Research / 2020 / Article

Research Article | Open Access

Volume 2020 |Article ID 3750314 | https://doi.org/10.34133/2020/3750314

Wenju Ren, Zongwei Mei, Shisheng Zheng, Shunning Li, Yuanmin Zhu, Jiaxin Zheng, Yuan Lin, Haibiao Chen, Meng Gu, Feng Pan, "Wavelength-Dependent Solar N2 Fixation into Ammonia and Nitrate in Pure Water", Research, vol. 2020, Article ID 3750314, 12 pages, 2020. https://doi.org/10.34133/2020/3750314

Wavelength-Dependent Solar N2 Fixation into Ammonia and Nitrate in Pure Water

Received12 Mar 2020
Accepted30 Apr 2020
Published29 May 2020

Abstract

Solar-driven N2 fixation using a photocatalyst in water presents a promising alternative to the traditional Haber-Bosch process in terms of both energy efficiency and environmental concern. At present, the product of solar N2 fixation is either NH4+ or NO3-. Few reports described the simultaneous formation of ammonia (NH4+) and nitrate (NO3-) by a photocatalytic reaction and the related mechanism. In this work, we report a strategy to photocatalytically fix nitrogen through simultaneous reduction and oxidation to produce NH4+ and NO3- by W18O49 nanowires in pure water. The underlying mechanism of wavelength-dependent N2 fixation in the presence of surface defects is proposed, with an emphasis on oxygen vacancies that not only facilitate the activation and dissociation of N2 but also improve light absorption and the separation of the photoexcited carriers. Both NH4+ and NO3- can be produced in pure water under a simulated solar light and even till the wavelength reaching 730 nm. The maximum quantum efficiency reaches 9% at 365 nm. Theoretical calculation reveals that disproportionation reaction of the N2 molecule is more energetically favorable than either reduction or oxidation alone. It is worth noting that the molar fraction of NH4+ in the total product (NH4+ plus NO3-) shows an inverted volcano shape from 365 nm to 730 nm. The increased fraction of NO3- from 365 nm to around 427 nm results from the competition between the oxygen evolution reaction (OER) at W sites without oxygen vacancies and the N2 oxidation reaction (NOR) at oxygen vacancy sites, which is driven by the intrinsically delocalized photoexcited holes. From 427 nm to 730 nm, NOR is energetically restricted due to its higher equilibrium potential than that of OER, accompanied by the localized photoexcited holes on oxygen vacancies. Full disproportionation of N2 is achieved within a range of wavelength from ~427 nm to ~515 nm. This work presents a rational strategy to efficiently utilize the photoexcited carriers and optimize the photocatalyst for practical nitrogen fixation.

1. Introduction

Ammonia (NH3) and nitrate are widely used for agricultural and chemical synthesis purposes [14]. Due to the environmental issues and energy crisis in recent years, NH3 has also gained growing interest as a liquid fuel for fuel cells due to its high energy density and easy storage [5]. However, the industrial production of NH3 is mainly based on the traditional Haber-Bosch process, which consumes nearly 2% of global energy and emits about 1% of greenhouse gases [6, 7]. Extra energy is needed for the production of nitrate from NH3 [1, 8]. As a green and environmentally friendly alternative for ammonia and nitrate synthesis, solar-driven nitrogen fixation in aqueous media using a photocatalyst at room temperature and atmospheric pressure presents a tantalizing approach [913]. However, the current efficiency of synthesizing NH3 and nitrate (NO3-) by a photocatalytic approach is still far from practical purpose [2].

Either NH4+ or NO3- as a solar N2 fixation product has been reported based on a tungsten oxide photocatalyst [8, 14], in which only the photogenerated electrons or holes are utilized. Few studies have exhibited the simultaneous coproduction of NH4+ and NO3- and a mechanism of wavelength-dependent solar N2 fixation. This process utilizes both photogenerated electrons and holes more efficiently. During the N2 reduction reaction (NRR) process, oxygen evolution reaction (OER) will participate in pure water without sacrificial reagents, which occurs throughout the photocatalytic N2 fixation process and consumes the photoexcited holes. Therefore, the overall reaction is as follows:

The voltage per electron is 1.13 V for the above reaction. Previous work has indicated that N2 can also be oxidized to NO3- over pothole-rich ultrathin WO3 nanosheets [8]. It is reasonable that NO3- and NH3 could be produced simultaneously, with N2 fixation proceeding through the following reactions:

The voltage per electron is 1.57 V for Reaction (2), meaning that this reaction route is thermodynamically unfavorable as compared to Reaction (1). However, nitrogen fixation on defected surfaces can be controlled by the reaction kinetics on the catalytic sites like oxygen vacancies on the surfaces of transition metal oxides. It should be noted that Reaction (3), i.e., the oxidation of NO, can occur spontaneously in aqueous solution [15]. Since the oxygen in Reaction (3) can only come from Reaction (1) when the external O2 is removed from the reaction system, the maximization of NO3- will correspond to the consumption of all produced oxygen, which gives the overall reaction as follows:

It is an established understanding that the rate-determining step for N2 fixation is the activation and dissociation of the extremely stable N≡N triple bond (bond strength of ~941 kJ mol-1) [14, 16, 17]. A key step for effective photocatalytic N2 fixation is to efficiently transfer the energetic photoexcited electrons to the rather inert N2 molecule [8]. The N≡N triple bond can be weakened and activated when electrons are injected from the solid-state catalysts into the empty antibonding -orbitals of the nitrogen molecule [8, 18]. For this purpose, abundant active sites with localized electrons should be created so that the N2 molecule can be chemisorbed for facile electron access. Such sites serve as the effective bridge between the energetic photoelectrons and the nitrogen molecule [14]. Surface vacancies with rich localized electrons due to the charge-transfer phenomenon [19] can effectively activate and weaken the N≡N triple bond by inducing chemisorption and electron injection [2, 20, 21]. Diversified surface vacancies, including oxygen [8, 9, 12, 14, 18], sulfur [22, 23], and nitrogen [2426], have been proven to promote the photocatalytic N2 fixation efficiency. On the other hand, the appropriate anion vacancies in the bulk and on the surfaces of semiconductor photocatalysts can facilitate the separation and migration of the photogenerated electrons and holes [27], which is also crucial for photocatalytic reactions [9, 12, 18]. The defects and disordered surfaces of a semiconductor alter the electronic structures by forming midgap states or band tail states [2830], thus extending the light absorption spectrum and resulting in enhanced light absorption capability.

In this work, ultrathin W18O49 nanowires with distorted surface structures containing abundant surface oxygen vacancies were synthesized using a simple solvothermal method and are intended as a prototype for studying the wavelength-controlled N2 fixation in the presence of surface defects. The as-synthesized sample showed photocatalytic activity for N2 fixation to NH4+ and NO3- in pure water from ultraviolet up to near the end of visible light (730 nm) and exhibited high performance under simulated solar light (AM 1.5G) irradiation. The quantum efficiency (QE) reached about 9% at 365 nm through the simultaneous generation of both NH4+ and NO3-. Both experimental and theoretical results suggested that surface oxygen vacancies serve as catalytic sites and are essential for the high efficiency of N2 fixation. The oxidation of N2 was found to be retarded at either sufficiently short or sufficiently long wavelengths, and full disproportionation of N2 through Reaction (4) is achieved during wavelength from ~427 nm to ~515 nm. A mechanism for wavelength-controlled N2 fixation via its simultaneous reduction and oxidation on defected surfaces was proposed, which sheds new light on the understanding of photocatalytic nitrogen fixation with different product selectivity and could provide guidelines for the design of future photocatalysts with higher utilization efficiency of photoexcited carriers.

2. Results and Discussion

W18O49 nanowires were prepared using a solvothermal method, and a reference sample was prepared by subsequently annealing the as-synthesized W18O49 nanowires at 300°C in air for 30 min to eliminate oxygen vacancies from the surface. The crystal structure and phase purity of the as-synthesized blue velvet-like product (Figure 1(a)) were revealed by X-ray diffraction (XRD) to be consistent with the standard monoclinic W18O49 (P2/m) (PDF 05-0392) as previously reported [3133]. No visible changes in XRD patterns can be observed after annealing (Figure S1). The scanning electron microscopy (SEM) images indicate that the as-synthesized W18O49 consists of ultrathin nanowires (Figure 1(b)), while the transmission electron microscopy (TEM) image (Figure 1(c)) further confirms that the diameters and the lengths of the as-synthesized nanowires are <10 nm and >2 μm, respectively. The selected area electron diffraction (SAED) pattern demonstrates that the diffraction rings belong to the (010, 020) planes of the W18O49 structure (inset in Figure 1(c)), consistent with XRD. The interplanar spacing of the (010) planes is ~3.8 Å, and the nanowire grows along the [010] direction (Figure 1(d)). After annealing, the nanowire becomes shorter and thicker (Figures S2a and S2b); however, the interplanar spacing remains 3.8 Å (Figure S2c), which further confirms that the crystal structure remains unchanged after annealing.

The photocatalytic nitrogen fixation performance of the W18O49 nanowires under simulated solar irradiation (AM 1.5G, 400 nm-1100 nm) is presented in Figure 2(a). The yield rates of NH4+ and NO3- within 12 h were about 22.8 μmol L-1 gcat-1 h-1 and 0.54 μmol L-1 gcat-1 h-1, respectively. W18O49 nanowires were also tested under irradiation of a 300 W xenon light (872 mW/cm2). The average NH3 production rate was about 65.2 μmol L-1 gcat-1 h-1, and the yield rate of NO3- was nearly 0.57 μmol L-1 gcat-1 h-1 for the as-synthesized W18O49 nanowires (Figure S3a). In comparison, the average NH3 production rate decreased to ~1.6 μmol L-1 gcat-1 h-1, and no NO3- was produced during the test process using the annealed photocatalyst at 300°C for 30 min (Figure S3b). The dependence of photocatalytic performances on oxygen vacancy concentrations is shown in Figure S4, and ultraviolet LED with high power was used to produce more NO3- for activity comparison. The yields of NH4+ and NO3- reduce as the annealing time (1.5, 3, and 30 min) increases at 300°C under a 300 W xenon lamp and 5 W of 370 nm LED illumination, respectively. The standard curves for quantifying ammonia and nitrate concentrations were calibrated using ion chromatography, which is a precise measurement for most anions and cations. The correlation coefficient values are 0.9993 and 0.9974 for ammonia and nitrate calibration curves (Figure S5), respectively. The peak signal from the lowest calibrated concentration (0.05 ppm) is very sharp and clean (Figures S5e and S5f), indicating that the instrument is capable of reliably measuring low concentrations at this level. Though some measured values for ammonia or nitrate listed in Table S1 are below the lowest calibrated concentration, the linear relationship between the peak area and the concentration of ammonia or nitrate should persist down to the origin. In other words, the calibrations in Figure S5 are acceptable to quantify the amount of ammonia and nitrate produced in the photocatalytic N2 fixation of this work.

To evaluate the photocatalytic stability, the suspension of the as-synthesized W18O49 nanowires was irradiated using a 300 W xenon lamp and tested for ten 12-hour cycles. After each cycle, the catalyst was carefully cleaned by filtration using a copious amount of distilled water to wash off the dissolved NH4+ and nitrate products. Even if a layer of NH4+ might strongly adsorb on the surface of the W18O49 catalyst and even if they are carried over to the next cycle, it is expected that the adsorbed NH4+ cannot easily desorb and only the NH4+ in the bulk solution can be extracted and measured. It is found that the as-synthesized W18O49 nanowires are relatively stable during the cycle test (Figure S3c). The UV-vis absorption spectrum of the sample after one cycle test (12 h) exhibits lower tail absorption intensity compared with that of the as-synthesized W18O49 nanowires (Figure S6), indicating that the concentration of oxygen vacancies decreases slightly. It can be speculated that the number of oxygen vacancies gradually decreased after each cycle according to the reduced photocatalytic activities as shown in Figure S3c. The total turnover number is greater than 114.7% after ten cycles, which confirms the photocatalytic reaction for N2 fixation.

In order to understand the wavelength-dependent catalytic process for N2 fixation, the as-synthesized photocatalyst suspension was irradiated under LED lights of different wavelengths: 365, 384.3, 400, 427, 468.4, 498, 515, 590, 620, 730, and 850 nm. It was found that the as-synthesized W18O49 nanowires can photocatalytically fix N2 to NH4+ and NO3- at wavelengths from ultraviolet up to 730 nm (Figure 2(b) and Figure S7 and Table S1). Although there have been some visible light-sensitive photofixation catalysts reported, our photocatalyst appears to be the one with the widest absorption range so far [14, 34, 35]. Figure 2(c) demonstrates the molar percentage ratio of NH4+ and NO3- to the total production. The ratio of NH4+ gradually decreases from 365 nm to 427 nm and increases from 427 nm to 730 nm with a high NO3- yield (35~40%) during 427~515 nm; the reason will be discussed herein below.

The wavelength-dependent quantum efficiencies (QE) can also be evaluated based on the amount of photofixation products under LED light illumination using the equations [1, 2] in the experimental part. The calculated QE values are closely related to the wavelength and the light absorption ability. The trend of the photon-to-product efficiency vs. wavelength follows an inverted volcano shape in the range from 365 to 498 nm (Figure 2(d)). In this range, the highest and lowest efficiencies are 9% at the wavelength of 365 nm and 5% at 427 nm, respectively, where the light absorption ability is weakest. On the other hand, the absorption edge of the as-synthesized W18O49 nanowires is about 428 nm because the band gap was close to the incident energy of the light source (Figure S8a). The absorption in longer wavelengths must be caused by the defect levels (DLs). According to the density of states (DOS) calculation, there are some defect levels located below the conduction band in the presence of oxygen vacancies in W18O49 (Figures S8b and S8c). The DLs explain why the as-synthesized W18O49 nanowires exhibit tail absorption in the UV-vis absorption spectrum transformed from the Kubelka-Munk formula (Figures S6 and S8a). Since the density of defects is expected to decrease after annealing, the absorption in the tail range for the annealed sample for 30 min is lower (Figure S8a), suggesting that the photocatalytic activity above 428 nm wavelength originates from the DLs. Photon-to-product efficiency gradually decreases in the wavelength range from 498 nm to 730 nm, possibly due to the relatively lower energy of the photoexcited carriers. This trend is proven by the real production of NH4+ and NO3- under different wavelength LED irradiation with approximate light intensity (Figure S7).

In order to determine the source of the oxygen element in the nitrate product, appropriate H218O in normal distilled water was used as the reagent. After 5 h of xenon lamp irradiation, the 18O isotope was quantified by the denitrification method using Delta V-Precon (Thermo Fisher Scientific, Germany, with the detailed measurement method described in Supplementary Materials) to be about 0.20% in NO3-. The measured peak position, peak area, and atomic percentage are shown in Figure S9 and Table S2, and all the calculation values were based on the equipped software on the test instrument. For additional confirmation, control experiments were carried out to prove the photocatalytic nitrogen fixation ability and exclude possible interference from any contaminants. Firstly, neither ammonia nor nitrate can be detected in pure water (100 mL) with the as-synthesized W18O49 photocatalyst (0.05 g) and Ar gas bubble under 300 W xenon lamp irradiation at 25°C (Figure S10a). This result demonstrates that the nitrogen element in the ammonia and nitrate is from the N2 flow through photocatalytic fixation. Secondly, ammonia and nitrate are also undetectable in pure water (100 mL) with the as-synthesized W18O49 photocatalyst (0.05 g) and N2 gas bubble without irradiation at 25°C (Figure S10b). Then, light illumination is required for N2 fixation and the fixation products are not from environmental contaminations. These results unequivocally confirm that the photocatalytic reaction of N2 fixation to ammonia and nitrate indeed happens in this process, and the oxygen element in nitrate originates from water.

The surface defects of the as-synthesized and annealed W18O49 nanowires are analyzed using atomic scale HAADF Z-contrast images. Lattice distortion, polycrystalline, and amorphous regions are revealed on the surface of the as-synthesized W18O49 nanowires (Figure 3(a)). However, all the surface irregularities disappear and the surface becomes smooth after annealing (Figure 3(b) and Figure S11). The electron energy loss spectroscopy (EELS) edges, which are sensitive to the unoccupied local density of states, provide useful information of local oxidation states and coordination chemistry of the W18O49 nanowires [36]. The W and O EELS results of the as-synthesized W18O49 nanowires are plotted in Figure 3(c), where peak A corresponds to the vacant density of states in the hybridized O 2p and W 5d orbitals. Therefore, the intensity of peak A is closely related to the oxidation state of W. As shown in the spectra collected inside the bulk and near the surface of the as-synthesized W18O49 nanowires, there is a significant drop in the intensity of peak A, which indicates a decrease in the valence state of W at the surface of W18O49 nanowires. For the W18O49 nanowires after annealing for 30 min, no visible change was observed in the EELS signals from the surface to the inside.

The chemical composition and the valence states of the as-synthesized and annealed W18O49 nanowires were examined with an X-ray photoelectron spectrometer (XPS). In the full range of XPS spectra, peaks at binding energies corresponding to O and W elements are clearly observed, and no impurities other than carbon are observed in the spectra (Figure S12a). The W 4f core-level spectrum of the as-synthesized sample could be fitted into two doublets with two different oxidation states. The main peaks of W 4f5/2 at 38 eV and W 4f7/2 at 36 eV are attributed to the W6+ oxidation state. The second doublet with a lower binding energy at 34.5 eV and 36.7 eV arises from W 4f5/2 and W 4f7/2 core levels of the W5+ oxidation state. These binding energies belong to the typical oxidation states found in W18O49 nanowires as reported previously [24, 3739]. The above results further confirm that the as-synthesized catalyst is W18O49 rather than WO3. However, the peaks attributed to the W5+ oxidation state disappear after annealing, indicating that the concentration of surface oxygen vacancies decreases significantly after annealing, which is consistent with the EELS (Figure 2(c)). The O 1s high-resolution XPS spectra of W18O49 can be described as the deconvolution into two peaks by the Gaussian distribution (Figures S12b and S12c), where the one at 530.7 and 530.8 eV can generally be assigned to the bridging oxygen on the W18O49 surface, while the peak at 531.5 eV is attributed to O-H in the oxygen defects [4042].

Electron paramagnetic resonance (EPR) spectroscopy and temperature-programmed desorption of N2 (N2-TPD) were used to investigate the oxygen vacancies and N2 adsorption on the surface oxygen vacancies. In Figure 3(e), a signal exists at around caused by oxygen vacancies in the as-synthesized and annealed sample for 30 min [12]. Thus, there are still some oxygen vacancies in the annealed sample. In Figure 3(f), a single desorption peak of N2 begins at 450 K and centers at 604.6 K, of which the positions suggest chemisorption of nitrogen on the surface. A first-order process was indicated with the N2 peak unchanged with coverage for adsorption at 545 K. N2 was the major desorption product, with a peak at 650 K on Ru/Al2O3 [4346]. Since the weights of samples were identical to these two TPD runs, the difference in peak areas can be used to compare the amount of N2 desorbed. For the annealed sample, the N2 TCD signal of the annealed W18O49 dramatically decreases in comparison with that of the as-synthesized sample, suggesting that there was a significant drop in the population of surface oxygen vacancies after annealing. In combination with the EPR and TCD results, it can be concluded that the surface oxygen vacancies are reduced but still expected to exist inside the annealed sample.

The PL emission spectra of the as-synthesized and annealed W18O49 nanowires for 30 min were examined in the wavelength range of 330-500 nm with an excitation wavelength of 280 nm at room temperature (Figure 3(g)). The blue emission peak at 421 nm is attributed to the oxygen vacancies in tungsten oxide nanowires and nanorods [47, 48], and the green emission peak centered at 483 nm is usually related to the intrinsic defect structures reduced particularly from oxygen deficiency [49]. The increased PL intensity demonstrates that more photogenerated electron-hole recombination occurs in the annealed sample. The PL result reveals that oxygen vacancies assist the separation of the photoexcited carriers, possibly by functioning as trapping sites.

The specific surface area is a key factor for the photocatalytic performance. The N2 adsorption and desorption isotherms of the as-synthesized and annealed W18O49 nanowires for 30 min were measured to evaluate their BET surface areas. The isotherms of both samples are of classical type IV with a hysteresis loop (Figure S13). The BET surface areas of the original W18O49 and annealed W18O49 were calculated to be 437.1 and 93.2 m2 g-1, respectively. Apparently, the annealing process resulted in a significant reduction in the surface area, most likely due to the change in the shape of the nanowires with the reconstruction of the rough surface. Although the specific surface area of the annealed sample was below one-fourth that of the as-synthesized sample, the photocatalytic activity turned out to be below 1/30 of the as-synthesized sample (Figure S3b). Therefore, the reduction in the specific surface area of the annealed sample alone cannot account for the significant decrease in the photocatalytic performance. It is the surface oxygen vacancy that primarily determines the performance of photofixation of N2.

In order to better understand the mechanisms for nitrogen fixation on the W18O49 nanowire from a microscopic point of view, density function theory calculation was carried out. The unit cell of W18O49 was optimized (Figure S14), and the relaxed lattice parameters (, , , and ) are consistent with pervious work [14]. The (001) surface was modelled with a supercell to investigate the N2 fixation process. We found that the N2 molecule cannot be chemically adsorbed on a perfect W18O49 surface (Figure S15), in which case the bond length of the adsorbed N2 is nearly identical to that of the gaseous phase. When an oxygen vacancy was introduced (Figure 4(a)), however, the Bader charge analysis showed that is localized on each of the two W atoms around the oxygen vacancy. Therefore, both N atoms of N2 could form strong bonds with the W atoms around the oxygen vacancy, and the N-N bond length is significantly stretched from 1.11 Å to 1.21 Å with adsorption energy of -1.70 eV, indicating the activation of the N2 triple bond. The charge difference analysis with Bader charge analysis was performed to analyze the charge of N2 adsorption configuration (Figure S16 and Table S3). It is found that the adsorbed N2 gain and the charges accumulate in the area between the bonded N and W atoms, while a charge depletion region is created between both N atoms, indicating that the N2 triple bond is weakened, thus facilitating the following nitrogen fixation reactions [50]. Free energy profiles toward different products were calculated under (Figures 4(b)–4(f)), and the optimized geometries for the reaction intermediates are presented in Figures S17–S22. For NRR, the most energetically favorable pathway is shown in Figure 4(b). In this process, the potential-determining step (PDS) is the last hydrogenation step (NH2-NH3) and the corresponding energy () is 1.71 eV. In NOR, N2 can be oxidized to produce two NO molecules (Figure 4(d)) with of 1.20 eV. Intriguingly, it is found that N2 can also disproportionate at a single oxygen vacancy, forming NH3 and NO successively (Figure 4(c)). The PDS is that the hydroxyl attacks the N intermediate to form the NOH intermediate and is 1.18 eV, demonstrating that such disproportionation reaction probably prevails in the N2 fixation process. The pathways for the OER on the W18O49 (001) with an oxygen vacancy and the pristine W18O49 (001) are conceived from studies of OER on other metal oxides and depicted in Figures 4(e) and 4(f). According to the calculated for both pathways, OER appears to be more feasible on a facet without an oxygen vacancy. The formation of H2O2 was also examined (Figure S23), and it turned out that H2O2 can hardly take part in the reactions of our system.

Although many factors may play a role in the entire photocatalytic process of N2 fixation on the W18O49 nanowires, we believe that the oxygen vacancies on the nanowire surface are essential in promoting the chemical adsorption of N2 molecules and providing catalytic-active sites for both ammonia and nitrate formations. Here, we propose the wavelength-controlled mechanism of photocatalytic N2 fixation on W18O49 nanowires (Figure 5). Under the whole range light irradiation from 365 to 730 nm used in this work, the photoexcited electrons transfer to the surface oxygen vacancies and reduce the chemisorbed N2 molecule to NH3 (Figure 5(a)). However, in the short wavelength range from 365 nm to around 427 nm, intrinsic absorption of light is valid and the photoexcited holes are generated in the bulk and near the surfaces. In this case, the highly mobile holes are delocalized over the surface regions and can reach the W sites either with or without a nearby oxygen vacancy. Owing to the favorable OER at the W sites without a nearby oxygen vacancy (Figures 4(e) and 4(f)), only a small portion of photoexcited holes are injected to the oxygen vacancies where NOR takes place. The OER takes more advantage under the shorter wavelength light irradiation (Figure 5(c)), which corresponds to the increase in the ratio of NO3- at wavelength from 365 nm to around 427 nm (Figure 2(c)). In the long wavelength from around 427 nm to 730 nm (Figure 5(d)), intrinsic absorption is not available; hence, all the reactions are most likely to occur on oxygen vacancies due to light absorption by DLs. Since the equilibrium potential of the N2/NO redox couple is 0.44 eV higher than that of the H2O/O2 redox couple (Figure 5(b)), OER will be more thermodynamically favorable than NOR at longer wavelength. As a result, the ratio of NO3- decreases in this range of wavelength (Figure 2(c)). It is worth mentioning that for wavelength from 427 nm to 515 nm, the ratio between the produced NH4+ and NO3- is close to 5 : 3. Given that the valence changes of N from N2 to NH4+ and NO3- are -3 and 5, respectively, we can deduce that nearly all the O2 molecules produced by OER are consumed by the oxidation of NO to NO3-, i.e., through Reaction (4). Under this condition, O2 should be regarded as a reaction intermediate rather than a reaction product, and all the photoexcited holes that participate in reactions will take part in the oxidation of N2. In other words, the photogenerated carriers are most efficiently utilized in this wavelength range.

3. Conclusions

In summary, we have developed oxygen vacancy-rich W18O49 ultrathin nanowires as an excellent photocatalyst for N2 fixation into ammonia and nitrate. Our investigation revealed that the oxygen vacancies promote the light absorption from the visible to the NIR region, improve the separation ability of the photoexcited electrons and holes, and also serve as the active sites for N2 chemisorption and the bridging between the photogenerated carriers from the catalyst to the N2 molecules. The total quantum efficiency can reach 9% at the irradiation wavelength of 365 nm. Theoretical results show that the oxygen vacancies are the catalytic sites for the formation of both ammonia and nitrate. Interestingly, the molar percentage ratio of NH4+ to the total production (NH4+ and NO3-) shows a gradual decrease from 365 nm to 427 nm, followed by an increase from 427 nm to 730 nm. This trend can be rationalized as follows: in the short wavelength range, the energetically favorable OER predominates at the W sites without a nearby oxygen vacancy due to the intrinsic absorption of the catalyst and the delocalized nature of the photoexcited holes; in long wavelength ranges, NOR becomes more energetically challenging as compared with OER at oxygen vacancies according to the equilibrium potential for both reactions. The photogenerated carriers are most efficiently utilized in the wavelength range from ~427 nm to ~515 nm. This work presents a new insight into the role of oxygen vacancies in the wavelength-dependent photocatalytic nitrogen fixation and demonstrates the underlying mechanisms that could guide the design of future photocatalysts of higher efficiencies.

Conflicts of Interest

The authors declare no competing interests.

Authors’ Contributions

Wenju Ren, Zongwei Mei, Shisheng Zheng, and Shunning Li contributed equally to this work.

Acknowledgments

The research was financially supported by the National Key R&D Program of China (2016YFB0700600), the Guangdong Innovation Team Project (No. 2013N080), the Soft Science Research Project of Guangdong Province (No. 2017B030301013), the Shenzhen Science and Technology Research Grant (ZDSYS201707281026184), and the Shenzhen Peacock Plan (KQTD2014062714543296).

Supplementary Materials

Preparation of W18O49 nanowires. Material characterizations. Photocatalytic activity evaluations. Isotope test. Figure S1: the XRD patterns of the original and annealed W18O49 nanowires. Figure S2: (a) SEM, (b) TEM, and (c) HRTEM images of the annealed W18O49 nanowires. Figure S3: the photocatalytic NH4+ and NO3- production ability of the (a) as-synthesized W18O49 nanowires and (b) annealed W18O49 nanowires for 30 min under a xenon lamp. (c) Stability of W18O49 nanowires for N2 reduction under a xenon lamp (light source: 300 W xenon lamp; 25°C). Figure S4: (a) NH4+ yield under 300 W of xenon lamp illumination and (b) NO3- yield under 5 W of 370 nm LED illumination for photocatalytic N2 fixation of the as-synthesized W18O49 nanowires (0.05 g) and the sample annealed for different times at 300°C. Figure S5: standard curve of (a) nitrate and (b) ammonia with high-performance ion chromatography. (c) Measurement curves of NH4+ from 0.05 ppm to 1 ppm tested by ion chromatography. (d) Measurement curves of NO3- and NO2- from 0.05 ppm to 1 ppm tested by ion chromatography. (e) The enlarged curves of (c). (f) The enlarged curves of (d). Figure S6: the UV-vis absorption spectra of the as-synthesized W18O49 nanowires and the as-synthesized sample after one cycle test of photocatalytic N2 fixation for 12 h. Figure S7: the photocatalytic NH4+ and NO3- production ability of the as-synthesized W18O49 nanowires under different wavelength LED irradiation: (a) 384.3 nm; (b) 400 nm; (c) 427 nm; (d) 468.4 nm; (e) 498 nm; (f) 515 nm; (g) 590 nm; (h) 620 nm; (i) 850 nm. Figure S8: (a) the UV/vis absorption spectra of W18O49 nanowires and annealed W18O49 nanowires. (b) Density of states for W18O49. (c) Density of states for W18O49 with an oxygen vacancy. Figure S9: the typical measurement peaks of different N2O. Figure S10: (a) the ammonia and nitrate concentration curves with the as-synthesized W18O49 photocatalyst (0.05 g) and Ar gas bubble in water (100 mL) under 300 W of xenon lamp irradiation (25°C). (b) The ammonia and nitrate concentration curves with the as-synthesized W18O49 (0.05 g) photocatalyst and N2 gas bubble in water (100 mL) without irradiation (25°C). Figure S11: the HRTEM image of W18O49 nanowires after annealing. Figure S12: (a) XPS full-range spectra of the original and annealed W18O49 nanowires for 30 min. High-resolution XPS O 1s of (b) W18O49 nanowires. (c) Annealed W18O49 nanowires for 30 min. Figure S13: N2 adsorption/desorption isotherms at 77 K of W18O49 nanowires and annealed W18O49 nanowires. Figure S14: the top view, side view, and lattice parameters of the bulk unit cell of W18O49. Figure S15: the adsorption configuration of the N2 molecule on the perfect W18O49 (001) facet without an oxygen vacancy. Figure S16: the different oxygen vacancies on the W18O49 nanowires. Figure S17: the charge difference analysis of the N2 adsorption configuration on the W18O49 (001) facet with one oxygen vacancy; blue represents charge accumulation while red indicates charge depletion. Figure S18: the optimized geometries for reaction intermediates during the nitrogen fixation process to the NH3 product over W18O49 (001) with one oxygen vacancy. Figure S19: the optimized geometries for reaction intermediates during the nitrogen fixation process to NH3 and NO products over W18O49 (001) with one oxygen vacancy. Figure S20: the optimized geometries for reaction intermediates during the nitrogen fixation process to the NO product over W18O49 (001) with one oxygen vacancy. Figure S21: the optimized geometries for reaction intermediates during the OER process over W18O49 (001) with one oxygen vacancy. Figure S22: the optimized geometries for reaction intermediates during the OER process over the pristine W18O49 (001). Figure S23: formation of H2O2 on the W18O49 (001) facet (a) with and (b) without one O vacancy and (c) the 2OH and O2H2 intermediates on the surface without an O vacancy. Table S1: the area of characteristic peak and corresponding ammonia and nitrate concentration. Table S2: the delta value of atomic percentage of 18O. Table S3: the formation energy of different oxygen vacancies. (Supplementary Materials)

References

  1. J. G. Chen, R. M. Crooks, L. C. Seefeldt et al., “Beyond fossil fuel–driven nitrogen transformations,” Science, vol. 360, no. 6391, article eaar6611, 2018. View at: Publisher Site | Google Scholar
  2. A. J. Medford and M. C. Hatzell, “Photon-driven nitrogen fixation: current progress, thermodynamic considerations, and future outlook,” ACS Catalysis, vol. 7, no. 4, pp. 2624–2643, 2017. View at: Publisher Site | Google Scholar
  3. J. A. Pool, E. Lobkovsky, and P. J. Chirik, “Hydrogenation and cleavage of dinitrogen to ammonia with a zirconium complex,” Nature, vol. 427, no. 6974, pp. 527–530, 2004. View at: Publisher Site | Google Scholar
  4. B. M. Lindley, R. S. Van Alten, M. Finger et al., “Mechanism of chemical and electrochemical N2 splitting by a rhenium pincer complex,” Journal of the American Chemical Society, vol. 140, no. 25, pp. 7922–7935, 2018. View at: Publisher Site | Google Scholar
  5. R. G. Boothroyd, “A proposed Australian transition to an anhydrous ammonia fuel transport economy to replace liquid petroleum fuels,” WIT Transactions on Ecology and the Environment, vol. 186, pp. 443–456, 2015. View at: Publisher Site | Google Scholar
  6. H. Liu, “Ammonia synthesis catalyst 100 years: practice, enlightenment and challenge,” Chinese Journal of Catalysis, vol. 35, no. 10, pp. 1619–1640, 2014. View at: Publisher Site | Google Scholar
  7. S. L. Foster, S. I. P. Bakovic, R. D. Duda et al., “Catalysts for nitrogen reduction to ammonia,” Nature Catalysis, vol. 1, no. 7, pp. 490–500, 2018. View at: Publisher Site | Google Scholar
  8. Y. Liu, M. Cheng, Z. He et al., “Pothole-rich ultrathin WO3 nanosheets that trigger N≡N bond activation of nitrogen for direct nitrate photosynthesis,” Angewandte Chemie, International Edition, vol. 58, no. 3, pp. 731–735, 2019. View at: Publisher Site | Google Scholar
  9. H. Hirakawa, M. Hashimoto, Y. Shiraishi, and T. Hirai, “Photocatalytic conversion of nitrogen to ammonia with water on surface oxygen vacancies of titanium dioxide,” Journal of the American Chemical Society, vol. 139, no. 31, pp. 10929–10936, 2017. View at: Publisher Site | Google Scholar
  10. G. N. Schrauzer and T. D. Guth, “Photolysis of water and photoreduction of nitrogen on titanium dioxide,” Journal of the American Chemical Society, vol. 99, no. 22, pp. 7189–7193, 2002. View at: Publisher Site | Google Scholar
  11. M. M. T. Khan, R. C. Bhardwaj, and C. Bhardwaj, “Catalytic fixation of nitrogen by the photocatalytic CdS/Pt/RuO2 particulate system in the presence of aqueous [Ru(Hedta)N2] complex,” Angewandte Chemie (International Ed. in English), vol. 27, no. 7, pp. 923–925, 1988. View at: Publisher Site | Google Scholar
  12. S. Wang, X. Hai, X. Ding et al., “Light-switchable oxygen vacancies in ultrafine Bi5O7Br nanotubes for boosting solar-driven nitrogen fixation in pure water,” Advanced Materials, vol. 29, no. 31, article 1701774, 2017. View at: Publisher Site | Google Scholar
  13. J. W. Peters, A. Rasmussen, L. C. Seefeldt et al., “Light-driven dinitrogen reduction catalyzed by a CdS:nitrogenase MoFe protein biohybrid,” Science, vol. 352, pp. 448–450, 2016. View at: Publisher Site | Google Scholar
  14. N. Zhang, A. Jalil, D. Wu et al., “Refining defect states in W18O49 by Mo doping: a strategy for tuning N2 activation towards solar-driven nitrogen fixation,” Journal of the American Chemical Society, vol. 140, no. 30, pp. 9434–9443, 2018. View at: Publisher Site | Google Scholar
  15. S. Yuan, J. Chen, Z. Lin, W. Li, G. Sheng, and H. Yu, “Nitrate formation from atmospheric nitrogen and oxygen photocatalysed by nano-sized titanium dioxide,” Nature Communications, vol. 4, no. 1, p. 2249, 2013. View at: Publisher Site | Google Scholar
  16. M. Kitano, Y. Inoue, Y. Yamazaki et al., “Ammonia synthesis using a stable electride as an electron donor and reversible hydrogen store,” Nature Chemistry, vol. 4, no. 11, pp. 934–940, 2012. View at: Publisher Site | Google Scholar
  17. D. Kumar, S. Pal, and S. Krishnamurty, “N2 activation on Al metal clusters: catalyzing role of BN-doped graphene support,” Physical Chemistry Chemical Physics, vol. 18, no. 40, pp. 27721–27727, 2016. View at: Publisher Site | Google Scholar
  18. H. Li, J. Shang, Z. Ai, and L. Zhang, “Efficient visible light nitrogen fixation with BiOBr nanosheets of oxygen vacancies on the exposed {001} facets,” Journal of the American Chemical Society, vol. 137, no. 19, pp. 6393–6399, 2015. View at: Publisher Site | Google Scholar
  19. Y. Liu, T. Zhou, Y. Zheng et al., “Local electric field facilitates high-performance Li-ion batteries,” ACS Nano, vol. 11, no. 8, pp. 8519–8526, 2017. View at: Publisher Site | Google Scholar
  20. X. Chen, N. Li, Z. Kong, W. J. Ong, and X. Zhao, “Photocatalytic fixation of nitrogen to ammonia: state-of-the-art advancements and future prospects,” Materials Horizons, vol. 5, no. 1, pp. 9–27, 2018. View at: Publisher Site | Google Scholar
  21. M. Li, H. Huang, J. Low, C. Gao, R. Long, and Y. Xiong, “Recent progress on electrocatalyst and photocatalyst design for nitrogen reduction,” Small Methods, vol. 3, no. 6, article 1800388, 2018. View at: Publisher Site | Google Scholar
  22. S. Hu, X. Chen, Q. Li, Y. Zhao, and W. Mao, “Effect of sulfur vacancies on the nitrogen photofixation performance of ternary metal sulfide photocatalysts,” Catalysis Science & Technology, vol. 6, no. 15, pp. 5884–5890, 2016. View at: Publisher Site | Google Scholar
  23. S. Hu, Y. Li, F. Li et al., “Construction of g-C3N4/Zn0.11Sn0.12Cd0.88S1.12 hybrid heterojunction catalyst with outstanding nitrogen photofixation performance induced by sulfur vacancies,” ACS Sustainable Chemistry & Engineering, vol. 4, pp. 2269–2278, 2016. View at: Publisher Site | Google Scholar
  24. H. Ma, Z. Shi, S. Li, and N. Liu, “Large-scale production of graphitic carbon nitride with outstanding nitrogen photofixation ability via a convenient microwave treatment,” Applied Surface Science, vol. 379, pp. 309–315, 2016. View at: Publisher Site | Google Scholar
  25. H. Ma, Z. Shi, Q. Li, and S. Li, “Preparation of graphitic carbon nitride with large specific surface area and outstanding N2 photofixation ability via a dissolve-regrowth process,” Journal of Physics and Chemistry of Solids, vol. 99, pp. 51–58, 2016. View at: Publisher Site | Google Scholar
  26. G. Wu, Y. Gao, and B. Zheng, “Template-free method for synthesizing sponge-like graphitic carbon nitride with a large surface area and outstanding nitrogen photofixation ability induced by nitrogen vacancies,” Ceramics International, vol. 42, no. 6, pp. 6985–6992, 2016. View at: Publisher Site | Google Scholar
  27. Y. C. Zhang, N. Afzal, L. Pan, X. Zhang, and J. J. Zou, “Structure-activity relationship of defective metal-based photocatalysts for water splitting: experimental and theoretical perspectives,” Advancement of Science, vol. 6, no. 10, article 1900053, 2019. View at: Publisher Site | Google Scholar
  28. G. Xi, S. Ouyang, P. Li et al., “Ultrathin W18O49 nanowires with diameters below 1 nm: synthesis, near-infrared absorption, photoluminescence, and photochemical reduction of carbon dioxide,” Angewandte Chemie, International Edition, vol. 51, no. 10, pp. 2395–2399, 2012. View at: Publisher Site | Google Scholar
  29. Z. Mei, B. Zhang, J. Zheng et al., “Tuning Cu dopant of Zn0.5Cd0.5S nanocrystals enables high-performance photocatalytic H2 evolution from water splitting under visible-light irradiation,” Nano Energy, vol. 26, pp. 405–416, 2016. View at: Publisher Site | Google Scholar
  30. S. Cong, F. Geng, and Z. Zhao, “Tungsten oxide materials for optoelectronic applications,” Advanced Materials, vol. 28, no. 47, pp. 10518–10528, 2016. View at: Publisher Site | Google Scholar
  31. H. Bai, W. Yi, J. Liu et al., “Large-scale synthesis of ultrathin tungsten oxide nanowire networks: an efficient catalyst for aerobic oxidation of toluene to benzaldehyde under visible light,” Nanoscale, vol. 8, no. 28, pp. 13545–13551, 2016. View at: Publisher Site | Google Scholar
  32. R. Ganesan, I. Perelshtein, and A. Gedanken, “Biotemplated synthesis of single-crystalline W18O49@C core-shell nanorod and its capacitance properties,” Journal of Physical Chemistry C, vol. 112, no. 6, pp. 1913–1919, 2008. View at: Publisher Site | Google Scholar
  33. M. Govender, L. Shikwambana, B. W. Mwakikunga, E. Sideras-Haddad, R. M. Eramus, and A. Forbes, “Formation of tungsten oxide nanostructures by laser pyrolysis: stars, fibres and spheres,” Nanoscale Research Letters, vol. 6, no. 1, p. 166, 2011. View at: Publisher Site | Google Scholar
  34. J. Ding, Q. Zhong, and H. Gu, “Iron-titanium dioxide composite nanoparticles prepared with an energy effective method for efficient visible-light-driven photocatalytic nitrogen reduction to ammonia,” Journal of Alloys and Compounds, vol. 746, pp. 147–152, 2018. View at: Publisher Site | Google Scholar
  35. Y. Zhao, Y. Zhao, G. I. N. Waterhouse et al., “Layered-double-hydroxide nanosheets as efficient visible-light-driven photocatalysts for dinitrogen fixation,” Advanced Materials, vol. 29, pp. 1–10, 2017. View at: Publisher Site | Google Scholar
  36. D. A. Muller, “Structure and bonding at the atomic scale by scanning transmission electron microscopy,” Nature Materials, vol. 8, no. 4, pp. 263–270, 2009. View at: Publisher Site | Google Scholar
  37. X. Chang, L. Dong, Y. Yin, and S. Sun, “A novel composite photocatalyst based on in situ growth of ultrathin tungsten oxide nanowires on graphene oxide sheets,” RSC Advances, vol. 3, no. 35, pp. 15005–15013, 2013. View at: Publisher Site | Google Scholar
  38. C. Guo, S. Yin, M. Yan, M. Kobayashi, M. Kakihana, and T. Sato, “Morphology-controlled synthesis of W18O49 nanostructures and their near-infrared absorption properties,” Inorganic Chemistry, vol. 51, no. 8, pp. 4763–4771, 2012. View at: Publisher Site | Google Scholar
  39. J. Guo, Y. Shi, H. Zhou, X. Wang, and T. Ma, “A novel composite of W18O49 nanorods on reduced graphene oxide sheets based on in situ synthesis and catalytic performance for oxygen reduction reaction,” RSC Advances, vol. 7, no. 4, pp. 2051–2057, 2017. View at: Publisher Site | Google Scholar
  40. T. Kunyapat, F. Xu, N. Neate et al., “Ce-doped bundled ultrafine diameter tungsten oxide nanowires with enhanced electrochromic performance,” Nanoscale, vol. 10, no. 10, pp. 4718–4726, 2018. View at: Publisher Site | Google Scholar
  41. H. C. Lin, C. Y. Su, Y. H. Yu, and C. K. Lin, “Non-catalytic and substrate-free method to titania-doped W18O49 nanorods: growth, characterizations, and electro-optical properties,” Journal of Nanoparticle Research, vol. 14, 2012. View at: Publisher Site | Google Scholar
  42. D. Y. Lu, J. Chen, S. Z. Deng, N. S. Xu, and W. H. Zhang, “The most powerful tool for the structural analysis of tungsten suboxide nanowires: Raman spectroscopy,” Journal of Materials Research, vol. 23, no. 2, pp. 402–408, 2008. View at: Publisher Site | Google Scholar
  43. J. A. Schwarz, “Temperature-programmed desorption and reaction: applications to supported catalysts,” Catalysis Reviews, vol. 25, pp. 141–227, 1983. View at: Publisher Site | Google Scholar
  44. Y. Chan, “Temperature-programmed desorption of N2, Ar, and CO2 encapsulated in 3A zeolite,” Journal of Catalysis, vol. 50, pp. 319–329, 2004. View at: Publisher Site | Google Scholar
  45. B. Fastrup, “Temperature-programmed adsorption and desorption of nitrogen on iron ammonia synthesis catalysts,” Journal of Catalysis, vol. 150, no. 2, pp. 345–355, 1994. View at: Publisher Site | Google Scholar
  46. B. Fastrup, M. Muhler, H. N. Nielsen, and L. P. Nielsen, “The interaction of H2 and N2 with iron catalysts used for NH3 synthesis: a temperature-programmed desorption and reaction study,” Journal of Catalysis, vol. 142, no. 1, pp. 135–146, 1993. View at: Publisher Site | Google Scholar
  47. W. Hu, Y. Zhao, Z. Liu, C. W. Dunnill, D. H. Gregory, and Y. Zhu, “Nanostructural evolution: from one-dimensional tungsten oxide nanowires to three-dimensional ferberite flowers,” Chemistry of Materials, vol. 20, no. 17, pp. 5657–5665, 2008. View at: Publisher Site | Google Scholar
  48. H. C. Lin, C. Y. Su, and C. K. Lin, “High-yield fabrication of W18O49@TiO2 core-shell nanoparticles: microstructures and optical-thermal properties,” Journal of Nanoparticle Research, vol. 13, no. 10, pp. 4549–4555, 2011. View at: Publisher Site | Google Scholar
  49. B. Moshofsky and T. Mokari, “Length and diameter control of ultrathin nanowires of substoichiometric tungsten oxide with insights into the growth mechanism,” Chemistry of Materials, vol. 25, no. 8, pp. 1384–1391, 2013. View at: Publisher Site | Google Scholar
  50. M. Falcone, L. Chatelain, R. Scopelliti, I. Živković, and M. Mazzanti, “Nitrogen reduction and functionalization by a multimetallic uranium nitride complex,” Nature, vol. 547, no. 7663, pp. 332–335, 2017. View at: Publisher Site | Google Scholar

Copyright © 2020 Wenju Ren et al. Exclusive Licensee Science and Technology Review Publishing House. Distributed under a Creative Commons Attribution License (CC BY 4.0).

 PDF Download Citation Citation
Views1529
Downloads921
Altmetric Score
Citations