Selective Hydrogenation of Dodecanoic Acid to Dodecane-1-ol Catalyzed by Supported Bimetallic Ni-Sn Alloy

Selective hydrogenation of dodecanoic acid over supported bimetallic Ni-Sn alloy catalysts into dodecane-1-ol is demonstrated. Bimetallic nickel-tin supported on titanium oxide (Ni-Sn(1.5)/TiO2) and gamma-alumina (Ni-Sn(1.5)/-Al2O3); 1.5 = Ni/Sn molar ratio) were synthesized via hydrothermal method in a sealed-Teflon autoclave reactor at 150 oC for 24 h, then followed by reducing with hydrogen gas at 400 oC for 1.5 h. The synthesized catalysts were characterized by means of XRD, IC-AES, N2-adsorption (BET method), H2-chemisorption, and NH3-TPD. Bimetallic Ni-Sn(1.5)/TiO2 catalyst was found to be effective for hydrogenation of dodecanoic acid (>99 % conversion) to dodecane-1-ol (93% yield) at 160 oC, 30 bar H2, and 20 h and the highest dodecane-1-ol (97 % yield) was obtained at initial pressure of H2, 50 bar. An increase of reaction temperature slightly enhanced the degree of hydrodeoxygenation of dodecanoic acid to produce dodecane over both Ni-Sn(1.5)/TiO2 and Ni-Sn(1.5)/-Al2O3 catalysts. Copyright © 2018 BCREC Group. All rights reserved


Introduction
The development of oleochemical products including surfactants, lubricants, plasticizers, coatings, and polymers has been made possible by remarkable advances made in organic syn-thesis, catalysis, and biotechnology [1].Fatty alcohol is one of the important basic oleochemicals which could be produced by hydrogenation of fatty acids or its esters by using heterogeneous catalysts (e.g.bimetallic copper chromite), but the reaction requires harsh temperature (e.g., up to 300 o C) and pressure (e.g., up to 300 bar of H2) in the presence of copper-chromite catalysts, which are potentially harmful for the environ-ment [2,3].In this regard, the development of catalysts to produce fatty alcohols under milder conditions has been long-standing industrial target.
Many previous attempts on the reduction of the reaction pressures (down to 20 bar H2) have been achieved with high conversion and selectivities by replacing Cu or Cr with noble metals such as Ru, Rh, Re, Pd, or Pt [4].For example, the hydrogenation of carboxylic acids to alcohols using a Pt/TiO2 catalyst as reported recently by Hardacre et al. at only 130 °C, 20 bar H2 and gave over 80 % conversion (with 90 % selectivity) after 12 h [5].Although these results are remarkable, it would be desirable to replace the expensive noble metals with more abundant and economic catalysts still capable of working at low H2 pressures and moderate reaction temperatures.The commercial Nibased catalysts such as: RANEY Ni/Kieselgur, or Ni/Al2O3 had been applied for fatty acids or fatty ester hydrogenation with high activity.However, the important drawbacks of these catalysts are low selectivity to desired products (i.e., produced the mixture of cis-and transproduct) and easily deactivated due to carbon deposited on the metal surface [6].
In previous reports, bimetallic Ru-based catalysts have also demonstrated high activity in hydrogenation of methyl oleate at relatively high temperature and H2 pressure (ca.270 ο C and 44 bar, respectively) within 7 h.However, the conversion of methyl oleate and selectivity of oleyl alcohol were only 80 % and 70 %, respectively.The addition of electropositive atom of Sn and B to Ru enhanced the dispersion of Ru and improved the electron density of Ru.The change of electron density Ru enhanced the affinity of Ru towards C=O bond of fatty acids which facilitated the reaction hydrogenation [12,13].Similar results have been also reported on the hydrogenation of dodecanoic acid and palmitic acid at 300 ο C, 60 bar H2, within 4 h over Ru-Sn/Al2O3 and Rh-Sn/Al2O3 catalysts and produced lauryl alcohol and hexadecane-1ol, respectively.On the other hand, Ru-Ge-B/Al2O3 catalyst gave only 20 % of oleyl alcohol with conversion of methyl oleic, ca.80 % [14].
In our previous investigations, we have reported the synthesis of both bulk and supported bimetallic Ni-Sn alloy catalysts from two types nickel precursors: first, from nickel salt (e.g.NiCl2 or NiCl24H2O) produced from both bulk and supported Ni-Sn alloys [15] and, second, from Raney Ni-supported on aluminium hydroxide (R-Ni/AlOH), which produced a nickel-tin alloy supported on aluminium hydroxide (Ni-Sn(x)/AlOH; x = Ni/Sn molar ratio) [16].We have also recently reported the catalytic performance of the Ni-Sn alloy during hydrogenation of biomass-derived levulinic acid in water to -valerolactone (GVL) [17].Over bulk Ni-Sn alloy catalysts, a relatively high reaction temperature (150 o C, 40 bar H2, 6 h) was applied to achieve both a high conversion and GVL yield (99 %) [18].Alternatively, a GVL yield of >99 % was obtained over Ni-Sn(x)/AlOH catalysts at a lower reaction temperature (120 o C) compared to the bulk catalysts [19].
In this present report, we continue our extend investigation of the hydrogenation of dodecanoic acid over supported bimetallic Ni-Sn alloy catalysts to the corresponding of lauryl alcohol.Bimetallic Ni-Sn alloy was supported on two types of supports c.a. titanium oxide (TiO2) and gamma-alumina (-Al2O3).Monometallic nickel supported on TiO2 (Ni/TiO2) and -Al2O3 (Ni/(-Al2O3) catalysts were also evaluate for comparison.

Synthesis of Ni-Sn(1.5)/TiO2
A typical procedure of the synthesis of titanium oxide supported Ni-Sn (1.5 is feeding ratio) alloy catalyst is described as follows [15]: NiCl2•6H2O (7.2 mmol) was dissolved in deionized water (denoted as solution A), and SnCl2•2H2O (4.8 mmol) was dissolved in ethanol/2-methoxy ethanol (2:1) (denoted as solution B) at room temperature.A one-gram titanium oxide (TiO2), solutions A, and B were mixed at room temperature; the temperature was subsequently raised to 50 o C and the mixture was stirred for 12 h.The pH of the mixture was adjusted to 12 through the dropwise addition of an aqueous solution of NaOH (3.1 M).The mixture was then placed into a sealed-Teflon autoclave for the hydrothermal reaction at 150 o C for 24 h.The resulting black precipitate was filtered, washed with distilled water, and then dried under vacuum overnight.Prior to the catalytic reaction, the obtained black powder was treated under hydrogen at 400 o C for 1.5 h.

Catalyst characterization
Powder X-ray diffraction (XRD) measurements were recorded on a Mac Science M18XHF instrument using monochromatic CuKα radiation (l = 0.15418 nm).The XRD was operated at 40 kV and 200 mA with a step width of 0.02 o and a scan speed of 4 o min -1 (α1 = 0.154057 nm, α2 = 0.154433 nm).Inductively coupled plasma-atomic emission spectroscopy (ICP-AES) measurements were performed on an SPS 1800H plasma spectrometer by Seiko Instruments Inc.Japan (Ni: 221.7162 nm and Sn: 189.898 nm).The Brunauer-Emmett-Teller surface area (SBET) and pore volume (V p ) were measured using N2 physisorption at -196 o C on a Belsorp Max (BEL Japan).The samples were degassed at 200 o C for 2 h to remove physisorbed gases prior to the measurement.The amount of nitrogen adsorbed onto the samples was used to calculate the BET surface area via the BET equation.The pore volume was estimated to be the liquid volume of nitrogen at a relative pressure of approximately 0.995 ac-cording to the Barrett-Joyner-Halenda (BJH) approach based on desorption data [20].
The H2 uptake was determined through irreversible H2 chemisorption.After the catalyst was heated at 100 o C under vacuum for 30 min, it was then heated at 400 o C under H2 for 30 min.The catalysts were subsequently cooled to room temperature under vacuum for 30 min.The H2 measurement was conducted at 0 o C, and the H2 uptake was calculated according to a method described in the literature [21,22].
The NH3-TPD was carried out on a Belsorp Max (BEL Japan).The samples were degassed at elevated temperature of 100-200 o C for 2 h to remove physisorbed gases prior to the measurement.The temperature was then kept at 200 o C for 2 h while flushed with He gas.NH3 gas (balanced NH3, 80 % ,and He, 20 %) was introduced at 100 o C for 30 min, then evacuated by helium gas to remove the physisorbed also for 30 min.Finally, temperature programmed desorption was carried out at temperature of 100-800 o C and the desorbed NH3 was monitored by TCD.

Catalytic reaction
Typical catalytic reaction procedure as follows.Catalyst (0.05 g), dodecanoic acid (1.8 mmol), internal standar of 1,6-hexanediol (0.03 mmol) and iso-PrOH (5 mL) as solvent were placed into a glass reaction tube, which fitted inside a stainless-steel reactor.The reactor was flushed with H2 for ~30 times.Finally, after H2 was introduced into the reactor with an initial H2 pressure of 30 bar at room temperature, the temperature of the reactor was increased to 160 o C.After 6 h of the reaction, both heating and stirring were stopped, the autoclave was removed from the hot plate and then cooled in ice-cold water.Once cooled, the contents of the autoclave were transferred into sample bin then centrifuged (~4000 rpm for 10 min) and ready for GC-FID analysis.The Ni-Sn(1.5)/TiO2 catalyst was easily separated using either simple centrifugation (4000 rpm for 10 min) or filtration, then finally dried overnight under vacuum at room temperature prior to re-usability testing.

Product analysis
Gas chromatography analysis of the reactant (dodecanoic acid) and products was performed on a Perkin Elmer Autosystem XL with a flame ionization detector with an InertCap 225 (i.d.0.25 mm, length 30 m, d.f.0.25 mm) capillary column of GL Science Inc. Tokyo Ja-pan.The products were confirmed by a comparison of their GC retention time, mass spectra with those of authentic samples.Gas chromatography-mass spectrometry (GC-MS) was performed on a Shimadzu GC-17B equipped with a thermal conductivity detector and an RT-βDEXsm capillary column.
The amount of remained reactant and products were determined from GC data using internal standard of 1,6-hexanediol that was introduced to the reactor before reaction run.Calibration curve was performed using known concentrations of internal standard, reactants and products in order to determine the correct response factors.The conversion of dodecanoic acid, yield, and selectivity of the products were calculated according to the following equations: where F 0 is the introduced mol reactant (dodecanoic acid), F t is the remaining mol reactant, and ΔF is the consumed mol reactant (introduced mol reactant -remained mol reactant), which are all obtained from GC analysis using an internal standard technique.

Catalyst characterization
The chemical compositions, BET specific surface area (SBET), and total pore volume (Vp) for -Al2O3 or TiO2 supported Ni-Sn(1.5)catalysts after H2 treatment at 400 o C for 1.5 h are summarized in Table 1.Based on the ICP-AES analyses, the compositions of the supported Ni-Sn alloy were approximately equivalent to the feeding ratios of each precursor and were reflected in the composition of each Ni-Sn alloy phase.The total loading amount of Ni-Sn was 2.3~2.4 mmol.g -1 for all of the supported Ni-Sn(1.5)samples (the composition (mol%) of Ni and Sn are listed in Table 1).The SBET and total pore volume (Vp) of each supported Ni-Sn(1.5)alloy samples were also summarized in Table 1.It is found that the SBET and Vp of Ni-Sn(1.5)/-Al2O3 were 94 m 2 g -1 and 0.39 cm 3 g -1 , respectively (entry 1), while for Ni-Sn(1.5)/TiO2 were 52 m 2 g -1 and 0.17 cm 3 g -1 (entry 2).It can be observed that SBET and total pore volume of TiO2 supported catalyst lower than that of -Al2O3 supported catalysts.It was reported that metal species lead more dispersible on support with lower porosity (e.g.TiO2) due to strong interaction between metal and support which will prevent metal aggregation or sintering during thermal treatment [15,23].
The H2 uptake for Ni-Sn(1.5)/TiO2 and Ni-Sn(1.5)/-Al2O3samples were 9.0 and 13.0 molg -1 , respectively and the average Ni-Sn alloy crystallite sizes derived from Scherrer equation of Ni3Sn2 (101) phase at 2 = 30.8ο were 16 nm and 6 nm, respectively.The H2 uptake supported Ni-Sn alloy in this work was much lower than that of the Raney Ni or supported Ni on aluminium hydroxide, suggesting that the presence of second metal of Sn reduced significantly the accessibility of hydrogen on the surface of Ni metal.Consequently, the catalytic reaction may be effectively proceeded at high pressure of H2 as its have been reported previously [15][16][17][18][19], The XRD patterns of Ni-Sn(1.5)/-Al2O3alloy catalyst after H2 treatment at 400 o C for 1.5 h are shown in Figure 1.It can be observed easily that Ni3Sn2 species as major alloy phase was formed on-Al2O3 with sharped peaks with very similar patterns to the JCPDS card standard of Ni3Sn2 of 6-414 [24].On the other hand, the XRD patterns Ni-Sn(1.5)/TiO2exhibited broadened peaks at 2 = 30.Ni3Sn2 (102), and Ni3Sn2 (110) diffraction peaks, respectively (Figure 2).These results suggest that the higher dispersions of Ni-Sn alloy on the TiO2 were formed as roughly depicted in the average Ni3Sn2 (101) crystallite sizes, which were 6 nm (Table 2).In conclusion, the XRD analysis and H2 measurement results confirm that Ni-Sn alloy phases were also formed on the supports and that their characteristics were consistent with the results observed for the bulk material as reported previously [15].The results of ammonia-temperature programmed desorption (NH3-TPD) measurements of the synthesized catalysts are summarized in Table 3 and the NH3-TPD profiles are shown in Figure 3 and Figure 4.It can be observed that Ni-Sn(1.5)/-Al2O3alloy catalyst consisted of three different peak positions of 178, 510, and 607 o C with acidic amount of 0.23, 0.03, and 0.09 mmol.g -1 , respectively (Figure 3 and Table 3, entry 1) with total number of acid sites = 0.35 mmol.g -1 .Therefore, it can be concluded that Ni-Sn(1.5)/-Al2O3catalyst has mainly weak acid sites.On the other hand, Ni-Sn(1.5)/TiO2displayed the peak position at 611 o C with total number of acidic sites = 0.29 mmol.g -1 that can be attributed to the strong acid sites (Figure 4 and Table 3, entry 2) [25,26].

Screening of catalysts
Results of selective hydrogenation of dodecanoic acid over supported bimetallic Ni-Sn(1.5)catalysts are summarized in Table 4 and the reaction pathways are shown in  It can be observed that by using the Ni-Sn(1.5)/-Al2O3catalysts, the conversion of dodecanoic acid was 59 % with dodecane-1-ol yield was 40 % while dodecane-1,1-diol and dodecane yields were 1 % and 4 %, respectively (entry 1).The conversion of dodecanoic acid reached 89 % when the reaction time was extended to 16 h and yielded 57 % and 4 % of dodecane-1-ol and dodecane, respectively (entry 2).Interestingly, Ni-Sn(1.5)/TiO2catalyst displayed a high conversion of dodecanoic acid (c.a.85 %) and high yield of dodecane-1-ol (c.a.80 %) (entry 3).The yield of dodecane-1-ol increased significantly to 90 % (with 97 % conversion) without significant formation of byproduct when the reaction time was extended to 16 h (entry 4).It can be observed that Ni-Sn(1.5)/TiO2catalyst showed higher catalytic performance than that of Ni-Sn(1.5)/-Al2O3.The high conversion of dodecanoic acid and the high yield of dodecan-1-ol over Ni-Sn(1.5)/TiO2 can be attributed to the relatively high dispersion of Ni-Sn alloy on TiO2 giving rise to active sites with a significantly higher catalytic activity.Alternatively, the high conversion and selectivity may be a result of the strong interactions between the active metals and TiO2 generating significant interactions between C=O groups and Ni-TiOx sites and leading to high selectivity to dodecane-1-ol [27].
The results of re-usability test confirmed that Ni-Sn(1.5)/TiO2catalyst was re-usable and the activity and selectivity of the catalyst can be restored to the initial after reducing with H2 Reaction conditions: Catalyst, 0.065 g; dodecanoic acid, 3.2 mmol; iso-PrOH, 5 mL, 160 o C; H2 pressure 30 bar, reaction time is 6 h. a The value in the parenthesis is Ni/Sn molar ratio.b Loading amount of Ni metal, determined by ICP-AES.c Conversion and selectivity of the main products were determined by GC using an internal standard technique.Product (1) = dodecane-1-ol, (2) dodecane-1,1-diol, and (3) dodecane.d Other products were mainly C11 hydrocarbon including undecane (C11H24), undecene (C11H22), and trace amount of cracking product like 1-pentylbenzene was identified by using GC-MS analysis.e Reaction time was 16 h.f Re-usability test (second reaction run) after the recovered Ni-Sn(1.5)/TiO2 was re-activated by reducing with H2 at 400 o C for 1.5 h.g The catalytic reaction in absence of H2 gas.h Reaction temperature was 180 o C, reaction time 6 h.

Profile of reaction kinetics
Kinetics of the hydrogenation of dodecanoic acid over bimetallic Ni-Sn(1.5)/TiO2 and Ni-Sn(1.5)/-Al2O3catalysts are shown in Figure 5.When Ni-Sn(1.5)/-Al2O3 catalyst was used, dodecanoic acid was slowly increased at the initial reaction time to achieve a 93 % yield after 24 h.Therefore, a similar explanation can be applied to the current results as follows.The induction periods could be associated with the slow formation of oxidic tin (Sn n+ ) from metallic tin (Sn 0 ), as reported by Sordelli et al. (Rh-Sn) [28] and Margitfalvi et al. (Pt-Sn) [29].Since the crystallite size or dispersion of Ni-Sn alloy could affect the length of the induction period, the induction period of Ni-Sn(1.5)/TiO2diminished, and a 100 % dodecanoic acid conversion (~90 % dodecane-1-ol yield) was achieved after 20 h.The excellent catalytic performance of Ni-Sn(1.5)/TiO2produces a promising candidate suitable for the large-scale production of dodecane-1-ol from the selective hydrogenation of dodecanoic acid.

Effect of reaction temperature
The effect of reaction temperature on the catalytic performance of Ni-Sn(1.5)/TiO2 in do-decanoic acid hydrogenation are also investigated and the results are shown in Figure 6.The conversion of dodecanoic acid gradually increased as the increase of reaction temperature and reached to completed reaction at 180 ο C within 6 h with dodecane-1-ol yield of 91 % (entry 4).A small amount of dodecane (4 % yield) and other products (total yield of 12 %) were observed at 100 o C and diminished at higher reaction temperature.These results indicated that further reaction of dodecane-1-ol did not occurred effectively over Ni-Sn(1.5)/TiO2catalyst.

Effect of initial pressure of H2
The effect of the initial H2 pressure on the dodecanoic acid conversion and product selectivity is shown in Figure 7.The conversion of dodecanoic acid and dodecane-1-ol yield increased, whereas the selectivity of ( 2), (3), and other products decreased smoothly to almost 0% between 40-60 bar.The highest yield of dodecane-1-ol (97 %) was achieved at initial pressure H2 of 50-60 bar, suggesting the hydrogenation reaction of dodecanoic acid to dodecane-1-ol over Ni-Sn(1.5)/TiO2catalyst can be proceed effectively under relatively high pressure of H2.These results are consistent with the low H2 uptake of bimetallic Ni-Sn(1.5)/TiO2.

a
8 o , 42.5 o , and 44.2 o , which correspond to the Ni3Sn2 (101), The value between the parenthesis is Ni/Sn molar ratio.b Molar composition of Ni or Sn, determined by ICP-AES.c Determined by N2 adsorption at -196 o C. The pore volume was estimated to be the liquid volume of nitrogen at a relative pressure of approximately 0.995 according to the Barrett-Joyner-Halenda (BJH) approach based on desorption data.

Table 4 .
Results of dodecanoic acid hydrogenation/hydrodeoxygenation over supported Ni-based catalysts