Single-step hydroconversion of triglycerides into biojet fuel using CO-tolerant PtRe catalyst supported on USY

Journal of Catalysis 379 (2019) 180–190

Contents lists available at ScienceDirect

Journal of Catalysis journal homepage: www.elsevier.com/locate/jcat

Single-step hydroconversion of triglycerides into biojet fuel using CO-tolerant PtRe catalyst supported on USY Kyungho Lee a, Mi-Eun Lee b, Jae-Kon Kim b, Byeongcheol Shin a, Minkee Choi a,⇑ a b

Department of Chemical & Biomolecular Engineering, KAIST, 291 Daehak-ro, Yuseong-gu, Daejeon 34141, Republic of Korea Alternative Fuel R&D Team, Korea Petroleum Quality & Distribution Authority, Cheongju-City, Chungcheongbuk-do 28115, Republic of Korea

a r t i c l e

i n f o

Article history: Received 22 July 2019 Revised 6 September 2019 Accepted 30 September 2019 Available online 16 October 2019 Keywords: CO-tolerance Biojet fuel Triglyceride Hydroconversion Hydrocracking PtRe

a b s t r a c t The direct hydroconversion of triglycerides into biojet fuel (i.e., cascade reactions including hydrogenation, deoxygenation, and hydrocracking) is challenging, even though it is beneficial in terms of process intensification. This is because the minute amount of CO produced during deoxygenation can poison the metal component of various metal–acid bifunctional catalysts, causing undesired overcracking and rapid catalyst deactivation through coke formation. To overcome this problem, we synthesized a COtolerant catalyst by supporting bimetallic PtRe on ultra-stable Y (USY) zeolite as acidic support. Compared to conventional catalyst (e.g., Pt/USY), PtRe/USY showed markedly enhanced tolerance to CO because it had weakened interaction with CO and also could rapidly convert the chemisorbed CO into less harmful methane and H2O through methanation. Consequently, overcracking and catalyst deactivation were greatly suppressed during the direct triglyceride hydroconversion. It is remarkable that PtRe/USY is intrinsically a much poorer hydrocracking catalyst than Pt/USY under pure H2 atmosphere because of its high hydrogenolysis activity. However, H2O and CO produced in situ during the deoxygenation of triglycerides selectively poisoned the active sites for undesired hydrogenolysis, thereby making PtRe/ USY a highly stable and selective catalyst for producing biojet fuel. Under the optimum condition, 41 wt% of biojet fuel with respect to palm oil could be produced through direct hydroconversion, which satisfied all the required fuel specifications. Ó 2019 Elsevier Inc. All rights reserved.

1. Introduction The International Air Transport Association (IATA) announced that fuel consumption in the aviation sector will increase by 5% annually [1]. To reduce anthropogenic CO2 emissions, IATA decided to achieve carbon-neutral growth by 2020 and 50% reduction by 2050 as compared to 2005 [2]. Biomass-derived jet fuel (or biojet fuel) has been considered one of the most promising solutions in the aviation industry for reducing the operating cost and the environmental impact. Four major catalytic processes have been investigated for producing biojet fuel [3]: (a) oil-to-jet (deoxygenation and hydrocracking of triglycerides) [4–9], (b) gas-to-jet (gasification and Fischer–Tropsch synthesis) [10–12], (c) alcohol-to-jet (alcohol dehydration and oligomerization) [13–17], and (d) sugar-to-jet (biological/catalytic upgrading of sugars) [18,19]. Among these technologies, the oil-to-jet technology is already

⇑ Corresponding author. E-mail address: [email protected] (M. Choi). https://doi.org/10.1016/j.jcat.2019.09.043 0021-9517/Ó 2019 Elsevier Inc. All rights reserved.

commercially available, and the produced biojet fuel has been tested for commercial and military flights [20–22]. The resultant biojet fuel is known to contain even lower aromatic and sulfur contents than those in petroleum-derived jet fuel [23–25], and it is currently allowed for blending into conventional jet fuel up to 50% [25]. Commercial oil-to-jet processes generally include multiple reaction steps and require at least two sequentially connected reactors (Scheme 1) [3–5]. In the first reactor, three major reactions occur over supported metal (e.g., Pt, Pd, or Ni) [5,6,26–28] or MoS2-type [29,30] catalysts: (1) hydrogenation of the C@C bonds in the unsaturated fatty acid units of triglycerides (reaction i in Scheme 1a), (2) hydrogenolysis of triglycerides into fatty acids (reaction ii), and (3) deoxygenation of the fatty acids into linear paraffins (reaction iii). After the reactions in the first reactor are complete, inorganic gas products (CO, CO2, and H2O) are removed, and the liquid paraffin products are injected into the second reactor for hydrocracking (reaction iv) over metal–acid bifunctional catalysts (Scheme 1b) [5,31,32]. The final hydrocarbon products are fractionated by distillation to collect iso-rich C8
C16 paraffins,

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

181

Scheme 1. (a) Reaction pathways in the hydroconversion of triglyceride into biojet fuel and (b) general reaction processes.

which are suitable for use as biojet fuel. In the oil-to-jet process, suppressing the overcracking of hydrocarbons in the second reactor is most important because it can cause the massive formation of light hydrocarbons (
2. Experimental 2.1. Catalyst preparation USY zeolite (H+ form, Zeolyst, CBV760) was used as acidic support. Monometallic Pt (2 wt%) was supported on USY by incipient wetness impregnation of an aqueous solution containing H2PtCl66H2O (Kojima Chemical). Bimetallic PtRe (2 wt% Pt, 2 wt% Re) was supported on USY by the co-impregnation of Pt (H2PtCl66H2O, Kojima Chemical) and Re (HReO4, Aldrich) precursors. After impregnation, the samples were dried at 373 K, calcined under dry air at 673 K (ramp: 2 K min
1) for 2 h, and reduced under H2 at 573 K for 2 h (ramp: 2 K min
1).

2.2. Catalyst characterization Elemental analyses of the catalysts were carried out by inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using iCAP 6300 (Thermo Scientific). N2 adsorption-desorption isotherms were collected using a BELSorp-max volumetric analyzer (BEL) at 77 K after pre-degassing at 673 K. The specific surface areas (SBET) were determined using the Brunauer–Emmett–Teller (BET) equation in the P/P0 range of 0.05–0.20. Powder X-ray diffraction (XRD) patterns were recorded using a D2-Phaser (Bruker) equipped with a Cu Ka radiation source (30 kV, 10 mA) and a LYNXEYE detector. Pyridine-IR spectra were recorded using a Nicolet iS50 (Thermo Fisher) with a lab-made in situ cell. Typically, a self-supporting wafer containing 20 mg of the samples was vacuum-degassed at 673 K for 4 h. Subsequently, pyridine vapor adsorption was carried out at 423 K for 1 h and evacuated at the same temperature for 2 h to remove weakly adsorbed pyridine. The amounts of Brønsted (nBAS) and Lewis acid sites (nLAS) were calculated from the bands at 1545 cm
1 (e = 1.67 cm lmol
1) and 1455 cm
1 (e = 2.22 cm lmol
1), respectively [33]. H2 and CO chemisorptions were measured at 323 K by using a pulse injection method with a BELCAT-B instrument (BEL) equipped with a thermal conductivity detector (TCD). Prior to the measurements, the catalysts were reduced under H2 atmosphere at 573 K for 2 h, cooled to 323 K under H2 flow, and purged with He at 323 K for 1 h. To investigate the effects of H2O pretreatment of the catalysts

182

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

on chemisorption, the catalysts were also pre-reduced under humid H2 containing 3 kPa H2O. High-angle annular dark field-scanning transmission electron microscopy (HAADF-STEM) and elemental mapping using energy dispersive X-ray spectroscopy (EDS) were carried out using a Titan cubed G2 60–300 instrument (FEI) operated at 300 kV. The surfacearea-weighted mean particle diameter (dTEM) was calculated using the following equation [34]:

dTEM ¼

X

3

ni di =

X

ni di

2

ð1Þ

X-ray absorption spectroscopy (XAS) data were collected over the Pt LIII (11564 eV) and Re LIII (10535 eV) edges at the Pohang Accelerator Laboratory (8C-Nano XAFS beamline, 3.0 GeV, 360 mA). Prior to the measurements, 0.25 g of the catalysts were pelletized (diameter: 13 mm) and mounted in a lab-made XAS cell equipped with Al foil windows (thickness: 0.05 mm). The catalysts were pre-reduced under dry H2 atmosphere at 573 K for 1 h before the measurements. XAS data were also collected after the prereduction of the catalysts under humid H2 atmosphere containing 3 kPa H2O. The Pt foil, Na2Pt(OH)6 (Aldrich), Re powder (Aldrich, 99.995%), ReO2 (Aldrich, 99.7%), and NH4ReO4 (Aldrich, 99%) were used as reference materials for XAS analysis. The X-ray absorption near edge structure (XANES) and extended X-ray absorption fine structure (EXAFS) data were processed using Athena software. The EXAFS were analyzed using Artemis implemented in a Demeter program package (interfaces for IFEFFIT and FEFF6 codes). After normalization and background removal, the k-space EXAFS data were Fourier-transformed into R-space after k3-weighting. The theoretical EXAFS signal was constructed using the FEFF6 code and fitted in R-space. The amplitude reduction factors (S20) for Pt–Pt and Pt–O shells were determined by analyzing the Pt foil and Na2Pt (OH)6, respectively. For the Re LIII edge, the Re powder and NH4ReO4 were used to determine S20 for Re–Re and Re–O shells, respectively. The obtained S20 values (0.88 for Pt–Pt, 0.90 for Pt–O, 0.97 for Re–Re, and 0.71 for Re–O) were consistent with the values reported in the literature [35–37]. Diffuse reflectance infrared Fourier transform spectroscopy (DRIFTS) was carried out using a Nicolet iS50 spectrometer (Thermo Fisher) and a Pike Diffuse IR cell accessory. Before recording the spectra, the catalysts were pre-reduced under dry or humid H2 (3 kPa H2O) atmospheres at 573 K for 1 h. CO adsorption was carried out at 303 K for 1 h, and then the gas flow was switched to dry H2. The spectra were recorded at various temperatures (303, 323, 373, 473, and 573 K) under H2 flow. 2.3. Catalytic reactions All reactions were carried out in a down-flow plug-flow reactor (stainless steel, 16.4 mm inner diameter). Typically, 1.2 g of the sieved catalysts (150–300 lm) were physically mixed with 10.8 g of acid-purified quartz particles, loaded into the reactor, and prereduced under H2 flow at 573 K for 2 h. The hydrocracking of nhexadecane (C16, TCI, >98%) was carried out in the temperature range of 493–583 K under a total pressure of 5 MPa. The H2 flow rate was fixed at 130 cm3 min
1 and C16 was introduced using an HPLC pump at the rate of 2.4 g h
1 (WHSV: 2.0 h
1). Hydrocracking of C16 in the presence of H2O, CO, or H2O/CO was also carried out in a similar manner, except for the co-injection of H2O (15 wt% of C16), CO (4.3 wt% of C16), and H2O/CO (15 wt%/4.3 wt% of C16) into 5 MPa H2, respectively. The liquid products collected using a liquid trap were analyzed using an offline gas chromatograph (GC) equipped with a flame ionization detector (FID) and a DB-5 column (Agilent, 30 m  0.32 mm  0.25 lm). The gas products were analyzed using an online GC equipped with an FID and a GS-GasPro

capillary column (Agilent, 30 m  0.25 mm), as well as TCD and a Carboxen-1000 column (Supelco, 1.5 m  1/8 in.). The single-step hydroconversion of palm oil (Sigma-Aldrich, analytical standard) was carried out similarly over 1.2 g of the sieved catalysts in the temperature range of 523–573 K under a total pressure of 5 MPa. The H2 flow rate was fixed at 232 cm3 min
1, and the palm oil was introduced using an HPLC pump at a rate of 2.4 g h
1 (WHSV: 2.0 h
1). All tubing between the palm oil container and the reactor (including a pump head) was thoroughly heated at 333 K to inhibit the solidification of palm oil (m.p.: 308 K). The liquid products were collected by a liquid trap gently heated at 353 K to inhibit the solidification of unconverted palm oil and fatty acids. The liquid hydrocarbon products were separated from condensed water by decantation. Afterward, they were analyzed using an offline GC equipped with an FID and a DB-5 column (Agilent, 30 m  0.32 mm  0.25 lm) after dissolution in chloroform and addition of n-eicosane (Aldrich, 99%) as an external standard. The product yields (wt%) were calculated on the basis of the weight of palm oil. The gas products were analyzed using an online GC system equipped with an FID and a GSGasPro capillary column (Agilent, 30 m  0.25 mm), as well as a TCD and a Carboxen-1000 column (Supelco, 1.5 m  1/8 in.). After the reaction, the reactor was purged with He (100 cm3 min
1) for 4 h at the reaction temperature and then cooled to room temperature to collect the used catalysts. The amount of coke in the used catalysts was analyzed by thermogravimetric analysis (TGA) using a TGA N-1000 instrument (Thermo). Temperature was increased from room temperature to 1073 K (ramp: 10 K min
1) under dry air flow. Pyridine-IR and HAADF-STEM investigations were also carried out with the used catalysts. 2.4. Separation of biojet fuel by distillation, and characterization For producing biojet fuel, a single-step hydroconversion of palm oil was carried out at 568 K under 5 MPa (WHSV: 2.0 h
1) by using PtRe/USY as a catalyst. Jet fuel-ranged hydrocarbons (b.p.: 416– 545 K) were fractionated by distillation using a batch distillation system equipped with a Vigreux distillation column. The distribution of hydrocarbons in the fractionated biojet fuel was analyzed using a 2D gas chromatograph with a time-of-flight mass spectrometer (GC/TOF-MS). A Pegasus 4D instrument (LECO Corp., St. Joseph, MI, USA) equipped with an Agilent 7890A GC, an RtxDHA-50 column (50 m  0.20 mm  0.50 lm, dimethyl polysiloxane), and a DB-17 column (1.5 m  0.15 mm  0.15 lm, diphenyl polysiloxane) was used for the analysis. The following properties of the biojet fuel were investigated: fuel distillation range (ASTM D2887), freezing point (ASTM D7153), flash point (ASTM D56), density (ASTM D4052), sulfur content (ASTM D4294), aromatic content (ASTM D6379), and net heat of combustion (ASTM D4809). 3. Results and discussion 3.1. Characterization of catalysts Two different metal–acid bifunctional catalysts, i.e., Pt/USY and PtRe/USY, were synthesized. Elemental analysis using ICP-AES showed that Pt/USY contained 2.0 wt% Pt, whereas PtRe/USY contained 2.0 wt% Pt and 2.1 wt% Re. HAADF-STEM images (Fig. 1a and b) showed that both Pt/USY and PtRe/USY contained mainly the metal particles having 1–2 nm diameters. The surface-areaweighted mean particle diameters of Pt/USY and PtRe/USY were determined to be 2.04 and 1.83 nm, respectively. In the XRD patterns of both samples (Fig. S1), only the characteristic peaks for the FAU zeolite structure were observed and no peak for the metal species was detected, confirming the high dispersion of the metal

183

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

pre-treatment in humid H2, while H2 chemisorption was much less affected. In result, PtRe/USY exhibited even lower CO chemisorption than Pt/USY. The result implied that Pt decorated with oxidized Re species can show significantly suppressed CO binding. Similar conclusions were derived with the PtRe catalysts prepared for the aqueous phase reforming (APR) [36,42]. In the case of Pt/ USY, both the H2 and CO chemisorption amounts decreased only slightly after pre-reduction under humid H2 atmosphere. To gain additional insight into the chemical state of the metal catalysts supported on USY, Pt LIII and Re LIII XAS analyses were carried out after the pre-reduction under dry or humid H2 atmospheres. In Fig. 2a and b, the oxidation states of Pt and Re of the catalysts were analyzed using the energy shift of absorption edge in the XANES (Fig. S2) [36,43,44]. The oxidation states of Pt in Pt/ USY and PtRe/USY were determined to be 1.0 and 1.3, respectively, after pre-reduction under dry H2 (Fig. 2a). After pre-reduction under humid H2, the oxidation states of Pt were only slightly increased to 1.4 in both catalysts. These results indicated that Pt species were always highly reduced and their oxidation state was rather insensitive to the presence of H2O vapor. In contrast, the

Fig. 1. HAADF-STEM images and particle size distributions of (a) Pt/USY and (b) PtRe/USY. (c and d) EDS elemental mapping of the PtRe/USY catalyst.

species. Elemental mapping using EDS indicated the uniform distribution of Pt and Re species in the PtRe/USY catalyst (Fig. 1c and d). No appreciable separate zoning of Pt and Re species was observed within the resolution of our investigation, indicating the formation of uniform bimetallic particles. The physical properties of the catalysts are summarized in Table 1. The number of Brønsted and Lewis acid sites in both Pt/USY and PtRe/USY samples decreased as compared to the USY zeolite. The results indicated that the metal particles were highly dispersed over the USY support through the strong interaction with acid sites [38,39]. After pre-reduction under dry H2 at 573 K, PtRe/USY showed increased CO chemisorption but decreased H2 chemisorption as compared to Pt/USY (Table 1). Earlier studies showed that reduced Re species can chemisorb CO, but not H2 [36,40,41]. Therefore, the lower H2 chemisorption of PtRe/USY than that of Pt/USY indicated that Pt and Re formed well-alloyed particles in PtRe/USY [40]. It was reported that Re species can be readily oxidized by H2O in the reaction medium [36,37,42], which can markedly alter their chemisorption properties. Because H2O can be generated in situ during the deoxygenation of triglycerides, the chemisorption behaviors of the catalysts were additionally investigated after pre-treatment in humid H2 atmosphere containing 3 kPa H2O (the values in parenthesis, Table 1). The results showed that CO chemisorption of PtRe/USY was significantly decreased after the

Fig. 2. Oxidation states of (a) Pt and (b) Re in Pt/USY and PtRe/USY catalysts pretreated under dry or humid H2 atmospheres. The oxidation states were calculated using the energy shift (E
E0) of Pt LIII and Re LIII absorption edges in XANES. (c) Pt LIII and (d) Re LIII k3-weighted Fourier transforms of the EXAFS of Pt/ USY and PtRe/USY pretreated under dry or humid H2 atmospheres (solid line: experimental data; dotted line: fitted data).

Table 1 Physical properties of the catalysts.

a b c d

Sample

SBETa (m2 g
1)

nH2b (lmol g
1)

nCOb (lmol g
1)

nBASd (lmol g
1)

nLASd (lmol g
1)

USY Pt/USY PtRe/USY

836 821 797

– 52 (46)c 45 (37)c

– 75 (65)c 100 (41)c

215 149 117

95 67 68

BET surface area determined by N2 adsorption at 77 K. H2 and CO chemisorption after pre-reduction under dry H2. H2 and CO chemisorption after pre-reduction under humid H2. Amounts of Brønsted and Lewis acid sites determined by pyridine-IR.

184

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

oxidation state of Re in PtRe/USY was significantly affected by the presence of H2O during pre-reduction (Fig. 2b). The oxidation state of Re was determined to be 1.6 after pre-reduction under dry H2; however, it significantly increased to 4.5 after the pre-reduction under humid H2. The Fourier transforms of k3-weighted EXAFS are shown in Fig. 2c and d, and the fitting results are summarized in Table 2. In the analysis of PtRe/USY, we denoted the scattering paths simply as Pt–M or Re–M (M = Pt, Re) because of the highly similar backscattering functions of Pt and Re [45]. After pre-reduction under dry H2, the EXAFS of all the catalysts could be well fitted by including only the Pt–M and Re–M scattering paths, indicating the negligible presence of Pt–O and Re–O bonds. After prereduction under humid H2 atmosphere, however, the EXAFS of PtRe/USY changed markedly, and it could be fitted successfully only after the inclusion of a Re–O scattering path. The average Re–O coordination number was determined to be 1.8, indicating that Re was substantially oxidized after pre-reduction under humid H2. These results were consistent with the XANES results. 3.2. Understanding the direct hydroconversion of triglyceride into biojet fuel Scheme 2. Hydrocracking in the (a) ideal regime and (b) non-ideal regime.

As shown in Scheme 1a, the direct hydroconversion of triglycerides into biojet fuel can be mainly considered as the hydrocracking of long-chain hydrocarbons (C15–C18) produced by the deoxygenation of fatty acid units of triglycerides. Therefore, to design the advanced hydroconversion catalysts for producing biojet fuel, it is essential to understand the reaction mechanism of the hydrocracking. The hydrocracking over metal–acid bifunctional catalysts generally consists of the following steps [32]: (i) dehydrogenation of paraffins into olefins on the metal sites; (ii) diffusion of olefins from the metal sites to the Brønsted acid sites; (iii) formation of carbenium ions on the Brønsted acid sites; (iv) isomerization and cracking (b-scission) of carbenium ions on the Brønsted acid sites; (v) diffusion of the produced olefins to the metal sites; and (vi) re-hydrogenation of olefins to paraffins on the metal sites. In hydrocracking, an optimum balance between the metal (required for hydrogenation/dehydrogenation) and acid (required for isomerization/cracking) functions is crucial for achieving ideal catalytic characteristics [31,46–50]. It is known that the metal function should be dominant over the acid function so that the acid-catalyzed reactions become rate-determining. Under such circumstances, the isomerization and cracking products are sequentially produced (Scheme 2a). Because the olefinic species converted on the acid sites are rapidly re-hydrogenated on the nearby metal sites, coke formation through olefin polymerization and thus catalyst deactivation are significantly suppressed. In contrast, if the acid function dominates the metal function, the olefins reside on the acid sites too long before their re-hydrogenation on the metal sites, which results in overcracking, as well as rapid catalyst deactivation through coke formation (Scheme 2b). Therefore, maintaining the dominance of metal function over acid function is the prerequisite for ideal hydrocracking.

Even though hydrocracking is the main reaction step in the direct hydroconversion of triglyceride into biojet fuel, there is a significant difference as compared to the hydrocracking of conventional hydrocarbons under pure H2 atmosphere. Namely, during the deoxygenation of triglycerides, minute amounts of CO and H2O are co-produced (Scheme 1a), which can poison the catalysts. This can disturb the balance between the metal and acid functions, causing the hydrocracking reaction to occur in a non-ideal regime. This means that a bifunctional catalyst having tolerance to these molecules should be developed for the successful direct hydroconversion of triglyceride into biojet fuel. Therefore, in the following sections, we first investigated the performances of Pt/USY and PtRe/USY catalysts in hydrocracking, with and without the coinjection of H2O, CO, or H2O/CO into the H2 flow. 3.3. Effects of H2O and CO on the hydrocracking of n-hexadecane over Pt/USY In Fig. S3–S6, the product distributions obtained during the hydrocracking of n-hexadecane (C16) over Pt/USY at different conditions are provided. The summary of the reaction results and representative data are shown in Fig. 3. Compared with the hydrocracking under pure H2, the reaction under H2O/H2 atmosphere exhibited reduced cracking activity and required ~20 K higher temperature for achieving a similar degree of C16 cracking (Fig. 3a). This can be understood by the fact that H2O is strongly adsorbed on the Brønsted acid sites [51]; thus, it decreases the rate of acid-catalyzed reactions (isomerization/ cracking), which are the rate-determining steps in ideal

Table 2 EXAFS fitting results of Pt/USY and PtRe/USY. Sample

Treatment

Shell

CN

Pt/USY

dry H2 humid H2 dry H2

Pt-Pt Pt-Pt Pt-M Re-M Pt-M Re-M Re-O

5.2 5.6 5.3 5.0 6.7 4.6 1.8

PtRe/USY

humid H2

(0.3) (0.5) (0.4) (0.4) (0.6) (0.5) (0.2)

R (Å)

r2 (Å2)

2.75 2.74 2.72 2.59 2.74 2.71 1.77

0.006 0.005 0.008 0.012 0.006 0.009 0.006

(0.01) (0.03) (0.01) (0.02) (0.04) (0.02) (0.01)

(0.001) (0.002) (0.001) (0.002) (0.001) (0.002) (0.001)

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

185

Fig. 3. (a) Cracking % of n-hexadecane (C16) as a function of temperature over the Pt/USY catalyst. (b) Yield and (c) iso/n ratio of C8 cracking products as a function of C16 cracking% measured at the initial reaction period (4 h). (d) C16 cracking % as a function of time-on-stream, and (e–h) initial product distributions in C16 hydrocracking (e) under pure H2 at 563 K, (f) under H2O/H2 at 583 K, (g) under CO/H2 at 543 K, and (h) under H2O/CO/H2 at 578 K.

hydrocracking. At similar C16 cracking degrees, however, both conditions gave almost similar product distributions. Even at large cracking degrees (97–98% C16 cracking), the products exhibited quite symmetric distributions centered at C8 (Fig. 3e–f), indicating suppressed overcracking [32,52]. Consequently, fairly high maximum yields of C8 hydrocarbons (48–51 wt%, Fig. 3b) with high iso/n ratios (6.0–7.5, Fig. 3c) could be produced under both atmospheres, which was promising for the jet fuel production. As shown in Fig. 3d, catalyst deactivation was not observed up to 48 h under both H2 and H2O/H2 atmospheres. Thermogravimetric analysis of the used catalysts in air showed insubstantial coke formation (<0.005 gcoke g
1 cat.). Such insignificant overcracking and catalyst deactivation supported that the hydrocracking reaction occurred ideally. The reaction under CO/H2 required a lower temperature for achieving a similar degree of C16 cracking (Fig. 3a), compared to the reaction under pure H2. At a 95% C16 cracking degree, very light cracking products (C4–C7) were majorly produced (Fig. 3g), indicating substantial overcracking. Consequently, the maximum C8 yield (30 wt%) was substantially lower than those (48–51 wt%) achievable under H2 and H2O/H2 atmospheres (Fig. 3b). Such overcracking could be attributed to the CO poisoning of Pt sites, which caused the long residence of the olefins on the acid sites before their re-hydrogenation. The analysis of the gas products revealed that the conversion of the initially injected CO to methane (C1) was minor (16%) under CO/H2 atmosphere, indicating the low methanation activity of Pt/USY. Similarly, the reaction under H2O/CO/H2 also resulted in substantial overcracking (Fig. 3h). The production of light products (C4–C7) was even more pronounced than the reaction under CO/H2. For achieving a similar degree of C16 cracking, the reaction under H2O/CO/H2 required higher temperature than that under CO/H2 (Fig. 3a). Again, this can be attributed to the poisoning of acid sites by H2O. Notably, the conversion of CO to C1 was almost completely suppressed (<1%) under H2O/ CO/H2, indicating that CO methanation became even more difficult on the Pt surface in the presence of H2O. The reactions under CO/H2 and H2O/CO/H2 atmospheres showed rapid catalyst deactivation within 48 h (Fig. 3d), wherein

the latter showed faster deactivation. The analysis of the used catalysts indicated 0.087 and 0.171 gcoke g
1 cat. coke deposition, respectively, which were much larger than the values obtained under H2 and H2O/H2 atmospheres (<0.005 gcoke g
1 cat.). The significant overcracking and rapid coke formation under CO-containing atmospheres supported that the poisoning of Pt by CO caused the hydrocracking to occur in a non-ideal regime (i.e., the acid function became dominant over the metal function, Scheme 2b). Notably, the more significant overcracking and coke formation under H2O/ CO/H2 than under CO/H2 indicated that the co-presence of H2O and CO led to more significant Pt poisoning than the sole presence of CO. This will be explained in the following section with CO DRIFTS data. 3.4. Effects of H2O and CO on the hydrocracking of n-hexadecane over PtRe/USY The product distributions obtained during C16 hydrocracking over PtRe/USY at different reaction conditions are provided in Fig. S7–S10. The summary of the reaction results and representative reaction data are shown in Fig. 4. As shown in Fig. 4a, the cracking activity decreased gradually (or higher temperature was required for achieving a similar cracking degree) as the H2, CO/ H2, H2O/H2, and H2O/CO/H2 atmospheres were used. Meanwhile, the maximum achievable C8 yields (Fig. 4b) and iso/n paraffin ratios (Fig. 4c) increased. These results indicated that the addition of H2O and CO decreased the overall cracking activity but increased the selectivity toward the desired products (i.e., iso-rich C8 hydrocarbons). This trend was markedly different from that of the Pt/USY catalyst. The product distribution obtained over PtRe/USY under pure H2 (95% C16 cracking degree) is shown in Fig. 4e. The most distinct feature was the significant formation of C1 (15.5 wt%) and C2 (3.4 wt%) and low iso-paraffin yields, which were significantly different from the reaction with Pt/USY (Fig. 3e). Propylene is the smallest product that can be formed through the b-scission of carbenium ions [32]; consequently, hydrocarbons smaller than C3 cannot be produced according to the conventional acid-catalyzed cracking

186

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

Fig. 4. (a) Cracking % of n-hexadecane (C16) as a function of temperature over the PtRe/USY catalyst. (b) Yield and (c) iso/n ratio of C8 cracking products as a function of the C16 cracking % measured at the initial reaction period (4 h). (d) C16 cracking % as a function of time-on-stream, and (e–h) initial product distributions in C16 hydrocracking (e) under pure H2 at 533 K, (f) under H2O/H2 at 568 K, (g) under CO/H2 at 543 K, and (h) under H2O/CO/H2 at 573 K.

mechanism. Therefore, the result indicated that cracking occurred not only through the metal–acid bifunctional mechanism involving the isomerization and subsequent b-scission of carbenium ions, but also through the direct hydrogenolysis (i.e., C–C bond cleavage with the addition of H) on the metal surface. It has been reported that reduced Re has significantly higher C–C hydrogenolysis activity than Pt [53,54]. Such hydrogenolysis is not a desirable reaction pathway in hydrocracking because it can enhance the formation of low-value light hydrocarbons. Notably, such undesired hydrogenolysis was suppressed by the injection of H2O (Fig. 4f) or CO (Fig. 4g) into the H2 stream, wherein the former was more effective. The co-injection of H2O and CO (Fig. 4h) was most effective in suppressing the hydrogenolysis. Therefore, under H2O/CO/H2 atmosphere, only a small amount of C1 (4.5 wt%) was produced (Fig. 4h). Under this condition, more than half of the C1 production could be attributed to the complete methanation of the initially injected CO (theoretical C1 yield through CO methanation is 2.5 wt%). It is notable that the initially injected CO was not detected under both CO/H2 and H2O/CO/H2 atmospheres, indicating the high methanation activity of PtRe/ USY. The suppression of hydrogenolysis should be why PtRe/USY exhibited increased maximum C8 yields (Fig. 4b) and significantly enhanced iso/n paraffin ratios (Fig. 4c) under H2O/CO/H2, compared to the reactions under other atmospheres. The suppression of hydrogenolysis by the presence of H2O and CO can be explained by the structural sensitivity of this reaction. It is known that coordinatively unsaturated metal sites (e.g., kink) are active for hydrogenolysis [55,56], and these sites can be preferentially blocked even by the minute presence of poisons [57–59]. Even in the presence of CO (i.e., under CO/H2 and H2O/CO/H2 atmospheres), the PtRe/USY catalyst showed cracking products having a quite symmetric distribution centered at C8, except the C1 produced through hydrogenolysis and CO methanation (Fig. 4g–h). This is in clear contrast to the reactions with Pt/USY (Fig. 3g–h), which showed significant overcracking (i.e., major production of C4–C7 light hydrocarbons) in the presence of CO. In addition, the PtRe/USY catalyst showed inappreciable deactivation

under both CO/H2 and H2O/CO/H2 atmospheres (Fig. 4d), in contrast to the reactions with Pt/USY (Fig. 3d). The used PtRe/USY catalysts showed negligible coke formation (<0.005 gcoke g
1 cat.), regardless of the gas atmospheres used for the reaction. These results strongly indicated that the C16 hydrocracking occurred over PtRe/USY in an ideal regime even under CO-containing atmospheres. The remarkable CO-tolerance of PtRe/USY was attributed to its high methanation activity, which could rapidly convert the chemisorbed CO into C1 and H2O. Methane (C1) is an inert species toward both of metal and acid functions, while H2O moderately poisons acid function. Therefore, PtRe/USY can maintain the dominance of metal function over acid function even in the CO-containing atmospheres (i.e., an ideal hydrocracking regime). It has been reported that the rate-limiting step of methanation is the C–O bond cleavage of the enol-like CHOH surface intermediates [60]. Earlier kinetic analysis showed that Pt has low methanation activity compared to other group VIII metals because of its excessively strong binding energy to CHOH and CO. This leads to very high surface coverage by these species, which limits the activation of H2 on the Pt surface [60]. On the other hand, it was reported that alloying Pt with oxophilic metal species can accelerate methanation by facilitating the C–O cleavage of the surface intermediate species [61–63]. Because Re is a representative oxophilic metal species, its close proximity to Pt might enhance the C–O cleavage of the surface intermediates, leading to the significantly increased methanation activity. Furthermore, as shown in our chemisorption data (Table 1), the decoration of Pt with partly oxidized Re could markedly suppress CO adsorption (or poisoning), which should be advantageous for more efficient H2 activation. To gain additional insight into the superior methanation activity of PtRe/USY to that of Pt/USY, we carried out CO DRIFTS investigations after the pre-reduction of the catalysts under dry or humid H2 atmospheres (Fig. 5). After the adsorption of CO, the spectra were collected while increasing the temperature under a dry H2 flow. In all the spectra, only the peak for linearly bound CO (2000–2100 cm
1) was detected. In the case of Pt/USY, the peak

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

187

Fig. 6. Product yields in the single-step hydroconversion of palm oil as a function of temperature over (a) Pt/USY and (b) PtRe/USY. Yields of hydrocarbons with different chain lengths as a function of temperature over (c) Pt/USY and (d) PtRe/ USY. Fig. 5. CO DRIFTS of (a) Pt/USY and (b) PtRe/USY, measured under H2 flow at 303, 323, 373, 473, and 573 K. Before measuring the CO spectra, the catalysts were prereduced under dry or humid H2 atmosphere at 573 K.

intensity decreased only slightly even after heating at 573 K (Fig. 5a), confirming its low methanation activity. Notably, the reduction of the peak intensity was smaller after the prereduction under humid H2 than under dry H2, indicating more suppressed methanation in the former case. This can be explained by the fact that CO binding on Pt surface becomes stronger in the presence of co-adsorbed H2O, because H2O donates electrons to Pt and this, in turn, enhances the p-backbonding between Pt and CO [64]. Such strengthened CO binding or enhanced CO coverage on Pt surface should result in suppressed H2 dissociation (i.e., low H coverage), which is unfavorable for methanation. This result is consistent with our earlier observation that C16 hydrocracking occurred more non-ideally (i.e., enhanced overcracking and coking) under H2O/CO/H2 than under CO/H2. In contrast, the CO peak intensity of PtRe/USY decreased rapidly upon increasing temperature (Fig. 5b) and completely disappeared at 573 K. The result confirmed the high methanation activity of PtRe/USY. Notably, the catalyst pre-reduced under humid H2 showed smaller and more rapidly decreasing CO peak intensity than that of the catalyst pre-reduced under dry H2. Such a result is consistent with the earlier chemisorption data (Table 1) indicating that the Pt decorated with oxidized Re shows weakened CO binding, which is favorable for methanation. 3.5. Direct hydroconversion of palm oil into biojet fuel The single-step hydroconversion of natural triglycerides (palm oil) into biojet fuel was carried out over the Pt/USY and PtRe/USY catalysts under a fixed space velocity (WHSV: 2.0 h
1). The palm oil is composed of mainly C16 (43 wt%) and C18 (56 wt%) fatty acid units (Table S1) [5]. Palmitic acid with completely saturated backbone was the only detected C16 acid, and oleic acid containing a single C@C bond was detected as a major C18 acid species. The reaction results are summarized in Fig. 6, and detailed hydrocar-

bon distributions obtained at different reaction conditions are provided in Fig. S11–S12. Under the reaction conditions tested, all the detected liquid products including hydrocarbons and oxygenates had fully saturated carbon backbones. The result confirmed that the hydrogenation of C@C bonds occurred prior to other reaction steps (Scheme 1a). Pt/USY exhibited almost complete deoxygenation of the triglyceride into hydrocarbons only above 553 K (Fig. 6a). When the reaction temperature was lower, however, significant amounts of oxygenates (mainly fatty acids + minor amounts of fatty alcohols and esters) were formed. Under the conditions enabling complete deoxygenation (>553 K), ~8 wt% of H2O and ~6 wt% of CO were produced, indicating that deoxygenation occurred through both hydrodeoxygenation and decarbonylation. Only 0.2–1.0 wt% C1 was formed, confirming the low methanation activity of Pt/USY (Fig. S11). On the other hand, PtRe/USY (Fig. 6b) showed much less formation of oxygenates than Pt/USY over the entire temperature range, indicating its higher deoxygenation activity. This can be attributed to the fact that the close proximity of oxophilic Re species to Pt can facilitate the cleavage of the C–O bond of fatty acids [65,66]. Under the conditions enabling complete deoxygenation (>553 K), PtRe/USY produced ~12 wt% of H2O, which was larger than that produced over Pt/USY (~8 wt%). At the same time, no CO was detected and a much larger amount of C1 (3.1–4.5 wt%, Fig. S12) was formed compared with Pt/USY (0.2–1.0 wt%). These results clearly indicate the complete methanation of the produced CO over the PtRe/USY catalyst. At the lowest temperature (523 K) where cracking was insubstantial, C15–C18 paraffins were detected as major hydrocarbon products (Fig. 6c and d) over both catalysts. Palm oil was originally composed of mainly C16 and C18 fatty acids; consequently, the formation of C15 and C17 paraffins could be attributed to decarbonylation (removal of O as CO, see Scheme 1) and/or decarboxylation (removal of O as CO2), whereas the formation of C16 and C18 paraffins could be explained by hydrodeoxygenation (removal of O as H2O). Because CO2 was not detected in the gas products, however, the contribution of the decarboxylation pathway could be ruled out under the present reaction conditions. As the reaction

188

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

temperature increased above 543 K, the yield of cracking products (C14) rapidly increased over both catalysts. Pt/USY produced significantly larger amounts of light 553 K). These results indicated significant overcracking over Pt/USY. In contrast, PtRe/USY produced more of C8–C14 hydrocarbons than
catalyst showed insignificant change of the product distributions (Fig. 7f) and negligible coke formation after the reaction (<0.005 gcoke g
1 cat.). Only the yields of undesired C1 and C2 were slightly decreased (compare Fig. 7b and d), which, in turn, resulted in increased yields of the C8–C14 cracking products (Fig. 7f). This could be attributed to the preferential deactivation of the active sites for hydrogenolysis (i.e., coordinatively unsaturated sites) even at very minor coke deposition [58]. Such a trend is actually desirable for the biojet fuel production. According to the pyridine-IR analysis, the decrease in the amount of acid sites was insubstantial (Table S2). In addition, no sintering of PtRe bimetallic particles was observed according to the HAADF-STEM investigation (Fig. S13b). The significant formation of light cracking products and rapid catalyst deactivation of Pt/USY indicated that hydrocracking occurred in a non-ideal regime during the direct hydroconversion of palm oil into biojet fuel. As demonstrated in Section 3.3, this can be explained by the strong poisoning of Pt sites by CO, which was produced by the decarbonylation of fatty acid units. In contrast, the suppressed overcracking and inappreciable deactivation of PtRe/USY indicated that hydrocracking occurred in an ideal regime. As explained in Section 3.4, this can be attributed to the high methanation activity of this catalyst, which can rapidly convert CO into much less detrimental C1 and H2O. It is interesting to remind that PtRe/USY is intrinsically a much poorer catalyst than Pt/USY for the ordinary hydrocracking of paraffins under pure H2 atmosphere. This is due to its high hydrogenolysis activity, which produces low-value light hydrocarbons. However, in the direct hydroconversion of triglycerides, the H2O and CO species produced in situ during deoxygenation could selectively poison the active sites for undesired hydrogenolysis, thereby making PtRe/USY a highly stable and selective catalyst for producing biojet fuel. 3.6. Yield and fuel specifications of the fractionated biojet fuel The liquid product obtained through the hydroconversion of palm oil over PtRe/USY at 568 K cannot directly satisfy all the specifications required for biojet fuel. The presence of light hydrocarbons (C7) reduces the flash point, and that of heavy hydrocarbons (C17) increases the freezing point out of the target fuel specifications. Therefore, we carried out distillation to fractionate the jet fuel-ranged hydrocarbons (b.p.: 416–545 K). The resultant hydrocarbon distribution is shown in Fig. 8. The results showed that the light (C7) and heavy hydrocarbons (C17) were almost completely removed. The iso/n paraffin ratio of the products was determined to be 5.1. The physical properties of the fractionated biojet fuel are summarized in Table 3, along with the required

Fig. 7. Hydrocarbon distributions obtained after the single-step hydroconversion of palm oil at 4 h over (a) Pt/USY and (b) PtRe/USY, and at 48 h over (c) Pt/USY and (d) PtRe/USY. Product yields plotted as a function of time-on-stream over (e) Pt/USY and (f) PtRe/USY (reaction temperature: 568 K).

Fig. 8. Hydrocarbon distribution of the final biojet fuel obtained after distillation.

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190 Table 3 Properties of biojet fuel obtained after distillation.

189

References

Property

Standard

Biojet fuel

Distillation temperature (K) 10% recovered 50% recovered 90% recovered Final boiling point (K) Freezing point (K) Flash point (K) Density at 288 K (kg m
3) Sulfur (mg kg
1) Aromatics (wt%) Net heat of combustion (MJ kg
1)

Max. 478 Report Report Max. 573 Max. 233 Min. 311 730–770 Max. 15 Max. 0.5 42.8

433 460 522 545 231 312 756 Not detected Not detected 46.9

fuel specifications. The final product satisfied all required fuel specifications [25], including boiling point, freezing point, flash point, density, aromatics and sulfur contents, and net heat of combustion. The final biojet fuel yield was calculated to be 41 wt% with respect to palm oil. 4. Conclusions We successfully produced high-quality biojet fuel through a single-step direct hydroconversion of triglycerides. The key idea was to design a CO-tolerant metal–acid bifunctional catalyst, which can carry out hydrocracking in an ideal regime even in the presence of CO produced by the deoxygenation of triglycerides. Such a catalyst was prepared by supporting bimetallic PtRe particles on USY as acidic support. Compared with a conventional monometallic Pt/USY catalyst, PtRe/USY showed markedly enhanced tolerance to CO because it had weakened interaction with CO and also could rapidly convert the chemisorbed CO into less harmful methane and H2O through methanation. Consequently, the PtRe/USY catalyst showed much suppressed overcracking and negligible catalyst deactivation through coke formation as compared to Pt/USY, which was strongly poisoned by CO. It is remarkable that PtRe/USY was intrinsically a much poorer hydrocracking catalyst than Pt/USY under pure H2 atmosphere because of its high hydrogenolysis activity. However, the small amounts of CO and H2O produced in situ during the deoxygenation of triglycerides selectively poisoned the active sites for undesired hydrogenolysis, thereby making PtRe/USY a highly stable and selective catalyst for the direct hydroconversion of triglycerides into biojet fuel. In the present study, we used the CO-tolerant PtRe/USY catalyst for simplifying the multi-step hydroconversion processes of biomass (triglycerides) into a single-step process. It is likely that a similar strategy can be applied to other catalytic processes involving cascade reactions, wherein CO is involved as a reaction intermediate. Acknowledgements This research was supported by the Advanced Biomass R&D Center (ABC) of Global Frontier Project funded by the Ministry of Science and ICT (ABC-2015M3A6A2066121), and Basic Science Research Program through the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (NRF2017R1A2B2002346). We acknowledge the Pohang Accelerator Laboratory (PAL) for the use of beamline. Appendix A. Supplementary material Supplementary data to this article can be found online at https://doi.org/10.1016/j.jcat.2019.09.043.

[1] F. Rosillo-Calle, S. Teelucksingh, D. Thrän, M. Seiffert, The potential and role of biofuels in commercial air transport – biojetfuel, IEA Bioenergy (2012). [2] IATA. Halving emissions by 2050 – Aviation brings its targets to Copenhagen, 2009 . [3] W.-C. Wang, L. Tao, Bio-jet fuel conversion technologies, Renew. Sustain. Energy Rev. 53 (2016) 801–822. [4] M.J. McCall, J.A. Kocal, A. Bhattacharyya, T.N. Kalnes, T.A. Brandvold, Production of aviation fuel from renewable feedstocks, US Patent 8039682B2, 2011. [5] M.Y. Kim, J.-K. Kim, M.-E. Lee, S. Lee, M. Choi, Maximizing biojet fuel production from triglyceride: Importance of the hydrocracking catalyst and separate deoxygenation/hydrocracking steps, ACS Catal. 7 (2017) 6256–6267. [6] H.J. Robota, J.C. Alger, L. Shafer, Converting algal triglycerides to diesel and HEFA jet fuel fractions, Energy Fuels 27 (2013) 985–996. [7] D. Verma, R. Kumar, B.S. Rana, A.K. Sinha, Aviation fuel production from lipids by a single-step route using hierarchical mesoporous zeolites, Energy Environ. Sci. 4 (2011) 1667–1671. [8] S. Liu, Q. Zhu, Q. Guan, L. He, W. Li, Bio-aviation fuel production from hydroprocessing castor oil promoted by the nickel-based bifunctional catalysts, Bioresour. Technol. 183 (2015) 93–100. [9] T. Li, J. Cheng, R. Huang, J. Zhou, K. Cen, Conversion of waste cooking oil to jet biofuel with nickel-based mesoporous zeolite Y catalyst, Bioresour. Technol. 197 (2015) 289–294. [10] R.J. Wittenbrink, P.J. Berlowitz, B.R. Cook, Synthetic jet fuel and process for its production, US Patent 6309432B1, 2001. [11] D. Lamprecht, Fischer
Tropsch fuel for use by the U.S. military as battlefielduse fuel of the future, Energy Fuels 21 (2007) 1448–1453. [12] Q. Yan, F. Yu, J. Liu, J. Street, J. Gao, Z. Cai, J. Zhang, Catalytic conversion wood syngas to synthetic aviation turbine fuels over a multifunctional catalyst, Bioresour. Technol. 127 (2013) 281–290. [13] B.G. Harvey, R.L. Quintana, Synthesis of renewable jet and diesel fuels from 2ethyl-1-hexene, Energy Environ. Sci. 3 (2010) 352–357. [14] M.W. Peters, J.D. Taylor, Renewable jet fuel blendstock from isobutanol, US Patent 8373012B2, 2013. [15] P.R. Gruber, M.W. Peters, J.M. Griffith, Y.A. Obaidi, L.E. Manzer, J.D. Taylor, D.E. Henton, Renewable compositions, US Patent 8378160B2, 2013. [16] D.J.L. Prak, M.H. Jones, P. Trulove, A.M. McDaniel, T. Dickerson, J.S. Cowart, Physical and chemical analysis of alcohol-to-jet (ATJ) fuel and development of surrogate fuel mixtures, Energy Fuels 29 (2015) 3760–3769. [17] M.E. Wright, Process for the dehydration of aqueous bio-derived terminal alcohols to terminal alkenes. US Patent 9242226B2, 2016. [18] R. Xing, A.V. Subrahmanyam, H. Olcay, W. Qi, G.P. van Walsum, H. Pendse, G. W. Huber, Production of jet and diesel fuel range alkanes from waste hemicellulose-derived aqueous solutions, Green Chem. 12 (2010) 1933–1946. [19] H. Olcay, A.V. Subrahmanyam, R. Xing, J. Lajoie, J.A. Dumesic, G.W. Huber, Production of renewable petroleum refinery diesel and jet fuel feedstocks from hemicellulose sugar streams, Energy Environ. Sci. 6 (2013) 205–216. [20] L. Rye, S. Blakey, C.W. Wilson, Sustainability of supply or the planet: a review of potential drop-in alternative aviation fuels, Energy Environ. Sci. 3 (2010) 17–27. [21] D. Chiaramonti, M. Prussi, M. Buffi, D. Tacconi, Sustainable bio kerosene: Process routes and industrial demonstration activities in aviation biofuels, Appl. Energy 136 (2014) 767–774. [22] R. Mawhood, E. Gazis, S. de Jong, R. Hoefnagels, R. Slade, Production pathways for renewable jet fuel: a review of commercialization status and future prospects, Biofuels Bioprod. Bioref. 10 (2016) 462–484. [23] M.N. Pearlson, A techno-economic and environmental assessment of hydroprocessed renewable distillate fuels, Master thesis, Massachusetts Institute of Technology, 2011. [24] W.-C. Wang, L. Tao, J. Markham, Y. Zhang, E. Tan, L. Batan, E. Warner, M. Biddy, Review of biojet fuel conversion technologies, Task No. BB14.4420, NREL, 2016. [25] American Society for Testing and Materials, ASTM D7566-19, Standard specification for aviation turbine fuel containing synthesized hydrocarbons, 2019. [26] P.T. Do, M. Chiappero, L.L. Lobban, D.E. Resasco, Catalytic deoxygenation of methyl-octanoate and methyl-stearate on Pt/Al2O3, Catal. Lett. 130 (2009) 9– 18. [27] T. Morgan, D. Grubb, E. Santillan-Jimenez, M. Crocker, Conversion of triglycerides to hydrocarbons over supported metal catalysts, Top. Catal. 53 (2010) 820–829. [28] B. Peng, Y. Yao, C. Zhao, J.A. Lercher, Towards quantitative conversion of microalgae oil to diesel-range alkanes with bifunctional catalysts, Angew. Chem. Int. Ed. 51 (2012) 2072–2075. [29] D. Kubicˇka, L. Kaluzˇa, Deoxygenation of vegetable oils over sulfided Ni, Mo and NiMo catalysts, Appl. Catal. A 372 (2010) 199–208. [30] O.I_. Sßenol, T.-R. Viljava, A.O.I. Krause,, Hydrodeoxygenation of aliphatic esters on sulphided NiMo/c-Al2O3 and CoMo/c-Al2O3 catalyst: the effect of water, Catal. Today 106 (2005) 186–189. [31] M. Guisnet, F. Alvarez, G. Giannetto, G. Perot, Hydroisomerization and hydrocracking of n-heptane on PtH zeolites. Effect of the porosity and of the distribution of metallic and acid sites, Catal. Today 1 (1987) 415–433.

190

K. Lee et al. / Journal of Catalysis 379 (2019) 180–190

[32] J. Weitkamp, Catalytic hydrocracking—mechanisms and versatility of the process, ChemCatChem 4 (2012) 292–306. [33] C.A. Emeis, Determination of integrated molar extinction coefficients for infrared absorption bands of pyridine adsorbed on solid acid catalysts, J. Catal. 141 (1993) 347–354. [34] M. Schneider, D.G. Duff, T. Mallat, M. Wildberger, A. Baiker, High surface area platinum-titania aerogels: Preparation, structural properties, and hydrogenation activity, J. Catal. 147 (1994) 500–514. [35] D.C. Koningsberger, B.L. Mojet, G.E. van Dorssen, D.E. Ramaker, XAFS spectroscopy; fundamental principles and data analysis, Top. Catal. 10 (2000) 143–155. [36] E.L. Kunkes, D.A. Simonetti, J.A. Dumesic, W.D. Pyrz, L.E. Murillo, J.G. Chen, D.J. Buttrey, The role of rhenium in the conversion of glycerol to synthesis gas over carbon supported platinum-rhenium catalysts, J. Catal. 260 (2008) 164–177. [37] A. Ciftci, D.A.J.M. Ligthart, A.O. Sen, A.J.F. van Hoof, H. Friedrich, E.J.M. Hensen, Pt-Re synergy in aqueous-phase reforming of glycerol and the water-gas shift reaction, J. Catal. 311 (2014) 88–101. [38] A. Corma, A. Martínez, V. Martínez-Soria, J.B. Montón, Hydrocracking of vacuum gasoil on the novel mesoporous MCM-41 aluminosilicate catalyst, J. Catal. 153 (1995) 25–31. [39] Y. Wu, S. Holdren, Y. Zhang, S.C. Oh, D.T. Tran, L. Emdadi, Z. Lu, M. Wang, T.J. Woehl, M. Zachariah, Y. Lei, D. Liu, Quantification of rhenium oxide dispersion on zeolite: effect of zeolite acidity and mesoporosity, J. Catal. 372 (2019) 128– 141. [40] D.A. Simonetti, E.L. Kunkes, J.A. Dumesic, Gas-phase conversion of glycerol to synthesis gas over carbon-supported platinum and platinum–rhenium catalysts, J. Catal. 247 (2007) 298–306. [41] C. Bolivar, H. Charcosset, R. Frety, M. Primet, L. Tournayan, C. Betizeau, G. LeClercq, R. Maurel, Platinum-rhenium-alumina catalysts: II. Study of the metallic phase after reduction, J. Catal. 45 (1976) 163–178. [42] Z. Wei, A. Karim, Y. Li, Y. Wang, Elucidation of the roles of Re in aqueous-phase reforming of glycerol over Pt
Re/C catalysts, ACS Catal. 5 (2015) 7312–7320. [43] F. Hilbrig, C. Michel, G.L. Haller, A XANES-TPR study of PtRe/Al2O3 catalysts, J. Phys. Chem. 96 (1992) 9893–9899. [44] J.G. Chen, C.M. Kim, B. Frühberger, B.D. DeVries, M.S. Touvelle, A NEXAFS determination of the oxidation state of vanadium carbide on V(110): Observation of charge transfer from vanadium to carbon, Surf. Sci. 321 (1994) 145–155. [45] O.M. Daniel, A. DeLaRiva, E.L. Kunkes, A.K. Datye, J.A. Dumesic, R.J. Davis, X-ray absorption spectroscopy of bimetallic Pt–Re catalysts for hydrogenolysis of glycerol to propanediols, ChemCatChem 2 (2010) 1107–1114. [46] H. Pichler, H. Schulz, H.O. Reitemeyer, J. Weitkamp, Uber das hydrocracken gesatterigter kohlenwasserstoffe Erdöl Kohle, Erdgas, Petrochem 25 (1972) 494–505. [47] J. Weitkamp, Hydrocracken, cracken und isomerisieren von kohlenwasserstoffen, Erdöl Kohle, Erdgas, Petrochem 31 (1978) 13–22. [48] G.E. Giannetto, G.R. Perot, M.R. Guisnet, Hydroisomerization and hydrocracking of n-alkanes. 1. Ideal hydroisomerization PtHY catalysts, Ind. Eng. Chem. Prod. Res. Dev. 25 (1986) 481–490.

[49] M. Guisnet, ‘‘Ideal” bifunctional catalysis over Pt-acid zeolites, Catal. Today 218–219 (2013) 123–134. [50] F. Regali, F.L. Liotta, A.M. Venezia, V. Montes, M. Boutonnet, S. Järås, Effect of metal loading on activity, selectivity and deactivation behavior of Pd/silica– alumina catalysts in the hydroconversion of n-hexadecane, Catal. Today 223 (2014) 87–96. [51] A. Jentys, G. Warecka, M. Derewinski, J.A. Lercher, Adsorption of water on ZSM5 zeolites, J. Phys. Chem. 93 (1989) 4837–4843. [52] K.P. de Jong, J. Zecˇevic´, H. Friedrich, P.E. de Jongh, M. Bulut, S. van Donk, R. Kenmogne, A. Finiels, V. Hulea, F. Fajula, Zeolite Y crystals with trimodal porosity as ideal hydrocracking catalysts, Angew. Chem. Int. Ed. 49 (2010) 10074–10078. [53] J.H. Sinfelt, Specificity in catalytic hydrogenolysis by metals, Adv. Catal. 23 (1973) 91–119. [54] J.H. Sinfelt, Catalytic hydrogenolysis on metals, Catal. Lett. 9 (1991) 159–172. [55] D.W. Blakely, G.A. Somorjai, The dehydrogenation and hydrogenolysis of cyclohexane and cyclohexene on stepped (high miller index) platinum surfaces, J. Catal. 42 (1976) 181–196. [56] L.M. Falicov, G.A. Somorjai, Correlation between catalytic activity and bonding and coordination number of atoms and molecules on transition metal surfaces: theory and experimental evidence, Proc. Natl. Acad. Sci. USA 82 (1985) 2207–2211. [57] G. Lietz, J. Völter, Low temperature hydrogenolysis and hydrogenation of alkylbenzenes on platinum catalysts, J. Catal. 45 (1976) 121–127. [58] J. Barbier, P. Marecot, N. Martin, L. Elassal, R. Maruel, Selective poisoning by coke formation on Pt/Al2O3, Stud. Surf. Sci. Catal. 6 (1980) 53–62. [59] P.G. Menon, G.B. Marin, G.F. Froment, Effect of sulfur poisoning on the hydrogenolysis activity of Pt in Pt-Al2O3 catalysts, Ind. Eng. Chem. Prod. Res. Dev. 21 (1982) 52–56. [60] M.A. Vannice, The catalytic synthesis of hydrocarbons from H2/CO mixtures over the group VIII metals: II. The kinetics of the methanation reaction over supported metals, J. Catal. 37 (1975) 462–473. [61] C.M. Tsang, S.M. Augustine, J.B. Butt, W.M.H. Sachtler, Synthesis and characterization of bimetallic clusters prepared by sublimation of Re2(CO)10 onto Pt/NaY, Appl. Catal. 46 (1989) 45–56. [62] L. Guczi, I. Kiricsi, Zeolite supported mono- and bimetallic systems: Structure and performance as CO hydrogenation catalysts, Appl. Catal. A 186 (1999) 375–394. [63] M.A. Vannice, The use of metal-support interactions to selectively activate carbonyl bonds, J. Mol. Catal. 59 (1990) 165–177. [64] G. Rupprechter, T. Dellwig, H. Unterhalt, H.-J. Freund, High-pressure carbon monoxide adsorption on Pt(111) revisited: a sum frequency generation study, J. Phys. Chem. B 105 (2001) 3797–3802. [65] M. Jin, M. Choi, Hydrothermal deoxygenation of triglycerides over carbonsupported bimetallic PtRe catalysts without an external hydrogen source, Mol. Catal. 474 (2019) 110419. [66] H.G. Manyar, C. Paun, R. Pilus, D.W. Rooney, J.M. Thompson, C. Hardacre, Highly selective and efficient hydrogenation of carboxylic acids to alcohols using titania supported Pt catalysts, Chem. Commun. 46 (2010) 6279–6281.