The preparation and characterization of nano-sized Pt–Pd/C catalysts and comparison of their superior catalytic activities for methanol and ethanol oxidation
- First Online:
- Cite this article as:
- Ozturk, Z., Sen, F., Sen, S. et al. J Mater Sci (2012) 47: 8134. doi:10.1007/s10853-012-6709-3
- 637 Downloads
In this study, two groups of carbon supported PtPd samples with different percentages of metals were prepared to examine the effects of Pd and stabilizing agents on the catalytic activity towards methanol and ethanol oxidation reactions. As a stabilizing agent, 1-hexanethiol and 1,1-dimethyl hexanethiol were used for group “a” and “b” catalysts, respectively. Cyclic voltammetry, chronoamperometry, X-ray diffraction (XRD), transmission electron microscopy (TEM), energy dispersive X-ray spectroscopy, and X-ray photoelectron spectroscopy (XPS) were employed to understand the nature of the prepared catalysts. TEM and XRD results indicated a similar size distribution of the metal nanoparticles with a narrow average crystallite size of 3.0–3.7 nm. XPS data revealed the presence of two different oxidation states for both platinum and palladium, being Pt(0), Pt(IV), Pd(0), and Pd(II). Electrochemical studies indicated that the group “b” type catalysts have a higher catalytic activity than group “a”. The most active catalyst was found to be a carbon supported 88 %Pt/12 %Pd prepared with 1,1-dimethyl hexanethiol, which has an activity of ~5 times (~0.450 A/mg Pt at 0.57 V for methanol) and ~14 times (~0.350 A/g Pt at 0.56 V for ethanol) greater than the commercial E-TEK catalyst.
The harmful products, such as sulfur dioxide (SO2), carbon monoxide (CO), and nitrogen oxides (NOx), formed during power generation from the combustion of fossil fuels and the rapid consumption rate of limited fossil fuel sources led to an interest in the development of fuel cells [1, 2]. Among the many kinds of fuel cells available, direct alcohol fuel cells are a promising future technology, e.g., direct methanol fuel cells (DMFCs) (a subcategory of direct alcohol fuel cells) have been receiving a great attention owing to high energy content per unit mass, ease of storage, transportation, handling, availability, low pollutant emission [3, 4, 5], and low cost of methanol [6, 7]. Ethanol can also be used as a fuel in direct alcohol fuel cells due to the same advantages stated above and the non-toxic feature of ethanol (usually called a green fuel) [8, 9, 10, 11]. Although methanol and ethanol are hopeful candidates as fuels for direct alcohol fuel cells, further developments are required, in terms of enhancement of catalyst performance, because these alcohols have no electrochemical activity in acidic solutions.
It is known that the improvement of more active catalysts is a key point for direct alcohol fuel cells and for that reason this is the focus of this study. Even though Pt-based catalysts are known as the best electrocatalysts for these types of fuel cells [12, 13, 14], methanol and ethanol oxidation reactions require a significant overpotential. In addition to this, high cost and limited platinum resources restrict the uses of these catalysts in practical applications [15, 16]. In order to overcome these problems, one of the most common approaches is to use another metal, such as Ru [17, 18, 19], Sn , Rh , and Mo , along with platinum and some of them have produced quite encouraging results. For this reason, in this study palladium was utilized as a second metal and PtPd nanoparticles were dispersed on a carbon support.
It is well known that the small size of nanoparticles increases the surface area of precious metals in the catalysts and consequently the activity of catalysts; ultimately decreasing the cost of the fuel cells. A problem with this approach is the agglomeration of PtPd nanoparticles. In order to prevent this in these nanoparticles [5, 23, 24, 25, 26], 1-hexanethiol (group “a” in our study) and 1,1-dimethyl hexanethiol (group “b”) were employed as stabilizing agents for the first time for this aim. Cyclic voltammetry (CV), chronoamperometry (CA), X-ray diffraction (XRD), transmission electron microscopy (TEM), energy dispersive X-ray spectroscopy (EDS), and X-ray photoelectron spectroscopy (XPS) were used for physical and electrochemical characterization of the prepared catalysts.
PtCl4 (99 %) and PdCl2 (59 %) were purchased from Alfa, tetrahydrofuran (THF, 99.5 %), methanol (≥99.5), ethanol (99.9 %), and HClO4 (60 %) were obtained from Merck, lithium triethylborohydride (superhydride, 1.0 M dissolved in THF) was bought from Aldrich, 1,1-dimethyl hexanethiol (C8H18S) (85 %), and hexanethiol (95 %) were acquired from ABCR, and carbon XC-72 was purchased from Cabot Europa Ltd.
Water was purified using a Millipore water purification system (18 MΩ) analytical grade.
Synthesis of catalyst Ia
A 0.24 mmol of PtCl4 and 0.050 mmol PdCl2 were dissolved in 25 ml of anhydrous tetrahydrofuran for 1 h, then 0.27 mmol of 1-hexanethiol was added to this solution and the mixture was stirred vigorously for 2.5 h. The platinum and palladium species were reduced by the drop wise addition of lithium triethylborohydride. All these steps were carried out under a pure argon atmosphere. An observed color change from yellowish to black was the indication of formation of Pt–Pd nanoparticles, therefore addition of superhydride was maintained till this color change was observed . The resulting solution was washed with dry ethanol in an ultrasonic bath, just before centrifugation, in order to remove excess thiols . This process was repeated until a clear filtrate solution was obtained. Afterwards, the solid precipitate was dried under vacuum at room temperature. A 1:10 ratio of metal nanoparticles to carbon XC-72 were mixed in ethanol for 2 days and the solvent removed under vacuum at room temperature.
Synthesis of catalysts IIa, IIIa, Ib, IIb, and IIIb
The name of catalysts, surfactants, and atomic percentage of platinum and palladium which are added to the reaction medium (A) and obtained from ICP results (B)
90 %Pt/10 %Pd
87.4 %Pt/12.6 %Pd
80 %Pt/20 %Pd
76.8 %Pt/23.2 %Pd
70 %Pt/30 %Pd
65.1 %Pt/34.9 %Pd
90 %Pt/10 %Pd
88.2 %Pt/11.8 %Pd
80 %Pt/20 %Pd
81.3 %Pt/18.7 %Pd
70 %Pt/30 %Pd
72.1 %Pt/27.9 %Pd
Transmission electron microscopy
A JEOL 200 kV TEM instrument was used to acquire TEM images. Sample preparation was performed through the suspension of ~0.5 mg catalyst in 3 ml of ethanol in an ultrasonic bath and this was followed by the placement of a drop of this solution on to a carbon covered 400-mesh copper grid. Finally, the solvent was evaporated at room temperature.
X-ray photoelectron spectroscopy
A Specs spectrometer was used for XPS analysis and Kα lines of Mg (1253.6 eV, 10 mA) were utilized as an X-ray source. All XPS peaks have been fitted using Gaussian function, and the C 1s line at 284.6 eV was considered as the reference line.
A Rigaku diffractometer with Ultima + theta–theta high resolution goniometer, the X-ray generator (Cu Kα radiation, λ = 1.54056 Å) with an operation conditions at 40 kV and 40 mA were employed for the XRD analysis.
A microcomputer-controlled potentiostat/galvanostat Solartron 1285 was used for CV and CA measurements. The saturated calomel electrode (SCE), glassy carbon and prepared catalysts on glassy carbon were employed as reference, counter and working electrodes, respectively.
Inductively coupled plasma spectroscopy
The amount of platinum and palladium in each catalyst was investigated by a Leeman Lab inductively coupled plasma spectroscopy (ICP).
Results and discussion
Peak positions, d, lattice parameters, atomic percentages of platinum and palladium, atomic fraction of Pd, percent palladium incorporated into PtPd alloy, and unmixed Pd %
3.927 ± 0.014
3.914 ± 0.012
3.906 ± 0.009
3.896 ± 0.007
3.925 ± 0.012
3.912 ± 0.010
3.906 ± 0.012
3.900 ± 0.011
Pt 4f7/2 and Pd 3d3/2 core binding energies, eV, for carbon supported PtPd catalysts (the numbers in parentheses are the relative percent intensities of the species)
Consequently, it is possible that the absolute numbers given here may incorporate some error, since absolute quantification in XPS involves a number of approximations, the relative amounts, however, should be unaffected. The main species is undoubtedly Pt(0) (between 85 and 67 %) and Pd(0) (between 93 and 74 %) for all catalysts, and the Pt(0)/Pt(IV) and Pd(0)/Pd(II) ratios are at a maximum for Catalysts Ia and Ib. The Pt(0)/Pt(IV) ratio decreases going from catalyst Ia (5.67) to IIa (2.62) to IIIa (2.05) and for catalyst Ib (6.94) to IIb (5.25) to IIIb (2.41) (Table 4).
The If/Ib ratio, onset and anodic peak potentials for all catalysts in MeOH (a) and EtOH (b) electrolyte solution
Max current (A/mg Pt)
Anodic peak (V)
Max current (A/mg Pt)
Anodic peak (V)
It is well known that If/Ib ratio, where If and Ib are the currents in the forward and reverse scan, respectively, can be used to define the performance of the catalysts. The higher the If/Ib ratio, the better the alcohol oxidation to the final product would be owing to the low accumulation of intermediate species on the surface of the catalysts for methanol and ethanol oxidation reactions. The If/Ib ratios are summarized in Table 5 and found that all PtPd catalysts have higher If/Ib ratios than Pt/C(a and b) and catalyst Ib has the highest ratio which indicates the best activity performance toward methanol and ethanol oxidation reactions as expected.
The most active catalyst, Ib, was found to have a current that is ~4.4 times greater than the catalyst, Pt/C(b), after a decay of 3,600 s. The results are consistent with those obtained from CV study. In addition to these outcomes, it is possible to state that the group “b” catalysts not only have higher stability but also higher poisoning tolerance when compared with group “a” catalysts for both methanol and ethanol oxidation reactions.
In summary, electrochemical studies indicated that catalysts Ia and Ib are the most active catalysts within each group. TEM and XRD data revealed that they have similar metal particle sizes and possess the highest percent alloy formation with respect to the amount of Pd that exists in the sample, between 87.3 and 89.0 %. The insertion of Pd into the Pt fcc structure changes between 72 and 82 % for catalysts IIa, IIb, IIIa, and IIIb and the rest of Pd remains unmixed, Table 3. As it is known that methanol and ethanol oxidation take place on platinum not on palladium, therefore it is clear we should observe a decrease in the performance of these catalysts compared to catalysts Ia and Ib. Besides TEM and XRD data, XPS and electrochemical results were also examined to find the reason for the high activity performance of these two catalysts. According to these data, catalysts Ia and Ib have an extraordinary high Pt(0)/Pt(IV) (6.94, 5.67), Pd(0)/Pd(II) (13.7, 7.4), Pt/Pd (8.9, 8.6), and If/Ib (18.8, 7.54 for methanol and 3.67, 3.39 for ethanol) ratios. It is also known that although alcohol oxidation reactions occur on platinum, second metal or metal oxide or hydroxide aid this reaction. Consequently, it is obvious that these two catalysts have very appropriate reaction media to form alcohol oxidation reaction which causes to have high performance of these catalysts.
In this article, two groups of carbon supported PtPd nanoparticle catalysts were synthesized by superhydride reduction method using 1-hexanethiol (group “a” catalysts) and 1,1-dimethyl hexanethiol (group “b” catalysts) as surfactants for the first time for this goal. XRD, XPS, TEM, CV, and CA were employed for physical and electrochemical characterization of the catalysts. TEM and XRD results revealed the uniform distribution of the PtPd alloy nanoparticles on carbon support with a narrow size ranging between 2.8 and 3.6 nm. Electrochemical studies indicated that group “b” catalysts have higher catalytic activities than that of group “a”, and that catalyst Ib is the most active one. Its performance is 5.0 and 14.0 times higher than that of commercially available Pt catalyst (E-TEK) for methanol and ethanol oxidation reactions, respectively. The reason for having a good performance could be due to the higher extent of alloy formation, CO tolerance, Pt/Pd, Pt(0)/Pt(IV), and Pd(0)/Pd(II) ratios.
The authors gratefully acknowledge Türkiye Bilimsel ve Teknik Araştırma Kurumu (TUBITAK), Grant 108T065 for the financial support. F. Şen and S. Şen thank the Middle East Technical University (METU) for Grant BAP-08-11-DPT2002K120510 and TUBITAK 2211 scholarships. The authors also thank Dr. W. Michael Pitcher for the proofreading of this manuscript.