Enabling the electrocatalytic fixation of N2 to NH3 by C-doped TiO2 nanoparticles under ambient conditions

The conventional Haber–Bosch process for industrial NH3 production from N2 and H2 is highly energy-intensive with a large amount of CO2 emissions and finding a more suitable method for NH3 synthesis under mild conditions is a very attractive topic. The electrocatalytic N2 reduction reaction (NRR) offers us an environmentally benign and sustainable route. In this communication, we report that C-doped TiO2 nanoparticles act as an efficient electrocatalyst for the NRR with excellent selectivity. In 0.1 M Na2SO4, it achieves an NH3 yield of 16.22 μg h−1 mgcat.−1 and a faradaic efficiency of 1.84% at −0.7 V vs. the reversible hydrogen electrode. Furthermore, this catalyst also shows good stability during electrolysis and recycling tests.

The conventional Haber-Bosch process for industrial NH 3 production from N 2 and H 2 is highly energy-intensive with a large amount of CO 2 emissions and finding a more suitable method for NH 3 synthesis under mild conditions is a very attractive topic. The electrocatalytic N 2 reduction reaction (NRR) offers us an environmentally benign and sustainable route. In this communication, we report that C-doped TiO 2 nanoparticles act as an efficient electrocatalyst for the NRR with excellent selectivity. In 0.1 M Na 2 SO 4 , it achieves an NH 3 yield of 16.22 mg h À1 mg cat.
the reversible hydrogen electrode. Furthermore, this catalyst also shows good stability during electrolysis and recycling tests.
NH 3 is an essential ingredient in the manufacture of fertilizers, medicaments, resins, dyes, explosives, etc. 1-4 In 2017, total worldwide NH 3 production exceeded 150 million tons, and the demand for NH 3 continues to grow. 5 Industrially, NH 3 is produced almost via the Haber-Bosch process. 6 In order to overcome the kinetic limitations of strong N^N triple bonds, elevated temperature (350-550 C) and high pressure (150-350 atm) are necessary throughout the whole process. 7-9 Moreover, it not only consumes a large amount of energy, but inevitably leads to signicant CO 2 emission. So, it is imperative to develop an environmentally friendly process for the sustainable conversion of N 2 to NH 3 . Electrochemical NH 3 synthesis from N 2 and H 2 O is a promising candidate for articial N 2 xation under ambient conditions due to its environment-friendly, convenient and low-cost characteristics. [10][11][12][13][14][15] Although electrochemical reduction is feasible for achieving the conversion of N 2 to NH 3 , it requires electrocatalysts for the N 2 reduction reaction (NRR) to meet the challenge associated with N 2 activation. Noble-metal catalysts such as Ru, 16 Au, 17,18 Ag, 19 and Rh 20 were reported as NRR catalysts with attractive catalytic performances, but the scarcity of these catalysts limits their wide application. Recently, transition metal oxides (TMOs) [21][22][23][24][25][26] have attracted much attention as NRR electrocatalysts, as they are inexpensive and can be easily prepared on a large scale. Therefore, it is still highly desirable to develop TMOs for the NRR. TiO 2 is nontoxic with a high thermal stability, 27 but its low electronic conductivity hinders its electrocatalytic application. 28 It has been reported that carbon doping can enhance the electronic conductivity of TiO 2 and facilitate charge transfer from the bulk to the surface region, 29 offering us a possible catalyst for the NRR, which, however, has not been explored before.
Herein, we report that C-doped TiO 2 nanoparticles (C-TiO 2 ) are effective for electrochemical N 2 conversion to NH 3 with excellent selectivity under ambient conditions. In 0.1 M Na 2 SO 4 , the catalyst achieves an NH 3 yield of 16  XPS spectra (Fig. 2d), three peaks can be deconvoluted at around 284.76, 286.15, and 289.12 eV for C-TiO 2 . The peak at 284.76 eV could be attributed to the surface adventitious carbon. 30 The two peaks at 286.15 and 289.12 eV are characteristic of the oxygen bound species C-O and Ti-O-C, respectively. 34 This result indicates that carbon atoms substitute for some of the lattice titanium atoms and form a Ti-O-C structure. 30 Compared with C-TiO 2 , only one C 1s XPS spectrum corresponding to C-C is observed for the TiO 2 sample, further conrming the existence of C in C-TiO 2 . In addition, the ultraviolet-visible (UV-vis) absorption spectra and the corresponding Kubelka-Munk plots of C-TiO 2 and TiO 2 are displayed in Fig. S3. † The band gap energies of C-TiO 2 (2.79 eV) and TiO 2 (2.96 eV) were determined by the intercept of the plots of (ahn) 1/2 versus photon energy (hn), 35 indicating a narrower band gap aer C doping. The enhancement of visible light absorption for C-TiO 2 and TiO 2 should be attributed to the carbon doping in the TiO 2 lattice, which would introduce a series of localized occupied states into the band gap of the TiO 2 lattice, leading to a strong visible light absorption. 36 All of the above results strongly support the successful preparation of C-TiO 2 nanoparticles.
The electrocatalytic NRR performance of C-TiO 2 was tested using a typical two-compartment and three-electrode device as the reaction vessel. C-TiO 2 was deposited on carbon paper (C-TiO 2 /CP with a C-TiO 2 loading of 0.10 mg) for the test. All of the potentials for the NRR were reported on the RHE scale. The produced NH 3 was detected by spectrophotometry with salicylic acid. 37 The relevant calibration curves are shown in Fig. S4. † The chronoamperometry curves at the corresponding potentials in N 2 -saturated 0.1 M Na 2 SO 4 are displayed in Fig. 3a, which can directly express the relationship between current density and time during the whole test process. Fig. 3b presents the UV-vis absorption spectra of the electrolyte stained with indophenol indicator aer 2 h electrolysis at a series of potentials, and the   values of absorbance at 660 nm were used to calculate the concentrations of the generated NH 3 at different applied potentials according to the calibration curve of NH 3 . Combined with the collected data, the nal results including the NH 3 yields and FEs under various potentials were calculated and are plotted in Fig. 3c. Both the NH 3 yields and FEs increase as the negative potential rises to À0.7 V, which is the optimum potential point when the NH 3 yield and FE are 16.22 mg h À1 mg cat. À1 and 1.84%, respectively. Aer that, as the potential continually increases, both the NH 3 yields and FEs decrease signicantly which is mainly caused by the competitive hydrogen evolution reaction. For comparison, the pure TiO 2 sample was tested under the same conditions and the corresponding results are presented in Fig. 3d. It is worth noting that the performance of C-TiO 2 is evidently better than that of pure TiO 2 . The superior NRR performance of C-TiO 2 can be rationally attributed to the C-TiO 2 nanoparticles having more exposed active sites (Fig. S5 †), enabling more effective utilization of them as electrocatalysts. The enhanced conductivity of C-TiO 2 also contributes to its higher catalytic activity. The charge transfer resistance related to the electrocatalytic kinetics can be determined from the diameter of the semicircles in the low frequency zone. 38 Electrochemical impedance spectroscopy data (Fig. S6 †) show that C-TiO 2 /CP possesses a smaller radius of the semicircle compared to TiO 2 /CP, suggesting that the C-TiO 2 sample has a lower charge transfer resistance 39 and thus faster NRR kinetics. Meanwhile, C-TiO 2 shows a higher performance than some of the previously reported NRR electrocatalysts. [40][41][42][43][44] More detailed comparisons are listed in Table S1. † To prove that NH 3 was generated via the N 2 reduction process of C-TiO 2 , three sets of control experiments were carried out: (1) immersing the samples in Ar-saturated solution at À0.7 V for 2 h; (2) immersing the samples in N 2 -saturated solution at an open circuit potential for 2 h; and (3) immersing the samples at À0.7 V with alternating 2 h cycles between N 2 -saturated and Ar-saturated solutions, for a total of 12 h. As shown in Fig. 4a and Fig. S7, † a trace amount of NH 3 production was detected under Ar-saturated solution and an open circuit potential. Combined with Fig. S8, † this result indicates that only N 2 provides the nitrogen source to NH 3 . Moreover, controlled trials were carried out to investigate the performance of bare CP. The relevant UV-vis absorption spectra are displayed in Fig. S9. † The results show the poor electrocatalytic activity of bare CP, indicating that C-TiO 2 is an active material for the NRR (Fig. 4b). In addition, stable performance is another important indicator for evaluating catalysts. Recycling tests were performed in N 2 -saturated 0.1 M Na 2 SO 4 6 times and the results are shown in Fig. 4c. The NH 3 yield and FE results show no obvious uctuation over the whole process, suggesting that C-TiO 2 possesses a stable NRR performance. Moreover, only a slight uctuation of current density is observed at À0.7 V aer 24 h electrolysis, further suggesting an excellent electrochemical stability.
Hydrazine (N 2 H 4 ), as a possible by-product in the NRR test, was detected by the method of Watt and Chrisp. 45 The relevant calibration curves are displayed in Fig. S10. † The UV-vis absorption spectra of N 2 H 4 aer 2 h electrolysis in a N 2 atmosphere at a series of potentials are shown in Fig. S11. † The concentrations of the possible by-product N 2 H 4 are determined according to the values of absorbance at 455 nm. The results demonstrated that no N 2 H 4 was detected at all potentials, implying the excellent selectivity of C-TiO 2 as an NRR electrocatalyst.
In summary, C-TiO 2 nanoparticles have been proven as an effective non-noble-metal electrocatalyst for the NRR at moderate temperatures and atmospheric pressure. This electrocatalyst achieves an NH 3 yield of 16.22 mg h À1 mg cat. À1 and a FE of 1.84% at À0.7 V vs. RHE in 0.1 M Na 2 SO 4 . It also exhibits excellent selectivity and satisfactory electrochemical stability during the process of electrochemical NH 3 synthesis under ambient conditions. This work not only offers us an attractive earth-abundant electrocatalyst for the NRR, but also opens up an exciting new avenue for the design and development of doped Ti-based catalysts 46,47 with enhanced performances toward electrocatalytic N 2 and nitrite 48 reduction for applications.

Conflicts of interest
There are no conicts to declare.