High selectivity and stability of Mg-doped Al-MCM-41 for in-situ catalytic upgrading fast pyrolysis bio-oil

High selectivity and stability of Mg-doped Al-MCM-41 for in-situ catalytic upgrading fast pyrolysis bio-oil

Energy Conversion and Management 142 (2017) 272–285 Contents lists available at ScienceDirect Energy Conversion and Management journal homepage: www...

5MB Sizes 0 Downloads 9 Views

Recommend Documents

No documents
Energy Conversion and Management 142 (2017) 272–285

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

High selectivity and stability of Mg-doped Al-MCM-41 for in-situ catalytic upgrading fast pyrolysis bio-oil Surachai Karnjanakom a, Thanyamai Suriya-umporn b,c, Asep Bayu a, Suwadee Kongparakul c, Chanatip Samart c, Chihiro Fushimi d, Abuliti Abudula a, Guoqing Guan a,b,⇑ a

Graduate School of Science and Technology, Hirosaki University, 1-Bunkyocho, Hirosaki 036-8560, Japan North Japan Research Institute for Sustainable Energy (NJRISE), Hirosaki University, 2-1-3, Matsubara, Aomori 030-0813, Japan Department of Chemistry, Faculty of Science and Technology, Thammasat University, Pathumtani 12120, Thailand d Department of Chemical Engineering, Tokyo University of Agriculture and Technology, 2-24-16 Naka-cho, Koganei, Tokyo, 184-8588, Japan b c

a r t i c l e

i n f o

Article history: Received 4 February 2017 Received in revised form 14 March 2017 Accepted 15 March 2017

Keywords: Biomass Bio-oil Mg-doped Al-MCM-41 Catalytic deoxygenation BTXs

a b s t r a c t In-situ catalytic upgrading of bio-oils derived from the fast pyrolysis of cellulose, lignin or sunflower stalk over Mg-doped Al-MCM-41 was investigated in details. It is found that Mg species with doping amounts ranged between 0.25 and 10 wt.% was well dispersed on Al-MCM-41, and that doping Mg on Al-MCM-41 effectively adjusted the acidity and basicity of the catalysts, resulting in significant improvement of biooil quality. Mg/Al-MCM-41 exhibited high selective conversion of bio-oils derived from cellulose, lignin or sunflower stalk to high value-added aromatic hydrocarbons via catalytic cracking, deoxygenation and aromatization. In the upgraded bio-oil, the relative total hydrocarbon amount reached up to approximately 80%, which consisted of aromatic hydrocarbon approximately 76% and aliphatic hydrocarbon approximately 4% for all feedstocks. The selectivity to the monocyclic aromatic hydrocarbons (MAHs) such as benzene, toluene and xylenes (BTXs) increased while the coke formed on the catalyst decreased with the increase in Mg doping amount. 1 wt.% Mg/Al-MCM-41 resulted in the highest relative total hydrocarbon amount in the upgraded bio-oil at lower catalytic deoxygenation temperature, and showed stable reusability for at least 5 cycles. It is expected that Mg/Al-MCM-41 can be widely applied for bio-oil upgrading in a practical process. Ó 2017 Elsevier Ltd. All rights reserved.

1. Introduction Aromatic hydrocarbons such as benzene, toluene, xylenes (BTXs), ethylbenzene, indenes and naphthalenes are considered as the basic chemicals in the petroleum chemistry, which can be also produced from renewable biomass source [1]. Fast pyrolysis of biomass has gained more extensive attention in recent years since it is one of the most promising technologies for bio-oil production with high yield [2]. However, the bio-oil obtained usually contains a large amount of oxygenated compounds such as phenols, ketones furans, acids and sugars so that it has some undesirable properties such as corrosiveness, instability, immiscibility and low heating value [3]. As such, it is necessary to improve the bio-oil quality by converting those oxygenated compounds to hydrocarbons such as BTXs chemicals. Catalytic cracking with deoxygenation of bio-oil is considered as a promising method since it has many advantages such as no ⇑ Corresponding author at: North Japan Research Institute for Sustainable Energy (NJRISE), Hirosaki University, 2-1-3, Matsubara, Aomori 030-0813, Japan. E-mail address: [email protected] (G. Guan). http://dx.doi.org/10.1016/j.enconman.2017.03.049 0196-8904/Ó 2017 Elsevier Ltd. All rights reserved.

H2 consumption, and cheap and facile operating condition at atmosphere pressure [4]. Many zeolites have been applied as solid acid catalysts for oxygen removal from bio-oil [5–7]. However, some bulky oxygenates were found to be difficult to pass through the small channel of zeolite, resulting in many oxygenate compounds remaining in the final bio-oil [8]. In particular, during lignin pyrolysis, large oxygenated molecules such as guaiacols, syringols, dimethoxyphenol and trimethoxyacetophenone are always formed, which cannot enter into the zeolite pores. Yu et al. [9] found that lignin-derived syringols cannot be effectively converted to aromatic hydrocarbons due to size exclusion or pore blockage. Park et al. [10] reported that low mass transfer rate is one of the problems by using HZSM-5 with micropore structure in the upgrading process. On the other hand, the strong acidity of zeolite easily leads to the coke formation, resulting in the blocking of pores and rapid deactivation of catalyst [11]. Thus, choosing suitable pore structure and avoiding the coke formation on the catalyst are necessary for developing effective catalyst for deoxygenation of bio-oil.

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Several mesoporous catalysts with adjustable uniform pore size in the range of 2–15 nm, have been developed in order to allow the interaction of large oxygenated molecules in the bio-oil with active sites in the pores [12,13]. MCM-41 is one of mesoporous materials with higher surface area, larger pore size and more accessible reaction sites than the traditional zeolites [14]. Its Brönsted and Lewis acidities can be also adjusted by aluminum incorporation. Adam et al. [15] found that the use of Al-MCM-41 resulted in the reduction of the carbonyl and acid yields, and the increase of hydrocarbons amount, especially aromatic hydrocarbons in the final upgraded bio-oil. Iliopoulou et al. [16] also reported that AlMCM-41 is a promising catalyst for production of high quality bio-oil. However, it is found that high amount of coke always deposited on this catalyst due to its specific properties such as high acidity and large pore volume. Moreover, even though high acidity and large pore of catalyst are benefit for the deoxygenation and mass transfer, it usually results in the formation of higher polycyclic aromatic hydrocarbons (PAHs) amount in bio-oil via further aromatization and alkylation of MAHs. It should be noted that PAHs are not desirable products due to their carcinogenicity. Especially, PAHs can serve as the precursors of coking, causing catalyst deactivation [17]. To solve these problems, various metals were doped on such catalysts. Vichaphund et al. [18] doped Ni, Ga, Mo, Pd or Co on HZSM-5 and found that the synergy of metal and HZSM-5 slightly improved anti-PAHs formation ability. Zheng et al. [19] used mesoporous Mo2N/c-Al2O3 for the catalytic fast pyrolysis of lignin, and found that the highest aromatic hydrocarbon yield reached 17.5%. Recently, alkali and alkaline earth metals (AAEM) were considered as the base catalyst for catalytic upgrading of bio-oil because of their high activity, abundant and low cost [20]. Lin et al. [21] confirmed that CaO can be applied for the catalytic deoxygenation reactions, which result in significant decreasing of oxygenated compounds such as laevoglucose, formic acid, and acetic acid in the bio-oil. Stefanidis et al. [22] used the basic MgO catalyst as an alternative of solid acid catalysts for in-situ upgrading of bio-oil, and found that MgO exhibited high decarboxylation ability to reduce acid components in bio-oil with releasing of CO2 gas. To date, only a few studies were reported on the doping of alkali earth metals such as Mg on mesoporous support for the selective productions of MAHs and PAHs [23]. In this study, Mg-doped Al-MCM-41 was prepared as an acid– base bifunctional mesoporous catalyst for upgrading of bio-oils derived from the fast pyrolysis of cellulose, lignin or sunflower stalk. The prepared catalyst was characterized using surface area and porosity measurements, X-ray diffraction (XRD), scanning electron microscopy with energy dispersive X-ray Spectroscopy (SEM-EDX), transmission electron microscopy (TEM), UV–Vis diffuse reflectance (UV–Vis) and NH3-Temperature-programmed desorption (NH3TPD), CO2-Temperature-programmed desorption (CO2-TPD) techniques. The product distributions in the upgraded bio-oils from different biomass feedstock were investigated by using Mg/Al-MCM41 with different Mg loading amounts at different reaction temperatures. The reaction pathways for catalytic upgrading of bio-oils to aromatic hydrocarbons were proposed. Furthermore, the reusability of catalyst was tested for 5 cycles in the cases without regeneration as well as with regeneration. To the best of our knowledge, such catalysts have not been reported in literature so far. It is expected that such catalysts can be widely used for production of high valueadded chemicals such as BTXs from bio-oils.

273

tively. Sunflower stalk collected from Aomori, Japan was milled to a size in the range of 0.5–1 mm. Prior to analysis, all these biomass feedstocks were dried at 105 °C for 12 h without any further treatment. Their proximate, ultimate analysis and thermal decomposition range were shown in Table S1 and Fig. S1, respectively. The details of biomass characterization can be found elsewhere [24]. 2.2. Catalyst preparation 2.2.1. Synthesis of Al-MCM-41 catalyst Al-MCM-41 (Si/Al = 30) was synthesized by sol-gel method following the procedure described in literature [25]. Here, tetraethyl orthosilicate (TEOS, Wako, Japan), aluminum isopropoxide (AIP, Wako, Japan) and cetyltrimethylammonium bromide (CTAB, Wako, Japan) were employed as silica source, aluminum source and structure-directing template, respectively in the synthesis. In a typical procedure, a certain amount of AIP was dissolved in 260 mL of deionized water, 32.4 mL of ethanol and 21 mL of ammonia solutions under vigorous stirring at 35 °C for 1 h. Then, 7.3 g of CTAB was added in the resultant clear solution and maintained at 35 °C under vigorous stirring until the template was fully dissolved. Thereafter, a certain amount of TEOS was added and stirred for 2 h. This procedure resulted in a precursor gel with a molar composition of 0.1TEOS:0.002AIP:0.02CTAB:2.4NH4OH:5.2EtOH:14.4H2O, which was transferred to a Teflon-lined stainless steel autoclave and heated at 100 °C for 24 h. Finally, the precipitated white product of MCM-41 was filtered, washed with deionized water, dried at 105 °C in oven and calcined at 550 °C in air atmosphere for 4 h. 2.2.2. Preparation of Mg-doped Al-MCM-41 catalyst Mg doped Al-MCM-41 s with various doping amounts (0.25– 10 wt.%) were prepared by impregnation method. In brief, a certain amount of Mg(NO3)26H2O (Wako, Japan) was dissolved in the deionized water and stirred until a homogeneous solution was obtained. Then, the Al-MCM-41 powder was added in the mixture and stirred at ambient temperature for 2 h. Thereafter, the slurry was dried at 80 °C and calcined at 650 °C in air atmosphere for 2 h. For comparison, pure mesoporous MgO material was also selected as a catalyst for the upgrading of bio-oil in this work. The mesoporous MgO was synthesized using hydrothermal method as described in the literature [26]. Prior to performance test, all the catalysts were pelleted, crushed and sieved to a particle size of 1–2.8 mm. 2.3. Catalyst characterization 2.3.1. Surface area and porosity measurements N2 adsorption/desorption isotherms were measured at 196 °C using a Quantachrome instrument (NOVA 4200e, USA). Before the N2 sorption measurement, the catalyst was degassed under vacuum condition at 200 °C for 2 h. The specific surface area and pore size of catalyst were calculated by Brumauer–Emmett–Teller (BET) method and Barrett–Joyner–Halanda (BJH) method, respectively. 2.3.2. X-ray diffraction (XRD) XRD pattern of catalyst structure was recorded using an X-ray diffractometer (XRD, Rigaku Smartlab, Japan) in 2h range of 25– 85° with a scanning step of 0.02° at each point using Cu Ka radiation (k = 0.1542 nm).

2. Experimental 2.1. Biomass feedstock Cellulose and lignin were purchased from Wako Pure Chemical Company (Japan) and Tokyo Chemical Industries (Japan), respec-

2.3.3. Acidity and basicity measurements The acidity and basicity of catalyst were evaluated by NH3Temperature-programmed desorption (NH3-TPD) and CO2Temperature-programmed desorption (CO2-TPD), respectively using a BET-CAT catalyst analyzer (BEL, Japan) equipped with a

274

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

thermal conductivity detector (TCD). In a typical acidity measurement, the catalyst was charged in a U-shaped quartz cell and preheated under helium flow at 750 °C for 1 h in order to remove out moisture and some impurities within the catalyst structure. Then, the preheated catalyst was saturated with a gas mixture of 5% NH3/95% He with a flow rate of 50 cm3/min at ambient temperature for 1 h. After stabilization, NH3 desorption was performed from room temperature to 800 °C with a heating rate of 10 °C/ min under He flow. Here, the NH3-desorption peak was detected using a TCD and the adsorbed NH3 concentration was quantified from the peak area by calibrating using the standard gas. Basicity measurement was carried out by the same measurement condition with NH3-TPD but only NH3 gas was changed into CO2. 2.3.4. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) The morphology of catalyst and the existence as well as the dispersion of Mg species on the surface of Al-MCM-41 were investigated by using a scanning electron microscope (SEM, SU8010, Hitachi, Japan) coupled with energy dispersive X-ray detector (EDX), and a JEM-2100F transmission electron microscope (TEM, JEOL, Japan). Before the SEM observation, the dried powder sample was dispersed on carbon tape supported on stub and then pretreated by Pt sputtering. In case of TEM, the sample was sonicated in ethanol for approximately 10–15 min, and then dropped on a copper grid and dried at room temperature.

2.5. Analysis of bio-oil and gas product Chemical compositions in obtained bio-oil was analyzed by using a gas chromatograph (GC-2010 Plus, Shimadzu, Japan)/mass spectrometry (GCMS-QP2010 Ultra, Shimadzu, Japan) with Ultra ALLOY+ 5 capillary column. Bio-oil sample was automatically injected into the column whose temperature was increased from 50 to 300 °C with a ramp rate 10 °C/min and held at 300 °C for 10 min. The ionization chamber of MS setup was set at 200 °C. The sample volume of 0.2 mL including a volume ratio of bio-oil to acetone from about 0.02 to 0.18 was injected into the machine. The compound was identified by comparison with the mass spectra in the NIST MS library, with a similarity above 85%. In addition, the available pure compounds (Wako, Japan) were also applied to confirm the peaks identified from NIST MS library based on matching of mass spectra and retention times. The yields of aromatic hydrocarbons such as benzene, toluene, xylenes, ethylbenzene, indenes and naphthalenes were quantified using external standards (Wako, Japan) and a four point calibration curve with the R2 > 0.99, which was constructed by plotting the peak area versus the concentration of each compound. Here, it should be noted that GC/MS technique could not provide the quantitative analyses for all of compounds due to the existence of many complex products and the lack of some external standards. Therefore, the relative

2.3.5. UV–Vis diffuse reflectance (UV–Vis) The coordination environment of Mg species on Al-MCM-41 structure was investigated by UV–Vis diffuse reflectance spectroscopy analysis using a JASCO V-650 spectrophotometer in a wavelength range of 190–340 nm. 2.3.6. Analysis of coke deposition amount Total coke amount deposited on the spent catalyst was quantified using a thermogravimatric analyzer (TGA, DTG-60H, Shimadzu, Japan). All the samples were preheated at 120 °C for 30 min to remove the moisture and then heated with a heating rate of 10 °C/min until a temperature of 800 °C was achieved under air flow. 2.4. Catalytic deoxgenation testing Fast pyrolysis of biomass and subsequently catalytic deoxygenation of bio-oil were conducted in a fixed-bed reactor system, in which an infrared image furnace (VHT series, ULVAC, Japan) was used, operating at atmospheric pressure. The reactor was made by quartz tubular with a height of 450 mm and an internal diameter of 5.5 mm. A schematic diagram of the experimental setup was presented elsewhere [27]. In a typical run, 0.4 g of biomass and 1 g of catalyst powder were separately packed with quartz wool in the reactor. During the reaction, N2 was used as the carrier gas with a flow rate of 100 cm3/min. Prior to the experiment, the reactor was purged with N2 gas flow for approximately 10 min to move out the inside air. The fast pyrolysis time and heating rate of furnace were fixed at 10 min and 1000 °C/min, respectively. The reaction temperature and heating rate of the experimental system were adjusted and controlled using a type K thermocouple coupled with a programmable temperature controller (TPC5000 ULVAC, Japan). The condensable bio-oil was trapped with acetone in ice-cooling bottle and the noncondensed gas was purified with a CaCl2 filter and collected in a gas bag for further analysis. After finishing the pyrolysis process, the residual solid in the reactor was immediately weighted by the end of reaction to determine the char yield, while the spent catalyst was analyzed by TGA to determine the coke yield.

Fig. 1. (A) N2 adsorption-desorption isotherms and (B) pore-size distributions of 0– 10 wt.% Mg/Al-MCM-41.

275

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285 Table 1 Physicochemical properties of as-prepared catalysts. Catalyst

Surface area (m2/g)

Pore volume (cm3/g)

Pore size (nm)

Acidity (mmol/g)

Basicity (mmol/g)

Al-MCM-41 0.25 wt.% Mg/Al-MCM-41 0.5 wt.% Mg/Al-MCM-41 1 wt.% Mg/Al-MCM-41 3 wt.% Mg/Al-MCM-41 5 wt.% Mg/Al-MCM-41 10 wt.% Mg/Al-MCM-41 MgO

1084 1038 1011 881 741 665 540 289

1.02 0.86 0.84 0.70 0.57 0.53 0.49 0.34

2.55 2.41 2.40 2.40 2.39 2.40 2.39 3.33

0.302 0.274 0.266 0.243 0.117 0.108 0.094 –

– 0.042 0.069 0.132 0.110 0.091 0.077 0.113

amount of each product derived from chromatographic peak area can be considered to be a good estimation or semi-quantitative because they indicate the concentrations of chemical products in bio-oil, and can be used to reveal the efficiency of catalysts and the changing in chemical products in bio-oil before and after the catalytic upgrading process [10,22,28–30]. Karl-Fisher Titration

method (MKS-500, KEM, Japan) using the ASTM E 203 was applied to measure water content in the obtained bio-oil. The collected non-condensed gas was analyzed by using a gas chromatograph (Agilent 7890A GC system, USA) equipped with TCD and 3 packed columns (1 molecular sieve 5A column + 1 HayeSep Q column + 1 molecular sieve 5A column) for separation of CO, CH4 and CO2 using He as a carrier gas, while a molecular sieve 5A for H2 detection using Ar as a carrier gas. The gas yield was quantitatively determined from external standards with peak areas obtained from GC. As such, the mass balance of products, i.e., gas, water, char and coke, was obtained from quantification, and bio-oil was determined from the difference. Each experiment was repeated at least three times under the same conditions in order to minimize error and ensure its repeatability. The reported data here were the average of the three tests with the relative standard deviation (RSD) in the range of 0.4–7.0%, indicating an overall good repeatability.

3. Results and discussion 3.1. Characterization of prepared catalyst

Fig. 2. XRD patterns of 0.25–10 wt.% Mg/Al-MCM-41. Inset: XRD pattern of MgO.

Fig. 1A shows N2 adsorption-desorption isotherms of Al-MCM41 and Mg/Al-MCM-41 catalysts. The isotherm of Al-MCM-41 presents the irreversible type IV sorption isotherm, which is a typical characteristic of ordered mesoporous material [14]. After Al-MCM41 was doped by Mg with various loading amounts, the similar

Fig. 3. TEM images of (A) 1 wt.% Mg/Al-MCM-41, (B) 5 wt.% Mg/Al-MCM-41 and (C) 10 wt.% Mg/Al-MCM-41.

276

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

isotherm as that of Al-MCM-41 was observed, implying that the ordered structure was retained. Fig. 1B shows the pore size distributions of catalysts. Here, all the catalysts present the narrow pore size distributions with an average pore size of approximately 2.4 nm. Moreover, a new second weak peak was observed in 10 wt.% Mg/Al-MCM-41, probably due to a slight pore collapse arising from the high Mg loading amount. As shown in Figs. S2A and B, it can be seen that the similar isotherm type with larger pore size ranged from 1 to 4 nm was obtained for MgO when compare to Al-MCM-41 or Mg/Al-MCM-41. The textural properties of catalysts are listed in Table 1. Al-MCM-41 and MgO exhibited the highest and the lowest surface areas, respectively. With the increase in Mg doping amount on Al-MCM-41, the surface area and pore volume decreased apparently, indicating that the dispersion and covering of Mg species on the surface of Al-MCM-41. Fig. 2 presents the XRD patterns of catalysts. No obvious diffraction peaks attributed to Mg species were observed even 10 wt.% of Mg was doped on Al-MCM-41. It indicates that Mg species can be dispersed on the Al-MCM-41 structure very well without any agglomeration or sintering during calcination. The synthesized MgO exhibited the main diffraction peaks at 36.9, 42.9, 62.3, 74.7 and 78.6°, which are labeled to (1 1 1), (2 0 0), (2 2 0), (3 1 1) and (2 2 2) crystal planes, respectively [31]. Fig. 3 shows TEM images of Mg/Al-MCM-41 catalysts with Mg loading amount of 1, 5 and 10 wt.%. One can see that all catalysts still present well-order hexagonal porous structure after Mg loading, indicating that Mg doping amount in this work had no effect on the framework of mesoporous MCM-41 material. Especially, no Mg particles were observed on the TEM images of all the catalysts, indicating that Mg species could homogeneously distributed on the Al-MCM-41 structure. Moreover, as shown in Fig. 3C, in the

case of an Mg loading amount of 10 wt.%, some collapsing sites can be found in the order hexagonal mesoporous structure. The SEM surface morphologies of catalysts are also shown in Fig. 4. It is found that no bulk Mg particles were formed on the surfaces of catalysts. Here, the homogeneous distribution of Mg element on the framework structure of Al-MCM-41 was also confirmed by EDX mapping, which is well agreement with the results of TEM, XRD and N2 sorption analyses. Fig. 5 shows UV–Vis diffuse reflectance spectra of catalysts. The 0.25 wt.% Mg/Al-MCM-41 exhibited three characteristic bands at 200, 230 and 290 nm, which could be assigned to the adsorption bands of charge-transfer in O-tometal related to Mg2+ stabilized in the support framework (MgAOASi), MgO and MgAO pair involving 3-coordinated O2 ions, respectively [32,33]. According to Fig. 5, the intensity of band at 200 nm obviously increased with the increase of Mg loading amount, especially for 3–10 wt.%, probably due to the domination of more Mg2+ and MgO species in the support framework. However, it should be noted that no any diffraction peaks of MgO in XRD patterns were observed for all samples (Fig. 2). This should be also due to the strong interaction between MgO species with Al-MCM-41 structure, resulting in the facile formation of MgAOASi bonds [34]. Fig. 6 shows NH3-TPD and CO2-TPD profiles of catalysts. As shown in Fig. 6A, two NH3 desorption peaks at low temperature of about 220 °C and high temperature of about 320 °C appeared in the parent Al-MCM-41, which are attributed to the weak acid sites (Lewis acid sites) and strong acid sites (Brønsted acid sites), respectively [35]. Interestingly, a single NH3-desorption peak at a wide temperature ranged from approximately 120 to 650 °C was found for 0.25 wt.% Mg/Al-MCM-41, 0.5 wt.% Mg/Al-MCM-41 and 1 wt.% Mg/Al-MCM-41, suggesting that the introduction of metal

Fig. 4. SEM and EDX mapping images of (A–C) 1 wt.% Mg/Al-MCM-41, (D–F) 5 wt.% Mg/Al-MCM-41 and (G–I) 10 wt.% Mg/Al-MCM-41.

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Fig. 5. Diffuse reflectance UV–Vis spectra of 0.25–10 wt.% Mg/Al-MCM-41. The parent Al-MCM-41 was used as the background.

277

the increase of Mg doping amount while the amount of weak acid sites at lower temperature decreased to some extent. It is possible that at this Mg loading amount, many kinds of Mg species such as MgO, MgAOASi and MgAO pairs were co-existed as can be seen in UV–Vis spectra (Fig. 5), which could easily interact with the active sites of Al-MCM-41, leading to the rearrangement of structure with the formation of new framework between dislocated Al and Mg species [37,38]. However, it also indicates that the basicity of Mg species could dominate the nature of the acid site in catalyst, leading to the reduction of acidity related with the total NH3desorption peak area [39]. Fig. 6B shows CO2-TPD analysis results. One can see that the amount of weak basic site at low temperature increased with the increase in Mg doping amount from 0.25 to 1 wt.%, and after the Mg doping amount was increased over 1 wt. %, basic sites also appear at high temperature. These results indicate that doping of Mg on Al-MCM-41 could result in the generation of new active sites, for example, new framework of Si-Mg-Al could be formed due to the interaction and substitution of proton sites by Mg [10,40]. CO2-TPD of mesoporous MgO is shown in Fig. S2C. The total acidity and basicity areas are summarized in Table 1. One can see that the increase of Mg doping amount from 0.25 to 10 wt.% resulted in the decrease of acidity from 0.274 to 0.094 mmol/g. As for basicity, the highest base site amount of 0.132 mmol/g was obtained in 1 wt.% Mg/Al-MCM-41 catalyst. The coexistence of new acid and basic sites should be beneficial for promoting the deoxygantion reactions as well as the coking resistance during reaction. 3.2. Catalytic performance

Fig. 6. (A) NH3-TPD profiles and (B) CO2-TPD profiles of 0.25–10 wt.% Mg/Al-MCM41.

spices could exchange with the proton of Brønsted acid sites, resulting in the obvious increase of amount of weak acid site [36]. After Mg doping amount was over 1 wt.%, NH3-desorption peak corresponding to the strong acid sites at high temperature was observed, where the areas still continuously increased with

3.2.1. In-situ catalytic upgrading of bio-oil from cellulose pyrolysis Fig. 7A shows the relative amounts of chemical products in the bio-oils derived from fast pyrolysis of cellulose at a reaction temperature of 600 °C in the absence and presence of various catalysts. The major chemicals in bio-oils determined by peak area percentage of GC–MS spectra can be classified into nine groups, i.e., aromatics, aliphatic hydrocarbons, phenols, ketones, aldehydes, furans, sugars, acids and others including ethers, alcohols and nitrogen/sulfur compounds. As can be seen in Fig. 7A, in the absence of catalyst, the relative total amount of oxygenated compounds produced in bio-oil was as high as 92.4%. Here, the identified oxygenated compounds were levoglucosan (LGA), 1,6-a nhydro-b-D-glucofuranose (AGF), hydroxyacetone and propanoic acid, which are usually derived via thermal decomposition and cleavage of glycosidic and b-alkyl-aryl bonds in cellulose structure [41]. In contrast, in the presence of catalyst, the relative total hydrocarbon amount in the upgraded bio-oil was increased from 7.6 to 37.9% by using 0.25 wt.% Mg/Al-MCM-41 catalyst. When 5 wt.% Mg/Al-MCM-41 was applied, the relative total hydrocarbon amount was further increased up to 77.1%. However, in the case of pure MgO, the relative total hydrocarbon amount was only 18.2%. These results indicate that surface area, acidity and basicity had great effect on upgrading quality during deoxygenation reactions including dehydration, decarboxylation and decarbonylation. In addition, as shown in Fig. S3A, with the increase in Mg doping amount on Al-MCM-41, the coke yield also decreased to some extent. This indicates that the coke formation on catalyst via condensation and polymerization reactions could be obstructed by Mg doping via promoting the hydrogen atom migration through C-H activation on the catalytically active sites. The possible reactions for the removal of coke on catalyst with the existence of alkali metal (Me) are proposed as follows [24,42]:

Me2 CO3 þ 2C ! 2Me þ 3CO

ð1Þ

Me2 O-C þ H2 O ! Me2 O2 -C þ H2

ð2Þ

278

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Fig. 7. (A) Chemical compositions and (B) aromatic selectivity in the upgraded bio-oils obtained from in-situ catalytic deoxygenation of bio-oil derived from fast pyrolysis of cellulose at reaction temperature of 600 °C by using various catalysts.

aMe2 O2 -C þ aC ! aMe2 O-C þ aCO

ð3Þ

ð1  aÞMe2 O2 -C þ ð1  aÞCO ! ð1  aÞMe2 O-C þ ð1  aÞCO2

ð4Þ

Overall reaction : aC þ H2 O ! H2 þ ð1  aÞCO2 þ ð2a  1ÞCO ð5Þ As shown in Fig. S3B, more CO was produced when Mg/Al-MCM-41 was used. It indicates that decarbonylation reaction should be the main reaction for deoxygenation of bio-oil in this case. Furthermore, it is found that the use of Mg/Al-MCM-41 not only favored the catalytic deoxgenation, but also promoted the aromatization, oligomerization, alkylation and Diels–Alder reactions since the relative aromatic hydrocarbon amount in the upgraded bio-oil was high. The possible reaction pathways for the catalytic upgrading of bio-oil derived from cellulose are proposed in Fig. 8. Firstly, anhydrosugars are formed via the thermal decomposition of polymer chain with glycosidic bond in cellulose. During this initial process,

CO2, CO and H2O are produced due to the cracking and dehydration. Then, these sugars further undergo dehydration and rearrangement reactions to form furans and small oxygenate compounds [43]. Thereafter, in the presence of catalyst, these intermediate oxygenated compounds are converted to hydrocarbons via deoxygenation reactions. Herein, the selectivity to hydrocarbons should be related to the physicochemical properties of catalyst. Since the ratio of produced aromatic hydrocarbon was high in the upgraded bio-oil, the distribution of aromatics was investigated. Here, the aromatic hydrocarbons were classified into two categories, i.e., (1) MAHs including benzene, toluene, xylenes and ethylbenzene, and (2) PAHs including indenes and naphthalenes. As shown in Fig. 7B and Table S2, with the increase in Mg doping amount, PAHs yield and selectivity to MAHs were decreased obviously. However, it should be noted that the selectivity to benzene was increased to some extent, indicating that Mg doping favored in promoting the aromatization especially for the formation of benzene. It is also possible that the alkylation

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

279

Fig. 8. Reaction pathways for aromatic hydrocarbons production from in-situ catalytic upgrading of bio-oil derived from fast pyrolysis of cellulose.

and aromatization of benzene to PAHs could be retarded by the basic sites resulted from Mg doping. Moreover, the existence of alkali metal and in-situ produced H2O during reaction could promote the generation of H2 via redox cycle reaction in Eq. (5). Here, the in-situ produced H2 might enhance dehydrogenation pathway in the presence of catalyst, leading to the increase of alkanes, which is a source of MAHs. In addition, it is reported that the generated reactive H radicals could promote deoxygenation via capping free radicals [44]. 3.2.2. In-situ catalytic upgrading of bio-oil from lignin pyrolysis Fig. 9A shows the relative amounts of chemical products in the bio-oils derived from fast pyrolysis of lignin at the reaction temperature of 600 °C in the absence and presence of various catalysts. In the absence of catalyst, the relative total hydrocarbon amount was only 14.9%. Here, the oxygenated compounds in original biooil were mainly phenolic compounds such as phenol, 2methoxylphenol, cresol, and so on, which were derived from the decomposition of CAO (b-O-4, a-O-4, 4-O-5 linking style) and CAC (b-5, 5-5, b-1, b-b linking style) bonds in lignin structure [45]. When pure MgO was used as the catalyst, the relative total hydrocarbon amount was 37.3% while the relative amount of produced phenol of 44.3% still existed in the upgraded bio-oil. When Mg/Al-MCM-41 was used, as shown in Fig. S4, the yields of H2O and H2 were higher than those in the case using cellulose as feedstock, indicating that the main deoxygenation reactions should be dehydration, dehydroxylation and cracking reactions. Here, the relative total hydrocarbon amounts in the upgraded bio-oils were increased to 79.6% and 90.6% when 0.25 wt.% Mg/Al-MCM-41 and 0.5 wt.% Mg/Al-MCM-41 were used respectively. However, the relative total hydrocarbon amount was almost unchanged when Mg doping amount was increased from 0.5 wt.% to the range of 1–10 wt.%. Furthermore, it should be noted that lower Mg dop-

ing amount was better for the upgrading of lignin-derived bio-oil than the upgrading of cellulose-derived bio-oil. It is possible that higher acidity with Lewis acid site of the catalyst was more beneficial for the catalytic cracking of the larger molecules as in ligninderived bio-oil. The possible reaction pathways for catalytic upgrading of lignin-derived bio-oil to aromatic hydrocarbons are proposed in Fig. 10. During lignin pyrolysis, large molecules of phenol alkoxy compounds are formed via thermal cracking, depolymerization and dehydration [19]. The lignin char is easily obtained with higher amount when compared with cellulose pyrolysis. In the presence of mesoporous catalyst, these oxygenated compounds diffuse into the catalyst layer, and contact with the active sites on the surfaces and inside the pores. Here, higher acidity is needed for the cleavage of CAO and CAC bonds before the deoxygenation reactions in order to obtain smaller hydrocarbon molecules such as BTXs. However, if the acidity is too high, further aromatization and polymerization of BTXs to form PAHs and coke could occur. Thus, it is necessary to tune the acid and basic sites on the catalyst, and introducing some basic sites on the acid catalyst like Al-MCM-41 is important to improve the stability of catalyst. Fig. 9B and Table S3 show the distribution of aromatic hydrocarbons in the upgraded bio-oils derived from fast pyrolysis of lignin using various catalysts. One can see that all catalysts exhibited the selectivity of BTXs up to 80%. The highest BTXs selectivity as high as 93.7% was achieved by using 10 wt.% Mg/Al-MCM-41. Interestingly, no indene and only a little amount of naphthalenes in upgraded bio-oil were found, indicating that this catalyst had very high selectivity for BTXs production. Comparing with the results by using cellulose as feedstock, the selectivity to PAHs was also lower. In general, more PAHs could result in more coke formed on catalyst [46]. That is why the coke yield in this case was lower than that when using cellulose as feedstock.

280

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Fig. 9. (A) Chemical compositions and (B) aromatic selectivity in the upgraded bio-oils obtained from in-situ catalytic deoxygenation of bio-oil derived from fast pyrolysis of lignin at reaction temperature of 600 °C by using various catalysts.

3.2.3. In-situ catalytic upgrading of bio-oil from sunflower stalk pyrolysis Fig. 11A shows the relative amounts of chemical products in the bio-oils derived from fast pyrolysis of sunflower stalk at the reaction temperature of 600 °C in the absence and presence of various catalysts. One can see that various oxygenated compounds, especially phenols, sugars and ketone compounds, were found in non-upgraded bio-oil since the practical biomass, i.e., sunflower stalk, is mainly consisted of cellulose, hemicellulose and lignin. In the presence of catalyst, it is found that almost all relative amounts of produced oxygenated compounds were reduced to some extent while the relative total hydrocarbon amount was increased. Interestingly, for the catalysts with Mg doping amounts in the range of 0.5–10 wt.%, no phenol was found in the upgraded bio-oil. This is probably because phenols could be easily converted to BTXs in the presence of Mg/Al-MCM-41. Fig. 11B and Table S4 present the distribution of aromatic hydrocarbons in the upgraded bio-

oils. One can see that the trend of aromatic distribution in the upgraded bio-oil was the similar as the cases using cellulose or lignin as the feedstock. Here, 1 wt.% Mg/Al-MCM-41 exhibited the highest activity and selectivity for catalytic upgrading of bio-oil from the fast pyrolysis of sunflower stalk with the relative total hydrocarbon amount of 86.2%. Therefore, in the following study, 1 wt.% Mg/Al-MCM-41 was also selected to investigate the effect of reaction temperature on the catalyst performance, comparing with the parent Al-MCM-41. As shown in Fig. 12A and C, using the parent Al-MCM-41 and 1 wt.% Mg/Al-MCM-41, the relative total hydrocarbon amount increased with the increase in the reaction temperature from 500–800 °C, suggesting that high temperature promoted the deoxygenation reactions. The maximum relative total hydrocarbon amount of 94.2% was obtained at 700 °C using 1 wt.% Mg/Al-MCM41. In contrast, when Al-MCM-41 was used, the maximum relative total hydrocarbon amount was only 86.5% at 800 °C. This indicates

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

281

Fig. 10. Reaction pathways for aromatic hydrocarbons production from in-situ catalytic upgrading of bio-oil derived from fast pyrolysis of lignin.

Fig. 11. (A) Chemical compositions and (B) aromatic selectivity in the upgraded bio-oils obtained from in-situ catalytic deoxygenation of bio-oil derived from fast pyrolysis of sunflower stalk at reaction temperature of 600 °C by using various catalysts.

282

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Fig. 12. The effect of reaction temperature on chemical compositions and aromatic selectivity in the upgraded bio-oils obtained from in-situ catalytic deoxygenation of bio-oil derived from fast pyrolysis of sunflower stalk by using (A,B) Al-MCM-41 and (C,D) 1 wt.% Mg/Al-MCM-41.

that Mg-doped Al-MCM-41 promoted catalytic deoxygenation at lower temperature when compared with the parent Al-MCM-41. Fig. 12B and D as well as Table S5 present the distribution of aromatic hydrocarbons in the upgraded bio-oils at different reaction temperatures using Al-MCM-41 and 1 wt.% Mg/Al-MCM-41. Considering with the relatively low temperature of 500 °C, 1 wt.% Mg/Al-MCM-41 exhibited high selectivity towards BTXs, and the selectivity of aromatic hydrocarbons in the upgraded bio-oil was in the order of benzene > toluene > xylenes > naphthalenes > ethylbenzene > indenes > others. In contrast, for Al-MCM-41, the order was changed to naphthalenes > indenes > toluene > xylenes > benzene > other > ethylbenzene, indicating that Al-MCM-41 without Mg doping favored in promoting the alkylation with MAHs to form PAHs. Recently, Stephanidis et al. [47] also reported the similar trend where PAHs were significantly increased by using Al-MCM41. Braga et al. [48] found that loading of 15–30% of WO3 on MCM-41 had a great effect on BTXs production from the catalytic upgrading of bio-oil. However, due to the higher loading amount and more expensive of WO3 when compared with Mg doping in this study, Mg/Al-MCM-41 should be currently considered as an alternative solid catalyst which has high efficiency for the upgrading of bio-oil and production of high value-added chemicals. In addition, with the increase of temperature from 500 to 700 °C, the selectivity towards MAHs such as benzene and toluene obviously increased. It is possible that the dealkylation of xylenes and PAHs occurred at high temperatures. On the basis of the above results, it is obvious that Mg doping and reaction temperature played significant roles in the product distribution of the upgraded bio-oil.

3.3. Catalyst reusability The reusability of 1 wt.% Mg/Al-MCM-41 catalyst was tested at an optimum temperature of 700 °C by using sunflower stalk as feedstock for the pyrolysis. Fig. 13A and B shows the results by reusing the catalyst for 5 cycles without any regeneration. One can see that 1 wt.% Mg/Al-MCM-41 exhibited high stability with only a slight reduction of the relative total hydrocarbon amount in the upgraded bio-oil in each reuse cycle, where the reduction percentage was less than 25% from the first to the fifth cycle. In the case of Al-MCM-41, serious reduction on the relative total hydrocarbon amount in the upgrade bio-oil was obviously observed. This reduction should be attributed to the coke deposition amount on active surface of catalyst during reaction, leading to the deactivation of catalyst. Table 2 shows the coke yield on the spent catalysts in each cycle. It is found that Al-MCM-41 without Mg doping had higher coke yield than Mg/Al-MCM-41. Moreover, the coke yield was obviously increased with the increase in the reuse cycle number. This should be the reason why the decreasing of hydrocarbon amount occurred in each cycle. Fig. 14 shows thermal decomposition amounts of coke on the surface of spent catalyst. As observed, two thermal decomposition peaks at low and high temperatures were attributed to the oxygenated coke or soft coke and the graphite-like coke or hard coke, respectively [49,50]. Here, one can see that Mg doping on Al-MCM-41 resulted in an obvious decrease in the thermal coke decomposition temperature range from ‘‘350–750 °C” to ‘‘250–650 °C”, suggesting that Mg favored the delaying of hard coke formation. It is possible that the changing of coke species formation might be related with the

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

283

Fig. 13. Reusability of (A) 1 wt.% Mg/Al-MCM-41 and (B) regenerated 1 wt.% Mg/Al-MCM-41 (after 5th reuse) for the in-situ catalytic deoxygenation of bio-oil derived from fast pyrolysis of sunflower stalk at reaction temperature of 700 °C.

existence of Mg species in the vicinity of support, which could have the synergy effect between proton sites with metal/metal oxide sites in the support structure [51]. From these results, the spent Mg-doped catalyst can be easily regenerated by calcination at lower temperature. Moreover, as shown in Fig. 13C, when the spent 1 wt.% Mg/AlMCM-41 catalyst (5th reuse) was calcined at 650 °C for 30 min

under air atmosphere, its activity was perfectly recovered. Moreover, when this regenerated spent catalyst was further reused for 5 cycles without regeneration. Interestingly, it showed better reusability with lower reduction percentage of the relative total hydrocarbon amount. Also, as shown in Table 2, the regenerated spent catalyst had lower coke yield in each cycle. This phenomenon could be explained from the existence of AAEM species

284

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285

Table 2 The amount of coke deposition on spent catalysts in each cycle after reaction (1st–5th reuse). Reused cycle

1st 2nd 3rd 4th 5th

Coke deposition amount (%)

reusability testing. It is expected that this kind of catalyst can be applied for the in-situ upgrading of bio-oil in a practical process. Acknowledgments

Al-MCM-41

1 wt.% Mg/Al-MCM-41

Regenerated 1 wt.% Mg/Al-MCM-41

5.3 6.4 7.9 9.5 11.3

2.8 4.7 6.1 7.4 8.9

2.2 3.1 4.5 5.8 6.7

This study is supported by Aomori City Government, Japan. S. Karnjanakom, and A. Bayu greatly acknowledge the Ministry of Education, Culture, Sports, Science, and Technology (MEXT) of Japan for the scholarship. The authors thanks Mr Yusuke Tsushima at Faculty of Agriculture and Life Science in Hirosaki University for TEM analysis. Appendix A. Supplementary material Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.enconman.2017. 03.049. References

Fig. 14. DTG profiles of the spent catalysts after reaction (5th reuse).

on the spent catalyst after reaction, which were derived from the decomposition of biomass [24,52,53]. Fig. S6 shows NH3-TPD and CO2-TPD profiles of the spent catalyst. The temperature range of NH3 or CO2 desorption of spent catalyst was shifted obviously when compared with the fresh ones. Moreover, as shown in Table S6, the acidity of spent catalyst also significantly decreased while the basicity increased when compared with the fresh one. This also identified that the contribution of AAEM on the basicity of the spent catalyst. In addition, after the spent catalyst was regenerated, the data showed the absence of a significant pore blockage as well as the slight decreasing of surface area, indicating that the carbon deposited on surface of catalyst can be removed easily. 4. Conclusions Catalytic activity, selectivity and stability of Al-MCM-41 catalyst were improved by doping with Mg species for in-situ upgrading of bio-oil derived from fast pyrolysis of biomass. The characterization results indicated that the doped Mg species showed good dispersion on Al-MCM-41 structure. The performance tests confirmed that bifunctional Mg/Al-MCM-41 promoted the catalytic deoxygenation as well as aromatization, resulting in the formation of aromatic hydrocarbons with high yield. Mg/AlMCM-41 showed high selectivity towards BTXs, especially towards benzene production. It is found that more BTXs were produced in the upgrading of bio-oil derived from fast pyrolysis of lignin than those from sunflower stalk and cellulose. The coke deposition amount on spent catalyst obviously reduced with the increase in Mg doping amount. No serious reduction of the ratio of produced hydrocarbon in the upgraded bio-oil was observed in the catalyst

[1] Zheng A, Zhao Z, Chang S, Huang Z, Wu H, Wang X, et al. Effect of crystal size of ZSM-5 on the aromatic yield and selectivity from catalytic fast pyrolysis of biomass. J Mol Catal A: Chem 2014;383–384:23–30. [2] Li F, Yuan Y, Huang Z, Chen B, Wang F. Sustainable production of aromatics from bio-oils through combined catalytic upgrading with in situ generated hydrogen. Appl Catal B 2015;165:547–54. [3] Wang J, Zhong Z, Song Z, Ding K, Deng A. Modification and regeneration of HZSM-5 catalyst in microwave assisted catalytic fast pyrolysis of mushroom waste. Energy Convers Manage 2016;123:29–34. [4] Cole DP, Lee YJ. Effective evaluation of catalytic deoxygenation for in situ catalytic fast pyrolysis using gas chromatography–high resolution mass spectrometry. J Anal Appl Pyrolysis 2015;112:129–34. [5] Widayatno WB, Guan G, Rizkiana J, Du X, Hao X, Zhang Z, et al. Selective catalytic conversion of bio-oil over high-silica zeolites. Biores Technol 2015;179:518–23. [6] Zhang B, Zhong ZP, Wang XB, Ding K, Song ZW. Catalytic upgrading of fast pyrolysis biomass vapors over fresh, spent and regenerated ZSM-5 zeolites. Fuel Process Technol 2015;138:430–4. [7] Galadima A, Muraza O. In situ fast pyrolysis of biomass with zeolite catalysts for bioaromatics/gasoline production: a review. Energy Convers Manage 2015;2015:338–54. [8] Li J, Li X, Zhou G, Wang W, Wang C, Komarneni S, et al. Catalytic fast pyrolysis of biomass with mesoporous ZSM-5 zeolites prepared by desilication with NaOH solutions. Appl Catal A 2014;470:115–22. [9] Yu Y, Li X, Su L, Zhang Y, Wang Y, Zhang H. The role of shape selectivity in catalytic fast pyrolysis of lignin with zeolite catalysts. Appl Catal A 2012;447– 448:115–23. [10] Park HJ, Heo HS, Jeon JK, Kim J, Ryoo R, Jeong KE, et al. Appl Catal B 2010; l95:365–73. [11] Mihalcik DJ, Mullen CA, Boateng AA. Screening acidic zeolites for catalytic fast pyrolysis of biomass and its components. J Anal Appl Pyrolysis 2011;92:224–32. [12] Custodis VBF, Karakoulia SA, Triantafyllidis KS, Bokhoven JAV. Catalytic fast pyrolysis of lignin over high-surface-area mesoporous aluminosilicates: effect of porosity and acidity. Chemsuschem 2016;9:1134–45. [13] Lu Q, Tang Z, Zhang Y, Zhu XF. Catalytic upgrading of biomass fast pyrolysis vapors with Pd/SBA-15 catalysts. Ind Eng Chem Res 2010;49:2573–80. [14] Naik SP, Ryu VBT, Miller JD, Zmierczak W. Al-MCM-41 as methanol dehydration catalyst. Appl Catal A 2010;381:183–90. [15] Adam J, Antonakou A, Lappas A, Stöcker M, Nilsen MH, Bouzga A, et al. In situ catalytic upgrading of biomass derived fast pyrolysis vapours in a fixed bed reactor using mesoporous materials. Microporous Mesoporous Mater 2006;96:93–101. [16] Iliopoulou EF, Antonakou EV, Karakoulia SA, Vasalos IA, Lappas AA, Triantafyllidis KS. Catalytic conversion of biomass pyrolysis products by mesoporous materials: effect of steam stability and acidity of Al-MCM-41 catalysts. Chem Eng J 2007;134:51–7. [17] Guisnet M, Magnoux P. Organic chemistry of coke formation. Appl Catal A 2001;212:83–96. [18] Vichaphund S, Aht-ong D, Sricharoenchaikul V, Atong D. Production of aromatic compounds from catalytic fast pyrolysis of Jatropha residues using metal/HZSM-5 prepared by ion-exchange and impregnation methods. Renewable Energy 2015;79:28–37. [19] Zheng Y, Chen D, Zhu X. Aromatic hydrocarbon production by the online catalytic cracking of lignin fast pyrolysis vapors using Mo2N/Al2O3. J Anal Appl Pyrolysis 2013;104:514–20. [20] Veses A, Aznar M, Martínez I, Martínez JD, López JM, Navarro MV, et al. Catalytic pyrolysis of wood biomass in an auger reactor using calcium-based catalysts. Biores Technol 2014;162:250–8.

S. Karnjanakom et al. / Energy Conversion and Management 142 (2017) 272–285 [21] Lin Y, Zhang C, Zhang M, Zhang J. Deoxygenation of bio-oil during pyrolysis of biomass in the presence of CaO in a fluidized-bed reactor. Energy Fuels 2010;24:5686–95. [22] Stefanidis SD, Karakoulia SA, Kalogiannis KG, Iliopoulou EF, Delimitis A, Yiannoulakis H, et al. Natural magnesium oxide (MgO) catalysts: A costeffective sustainable alternative to acid zeolites for the in situ upgrading of biomass fast pyrolysis oil. Appl Catal B 2016;196:155–73. [23] Zabeti M, Nguyen TS, Lefferts L, Heeres HJ, Seshan K. In situ catalytic pyrolysis of lignocellulose using alkali-modified amorphous silica alumina. Biores Technol 2012;118:374–81. [24] Karnjanakom S, Guan G, Asep B, Du X, Hao X, Samart C, et al. Catalytic steam reforming of tar derived from steam gasification of sunflower stalk over ethylene glycol assisting prepared Ni/MCM-41. Energy Convers Manage 2015;98:359–68. [25] Roik NV, Belyakova LA. Sol–gel synthesis of MCM-41 silicas and selective vapor-phase modification of their surface. J Solid State Chem 2013;207:194–202. [26] Peng W, Li J, Chen B, Wang N, Luo G, Wei F. Mesoporous MgO synthesized by a homogeneous-hydrothermal method and its catalytic performance on gasphase acetone condensation at low temperatures. Catal Commun 2016;74:39–42. [27] Karnjanakom S, Guan G, Asep B, Hao X, Kongparakul S, Samart C, et al. Catalytic upgrading of bio-oil over Cu/MCM-41 and Cu/KIT-6 prepared by bcyclodextrin-assisted coimpregnation method. J Phys Chem C 2016;120:3396–407. [28] Sun L, Zhang X, Chen L, Zhao B, Yang S, Xie X. Comparison of catalytic fast pyrolysis of biomass to aromatic hydrocarbons over ZSM-5 and Fe/ZSM-5 catalysts. J Anal Appl Pyrolysis 2016;121:342–6. [29] Dong CQ, Zhang ZF, Lu Q, Yang YP. Characteristics and mechanism study of analytical fast pyrolysis of poplar wood. Energy Convers Manage 2012;57:49–59. [30] Shen D, Zhao J, Xiao R. Catalytic transformation of lignin to aromatic hydrocarbons over solid-acid catalyst: effect of lignin sources and catalyst species. Energy Convers Manage 2016;124:61–72. [31] Fakhri A, Behrouz S. Comparison studies of adsorption properties of MgO nanoparticles and ZnO–MgO nanocomposites for linezolid antibiotic removal from aqueous solution using response surface methodology. Process Saf Environ Prot 2015;94:37–43. [32] Janssens W, Makshina EV, Vanelderen P, Clippel FD, Houthoofd K, Kerkhofs S, et al. Ternary Ag/MgO-SiO2 catalysts for the conversion of ethanol into butadiene. Chemsuschem 2015;8:994–1008. [33] Angelici C, Velthoen MEZ, Weckhuysen BM, Bruijnincx PCA. Effect of preparation method and CuO promotion in the conversion of ethanol into 1,3-butadiene over SiO2–MgO catalysts. Chemsuschem 2014;7:2505–15. [34] Fujita SI, Segawa S, Kawashima K, Nie X, Erata T. One-pot room-temperature synthesis of Mg containing MCM-41 mesoporous silica for aldol reactions. J Mater Sci Technol 2016. http://dx.doi.org/10.1016/j.jmst.2016.08.02. [35] Zhang D, Wang R, Yang X. Beckmann rearrangement of cyclohexanone oxime over Al-MCM-41 and P modified Al-MCM-41 molecular sieves. Catal Commun 2011;12:399–402.

285

[36] Tshabalala TE, Scurrell MS. Aromatization of n-hexane over Ga, Mo and Zn modified H-ZSM-5 zeolite catalysts. Catal Commun 2015;72:49–52. [37] Rizkiana J, Guan G, Widayatno WB, Yang J, Hao X, Matsuoka K, et al. Mgmodified ultra-stable Y type zeolite for the rapid catalytic co-pyrolysis of lowrank coal and biomass. RSC Adv 2016;6:2096–105. [38] Intana T, Föttinger K, Rupprechter G, Kongkachuichay P. Physicochemical properties of Cu loaded onto core–shell Al-MCM-41: effect of loading methods. Colloids Surf A 2015;467:157–65. [39] Asthana S, Samanta C, Bhaumik A, Banerjee B, Voolapalli RK, Saha B. Direct synthesis of dimethyl ether from syngas over Cu-based catalysts: enhanced selectivity in the presence of MgO. J Catal 2016;334:89–101. [40] Iliopoulou EF, Stefanidis SD, Kalogiannis KG, Delimitis A, Lappas AA, Triantafyllidis KS. Catalytic upgrading of biomass pyrolysis vapors using transition metal-modified ZSM-5 zeolite. Appl Catal B 2012;127:281–90. [41] Muley PD, Henkel C, Abdollahi KK, Marculescu C, Boldor D. A critical comparison of pyrolysis of cellulose, lignin, and pine sawdust using an induction heating reactor. Energy Convers Manage 2016;117:273–80. [42] Mckee DW. Mechanisms of the alkali metal catalysed gasification of carbon. Fuel 1983;62:170–5. [43] Rezaei PS, Shafaghat H, Daud WMAW. Production of green aromatics and olefins by catalytic cracking of oxygenate compounds derived from biomass pyrolysis: A review. Appl Catal A 2014;469:490–511. [44] Melligan F, Hayes MHB, Kwapinski W, Leahy JJ. Hydro-pyrolysis of biomass and online catalytic vapor upgrading with Ni-ZSM-5 and Ni-MCM-41. Energy Fuels 2012;26:6080–90. [45] Britt PF, Iii ACB, Cooney MJ, Martineau DR. Flash vacuum pyrolysis of methoxysubstituted lignin model compounds. J Org Chem 2000;65:1376–89. [46] Kelkar S, Saffron CM, Andreassi K, Li Z, Murkute A, Miller DJ, et al. A survey of catalysts for aromatics from fast pyrolysis of biomass. Appl Catal B 2015;174– 175:85–95. [47] Stephanidis S, Nitsos C, Kalogiannis K, Iliopoulou EF, Lappas AA, Triantafyllidis KS. Catalytic upgrading of lignocellulosic biomass pyrolysis vapours: effect of hydrothermal pre-treatment of biomass. Catal Today 2011;167:37–45. [48] Braga RM, Melo DMA, Sobrinho EV, Barros JMF, Melo MAF, Carvalho AFM, et al. Catalytic upgrading of Elephant grass (Pennisetum purpureum Schum) pyrolysis vapor using WO3 supported on RHA and RHA-MCM-41. Catal Today 2017;279:224–32. [49] Fan Y, Cai Y, Li X, Yin H, Xia J. Coking characteristics and deactivation mechanism of the HZSM-5 zeolite employed in the upgrading of biomassderived vapors. J Ind Eng Chem 2017;46:139–49. [50] Zhang H, Shao S, Xiao R, Shen D, Zeng J. Characterization of coke deposition in the catalytic fast pyrolysis of biomass derivates. Energy Fuels 2014;28:52–7. [51] Widayatno WB, Guan G, Rizkiana J, Yang J, Hao X, Tsutsumi A, et al. Upgrading of bio-oil from biomass pyrolysis over Cu-modified b-zeolite catalyst with high selectivity and stability. Appl Catal B 2016;186:166–72. [52] Guan G, Chen G, Kasai Y, Lim EWC, Hao X, Kaewpanha M, et al. Catalytic steam reforming of biomass tar over iron-or nickel-based catalyst supported on calcined scallop shell. Appl Catal B 2012;115–116:159–68. [53] Paasikallio V, Lindfors C, Kuoppala E, Solantausta Y, Oasmaa A, Lehto J, et al. Green Chem 2014;16:3549–59.