Dehydration of glucose to 5-Hydroxymethlyfurfural on bifunctional carbon catalysts

Depósito de investigación de la Universidad de Sevilla https://idus.us.es/ “This is an Accepted Manuscript of an article published by Elsevier in: APPLIED CATALYSIS B: ENVIRONMENTAL on 5 June 2021, available at: https://doi.org/10.1016/j.apcatb.2021.119938 .” 1 Dehydration of glucose to 5-Hydroxymethlyfurfural on bifunctional carbon catalysts Charf Eddine Bounoukta 1,2, Cristina Megías-Sayago1,3, Fatima Ammari 2, Svetlana Ivanova1*, Antonio Monzon 3, Miguel Angel Centeno1, Jose Antonio Odriozola1 1 Departamento de Química Inorgánica e Instituto de Ciencia de Materiales de Sevilla-Centro Mixto CSIC/Universidad de Sevilla, Avda. Américo Vespucio, 49, 41092, Sevilla, Spain. 2 Laboratoire de Génie des Procédés Chimiques- LGPC, Département de Génie des Procédés, Faculté de Technologie, Université FERHAT ABBAS SETIF-1,19000 SETIF, Algérie. 3 Departamento de Ingeniería Química y Tecnologías del Medio Ambiente, Instituto de Nanociencia y Materiales de Aragón (INMA), CSIC-Universidad de Zaragoza, 50018, Zaragoza, España. *E-mail: [email protected] Abstract The proposed study tries to reply on one important question concerning glucose dehydration: What is the role of bare or tandem Lewis/Brønsted acid sites in the reaction and which are better? A series of mono and bifunctional catalyst are designed and screened for the glucose dehydration reaction. The results clearly reveal that catalyst activity is a function of catalyst composition. The presence of Lewis sites the reaction toward first step isomerization, while the Brønsted acid dehydrate directly glucose to HMF via levoglucosane intermediate. This study proposed also a kinetic modelling of the included reactions and their contrast with the empirical observations. Keywords: HMF production, glucose dehydration, biomass conversion, activated carbon, bifunctional catalysts 2 1. Introduction The gradual depletion of fossil fuel sources together with the increasing demand for fuels and fine chemicals oriented important scientific efforts in the last decades towards the exploitation of new renewable resources [1,2]. Within the renewable sources, vegetal (lignocellulosic) biomass awakes special interest due to its natural availability and carbohydrate-rich composition allowing its transformation in a second generation fuels and/or platform molecules [3]. The transformation of glucose to 5-hydroxymehylfurfural (HMF), positioned in Top 12 of highpotential bio-based products by the US Department of Energy, is one of all available possibilities [4]. This molecule present great commercial and technical opportunities due to the high number of useful chemicals that can originate, for example, 2,5-furandicarboxylic acid (FDCA), produced after HMF oxidation stands out as a promising next generation polyester building block being considered the future substitute of terephthalic and isophthalic acids [5– 8]. HMF can be converted also in 2,5-dimethylfurane (DMF) a biofuel with important energy efficiency, in 2,5-di(hydroxymethyl)furan (DHMF), a useful building block in the production of polyurethane foams [9–11], or in 1,6-hexanediol and caprolactam, important polymer precursors [12–14]. Furthermore, HMF has been also reported to produce liquid alkanes from renewable biomass sources [15,16]. Nevertheless, HMF production is not completely operative at industrial scale due to some economic constraints [17]. Glucose and especially fructose are used as starting chemicals to obtain HMF from biomass. Its production from fructose entails a simple dehydration step at low temperatures [18], expensive and unattractive process, as fructose is not abundant naturally and must be previously produced. On the other hand, glucose is a highly multifunctional sixcarbon monosaccharide considered an ideal low-cost feedstock and preferred to synthesize a 3 huge amount of valuable chemicals as HMF, levoglucosan (LGA), formic acid (FA) and levulinic acid (LA) among others [19]. Nevertheless, glucose dehydration still presents several drawbacks [20]: use of expensive solvents such ionic liquids, separation difficulties and impossibility to reuse the catalysts in homogeneous conditions, high energy consumption and participation in side reactions. Examples of the latter are HMF/glucose condensation reactions to humins [21] and/or HMF hydration to levulinic and formic acids (Lev+For) [22]. Generally, HMF is formed in presence of tandem Lewis/Brønsted acid sites [23], via Lewiscatalyzed glucose isomerization to fructose and consecutive HMF production over Brønsted acid sites. Recently, Huber et al. [24], among others [25], proposed a new mechanistic pathway for the formation of HMF from cellulose/glucose over Brønsted acid catalysts, based on direct dehydration forming LGA as intermediate. The involved reactions are summarized in Scheme 1. Scheme 1. Glucose dehydration pathways 4 Among the reported heterogeneous catalysts for biomass valorization processes carbons [26,27], zeolites [28], mesoporous silicas [29], metal oxides [30], heteropolyacids [31,32], sulfonated resins [33] and metal-organic frameworks [34] can be found. The group of activated carbons generated from low-cost precursors, appears as the most versatile materials for the preparation of biomass valorization catalysts. They exhibit high specific surface area, adaptable pore structures, tailored functional surface chemistry and good thermal stability at high temperatures [35]. Several studies have focused on glucose isomerization and dehydration reactions over functionalized carbon catalysts where the highest HMF yields are obtained using dimethyl sulfoxide (DMSO) [39, 40] as a solvent. Nevertheless, the high boiling point of this solvent makes difficult HMF separation. Iris K. M. Yu et al. [38] studied the catalytic activity of Al/biochars in microwave-assisted glucose to fructose isomerization obtaining fructose yields comparable to those reported over inorganic catalysts. Li et al. [39] used niobia/carbon composites in glucose to 5-hydroxymethylfurfural reaction at 160°C obtaining 97.6% conversion, with 59.3% HMF and 35.4% humins yields. Activated carbon (AC) also demonstrated important activity, showing around 51% HMF yield with 78% selectivity in NaCl saturated reaction media [40]. The simple chemical structure and ease of functionalization converts the activated carbons in very versatile materials for active sites modification and study. That is why in this study a commercial activated carbon (AC) was used as raw material for the preparation of bifunctional Bronsted/Lewis tandem site solid catalysts with the final objective to study their efficiency in the conversion of glucose to HMF in biphasic water - methyl isobutyl ketone (MIBK) systems. The functionalities of the catalysts are modified to achieve optimal HMF yields for low humins formation. The study includes also a kinetic modelling of all possible reactions taking place. Catalyst recyclability and stability was likewise evaluated. 5 2. Experimental 2.1 Materials All reactants, p-toluenesulfonic acid monohydrate (pTSa, H2O), activated charcoal (DARCO 100 m size) (AC), zinc chloride (ZnCl2), methyl isobutyl ketone (MIBK) and glucose (C6H12O6) purchased from Sigma Aldrich and calcium chloride dehydrate (CaCl2. 2H2O) from Alfa Aesar were used as received. 2.2 Catalysts preparation The bifunctionality of the catalysts was assured by the introduction of Brønsted and/or Lewis acid sites over activated carbon in two steps. Firstly, the Lewis site, i.e. metal chloride, (CaCl2 or ZnCl2) was impregnated over 1 g of AC using 20 mL solution containing 20 wt.% of metal chloride. After 3 hours of aging, the excess of solvent was removed in rotary evaporator under reduced pressure. After that, the sample was dried at 110 °C and the resulting solid activated 2h at 400 °C (10°C.min-1 ramp) under nitrogen flow (100 mL.min-1). The final solid was washed abundantly with distilled water and dried at 110°C. Two samples were originated after this preparation, Ca/AC and Zn/AC, respectively. The second step introduces Brønsted acidity by sulfonation of the as prepared Ca/AC and Zn/AC and a bare AC samples with p-toluenesulfonic acid (p-TSA). The solids (1g of catalysts or AC, respectively) were treated with 0.5 M p-TSA solution during 24 h followed by solvent evaporation at 80°C under reduced pressure in rotary evaporator. The solids were dried at 80 °C for 12 h, washed abundantly with water and dried again at 110 °C. The obtained samples are pTSA-AC, pTSA-Zn/AC and pTSA-Ca/AC, respectively. 2.3 Catalytic tests 6 Glucose dehydration was carried out in a 50 mL glass reactor equipped with a Young valve and stirrer flea. Glucose (1mmol) and catalyst (40 mg) were mixed in a MIBK/H2O (6/1 v/v ratio) biphasic system containing 1.5 mL of water. The mixture was introduced in preheated paraffin oil bath at the temperature of reaction for a desired time under stirring at approximately 600 rpm. After each experiment, the reactor was cooled down in an ice bath and the mixture microfiltered with 0.45 µL syringe nylon filter. The obtained products were analyzed by High- Performance Liquid Chromatography (HPLC) equipped with refractive index detector (Varian 360-LC) and Hi-Plex H column (300 × 7.7 mm) using a 0.005 M H2SO4 solution as mobile phase. Glucose conversion and HMF selectivity/yield were calculated according to the following equations (1-3). All insoluble products, in both, water and MIBK, were considered humins. () (%) 100 1 () final initial nGlucose Conversion nGlucose     (1) () (%) 100 () final i i final i n Product Selectivity n Products        (2) () (%) 100 () final i i initial n Product Yield nGlucose     (3) where ni and nf are the initial and final moles of the analyzed compound. 2.4 Characterization Thermogravimetric analysis was carried out on a TA Instrument Q600 in air from room temperature to 900 ºC with a heating rate of 10º. min-1. 7 Raman spectroscopy measurements were performed on a Horiba Jobin Yvon dispersive microscopy (HR800) with confocal aperture of 1000 μm, laser spot diameter of 0.72 μm and spatial resolution of 360 nm. The microscope is equipped with a diffraction grating of 600 tours.mm-1, CCD detector, green laser (λ = 532.14 nm, maximum power 20 mW) and a 100× objective. The textural properties were analyzed by N2 adsorption-desorption measurements at liquid nitrogen temperature on a Micrometrics ASAP 2010 equipment. Prior to analysis, the samples were degassed in vacuum during 2 h at 250 °C in order to remove surface impurities and occluded fluids. The Brunauer-Emmett-Teller (BET) method was used to calculate the specific surface area whilst pore-size distributions were determined using the Barett–Joyner–Halenda (BJH) method. XRD measurements were performed at room temperature on a Panalitycal X’Pert Pro diffractometer, equipped with Cu anode. The diffractograms were acquired with 0.05° step size and 300 s time of acquisition within 10–90° 2θ range. SEM images were acquired in a HITACHI S-4800 microscope equipped with secondary and electron backscattered detectors. The micrographs were taken at working distance of 4 mm and 2 kV voltage. The compositional analysis of the samples was performed by energy dispersive X-Ray spectroscopy (EDS), operating at 15 mm working distance and 20 kV voltage. 3. Results and discussion The thermal stability of all prepared samples analyzed by TGA is presented in Figure 1. For all samples mass loss within the 25-110 °C temperature range is observed and ascribed to physically adsorbed water release. A difference in this water content is obvious within the 8 series, with all sulfonated samples presenting higher water content in comparison to bare AC (less hydrophilic) and metallic halides treated samples. This observation allows us to conclude that the sulfonating treatment increases significantly surface hydrophilicity by introducing sulfonic groups. As for the metal halide treated carbons, higher water release is observed for Ca/AC which may be attained to the presence of structural water in the used calcium precursor. The combination of Lewis/Brønsted sites increases even more the hydrophilicity of the samples. With the temperature, all samples show a slow but continuous weight loss, more noticeable for the sulfonated catalysts due to the decomposition of the sulfonic groups bonded to the carbon layers around 340 °C [26]. Without exception, the oxidation temperature of functionalized samples is lower to that of commercial AC suggesting more oxidized surface for the functionalized samples and/or increased hydrophilicity leading to faster degradation with the temperature. The highest rate of oxidation of the Zn-containing samples can be attributed to the transformation of chloride to oxide at temperatures inferior to 500 ºC and its participation in carbon reduction reaction [41]. Figure 1. Thermogravimetric analysis of prepared samples. 100 200 300 400 500 600 700 0 10 20 30 40 50 60 70 80 90 100 Weight loss (%) Temperature (°C) AC Zn/AC Ca/AC pTSA-AC pTSA-Zn/AC pTSA-Ca/AC 15 Figure 2. Catalysts screening (mglucose=180 mg, mcat= 40 mg, T=180 °C, time=24 h, Vtotal=10.5 mL, MIBK/H2O, 6/1 v/v) A) Glucose conversion, HMF yield B) Product selectivity. The functionalization of AC increases glucose conversion almost twice no matter the nature of the introduced sites (Lewis, Brønsted or both), Figure 5A. The low conversion of AC sample can be attributed to the presence of insufficient Brønsted acid sites such as carboxylic (-COOH) and hydroxyl groups (OH) on carbon surface and on its small average pores size and volume. Upon functionalization, glucose dehydration activity improves significantly, especially for the samples containing Ca. The bifunctional pTSA-Ca/AC catalyst reaches almost full glucose conversion and an important HMF yield (57 %). Nevertheless, the yield of HMF by itself is less informative and it is better to take into consideration the product selectivity (Figure 5B). In addition to the already cited products, (fructose, Fru), HMF and its rehydration productslevulinic and formic acids (Lev+for)) it is worth to introduce humins selectivity, indirectly evaluated as a measure of the incomplete C-balance during the reaction. All matter not converted in analyzable products is considered as humins, products of aldehyde and alcohol condensation reactions. What is the function of the acid sites in this reaction? Both types of sites (Lewis and Brønsted) catalyze furfurals condensation to humins, but only Lewis sites catalyze the crossed condensation between monosaccharide’s and furfurals to the same macromolecule [47]. The case of Zn/AC sample is very interesting and very different from the results obtained for Ca/AC catalyst. Although prepared as Lewis site catalyst, it is behaving as Brønsted acid catalyst, i.e. its product distribution and selectivity are much more similar to pTSA/AC catalyst than to Ca/AC sample. We can tentatively explain this with the amphoteric character of Zn2+ in aqueous solution and its hydration to oxo-hydroxo compounds that will behave rather as Brønsted than Lewis acid. For this catalyst a decrease of the HMF selectivity 16 is observed and related to higher humins selectivity at high glucose conversion. If we suppose that Zn/AC catalyst presents more Brønsted sites in comparison to Ca/AC sample and that Ca2+ Lewis sites present less harder-acid character than Zn2+ sites (higher charge/atomic radii ratio for the latter) the Ca-containing samples will present less centers in number and/or in hardness which should result in minor degree of HMF/glucose polymerization, lower humins concentration and higher HMF selectivity [48]. For the bifunctional catalysts, the HMF selectivity is a function of the Brønsted/Lewis sites ratio. The higher the ratio the higher the selectivity towards HMF. We can assume that pTSA- Ca/AC catalyst present a fine site balance and results in high HMF and low humins selectivity. The higher Brønsted to Lewis ratio for this sample is confirmed also by the higher Lev+For selectivity, being the HMF hydration reaction catalyzed only by Brønsted sites. All catalysts present low fructose and levoglucosane (LGA) selectivity indicating that HMF production includes both pathways in a parallel manner and that their transformation is faster reaction than their formation. In summary, Ca containing samples give the best activity/HMF selectivity ratio and the time effect and kinetic modelling was made over pTSA-Ca/AC sample in comparison with pTSA- AC and Ca/AC. 3.1.2. Time effect The kinetics of the reaction for pTSA/AC, Ca/AC and pTSA-Ca/AC are presented in Figure 3. 17 Figure 3. Evolution along time of A) Glucose conversion; and B) HMF selectivity over pTSa- Ca/AC, Ca/AC and pTSA/AC (m glucose = 180 mg, time= 24 h, v H2O =1,5mL MIBK/H 2 O, 6/1 v/v). 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 30 Glucose conversion (%) Time (h) A pTSA-Ca/AC pTSA/AC Ca/AC 0 10 20 30 40 50 60 0 5 10 15 20 25 30 HMF yield, % Time, h Ca/AC pTSA/AC pTSA-Ca/AC B 18 Glucose conversion increases gradually with reaction time no matter the nature of the active sites. The HMF yield presents, however, an optimal value reached earlier for pTSA/AC catalysts and later for the Ca containing systems. After that value, the HMF yield decreases due to the participation of furfural in rehydration and condensation reactions. An interesting observation arises after considering all reaction products (Figure 4). When the Brønsted sites prevail (pTSA/AC, Figure 4 B) the LGA intermediate is the one present in the first 12 h of reaction, suggesting that the HMF formation over this catalyst goes via LGA pathway (Scheme 1). For the catalysts containing Lewis sites (Ca/AC, Figure 4 A) or tandem Brønsted/Lewis sites (pTSA-Ca/AC, Figure 4C) the HMF formation proceeds simultaneously via both, LGA and fructose pathways. 0 10 20 30 40 50 60 70 80 90 100 1 2 3 6 12 18 24 30 Selectivity,% Time, h FRU LGA lev+for HMF humins A 19 Figure 4. Product selectivity distribution over A) Ca/AC, B) pTSA/AC, and C) pTSa-Ca/AC catalyst, (m glucose= 180 mg, time= 24 h, vH20= 1,5 mL, MIBK/H2O, 6/1 v/v). 0 10 20 30 40 50 60 70 80 90 100 123612182430 Selectivity, % Time, h FRU lga lev+for HMF humins B 0 10 20 30 40 50 60 70 80 90 100 123612182430 Selectivity, % Time, h C FRU LGA LEV+FOR HMF HUMINS 20 Another important observation is the appearance of humins in the early stage of reaction proceeding via LGA pathway (pTSA/AC, Figure 4 B). Humins presence suggests an important rate of glucose-glucose and HMF-HMF selfpolymerization and even some HMF-glucose cross-polymerization. On the contrary the fructose mediated route (Figure 4C) includes formation of di-fructose-di-anhydrides with groups blocked for condensation [49]. In general, the mechanism of reaction seems to be a function of Brønsted/Lewis sites ratio. It is initiated by the protonation of glucose’ C2-OH over AC and -SO3H Brønsted sites. On first step an enol intermediate is generated and it transforms subsequently either in fructose via hydride transfer from C2 to C1 facilitated by Lewis sites or directly in 5-HMF via successive dehydration and cyclization steps, catalyzed by Brønsted sites [50]. In the same time, the protonation of glucose intermediates is greatly influenced by the solvent system. In particular, it is plausible to suggest that MIBK as an aprotic solvent, like dimethylsulfoxide or THF, enhances the rate of C2-O bond breaking, leading to an improvement of the rate of HMF formation [51]. Thus, the aprotic solvent addition influences the conformation of glucose, which can be responsible for the change of the reaction pathway. In previous study Yun Yu et al. [25]. reported that the presence of aprotic solvent provokes a change in glucose tautomer’s balance (β-D-glucopyranose as major tautomer) and leads to a direct HMF formation instead of usually preferred isomerization to fructose or production of β- D-anhydroglucopyranose (LGA). This means that in presence of aprotic solvent and strong Brønsted sites the isomerization steps is suppressed and the HMF is formed either via LGA or by a direct HMF formation (pTSA-AC case). However in presence of Lewis site (Ca-containing catalyst) glucose dehydration pathway swiches to fructose production by lowering the energy of C2-C1 hydride transfer. The appearence of this route decreases the rate of humins formation, thus improving the selectivity 21 of the process. But what is the effect of Cl-? It has been reported [52], that Clparticipates as proton aceptor during the H bonding reactions and facilitates the transformation of β-D- glucopyranose in α-D-glucopyranose being the later more easily transformable in enol form (first reaction step) thus lowering the energy of the rate-determining step in HMF formation (glucose and/or fructose enolization). Nevertheless, more detailed study is necessary to elucidate the effect of the Cl-. If we focus on the catalysts containing Ca, the added Brønsted acidity of the SO3H group in pTSA-Ca/AC facilitates C2-O bond breaking giving higher fructose yields and faster successive dehydration and rehydration reactions. In contast, poor Brønsted acidity like in the case of Ca/AC sample, makes the reaction pathways energetically challenging between the oficial routes and direct HMF formation in the presence of aprotic solvents with metal halides accelerating the rate-determining step. As a result HMF with moderate to high yields is formed with minimum humins formation. This clearly demonstrates that Ca/AC catalysts is as suitable catalyst as the bifunctional catalyst pTSA-Ca/AC in terms of HMF production with low carbon lost. Kinetic modelling Based on the literature we have assumed a multi-step pathway model (Scheme 2) to analyze the experimental results. Two models are proposed including coincident main steps and different pathways for the humins generation. The model 1 considers humins generation from both, glucose and HMF, in separate reactions (kHumF and kHumG); while the model 2 considers the participation of both components in a cross polymerization reaction (kHum), see Scheme 1. 22 Scheme 2. Main reactions involved in kinetic models 1 and 2. The initial glucose can isomerize to fructose or dehydrate to levoglucosan and both products dehydrate to HMF. HMF can further dehydrate or participate in polymerization reaction, either alone or cross polymerizing with glucose. As discussed above dehydration is better catalyzed by Brønsted sites while the isomerization goes improved in presence of Lewis sites. As a first approach, we have assumed a simple power-law kinetics of first order for all the reactions considered in the above scheme, i.e. n i =1. Therefore, the net rates of each component are given by the following expressions: Kinetic model 1: 12 12 [] [][] [] HumG G n nn Glucose Hum dGlucose r k Glucose k Glucose k Glucose dt  (4) 3 1 13 [] [][ ] n n Fructose dFructose r k Glucose k Fructose dt  (5) 34 5 34 5 [][][ ] [] [] HumF F nn HMF nn Hum dHMF r k Fructose k Levoglucosan dt kHMF k HMF       (6) 24 24 [] [][ ] nn Levoglucosan d Levoglucosan r k Glucose k Levoglucosan dt  (7) 5 5 [] [] n Levulinic dLevulinic rkHMF dt  (8) 23 [] [] [] H umG HumF GF nn Humins Hum Hum dHumins r k Glucose k HMF dt   (9) Kinetic model 2: According to the previous consideration for the humins formation, the net rates of reaction for Glucose, HMF and humins are now given by the following expressions: 12 12 [] [][] [][] nn Glucose Hum dGlucose r k Glucose k Glucose k Glucose HMF dt  (10) 34 5 34 5 [][][ ] [] [ ][] nn HMF n Hum dHMF r k Fructose k Levoglucosan dt kHMF k GlucoseHMF        (11) [][][] Humins Hum dHumins r k Glucose HMF dt  (12) The rest of equations remain unchanged. The kinetic parameters were calculated by non-linear least squared multivariable regression, minimizing the total weighted sum of squared residuals for each component: 2 , ,exp , ,calc ,calc , ,exp ,calc 1 ; ij ij iij ij i yy WSSR y y n y        (13) The i index stands of each one of the products detected in the reaction and n is the number of experimental point of each curve of Figure 5. The weighting factor taken is the average value of the yield of each component. The first approximation of the model (not shown) with fixed first-order for all considered reactions shows very poor concordance with the experimental data for the pTSA containing systems (especially for the pTSA/AC). So, it was decided to include the n1 order corresponding to glucose to fructose reaction as a variable parameter. The as calculated parameters are summarized in Table 3 and compared to the experimental values in Figure 5 (experimental 24 values represented by points, proposed model 1 by dashed lines and proposed model 2 by solid lines). Table 3. Values of the parameters of kinetic models. Model 1 Parameters pTSA-Ca/AC pTSA/AC Ca/AC k1 2.17E-04 2.06E+02 8.13E-02 k2 2.47E-10 2.41E-02 3.66E-11 k3 1.63E-01 1.00E-09 4.22E-01 k4 3.94E-02 2.35E-01 6.09E-09 k5 8.82E-03 1.40E-02 1.64E-02 kHumF 1.60E-02 7.43E-03 8.46E-03 kHumG 1.19E-02 3.40E-02 7.06E-10 n1 2.58 -1.92 0.88 Total WSSR 19.60 33.76 20.45 Model 2 Parameters pTSA-Ca/AC pTSA/AC Ca/AC k1 3.02E-03 8.09E+00 6.73E-03 k2 6.76E-04 5.79E-02 3.33E-02 k3 1.93E-01 1.00E-09 1.98E-01 k4 6.57E-02 4.05E-01 2.29E-01 k5 8.33E-03 1.32E-02 1.60E-02 kHum 1.33E-03 1.57E-03 2.00E-04 n1 1.96 -1.24 1.33 Total WSSR 37.44 53.32 17.43 The calculated orders vary from positive to negative and fits fairly well the experimental data. No matter the model, Ca/AC shows n1 order close to 1 and it could be modeled under pseudofirst order kinetics for all reactions. Nevertheless, the pTSA/AC sample shows a negative order close to -2 indicating that the transformation glucose to fructose in strongly inhibited and the isomerization reaction occurs in very low extension in comparison with the Ca promoted catalysts. Although a high value for the k1 constant obtained for pTSA/AC sample, its input in the overall kinetics is in fact negligible due to the calculated negative order. Although a high k1 constant obtained for this sample its input in the overall kinetics is negligible due to the calculated negative order. For the bifunctional catalyst (pTSA-Ca/AC) the kinetic order n1 is positive and fructose is formed with a lower rate (k1 constant) in comparison to the pure Lewis 31 [17] G. Portillo Perez, A. Mukherjee, M.-J. Dumont, Insights into HMF catalysis, Journal of Industrial and Engineering Chemistry. 70 (2019) 1–34. https://doi.org/10.1016/j.jiec.2018.10.002. [18] T. Wang, M.W. Nolte, B.H. Shanks, Catalytic dehydration of C 6 carbohydrates for the production of hydroxymethylfurfural (HMF) as a versatile platform chemical, Green Chem. 16 (2014) 548– 572. https://doi.org/10.1039/C3GC41365A. [19] H. Xin, T. Zhang, W. Li, M. Su, S. Li, Q. Shao, L. Ma, Dehydration of glucose to 5- hydroxymethylfurfural and 5-ethoxymethylfurfural by combining Lewis and Brønsted acid, RSC Adv. 7 (2017) 41546–41551. https://doi.org/10.1039/C7RA07684C. [20] B. Agarwal, K. Kailasam, R.S. Sangwan, S. Elumalai, Traversing the history of solid catalysts for heterogeneous synthesis of 5-hydroxymethylfurfural from carbohydrate sugars: A review, Renewable and Sustainable Energy Reviews. 82 (2018) 2408–2425. https://doi.org/10.1016/j.rser.2017.08.088. [21] H. Xia, H. Hu, S. Xu, K. Xiao, S. Zuo, Catalytic conversion of glucose to 5-hydroxymethyfural over Fe/β zeolites with extra-framework isolated Fe species in a biphasic reaction system, Biomass and Bioenergy. 108 (2018) 426–432. https://doi.org/10.1016/j.biombioe.2017.12.007. [22] I. Elsayed, M. Mashaly, F. Eltaweel, M.A. Jackson, E.B. Hassan, Dehydration of glucose to 5- hydroxymethylfurfural by a core-shell Fe3O4@SiO2-SO3H magnetic nanoparticle catalyst, Fuel. 221 (2018) 407–416. https://doi.org/10.1016/j.fuel.2018.02.135. [23] S.S. Chen, T. Maneerung, D.C.W. Tsang, Y.S. Ok, C.-H. Wang, Valorization of biomass to hydroxymethylfurfural, levulinic acid, and fatty acid methyl ester by heterogeneous catalysts, Chemical Engineering Journal. 328 (2017) 246–273. https://doi.org/10.1016/j.cej.2017.07.020. [24] R. Weingarten, A. Rodriguez-Beuerman, F. Cao, J.S. Luterbacher, D.M. Alonso, J.A. Dumesic, G.W. Huber, Selective Conversion of Cellulose to Hydroxymethylfurfural in Polar Aprotic Solvents, ChemCatChem. 6 (2014) 2229–2234. https://doi.org/10.1002/cctc.201402299. [25] B. Song, Y. Yu, H. Wu, Tuning glucose decomposition in hot-compressed gammavalerolactone/water mixtures: From isomerization to dehydration reactions, Fuel. 238 (2019) 225– 231. https://doi.org/10.1016/j.fuel.2018.10.065. [26] R. Zhong, B.F. Sels, Sulfonated mesoporous carbon and silica-carbon nanocomposites for biomass conversion, Applied Catalysis B: Environmental. 236 (2018) 518–545. https://doi.org/10.1016/j.apcatb.2018.05.012. [27] X. Xiong, I.K.M. Yu, L. Cao, D.C.W. Tsang, S. Zhang, Y.S. Ok, A review of biochar-based catalysts for chemical synthesis, biofuel production, and pollution control, Bioresource Technology. 246 (2017) 254–270. https://doi.org/10.1016/j.biortech.2017.06.163. [28] C.G.S. Lima, E.Y.C. Jorge, L.G.S. Batinga, T. de M. Lima, M.W. Paixão, ZSM-5 zeolite as a promising catalyst for the preparation and upgrading of lignocellulosic biomass-derived chemicals, Current Opinion in Green and Sustainable Chemistry. 15 (2019) 13–19. https://doi.org/10.1016/j.cogsc.2018.08.001. [29] M. Stöcker, Biofuels and Biomass-To-Liquid Fuels in the Biorefinery: Catalytic Conversion of Lignocellulosic Biomass using Porous Materials, Angewandte Chemie International Edition. 47 (2008) 9200–9211. https://doi.org/10.1002/anie.200801476. [30] Q. Hou, M. Zhen, L. Liu, Y. Chen, F. Huang, S. Zhang, W. Li, M. Ju, Tin phosphate as a heterogeneous catalyst for efficient dehydration of glucose into 5-hydroxymethylfurfural in ionic liquid, Applied Catalysis B: Environmental. 224 (2018) 183–193. https://doi.org/10.1016/j.apcatb.2017.09.049. [31] S.-S. Wang, G.-Y. Yang, Recent Advances in Polyoxometalate-Catalyzed Reactions, Chem. Rev. 115 (2015) 4893–4962. https://doi.org/10.1021/cr500390v. [32] X. Zhang, D. Zhang, Z. Sun, L. Xue, X. Wang, Z. Jiang, Highly efficient preparation of HMF from cellulose using temperature-responsive heteropolyacid catalysts in cascade reaction, Applied Catalysis B: Environmental. 196 (2016) 50–56. https://doi.org/10.1016/j.apcatb.2016.05.019. [33] B.M.E. Russbueldt, W.F. Hoelderich, New sulfonic acid ion-exchange resins for the preesterification of different oils and fats with high content of free fatty acids, Applied Catalysis A: General. 362 (2009) 47–57. https://doi.org/10.1016/j.apcata.2009.04.019. [34] N. Aljammal, C. Jabbour, J.W. Thybaut, K. Demeestere, F. Verpoort, P.M. Heynderickx, Metalorganic frameworks as catalysts for sugar conversion into platform chemicals: State-of-the-art and 32 prospects, Coordination Chemistry Reviews. 401 (2019) 213064. https://doi.org/10.1016/j.ccr.2019.213064. [35] X. Zhao, J. Xu, A. Wang, T. Zhang, Porous carbon in catalytic transformation of cellulose, Chinese Journal of Catalysis. 36 (2015) 1419–1427. https://doi.org/10.1016/S1872-2067(15)60942-1. [36] B. Zou, X. Chen, C. Zhou, X. Yu, H. Ma, J. Zhao, X. Bao, Highly-efficient and low-cost synthesis of 5-hydroxymethylfurfural from monosaccharides catalyzed by surface treated biomass, Can. J. Chem. Eng. 96 (2018) 1337–1344. https://doi.org/10.1002/cjce.23077. [37] R.S. Thombal, V.H. Jadhav, Biomass derived β-cyclodextrin-SO 3 H carbonaceous solid acid catalyst for catalytic conversion of carbohydrates to 5-hydroxymethylfurfural, Applied Catalysis A: General. 499 (2015) 213–216. https://doi.org/10.1016/j.apcata.2015.04.021. [38] Aluminium-biochar composites as sustainable heterogeneous catalysts for glucose isomerisation in a biorefinery - Green Chemistry (RSC Publishing), (n.d.). https://pubs.rsc.org/en/content/articlelanding/2019/gc/c8gc02466a/unauth#!divAbstract (accessed November 13, 2019). [39] X. Li, K. Peng, Q. Xia, X. Liu, Y. Wang, Efficient conversion of cellulose into 5- hydroxymethylfurfural over niobia/carbon composites, Chemical Engineering Journal. 332 (2018) 528–536. https://doi.org/10.1016/j.cej.2017.06.105. [40] A. Rusanen, R. Lahti, K. Lappalainen, J. Kärkkäinen, T. Hu, H. Romar, U. Lassi, Catalytic conversion of glucose to 5-hydroxymethylfurfural over biomass-based activated carbon catalyst, Catalysis Today. (2019). https://doi.org/10.1016/j.cattod.2019.02.040. [41] A. Berman, M. Epstein, The kinetic model for carboreduction of zinc oxide, J. Phys. IV France. 09 (1999) Pr3-319-Pr3-324. https://doi.org/10.1051/jp4:1999349. [42] S.A. Schmidt, M. Peurla, N. Kumar, K. Eränen, D.Yu. Murzin, T. Salmi, Preparation of selective ZnCl 2 /alumina catalysts for methyl chloride synthesis: Influence of pH, precursor and zinc loading, Applied Catalysis A: General. 490 (2015) 117–127. https://doi.org/10.1016/j.apcata.2014.11.008. [43] S.A. Schmidt, N. Kumar, A. Shchukarev, K. Eränen, J.-P. Mikkola, D.Yu. Murzin, T. Salmi, Preparation and characterization of neat and ZnCl2 modified zeolites and alumina for methyl chloride synthesis, Applied Catalysis A: General. 468 (2013) 120–134. https://doi.org/10.1016/j.apcata.2013.08.039. [44] A.C. Ferrari, J. Robertson, Interpretation of Raman spectra of disordered and amorphous carbon, Phys. Rev. B. 61 (2000) 14095–14107. https://doi.org/10.1103/PhysRevB.61.14095. [45] J.L. Santos, M. Alda-Onggar, V. Fedorov, M. Peurla, K. Eränen, P. Mäki-Arvela, M.Á. Centeno, D.Yu. Murzin, Hydrodeoxygenation of vanillin over carbon supported metal catalysts, Applied Catalysis A: General. 561 (2018) 137–149. https://doi.org/10.1016/j.apcata.2018.05.010. [46] C.M. Subramaniyam, N.R. Srinivasan, Z. Tai, H.K. Liu, S.X. Dou, Enhanced capacity and cycle life of nitrogen-doped activated charcoal anode for the lithium ion battery: a solvent-free approach, RSC Adv. 7 (2017) 16505–16512. https://doi.org/10.1039/C6RA27836A. [47] R. Weingarten, G.A. Tompsett, W.C. Conner, G.W. Huber, Design of solid acid catalysts for aqueous-phase dehydration of carbohydrates: The role of Lewis and Brønsted acid sites, Journal of Catalysis. 279 (2011) 174–182. https://doi.org/10.1016/j.jcat.2011.01.013. [48] Y.J. Pagán-Torres, T. Wang, J.M.R. Gallo, B.H. Shanks, J.A. Dumesic, Production of 5- Hydroxymethylfurfural from Glucose Using a Combination of Lewis and Brønsted Acid Catalysts in Water in a Biphasic Reactor with an Alkylphenol Solvent, ACS Catal. 2 (2012) 930–934. https://doi.org/10.1021/cs300192z. [49] B.F.M. Kuster, 5-Hydroxymethylfurfural (HMF). A Review Focussing on its Manufacture, Starch/Stärke. 42 (1990) 314–321. https://doi.org/10.1002/star.19900420808. [50] W. Yang, C. Zhang, C. Li, Z.Y. Huang, X. Miao, Pathway of 5-hydroxymethyl-2-furaldehyde formation in honey, J Food Sci Technol. 56 (2019) 2417–2425. https://doi.org/10.1007/s13197-019- 03708-7. [51] V. Vasudevan, S.H. Mushrif, Insights into the solvation of glucose in water, dimethyl sulfoxide (DMSO), tetrahydrofuran (THF) and N,N-dimethylformamide (DMF) and its possible implications on the conversion of glucose to platform chemicals, RSC Adv. 5 (2015) 20756–20763. https://doi.org/10.1039/C4RA15123B. 33 [52] C. Zhou, J. Zhao, A.E.A. Yagoub, H. Ma, X. Yu, J. Hu, X. Bao, S. Liu, Conversion of glucose into 5-hydroxymethylfurfural in different solvents and catalysts: Reaction kinetics and mechanism, Egyptian Journal of Petroleum. 26 (2017) 477–487. https://doi.org/10.1016/j.ejpe.2016.07.005. [53] W. Weiqi, W. Shubin, Experimental and kinetic study of glucose conversion to levulinic acid catalyzed by synergy of Lewis and Brønsted acids, Chemical Engineering Journal. 307 (2017) 389– 398. https://doi.org/10.1016/j.cej.2016.08.099. [54] K. Kumar, S. Pathak, S. Upadhyayula, 2nd generation biomass derived glucose conversion to 5- hydroxymethylfurfural and levulinic acid catalyzed by ionic liquid and transition metal sulfate: Elucidation of kinetics and mechanism, Journal of Cleaner Production. 256 (2020) 120292. https://doi.org/10.1016/j.jclepro.2020.120292. [55] W. Wei, S. Wu, Experimental and kinetic study of glucose conversion to levulinic acid in aqueous medium over Cr/HZSM-5 catalyst, Fuel. 225 (2018) 311–321. https://doi.org/10.1016/j.fuel.2018.03.120. [56] G. Delgado Martin, C.E. Bounoukta, F. Ammari, M.I. Domínguez, A. Monzón, S. Ivanova, M.Á. Centeno, Fructose dehydration reaction over functionalized nanographitic catalysts in MIBK/H2O biphasic system, Catal. Today. (2020) S0920586120301334. https://doi.org/10.1016/j.cattod.2020.03.016.