Continuous Dehydrogenation of Aqueous Formic Acid ...

Download PDF
4 downloads 0 Views 1MB Size Report
Comment
Rahat Javaid, Shin-ichiro Kawasaki, Ryuto Ookawara, Koichi Sato, Masateru ... Hiromu Yoshiura, Miki Tamura, Makoto Aso, Noriho Kamiya and Masahiro Goto ...
VOL. NO.

46 11

NOVEMBER

2

0

1

3

JOURNAL OF CHEMICAL ENGINEERING O F J A PA N

[ VOL. 46, NO. 11, NOVEMBER 2013 ]

▶ Transport Phenomena and Fluid Engineering Micromixing Characteristics in the Impeller Discharging Area in Aerated Stirred Tank Wanbo Li, Xingye Geng, Yuyun Bao and Zhengming Gao ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 717 A Front-Tracking Method for Three-Phase Computations of Solidification with Volume Change Truong V. Vu, Gretar Tryggvason, Shunji Homma, John C. Wells and Hideyuki Takakura ––––––––––––––––––––––––––––––––– 726

▶ Chemical Reaction Engineering Reaction Yield of Acrolein Production Process Using Concentrated Glycerol and Supercritical Water Takeyuki Kondo, Toshiaki Matsuo and Yasunari Sase –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 732 Preparation and Photocatalytic Properties of Co-doped TiO2/SiO2–MWCNTs Thin Films for Methylene Blue Removal under Visible Light Irradiation Charuwan Kaewtip and Virote Boonamnuayvitaya ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 738

The Society of Chemical Engineers, Japan

Effect of Electrode Configuration on NO Removal in a Coaxial Dielectric Barrier Discharge Reactor Baomin Sun, Tao Wang, Bin Yang, Xiangyu Zhu, Dong Wang and Haiping Xiao –––––––––––––––––––––––––––––––––––––––––– 746 Continuous Dehydrogenation of Aqueous Formic Acid under Sub-Critical Conditions by Use of Hollow Tubular Reactor Coated with Thin Palladium Oxide Layer Rahat Javaid, Shin-ichiro Kawasaki, Ryuto Ookawara, Koichi Sato, Masateru Nishioka, Akira Suzuki and Toshishige M. Suzuki –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 751

▶ Process Systems Engineering and Safety Process Design of Natural Gas Liquid Recovery Processes Sekwang Yoon, Hyun Jun Cho, Yeong-Koo Yeo and Jin-Kuk Kim –––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 759

▶ Micro and Nano Systems Evaluation for the Mixing Performance of a Microreactor with Inline Measurement Yukako Asano, Shigenori Togashi and Yoshishige Endo –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 770

▶ Materials Engineering and Interfacial Phenomena Preparation of Uniform-Sized Poly[methacryloxypropyl Tris(trimethylsiloxy)silane] Microspheres via Shirasu Porous Glass Membrane Emulsification Technique Yao-Tong Lai, Seiichi Ohta, Kazuki Akamatsu, Shin-ichi Nakao, Yasuyuki Sakai and Taichi Ito –––––––––––––––––––––––––––– 777 The Purity of Methacrylic Acid Crystals under the Presence of Maleic Acid and Methanol in Melt Crystallization Tomomichi Hino, Shinpei Kato and Hiroshi Takiyama –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 785 The Effect of Morphology of Crystals in Suspension on Scale Formation Dynamics in Cooling Crystallization for Organic Compound Shoji Kudo, Tomomichi Hino and Hiroshi Takiyama –––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 790

JCEJAQ 46(11) 717-796(2013) ISSN 0021-9592

Ionic Liquid-in-Oil Microemulsions as Potential Carriers for the Transdermal Delivery of Methotrexate (SC) Hiromu Yoshiura, Miki Tamura, Makoto Aso, Noriho Kamiya and Masahiro Goto ––––––––––––––––––––––––––––––––––––––––– 794

This article appeared in the Journal of Chemical Engineering of Japan. The attached copy is provided to the author for non-commercial research, education use and sharing with colleagues. Other uses listed below are prohibited: - Reproduction, - Commercial use, - Posting to personal, institutional or third party websites.

Research Paper

Journal of Chemical Engineering of Japan, Vol. 46, No. 11, pp. 751–758, 2013

Continuous Dehydrogenation of Aqueous Formic Acid under Sub-Critical Conditions by Use of Hollow Tubular Reactor Coated with Thin Palladium Oxide Layer Rahat Javaid, Shin-ichiro Kawasaki, Ryuto Ookawara, Koichi Sato, Masateru Nishioka, Akira Suzuki and Toshishige M. Suzuki Research Center for Compact Chemical System, National Institute of Advanced Industrial Science and Technology, AIST, 4-2-1 Nigatake, Miyagino-ku, Sendai-shi, Miyagi 983-8551, Japan Keywords: Formic Acid, Dehydrogenation, Tubular Reactor, Thin Catalyst Layer, Palladium Oxide We have fabricated the catalytic tubular reactor whose inner wall is coated with thin palladium oxide (PdO) layer (ca. 3 µm). Continuous decomposition of aqueous formic acid (HCO2H) was examined by the flow reaction system installed with the catalytic tubular reactor. Reactor with PdO inner surface showed superior catalytic performance than that obtained by the reactor having metallic Pd surface. Higher temperature and longer residence time increased the reaction conversion but the effect of pressure was small. Steady generation of hydrogen (H2) was achieved keeping >99% conversion of formic acid at 300°C (10 MPa) within 4.3 s residence time. H2 and carbon dioxide (CO2) were major products with ca. 1 : 1 molar ratio and the generation of carbon monoxide (CO) was minor. Generation of CO was considerably suppressed by increase of sodium formate ratio i.e., formate/(formate+formic acid) ratio. Long term test revealed that steady productivity of H2 was maintained more than 600 h reaction. The XPS spectrum of catalytic surface after long time use evidenced the presence of PdO in large excess along with small amount of metallic Pd indicating that reduction of PdO surface proceeded quite slowly.

Introduction Storage and on-demand generation of hydrogen (H2) from safe and transportable sources are of high significance particularly for mobile and small scale applications. Formic acid (HCO2H) which contains 4.4% of H2 has been considered as one of the possible options for convenient H2 carrier because it is liquid at ambient temperature/pressure, soluble in water and safely transportable (Joó, 2008; Johnson et al., 2010). Moreover formic acid can be easily obtained from carbon monoxide (CO) or biomass processing (Gu et al., 2011; Solymosi et al., 2011). Production of H2 from formic acid has been extensively studied using homogeneous and heterogeneous metal catalysts. Homogeneous ruthenium complexes (Loges et al., 2008; Boddien et al., 2009; Gan et al., 2009) and rhodium complexes (Loges et al., 2010) actively worked at moderate operating temperatures. A great deal of research is being undertaken to develop heterogeneous catalyst systems because they have advantages in terms of separation and reusability of the catalyst and applicability as the fixed bed in flow-through reactors. Platinum group metals (Pt, Rh, Ir) (Solymosi et al., 2011) and palladium (Pd) supported on activated carbon or metal oxides have actively decomposed Received on August 1, 2013; accepted on August 26, 2013 DOI: 10.1252/jcej.13we184 Correspondence concerning this article should be addressed to T. M. Suzuki (E-mail address: [email protected]). Vol. 46  No.©11  Copyright 2013  2013 The Society of Chemical Engineers, Japan

formic acid into H2 and carbon dioxide (CO2) (Zhao et al., 2011). Recently, sophisticated catalysts including nanoparticles, metal alloys and core–shell type bimetals supported on metal oxides have shown remarkable enhancement of reaction rate and effectively released H2 at ambient temperatures (Zhou et al., 2008, 2010; Ojeda and Iglesia, 2009; Gu et al., 2011; Tedsree et al., 2011; Bi et al., 2012). In contrast to batch reaction, flow reactor system enables continuous production of H2 with constant quantity and quality for long time without renewal of reactants. Flow reaction of formic acid decomposition was initially conducted under hydrothermal conditions using high pressure and high temperature water (HPHT–H2O) medium (Yu and Savage, 1998; Zhang et al., 2010). In general, harsh temperature conditions (500–650°C) and high pressure (24–30 MPa) are required to attain complete conversion of formic acid by the non-catalytic processes (Yu and Savage, 1998; Zhang et al., 2010). To combine the advantages of flow reaction and heterogeneous catalysts, packed bed reactors were studied using supported Pd and Pt catalysts (Hyde and Poliakoff, 2004; Bulushev et al., 2010). However smooth material flow is tend to be obstructed by the packed bed leading to undesirable pressure drop and clogging of reactor tube. In contrast hollow tubular reactor minimizes the mass transfer resistance and therefore ensures smooth flow of reactants even in bi-phase reactions of gas/liquid mixture and solid/liquid slurry. Recently we developed the micro-tubular reactors of less than 0.5 mm inner diameter whose interior walls are 751

completely coated with thin metal layer (1–2 µm) of Pd and Pt (Javaid et al., 2009, 2010, 2011, 2013). The micro-tubular reactor offers large surface to volume ratio, good mixing and heat transfer properties that enhances reaction rate. In fact we found the Pd and PdO coated catalytic reactor continuously decomposed aqueous formic acid in a remarkably short residence time. Herein we present the performance of the catalytic reactors for the steady production of H2 from formic acid, particularly with respect to the oxidation state of catalytic surface, reaction temperature/pressure, pH of solution, residence time and long term stability of the reactors.

1. Experimental 1.1 Materials Reagent grade of formic acid (98%), sodium formate, palladium chloride, disodium ethylenediamine tetraacetate (EDTA–Na2), ammonia (28%), hydrazine monohydrate and hydrogen peroxide (30%) were purchased from Wako Pure Chemical Industries, Ltd. and used without further purifications. 1.2 Fabrication of catalytic micro-tubular reactor Metallic tube of Inconel 625 (o.d. 1.6 mm, i.d. 0.5 mm, length 1000 mm) with inner titanium (Ti) layer (thickness 120 µm) was used as the reactor. This double layered tube was fabricated by E.S.Q. Co., Japan by elongation of Ti inlaid Inconel 625 piece by stretch draw process (Javaid et al., 2011). Inner Ti wall was oxidized to TiO2 by feeding of 0.3 wt% hydrogen peroxide solution at supercritical water condition (450°C and 30 MPa) at a flow rate of 20 mL min−1 for 2 h. The TiO2 surface was then modified with ligand by the introduction of toluene solution containing 5% triethoxysilylpropyldiethylenetriamine and then heated at 70°C for 24 h. Pd metal particles as seed were immobilized by ligand exchange reaction of diethylenetriamine moiety with [Pd(NH3)4]2+, followed by reduction with hydrazine solution. Electroless Pd plating of inner wall was carried out by continuous feed of plating solution (0.5 mL min−1) into the reactor tube at 50°C for 5 h. A mixture of 4 mM [Pd(NH3)4]Cl2, 0.15 M EDTA–Na2, 1 M ammonia and 10 mM hydrazine monohydrate were used as the plating solution (Javaid et al., 2009, 2010, 2011, 2013). The plating solution after passing through the reactor was collected in fractions and analyzed by inductively coupled plasma-atomic emission spectroscopy (ICP–AES), SII Model SPS 3100. The amount of Pd deposited on inner surface of tube was calculated by the difference of the initial amount of Pd supplied and that remained in the plating solution after passing through the reactor. The Pd surface was oxidized to PdO by calcination of the reactor tube at 750°C for 2 h under air flow. The Pd surface was restored by reduction of PdO tube at 500°C for 3 h under H2 flow. Inner surface of the catalytic tubular reactors was characterized by scanning electron microscopy (SEM) equipped with energy-dispersive X-ray spectroscopy (EDX) (XL30S, Phillips Co.). Oxidation state of catalyst surface was analyzed 752

Fig. 1 Schematic diagram of the experimental setup

by X-ray photoelectron spectroscopy (XPS) analysis (ESCA 5600 XPS, PHI) under ultra-vacuum with MgKα excitation wavelength. 1.3 Catalytic decomposition of formic acid A schematic diagram of the instrumental setup is shown in Figure 1. The metallic tubular reactor of 1000 mm length with inner volume of 196 µL coated with Pd or PdO catalyst was placed in an oven at constant temperature within ±0.1°C. Air was expelled from the tubular reactor system by sweep with nitrogen gas prior to start the experiments. Most of the cases 0.5 M formic acid solution was introduced to syringe pump (ISCO, model 260D) and mixed in a T-mixer (0.6 mL min−1 flow rate) with HPHT–H2O (1.4 mL min−1 flow rate) so that to be 0.15 M formic acid solution. The formic acid solution was fed into the tubular reactor coated with metal catalyst. The temperature and pressure at each point were measured by thermo-couples (K-type) and digital pressure gauge, respectively. After passing through the reactor, the solution was rapidly cooled in a heat exchanger and depressurized by a back pressure regulator. The effluent was separated into gas and liquid phases in a gas/liquid separator at ambient conditions. Gas flow rate was measured by a soap-film flow meter (VP-2U, Horiba STEC Co.). Collected gaseous products were analyzed by gas chromatograph (GC-8A, Shimadzu Corp.) equipped with thermal conductivity detector (TCD). Molecular sieve 5A column (3 mm×6 m) was used for H2 and CO analysis with argon as the carrier gas while Porapak-Q column (3 mm×3 m) was used coupled with helium as the carrier gas for the CO2 analysis. Inorganic carbon (IC) and total organic carbon (TOC) contents in the liquid phase were determined by a total carbon analyzer (TOC-5000A, Shimadzu Corp.). The conversion of formic acid was calculated by the difference in TOC value between inlet and outlet of the reactor. Production of CO2 was calculated by the sum of CO2 in gas phase (GC–TCD) and that dissolved in aqueous phase (IC). Leaching of Pd from the reactor tube during the reaction was checked by ICP–AES analysis of the liquid phase. Formic acid conversions were calculated by the difference of TOC content in the aqueous phase before and after the reaction. Residence time was calculated by the inner volume of the reactor divided by the volumetric flow rate of water which was given by mass flow rate of water divided by water density at the reaction conditions. The blank test was performed by use of Inconel 625 with inlaid TiO2/Ti as reactor Journal of Chemical Engineering of Japan

Fig. 3 The XPS spectra of the inner surface of the catalytic tube as deposited Pd surface (solid line) by electroless plating and that calcined at 750°C under air stream (broken line)

Fig. 2 Images of catalytic tubular reactor: a) Outer appearance of Inconel tube; b) Diagonal cross sectional image of Inconel 625 tube with TiO2/Ti secondary layer coated with thin Pd layer; c) SEM image of Pd particles deposited over inner surface of the reactor tube; d) The catalyst layer after calcination at 750°C under air flow for 2 h

tube without Pd plated layer.

2. Results and Discussion 2.1 Fabrication of the Pd coated metallic reactor Nickel alloy (Inconel 625) tube with Ti inner layer (Figure 2(a)) was used as the reactor substrate in order to ensure the durability at high temperature/pressure (Inconel 625) and resistance against the corrosion condition (Ti). Oxidation of Ti surface to TiO2 was done because oxide surface is desirable for the deposition of Pd particles as seeds prior to electroless plating. In addition, TiO2 layer acts as a barrier that prevents inter-metallic diffusion of Ti and the plated Pd layer (Javaid et al., 2010, 2013). The inner wall of the reactor tube was coated with thin Pd layer by continuous supply of Pd plating solution. The cross sectional image of the freshly plated reactor shows that the inner surface is completely coated with thin Pd layer that locates over the Ti layer (Figure 2(b)). Magnified image (Figure 2(c)) shows numerous round shaped Pd crystals that form rough catalytic surface thereby reaction can be promoted. The thickness of the Pd layer was estimated to 3.6 µm by the amount of deposited Pd (67.3 mg) over inner surface of 0.5×1000 mm tube. The morphology of catalyst surface did not change after calcination at 750°C under air flow (Figure 2(d)). These catalytic tubular reactors provide a remarkably high surface area and volume ratio of 0.8×104 m2 m−3 [Calculated surface area of reactor= 2πrl=1.57×10−3 m2, Calculated volume of reactor= πr2l=1.96×10−7 m3 (where r=radius of reactor and l=length of reactor]. Figure 3 shows the XPS spectra of inner Pd surface of the Vol. 46  No. 11  2013

reactors before and after calcination in air. The XPS spectrum of freshly plated Pd surface gave 3d3/2 and 3d5/2 peaks at 340.5 eV and 335.2 eV, respectively which correspond to the peaks of typical metal Pd. Calcination of Pd coated reactor tube gave distinct high energy shift of the peaks to 341.9 eV and 336.6 eV accompany with the disappearance of Pd(0) peaks. These new peaks are assigned to those of PdO. This confirms that the surface of electrolessly plated metallic Pd was completely oxidized to Pd(II) oxide by calcination under air. 2.2 Continuous decomposition of formic acid We assembled the syringe pumps, pre-heater, catalytic tubular reactor, oven, cooler and gas/liquid separator with the well monitored temperature and pressure control devices as depicted in Figure 1. Reactor tubes with as plated Pd surface and that calcined under air were used in the present work. A continuous supply of formic acid for 6 h was carried out in each experimental run. During 6 h continuous reaction, products in gas and liquid phases were collected each 15–30 min. The volume of evolved gas and each component in gas/liquid phases was analyzed by GC–TCD and TOC/IC. 2.2.1 Effect of temperature Table 1 displays the experimental data for the decomposition of 0.15 M formic acid at temperatures ranging from 260 to 300°C. All the experiments were carried out keeping the pressure at 10 MPa. The main products were H2 and CO2 which correspond to more than 98% of the total evolved gas. Small amount of CO was observed but methane and methanol were not found in the products. The ratio of H2 and CO2 is close to unity in these experiments (Table 1; Entries 2–11). Higher temperature facilitates the conversion of formic acid to H2 and CO2 along with a decrease in CO evolution (Table 1; Entries 2–6 and 7–11). Catalytic layer is essentially required for the enhancement of formic acid decomposition. Neither gas evolution nor conversion of formic acid was observed in the absence of Pd layer at 300°C (Table 1; Entry 1). Presence of catalytic surface undoubtedly contributed to facilitate the reaction. While in the absence of catalytic surface, much higher temperature and pressure conditions like supercritical water are required for complete decomposition of formic acid 753

Table 1 Formic acid conversion and volume % of product gas from the decomposition of formic acida) Entry 1 2 3 4 5 6 7 8 9 10 11

Reactor

Product gas [vol%]

Temperature [°C]

Residence time [s]

Formic acid conversionb) [%]

H2

CO2c)

CO

300 300 290 280 270 260 300 290 280 270 260

4.3 4.3 4.4 4.5 4.6 4.7 4.3 4.4 4.5 4.6 4.7

No reaction 99.2 70.9 38.9 28.2 18.9 69.8 39.5 26.4 20.6 16.0

— 49.7 49.6 49.6 49.2 48.5 49.3 49.0 48.8 48.7 48.5

— 49.9 49.2 48.8 48.7 48.4 49.0 48.9 48.4 48.1 47.4

— 0.4 1.2 1.6 2.1 3.1 1.7 2.1 2.8 3.2 4.1

d)

Blank (without Pd layer) PdO inner surface

Pd inner surface

a) For all experiments 0.15 M formic acid was used and pressure was kept constant at 10 MPa. b) Calculated based on the TOC change in the aqueous phase before and after reaction. c) Sum of CO2 in the gas phase by GC–TCD and aqueous phase by TOC (IC measurement). d) The blank test was performed by use of Inconel 625 with inlaid TiO2/Ti as reactor tube without Pd plated layer.

under similar orders of residence time (Yu and Savage, 1998; Zhang et al., 2010). The essential pathways of formic acid decomposition are dehydrogenation accompany with generation of CO2 (Eq. (1)) and dehydration with generation of CO (Eq. (2)) (Ojeda and Iglesia, 2009).

HCO2H = H2 +CO2 ΔG = − 48.4 kJ mol −1

(1)

(2) HCO2H = H2O+CO ΔG = − 28.5 kJ mol Evolution of CO is undesirable because CO molecule strongly binds to metal surface and inhibits the catalytic reactions as observed for the fuel cell anode catalysts (Kordesch and Simader, 1995). In our catalytic reactor system, evolution of H2 and CO2 are always at least more than one order of magnitude higher than that of CO. This suggests that dehydrogenation (Eq. (1)) is the dominant reaction pathway as compared to dehydration rout (Eq. (2)). Present catalytic reaction in aqueous solution is in consistent with the general trend of hydrothermal decomposition of formic acid where dehydrogenation is predominant (Bjerre and Sørensen, 1992; Brill et al., 1996). Shift of equilibrium towards product side in Eq. (2) must be suppressed in hydrothermal reaction where large excess of water is present. In contrast, dehydration is major reaction path in the gas-phase decomposition resulting higher CO content (Blake and Hinshelwood, 1960; Blake et al., 1971; Hsu et al., 1982; Saito et al., 1984). According to Yu and Savage (1998), water plays a key role in decomposition of formic acid under hydrothermal conditions. Rueller and co-authors (Ruelle et al., 1986; Ruelle, 1987) and Melius et al. (1990) considered the potential catalytic effect of water on the formic acid decomposition. 2.2.2 Oxidation state of Pd surface It is of note, oxidized Pd surface (PdO) always gave better performance than that by metallic Pd with respect to reaction conversion and H2 production selectivity. More than 99% conversion of formic acid along with 0.4% CO content was attained by PdO at 300°C within 4.3 s residence time (Table 1; Entry 2). Con−1

754

Fig. 4 Experimental runs by alternate change of reactor surface i.e., PdO and Pd. The surface of the catalyst was oxidized by passing air at 750°C and reduced by hydrogen gas to the reactor tube at 500°C alternately in each 6 h run. Temperature, pressure, residence time and concentration of formic acid were kept constant at 300°C, 10 MPa, 4.3 s and 0.15 M, respectively. During each run (6 h) gas/liquid sample analysis was carried out every 30 min interval

trary, conversion by the metallic Pd surface remained 69.8% under the same reaction conditions with relatively higher CO content (1.7%) (Table 1; Entry 7). The oxidation state of Pd surface can be reversibly changed by oxidation and reduction of the reactor tube. Figure 4 shows the reaction conversion profile at 300°C by alternate change of catalytic surface i.e., Pd and PdO. Obviously the reaction efficiency in terms of formic acid conversion is restored by the repeated change of the oxidation state of metal surface demonstrating the reproducibility and revision of catalytic activity. Zhao et al. (2011), described Pd(II) is an active state for formic acid decomposition by the Pd(II) complex catalyst immobilized on SiO2 via sulfur containing ligand (SiO2–S– Pd(II)). According to their proposed mechanism, Pd(II) surface is more susceptible to formate (HCO2−) interaction Journal of Chemical Engineering of Japan

Table 2 Effect of different parameters on continuous generation of H2 from formic acid decomposition using PdO as catalyst Entry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

Parameter Pressure

Residence time

Concentration

Pressure [MPa]

Formic acid concentration [M]

Residence time [s]

Formic acid conversion [%]

H2

CO2

CO

300 300 300 280 280 280 260 260 260 300 300 300 300 300 300

10 15 20 10 15 20 10 15 20 10 10 10 10 10 10

0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.15 0.30 0.60

4.29 4.35 4.41 4.54 4.58 4.62 4.74 4.78 4.81 4.3 2.0 1.0 4.3 4.3 4.3

99.2 99.0 98.5 38.9 36.1 35.3 18.9 17.1 15.6 99.2 91.0 62.3 99.2 98.7 90.2

49.7 49.7 49.7 49.6 49.5 49.5 4.85 48.4 48.3 49.7 49.9 49.7 49.7 49.7 49.6

49.9 49.9 49.8 48.8 48.8 48.8 48.4 48.5 48.5 49.9 49.7 49.5 49.9 49.5 48.9

0.4 0.4 0.5 1.6 1.7 1.7 3.1 3.1 3.2 0.4 0.4 0.8 0.4 0.8 1.5

by direct coordination thereby decarboxylation and subsequent dehydrogenation takes place. Pd(II) species in homogeneous ethanethiol complex (PdCl2/EtSH) solution were rapidly reduced to metallic Pd under the presence of H2 and formic acid. Contrary, Pd(II) immobilized on solid support (SiO2–S–Pd(II)) was considerably stable (Zhao et al., 2011). Our PdO coated reactor also revealed a long life keeping the catalytic activity for more than 600 h (vide infra). 2.2.3 Effects of pressure and residence time Table 2 shows the reaction data by alternation of parameters including pressure, residence time and concentration of formic acid. Unlike temperature, effect of pressure was very small in the present pressure range (10–20 MPa) at 300, 280 and 260°C (Table 2; Entries 1–9). Non-catalytic decomposition in the supercritical water (>380°C), pressure played more significant role where formic acid conversion and H2 production were increased with increase of pressure (Yu and Savage, 1998). This observation was explained correlated with the sensitive change of water density at supercritical condition. Contrary, the difference of water densities is rather small in the present pressure range; for example at 280°C, i.e., 10 MPa (0.76 mL−1) and 20 MPa (0.77 mL−1). Residence time gave more marked effect (Table 2; Entries 10–12). At 300°C and 10 MPa, more than 90% of formic acid conversion was maintained by decrease of the residence time to 2.0 s but it rapidly declined to 62% at 1 s reaction presumably due to insufficient contact time with catalytic surface. Increase of formic acid concentration decreased the reaction conversion along with increase of CO content (Table 2; Entries 13–15). Similar trend observed in the supercritical water has been attributed to the dehydration besides the dehydrogenation (Zhang et al., 2010). Besides the Eqs. (1) and (2), water gas shift reaction expressed in Eq. (3) can be participated in where CO, H2, CO2 and H2O are present. This reaction also has contributed to control the CO/CO2 ratio.

Vol. 46  No. 11  2013

Product gas [vol%]

Temperature [°C]

Fig. 5 Effect of formate/(formate+formic acid) ratio on the reaction conversion (line pattern) and the vol% of CO generated (bar graph) at 280°C and 10 MPa, using PdO coated tubular reactor

CO + H2O = H2 + CO2

(3)

2.2.4 Effect of addition of sodium formate Since we considered that formate ion is the key species particularly for the formation of coordinated intermediate with Pd(II) (PdO), we examined the effect of sodium formate added to formic acid. Figure 5 shows the reaction conversion and CO % in the product at 280°C (10 MPa) as the function of formate/(formate+formic acid) ratio. The reaction conversion increased with the addition of formate. For example, initial conversion of 38.9% increased to 71.8% with formate/ (formate+formic acid) ratio of 0.4. However further increase of formate ratio gave negative effect to the reaction conversion. The pH of reaction solution varied from 2.3 to finally 7.0 accompanied with the increase of formate ratio (Figure 5). Since pKa value of formic acid is 3.6, formate anion is the dominant species at the pH >3.6. Observed enhancement of conversion at initial stage might be attrib755

Fig. 7 XPS spectrum of Pd 3d5/2 peaks obtained for the surface of catalyst layer after more than 600 h reaction. The peak was fit to two components at 336.8 eV and 335.1 eV which correspond to PdO and metallic Pd, respectively Fig. 6 Long term test of the reactor performance for the decomposition of 0.15 M formic acid conducted at (▲) 300°C and (■) 260°C under 10 MPa. The plot of each run is the conversion % of formic acid for 6 h continuous flow reaction. The total reaction time is 72 h at each temperature

uted to the increase of formate species which can favorably interact with catalytic surface (Pd–OH). However the surface charge of Pd–OH decreases with increase of pH due to deprotonation of the hydroxyl group (Li et al., 2011). According to the zeta potential measurement, the isoelectric point of PdO locates between pH 5.0 to 6.0 (Li et al., 2011), where the surface charge switches from positive to negative. Therefore the catalytic surface is less susceptible to formate anion with increase of pH leading to suppress the reaction conversion. The CO selectivity of the present system is lower than that of conventional reforming processes of formic acid, methanol, ethanol and methane etc. (Deluga et al., 2004; Hoang et al., 2006; Shin and Besser, 2007). However it is still the level of poison the Pt catalysts used in the fuel cells. Notably, addition of sodium formate effectively suppressed the CO formation; content of CO in the product gas was decreased from initial 1.57% to finally non detectable level by GC– TCD. This is because CO generated is readily converted to formate ion in basic medium by the reaction with hydroxyl ion as given in Eq. (4) (Vandenberg et al., 1984; Ford and Rokicki, 1988; Fachinetti et al., 1996; Ford et al., 2002). This minimizes the CO poisoning of catalyst surface for further application of the produced H2 to fuel cell systems.

CO+OH− = HCO2 −

(4)

2.2.5 Long term stability of the reactor performance In order to evaluate the stability/durability of the reactor under long term reactions, twelve experimental runs were successively conducted at 300°C and then another twelve runs were done at 260°C keeping the pressure 10 MPa (Figure 6). Each run consists of a continuous reaction of 6 h and therefore 72 h continuous reaction was conducted at each 300°C and 260°C temperatures. We did not observe the decrease of H2 generation during twelve time repeat of each experimental run at 300°C and 260°C (Figure 6). Leaching of Pd from 756

the reactor was negligible during these experimental runs. Figure 7 shows the XPS spectrum for Pd 3d5/2 of catalytic surface after more than 600 h use. Appearance of intense peak at 336.8 eV along with the shoulder at 335.1 eV corresponds to PdO and metallic Pd, respectively. This indicates the partial reduction of PdO to metallic Pd phase after long time exposure in a reducing environment. According to Gaussian analysis of the spectral curve the distribution of metallic Pd was around 18%. Therefore reduction of PdO surface proceeded considerably slow. Complete PdO surface can be retrieved simply by calcination of the reactor at 750°C under air flow. Additionally, long life time of catalyst and reproducibility of results imply that CO did not accumulate on the catalyst surface by adsorption and desorption equilibrium under high temperature high pressure flow (Graf et al., 2009). Therefore catalytic activity was not inhibited in the course of the continuous reaction.

Conclusion Continuous decomposition of aqueous formic acid was achieved in a hollow tubular reactor coated with thin Pd or PdO layer. Dehydrogenation is the dominant reaction pathway than the dehydration rout. Nearly equal quantity of H2 and CO2 were generated as the major products while CO generation was minor. Addition of sodium formate of increase of pH significantly suppressed the CO formation. This can minimize the CO inhibition of catalyst surface. PdO surface gave better catalytic performance than the metallic Pd counterpart in accord with the mechanism where Pd(II) surface is more susceptible to formate (HCO2−) interaction by direct coordination. The relative number of surface Pd–OH and Pd–O by varying formate/(formate+ formic acid) ratio seemed to account for the difference in the catalytic activity observed. Formic acid conversion of more than 99% was attained at 300°C, with remarkably short residence time (4.3 s). No significant sign of deactivation was observed after more than 600 h of reaction keeping stable conversion of formic acid. This is in consistence with the considerably slow reduction process of PdO surface monitored by XPS spectra. Present hollow tubular reactor Journal of Chemical Engineering of Japan

ensured the long term stability of the catalytic reactor keeping the steady generation of H2 due to the high robustness of catalytic reactor, low CO poisoning and smooth material flow. Literature Cited Bi, Q.-Y., X.-L. Du, Y.-M. Liu, Y. Cao, H.-Y. He and K.-N. Fan; “Efficient Subnanometric Gold-Catalyzed Hydrogen Generation via Formic Acid Decomposition under Ambient Conditions,” J. Am. Chem. Soc., 134, 8926–8933 (2012) Bjerre, A. B. and E. Sørensen; “Thermal Decomposition of Dilute Aqueous Formic Acid Solutions,” Ind. Eng. Chem. Res., 31, 1574–1577 (1992) Blake, P. G. and C. Hinshelwood; “The Homogeneous Decomposition Reactions of Gaseous Formic Acid,” Proc. R. Soc. A, 255, 444–455 (1960) Blake, P. G., H. H. Davies and G. E. Jackson; “Dehydration Mechanisms in the Thermal Decomposition of Gaseous Formic Acid,” J. Chem. Soc. B, 1923–1925 (1971) Boddien, A., B. Loges, H. Junge, F. Gärtner, J. R. Noyes and M. Beller; “Continuous Hydrogen Generation from Formic Acid: Highly Active and Stable Ruthenium Catalysts,” Adv. Synth. Catal., 351, 2517–2520 (2009) Brill, T. B., J. W. Schoppelrei, P. G. Maiella and A. Belsky; “Spectrokinetics of Solvo-Thermal Reactions,” Proc. 2nd Int. Conf. Solvotherm. Reactions, pp. 5–8, Takamatsu, Japan (1996) Bulushev, D. A., S. Beloshapkin and J. R. H. Rossa; “Hydrogen from Formic Acid Decomposition over Pd and Au Catalysts,” Catal. Today, 154, 7–12 (2010) Deluga, G. A., J. R. Salge, L. D. Schmidt and X. E. Verykios; “Renewable Hydrogen from Ethanol by Autothermal Reforming,” Science, 303, 993–997 (2004) Fachinetti, G., T. Funaioli, L. Lecci and F. Marchetti; “Ru3(CO)12 in Acidic Media. Intermediates of the Acid-Cocatalyzed Water Gas Shift Reaction (WGSR),” Inorg. Chem., 35, 7217–7224 (1996) Ford, P. C. and A. Rokicki; “Nucleophilic Activation of Carbon Monoxide: Applications to Homogeneous Catalysis by Metal Carbonyls of the Water Gas Shift and Related Reactions,” Adv. Organomet. Chem., 28, 139–217 (1988) Ford, P. C., E. Trabuco and M. M. Mdleleni; Encyclopedia of Catalysis, 1st ed., Wiley-VCH Publishers, New York, U.S.A. (2002) Gan, W., P. J. Dyson and G. Laurenczy; “Hydrogen Storage and Delivery: Immobilization of a Highly Active Homogeneous Catalyst for the Decomposition of Formic Acid to Hydrogen and Carbon Dioxide,” React. Kinet. Catal. Lett., 98, 205–213 (2009) Graf, P. O., D. J. M. de Vlieger, B. L. Mojet and L. Lefferts; “New Insights in Reactivity of Hydroxyl Groups in Water Gas Shift Reaction on Pt/ZrO2,” J. Catal., 262, 181–187 (2009) Gu, X., Z. H. Lu, H. L. Jiang, T. Akita and Q. Xu; “Synergistic Catalysis of Metal–Organic Framework-Immobilized Au–Pd Nanoparticles in Dehydrogenation of Formic Acid for Chemical Hydrogen Storage,” J. Am. Chem. Soc., 133, 11822–11825 (2011) Hoang, D. L., S. H. Chan and O. L. Ding; “Hydrogen Production for Fuel Cells by Autothermal Reforming of Methane over Sulfide Nickel Catalyst on a Gamma Alumina Support,” J. Power Sources, 159, 1248–1257 (2006) Hsu, D. S. Y., W. M. Shaub, M. Blackburn and M. C. Lin; “Thermal Decomposition of Formic Acid at High Temperatures in Shock Waves,” Proc. 19th Int. Symp. Combust., pp. 89–96, Hifa, Israel (1982) Hyde, J. R. and M. Poliakoff; “Supercritical Hydrogenation and AcidVol. 46  No. 11  2013

Catalyzed Reactions without Gases,” Chem. Commun., 1482–1483 (2004) Javaid, R., D. A. P. Tanaka, H. Kawanami and T. M. Suzuki; “Silica Capillary with Metal (Pd and Pt) Inner Wall: Application to Continuous Decomposition of Hydrogen Peroxide,” Chem. Lett., 38, 146–147 (2009) Javaid, R., H. Kawanami, M. Chatterjee, T. Ishizaka, A. Suzuki and T. M. Suzuki; “Fabrication of Microtubular Reactors Coated with Thin Catalytic Layer (M=Pd, Pd–Cu, Pt, Rh, Au),” Catal. Commun., 11, 1160–1164 (2010) Javaid, R., H. Kawanami, M. Chatterjee, T. Ishizaka, A. Suzuki and T. M. Suzuki; “Sonogashira C–C Coupling Reaction in Water Using Tubular Reactors with Catalytic Metal Inner Surface,” Chem. Eng. J., 167, 431–435 (2011) Javaid, R., S. Kawasaki, A. Suzuki and T. M. Suzuki; “Simple and Rapid Hydrogenation of p-Nitrophenol with Aqueous Formic Acid in Catalytic Flow Reactors,” Beilstein J. Org. Chem., 9, 1156–1163 (2013) Johnson, T. C., D. J. Morris and M. Wills; “Hydrogen Generation from Formic Acid and Alcohols Using Homogeneous Catalysts,” Chem. Soc. Rev., 39, 81–88 (2010) Joó, F.; “Breakthroughs in Hydrogen Storage-Formic Acid as a Sustainable Storage Material for Hydrogen,” ChemSusChem, 1, 805–808 (2008) Kordesch, K. V. and G. R. Simader; “Environmental Impact of Fuel Cell Technology,” Chem. Rev., 95, 191–207 (1995) Li, W., Z. Qiang, T. Zhang, X. Bao and X. Zhao; “Efficient Degradation of Pyruvic Acid in Water by Catalytic Ozonation with PdO/CeO2,” J. Mol. Catal. Chem., 348, 70–76 (2011) Loges, B., A. Boddien, H. Junge and M. Beller; “Controlled Generation of Hydrogen from Formic Acid Amine Adducts at Room Temperature and Application in H2/O2 Fuel Cells,” Angew. Chem. Int. Ed., 47, 3962–3965 (2008) Loges, B., A. Boddien, F. Gärtner, H. Junge and M. Beller; “Catalytic Generation of Hydrogen from Formic Acid and Its Derivatives: Useful Hydrogen Storage Materials,” Top. Catal., 53, 902–914 (2010) Melius, C. F., N. E. Bergan and J. E. Shepherd; “Effects of Water on Combustion Kinetics at High Pressure,” Proc. 23rd Int. Symp. Combust., pp. 217–223, Orleans, France (1990) Ojeda, M. and E. Iglesia; “Formic Acid Dehydrogenation on Au-Based Catalysts at Near-Ambient Temperatures,” Angew. Chem. Int. Ed., 48, 4800–4803 (2009) Ruelle, P.; “Ab initio Study of the Unimolecular Pyrolysis Mechanisms of Formic Acid: Additional Comments Based on Refined Calculations,” J. Am. Chem. Soc., 109, 1722–1725 (1987) Ruelle, P., U. W. Kesselring and H. Nam-Tran; “Ab initio QuantumChemical Study of the Unimolecular Pyrolysis Mechanisms of Formic Acid,” J. Am. Chem. Soc., 108, 371–375 (1986) Saito, K., T. Kakumoto, H. Kuroda, S. Torii and A. Imamura; “Thermal Unimolecular Decomposition of Formic Acid,” J. Chem. Phys., 80, 4989–4996 (1984) Shin, W. C. and R. S. Besser; “Towards Autonomous Control of Microreactor System for Steam Reforming of Methanol,” J. Power Sources, 164, 328–335 (2007) Solymosi, F., Á. Koós, N. Liliom and I. Ugrai; “Production of CO–Free H2 from Formic Acid. A Comparative Study of the Catalytic Behavior of Pt Metals on a Carbon Support,” J. Catal., 279, 213–219 (2011) Tedsree, K., T. Li, S. Jones, C. W. A. Chan, K. M. K. Yu, P. A. J. Bagot, E. A. Marquis, G. D. W. Smith and S. C. E. Tsang; “Hydrogen Production from Formic Acid Decomposition at Room Temperature 757

Using a Ag–Pd Core–Shell Nanocatalyst,” Nat. Nanotechnol., 5, 302–307 (2011) Vandenberg, D. M., T. M. Suzuki and P. C. Ford; “Homogeneous Catalysis of the Water Gas Shift Reaction by Iridium Carbonyl in Alkaline Solution,” J. Organomet. Chem., 272, 309–320 (1984) Yu, J. and P. E. Savage; “Decomposition of Formic Acid under Hydrothermal Conditions,” Ind. Eng. Chem. Res., 37, 2–10 (1998) Zhang, Y., J. Zhang, L. Zhao and C. Sheng; “Decomposition of Formic Acid in Supercritical Water,” Energy Fuels, 24, 95–99 (2010) Zhao, Y., L. Deng, S. Tang, D. Lai, B. Liao, Y. Fu and Q. Guo; “Selective

758

Decomposition of Formic Acid over Immobilized Catalysts,” Energy Fuels, 25, 3693–3697 (2011) Zhou, X., Y. Huang, W. Xing, C. Liu, J. Liao and T. Lu; “High-Quality Hydrogen from the Catalyzed Decomposition of Formic Acid by Pd–Au/C and Pd–Ag/C,” Chem. Commun., 3540–3542 (2008) Zhou, X., Y. Huang, C. Liu, J. Liao, T. Lu and W. Xing; “Available Hydrogen from Formic Acid Decomposed by Rare Earth Elements Promoted Pd–Au/C Catalysts at Low Temperature,” ChemSusChem, 3, 1379–1382 (2010)

Journal of Chemical Engineering of Japan