CHEMICAL SCIENCE

Ultra-low platinum load proton exchange membrane fuel cells


On July 25, 2022, the research group of Professor Huang Yu, chair of the Department of Materials Science and Engineering at the University of California, Los Angeles (UCLA), published an article in the journal Nature Nanotechnology titled “Graphene-nanopocket-encaged PtCo nanocatalysts for highly durable fuel cell operation under.” demanding ultralow-Pt-loading condition”.

The results report a PtCo nanocatalyst protected by graphene nanocapsules and applied to proton exchange membrane fuel cells as cathode catalysts, thereby achieving fuel cell power density and durability under challenging ultra-low platinum load conditions while reaching the international advanced level, it is expected that the application of this report catalyst can reduce the amount of platinum group metals required for fuel cell vehicles to the same level as the level required for internal combustion engine vehicle exhaust gas treatment. This means that the platinum group metals required for fuel cells are greatly reduced and their large-scale application is no longer limited by the extremely limited reserves and production of platinum group metals, paving the way for further large-scale applications of fuel cells.

The corresponding author of the paper is Professor Huang Yu; The first author, Professor Zhao Zipeng, is currently working at Beijing Institute of Technology; First author Zeyan Liu is currently studying at the University of California, Los Angeles (UCLA).

Research implications

As a green power device that can replace the internal combustion engine, proton exchange membrane fuel cells (PEMFCs) can theoretically not rely on carbon-containing fossil energy, so the development and application of PEMFCs is of great significance to the realization of the national development goals of “carbon peaking” and “carbon neutrality”. For currently commercially available PEMFCs, platinum group metals (PGM) are irreplaceable catalyst materials, especially those used to accelerate the slow oxygen reduction reaction (ORR) of the cathode. Platinum group metals are also needed for combustion engine cars and used in exhaust gas catalytic treatment. At present, the demand for the automobile industry accounts for nearly half of the global PGM production, and the current fuel cell vehicles need 5-10 times the platinum group metal to internal combustion engine vehicles, with the current fuel cell PGM demand, fuel cell vehicles to replace the internal combustion engine vehicle is bound to lead to a shortage of platinum group metals, and because the global reserves and production of platinum group metals are extremely limited, which is a great obstacle to the large-scale application of fuel cells. In previous application scenarios, the reduction of platinum group metal loads often leads to the sacrifice of fuel cell device performance and poor stability, which has prompted researchers around the world to develop catalysts with higher catalytic activity and more stable to achieve the reduction of the demand for platinum group metals for fuel cell vehicles to the level of internal combustion engine vehicles.

Difficult work

In ordinary studies, the intrinsic mass activity (MA) of the PGM catalyst is determined by the specific area activity (SA) and the electrochemical surface area (ECSA). On the one hand, increasing the SA of the PGM catalyst, that is, by increasing the conversion frequency (TOF) of a single active site, can ensure the high catalytic activity of the entire fuel cell unit at ultra-low PGM loading. However, the increase in TOF will undoubtedly increase the pressure of the material transport, because this means that the reactants need to be transported to a single active site faster and the product needs to be removed more quickly. On the other hand, reducing the size of the catalyst and constructing ultra-fine nanocatalysts to have a high ECSA is also an idea to solve the problem of limited catalytic activity at ultra-low PGM loads. However, ultra-fine nanoparticles with a high surface area-to-volume ratio are thermodynamically substable, so that nanoparticles tend to grow significantly through physical aggregation or Ostwald curing process, which will cause gradual losses of ECSA and MA, and stability is difficult to meet the requirements. Therefore, the realization of long-range stability of PEMFCs with ultra-low PGM loads is a very serious challenge, which requires the ultra-small size of the catalyst and excellent stability in the design of the nanocatalyst.

Innovative ideas

In light of this, the journal Nature Nanotechnology recently reported the design and synthesis of a very stable ultra-fine PtCo nanocatalyst protected by graphene nanobuffs for PEMFCs with ultra-low precious metal loads (Figure 1a). Through this design, the ultra-fine nanocatalyst is protected in the graphene nano pocket while ensuring electrochemical reachability, but also limits the aggregation of the catalyst, and alleviates the oxidative dissolution of the catalyst and the Ostwald curing process. Even under very harsh conditions at ultra-low PGM loads, this special structure ensures excellent activity and excellent catalytic stability. Fuel cells with this catalyst exhibit excellent performance and durability in challenging ultra-low PGM load conditions.

Research content

Figure 1: Schematic diagram of protective nano pocket design and PtCo@Gnp characterization.

PtCo@Gnp catalyst is prepared by loading the precursors of organometals Pt and Co (Pt(acac)2 and Co(acac)2) onto the carbon matrix, followed by annealing at 700 °C, and finally the pyrolysis catalyst is pickled in sulfuric acid to remove the more easily dissolved Co. Transmission electron microscopy (TEM) studies have shown that ultrafine nanoparticles are evenly distributed on carbon carriers with mass-weighted average sizes of 3.0±0.8 nm (Figures 1b, c). Powder X-ray diffraction studies have shown that the resulting catalyst is a face-centered cubic structure, the same as Pt/C. Analysis of the overall composition by inductively coupled plasma atomic emission spectroscopy (ICP-AES) yields a Pt:Co atomic ratio of 80.8:19.2, while surface-sensitive X-ray photoelectron spectroscopy shows a higher Pt:Co ratio (89.4:10.6), suggesting that the catalyst has a shell-rich core-shell structure, which has also been confirmed by scanning transmission electron microscopy (STEM) and corresponding energy dispersive X-ray spectroscopy (EDS) element mapping studies (Figure 1d). With the help of high-resolution STEM, further analysis of PtCo@Gnp. A closer comparison of high-angle circular darkfield and brightfield STEM images revealed that ultrafine PtCo nanoparticles were clearly wrapped in nanoscale pockets consisting of a single or several layers of graphene shells (Figure 1e). In particular, the brightfield images also clearly show that PtCo nanoparticles are well supported on carbon skeleton structures with clearly resolvable graphite layers. It’s also worth noting that additional layers of graphite (usually single layers of graphene) surrounding the PtCo nanoparticles are also clearly visible. After a thorough analytical examination, the results showed that all PtCo nanoparticles were encased in similar graphene nanobags (Figure 1e). The presence of a graphene outer layer around PtCo nanoparticles was also confirmed by electron energy loss spectroscopy (EELS) mapping. It is worth noting that there is usually a nanoscale spacing (~0.4-1.0 nm) between this graphene nanocapsule and the surface of PtCo, that is, a non-contact shell is formed, which makes the PtCo@Gnp electrochemically accessible. In addition, the PtCo@Gnp has excellent ECSA (68.7 m2 gPGM-1), significantly higher than Pt/C (34.7 m2 gPGM-1), which is in line with the expected ECSA height corresponding to the measured particle size. Similarly, the oxygen reduction reaction SA of PtCo@Gnp (1.62 mA cm-2) measured using rotating disc electrodes is also 2.7 times that of Pt/C (0.61 mA cm-2). The above features clearly indicate that graphene nanocapsules may be porous, which makes the catalyst surface accessible and therefore PtCo@Gnp electrochemically active.

Figure 2: MEA performance of Pt/C, c-PtCo/C and PtCo@Gnp catalysts and representative catalyst comparison in the literature.

The researchers evaluated the performance of PtCo@Gnp in fuel cells with ultra-low precious metal loads against the DOE 2020 performance target, and the researchers prepared two different cathodes with PGM loads of 0.090 and 0.060 mgPGM cm-2, respectively. Together with an anode load of 0.010 mgPGM cm-2 (commercial Pt/C), the total MEA PGM load is 0.100 and 0.070 mgPGM cm-2, respectively. The 0.070 mgPGM cm-2 PGM load is only 19% of the PGM load (0.365 mgPGM cm-2) in the Toyota Mirai fuel cell vehicle and is close to the ultimate target of doe platinum group metal load (0.0625 mgPGM cm-2). Under cathode PGM loadings of 0.090 mgPGM cm-2 and 0.060 mgPGM cm-2, respectively, the initial MA of the PtCo@Gnp was 1.14 A mgPGM-1 and 1.21 APGM-1 (Figure 2), significantly higher than Pt/C (0.42 and 0.40 A mgPGM-1), commercial PtCo/C catalyst (c-PtCo) (0.57 A mgPGM-1) and DOE target (0.44 A). mgPGM-1)。 The researchers also performed the ADT test, referring to the latest 30,000-cycle square-wave test, which held the cathode at 0.6 V and 0.95 V for 3 s in each cycle. Even at ultra-low loads (0.070 mgPGM cm-2), PtCo@Gnp retained 73% of its initial MA after the more challenging square-wave ADT, which was much higher than the same load amounts of Pt/C (25%) and c-PtCo/C (30%). As a result, PtCo@Gnp based on harsh ultra-low PGM load levels exhibit very high durability in square wave ADTs (Figure 2d). In addition, it is worth noting that the EOL MA (0.89 A mgPGM-1) of the PtCo@Gnp is already 3 times higher than the DOE target (0.264 A mgPGM-1), 5 times higher than c-PtCo/C (0.17 A mgPGM-1), and nearly 9 times that of Pt/C (0.10 A mgPGM-1) (Table 1), which is one of the highest EOL MA reported in MEA tests to date (Figure 2d).

Table 1: Comparison of MEA key performance in PtCo@Gnp-based MEA and related literature.

Figure 3: MEA polarization curve under ultra-low PGM load (0.07 mgPGM cm-2 total cathode and anode) tested in a hydrogen/oxygen environment.

Although the cathode MA test under pure oxygen better assesses intrinsic activity that is less affected by mass transfer problems, the power rating test using air as an oxygen source directly reflects the actual performance of PEMFC in the operating environment. Following the recommended DOE test method, the MEA’s power rating is evaluated at 0.67V and the fuel cell operates at 94°C。 In the range of current densities from low to high, PtCo@Gnp exhibit significantly better performance than Pt/C or c-PtCo/C (Figure 3a). In particular, PtCo@Gnp provide mass normalized power ratings of 10.1 and 13.2 W mgPGM-1 at PGM loads of 0.100 and 0.070 mgPGM cm-2 (Figure 3d), both of which greatly exceed the DOE target (8 W mgPGM-1) (Figure 3e). Impressively, PtCo@Gnp exhibits excellent durability, far superior to Pt/C and c-PtCo/C (Table 1), which can also be reflected in the EOL power rating performance after ADT. Especially at ultra-low PGM loads of 0.070 mgPGM cm-2, the EOL rating for PtCo@Gnp is 11.4 W mgPGM-1, far exceeding Pt/C (3.8 W mgPGM-1) (Figures 3e and Table 1). Similarly, even at ultra-low PGM loads of 0.070 mgPGM cm-2, the voltage loss PtCo@Gnp at 0.8 A cm-2 is only 18.8 mV (meeting the DOE target loss < 30 mV), nearly an order of magnitude smaller than Pt/C (163 mV) or c-PtCo/C (100.8 mV) at the same load (Figures 3f and Table 1), which highlights a significant improvement in PtCo@Gnp stability. The researchers also further investigated area normalization power ratings, another key parameter for calibrating performance in practical applications. At a PGM load of 0.100 mgPGM cm-2, the PtCo@Gnp can deliver a rated power rating of 1.01 W cm-2, meeting the DOE target (1.0 W cm-2) and outperforming Pt/C (0.91 W cm-2). PtCo@Gnp’s superior durability is particularly evident in its excellent EOL area power rating. After ADT, PtCo@Gnp still has a 0.87 W cm-2 EOL rating at a PGM load of 0.100 mgPGM cm-2, significantly better than Pt/C (0.57 W cm-2). In addition, at an ultra-low PGM load of 0.070 mgPGM cm-2, PtCo@Gnp MEA exhibits an impressive EOL area rating of 0.80 W cm-2, far exceeding Pt/C (0.27 W cm-2) or c-PtCo/C (0.52 W cm-2) at the same load (Table 1), indicating that PtCo@Gnp have extraordinary durability even under more demanding ultra-low PGM loads. Therefore, when the PtCo@Gnp catalyst operates under demanding ultra-low load conditions, all of its EOL performance specifications exceed the DOE target (Table 1). This significantly improved EOL performance provides a more uniform power output throughout its lifecycle, which is much needed for practical applications. In addition, based on linear extrapolation of MA degradation or voltage loss, this significantly improved durability that has exceeded the DOE target can significantly extend the life of the fuel cell by approximately 50%.

Figure 4: Catalyst characterization, particle size distribution analysis, and corresponding MEA test results for EOL

To understand the significant stability differences between the different catalysts, the researchers characterized and compared the different catalysts after undergoing ADT in MEA. Firstly, the change of nanoparticle size distribution after ADT was analyzed. In order to correctly reflect the mass fraction of particles of different sizes, mass-weighted size distributions are plotted. Overall, the mass-weighted size of Pt/C increased sharply from 5.7 ±2.0 nm at cycle start (BOL) to 12.2±5.6 nm in EOL (Figure 4a). A similar size increase was observed in c-PtCo/C (Figure 4b). This increase in nanoparticle size is mainly attributable to: (1) nanoparticle separation, movement, and then aggregation, and (2) oxidative dissolution, diffusion, and Oosterwald curing processes. To this end, the graphene shell can effectively prevent the movement and aggregation of nanoparticles, and greatly delay oxidative dissolution and diffusion, because it largely retains the dissolved Pt atoms in the pocket of graphene, which can be re-deposited on PtCo nanoparticles, thus helping to maintain the size of ultra-fine PtCo nanoparticles. In fact, the mass-weighted average size of PtCo nanoparticles in the PtCo@Gnp has only moderately increased from 3.4±1.1 nm of BOL to 5.1±1.7 nm of EOL (Figure 4c), which clearly indicates that the presence of graphene nanocapsules effectively inhibits size growth and improves the overall durability of PtCo nanocatalysts. High-resolution STEM images and EDS plots confirm that the PtCo nanoparticles in the PtCo@Gnp still retain their core-shell structure and Pt-rich shell when EOL (Figure 4d). It is also worth noting that although protective graphene nanocapsules remain on PtCo nanoparticles, the space between them has decreased (Figure 4e, f). This is most likely because graphene nanocapsacks become more hydrophilic after ADT, collapsing onto PtCo nanoparticles due to capillary forces during TEM sample preparation. The small increase in mass-weighted size in the PtCo@Gnp (from 3.4 to 5.1 nm) allows it to retain a relatively high ECSA (32.4 m2 gPGM-1) at EOL, significantly higher than Pt/C (13.2 m2 gPGM-1) and c-PtCo/C (13.6 m2 gPGM-1) (Figure 4g). It can also be demonstrated through oxygen transport resistance studies PtCo@Gnp can maintain a high ECSA during fuel cell operation. Because the pressure-independent oxygen transfer resistance (RP-Ind) in MEA is inversely proportional to the number of active sites per unit area of the electrode. Compared with significant increases in RP-Ind in Pt/C (640%) and c-PtCo/C (265%) MEAs, PtCo@Gnp MEA showed a much smaller increase after ADT, at only about 100% (Figure 4h). The sharp increase in RP-Ind of the Pt/C electrode can be attributed to a significant increase in particle size (12.2 nm) and a significant decrease in particle density (from 5.2×1013 to 8.1×1012 cm-2), which results in a significant increase in the resistance of oxygen diffusion to the active site. In contrast, PtCo@Gnp retained a relatively small size (5.1 nm) and a high particle density (5.7×1013 cm-2) at EOL to ensure a sufficiently low RP-Ind, which is critical for long-term stability (Figure 4i).

Summarize the outlook

By wrapping the ultra-fine nanocatalyst in graphene nanocapsacks, the researchers developed a new PtCo@Gnp design that shows excellent activity and stability in actual MEA applications.

1) The catalyst guarantees highly robust performance of PEMFCs at ultra-low PGM loading (0.070 mgPGM cm-2), ma up to 1.21 A gPGM-1, rated power up to 13.2 W mgPGM-1 and excellent durability (MA retention after ADT is 73%, voltage loss at 0.8 A cm-2 is only 18.8 mV), all of which exceed the relevant DOE 2020 target.

2) With the high power rating and high durability of the MEA at ultra-low PGM loads, the PtCo@Gnp is expected to significantly reduce the PGM required for 90 kW fuel cell vehicles to about 6.8 grams, which is comparable to the PGM (2-8 grams) load in the exhaust gas catalytic converter of internal combustion engine-powered vehicles.

The catalyst reported in this article achieves both high activity and excellent stability under more challenging ultra-low load conditions, and the catalyst reported in the application article can greatly reduce the amount of platinum group metals used in peMFCs in practical applications, and it can be expected that the total mass of platinum group metals required for a fuel cell vehicle is roughly equivalent to that of an internal combustion engine vehicle, thereby significantly reducing costs in large-scale production. And in the process of large-scale replacement of internal combustion engine vehicles by fuel cell vehicles, the impact and restrictions of platinum group metal reserves and market supply will be mitigated. The work covered in this article will be a milestone step in the large-scale rollout of PEMFCs. (Source: Science Network)

Related paper information:https://doi.org/10.1038/s41565-022-01170-9



Source link

Related Articles

Leave a Reply

Your email address will not be published. Required fields are marked *

Back to top button