Catalytic and Electrocatalytic Oxidation of Ethanol over Palladium

Jul 10, 2013 - ... ethanol can be obtained from renewable resources like sugar cane, ...... Du , W.; Mackenzie , K. E.; Milano , D. F.; Deskins , N. A...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Langmuir

Catalytic and Electrocatalytic Oxidation of Ethanol over PalladiumBased Nanoalloy Catalysts Jun Yin,† Shiyao Shan,† Mei Shan Ng,† Lefu Yang,† Derrick Mott,‡ Weiqin Fang,† Ning Kang,† Jin Luo,† and Chuan-Jian Zhong*,† †

Department of Chemistry, State University of New York at Binghamton, Binghamton, New York 13902, United States School of Materials Science, Japan Advanced Institute of Science and Technology, 1-1 Asahidai, Nomi, 923-1292 Ishikawa, Japan

‡

S Supporting Information *

ABSTRACT: The control of the nanoscale composition and structure of alloy catalysts plays an important role in heterogeneous catalysis. This paper describes novel findings of an investigation for Pd-based nanoalloy catalysts (PdCo and PdCu) for ethanol oxidation reaction (EOR) in gas phase and alkaline electrolyte. Although the PdCo catalyst exhibits a mass activity similar to Pd, the PdCu catalyst is shown to display a much higher mass activity than Pd for the electrocatalytic EOR in alkaline electrolyte. This finding is consistent with the finding on the surface enrichment of Pd on the alloyed PdCu surface, in contrast to the surface enrichment of Co in the alloyed PdCo surface. The viability of C−C bond cleavage was also probed for the PdCu catalysts in both gas-phase and electrolyte-phase EOR. In the gas-phase reaction, although the catalytic conversion rate for CO2 product is higher over Pd than PdCu, the nanoalloy PdCu catalyst appears to suppress the formation of acetic acid, which is a significant portion of the product in the case of pure Pd catalyst. In the alkaline electrolyte, CO2 was detected from the gas phase above the electrolyte upon acid treatment following the electrolysis, along with traces of aldehyde and acetic acid. An analysis of the electrochemical properties indicates that the oxophilicity of the base metal alloyed with Pd, in addition to the surface enrichment of metals, may have played an important role in the observed difference of the catalytic and electrocatalytic activities. In comparison with Pd alloyed with Co, the results for Pd alloyed with Cu showed a more significant positive shift of the reduction potential of the oxygenated Pd species on the surface. These findings have important implications for further fine-tuning of the Pd nanoalloys in terms of base metal composition toward highly active and selective catalysts for EOR.

■

INTRODUCTION The development of direct ethanol fuel cells (DEFCs) as a power source has drawn a surge of interest in recent years due to its biofuel characteristics, its ability to eliminate the toxicity issue of methanol as in direct methanol fuel cells, and its higher energy density than that of direct methanol fuel cells. Ethanol is an attractive alternative fuel to methanol because ethanol is a hydrogen-rich liquid, and it has a higher energy density (8.0 kWh/kg) compared to that for methanol (6.1 kWh/kg). Ethanol also has a 33% higher energy density than that for methanol. Importantly, ethanol can be obtained from renewable resources like sugar cane, wheat, or corn, which has the potential to significantly increase the fuel utilization and fuel cell performance, lower the cost, improve the safety, and enhance the fuel sustainability. To enable ethanol based fuel cells, one of the key problems is the need of highly effective catalysts, which is considered to be much more challenging than the study of catalysts for direct methanol fuel cells (DMFCs).1,2 This is largely due to the difficulty in C−C bond cleavage for the complete oxidation of ethanol to CO2. In addition to the extensively studied platinum catalysts, palladium-based systems have received increased interest. The mechanism of the ethanol oxidation reaction on Pd in alkaline media is believed to involve the removal of the adsorbed ethoxi by the adsorbed hydroxyl as the rate-determining step.3,4 Alloys of Pd have been shown to enhance the catalytic activity and © XXXX American Chemical Society

stability of Pd for ethanol oxidation reaction (EOR) in alkaline media. Examples include PdIrNi/C,5 PdPt/C,6,7 PdCu films,8 PdNi/C,9 PdNi/CNF,10 PdAu/C,11,12 and PtSnPd/C.13 PdAg thin films on reduced graphene oxide14 were shown to exhibit an enhanced performance over other support materials such as carbon black and carbon nanotubes.15 PdSn catalysts were shown to display an enhanced performance over commercial Pd/C.16 PdCu catalyst was also demonstrated with high activity for electrooxidation of 2-propanol recently.17 Despite increasing demonstrations of the catalysts for ethanol oxidation, the understanding of the catalyst design and structures for achieving C−C cleavage of ethanol in its electrooxidation remains elusive. Pt/Rh/SnO2 electrocatalysts was recently shown to be capable of oxidizing ethanol to CO2 in an acidic electrolyte,18 which is believed to involve weakly bonded Rh− OH which can adsorb ethanol and break the C−C bond. In another study,19 selective C−C bond cleavage of ethanol was shown to form carbon dioxide in alkaline solutions using Pb(IV) acetate in the solution phase as a cocatalyst of Pt electrode. The cocatalyst (underpotentially deposited Pb) is believed to form Pb(OH)2 on the Pt surface to favor OH adsorption at lower potentials to oxidize the CO and other Received: May 15, 2013 Revised: June 22, 2013

A

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

h yielding 40% loading. The commercial Pd/C contains 50% water in the form of wet paste. After the reduction treatment, the water was removed, and the loading was thus doubled. Instrumentation and Measurements. Transmission Electron Microscopy (TEM). TEM was performed on a Hitachi H-7000 electron microscope (100 kV). For TEM measurements, the catalyst samples were suspended in hexane solution and were drop cast onto a carboncoated copper grid followed by solvent evaporation in air at room temperature. Direct Current Plasma−Atomic Emission Spectrometer (DCP− AES). The composition was analyzed using the direct current plasma− atomic emission spectroscopy, which was performed using an ARL Fisons SS-7 dc plasma−atomic emission spectrometer (DCP−AES). Thermogravimetric Analysis (TGA). TGA was performed on a Perkin-Elmer Pyris 1-TGA for determining the metal loading on the carbon support, details of which were described in our previous reports.25 Typical samples weighed ∼4 mg and were heated in a platinum pan. Samples were heated in 20% O2 at a rate of 10 °C min−1. Electrochemical Measurements. A glassy carbon disk electrode with a geometric surface area of 0.196 cm2 was coated with the catalyst layer using modified method from previous reports.25 Briefly, a suspension of the catalyst was prepared by suspending 1.0 mg of catalyst in 1 mL of water with a diluted concentration (5 vol. %) of Nafion polymer (5 wt %). The suspension was ultrasonicated using a pulse ultrasonic probe for 10 min or until a dark, uniform ink was achieved. A total of 10 μL of the suspension was then transferred and uniformly distributed over the surface of the polished glassy carbon disk. The electrode was dried overnight at room temperature. The electrochemical activity for ethanol oxidation reaction was measured using the hydrodynamic rotating disk electrode technique. The standard three-electrode configuration was used for the cell, and the reference and counter electrodes were in separate compartments of the electrochemical cell. Cyclic voltammetry (CV) was performed at room temperature to clean the catalyst surface on the electrode. Optimal grade KOH pellets were diluted with Milli-Q water to produce a concentration of 0.5 M as the electrolyte. The electrolyte was deaerated with high purity nitrogen before the measurement. All potentials are given with respect to the reference electrode of Ag/AgCl with saturated KCl, which was +0.20 V with respect to NHE reference electrode. For the electrolysis experiment, high surface area vitreous carbon was used as the working electrode, on which the catalyst was loaded in a similar fashion as loading catalyst on the glassy carbon electrode. The electrolysis experiment were all carried out in a well-sealed cell with the electrolyte containing 0.5 M KOH/0.5 M ethanol. A constant potential (0.24−0.26 V) was applied to the catalyst-loaded glassy carbon electrode, and the current was recorded. After electrolysis, the solution was neutralized by an N2-purged concentrated H2SO4 solution, adjusting to a pH of 2. The gas sample was taken using a syringe of 30 mL volume and was analyzed by GC. Gas Phase Analysis. The catalytic activity of the catalysts for gas phase ethanol (∼1.5 vol. %, 15 vol. % O2 balanced by N2) oxidation reaction was measured using a customer-built system including a temperature-controlled reactor, gas flow/mixing/injection controllers, and an online gas chromatograph (GC, Shimadzu 8A) equipped with 5A molecular sieve and Porapak Q packed columns and a thermal conductivity detector. The catalytic activity was determined by analyzing the composition of the tail gas effusing from the quartz micro reactor packed with catalyst fixed bed. The GC was also used for the analysis of the composition of the gas phase above the electrolyte in a sealed electrolysis cell.

carbonyl species and to form a complex with ethanol to control the mode of the ethanol adsorption on the Pt catalyst for a favorable activation of the C−C bond. These studies have provided some useful information for the design of catalysts for achieving C−C bond cleavage. The exploration of Pt- and Pd-based alloy catalysts clearly holds promises for the design of catalysts for electrocatalytic oxidation of alcohols. However, little has been established in the understanding of how the detailed structures in alloying Pd with different base metals influence the catalytic or electrocatalytic activity. In view of recent understanding of the correlation of alloying structures of binary/ternary nanoalloys in a number of related catalytic or electrocatalytic reactions,20−23 the ability to control the nanoscale surface, composition, and structure is essential for achieving the synergy of multifunctionality in terms of catalytic activation of C−C and C−H bonds and oxygenation. In this report, we describe novel findings of an investigation of nanoalloy catalysts of Pd alloyed with oxophilic transition metals such as Co and Cu for electrocatalytic oxidation of ethanol in alkaline electrolytes. The focus is to demonstrate that the nanoscale surface structure and the oxophilicity of the alloyed base metals produce a synergistic electrocatalytic activity of the Pd nanoalloy catalysts for EOR.

■

EXPERIMENTAL SECTION

Chemicals. Platinum(II) acetylacetonate (Pt(acac)2, 97%), potassium palladium(II) chloride (K2PdCl4, 99.99%), copper(II) acetylacetonate (Cu(acac)2, 97%), Co(III) acetylacetonate (Co(acac)3, 99.95%), octyl ether ([CH 3 (CH 2 ) 7 ] 2 O, 99%), oleylamine (CH 3 (CH 2 ) 7 CH CH(CH 2 ) 8 NH 2 , 70%), and oleic acid (CH3(CH2)7CH CH(CH2)7COOH, 99+%) were purchased from Aldrich and used as received. Other chemicals such as ethanol, methanol, hexane, and concentrated H2SO4 (98 wt %) were purchased from Fisher Scientific. Vulcan carbon XC-72 was from Cabot. Reticulated vitreous carbon (RVC, pore # 30 PPI) was obtained from Ultramet. Pd (20% on activated carbon (Pearlman’s catalyst), unreduced, 50% water wet paste (Escat 1951, BASF Kit)) was obtained from Strem Chemicals. Synthesis of Nanoalloys and Preparation of Catalysts. The synthesis of PdCu and PdCo nanoparticles followed the protocol reported in our previous reports24 with slight modifications. Briefly, palladium(II) acetylacetonate and copper(II) acetylacetonate in a controlled molar ratio were dissolved into octyl ether solvent. 1,2Hexadecanediol was added as the reducing agent. With the elevated temperature, the metal precursors started to decompose, and the solution became dark. When it was heated to 105 °C, oleic acid and oleylamine was added as the capping agent under N2 atmosphere. Then, N2 purging was stopped and the mixture was heated up to 220 °C with reflux for 0.5 h. The color of the solution appeared black and was collected until it was cool to room temperature. The nanoparticles were precipitated out by adding ethanol and centrifuging. The sample was dispersed in hexane solvent for further use. The PdCo nanoparticles were synthesized using a similar protocol. The catalysts were prepared from these as-synthesized nanoparticles for the characterizations, including assembling, activation, ink preparation, and coating on the electrode. Carbon black XC-72 was used as support materials which were suspended in hexane solvent and sonicated in ice bath for 3 h. A controlled amount of as-synthesized nanoparticles was added into the solution followed by sonication and overnight stirring. The carbon-supported nanoparticles were obtained by removing the solvent. The carbon-supported nanoparticles were activated by thermochemical processing, details of which were reported previously.25 Typically, the carbon-supported sample was treated at 260 °C under 20%O2/80%N2 for 1 h for removing the organic shells and at 400 °C under 15%H2/85%N2 for 2 h for calcining the catalysts. The commercial Pd/C was treated at 400 °C under 15%H2/85%N2 for 1

■

RESULTS AND DISCUSSION Structural Characterization of the Catalysts. The bimetallic composition of the Pd nanoalloys was analyzed using the DCP-AES method. The composition in the Pd alloy nanoparticles, including PdCo and PdCu, can be well controlled by controlling the feeding composition in the

B

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

enrichment between the two alloy particles is important because it provides information for assessing the electrocatalytic activity, as discussed in the next subsection. Figure 3A shows the XRD pattern for a sample of Pd63Co27 /C after thermal treatment, indicating the formation of an alloy. The observed diffraction peaks with 2θ value can be indexed as 41.1 (111), 47.6 (200), 69.9 (220), and 84.4 (311) reflection of fcc structure for Pd63Co27 /C samples. Compared to the database for single metal systems (e.g., 2θ value of 40.1 (111), 46.7 (200), 68.2 (220), and 82.1 (311) for Pd and 44.3 (111), 51.6 (200), and 75.9 (220) for Co), the diffraction peaks for Pd63Co27 shift to higher diffraction angles than those for Pd, indicating a reduction of lattice constant. The replacement of a Pd atom by Co atom with small radius is believed to lead to the observation of lattice constant reduction. The fcc-type lattice constant is found to be 0.381 nm. The particle size determined from the peak width in the XRD pattern is 7.1 nm, slightly larger than the TEM-determined size. Figure 3B shows XRD patterns for sample Pd 21 Cu 79 supported on carbon after thermal treatment. This pattern is mostly characteristic of Pd21Cu79 alloy phase with fcc structure. However, the presence of copper oxide (2θ ≈ 35.6 and 38.6) was detected, which results from the thermal treatment. The observed diffraction peaks with 2θ value can be indexed as 42.3 (111), 49.0 (200), and 72.0 (220) reflection of fcc structure for Pd21Cu79 /C samples. Similar as observations of Pd63Co27 diffraction peaks, the diffraction peaks for Pd21Cu79 shift to higher diffraction angles than those for Pd compared to the database for single metal systems (e.g., 2θ value of 40.1 (111), 46.7 (200), 68.2 (220), and 82.1 (311) for Pd and 43.5 (111), 50.4 (200), and 74.0 (220) for Cu), which indicates a reduction of lattice constant. The fcc-type lattice constant is found to be 0. 371 nm. The replacement of a Pd atom by Cu atom with small radius is believed to lead to the observation of lattice constant reduction. Note that the presence of CuOx was also

synthesis. We have synthesized Co and PdCu with a series of bimetallic compositions. In this paper, we focused on the studies of one composition for each for assessing the effect of alloying of Pd with Cu or Co on the atomic configurations and the catalytic properties. Figure 1 shows a set of TEM images for the as-synthesized Pd63Co37 and Pd21Cu79 nanoparticles. The Pd63Co37 nano-

Figure 1. TEM micrographs of as-synthesized Pd63Co37 nanoparticles (4.9 ± 0.7 nm; A) and Pd21Cu79 nanoparticles (∼9.7 ± 0.6 nm; B).

particles exhibit an average size of 4.9 ± 0.7 nm. The Pd21Cu79 nanoparticles show an average size of 9.7 ± 0.6 nm. The nanoparticles were supported on carbon materials, and followed by thermal treatment to calcine the alloy nanoparticles. The detailed crystallinity and bimetallic distribution of these alloy catalysts were further examined using HAADFSTEM and EDS mapping techniques. As shown in Figure 2, both particles exhibit crystalline facets, and there is some degree of surface enrichment. In the case of Pd21Cu79 there is some surface enrichment of Pd in a very thin layer of the alloyed surface. In the case of Pd63Co37, there is clear surface enrichment of Co in a thick layer of the alloyed surface which does not seem to fully cover the surface of the nanoparticles. The finding of this sharp contrast in surface

Figure 2. HAADF-STEM image (the left image) and EDS composition mapping (the right three images). (A) Pd63Co37 /C after thermal treatment at 400 °C. (B) Pd21Cu79 /C after thermal treatment at 400 °C. C

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 3. XRD patterns for Pd63Co37/C (A) and Pd21Cu79/C (B) catalysts after thermal treatment at 400 °C.

Figure 4. (A) Comparisons of Pd mass specific ethanol oxidation rates of Pd21Cu79/C with those of Pd/C catalysts at two different states: asprepared [PdCu/C (a) and Pd/C (c)] and oxidized states [PdCu/C (b) and Pd/C (d)]. The Pd mass specific ethanol oxidation rate is normalized against the metal loading on carbon and the Pd composition in the nanoparticles. (The standard deviation of the data is ±0.06 × 10−6 mol gPd−1 s−1, 0.4% (RSD)). The flow rate of ethanol was about 0.5 mL/min (1.5 vol. % ethanol, 15 vol. % O2 balanced by N2). (B) Comparisons of normalized selectivity (Nor’d) CO2 and CH3CHO for Pd/C (I) and Pd21Cu79/C (II) catalysts at two different states as-prepared (Nor’d CO2 selectivity (curve a′), Nor’d CH3CHO selectivity (curve a)) and oxidized states ((Nor’d CO2 selectivity (curve b′), Nor’d CH3CHO selectivity (curve b)). Note that a fraction of acetic acid was also detected in the products for the case of Pd/C catalyst (I), where little acetic acid was detected in the case of PdCu/C catalyst (II). The Nor’d selectivity was calculated as follows: Nor’d CO2 = CO2 conversion/(CO2 conversion + CH3CHO conversion) × 100% and Nor’d CH3CHO = CH3CHO conversion/(CO2 conversion + CH3CHO conversion) × 100%, assuming in each case that CO2 and CH3CHO are the only two products. This assumption was only for the purpose of comparing the relative change of these two products.

detected by the observation of the diffraction peaks at 2θ = 35.4° and 38.6°, which reflected a certain degree of oxidation of copper due to the high percentage of copper in the Pd21Cu79 nanoalloys. The particle size determined from the peak width in the XRD pattern is 6.6 nm, slightly smaller than the TEMdetermined size. Catalytic Oxidation of Ethanol in the Gas Phase. The catalytic oxidation of ethanol in gas phase over Pd alloy catalysts was first examined to evaluate activity and selectivity. Figure 4 compares Pd mass specific ethanol oxidation rates between Pd21Cu79/C with those of Pd/C catalysts at asprepared and oxidized states. In gas phase, the Pd/C catalyst shows higher catalytic activities for ethanol conversion than those over the PdCu/C catalyst. In both cases, the activity drops upon the oxidation treatment of the catalysts. However, the alloyed catalyst shows a less degree of decrease upon the

oxidation treatment, suggesting that the fresh alloy catalyst is likely partially oxidized. In Figure 4B, the catalytic selectivity data are compared in terms of CO2 and CH3CHO products (and CH3COOH, not shown) detected for the catalytic oxidation of ethanol over Pd/ C and Pd21Cu79/C catalysts. For Pd/C, the selectivity for forming CO2 is about the same as that for forming CH3CHO for both fresh and oxidized states. Note however that a significant fraction of CH3COOH (∼50%) was detected in the gas phase as soon as CO2 and CH3CHO were detected. In comparison, a similar trend is observed for Pd21Cu79/C, but at a higher temperature range, indicating a lower catalytic activity. However, almost no CH3COOH was detected when CO2 and CH3CHO were detected. Although the catalytic conversion rate is higher for Pd/C than PdCu/C, the selectivity appears to be quite different. It is evident that, although aldehyde is involved in both cases, the alloy PdCu catalyst appears to suppress the D

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Figure 5. (A) Cyclic voltammetric curves for a 20% Pd/C catalyst (on GC electrode, 0.196 cm2) in 0.5 M KOH with 0.5 M ethanol. Scan rate: 20 mV/s. Insert: CV curve in the absence of ethanol. (B) Cyclic voltammetric curves for a 27% Pd63Co37/C catalyst (on GC electrode, 0.196 cm2) in 0.5 M KOH with 0.5 M ethanol. Insert: CV curve in the absence of ethanol. Scan rate: 20 mV/s. Reference electrode: Ag/AgCl with saturated KCl.

Figure 6. Cyclic voltammetric curves for Pd/C (20%) and Pd63Co37/C (27%), (on GC electrode, 0.196 cm2) in 0.5 M KOH with 0.5 M methanol ([A] Pd/C (a) and Pd63Co37/C (b) and 0.5 M ethanol ([B] Pd/C (a) and Pd63Co37/C (b)). Inserts: comparisons of mass activities. Scan rate: 20 mV/s. Reference electrode: Ag/AgCl with saturated KCl.

Electrocatalytic Oxidation of Ethanol in Alkaline Electrolyte. As shown in Figure 5A for commercial Pd/C in 0.5 M KOH with 0.5 M ethanol, the peak current for the ethanol oxidation is observed at −0.20 V, with an onset potential at ∼−0.6 V. The anodic wave observed in the reversed sweep is attributed to the oxidation of ethanol upon the removal of oxygenated species such as OH− and possible intermediate species such as CO on the Pd surface formed. This ethanol oxidation wave is observed at −0.34 V, which indeed occurs at the potential near the reduction peak potential for oxygenated species on Pd surface, i.e., −0.34 V as shown by the insert in Figure 5A. This feature is characterized by a steep rise of current at −0.30 V and a gradual fall with the end potential approaching the onset potential, and a narrower peak width as well. The magnitude of the peak currents for the reverse wave is about the same as that for the forward wave. Figure 5B shows a typical CV curve of Pd63Co37/C catalyst in 0.5 M KOH with 0.5 M ethanol. In this case, the forward peak is observed at −0.19 V, whereas the reverse wave is at −0.32 V,

formation of acetic acid, which is the significant portion of the product in the case of Pd catalyst. These differences are indicated by the relative percentages of products detected in the following reaction schemes (1a, 1b) for both oxidized catalysts (at 200 °C). Pd/C

CH3CH 2OH ⎯⎯⎯⎯⎯→ CO2 (∼ 6%) + CH3CHO(∼ 17%) + CH3COOH(∼ 77%)

(1a)

PdCu/C

CH3CH 2OH ⎯⎯⎯⎯⎯⎯⎯⎯→ CO2 (∼ 6%) + CH3CHO(∼ 94%) + CH3COOH(∼ 0(< 5%))

(1b)

The reaction product distribution showed a clear difference between the Pd/C and PdCu/C catalysts. These results were supported by multiple GC analyses under varied column temperatures (Figure S1). No CO and H2 were detected for PdCu catalysts, but a small amount of CO ( E Cu(OH)2 ≫ E Co(OH)2. This difference, as documented by studies of Pd, Cu, and Co electrode in alkaline electrolytes,26−28 suggests that the propensity of Co to form Co(OH)2 is perhaps too strong than Cu to form Cu(OH)2 so that Pd−OHads species can not be easy to be removed by Co alloyed in Pd. In contrast, Cu alloyed in Pd may have the ability to transfer OHads between Pd to Cu sites, thus facilitating the process of providing activated −OHads species or remove unwanted OHads species, which play an important role in the rate determining step (eq 4). The difference of this type of oxophilicity of the base metals alloyed in Pd provides one possible explanation for the activity enhancement of Cu over Co in the Pd nanoalloy catalysts for the electrocatalytic oxidation of ethanol in alkaline electrolyte. We also attempted to analyze the product of CO2 by sampling the gas sample from the sealed top compartment of the electrolysis cell with PdCu/C using high surface area VC (vitreous carbon, pore # 30 PPI) as an electrode support. The electrolysis was performed in 0.5 M KOH/0.5 M EtOH at 0.24−0.26 V (vs Ag/AgCl, Sat’d KCl) for 20−22 h. Control experiments were performed to ensure that the detected CO2 was not from ambient air. The electrolyte was an 0.5 M KOH, which was prepared by either freshly dissolving 2.8 g of KOH pellets in 100 mL of N2-purged H2O (trial A), or overnight N2purging an aged 0.5 M KOH electrolyte (trial B). Figure 10 shows two sets of gas chromatograms for the gas sample from the sealed top compartment of the electrolysis cell and control cell. Although CO2 was detected in both the electrolysis and control cells, the CO2 peak is clearly much higher from the electrolysis cell than that in the control. This difference is substantiated by the similar amount of water and ethanol being detected between the electrolysis and the control cells. The electrolysis in both trials was run for a similar period of time.

Pd − (CH3CH 2OH)ads + 3OH− → Pd − (CH3CO)ads + 3H 2O + 3e−

Pd(OH)2 + 2e ↔ Pd + 2OH−

(4) H

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

Table 3. GC Detected CO2 in Comparison with CO2 Calculated from the Electrolysis Charge

a

PdCu/C

CO2 (integrated GC peak)

CO2 (vol. %)

CO2 conc. (×10−7)a

electrolysis (C)

CO2 conc. (×10−7)b

trial 1 (control 1) trial 2 (control 2)

0.97 (0.26) 0.83 (0.38)

1.31 (0.36) 1.13 (0.52)

3.25

6.75

0.93

2.71

18.48

2.13

Calculated from GC peak (mol/mL). bCalculated from electrolysis charge (mol/mL) (assuming 100%).

■

However, the amount of catalysts loaded on the RVC electrode for trail B (5 mg) was more than that in trail A by a factor of 2. As shown in Table 3, almost no aldehyde was detected, but trace of CH3COOH was detected. Since CH3COOH is soluble in the acid-treated electrolyte, the detected trace of CH3COOH in the gas phase is a fraction of the total CH3COOH produced, suggesting CH3COOH is part of the products. While no CH3COOH was detected, trace of aldehyde was detected in both trials. Although quantitative aspects of the electrolysis data and the GC data are yet to be further delineated, the data did provide an indication that CO2 was detected. In both trials, the detected amounts of CO2 were above the control experiments. This finding suggests the possibility of the electrocatalytic oxidation involving C−C cleavage over the PdCu catalysts. However, the confirmation of this finding still needs a detailed analysis of the products, including radio labeling of the ethanol molecules or FTIR analysis, which is part of our ongoing investigations.

ASSOCIATED CONTENT

* Supporting Information S

Addition figure (Figure S1) as discussed in the text. This material is available free of charge via the Internet at http:// pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The work is supported in part by NYSERDA (No. 18514), and the National Science Foundation (CHE 0848701, CMMI 1100736).

■

■

CONCLUSIONS In conclusions, the results have demonstrated that the enhancement of the catalytic and electrocatalytic activity of Pd alloyed with transition metals (Co and Cu) for EOR depends on the chemical nature of the nanaoalloy. In comparison with Pd, the PdCo catalyst exhibits a similar mass activity for the electrocatalytic EOR in alkaline electrolyte. However, the PdCu catalyst is shown to display a much higher mass activity. This finding is consistent with the finding on the surface enrichment of Pd on the alloyed PdCu surface, in contrast to the surface enrichment of Co in the alloyed PdCo surface. In the gas-phase reaction, although the catalytic conversion rate for CO2 product is higher over Pd than PdCu, the catalytic activity over the nanoalloy PdCu catalyst is shown to suppress the formation of acetic acid, which is a significant portion of the product in the case of Pd catalyst. In the alkaline electrolyte, CO2 was detected from the gas phase above the electrolyte upon acid treatment following the electrolysis, along with traces of aldehyde and acetic acid. An analysis of the electrochemical properties of the metal components indicates that the oxophilicity of the base metals (Cu and Co) alloyed with Pd, in addition to the surface enrichment of metals, may have played an important role in the observed difference of the catalytic and electrocatalytic activities. In comparison with Pd alloyed with Co, the results for Pd alloyed with Cu showed a more significant positive shift of the reduction potential of the oxygenated Pd species on the surface. Although the current work has focused on demonstrating the viability of Pd alloy for ethanol oxidation, our future work will determine the mechanistic details of this reaction.29 A quantitative delineation of the correlation between the catalytic electrocatalytic activity and the base metal’s oxophilicity in the nanoalloy catalyst is also part of our ongoing work.

REFERENCES

(1) Zhou, W. J.; Zhou, B.; Li, W. Z.; Song, S. Q.; Sun, G. Q.; Xin, Q.; Douvartzides, S.; Goula, M.; Tsiakaras, P. Performance comparison of low-temperature direct alcohol fuel cells with different anode catalysts. J. Power Source 2004, 126, 16−22. (2) Tadanaga, K.; Furukawa, Y.; Hayashi, A.; Tatsumisago, M. Direct Ethanol Fuel Cell Using Hydrotalcite Clay as a Hydroxide Ion Conductive Electrolyte. Adv. Mater. 2010, 22, 4401−4404. (3) Dimos, M. M.; Blanchard, G. J. Evaluating the Role of Pt and Pd Catalyst Morphology on Electrocatalytic Methanol and Ethanol Oxidation. J. Phys. Chem. C 2010, 114, 6019−6026. (4) Liang, Z. X.; Zhao, T. S.; Xu, J. B.; Zhu, L. D. Mechanism study of the ethanol oxidation reaction on palladium in alkaline media. Electrochim. Acta 2009, 54, 2203−2208. (5) Shen, S.; Zhao, T. S.; Xu, J.; Li, Y. High performance of a carbon supported ternary PdIrNi catalyst for ethanol electro-oxidation in anion-exchange membrane direct ethanol fuel cells. Energy Environ. Sci. 2011, 4, 1428−1433. (6) Mahapatra, S. S.; Dutta, A.; Datta, J. Temperature effect on the electrode kinetics of ethanol oxidation on Pd modified Pt electrodes and the estimation of intermediates formed in alkali medium. Electrochim. Acta 2010, 55, 9097−9104. (7) Wu, H.; Li, H.; Zhai, Y.; Xu, X.; Jin, Y. Facile Synthesis of FreeStanding Pd-Based Nanomembranes with Enhanced Catalytic Performance for Methanol/Ethanol Oxidation. Adv. Mater. 2012, 24, 1594−1597. (8) Jou, L.-S.; Chang, J.-K.; Twhang, T.-J.; Sun, I.-W. Electrodeposition of Palladium-Copper Films from 1-Ethyl-3-methylimidazolium Chloride-Tetrafluoroborate Ionic Liquid on Indium Tin Oxide Electrodes. J. Electrochem. Soc. 2009, 156, D193−D197. (9) Zhang, Z.; Xin, L.; Sun, K.; Li, W. Pd-Ni electrocatalysts for efficient ethanol oxidation reaction in alkaline electrolyte. Int. J. Hydrogen Energy 2011, 36, 12686−12697. (10) Maiyalagan, T.; Scott, K. Performance of carbon nanofiber supported Pd-Ni catalysts for electro-oxidation of ethanol in alkaline medium. J. Power Source 2010, 195, 5246−5251. (11) Nie, M.; Tang, H.; Wei, Z.; Jiang, S. P.; Shen, P. K. Highly efficient AuPd-WC/C electrocatalyst for ethanol oxidation. Electrochem. Commun. 2007, 9, 2375−2379. I

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX

Langmuir

Article

(12) Lee, Y. W.; Kim, M.; Kim, Y.; Han, S. W. Synthesis and Electrocatalytic Activity of Au-Pd Alloy Nanodendrites for Ethanol Oxidation. J. Phys. Chem. C 2010, 114, 7689−7693. (13) Lee, E.; Park, I.-S.; Manthiram, A. Synthesis and Characterization of Pt-Sn-Pd/C Catalysts for Ethanol Electro-Oxidation Reaction. J. Phys. Chem. C 2010, 114, 10634−10640. (14) Oliveira, M. C.; Rego, R.; Fernandes, L. S.; Tavares, P. B. Evaluation of the catalytic activity of Pd−Ag alloys on ethanol oxidation and oxygen reduction reactions in alkaline medium. J. Power Source 2011, 196, 6092−6098. (15) Liu, J.; Zhou, H.; Wang, Q.; Zeng, F.; Kuang, Y. Reduced graphene oxide supported palladium-silver bimetallic nanoparticles for ethanol electro-oxidation in alkaline media. J. Mater. Sci. 2012, 47, 2188−2194. (16) Du, W.; Mackenzie, K. E.; Milano, D. F.; Deskins, N. A.; Su, D.; Teng, X. Palladium-Tin Alloyed Catalysts for the Ethanol Oxidation Reaction in an Alkaline Medium. ACS Catal. 2012, 2, 287−297. (17) Serov, A.; Martinez, U.; Falase, A.; Atanassov, P. Highly active Pd-Cu catalysts for electrooxidation of 2-propanol. Electrochem. Commun. 2012, 22, 193−196. (18) Kowal, A.; Li, M.; Shao, M.; Sasaki, K.; Vukmirovic, M. B.; Zhang, J.; Marinkovic, N. S.; Liu, P.; Frenkel, A. I.; Adzic, R. R. Ternary Pt/Rh/SnO2 electrocatalysts for oxidizing ethanol to CO2. Nat. Mater. 2009, 8, 325−330. (19) He, Q.; Shyam, B.; Macounová, K.; Krtil, P.; Ramaker, D.; Mukerjee, S. Dramatically Enhanced Cleavage of the C-C Bond Using an Electrocatalytically Coupled Reaction. J. Am. Chem. Soc. 2012, 134, 8655−8661. (20) Wanjala, B. N.; Fang, B.; Shan, S.; Petkov, V.; Zhu, P.; Loukrakpam, R.; Chen, Y.; Luo, J.; Yin, J.; Yang, L.; Shan, S.; Shao, M.; Zhong, C. J. Design of Ternary Nanoalloy Catalysts: Effect of Nanoscale Alloying and Structural Perfection on Electrocatalytic Enhancement. Chem. Mater. 2012, 24, 4283−4293. (21) Wanjala, B.; Fang, B.; Luo, J.; Chen, Y.; Yin, J.; Engelhard, M.; Loukrakpam, R.; Zhong, C. J. Correlation between Atomic Coordination Structure and Enhanced Electrocatalytic Activity for Trimetallic Alloy Catalysts. J. Am. Chem. Soc. 2011, 133, 12714− 12727. (22) Yang, L.; Shan, S.; Loukrakpam, R.; Petkov, V.; Ren, Y.; Wanjala, B.; Engelhard, M.; Luo, J.; Yin, J.; Chen, Y.; Zhong, C. J. Role of support-nanoalloy interactions in the atomic-scale structural and chemical ordering for tuning catalytic sites. J. Am. Chem. Soc. 2012, 134, 15048−15060. (23) Loukrakpam, R.; Luo, J.; He, T.; Chen, Y. S.; Xu, Z.; Njoki, P.; Wanjala, B.; Fang, B.; Mott, D.; Yin, J.; Klar, J.; Powell, B.; Zhong, C. J. Nanoengineered PtCo and PtNi Catalysts for Oxygen Reduction Reaction: An Assessment of the Structural and Electrocatalytic Properties. J. Phys. Chem. C 2011, 115, 1682−1694. (24) Luo, J.; Han, L.; Kariuki, N.; Wang, L.; Mott, D.; Zhong, C. J.; He, T. Synthesis and Characterization of Monolayer-Capped PtVFe Nanoparticles with Controllable Sizes and Composition. Chem. Mater. 2005, 17, 5282−5290. (25) Luo, J.; Wang, L.; Mott, D.; Njoki, P.; Kariuki, N.; Zhong, C. J.; He, T. Ternary Alloy Nanoparticles with Controllable Sizes and Composition and Electrocatalytic Activity. J. Mater. Chem. 2006, 16, 1665−1673. (26) Shim, H.-S.; Shinde, V. R.; Kim, H. J.; Sung, Y.-E.; Kim, W. B. Porous cobalt oxide thin films from low temperature solution phase synthesis for electrochromic electrode. Thin Solids Films 2008, 516, 8573−8578. (27) Birss, V. I.; Chan, M.; Phan, T.; Vanýsek, P.; Zhang, A. An electrochemical study of the composition of thin, compact Pd oxide films. J. Chem. Soc., Faraday Trans. 1996, 92, 4041−4047. (28) Lide, D. R. Handbook of Chemistry and Physics, 77th ed.; CRC Press, Inc.: Boca Raton FL, 1996. (29) Lai, S. C. S.; Kleijn, S. E. F.; Ö ztürk, F. T. Z.; Vellinga, V. C. v. R.; Koning, J.; Rodriguez, P.; Koper, M. T. M. Effects of electrolyte pH and composition on the ethanol electro-oxidation reaction. Catal. Today 2010, 154, 92−104. J

dx.doi.org/10.1021/la401839m | Langmuir XXXX, XXX, XXX−XXX