Skip to main content
NCBI home page
iScience logoLink to iScience
. 2020 Sep 28;23(10):101613. doi: 10.1016/j.isci.2020.101613

Few-Atomic-Layers Iron for Hydrogen Evolution from Water by Photoelectrocatalysis

Baowen Zhou 1,2,7, Pengfei Ou 3,7, Roksana Tonny Rashid 2,7, Srinivas Vanka 1, Kai Sun 4, Lin Yao 5, Haiding Sun 6, Jun Song 3,∗, Zetian Mi 1,2,8,∗∗
PMCID: PMC7559863  PMID: 33089102

Summary

The carbon-free production of hydrogen from water splitting holds grand promise for the critical energy and environmental challenges. Herein, few-atomic-layers iron (FeFAL) anchored on GaN nanowire arrays (NWs) is demonstrated as a highly active hydrogen evolution reaction catalyst, attributing to the spatial confinement and the nitrogen-terminated surface of GaN NWs. Based on density functional theory calculations, the hydrogen adsorption on FeFAL:GaN NWs is found to exhibit a significantly low free energy of −0.13 eV, indicative of high activity. Meanwhile, its outstanding optoelectronic properties are realized by the strong electronic coupling between atomic iron layers and GaN(10ī0) together with the nearly defect-free GaN NWs. As a result, FeFAL:GaN NWs/n+-p Si exhibits a prominent current density of ∼ −30 mA cm−2 at an overpotential of ∼0.2 V versus reversible hydrogen electrode with a decent onset potential of +0.35 V and 98% Faradaic efficiency in 0.5 mol/L KHCO3 aqueous solution under standard one-sun illumination.

Subject Areas: Catalysis, Electrochemical Energy Production, Nanoelectrochemistry, Atomic Electronic Structure, Nanomaterials

Graphical Abstract

graphic file with name fx1.jpg

Open in a new tab

Highlights

  • •

    Few-atomic-layers iron was anchored on GaN nanowires as an efficient HER catalyst

  • •

    The spatial-confinement and N-rich GaN is essential for forming atomic iron layers

  • •

    Low hydrogen absorption free energy is theoretically revealed over Fe3L:GaN

  • •

    The device exhibits a prominent performance for PEC water splitting to H2

Catalysis; Electrochemical Energy Production; Nanoelectrochemistry; Atomic Electronic Structure; Nanomaterials

Introduction

Photoelectrocatalytic (PEC) water splitting by utilization of solar energy and electricity presents one carbon-free route for the production of hydrogen, which holds grand promise for the worldwide challenges that we face today (e.g., energy shortage, global warming, and environmental issues) (Vijselaar et al., 2018). An efficient HER catalyst is at the core of a PEC cell for solar-driven water splitting. Up to now, platinum is well known as the state-of-the-art HER electrocatalyst (Cheng et al., 2016). Unfortunately, the high price and low abundance severely limit its large-scale applications. In this context, the pursuit of noble-metal-free HER electrocatalysts has attracted considerable interest over the past decades (Hui et al., 2018; Ye et al., 2018; Gao et al., 2017; Arif et al., 2018). So far, a wide variety of earth-abundant materials such as phosphides (Wexler et al., 2018; Shi et al., 2016), and chalcogenides (Feng et al., 2018; Zhou et al., 2018; Meng et al., 2015) have been developed as promising substitutes for platinum arising from their low cost and high activity. However, their overall efficiencies are still far below the demand of commercial applications and acidic media is generally a prerequisite for these materials to produce hydrogen with high efficiency (Ji et al., 2015; Zhang et al., 2016). From the viewpoint of environmental accountability and sustainability, hydrogen evolution in near-neutral/alkaline medium is relatively benign compared with that conducted in strong acid aqueous solution (Zhao et al., 2018; Jin et al., 2014; Hashemi et al., 2015). It is therefore necessary to develop a new, efficient, and inexpensive electrocatalyst for PEC water splitting, especially in near-neutral/alkaline aqueous medium.

[FeFe]-hydrogenase, as one homogeneous metalloenzyme from green plant, is the most efficient HER biocatalyst, owing to its unique atomic structure, well-defined catalytic centers, and superior metal-utilization efficiency (Camara and Rauchfuss, 2012; Wodrich and Hu, 2018; Tard and Pickett, 2009; Pullen, et al., 2013; Le Goff et al., 2009). Inspired by this masterpiece of nature, tremendous efforts have been devoted to exploring iron-based hydrogenase mimics for water splitting toward hydrogen. For example, Klumer and co-workers reported an assembly of chromophores to a bis(thiolate)-bridged diiron ([2Fe2S]) as catalyst for hydrogen production by using a modular supramolecular approach (Kluwer et al., 2009). Wu et al. developed a set of water-soluble [FeFe]-hydrogenase mimics and, by integrating with CdSe quantum dots, demonstrated superior activity for photocatalytic hydrogen production in water (Wang et al., 2013; Li et al., 2013). In spite of significant progress, these homogeneous hydrogenase mimics are restricted by a series of shortcomings, including complex fabrication, inherent fragility, and great difficulty in scaling up to industrial applications.

Atomically dispersed metals are emerging as a rising star of heterogeneous catalysts with impressive homogeneous features such as well-defined catalytic centers, low-coordination environment, and high-efficiency atom utilization (Wang et al., 2018). In addition, atomic-level catalysts possess strong metal-support interactions and high surface energy, thus presenting great promise to achieve high performance for various chemical reactions. It is worth mentioning that atomically dispersed metals have grand potential in water splitting (Xue et al., 2018; Hui et al., 2019; Yu et al., 2019; Zhang et al., 2018). For example, atomic-scale cobalt was supported on nitrogen-doped graphene as an efficient and inexpensive electrocatalyst for hydrogen generation from water splitting, wherein catalytical sites are associated with the metal centers coordinated to nitrogen (Fei et al., 2015). Similarly, tuned by electrochemical methods, atomically dispersed nickel species were anchored on graphitized carbon for electrocatalytic water splitting toward hydrogen (Fan, et al., 2016). Despite great promise, the area of atomically dispersed metals in catalyzing hydrogen evolution reaction is still in the infant stage, and the exploration of heterogeneous atomic-level iron catalyst for hydrogen formation is of significance but particular challenge (Cao et al., 2017).

In this work, by utilizing the unique spatial confinement and N-terminated feature of GaN nanowire arrays (NWs), few-atomic-layers iron (FeFAL) is anchored onto the lateral surface of GaN NWs on wafer-scale n+-p silicon junction as an efficient catalyst for PEC water splitting toward hydrogen. Theoretically, density functional theory (DFT) calculations suggest that few-atomic-layers iron is remarkably favorable for hydrogen evolution with an extremely low free energy of hydrogen adsorption. Moreover, strong electronic coupling between few-atomic-layers iron and GaN(10ī0) together with the well-defined GaN NWs with nearly defect-free structure enables superior optoelectronic properties. Experimentally, the monolithically integrated FeFAL:GaN NWs/n+-p Si demonstrates high activity for water splitting toward hydrogen. A high photocurrent density of −15.6 mA cm−2 is acquired at 0 V versus RHE with a decent onset potential of +0.35 V and high Faradaic efficiency of 98% in 0.5 M KHCO3 aqueous solution at argon atmosphere under standard one-sun illumination (AM 1.5 G, 100 mW cm−2). It is worth noting that a nearly saturated and high current density of ∼ −30 mA cm−2 is achieved at a minor overpotential of ∼0.2 V. To our best knowledge, this is the utilization of atom-scale iron as an inexpensive and efficient catalyst for hydrogen production in near-neutral/alkaline aqueous medium for the first time.

Results and Discussion

Fabrication and Characterization of the Photocathodes

Vertically aligned GaN NWs were grown on 2-inch n+-p silicon wafer (GaN NWs/n+-p Si) by radio frequency plasma-assisted molecular beam epitaxy (PA-MBE) (please see Figures 1 and S1, and Transparent Methods section). By tailoring the growth conditions, e.g., growing under N-rich conditions, the epitaxial GaN NW surfaces were engineered to be terminated with abundant nitrogen atoms, not only for their top c-plane but also for the lateral nonpolar (10ī0) surfaces (Kibria et al., 2016). Such abundant nitrogen coordinating sites provide sufficient anchors for stabilizing atomically dispersed metals (Zhang et al., 2017; Li et al., 2018). What is more, the spatial confinement arising from nanowire arrays is favorable for atomically dispersed iron (Chen et al., 2018). As such, few-atomic-layers iron was readily deposited onto the lateral m-plane of GaN NWs by a simple electrocatalytic process (Figure 1). The morphology of Fex:GaN NWs/n+-p could be tailored by modulating the electrodeposition cycles (x denotes the number of deposition cycles). In the absence of GaN NWs, the iron cocatalyst was directly loaded on silicon through the identical process for comparison.

Figure 1.

Figure 1

Open in a new tab

Schematic Diagram for the Decoration of Few-Atomic-Layers Iron onto N-Terminated GaN NWs by a Combination of Plasma-Assisted Molecular Beam Epitaxy with Electrodeposition

The structure and composition were characterized using scanning electron microscopy (SEM), scanning transmission electron microscopy (STEM), and X-ray photoelectron spectroscopy (XPS). The results are illustrated in Figure 2. SEM images in Figures 2A and S2 show that, in the absence of GaN NWs, the iron cocatalyst on silicon substrate (Fe/n+-p Si) exhibits a nanosheet-like morphology at hundred-nanometers level owing to the lack of nitrogen coordinating sites and spatial confinement. By the utilization of molecular beam epitaxy, N-terminated GaN NWs was epitaxially introduced onto silicon substrate with an average length of ∼300 nm and diameter of ca. 50 nm (Figure S3). Moreover, as shown in the top-view SEM of GaN/n+-p Si in Figure 2B, the epitaxial GaN nanowire arrays are vertically aligned on silicon with relatively uniform spatial confinement. Such spatial confinement is essential for dispersing cocatalysts onto the lateral plane of GaN at atomic level. Using these nanowires as scaffolds, the iron cocatalyst could be loaded by electrodeposition, which does not significantly alter the nanowire arrays (Figure 2C). The low-angle annular dark-field scanning transmission electron microscopy (STEM-LAADF) image shows that the lateral surface of GaN nanowire is covered by few-atomic-layers iron with an intimate core/shell structure as highlighted in Figure 2D. The core with brighter intensity is attributed to be Ga atoms, whereas the dark layer is likely to be iron atoms since the image provides Z-contrast, wherein Z is the efficient atomic number. The bright-field scanning transmission electron microscopy (STEM-BF) image (Figure 2E) and corresponding composition extracted summed spectrum further verifies that the outer layers actually consisted of iron species (Figures 2F). High-resolution STEM images acquired from different models also corroborate the atomically dispersed iron on the lateral surface of GaN NWs (Figures S4 and S5). However, on the top of GaN NWs, despite of N-rich surface, Fe cocatalysts easily formed nanoclusters, owing to the lack of spatial confinements (Figure S6). Based on these characterizations, it is reasonable to conclude that both nitrogen-rich surface and spatial confinement of GaN NWs are indispensable for the formation of iron atomic layer (Chen et al., 2018). Herein, the morphological information is acquired from the as-synthesized sample with 80 cycles of iron electrodeposition (Fe80:GaN NWs/n+-p Si). Hereafter, the as-synthesized electrode at 80 cycles of iron electrodeposition is denoted as FeFAL:GaN NWs/n+-p Si unless specially points out. XPS measurement confirmed that the iron species were successfully decorated on GaN/n+-p Si (Figure 2G). Iron was found to exist in oxidized states, which was presumably due to the unavoidable oxidation for iron in water and air, particularly for those at nanoscale (Figure S7). X-ray diffraction spectroscopy characterization reveals the appearance of the feature peak of Fe2O3 (hematite) at 2-theta of 33.2° (JCPDS Data: PDF-33-0664) after the electrodeposition, whereas the featured peaks of GaN (002) at 2-theta of 34.5° remain intact (Figure S8). The successful decoration of iron catalysts does not only offer electron sinks for effectively extracting photoinduced electrons, but also provide active sites for catalyzing hydrogen evolution reaction, which is critical for high-efficiency HER. In addition, high-resolution STEM-HADDF image in Figure 2H shows that the inter-planar lattice spacing of GaN (002) is 0.26 nm, suggesting the c-axis growth direction of the nanowire. The GaN core is nearly defect-free and could favor efficient charge carrier transport. It has been well studied by our previous work and will be further discussed in the following section. What is more, the well-defined GaN NWs could maximize the catalytic iron centers, further enhancing the performance (Deng et al., 2019).

Figure 2.

Figure 2

Open in a new tab

Structure and Composition Characterization

SEM images of Fe/n+-p Si (A), GaN NWs/n+-p Si (B), and FeFAL:GaN NWs/n+-p Si (C). STEM-LAADF, STEM-BF, and STEM-HAADF images of GaN NWs decorated with few-atomic-layers iron (D, E, and H), respectively. Extracted summed spectrum of few-atomic-layers iron decorated GaN NWs (F). XPS survey spectrum of FeFAL:GaN NWs/n+-p Si (G).

The Evaluation of PEC Water Splitting Performance

The PEC water splitting performance of FeFAL:GaN NWs/n+-p Si as well as other working electrodes was tested using a three-electrode-configuration chamber (Figure S9). The working electrode was immersed in argon-purged 0.5 M KHCO3 aqueous solution, and solar simulator was used as the light source (AM 1.5 G, 100 mW cm−2). Both Pt counter electrode and Ag/AgCl reference electrode were separated from the working electrode by Nafion membranes to exclude the possibility that Pt was redeposited onto the working electrode. As illustrated in Figure 3A, the onset potential of bare n+-p silicon is negatively as low as −0.4 V with a low current density of −6 mA cm−2 at −0.8 V. The inferior performance could be attributed to ineffective incident light collection, severe charge carriers' recombination, as well as sluggish reaction kinetics. The incorporation of GaN NWs would reduce the strong reflection of the planar silicon substrate (Boettcher et al., 2011) and facilitate the electron extraction (Zhou et al., 2019), thus improving the activity to some extent. However, owing to the lack of catalytic centers, GaN NWs/n+-p Si still suffered from limited activity. The nanosheet-like Fe/n+-p Si showed a similar J-V curve as GaN NWs/n+-p Si, indicating the limited activity of nanosheet-like Fe at hundred-nanometers level. Surprisingly, FeFAL:GaN NWs/n+-p Si at 80 cycles of iron electrodeposition shows a superior PEC behavior compared with that of both Fe/n+-p Si and GaN NWs/n+-p Si. The onset potential is +0.35 V versus RHE, with a prominent current density of −15.6 mA cm−2 at 0 V versus RHE. It is worth noting that a nearly saturated and high current density of ∼ −30 mA cm−2 is achieved at a minor overpotential of ∼0.2 V, which is approaching to the current density limit of silicon-based photocathode under standard one-sun illumination. The highest applied bias photo-to-current efficiency (ABPE) of 0.9% is acquired at an underpotential of 0.11 V with current density of −8 mA cm−2 (Figure S10). The number of the electrodeposition cycles of Fe cocatalyst affected the J-V curve significantly. The increase in electrodeposition, from 20 to 40 cycles, would lead to a gradual improvement in the PEC performance, ascribing to the increasing catalytic centers. An optimal activity is achieved at 80 cycles. However, at a higher loading, Fe200:GaN NWs/n+-p Si exhibits an evidently reduced activity with more negative onset potential and lower saturated current density (Figure 3B), in line with our expectation. It is likely due to the low catalytic activity of thick iron-base layer (Figure S11) (Fei et al., 2015). Based on these findings, it is reasonable to conclude that the high HER activity mainly arise from the few-atomic-layers iron on the lateral surface of GaN NWs, which is consistent with the results of density functional theoretical calculation described below, whereas the activity of larger-size iron is very limited. By comparing with the recent state-of-the-art HER catalysts (Table S1), we discovered that the Pt-based catalyst illustrated a higher positive onset potential of +0.4 V in contrast to that of +0.35 V for FeFAL:GaN NWs/n+-p Si with a relatively better fill factor due to the accelerated kinetics (Figure S12). Nevertheless, this result did not change the conclusion that few-atomic-layers iron could be used as a promising HER catalyst resulting from its much lower price than that of noble metals like Pt. Under dark, the current density of FeFAL:GaN NWs/n+-p Si is negligible, revealing that solar light is the energy force of the reaction. Additionally, no hydrogen was detected without external circuit regardless of illumination, indicating that this is a photoelectrocatalytic process.

Figure 3.

Figure 3

Open in a new tab

Photoelectrochemical Activity

(A) J-V curves of n+-p Si, Fe/n+-p Si, GaN NWs/n+-p Si, and FeFAL:GaN NWs/n+-p Si with 80 cycles of iron electrodeposition in argon-purged 0.5 M KHCO3 aqueous solution under standard one-sun illumination. The black line is the J-V curve of FeFAL:GaN NWs/n+-p Si under the identical conditions without illumination.

(B) Influence of the electro-depositing cycles of the iron catalyst on the performance.

(C) Current density of FeFAL:GaN NWs/n+-p Si at 0 V versus RHE.

(D) The productivity of hydrogen and oxygen and Faradaic efficiency of FeFAL:GaN NWs/n+-p Si for hydrogen evolution reaction. Experimental conditions: 0.5 M KHCO3 aqueous solution, argon atmosphere, AM 1.5 G 100 mW cm−2.

Under the optimized conditions, the productivity and Faradaic efficiency of FeFAL:GaN NWs/n+-p Si for hydrogen evolution reaction were evaluated at 0 V versus RHE. Under standard one-sun irradiation for 3,000 s, a relatively stable current density of about −16 mA cm−2 is achieved for hydrogen evolution (Figure 3C). As shown in Figures S13 and S14, STEM and XPS characterization revealed that the morphology and chemical oxidation of the as-prepared FeFAL:GaN NWs did not vary notably after the stability testing, suggesting the relative stability of the photo/electrocatalysts. Gas chromatography measurement indicated that the gas evolved from the working electrode was hydrogen from water reduction. Based on the calculation, the hydrogen evolution rate is as high as 306 μmol cm−2 h−1 based on geometric surface of the photocathode with high Faradaic efficiency of 98% (Figure 3D). At the same time, the counter electrode of Pt wire produced oxygen stoichiometrically from water oxidation. At the same time, trace amount of carbon monoxide was produced with a tiny Faradaic efficiency of <1%, which might originate from the reduction of HCO3− in the electrolyte. GC and H-NMR analyses suggested that no other carbon-based liquid and gaseous products were detected. These results provided a direct evidence that a large proportion of the photoinduced electrons were consumed for converting protons to hydrogen with high efficiency.

Density Functional Theory Calculations

To theoretically elucidate the enhanced PEC activity over FeFAL:GaN NWs/n+-p Si at the atomic scale, DFT calculations were first conducted to investigate the geometry and interaction between the atomic Fe layers and GaN (please see Computational Methods for more details). Based on the characterization results, a slab model containing three-layer (3L) atomic Fe on the N-terminated surface of wurtzite-GaN(10ī0) was established to study the surface and interface properties of Fe3L:GaN(10ī0). Top views of the optimized geometry for each layer of atomic Fe in Fe3L:GaN(10ī0) are presented in Figure 4A. It was discovered that the first layer of Fe atoms prefers to sit at the top of N atoms on GaN(10ī0) surface with a Fe−N bond length of 1.98 Å, followed by the second and third layers favoring the hollow and Ga top sites, respectively (the layer-by-layer atomic configurations and corresponding relative energy are summarized in Figure S15 and Table S2). The Ga−N dimer formed from surface reconstruction on pristine GaN(10ī0) is flattened after the deposition of Fe atoms. Notably, the atomic geometry of Fe3L:GaN(10ī0) optimized from DFT calculations is well matched with the STEM-HAADF characterization in Figure 2D, indicating the accuracy of our simulation model and method.

Figure 4.

Figure 4

Open in a new tab

Density Functional Theory Calculations

(A) Top view of atomic geometries for the adsorbed first, second, and third Fe layers in the Fe3L:GaN(10ī0) system along the c axis.

(B) Perspective views of the contour plots for the charge density redistribution near the interface region of Fe3L:GaN(10ī0) along the b (left) and a (right) axis, respectively. Light yellow and light blue regions indicate the accumulation and depletion of electron density, respectively.

(C) Calculated free energy diagram of HER for pristine GaN(10ī0) and Fe3L:GaN(10ī0) at zero applied potential.

Fe, blue; Ga, yellow; N, pink; H, white.

Besides, the electronic properties of Fe3L:GaN(10ī0) were also investigated. As shown in Figure 4B, remarkable charge density redistribution is observed near the interface, suggesting a strong electronic coupling between Fe3L and GaN(10ī0). Remarkable electron reduction (green color) occurs near the Fe atoms, whereas electron accumulation (yellow color) is observed around the neighboring N atoms, which is in good consistency with the XPS observation of a minor positive shift for FeFAL:GaN NWs/n+-p Si compared with Fe/n+-p Si (Figure S7). On the other hand, notable electron accumulation emerges around the middle region of the Fe and Ga atoms, which indicates the formation of covalent-like Fe−Ga bonds. These theoretical results regarding electronic properties reveal the formation of an efficient electron-transition channel that is highly favorable for the separation/migration of electron-hole pairs and for reducing the voltage loss during the reaction, which is in good agreement with the electrochemical measurement in the following.

It is noted that, in this study, FeFAL:GaN NWs/n+-p Si was applied as a photocathode for water reduction toward hydrogen, whereas platinum wire was used as an anode for water oxidation toward oxygen. Therefore, the theoretical calculations were focused on the free energy of H adsorption on the catalysts surface of FeFAL:GaN NWs/n+-p Si, i.e., G∗H, which is considered to be a key metric in quantitatively assessing the HER catalytic activity. Based on the computational hydrogen electrode (CHE) model suggested by Nørskov et al. (2004), a thermo-neutral value of ΔG∗H (i.e., 0 eV) at which the H adsorption is neither too strong or too weak on a catalyst surface is highly preferred for an ideal HER process. Employing DFT calculations, the free energy diagram of HER for both GaN(10ī0) and Fe3L:GaN(10ī0) were mapped out, as illustrated in Figure 4C. As seen from the figure, ΔG∗H shows a value of −0.32 eV on the pristine GaN(10ī0) surface, indicating a rather strong binding of hydrogen with the N atoms on the surface. Such strong binding suggests that HER would be unfavorable to occur on a pristine GaN. On the other hand, the binding strength of H on the surface of the Fe3L:GaN(10ī0) is significantly weakened, resulting in an optimal ΔG∗H value of −0.13 eV, comparable with that of a state-of-the-art HER catalyst of platinum (Nørskov et al., 2005). Additionally, to consider the possible oxidation that occurred on the surfaces, we conducted further theoretical calculation to study the effect of partially oxidized surface of FeFAL:GaN NWs/n+-p Si on the H adsorption. We found that the ΔG∗H becomes even more thermo-neutral when the surface iron atoms are oxidized at different ratios (O coverages ranging from 8% to 33%), which indicates the formation of iron oxide would further enhance the superior performance of FeFAL:GaN NWs/n+-p Si (Figures S16 and S17). Overall, the superior HER performance of FeFAL:GaN NWs/n+-p Si is attributed to the synergetic effect between the strong electronic coupling at the interface region and an optimal H adsorption strength.

To gain more insights into the superior performance, we carried out a series of optoelectronic measurements. Firstly, ultraviolet-visible (UV-vis) diffuse reflectance spectroscopy in Figure 5A illustrates that compared with planar silicon, GaN NWs/n+-p Si demonstrates a remarkably improved light absorption due to the light-trapping effect. The integration of iron catalyst with GaN NWs further enhances the optical collection efficiency, rendering an excellent optical property for the reaction. Furthermore, the electronic property of FeFAL:GaN NWs/n+-p Si photocathode was studied by electrochemical impedance spectroscopy. The fitting equivalent circuit model of FeFAL:GaN NWs/n+-p Si is shown in Figure S18. Electrochemical impedance spectroscopy in Figure 5B shows that the radius of FeFAL:Ga NWs/n+-p Si is much smaller than that of Fe/n+-p Si, suggesting that the electron resistance of FeFAL:GaN NWs/n+-p Si is much lower than that of the nanosheet-like Fe/n+-p Si (Ding et al., 2014). It reveals that the epitaxial GaN NWs can serve as an ideal electron-transition channel for charge carrier separation, which is verified by room temperature photoluminescence (PL) spectroscopy (Figure S19). Compared with GaN NWs/n+-p Si, a dramatic reduction in PL intensity of FeFAL:GaN NWs/n+-p Si suggests that the few-atomic-layers Fe reduces the radiative recombination of electrons and holes. It is well consistent with our previous observations (Zhou et al., 2020). The energy bandgap diagram of FeFAL:GaN NWs/n+-p Si is shown in Figure 5C. Herein, the conduction band alignment between GaN and Si is almost negligible, as confirmed by our recent study (Vanka et al., 2018). Under illumination, the upward bending of surface is reduced by the accumulated photogenerated electrons. Hence, the photoinduced electrons can be easily extracted from n+-Si to nearly dislocation-free n-GaN grown by highly controlled molecular beam epitaxy technology, which further migrate to atomic Fe layers with greatly reduced voltage loss. What is more, in such a unique nanoarchitecture, GaN nanowire is capable of maximizing catalytic centers. Together with the superior ΔG∗H, FeFAL:GaN NWs/n+-p Si is thus highly active for hydrogen production. Meanwhile, the holes migrate to the counter electrode via external circuit for oxygen evolution from water oxidation (Figure 5D).

Figure 5.

Figure 5

Open in a new tab

Proposed Mechanism

(A) UV-vis diffuse reflectance spectroscopy of n+-p Si, Fe/n+-p Si, GaN NWs/n+-p Si, and FeFAL:GaN NWs/n+-p Si.

(B) Electrochemical impedance spectroscopy of Fe/n+-p Si and FeFAL:GaN NWs/n+-p Si.

(C) The energy diagram of FeFAL:GaN NWs/n+-p Si. Blue spot: electron, red spot: hole.

(D) Schematic illustration of FeFAL:GaN NWs/n+-p Si for PEC water splitting toward hydrogen and oxygen under standard one-sun illumination.

Conclusion

In summary, FeFAL was anchored on GaN NWs/Si by a facile electrodeposition approach. Density functional theory calculations reveal an impressive hydrogen adsorption free energy of −0.13 eV on FeFAL as an indicator of high HER activity. Moreover, the strong electronic interaction between defect-free GaN and FeFAL provides an efficient electron-transition channel for charge carriers separation. Consequently, FeFAL:GaN NWs/n+-P Si demonstrates a prominent current density of −15.6 mA cm−2 at 0 V with a decent onset potential of +0.35 V under standard one-sun illumination. The hydrogen evolution rate is as high as 306 μmol cm−2 h−1 with ∼98% Faradaic efficiency. The device, consisting of the two most produced semiconductors (Si and GaN) and earth-abundant material of iron as cocatalyst, is manufactured by mature epitaxial technology and electrodeposition. As such, this study presents a viable strategy for achieving economic, large-scale, and carbon-free hydrogen production from PEC water splitting using solar energy.

Limitations of the Study

This study has demonstrated that few-atomic-layers iron could be deposited on the N-rich m-plane surface of GaN NWs epitaxially grown on silicon substrate as an efficient photocathode for PEC water splitting toward hydrogen. However, the overall performance of the as-synthesized electrode is still relatively low compared with that of the state-of-the-art devices (Table S1). What is more, the morphology of the iron-based catalyst on the top surface of GaN NWs is difficult to be engineered to form atomic layers owing to the lack of spatial confinement, which is the most likely reason for the limited performance of the device. Further highly controlled synthesis of atomic-scale iron catalyst covered the entire GaN NWs will be conducted to improve the overall performance.

Resource Availability

Lead Contact

Further requests for resources and materials should be directed to and will be fulfilled by the Lead Contact, Zetian Mi (ztmi@umich.edu).

Materials Availability

This study did not yield new unique reagents.

Data and Code Availability

This study did not produce datasets/code.

Methods

All methods can be found in the accompanying Transparent Methods supplemental file.

Acknowledgments

B.Z. and Z.M. are thankful for the financial support from Emissions Reduction Alberta and from The University of Michigan College of Engineering Blue Sky Research Program. P.O. and J.S. acknowledge the financial support of the Natural Science and Engineering Research Council of Canada (NSERC) Discovery grant (grant # RGPIN-2017-05187) and McGill Engineering Doctoral Award (MEDA) and computational resources provided by Compute Canada. We also thank the technical support from the Michigan Center for Materials Characterization.

Author Contributions

B.Z. and Z.M. conceived and designed the project. B.Z. and R.T.R. fabricated and characterized the samples and conducted the PEC testing. K.S. conducted the STEM characterizations. P.O. and J.S. performed DFT calculations and analyzed the calculation results. R.T.R. and S.V. carried out the nanowire arrays growth. L.Y. and H.S. contributed to result analysis and discussions. The manuscript was written by B.Z., P.O., J.S. and Z.M. with contributions from other co-authors.

Declaration of Interests

Some IP related to GaN nanowire synthesis was licensed to NS Nanotech, Inc., which was co-founded by Z.M.

Published: October 23, 2020

Footnotes

Supplemental Information can be found online at https://doi.org/10.1016/j.isci.2020.101613.

Contributor Information

Jun Song, Email: jun.song2@mcgill.ca.

Zetian Mi, Email: ztmi@umich.edu.

Supplemental Information

Document S1. Transparent Methods and Figures S1–S19, and Tables S1 and S2
mmc1.pdf (2MB, pdf)

References

  1. Arif M., Yasin G., Shakeel M., Fang X.Y., Gao R., Ji S.F., Yan D.P. Coupling of bifunctional CoMn-layered double hydroxide@graphitic C3N4 nanohybrids towards efficient photoelectrochemical overall water splitting. Chem. Asian J. 2018;13:1045–1052. doi: 10.1002/asia.201800016. [DOI] [PubMed] [Google Scholar]
  2. Boettcher S.W., Warren E.L., Putnam M.C., Santori E.A., Turner-Evans D., Kelzenberg M.D., Walter M.G., McKone J.R., Brunschwig B.S., Atwater H.A., Lewis N.S. Photoelectrochemical hydrogen evolution using Si microwire arrays. J. Am. Chem. Soc. 2011;133:1216–1219. doi: 10.1021/ja108801m. [DOI] [PubMed] [Google Scholar]
  3. Camara J.M., Rauchfuss T.B. Combining acid-base, redox and substrate binding functionalities to give a complete model for the [FeFe]-hydrogenase. Nat. Chem. 2012;4:26–30. doi: 10.1038/nchem.1180. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cao Y.J., Chen S., Luo Q.Q., Yan H., Lin Y., Liu W., Cao L.L., Lu J.L., Yang J.L., Yao T., Wei S.Q. Atomic-level insight into optimizing the hydrogen evolution pathway over a Co1-N4 single-site photocatalyst. Angew. Chem. Int. Ed. 2017;56:12191–12196. doi: 10.1002/anie.201706467. [DOI] [PubMed] [Google Scholar]
  5. Chen Y.J., Ji S.F., Chen C., Peng Q., Wang D.S., Li Y.D. Single-atom catalysts: synthetic strategies and electrochemical applications. Joule. 2018;2:1242–1264. [Google Scholar]
  6. Cheng N.C., Stambula S., Wang D., Bains M.N., Liu J., Riese A., Xiao B.W., Li R.Y., Sham T.K., Liu L.M. Platinum single-atom and cluster catalysis of the hydrogen evolution reaction. Nat. Commun. 2016;7:13638. doi: 10.1038/ncomms13638. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Deng J., Su Y.D., Liu D., Yang P.D., Liu B., Liu C. Nanowire photoelectrochemistry. Chem. Rev. 2019;119:9221–9259. doi: 10.1021/acs.chemrev.9b00232. [DOI] [PubMed] [Google Scholar]
  8. Ding Q., Meng F., English C.R., Caban-Acevedo M., Shearer M.J., Liang D., Daniel A.S., Hamer R.J., Jin S. Efficient photoelectrochemical hydrogen generation using heterostructures of Si and chemically exfoliated metallic MoS2. J. Am. Chem. Soc. 2014;136:8504–8507. doi: 10.1021/ja5025673. [DOI] [PubMed] [Google Scholar]
  9. Fan L.L., Liu P.F., Yan X.C., Gu L., Yang Z.Z., Yang H.G., Qiu S.L., Yao X.D. Atomically isolated nickel species anchored on graphitized carbon for efficient hydrogen evolution electrocatalysis. Nat. Commun. 2016;7:10667. doi: 10.1038/ncomms10667. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Fei H.L., Dong J.C., Arellano-Jimenez M.J., Ye G.L., Kim N.D., Samuel E.L.G., Peng Z.W., Zhu Z., Qin F., Bao J.M. Atomic cobalt on nitrogen-doped graphene for hydrogen generation. Nat. Commun. 2015;6:8668. doi: 10.1038/ncomms9668. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Feng J.X., Wu J.Q., Tong Y.X., Li G.R. Efficient hydrogen evolution on Cu nanodots-decorated Ni3S2 nanotubes by optimizing atomic hydrogen adsorption and desorption. J. Am. Chem. Soc. 2018;140:610–617. doi: 10.1021/jacs.7b08521. [DOI] [PubMed] [Google Scholar]
  12. Gao R., Zhang H., Yan D.P. Iron diselenide nanoplatelets: stable and efficient water-electrolysis catalysts. Nano Energy. 2017;31:90–95. [Google Scholar]
  13. Le Goff A., Artero V., Jousselme B., Dinh Tran P., Guillet N., Metaye R., Fihri A., Palacin S., Fontecave M. From hydrogenases to noble metal-free catalytic nanomaterials for H2 production and uptake. Science. 2009;326:1384–1387. doi: 10.1126/science.1179773. [DOI] [PubMed] [Google Scholar]
  14. Hashemi S.M.H., Modestino M.A., Psaltis D. A membrane-less electrolyzer for hydrogen production across the pH scale. Energ. Environ. Sci. 2015;8:2003–2009. [Google Scholar]
  15. Hui L., Xue Y.R., Huang B.L., Yu H.D., Zhang C., Zhang D.Y., Jia D.Z., Zhao Y.J., Li Y.J., Liu H.B., Li Y.L. Overall water splitting by graphdiyne-exfoliated and -sandwiched layered double-hydroxide nanosheet arrays. Nat. Commun. 2018;9:5309. doi: 10.1038/s41467-018-07790-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hui L., Xue Y.R., Yu H.D., Liu Y.X., Fang Y., Xing C.Y., Huang B.L., Li Y.L. Highly efficient and selective generation of ammonia and hydrogen on a graphdiyne-based catalyst. J. Am. Chem. Soc. 2019;141:10677–10683. doi: 10.1021/jacs.9b03004. [DOI] [PubMed] [Google Scholar]
  17. Ji L., McDaniel M.D., Wang S.J., Posadas A.B., Li X.H., Huang H.Y., Lee J.C., Demkov A.A., Bard A.J., Ekerdt J.G., Yu E.T. A silicon-based photocathode for water reduction with an epitaxial SrTiO3 protection layer and a nanostructured catalyst. Nat. Nanotechnol. 2015;10:84–90. doi: 10.1038/nnano.2014.277. [DOI] [PubMed] [Google Scholar]
  18. Jin J., Walczak K., Singh M.R., Karp C., Lewis N.S., Xiang C.X. An experimental and modeling/simulation-based evaluation of the efficiency and operational performance characteristics of an integrated, membrane-free, neutral pH solar-driven water-splitting system. Energ. Environ. Sci. 2014;7:3371–3380. [Google Scholar]
  19. Kibria M.G., Qiao R.M., Yang W.L., Boukahil I., Kong X.H., Chowdhury F.A., Trudeau M.L., Ji W., Guo H., Himpsel F.J. Atomic-scale origin of long-term stability and high performance of p-GaN nanowire arrays for photocatalytic overall pure water splitting. Adv. Mater. 2016;28:8388–8397. doi: 10.1002/adma.201602274. [DOI] [PubMed] [Google Scholar]
  20. Kluwer A.M., Kapre R., Hartl F., Lutz M., Spek A.L., Brouwer A.M., van Leeuwen P.W.N.M., Reek J.N.H. Self-assembled biomimetic [2Fe2S]-hydrogenase-based photocatalyst for molecular hydrogen evolution. Proc. Natl. Acad. Sci. U S A. 2009;106:10460–10465. doi: 10.1073/pnas.0809666106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li C.B., Li Z.J., Yu S., Wang G.X., Wang F., Meng Q.Y., Chen B., Feng K., Tung C.H., Wu L.Z. Interface-directed assembly of a simple precursor of [FeFe]-H2ase mimics on CdSe QDs for photosynthetic hydrogen evolution in water. Energ. Environ. Sci. 2013;6:2597–2602. [Google Scholar]
  22. Li Q.H., Chen W.X., Xiao H., Gong Y., Li Z., Zheng L.R., Zheng X.S., Yan W.S., Cheong W.C., Shen R.A. Fe isolated single atoms on S, N codoped carbon by copolymer pyrolysis strategy for highly efficient oxygen reduction reaction. Adv. Mater. 2018;30:1800588. doi: 10.1002/adma.201800588. [DOI] [PubMed] [Google Scholar]
  23. Meng C.H., Liu Z.Y., Zhang T.R., Zhai J. Layered MoS2 nanoparticles on TiO2 nanotubes by a photocatalytic strategy for use as high-performance electrocatalysts in hydrogen evolution reactions. Green Chem. 2015;17:2764–2768. [Google Scholar]
  24. Nørskov J.K., Rossmeisl J., Logadottir A., Lindqvist L., Kitchin J.R., Bligaard T., Jonsson H. Origin of the overpotential for oxygen reduction at a fuel-cell cathode. J. Phys. Chem. B. 2004;108(46):17886–17892. [Google Scholar]
  25. Nørskov J.K., Bligaard T., Logadottir A., Kitchin J.R., Chen J.G., Pandelov S., Stimming U. Trends in the exchange current for hydrogen evolution. J. Electrochem. Soc. 2005;152:J23–J26. [Google Scholar]
  26. Pullen S., Fei H.H., Orthaber A., Cohen S.M., Ott S. Enhanced photochemical hydrogen production by a molecular diiron catalyst incorporated into a metal-organic framework. J. Am. Chem. Soc. 2013;135:16997–17003. doi: 10.1021/ja407176p. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Shi L., Li P., Zhou W., Wang T., Chang K., Zhang H.B., Kako T., Liu G.G., Ye J.H. N-type boron phosphide as a highly stable, metal-free, visible-light-active photocatalyst for hydrogen evolution. Nano Energy. 2016;28:158–163. [Google Scholar]
  28. Tard C., Pickett C.J. Structural and functional analogues of the active sites of the [Fe]-, [NiFe]-, and [FeFe]-hydrogenases. Chem. Rev. 2009;109:2245–2274. doi: 10.1021/cr800542q. [DOI] [PubMed] [Google Scholar]
  29. Vanka S., Arca E., Cheng S.B., Sun K., Botton G.A., Teeter G., Mi Z.T. High efficiency Si photocathode protected by multifunctional GaN nanostructures. Nano Lett. 2018;18:6530–6537. doi: 10.1021/acs.nanolett.8b03087. [DOI] [PubMed] [Google Scholar]
  30. Vijselaar W., Westerik P., Veerbeek J., Tiggelaar R.M., Berenschot E., Tas N.R., Gardeniers H., Huskens J. Spatial decoupling of light absorption and catalytic activity of Ni-Mo-loaded high-aspect-ratio silicon microwire photocathodes. Nat. Energy. 2018;3:185–192. [Google Scholar]
  31. Wang F., Liang W.J., Jian J.X., Li C.B., Chen B., Tung C.H., Wu L.Z. Exceptional poly(acrylic acid)-based artificial [FeFe]-hydrogenases for photocatalytic H2 production in water. Angew. Chem. Int. Ed. 2013;52:8134–8138. doi: 10.1002/anie.201303110. [DOI] [PubMed] [Google Scholar]
  32. Wang A.Q., Li J., Zhang T. Heterogeneous single-atom catalysis. Nat. Rev. Chem. 2018;2:65–81. [Google Scholar]
  33. Wexler R.B., Martirez J.M.P., Rappe A.M. Chemical pressure-driven enhancement of the hydrogen evolving activity of Ni2P from nonmetal surface doping interpreted via machine learning. J. Am. Chem. Soc. 2018;140:4678–4683. doi: 10.1021/jacs.8b00947. [DOI] [PubMed] [Google Scholar]
  34. Wodrich M.D., Hu X.L. Natural inspirations for metal-ligand cooperative catalysis. Nat. Rev. Chem. 2018;2:0099. [Google Scholar]
  35. Xue Y.R., Huang B.L., Yi Y.P., Guo Y., Zuo Z.C., Li Y.J., Jia Z.Y., Liu H.B., Li Y.L. Anchoring zero valence single atoms of nickel and iron on graphdiyne for hydrogen evolution. Nat. Commun. 2018;9:1460. doi: 10.1038/s41467-018-03896-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Ye W., Fang X.Y., Chen X.B., Yan D.P. A three-dimensional nickel-chromium layered double hydroxide micro/nanosheet array as an efficient and stable bifunctional electrocatalyst for overall water splitting. Nanoscale. 2018;10:19481–19491. doi: 10.1039/c8nr05974h. [DOI] [PubMed] [Google Scholar]
  37. Yu H.D., Xue Y.R., Huang B.L., Hui L., Zhang C., Fang Y., Liu Y.X., Zhao Y.J., Li Y.J., Liu H.B., Li Y.L. Ultrathin nanosheet of graphdiyne-supported palladium atom catalyst for efficient hydrogen production. iScience. 2019;11:31–41. doi: 10.1016/j.isci.2018.12.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Zhang H.X., Ding Q., He D.H., Liu H., Liu W., Li Z.J., Yang B., Zhang X.W., Lei L.C., Jin S. A p-Si/NiCoSex core/shell nanopillar array photocathode for enhanced photoelectrochemical hydrogen production. Energ. Environ. Sci. 2016;9:3113–3119. [Google Scholar]
  39. Zhang M.L., Wang Y.G., Chen W.X., Dong J.C., Zheng L.R., Luo J., Wan J.W., Tian S.B., Cheong W.C., Wang D.S., Li Y.D. Metal (Hydro)oxides@polymer core-shell strategy to metal single-atom materials. J. Am. Chem. Soc. 2017;139:10976–10979. doi: 10.1021/jacs.7b05372. [DOI] [PubMed] [Google Scholar]
  40. Zhang H.B., An P.F., Zhou W., Guan B.Y., Zhang P., Dong J.C., Lou X.W. Dynamic traction of lattice-confined platinum atoms into mesoporous carbon matrix for hydrogen evolution reaction. Sci. Adv. 2018;4:eaao6657. doi: 10.1126/sciadv.aao6657. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Zhao Z.P., Liu H.T., Gao W.P., Xue W., Liu Z.Y., Huang J., Pan X.Q., Huang Y. Surface-engineering PtNi-O nanostructured with record-high performance for electrocatalytic hydrogen evolution reaction. J. Am. Chem. Soc. 2018;140:9046–9050. doi: 10.1021/jacs.8b04770. [DOI] [PubMed] [Google Scholar]
  42. Zhou B.W., Kong X.H., Vanka S., Chu S., Ghamari P., Wang Y.C., Pant N., Shih I., Guo H., Mi Z.T. Gallium nitride nanowire as a linker of molybdenum sulfides and silicon for photoelectrocatalytic water splitting. Nat. Commun. 2018;9:3856. doi: 10.1038/s41467-018-06140-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhou B.W., Kong X.H., Vanka S., Cheng S.B., Pant N., Chu S., Ghamari P., Wang Y.C., Botton G.A., Guo H., Mi Z.T. A GaN:Sn nanoarchitecture integrated on a silicon platform for converting CO2 to HCOOH by photoelectrocatalysis. Energ. Environ. Sci. 2019;12:2842–2848. [Google Scholar]
  44. Zhou B.W., Ou P.F., Pant N., Cheng S.B., Vanka S., Chu S., Rashid R.T., Botton G., Song J., Mi Z.T. Highly efficient binary copper-iron catalyst for photoelectrochemical carbon dioxide reduction toward methane. Proc. Natl. Acad. Sci. U S A. 2020;117:1330–1338. doi: 10.1073/pnas.1911159117. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Transparent Methods and Figures S1–S19, and Tables S1 and S2
mmc1.pdf (2MB, pdf)

Data Availability Statement

This study did not produce datasets/code.

Articles from iScience are provided here courtesy of Elsevier

RESOURCES