1. Introduction

Platinum group metals (pgms) as NCs are widely studied for many applications, including spectroscopy (1–5), biology (6–8) and catalysis (9–15). Specifically of interest in catalysis is using NCs containing elements in addition to the pgms to reduce the amount of precious metal needed in the system. The use of solvothermally synthesised platinum-cobalt NCs to catalyse the electrochemical oxygen reduction reaction (ORR) is well reported as a method of providing high activity while minimising the precious metal loading of the catalyst (1, 16–20). However, many of the reports that propose these systems for use as cathode material in proton exchange membrane fuel cells (PEMFCs) (21, 22) utilise high-temperature reactions of up to 300°C and environmentally unfriendly solvents (1, 16, 20). Additionally, many of the chemicals typically used in the solvothermal colloidal synthesis are hazardous, and often require specialised and costly equipment for storage and handling like a glovebox; these include, for example, oleylamine, trioctylphosphine and cobalt(0) pentacarbonyl. For PEMFCs (23) to act as a viable, clean source of energy, as many steps of production as possible must follow proper green chemistry principles (24) and minimise potential hazards. Many reports have discussed methods to produce NCs following green methods; however, these often lack uniformity in size and morphology (25–31). In this report, we demonstrate a room temperature, aqueous synthesis of near monodisperse platinum-cobalt NCs using green electrochemical methods.

Electrochemical methods for the synthesis of single-component NC systems are well understood (27, 29, 31–45), and bimetallic NC systems are currently being investigated in the literature (27, 46–51). However, to the best of our knowledge, these systems lack uniformity in shape, size and composition (8, 27, 46–49, 51). In this report, we describe an electrochemical method capable of producing highly monodisperse and colloidally stable, sub-5 nm platinum-cobalt alloy NCs, which are catalytically superior in activity compared to pure platinum NCs (16, 20). The ability to produce NCs through green chemistry methods is key to being able to synthesise nanocatalysts with minimal environmental impact and at a lower cost. By creating more uniform species, we allow for the NCs to maintain a higher catalytic surface area and for a more uniform morphology of the NCs to be studied.

2. Methods

To perform the desired electrochemical synthesis of the alloy NCs, procedures were developed through the modification of previously reported methods (31, 37, 51). An overview of the synthetic procedure is shown in Figure 1. As shown in Figure 1(a), 60 mM of dimethylformamide (DMF) is added to the aqueous solution containing both the precious metal precursor and the cobalt precursor. A complete list of materials is included in Supporting Information Section S1. As seen in Figure 1(b), the potential cycle taken to undergo the synthesis involves one high potential segment allowing for nucleation, followed by a milder hold segment at –500 mV to allow for NC growth. The difference in reduction potential between the cobalt and platinum precursors can be overcome through the addition of small amounts of DMF, which selectively coordinates to the platinum surfaces allowing for the standard reduction potential of the platinum species (–730 mV vs. RHE) to approach that of cobalt (–280 mV vs. RHE) by nearly 200 mV (51). By briefly holding the electrolyte solution at –650 mA, the reduction potential of both species is surpassed, allowing for the reduction of the ionic species and subsequent nucleation.

Figure 1(c) shows that the electrolyte solution takes on an orange hue before synthesis occurs. The electrolyte solution is magnetically stirred at 300 rpm during the electrosynthesis to avoid any concentration gradients in the electrolyte solution or overpotential barriers. Additionally, the electrolyte solution is purged with nitrogen before the reaction to minimise the effects of trace gas compounds that may be present and affect the reaction. The currents obtained during synthesis are shown in Figure S1 in the Supporting Information and demonstrate that our method of constant potential electrolysis for synthesis causes no unexpected variations in current density, but instead produces a uniform current throughout the synthetic process.

As stated in Figure 1(a) and shown in Figure 1(d)–1(f), the working electrode used is either a platinum wire or a transparent conducting oxide (TCO) layer of indium tin oxide (ITO) on a glass substrate. For the cases of the platinum wire working electrode, we explore both syntheses without a surfactant (Figure 1(d)) and with the addition of 100 mM cetyltrimethylammonium bromide (CTAB) as a surfactant (Figure 1(e)). Without surfactant, the NCs do not aggregate, likely due to ionic stabilisation. We show that CTAB acts as a stabilising ligand during the reaction, allowing for colloidally stable NCs. In Figure 1(f), the case of an ITO working electrode is shown, without the presence of surfactant, yielding a NC coating on the electrode. This case was tested without the presence of a surfactant to allow electrodeposition to occur onto the TCO layer.

3. Results and Discussion

We can compare the synthetic results of the platinum-cobalt alloy NCs formed in the cases shown in Figure 1(d)–1(f); these include electrochemical synthesis without a stabilising agent, electrochemical synthesis with a stabilising agent and electrochemical deposition directly onto the electrode surface, respectively. We quantitatively compare monodispersity through size distribution from transmission electron microscopy (TEM), while a qualitative comparison is possible through the observation of change in the electrolytic solution. As shown in Figure 2(c), the electrolyte solution begins as a transparent orange solution; as shown in Figure S3 in the Supporting Information, while the electrolyte solution with CTAB present is the same hue but cloudy due to the addition of a surfactant. Figure 2 shows the results of electrochemical synthesis both without (Figure 2(a) and 2(b)) and with (Figure 2(c) and 2(d)) CTAB.

From Figure 2(a), when lacking the presence of a surfactant, the NCs are unable to form a stable colloidal suspension. The NCs, as shown in Figure 2(b), can also be seen to be sparse in the colloidal solution and not well-formed; the crystals have a dispersion of approximately 57%. In contrast, when the CTAB is added as a stabilising agent, the NCs form a stable, dilute, colloidal solution, as shown in Figure 2(c). This signifies that the CTAB is not only able to act as a surfactant but also acts in the reaction as the ligand species. By having a stabilising ligand present during the reaction, we can produce NCs with a significantly higher level of monodispersity than previously reported in the literature (31, 47, 48, 51). As shown in Figure 2(d), the NCs display a uniform spherical morphology, with a dispersion of approximately 29%. This demonstrates that the addition of stabilising ligands into the electrochemical solution before potential onset provides NCs with nearly double the monodispersity of NCs formed without CTAB.

In addition to the NCs produced through direct electrochemical synthesis, we tested the applicability of this synthetic method to electrodeposition. Electrodeposition provides the additional benefit of forming the NCs directly onto a useful substrate without the need for additional purification and transfer steps. In this work, we use ITO as a TCO electrode in place of the platinum working electrode used in electrosynthesis. This can be seen in Figure 1(a) and 1(f). The results obtained are summarised in Figure 3.

As seen in Figure 3(a), the NCs are smaller through electrodeposition than those formed through electrosynthesis, which is beneficial to catalytic purposes where a high surface area is desirable. However, at 35%, the dispersion of the NCs is higher than the NCs synthesised using CTAB. Again, we attribute this to the increase in the uniformity of the NCs provided by a stabilising ligand. The ITO slide after deposition, as shown in Figure 3(b), indicates that while not completely uniform, the NCs can form a layer directly on the ITO slide. This, in combination with the colour change in the electrolytic solution from orange to yellow seen in Figure 3(c), indicated that NCs were able to form and that the precursors used had been removed from the electrolytic solution during the deposition, indicating NCs formed with both platinum and cobalt, as neither forms a yellow solution individually. We attribute this to the synthesis of bimetallic NCs. To confirm that this is the case, we characterise the samples as shown in Figure 4. For this characterisation, we focus on the sample with the greatest synthetic uniformity: the electrosynthetic samples formed with CTAB as a stabilising agent.

From Figure 4(a), we can see diffraction peaks indicative of the random alloy face-centered-cubic crystal phase of platinum-cobalt NCs. These peaks are in correspondence to previous reports on this alloy (16, 20). The large background peak around 43° 2θ is believed to be due to either stacking faults in the NCs, or some existence of a pure cobalt phase. No major peaks indicative of either pure platinum or cobalt are observed, implying no significant phase segregation during the synthesis. In the high-resolution scanning transmission electron microscopy (STEM) micrograph in Figure 4(b), a face-centered cubic (fcc) lattice is observed, which is in correspondence with the crystal structure of the alloy NCs. This is also seen in Figure 4(c)–4(d), which shows individual particle-resolved STEM micrographs. These micrographs show lattice correspondence to a fcc A1 crystal phase (16, 18, 20). A bimetallic structure is again observed in Figure 4(g) through energy dispersive spectroscopy (EDS), where the cobalt and platinum are uniform throughout the NCs. This implies that the NCs are indeed uniform alloys and do not suffer from any phase segregation. The presence of both platinum and cobalt on a larger scale at a near 1:1 ratio can be seen in the EDS spectra in Figure S2 in the Supporting Information.

4. Conclusion

We have shown a green electrochemical method of NC synthesis capable of producing more uniform bimetallic alloy platinum-cobalt NCs than previously reported. The ability to synthesise catalytically relevant materials such as platinum-cobalt under green conditions is vital to ensure that in the process of producing catalysts to perform environmentally conscious tasks, we do not create harmful byproducts such as significant energy consumption to reach high temperatures or the release of harmful solvents into the environment. By producing small uniform NCs, we pave the way for pgm catalysts to be more economically viable to act as catalysts for applications such as PEMFCs. We believe that this method of NC synthesis will be readily applicable to many systems, allowing for green catalyst production for various applications. Through the understanding of electrochemical synthesis for both a direct synthesis method and an electrodeposition method, this work can be applied to systems where the catalyst can be directly electro-impregnated onto a support material. Through direct catalytic impregnation under green conditions, the methodology developed here will increase the commercial viability of pgm NCs for catalytic applications.