Shuo
Li
a,
Guoli
Zhang
*a,
Qianrui
Li
a,
Tianshun
He
a and
Xiaoqi
Sun
*ab
aDepartment of Chemistry, Northeastern University, Shenyang 110819, China. E-mail: 2210076@stu.neu.edu.cn; sunxiaoqi@mail.neu.edu.cn
bNational Frontiers Science Center for Industrial Intelligence and Systems Optimization, Northeastern University, 3-11 Wenhua Road, Shenyang, 110819, China
First published on 10th October 2024
Small quinone molecules are promising cathode materials for aqueous zinc batteries. However, they experience fast capacity decay due to dissolution in electrolytes. Herein, we introduce a simple methyl group to a naphthoquinone (NQ) cathode and demonstrate a facile self-saturation strategy. The methyl group exhibits hydrophobic properties together with light weight and a weak electron-donation effect, which allows a good balance among cycling stability, capacity and voltage for cathode materials. The resulting menadione (Me-NQ) presents around one-third solubility of NQ. The former thus rapidly reaches saturation in the electrolyte during cycling, which suppresses subsequent dissolution. Thanks to this process, the Me-NQ cathode preserves 146 mA h g−1 capacity after 3500 cycles at 5 A g−1, far exceeding 88 mA h g−1 for NQ. Me-NQ also delivers a stabilized capacity of 316 mA h g−1 at 0.1 A g−1 with only 0.05 V lower average redox voltage than NQ. The co-storage of Zn2+ and H+ with the redox reactions on the carbonyl sites of Me-NQ is revealed.
Among organic cathode candidates, small molecule quinones with carbonyl groups as active sites present the ease of synthesis, high capacity and high reaction reversibility. For example, Kundu et al. proposed a p-chloranil cathode in aqueous zinc batteries, delivering a specific capacity of 205 mA h g−1 at 0.2C.32 Zhang et al. uncovered the effect of hydrogen bonding for cation storage capability of quinone-based small molecules and achieved a specific capacity of 180.3 mA h g−1 at 5 A g−1 with an ultra-high loading of 66.2 mg cm−2 for a 2,5-diaminocyclohexa-2,5-diene-1,4-dione (DABQ) cathode.35 Our group revealed facile proton transport through the Grotthuss-type mechanism in tetraamino-p-benzoquinone (TABQ), which reached 308 mA h g−1 capacity at 0.1 A g−1 and retained 213 mA h g−1 at 5 A g−1.36 We also showed that the substitution sites of methoxy groups on quinone affected the charge distribution and furthermore their electrochemical performance.37 Despite the promising electrochemical activity of small molecule quinones in aqueous zinc cells, they exhibit relatively high solubility in electrolytes. It results in the facile loss of active material from the cathode and thus fast capacity decay especially at lower current densities. Current solutions to the dissolution problem include polymerization,38,39 compositing with other materials,40 introducing hydrophobic sites,14etc. Among them, molecular structure regulation would fundamentally solve the related issues. Nevertheless, the possible introduction of dead weight and influence on the redox potential of carbonyl sites should also be taken into consideration for the electrochemical performance of designed cathode molecules.
Balancing the overall requirements of high cycling stability, capacity and redox potential for cathode materials, the hydrophobic methyl group with relatively light weight and a weak electron-donation effect is a desired substitution candidate. Herein, we introduce a methyl group on the 1,4-naphthoquinone (NQ) molecule to form a menaquinone (Me-NQ) cathode for aqueous zinc cells. As expected, the solubility of Me-NQ is effectively reduced to around one-third of NQ, together with only 8% reduction in theoretical capacity and a 0.05 V decrease in average redox voltage. In zinc cells, the low solubility of Me-NQ ensures its facile self-saturation in the electrolyte, as confirmed by in situ UV-vis analysis, which suppresses further active material dissolution from the cathode. Therefore, the capacity of Me-NQ stabilizes at 316 mA h g−1 at 0.1 A g−1, and a good capacity of 146 mA h g−1 is retained after 3500 cycles at 5 A g−1. For NQ, in contrast, the dissolution continues and capacity rapidly drops to 203 mA h g−1 after only 5 cycles at 0.1 A g−1 and 88 mA h g−1 after 3500 cycles at 5 A g−1. Further studies confirm the facilitated reaction kinetics with the Me-NQ cathode and the redox reactions on carbonyl sites together with the co-storage of Zn2+ and protons.
The electrochemical performance of the NQ and Me-NQ cathode materials is studied in aqueous zinc cells. Fig. 2a–c show the capacity and charge–discharge curves at different current densities. Notably, the capacity of the NQ cathode decays rapidly over cycling. It only retains 203 mA h g−1 capacity after 5 cycles at 0.1 A g−1, which further decreases to 175 mA h g−1, 128 mA h g−1, 106 mA h g−1, 93 mA h g−1, 88 mA h g−1, and 85 mA h g−1 with the increase of current density to 0.2 A g−1, 0.5 A g−1, 1 A g−1, 2 A g−1, 3 A g−1 and 5 A g−1, respectively. A low capacity of 175 mA h g−1 is obtained when the current density returns to 0.1 A g−1. With the Me-NQ cathode, in contrast, the capacity stabilizes fast within the initial couple cycles and reaches 316 mA h g−1 after 5 cycles at 0.1 A g−1. It also shows stable evolution at the same current densities during the subsequent rate tests and delivers 287 mA h g−1, 248 mA h g−1, 233 mA h g−1, 224 mA h g−1, 218 mA h g−1 and 213 mA h g−1 at 0.2 A g−1, 0.5 A g−1, 1 A g−1, 2 A g−1, 3 A g−1 and 5 A g−1, respectively. When the current density returns to 0.1 A g−1, the capacity retrieves to 304 mA h g−1. The self-discharge behaviors of the two cathodes are measured by resting the cells at the charged state for 48 h (Fig. 2d and e). The NQ cathode only preserves 75.3% capacity after the rest period, and it largely increases to 97.3% with the introduction of the methyl group. Meanwhile, thanks to the weak electron-donation effect of the methyl group, the Me-NQ cathode only exhibits slightly reduced average redox voltage at 0.83 V in comparison to 0.88 V for NQ (Fig. S3†). As a result, the energy densities of the Me-NQ cathode largely exceed those for NQ as compared in the Ragone plots in Fig. 2f.
The long-term cycling performance of the two cathodes is studied (Fig. 2g and S4†). The NQ cathode exhibits rapid capacity decays especially during the early stages of cycling. As a result, it only retains 153 mA h g−1 capacity after 100 cycles at a low current density of 0.2 A g−1 and 88 mA h g−1 after 3500 cycles at a high current density of 5 A g−1. In contrast, much more stable cycling performance is obtained with the Me-NQ cathode. It preserves a good capacity of 236 mA h g−1 after 100 cycles at 0.2 A g−1 and 146 mA h g−1 after 3500 cycles at 5 A g−1. This performance is comparable with that of previously reported organic cathode materials (Table S1†). Importantly, the Fourier-transform infrared (FT-IR) spectrum and SEM image of the Me-NQ cathode after long-term cycling suggest the preservation of characteristic vibration peaks and morphology (Fig. S5†). This confirms its excellent structural and morphological stability during electrochemical processes.
The above electrochemical tests demonstrate the superior cycling stability of the Me-NQ cathode in comparison to NQ. In order to reveal the underlying mechanism, in situ UV-vis spectroscopy is carried out for the electrolytes with Me-NQ and NQ cathodes, respectively. As shown in Fig. 3a, an absorption band is noted at around 250 nm at the beginning of the discharge process of the NQ cathode. It keeps increasing together with the appearance and growth of another band at 330 nm upon the subsequent discharge and charge. They are attributed to the continuously dissolved active material from the cathode. Interestingly, the spectra of the electrolyte with the Me-NQ cathode show similar evolutions of absorption bands, though with much weaker peak intensities (Fig. 3b). This suggests that the Me-NQ active material also experiences dissolution at the beginning of cycling in zinc cells; nevertheless, the process is effectively suppressed thanks to its much lower solubility than NQ. In situ UV-vis analysis is further carried out for the two electrolytes with the cathodes cycled for 4 cycles. As shown in Fig. 3c, the peak intensity with the NQ cathode keeps increasing upon cycling and reaches the highest absorbance of 1.3 after 4 cycles. This suggests the severe and continuous dissolution of the active material, which results in fast capacity decay. In the electrolyte with the Me-NQ cathode, in contrast, the peak intensity quickly reaches a maximum absorbance of 0.33 after only two cycles and remains stable (Fig. 3d). This corresponds to the facile saturation of the active material in the electrolyte and equilibrium of the dissolution process. This self-saturation behavior inhibits the subsequent continuous dissolution. Therefore, the majority of the active material stays on the cathode and delivers electrochemical activity, ensuring the stabilization of capacity in zinc cells.
To further demonstrate this self-saturation behavior, the active material loading is increased to 4 mg cm−2, which reduces the percentage of dissolved parts with respect to the total material. The cycling test is carried out at 0.2 A g−1. As shown in Fig. 3e and f, the NQ cathode still experiences facile capacity decay within the first 10 cycles and retains only 79 mA h g−1 capacity after 100 cycles. For the Me-NQ cathode, in contrast, a much more stable capacity retention is realized. It preserves a much higher capacity of 258 mA h g−1 after 100 cycles together with excellent preservation of charge–discharge curve shapes. This is also higher than 236 mA h g−1 with about 1.1 mg cm−2 active material loading, confirming the self-saturation mechanism.
The reaction kinetics of the two cathode materials in zinc cells are further studied. Electrochemical impedance spectroscopy (EIS) is carried out, and the Nyquist plots are fitted with the typical equivalent circuit shown in the inset (Fig. 4a). This results in a charge transfer resistance (Rct) of 317 Ω for NQ, which is reduced to 185 Ω for Me-NQ. This suggests the facilitated reaction kinetics for Me-NQ. Galvanostatic intermittent titration technique (GITT) measurements are carried out to study the cation diffusion behaviors (Fig. 4b and c). The diffusion coefficient of NQ starts at the order of 10−10 cm2 s−1 at the beginning of discharge, which decays and drops to the order of 10−12 cm2 s−1 during the final stage of discharge. It evolves within the orders of 10−10 to 10−11 cm2 s−1 during charge. For Me-NQ, the diffusion coefficients are mostly in the order of 10−10 cm2 s−1 during both the discharge and charge processes. This confirms the faster cation diffusion in Me-NQ in comparison to NQ. Cyclic voltammetry (CV) tests are further carried out to separate the charge storage processes (Fig. 4d and e). The non-diffusion and diffusion-controlled percentages are calculated according to the equation i = k1v + k2v1/2 and the results are summarized in Fig. 4f. With the NQ cathode, the non-diffusion-controlled process contributes to 53% of overall capacity at 1.0 mV s−1, which increases to 57%, 60%, 63%, 65% and 68% with the increase of scan rates to 1.2, 1.4, 1.6, 1.8 and 2.0 mV s−1, respectively. In comparison, higher percentages of 66%, 69%, 71%, 73%, 74% and 76% are obtained with the Me-NQ cathode at the corresponding scan rates. The above results confirm the facilitated reaction kinetics of Me-NQ compared to NQ. Therefore, the Me-NQ cathode delivers higher practical capacity at various current densities than NQ despite the slightly lower theoretical capacity of the former. Besides, the lower solubility and faster self-saturation process of Me-NQ also retain the majority of the active material at the cathode and ensure higher capacity.
The energy storage mechanism of the Me-NQ cathode in aqueous zinc cells is investigated. The redox active sites are studied by FT-IR. As shown in Fig. 5a, the characteristic peak of CO at 1664 cm−1 disappears during discharge and appears upon charge (green region). Simultaneously, the peak at 1103 cm−1, which is attributed to the C–O bond (pink region),43,44 also shows reversible appearance and disappearance upon discharge and charge. This suggests the reduction of CO on Me-NQ to C–O during discharge and reversible oxidation during subsequent charge. In accordance, the molecular electrostatic potential (ESP) map of Me-NQ in Fig. 5b shows the most negative regions at carbonyl groups, confirming them as the electrochemically active sites.
The cations stored in Me-NQ are studied by testing the cathodes in different electrolytes of ZnSO4 with pH = 4, H2SO4 with pH = 1, and ZnSO4 with pH adjusted to 1 using H2SO4. Fig. 5c compares the CV curves at 0.1 mV s−1. Interestingly, the two ZnSO4 electrolytes with different pH values show close peak positions. This suggests that the Zn2+ cation is favorable. Meanwhile, in the Zn2+ free H2SO4 electrolyte, a pair of redox peaks is observed at higher potential in comparison to the ones in zinc electrolytes. It demonstrates the proton storage capability of Me-NQ. The peak shift is attributed to the Nernst shift as well as the different reaction potentials with Zn2+ and H+ cations. Fig. 5d compares the XRD patterns of the cathode at different states. The Me-NQ active material exhibits poor crystallinity, and diffractions from KB, PTFE and filter paper adhered to the cathode are observed in the charged electrode. At the end of discharge, extra peaks attributing to Zn4SO4(OH)6·3/4H2O show up. In the SEM image of the discharged electrode, flakes are found over the active material particles (Fig. 5e), which is the typical morphology of zinc basic salts. They disappear upon charge. Their formation/dissolution result from the pH decrease/increase in the cathode region,35 which is further attributed to the reversible proton insertion/de-insertion in Me-NQ. Besides, the energy dispersive X-ray spectroscopy (EDS) mappings of the discharged cathode show the homogeneous distributions of C, O, S and Zn elements (Fig. 5f). The Zn signal results from zinc basic salts as well as Zn stored in the cathode. Density functional theory (DFT) calculations of Me-NQ binding with H+ or Zn2+ provide further insights into the cation storage behavior. As summarized in Fig. 5g, the binding becomes stronger with more proton interactions in comparison to Zn2+. This suggests an easier proton insertion in Me-NQ. Nevertheless, since the concentration of Zn2+ in the ZnSO4 electrolyte is much higher than that of H+, Zn2+ cations are also involved in the energy storage process. Overall, the above analysis confirms the co-storage of Zn2+ and H+ on the carbonyl sites of the Me-NQ cathode with excellent reversibility in aqueous zinc cells. Notably, characterization studies also reveal the reversible change of the carbonyl group and cation co-storage capability with the NQ cathode (Fig. S7†), demonstrating that the reaction mechanism is not affected with the introduction of the methyl group.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4sc04685d |
This journal is © The Royal Society of Chemistry 2024 |