A highly [001]-textured Sb2Se3 photocathode for efficient photoelectrochemical water reduction

Hongpeng Zhou a, Menglei Feng a, Kena Song a, Bin Liao b, Yichang Wang b, Ruchuan Liu a, Xiangnan Gong c, Dingke Zhang *d, Lingfei Cao *b and Shijian Chen *a
aChongqing Key Laboratory of Soft Condensed Matter Physics and Smart Materials, College of Physics, Chongqing University, Chongqing, 401331, China. E-mail: sjchen@cqu.edu.cn
bInternational Joint Laboratory for Light Alloys (Ministry of Education), College of Materials Science and Engineering, Chongqing University, Chongqing 400044, China. E-mail: caolingfei@cqu.edu.cn
cAnalytical and Testing Center of Chongqing University, Chongqing 400044, China
dCollege of Physics and Electronic Engineering, Chongqing Normal University, Chongqing, 401331, China. E-mail: zhangdk@cqnu.edu.cn

Received 10th October 2019 , Accepted 6th November 2019

First published on 6th November 2019


Abstract

Anisotropic Sb2Se3 is an emerging earth-abundant photocathode for photoelectrochemical water splitting. However, controlling the growth of the Sb2Se3 film with optimal [001] crystallographic orientation is still the most challenging issue. Here, we successfully synthesized [001]-oriented Sb2Se3via a reliable and facile method. The [001]-oriented Sb2Se3 film could provide an excellent carrier-migration efficiency. Consequently, we achieved a record-high photocurrent density of −20.2 mA cm−2 at 0 VRHE and a very high half-cell solar-to-hydrogen efficiency of 1.36% under 1-sun simulated solar illumination in a TiO2/[001]-Sb2Se3 photocathode. This work provides an effective strategy and important guidelines for rationally designing optoelectronic devices based on the [001]-oriented Sb2Se3 film.


Introduction

One of the urgent scientific and technological challenges of humanity is to explore the secure supply of clean and renewable energy forms.1–4 In the past decades, tremendous research efforts have been focused on designing a high-efficiency artificial photosynthesis system to convert inexhaustible and highly intermittent sunlight energy into energy-dense chemical bonds of storable and transportable fuel molecules.5–8 Photoelectrochemical (PEC) water splitting cell is an attractive solar-to-chemical energy conversion approach to generate storable and transportable clean hydrogen fuel from water via a solar-driven semiconductor-liquid junction (SCLJ) photoelectrode.4,9,10 In the last decade, the STH conversion efficiency milestone had been achieved by the wafer-based III–V semiconductor material system owing to its highly compatible optoelectronic properties, mature crystalline manufacturing system and the defect passivation strategy.5,6,8 However, epitaxial wafer-scale film growth involves a high cost, high material complexity and a relatively low yield.11–13 Another limitation of the conventional wafer-based technology is that it only can fabricate a device with a small cell area. Thus, the realization of a new low-cost and high performance material is indispensable to pave the way for large-scale practical applications.13,14

Antimony selenide (Sb2Se3) is an emerging earth-abundant semiconductor that is widely used as a light absorbing material.14,15 It is highly attractive for the PEC photoelectrode due to its narrow bandgap covering most of the solar spectrum and the appropriate band levels with respect to the water reduction redox potential; it also possesses a large absorption coefficient (>105 cm−1) and a moderate minority carrier mobility (∼10 cm2 V−1 s−1), and it is chemically stable in harsh environments, such as strong acid electrolytes owing to its intrinsic photocorrosion-resistant properties.16–24 The unique two-dimensional layered orthorhombic Sb2Se3 consists of parallel one-dimensional (1D) polymerized (Sb4Se6)n ribbon chain units stacked along the [001]-direction (or ribbon-direction) via Sb–Se covalent bonds, which are linked by van der Waals (vdWs) bonds along the [100]-direction and [010]-direction, as shown in Fig. S1.[thin space (1/6-em)]25,26 The interstitial spaces of parallelly-stacked ribbons bonded via vdWs interactions are free of unsaturated chemical bonds and can effectively minimize carrier recombination. This self-passivation feature shows the possibility of fabricating a Sb2Se3 film with a high electronic quality and a low detrimental defect density. Moreover, as illustrated in Fig. 1a, the anisotropic structure renders its high direction-dependent carrier transport efficiency.27–32 The previously reported hole mobilities are 1.17, 0.69 and 2.59 cm2 V−1 s−1 along the [100], [010] and [001] directions, respectively.33 The [001]-direction exhibits the highest mobility due to the strong covalent bond interactions along the (Sb4Se6)n ribbon direction compared to vdW bonded [100]/[010] ([hk0]) and all other's crystallographic direction ([hk1]). Therefore, if we can grow highly [001]-oriented Sb2Se3 films, the maximum transport efficiency can be achieved due to its high carrier transport efficiency along the ribbon orientation and the low recombination loss between the vdWs gap during carrier extraction. Indeed, a few trials have shown that a slightly-tilted ribbon-direction (such as [221] or [211] orientations, Fig. 1a) exhibited a much better transport efficiency over horizontally oriented ribbons (such as [120] orientation).14,15,34,35


image file: c9nr08700a-f1.tif
Fig. 1 (a) Schematic illustration for the relationship between photogenerated charge carriers’ transport and growth orientations of Sb2Se3 films. θ was defined as the angle between the ribbon-direction (or [001]-direction) and the substrate plane. (b) Typical XRD patterns of Sb2Se3 films. (c) The texture coefficients of 10 selected dominant diffraction peaks of Sb2Se3 films. The (hkl) planes are arranged according to the angle between the [001]-direction (ribbon-direction) and the substrate plane. (d) XRD pole figure of (002) plane preferred orientation, radial lines (purple) in the pole figure represent the tilt angle (χ) with an increment of 30°.

In this work, highly [001]-oriented Sb2Se3 films were prepared by selenizing [003]-oriented Sb films in vacuum sealed quartz tubes. We demonstrated that Sb2Se3 with preferred [001]-orientation provides an efficient charge carrier transport channel. Thus, our Sb2Se3 photoelectrodes exhibit a remarkable PEC performance, reaching a record high saturation photocurrent density of −25.4 mA cm−2 at −0.2 V versus the reversible hydrogen electrode (VRHE) under 1-sun simulated air mass 1.5 global (AM 1.5 G) illumination. Our experimental results provide a promising way for the development of a highly efficient [001]-oriented Sb2Se3 photocathode.

Results and discussion

In this work, we prepared [001]-oriented Sb2Se3 thin films on Mo substrates as an essential step for the fabrication of a highly efficient photocathode for photoelectrochemical water reduction. Fig. S2a schematically illustrates our reliable and facile synthesis procedures. Sb2Se3 with [001] preferred orientation was prepared by selenizing Sb films with [003] preferential orientation, which were sputtered on a Mo [110] substrate (Fig. S2b). Our optimized reaction was conducted in a small-sized (about 11 cm3) and high vacuum-sealed (1.2 × 10−4 Pa) quartz tube at 325 °C for a very long duration (8 hours). The XRD pattern (Fig. 1b) of the 750 nm Sb2Se3 can be well indexed to the orthorhombic phase Sb2Se3 (JCPDS 15-0861). The absence of any impurity phases, such as Sb and Se, indicates the high purity of the film. Notably, the obtained Sb2Se3 film exhibits a very high ratio of (002) peak and (221) peak compared to the non-preferentially oriented Sb2Se3 powder reference (JCPDS 15-0861) and common [221]-oriented Sb2Se3 (Fig. S3). This result indicates that the 750 nm Sb2Se3 film shows a highly preferred [001] orientation.

Here, we further calculated the crystallographic texture coefficient (TC, see Method section) for selected diffraction peaks to quantify the degree of preferred oriented growth.36 A TC value of greater than 1 for the (hkl) plane represents that the film has [hkl] preferential orientation. Fig. 1c plots the TC distributions of 10 selected dominant XRD peaks for the Sb2Se3 films arranged according to the angle (θ) between the [001]-direction (ribbon-direction) and the substrate plane.35 It is clearly shown that the TC values for all the selected (hkl) planes monotonously increase from 0.3 to 4.1 with the increase of θ from 0° to 90°. The (002) plane has a maximum TC value, which is much larger than the TC values of both (hk0) and (hk1) planes. These results further demonstrate that the grains of the Sb2Se3 film are mainly [001] preferred orientation growth. To further verify the nature of preferred [001] orientation,37,38 XRD pole figures were measured (Fig. S4) at (120), (211), (221) and (002) reflection planes of the film to examine the global orientation distribution. The restructured complete XRD pole figure of the (002) facet from experimental data (Fig. 1d) shows a sharp single-pole with a very narrow tilt angle of 10°. On the contrary, pole figure of the (120) reflection plane exhibits a non-preferred orientation as evidenced by the random intensity distribution and the pole figures of (211) and (221) reflection planes show a lower degree of preferred orientation than that of the (002) reflection plane (Fig. S4). These results provide strong proof for the highly preferred [001] orientation of our Sb2Se3 film. Furthermore, we found that a Sb2Se3 film with a larger thickness (such as 3000 nm, blue line in Fig. 1b) prepared at identical conditions exhibits an even stronger degree of preferred [001] orientation, indicating that our strategy is effective for the fabrication of the preferred [001] orientation. Moreover, our strategy can easily fabricate Sb2Se3 with a highly [001]-preferred orientation over a broad temperature range (275–350 °C) and ratios of [Se]/[Sb] (1.5–2.9) (Fig. S5, using 750 nm thickness), evidently indicating the reliability of our synthesis strategy. Since a very thick film is unfavorable for the charge carrier extraction, an optimized thickness of 750 nm that can balance the light absorption and charge carrier transport was chosen for further studies.

The preferential growth of Sb2Se3 generally results from the inherent anisotropic nature of the V2VI3 type stibnite-type structure,17,39–41 which thermodynamically facilitates the [001]-oriented crystal growth. Moreover, kinetic conditions, such as the type of substrate also play an important role in the film growth process. Mai et al. have demonstrated that a Mo substrate with preferred [110] orientation can facilitate the [001]-oriented growth of a Sb2Se3 film.15 Their DFT calculation revealed that a Mo substrate with preferred [110]-orientation can facilitate and stabilize the [001]-oriented growth of Sb2Se3 grains. In our experiments, preferred [003]-oriented Sb film was first grown on a Mo (110) substrate (Fig. S3) and our obtained Sb2Se3 film had a much higher degree of preferred [001]-oriented growth. Thus, the [003]-oriented Sb film may play a key role in preferred [001]-orientation of Sb2Se3. We noticed that the Sb (003) surface has a hexagonal cellular structure with a size of 5.8 Å in the crystal structure (the inset of Fig. S2b). The big cellular structure enables Se atoms to diffuse through it to react with Sb atoms at an elevated high temperature. It is known that the Se atmosphere in the sealed small-sized quartz tube (about 11 cm3) always retains its saturated vapour pressure (excessive Se, [Se]/[Sb] = 2.2) at a given temperature throughout the entire reaction while the low volatile Sb film is maintained as a solid (Fig. S6). The Sb film is converted into a Sb2Se3 film via the reaction with condensed Se atoms that continuously diffuse toward the Mo substrate. We speculate the possible growth process of ribbon-orientated Sb2Se3 films as: Se atoms were initially adsorbed on the solid Sb surface at an elevated temperature and reacted with Sb atoms to gradually form a layer of stable Sb2Se3 along ribbon orientation at the Sb(003)/Se interface due to the higher kinetic rate stemming from the anisotropic nature. Then, Se atoms continuously diffused to the Sb(003)/Sb2Se3(002) interface through the vdWs gap between ribbons and eventually lengthened the (Sb4Se6)n ribbon. We naturally infer that the [003]-oriented Sb film may facilitate the initial preferred [001]-oriented growth of Sb2Se3 at the initial selenization stage on the surface of the Sb film, which may be the origin of our Sb2Se3 on an identical Mo substrate, possessing a higher degree of preferred [001]-oriented growth. Then, the Mo substrate with [110]-orientation can possibly stabilize the [001]-oriented Sb2Se3 when the Sb2Se3 film with preferred [001] orientation finally extends to the Mo interface. Moreover, our reaction was conducted in a high vacuum (1.2 × 10−4 Pa) at 325 °C for a very long duration (8 hours). Such a high vacuum can avoid detrimental impurity defects that may affect the nucleation and growth, thus benefiting the thermodynamically favorable [001]-oriented growth. Long sintering duration, similar to conventional high temperature solid-state reaction method, also improves the quality of crystallinity along the [001] orientation.

The microstructure of the obtained Sb2Se3 film was observed via scanning electron microscopy (SEM). The obtained Sb2Se3 film shows compact grains and a pinhole-free morphology (Fig. 2a). Fig. 2b shows the corresponding cross-sectional SEM image of the Sb2Se3/Mo film. The well-oriented columnar Sb2Se3 grains extending from the surface to the back contact can potentially reduce the number of grain boundaries (GBs) and the recombination at the GBs. The cross-sectional transmission electron microscopy (TEM) image of Sb2Se3 (Fig. 2c) shows a very low defect density, such as dislocations. High resolution TEM (Fig. 2d) image verifies a layer of Mo–Se intermediate phase formed at the Sb2Se3/Mo interface,42 which was also confirmed by a cross-sectional Raman spectrum (Fig. S7). This Mo–Se intermediate layer may improve hole carriers’ extraction at the back contact electrode.43 The EDS elemental mappings (Fig. 2e–h) indicate the homogeneous distribution of Sb and Se across the entire view filed. All the results suggest that our Sb2Se3 film possesses a high crystal quality.


image file: c9nr08700a-f2.tif
Fig. 2 Morphology and phase characteristics (a and b) surface morphology and cross-sectional SEM images of Sb2Se3 photoelectrode. Fake-color was used to aid visualization of the layers. (c and d) High-resolution TEM images of a Sb2Se3/Mo interface. (e–h) STEM-HAADF image of the Sb2Se3 film and corresponding elemental distribution mapping images of Sb and Se.

Optical properties of the Sb2Se3 film was characterized by a transmission spectrum. The bandgap of Sb2Se3 was determined to be 1.2 eV (Fig. S8), which covers a large portion of the solar spectrum between ultraviolet and near-infrared regions. The theoretical maximum photocurrent density of the Sb2Se3 film with a sufficient thickness was calculated to be 40.9 mA cm−2 under AM1.5G solar light illumination for 1.2 eV bandgap (Fig. S9). However, to achieve the theoretical maximum photocurrent density, a absorber with proper thickness is required to harvest most photons beyond bandgap energy in solar spectrum and is simultaneously collected within charge transport distance.4 Thus, it is more important to know the optimum absorber thickness rather than its theoretical maximum. The theoretical calculations indicate that the charge carriers are mainly generated within a depth of 400 nm from the surface (corresponding to a theoretical photocurrent density of 38.3 mA cm−2) and gradually reach their maximum photocurrent density of 40.9 mA cm−2 at about 800 nm depth from the surface (Fig. S10). In this work, our optimized thickness of 750 nm that balances light absorption and charge carrier transport is enough to produce a theoretical photocurrent density of 39.6 mA cm−2 and it can reach 97% of its theoretical maximum photocurrent density. Such a large theoretical photocurrent density indicates that Sb2Se3 has an outstanding light absorption capability and a promising future in photoelectrochemical water splitting.

The PEC performance of the obtained Sb2Se3 film was examined by linear sweep voltammetry (LSV) in a 1 M H2SO4 electrolyte under 1-sun AM1.5G illumination as shown in Fig. 3. As-prepared Sb2Se3 with preferred [001]-orientation deposited with Pt cocatalyst deposition (Pt/Sb2Se3) shows an extremely high saturation photocurrent density of −17.7 mA cm−2 at −0.2 VRHE. The photocurrent density at 0 VRHE rapidly decreases to −1.3 mA cm−2. However, in comparison, the Sb2Se3 with common optimal [221] preferred orientation (Fig. S11) shows a much lower saturation photocurrent density of −5.9 mA cm−2 at −0.2 VRHE and a photocurrent density of −0.37 mA cm−2 at 0 VRHE. To further elucidate this, we conducted electrochemical impedance spectroscopy measurements. As shown in Fig. S11b, compared to [221]-oriented Sb2Se3, [001]-oriented Sb2Se3 shows a much smaller transport impedance, which reveals a better charge transport of preferred [001]-orientation over other common orientations.


image file: c9nr08700a-f3.tif
Fig. 3 Photoelectrochemical performance characteristics. JV curves of Pt/[001]-Sb2Se3, Pt/annealed [001]Sb2Se3 and Pt/TiO2/annealed [001]-Sb2Se3 photocathode under chopped AM1.5G simulated illumination in 1 M H2SO4 electrolyte.

Although a maximum high saturation photocurrent density of −17.7 mA cm−2 is achieved, this photocurrent density is only approximately 45% of its theoretical maximum photocurrent density (−39.6 mA cm−2). Therefore, there is still plenty of room for further improvements of the PEC performance. To further enhance the PEC performance, we carried out a post-annealing treatment for the sample. As shown in Fig. 3, the saturation photocurrent at −0.2 VRHE is greatly increased from −17.7 mA cm−2 to −21.2 mA cm−2 for the Pt/annealed Sb2Se3 photocathode, which is about 54% of its theoretical maximum photocurrent density. Furthermore, the photocurrent density at 0 VRHE is also dramatically enhanced from −1.3 mA cm−2 to −12.2 mA cm−2, which is higher than or comparable to most of the reported Sb2Se3-based photocathodes with better heterojunction structures (Table S1) and emerging low-cost and commercial absorber photocathodes (listed in Table S1). These results highlight the promising future of the [001]-oriented Sb2Se3-based photocathodes for high efficiency PEC devices.

To elucidate the effect of the post annealing treatment, we carried out SEM (Fig. S12a and 12b), XPS (Fig. S12c and S12d), Raman (Fig. S12e) and XRD (Fig. S12f) measurements. These results indicate that the post annealing treatment has an insignificant effect on the morphology, chemical compositions and the crystal structure. However, we noticed that the (002) peak of the enlarged XRD pattern (Fig. S12g) shows a slight shift toward larger angles due to lattice contraction. So, we suppose that the dramatically enhanced performance with the post annealing treatment may have originated from the improvement of electronic properties due to changes in defect occupation in the lattice structure. Lattice defects can be generally categorized into three main types: intrinsic point defects (e.g. vacancy, interstitial defect and anti-site defect), extrinsic defects (e.g. foreign dopant) and interface defects (e.g. dangling bond).11,44–46 Bearing in mind that our Sb2Se3 films were fabricated by selenizing the Sb films via Se diffusion under high temperature and the crystal structure of Sb2Se3 contains very large vdWs interstitial space (3.74 Å, Fig. S1), we inferred that the most possible detrimental defects should be the excess Se interstitial-type (Sei) defects that are intercalated into the vdWs gap.42,47,48 Therefore, we observed a noticeable lattice contraction due to the removal of intercalated Se. In addition, the band edge in transmission spectra (Fig. S12h) shows a blue shift after the post-annealing treatment, which indicates that possible Sei defects may introduce a defect state within the bandgap of Sb2Se3. First-principle calculations based on density functional theory (DFT) indicate that there are four typical crystallographic sites for Sei defects (Fig. 4a and c) with formation energies of −2.01, −5.38, −2.02 eV and 2.0 eV, respectively (Table S2). The density of states (DOS) (Fig. 4b and d) shows that interstitial Se defects in vdWs gaps (vdWs I, vdWs II) create band tailing by generating defective states near band edges and the interstitial Se defects in lattice (lattice I, lattice II) lead to band tailing and even defective bands within the bandgap, which will cause severe charge recombination due to the fluctuation of the band potential.49,50 This result corroborates the origin of the band edge blue shift in transmission spectra (Fig. S12h). The post-annealing treatment can provide sufficient kinetic energy to remove excess intercalated Se atoms in the vdWs gap/lattice, enabling a dramatic improvement in the PEC performance of the Sb2Se3 photocathode.


image file: c9nr08700a-f4.tif
Fig. 4 (a–c) Calculation models of typical interstitial Sei defects at vdWs gaps and in the lattice of the Sb2Se3 structure, (b–d) density of states of two typical interstitial Sei defects and pristine Sb2Se3.

SKPM technique is a powerful experimental tool to study the electrostatic potential fluctuation at the sample surface with nanoscale spatial resolution, which can effectively reflect the electronic properties of the sample.51 As shown in Fig. 5, the contact potential difference (CPD) fluctuation of the as-prepared Sb2Se3 film is about 35 mV (Fig. 5c and e), which is much smaller than that of Cu(In,Ga)Se2 and CZTS films (generally >100 mV)52,53 due to the absence of unsaturated chemical bonds between vdWs gap. Upon annealing, the film exhibits a lower CPD fluctuation of 15 mV (Fig. 5d and e). The lower fluctuation of CPD reveals a lower surface charged defect density, manifesting that the Sei defects are really eliminated through the post annealing treatment.53–56 Furthermore, the statistical distribution peak for the CPD (Fig. 5e) exhibits a shift of about 105 mV toward the lower potential direction upon post-annealing treatment, corresponding to the shift of the Fermi level toward the lower potential by 105 mV. The down-shifted Fermi level can induce a larger built-in field at the space charge region (SCR) due to larger band bending, which will favor the charge carrier extraction.57,58 As expected, the SKPM measurement also indicates that the post annealing treatments indeed diminish defective states among vdW gaps. So, the annealing treatment is very critical for improving the electronic quality of our Sb2Se3 film with [001] preferred orientation and then dramatically enhancing the water reduction performance.


image file: c9nr08700a-f5.tif
Fig. 5 Two-dimensional topography spatial maps and corresponding contact potential difference (CPD) spatial maps of the Sb2Se3/Mo/SLG photocathode without (a and b) and with annealing (c and d). Histogram of CPD distribution (e).

Although our ribbon-oriented Sb2Se3 photocathode exhibits a very high saturation photocurrent density of −21.2 mA cm−2, the half-cell solar-to-hydrogen efficiency (HC-STH) is only 0.534% due to the low photocurrent density at the higher potential region (Fig. S13 and Fig. 3), which can be ascribed to the insufficient band bending of the Sb2Se3/water junction (SCLJ), which leads to insufficient carrier extraction at positive potentials. To enhance charge carriers’ extraction at positive potentials, a layer of 22 nm amorphous TiO2 was purposely deposited on Sb2Se3 to construct p–n heterojunctions via atomic layer deposition (ALD).

As shown in Fig. 3, the saturation photocurrent density of Pt/TiO2/annealed Sb2Se3 noticeably increased to −25.4 mA cm−2, clearly indicating that ALD-TiO2 can effectively improve charge extraction. Moreover, the photocurrent density of Pt/TiO2/annealed Sb2Se3 at 0 VRHE dramatically increased from −12.2 mA cm−2 to −20.2 mA cm−2 and the onset potential increased from 0.20 VRHE to 0.57 VRHE (the onset potential of photocurrent Eonset here is defined as the potential at which a photocurrent density of 20 μA cm−2 is observed). Such astonishing improvements enable our Pt/TiO2/annealed Sb2Se3 photocathode to deliver a very high HC-STH of 1.36% (Fig. S13). The photocurrent density of −20.2 mA cm−2 at 0 VRHE is much larger than that of optimal commonly reported [120]-oriented Sb2Se3 (−5 mA cm−2) and [221]-oriented Sb2Se3 (−12.5 mA cm−2) with the identical Pt/TiO2/Sb2Se3 heterojunction,30 undoubtedly indicating the superior advantages of our preferred [001]-oriented Sb2Se3 photocathode owing to its optimal charge extraction efficiency (Fig. 1a). Moreover, the incident photon-to-current conversion efficiency (IPCE) was measured at 0 VRHE (Fig. S14a). The IPCE spectrum of Pt/TiO2/annealed Sb2Se3 exhibits a broad spectral response from 300 nm to 1000 nm. The photocurrent density reaches 20.0 mA cm−2 at 0 VRHE by integrating the IPCE with the standard AM 1.5G spectrum, which is well consistent with the photocurrent density obtained from LSV measurements (Fig. 3). The long-term stability is an important fundamental research challenge for a PEC water splitting device. Our Sb2Se3 photocathode exhibits an impressive stability in a 1 M H2SO4 electrolyte (Fig. S14b), with less than 15% decay over two hours. Continuous and stable chronoamperometry tests with large currents and the generation of hydrogen bubbles (Fig. S14c) indicate that the current is mostly derived from the hydrogen reduction reaction, which is consistent with the fact that the faradaic efficiency of the hydrogen reduction reaction on the Pt/ALD-TiO2 surface is typically larger than 90%.30 To the best of our knowledge, both the saturation photocurrent density of −25.4 mA cm−2 at 0.2 VRHE and the photocurrent density of −20.2 mA cm−2 at 0 VRHE are the highest photocurrent densities achieved by Sb2Se3 photocathodes. Clearly, these results can be recognized as a noticeable achievement compared to previous Sb2Se3-based heterojunction photocathodes, thanks to our reliable and facile method of synthesizing Sb2Se3 with highly [001] preferred orientation.

Conclusion

In summary, we have successfully fabricated a highly [001]-oriented Sb2Se3 film coated with an ALD-TiO2 layer as a promising photocathode. Combined experiments demonstrated that the Sb2Se3 film with [001] preferred orientation can provide an outstanding charge transport efficiency, thus enabling an exceptionally high efficiency for PEC water reduction. We also revealed that post annealing treatments can eliminate the excess Se defects from the preparation process and maximize the advantage of ribbon-assisted carrier transport. Furthermore, by further fabricating TiO2/Sb2Se3 heterojunctional structures, a record-high saturation photocurrent density of −25.4 mA cm−2 at −0.2 VRHE, a record high photocurrent density of −20.2 mA cm−2, and a milestone solar-to-hydrogen efficiency of 1.36% were achieved under 100 mW cm−2 simulated solar illumination. The material design and synthesis strategy demonstrated in this work not only demonstrate the potential application of [001]-oriented Sb2Se3 for PEC water reduction but also provide a new platform for designing future functionalized optoelectronic devices based on [001]-oriented Sb2Se3.

Experimental

Sputtering of Sb/Mo/SLG precursor film

Bilayer (first layer of 670 nm sputtering at 3 Pa, second layer of 730 nm sputtering at 0.25 Pa) metal Mo conductor films were deposited on commercial soda-lime glasses (SLG, 16 mm × 10 mm × 1.5 mm) using the DC sputtering method at ∼9.87 W cm−2 power density.59 Afterwards, ∼335 nm Sb precursor layer was deposited on the partly masked Mo/SLG film (defined active area 8 mm × 10 mm, Fig. S2) by the sputtering method at ∼1.97 W cm−2 at 250 °C of substrate temperature.

Preparation of ribbon-oriented Sb2Se3 films

The Sb/Mo/SLG film and high-purity Se powder (4 N, Aladdin) (1.0–2.0 mg) were vacuum sealed in a quartz tube (ID = 10 mm, OD = 12 mm and L = 200 mm) under 1.2 × 10−4 Pa. Then, the sealed quartz tube was calcined in a muffle furnace at elevated temperatures for 8 hours at 325 °C with 1.5 mg of Se. The sample with post-annealing treatment was taken out from the quartz tube and annealed at 350 °C for 1 hour in a horizontal tube furnace with an Ar flux of 200 sccm.

Preparation of [221]-preferential oriented growth Sb2Se3 films

The control sample prepared in a horizontal tube furnace (ID = 58 mm, OD = 60 mm and L = 1200 mm) was synthesized by selenizing our Sb/Mo precursor film at 350 °C for 30 min using 30 mg of Se powder under an inert Ar atmosphere.16

Fabrication of Sb2Se3/Mo/SLG photoelectrode

Briefly, a PVC insulated Cu wire was attached to the metal Mo side of the Mo/SLG film using Ag paint (SPI colloidal silver). Then, a PTFE O-ring (ID = 7 mm) was pasted on the Sb2Se3 region to define the active area during the photoelectrochemical measurements. The active area was calculated by the ImageJ software. Finally, the photoelectrodes were embedded in an insulating 9462 Hysol epoxy resin. A sealed PTFE cell with a quartz window was used for photoelectrochemical tests.

Deposition of the Pt cocatalyst

Pt cocatalyst nanoparticles for kinetically enhancing the hydrogen evolution reaction (HER) were deposited by galvanostatic photo-assisted electrodeposition in a 1 mM H2PtCl6 salt solution. Typically, the capsulated Sb2Se3/Mo/SLG photoelectrode was platinized at a constant current density of −8.5 μA cm−2 for 15 minutes under 100 mW cm−2 simulated sunlight illumination. After the electrodeposition process, the photocathode was immediately rinsed with deionized water for several times.

Atomic layer deposition of the TiO2 layer

TiO2 films were deposited using an MNT-100 (Wuxi MNT Micro and Nanotech Co.) atomic-layer deposition system. The substrate temperature was set to 150 °C during the TiO2 deposition. Tetrakisdimethylamidotitanium (TDMAT) and H2O were used as Ti and O sources, respectively. A total of 500 ALD cycles were carried out. Each ALD cycle consisted of a 0.2 s pulse of TDMAT followed by 15 s of N2 purging and a 0.2 s pulse of H2O followed by 15 s of N2 purging. ALD chamber was maintained under a constant 20 sccm N2 purging and had a background pressure of 49 Pa.

PEC measurements

The photoelectrochemical performance of the obtained photoelectrode was studied using a three-electrode configuration with our prepared Sb2Se3 photoelectrode as the working electrode, a Pt wire as the counter electrode and Ag/AgCl (KCl saturated) as the reference electrode. The working electrolyte was a 1 M H2SO4 aqueous solution (pH = 0). The potential of the working electrode obtained from each PEC measurement was converted into an RHE reference using the following equation:
 
ERHE = EAg/AgCl + 0.059 V × pH + 0.197.(1)

An electrochemical workstation (Compactstat.e10800, IVIUM) was utilized to measure PEC properties. The JV curves were recorded at a scan rate of 10 mV s−1 under chopped AM1.5G simulated sunlight (Newport, LCS-100 94011A). The LCS 100 was certified as an ABB class solar simulator. The light intensity was calibrated to 100 mW cm−2 (1 sun) using a 91[thin space (1/6-em)]150 V Si reference solar cell (Oriel Instrument).

The half-cell solar-to-hydrogen efficiency (HC-STH) of the photoelectrode under AM1.5 simulated sunlight illumination can be evaluated using the following equation:10,60

 
image file: c9nr08700a-t1.tif(2)
where jph is the photocurrent density, ERHE is the RHE potential of the working electrode, EH+/H2 is the equilibrium potential for hydrogen evolution (0 VRHE), ηF is the faradaic efficiency (assuming 100%) and Ptoal is the incident simulated sunlight intensity (100 mW cm−2). It should be noted that HC-STH is a convenient maximum estimation of the applied-bias solar-to-hydrogen efficiency (AB-STH) in the three-electrode cell configuration and was employed to evaluate the photoelectrochemical efficiency in this work.

Structural and chemical characterization

The structure, crystallinity and growth orientation were investigated by X-ray diffraction (XRD) (PANalytical X'pert powder diffractometer equipped with Cu Kα radiation). Macrotextures of the samples were determined by the Schulz reflection method with a Cu Kα radiation source using an X-ray diffractometer (D/max-2500, Rigaku). Four incomplete experimental pole figures of (120), (221), (211) and (002) planes were collected with tilt angle (χ, step size 5°) 0°–70° and azimuthal angle 0°–360° (ϕ, step size 5°). The complete pole figures were restructured using the mtex toolbox.61 Raman spectroscopy (LabRAM HR Evolution, HORIBA Jobin Yvon) using a 532 nm laser as the excitation source was also employed as a supplementary technique to identity the structural information and possible surface chemical transitions during the experimental optimization process. UV-VIS-NIR transmission spectra were recorded on a Shimadzu UV-3600 spectrophotometer equipped with an integrating sphere. Surface morphology evolution and cross-sectional images of Sb2Se3 films were acquired using a field emission scanning electron microscope (FESEM, MIRA3 TESCAN). The microstructure of the Sb2Se3/Mo chemical intermixing interface was investigated via a Talos F200S (Thermo Fisher Scientific) transmission electron microscope (TEM). Energy dispersive X-ray spectroscopy (EDS) was also performed on this TEM. Surface chemical state information was obtained on an ESCALAB250Xi X-ray photoelectron spectrometer (XPS, Thermo Scientific) with monochromatic Al Kα radiation as the excitation source. Scanning Kelvin probe microscopy (SKPM) measurements were recorded on an MFP-3D-BIO (Asylum Research) atomic force microscope (AFM) using a Pt/Ir coated Si conductive probe (Econo-SCM-PIT, Asylum Research). Gold films (work function 5.1 eV) deposited on SLG glass were employed to calibrate the work function of the Si tip. All the measurements were conducted using an identical conductive probe.

Calculating the texture coefficient

The texture coefficient (TC) of 10 selected dominant XRD peaks for the Sb2Se3 films fabricated at different temperatures with different Se contents were calculated by the following equation:
 
image file: c9nr08700a-t2.tif(3)
where I(hkl) is the measured intensity of the (hkl) reflection plane in the XRD patterns for Sb2Se3 film samples, I0(hkl) is the corresponding intensity of the (hkl) reflection plane in standard data (JCPDS 15-0861). The N is the total number of diffraction peaks used in the calculation.

First principle calculation

The neutral defect calculations were based on the density functional theory (DFT) using the Vienna ab initio simulation package (VASP) code.62 The electron–ion interaction was treated by Projector-Argument Wave (PAW) potential and the exchange correlation interaction was described by Generalized Gradient Approximation (GGA) with the Perdew–Burke–Ernzerhof (PBE) exchange correlation function.63 The cut-off energy was 400 eV. A 9 × 11 × 11 Monkhorst–pack k-mesh was employed in electronic structure calculations. The cell shape, volume and ionic positions were fully relaxed until the maximum force on each atom (Hellmann–Feynman force) was less than 0.05 eV Å−1. The tetrahedron method with Bloch correction was applied to calculate the density of states (DOS) and the energy convergence criterion for the self-consistent energy was 1 × 10−4 eV per atom.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (grants 51672031) and Natural Science Foundation of Chongqing, China (Grant No. cstc2019jcyj-msxmX0572). We also appreciate the projects no. 2018CDJDWL0011 and no. 2019CDXZWL002 from the Fundamental Research Funds for the Central Universities and the sharing fund of large-scale equipment of Chongqing University.

Notes and references

  1. N. S. Lewis and D. G. Nocera, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 15729–15735 CrossRef CAS PubMed.
  2. J. A. Turner, Science, 2004, 305, 972–974 CrossRef CAS.
  3. J. A. Turner, Science, 1999, 285, 687–689 CrossRef CAS.
  4. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473 CrossRef CAS PubMed.
  5. O. Khaselev and J. A. Turner, Science, 1998, 280, 425–427 CrossRef CAS PubMed.
  6. J. L. Young, M. A. Steiner, H. Doescher, R. M. France, J. A. Turner and T. G. Deutsch, Nat. Energy, 2017, 2, 17028 CrossRef CAS.
  7. M. M. May, H.-J. Lewerenz, D. Lackner, F. Dimroth and T. Hannappel, Nat. Commun., 2015, 6, 8286 CrossRef CAS PubMed.
  8. W.-H. Cheng, M. H. Richter, M. M. May, J. Ohlmann, D. Lackner, F. Dimroth, T. Hannappel, H. A. Atwater and H.-J. Lewerenz, ACS Energy Lett., 2018, 3, 1795–1800 CrossRef CAS.
  9. H. Doescher, J. L. Young, J. F. Geisz, J. A. Turner and T. G. Deutsch, Energy Environ. Sci., 2016, 9, 74–80 RSC.
  10. T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 43, 7520–7535 RSC.
  11. A. Walsh and A. Zunger, Nat. Mater., 2017, 16, 964–967 CrossRef CAS PubMed.
  12. J. Jean, P. R. Brown, R. L. Jaffe, T. Buonassisi and V. Bulovic, Energy Environ. Sci., 2015, 8, 1200–1219 RSC.
  13. B. Zhou, X. Kong, S. Vanka, S. Chu, P. Ghamari, Y. Wang, N. Pant, I. Shih, H. Guo and Z. Mi, Nat. Commun., 2018, 9, 3856 CrossRef.
  14. L. Wang, D.-B. Li, K. Li, C. Chen, H.-X. Deng, L. Gao, Y. Zhao, F. Jiang, L. Li, F. Huang, Y. He, H. Song, G. Niu and J. Tang, Nat. Energy, 2017, 2, 17046 CrossRef CAS.
  15. Z. Li, X. Liang, G. Li, H. Liu, H. Zhang, J. Guo, J. Chen, K. Shen, X. San, W. Yu, R. E. I. Schropp and Y. Mai, Nat. Commun., 2019, 10, 125 CrossRef PubMed.
  16. R. R. Prabhakar, W. Septina, S. Siol, T. Moehl, R. Wick-Joliat and S. D. Tilley, J. Mater. Chem. A, 2017, 5, 23139–23145 RSC.
  17. J. Kim, W. Yang, Y. Oh, H. Lee, S. Lee, H. Shin, J. Kim and J. Moon, J. Mater. Chem. A, 2017, 5, 2180–2187 RSC.
  18. J. Tan, W. Yang, Y. Oh, H. Lee, J. Park and J. Moon, ACS Appl. Mater. Interfaces, 2018, 10, 10898–10908 CrossRef CAS PubMed.
  19. L. Zhang, Y. Li, C. Li, Q. Chen, Z. Zhen, X. Jiang, M. Zhong, F. Zhang and H. Zhu, ACS Nano, 2017, 11, 12753–12763 CrossRef CAS PubMed.
  20. W. Yang, J. Ahn, Y. Oh, J. Tan, H. Lee, J. Park, H.-C. Kwon, J. Kim, W. Jo, J. Kim and J. Moon, Adv. Energy Mater., 2018, 8, 1702888 CrossRef.
  21. J. Park, W. Yang, Y. Oh, J. Tan, H. Lee, R. Boppella and J. Moon, ACS Energy Lett., 2019, 4, 517–526 CrossRef CAS.
  22. H. Lee, W. Yang, J. Tan, Y. Oh, J. Park and J. Moon, ACS Energy Lett., 2019, 4, 995–1003 CrossRef CAS.
  23. J. Tan, W. Yang, Y. Oh, H. Lee, J. Park, R. Boppella, J. Kim and J. Moon, Adv. Energy Mater., 2019, 9, 1900179 CrossRef.
  24. W. Yang, S. Lee, H.-C. Kwon, J. Tan, H. Lee, J. Park, Y. Oh, H. Choi and J. Moon, ACS Nano, 2018, 12, 11088–11097 CrossRef CAS PubMed.
  25. R. Caracas and X. Gonze, Phys. Chem. Miner., 2005, 32, 295–300 CrossRef CAS.
  26. H. Song, T. Li, J. Zhang, Y. Zhou, J. Luo, C. Chen, B. Yang, C. Ge, Y. Wu and J. Tang, Adv. Mater., 2017, 29, 1700441 CrossRef PubMed.
  27. H. G. Cha, M. J. Kang, I. C. Hwang, H. Kim, K. B. Yoon and Y. S. Kang, Chem. Commun., 2015, 51, 6407–6410 RSC.
  28. C. W. Kim, S. Ji, M. J. Kang, H. Park, F. Li, H.-M. Cheng and Y. S. Kang, Mater. Today Energy, 2019, 13, 205–213 CrossRef.
  29. J. Y. Zheng, A. U. Pawar, C. W. Kim, Y. J. Kim and Y. S. Kang, Appl. Catal., B, 2018, 233, 88–98 CrossRef CAS.
  30. M. J. Kang, C. W. Kim, A. U. Pawar, H. G. Cha, S. Ji, W.-B. Cai and Y. S. Kang, ACS Energy Lett., 2019, 4, 1549–1555 CrossRef CAS.
  31. C. W. Kim, S. J. Yeob, H.-M. Cheng and Y. S. Kang, Energy Environ. Sci., 2015, 8, 3646–3653 RSC.
  32. C. W. Kim, Y. S. Son, M. J. Kang, D. Y. Kim and Y. S. Kang, Adv. Energy Mater., 2016, 6, 1501754 CrossRef.
  33. C. Chen, D. C. Bobela, Y. Yang, S. Lu, K. Zeng, C. Ge, B. Yang, L. Gao, Y. Zhao, M. C. Beard and J. Tang, Front. Optoelectron., 2017, 10, 18–30 CrossRef.
  34. Y. Zhou, L. Wang, S. Chen, S. Qin, X. Liu, J. Chen, D.-J. Xue, M. Luo, Y. Cao, Y. Cheng, E. H. Sargent and J. Tang, Nat. Photonics, 2015, 9, 409 CrossRef CAS.
  35. L. Guo, B. Zhang, Y. Qin, D. Li, L. Li, X. Qian and F. Yan, Solar RRL, 2018, 2, 1800128 CrossRef.
  36. G. B. Harris, Philos. Mag., 1952, 43, 113–123 Search PubMed.
  37. H. S. Han, S. Shin, D. H. Kim, I. J. Park, J. S. Kim, P.-S. Huang, J.-K. Lee, I. S. Cho and X. Zheng, Energy Environ. Sci., 2018, 11, 1299–1306 RSC.
  38. N. K. Mahenderkar, Q. Chen, Y.-C. Liu, A. R. Duchild, S. Hofheins, E. Chason and J. A. Switzer, Science, 2017, 355, 1203–1206 CrossRef CAS.
  39. Y. H. Kwon, M. Jeong, H. W. Do, J. Y. Lee and H. K. Cho, Nanoscale, 2015, 7, 12913–12920 RSC.
  40. J. Ma, Y. Wang, Y. Wang, Q. Chen, J. Lian and W. Zheng, J. Phys. Chem. C, 2009, 113, 13588–13592 CrossRef CAS.
  41. R. Kondrotas, J. Zhang, C. Wang and J. Tang, Sol. Energy Mater. Sol. Cells, 2019, 199, 16–23 CrossRef CAS.
  42. Z.-W. Wang, W.-K. Huang, K.-W. Zhang, D.-J. Shu, M. Wang and S.-C. Li, Phys. Rev. B: Condens. Matter Mater. Phys., 2017, 95, 121403 CrossRef.
  43. Z. Li, X. Chen, H. Zhu, J. Chen, Y. Guo, C. Zhang, W. Zhang, X. Niu and Y. Mai, Sol. Energy Mater. Sol. Cells, 2017, 161, 190–196 CrossRef CAS.
  44. J.-S. Park, S. Kim, Z. Xie and A. Walsh, Nat. Rev. Mater., 2018, 3, 194–210 CrossRef CAS.
  45. D. Shin, B. Saparov and D. B. Mitzi, Adv. Energy Mater., 2017, 7, 1602366 CrossRef.
  46. X. Liu, X. Xiao, Y. Yang, D.-J. Xue, D.-B. Li, C. Chen, S. Lu, L. Gao, Y. He, M. C. Beard, G. Wang, S. Chen and J. Tang, Prog. Photovoltaics, 2017, 25, 861–870 CAS.
  47. S. Hu, K. Gopinadhan, A. Rakowski, M. Neek-Amal, T. Heine, I. V. Grigorieva, S. J. Haigh, F. M. Peeters, A. K. Geim and M. Lozada-Hidalgo, Nat. Nanotechnol., 2018, 13, 468 CrossRef CAS PubMed.
  48. S. Yao, J. Cui, Z. Lu, Z.-L. Xu, L. Qin, J. Huang, Z. Sadighi, F. Ciucci and J.-K. Kim, Adv. Energy Mater., 2017, 7, 1602149 CrossRef.
  49. C. Yan, K. Sun, J. Huang, S. Johnston, F. Liu, B. Puthen, B. P. Veettil, K. Sun, A. Pu, F. Zhou, J. A. Stride, M. A. Green and X. Hao, ACS Energy Lett., 2017, 2, 930–936 CrossRef CAS.
  50. T. Gokmen, O. Gunawan, T. K. Todorov and D. B. Mitzi, Appl. Phys. Lett., 2013, 103, 103506 CrossRef.
  51. I. Visoly-Fisher, S. R. Cohen, K. Gartsman, A. Ruzin and D. Cahen, Adv. Funct. Mater., 2006, 16, 649–660 CrossRef CAS.
  52. R. Baier, C. Leendertz, D. Abou-Ras, M. C. Lux-Steiner and S. Sadewasser, Sol. Energy Mater. Sol. Cells, 2014, 130, 124–131 CrossRef CAS.
  53. J. B. Li, V. Chawla and B. M. Clemens, Adv. Mater., 2012, 24, 720 CrossRef CAS PubMed.
  54. Q. Chen, H. Zhou, Y. Fang, A. Z. Stieg, T.-B. Song, H.-H. Wang, X. Xu, Y. Liu, S. Lu, J. You, P. Sun, J. McKay, M. S. Goorsky and Y. Yang, Nat. Commun., 2015, 6, 7269 CrossRef CAS PubMed.
  55. W.-Q. Wu, Q. Wang, Y. Fang, Y. Shao, S. Tang, Y. Deng, H. Lu, Y. Liu, T. Li, Z. Yang, A. Gruverman and J. Huang, Nat. Commun., 2018, 9, 1625 CrossRef.
  56. R. Chen, F. Fan, T. Dittrich and C. Li, Chem. Soc. Rev., 2018, 47, 8238–8262 RSC.
  57. P. Dai, W. Li, J. Xie, Y. He, J. Thorne, G. McMahon, J. Zhan and D. Wang, Angew. Chem., Int. Ed., 2014, 53, 13493–13497 CrossRef CAS.
  58. M. T. Mayer, Curr. Opin. Electrochem., 2017, 2, 104–110 CrossRef CAS.
  59. J. H. Scofield, A. Duda, D. Albin, B. L. Ballard and P. K. Predecki, Thin Solid Films, 1995, 260, 26–31 CrossRef CAS.
  60. R. H. Coridan, A. C. Nielander, S. A. Francis, M. T. McDowell, V. Dix, S. M. Chatman and N. S. Lewis, Energy Environ. Sci., 2015, 8, 2886–2901 RSC.
  61. F. Bachmann, R. Hielscher and H. Schaeben, Solid State Phenom., 2010, 160, 63 CAS.
  62. G. Kresse and J. Furthmuller, Comput. Mater. Sci., 1996, 6, 15–50 CrossRef CAS.
  63. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1997, 78, 1396–1396 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9nr08700a

This journal is © The Royal Society of Chemistry 2019