate of glucose conversion into fructose surely exists such as maltose, as reported in a previous study [16].
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
4
Fig. 4. Fructose selectivity against glucose conversion: (a): with Ti-7.5, (b): with Ti-15. Symbols : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
Figs. 4 and 5 show the relationship between fructose selectivity and glucose conversion with TiO2 -doped ZrO2 and CaO-doped ZrO2 , respectively. Figs. S5–S9 show the time profile of the reactions. For the case of TiO2 -doped ZrO2 (Ti-7.5 and Ti-15), fructose selectivity was always lower than that for ZrO2 and higher than that for TiO2 at all glucose conversions for a given reaction temperature. By doping CaO in ZrO2 matrix, fructose selectivity was enhanced. In the presence of Ca-4.3, the maximum fructose selectivity was 86% at 25% glucose conversion at 140 ◦ C for 15 min reaction time. Fig. 6 shows a comparison between data in this work (Ca-14 and Ca-24) and that of the literature [9,10]. Yue et al. [9] conducted glucose isomerization in the presence of zirconosilicate in a glass reactor at temperatures from 80 to 160 ◦ C for 1 h reaction time. Son et al. [10] studied the catalytic effect of zirconium carbonate and zirconium phosphate on glucose isomerization at conditions of 120 ◦ C for 20 min reaction time. Except for one datum obtained by Son et al. [10] that had a 76% of fructose selectivity for 45% of glucose conversion all the data were lower than this work. From results in this work, it could be confirmed that zirconium doped alkaline-earth metals are effective and are tolerate to water for the isomerization of glucose. It can be expected that the catalytic activity of zirconium solid solutions can be improved if the relationship between surface property and catalytic activity can be clarified.
Fig. 5. Fructose selectivity against glucose conversion: a): with Ca-4.3, b): with Ca14, C): with Ca-24. Symbols: : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
3.2. Reaction model and kinetic analysis To analyze the catalytic activity, correlation between physical properties of catalysts and reactivity (fructose selectivity at a glucose conversion) was attempted by assuming a simple reaction network model (Fig. 7). As shown in Fig. 7, based on the experimental results given in the previous section, there are three reaction routes: (i) reversible isomerization between glucose and
Fig. 6. Fructose selectivity against glucose conversion for Ca doped ZrO2 for this work and values reported by other researches. Symbols: 䊉: Yue et al. [9], : Son et al. [10], : Ca-24, 160 ◦ C, : Ca-24, 140 ◦ C, 䊐: Ca-24, 120 ◦ C, 䊏: Ca-14, 120 ◦ C, : Ca-14, 140 ◦ C.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
kgf
glucose kgp
kfg
5
fructose k pf
kpg
k fp
other products Fig. 7. Simple reaction network model for glucose isomerization. Reaction rate constants k defined as first order. Here subscripts indicate the component; g is corresponded to glucose, f is fructose, and p is other products; and combination of two characters in subscripts represents direction of reaction; gf is glucose into fructose, fg is fructose into glucose and so on.
fructose, (ii) reversible conversion of glucose into other products, (iii) reversible conversion of fructose into other products. In Fig. 7, the rate constant of each pathway is described by kgf , kfg , kgp , kpg , kpf , and kfp and the units of these rate constants are all s−1 because all pathways are assumed to be first order. Here subscripts indicate the component; g is corresponded to glucose, f is fructose, and p is other products; and combination of two characters in subscripts represents direction of reaction; gf is glucose into fructose, fg is fructose into glucose and so on. Based on the simple network model (Fig. 7) and the kinetic constants, the rate of glucose (Glu) reaction is expressed by Eq. (1), that of fructose (Fru) is by Eq. (2) and other products (P) by Eq. (3).
D [Glu] = − kgf + kgf [Glu] + kfg [Fru] + kpg [P] dt d [Fru] = kgf [Glu] − kfg + kfp [fru] + kpf [P] dt d [P] = kgp [Glu] + kfp [Fru] − kpg + kpf [P] dt
(1) (2) (3)
These three equations were solved by fourth order Runge–Kutta method and the rate constants in the equations were determined by fitting experimental results at all reaction conditions, to minimize sum of squared residue (SSR) between experimental data (Expi ) and calculation (Calci ) based on Eq. (4). SSR =
n
(Calci − Expi )2
(4)
i=1
Table 2 shows the rate constants at all reaction temperatures for each experimental system (with and without catalysts) with their SSR value. The fitting results are also shown in Figs. S1–S9. As shown in Table 2, kpg was found to be equal to zero for all reactions and kpf was determined to have a positive value. These results show that the intermediate that exists in the pathway from other products (including intermediate) to glucose, kpg , can be neglected. In Table 2, the rate constant of fructose formation (kgf + kpf ) and that of fructose decomposition (kfg + kfp ) are also shown. As shown in Table 2, the order of each rate constant typically followed: kfg > kgf > kpf > kgp
(5)
This trend shows the rate of fructose into glucose was higher than that of glucose into fructose at the experimental condition that were investigated. However, for the case of some of the CaO-doped ZrO2 catalysts at lower temperatures, fructose formation rate (kgf + kpf ) was higher than fructose disappearance rate (kfg + kfp ). This indicates that glucose isomerization into fructose is favored at lower temperatures for a high basicity catalyst. 3.3. Correlation of rate constants with acidity-basicity of catalysts In this study, the reaction rate on the surface of catalysts depended directly on the rate of chemical reaction because the reaction temperature was below 160 ◦ C and the mass transfer of
Fig. 8. Correlation between base/acid mole ratio and fructose formation/disappearance ratio at 160 ◦ C for ZrO2 , TiO2 , Ca-4.3, Ca-14, Ca-24, Ti-7.5 and Ti-15catalysts. (a): relationship between rate constants and base/acid mole ratio, Symbols: : kfg , 䊐: kgf , : kpf , : kgp , 䊉: kfp , (b) relationship between fructose formation/disappearance ratio and base/acid ratio.
the reactants (glucose, fructose and an intermediate) was fast compared with the reaction rates. Fig. 8 shows the correlation of catalytic activity and acidity–basicity. Fig. 8a shows the relationship between base/acid ratio and the rate constants at 160 ◦ C. Fig. 8b shows the relationship between the fructose formation rate (kgf + kpf ) and fructose disappearance rate (kfg + kfp ) at 160 ◦ C as they depend on the base/acid ratio on the catalysts. In both figures, the rate constants and fructose formation/disappearance ratio are linearly proportional to the base/acid ratio of the catalyst even though there is some scatter in the plots. As shown in Fig. 8, to enhance the catalytic activity for glucose isomerization into fructose, the ratio of base site to acid site should be increased. Finally, the temperature dependence of the glucose isomerization with Ca-doped ZrO2 is discussed. Fig. 9 shows the temperature dependence of kgp , which is an important rate constant for glucose isomerization, because inhibition of glucose conversion to other products is main way to improve fructose selectivity especially since the rate of fructose conversion into other products is negligible at the experimental conditions tested in this study. As shown in Fig. 9, kgp becomes lower with decreasing temperature so that the increase of kgp with increasing base/acid ratio is small at 120 ◦ C. Based on this result and previous discussion, an appropriate catalyst for glucose isomerization should have a high base/acid ratio of at least 1.0 or more, and an appropriate reaction condition such as a reaction temperature of 120 ◦ C or less. For the further consideration of the usefulness and recyclable of the ZrO2 catalysts, long duration test of the catalyst must be performed and surface characteristics before and after the reactions such as TPD, BET, XRD, SEM and so on, has to be conducted. These will be studied in the next stage for practical realization.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
6 Table 2 Rate constants of network model. Catalyst
Temp., ◦ C
k × 103 [s−1 ] gp gf
pg
SSR [–] pf
fp
fg
gf + pf
fg + fp
No catalyst
140 160 180
0.35 0.28 0.27
0.00 0.07 0.12
0.00 0.00 0.00
0.00 0.84 0.21
0.00 0.00 0.00
7.42 5.31 3.37
0.35 1.11 0.48
7.42 5.31 3.37
3.5 0.8 1.7
TiO2
120 140 160
0.27 0.74 2.42
0.08 0.16 0.56
0.00 0.00 0.00
1.51 0.59 0.71
0.00 0.00 0.00
3.79 5.73 13.01
1.79 1.33 3.13
3.79 5.73 13.01
1.5 2.3 6.9
Ti-15(700)
120 140 160
0.22 1.10 2.30
0.10 0.00 0.35
0.00 0.00 0.00
1.05 3.61 0.89
0.00 4.70 1.92
2.16 4.59 7.07
1.27 4.71 3.19
2.16 9.30 8.98
0.6 4.0 17.0
Ti-7.5(700)
120 140 160
0.30 0.88 1.92
0.06 0.18 0.63
0.00 0.00 0.00
0.00 0.73 0.69
0.10 0.11 0.00
2.09 4.39 6.81
0.30 1.61 2.61
2.19 4.51 6.81
7.0 7.0 11.3
ZrO2
120 140 160
0.31 1.62 2.35
0.76 0.43 1.34
0.00 0.00 0.00
11.60 1.05 1.17
3.56 0.00 0.00
5.18 5.89 7.07
11.92 2.67 3.53
8.73 5.89 7.07
9.6 8.3 15.6
Ca-4.3(700)
120 140 160
0.63 1.86 2.71
0.17 0.34 0.62
0.00 0.00 0.00
1.43 1.42 1.20
0.00 0.00 0.00
3.23 7.44 7.16
2.06 3.28 3.91
3.23 7.44 7.16
3.9 7.3 25.5
Ca-14(700)
120 140 160
0.41 1.41 3.35
0.07 0.37 1.00
0.00 0.00 0.00
1.71 1.51 1.82
0.51 0.00 0.00
1.56 4.43 8.22
2.12 2.91 5.17
2.06 4.43 8.22
8.0 5.7 42.1
Ca-24(700)
120 140 160
0.45 1.20 3.14
0.14 0.49 1.59
0.00 0.00 0.00
0.94 0.91 0.72
0.00 0.10 0.00
1.64 2.28 4.09
1.39 2.11 3.87
1.64 2.38 4.09
19.3 57.1 11.6
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cattod.2016.01. 049. References
Fig. 9. Base/acid ratio dependence of kgp at different temperatures. Symbols: : kgf 160 ◦ C, 䊐: kgf 140 ◦ C, : kgf 120 ◦ C.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
4. Conclusions
[12]
The effect of solid base catalysts (TiO2 , ZrO2 and CaO- or TiO2 doped ZrO2 ) on the isomerization of glucose to fructose in water was examined with microwave-assisted heating. For all catalysts that were tested in this study, glucose conversion and fructose selectivity was improved over non-catalytic system. The CaOdoped ZrO2 selectively promoted fructose formation from glucose. The correlation between the surface properties of metal oxides and reactivity revealed that the base/acid mole ratio on the catalyst is strongly related to the catalytic activity for glucose isomerization. Solid catalysts with a base/acid ratio greater than 1.0 would increase the fructose selectivity.
[13] [14] [15] [16]
C. Liu, J.M. Carraher, J.L. Swedberg, ACS Catal. 4 (2014) 4295–4298. M. Moliner, Y. Roman-Leshkov, M.E. Davis, PNAS 107 (2010) 6164–6168. C. Moreau, R. Durand, A. Roux, Appl. Catal. A 193 (2000) 257–264. J. Lecomte, A. Finiels, C. Moreau, Starch Starke 54 (2002) 75–79. C. Moreau, J. Lecomte, A. Roux, Catal. Commun. 7 (2006) 941–944. I. Delidovich, R. Palkovits, Catal. Sci. Technol. 4 (2014) 4322–4329. S. Lima, A.S. Dias, Z. Lin, Appl. Catal. A 339 (2008) 21–27. R.O.L. Souza, D.P. Fabiano, C. Feche, Catal. Today 195 (2012) 114–119. C. Yue, M.S. Rigutto, E.J.M. Hensen, Catal. Lett. 144 (2014) 2121–2128. P.A. Son, S. Nishimura, K. Ebitani, React. Kinet. Mech. Catal. 111 (2014) 183–197. M. Watanabe, Y. Aizawa, T. Iida, T.M. Aida, C. Levy, K. Sue, H. Inomata, Carbohydr. Res. 340 (2005) 1925–1930. M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura, H. Inomata, Appl. Catal. A 295 (2005) 150–156. H.J. Lee, W. Joe, J.C. Jung, Korean J. Chem. Eng. 29 (2012) 1019–1024. X. Qi, M. Watanabe, T.M. Aida, R.L. Smith Jr., Catal. Commun. 9 (2008) 2244–2249. M. Watanabe, X. Qi, T.M. Aida, R.L. Smith, The Development and Application of Microwave Heating, in: W. Cao (Ed.), InTech, Croatia, 2012, pp. 163–178. R. Gounder, M.E. Davis, J. Catal. 308 (2013) 176–188.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
Catalysis Today xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today journal homepage: www.elsevier.com/locate/cattod
Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 Haruyuki Kitajima a , Yoshimasa Higashino a , Shiho Matsuda a , Heng Zhong b , Masaru Watanabe a,b,∗ , Taku M. Aida a , Richard Lee Smith Jr. a,b a b
Graduate School of Environmental Studies, Tohoku University, 6-6-11 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan Research Center of Supercritical Fluid Technology, Tohoku University, 6-6 Aoba, Aramaki, Aoba-ku, Sendai 980-8579, Japan
a r t i c l e
i n f o
Article history: Received 27 October 2015 Received in revised form 5 January 2016 Accepted 19 January 2016 Available online xxx Keywords: Glucose Fructose Hydrothermal Isomerization ZrO2 solid solution
a b s t r a c t Catalytic activity of TiO2 , ZrO2 and ZrO2 solid solutions (with CaO or TiO2 ) for the isomerization of glucose into fructose under hydrothermal conditions (120–180 ◦ C for 5–15 min reaction time) was evaluated by experimental kinetic studies and surface acidity–basicity measurements. Kinetic studies were conducted with batch reactors heated by microwave. Under hydrothermal conditions regardless of the temperature, fructose yield was always below 10% and fructose selectivity was rapidly decreased with increasing glucose conversion. By adding 10 mM NaOH, fructose yield reached 20% at 160 ◦ C for 5 min. In the presence of TiO2 , fructose yield and selectivity were similar (160 ◦ C, 47% of glucose conversion and 14% fructose yield). ZrO2 showed higher catalytic activity (160 ◦ C, 63% glucose conversion and 21% fructose yield) compared with TiO2 . The catalytic activity of TiO2 doped ZrO2 was between TiO2 and ZrO2 . To increase the basicity of ZrO2 , CaO was doped into the ZrO2 matrix. For experiments with the 24 wt% CaO doped ZrO2 , fructose selectivity was higher than 70% even at 30% glucose conversion at 160 ◦ C for 15 min. For systematical understanding catalytic activity of the metal oxides used, acidity and basicity of the catalysts were measured by temperature programmed desorption (TPD) using either CO2 or NH3 . It was found that basicity on the surface increased with increasing the amount of CaO in the ZrO2 solid solution while acidity increased with amounts of TiO2 . For a simple network model of glucose reactions, rate constants were fitted to the data assuming a simple network model and they were completed with the acid-base properties of TiO2 , ZrO2 and CaO doped and TiO2 doped ZrO2 . As a result, the reactivity was found to be correlated with the ratio of the base to acid sites on the surface and fructose formation was linearly proportional to the base/acid mole ratio on the surface of the catalysts. © 2016 Elsevier B.V. All rights reserved.
1. Introduction Glucose is widely found in natural resources such as carbohydrates, saccharides and cellulosic materials and it is an important chemical for food and biomass chemistry. Transformation of glucose into its isomer, namely fructose, is important as a primary step for modifying of sugars for dietary purposes or for as chemical use. It is well known that the isomerization of glucose into fructose is catalyzed by alkali (or solid base) catalysts through proton transfer or Lewis acid catalysts through hydride transfer. Liu et al. [1] found high catalytic activity of organic amines on glucose isomerization in hot water at 100 ◦ C for 30 min reaction time, where glucose conversion was 51% and fructose yield was 32% (fructose selectivity:
∗ Corresponding author. E-mail address: [email protected] (M. Watanabe).
63%). However, homogeneous catalysts have disadvantages in their reuse and recycle due to the difficulty of their separation from the reaction products. To overcome the difficulty of reuse and recycle homogenous catalysts, solid catalysts are preferable. Tin coated zeolite, which is a solid Lewis acid catalyst, has been demonstrated to have high catalytic activity for glucose isomerization in water at 110–140 ◦ C, giving around 30% of fructose yield at 50% of glucose conversion [2]. Tin-coated zeolite, on the other hand, requires a complicated and long-period production process (hydrothermal treatment at 180 ◦ C for up to 40 d). Alkali-metal ion exchange zeolite has also been demonstrated to have activity for glucose isomerization [3–5] but metal cations tend to leach from the surface of the zeolites under hydrothermal conditions (90–95 ◦ C). Talcites have been tested for glucose isomerization in hot water [3,5,6] and leaching of magnesium ion in the talcite was observed [6].
http://dx.doi.org/10.1016/j.cattod.2016.01.049 0920-5861/© 2016 Elsevier B.V. All rights reserved.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
2
As a water-tolerate solid base catalyst, metallosilicates [7], zirconia promoted with Cs [8], zironosilicate [9] and zirconium carbonate [10] have been tested to isomerize glucose into fructose under hydrothermal conditions at 80–150 ◦ C. Zirconiumsilicate and zirconium carbonate have good stability under hydrothermal conditions and their catalytic activity is also high (for zirconiumsilicate, glucose conversion was 46% and fructose selectivity was 52% at 80 ◦ C for 1 h; for ziroconium carbonate, glucose conversion was 45% and fructose selectivity was 76% at 120 ◦ C for 20 min). Surface basicity of these ziroconium-based catalysts was evaluated by CO2 -TPD [9] or titration method [10] and these basic properties were considered to be related to the catalytic activity, but the correlation between surface properties and catalytic activity was not clarified. Previously, anatase-TiO2 and tetragonal ZrO2 , which are stable under hydrothermal conditions at 200 ◦ C, were used to study the of isomerization glucose [11,12] and it was found that zirconium could promote the reaction. Surface basicity and acidity of the zirconium was evaluated by CO2 - and NH3 -TPD and it was considered that higher basicity of the zirconia was closely related to its higher catalytic activity than anatase-TiO2 , of which its basicity was weaker [12]. For the design of an appropriate solid base catalyst for glucose isomerization having water-tolerance, zirconium-based catalyst seems to be preferable. According to Moreau et al. [5], the rate of glucose isomerization depends on pH (NaOH amount) of water media in hot water (90 ◦ C). Lee et al. [13] clearly showed that the acidity–basicity of zirconium can be controlled by doped-metal cation and their amount. In this work, catalytic activity of CaO- or TiO2 -solid solutions with zirconium for glucose isomerization in hydrothermal water at 120–180 ◦ C was evaluated with experimental reaction studies and correlation of the rates with the catalyst surface acidity–basicity. For comparison, anatase-TiO2 and tetragonal-ZrO2 were tested as catalysts for glucose isomerization along with glucose reaction with and without NaOH. To determine the acidity–basicity of the metal oxides used in this study, temperature programmed desorption method (TPD: NH3 -TPD for acidity and CO2 -TPD for basicity) was used. Based on the kinetic data, a simple network kinetic model was developed and the rate constant of each reaction step was determined by fitting a model to the experimental data. The correlation between the rate constant and acidity-basicity of the metal oxides was elucidated.
2. Experimental 2.1. Materials Anatase TiO2 was purchased from Wako Co., Ltd., and used as it is. Tetragonal ZrO2 was prepared from Zr(OH)4 obtained from Nakarai Tesque (Kyoto, Japan) by calcination at 400 ◦ C. The crystal structure of the calcined ZrO2 was found to be mixture of monoclinic and tetragonal from determination by XRD analysis [12]. ZrO2 solid solutions (CaO or TiO2 were doped into the ZrO2 matrix) was provided by Daiichi Kigenso Co., Ltd. (Osaka, Japan). The doped amounts of CaO were 4.3, 14, and 24 mol% denoted as Ca-4.3, Ca14, and Ca-24, respectively and those of TiO2 were 7.5 and 15 mol% denoted as Ti-7.5 and Ti-15, respectively. These CaO- and TiO2 doped ZrO2 were calcined at 700 ◦ C for 1 h. The structure of all the doped ZrO2 was cubic (from XRD analysis, not shown here) regardless of the kind and amount of doped metal oxide. Table 1 shows some physical properties of the metal oxides that are used in this study. The amounts of acid and base sites were measured by a temperature programmed desorption (TPD) method. Probe substrate for acid and base sites were NH3 and CO2 respectively. The procedures and the conditions of TPD measurement were described below. In summary, ZrO2 has the similar amount of acid and base
sites. The amount of base sites on CaO-doped ZrO2 was higher than that of acid sites and the basicity increased with increasing the doped amount of CaO. On the TiO2 -doped ZrO2 and the anataseTiO2 , the amount of acid sites was higher than that of base sites. Glucose and fructose were purchased from Wako-Pure Chemicals, Co., Ltd. (Osaka, Japan) and used without further purification. Pure water was obtained by ion exchange and distillation equipment (Yamato). 2.2. Apparatus and procedures A microwave irradiation apparatus containing high-pressure glass reactor was employed to perform the glucose isomerization experiments. The detail of the apparatus is given in previous reports [14,15]. Briefly, multi-mode microwave oven (m-reactor, Shikoku Keisoku, Co., Ltd.) was used to heat a high-pressure glass reactor (inner volume: 10 mL, maximum pressure: 10 MPa, Hiper glasstor, Takatsu Techne Co., Ltd.) The glass reactor was covered with polycarbonate outer tube and PEEK caps. A vacuum was used between the outer cover and the glass reactor to reduce heat losses by convection. Water was fed into the evacuated space to assist in the termination of the reaction after the reaction completed. An aqueous solution (5 g) of glucose (2 wt%) and catalyst (0.1 g) were loaded into the glass reactor. After sealing the glass reactor, 1.3 MPa of nitrogen gas was introduced to inhibit solution boiling. For comparison, glucose reaction experiments without catalyst and with 10 mM of NaOH instead of pure water were conducted. The loaded reactor was placed in the microwave oven set at 120–180 ◦ C and temperature was maintained at the settings for 3–30 min. After the reaction, the product solution consisting of the reactants, products, water and catalyst was recovered with rinsing the reactor by water after cooling and depressurization. Solid catalysts and water soluble products were separated by vacuum filtration. 2.3. Analysis and definition Liquid products were analyzed by HPLC (column: Shodex SH1011, detector: RI, effluent: 1 mL/min of 5 mM H2 SO4 sulfuric acid aqueous solution, column oven: 60 ◦ C) and yields were defined on a molar basis. The detected components were glucose, fructose, 5-hydroxymethyl-furfural (5-HMF), glyceraldehyde, erythrose, glycoaldehyde, lactic acid, formic acid, acetic acid, and levulinic acid. In this study, the focus of the research was on glucose isomerization thus glucose yield, fructose yield and fructose selectivity (fructose yield/glucose conversion) were mainly evaluated (Fig. 1). 3. Results and discussion 3.1. Experimental data Time profiles of glucose reactions are shown in Fig. S1 (Supplementary material). Glucose was relatively stable at 140–180 ◦ C for 30 min and the maximum conversion was 23% at 180 ◦ C for 30 min and then fructose yield was 7%. Fig. S2 shows the glucose reaction with 10 mM NaOH at 140 ◦ C and 160 ◦ C for 3–15 min. The maximum conversion was 30% and then fructose yield was about 20%. For both cases (without catalyst and with NaOH), glucose conversion and fructose yield reached a plateau that resembled chemical equilibrium. Fig. 2 shows the fructose selectivity against glucose conversion with and without NaOH. Without catalyst, fructose selectivity drastically dropped and the formation into the other products was formed with progressing glucose reaction. In the presence of NaOH, selectivity of glucose isomerization into fructose was enhanced, but the selectivity was still low with fructose
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
3
Table 1 Physical properties of TiO2 , ZrO2 and doped ZrO2 solid solutions used in this study. Catalyst
Crystal structure
BET (m2 /g)
Amount of acid site ( mol/g)
Amount of base site ( mol/g)
Density of acid site ( mol/m2 )
Density of base site ( mol/m2 )
ZrO2 TiO2 Ca-4.3(700) Ca-14(700) Ca-24(700) Ti-7.5(700) Ti-15(700)
Monoclinic/tetragonal Anatase Monoclinic Monoclinic Monoclinic Monoclinic/tetragonal Monoclinic/tetragonal
110 4.7 46.8 54.7 79.2 55.4 197.1
670 79 410 490 480 560 400
550 42 530 570 510 240 230
6.1 16.8 8.8 9.0 6.1 10.1 2.0
5.0 8.9 11.3 10.4 6.4 4.3 1.2
All calcination times were 1 h. Hyphenated values refer to the amount of dopant in weight percent doped into ZrO2 .
Fig. 1. Microwave irradiation apparatus with high-pressure glass reactor. (1): microwave oven, (2): polycarbonate outer tube, (3): thick-walled glass reactor, (4): thermocouple, (5): PEEK cap, (6): Ar gas, (7): valve, (8): valve for gas replacement, (9): cooling water, (10): vacuum pump, (11): pressure indicater.
Fig. 3. Fructose selectivity against glucose conversion: (a) with tetragonal/monoclinic ZrO2 catalyst, (b) with anatase TiO2 catalyst. Symbols : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
Fig. 2. Fructose selectivity against glucose conversion with and without 10 mM NaOH. Symbols: : no catalyst 140 ◦ C, 䊐: no catalyst 160 ◦ C,: no catalyst 180 ◦ C, 䊏: NaOH 140 ◦ C, : NaOH 160 ◦ C.
selectivity being about 70% at 30% of glucose conversion, 160 ◦ C for 30 min reaction time. Figs. S3 and S4 show the fructose yield versus reaction time for the glucose isomerization reaction with ZrO2 or TiO2 catalysts at
120–160 ◦ C under hydrothermal conditions for 3–30 min reaction time. Figs. S3 and S4, 5-HMF yields are also shown because 4–7% yields of 5-HMF were formed while 5-HMF formation was negligible with and without NaOH. The formation of 5-HMF is evidence that some amounts of acid sites exist in ZrO2 and TiO2 catalysts and that the acidity of TiO2 was most likely higher than that of ZrO2 . Catalytic activity of glucose isomerization of ZrO2 was higher (maximum fructose yield and glucose conversion were 21 and 63%, respectively) than that of TiO2 (maximum fructose yield and glucose conversion were 14% and 47%, respectively). Fig. 3 shows the fructose selectivity against glucose conversion in the presence of ZrO2 and TiO2 at 120–160 ◦ C. Except for the reaction at 120 ◦ C in the presence of ZrO2 , fructose selectivity decreased with increasing glucose conversion. The maximum selectivity of fructose was 71%, which was achieved at the reaction with TiO2 , but the fructose selectivity drastically decreased with the increase of glucose conversion. The reaction behavior at 120 ◦ C in the presence of ZrO2 implies that an intermediate of glucose conversion into fructose surely exists such as maltose, as reported in a previous study [16].
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model CATTOD-10021; No. of Pages 6
ARTICLE IN PRESS H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
4
Fig. 4. Fructose selectivity against glucose conversion: (a): with Ti-7.5, (b): with Ti-15. Symbols : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
Figs. 4 and 5 show the relationship between fructose selectivity and glucose conversion with TiO2 -doped ZrO2 and CaO-doped ZrO2 , respectively. Figs. S5–S9 show the time profile of the reactions. For the case of TiO2 -doped ZrO2 (Ti-7.5 and Ti-15), fructose selectivity was always lower than that for ZrO2 and higher than that for TiO2 at all glucose conversions for a given reaction temperature. By doping CaO in ZrO2 matrix, fructose selectivity was enhanced. In the presence of Ca-4.3, the maximum fructose selectivity was 86% at 25% glucose conversion at 140 ◦ C for 15 min reaction time. Fig. 6 shows a comparison between data in this work (Ca-14 and Ca-24) and that of the literature [9,10]. Yue et al. [9] conducted glucose isomerization in the presence of zirconosilicate in a glass reactor at temperatures from 80 to 160 ◦ C for 1 h reaction time. Son et al. [10] studied the catalytic effect of zirconium carbonate and zirconium phosphate on glucose isomerization at conditions of 120 ◦ C for 20 min reaction time. Except for one datum obtained by Son et al. [10] that had a 76% of fructose selectivity for 45% of glucose conversion all the data were lower than this work. From results in this work, it could be confirmed that zirconium doped alkaline-earth metals are effective and are tolerate to water for the isomerization of glucose. It can be expected that the catalytic activity of zirconium solid solutions can be improved if the relationship between surface property and catalytic activity can be clarified.
Fig. 5. Fructose selectivity against glucose conversion: a): with Ca-4.3, b): with Ca14, C): with Ca-24. Symbols: : 120 ◦ C, 䊐: 140 ◦ C, : 160 ◦ C.
3.2. Reaction model and kinetic analysis To analyze the catalytic activity, correlation between physical properties of catalysts and reactivity (fructose selectivity at a glucose conversion) was attempted by assuming a simple reaction network model (Fig. 7). As shown in Fig. 7, based on the experimental results given in the previous section, there are three reaction routes: (i) reversible isomerization between glucose and
Fig. 6. Fructose selectivity against glucose conversion for Ca doped ZrO2 for this work and values reported by other researches. Symbols: 䊉: Yue et al. [9], : Son et al. [10], : Ca-24, 160 ◦ C, : Ca-24, 140 ◦ C, 䊐: Ca-24, 120 ◦ C, 䊏: Ca-14, 120 ◦ C, : Ca-14, 140 ◦ C.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
kgf
glucose kgp
kfg
5
fructose k pf
kpg
k fp
other products Fig. 7. Simple reaction network model for glucose isomerization. Reaction rate constants k defined as first order. Here subscripts indicate the component; g is corresponded to glucose, f is fructose, and p is other products; and combination of two characters in subscripts represents direction of reaction; gf is glucose into fructose, fg is fructose into glucose and so on.
fructose, (ii) reversible conversion of glucose into other products, (iii) reversible conversion of fructose into other products. In Fig. 7, the rate constant of each pathway is described by kgf , kfg , kgp , kpg , kpf , and kfp and the units of these rate constants are all s−1 because all pathways are assumed to be first order. Here subscripts indicate the component; g is corresponded to glucose, f is fructose, and p is other products; and combination of two characters in subscripts represents direction of reaction; gf is glucose into fructose, fg is fructose into glucose and so on. Based on the simple network model (Fig. 7) and the kinetic constants, the rate of glucose (Glu) reaction is expressed by Eq. (1), that of fructose (Fru) is by Eq. (2) and other products (P) by Eq. (3).
D [Glu] = − kgf + kgf [Glu] + kfg [Fru] + kpg [P] dt d [Fru] = kgf [Glu] − kfg + kfp [fru] + kpf [P] dt d [P] = kgp [Glu] + kfp [Fru] − kpg + kpf [P] dt
(1) (2) (3)
These three equations were solved by fourth order Runge–Kutta method and the rate constants in the equations were determined by fitting experimental results at all reaction conditions, to minimize sum of squared residue (SSR) between experimental data (Expi ) and calculation (Calci ) based on Eq. (4). SSR =
n
(Calci − Expi )2
(4)
i=1
Table 2 shows the rate constants at all reaction temperatures for each experimental system (with and without catalysts) with their SSR value. The fitting results are also shown in Figs. S1–S9. As shown in Table 2, kpg was found to be equal to zero for all reactions and kpf was determined to have a positive value. These results show that the intermediate that exists in the pathway from other products (including intermediate) to glucose, kpg , can be neglected. In Table 2, the rate constant of fructose formation (kgf + kpf ) and that of fructose decomposition (kfg + kfp ) are also shown. As shown in Table 2, the order of each rate constant typically followed: kfg > kgf > kpf > kgp
(5)
This trend shows the rate of fructose into glucose was higher than that of glucose into fructose at the experimental condition that were investigated. However, for the case of some of the CaO-doped ZrO2 catalysts at lower temperatures, fructose formation rate (kgf + kpf ) was higher than fructose disappearance rate (kfg + kfp ). This indicates that glucose isomerization into fructose is favored at lower temperatures for a high basicity catalyst. 3.3. Correlation of rate constants with acidity-basicity of catalysts In this study, the reaction rate on the surface of catalysts depended directly on the rate of chemical reaction because the reaction temperature was below 160 ◦ C and the mass transfer of
Fig. 8. Correlation between base/acid mole ratio and fructose formation/disappearance ratio at 160 ◦ C for ZrO2 , TiO2 , Ca-4.3, Ca-14, Ca-24, Ti-7.5 and Ti-15catalysts. (a): relationship between rate constants and base/acid mole ratio, Symbols: : kfg , 䊐: kgf , : kpf , : kgp , 䊉: kfp , (b) relationship between fructose formation/disappearance ratio and base/acid ratio.
the reactants (glucose, fructose and an intermediate) was fast compared with the reaction rates. Fig. 8 shows the correlation of catalytic activity and acidity–basicity. Fig. 8a shows the relationship between base/acid ratio and the rate constants at 160 ◦ C. Fig. 8b shows the relationship between the fructose formation rate (kgf + kpf ) and fructose disappearance rate (kfg + kfp ) at 160 ◦ C as they depend on the base/acid ratio on the catalysts. In both figures, the rate constants and fructose formation/disappearance ratio are linearly proportional to the base/acid ratio of the catalyst even though there is some scatter in the plots. As shown in Fig. 8, to enhance the catalytic activity for glucose isomerization into fructose, the ratio of base site to acid site should be increased. Finally, the temperature dependence of the glucose isomerization with Ca-doped ZrO2 is discussed. Fig. 9 shows the temperature dependence of kgp , which is an important rate constant for glucose isomerization, because inhibition of glucose conversion to other products is main way to improve fructose selectivity especially since the rate of fructose conversion into other products is negligible at the experimental conditions tested in this study. As shown in Fig. 9, kgp becomes lower with decreasing temperature so that the increase of kgp with increasing base/acid ratio is small at 120 ◦ C. Based on this result and previous discussion, an appropriate catalyst for glucose isomerization should have a high base/acid ratio of at least 1.0 or more, and an appropriate reaction condition such as a reaction temperature of 120 ◦ C or less. For the further consideration of the usefulness and recyclable of the ZrO2 catalysts, long duration test of the catalyst must be performed and surface characteristics before and after the reactions such as TPD, BET, XRD, SEM and so on, has to be conducted. These will be studied in the next stage for practical realization.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049
G Model
ARTICLE IN PRESS
CATTOD-10021; No. of Pages 6
H. Kitajima et al. / Catalysis Today xxx (2016) xxx–xxx
6 Table 2 Rate constants of network model. Catalyst
Temp., ◦ C
k × 103 [s−1 ] gp gf
pg
SSR [–] pf
fp
fg
gf + pf
fg + fp
No catalyst
140 160 180
0.35 0.28 0.27
0.00 0.07 0.12
0.00 0.00 0.00
0.00 0.84 0.21
0.00 0.00 0.00
7.42 5.31 3.37
0.35 1.11 0.48
7.42 5.31 3.37
3.5 0.8 1.7
TiO2
120 140 160
0.27 0.74 2.42
0.08 0.16 0.56
0.00 0.00 0.00
1.51 0.59 0.71
0.00 0.00 0.00
3.79 5.73 13.01
1.79 1.33 3.13
3.79 5.73 13.01
1.5 2.3 6.9
Ti-15(700)
120 140 160
0.22 1.10 2.30
0.10 0.00 0.35
0.00 0.00 0.00
1.05 3.61 0.89
0.00 4.70 1.92
2.16 4.59 7.07
1.27 4.71 3.19
2.16 9.30 8.98
0.6 4.0 17.0
Ti-7.5(700)
120 140 160
0.30 0.88 1.92
0.06 0.18 0.63
0.00 0.00 0.00
0.00 0.73 0.69
0.10 0.11 0.00
2.09 4.39 6.81
0.30 1.61 2.61
2.19 4.51 6.81
7.0 7.0 11.3
ZrO2
120 140 160
0.31 1.62 2.35
0.76 0.43 1.34
0.00 0.00 0.00
11.60 1.05 1.17
3.56 0.00 0.00
5.18 5.89 7.07
11.92 2.67 3.53
8.73 5.89 7.07
9.6 8.3 15.6
Ca-4.3(700)
120 140 160
0.63 1.86 2.71
0.17 0.34 0.62
0.00 0.00 0.00
1.43 1.42 1.20
0.00 0.00 0.00
3.23 7.44 7.16
2.06 3.28 3.91
3.23 7.44 7.16
3.9 7.3 25.5
Ca-14(700)
120 140 160
0.41 1.41 3.35
0.07 0.37 1.00
0.00 0.00 0.00
1.71 1.51 1.82
0.51 0.00 0.00
1.56 4.43 8.22
2.12 2.91 5.17
2.06 4.43 8.22
8.0 5.7 42.1
Ca-24(700)
120 140 160
0.45 1.20 3.14
0.14 0.49 1.59
0.00 0.00 0.00
0.94 0.91 0.72
0.00 0.10 0.00
1.64 2.28 4.09
1.39 2.11 3.87
1.64 2.38 4.09
19.3 57.1 11.6
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.cattod.2016.01. 049. References
Fig. 9. Base/acid ratio dependence of kgp at different temperatures. Symbols: : kgf 160 ◦ C, 䊐: kgf 140 ◦ C, : kgf 120 ◦ C.
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11]
4. Conclusions
[12]
The effect of solid base catalysts (TiO2 , ZrO2 and CaO- or TiO2 doped ZrO2 ) on the isomerization of glucose to fructose in water was examined with microwave-assisted heating. For all catalysts that were tested in this study, glucose conversion and fructose selectivity was improved over non-catalytic system. The CaOdoped ZrO2 selectively promoted fructose formation from glucose. The correlation between the surface properties of metal oxides and reactivity revealed that the base/acid mole ratio on the catalyst is strongly related to the catalytic activity for glucose isomerization. Solid catalysts with a base/acid ratio greater than 1.0 would increase the fructose selectivity.
[13] [14] [15] [16]
C. Liu, J.M. Carraher, J.L. Swedberg, ACS Catal. 4 (2014) 4295–4298. M. Moliner, Y. Roman-Leshkov, M.E. Davis, PNAS 107 (2010) 6164–6168. C. Moreau, R. Durand, A. Roux, Appl. Catal. A 193 (2000) 257–264. J. Lecomte, A. Finiels, C. Moreau, Starch Starke 54 (2002) 75–79. C. Moreau, J. Lecomte, A. Roux, Catal. Commun. 7 (2006) 941–944. I. Delidovich, R. Palkovits, Catal. Sci. Technol. 4 (2014) 4322–4329. S. Lima, A.S. Dias, Z. Lin, Appl. Catal. A 339 (2008) 21–27. R.O.L. Souza, D.P. Fabiano, C. Feche, Catal. Today 195 (2012) 114–119. C. Yue, M.S. Rigutto, E.J.M. Hensen, Catal. Lett. 144 (2014) 2121–2128. P.A. Son, S. Nishimura, K. Ebitani, React. Kinet. Mech. Catal. 111 (2014) 183–197. M. Watanabe, Y. Aizawa, T. Iida, T.M. Aida, C. Levy, K. Sue, H. Inomata, Carbohydr. Res. 340 (2005) 1925–1930. M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura, H. Inomata, Appl. Catal. A 295 (2005) 150–156. H.J. Lee, W. Joe, J.C. Jung, Korean J. Chem. Eng. 29 (2012) 1019–1024. X. Qi, M. Watanabe, T.M. Aida, R.L. Smith Jr., Catal. Commun. 9 (2008) 2244–2249. M. Watanabe, X. Qi, T.M. Aida, R.L. Smith, The Development and Application of Microwave Heating, in: W. Cao (Ed.), InTech, Croatia, 2012, pp. 163–178. R. Gounder, M.E. Davis, J. Catal. 308 (2013) 176–188.
Please cite this article in press as: H. Kitajima, et al., Isomerization of glucose at hydrothermal condition with TiO2 , ZrO2 , CaO-doped ZrO2 or TiO2 -doped ZrO2 , Catal. Today (2016), http://dx.doi.org/10.1016/j.cattod.2016.01.049