Background

Formic acid and methanol are potential in the conversions between chemical energy and electric energy as the chemical fuels. During earlier investigations, researchers have dismissed formic acid as a practical fuel because of the high overpotential evidenced by experiments, indicative that the reaction starting from formic acid appeared to be too difficult. The pristine Pt-based electrocatalysts have lowered the chemical barriers for conversion. Therefore, the bimetallic Pt-based nanomaterials were introduced as anode catalysts for methanol and formic acid oxidation [17]. Furthermore, Pt-M alloy and bimetallic core-shell nanoparticles promised for electrocatalytic reactions have been developed [57], which could enhance the electrocatalytic abilities by high CO-tolerance capability during the electrocatalysis. As a one-carbon-containing molecule similar to methanol, formic acid is a suitable organic molecule to be used directly in fuel cells, removing the need of complicated catalytic reforming. With a boiling point of 100.8 °C, formic acid is a liquid at standard temperature and pressure. Formic acid is accordingly much easier and much safer than hydrogen and methanol in energy storage because it does not need the high pressure or the low temperature operation. The efficiency of formic acid fuel cells can be engineered to a much higher level than that of methanol in addressing the crossover issue with the polymer membrane [8]. An alternative pathway leading to CO intermediate for the methanol or formic acid oxidation results in the poisoning of the active Pt surfaces with covalent CO bonding. To avoid this situation, two ways of improvement are considered usually to enhance the Pt activity with catalytic performance: (1) synthesis of the Pt-M bimetallic electrocatalysts and (2) addition of the carbon materials as support [911].

The carbon materials (nanoscale active carbon, carbon nanotubes, carbon fibers, and graphene) are used as support to the metal nanoparticles in order to enhance the electrochemical catalytic activities [1217]. Graphene is one of the most eye-catching materials for researchers in a variety of fields because of its high conductivity and superior mechanical/electrochemical property [1823]. However, with different methods in synthesized graphenes, these materials are produced with different properties [2429]. The drawbacks for graphene are its hydrophobicity and the aggregation behavior, which have limited the wet processes to a great extent. Some research groups were working on modification of the graphene surface by ionic liquids and/or polymeric materials [30, 31]. Poly(diallyldimethylammonium) chloride, abbreviated as PDDA, was introduced in this study because the functional groups and the non-covalent interactions of polymer with the graphene surface maintain the nice property of graphene and in addition enhance the electrocatalytic capability [32, 33]. Herein, we used a hydrothermal method to produce the Au nanoparticles and the bimetallic PtAu nanoparticles on PDDA-modified graphene sheets (Au/PPDA-G and PtAu/PDDA-G, respectively) and would use these novel nanocomposites to make formic acid oxidation processes.

Methods

PDDA-modified Graphene (PDDA-G)

The preparation of graphene followed the modified Hummer’s method. Into a 250-mL round-bottomed flask held in iced bath, it was introduced with concentrated sulfuric acid (25 mL) then fumed with nitric acid (10 mL in droplets) for 15 min. With iced bath, graphite powder (1 g) (Sigma-Aldrich, Ltd) was introduced into the round-bottomed flask under vigorous stirring. After being well mixed, the solution was added with potassium permanganate (22 g) for 30 min before the iced bath was removed. The solution was then stirred at room temperature for 96 h. Suitable (300 ~ 500 mL) amounts of doubly ion-exchanged (D.I.) water were added to the product mixture while kept in ice bath then the solution was centrifuged for removal of the supernatant. This procedure was performed for three times. The collected precipitates were rinsed with methanol for three times. The muddy-like solid was dried at 80 °C for 12 h to yield the graphite oxide.

In a 100-mL Teflon-lined autoclave with 65 mL D.I. water, 193.8 mg of graphite oxide and 2.54 mL of PDDA (35 wt. %) were added. Then, the autoclave was heated in an oven at 90 °C for 5 h then cooled to room temperature, before removal of the upper clear solution and the solid materials washed with D.I. water for three times. The muddy-like remains were dried at 80 °C overnight. The PDDA-G was then ready to proceed to the deposition of nanoparticles.

Nanoparticles Deposition on PDDA-G

Fifty milliliters of D.I. water, 100 mg PDDA-G, 131.1 mg H2AuCl4 (Sigma-Aldrich, Ltd), and 8 mL hydrazine monohydrate (98 %, Sigma-Aldrich, Ltd) were added and then the mixture was sonicated for 30 min before the mixture was transferred into a 100-mL Teflon-lined autoclave apparatus and heated at 90 °C for 24 h. The composite of Au nanoparticles on PDDA-modified graphene (Au/PDDA-G) as a solution was centrifuged three times with D.I. water followed with removal of supernatants. The residuals were dried under vacuum at 90 °C for 24 h and then collected.

The composite of bimetallic PtAu nanoparticles on PDDA-modified graphene (PtAu/PDDA-G) was obtained in a similar way by mixing 99 mg PDDA-modified graphene, 215.3 mg K2PtCl6 (~0.456 mmol), and 133 mg H2AuCl4 (~0.456 mmol) with 50 mL D.I. water and sonificated for 30 min, followed by addition of 10 mL hydrazine monohydrate (80 %) and sonificated further for 10 min. After that, the mixture was transferred to a 100-mL Teflon-lined autoclave apparatus and heated at 90 °C for 24 h. The collected PtAu/PDDA-G solutions were centrifuged three times with D.I. water followed with removal of supernatants. The residuals were dried under vacuum at 90 °C for 24 h and then collected. The cubic Pt/G was synthesized in the previous method [33].

X-ray Diffraction (XRD) Characterization

XRD patterns of nanocomposites were examined by Bruker D8 with CuKα X-ray source. The average crystalline sizes of Au/PDDA-G and Pt Au/PDDA-G were estimated by Debye-Scherrer equation (1)

$$ D=K\lambda /\beta \cos \theta $$
(1)

where D is the average crystalline sizes in nanometers, K is shape factor (0.94 for this case), λ is the wavelength of X-rays, Cu = 0.154 nm is the wavelength (nm), and β is full width at half maximum of peaks at 2θ in the pattern [34].

Transmission Electron Microscopy (TEM) Analysis

The TEM analysis employed a JOEL 2100 LaB6 light source, 200 keV, and equipped with EDX. The TEM samples were prepared with ethanol dispersion dropped on the grid Cu net for analysis. The selected area electron diffraction (SAED) was employed to reveal Au/PDDA-G and PtAu/PDDA-G electronic diffraction pattern and the lattice d-spacing.

Thermal Gravimetric Analysis (TGA)

TGA was examined with the Perkin Elmer Pyris 1 TGA from room temperature to 800 °C at a heating rate of 10 °C min−1 with about 5 mg of sample loading on the platinum plate.

X-ray Photoelectron Spectroscopic (XPS) Analysis

The XPS analysis was measured by Physical Electronics PHI Quantra XPS microprobe equipped with an Al monochromatic X-ray source (1486.6 eV, 45° incident angle).

Electrochemical Analysis

The electrochemical catalysts (11.3 mg Au/PDDA-G and 13.0 mg PtAu/PDDA-G, separately) were dispersed in 5 mL D.I. water with super sonification for 30 min. The electrochemical electrode was prepared by dropping 5 uL of the above dispersions on the glass-carbon electrode (0.0706 cm2) with constant-volume pipet and dried at room temperature. The procedure was repeated for three times. The electrochemical properties were recorded with an Autolab PGSTAT30 potentiostat/galvanostat. These specimens were investigated with the cycle voltammetry (CV, voltage scan range from −0.25 to 1.0 V, at a scan rate of 50 mV/s) and electrochemical impedance spectroscopy (EIS, scan frequency from 10 kHz to 100 mHz at potential 0.3 V with 5 mV amplified) [35].

Results and Discussion

Au/PDDA-G and PtAu/PDDA-G were synthesized easily in a one-pot process using the hydrothermal method at 90 °C for 24 h. These nanocomposites were with characteristic crystalline structures of the metal nanoparticles. Figure 1 shows the XRD results of the nanocomposite Au/PDDA-G, e.g., 2θ of Au plane (111) at 32.23°, (200) at 44.37°, (211) at 64.67°, (311) at 77.52°, and (222) at 81.88°, respectively The bimetallic nanocomposite of PtAu/PDDA-G exhibits 2θ of Pt plane (111) at 39.28°, (200) at 46.22°, and (220) at 67.35° in XRD patterns, respectively (Fig. 2). The estimation for the nanoparticle sizes was followed with Debye-Scherrer equation, shown in formula (1). The roughly calculated particle size for Au/PDDA-G is 32.9 nm and PtAu/PDDA-G is 1.8 nm, respectively. Compared to TEM results, the size distributions of the Au nanoparticles and of the bimetallic PtAu on PDDA-modified graphene sheets are 20–50 and 5–10 nm, similar with the XRD results, as shown in Figs. 1ac and 2ac, respectively. The PDDA-modified graphene sheet (PDDA-G) shows an XRD peak at 2θ = 21.35°, with layer distance seems to be longer than that of pristine graphene sheets (200) at 2θ 25°–26° [36], which means that PDDA has immobilized on the graphene sheets.

Fig. 1
figure 1

TEM and XRD results of Au/PDDA-G electrocatalyst, at a low magnification; b high-resolution TEM and lattice image analysis; c intermediate magnification; d XRD pattern

Fig. 2
figure 2

TEM and XRD results of PtAu/PDDA-G electrocatalyst, at a low magnification; b high-resolution TEM image; c SAED pattern; d XRD pattern

The thermal analysis (TGA) results in Fig. 3 indicate the metal nanoparticles loading on the PDDA-modified graphene. The PtAu/PDDA-G nanocomposite exhibited weight loss between 350 and 550 °C and the Au/PPDA-G was observed in 450–500 °C. The better the thermal stability attributed to the alloyed PtAu attached on the PDDA-G surface could absorb, the higher the thermal energy of Au/PDDA-G, which increases the temperature of the decomposition. Furthermore, the inorganic content (metal + graphene) of Au/PDDA-G and bimetallic PtAu/PDDA-G is about 54.2 and 76.2 wt. %, respectively. If the content of PDDA and graphene is constant, TGA results show that higher ratios of the bimetallic PtAu nanoparticles were adsorbed on the surface of PDDA-graphene substrate, compared to the Au/PDDA-G system.

Fig. 3
figure 3

TGA results of Au/PDDA-G and PtAu/PDDA-G electrocatalysts

The metal nanoparticles deposited on the PDDA-modified graphene were characterized also with XPS. The Au 4f and Pt 4f binding energies are shown in Table 1, Figs. 4 and 5, respectively. The Au0 4f binding energies are 83.9 ~ 83.7 and 87.7 ~ 87.5 eV in the Au/PDDA-G sample and in the bimetallic PtAu/PDDA-G sample, respectively. The Pt 4f binding energies were curve fitted to resolve the Pt0 and PtII states. The binding energy for Au is shifted by 0.2 eV from Au/PDDA-G to PtAu/PDDA-G, consistent to a PtAu alloy chemical environment. The XPS spectral deconvolution results on C1s, O1s, and N1s for Au/PDDA-G and PtAu/PDDA-G were shown in Table 1. The N-1s binding energies decrease from PDDA-G, Au/PDDA-G to PtAu/PDDA-G, attributed to metal deposition on the surface of PDDA-G.

Table 1 XPS spectral fittings of Au/PDDA-G, PtAu/PDDA-G, and PDDA-G electrocatalysts
Fig. 4
figure 4

XPS spectra of Au 4f and Pt 4f binding energies in the Au/PDDA-G and PtAu/PDDA-G electrocatalyst

Fig. 5
figure 5

The electrochemical study in aqueous 0.5 M H2SO4 + 0.5 M HCOOH. a The cyclic voltammetric study in 0.5 M aqueous H2SO4. The inset is the same electrical chemical study for Au nanoparticles. b The HCOOH oxidation test for three electrocatalysts in 0.0–1.0 V range. The numbers are the area ratios with respect to the dash baselines. The inset is the test for Au/PDDA and Au, respectively

The Au/PDDA-G and PtAu/PDDA-G were studied electrochemically in aqueous solutions made up of 0.5 M H2SO4 only and made up of 0.5 M H2SO4 + 0.5 M HCOOH, for determination of electrocatalytical activities. Figure 5a exhibits the cyclic voltammetric results of Au, Au/PDDA-G, and bimetallic PtAu/PDDA-G. The PtAu/PDDA-G specimen clearly demonstrates the hydrogen redox behavior but not the Au-based nanocomposites. The Au specimen works similarly to the Au/PDDA-G specimen, with preference to oxygen reduction reaction, however, the hydrogen redox abilities do not work similarly to the Pt-based electrocatalysts.

The formic acid oxidation on Pt usually follows the so-called dual pathway [37].

I: Direct dehydrogenation producing CO2. (iPI)

$$ \mathrm{HCOOH}\to {\mathrm{H}\mathrm{COO}}_{\mathrm{ads}}+{\mathrm{H}}^{+} + {\mathrm{e}}^{-} $$
(2)
$$ {\mathrm{H}\mathrm{COO}}_{\mathrm{ads}}\to {\mathrm{CO}}_2 + {\mathrm{H}}^{+} + {\mathrm{e}}^{-} $$
(3)

II: Dehydration generating CO (poisoning intermediate). (iPII)

$$ \mathrm{HCOOH}\to \mathrm{C}\mathrm{O} + {\mathrm{H}}_2\mathrm{O}\to {\mathrm{CO}}_2 + 2{\mathrm{H}}^{+}+2{\mathrm{e}}^{-} $$
(4)

The peaks at around 0.5 V (iPI) are attributed to the oxidation of HCOOH to CO2; the peaks at about 0.92 V (iPII) are due to the presence of CO on surface of metal catalysts (denoted as COads, as generated from the dissociative adsorption step, whose intensity indicates the amount of COads on the surface of electrocatalysts) [37, 24]. A high iPI/iPII ratio (5.36 vs. 0.84) together with a low onset potential (0.20 V vs. 0.25 V) for PtAu/PDDA-G vs. Au/PDDA-G, respectively, indicates that the PtAu/PDDA-G prefers the direct dehydrogenation branch in formic acid oxidation, see Fig. 5b.

The electrochemical impedance spectroscopy study at constant 0.3 V revealed the resistance of HCOOH oxidation for the Au/PDDA-G and for the PtAu/PDDA-G nanocomposites. The Nyguist plots are shown in Fig. 6. The electrochemical impedance of Au/PDDA-G was apparently much greater than that of the PtAu/PDDA-G, both being with PDDA-modified graphene sheets as support. The HCOOH oxidation utilizing Au/PDDA-G has very low activity and very high resistance when compared to that utilizing PtAu/PDDA-G (Figs. 5 and 6). The equivalent circuit diagrams for Au/PDDA-G and for PtAu/PDDA-G were shown in Fig. 6, applying the forms of R s (R p C) and R s (R p1 C 1 )(R p2 C 1 ), respectively, where R s is for the resistance of electrode, R p is for the chemical reaction/electron resistance, and C is for the electrochemical capacitance attributed to the double layers or interfaces. The R s of Au/PDDA-G is 9.75 ohm, smaller than that of PtAu/PDDA-G, 12.3 ohm. Thus, the resistance of electrode for Au/PDDA-G is smaller than that for PtAu/PDDA-G. On the other hand, the chemical reaction/electron resistance is reverse in order: R p1 (51.0 ohm) and R p2 (22.2 kohm) for PtAu/PDDA are both smaller than R p (26.6 kohm) for Au/PDDA-G.

Fig. 6
figure 6

EIS and fitting results for Au/PDDA-G and PtAu/PDDA-G electrocatalysts. The upper and lower insertions depict equivalent circuit simulations for Au/PDDA-G and PtAu/PDDA-G electrocatalysts, respectively

For catalytic HCOOH oxidation activities, the cubic Pt/G was found very active but the activity followed a hyperbolic decay to ~5 mA/mgcm2 in 1800 s. The PtAu/PDDA-G was found very active during the initial 10 s and then the activity followed a slow, linear drop to 18 mA/mgcm2 after 1800 s. However, the Au/PDDA-G was found to exhibit bad performance on oxidation and bad stability. The long-term stability in decreasing order shows as follows: PtAu/PDDA-G > cubic Pt/G > Au/PDDA-G, as shown in Fig. 7.

Fig. 7
figure 7

The electrochemical catalysts durability test for formic acid oxidation at the potential 300 mV for about 3000 s. The blue line is the result for PtAu/PDDA-G, the black line is for cubic Pt (60 wt.%)/G, and the yellow line is for Au/PDDA-G

Conclusions

We have successfully used a one-pot hydrothermal method to synthesize the Au/PDDA-G and the bimetallic PtAu/PDDA-G nanocomposites for formic acid oxidation. TEM, XRD, and XPS results illustrate that the Au (20–50 nm) and bimetallic PtAu (5–10 nm) nanoparticle could grow well on the graphene sheets. From the oxidation activity test, PtAu/PDDA-G exhibits the highest iPI/iPII ratio, represents the best capability in anti-CO poisoning, and hence displays the greatest stability in long-term operation. The novel PtAu/PDDA-G is potential to be applied in the electrocatalyst for formic acid oxidation.