- Email: [email protected]
Highly durable Pt-supported niobia–silica aerogel catalysts in the aqueous-phase hydrodeoxygenation of 1-propanol
Catalysis Communications 29 (2012) 40–47
Contents lists available at SciVerse ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Short Communication
Highly durable Pt-supported niobia–silica aerogel catalysts in the aqueous-phase hydrodeoxygenation of 1-propanol Jihye Ryu a, b, Sung Min Kim c, Jae-Wook Choi a, Jeong-Myeong Ha a, Dong June Ahn b, Dong Jin Suh a,⁎, Young-Woong Suh c,⁎⁎ a b c
Clean Energy Research Center, Korea Institute of Science and Technology, Seoul 136‐791, Republic of Korea Department of Chemical and Biological Engineering, Korea University, Seoul 136‐701, Republic of Korea Department of Chemical Engineerling, Hanyang University, Seoul 133‐791, Republic of Korea
a r t i c l e
i n f o
Article history: Received 11 August 2012 Received in revised form 15 September 2012 Accepted 18 September 2012 Available online 24 September 2012 Keywords: 1-Propanol Aqueous-phase processing Hydrodeoxygenation Aerogel Pt/Nb2O5–SiO2
a b s t r a c t The aqueous-phase hydrodeoxygenation (APHDO) of 1-propanol at 230 °C and 35 bar was studied over 1 wt.% Pt catalysts supported on several metal oxides. Pt catalysts supported on amorphous silica, alumina and niobia aerogels, and crystalline niobic acid calcined at 500 °C showed low activities or deactivation. Under the APHDO condition, these supports experienced a structure transformation to crystalline quartz, boehmite and niobia TT phase. Pt/Nb2O5–Al2O3 aerogel also suffered from the same crystallization behavior. In contrast, Pt/Nb2O5–SiO2 aerogels with different Nb/(Nb + Si) ratios maintained a good catalytic performance for prolonged reaction periods. Through characterizations of spent catalysts, Nb2O5–SiO2 aerogels were found to retain X-ray amorphous and porous structure. Also, their acid site densities were negligibly changed during the reaction. The catalysts with the Nb / (Nb + Si) ratio of 0.500 and 0.575 reached the molar propane/ethane ratio of about 1.0, indicating that Pt/Nb2O5–SiO2 aerogel catalysts are highly active in the presence of water. © 2012 Elsevier B.V. All rights reserved.
1. Introduction An efficient conversion of biomass-derived molecules into fuels and chemicals has been studied in recent years [1–6]. Particularly, much attention has been paid to aqueous-phase processing of C3 alcohols to produce H2 and alkanes [7–9]. Huber and coworkers investigated the aqueous-phase hydrodeoxygenation (APHDO) of sorbitol and sugar solutions towards understanding the reaction chemistry [10–12]. In the APHDO process, a series of dehydration and hydrogenation steps takes place over bifunctional catalysts in which an acid is responsible for the dehydration of the reactant and a metal for subsequent hydrogenation. The carbon loss of reactant can be minimized in this case. However, if metal contribution is more dominant than acid, lighter alkanes are produced through dehydrogenation and decarbonylation reactions on metal surfaces. Therefore, the adjustment in the concentration and strength of acid sites is of importance to selectively produce heavier alkanes over bifunctional catalysts. The similar assessment was done on the aqueous-phase reforming and hydrodeoxygenation of C3 alcohols [7–10,13,14]. In this work, 1-propanol (1-PrOH) was used as a model reactant for oxygenated biomass-derived compounds, due to the difficulty to ⁎ Corresponding author. Tel.: +82 2 958 5192; fax: +82 2 958 5209. ⁎⁎ Corresponding author. Tel.: +82 2 2220 2329; fax: +82 2 2298 4101. E-mail addresses: [email protected] (D.J. Suh), [email protected] (Y.-W. Suh). 1566-7367/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.catcom.2012.09.022
dehydrate 1-PrOH with a terminal OH among C3 alcohols along with the simple product distribution. Major products in the APHDO of 1-PrOH are CO, CO2, CH4, C2H6 and C3H8 in gas phase. When the dehydration and hydrogenation of 1-propanol occur in sequence (C\O bond cleavage), propane is the main product. In contrast, ethane and CO are formed by the dehydrogenation and decarbonylation (C\C bond cleavage). CO further undergoes the water–gas shift into CO2 and methanation into CH4. Therefore, if it is desired to lower the oxygen content in products and minimize carbon loss by suppressing the cleavage of C\C bond, an acid part of bifunctional catalyst responsible for the dehydration of 1-PrOH is then crucial, as in the case of sorbitol [10]. Niobium oxides have been found to be active in acid-catalyzed reactions, even in the presence of water [15–18]. It was reported that amorphous niobium oxide is composed of distorted NbO6 octahedra and NbO4 tetrahedra [19,20]; however, it suffers from poor hydrothermal stability, resulting in both a poor activity and low surface area after use [21,22]. This predicts that the texture, structure or acidity of niobia may be changed under aqueous-phase reaction conditions due to high temperature and H2 pressure in water. The prediction is supported by the recent report of Hensen and coworkers that boehmite formation is observed for spent Pt/γ-Al2O3 and Pt/SiO2–Al2O3 catalysts in the aqueous-phase glycerol reforming [7]. Therefore, the catalyst stability is an important issue in the APHDO reaction, besides a superior catalytic performance.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
Fig. 1. Conversion of 1-PrOH (A) and carbon selectivity of gas-phase products obtained over 1 wt.% Pt-supported catalysts (B–G):
41
C3H8,
C2H6,
CO2,
CO, and
CH4.
42
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
In this work, we focused on examining the activity, product selectivity and stability of Pt catalysts supported on commercial niobic acid (Nb2O5·xH2O), and SiO2, Al2O3 and Nb2O5 aerogels. Herein, aerogel-type metal oxides were chosen because it is easy to monitor the change of catalyst properties during the APHDO reaction, owing to an amorphous character, highly porous structure and large surface area. Furthermore, Nb2O5–Al2O3 aerogel as well as Nb2O5–SiO2 aerogels with different Nb amounts were prepared and used for the APHDO reaction of 1-PrOH, since niobia-containing mixed oxides are more resistant to heat and pressure [21,22]. 2. Experimental
X-ray diffraction (XRD) analysis was conducted with a Rigaku D/ Max-2500 diffractometer using Cu Kα (λ = 0.1541 nm) as a radiation source. XRD patterns were collected with a step size of 0.02° in the 2θ range of 5°–90°. N2 adsorption–desorption analysis was performed using a Micromeritics ASAP 2020 and ASAP 2010. N2 BET (Brunauer– Emmett–Teller) specific surface areas of fresh and spent catalysts were measured from the adsorption isotherm branch and pore size distributions were calculated by BJH (Barrett–Joyner–Halenda) method with desorption isotherm branches. Before measurement, samples were degassed at 120 °C under vacuum overnight and subsequently purged with He. H2 chemisorption was performed by a BEL-CAT instrument. About 50 mg of catalyst was successively pretreated in flowing 10% O2/He
2.1. Preparation of catalysts
2.2. Catalyst characterization NH3 temperature programmed desorption (NH3-TPD) experiments were carried out using a BEL-CAT instrument (BEL Japan, Inc.) equipped with a thermal conductivity detector (TCD). In a typical experiment, about 30 mg of catalyst was loaded and reduced in H2 at 400 °C for 30 min. It was then purged in He with ramping the temperature to 500 °C and cooled to 100 °C for NH3 adsorption. After exposure of the reduced catalyst to a flow of 5% NH3/He (50 mL/min) for 30 min, NH3 physically adsorbed was removed by purging pure He for 1 h at 100 °C. NH3-TPD profiles were obtained by ramping the temperature at a heating rate of 5 °C/min under pure He (30 mL/min) to 900 °C. Used catalysts were also subjected to TPD analysis, for which drying under vacuum overnight and pretreatment in He at 500 °C were conducted in advance.
A 1 wt.% Pt/Nb2O5-Al2O3 1 wt.% Pt/Al2O3 1 wt.% Pt/Nb2O5-SiO2
Intensity (a.u.)
1 wt.% Pt/SiO2 1 wt.% Pt/Nb2O5
1 wt.% Pt/Nb2O5.xH2O 20
40
60
80
2θ (deg.)
B 1 wt.% Pt/Nb2O5-Al2O3 1 wt.% Pt/Al2O3 1 wt.% Pt/Nb2O5-SiO2
Intensity (a.u.)
Nb2O5 gel was synthesized by the hydrolysis and condensation of niobium chloride (NbCl5, 99%) in ethanol (ethanol/Nb= 47 mol/mol). Nitric acid (HNO3/Nb =0.2 mol/mol, 70%) and deionized water (H2O/ Nb =5 mol/mol) were added for hydrolysis, and propylene oxide (propylene oxide/Nb= 6.3 mol/mol, 99%) as a gelation promoter was then added with stirring. During gel formation, the viscosity increased with heat evolution. Al2O3 gel was also prepared using AlCl3·6H2O (min. 97.0%), HNO3, water, propylene oxide and ethanol in a molar ratio of 1:0.1:3:3:35. In the case of SiO2 gel, the sol–gel process of tetraethyl orthosilicate (TEOS) was catalyzed by NH4OH (min. 28.0%) and NH4F (min. 97.0%), where a molar ratio of TEOS, methanol, H2O, NH4OH and NH4F was 1:12.5:4:0.005:0.001. Nb2O5–SiO2 gels were prepared by simply adding TEOS into the niobium solution, prior to the hydrolysis. The typical molar ratio of (Nb+Si), HNO3, water, propylene oxide and ethanol was 2:0.2:6:6:45. For the preparation of niobia–silica with different Nb/(Nb+Si) ratios, the amount of propylene oxide added for gel formation was proportional to that of niobium while nitric acid and water were not changed. For Nb2O5–Al2O3 gel, aluminum chloride hexahydrate (Al/Nb=1 mol/mol) was dissolved in the mixture of nitric acid (HNO3/Nb=0.3 mol/mol) and ethanol (ethanol/Nb=45 mol/mol). After the water addition (H2O/ Nb=8 mol/mol) to completely dissolve the aluminum precursor, niobium chloride was added. The mixture was transformed into gel by propylene oxide (propylene oxide/Nb=8 mol/mol). The resulting wet gels were kept at room temperature for 3 days except Nb2O5 gel (1 day). For supercritical drying, ethanol in wet gel pores was exchanged with liquid CO2 at a pressure of 2000 psi and dried at 70 °C. To remove the remaining organic materials before Pt loading, single metal oxide aerogels and niobic acid (Soekawa chemical) were calcined at 500 °C for 4 h, whereas mixed oxide aerogels were treated at 500 °C for 8 h. Catalysts containing 1 wt.% Pt were prepared by incipient wetness impregnation with an aqueous solution of H2PtCl6·6H2O (≥ 37.5% Pt basis), dried in an oven at 105 °C overnight and calcined at 400 °C for 2 h in flowing air.
1 wt.% Pt/SiO2
1 wt.% Pt/Nb2O5
1 wt.% Pt/Nb2O5.xH2O 20
40
60
80
2θ (deg.) Fig. 2. XRD patterns of reduced (A) and spent catalysts after the reaction for 10 or 24 h (B); (a) Pt/Nb2O5·xH2O, (b) Pt/Nb2O5, (c) Pt/SiO2, (d) Pt/Nb2O5–SiO2, (e) Pt/Al2O3, (f) Pt/Nb2O5–Al2O3; Pt, TT–Nb2O5, quartz, and boehmite.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
43
Table 1 Textural and acid properties of fresh and spent Pt-supported catalysts. Catalyst
Pt/Nb2O5·xH2O Pt/Nb2O5
Pt/SiO2 Pt/Al2O3 Pt/Nb2O5–Al2O3
Pt/Nb2O5–SiO2 0.500 Pt/Nb2O5–SiO2 0.575 Pt/Nb2O5–SiO2 0.625 Pt/Nb2O5–SiO2 0.675 a
TOS
Fresh 10 h Fresh 10 h 24 h Fresh 24 h Fresh 24 h Fresh 10 h 24 h Fresh 10 h 24 h Fresh 24 h Fresh 24 h Fresh 24 h
Surface area (m2/g)
Pore volumea (cm3/g)
17 17 198 42 32 751 4 313 66 213 117 68 365 263 210 326 252 345 259 299 259
0.067 0.074 0.828 0.167 0.229 0.829 0.022 0.288 0.458 1.236 0.992 0.317 1.145 1.010 0.822 0.985 0.872 1.087 0.856 0.934 0.799
NH3-TPD (μmol/m2) T b 450 °C
T > 450 °C
Total
1.454 – 2.762 2.717 3.272 0.086 – 2.202 3.425 2.287 2.064 3.048 1.853 1.851 2.123 2.128 2.234 2.091 2.079 2.177 1.979
63.171 47.954 0.009 0.697 8.254 1.217 22.842 4.699 14.262 1.015 2.334 6.069 0.339 0.923 1.338 0.376 0.848 0.293 0.845 0.280 0.907
64.625 47.954 2.771 3.414 11.526 1.302 22.842 6.900 17.687 3.303 4.399 9.117 2.192 2.774 3.461 2.505 3.082 2.384 2.924 2.458 2.886
BJH desorption cumulative pore volume.
for 15 min and 10% H2/Ar at 400 °C for 30 min, and then cooled to 55 °C in Ar. H2 adsorption took place at 55 °C with 10% H2/Ar. 2.3. Catalytic activity test Activity tests were carried out in an upflow stainless steel reactor to which a gas–liquid separator (internal volume: 65 mL) was connected at an upper position (Fig. S1). The catalysts were packed with glass beads between glass wool and gasket. Pt-supported catalysts were reduced at 400 °C for 2 h under flowing H2 (100 mL/min). After the reactor was cooled to room temperature, the system was pressurized with H2 (200 mL/min) to 35 bar, and heated to 230 °C again with H2 flow of 13 mL/min. The total pressure in the reactor and separator was controlled by a back-pressure regulator (BPR) attached to the exit line of the separator. An aqueous solution of 10 wt.% 1-PrOH was then fed at a flow rate of 0.05 mL/min. The time was counted when liquid products were visible in the bottom of the separator (ca. 2 h). During the catalytic run, gas-phase products released through the BPR were analyzed using an online GC equipped with FID installed with Hayesep D column. Because the SUS line connected between the reactor and the separator was air-cooled with a fan, the temperature of liquid sample in the separator during the reaction was about 30 °C, preventing volatile liquid molecules from passing through the separator and BPR. The liquid sample collected in the separator was drained every 2 h and analyzed using a HPLC equipped with RID and Bio-Rad HPX-87H column. Due to the pressure difference between the reaction pressures (35 bar) and atmospheric pressure, much care was taken to trap most of liquid molecules using a loosely tight glass vial. The conversion of 1-PrOH was determined from the difference of 1-PrOH concentration by HPLC analysis. Other liquid products, such as propionic acid, propanal, 2-PrOH and propyl propanoate, were detected in negligible amounts. Thus, the carbon selectivity (C%) of only gas-phase products was calculated based on the amount detected by GC analysis, as follows: Carbon selectivityðC% Þ ¼
ðmole of a specific gas productÞ ðnumber of carbon atom in a moleculeÞ ðmoles of CO; CO2 and CH4 Þ þ 2 ðmole of C2 H6 Þ þ 3 ðmole of C3 H8 Þ 100:
In the beginning of the reaction, C3H8, C2H6, CO, CO2 and CH4 were detected, but the formation of CO and CH4 sharply diminished over the time-on-stream, probably due to the fast water–gas shift reaction into CO2 and H2 as well as the decline of methanation reaction. Note that there is a certain extent of sample loss when the liquid sample was drained, although several precautions were undertaken in order to achieve the perfect material balance. When the absolute amount of 1-propanol consumed measured by HPLC analysis was compared to that of gaseous products detected by a GC, the observed difference was in the range of ±5%. Thus, the same experiment was repeated at least 2–3 times for reproducibility. 3. Results and discussion The APHDO reaction at 230 °C and 35 bar was conducted using Pt catalysts supported on different metal oxides, such as niobic acid (Nb2O5·xH2O), and SiO2, Al2O3, Nb2O5, Nb2O5–SiO2 and Nb2O5– Al2O3 aerogels. Fig. 1 shows the conversion of 1-PrOH and carbon selectivity of gas-phase products. In the blank test using glass beads
Table 2 The conversion of 1-propanol and gas-phase product selectivities using Pt/Nb2O5–SiO2 aerogel catalysts. Nb/(Nb + Si)
TOS
Conversion (%)
Carbon selectivity (C%) CO
CO2
CH4
C2H6
C3H8
0.500
2h 10 h 24 h 2h 10 h 24 h 2h 10 h 24 h 2h 10 h 24 h
44.2 43.2 45.8 47.9 46.2 49.3 65.0 54.7 53.8 62.7 54.2 50.4
0.18 0.13 0.09 0.19 0.14 0.12 0.14 0.10 0.10 0.16 0.12 0.09
16.95 16.54 15.00 16.78 17.02 15.36 16.30 16.61 15.71 17.10 17.45 17.16
0.14 0.09 0.06 0.20 0.09 0.08 0.20 0.09 0.09 0.21 0.09 0.09
38.75 37.98 37.08 36.60 37.62 33.65 37.34 38.27 35.62 37.60 38.74 38.24
43.98 45.26 50.77 46.22 45.13 50.79 46.02 44.94 48.49 44.94 43.59 44.42
0.575
0.625
0.675
C3/C2a (mol/mol) 0.757 0.794 0.913 0.842 0.800 1.006 0.822 0.783 0.907 0.797 0.750 0.774
a Calculated from the equation: C3/C2 = [2 × (carbon selectivity of propane)] / [3 × (carbon selectivity of ethane)].
44
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
under this condition, the 1-PrOH conversion calculated by HPLC analysis was 3%–4%, where only trace amounts of liquid products were formed and gaseous products were not detected at all. Among the six catalysts, the lowest conversion was obtained over Pt/SiO2 aerogel (ca. 3.6% at 24 h) along with more selective formation of ethane than propane. This is due to relatively neutral SiO2 surface, even though the surface silanol species can act as Brönsted acid. Pt/Al2O3 aerogel also showed the conversion of less than 20% and the carbon selectivity to propane lower than that to ethane, although the cleavage of C\O bond is expected to be more favorable due to acid sites of Al2O3. It is speculated that the acidity of Al2O3 are not strong enough to catalyze the dehydration of 1-PrOH due to the competitive adsorption of H2O on Al2O3 [14]. From observations that Pt/Al2O3 showed high production of H2 and CO2 during aqueous-phase reforming of
oxygenated compounds [7,8,23], it can be inferred that the cleavage of C\C bond is favorable on Pt/Al2O3. In the case of Pt/Nb2O5 aerogel, the conversion of 92% and propane selectivity of 52 C% were initially observed; however, both declined to 49% and 42 C% in 10 h. This is explained from the activity and characteristics of Pt/Nb2O5·xH2O of which the phase was already transformed from amorphous to crystalline TT phase (TT–Nb2O5) during calcination at 500 °C (Fig. 2A). Since crystalline niobia has less acid sites and lower specific surface areas than amorphous one [16,19,20], Pt/Nb2O5·xH2O showed the conversion of 10.5% and the ethane selectivity of 54.6 C% at 10 h. By comparing fresh and spent Pt/Nb2O5 aerogel catalysts, it was noticed that Nb2O5 aerogel was crystallized into TT–Nb2O5 during the reaction at 230 °C for 10 h (Fig. 2) and its BET specific surface area was significantly reduced from 198 to 42 m2/g (Table 1). NH3-TPD
6
6
1 wt.% Pt/Nb2O5.xH2O
1 wt.% Pt/Nb2O5 5
Pore volume (cm3/g)
Pore volume (cm3/g)
5 4 3 2
4 3 2 1
1
0
0 1
10
100
1
Pore diameter (nm)
1 wt.% Pt/SiO2
1 wt.% Pt/Nb2O5-SiO2 5
Pore volume (cm3/g)
5
Pore volume (cm3/g)
100
6
6
4 3 2
4 3 2 1
1
0
0 1
10
1
100
10
100
Pore diameter (nm)
Pore diameter (nm) 6
6
1 wt.% Pt/Al2O3
1 wt.% Pt/Nb2O5-Al2O3 5
Pore volume (cm3/g)
5
Pore volume (cm3/g)
10
Pore diameter (nm)
4 3 2
4 3 2 1
1
0
0 1
10
100
1
Fig. 3. Pore size distribution (PSD) curves of Pd-supported catalysts;
10
100
Pore diameter (nm)
Pore diameter (nm) after calcination,
after the reaction for 10 h, and
after the reaction for 24 h.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
peak detected below 450 °C was also very little in the spent Pt/Nb2O5 (Fig. S2: NH3-TPD profiles). Therefore, the phase modification of Pt/ Nb2O5 during the reaction leads to the decrease of both the 1-PrOH conversion and propane selectivity. In terms of the structure change under the hydrothermal reaction condition, Pt/Al2O3 was found to be transformed into crystalline boehmite (AlOOH) with the lower specific surface area of 66 m 2/g (fresh catalyst: 313 m 2/g) and larger pore diameter. This is in accordance with the report of Hensen and coworkers [7], which is related to the hydroxylation of alumina domains [14,24]. However, it did
45
not show a drop of catalytic activity at all. This may be accounted for by the fact that acid sites of the spent Pt/Al2O3 were still intact (Table 1). The above behavior was similarly observed in Pt/SiO2; after the reaction for 24 h, the specific surface area decreased from 751 to 4 m 2/g and crystalline quartz was formed. Therefore, unique properties of the prepared SiO2, Nb2O5 and Al2O3 aerogels, including extremely porous structure and large surface area, disappeared due to the severe APHDO reaction condition. In general, the aerogel synthesis does not produce a new and thermodynamically stable phase, but simply stabilizes a porous network by
Fig. 4. XRD patterns (A), PSD curves (B), and NH3-TPD profiles (C) of fresh (reduced for XRD and NH3-TPD, and calcined for PSD) and spent Pt/Nb2O5–SiO2 aerogel catalysts with the Nb/(Nb + Si) ratio of 0.500, 0.575, 0.625 and 0.675.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
lowering the mobility necessary for metal atoms to sinter and crystallize [19]. Nb2O5 homogeneously dispersed in SiO2 and Al2O3 holds low mobility and improved acidity [20,25,26]. From the literature, the acid strength of Nb structural units is decreased in the following order: tetrahedral NbO4 with Nb_O bonds>highly distorted octahedral NbO6 with Nb_O bonds>slightly distorted octahedral NbO6 without Nb_O bonds [20], where the last structural unit is dominant in the crystalline niobia (TT–Nb2O5). Therefore, preserving amorphous niobia phase is beneficial for maintaining the acid site strength during the reaction. Thus, Nb2O5–Al2O3 and Nb2O5–SiO2 aerogels with the molar Al/Nb or Si/Nb ratio of 1 were prepared and tested in the APHDO of 1-PrOH. As a result, Pt/Nb2O5–SiO2 showed the stable activity (conversion: 45.8%) and higher propane selectivity (43.9–50.8 C%), while the activity of Pt/Nb2O5–Al2O3 gradually increased with the TOS (Fig. 1). While the specific surface area of the latter catalyst was dropped to 68 from 213 m 2/g, it underwent the crystallization of niobia and alumina to TT–Nb2O5 and boehmite during the reaction, respectively (Fig. 2). Therefore, the gradual increase in the activity of Pt/Nb2O5–Al2O3 is explained by the formation of boehmite in the course of the reaction (Fig. S3: XRD patterns). According to the report of Hensen and coworkers [7], Pt in interaction with boehmite is more active in hydrogenation reactions than Pt in interaction with γ-Al2O3 and higher surface acidity is obtained as a result of boehmite formation. In sharp contrast, Pt/Nb2O5–SiO2 kept relatively amorphous and original pore structure (Fig. 3), where the XRD peak of crystalline TT–Nb2O5 was dimly visible on the spent catalyst and the decrease of surface area from 365 to 210 m 2/g was confirmed. Also, its acid site density was retained (Table 1). Pt/Nb2O5–SiO2 aerogel was therefore thought to be a highly durable catalyst at high H2 pressures and temperatures in the presence of water. For further confirmation, 1 wt.% Pt/Nb2O5–SiO2 aerogels with the Nb/(Nb + Si) ratio in the range of 0.575 and 0.675 were prepared and tested. The conversion of 1-PrOH and gas-phase product selectivities obtained for the reaction period of 24 h are summarized in Table 2. The catalytic performance of all Pt/Nb2O5–SiO2 catalysts nearly approached to steady state at the TOS of 10 h and maintained until 24 h. As the Nb amount in Pt/Nb2O5–SiO2 catalysts varied, the conversion was not much different (50% ± 4% at 24 h) whereas the selectivity to propane decreased with the higher amounts of Nb. XRD analysis indicated that all fresh catalysts were completely amorphous. In the case of spent catalysts, the formation of TT– Nb2O5 was not important (Fig. 4A). The surface area and pore volume of spent catalysts are a little lower than those of the fresh ones, as summarized in Table 1, but mesoporous aerogel structures were still retained (Fig. 4B). A negligible change in the amount of acid sites confirmed by NH3-TPD profiles (Fig. 4C) proves the excellent stability of the catalysts composed of Nb2O5–SiO2 aerogels. It was reported that highly acidic tetrahedral niobia units containing Nb_O bonds could be stabilized in silica matrix [19,20]. It is also possible to stabilize niobia in a structure containing Nb_O bonds on silica with strong Brönsted acidity which comes from a terminal oxygen effect and an inductive effect of the support that combine to delocalize the charge upon proton removal. [27]. It is hence thought that NH3 desorption above 450 °C depicting a presence of strong acid sites is attributed to the niobia–silica interaction. Based on the above results, Nb2O5–SiO2 aerogels prepared in this work are considered to be relatively robust under the APHDO reaction condition. It is presumed that the two oxides are intimately mixed, which inhibits the transformation of amorphous Nb2O5 to crystalline phase and prevents sintering and agglomeration during reaction. Although there may be Nb- or Si-enriched domain and/or Nb2O5 may be dispersed in the bulk of the SiO2 matrix as aggregated nanosized particles due to the high content of Nb, it is thought that Nb2O5 and SiO2 are mixed well enough to endure the severe reaction condition. In order to inhibit the carbon loss from 1-PrOH through the APHDO reaction, the desired initial reaction step is dehydration by acid sites,
followed by hydrogenation of propene to propane on Pt particles. Thus, the relative contribution of each function, metal and acid, determines the product distribution. However, on Nb2O5, it is difficult to find a relationship between the catalytic performance and the ability of Pt to chemisorb H2. The amount of adsorbed H2 decreased with the Nb amount increasing (Fig. S4: H2 amount adsorbed), probably due to strong metal support interactions (SMSI) suppressing the ability to chemisorb H2 or CO in agreement with the reports that Nb2O5 appears to have reverse SMSI behavior [28–30]. However, the catalytic activity showed the opposite trend. It is presumed that the difference between the measurement condition and the reaction condition makes it difficult to evaluate the Pt activity. In terms of acidity, the selectivity to propane increased with the total acid site density (Fig. 5). The catalysts with the Nb/(Nb+ Si) ratio of 0.500 and 0.575 reached the propane/ethane ratio obtained at 24 h (Table 2) of about 1.0. It means that 50 mol% of 1-PrOH is initially changed through dehydration. Pt/Nb2O5–SiO2 aerogel catalysts appeared to be highly active in the presence of water. 4. Conclusions Unlike many previous reports focusing on the development of an active and selective catalyst for the APHDO reaction, this work investigated the catalyst durability due to the severe reaction condition. Thus, Pt catalysts supported on crystalline Nb2O5·xH2O, and amorphous SiO2, Al2O3 and Nb2O5 aerogels were tested for their activity in the APHDO of 1-PrOH at 230 °C and 35 bar. While the first three catalysts showed low conversions, their support materials were structurally transformed into crystalline TT–Nb2O5, quartz and boehmite during the reaction. Although Pt/Nb2O5 aerogel also showed the crystallization into TT–Nb2O5, the initial activity was pretty high along with the deactivation behavior. Thus, Pt/Nb2O5–Al2O3 and Pt/Nb2O5–SiO2 aerogels were prepared and tested; however, the former catalyst experienced the crystallization into TT–Nb2O5 and boehmite under the hydrothermal reaction condition. In the case of Pt/Nb2O5–SiO2 aerogels with varying Nb/(Nb+ Si) ratios, the pore collapse and surface area decrease were largely diminished, while the structure is fairly X-ray amorphous. The acid site density of Pt/Nb2O5–SiO2 aerogels was also maintained. In terms of the catalytic performance, Pt/Nb2O5–SiO2 aerogel catalysts exhibited the similar conversion around 50% up to 24 h and the molar propane/ethane ratio increased to about 1.0 with the acid site density. Consequently,
1.2
1.1
Propane/ehtane (mol/mol)
46
0.575 1.0
0.500 0.625
0.9
0.8
0.675 0.7
0.6 2.8
3.0
3.2
Acid site density
3.4
3.6
(μmol/m2)
Fig. 5. Propane/ethane molar ratio vs. acid site density of 1 wt.% Pt/Nb2O5–SiO2 aerogel catalysts with the Nb/(Nb + Si) ratio of 0.500, 0.575, 0.625 and 0.675 obtained after the reaction for 24 h.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
Pt/Nb2O5–SiO2 aerogels are considered to be a highly durable and active catalyst in the APHDO reaction of 1-PrOH. Acknowledgments This research was financially supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology of Korea (Grant no. 2012R1A1A1007866) and Industrial Source Technology Development Program of the Ministry of Knowledge and Economy (MKE) of Korea (Grant no. 2012‐10042591). Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.catcom.2012.09.022. References [1] D.M. Alonso, J.Q. Bond, J.A. Dumesic, Green Chemistry 12 (2010) 1493–1513. [2] G.W. Huber, J.A. Dumesic, Catalysis Today 111 (2006) 119–132. [3] R.M. West, M.H. Tucker, D.J. Braden, J.A. Dumesic, Catalysis Communications 10 (2009) 1743–1746. [4] J.C. Serrano-Ruiz, J.A. Dumesic, Green Chemistry 11 (2009) 1101–1104. [5] R. Xing, A.V. Subrahmanyam, H. Olcay, W. Qi, G.P. van Walsum, H. Pendse, G.W. Huber, Green Chemistry 12 (2010) 1933–1946. [6] R. Weingarten, G.A. Tompsett, W.C. Conner Jr., G.W. Huber, Journal of Catalysis 279 (2011) 174–182. [7] A. Ciftci, B. Peng, A. Jentys, J.A. Lercher, E.J.M. Hensen, Applied Catalysis A 431–432 (2012) 113–119.
47
[8] A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A.A. Lemonidou, J.A. Lercher, Journal of Catalysis 269 (2010) 411–420. [9] M. Balaraju, V. Rekha, P.S. Sai Prasad, B.L.A. Prabhavathi Devi, R.B.N. Prasad, N. Lingaiah, Applied Catalysis A 354 (2009) 82–87. [10] N. Li, G.A. Tompsett, G.W. Huber, ChemSusChem 3 (2010) 1154–1157. [11] N. Li, G.W. Huber, Journal of Catalysis 270 (2010) 48–59. [12] N. Li, G.A. Tompsett, T. Zhang, J. Shi, C.E. Wyman, G.W. Huber, Green Chemistry 13 (2011) 91–101. [13] L. Chen, Y. Zhu, H. Zheng, C. Zhang, B. Zhang, Y. Li, Journal of Molecular Catalysis A 351 (2011) 217–227. [14] B. Peng, C. Zhao, I. Mejia-Centeno, G.A. Fuentes, A. Jentys, J.A. Lercher, Catalysis Today 183 (2012) 3–9. [15] P. Carniti, A. Gervasini, M. Marzo, Catalysis Today 152 (2010) 42–47. [16] I. Nowak, M. Ziolek, Chemical Reviews 99 (1999) 3603–3624. [17] P. Batamack, R. Vincent, J. Fraissard, Catalysis Today 28 (1996) 31–39. [18] K. Nakajima, Y. Baba, R. Noma, M. Kitano, J.N. Kondo, S. Hayashi, M. Hara, Journal of the American Chemical Society 133 (2011) 4224–4227. [19] S.M. Maurer, E.I. Ko, Journal of Catalysis 135 (1992) 125–134. [20] S.M. Maurer, D. Ng, E.I. Ko, Catalysis Today 16 (1993) 319–331. [21] Y.J. Pagán-Torres, J.M.R. Gallo, D. Wang, H.N. Pham, J.A. Libera, C.L. Marshall, J.W. Elam, A.K. Datye, J.A. Dumesic, ACS Catalysis 1 (2011) 1234–1245. [22] H.N. Pham, Y.J. Pagan-Torres, J.C. Serrano-Ruiz, D. Wang, J.A. Dumesic, A.K. Datye, Applied Catalysis A 397 (2011) 153–162. [23] J.W. Shabaker, G.W. Huber, R.R. Davda, R.D. Cortright, J.A. Dumesic, Catalysis Letters 88 (2003) 1–8. [24] S.C. Shen, Q. Chen, P.S. Chow, G.H. Tan, X.T. Zeng, Z. Wang, R.B.H. Tan, Journal of Physical Chemistry C 111 (2007) 700–707. [25] M.S.P. Francisco, R. Landers, Y. Gushikem, Journal of Solid State Chemistry 177 (2004) 2432–2439. [26] J.G. Weissman, Catalysis Today 28 (1996) 159–166. [27] E.C. DeCanio, V.P. Nero, E.I. Ko, Journal of Catalysis 146 (1994) 317–319. [28] R. Brown, C. Kemball, Journal of the Chemical Society, Faraday Transactions 92 (1996) 281–288. [29] T. Uchijima, Catalysis Today 28 (1996) 105–117. [30] D.A.G. Aranda, M. Schmal, Journal of Catalysis 171 (1997) 398–405.
Contents lists available at SciVerse ScienceDirect
Catalysis Communications journal homepage: www.elsevier.com/locate/catcom
Short Communication
Highly durable Pt-supported niobia–silica aerogel catalysts in the aqueous-phase hydrodeoxygenation of 1-propanol Jihye Ryu a, b, Sung Min Kim c, Jae-Wook Choi a, Jeong-Myeong Ha a, Dong June Ahn b, Dong Jin Suh a,⁎, Young-Woong Suh c,⁎⁎ a b c
Clean Energy Research Center, Korea Institute of Science and Technology, Seoul 136‐791, Republic of Korea Department of Chemical and Biological Engineering, Korea University, Seoul 136‐701, Republic of Korea Department of Chemical Engineerling, Hanyang University, Seoul 133‐791, Republic of Korea
a r t i c l e
i n f o
Article history: Received 11 August 2012 Received in revised form 15 September 2012 Accepted 18 September 2012 Available online 24 September 2012 Keywords: 1-Propanol Aqueous-phase processing Hydrodeoxygenation Aerogel Pt/Nb2O5–SiO2
a b s t r a c t The aqueous-phase hydrodeoxygenation (APHDO) of 1-propanol at 230 °C and 35 bar was studied over 1 wt.% Pt catalysts supported on several metal oxides. Pt catalysts supported on amorphous silica, alumina and niobia aerogels, and crystalline niobic acid calcined at 500 °C showed low activities or deactivation. Under the APHDO condition, these supports experienced a structure transformation to crystalline quartz, boehmite and niobia TT phase. Pt/Nb2O5–Al2O3 aerogel also suffered from the same crystallization behavior. In contrast, Pt/Nb2O5–SiO2 aerogels with different Nb/(Nb + Si) ratios maintained a good catalytic performance for prolonged reaction periods. Through characterizations of spent catalysts, Nb2O5–SiO2 aerogels were found to retain X-ray amorphous and porous structure. Also, their acid site densities were negligibly changed during the reaction. The catalysts with the Nb / (Nb + Si) ratio of 0.500 and 0.575 reached the molar propane/ethane ratio of about 1.0, indicating that Pt/Nb2O5–SiO2 aerogel catalysts are highly active in the presence of water. © 2012 Elsevier B.V. All rights reserved.
1. Introduction An efficient conversion of biomass-derived molecules into fuels and chemicals has been studied in recent years [1–6]. Particularly, much attention has been paid to aqueous-phase processing of C3 alcohols to produce H2 and alkanes [7–9]. Huber and coworkers investigated the aqueous-phase hydrodeoxygenation (APHDO) of sorbitol and sugar solutions towards understanding the reaction chemistry [10–12]. In the APHDO process, a series of dehydration and hydrogenation steps takes place over bifunctional catalysts in which an acid is responsible for the dehydration of the reactant and a metal for subsequent hydrogenation. The carbon loss of reactant can be minimized in this case. However, if metal contribution is more dominant than acid, lighter alkanes are produced through dehydrogenation and decarbonylation reactions on metal surfaces. Therefore, the adjustment in the concentration and strength of acid sites is of importance to selectively produce heavier alkanes over bifunctional catalysts. The similar assessment was done on the aqueous-phase reforming and hydrodeoxygenation of C3 alcohols [7–10,13,14]. In this work, 1-propanol (1-PrOH) was used as a model reactant for oxygenated biomass-derived compounds, due to the difficulty to ⁎ Corresponding author. Tel.: +82 2 958 5192; fax: +82 2 958 5209. ⁎⁎ Corresponding author. Tel.: +82 2 2220 2329; fax: +82 2 2298 4101. E-mail addresses: [email protected] (D.J. Suh), [email protected] (Y.-W. Suh). 1566-7367/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.catcom.2012.09.022
dehydrate 1-PrOH with a terminal OH among C3 alcohols along with the simple product distribution. Major products in the APHDO of 1-PrOH are CO, CO2, CH4, C2H6 and C3H8 in gas phase. When the dehydration and hydrogenation of 1-propanol occur in sequence (C\O bond cleavage), propane is the main product. In contrast, ethane and CO are formed by the dehydrogenation and decarbonylation (C\C bond cleavage). CO further undergoes the water–gas shift into CO2 and methanation into CH4. Therefore, if it is desired to lower the oxygen content in products and minimize carbon loss by suppressing the cleavage of C\C bond, an acid part of bifunctional catalyst responsible for the dehydration of 1-PrOH is then crucial, as in the case of sorbitol [10]. Niobium oxides have been found to be active in acid-catalyzed reactions, even in the presence of water [15–18]. It was reported that amorphous niobium oxide is composed of distorted NbO6 octahedra and NbO4 tetrahedra [19,20]; however, it suffers from poor hydrothermal stability, resulting in both a poor activity and low surface area after use [21,22]. This predicts that the texture, structure or acidity of niobia may be changed under aqueous-phase reaction conditions due to high temperature and H2 pressure in water. The prediction is supported by the recent report of Hensen and coworkers that boehmite formation is observed for spent Pt/γ-Al2O3 and Pt/SiO2–Al2O3 catalysts in the aqueous-phase glycerol reforming [7]. Therefore, the catalyst stability is an important issue in the APHDO reaction, besides a superior catalytic performance.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
Fig. 1. Conversion of 1-PrOH (A) and carbon selectivity of gas-phase products obtained over 1 wt.% Pt-supported catalysts (B–G):
41
C3H8,
C2H6,
CO2,
CO, and
CH4.
42
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
In this work, we focused on examining the activity, product selectivity and stability of Pt catalysts supported on commercial niobic acid (Nb2O5·xH2O), and SiO2, Al2O3 and Nb2O5 aerogels. Herein, aerogel-type metal oxides were chosen because it is easy to monitor the change of catalyst properties during the APHDO reaction, owing to an amorphous character, highly porous structure and large surface area. Furthermore, Nb2O5–Al2O3 aerogel as well as Nb2O5–SiO2 aerogels with different Nb amounts were prepared and used for the APHDO reaction of 1-PrOH, since niobia-containing mixed oxides are more resistant to heat and pressure [21,22]. 2. Experimental
X-ray diffraction (XRD) analysis was conducted with a Rigaku D/ Max-2500 diffractometer using Cu Kα (λ = 0.1541 nm) as a radiation source. XRD patterns were collected with a step size of 0.02° in the 2θ range of 5°–90°. N2 adsorption–desorption analysis was performed using a Micromeritics ASAP 2020 and ASAP 2010. N2 BET (Brunauer– Emmett–Teller) specific surface areas of fresh and spent catalysts were measured from the adsorption isotherm branch and pore size distributions were calculated by BJH (Barrett–Joyner–Halenda) method with desorption isotherm branches. Before measurement, samples were degassed at 120 °C under vacuum overnight and subsequently purged with He. H2 chemisorption was performed by a BEL-CAT instrument. About 50 mg of catalyst was successively pretreated in flowing 10% O2/He
2.1. Preparation of catalysts
2.2. Catalyst characterization NH3 temperature programmed desorption (NH3-TPD) experiments were carried out using a BEL-CAT instrument (BEL Japan, Inc.) equipped with a thermal conductivity detector (TCD). In a typical experiment, about 30 mg of catalyst was loaded and reduced in H2 at 400 °C for 30 min. It was then purged in He with ramping the temperature to 500 °C and cooled to 100 °C for NH3 adsorption. After exposure of the reduced catalyst to a flow of 5% NH3/He (50 mL/min) for 30 min, NH3 physically adsorbed was removed by purging pure He for 1 h at 100 °C. NH3-TPD profiles were obtained by ramping the temperature at a heating rate of 5 °C/min under pure He (30 mL/min) to 900 °C. Used catalysts were also subjected to TPD analysis, for which drying under vacuum overnight and pretreatment in He at 500 °C were conducted in advance.
A 1 wt.% Pt/Nb2O5-Al2O3 1 wt.% Pt/Al2O3 1 wt.% Pt/Nb2O5-SiO2
Intensity (a.u.)
1 wt.% Pt/SiO2 1 wt.% Pt/Nb2O5
1 wt.% Pt/Nb2O5.xH2O 20
40
60
80
2θ (deg.)
B 1 wt.% Pt/Nb2O5-Al2O3 1 wt.% Pt/Al2O3 1 wt.% Pt/Nb2O5-SiO2
Intensity (a.u.)
Nb2O5 gel was synthesized by the hydrolysis and condensation of niobium chloride (NbCl5, 99%) in ethanol (ethanol/Nb= 47 mol/mol). Nitric acid (HNO3/Nb =0.2 mol/mol, 70%) and deionized water (H2O/ Nb =5 mol/mol) were added for hydrolysis, and propylene oxide (propylene oxide/Nb= 6.3 mol/mol, 99%) as a gelation promoter was then added with stirring. During gel formation, the viscosity increased with heat evolution. Al2O3 gel was also prepared using AlCl3·6H2O (min. 97.0%), HNO3, water, propylene oxide and ethanol in a molar ratio of 1:0.1:3:3:35. In the case of SiO2 gel, the sol–gel process of tetraethyl orthosilicate (TEOS) was catalyzed by NH4OH (min. 28.0%) and NH4F (min. 97.0%), where a molar ratio of TEOS, methanol, H2O, NH4OH and NH4F was 1:12.5:4:0.005:0.001. Nb2O5–SiO2 gels were prepared by simply adding TEOS into the niobium solution, prior to the hydrolysis. The typical molar ratio of (Nb+Si), HNO3, water, propylene oxide and ethanol was 2:0.2:6:6:45. For the preparation of niobia–silica with different Nb/(Nb+Si) ratios, the amount of propylene oxide added for gel formation was proportional to that of niobium while nitric acid and water were not changed. For Nb2O5–Al2O3 gel, aluminum chloride hexahydrate (Al/Nb=1 mol/mol) was dissolved in the mixture of nitric acid (HNO3/Nb=0.3 mol/mol) and ethanol (ethanol/Nb=45 mol/mol). After the water addition (H2O/ Nb=8 mol/mol) to completely dissolve the aluminum precursor, niobium chloride was added. The mixture was transformed into gel by propylene oxide (propylene oxide/Nb=8 mol/mol). The resulting wet gels were kept at room temperature for 3 days except Nb2O5 gel (1 day). For supercritical drying, ethanol in wet gel pores was exchanged with liquid CO2 at a pressure of 2000 psi and dried at 70 °C. To remove the remaining organic materials before Pt loading, single metal oxide aerogels and niobic acid (Soekawa chemical) were calcined at 500 °C for 4 h, whereas mixed oxide aerogels were treated at 500 °C for 8 h. Catalysts containing 1 wt.% Pt were prepared by incipient wetness impregnation with an aqueous solution of H2PtCl6·6H2O (≥ 37.5% Pt basis), dried in an oven at 105 °C overnight and calcined at 400 °C for 2 h in flowing air.
1 wt.% Pt/SiO2
1 wt.% Pt/Nb2O5
1 wt.% Pt/Nb2O5.xH2O 20
40
60
80
2θ (deg.) Fig. 2. XRD patterns of reduced (A) and spent catalysts after the reaction for 10 or 24 h (B); (a) Pt/Nb2O5·xH2O, (b) Pt/Nb2O5, (c) Pt/SiO2, (d) Pt/Nb2O5–SiO2, (e) Pt/Al2O3, (f) Pt/Nb2O5–Al2O3; Pt, TT–Nb2O5, quartz, and boehmite.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
43
Table 1 Textural and acid properties of fresh and spent Pt-supported catalysts. Catalyst
Pt/Nb2O5·xH2O Pt/Nb2O5
Pt/SiO2 Pt/Al2O3 Pt/Nb2O5–Al2O3
Pt/Nb2O5–SiO2 0.500 Pt/Nb2O5–SiO2 0.575 Pt/Nb2O5–SiO2 0.625 Pt/Nb2O5–SiO2 0.675 a
TOS
Fresh 10 h Fresh 10 h 24 h Fresh 24 h Fresh 24 h Fresh 10 h 24 h Fresh 10 h 24 h Fresh 24 h Fresh 24 h Fresh 24 h
Surface area (m2/g)
Pore volumea (cm3/g)
17 17 198 42 32 751 4 313 66 213 117 68 365 263 210 326 252 345 259 299 259
0.067 0.074 0.828 0.167 0.229 0.829 0.022 0.288 0.458 1.236 0.992 0.317 1.145 1.010 0.822 0.985 0.872 1.087 0.856 0.934 0.799
NH3-TPD (μmol/m2) T b 450 °C
T > 450 °C
Total
1.454 – 2.762 2.717 3.272 0.086 – 2.202 3.425 2.287 2.064 3.048 1.853 1.851 2.123 2.128 2.234 2.091 2.079 2.177 1.979
63.171 47.954 0.009 0.697 8.254 1.217 22.842 4.699 14.262 1.015 2.334 6.069 0.339 0.923 1.338 0.376 0.848 0.293 0.845 0.280 0.907
64.625 47.954 2.771 3.414 11.526 1.302 22.842 6.900 17.687 3.303 4.399 9.117 2.192 2.774 3.461 2.505 3.082 2.384 2.924 2.458 2.886
BJH desorption cumulative pore volume.
for 15 min and 10% H2/Ar at 400 °C for 30 min, and then cooled to 55 °C in Ar. H2 adsorption took place at 55 °C with 10% H2/Ar. 2.3. Catalytic activity test Activity tests were carried out in an upflow stainless steel reactor to which a gas–liquid separator (internal volume: 65 mL) was connected at an upper position (Fig. S1). The catalysts were packed with glass beads between glass wool and gasket. Pt-supported catalysts were reduced at 400 °C for 2 h under flowing H2 (100 mL/min). After the reactor was cooled to room temperature, the system was pressurized with H2 (200 mL/min) to 35 bar, and heated to 230 °C again with H2 flow of 13 mL/min. The total pressure in the reactor and separator was controlled by a back-pressure regulator (BPR) attached to the exit line of the separator. An aqueous solution of 10 wt.% 1-PrOH was then fed at a flow rate of 0.05 mL/min. The time was counted when liquid products were visible in the bottom of the separator (ca. 2 h). During the catalytic run, gas-phase products released through the BPR were analyzed using an online GC equipped with FID installed with Hayesep D column. Because the SUS line connected between the reactor and the separator was air-cooled with a fan, the temperature of liquid sample in the separator during the reaction was about 30 °C, preventing volatile liquid molecules from passing through the separator and BPR. The liquid sample collected in the separator was drained every 2 h and analyzed using a HPLC equipped with RID and Bio-Rad HPX-87H column. Due to the pressure difference between the reaction pressures (35 bar) and atmospheric pressure, much care was taken to trap most of liquid molecules using a loosely tight glass vial. The conversion of 1-PrOH was determined from the difference of 1-PrOH concentration by HPLC analysis. Other liquid products, such as propionic acid, propanal, 2-PrOH and propyl propanoate, were detected in negligible amounts. Thus, the carbon selectivity (C%) of only gas-phase products was calculated based on the amount detected by GC analysis, as follows: Carbon selectivityðC% Þ ¼
ðmole of a specific gas productÞ ðnumber of carbon atom in a moleculeÞ ðmoles of CO; CO2 and CH4 Þ þ 2 ðmole of C2 H6 Þ þ 3 ðmole of C3 H8 Þ 100:
In the beginning of the reaction, C3H8, C2H6, CO, CO2 and CH4 were detected, but the formation of CO and CH4 sharply diminished over the time-on-stream, probably due to the fast water–gas shift reaction into CO2 and H2 as well as the decline of methanation reaction. Note that there is a certain extent of sample loss when the liquid sample was drained, although several precautions were undertaken in order to achieve the perfect material balance. When the absolute amount of 1-propanol consumed measured by HPLC analysis was compared to that of gaseous products detected by a GC, the observed difference was in the range of ±5%. Thus, the same experiment was repeated at least 2–3 times for reproducibility. 3. Results and discussion The APHDO reaction at 230 °C and 35 bar was conducted using Pt catalysts supported on different metal oxides, such as niobic acid (Nb2O5·xH2O), and SiO2, Al2O3, Nb2O5, Nb2O5–SiO2 and Nb2O5– Al2O3 aerogels. Fig. 1 shows the conversion of 1-PrOH and carbon selectivity of gas-phase products. In the blank test using glass beads
Table 2 The conversion of 1-propanol and gas-phase product selectivities using Pt/Nb2O5–SiO2 aerogel catalysts. Nb/(Nb + Si)
TOS
Conversion (%)
Carbon selectivity (C%) CO
CO2
CH4
C2H6
C3H8
0.500
2h 10 h 24 h 2h 10 h 24 h 2h 10 h 24 h 2h 10 h 24 h
44.2 43.2 45.8 47.9 46.2 49.3 65.0 54.7 53.8 62.7 54.2 50.4
0.18 0.13 0.09 0.19 0.14 0.12 0.14 0.10 0.10 0.16 0.12 0.09
16.95 16.54 15.00 16.78 17.02 15.36 16.30 16.61 15.71 17.10 17.45 17.16
0.14 0.09 0.06 0.20 0.09 0.08 0.20 0.09 0.09 0.21 0.09 0.09
38.75 37.98 37.08 36.60 37.62 33.65 37.34 38.27 35.62 37.60 38.74 38.24
43.98 45.26 50.77 46.22 45.13 50.79 46.02 44.94 48.49 44.94 43.59 44.42
0.575
0.625
0.675
C3/C2a (mol/mol) 0.757 0.794 0.913 0.842 0.800 1.006 0.822 0.783 0.907 0.797 0.750 0.774
a Calculated from the equation: C3/C2 = [2 × (carbon selectivity of propane)] / [3 × (carbon selectivity of ethane)].
44
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
under this condition, the 1-PrOH conversion calculated by HPLC analysis was 3%–4%, where only trace amounts of liquid products were formed and gaseous products were not detected at all. Among the six catalysts, the lowest conversion was obtained over Pt/SiO2 aerogel (ca. 3.6% at 24 h) along with more selective formation of ethane than propane. This is due to relatively neutral SiO2 surface, even though the surface silanol species can act as Brönsted acid. Pt/Al2O3 aerogel also showed the conversion of less than 20% and the carbon selectivity to propane lower than that to ethane, although the cleavage of C\O bond is expected to be more favorable due to acid sites of Al2O3. It is speculated that the acidity of Al2O3 are not strong enough to catalyze the dehydration of 1-PrOH due to the competitive adsorption of H2O on Al2O3 [14]. From observations that Pt/Al2O3 showed high production of H2 and CO2 during aqueous-phase reforming of
oxygenated compounds [7,8,23], it can be inferred that the cleavage of C\C bond is favorable on Pt/Al2O3. In the case of Pt/Nb2O5 aerogel, the conversion of 92% and propane selectivity of 52 C% were initially observed; however, both declined to 49% and 42 C% in 10 h. This is explained from the activity and characteristics of Pt/Nb2O5·xH2O of which the phase was already transformed from amorphous to crystalline TT phase (TT–Nb2O5) during calcination at 500 °C (Fig. 2A). Since crystalline niobia has less acid sites and lower specific surface areas than amorphous one [16,19,20], Pt/Nb2O5·xH2O showed the conversion of 10.5% and the ethane selectivity of 54.6 C% at 10 h. By comparing fresh and spent Pt/Nb2O5 aerogel catalysts, it was noticed that Nb2O5 aerogel was crystallized into TT–Nb2O5 during the reaction at 230 °C for 10 h (Fig. 2) and its BET specific surface area was significantly reduced from 198 to 42 m2/g (Table 1). NH3-TPD
6
6
1 wt.% Pt/Nb2O5.xH2O
1 wt.% Pt/Nb2O5 5
Pore volume (cm3/g)
Pore volume (cm3/g)
5 4 3 2
4 3 2 1
1
0
0 1
10
100
1
Pore diameter (nm)
1 wt.% Pt/SiO2
1 wt.% Pt/Nb2O5-SiO2 5
Pore volume (cm3/g)
5
Pore volume (cm3/g)
100
6
6
4 3 2
4 3 2 1
1
0
0 1
10
1
100
10
100
Pore diameter (nm)
Pore diameter (nm) 6
6
1 wt.% Pt/Al2O3
1 wt.% Pt/Nb2O5-Al2O3 5
Pore volume (cm3/g)
5
Pore volume (cm3/g)
10
Pore diameter (nm)
4 3 2
4 3 2 1
1
0
0 1
10
100
1
Fig. 3. Pore size distribution (PSD) curves of Pd-supported catalysts;
10
100
Pore diameter (nm)
Pore diameter (nm) after calcination,
after the reaction for 10 h, and
after the reaction for 24 h.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
peak detected below 450 °C was also very little in the spent Pt/Nb2O5 (Fig. S2: NH3-TPD profiles). Therefore, the phase modification of Pt/ Nb2O5 during the reaction leads to the decrease of both the 1-PrOH conversion and propane selectivity. In terms of the structure change under the hydrothermal reaction condition, Pt/Al2O3 was found to be transformed into crystalline boehmite (AlOOH) with the lower specific surface area of 66 m 2/g (fresh catalyst: 313 m 2/g) and larger pore diameter. This is in accordance with the report of Hensen and coworkers [7], which is related to the hydroxylation of alumina domains [14,24]. However, it did
45
not show a drop of catalytic activity at all. This may be accounted for by the fact that acid sites of the spent Pt/Al2O3 were still intact (Table 1). The above behavior was similarly observed in Pt/SiO2; after the reaction for 24 h, the specific surface area decreased from 751 to 4 m 2/g and crystalline quartz was formed. Therefore, unique properties of the prepared SiO2, Nb2O5 and Al2O3 aerogels, including extremely porous structure and large surface area, disappeared due to the severe APHDO reaction condition. In general, the aerogel synthesis does not produce a new and thermodynamically stable phase, but simply stabilizes a porous network by
Fig. 4. XRD patterns (A), PSD curves (B), and NH3-TPD profiles (C) of fresh (reduced for XRD and NH3-TPD, and calcined for PSD) and spent Pt/Nb2O5–SiO2 aerogel catalysts with the Nb/(Nb + Si) ratio of 0.500, 0.575, 0.625 and 0.675.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
lowering the mobility necessary for metal atoms to sinter and crystallize [19]. Nb2O5 homogeneously dispersed in SiO2 and Al2O3 holds low mobility and improved acidity [20,25,26]. From the literature, the acid strength of Nb structural units is decreased in the following order: tetrahedral NbO4 with Nb_O bonds>highly distorted octahedral NbO6 with Nb_O bonds>slightly distorted octahedral NbO6 without Nb_O bonds [20], where the last structural unit is dominant in the crystalline niobia (TT–Nb2O5). Therefore, preserving amorphous niobia phase is beneficial for maintaining the acid site strength during the reaction. Thus, Nb2O5–Al2O3 and Nb2O5–SiO2 aerogels with the molar Al/Nb or Si/Nb ratio of 1 were prepared and tested in the APHDO of 1-PrOH. As a result, Pt/Nb2O5–SiO2 showed the stable activity (conversion: 45.8%) and higher propane selectivity (43.9–50.8 C%), while the activity of Pt/Nb2O5–Al2O3 gradually increased with the TOS (Fig. 1). While the specific surface area of the latter catalyst was dropped to 68 from 213 m 2/g, it underwent the crystallization of niobia and alumina to TT–Nb2O5 and boehmite during the reaction, respectively (Fig. 2). Therefore, the gradual increase in the activity of Pt/Nb2O5–Al2O3 is explained by the formation of boehmite in the course of the reaction (Fig. S3: XRD patterns). According to the report of Hensen and coworkers [7], Pt in interaction with boehmite is more active in hydrogenation reactions than Pt in interaction with γ-Al2O3 and higher surface acidity is obtained as a result of boehmite formation. In sharp contrast, Pt/Nb2O5–SiO2 kept relatively amorphous and original pore structure (Fig. 3), where the XRD peak of crystalline TT–Nb2O5 was dimly visible on the spent catalyst and the decrease of surface area from 365 to 210 m 2/g was confirmed. Also, its acid site density was retained (Table 1). Pt/Nb2O5–SiO2 aerogel was therefore thought to be a highly durable catalyst at high H2 pressures and temperatures in the presence of water. For further confirmation, 1 wt.% Pt/Nb2O5–SiO2 aerogels with the Nb/(Nb + Si) ratio in the range of 0.575 and 0.675 were prepared and tested. The conversion of 1-PrOH and gas-phase product selectivities obtained for the reaction period of 24 h are summarized in Table 2. The catalytic performance of all Pt/Nb2O5–SiO2 catalysts nearly approached to steady state at the TOS of 10 h and maintained until 24 h. As the Nb amount in Pt/Nb2O5–SiO2 catalysts varied, the conversion was not much different (50% ± 4% at 24 h) whereas the selectivity to propane decreased with the higher amounts of Nb. XRD analysis indicated that all fresh catalysts were completely amorphous. In the case of spent catalysts, the formation of TT– Nb2O5 was not important (Fig. 4A). The surface area and pore volume of spent catalysts are a little lower than those of the fresh ones, as summarized in Table 1, but mesoporous aerogel structures were still retained (Fig. 4B). A negligible change in the amount of acid sites confirmed by NH3-TPD profiles (Fig. 4C) proves the excellent stability of the catalysts composed of Nb2O5–SiO2 aerogels. It was reported that highly acidic tetrahedral niobia units containing Nb_O bonds could be stabilized in silica matrix [19,20]. It is also possible to stabilize niobia in a structure containing Nb_O bonds on silica with strong Brönsted acidity which comes from a terminal oxygen effect and an inductive effect of the support that combine to delocalize the charge upon proton removal. [27]. It is hence thought that NH3 desorption above 450 °C depicting a presence of strong acid sites is attributed to the niobia–silica interaction. Based on the above results, Nb2O5–SiO2 aerogels prepared in this work are considered to be relatively robust under the APHDO reaction condition. It is presumed that the two oxides are intimately mixed, which inhibits the transformation of amorphous Nb2O5 to crystalline phase and prevents sintering and agglomeration during reaction. Although there may be Nb- or Si-enriched domain and/or Nb2O5 may be dispersed in the bulk of the SiO2 matrix as aggregated nanosized particles due to the high content of Nb, it is thought that Nb2O5 and SiO2 are mixed well enough to endure the severe reaction condition. In order to inhibit the carbon loss from 1-PrOH through the APHDO reaction, the desired initial reaction step is dehydration by acid sites,
followed by hydrogenation of propene to propane on Pt particles. Thus, the relative contribution of each function, metal and acid, determines the product distribution. However, on Nb2O5, it is difficult to find a relationship between the catalytic performance and the ability of Pt to chemisorb H2. The amount of adsorbed H2 decreased with the Nb amount increasing (Fig. S4: H2 amount adsorbed), probably due to strong metal support interactions (SMSI) suppressing the ability to chemisorb H2 or CO in agreement with the reports that Nb2O5 appears to have reverse SMSI behavior [28–30]. However, the catalytic activity showed the opposite trend. It is presumed that the difference between the measurement condition and the reaction condition makes it difficult to evaluate the Pt activity. In terms of acidity, the selectivity to propane increased with the total acid site density (Fig. 5). The catalysts with the Nb/(Nb+ Si) ratio of 0.500 and 0.575 reached the propane/ethane ratio obtained at 24 h (Table 2) of about 1.0. It means that 50 mol% of 1-PrOH is initially changed through dehydration. Pt/Nb2O5–SiO2 aerogel catalysts appeared to be highly active in the presence of water. 4. Conclusions Unlike many previous reports focusing on the development of an active and selective catalyst for the APHDO reaction, this work investigated the catalyst durability due to the severe reaction condition. Thus, Pt catalysts supported on crystalline Nb2O5·xH2O, and amorphous SiO2, Al2O3 and Nb2O5 aerogels were tested for their activity in the APHDO of 1-PrOH at 230 °C and 35 bar. While the first three catalysts showed low conversions, their support materials were structurally transformed into crystalline TT–Nb2O5, quartz and boehmite during the reaction. Although Pt/Nb2O5 aerogel also showed the crystallization into TT–Nb2O5, the initial activity was pretty high along with the deactivation behavior. Thus, Pt/Nb2O5–Al2O3 and Pt/Nb2O5–SiO2 aerogels were prepared and tested; however, the former catalyst experienced the crystallization into TT–Nb2O5 and boehmite under the hydrothermal reaction condition. In the case of Pt/Nb2O5–SiO2 aerogels with varying Nb/(Nb+ Si) ratios, the pore collapse and surface area decrease were largely diminished, while the structure is fairly X-ray amorphous. The acid site density of Pt/Nb2O5–SiO2 aerogels was also maintained. In terms of the catalytic performance, Pt/Nb2O5–SiO2 aerogel catalysts exhibited the similar conversion around 50% up to 24 h and the molar propane/ethane ratio increased to about 1.0 with the acid site density. Consequently,
1.2
1.1
Propane/ehtane (mol/mol)
46
0.575 1.0
0.500 0.625
0.9
0.8
0.675 0.7
0.6 2.8
3.0
3.2
Acid site density
3.4
3.6
(μmol/m2)
Fig. 5. Propane/ethane molar ratio vs. acid site density of 1 wt.% Pt/Nb2O5–SiO2 aerogel catalysts with the Nb/(Nb + Si) ratio of 0.500, 0.575, 0.625 and 0.675 obtained after the reaction for 24 h.
J. Ryu et al. / Catalysis Communications 29 (2012) 40–47
Pt/Nb2O5–SiO2 aerogels are considered to be a highly durable and active catalyst in the APHDO reaction of 1-PrOH. Acknowledgments This research was financially supported by Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science and Technology of Korea (Grant no. 2012R1A1A1007866) and Industrial Source Technology Development Program of the Ministry of Knowledge and Economy (MKE) of Korea (Grant no. 2012‐10042591). Appendix A. Supplementary data Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.catcom.2012.09.022. References [1] D.M. Alonso, J.Q. Bond, J.A. Dumesic, Green Chemistry 12 (2010) 1493–1513. [2] G.W. Huber, J.A. Dumesic, Catalysis Today 111 (2006) 119–132. [3] R.M. West, M.H. Tucker, D.J. Braden, J.A. Dumesic, Catalysis Communications 10 (2009) 1743–1746. [4] J.C. Serrano-Ruiz, J.A. Dumesic, Green Chemistry 11 (2009) 1101–1104. [5] R. Xing, A.V. Subrahmanyam, H. Olcay, W. Qi, G.P. van Walsum, H. Pendse, G.W. Huber, Green Chemistry 12 (2010) 1933–1946. [6] R. Weingarten, G.A. Tompsett, W.C. Conner Jr., G.W. Huber, Journal of Catalysis 279 (2011) 174–182. [7] A. Ciftci, B. Peng, A. Jentys, J.A. Lercher, E.J.M. Hensen, Applied Catalysis A 431–432 (2012) 113–119.
47
[8] A. Wawrzetz, B. Peng, A. Hrabar, A. Jentys, A.A. Lemonidou, J.A. Lercher, Journal of Catalysis 269 (2010) 411–420. [9] M. Balaraju, V. Rekha, P.S. Sai Prasad, B.L.A. Prabhavathi Devi, R.B.N. Prasad, N. Lingaiah, Applied Catalysis A 354 (2009) 82–87. [10] N. Li, G.A. Tompsett, G.W. Huber, ChemSusChem 3 (2010) 1154–1157. [11] N. Li, G.W. Huber, Journal of Catalysis 270 (2010) 48–59. [12] N. Li, G.A. Tompsett, T. Zhang, J. Shi, C.E. Wyman, G.W. Huber, Green Chemistry 13 (2011) 91–101. [13] L. Chen, Y. Zhu, H. Zheng, C. Zhang, B. Zhang, Y. Li, Journal of Molecular Catalysis A 351 (2011) 217–227. [14] B. Peng, C. Zhao, I. Mejia-Centeno, G.A. Fuentes, A. Jentys, J.A. Lercher, Catalysis Today 183 (2012) 3–9. [15] P. Carniti, A. Gervasini, M. Marzo, Catalysis Today 152 (2010) 42–47. [16] I. Nowak, M. Ziolek, Chemical Reviews 99 (1999) 3603–3624. [17] P. Batamack, R. Vincent, J. Fraissard, Catalysis Today 28 (1996) 31–39. [18] K. Nakajima, Y. Baba, R. Noma, M. Kitano, J.N. Kondo, S. Hayashi, M. Hara, Journal of the American Chemical Society 133 (2011) 4224–4227. [19] S.M. Maurer, E.I. Ko, Journal of Catalysis 135 (1992) 125–134. [20] S.M. Maurer, D. Ng, E.I. Ko, Catalysis Today 16 (1993) 319–331. [21] Y.J. Pagán-Torres, J.M.R. Gallo, D. Wang, H.N. Pham, J.A. Libera, C.L. Marshall, J.W. Elam, A.K. Datye, J.A. Dumesic, ACS Catalysis 1 (2011) 1234–1245. [22] H.N. Pham, Y.J. Pagan-Torres, J.C. Serrano-Ruiz, D. Wang, J.A. Dumesic, A.K. Datye, Applied Catalysis A 397 (2011) 153–162. [23] J.W. Shabaker, G.W. Huber, R.R. Davda, R.D. Cortright, J.A. Dumesic, Catalysis Letters 88 (2003) 1–8. [24] S.C. Shen, Q. Chen, P.S. Chow, G.H. Tan, X.T. Zeng, Z. Wang, R.B.H. Tan, Journal of Physical Chemistry C 111 (2007) 700–707. [25] M.S.P. Francisco, R. Landers, Y. Gushikem, Journal of Solid State Chemistry 177 (2004) 2432–2439. [26] J.G. Weissman, Catalysis Today 28 (1996) 159–166. [27] E.C. DeCanio, V.P. Nero, E.I. Ko, Journal of Catalysis 146 (1994) 317–319. [28] R. Brown, C. Kemball, Journal of the Chemical Society, Faraday Transactions 92 (1996) 281–288. [29] T. Uchijima, Catalysis Today 28 (1996) 105–117. [30] D.A.G. Aranda, M. Schmal, Journal of Catalysis 171 (1997) 398–405.