Aqueous phase reforming and hydrodeoxygenation of ethylene glycol on Pt/SiO₂–Al₂O₃: effects of surface acidity on product distribution

Abstract

Aqueous-phase reforming (APR) and aqueous-phase hydrodeoxygenation (APH) reactions of ethylene glycol (EG) were investigated using platinum supported solid-acid SiO₂–Al₂O₃ catalysts with different Si/Al molar ratios. The molar ratio of Si/Al on the SiO₂–Al₂O₃ mixed metal oxides largely altered the surface area due to changes to the acidity, as well as changing the reduction behavior of the supported platinum nanoparticles. The Pt/SiO₂–Al₂O₃ catalysts with a Si/Al molar ratio of 0.1 showed a higher activity for APR as well as APH. Among the various properties of Pt/SiO₂–Al₂O₃, the amount of acid sites on the SiO₂–Al₂O₃ supports changed the EG conversion and production distribution with different coke depositions. The acidic property was a more dominant factor for the catalytic activity than the affects of the platinum crystallite size on the reduction behavior. The easy and simultaneous cleavages of C–C as well as the C–O bonds in EG on the Bronsted acid sites of Pt/SiO₂–Al₂O₃ catalysts were responsible for a higher EG conversion and hydrocarbon formation. A larger number of weak acid sites was also related to the formation of larger hydrocarbons and a lower coke deposition. Compared with Pt/Al₂O₃, improved catalytic acidity with a low coke deposition was observed for Pt/SiO₂–Al₂O₃ at a Si/Al molar ratio of 0.1. This can mainly be attributed to the easy control of weak and strong acid sites with a high dispersion of platinum crystallites by simply changing the Si/Al molar ratio of the SiO₂–Al₂O₃ mixed metal oxides.

---

1. Introduction

Traditional fossil fuels such as coal, crude oil and natural or shale gases can generate energy with the emission of environmental pollutants. Renewable energy sources such as biomass-derived hydrogen can be one of alternative candidates to overcome the environmental problems of fossil fuels and limited resources. Biomass-derived fuels show promise for complete carbon recycling through transformation processes such as gasification, pyrolysis, fermentation, chemical conversion and so on.^1–3 Among them, the aqueous-phase reforming (APR) reaction of biomass-derived intermediates seems to have many advantages due to its low operating temperature and high purity hydrogen production with a low concentration of CO. Generally, some supported novel metals such as Pt, Ru, Ir, Rh, Ni or Pd are known to be active for the APR reaction with a high selectivity to hydrogen.⁴ In particular, Pt-supported catalysts on various supports such as Al₂O₃, SiO₂, zeolites and Al₂O₃–SiO₂ and so on have been investigated for the APR reaction through characteristic reactions such as the main APR reaction, water–gas shift (WGS) reaction, dehydrogenation, methanation and decomposition, as represented in eqn (1)–(4) using a biomass-derived model component of ethylene glycol (EG).^5–7

C₂H₆O₂ + 2H₂O → 2CO₂ + 5H₂ (APR) (1)

CO + H₂O → CO₂ + H₂ (WGS) (2)

C₂H₆O₂ → C₂H_6−xO₂ + (x/2)H₂ (dehydrogenation) (3)

CO + 3H₂ → CH₄ + H₂O (methanation) (4)

The reaction pathways of the APR reaction for hydrogen generation using a EG feedstock are generally known to proceed first by the dehydrogenation reaction step of EG followed by simultaneous C–C cleavage and WGS reaction. The methanation reaction is one of the main side reactions, which can also be strongly related to the extent of coke formation.^8,9 Moreover, the byproducts such as ethylene, CO and other oxygenates can also cause severe coke formation on the catalyst surface.^10–13 Hydrogen and alkanes can be produced by separate C–C and C–O bond cleavages during the APR reaction.^14–17 The oxygenates, such as alcohols and acetaldehydes, can also be formed by simultaneous dehydration on Ru-based APR catalysts; however, these oxygenated intermediates can easily be transformed to hydrogen and CO₂ by a facile WGS reaction as well.¹⁸ The selective C–C or C–O bond cleavages have been known to generate various products during the APR reaction, where C–C bond cleavage (eqn (5)) can selectively produce hydrogen and CO₂ with a combination of the WGS reaction and C–O bond cleavage (eqn (6)), which can selectively produce hydrocarbons as well. The simultaneous C–O and C–C cleavages (eqn (7)) can also happen easily since these cleavages are very competitive reactions during the APR reaction. The preferential activities of separate C–C or C–O bond cleavages are strongly dependent on the types of active metals, which seems to be an important catalyst design factor due to the high O/C ratio of the biomass-derived intermediates.¹⁹

C₂H₆O₂ → 2CO + 3H₂ (C–C cleavage) (5)

C₂H₆O₂ → C₂H₄ + C₂H₆ + 2H₂O (C–O cleavage) (6)

C₂H₆O₂ → CH₄ + CO + CH₂O + CH₃OH (C–C with simultaneous C–O cleavage) (7)

In addition, the aqueous-phase hydrodeoxygenation (APH) reaction is a known efficient production method of alkanes and petrochemicals using biomass-derived chemicals with a similar catalytic system to APR catalysts.^20–22 However, the APH reaction requires a much higher activity for C–C cleavage to remove oxygen atoms by dehydration and add hydrogen atoms through hydrogenation and hydrogenolysis. In general, supported metal catalysts containing active palladium species are known to be active for C=O and C=C bond cleavage in furfural by hydrogenation,²³ and ruthenium-based catalysts have been reported as the most active catalysts for the APH reaction based activation of non-furanic carbonyl groups, which also requires different active sites compared to the APR reaction. Therefore, APH catalysts generally require different active sites for C–C cleavage through a retro-aldol condensation with decarbonylation on the metallic sites and C–O cleavage by selective dehydration on the acidic sites.²⁴ Based on the previous studies,^{15,22,25–30} the C–C and C–O bond cleavages are competitive reactions at the very beginning of the retro-aldol condensation, and the dehydration step on the metallic sites of the novel metals, as well as on the acid sites of solid-acid surfaces.^15,22,25,26 In addition, the formed alcohols, such as methanol, during C–C and C–O bond cleavage can be converted to CO₂ and hydrogen by dehydrogenation followed by the WGS reaction, which seems to be a competitive reaction on metallic platinum sites, such as Pt(111) surfaces.^27–30 Among the competitive and complicated reactions during the APR and APH reactions, the hydrogenation and dehydration steps are known to be essential steps for an easy transformation of the biomass-derived intermediates. Those reactions selectively occur on heterogeneous or homogeneous catalytic systems when using bimetal or metal complexes with a combination of solid-acid zeolites and metal oxides.^31–34 In addition, hydrogenolysis and dehydration with a subsequent dehydrogenation reaction have also been investigated using some solid-acid catalysts to efficiently remove oxygen atoms in biomass-derived reactants.^35–40

In the present investigation, simple and prototype Pt/SiO₂–Al₂O₃ catalysts with different acidic properties due to different Si/Al ratios were investigated for APR and APH reactions to verify how the amount and type of acidic sites affects the different catalytic activities and product distributions. Even though some Pt-supported solid acid catalysts have been extensively reported, the comparative studies of the APR and APH reactions using the same Pt/SiO₂–Al₂O₃ catalysts have been scarcely reported until now as far as we know, especially when comparing the different hydrocarbon distributions according to the surface acidic properties.

2. Experimental section

2.1. Catalyst preparation and activity test of the Pt/SA catalysts

SiO₂–Al₂O₃ mixed metal oxides (SA) with a different Si/Al molar ratios from 0 to 1.0 were synthesized by the coprecipitation method. In more detail, the SiO₂–Al₂O₃ solid acid supports were prepared using aluminum nitrate nonahydrate (Al(NO₃)₃·9H₂O, Daejung) and tetraethylorthosilicate (TEOS, Si(OC₂H₅)₄, Alfa) as precursors after dissolving the separate precursors in a mixed solvent of deionized water and ethanol. The dissolved separate metal precursors were vigorously stirred for 15 minutes separately, and the separate metal precursor solutions were then carefully mixed together. For the subsequent coprecipitation of the above two solutions, 30 vol% ammonia solution was dropped slowly into the mixed metal precursor solutions which were stirred for 3 h at ambient temperature. The as-prepared gel-like slurry was dried under vacuum conditions at 60 °C for 2 h followed by drying at 110 °C overnight. Finally, the as-prepared sample was calcined at 300 °C for 4 h to prepare a solid acid mixed metal oxide of SiO₂–Al₂O₃.⁴⁰ To prepare the platinum supported catalysts using the previously prepared SA supports, tetraammineplatinum(II) nitrate (Pt(NO₃)₄(NO₃)₂) precursor was loaded over the as-prepared SA support by the incipient wetness impregnation method at a fixed platinum metal amount from 1 to 7 wt% using deionized water. The platinum supported SA catalysts were dried overnight at 80 °C followed by calcining at 350 °C for 4 h. The as-prepared catalyst was denoted as Pt/SA(x), where SA stands for the solid-acid SiO₂–Al₂O₃ support and x for different molar ratios of Si/Al. In addition, Pt(y)/SA(0.1) catalysts were also prepared similarly and y represents the weight% of the platinum metal on the SA(0.1) support.

The APR reaction was carried out in a fixed-bed tubular reactor with an outer diameter of 9.5 mm at the reaction conditions of T = 250 °C, P = 4.5 MPa and weight hourly space velocity (WHSV) = 2.0 h⁻¹ for around 20 h using 0.3 g catalyst. Prior to the activity test, the catalyst was reduced at 350 °C for 2 h under a flow of 5 vol% H₂ balanced with N₂. The reaction temperature was adjusted using a PID controller with a K-type thermocouple and the pressure was controlled using a back pressure regulator. The reactant, ethylene glycol (EG), at 10 wt% in an aqueous solution (molar ratio of EG/H₂O = 0.383) was fed into the reactor at a flow rate of 0.1 mL min⁻¹ using a HPLC pump (SP930D, YoungLin) with an N₂ flow with a flow rate of 30 mL min⁻¹, which was used as a carrier and an internal standard gas. The APH reaction was also carried out at similar reduction and reaction conditions with 0.5 g catalyst using a mixed gas of 80 vol% H₂ balanced with N₂. The catalytic activity was measured at the reaction conditions of T = 260 °C, P = 5.0 MPa and WHSV = 0.6 h⁻¹ for around 20 h. The effluent gases from the reactor were analyzed in situ using a gas chromatograph (6500 GC, YoungLin) equipped with a thermal conductivity detector (TCD) connected to a Carboxen 1000 column as well as a flame ionization detector (FID) connected with a HP-PLOT/Q capillary column. The conversion of EG and product distributions were calculated using the following equations based on an internal standard gas of N₂, and the production rate of H₂ was also calculated using the produced moles of H₂ with a unit of mL (g_cat h)⁻¹.

- EG conversion to gaseous products (mol%) = (moles of C atoms in a gas-phase/moles of C atoms in the EG feed) × 100.

- Selectivity of gaseous products (mol%) = (moles of selected product in a gas-phase/total moles of carbons formed in a gas-phase) × 100.

2.2. Catalyst characterizations

Powder X-ray diffraction (XRD) patterns of the fresh and used Pt/SA catalysts were obtained using D8 ADVANCE (Bruker) instrument with a Cu-Kα radiation of wavelength of 15.406 nm at an operating voltage of 40 kV with a scanning rate of 4° min⁻¹ in the diffraction range of 2θ = 20–70° to verify the crystalline phases and crystallite sizes. The crystallite sizes of the supported platinum nanoparticles were calculated using a value of full width at half maximum (FWHM) by the Debye–Scherrer equation with the most intensive diffraction peak of metallic platinum at 2θ = 40.0°. Thermogravimetric analysis (TGA) of the used Pt/SA catalysts was carried out using a Seiko Exstar 6000 (TG/DTA6100) to confirm the quantity of the deposited coke after 20 h of the APR reaction on stream. The Si/Al molar ratio of the mixed metal oxides was further characterized by X-ray fluorescence (XRF) analysis using a Bruker AXS S4 Pioneer that operated at 60 kV and 150 mA.

The specific surface area, pore volume and average pore diameter were measured by N₂ adsorption–desorption analyses using a Micromeritics TriStar II instrument working at liquid nitrogen temperature of −196 °C. The specific surface area of the fresh Pt/SA catalysts was calculated by the Brunauer–Emmett–Teller (BET) method, and the average pore diameter and its pore size distribution were measured by the Barrett–Joyner–Halenda (BJH) method from the desorption isotherm. Prior to analysis, the sample was pretreated by degassing at 90 and 350 °C for 4 h consecutively. In addition, H₂ chemisorption analysis of the Pt/SA catalysts was also carried out using a Micromeritics ASAP 2020 instrument. The metal surface area and dispersion of the supported platinum nanoparticles were obtained using the amount of hydrogen adsorbed on the reduced Pt/SA catalyst at 35 °C. Prior to analysis, the sample was pretreated under vacuum conditions of 10 mmHg at 300 °C for 2 h to remove any impurities on the Pt/SA surfaces. The sample was subsequently reduced at 350 °C for 2 h, which are the same conditions as the reduction step before the APR reaction. The dispersion, metal surface area and crystallite size of the supported platinum nanoparticles were calculated with an assumption of a stoichiometry number (H/Pt) of 1.0.

The temperature-programmed reduction with H₂ (TPR) and the temperature-programmed desorption of ammonia (NH₃-TPD) were separately measured using a Belcat-M (Bel Japan) instrument equipped with TCD. Prior to TPR analysis, 30 mg of the fresh Pt/SA catalyst was pretreated at 250 °C under a flow of argon to remove any impurities or water adsorbed on the surfaces. TPR patterns were obtained with a ramping rate of 10 °C min⁻¹ under a flow of 10 vol% H₂ balanced with argon. The formed water from the TPR experiment was selectively removed using a molecular sieve trap, and the hydrogen consumption was measured by TCD. For the analysis of NH₃-TPD, 30 mg of the fresh Pt/SA catalyst was loaded in a quart tube reactor and the sample was pretreated at 250 °C for 2 h under a flow of He. After that, ammonia gas was adsorbed at 50 °C for 30 minutes and the physisorbed ammonia was removed by flushing it under a He flow at the same temperature for 1 h. Finally, the amount of the desorbed ammonia was measured by TCD under He flow in the temperature range of 100–600 °C at a ramping rate of 10 °C min⁻¹. The surface morphologies and the variation of crystallite sizes on the fresh and used Pt/SA(0) and fresh Pt/SA(0.1) after the APR reaction were characterized using transmission electron microscopy (TEM) with a TECNAI G2 instrument operated at an accelerating voltage of 200 kV.

Fourier transformed infrared (FT-IR) spectroscopy of the adsorbed pyridine (Py-IR) was used to characterize the types of acid sites, such as Bronsted and Lewis sites, on the fresh Pt/SA catalysts. Using an in situ IR cell equipped with a CaF₂ window and a shelf-supported thin pellet of the Pt/SA catalyst which was previously mixed and pelletized with an excess KBr powder at a mass ratio of 100 : 1, the Py-IR experiment was carried out. For a selective adsorption of pyridine probe molecules on the acid sites, the pyridine was previously flowed onto the pellet using an N₂ gas carrier at room temperature for 30 min. After the physisorbed pyridine was selectively evacuated for 45 min under vacuum conditions, the chemically adsorbed pyridine absorption band spectra were obtained. The background peaks were subtracted before the adsorption of pyridine onto the fresh Pt/SA catalyst.

3. Results and discussion

3.1. Catalytic activities of the APR and APH reaction

The catalytic activities for the APR reaction of EG on the Pt/SA catalysts at a fixed platinum metal content of 5 wt% are summarized in Table 1 in terms of the conversion of EG to gaseous products and product distributions, where SA(x) represents the solid-acid SiO₂–Al₂O₃ for SA with x digit for the molar ratio of Si/Al, where SA(0) is the Al₂O₃ support. The EG conversions to gaseous products were maximized on Pt/SA(0.1) with 43.4% and a maximum H₂ production rate of 1691 mL (g_cat h)⁻¹, which were the averaged values at a steady-state after the reaction occurred for 15 h on stream. A typical volcano pattern of EG conversions and product distributions was observed on the Pt/SA catalysts. The lowest EG conversion was observed on the Pt/SA(1.0) with a value of 18.4%, which seems to be responsible for the observed high CO selectivity of 5.2% and lower CO₂ selectivity due to the suppressed WGS activity of the intermediates derived from C–C bond cleavage of EG. Interestingly, the hydrocarbon selectivity decreased with an increase of the Si/Al molar ratios of the Pt/SA catalysts from 0.9% on the Pt/SA(0.1) to 2.5% on the Pt/SA(1.0), except for the Pt/SA(0). This observation seems to be attributed to a higher dehydrogenation activity for the C–C bond cleavage on the catalysts having a larger amount of Bronsted acid sites.^8,9 The most interesting observation is a significantly different hydrocarbon distribution, showing an increasing trend of CH₄ selectivity from 20.5 to 64.5% and vice versa for C₂H₆ from 79.5 to 35.3% with increasing Si/Al ratios. This observation can be attributed to the different acidity of the SiO₂–Al₂O₃ surfaces which change the intrinsic activities of C–C and C–O bond cleavages. Based on these observations, a higher conversion of EG to gaseous products is responsible for a higher H₂ production rate with a lower selectivity of CO and hydrocarbons with the help of the enhanced WGS activity, except for the Pt/SA(0). Even though the Pt/SA(0) showed a lower conversion of EG, around 27.5%, a higher selectivity to H₂ and a lower selectivity to CO were observed due to possible coke deposition on the too strong acidic sites with higher WGS activities on the largely exposed platinum metal surfaces. Compared to other Pt/SA catalysts which showed a stable EG conversion to gaseous products even after 15 h on stream, a significant decrease of EG conversion on the Pt/SA(0) during 15 h on stream can be possibly attributed to severe coke deposition, as shown in Fig. 1. The steady increase of the EG conversion to gaseous products for 15 h on stream can be due to a possible surface rearrangement of the supported platinum crystallites and an easy phase transformation of alumina species. It can be also attributed to a hydrogen gas residue used for reduction step of Pt/SA catalysts and for a pressure adjustment step at 5.0 MPa before the APR reaction. However, the comparisons of the catalytic activities at steady-state after 15 h on stream seem to be reasonable since similar catalytic activity changes were observed according to the Si/Al ratio on the Pt/SA catalysts at similar EG conversions, as summarized in ESI Table S1.† The selectivity to hydrocarbons on the Pt/SA catalysts seems to be mainly caused by the presence of the acidic sites with proper acid strength, which seems to be increased with an increase of the surface acid sites.^35–39 The selectivity of hydrocarbons such as the C₁ and C₂ paraffinic species seems to increase preferentially on the weak acid sites of SiO₂–Al₂O₃ by a possible decomposition of the intermediates derived from EG, which can be dehydration or retro-aldol condensation reactions. To verify the effects of acid sites during the consecutive reactions of EG conversion, the most active Pt/SA(0.1) catalyst was further examined at different space velocities such as 1.0 and 0.5 h⁻¹ and the results are summarized in Table 1. The conversions of EG to gaseous products at different space velocities are displayed in ESI Fig. S1.† With a decrease of space velocities from 2.0 to 0.5 h⁻¹ on the Pt/SA(0.1) at T = 250 °C and P = 4.5 MPa, the conversion of EG steadily increased from 43.4 to 78.6% without significant changes of product distributions except for hydrocarbons, which changed from 0.9 to 2.0% with a significant variation of C₁ from 36.2 to 78.1% and C₂ from 63.8 to 21.9%. These results strongly suggest that the intermediates such as aldehydes and alcohols formed from the APR reaction of EG can be further reacted to form hydrocarbons, especially for CH₄ hydrocarbon at a longer residence time. It also revealed that the Bronsted acidic sites with weak acid strengths and small metallic platinum crystallites (confirmed by NH₃-TPD and TPR in the next section) are responsible for the further transformation of the intermediates to light hydrocarbons, which seems to be similar to the effects of longer residence times of the intermediates formed by simultaneous C–O and C–C bond cleavage of EG. Interestingly, the hydrocarbon formation rate on the Pt/SA catalysts with a Si/Al ratio above 0.4 was stabilized after 10 h on stream, as shown in ESI Fig. S2,† which can be possibly attributed to the presence of a relatively increased number of Bronsted acid sites. As summarized in ESI Table S2,† a higher hydrocarbon formation was observed for the catalyst containing a large amount of Pt, Pt(7)/SA(0.1), with a value of 5.0% and a CH₄ selectivity of 88.6%. The large amount of the metallic platinum content seems to be responsible for the further reactions of intermediates to hydrocarbons. However, conversion of EG to gaseous products was not significantly increased on the Pt/SA(0.1) which has a large amount of platinum above 3 wt%, except for the Pt(1)/SA(0.1) due to the easy aggregation of the supported platinum crystallites at high platinum amounts on the SA(0.1) support with facile phase transformations of alumina.

Table 1 Aqueous phase reforming (APR)^a and aqueous phase hydrodeoxygenation (APH)^b reaction over the Pt/SA catalysts

Notation	Aqueous phase reforming (APR)							Aqueous phase hydrodeoxygenation (APH)
	Conversion to gases (%)/production rate of H2 (mL (gcat h)^−1)	Selectivity of products (mol%)				Selectivity of hydrocarbons (mol%)		Conversion to gases (%)/production rate of HC^c (mL (gcat h)^−1)	Selectivity of products (mol%)			Selectivity of hydrocarbons (mol%)
		H2	CO	CO2	HC^c	CH4	C2		CO	CO2	HC^c	CH4	C2	C3
a APR reaction was carried out in a fixed-bed tubular reactor at the reaction conditions of T = 250 °C, P = 4.5 MPa and weight hourly space velocity (WHSV) = 2.0 h^−1 with 0.3 g catalyst using 10 wt% ethylene glycol (EG) in an aqueous solution. The conversion and selectivity are the averaged values at steady-state after the reaction of 15 h on stream.b APH reaction was carried out in a fixed-bed tubular reactor at the reaction conditions of T = 260 °C, P = 5.0 MPa and weight hourly space velocity (WHSV) = 0.6 h^−1 with 0.5 g catalyst using 10 wt% ethylene glycol (EG) in an aqueous solution. The conversion and selectivity are the averaged values at steady-state after the reaction of 15 h on stream.c HC stands for the hydrocarbons formed during APR and APH reaction, which are mainly paraffinic hydrocarbons such as methane (CH4), ethane (C2) and propane (C3).d APR reaction was carried out in a fixed-bed tubular reactor at the reaction conditions of T = 250 °C, P = 4.5 MPa and weight hourly space velocity (WHSV) = 1.0 h^−1 with 0.3 g catalyst using 10 wt% ethylene glycol (EG) in an aqueous solution.e APR reaction was carried out in a fixed-bed tubular reactor at the reaction conditions of T = 250 °C, P = 4.5 MPa and weight hourly space velocity (WHSV) = 0.5 h^−1 with 0.3 g catalyst using 10 wt% ethylene glycol (EG) in an aqueous solution.
Pt/SA(0)	27.5/1076	73.1	0.4	25.3	1.2	20.5	79.5	42.7/1.4	1.1	96.2	2.7	46.0	50.9	3.1
Pt/SA(0.1)	43.4/1691	72.9	0.8	25.4	0.9	36.2	63.8	47.9/2.0	2.6	93.8	3.6	52.7	47.3	0.0
Pt/SA(0.4)	30.7/1178	72.7	1.8	23.9	1.6	57.7	42.3	46.8/4.0	2.0	90.9	7.1	53.1	45.9	1.0
Pt/SA(1.0)	18.4/657	71.3	5.2	21.0	2.5	64.5	35.5	29.0/3.1	14.7	76.3	9.0	56.1	42.8	1.1
Pt/SA(0.1)^d	64.6/1251	72.8	0.3	25.5	1.4	47.9	52.1	—	—	—	—	—	—	—
Pt/SA(0.1)^e	78.6/730	72.0	1.5	24.5	2.0	78.1	21.9	—	—	—	—	—	—	—

---

Fig. 1 Conversion of EG to gaseous products with time on stream on the Pt/SA catalysts for APR reaction at the reaction conditions of T = 250 °C, P = 4.5 MPa and weight hourly space velocity (WHSV) = 2.0 h⁻¹ with 0.3 g catalyst using 10 wt% EG in an aqueous solution.

To further verify the role of the acidity of the SA supports according to the Si/Al ratios to the hydrocarbon selectivity and EG conversion, the APH reaction was carried out on all Pt/SA catalysts and the results are summarized in Table 1. The variations of EG conversion to gaseous products were found to be similar with those of the APR reaction with the highest conversion of 47.9% on the Pt/SA(0.1) and the lowest on the Pt/SA(1.0) with 29.0% at steady-state. CO₂ selectivity decreased with an increase in the Si/Al ratio from 96.2% on the Pt/SA(0) to 76.3% on the Pt/SA(1.0), possibly due to the lower WGS activity of the intermediates formed from C–C bond cleavage by the dehydration of EG. However, CO selectivity was increased from 1.1% on the Pt/SA(0) to 14.7% on the Pt/SA(1.0) with the increasing Si/Al ratio of the Pt/SA catalysts. This observation suggests that the catalytic activities of the APR and WGS reaction are strongly influenced by the amount of active metal surface area as well as the acid sites during the aqueous-phase reaction. At a higher partial H₂ pressure during APH reaction, hydrogenation activity can be increased by enhancing the C–C bond cleavage of EG due to the additional H₂ formed by the WGS reaction. Therefore, the increased hydrocarbon selectivity from 2.7 to 9.0% and its production rate from 1.4 to 3.1 mL (g_cat h)⁻¹ with an increase in the Si/Al molar ratio during the APH reaction can support an enhanced hydrogenation activity, which was significantly increased with an increase in the Si/Al ratio on Pt/SA catalysts. As shown in Fig. 2, the catalytic activity for the APH reaction with time on stream seems to be more stable than the APR reaction without showing a significant deactivation after 10 h on stream. Interestingly, the variations of C₁–C₃ paraffinic hydrocarbon selectivity for the APH reaction were found to be similar with those of the APR reaction with the same trends, showing a higher selectivity of hydrocarbons on the Pt/SA catalysts with larger Bronsted acid sites. The increase of C₁ selectivity from 46.0 to 56.1% and the decrease of C₂ selectivity from 50.9 to 42.8% were observed with an increasing Si/Al ratio on the Pt/SA catalysts, which suggests that the different acidity of the SiO₂–Al₂O₃ surfaces and the metallic surface area of platinum crystallites can largely alter the C–C and C–O bond cleavages. The production rates of the hydrocarbons on the Pt/SA catalysts are displayed in ESI Fig. S3.† At a high Si/Al ratio above 0.4, the significant stability of the hydrocarbon formation rate with time on stream seems to be attributed to a lower coke deposition in addition to a large number of Bronsted acid sites. From the observed catalytic activities for APR as well as the APH reaction on the Pt/SA catalysts with different Pt loadings and space velocities, an optimal Pt/SA(0.1) was found which achieves a higher conversion of EG and hydrogen production rate. The product distribution was significantly affected by the surface properties such as the density and type of acid sites in addition to the surface area of the active platinum metal nanoparticles, which can also change the hydrocarbon distributions through C–C and C–O bond cleavages of intermediates formed from EG.

Fig. 2 Conversion of EG to gaseous products with time on stream on the Pt/SA catalysts for APH reaction at the reaction conditions of T = 260 °C, P = 5.0 MPa and weight hourly space velocity (WHSV) = 0.6 h⁻¹ with 0.5 g catalyst using 10 wt% EG in an aqueous solution.

3.2. Physicochemical surface properties of the Pt/SA catalysts

The physicochemical properties such as specific surface areas, pore volumes and average pore diameters of the Pt/SA catalysts and SA supports measured by N₂ adsorption–desorption methods are summarized in Table 2. The Pt/SA catalysts prepared by the coprecipitation method of the SA supports and the subsequent impregnation of platinum metal precursor showed surface areas above 300 m² g⁻¹, increasing from 302 m² g⁻¹ for Pt/SA(0) to 384 m² g⁻¹ for Pt/SA(1.0) in line with the increasing Si/Al molar ratio. The actual molar Si/Al ratio on the SA supports was found to be 0.12, 0.46 and 0.89 on the SA(0.1), SA(0.4) and SA(1.0), respectively, which were measured by XRF analysis. From the observed variations of surface areas of the SA supports from 370 to 496 m² g⁻¹, the decreased surface area of the Pt/SA catalysts was attributed to partial pore blockage of the pore mouths through platinum metal deposition. These aggregation phenomena are well-known segregation processes caused by deposition of the impregnated metal nanoparticles on the outer mesopores, which can be induced during drying or calcination steps.^18,41 The pore volumes of the Pt/SA catalysts increased with an increasing surface area in the range of 0.24–0.55 cm³ g⁻¹, and the average pore diameter was found to be larger on the Pt/SA catalysts with a size of 4.6–5.2 nm than for Pt/AS(0) due to possible intra-particular pore formation on the SiO₂–Al₂O₃ support. Pore size distribution on all the Pt/SA catalysts showed a unimodal pore size distribution, as shown in ESI Fig. S4.† Since the specific surface area and average pore diameter were not significantly altered on all fresh Pt/SA catalysts, these physical properties seem too insignificant to change the catalytic activity of the APR⁴¹ and APH reactions.

Table 2 Physicochemical properties of the Pt/SA catalysts before and after APR reaction

Notation	XRF (Si/Al molar ratio)	N2 sorption^a				H2 chemisorption^b			TPR^c (μmolH2 gcat^−1)			NH3-TPD^d (mmolNH3 gcat^−1)			Py-IR^e	Coke (%) from TGA
		Sg (m^2 g^−1)	Pv (cm^3 g^−1)	Ps (nm)	Sg(sup) (m^2 g^−1)	D (%)	Sg (m^2 g^−1)	Dp (nm)	TL	TM	TH	Tw	TS	Ttot	B/L
a Sg, Pv, and Ps measured by N2 adsorption–desorption analysis represent the surface area (m^2 g^−1), pore volume (cm^3 g^−1) and average pore diameter (nm) of the Pt/SA catalysts, respectively. The Sg(sup) represents the surface area (m^2 g^−1) of the SA support itself.b D, Sg and Dp measured by H2 chemisorption represent the dispersion (%), surface area (m^2 g^−1) and average crystallite size (nm) of metallic platinum on the Pt/SA catalysts, respectively.c The amount of consumed hydrogen from TPR was denoted as TL, TM and TH for the consumed amount of hydrogen in the temperature range of <200, 200–500, and >500 °C, respectively with a unit of μmolH2 gcat^−1.d The amount of acid sites measured by NH3-TPD with a unit of mmolNH3 gcat^−1 was denoted as Tw, TS and Ttot for weak, strong acid sites with total amount of acid sites in the temperature range of <300 and 300–500 °C, respectively.e The characteristic absorption peaks of pyridine molecules were assigned to Bronsted acidic site (B) at a wave number of 1550 cm^−1 and Lewis acidic site (L) at that of 1450 cm^−1, and the ratio of B/L on the Pt/SA catalysts was calculated using those relative integrated areas.
Pt/SA(0)	0	302	0.24	3.1	370	45.7	5.64	2.54	157	104	72	0.45	5.94	6.39	0	10.0
Pt/SA(0.1)	0.12	321	0.46	4.6	415	61.9	7.65	1.86	52	87	—	0.58	5.18	5.76	0.61	3.9
Pt/SA(0.4)	0.46	338	0.53	5.2	462	63.9	7.89	1.81	38	181	—	0.77	4.22	4.99	0.95	5.0
Pt/SA(1.0)	0.89	384	0.55	4.6	496	63.5	7.85	1.82	23	188	—	0.75	3.80	4.55	1.22	5.9

---

Therefore, the dispersion, surface area and crystallite size of the supported metallic platinum on the fresh Pt/SA catalysts were further measured by H₂ chemisorption, and the results summarized in Table 2. The dispersions of metallic platinum crystallites were found to be in the range of 45.7–63.9%. Pt/SA(0) showed the lowest value of 45.7% and other Pt/SA catalysts had values of above 60%. In addition, a larger crystallite size of 2.54 nm and a smaller surface area of metallic platinum of 5.64 m² g_Pt⁻¹ were also observed on the Pt/SA(0), however the values were found to be similar on the other Pt/SA catalysts in the ranges of 1.81–1.86 nm and 7.65–7.89 m² g_Pt⁻¹, respectively. Generally, a larger crystallite size and smaller surface area of metallic platinum crystallites are known to be responsible for a lower hydrogenolysis and WGS activity.^18,41 The variations of platinum crystallite sizes can significantly alter the activity as well as the deactivation rate, and some active metal particles are preferentially sintered during the reaction, which results in a change in the catalytic stability.^42–48 Therefore, the observed high selectivity of CH₄ on the Pt/SA(0.4) and Pt/SA(1.0) compared with the Pt/SA(0) can be attributed to a higher surface area of metallic platinum crystallites, possibly due to methanation activity after C–C and C–O bond cleavage of the intermediates.^{5–9,14–17} However, since there was no direct correlation of EG conversion with the surface area of metallic platinum crystallites, we believe that the observed difference in catalytic activity and product distribution seems to originate preferentially from the other surface properties of the Pt/SA catalysts, such as the density and type of the acid site, aggregation of active metals, coke deposition etc.

The crystalline structures and phases of the Pt/SA catalysts measured by XRD analyses are displayed in Fig. 3 and in ESI Fig. S5†, which show the used and fresh Pt/SA catalysts, respectively. Some crystalline phases of platinum crystallites and boehmite-phase alumina (γ-AlOOH) were clearly observed on the fresh Pt/SA(0), however, only platinum crystallite phases were observed on the fresh Pt/SA(0) and Pt/SA(0.1) due to the presence of the amorphous phases of the SiO₂–Al₂O₃ mixed metal oxides with highly dispersed platinum crystallites.^42,49 A larger characteristic peak intensity of platinum crystallites was observed on the fresh Pt/SA(0) and Pt/SA(0.1) compared with other Pt/SA catalysts, which strongly suggests larger platinum crystallite formation compared with other Pt/SA catalysts, and these results were also supported by H₂-chemisorption and TEM analysis. Interestingly, the fresh Pt/SA(0) showed characteristic γ-AlOOH and Pt(111) phases at the diffraction peak positions of 2θ = 40.0, 47.5 and 68.0, which revealed a phase transformation of γ-alumina to the boehmite phase, which interacted strongly with the supported platinum nanoparticles.⁴² As shown in Fig. 3, the aggregation of platinum crystallites after the APR reaction measured by XRD analysis was significant on the used Pt/SA catalysts with an average platinum crystallite size of 5.8, 2.2, 2.5, and 2.8 nm on the Pt/SA(0), Pt/SA(0.1), Pt/SA(0.4) and Pt/SA(1.0), respectively. In addition, the extent of transformation to the γ-AlOOH phases was much more significant on the Pt/SA(0) after the APR reaction, and a significant coke deposition was only observed on the Pt/SA(0), showing the characteristic peak of the coke precursor at around 2θ = 23°. The larger aggregation phenomena of platinum crystallites on the Pt/SA(0) seems to be mainly attributed to a smaller amount of surface acidic sites and the easy transformation to the boehmite phase through the weak interaction of platinum crystallites compared with other Pt/SA catalysts. In addition, the observed increased coke deposition on the Pt/SA(0) may also be due to the presence of the strong acid sites on the Al₂O₃ or boehmite phases, which can greatly alter the EG conversion and product distribution during the APR and APH reactions.^25–30 The extent of coke deposition on the used Pt/SA catalysts after an APR reaction of 20 h was further characterized by thermogravimetric analysis (TGA). In general, a weight loss (%) measured in a temperature range of 300–600 °C can be attributed to the oxidation of deposited graphitic carbons.⁴⁹ As shown in Table 2 and ESI Fig. S6,† a significant coke deposition of 10.0 wt% was observed on the Pt/SA(0). Relatively smaller amounts of coke deposition in the range of 3.9–5.5 wt% were observed on the other Pt/SA catalysts with the lowest value of 3.9 wt% on the Pt/SA(0.1), which was the most active catalyst for APR and APH reaction. The inactive coke precursors can generally be formed over strong acidic sites from unstable intermediates during the APR reaction.¹² However, a facile hydrogen spillover characteristic on the surfaces can suppress coke formation by hydrogenolysis.⁵⁰ The ratio of Bronsted/Lewis acid sites can also significantly alter the adsorption characteristics of the formed oxygenates, which can also promote coke depositions and deactivate them, especially for strong Bronsted acidic sites.^10,51

Fig. 3 XRD patterns of the used Pt/SA catalysts after APR reaction for 20 h.

The reduction of platinum crystallites on the Pt/SA catalysts was further verified by TPR experiments. The reduction behaviors of those catalysts are displayed in Fig. 4, and the consumed amount of hydrogen is summarized in Table 2. The broad reduction peaks on the Pt/SA catalysts were observed at a maximum temperature at around 350 °C with a shoulder peak at around 430 °C, which suggests a relatively homogeneous distribution of platinum nanoparticles, except for the Pt/SA(0). However, multiple reduction peaks were observed on the Pt/SA(0) which suggests the distribution of various platinum crystallite sizes on the Al₂O₃ support. A lower reduction temperature peak at around 180 °C on Pt/SA(0) seems to be responsible for the decomposition of the platinum metal precursor with an easy reduction character (assigned to T_L). In addition, a medium reduction temperature peak at around 350 °C (assigned to T_M) on the Pt/SA catalysts can be assigned to the reduction of well-dispersed platinum nanoparticles on the SA surfaces. The reduction behaviors of supported platinum crystallites are strongly affected by the surface properties of silica-alumina mixed oxides through changing the platinum-support interactions.^42,52 Therefore, a higher temperature reduction peak (assigned to T_H) shows the strongly interacting platinum nanoparticles solely on the Al₂O₃ support.^41,53 Interestingly, the hydrogen consumption assigned to T_M peak increased with an increase in the Si/Al ratio from 87 to 188 μmol g_cat⁻¹ without a significant variation in the T_L peak except for Pt/SA(0). It seems to be attributed to the enhanced acidic sites of the bare SA supports by forming strongly interacting and well-dispersed platinum nanoparticles on strong acidic SiO₂–Al₂O₃ supports.^42,52 These characteristic reduction behaviors of the supported platinum nanoparticles also significantly alter the WGS activity and C–C and C–O bond cleavages^{20–23,27–30} by the hydrogenation and hydrogenolysis reaction of intermediates formed by EG conversion, especially when the hydrocarbon selectivity is altered.

Fig. 4 H₂-TPR profiles of the fresh Pt/SA catalysts.

To further verify the effects of acid sites on the platinum metal dispersion and reduction behavior on the solid acid SA supports, NH₃-TPD analysis was carried out on fresh Pt/SA catalysts and the desorption patterns are displayed in Fig. 5 and the amount of desorbed NH₃ is summarized in Table 2. The desorption spectra of NH₃ showed two characteristic peaks at a maximum desorption temperature of around 200 °C (assigned to T_w for the desorbed amount of NH₃ below 300 °C) and 450 °C (assigned to T_S for the desorbed amount of NH₃ between 300 and 500 °C), which can be also assigned to weak and strong acidic sites, respectively.^54–56 With an increase of the Si/Al molar ratio on the Pt/SA catalysts, the desorbed amount of NH₃ from weak acidic sites (T_w) increased and decreased steadily for strong acidic sites (T_S) and total acidic sites (T_tot). In general, the Si–OH and Al³⁺ ions on the mixed metal oxides are known to be responsible for Brønsted (B) and Lewis acidic sites (L), respectively.^{35–40,54–56} As summarized in Table 2, the number of weak acid sites increased and finally approach a constant value from 0.45 mmol g_cat⁻¹ on the Pt/SA(0) to 0.75 mmol g_cat⁻¹ on the Pt/SA(1.0). However, the amount of strong acid sites decreased from 5.94 mmol g_cat⁻¹ on the Pt/SA(0) to 3.80 mmol g_cat⁻¹ on the Pt/SA(1.0), this is similar for total acid sites (T_tot) with an increasing Si/Al ratio on the Pt/SA catalysts. Therefore, the changes of the molar ratio of Si/Al on the mixed SiO₂–Al₂O₃ metal oxides largely altered the surface acidic properties and the concentrations of Brønsted and Lewis acid sites. NH₃-TPD was used to measure the total amount and strength of acid sites on the Pt/SA catalysts and the types of acid sites, such as Brønsted and Lewis acid sites, were further characterized by Fourier-transformed infrared analysis of adsorbed pyridine molecules (Py-IR). To further verify the B/L ratio on the Pt/SA catalysts, the characteristic absorption bands of pyridine molecules assigned to Bronsted acid sites (B) at a wavenumber of 1550 cm⁻¹, Lewis acid sites (L) at 1450 cm⁻¹, and combined Bronsted and Lewis acid sites (B + L)^54,55 at 1480 cm⁻¹ were integrated and compared of the each pyridine peak. The B/L ratio on the Pt/SA catalysts, summarized in Table 2, was increased with an increase in the Si/Al ratio, showing the ratio from 0 to 1.22. This observation strongly suggests that the amounts of Bronsted acid sites increase with the increase in Si/Al molar ratio on the Pt/SA catalysts. These different amounts and types of acid sites on the Pt/SA catalysts can largely alter the dispersion of platinum nanoparticles. This variation also changes the product distribution and EG conversion to gaseous products during the APR and APH reactions by significantly altering the activities of C–C and C–O bond cleavages.^{20–23,27–30}

Fig. 5 NH₃-TPD profiles on the fresh Pt/SA catalysts.

3.3. Roles of acidic sites and metal surface area on hydrocarbon distribution

The APR and APH reactions of biomass-derived intermediates have generally been known to proceed through complicated reaction pathways such as dehydrogenation, C–C and C–O cleavages and the WGS reaction on both the active metals and acid sites.^10,14,43,57 Therefore, a higher CO₂ selectivity and H₂ production rate can be obtained using hybrid catalysts with highly dispersed active metallic sites on the solid-acid sites of some mixed metal oxides. In addition, the solid-acid catalysts after properly modifying Bronsted and Lewis acid sites can also largely promote the dehydration activity of the biomass-derived intermediates.^17,25,38,58 From our previous work,⁴¹ conversion of EG to gaseous products for the APR reaction was mainly dependent on the dispersion of platinum crystallites on the mixed metal oxides of CeO₂–ZrO₂ and the highly dispersed and increased surface area of metallic platinum crystallites on ceria-zirconia support showed a higher EG conversion and hydrogen productivity.⁴¹ However, the catalytic activity on the solid-acid hybrid Pt/SA catalysts showed somewhat different catalytic behaviors by showing the significant effects of the acidic sites on the solid-acid SiO₂–Al₂O₃, which also modified the interactions of platinum nanoparticles with the acidic sites on the solid-acid SiO₂–Al₂O₃ support as well.

The observed higher selectivity to CO₂ and H₂ with higher EG conversion to gaseous products on Pt/SA(0.1) with a lower alkane formation with values of 43.3% for the APR reaction and 47.9% for the APH reaction were mainly attributed to the highly dispersed platinum nanoparticles with an abundance of total acidic sites, as summarized in Tables 1 and 2. However, the observed lower activity on the Pt/SA(0) was mainly attributed to significant coke deposition due to the abundant strong acid sites which show strong adsorption of intermediates. The lower alkane selectivity for the APR reaction on Pt/SA(0.1) seems to also be attributed to a higher activity for C–O and C–C cleavage on the metallic platinum crystallites as well as on the acidic sites of the SiO₂–Al₂O₃ surfaces. Interestingly, the hydrocarbon selectivity, especially for CH₄, was increased with an increase in the Si/Al ratio and vice versa for C₂ hydrocarbons. Therefore, to further verify the effects of acid sites on the hydrocarbon selectivity, a supplementary APH reaction was also carried out. The variations of EG conversion and product distribution were similar to the results of the APR reaction except for the higher hydrocarbon productivity. Hydrocarbon selectivity correlated well with the amount of weak Bronsted acid sites, and EG conversion correlated to the amount of strong acid sites, which seems to be reasonable due to selective C–C and C–O cleavage on the strong acidic sites followed by hydrogenation and WGS activity on the platinum metallic sites, as well as the weak acidic sites, except for Pt/SA(0) due to fast deactivation by severe coke deposition, as confirmed by NH₃-TPD, Py-IR, and TGA analyses. In addition, the observed increased CO₂ and H₂ selectivity on the Pt/SA(0.1) during the APR and APH reactions seems to be attributed to highly dispersed platinum nanoparticles with a proper interaction with SiO₂–Al₂O₃ surfaces, as confirmed by TPR and H₂ chemisorption. We believe that an increased selectivity of CH₄ on the Pt/SA(0.1) may be attributed to the highly dispersed platinum nanoparticles as well as the abundant weak Bronsted acid sites due to the enhanced WGS activity on the active metallic sites. The TEM images on the selected Pt/SA catalysts before and after the APR reaction are displayed in Fig. 6, and a decreased aggregation of the well-dispersed platinum nanoparticles below 5 nm was observed on the Pt/SA(0.1) (Fig. 6(B-1) and (B-2)). This revealed less aggregation of platinum crystallites on the Pt/SA(0.1) compared with the Pt/SA(0) and significant aggregation of platinum crystallites between 5 and 10 nm (Fig. 6(A-1) and (A-2)), which was well matched with the results of the XRD analysis. In general, the acid site density plays an important role in enhancing the heat of adsorption and the interaction between the biomass-derived intermediates on the acid sites, which can largely change the product distribution and conversion of the biomass-derived chemicals.^59–61 These phenomena were also confirmed by changing the residence time of the reactants and the platinum metal loading, as shown in Table 1 and ESI Table S2,† and a longer residence time and a high loading of platinum metals are responsible for a high formation rate of CH₄ as well as H₂ and CO₂ through enhanced hydrogenolysis and WGS activity. These activity changes can be also attributed to a further dehydration of the intermediates formed though dehydration and hydrogenation of EG reactant. Therefore, the hydrocarbon formation during the APR reaction seems to follow consecutive reaction pathways through dehydrogenation over metallic platinum nanoparticles followed by hydrogenolysis of the dehydrogenated intermediates through C–C and C–O cleavage, mainly on the acid sites, and further WGS reaction on the platinum metallic sites. Since the C₂ hydrocarbon can be formed by selective C–C and C–O cleavage followed by hydrogenation on the metallic sites, C₂ selectivity was decreased with an increase of Si/Al molar ratio due to a decreased amount of the strong acid sites on the Pt/SA catalysts with a larger Si/Al molar ratio.

Fig. 6 TEM images on the selected Pt/SA catalysts; (A-1) fresh Pt/SA(0), (A-2) used Pt/SA(0), (B-1) fresh Pt/SA(0.1) and (B-2) used Pt/SA(0.1).

According to the previous study of the APR reaction combined with the Fischer–Tropsch synthesis reaction in a batch reacting system,⁶² the acid component such as sulfuric acid strongly influences the CO production rate and subsequently changes the activity of the Fischer–Tropsch synthesis. The conversion of biomass-derived reactants to gaseous products has been known to be influenced by an initial step activity of dehydrogenation and dehydration, and the dehydrogenated intermediates can further proceed by the dehydrogenation reaction to form hydrocarbons.^57,58,63 In addition, the solid-acid metal oxides after a modification of Bronsted and Lewis acid sites can largely change the dehydration activity, as reported previously by many researchers.^38,64 It is known that an increase of Si content on the solid-acid Al₂O₃–SiO₂ catalysts can increase the B/L ratio by producing an active Si–OH species on the surface.⁶⁵ In addition, a decrease of surface acidity can be attributed to a lower metal–oxygen bond energy in the order of Al–O > Zr–O > Si–O, which seems to be related to a large formation of oxygen vacant sites in the mixed metal oxides, which are assigned to Lewis acid sites.^66,67 Therefore, a higher Si/Al ratio on the present Pt/SA catalysts was responsible for a higher ratio of B/L which also changes the hydrocarbon distribution significantly, especially for CH₄ hydrocarbons during the APR and APH reactions. The reducibility of the supported platinum nanoparticles was also influenced by the surface acidity due to the change in metal–support interactions, and the strong interaction of the platinum nanoparticles with acid sites can be changed by shifting a reduction peak to a higher temperature and forming strong metal–support interactions on the strong acid sites of the solid-acid surfaces.^43,68–74 Based on the present observation, the surface acidity of the SiO₂–Al₂O₃ mixed metal oxides largely altered the dispersion of platinum nanoparticles through their interaction with the solid acid surfaces, which can also significantly alter the EG conversion and hydrocarbon selectivity. In addition, the selective coke deposition on the strong acid sites on the Pt/SA(0) also altered the catalytic activity and product distribution, especially for hydrocarbon selectivity.

In summary, EG conversion and hydrocarbon distribution were strongly affected by the amount and type of acidic sites with the dispersion and reducibility of the supported platinum nanoparticles, which have a significant interaction with the solid-acid SiO₂–Al₂O₃ surfaces, as shown in Fig. 7. The stronger acid sites (or density of total acid sites) on the Pt/SA catalysts are preferentially responsible for a higher EG conversion through the APR and APH reactions, i.e., EG conversion was increased with the increase of the strong acid sites through a first step of hydrogenolysis and dehydration of EG. However, Pt/SA(0) with too strong acid sites is responsible for significant coke deposition which rapidly deactivates the Pt/SA(0). In addition, a larger amount of weak Bronsted acid sites (or a higher ratio of Bronsted/Lewis acid sites (B/L)), seems to interact with the supported metallic platinum nanoparticles by forming a smaller platinum crystallite size of below 5 nm, and is responsible for a higher CH₄ hydrocarbon selectivity through subsequent hydrogenation and WGS activity to form CO₂ simultaneously during the APR and APH reactions. Therefore, proper controls of the amount and type of acid sites by adjusting the metal-support interaction on the Pt/SA catalysts are important to the enhance catalytic activity for a transformation of biomass-derived intermediates during the APR and APH reactions, which can be simply obtained by changing the Si/Al molar ratio on the Pt/SA catalysts.

Fig. 7 Conversion of EG to gaseous products and hydrocarbon (HC) selectivity for APR and APH reaction as well as the density of total acid sites and the ratio of Bronsted/Lewis acid sites in terms of the Si/Al molar ratios on the Pt/SA catalysts.

4. Conclusions

The platinum supported SiO₂–Al₂O₃ catalysts were found to be an efficient and simple catalytic system for the chemical transformation of biomass-derived intermediates to useful hydrogen and paraffinic hydrocarbons. The larger metallic surface area of the platinum nanoparticles and the interaction with Bronsted acid sites on the SiO₂–Al₂O₃ support was responsible for the large hydrocarbon selectivity through hydrogenation and WGS reactions. In addition, the stronger Bronsted acid sites play an important role to enhance the EG conversion to gaseous hydrogen product during the APR and APH reactions. The APR and APH reactions proceeded through the consecutive reaction pathways, where the strong acid sites on the Pt/SA catalysts act as active sites for the hydrogenolysis of C–C and C–O bond cleavage. The weak Bronsted acid sites are related to the dispersion of the supported platinum nanoparticles, which are also responsible for a higher reducibility and WGS activity. Even though the presence of total acid sites and the reducibility of platinum nanoparticles correlated with the catalytic activity of APR and APH, the catalyst deactivation on the Pt/SA(0) during the APR reaction was strongly affected by heavy coke deposition on the stronger acid sites. With an increase of Si/Al on the SiO₂–Al₂O₃ mixed metal oxides, the EG conversion was decreased due to the decreased first step activity such as hydrogenolysis and dehydration of EG. This originates from the smaller number of stronger acid sites and Pt/SA(0.1) was found to be an optimal catalyst for the APR and APH reactions, which can be easily prepared by simply changing the Si/Al molar ratio of the Pt/SA catalysts.

Acknowledgements

This work was supported by the Korea Institute of Energy Tech-nology Evaluation and Planning (KETEP) with Project the number 20132010201750. The authors would like to acknowledge the financial support from the National Research Foundation of Korea (NRF) grant funded by the Korea government (NRF-2014R1A1A2A16055557 and NRF-2016M3D3A1A01913253). This work was supported by an institutional program Grant (2E26570-16-037) from Korean Institute of Science and Technology.

References

1. J. C. Serrano-Ruiz and J. A. Dumesic, Energy Environ. Sci., 2011, 4, 83–99.
2. A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J. Cairney, C. A. Eckert, W. J. Frederick, J. P. Hallett, D. J. Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and T. Tschaplinski, Science, 2006, 311, 484–489.
3. J. R. Rostrup-Nielsen, Science, 2005, 308, 1421–1422.
4. R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright and J. A. Dumesic, Appl. Catal., B, 2003, 43, 13–26.
5. R. R. Davda, J. W. Shabaker, G. W. Huber, R. D. Cortright and J. A. Dumesic, Appl. Catal., B, 2005, 56, 171–186.
6. J. W. Shabaker, G. W. Huber, R. R. Davda, R. D. Cortright and J. A. Dumesic, Catal. Lett., 2003, 88, 1–8.
7. J. W. Shabaker and J. A. Dumesic, Ind. Eng. Chem. Res., 2004, 43, 3105–3112.
8. C. H. Bartholomew, Appl. Catal., A, 2001, 212, 17–60.
9. C. H. Bartholomew, Catal. Rev., 1982, 24, 67–112.
10. D. J. M. de Vlieger, B. L. Mojet, L. Lefferts and K. Seshan, J. Catal., 2012, 292, 239–245.
11. X. Hu and G. Lu, Appl. Catal., B, 2009, 88, 376–385.
12. P. N. Kechagiopoulos, S. S. Voutetakis, A. A. Lemonidou and I. A. Vasalos, Energy Fuels, 2006, 20, 2155–2163.
13. D. Mei, V. Lebarbier Dagle, R. Xing, K. O. Albrecht and R. A. Dagle, ACS Catal., 2016, 6, 315–325.
14. D. M. Alonso, S. G. Wettstein and J. A. Dumesic, Chem. Soc. Rev., 2012, 41, 8075–8098.
15. X.-K. Gu, B. Liu and J. Greeley, ACS Catal., 2015, 5, 2623–2631.
16. M. Salciccioli and D. G. Vlachos, ACS Catal., 2011, 1, 1246–1256.
17. S. Kandoi, J. Greeley, D. Simonetti, J. Shabaker, J. A. Dumesic and M. Mavrikakis, J. Phys. Chem. C, 2011, 115, 961–971.
18. T. Nozawa, Y. Mizukoshi, A. Yoshida and S. Naito, Appl. Catal., B, 2014, 146, 221–226.
19. R. Rinaldi and F. Schuth, Energy Environ. Sci., 2009, 2, 610–626.
20. J. Lee, Y. Xu and G. W. Huber, Appl. Catal., B, 2013, 140–141, 98–107.
21. G. W. Huber, R. D. Cortright and J. A. Dumesic, Angew. Chem., Int. Ed., 2004, 43, 1549–1551.
22. P. Ferrin, D. Simonetti, S. Kandoi, E. Kunkes, J. A. Dumesic, J. K. Nørskov and M. Mavrikakis, J. Am. Chem. Soc., 2009, 131, 5809–5815.
23. B. M. Moreno, N. Li, J. Lee, G. W. Huber and M. T. Klein, RSC Adv., 2013, 3, 23769–23784.
24. N. Li and G. W. Huber, J. Catal., 2010, 270, 48–59.
25. M. Salciccioli, W. Yu, M. A. Barteau, J. G. Chen and D. G. Vlachos, J. Am. Chem. Soc., 2011, 133, 7996–8004.
26. K. L. Deutsch, D. G. Lahr and B. H. Shanks, Green Chem., 2012, 14, 1635–1642.
27. J. Greeley and M. Mavrikakis, J. Am. Chem. Soc., 2004, 126, 3910–3919.
28. J. Greeley and M. Mavrikakis, J. Am. Chem. Soc., 2002, 124, 7193–7201.
29. W. Luo and A. Asthagiri, J. Phys. Chem. C, 2014, 118, 15274–15285.
30. S. Sa, H. Silva, L. Brandão, J. M. Sousa and A. Mendes, Appl. Catal., B, 2010, 99, 43–57.
31. J. Lee, Y. T. Kim and G. W. Huber, Green Chem., 2014, 16, 708–718.
32. J. Yang, C. L. Williams, A. Ramasubramaniam and P. J. Dauenhauer, Green Chem., 2014, 16, 675–682.
33. C. Zhang, J. Xing, L. Song, H. Xin, S. Lin, L. Xing and X. Li, Catal. Today, 2014, 234, 145–152.
34. D. B. Lao, A. C. E. Owens, D. M. Heinekey and K. I. Goldberg, ACS Catal., 2013, 3, 2391–2396.
35. Y. Nakagawa and K. Tomishige, Catal. Sci. Technol., 2011, 1, 179–190.
36. Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and K. Tomishige, Appl. Catal., B, 2011, 105, 117–127.
37. Y. T. Kim, K.-D. Jung and E. D. Park, Appl. Catal., B, 2011, 107, 177–187.
38. R. Weingarten, G. A. Tompsett, W. C. Conner Jr and G. W. Huber, J. Catal., 2011, 279, 174–182.
39. L. Yang, G. Tsilomelekis, S. Caratzoulas and D. G. Vlachos, ChemSusChem, 2015, 8, 1334–1341.
40. K. Okada, T. Tomita, Y. Kameshima, A. Yasumori and K. J. D. MacKenzie, J. Mater. Chem., 1999, 9, 1307–1312.
41. S. Jeon, H. W. Ham, Y. W. Suh and J. W. Bae, RSC Adv., 2015, 5, 54806–54815.
42. A. Ciftci, B. Peng, A. Jentys, J. A. Lercher and E. J. M. Hensen, Appl. Catal., A, 2012, 431–432, 113–119.
43. A. Ciftci, D. A. J. M. Ligthart and E. J. M. Hensen, Appl. Catal., B, 2015, 174–175, 126–135.
44. K. Lehnert and P. Claus, Catal. Commun., 2008, 9, 2543–2546.
45. J. W. Bae, I. G. Kim, J. S. Lee, K. H. Lee and E. J. Jang, Appl. Catal., A, 2003, 240, 129–142.
46. M. Besson and P. Gallezot, Catal. Today, 2003, 81, 547–559.
47. A. Ciftci, D. A. J. M. Ligthart, A. O. Sen, A. J. F. van Hoof, H. Friedrich and E. J. M. Hensen, J. Catal., 2014, 311, 88–101.
48. A. Ciftci, D. A. J. M. Ligthart and E. J. M. Hensen, Green Chem., 2014, 16, 853–863.
49. B. Sanchez, M. S. Gross, B. D. Costa and C. A. Querini, Appl. Catal., A, 2009, 364, 35–41.
50. A. Iriondo, J. F. Cambra, V. L. Barrio, M. B. Guemez, P. L. Arias, M. C. Sanchez-Sanchez, R. M. Navarro and J. L. G. Fierro, Appl. Catal., B, 2011, 106, 83–93.
51. Z. Zhong, H. Ang, C. Choong, L. Chen, L. Huang and J. Lin, Phys. Chem. Chem. Phys., 2009, 11, 872–880.
52. M. Niwa, K. Awano and Y. Murakami, Appl. Catal., 1983, 7, 317–325.
53. U. Nylén, J. F. Delgado, S. Järås and M. Boutonnet, Appl. Catal., A, 2004, 262, 189–200.
54. J. H. Park, C. Pang, C. H. Chung and J. W. Bae, J. Nanosci. Nanotechnol., 2016, 16(5), 4626–4630.
55. S. Y. Park, C. H. Shin and J. W. Bae, Catal. Commun., 2016, 75, 28–31.
56. J. W. Jung, Y. J. Lee, S. H. Um, P. J. Yoo, D. H. Lee, K. W. Jun and J. W. Bae, Appl. Catal., B, 2012, 126, 1–8.
57. A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A. A. Lemonidou and J. A. Lercher, J. Catal., 2010, 269, 411–420.
58. G. S. Foo, D. Wei, D. S. Sholl and C. Sievers, ACS Catal., 2014, 4, 3180–3192.
59. G. Yaluris, R. B. Larson, J. M. Kobe, M. R. González, K. B. Fogash and J. A. Dumesic, J. Catal., 1996, 158, 336–342.
60. C. D. Baertsch, K. T. Komala, Y.-H. Chua and E. Iglesia, J. Catal., 2002, 205, 44–57.
61. M. Watanabe, Y. Aizawa, T. Iida, R. Nishimura and H. Inomata, Appl. Catal., A, 2005, 295, 150–156.
62. V. V. Ordomsky and A. Y. Khodakov, Green Chem., 2014, 16, 2128–2131.
63. Y. T. Kim, J. A. Dumesic and G. W. Huber, J. Catal., 2013, 304, 72–85.
64. K. Nakajima, Y. Baba, R. Noma, M. Kitano, J. N. Kondo, S. Hayashi and M. Hara, J. Am. Chem. Soc., 2011, 133, 4224–4227.
65. S. Rajagopal, J. A. Marzari and R. Miranda, J. Catal., 1995, 151, 192–203.
66. L. Wang, H. Wan, S. Jin, X. Chen, C. Li and C. Liang, Catal. Sci. Technol., 2015, 5, 465–474.
67. D. G. Rethwisch and J. A. Dumesic, Langmuir, 1986, 2, 73–79.
68. C. W. Chou, S. J. Chu, H. J. Chiang, C. Y. Huang, C. Lee, S. R. Sheen, T. P. Perng and C. Yeh, J. Phys. Chem. B, 2001, 105, 9113–9117.
69. P. Panagiotopoulou, A. Christodoulakis, D. I. Kondarides and S. Boghosian, J. Catal., 2006, 240, 114–125.
70. A. S. Ivanova, E. M. Slavinskaya, R. V Gulyaev, V. I. Zaikovskii, O. A. Stonkus, I. G. Danilova, L. M. Plyasova, I. A. Polukhina and A. I. Boronin, Appl. Catal., B, 2010, 97, 57–71.
71. J. N. Kuhn, W. Huang, C. K. Tsung, Y. Zhang and G. A. Somorjai, J. Am. Chem. Soc., 2008, 130, 14026–14027.
72. Q. Liu, U. A. Joshi, K. Uber and J. R. Regalbuto, Phys. Chem. Chem. Phys., 2014, 16, 26431–26435.
73. C. Serre, F. Garin, G. Belot and G. Maire, J. Catal., 1993, 141, 1–8.
74. G. Jacobs, U. M. Graham, E. Chenu, P. M. Patterson, A. Dozier and B. H. Davis, J. Catal., 2005, 229, 499–512.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra09522d

---