- Email: [email protected]
palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction
Accepted Manuscript Title: Novel graphitic carbon nitride/graphite carbon/palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction Author: Zesheng Li Runsheng Lin Zhisen Liu Dehao Li Hongqiang Wang Qingyu Li PII: DOI: Reference:
S0013-4686(16)30126-8 http://dx.doi.org/doi:10.1016/j.electacta.2016.01.124 EA 26506
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
2-12-2015 14-1-2016 16-1-2016
Please cite this article as: Zesheng Li, Runsheng Lin, Zhisen Liu, Dehao Li, Hongqiang Wang, Qingyu Li, Novel graphitic carbon nitride/graphite carbon/palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2016.01.124 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel graphitic carbon nitride/graphite carbon/palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction Zesheng Li a,*, Runsheng Lin a, Zhisen Liu a, Dehao Li a, Hongqiang Wang b and Qingyu Li b,* a
Development Center of Technology for Petrochemical Pollution Control and Cleaner Production of Guangdong Universitites,
College of Chemical Engineering, Guangdong University of Petrochemical Technology, Maoming, 525000, China b
Guangxi Key Laboratory of Low Carbon Energy Materials, School of Chemical and pharmaceutical Sciences,
Guangxi Normal University, Guilin, 541004, China.
Corresponence author: Zesheng Li, E-mail: [email protected]; Qingyu Li E-mail: [email protected].
Graphic abstract
Novel graphitic carbon nitride/graphite carbon/palladium (Pd@g-C3N4/GC) electrocatalyst with excellent electrocatalytic performance and good electrochemical stability are developed for the ethanol oxidation reaction in alkaline medium.
Highlights (1) Graphite carbon support was massively prepared by a cost-effective catalytic strategy. (2) Ultra-thin g-C3N4 was successfully deposited on the surface of graphite carbon support. (3) Three-phase boundary from graphite carbon, g-C3N4 and Pd was successfully formed. (4) Nitrogen-rich g-C3N4 effectively promoted the electrochemical activity of Pd metal.
Abstract: Graphitic carbon nitride (g-C3N4) possesses high nitrogen content with excellent chemical and thermal stability, which also shows inherent electrochemical properties. Due to the extremely low electronic conductivity, pure g-C3N4 is usually restricted to electrocatalytic applications. In this study, we report the synthesis of graphite carbon (GC) supported g-C3N4 and palladium (Pd) electrocatalyst with excellent electronic conductivity for the ethanol oxidation reaction. In virtue of the ultra-thin structure of g-C3N4 deposited on the surface of GC, the crucial three-phase boundary from GC, g-C3N4 and Pd component are successfully formed in Pd@g-C3N4/GC electrocatalyst. The electrocatalytic performances on ethanol electrooxidation of Pd@g-C3N4/GC, Pd@g-C3N4, Pd@GC and Pd@amorphous carbon (AC) have been systematically investigated by the techniques of cyclic voltammetry and electrochemical impedance spectroscopy. The Pd@g-C3N4/GC displays the best electrocatalytic performances with a high oxidation peak current density of 2156 A/g Pd and low on-set potential of 0.32 V for the ethanol oxidation. The peak current density of Pd@g-C3N4/GC maintains at 1904 A/g Pd after 200 continuous cyclic voltammetry cycles. The Pd@g-C3N4/GC composite material might be one of candidate electrocatalysts for the ethanol oxidation reaction in direct ethanol fuel cells. Keywords: Graphitic carbon nitride, Graphite carbon, Palladium, Electrocatalyst, Ethanol oxidation
1. Introduction Platinum-free electrocatalyst based on palladium (Pd) has received increasing attention as one of promising anode electrocatalysts for ethanol oxidation reaction (EOR) in direct ethanol fuel cells (DEFCs), due to their inherent high catalytic activity and low cost of raw material [1-3]. At present, improving the catalytic efficiency and utilization rate of Pd, reducing its usage and improving its stability for the ethanol oxidation, is the core problem for the anode electrocatalysts of DEFCs [4, 5]. It is well known that, the support material plays a key framework role in the field of electrocatalysis, which can provide large specific surface area to accommodate catalysts nanoparticles [6]. Usually, the application of suitable support material is conducive to the dispersion of Pd nanoparticles and the reduction of agglomeration, so as to improve the catalytic efficiency and utilization rate, as well as reduce the use of Pd [7, 8]. Particularly, some atoms or atomic groups in the support material can be coupled with the Pd and release catalytic active sites, such promoting the conversion of ethanol oxidation intermediates and improving the catalytic efficiency [9, 10]. Relative to the conductive metal oxides and conducting polymers, carbon materials are one of the most common support materials in electrochemical applications [11, 12]. As support materials for electrocatalysts, carbon materials possess many advantages, for example, excellent electrical conductivity, low density, low expansion rate, high chemical stability, wide range of sources and very cheap price. Graphene, carbon nanotubes, Vulcan XC-72 and mesoporous carbon are considered to be excellent carbon supports for electrocatalysts, which not only have high specific surface area but also have good electrical conductivity and favorable electron-transfer network structure [13-15]. Because of its ordered array structure and less defects in lattice, graphite carbon displays better electrical conductivity and larger current output over amorphous carbon [16,17]. In general, the amorphous carbon can be changed into highly ordered graphite carbon at very high temperature (2200-3000℃), which requires a special high-temperature equipment and enormous
energy consumption. Fortunately, the graphite carbon can be also synthesized by catalytic process in lower temperature (<1000℃) [18-20], which not only reduces the production temperature and energy consumption but also reduces the cost of equipment and operation. Graphitic carbon nitride (g-C3N4) is a new type of conjugated semiconductor material, which is seized of ordered arrangements of tri-s-triazine subunits and tertiary amino groups, and laminated structures connected with weak van der Waals force [21]. Due to its convenient preparation, low cost, environmental friendliness, nitrogen-rich composition and two-dimensional graphite-like structure, g-C3N4 has been recently attracted greatly attention in the field of both photocatalysis and electrocatalysis [22,23]. Particularly, g-C3N4-based materials have been used as electrocatalysts for hydrogen evolution [24], oxygen evolution [25], oxygen reduction reaction [26] as well as methanol oxidation [27]. However, the pristine g-C3N4 is severely restricted in the applications of various electrochemical techniques, due to its extremely low electronic conductivity [28]. The g-C3N4 is always obtained by thermal condensation process along with the volume shrinkage, which results in numerous irregular holes within the bulk g-C3N4 [29]. The result further leads to a decreased conductivity of g-C3N4, so its electrochemical performance in electro-oxidation of alcohols is not so satisfactory. Therefore, it is of fundamental interest to develop synthetic strategies for depositing thin g-C3N4 layer onto a conducting support in order to improve its conductivity essentially [30]. In the present investigation, we report a novel graphitic carbon nitride/graphite carbon/palladium nanocomposite (Pd@g-C3N4/GC) as an efficient electrocatalyst for the ethanol electro-oxidation. To best of our knowledge, this is the first report about the g-C3N4-promote Pd electrocatalyst for the application of ethanol electro-oxidation. For this composite system, several features have become apparent over previous reports: (i) the graphite carbon support with nanosheet-like structure is prepared by a cost-effective low-temperature (850 ℃) nickel catalytic route from inexpensive ion-exchange resin; (ii) the ultra-thin g-C3N4 “nano-islands” (5 nm) is successfully deposited on the
surface of graphite carbon support, which is expressly favorable for the electron transfer and ion diffusion of electrolyte; (iii) the electron-coupling effect between g-C3N4 and GC can promote the fixation of Pd nanoparticles on to the support [26]; (iv) the pyridine-type nitrogen-rich structure units of g-C3N4 can induce a large amount of electrochemical active points when interacted with the Pd metal [27]. With these merits, we investigated that the ethanol oxidation electrocatalyst with excellent performances, including high activity and stability, could be designed based on the Pd@g-C3N4/GC composite architectures.
2. Experimental 2.1. Preparation of graphite carbon (GC) support The GC support was prepared by low-temperature catalytic graphitization with acrylic acid-based cation exchange resin (D113) as a carbon source and nickel acetate as a precursor of nickel catalyst. 20 g of cation exchange resin was added into a beaker containing 200 ml of distilled water, and 2.5 g of nickel acetate was dissolved in 50 ml of distilled water with anther beaker. The nickel acetate aqueous solution was slowly dripped into the first beaker with cation exchange resin under high-speed stirring. After 3h exchange process, the nickel ions were basically adsorbed onto the surface of the resin. The resin-nickel ions complex was collected by filtration and dried at 80℃ for 24 h. The dried resin-nickel ions complex was put into a tube furnace with nitrogen atmosphere, and was heated to 850 ℃ in 2℃/min, holding for 2 h. The GC support was finally obtained after the Ni removal with 1mol/L HCl aqueous solution and smashing by high speed grinding. And the amorphous carbon (AC) was also prepared by similar steps without the use of nickel acetate. 2.2. Preparation of g-C3N4/GC composite support The g-C3N4/GC composite support was prepared by in-situ thermal polymerization of melamine on GC support. Typically, five mixtures of GC and melamine with different mass ratio of 1:0.0625,
1:0.125, 1:0.25, 1:0.5, and 1:1 were put into five beakers. 100 ml of distilled water and 50 ml of ethanol were added into each beaker. The five mixed solutions were then treated by ultrasonic dispersion at a temperature of 80 ℃ until the solvent is volatilized basically. The five mixed products were put into five corundum crucibles, sealed with aluminum papers, and were heated at 550 ℃ in nitrogen atmosphere for 4h. The g-C3N4 (n)/GC composite support (“n” stands for different mass ratio of melamine) were finally obtained after grinding. The pure g-C3N4 was also prepared by thermal polymerization of melamine at 550 ℃ in nitrogen atmosphere for 4h. 2.3. Preparation of Pd@g-C3N4/GC composite electrocatalyst The Pd@g-C3N4/GC composite electrocatalyst was prepared by an ethylene glycol-microwave heating reduction method form the Na2PdCl4 precursor. Typically, 80mg of g-C3N4/GC composite support, quantified Na2PdCl4 solution (containing 20 mg Pd) and 100ml of ethylene glycol were put into a beaker and treated by ultrasonic dispersion for 30 min at room temperature. The mixed solution was put into a microwave oven, and was heated to the boiling point and maintained for 30 seconds. The heated mixed solution was then naturally cooled down to the room temperature. The Pd@g-C3N4/GC composite electrocatalyst was finally achieved after centrifugal separation, washing and vacuum drying. At the same time, the Pd/GC, Pd@g-C3N4 and Pd/AC electrocatalyst were also prepared by the same parameter and procedure, with single GC, g-C3N4 and AC as support, respectively. 2.4. Material characterization The crystallographic textures and morphologies of the as-prepared sample were observed by X-ray diffractometer (XRD, Rigaku, D/max 2500 v/pc) with Cu-Ka1 radiation (λ=1.5406 Å), thermal field emission environment scanning electron microscopy (FE-SEM, FEI, Quanta 400), and transmission electron microscopy (TEM, JEM-2010HR). Fourier transformation infrared (FTIR) spectrum was recorded on Nicolet 6700 FTIR spectrometer (Madison, Wisconsin) with potassium
bromide as compressed slices. 2.5. Electrochemical characterization Electrochemical measurements were performed in conventional three-electrode system. The working electrode was prepared by mixing 5 mg of each catalyst, 900 µL ethanol, and 100 µL 0.5 wt% Nafion (DuPont, USA) to form slurry, followed by dipping 5 µL of the slurry onto a glass carbon electrode (0.785 cm2) (5 µg Pd for each electrode) and drying under the IR lamp. The three-electrode system was tested in a 1 mol/L KOH aqueous solution or 1 mol/L KOH + 1 mol/L C2H5OH mixed aqueous solution, with a Pt foil as the counter electrode, a reversible hydrogen electrode (RHE) as the reference electrode. The cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were recorded by an electrochemical work station (CHI 660). The CV was performed within voltage range of 0.0~1.2 V at a scan rate of 50 mV/s. The EIS was measured in the frequency range of 100 mHz to 100 kHz at an open-circuit potential with an AC perturbation of 5 mV.
3. Results and discussion 3.1. Morphology and structure The preparation process involving the GC, g-C3N4/GC as well as Pd@g-C3N4/GC nanocomposite is schematically shown in Fig. 1. Firstly, the Ni2+ can be uniformly adsorbed onto the surface and internal structures of the resin by the ion exchange in aqueous solution. Secondly, the bulk porous graphite foam was achieved after the heating treatment and Ni removal. In order to obtain nanosheet-like GC support, a high speed grinding procedure was further carried out. Thirdly, the g-C3N4/GC composite was synthesized by in-situ thermal polymerization of melamine on GC support. Expressly, ultra-thin g-C3N4 with a “nano-islands”-like discrete distribution on the surface of GC support is one striking feature for the g-C3N4/GC composite. Finally, the Pd@g-C3N4/GC composite electrocatalyst was rapidly prepared by an ethylene glycol-microwave heating reduction
method form the Na2PdCl4 precursor. Notably, the crucial three-phase boundary from GC, g-C3N4 and Pd component was successfully formed in the Pd@g-C3N4/GC composite architecture. Fig. 2 (a) reveals the general morphology of the Pd@g-C3N4/GC composite by field-emission scan electron microscopy (FE-SEM). It indicates that the sample has a relatively uniform distribution without apparent massive structures or agglomerated particles. Interestingly, the local magnification reveals that the sample is composed of a number of nanosheet-like structures with width of about 200 nm and thickness of about 15 nm (Fig. 2 (b)). The formation of the Pd nanoparticles decorated on the nanosheet-like GC support is further confirmed using transmission electron microscopy (TEM) (Fig. 2 (c)). The diameters of these Pd nanoparticles were from 8 nm to 20 nm. The high resolution (HR) TEM image (Fig. 2 (d)) distinctly shows the lattice fringes with spacings of 0.22 nm and 0.33 nm, which correspond to the (111) plane of Pd and (002) plane of GC, respectively. Moreover, a small spot of g-C3N4 (< 5 nm in thickness) deposited on GC support can be observed from Fig. 2 (d). Most importantly, the three-phase solid boundary from the GC, g-C3N4 and Pd three different components has been demonstrated via the HR-TEM image. To show further insight into the detailed nanostructure of the GC and the formation condition of three-phase solid boundary, the comparative TEM images of GC and g-C3N4(0.0625)/GC are presented in Fig. 3. The TEM images of GC clearly reveal the sample has irregular nanosheet-like structures (Fig. 3 (a)) and the thickness of nanosheet is about 10 nm (Fig. 3 (b)). The formation process of the nanosheet-like graphite carbon is mainly due to the nickel catalyst with larger grain size, in which the large and flat crystallographic plane of nickel plays a role in templating two-dimensional crystal structure of graphite carbon [19]. For g-C3N4(0.0625)/GC, some similar nanosheet-like structures can be observed (Fig. 3 (c)), differently, a small quantity of g-C3N4 is deposited on the surface of GC nanosheets (Fig. 3 (d)). Remarkably, such ultra-thin g-C3N4 show a typical “nano-islands”-like discrete distribution on the GC, which should be a critical factor for the
formation of three-phase solid boundary from the GC, g-C3N4 and subsequent Pd component. Fig. 4 displays the Fourier transform infrared spectroscopy (FT-IR) spectra of the as-prepared samples. The FT-IR spectra of the GC and AC samples are shown in Fig. 4 (a). The peak at around 3450 cm-1 corresponds to the O-H vibrating modes of absorbed water. The peak at around 1100 cm-1 corresponds to the C-O vibrating modes of oxygen containing groups. The peaks from 1300 cm-1 to 1700 cm-1 reflect the benzene-ring C=C signals of carbon. For the pure g-C3N4 (Fig. 4 (b)), the peaks at 1642, 1565, 1463, and 1411 cm-1 correspond to the typical conjugated groups of C=N in aromatic ring. The peaks at 1323 and 1241 cm-1 correspond to the C-N vibrating mode of aromatic amine. Expressly, the characteristic breathing mode of “triazine units” at 807 cm-1 is clearly observed [31]. Fig. 4 (c) and (d) show the FT-IR spectra of g-C3N4(n)/GC samples with different “n” from 1 to 0.0625. It can be seen that the characteristic peak of “triazine units” at 807 cm-1 and the series of peaks from 1241 cm-1 to 1642 cm-1 are clearly visible for the g-C3N4(1)/GC sample. With the reduction of “n”, the characteristic peak of “triazine units” at 807 cm-1 is gradually weakened due to the decrease of g-C3N4 content. The crystalline structures of as-prepared samples were further characterized by the X-ray diffraction (XRD), with the result shown in Fig. 5. The XRD pattern of pure g-C3N4 is shown in Fig. 5 (a), which reveals two characteristic peaks at 13.2 ° and 27.4° corresponding to the (100) and (002) crystal planes. All the reflections of the sample can be indexed to a pure hexagonal phase g-C3N4 (JCPDS 87-1526) [32]. The strong peak at 27.4° is originated from lamellar stacking of conjugated aromatic system, which belongs to one unique (002) crystal plane peak of graphitic materials [33]. The XRD pattern of GC support (Fig. 5 (b)) shows a very strong (002) peak at 26.5° and a visible (101) peak at 44.5°, indicating that the carbon material with a high degree of graphitization has been generated from resin precursor by the nickel catalysis at a lower temperature of 850 ℃[19]. For the AC support (Fig. 5 (c)), a broad peak (002) peak at 23.9° is observed, which indicates the
low crystallinity characteristic of amorphous carbon materials. The XRD patterns of g-C3N4(n)/GC samples (Fig. 5 (d)) demonstrate a series of obvious (002) peaks between 26.5° and 27.4°, which are derived from the consolidation of (002) diffraction peaks from the GC and g-C3N4. With the increase of “n”, the position of (002) peaks are gradually shifted from 26.5° to 27.4°, suggesting the increasing weight of the g-C3N4. The XRD patterns of Pd/GC and Pd/AC are shown in Fig. 5 (e). Besides the (002) diffraction peaks of carbon, another five characteristic peaks from left to right corresponding to (111), (200), (220), (311) and (222) planes of the face centered cubic (fcc) structure of pure Pd (JCPDS, 87-0639) are easily observed. For the Pd@g-C3N4(n)/GC samples (Fig. 5 (f)), similar characteristic peaks of Pd are also observed. The average grain size of these Pd nanocrystals for all these samples have an equal value of about 13 nm, which is evaluated from (111) plane of Pd by the Scherer equation. These results show that the Pd metals with fine nanostructures have been successfully deposited on the GC support materials. In order to investigate the influence of g-C3N4 amounts on the catalyst's structures, the comparative SEM images have been given for all catalysts samples in Fig. 6. For the Pd/GC, a highly uniform distribution without agglomeration can be observed (Fig. 6 (a)). For the Pd@g-C3N4(0.0625)/GC and Pd@g-C3N4(0.125)/GC with lower g-C3N4 amounts, the relatively uniform distribution can be also observed (Fig. 6 (b) and (c)). However, with the further increase of the g-C3N4 amounts, the catalyst's structures become more and more uneven and massive agglomeration are formed (Fig. 6 (d)~(f)). The results show the lower amount of g-C3N4 offers the better distribution structures for the Pd@g-C3N4/GC, which is very important to enhance their catalytic performances towards the ethanol oxidation reaction. 3.2. Electrochemical performances The electrochemical performances of as-prepared samples have been systematically investigated by CV and EIS testing techniques in three-electrode system. Fig. 7 (a) and (b) show the CV curves
of different samples including the (a) Pd/GC, Pd/AC, Pd@g-C3N4, Pd@g-C3N4(0.0625)/GC and (b) Pd@g-C3N4(n=1~0.0625)/GC in 1mol/L KOH aqueous solution at 50 mV/s. It is obvious that all these CV curves (except for Pd@g-C3N4) reveal the common characteristics of the adsorption and desorption of hydrogen, electric double layer, the formation of palladium oxides and its reduction [34]. In the positive scan response, the potential peaks at about 0.18V, 0.40 V and the region from 0.80 V to 1.20 V indicate the signals of hydrogen adsorption, adsorption of hydroxyl transition state compounds, and the oxidation of palladium, respectively. In the negative scan response, the potential peaks at about 0.72V and 0.20V indicate the signals of reduction of palladium oxides and hydrogen desorption, respectively. The absence of CV peaks for the Pd@g-C3N4 is largely because of the poor electrical conductivity of g-C3N4. Fig. 7 (c) and (d) show the CV curves of different samples including the Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s. In the positive scan response, the potential peaks from 0.4V～1.0 V show the oxidation of ethanol under Pd catalysis, and the potential peaks from 0.4V～0.8 V in the negative scan response show the further oxidation of intermediate compounds [35]. Evidently, from Fig. 7 (c), it can been seen that the positive scan response peak current densities (Imax) of these samples are, in order, Pd@g-C3N4(0.0625)/GC (2156 A/g Pd) > Pd/GC (1150 A/g Pd) > Pd/AC (1018 A/g Pd) > Pd@g-C3N4 (8.6 A/g Pd). In alkaline media, Pd-based electrocatalysts once achieved an unusual max-mass specific current density (442 mA/µg Pd), however, most of the Pd-based electrocatalysts exhibited much lower values [36]. Therefore, the specific current density (2156 A/g Pd) of Pd@g-C3N4/GC electrocatalyst would be a comparable catalytic activity for the ethanol oxidation reaction. Furthermore, lower potentials in the anodic oxidation are constructive in achieving a higher cell voltage and energy density in DEFCs. For this purpose, linear-sweep voltammograms (i.e. partial enlargement curves from Fig. 7 (c)) of the as-prepared samples were further obtained
(see Fig. 7 (d)). The on-set potentials of these samples are, in order, Pd@g-C3N4(0.0625)/GC (0.320 V) < Pd/GC (0.370 V) < Pd/AC (0.372 V) < Pd@g-C3N4 (0.465 V). To sum up, for both the peak current densities and on-set potentials, the Pd@g-C3N4(0.0625)/GC presents the much better performances than the Pd/GC, which demonstrates that the introduction of appropriate amount of g-C3N4 is conducive to the improvement of the electrocatalytic performances of Pd on the ethanol oxidation reaction. The poor electrocatalytic performances for the Pd@g-C3N4 are also mainly due to the poor electrical conductivity of g-C3N4 (Fig. 7 (e)). Fig. 7 (f) and (g) show the CV curves of the Pd@g-C3N4(n=1~0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s. With the increase of “n” (namely the content of g-C3N4), the on-set potentials of the Pd@g-C3N4(n=1~0.0625)/GC samples present a changing trend of positive shift after the first negative shift. The peak current density generally shows a decreasing trend with the increase of g-C3N4 content. The above analysis result shows that appropriate amount of g-C3N4 can promote the catalytic effect of Pd, but too much will be counterproductive. The reasons are as follows: (i) The N atoms in g-C3N4 would be combined with the Pd atoms to change the d orbitals of Pd (may called electron-coupling effect), which can reduce the energy required for Pd atoms from their ground state to excited state, and thus promote the ethanol oxidation with appropriate amount of g-C3N4 [26,27]; (ii) The increase of g-C3N4 content will lead to the closely distribution of g-C3N4 on GC , and further blocking a part of electronic transmission channel for the catalyst [28]; (iii) If the g-C3N4 content is excessive, the nitrogen may give a priority to combine with ethanol by hydrogen bonds, which can hinder the contact of ethanol and palladium and depress the catalytic oxidation of ethanol [37]. Among all these Pd@g-C3N4(n=1~0.0625)/GC samples, the Pd@g-C3N4(0.0625)/GC demonstrates a maximal oxidation current density at the entire potential range, although its on-set potential is not the lowest. For the Pd@g-C3N4/GC hybrid system, the catalytic activity has strong ties to the three-phase
solid boundary of the GC, g-C3N4 and Pd components. With regard to the Pd@g-C3N4(0.0625)/GC sample, the ultra-thin g-C3N4 with “nano-islands”-like discrete distribution guarantees the formation of well three-phase solid boundary from the GC, g-C3N4 and Pd components (see Fig. 2 (d) and Fig. 8 (a) for details). The appropriate nitrogen species of g-C3N4 and available conductive boundary of GC provide both highly active sites of palladium and good electronic transmission channels for the catalytic system. However, with the increase of the g-C3N4 amounts, the catalyst displays massive agglomeration due to the excessive g-C3N4 (see Fig. 6), such leading to poor three-phase solid boundary of the GC, g-C3N4 and Pd components (see Fig. 8 (b)). The widely accepted catalytic mechanism for the ethanol electrooxidation in alkaline medium can be summarized as follows:[1] Pd + OH- → Pd-OHads + e-
(1)
Pd + C2H5OH → Pd-(C2H5OH)ads
(2)
Pd-(C2H5OH)ads +3OH- → Pd-(CH3CO) ads + 3H2O +3e-
(3)
Pd-(CH3CO) ads + Pd-OHads → Pd-CH3COOH + Pd
(4)
It is suggested that the rate-determining step is the removal of the adsorbed acyl groups by the adsorbed hydroxyl groups (eq. (4)) [1]. So an adequate amount of adsorbed hydroxyl groups is significant to achieve a high electrooxidation current. The beneficial effect by adding catalyst promoter (e.g. metal oxides [38] and metal carbides [7]) into the noble metal catalyst system has been assimilated to that of Ru in Pt-Ru catalysts, namely the well-known “bifunctional mechanism” [39]. Particularly, these foreign promoters could increase the concentration of OHads species on the surface of noble metal catalysts, favoring the transformation of aldehyde into corresponding carboxylic acid [1]. With regard to the present Pd@g-C3N4/GC catalysts with g-C3N4 as catalyst promoter, the “bifunctional mechanism” [39] and “electron-coupling effect” [26, 27] may together contribute to its high catalytic activity for the ethanol oxidation reaction.
Electrochemical impedance spectroscopy (EIS) is another important method for electrochemical measurement. Fig. 9 shows the EIS spectra of different samples including the Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution. Usually, the charge transfer in the surface of one catalyst can be effectively studied from the EIS spectra. In the high frequency region, the curvature radius of the semicircle is relative to the kinetics of electrochemical systems. The smaller curvature radius suggests the easier charge transfer, and thus quick dynamics of the electrochemical system is released [40]. It can be clearly observed from Fig. 9 (a) that the curvature radius of semicircle is very large for the Pd@g-C3N4 electrode, which is due to the extremely low electronic conductivity of g-C3N4. Fig. 9 (b) shows that the curvature radii of semicircles for these electrodes are in turn of Pd/GC < Pd@g-C3N4(0.0625)/GC < Pd/AC electrodes. Because of its high graphite structure and less crystal defects of GC, the Pd/GC electrode can offer high electrical conductivity and fast dynamics. Most of the carbon atoms of AC are stacked with irregular structures, and the graphitization degree of AC is lower, which can greatly reduce the electron transfer capability for the Pd/AC electrode. The curvature radius of Pd@g-C3N4(0.0625)/GC is between Pd/GC and Pd/AC, indicating that the Pd@g-C3N4(0.0625)/GC have enough electrical conductivity and desirable dynamics. The reasons are as follows: (i) the high electrical conductivity of GC ensures the efficient charge transfer of the Pd@g-C3N4(0.0625)/GC, and (ii) the discrete distribution of moderate g-C3N4 “nano-islands” on the surface of GC support is beneficial to the electron transfer and ion diffusion of electrolyte. The electrochemical stability of the Pd/GC, Pd/AC and Pd@g-C3N4(0.0625)/GC electrodes has been characterized by 200 CV cycles in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s, with the result shown in Fig. 10. The peak current densities of the Pd/GC, Pd/AC and Pd@g-C3N4(0.0625)/GC can be clearly observed from Fig. 10 (a), Fig. 10 (c) and Fig. 10 (e), which displays a common trend to increase firstly and then decrease with the increase of CV cycles. The
main reason for the increased trend is that the active points of Pd reach the maximum value needs an electrochemical activation process. The reasons for the decreased trend include two aspects: (i) in each CV cycle, the potential palladium oxides can not be completely reduced to be metal palladium and the number of active points of palladium would be decreased, and (ii) in the catalytic reaction, the decrease of ethanol concentration and the accumulation of catalysate will affect the output of peak current density in a certain extent. Usually, the second reason can be negligible due to its equivalence property for different samples. Obviously, the peak current density of Pd@g-C3N4(0.0625)/GC (46th cycle, Imax=2156 A/g Pd) can reach the maximum in a shorter time with respect to Pd/GC (59th cycle, Imax=1050 A/g Pd) and Pd/AC (58th cycle, Imax=1018 A/g Pd). The results show that, compared with Pd/GC and Pd/AC, the g-C3N4 component in the Pd@g-C3N4(0.0625)/GC can promote Pd catalytic oxidation of oxidation, and the peak current density can reach the maximum in a shorter cycle time. After 200 CV cycles, the peak current density of Pd@g-C3N4(0.0625)/GC (I200=1904 A/g Pd) is about two times of Pd/GC (I200=966 A/g Pd) and Pd/AC (I200=954 A/g Pd). The electrochemical stability of the catalyst samples can be evaluated by the potential difference at a same current density [7,17], before and after the 200 CV cycles. At the same current density of 100 A/g Pd, the potential difference of Pd/GC (Fig. 10 (b)), Pd@g-C3N4(0.0625)/GC (Fig. 10 (d)) and Pd/AC (Fig. 10 (f)) are about 6 mV, 18 mV and 25 mV, respectively. These results demonstrate that the Pd@g-C3N4(0.0625)/GC electrode has a desirable electrochemical stability for the ethanol oxidation reaction. The enhanced electrochemical stability of Pd@g-C3N4(0.0625)/GC electrode can be attributed to two aspects: (i) the graphite carbon has a more stable crystal structure relative to the amorphous carbon [16,17], which can provide a more stable electrochemical environment for the Pd and g-C3N4 components; (ii) the electron-coupling effect between g-C3N4 and GC would be occurred [26,27], which can promote the fixation of Pd nanoparticles on to the support and stable three-phase boundary of GC,
g-C3N4 and Pd was formed.
4. Conclusions In summary, graphitic carbon nitride/graphite carbon/palladium (Pd@g-C3N4/GC) electrocatalyst is successfully fabricated for the first time towards the ethanol oxidation reaction in the alkaline medium. For this Pd@g-C3N4/GC hybrid architecture, ultra-thin g-C3N4 with a “nano-islands”-like discrete distribution on the surface of GC support has been achieved, where the crucial three-phase boundary from GC, g-C3N4 and Pd component was successfully formed. Owing to the promotion effects from the GC support and g-C3N4 component, the as-prepared Pd@g-C3N4/GC electrocatalyst displays excellent electrocatalytic performance and good electrochemical stability for the ethanol oxidation reaction. The maximal peak current densities of Pd@g-C3N4/GC can reach 2156 A/g Pd, which is 1.87 times as high as that of Pd/GC (1150 A/g Pd). After 200 CV cycles, the peak current density of Pd@g-C3N4/GC can retain 1904 A/g Pd, while that of Pd/GC is only 966 A/g Pd. All these results persuade us to propose the novel Pd@g-C3N4/GC nanocomposite as one of promising electrocatalyst for the ethanol oxidation reaction of DEFCs.
Acknowledgements Financial support by the National Natural Science Foundation of China (21443006), Provincial Natural Science Foundation of Guangdong (2014A030310179), Young Creative Talents Project of Guangdong Province Education Department (2014KQNCX200), Science and Technology Project of Maoming (2014006) and Doctor Startup Project of local University (513086) are gratefully acknowledged.
References [1] C. Bianchini, P.K. Shen, Palladium-based electrocatalysts for alcohol oxidation in half cells and in direct alcohol fuel cells, Chemical Reviews, 109 (2009) 4183–4206. [2] M.Z.F. Kamarudin, S.K. Kamarudin, M.S. Masdar, W.R.W. Daud, Review: Direct ethanol fuel cells, International Journal of Hydrogen Energy, 38 (2013) 9438–9453. [3] J. Noborikawa, J. Lau, J. Ta, S. Hu, L. Scudiero, S. Derakhshan, S. Ha, J.L. Haan, Palladium-copper electrocatalyst for promotion of oxidation of formate and ethanol in alkaline media, Electrochimica Acta, 137 (2014) 654–660. [4] R. Carrera-Cerritos, R. Fuentes-Ramírez, F.M. Cuevas-Muñiz, J. Ledesma-García, L.G. Arriaga, Performance and stability of Pd nanostructures in an alkaline direct ethanol fuel cell, Journal of Power Sources, 269 (2014) 370–378. [5] P. Mukherjee, P.S. Roy, K. Mandal, D. Bhattacharjee, S. Dasgupta, S.K. Bhattacharya, Improved catalysis of room temperature synthesized Pd-Cu alloy nanoparticles for anodic oxidation of ethanol in alkaline media, Electrochimica Acta, 154 (2015) 447–455. [6] S. Sharma, B.G. Pollet, Support materials for PEMFC and DMFC electrocatalysts—A review, Journal of Power Sources, 208 (2012) 96–119. [7] Z. Li, S. Ji, B.G. Pollet, P.K. Shen, A Co3W3C promoted Pd catalyst exhibiting competitive performance over Pt/C catalysts towards the oxygen reduction reaction, Chemical Communications, 50 (2014) 566–568. [8] M. Liu, C. Peng, W. Yang, J. Guo, Y. Zheng, P. Chen, T. Huang, J. Xua, Pd nanoparticles supported on three-dimensional graphene aerogels as highly efficient catalysts for methanol electrooxidation, Electrochimica Acta,178 (2015) 838–846. [9] P. Wu, Y. Huang, L. Zhou, Y. Wang, Y. Bu, J. Yao, Nitrogen-doped graphene supported highly dispersed palladium-lead nanoparticles for synergetic enhancement of ethanol electrooxidation
in alkaline medium, Electrochimica Acta, 152 (2015) 68–74. [10] H. Chen, Y. Huang, D. Tang, T. Zhang, Y. Wang, Ethanol oxidation on Pd/C promoted with CaSiO3 in alkaline medium, Electrochimica Acta, 158 (2015) 18–23. [11] Y. Wang, D.P. Wilkinson, J. Zhang, Noncarbon support materials for polymer electrolyte membrane fuel cell electrocatalysts, Chemical Reviews, 111 (2011) 7625–7651. [12] P. Trogadas, T.F. Fuller, P. Strasser, Carbon as catalyst and support for electrochemical energy conversion, Carbon, 75 (2014) 5–42. [13] J. Lv, S. Li, A. Wang, L. Mei, J. Chen, J. Feng, Monodisperse Au-Pd bimetallic alloyed nanoparticles supported on reduced graphene oxide with enhanced electrocatalytic activity towards oxygen reduction reaction, Electrochimica Acta, 136 (2014) 521–528. [14] H. Heli, J. Pishahang, Cobalt oxide nanoparticles anchored to multiwalled carbon nanotubes: Synthesis and application for enhanced electrocatalytic reaction and highly sensitive nonenzymatic detection of hydrogen peroxide, Electrochimica Acta, 123 (2014) 518–526. [15] Z. Li, D. Li, Z. Liu, B. Li, C. Ge, Y. Fang, Mesoporous carbon microspheres with high capacitive performances for supercapacitors, Electrochimica Acta, 158 (2015) 237–245. [16] H. Huang, X. Wang, Recent progress on carbon-based support materials for electrocatalysts of direct methanol fuel cells. Journal of Materials Chemistry A, 2 (2014) 6266–6291. [17] Z. Li, Y. Li, S.P. Jiang, G. He, P.K. Shen, Novel graphene-like nanosheet supported highly active electrocatalysts with ultralow Pt loadings for oxygen reduction reaction, Journal of Materials Chemistry A, 2 (2014)16898–16904. [18] Z. Yan, M. Cai, P.K. Shen, Low temperature formation of porous graphitized carbon for electrocatalysis, Journal of Materials Chemistry, 22 (2012) 2133–2139. [19] Y. Li Z. Li, P.K. Shen, Simultaneous formation of ultrahigh surface area and three-dimensional hierarchical porous grapheme-like networks for fast and highly stable supercapacitors.
Advanced Materials, 25 (2013) 2474–2480. [20] X. Xu, Y. Xu, H. Zhang, M. Ji, H. Dong, The effect of NiO as graphitization catalyst on the structure and electrochemical performance of LiFePO4/C cathode materials, Electrochimica Acta, 158 (2015) 348–355. [21] Y. Wang, X. Wang, M. Antonietti, Polymeric graphitic carbon nitride as a heterogeneous organocatalyst: from photochemistry to multipurpose catalysis to sustainable chemistry, Angewandte Chemie International Edition, 51 (2012) 68–89. [22] S. Cao, J. Low, J. Yu, M. Jaroniec, Polymeric photocatalysts based on graphitic carbon nitride, Advanced Materials, 27 (2015) 2150–2176. [23] Y. Zheng, J. Liu, J. Liang, M. Jaroniec, S.Z. Qiao, Graphitic carbon nitride materials: controllable synthesis and applications in fuel cells and photocatalysis, Energy & Environmental Science, 5 (2012) 6717–6731. [24] J. Duan, S. Chen, M. Jaroniec, S.Z. Qiao, Porous C3N4 nanolayers@N-graphene films as catalyst electrodes for highly efficient hydrogen evolution, ACS nano, 9 (2015) 931–940. [25] T.Y. Ma, J. Ran, S. Dai, M. Jaroniec, S.Z. Qiao, Phosphorus-doped graphitic carbon nitrides grown in situ on carbon-fiber paper: flexible and reversible oxygen electrodes, Angewandte Chemie International Edition, 54 (2015) 4646–4650. [26] J. Tian, R. Ning, Q. Liu, A.M. Asiri, A.O. Al-Youbi, X. Sun, Three-dimensional porous supramolecular architecture from ultrathin g‑C3N4 nanosheets and reduced graphene oxide: solution self-assembly construction and application as a highly efficient metal-free electrocatalyst for oxygen reduction reaction, ACS Applied Materials & Interfaces, 6 (2014) 1011–1017. [27] H. Huang, S. Yang, R. Vajtai, X. Wang, P.M. Ajayan, Pt-decorated 3D architectures built from graphene and graphitic carbon nitride nanosheets as efficient methanol oxidation catalysts,
Advanced Materials, 26 (2014) 5160–5165. [28] Y. Zheng, Y. Jiao, J. Chen, J. Liu, J. Liang, A. Du, W. Zhang, Z. Zhu, S.C. Smith, M. Jaroniec, G.Q. Lu, S.Z. Qiao, Nanoporous graphitic-C3N4@carbon metal-free electrocatalysts for highly efficient oxygen reduction, Journal of the American Chemical Society, 133 (2011) 20116–20119. [29] J. Yuan, Q. Gao, X. Li, Y. Liu, Y. Fang, S. Yang, F. Peng, X. Zhou, Novel 3-D nanoporous graphitic-C3N4 nanosheets with heterostructured modification for efficient visible-light photocatalytic hydrogen production, RSC Advances, 4 (2014) 52332–52337. [30] G. Wang, S. Kuang, J. Zhang, S. Hou, S. Nian, Graphitic carbon nitride/multiwalled carbon nanotubes composite as Pt-free counter electrode for high-efficiency dye-sensitized solar cells, Electrochimica Acta, 2015, doi:10.1016/j.electacta.2015.11.039. [31] Z. Li, S. Yang, J. Zhou, D. Li, X. Zhou, C. Ge, Y. Fang, Novel mesoporous g-C3N4 and BiPO4 nanorods hybrid architectures and their enhanced visible-light-driven photocatalytic performances, Chemical Engineering Journal, 241 (2014) 344–351. [32] W. Wang, C.Y.Jimmy, Z. Shen, D.K. Chan, T. Gu, g-C3N4 quantum dots: direct synthesis, upconversion properties and photocatalytic application, Chemical Communications, 50 (2014) 10148–10150. [33] Z. Li, B. Li, S. Peng, D. Li, S. Yang, Y. Fang, Novel visible light-induced g-C3N4 quantum dot/BiPO4 nanocrystal composite photocatalysts for efficient degradation of methyl orange, RSC Advances, 4 (2014) 35144–35148. [34] J. Datta, A. Dutta, S. Mukherjee, The beneficial role of the cometals Pd and Au in the carbon-supported PtPdAu catalyst toward promoting ethanol oxidation kinetics in alkaline fuel cells: temperature effect and reaction mechanism, Journal of Physical Chemistry C, 115 (2011) 15324–15334.
[35] A.M. Hofstead-Duffy, D. Chen, S. Sun, Y. Ye, J. Tong, Origin of the current peak of negative scan in the cyclic voltammetry of methanol electro-oxidation on Pt-based electrocatalysts: a revisit to the current ratio criterion, Journal of Materials Chemistry, 22 (2012) 5205–5208. [36] A. Brouzgou, A. Podias, P. Tsiakaras, PEMFCs and AEMFCs directly fed with ethanol: a current status comparative review, Journal of Applied Electrochemistry, 43 (2013) 119–136. [37] S. Cao, J. Yu, g‑C3N4‑based photocatalysts for hydrogen generation, Journal of Physical Chemistry Letters, 5 (2014) 2101−2107. [38] K. Wang, B. Wang, J. Chang, L. Feng, W. Xing, Formic acid electrooxidation catalyzed by Pd/SmOx-C hybrid catalyst in fuel cells, Electrochimica Acta, 150 (2014) 329–336. [39] T. Yajima, N. Wakabayashi, H. Uchida, M. Watanabe, Adsorbed water for the electro-oxidation of methanol at Pt–Ru alloy, Chemical Communications, (2003) 828–829. [40] A. Battistel1, M. Fan, J. Stojadinović, F.L. Mantia, Analysis and mitigation of the artefacts in electrochemical impedance spectroscopy due to three-electrode geometry, Electrochimica Acta, 135 (2014) 133–138.
Figure captions Fig. 1 Schematic process for the preparation of Pd@g-C3N4/GC nanocomposite.
Fig. 2 Typical SEM images (a,b) of g-C3N4(0.0625)/GC composite and TEM images (c,d) of Pd@g-C3N4(0.0625)/GC composite.
Fig. 3 Comparative TEM images of GC (a,b) and g-C3N4(0.0625)/GC composite (c,d).
Fig. 4 FT-IR spectra of AC and GC (a), g-C3N4(b), g-C3N4(n=1~0.25)/GC (c) and g-C3N4(n=0.125~0.0625)/GC (d).
Fig. 5 XRD patterns of g-C3N4 (a), GC(b), AC (c), g-C3N4(n=1~0.0625)/GC (d), Pd/GC and Pd/AC (e)、Pd@g-C3N4(n=1~0.0625)/GC (f).
Fig. 6 Comparative SEM images of Pd/GC, Pd@g-C3N4(0.0625)/GC, Pd@g-C3N4(0.125)/GC, Pd@g-C3N4(0.25)/GC, Pd@g-C3N4(0.5)/GC, Pd@g-C3N4(1)/GC.
Fig. 7 CV curves in 1mol/L KOH aqueous solution at 50 mV/s for (a) Pd/GC, Pd/AC, Pd@g-C3N4, Pd@g-C3N4(0.0625)/GC, (b) Pd@g-C3N4(n=1~0.0625)/GC; CV curves in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s for (c, d and e) Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC ((d) and (e) is the partial enlargement curves from (c)), (f and g) Pd@g-C3N4(n=1~0.0625)/GC ((g) is the partial enlargement curves from (f)).
Fig. 8 Schematic diagram of three-phase boundary for the Pd@g-C3N4/GC with lower (a) and higher (b) amount of g-C3N4.
Fig. 9 Nyquist plots (a) of EIS and its expanded high frequency region (b) for Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC.
Fig. 10 CV cycling test (200 cycles) in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s for Pd@g-C3N4(0.0625)/GC(a,b), Pd/GC(c,d), Pd/AC(e,f) ((b), (d) and (f) are the partial enlargement curves from (a), (c) and (e), respectively).
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
S0013-4686(16)30126-8 http://dx.doi.org/doi:10.1016/j.electacta.2016.01.124 EA 26506
To appear in:
Electrochimica Acta
Received date: Revised date: Accepted date:
2-12-2015 14-1-2016 16-1-2016
Please cite this article as: Zesheng Li, Runsheng Lin, Zhisen Liu, Dehao Li, Hongqiang Wang, Qingyu Li, Novel graphitic carbon nitride/graphite carbon/palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction, Electrochimica Acta http://dx.doi.org/10.1016/j.electacta.2016.01.124 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Novel graphitic carbon nitride/graphite carbon/palladium nanocomposite as a high-performance electrocatalyst for the ethanol oxidation reaction Zesheng Li a,*, Runsheng Lin a, Zhisen Liu a, Dehao Li a, Hongqiang Wang b and Qingyu Li b,* a
Development Center of Technology for Petrochemical Pollution Control and Cleaner Production of Guangdong Universitites,
College of Chemical Engineering, Guangdong University of Petrochemical Technology, Maoming, 525000, China b
Guangxi Key Laboratory of Low Carbon Energy Materials, School of Chemical and pharmaceutical Sciences,
Guangxi Normal University, Guilin, 541004, China.
Corresponence author: Zesheng Li, E-mail: [email protected]; Qingyu Li E-mail: [email protected].
Graphic abstract
Novel graphitic carbon nitride/graphite carbon/palladium (Pd@g-C3N4/GC) electrocatalyst with excellent electrocatalytic performance and good electrochemical stability are developed for the ethanol oxidation reaction in alkaline medium.
Highlights (1) Graphite carbon support was massively prepared by a cost-effective catalytic strategy. (2) Ultra-thin g-C3N4 was successfully deposited on the surface of graphite carbon support. (3) Three-phase boundary from graphite carbon, g-C3N4 and Pd was successfully formed. (4) Nitrogen-rich g-C3N4 effectively promoted the electrochemical activity of Pd metal.
Abstract: Graphitic carbon nitride (g-C3N4) possesses high nitrogen content with excellent chemical and thermal stability, which also shows inherent electrochemical properties. Due to the extremely low electronic conductivity, pure g-C3N4 is usually restricted to electrocatalytic applications. In this study, we report the synthesis of graphite carbon (GC) supported g-C3N4 and palladium (Pd) electrocatalyst with excellent electronic conductivity for the ethanol oxidation reaction. In virtue of the ultra-thin structure of g-C3N4 deposited on the surface of GC, the crucial three-phase boundary from GC, g-C3N4 and Pd component are successfully formed in Pd@g-C3N4/GC electrocatalyst. The electrocatalytic performances on ethanol electrooxidation of Pd@g-C3N4/GC, Pd@g-C3N4, Pd@GC and Pd@amorphous carbon (AC) have been systematically investigated by the techniques of cyclic voltammetry and electrochemical impedance spectroscopy. The Pd@g-C3N4/GC displays the best electrocatalytic performances with a high oxidation peak current density of 2156 A/g Pd and low on-set potential of 0.32 V for the ethanol oxidation. The peak current density of Pd@g-C3N4/GC maintains at 1904 A/g Pd after 200 continuous cyclic voltammetry cycles. The Pd@g-C3N4/GC composite material might be one of candidate electrocatalysts for the ethanol oxidation reaction in direct ethanol fuel cells. Keywords: Graphitic carbon nitride, Graphite carbon, Palladium, Electrocatalyst, Ethanol oxidation
1. Introduction Platinum-free electrocatalyst based on palladium (Pd) has received increasing attention as one of promising anode electrocatalysts for ethanol oxidation reaction (EOR) in direct ethanol fuel cells (DEFCs), due to their inherent high catalytic activity and low cost of raw material [1-3]. At present, improving the catalytic efficiency and utilization rate of Pd, reducing its usage and improving its stability for the ethanol oxidation, is the core problem for the anode electrocatalysts of DEFCs [4, 5]. It is well known that, the support material plays a key framework role in the field of electrocatalysis, which can provide large specific surface area to accommodate catalysts nanoparticles [6]. Usually, the application of suitable support material is conducive to the dispersion of Pd nanoparticles and the reduction of agglomeration, so as to improve the catalytic efficiency and utilization rate, as well as reduce the use of Pd [7, 8]. Particularly, some atoms or atomic groups in the support material can be coupled with the Pd and release catalytic active sites, such promoting the conversion of ethanol oxidation intermediates and improving the catalytic efficiency [9, 10]. Relative to the conductive metal oxides and conducting polymers, carbon materials are one of the most common support materials in electrochemical applications [11, 12]. As support materials for electrocatalysts, carbon materials possess many advantages, for example, excellent electrical conductivity, low density, low expansion rate, high chemical stability, wide range of sources and very cheap price. Graphene, carbon nanotubes, Vulcan XC-72 and mesoporous carbon are considered to be excellent carbon supports for electrocatalysts, which not only have high specific surface area but also have good electrical conductivity and favorable electron-transfer network structure [13-15]. Because of its ordered array structure and less defects in lattice, graphite carbon displays better electrical conductivity and larger current output over amorphous carbon [16,17]. In general, the amorphous carbon can be changed into highly ordered graphite carbon at very high temperature (2200-3000℃), which requires a special high-temperature equipment and enormous
energy consumption. Fortunately, the graphite carbon can be also synthesized by catalytic process in lower temperature (<1000℃) [18-20], which not only reduces the production temperature and energy consumption but also reduces the cost of equipment and operation. Graphitic carbon nitride (g-C3N4) is a new type of conjugated semiconductor material, which is seized of ordered arrangements of tri-s-triazine subunits and tertiary amino groups, and laminated structures connected with weak van der Waals force [21]. Due to its convenient preparation, low cost, environmental friendliness, nitrogen-rich composition and two-dimensional graphite-like structure, g-C3N4 has been recently attracted greatly attention in the field of both photocatalysis and electrocatalysis [22,23]. Particularly, g-C3N4-based materials have been used as electrocatalysts for hydrogen evolution [24], oxygen evolution [25], oxygen reduction reaction [26] as well as methanol oxidation [27]. However, the pristine g-C3N4 is severely restricted in the applications of various electrochemical techniques, due to its extremely low electronic conductivity [28]. The g-C3N4 is always obtained by thermal condensation process along with the volume shrinkage, which results in numerous irregular holes within the bulk g-C3N4 [29]. The result further leads to a decreased conductivity of g-C3N4, so its electrochemical performance in electro-oxidation of alcohols is not so satisfactory. Therefore, it is of fundamental interest to develop synthetic strategies for depositing thin g-C3N4 layer onto a conducting support in order to improve its conductivity essentially [30]. In the present investigation, we report a novel graphitic carbon nitride/graphite carbon/palladium nanocomposite (Pd@g-C3N4/GC) as an efficient electrocatalyst for the ethanol electro-oxidation. To best of our knowledge, this is the first report about the g-C3N4-promote Pd electrocatalyst for the application of ethanol electro-oxidation. For this composite system, several features have become apparent over previous reports: (i) the graphite carbon support with nanosheet-like structure is prepared by a cost-effective low-temperature (850 ℃) nickel catalytic route from inexpensive ion-exchange resin; (ii) the ultra-thin g-C3N4 “nano-islands” (5 nm) is successfully deposited on the
surface of graphite carbon support, which is expressly favorable for the electron transfer and ion diffusion of electrolyte; (iii) the electron-coupling effect between g-C3N4 and GC can promote the fixation of Pd nanoparticles on to the support [26]; (iv) the pyridine-type nitrogen-rich structure units of g-C3N4 can induce a large amount of electrochemical active points when interacted with the Pd metal [27]. With these merits, we investigated that the ethanol oxidation electrocatalyst with excellent performances, including high activity and stability, could be designed based on the Pd@g-C3N4/GC composite architectures.
2. Experimental 2.1. Preparation of graphite carbon (GC) support The GC support was prepared by low-temperature catalytic graphitization with acrylic acid-based cation exchange resin (D113) as a carbon source and nickel acetate as a precursor of nickel catalyst. 20 g of cation exchange resin was added into a beaker containing 200 ml of distilled water, and 2.5 g of nickel acetate was dissolved in 50 ml of distilled water with anther beaker. The nickel acetate aqueous solution was slowly dripped into the first beaker with cation exchange resin under high-speed stirring. After 3h exchange process, the nickel ions were basically adsorbed onto the surface of the resin. The resin-nickel ions complex was collected by filtration and dried at 80℃ for 24 h. The dried resin-nickel ions complex was put into a tube furnace with nitrogen atmosphere, and was heated to 850 ℃ in 2℃/min, holding for 2 h. The GC support was finally obtained after the Ni removal with 1mol/L HCl aqueous solution and smashing by high speed grinding. And the amorphous carbon (AC) was also prepared by similar steps without the use of nickel acetate. 2.2. Preparation of g-C3N4/GC composite support The g-C3N4/GC composite support was prepared by in-situ thermal polymerization of melamine on GC support. Typically, five mixtures of GC and melamine with different mass ratio of 1:0.0625,
1:0.125, 1:0.25, 1:0.5, and 1:1 were put into five beakers. 100 ml of distilled water and 50 ml of ethanol were added into each beaker. The five mixed solutions were then treated by ultrasonic dispersion at a temperature of 80 ℃ until the solvent is volatilized basically. The five mixed products were put into five corundum crucibles, sealed with aluminum papers, and were heated at 550 ℃ in nitrogen atmosphere for 4h. The g-C3N4 (n)/GC composite support (“n” stands for different mass ratio of melamine) were finally obtained after grinding. The pure g-C3N4 was also prepared by thermal polymerization of melamine at 550 ℃ in nitrogen atmosphere for 4h. 2.3. Preparation of Pd@g-C3N4/GC composite electrocatalyst The Pd@g-C3N4/GC composite electrocatalyst was prepared by an ethylene glycol-microwave heating reduction method form the Na2PdCl4 precursor. Typically, 80mg of g-C3N4/GC composite support, quantified Na2PdCl4 solution (containing 20 mg Pd) and 100ml of ethylene glycol were put into a beaker and treated by ultrasonic dispersion for 30 min at room temperature. The mixed solution was put into a microwave oven, and was heated to the boiling point and maintained for 30 seconds. The heated mixed solution was then naturally cooled down to the room temperature. The Pd@g-C3N4/GC composite electrocatalyst was finally achieved after centrifugal separation, washing and vacuum drying. At the same time, the Pd/GC, Pd@g-C3N4 and Pd/AC electrocatalyst were also prepared by the same parameter and procedure, with single GC, g-C3N4 and AC as support, respectively. 2.4. Material characterization The crystallographic textures and morphologies of the as-prepared sample were observed by X-ray diffractometer (XRD, Rigaku, D/max 2500 v/pc) with Cu-Ka1 radiation (λ=1.5406 Å), thermal field emission environment scanning electron microscopy (FE-SEM, FEI, Quanta 400), and transmission electron microscopy (TEM, JEM-2010HR). Fourier transformation infrared (FTIR) spectrum was recorded on Nicolet 6700 FTIR spectrometer (Madison, Wisconsin) with potassium
bromide as compressed slices. 2.5. Electrochemical characterization Electrochemical measurements were performed in conventional three-electrode system. The working electrode was prepared by mixing 5 mg of each catalyst, 900 µL ethanol, and 100 µL 0.5 wt% Nafion (DuPont, USA) to form slurry, followed by dipping 5 µL of the slurry onto a glass carbon electrode (0.785 cm2) (5 µg Pd for each electrode) and drying under the IR lamp. The three-electrode system was tested in a 1 mol/L KOH aqueous solution or 1 mol/L KOH + 1 mol/L C2H5OH mixed aqueous solution, with a Pt foil as the counter electrode, a reversible hydrogen electrode (RHE) as the reference electrode. The cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) were recorded by an electrochemical work station (CHI 660). The CV was performed within voltage range of 0.0~1.2 V at a scan rate of 50 mV/s. The EIS was measured in the frequency range of 100 mHz to 100 kHz at an open-circuit potential with an AC perturbation of 5 mV.
3. Results and discussion 3.1. Morphology and structure The preparation process involving the GC, g-C3N4/GC as well as Pd@g-C3N4/GC nanocomposite is schematically shown in Fig. 1. Firstly, the Ni2+ can be uniformly adsorbed onto the surface and internal structures of the resin by the ion exchange in aqueous solution. Secondly, the bulk porous graphite foam was achieved after the heating treatment and Ni removal. In order to obtain nanosheet-like GC support, a high speed grinding procedure was further carried out. Thirdly, the g-C3N4/GC composite was synthesized by in-situ thermal polymerization of melamine on GC support. Expressly, ultra-thin g-C3N4 with a “nano-islands”-like discrete distribution on the surface of GC support is one striking feature for the g-C3N4/GC composite. Finally, the Pd@g-C3N4/GC composite electrocatalyst was rapidly prepared by an ethylene glycol-microwave heating reduction
method form the Na2PdCl4 precursor. Notably, the crucial three-phase boundary from GC, g-C3N4 and Pd component was successfully formed in the Pd@g-C3N4/GC composite architecture. Fig. 2 (a) reveals the general morphology of the Pd@g-C3N4/GC composite by field-emission scan electron microscopy (FE-SEM). It indicates that the sample has a relatively uniform distribution without apparent massive structures or agglomerated particles. Interestingly, the local magnification reveals that the sample is composed of a number of nanosheet-like structures with width of about 200 nm and thickness of about 15 nm (Fig. 2 (b)). The formation of the Pd nanoparticles decorated on the nanosheet-like GC support is further confirmed using transmission electron microscopy (TEM) (Fig. 2 (c)). The diameters of these Pd nanoparticles were from 8 nm to 20 nm. The high resolution (HR) TEM image (Fig. 2 (d)) distinctly shows the lattice fringes with spacings of 0.22 nm and 0.33 nm, which correspond to the (111) plane of Pd and (002) plane of GC, respectively. Moreover, a small spot of g-C3N4 (< 5 nm in thickness) deposited on GC support can be observed from Fig. 2 (d). Most importantly, the three-phase solid boundary from the GC, g-C3N4 and Pd three different components has been demonstrated via the HR-TEM image. To show further insight into the detailed nanostructure of the GC and the formation condition of three-phase solid boundary, the comparative TEM images of GC and g-C3N4(0.0625)/GC are presented in Fig. 3. The TEM images of GC clearly reveal the sample has irregular nanosheet-like structures (Fig. 3 (a)) and the thickness of nanosheet is about 10 nm (Fig. 3 (b)). The formation process of the nanosheet-like graphite carbon is mainly due to the nickel catalyst with larger grain size, in which the large and flat crystallographic plane of nickel plays a role in templating two-dimensional crystal structure of graphite carbon [19]. For g-C3N4(0.0625)/GC, some similar nanosheet-like structures can be observed (Fig. 3 (c)), differently, a small quantity of g-C3N4 is deposited on the surface of GC nanosheets (Fig. 3 (d)). Remarkably, such ultra-thin g-C3N4 show a typical “nano-islands”-like discrete distribution on the GC, which should be a critical factor for the
formation of three-phase solid boundary from the GC, g-C3N4 and subsequent Pd component. Fig. 4 displays the Fourier transform infrared spectroscopy (FT-IR) spectra of the as-prepared samples. The FT-IR spectra of the GC and AC samples are shown in Fig. 4 (a). The peak at around 3450 cm-1 corresponds to the O-H vibrating modes of absorbed water. The peak at around 1100 cm-1 corresponds to the C-O vibrating modes of oxygen containing groups. The peaks from 1300 cm-1 to 1700 cm-1 reflect the benzene-ring C=C signals of carbon. For the pure g-C3N4 (Fig. 4 (b)), the peaks at 1642, 1565, 1463, and 1411 cm-1 correspond to the typical conjugated groups of C=N in aromatic ring. The peaks at 1323 and 1241 cm-1 correspond to the C-N vibrating mode of aromatic amine. Expressly, the characteristic breathing mode of “triazine units” at 807 cm-1 is clearly observed [31]. Fig. 4 (c) and (d) show the FT-IR spectra of g-C3N4(n)/GC samples with different “n” from 1 to 0.0625. It can be seen that the characteristic peak of “triazine units” at 807 cm-1 and the series of peaks from 1241 cm-1 to 1642 cm-1 are clearly visible for the g-C3N4(1)/GC sample. With the reduction of “n”, the characteristic peak of “triazine units” at 807 cm-1 is gradually weakened due to the decrease of g-C3N4 content. The crystalline structures of as-prepared samples were further characterized by the X-ray diffraction (XRD), with the result shown in Fig. 5. The XRD pattern of pure g-C3N4 is shown in Fig. 5 (a), which reveals two characteristic peaks at 13.2 ° and 27.4° corresponding to the (100) and (002) crystal planes. All the reflections of the sample can be indexed to a pure hexagonal phase g-C3N4 (JCPDS 87-1526) [32]. The strong peak at 27.4° is originated from lamellar stacking of conjugated aromatic system, which belongs to one unique (002) crystal plane peak of graphitic materials [33]. The XRD pattern of GC support (Fig. 5 (b)) shows a very strong (002) peak at 26.5° and a visible (101) peak at 44.5°, indicating that the carbon material with a high degree of graphitization has been generated from resin precursor by the nickel catalysis at a lower temperature of 850 ℃[19]. For the AC support (Fig. 5 (c)), a broad peak (002) peak at 23.9° is observed, which indicates the
low crystallinity characteristic of amorphous carbon materials. The XRD patterns of g-C3N4(n)/GC samples (Fig. 5 (d)) demonstrate a series of obvious (002) peaks between 26.5° and 27.4°, which are derived from the consolidation of (002) diffraction peaks from the GC and g-C3N4. With the increase of “n”, the position of (002) peaks are gradually shifted from 26.5° to 27.4°, suggesting the increasing weight of the g-C3N4. The XRD patterns of Pd/GC and Pd/AC are shown in Fig. 5 (e). Besides the (002) diffraction peaks of carbon, another five characteristic peaks from left to right corresponding to (111), (200), (220), (311) and (222) planes of the face centered cubic (fcc) structure of pure Pd (JCPDS, 87-0639) are easily observed. For the Pd@g-C3N4(n)/GC samples (Fig. 5 (f)), similar characteristic peaks of Pd are also observed. The average grain size of these Pd nanocrystals for all these samples have an equal value of about 13 nm, which is evaluated from (111) plane of Pd by the Scherer equation. These results show that the Pd metals with fine nanostructures have been successfully deposited on the GC support materials. In order to investigate the influence of g-C3N4 amounts on the catalyst's structures, the comparative SEM images have been given for all catalysts samples in Fig. 6. For the Pd/GC, a highly uniform distribution without agglomeration can be observed (Fig. 6 (a)). For the Pd@g-C3N4(0.0625)/GC and Pd@g-C3N4(0.125)/GC with lower g-C3N4 amounts, the relatively uniform distribution can be also observed (Fig. 6 (b) and (c)). However, with the further increase of the g-C3N4 amounts, the catalyst's structures become more and more uneven and massive agglomeration are formed (Fig. 6 (d)~(f)). The results show the lower amount of g-C3N4 offers the better distribution structures for the Pd@g-C3N4/GC, which is very important to enhance their catalytic performances towards the ethanol oxidation reaction. 3.2. Electrochemical performances The electrochemical performances of as-prepared samples have been systematically investigated by CV and EIS testing techniques in three-electrode system. Fig. 7 (a) and (b) show the CV curves
of different samples including the (a) Pd/GC, Pd/AC, Pd@g-C3N4, Pd@g-C3N4(0.0625)/GC and (b) Pd@g-C3N4(n=1~0.0625)/GC in 1mol/L KOH aqueous solution at 50 mV/s. It is obvious that all these CV curves (except for Pd@g-C3N4) reveal the common characteristics of the adsorption and desorption of hydrogen, electric double layer, the formation of palladium oxides and its reduction [34]. In the positive scan response, the potential peaks at about 0.18V, 0.40 V and the region from 0.80 V to 1.20 V indicate the signals of hydrogen adsorption, adsorption of hydroxyl transition state compounds, and the oxidation of palladium, respectively. In the negative scan response, the potential peaks at about 0.72V and 0.20V indicate the signals of reduction of palladium oxides and hydrogen desorption, respectively. The absence of CV peaks for the Pd@g-C3N4 is largely because of the poor electrical conductivity of g-C3N4. Fig. 7 (c) and (d) show the CV curves of different samples including the Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s. In the positive scan response, the potential peaks from 0.4V～1.0 V show the oxidation of ethanol under Pd catalysis, and the potential peaks from 0.4V～0.8 V in the negative scan response show the further oxidation of intermediate compounds [35]. Evidently, from Fig. 7 (c), it can been seen that the positive scan response peak current densities (Imax) of these samples are, in order, Pd@g-C3N4(0.0625)/GC (2156 A/g Pd) > Pd/GC (1150 A/g Pd) > Pd/AC (1018 A/g Pd) > Pd@g-C3N4 (8.6 A/g Pd). In alkaline media, Pd-based electrocatalysts once achieved an unusual max-mass specific current density (442 mA/µg Pd), however, most of the Pd-based electrocatalysts exhibited much lower values [36]. Therefore, the specific current density (2156 A/g Pd) of Pd@g-C3N4/GC electrocatalyst would be a comparable catalytic activity for the ethanol oxidation reaction. Furthermore, lower potentials in the anodic oxidation are constructive in achieving a higher cell voltage and energy density in DEFCs. For this purpose, linear-sweep voltammograms (i.e. partial enlargement curves from Fig. 7 (c)) of the as-prepared samples were further obtained
(see Fig. 7 (d)). The on-set potentials of these samples are, in order, Pd@g-C3N4(0.0625)/GC (0.320 V) < Pd/GC (0.370 V) < Pd/AC (0.372 V) < Pd@g-C3N4 (0.465 V). To sum up, for both the peak current densities and on-set potentials, the Pd@g-C3N4(0.0625)/GC presents the much better performances than the Pd/GC, which demonstrates that the introduction of appropriate amount of g-C3N4 is conducive to the improvement of the electrocatalytic performances of Pd on the ethanol oxidation reaction. The poor electrocatalytic performances for the Pd@g-C3N4 are also mainly due to the poor electrical conductivity of g-C3N4 (Fig. 7 (e)). Fig. 7 (f) and (g) show the CV curves of the Pd@g-C3N4(n=1~0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s. With the increase of “n” (namely the content of g-C3N4), the on-set potentials of the Pd@g-C3N4(n=1~0.0625)/GC samples present a changing trend of positive shift after the first negative shift. The peak current density generally shows a decreasing trend with the increase of g-C3N4 content. The above analysis result shows that appropriate amount of g-C3N4 can promote the catalytic effect of Pd, but too much will be counterproductive. The reasons are as follows: (i) The N atoms in g-C3N4 would be combined with the Pd atoms to change the d orbitals of Pd (may called electron-coupling effect), which can reduce the energy required for Pd atoms from their ground state to excited state, and thus promote the ethanol oxidation with appropriate amount of g-C3N4 [26,27]; (ii) The increase of g-C3N4 content will lead to the closely distribution of g-C3N4 on GC , and further blocking a part of electronic transmission channel for the catalyst [28]; (iii) If the g-C3N4 content is excessive, the nitrogen may give a priority to combine with ethanol by hydrogen bonds, which can hinder the contact of ethanol and palladium and depress the catalytic oxidation of ethanol [37]. Among all these Pd@g-C3N4(n=1~0.0625)/GC samples, the Pd@g-C3N4(0.0625)/GC demonstrates a maximal oxidation current density at the entire potential range, although its on-set potential is not the lowest. For the Pd@g-C3N4/GC hybrid system, the catalytic activity has strong ties to the three-phase
solid boundary of the GC, g-C3N4 and Pd components. With regard to the Pd@g-C3N4(0.0625)/GC sample, the ultra-thin g-C3N4 with “nano-islands”-like discrete distribution guarantees the formation of well three-phase solid boundary from the GC, g-C3N4 and Pd components (see Fig. 2 (d) and Fig. 8 (a) for details). The appropriate nitrogen species of g-C3N4 and available conductive boundary of GC provide both highly active sites of palladium and good electronic transmission channels for the catalytic system. However, with the increase of the g-C3N4 amounts, the catalyst displays massive agglomeration due to the excessive g-C3N4 (see Fig. 6), such leading to poor three-phase solid boundary of the GC, g-C3N4 and Pd components (see Fig. 8 (b)). The widely accepted catalytic mechanism for the ethanol electrooxidation in alkaline medium can be summarized as follows:[1] Pd + OH- → Pd-OHads + e-
(1)
Pd + C2H5OH → Pd-(C2H5OH)ads
(2)
Pd-(C2H5OH)ads +3OH- → Pd-(CH3CO) ads + 3H2O +3e-
(3)
Pd-(CH3CO) ads + Pd-OHads → Pd-CH3COOH + Pd
(4)
It is suggested that the rate-determining step is the removal of the adsorbed acyl groups by the adsorbed hydroxyl groups (eq. (4)) [1]. So an adequate amount of adsorbed hydroxyl groups is significant to achieve a high electrooxidation current. The beneficial effect by adding catalyst promoter (e.g. metal oxides [38] and metal carbides [7]) into the noble metal catalyst system has been assimilated to that of Ru in Pt-Ru catalysts, namely the well-known “bifunctional mechanism” [39]. Particularly, these foreign promoters could increase the concentration of OHads species on the surface of noble metal catalysts, favoring the transformation of aldehyde into corresponding carboxylic acid [1]. With regard to the present Pd@g-C3N4/GC catalysts with g-C3N4 as catalyst promoter, the “bifunctional mechanism” [39] and “electron-coupling effect” [26, 27] may together contribute to its high catalytic activity for the ethanol oxidation reaction.
Electrochemical impedance spectroscopy (EIS) is another important method for electrochemical measurement. Fig. 9 shows the EIS spectra of different samples including the Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution. Usually, the charge transfer in the surface of one catalyst can be effectively studied from the EIS spectra. In the high frequency region, the curvature radius of the semicircle is relative to the kinetics of electrochemical systems. The smaller curvature radius suggests the easier charge transfer, and thus quick dynamics of the electrochemical system is released [40]. It can be clearly observed from Fig. 9 (a) that the curvature radius of semicircle is very large for the Pd@g-C3N4 electrode, which is due to the extremely low electronic conductivity of g-C3N4. Fig. 9 (b) shows that the curvature radii of semicircles for these electrodes are in turn of Pd/GC < Pd@g-C3N4(0.0625)/GC < Pd/AC electrodes. Because of its high graphite structure and less crystal defects of GC, the Pd/GC electrode can offer high electrical conductivity and fast dynamics. Most of the carbon atoms of AC are stacked with irregular structures, and the graphitization degree of AC is lower, which can greatly reduce the electron transfer capability for the Pd/AC electrode. The curvature radius of Pd@g-C3N4(0.0625)/GC is between Pd/GC and Pd/AC, indicating that the Pd@g-C3N4(0.0625)/GC have enough electrical conductivity and desirable dynamics. The reasons are as follows: (i) the high electrical conductivity of GC ensures the efficient charge transfer of the Pd@g-C3N4(0.0625)/GC, and (ii) the discrete distribution of moderate g-C3N4 “nano-islands” on the surface of GC support is beneficial to the electron transfer and ion diffusion of electrolyte. The electrochemical stability of the Pd/GC, Pd/AC and Pd@g-C3N4(0.0625)/GC electrodes has been characterized by 200 CV cycles in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s, with the result shown in Fig. 10. The peak current densities of the Pd/GC, Pd/AC and Pd@g-C3N4(0.0625)/GC can be clearly observed from Fig. 10 (a), Fig. 10 (c) and Fig. 10 (e), which displays a common trend to increase firstly and then decrease with the increase of CV cycles. The
main reason for the increased trend is that the active points of Pd reach the maximum value needs an electrochemical activation process. The reasons for the decreased trend include two aspects: (i) in each CV cycle, the potential palladium oxides can not be completely reduced to be metal palladium and the number of active points of palladium would be decreased, and (ii) in the catalytic reaction, the decrease of ethanol concentration and the accumulation of catalysate will affect the output of peak current density in a certain extent. Usually, the second reason can be negligible due to its equivalence property for different samples. Obviously, the peak current density of Pd@g-C3N4(0.0625)/GC (46th cycle, Imax=2156 A/g Pd) can reach the maximum in a shorter time with respect to Pd/GC (59th cycle, Imax=1050 A/g Pd) and Pd/AC (58th cycle, Imax=1018 A/g Pd). The results show that, compared with Pd/GC and Pd/AC, the g-C3N4 component in the Pd@g-C3N4(0.0625)/GC can promote Pd catalytic oxidation of oxidation, and the peak current density can reach the maximum in a shorter cycle time. After 200 CV cycles, the peak current density of Pd@g-C3N4(0.0625)/GC (I200=1904 A/g Pd) is about two times of Pd/GC (I200=966 A/g Pd) and Pd/AC (I200=954 A/g Pd). The electrochemical stability of the catalyst samples can be evaluated by the potential difference at a same current density [7,17], before and after the 200 CV cycles. At the same current density of 100 A/g Pd, the potential difference of Pd/GC (Fig. 10 (b)), Pd@g-C3N4(0.0625)/GC (Fig. 10 (d)) and Pd/AC (Fig. 10 (f)) are about 6 mV, 18 mV and 25 mV, respectively. These results demonstrate that the Pd@g-C3N4(0.0625)/GC electrode has a desirable electrochemical stability for the ethanol oxidation reaction. The enhanced electrochemical stability of Pd@g-C3N4(0.0625)/GC electrode can be attributed to two aspects: (i) the graphite carbon has a more stable crystal structure relative to the amorphous carbon [16,17], which can provide a more stable electrochemical environment for the Pd and g-C3N4 components; (ii) the electron-coupling effect between g-C3N4 and GC would be occurred [26,27], which can promote the fixation of Pd nanoparticles on to the support and stable three-phase boundary of GC,
g-C3N4 and Pd was formed.
4. Conclusions In summary, graphitic carbon nitride/graphite carbon/palladium (Pd@g-C3N4/GC) electrocatalyst is successfully fabricated for the first time towards the ethanol oxidation reaction in the alkaline medium. For this Pd@g-C3N4/GC hybrid architecture, ultra-thin g-C3N4 with a “nano-islands”-like discrete distribution on the surface of GC support has been achieved, where the crucial three-phase boundary from GC, g-C3N4 and Pd component was successfully formed. Owing to the promotion effects from the GC support and g-C3N4 component, the as-prepared Pd@g-C3N4/GC electrocatalyst displays excellent electrocatalytic performance and good electrochemical stability for the ethanol oxidation reaction. The maximal peak current densities of Pd@g-C3N4/GC can reach 2156 A/g Pd, which is 1.87 times as high as that of Pd/GC (1150 A/g Pd). After 200 CV cycles, the peak current density of Pd@g-C3N4/GC can retain 1904 A/g Pd, while that of Pd/GC is only 966 A/g Pd. All these results persuade us to propose the novel Pd@g-C3N4/GC nanocomposite as one of promising electrocatalyst for the ethanol oxidation reaction of DEFCs.
Acknowledgements Financial support by the National Natural Science Foundation of China (21443006), Provincial Natural Science Foundation of Guangdong (2014A030310179), Young Creative Talents Project of Guangdong Province Education Department (2014KQNCX200), Science and Technology Project of Maoming (2014006) and Doctor Startup Project of local University (513086) are gratefully acknowledged.
References [1] C. Bianchini, P.K. Shen, Palladium-based electrocatalysts for alcohol oxidation in half cells and in direct alcohol fuel cells, Chemical Reviews, 109 (2009) 4183–4206. [2] M.Z.F. Kamarudin, S.K. Kamarudin, M.S. Masdar, W.R.W. Daud, Review: Direct ethanol fuel cells, International Journal of Hydrogen Energy, 38 (2013) 9438–9453. [3] J. Noborikawa, J. Lau, J. Ta, S. Hu, L. Scudiero, S. Derakhshan, S. Ha, J.L. Haan, Palladium-copper electrocatalyst for promotion of oxidation of formate and ethanol in alkaline media, Electrochimica Acta, 137 (2014) 654–660. [4] R. Carrera-Cerritos, R. Fuentes-Ramírez, F.M. Cuevas-Muñiz, J. Ledesma-García, L.G. Arriaga, Performance and stability of Pd nanostructures in an alkaline direct ethanol fuel cell, Journal of Power Sources, 269 (2014) 370–378. [5] P. Mukherjee, P.S. Roy, K. Mandal, D. Bhattacharjee, S. Dasgupta, S.K. Bhattacharya, Improved catalysis of room temperature synthesized Pd-Cu alloy nanoparticles for anodic oxidation of ethanol in alkaline media, Electrochimica Acta, 154 (2015) 447–455. [6] S. Sharma, B.G. Pollet, Support materials for PEMFC and DMFC electrocatalysts—A review, Journal of Power Sources, 208 (2012) 96–119. [7] Z. Li, S. Ji, B.G. Pollet, P.K. Shen, A Co3W3C promoted Pd catalyst exhibiting competitive performance over Pt/C catalysts towards the oxygen reduction reaction, Chemical Communications, 50 (2014) 566–568. [8] M. Liu, C. Peng, W. Yang, J. Guo, Y. Zheng, P. Chen, T. Huang, J. Xua, Pd nanoparticles supported on three-dimensional graphene aerogels as highly efficient catalysts for methanol electrooxidation, Electrochimica Acta,178 (2015) 838–846. [9] P. Wu, Y. Huang, L. Zhou, Y. Wang, Y. Bu, J. Yao, Nitrogen-doped graphene supported highly dispersed palladium-lead nanoparticles for synergetic enhancement of ethanol electrooxidation
in alkaline medium, Electrochimica Acta, 152 (2015) 68–74. [10] H. Chen, Y. Huang, D. Tang, T. Zhang, Y. Wang, Ethanol oxidation on Pd/C promoted with CaSiO3 in alkaline medium, Electrochimica Acta, 158 (2015) 18–23. [11] Y. Wang, D.P. Wilkinson, J. Zhang, Noncarbon support materials for polymer electrolyte membrane fuel cell electrocatalysts, Chemical Reviews, 111 (2011) 7625–7651. [12] P. Trogadas, T.F. Fuller, P. Strasser, Carbon as catalyst and support for electrochemical energy conversion, Carbon, 75 (2014) 5–42. [13] J. Lv, S. Li, A. Wang, L. Mei, J. Chen, J. Feng, Monodisperse Au-Pd bimetallic alloyed nanoparticles supported on reduced graphene oxide with enhanced electrocatalytic activity towards oxygen reduction reaction, Electrochimica Acta, 136 (2014) 521–528. [14] H. Heli, J. Pishahang, Cobalt oxide nanoparticles anchored to multiwalled carbon nanotubes: Synthesis and application for enhanced electrocatalytic reaction and highly sensitive nonenzymatic detection of hydrogen peroxide, Electrochimica Acta, 123 (2014) 518–526. [15] Z. Li, D. Li, Z. Liu, B. Li, C. Ge, Y. Fang, Mesoporous carbon microspheres with high capacitive performances for supercapacitors, Electrochimica Acta, 158 (2015) 237–245. [16] H. Huang, X. Wang, Recent progress on carbon-based support materials for electrocatalysts of direct methanol fuel cells. Journal of Materials Chemistry A, 2 (2014) 6266–6291. [17] Z. Li, Y. Li, S.P. Jiang, G. He, P.K. Shen, Novel graphene-like nanosheet supported highly active electrocatalysts with ultralow Pt loadings for oxygen reduction reaction, Journal of Materials Chemistry A, 2 (2014)16898–16904. [18] Z. Yan, M. Cai, P.K. Shen, Low temperature formation of porous graphitized carbon for electrocatalysis, Journal of Materials Chemistry, 22 (2012) 2133–2139. [19] Y. Li Z. Li, P.K. Shen, Simultaneous formation of ultrahigh surface area and three-dimensional hierarchical porous grapheme-like networks for fast and highly stable supercapacitors.
Advanced Materials, 25 (2013) 2474–2480. [20] X. Xu, Y. Xu, H. Zhang, M. Ji, H. Dong, The effect of NiO as graphitization catalyst on the structure and electrochemical performance of LiFePO4/C cathode materials, Electrochimica Acta, 158 (2015) 348–355. [21] Y. Wang, X. Wang, M. Antonietti, Polymeric graphitic carbon nitride as a heterogeneous organocatalyst: from photochemistry to multipurpose catalysis to sustainable chemistry, Angewandte Chemie International Edition, 51 (2012) 68–89. [22] S. Cao, J. Low, J. Yu, M. Jaroniec, Polymeric photocatalysts based on graphitic carbon nitride, Advanced Materials, 27 (2015) 2150–2176. [23] Y. Zheng, J. Liu, J. Liang, M. Jaroniec, S.Z. Qiao, Graphitic carbon nitride materials: controllable synthesis and applications in fuel cells and photocatalysis, Energy & Environmental Science, 5 (2012) 6717–6731. [24] J. Duan, S. Chen, M. Jaroniec, S.Z. Qiao, Porous C3N4 nanolayers@N-graphene films as catalyst electrodes for highly efficient hydrogen evolution, ACS nano, 9 (2015) 931–940. [25] T.Y. Ma, J. Ran, S. Dai, M. Jaroniec, S.Z. Qiao, Phosphorus-doped graphitic carbon nitrides grown in situ on carbon-fiber paper: flexible and reversible oxygen electrodes, Angewandte Chemie International Edition, 54 (2015) 4646–4650. [26] J. Tian, R. Ning, Q. Liu, A.M. Asiri, A.O. Al-Youbi, X. Sun, Three-dimensional porous supramolecular architecture from ultrathin g‑C3N4 nanosheets and reduced graphene oxide: solution self-assembly construction and application as a highly efficient metal-free electrocatalyst for oxygen reduction reaction, ACS Applied Materials & Interfaces, 6 (2014) 1011–1017. [27] H. Huang, S. Yang, R. Vajtai, X. Wang, P.M. Ajayan, Pt-decorated 3D architectures built from graphene and graphitic carbon nitride nanosheets as efficient methanol oxidation catalysts,
Advanced Materials, 26 (2014) 5160–5165. [28] Y. Zheng, Y. Jiao, J. Chen, J. Liu, J. Liang, A. Du, W. Zhang, Z. Zhu, S.C. Smith, M. Jaroniec, G.Q. Lu, S.Z. Qiao, Nanoporous graphitic-C3N4@carbon metal-free electrocatalysts for highly efficient oxygen reduction, Journal of the American Chemical Society, 133 (2011) 20116–20119. [29] J. Yuan, Q. Gao, X. Li, Y. Liu, Y. Fang, S. Yang, F. Peng, X. Zhou, Novel 3-D nanoporous graphitic-C3N4 nanosheets with heterostructured modification for efficient visible-light photocatalytic hydrogen production, RSC Advances, 4 (2014) 52332–52337. [30] G. Wang, S. Kuang, J. Zhang, S. Hou, S. Nian, Graphitic carbon nitride/multiwalled carbon nanotubes composite as Pt-free counter electrode for high-efficiency dye-sensitized solar cells, Electrochimica Acta, 2015, doi:10.1016/j.electacta.2015.11.039. [31] Z. Li, S. Yang, J. Zhou, D. Li, X. Zhou, C. Ge, Y. Fang, Novel mesoporous g-C3N4 and BiPO4 nanorods hybrid architectures and their enhanced visible-light-driven photocatalytic performances, Chemical Engineering Journal, 241 (2014) 344–351. [32] W. Wang, C.Y.Jimmy, Z. Shen, D.K. Chan, T. Gu, g-C3N4 quantum dots: direct synthesis, upconversion properties and photocatalytic application, Chemical Communications, 50 (2014) 10148–10150. [33] Z. Li, B. Li, S. Peng, D. Li, S. Yang, Y. Fang, Novel visible light-induced g-C3N4 quantum dot/BiPO4 nanocrystal composite photocatalysts for efficient degradation of methyl orange, RSC Advances, 4 (2014) 35144–35148. [34] J. Datta, A. Dutta, S. Mukherjee, The beneficial role of the cometals Pd and Au in the carbon-supported PtPdAu catalyst toward promoting ethanol oxidation kinetics in alkaline fuel cells: temperature effect and reaction mechanism, Journal of Physical Chemistry C, 115 (2011) 15324–15334.
[35] A.M. Hofstead-Duffy, D. Chen, S. Sun, Y. Ye, J. Tong, Origin of the current peak of negative scan in the cyclic voltammetry of methanol electro-oxidation on Pt-based electrocatalysts: a revisit to the current ratio criterion, Journal of Materials Chemistry, 22 (2012) 5205–5208. [36] A. Brouzgou, A. Podias, P. Tsiakaras, PEMFCs and AEMFCs directly fed with ethanol: a current status comparative review, Journal of Applied Electrochemistry, 43 (2013) 119–136. [37] S. Cao, J. Yu, g‑C3N4‑based photocatalysts for hydrogen generation, Journal of Physical Chemistry Letters, 5 (2014) 2101−2107. [38] K. Wang, B. Wang, J. Chang, L. Feng, W. Xing, Formic acid electrooxidation catalyzed by Pd/SmOx-C hybrid catalyst in fuel cells, Electrochimica Acta, 150 (2014) 329–336. [39] T. Yajima, N. Wakabayashi, H. Uchida, M. Watanabe, Adsorbed water for the electro-oxidation of methanol at Pt–Ru alloy, Chemical Communications, (2003) 828–829. [40] A. Battistel1, M. Fan, J. Stojadinović, F.L. Mantia, Analysis and mitigation of the artefacts in electrochemical impedance spectroscopy due to three-electrode geometry, Electrochimica Acta, 135 (2014) 133–138.
Figure captions Fig. 1 Schematic process for the preparation of Pd@g-C3N4/GC nanocomposite.
Fig. 2 Typical SEM images (a,b) of g-C3N4(0.0625)/GC composite and TEM images (c,d) of Pd@g-C3N4(0.0625)/GC composite.
Fig. 3 Comparative TEM images of GC (a,b) and g-C3N4(0.0625)/GC composite (c,d).
Fig. 4 FT-IR spectra of AC and GC (a), g-C3N4(b), g-C3N4(n=1~0.25)/GC (c) and g-C3N4(n=0.125~0.0625)/GC (d).
Fig. 5 XRD patterns of g-C3N4 (a), GC(b), AC (c), g-C3N4(n=1~0.0625)/GC (d), Pd/GC and Pd/AC (e)、Pd@g-C3N4(n=1~0.0625)/GC (f).
Fig. 6 Comparative SEM images of Pd/GC, Pd@g-C3N4(0.0625)/GC, Pd@g-C3N4(0.125)/GC, Pd@g-C3N4(0.25)/GC, Pd@g-C3N4(0.5)/GC, Pd@g-C3N4(1)/GC.
Fig. 7 CV curves in 1mol/L KOH aqueous solution at 50 mV/s for (a) Pd/GC, Pd/AC, Pd@g-C3N4, Pd@g-C3N4(0.0625)/GC, (b) Pd@g-C3N4(n=1~0.0625)/GC; CV curves in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s for (c, d and e) Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC ((d) and (e) is the partial enlargement curves from (c)), (f and g) Pd@g-C3N4(n=1~0.0625)/GC ((g) is the partial enlargement curves from (f)).
Fig. 8 Schematic diagram of three-phase boundary for the Pd@g-C3N4/GC with lower (a) and higher (b) amount of g-C3N4.
Fig. 9 Nyquist plots (a) of EIS and its expanded high frequency region (b) for Pd/GC, Pd/AC, Pd@g-C3N4 and Pd@g-C3N4(0.0625)/GC.
Fig. 10 CV cycling test (200 cycles) in 1 mol/L KOH + 1 mol/L C2H5OH aqueous solution at 50 mV/s for Pd@g-C3N4(0.0625)/GC(a,b), Pd/GC(c,d), Pd/AC(e,f) ((b), (d) and (f) are the partial enlargement curves from (a), (c) and (e), respectively).
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10