Surface plasmon mediates the visible light–responsive lithium–oxygen battery with Au nanoparticles on defective carbon nitride

Significance

Deployment of electric vehicles with long driving ranges demands new electrochemistry beyond the current Li-ion batteries. Aprotic lithium–oxygen batteries have gained considerable attention and promise to provide energy density of approximately three to five times that of state-of-the-art Li-ion batteries, but they are plagued by sluggish reaction kinetics at cathodes and induced large-voltage hysteresis. A photo-mediated strategy has been attempted to reduce the discharge/charge overvoltage, but it is obstructed by the limited ultraviolet light absorption and the inevitably high carrier recombination on semiconductor photocathodes. Here, a plasmonic heterojunction of gold nanoparticle-decorated carbon nitride with nitrogen vacancies is used as a bifunctional catalyst to accelerate the oxygen cathode reactions of the lithium–oxygen battery under visible light.

Abstract

Aprotic lithium-oxygen (Li-O₂) batteries have gained extensive interest in the past decade, but are plagued by slow reaction kinetics and induced large-voltage hysteresis. Herein, we use a plasmonic heterojunction of Au nanoparticle (NP)–decorated C₃N₄ with nitrogen vacancies (Au/N_V-C₃N₄) as a bifunctional catalyst to promote oxygen cathode reactions of the visible light–responsive Li-O₂ battery. The nitrogen vacancies on N_V-C₃N₄ can adsorb and activate O₂ molecules, which are subsequently converted to Li₂O₂ as the discharge product by photogenerated hot electrons from plasmonic Au NPs. While charging, the holes on Au NPs drive the reverse decomposition of Li₂O₂ with a reduced applied voltage. The discharge voltage of the Li-O₂ battery with Au/N_V-C₃N₄ is significantly raised to 3.16 V under illumination, exceeding its equilibrium voltage, and the decreased charge voltage of 3.26 V has good rate capability and cycle stability. This is ascribed to the plasmonic hot electrons on Au NPs pumped from the conduction bands of N_V-C₃N₄ and the prolonged carrier life span of Au/N_V-C₃N₄. This work highlights the vital role of plasmonic enhancement and sheds light on the design of semiconductors for visible light–mediated Li-O₂ batteries and beyond.

The aprotic lithium-oxygen (Li-O₂) battery promises ultrahigh theoretical energy density (∼3,600 Wh·kg⁻¹) and is operated with oxygen reduction to generate the product of Li₂O₂ and its reverse oxidation (2Li⁺ + O₂ + 2e⁻ ↔ Li₂O₂, E⁰ = 2.96 V) (1–5). The sluggish oxygen cathode reactions, including the oxygen evolution reaction (OER) and the oxygen reduction reaction (ORR), lead to a high discharge/charge overvoltage (∼1.0 V) during cycles and low round-trip efficiency (6–9). Since the pioneering work on the photoinvolved Li-O₂ battery using TiO₂ (10) or C₃N₄ (11) under ultraviolet (UV)-light irradiation, reduction of the charge/discharge overvoltage via a photomediated strategy has been extensively studied and is anticipated to solve the kinetic issues of the Li-O₂ battery (12–18). However, the light absorption of most semiconductors used is confined in the region of UV light, accounting for only ca. 4% of the solar spectrum (14–16). Expanding the light harvesting from UV to visible light is the long-term goal and challenge of photocatalysis (17–20). Simultaneously, high carrier recombination consumes the majority of photoelectrons and holes before catalyzing the targeted reactions, resulting in a mismatch between the carrier lifetime and kinetics of ORR or OER (19–21). This necessitates a structural design of semiconducting materials for visible-light harvesting to accelerate the cathode reactions in Li-O₂ batteries.

Localized surface plasmon resonance (LSPR), which refers to the collective oscillation of conduction band (CB) electrons in metal nanocrystals under resonant excitation, has recently gained much attention (22–25). The decay of excited LSPR can produce hot electrons and holes, which initiate various chemical reactions (22, 23). Intriguingly, when plasmonic metal (e.g., Au, Ag) nanoparticles (NPs) come into contact with a semiconductor such as MoS₂, TiO₂, etc., an interfacial Schottky barrier forms; this barrier functions as a filter to force the energetic electrons or holes to migrate across the interface while inhibiting their reverse movement, thereby leading to effective electron–hole separation and suppressed charge–carrier recombination (26–30). LSPR systems generally are composed of plasmonic metal and semiconductors and exhibit the benefits of a low electron–hole recombination rate, enhanced light harvesting, and tailored response wavelengths from the visible to the near-infrared region (22). Recently, Au/CdSe (31) and Au/Ni(OH)₂ (32) heterojunctions have been attempted for a photocatalytic hydrogen evolution reaction and OER with the aid of hot electrons and holes under visible light. Coupling the plasmonic metal with suitable semiconductors for broadened light harvesting and a plasmon-enhanced effect is highly desirable for both ORR and OER in the Li-O₂ battery.

Herein, we report defective C₃N₄ (Au/N_V-C₃N₄) decorated with plasmonic Au NPs as a bifunctional heterojunction catalyst that promotes cathode reactions of the Li-O₂ battery under visible light. The N_V on N_V-C₃N₄ is prone to adsorb and activate O₂, and the plasmon-excited electrons on Au migrate to the CB of N_V-C₃N₄ and relax to the N_V-induced defect band (DB) for O₂ reduction to LiO₂; then it undergoes electron reduction to Li₂O₂. Reversely, the Li₂O₂ is removed by the holes on the Au NPs driven by the applied voltage. The discharge voltage is raised to 3.16 V, and the charge voltage is lowered to 3.26 V at 0.05 mA·cm⁻² with a good rate capability and cycle stability. This investigation integrates a plasmonic heterojunction into the aprotic Li-O₂ battery and illustrates photoenergy conversion and storage under visible light.

Results and Discussion

Characterization of Au/N_V-C₃N₄.

The N_V-C₃N₄ and C₃N₄ were synthesized by heating guanidine hydrochloride at 600 °C and 550 °C, respectively, in an argon atmosphere (33). They were verified by X-ray diffraction (XRD) patterns (Fig. 1A) and Fourier-transform infrared (FTIR) spectra (SI Appendix, Fig. S1), in accordance with the molecular structure of C₃N₄ as shown in SI Appendix, Fig. S2. The presence of N_V on N_V-C₃N₄ is revealed by electron paramagnetic resonance (EPR) analysis. Compared with C₃N₄, which has rare defects, an enhanced EPR signal at g = 2.003 is visible on N_V-C₃N₄, as shown in Fig. 1B, assigned to the unpaired electrons on the carbon atoms. This indicates rich defects on N_V-C₃N₄. The N_V position on N_V-C₃N₄ is further ascertained by X-ray photoelectron spectroscopy (XPS). As shown in SI Appendix, Fig. S3, the ratio between N_2c (two-coordinated N) and N_3c (three-coordinated N) in the N 1s spectrum drops drastically to 1.2 from 3.7 in C₃N₄, indicating that the N_V is preferentially located at N_2c sites, as depicted in Fig. 1B, Inset. An additional C-N_V peak at 289.3 eV (33), as shown in SI Appendix, Fig. S3, also confirms the presence of vacant N_2c atoms for the formation of two-coordinated carbon.

Fig. 1.

(A) XRD patterns of Au/N_V-C₃N₄ and N_V-C₃N₄. (B) EPR spectra of C₃N₄, N_V-C₃N₄, and Au/N_V-C₃N₄. Inset shows the schematic of N_V-engineered C₃N₄. (C) Au 4f and N 1s XPS spectra of Au/N_V-C₃N₄. (D) C 1s XPS spectra of N_V-C₃N₄ and Au/N_V-C₃N₄. (E and F) Scanning transmission electron microscopy and high-resolution transmission electron microscopy images of Au/N_V-C₃N₄.

Au NPs were prepared by a sodium citrate reduction method and deposited on N_V-C₃N₄ to form Au/N_V-C₃N₄. The characteristic peaks at 38.2° and 44.5° are assigned to the (111) and (200) planes of Au shown in Fig. 1A, indicating the loading of Au NPs on N_V-C₃N₄. Fig. 1C shows the Au 4f and N 1s XPS spectra of Au/N_V-C₃N₄, in which the Au 4f_5/2 peak at 87.1 eV and the 4f_7/2 peak at 83.5 eV further reveal the presence of Au NPs. The N 1s spectrum is deconvoluted to N-(C)₃ (N_3c, 399.8 eV), C-N = C (N_2c, 398.6 eV), and C-N-H (400.9 eV), of which the ratio of N_2c to N_3c equals 1.1 and is similar to that of N_V-C₃N₄, as shown in SI Appendix, Fig. S3. These findings suggest the successful preparation of Au/N_V-C₃N₄. In addition, the C 1s XPS spectra in Fig. 1D show that the binding energies of N = C-N₂ and C-N_V on Au/N_V-C₃N₄ are negatively shifted about 0.4 and 0.2 eV, respectively, as compared with those on N_V-C₃N₄. The binding energy of Au 4f in Au/N_V-C₃N₄ is positively shifted by ca. 0.3 eV, compared with that of pure Au NPs as shown in SI Appendix, Fig. S4. These findings confirm that the electrons transfer from Au to N_V-C₃N₄ (30). Fig. 1E displays the high-angle annular dark-field scanning transmission electron microscopy image of Au/N_V-C₃N₄, indicating that the Au NPs with an average size of 30 nm are well dispersed on N_V-C₃N₄ without aggregation, in accordance with the high-resolution transmission electron microscopy images of Au/N_V-C₃N₄ in SI Appendix, Fig. S5. The lattice spacing of 0.23 nm in Fig. 1F is well matched with the (111) plane of Au NPs, and their intimate contact with N_V-C₃N₄ ensures facile interfacial electron transfer. Thermogravimetric analysis of Au/N_V-C₃N₄, as shown in SI Appendix, Fig. S6, indicates that the loading of Au NPs on Au/N_V-C₃N₄ is 1.5 wt %.

Fig. 2A displays the ultraviolet-visible (UV-Vis) absorption spectra of Au/N_V-C₃N₄ and N_V-C₃N₄. The N_V-C₃N₄ shows an absorption edge of 486 nm. The converted Tauc plot shows a band gap (E_g) of 2.67 eV for N_V-C₃N₄, as shown in SI Appendix, Fig. S7A. Compared with C₃N₄ in SI Appendix, Figs. S8 and S9, a decaying tail over the visible region originates from the N_V-induced defect states within the band gap (34), which is consistent with the calculated projected density of state (PDOS) of N_V-C₃N₄ as shown in Fig. 2B. The N_V-induced DB value of N_V-C₃N₄ is estimated to be 2.35 eV according to the converted Urbach plot in SI Appendix, Fig. S7B. The UV photoelectron spectrum of N_V-C₃N₄ in SI Appendix, Fig. S10 indicates that the maximum valence band (VB) value is 4.40 V vs. Li⁺/Li in N_V-C₃N₄ (35). Correspondingly, based on the formula E_CB = E_VB − E_g, the CB value is calculated to be 1.73 V vs. Li⁺/Li. The standard potential of O₂/Li₂O₂ is located between the DB and VB of N_V-C₃N₄ (SI Appendix, Fig. S10), which fulfills the thermodynamic prerequisites of photocatalytic ORR and OER. In contrast to N_V-C₃N₄, Au/N_V-C₃N₄ presents another strong UV/Vis absorption peak at 540 nm as shown in Fig. 2A, which is caused by the plasmonic excitation of Au NPs. The LSPR peak of colloidal Au NPs is at 533 nm, as shown in SI Appendix, Fig. S11. The clear red shift also provides evidence of electron interaction between Au NPs and N_V-C₃N₄ (26), consistent with the XPS spectra in Fig. 1F.

Fig. 2.

(A) UV/Vis absorption spectra of Au/N_V-C₃N₄ and N_V-C₃N₄. (B) Calculated PDOS of N_V-C₃N₄. (C) FDTD simulations of Au/N_V-C₃N₄ and N_V-C₃N₄. E₀ and E represent the incident and localized electric field, respectively. (D) Time-resolved PL spectra of Au/N_V-C₃N₄ and N_V-C₃N₄.

The electric-field intensities around Au/N_V-C₃N₄ and N_V-C₃N₄ are simulated under illumination wavelengths of 350 to 800 nm by the finite difference time domain (FDTD) method, as shown in Fig. 2C. Compared with N_V-C₃N₄, the electric-field intensity of Au/N_V-C₃N₄ is increased by 10 times near the Au NPs with spatially confined “hot spots,” suggesting that the plasmon-induced hot electrons and holes are generated at the interface of N_V-C₃N₄ and Au NPs. Steady-state and time-resolved photoluminescence (PL) spectroscopies were conducted to analyze the charge separation on N_V-C₃N₄ and Au/N_V-C₃N₄. The PL emission of Au/N_V-C₃N₄ is much weaker than that of N_V-C₃N₄, as shown in SI Appendix, Fig. S12, indicating an inhibited charge recombination on Au/N_V-C₃N₄. The PL lifetimes of Au/N_V-C₃N₄ and N_V-C₃N₄ are calculated to be 3.23 and 1.76 ns, respectively, as shown in Fig. 2D. The prolonged PL lifetime also reflects the inhibitive charge recombination after loading Au onto N_V-C₃N₄. Surface photovoltage spectroscopy (SPV) is used to analyze the charge separation degree at nanoscale and to record the surface voltage change under light irradiation. The stronger SPV signal on Au/N_V-C₃N₄ in SI Appendix, Fig. S13 reveals the efficient charge separation and higher carrier concentration of Au/N_V-C₃N₄.

Electrochemical Performances.

The cathodes were prepared by spraying the suspension of Au/N_V-C₃N₄, N_V-C₃N₄, Au/C₃N₄, or C₃N₄ onto carbon paper, as shown in SI Appendix, Fig. S14. The photoinvolved Li-O₂ battery is composed of an Au/N_V-C₃N₄ cathode, an electrolyte of 1.0 M lithium bis(trifluoromethane) sulfonimide (LiTFSI) in tetraglyme (G4), and a Li metal anode, which are assembled in a quartz glass cell (SI Appendix, Fig. S15). Fig. 3A displays the galvanostatic discharge/charge profiles of the Li-O₂ battery with Au/N_V-C₃N₄ and N_V-C₃N₄ with or without illumination at 0.05 mA·cm⁻². The Li-O₂ battery with Au/N_V-C₃N₄ under illumination (Au/N_V-C₃N₄-L) shows an increased discharge voltage of 3.16 V and a lowered charge voltage of 3.26 V, corresponding to an ultrahigh round-trip efficiency of 97.0%. Without illumination (Au/N_V-C₃N₄-D), it exhibits discharge and charge voltages of 2.68 and 4.22 V, respectively, with a round-trip efficiency of 63.5%. The increased round-trip efficiency under illumination is ascribed to solar energy conversion and storage during discharge and charge, respectively. In contrast, the Li-O₂ battery with N_V-C₃N₄ under illumination (N_V-C₃N₄-L) presents charge and discharge voltages of 3.43 and 3.05 V, respectively, with an energy efficiency of 88.9%, as shown in SI Appendix, Fig. S16A. The high efficiency of Au/N_V-C₃N₄ is attributed to the migration of hot electrons from plasmonic Au to N_V-C₃N₄. In addition, the Li-O₂ battery with Au/C₃N₄ under illumination (Au/C₃N₄-L) exhibits discharge and charge voltages of 3.02 and 3.39 V, respectively, possessing a round-trip efficiency of 89.1%, as shown in SI Appendix, Fig. S16B. The higher efficiency of Au/N_V-C₃N₄ compared with Au/C₃N₄ indicates that the existence of N_V promotes the oxygen cathode reaction kinetics upon discharging and charging with extended visible-light absorption, as well as C₃N₄, as shown in SI Appendix, Fig. S16C. In addition, the discharge/charge profile of Au/N_V-C₃N₄ with illumination in an Ar atmosphere (SI Appendix, Fig. S17) excludes the intrinsic redox of Au/N_V-C₃N₄ and the good stability of battery materials.

Fig. 3.

(A) Discharge/charge profiles of the Li-O₂ battery with Au/N_V-C₃N₄ and N_V-C₃N₄ with or without illumination at 0.05 mA·cm⁻². (B) Discharge and charge profiles of Au/N_V-C₃N₄ and N_V-C₃N₄ at various current densities under illumination. (C) Cycling performance with Au/N_V-C₃N₄ and N_V-C₃N₄ cathodes under illumination. (D) IPCE of Au/N_V-C₃N₄ and N_V-C₃N₄ at 420, 520, 550, 600, and 700 nm and absorption spectrum of Au NPs.

When the current density is raised to 0.10, 0.15, and 0.25 mA·cm⁻², the charge/discharge overvoltages of the Li-O₂ battery with Au/N_V-C₃N₄-L are 0.14, 0.20, and 0.27 V, respectively, as shown in Fig. 3B. In the dark, the charge/discharge overpotentials rise quickly under the same current densities as those in SI Appendix, Fig. S18. The reduced polarization under illumination benefits from the photogenerated hot electrons and holes on Au/N_V-C₃N₄. Accordingly, the charge/discharge overvoltages of N_V-C₃N₄-L are 0.45, 0.54, and 0.68 V at 0.10, 0.15, and 0.25 mA·cm⁻², respectively. As the current densities are raised from 0.05 to 0.25 mA·cm⁻², the difference in charge/discharge overvoltages of the Li-O₂ battery between Au/N_V-C₃N₄ and N_V-C₃N₄ becomes more obvious. This suggests that the life span of charge carriers, rather than oxygen cathode reaction kinetics, is the rate-determining step, in good agreement with the time-resolved PL spectra shown in Fig. 2D. Cycling performances of the Li-O₂ battery with Au/N_V-C₃N₄ and N_V-C₃N₄ at 0.15 mA·cm⁻² are displayed in Fig. 3C. With light irradiation, the Li-O₂ battery with Au/N_V-C₃N₄-L displays low overvoltages upon discharging and charging for 50 cycles, in sharp contrast to the greater charge/discharge overvoltages of N_V-C₃N₄-L. This suggests that the plasmonic excitation on illuminated Au/N_V-C₃N₄ is favorable for ORR and OER kinetics and cycling stability.

The increased voltage polarization after 50 cycles may be ascribed to the minimal side reactions at the cathode (1, 4). Iodometric titration is used to quantify the generated Li₂O₂ on the discharged cathode and to further analyze the proportion of side reaction. The discharge product of Li₂O₂ reacts with H₂O via Li₂O₂ + 2H₂O → H₂O₂ + 2LiOH; next, H₂O₂ oxidizes iodide to iodine, which is titrated by Na₂S₂O₃ (SI Appendix, Fig. S19 and Supplementary Method). It reveals the highly reversible 1.99 e⁻/O₂ reaction, compared with the theoretical 2.00 e⁻/O₂. The minimal side reaction products, such as carbonates, may induce increased voltage polarization during cycles. On the other hand, upon charging, the remaining little Li₂O₂ particles away from the cathode/catalyst surface may lose electric contact and result in an increased charge overvoltage. The round-trip efficiency of the Li-O₂ battery with Au/N_V-C₃N₄ still reaches 92.5% after 50 cycles. In contrast, in the reported literature, the discharge/charge voltage polarization after cycles obviously increased, with low round-trip efficiencies (below 85%) due to the high charge carrier recombination and side reactions at high voltages (14–16). In addition, SI Appendix, Fig. S20A shows the discharge/charge profiles of the Li-O₂ battery with Au/N_V-C₃N₄, where illumination is only applicable for charge. It exhibits a lowered charge voltage of 3.30 V with a round-trip efficiency of 82.0%, which outperforms the other competitive cathodes for the Li-O₂ battery (4, 5). A high capacity of 2.50 mAh·cm⁻² for the photoinvolved Li-O₂ battery has been produced, as depicted in SI Appendix, Fig. S20B, and this suggests its practicality.

In addition, the plasmonic effect strongly depends on the wavelength of incident light (36). To verify that the enhanced battery performance of Au/N_V-C₃N₄ benefits from the plasmonic effect of Au NPs, the incident photon to current efficiency (IPCE) at different wavelengths on Au/N_V-C₃N₄ and N_V-C₃N₄ was evaluated (28, 29), as depicted in Fig. 3D and SI Appendix, Fig. S21 and Table S1. The IPCE of N_V-C₃N₄ is much lower than that of Au/N_V-C₃N₄, which reveals efficient plasmonic excitation on Au/N_V-C₃N₄. The highest IPCE of Au/N_V-C₃N₄ is at ∼550 nm, near the LSPR absorption peak of Au NPs, as shown in Fig. 3D. The IPCE of Au/N_V-C₃N₄ at varied wavelengths follows the trend of extinction spectrum of Au NPs, confirming the wavelength-dependent performance of the Li-O₂ battery on the plasmonic effect of Au NPs. A tabular form summarizing the cathode catalysts for the photoinvolved Li-O₂ battery has been provided in SI Appendix, Table S2. The Li-O₂ battery with an Au/N_V-C₃N₄ cathode displays extended visible-light absorption, prolonged carrier lifetime, and better electrochemical performance.

Reaction Mechanism.

The discharge product of Li₂O₂ is identified by XRD in Fig. 4A in the Li-O₂ battery with Au/N_V-C₃N₄-L, in line with the Raman spectra in SI Appendix, Fig. S22A with a characteristic O-O stretching vibration of Li₂O₂ at 780 cm⁻¹ (37). XPS spectra also confirm the formation of Li₂O₂ in the discharged Au/N_V-C₃N₄ electrode, according to the Li 1s signal at 55.0 eV in Fig. 4B and the O 1s signal at 531.2 eV in Fig. 4C (11). The C 1s spectrum in SI Appendix, Fig. S22B indicates no visible parasitic products upon discharging and charging, which is consistent with the iodometric titration results in SI Appendix, Supplementary Method. After charging, the product of Li₂O₂ is reversibly decomposed, as verified by the elimination of its XPS signals, Raman bands, and diffraction peaks, as shown in Fig. 4 and SI Appendix, Fig. S22. As shown in Fig. 4D and SI Appendix, Fig. S23, numerous toroidal Li₂O₂ particles are uniformly covered on cathodes after discharge and are reversibly decomposed upon charging. The reversible behavior is recorded by scanning electron microscopy images for 50 cycles, as shown in SI Appendix, Fig. S23. These indicate the good reversibility of formation and oxidation of Li₂O₂ in the Li-O₂ battery with Au/N_V-C₃N₄-L. The ¹H NMR spectra in SI Appendix, Fig. S24 imply inconspicuous electrolyte decomposition after discharge and charge with illumination.

Fig. 4.

(A) XRD patterns. (B) Li 1s and (C) O 1s XPS spectra. (D) Scanning electron microscopy images of discharged and charged Au/N_V-C₃N₄ cathodes under illumination.

Density functional theory calculations are further performed to simulate the oxygen redox chemistry on Au/N_V-C₃N₄ under illumination. Fig. 5 A and B shows the differential charge densities of O₂-adsorbed N_V-C₃N₄ and C₃N₄. O₂ is preferentially adsorbed on N_V of N_V-C₃N₄ with a larger adsorption energy (−3.33 eV) than that of C₃N₄ (−1.65 eV), which is also confirmed by the large O₂ desorption capacity in the temperature-programmed desorption (TPD) profile of N_V-C₃N₄ relative to C₃N₄ (SI Appendix, Fig. S25). As external electrons are injected into O₂ molecules, the O-O bond can be weakened to present an elongated bond length of 1.38 Å, compared with 1.23 Å in free O₂, which indicates the activation of O₂. Free energy diagrams are plotted to describe the initial formation or dissociation of Li₄O₄ at different applied potentials, as shown in Fig. 5C. The adsorption configuration of N_V-C₃N₄-O₂ clearly shows that the C atoms around N_2c defect sites strongly bond with O₂ and then combine with Li⁺ for N_V-C₃N₄-LiO₂, as shown in SI Appendix, Fig. S26. The LiO₂ intermediate is further confirmed by the characteristic peak at 1,088 cm⁻¹ in the Raman spectrum of discharged cathode with a cutoff capacity of 0.20 mAh·cm⁻², as shown in SI Appendix, Fig. S27. The entire steps of Li₄O₄ formation are exothermal at an open circuit potential of 0 V (U = 0 V), suggesting a spontaneous process on N_V-C₃N₄. Such an energy path can stay downhill until the applied voltage reaches 2.59 V (U_DC = 2.59 V), while the reverse dissociation of Li₄O₄ is thermodynamically favorable with the applied voltage exceeding 3.79 V (U_C = 3.79 V).

Fig. 5.

(A and B) Differential charge densities of O₂-adsorbed N_V-C₃N₄ and C₃N₄. The yellow and blue areas represent charge accumulation and depletion, respectively. The value of isosurface is set to 0.003 e·bohr⁻³. (C) Free energy diagrams of battery reactions. U_DC is the theoretically maximum discharge voltage, and U_C is the theoretically minimum charge voltage. (D) PDOS of N_V-C₃N₄-O₂. Inset shows the partial charge density of the defect energy level of N_V-C₃N₄. (E) PDOS of N_V-C₃N₄ after loading with Au₄ cluster. (F) Schematic of plasmonic hot electron injection and hole transfer for the respective discharge and charge processes in a photoinvolved Li-O₂ battery with Au/N_V-C₃N₄. E_F, Fermi level; SPR, surface plasmon resonance.

Moreover, the PDOS of N_V-C₃N₄-O₂ is calculated to explain the interaction of O₂ and the N_V of N_V-C₃N₄ under illumination, as depicted in Fig. 5D. It shows that midgap states appear in the vicinity of the Fermi level (E_F) of N_V-C₃N₄. As O₂ is adsorbed onto N_V-C₃N₄, there is an apparent peak overlap between the p states of O and the defect states of N_V-C₃N₄, suggesting a strong orbital hybridization, which is consistent with the differential charge density of N_V-C₃N₄-O₂ shown in Fig. 5A. The generation and relaxation of photoelectrons on N_V-C₃N₄ with illumination are depicted in SI Appendix, Fig. S28. The defect energy levels, located between the VB and CB of N_V-C₃N₄, can capture the photogenerated electrons and further facilitate the O₂ reduction. The Au₄ cluster is adopted as a simplified model to analyze the hybridization between the d states of Au clusters and the defect level of N_V-C₃N₄. After loading the Au₄ cluster onto N_V-C₃N₄, a notable hybridization between the d bands of the Au₄ cluster and the defect states of N_V-C₃N₄ is observed, as shown in Fig. 5E. This means that the Au₄ cluster can provide extra d electrons to participate in ORR and enhance the catalytic activity of N_V-C₃N₄.

The reaction mechanism on the Au/N_V-C₃N₄ cathode of the Li-O₂ battery is described in Fig. 5F and SI Appendix, Fig. S29. On discharge, O₂ is first adsorbed and activated at the N_V sites of N_V-C₃N₄. The maximum absorption edge of Au/N_V-C₃N₄ is about 620 nm, corresponding to the photon energy of 2.00 eV. The Au/N_V-C₃N₄ absorbs visible-light photons beyond 2.00 eV to generate hot electrons and holes, which are separated by the space charge region at the interface of Au and N_V-C₃N₄ (38, 39). The hot electrons are transferred to the CB of N_V-C₃N₄, relax to the N_V-induced DB, and then are injected into the molecular orbitals of O₂ for reduction to LiO₂, which further undergoes electron reduction to Li₂O₂. The theoretical discharge voltage equals the potential difference between the E_F and Li⁺/Li (SI Appendix, Fig. S29A). According to the work function (W) of Au (∼5.0 eV) and the formula E_F = W − 4.5 (22, 40), the E_F is calculated to be ∼0.5 V vs. the normal hydrogen electrode, namely ∼3.54 V vs. Li⁺/Li. Therefore, the theoretical discharge potential considering the contribution of light energy is 3.54 V, higher than the equilibrium voltage of 2.96 V, meaning light energy conversion to electricity. In a reverse charge, the holes on Au NPs are driven to oxidize the Li₂O₂ by an applied voltage. The theoretical charge voltage is determined by the voltage gap between the DB and Li⁺/Li to be 2.35 V, lower than the equilibrium voltage of 2.96 V (SI Appendix, Fig. S29B), and corresponding to light energy conversion. The achievable discharge and charge voltages of the Li-O₂ battery with Au/N_V-C₃N₄ are 3.16 and 3.26 V, respectively, at 0.05 mA·cm⁻², and they deviate from the theoretical values (3.54 and 2.35 V, respectively) owing to the low utilization efficiency of light energy and the transport barrier between the catalyst and the Li₂O₂. Controlling the realistic loading of plasmonic metal NPs within a rationally designed three-dimensional porous cathode may improve the utilization efficiency of light and further promote the cathode reaction kinetics in the photoinvolved Li-O₂ battery.

Conclusions

A plasmonic heterojunction, namely Au/N_V-C₃N₄, is used as the bifunctional catalyst for a visible light–responsive Li-O₂ battery to facilitate the formation and oxidation of Li₂O₂. The hot electrons generated on Au NPs transfer to N_V sites for O₂ reduction to LiO₂, which subsequently undergoes reduction to Li₂O₂. Reversely, the holes on Au NPs moved by an applied voltage drive the Li₂O₂ oxidation to O₂ and Li⁺. The achievable discharge voltage of the Au/N_V-C₃N₄ cathode is up to 3.16 V, surpassing the equilibrium voltage by about 200 mV; the charge voltage is decreased to 3.26 V at 0.05 mA·cm⁻² from 4.22 V in the dark, corresponding to an ultrahigh round-trip efficiency of 97.0%. Even at 0.25 mA·cm⁻², the discharge and charge voltages of the Au/N_V-C₃N₄ cathode reach 3.08 and 3.35 V, respectively. The stable discharge and charge of the Au/N_V-C₃N₄ cathode are sustained for 50 cycles. The superior rate capability and cycle stability result in the plasmon-induced hot carrier transfer from Au NPs to N_V-C₃N₄ and the good matching of the carrier life span with oxygen redox kinetics on Au/N_V-C₃N₄. This investigation will promote the rational design of semiconductor electrodes for photoinvolved Li-O₂ batteries and beyond.

Materials and Methods

Preparation of the N_V-C₃N₄ and C₃N₄.

The N_V-C₃N₄ and C₃N₄ were synthesized by heating guanidine hydrochloride (Aldrich; 99%) in an open crucible at 600 °C and 550 °C, respectively, for 4 h, with an Ar flow of 100 mL·min⁻¹ at a ramp rate of 4 °C·min⁻¹. The resultant yellow agglomerates were milled into powder in a mortar. To increase the hydrophilicity of N_V-C₃N₄, 1 g of N_V-C₃N₄ was vigorously stirred in 30 mL of concentrated hydrochloric acid for 1 h at room temperature. The mixture was filtrated and washed with deionized water repeatedly to remove the superfluous acid.

Preparation of Au/N_V-C₃N₄ and Au/C₃N₄.

One milliliter of 1.0 wt % HAuCl₄ solution was added to 96 mL deionized water and then heated to boil under vigorous stirring; 1.5 mL of 1.0 wt % sodium citrate solution was injected into the above solution until the color changed to wine red, indicating the formation of Au NPs. The acid-processed N_V-C₃N₄ powder was dispersed in 20 mL of deionized water under vigorous stirring, into which the as-prepared Au colloidal solution was added dropwise. After stirring for 2 h, the resulting sample (denoted as Au/N_V-C₃N₄) was collected, washed with deionized water several times, and dried in a vacuum at 100 °C for 10 h. The preparation of Au/C₃N₄ was similar to that of Au/N_V-C₃N₄, except for the replacement of N_V-C₃N₄ by C₃N₄.

Battery Assembly and Electrochemical Measurement.

The cathode suspension was prepared by mixing Au/N_V-C₃N₄, N_V-C₃N₄, Au/C₃N₄, or C₃N₄ powder and sodium carboxymethylcellulose with a weight ratio of 9:1. Next, the suspension was sprayed onto carbon paper and dried at 110 °C for 12 h to obtain the Au/N_V-C₃N₄ electrode. The average mass loading of active material in each electrode was ∼0.4 mg·cm⁻². The battery assembly was performed in an Ar-filled glove box with both H₂O and O₂ contents below 0.1 ppm. The Li-O₂ battery was assembled in a sealed quartz glass cell, ensuring sufficient illumination of light on the electrodes with ignored light absorption by the glass. After purging with O₂ into the cell for 30 min, galvanostatic discharge/charge cycles were conducted. The electrolyte was a solution of 1.0 M LiTFSI/G4, and Li metal foil was used as a negative electrode. The effective area of each electrode for illumination was 0.4 cm². A 300-W Xe lamp was used as the light source with a wavelength range of 350 to 800 nm.

Characterization.

XRD (Rigaku MiniFlex600 X-ray generator; λ = 1.5406 Å, Cu Kα radiation) was used to conduct materials analyses. XPS measurements were performed on a Perkin-Elmer PHI 1600 ESCA system to characterize the cathode surface composition. Transmission electron microscopy (FEI Talos F200X G2; AEMC) was conducted to analyze the morphology and microstructure of the Au/N_V-C₃N₄. Scanning electron microscopy (JEOL-JSM7500F) was conducted to observe the morphologies of the discharged and charged cathodes. Before the tests, dehydrated dimethoxyethane was used to wash the discharged/charged cathodes to remove residual electrolytes. The UV-vis absorption measurements were performed on a Thermo Scientific Evolution 220 UV-vis spectrophotometer. FTIR spectra were collected on a Bruker Tensor II Sample Compartment RT-DLaTGS. Raman spectra were recorded with a customized LabRAM HR800 confocal Raman microscope (Horiba Jobin Yvon). O₂ TPD profiles were recorded by a Quantachrome CHEMBET3000 analyzer. ¹H NMR spectra were obtained on a Bruker spectrometer (400 MHz).

Simulations of Plasmonic Near-Field Maps.

Plasmonic near-field maps were simulated at the LSPR wavelengths using the FDTD solution (Lumerical). The diameter of Au NPs was fixed at 30 nm. The scale of N_V-C₃N₄ was set to be 15 × 60 × 150 nm³. The refractive index of N_V-C₃N₄ was assumed to be 1.0, and the permittivity of Au was obtained by fitting experimental data from the literature.

Computational Details.

All the calculations were carried out using the Vienna Ab initio Simulation Package code (41–43). The generalized gradient approximation within the Perdew–Burke–Ernzerhof functional was adopted to describe the electronic exchange and correlation effects (43). The cutoff energy was set to 450 eV. The convergence criteria of the total energy and force were set to 10⁻⁵ eV and 0.02 eV·Å⁻¹, respectively. The model of defective C₃N₄ adopted a 2 × 2 C₃N₄ supercell by removing a pyridine N atom. A vacuum layer of about 20 Å was used to prevent interaction between the periodic slabs. More calculation results are provided in SI Appendix, Note S1 and Table S3.

Data Availability

All study data are included in the article and/or SI Appendix.