The Zn dendrites and other side reactions (hydrogen evolution, corrosion) are closely related to EEI. Interfacial engineering, such as electrolyte additives, can effectively inhibit the Zn dendrites and other side reactions by regulating the Zn deposition behavior37,38. Specifically, during the Zn2+ plating/stripping process, as shown in Fig. 1a. Unfortunately, Zn2+ in the electrolyte often exists in the form of hydrated Zn2+, with each Zn2+ surrounded by six combining H2O. Therefore, the solvated [Zn(H2O)6]2+ must consume a certain amount of energy to de-solvate and release a large amount of solvated H2O to form Zn2+, and the released solvated H2O forms active free H2O. The H–O bonds of active free H2O are relatively weak, thus leading to the deprotonation of H2O, promoting the decomposition of active free H2O and producing H2. Subsequently, the H+ is continuously consumed, increasing the local pH value, which Zn2+ and [Zn(H2O)6]2+ on the Zn anode surface are more likely to participate in side reactions, forming low conductivity by-products (ZHS). These low conductivity by-products grow recklessly on the Zn anode surface, making the surface rough and changing the initial uniform distribution of current, triggering the growth of Zn dendrites16,39. In this situation, it is possible to regulate the structure of [Zn(H2O)6]2+, reduce the amount of active free H2O, and achieve a dendrite-free zinc ion battery with long-term electrochemical stability40.Fig. 1: Schematics of the Zn2+ plating/stripping behaviors.a The Zn anode surface in ZS. b The Zn anode surface in NDs-DMSO-ZS.Therefore, we attempted to introduce NDs-DMSO additive into 2 M ZnSO4 electrolyte. As shown in Fig. 1b, this additive plays a crucial role in inhibiting Zn dendrites and other side reactions. Due to the organic functional group (S = O), DMSO can act as a hydrogen bonding acceptor (HB acceptor) to form strong coordination interactions with H2O, reducing the amount of active free H2O15. In addition, DMSO can also participate in the [Zn(H2O)6]2+ structure and optimize the Zn2+ solvation sheath, which can regulate the Zn deposition behavior and inhibit the growth of Zn dendrites and other side reactions17. However, as a highly polar organic additive, DMSO has a strong adsorption capacity for Zn2+, leading to severe electrode polarization and cell failure26. NDs terminated by various oxygen-containing surface functional groups have shown enormous potential in energy storage applications. The surface chemical properties of NDs were characterized by FTIR, as shown in Fig. S1. Several strong absorption peaks at approximately 1089, 1630/1758, and 3433 cm−1, correspond to the stretching vibrations of C-O, COO-, and O-H bonds29,41,42. The FTIR results indicate the existence of many -OH and -COOH functional groups on the NDs’ surface. Due to the presence of surface functional groups like -OH, -COOH, and C = O, NDs can act as both hydrogen bonding donor (HB donor) and hydrogen bonding acceptor (HB acceptor), which NDs can form HBs with DMSO to better assist in Zn deposition and reduce electrode polarization34,36. Specifically, during the Zn2+ de-solvation process, the [Zn(H2O)mDMSOn]2+ must decouple and release DMSO and water molecules to form Zn2+. However, DMSO has a strong adsorption capacity for Zn2+, so the de-solvation process requires more energy consumption, resulting in slower reaction kinetics and increased electrode polarization. Fortunately, OH and -COOH on the surface of NDs as HB donor groups and the S = O groups of DMSO as HB acceptor groups to form the HBs, and DMSO molecules can quickly detach from the [Zn(H2O)mDMSOn]2+ structure under the action of HBs. The strong intermolecular HBs decrease unnecessary energy consumption during the de-solvation process, improve reaction kinetics, and reduce electrode polarization. As an electrolyte additive, NDs-DMSO optimizes the electrolyte, regulates the Zn deposition behavior, and achieves a dense and uniform Zn anode surface.Scanning electron microscopy (SEM) was performed to clarify the morphology evolution of the Zn anode surface in Zn | |Zn symmetric cells with different electrolytes and cycles. As shown in Fig. 2a, after cycling for 2 hours under a current density of 10 mA cm−2, the Zn anode surface using ZS is covered by numerous plate-like protrusions, which indicate uneven Zn deposition behavior. During the charging-discharging process, these plate-like Zn dendrites continuously grow along the separator. Eventually, the Zn dendrites may penetrate the separator to cause short-circuiting and separate from the Zn anode surface to form so-called “dead Zn”15. The corresponding energy-dispersive X-ray spectroscopy (EDS) results indicate a higher content of the O and S elements (Fig. 2a1–2a3, Fig. S2a), further confirming the formation of the by-products (ZHS). More importantly, the by-products of low conductivity usually form a passivation layer on the Zn anode surface, thereby reducing the Zn utilization and affecting the Zn deposition kinetics43. By contrast, the Zn anode surface using NDs-DMSO-ZS doesn’t exhibit particularly prominent plate-like Zn dendrites during the cycle process (Fig. 2b), suggesting relatively steady Zn deposition behavior. The EDS results also show that there is uniform Zn elements distribution and fewer by-products. (Fig. 2b1–2b3, Fig. S2b). Fig. S3 further investigates the ZHS of these different electrolytes using the X-ray diffraction (XRD) tests. During the discharge phase, the HER causes an increase in OH- concentration, promoting the production of ZHS at EEI. During the charging phase, the disappearance of ZHS indicates good reversibility of the cathode materials44,45,46. For Mn‐based cathode material, the formation of ZHS in the above process can be summarized as follows47:$$4{Zn}^{2+}+6{{OH}}^{-}+4{{SO}}_{4}^{2-}+x{H}_{2}O\leftrightarrow {{Zn}}_{4}{({OH})}_{6}{{SO}}_{4}\cdot x{H}_{2}O$$
(1)
Fig. 2: Morphology change of the Zn anode surface after 2 h cycling at 10 mA cm−2 for 10 mAh cm−2.a, a1–a3 SEM images and the corresponding EDS elemental maps of the Zn anode surface in ZS. b, b1–b3 SEM images and the corresponding EDS elemental maps of the Zn anode surface in NDs-DMSO-ZS. The SEM images of the Zn anode surface after 20 h cycling at 1 mA cm−2 for 1 mAh cm−2 in (c) ZS, (d) DMSO-ZS, and (e) NDs-DMSO-ZS.At the first full discharge (0.8 V), as shown in Fig. S3a, ZHS was formed on the cathode of the three electrolytes, but the diffraction peak intensity of NDs-DMSO-ZS was the weakest, indicating a lower content of ZHS (approximately located at 8.2°). At the first full charge (1.9 V), as shown in Fig. S3b, only the diffraction peak intensity of NDs-DMSO-ZS significantly decreased. When subjected to the second discharge (1.6 V), as shown in Fig. S3c, the diffraction peak intensity of NDs-DMSO-ZS increased but remained the weakest. By comparing different potentials, the cathode surface of DMSO-ZS and ZS accumulated a large number of ZHS. The intensity change of the ZHS diffraction peak of NDs-DMSO-ZS indicates that it has good cycling stability. In addition, Zn | |Zn symmetric cells were assembled using three different electrolytes and tested under a current density of 1 mA cm−2. After cycling, the morphology of their Zn anodes was characterized by SEM, as shown in Fig. 2c, d, e. The SEM results show that after 10 cycles of the cells, the Zn anode surface using ZS is covered with disordered and irregular dendrites, while the DMSO-ZS is also covered with a small number of dendrites, indicating that uneven Zn deposition leads to the formation of dendrites, which seriously affects the electrochemical performance and ultimately makes the cell failure48,49. In contrast, the surface of the Zn anode of NDs-DMSO-ZS is smooth, indicating that the Zn2+ plating/stripping process is more uniform during the cycle process, which is conducive to the long-term cycle stability of the cell.To investigate the influence of NDs-DMSO on the Zn deposition kinetics, we subsequently performed electrochemical performance tests on different cells. As shown in Fig. 3a, the rate performance of Zn | |Zn symmetric cells with a fixed areal capacity of 1 mAh cm−2. Compared with the cells using ZS alone, the symmetric cells using DMSO-ZS exhibit higher overpotential. When DMSO participates in the Zn2+ solvation sheath, the hydrated Zn2+ exists in the form of [Zn(H2O)m(DMSO)n]2+, and the radius of Zn2+ solvation sheath increases, which may lead to the slower Zn2+ diffusion rate in the electrolyte, resulting in significantly increased overpotential. It has been reported that proper overpotential can suppress HER, but too high overpotential will affect Zn deposition kinetics15,50. However, the introduction of NDs slightly reduces the overpotential. The reason may be that NDs provide abundant active sites (-OH, -COOH, C = O) for Zn deposition, especially the surface -COOH groups that can strongly attract cationic Zn2+, increasing the concentration of Zn2+ at EEI, which is beneficial to reducing overpotential51,52. In addition, due to NDs can provide more zoophilic nucleation sites for the Zn2+ plating/stripping, which is also favorable for reaction kinetics53. For a more comprehensive assessment of the Zn deposition kinetics, the exchange current density is calculated by Eq. (2)54:$${{\rm{i}}} \, \approx \, {i}_{0}\frac{F}{{RT}}\frac{\eta }{2}$$
(2)
where \(i\) is the current density, \({i}_{0}\) is the exchange current density, \(\eta\) is the total overpotential, \(F\) is the Faraday constant, \(R\) is the gas constant, and \(T\) is the absolute temperature. As shown in Fig. 3b, DMSO is added to the electrolyte only, and the exchange current density value is the smallest (7.0 mA cm−2), indicating relatively slow Zn deposition kinetics. Simultaneously, NDs slightly increase the exchange current density (7.6 mA cm−2), which is conducive to improving the Zn deposition kinetics53. To study the initial stage of the Zn deposition behavior, the Zn//Ti asymmetric cells were characterized by cyclic voltammetry (CV). In the initial stages of Zn nucleation and growth, the nucleation overpotential (NOP) serves as a parameter to explain the extent of electrode polarization. Generally, the relationship between the critical Zn nucleus radius (\({r}_{{crit}}\)) and the NOP (\({{\rm{\eta }}}\)) is by Eq. (3)55:$${r}_{{crit}}=2\frac{\gamma {V}_{m}}{F{{\rm{|}}}\eta {{\rm{|}}}}$$
(3)
Here, \({{\rm{\gamma }}}\) represents the surface energy at the electrode-electrolyte interface, \({V}_{m}\) is the Zn molar volume, and \({{\rm{F}}}\) is the Faraday’s constant. According to Eq. (3), \({r}_{{crit}}\) shows a clear inverse relationship with \({{\rm{\eta }}}\)56. In the ZS, a low NOP (59.41 mV) indicates that the initial Zn nuclei are large, which is not beneficial to uniform Zn deposition, leading to the growth of Zn dendrites. In contrast, DMSO can promote Zn nucleus fineness, which is beneficial to achieving dense and uniform Zn deposition behavior (Fig. 3c)57. After adding NDs, NOP decreased slightly, which may be due to the regulation of HBs. Under the action of hydrogen bonding force, DMSO and H2O in [Zn(H2O)m(DMSO)n]2+ smoothly leave the solvation sheath, reducing energy consumption and lowering the Zn2+ nucleation barrier during the Zn2+ de-solvation process, increasing kinetics. Subsequently, we tested the charge-transfer resistance (Rct) of Zn | |Zn symmetric cells at different temperatures and calculated the de-solvation activation energy of different electrolytes by the Arrhenius equation. The Rct at different temperatures (45-65 曟) can be obtained from the EIS plots (Fig. S4). Generally speaking, a lower Ea value is beneficial for promoting the de-solvation process of hydrated Zn2+48. As shown in Fig. 3d, the Zn | |Zn symmetric cells using NDs-DMSO-ZS have the smallest Ea value (43.86 KJ mol−1), indicating that the NDs-DMSO additive has a positive effect on the kinetics during Zn plating. This can be attributed to HBs, which significantly reduce the activation energy barrier of Zn2+ de-solvation under the action of hydrogen bonding force. DMSO and H2O in [Zn(H2O)m(DMSO)n]2+ smoothly leave the solvation sheath, reducing energy consumption and promoting the reduction of Zn2+ to Zn58.Fig. 3: The electrochemical performance of Zn||Zn symmetric cells and Zn//Ti asymmetric cells in different electrolytes.a Rate performances of the Zn||Zn symmetric cells in different current densities with a fixed capacity of 1 mAh cm−2. b Exchange current density from curves in (a). c Cyclic voltammograms (CV) for Zn nucleation on bare Ti foil in different current densities under the scan rate of 1 mV s−1. d Arrhenius curves of ZS, DMSO-ZS, and NDs-DMSO-ZS. e Tafel test of the Zn||Zn symmetric cells in different electrolytes. f Chronoamperograms (CAs) at an overpotential of −150 mV.Additionally, H2O in the Zn2+ solvation sheath can trigger various side reactions, including hydrogen evolution and corrosion. The linear sweep voltammetry (LSV) test results of Zn//Ti asymmetric cells indicate a shift towards the negative direction in the hydrogen evolution potential of NDs-DMSO-ZS (Fig. S5), which may be due to the surface groups (-COOH) increases the overpotential of H+ reduction, inhibiting HER21,52. As shown in Fig. 3e, the corrosion current densities for three different electrolytes were determined through Tafel fitting: NDs-DMSO-ZS (0.145 mA cm−2), DMSO-ZS (0.339 mA cm−2), and pure ZS (1.905 mA cm−2). The results show that the corrosion current density of NDs-DMSO-ZS is the smallest, meaning a lower corrosion reaction rate on its Zn anode surface. The lower corrosion reaction rate indicates that the Zn anode has good anti-corrosion, which helps to suppress the formation of Zn dendrites59. In particular, due to the various surface functional groups on NDs, they also interact with H2O through the HBs while interacting with DMSO. This interaction also reduces the quantity of active free H2O on the surface of the Zn anode, thereby enhancing anti-corrosion21,22. Finally, as shown in Fig. 3f, the working mechanism of regulating Zn deposition behavior is studied by chronoamperometry (CA) testing. At an overpotential of −150 mV, the current density of the cells using ZS exhibits a linear increasing trend throughout the entire testing period. The Zn2+ is influenced by the tip effect, leading to their preferential accumulation at the tips, ultimately evolving into Zn dendrites, indicating that Zn2+ undergoes uncontrolled diffusion along the two-dimensional (2D) direction. In the cells using DMSO-ZS and NDs-DMSO-ZS, their current densities instantaneously increase within the first 20 to 80 s of the process, indicating a rapid 2D diffusion process at the initiation of Zn deposition. Subsequently, the current densities tend to stabilize, suggesting that the Zn2+ diffusion has shifted to a three-dimensional (3D) phase. In general, the Zn nucleus can form rapidly in the initial stages of Zn nucleation, and the stability of Zn deposition kinetics will not be affected by the nanoparticles (NDs)46,60.Next, we tested the CE of Zn//Cu asymmetric cells, as illustrated in Fig. 4a. The Zn//Cu asymmetric cells using DMSO-ZS and NDs-DMSO-ZS exhibit stable initial cycling, with no significant fluctuations observed. As the cycle progresses, the Zn//Cu asymmetric cells using DMSO-ZS maintain less than 200 cycles, and the CE curve indicates cell failure59. By contrast, the Zn//Cu asymmetric cells introducing NDs-DMSO provide more than 800 effective cycles, maintaining a high cycling stability (99.8%). Benefiting from the rich surface functional groups, the interaction between DMSO and NDs improves the Zn2+ diffusion rate; NDs also interact with H2O to enhance anti-corrosion. Therefore, NDs are very important for the long-term cycling stability of the cells. Figure 4b illustrates the initial CE of Zn//Cu asymmetric cells using ZS, DMSO-ZS, and NDs-DMSO-ZS, which are 93.7%, 98.3%, and 98.9%, respectively6,15. To further investigate the influence of NDs-DMSO on the long-term stability and reversibility of AZIBs, we conducted repeated charge-discharge tests on the Zn | |Zn symmetric cells. As shown in Fig. 4d, the Zn | |Zn symmetric cells using NDs-DMSO-ZS exhibit impressive cycling stability, surpassing 1500 hours at a current density of 1 mA cm−2. By contrast, the Zn | |Zn symmetric cells using ZS and DMSO-ZS can only sustain cycles for 342 hours and 733 hours, respectively. As expected, as the cycle progresses, the cells using DMSO-ZS exhibit significant electrode polarization, ultimately leading to cell failure. This may be due to the shorter chain lengths and smaller end steric hindrances of DMSO molecules, which facilitate the formation of dense structures and coordination interaction organic interface layer at EEI. The organic interface layer will slow down the reduction reaction kinetics of Zn2+, hinder the Zn2+ de-solvation process, and lead to intensified interface polarization20,61,62,63. In general, electrode polarization is more prominent at high current density, but the current density is increased to 3 mA cm−2 (corresponding to a surface capacity of 3 mAh cm−2), as shown in Fig. 4e. The Zn | |Zn symmetric cells using NDs-DMSO-ZS show cycling stability (800 h), which is an exciting result. After the cells failed, we disassembled them and characterized the morphology of the Zn anode surface (Fig. 4c, Fig. S6). In stark contrast, the Zn | |Zn symmetrical cells using NDs-DMSO-ZS exhibit a clear hexagonal structure growing outward on the Zn anode surface. As reported, outward growth deposition can prevent the formation of Zn dendrites, effectively prolonging the lifetime of the Zn2+ plating/stripping64. Under the synergistic effect of DMSO and NDs, Zn dendrites and other side reactions are effectively suppressed, reducing the effects caused by electrode polarization and promoting the long-term cycling stability of the cells65,66,67.Fig. 4: The performance of Zn//Cu asymmetric cells and Zn||Zn symmetric cells in different electrolytes.a Coulombic efficiency (CE) of the Zn//Cu asymmetric cells in different electrolytes. b Galvanostatic Charge-Discharge (GCD) profiles of the Zn//Cu asymmetric cells in different electrolytes. c SEM of the Zn anode after the cells failure at 3 mA cm−2 for 3 mAh cm−2 in NDs-DMSO-ZS. Cycling performance of Zn||Zn symmetric cells in different electrolytes at (d) 1 mA cm−2 for 1 mAh cm−2 and (e) 3 mA cm−2 for 3 mAh cm−2.To demonstrate the feasibility of the NDs-DMSO additive in a practical battery system, the Zn//MnO2 full cells were assembled using different electrolytes. As shown in Fig. 5a, cyclic voltammetry (CV) testing results indicate that the Zn//MnO2 full cells using NDs-DMSO-ZS demonstrate two noticeable redox peaks at a scanning rate of 0.1 mV s−1. The redox peaks gradually broaden as the scan rate increases from 0.1 mV s−1 to 1 mV s−1. Nevertheless, the overall shape of the CV curves remains consistent throughout this process. Assuming that the relationship between the current and the scan rate according to the following equation68:$${{\rm{i}}}={{\rm{a}}}{\nu }^{b}$$
(4)
it can be rewritten as$$\log \left({{\rm{i}}}\right)={{\rm{blog}}}\left({{\rm{\nu }}}\right)+\log \left({{\rm{a}}}\right)$$
(5)
where \(i\) refers to current, \(v\) stands for scan rate, and the corresponding adjustable parameters are \(a\) and \(b\). Through fitting \(\log (i)\) versus \(\log (v)\), the coefficient \(b\) for peaks 1 and 4 can be determined based on the slope of the linear regression lines. The calculated \(b\) values for peaks 1 and 4 are theoretically 0.76 and 0.77, respectively (Fig. 5b). When the parameter “\(b\)” exceeds 0.5, it indicates that the pseudocapacitive redox reaction plays a predominant role throughout the whole storage process, which determines the high rate performance of the cells68,69. Compared to the cells using NDs-DMSO-ZS, the others exhibit relatively small “\(b\)” values, indicating a lower contribution from their pseudocapacitive behavior(Fig. S7). Additionally, as shown in Fig. 5c, the charge transfer kinetics at EEI was studied by electrochemical impedance spectroscopy (EIS). The semicircle in the high-frequency region represents the interfacial charge transfer resistance (Rct), while the straight line in the low-frequency region represents the diffusion resistance70. The results show that the Zn//MnO2 full cells using NDs-DMSO-ZS exhibit larger Rct before the first charge-discharge cycle (see insets for EIS graphs). It may be that NDs play a certain degree of particle obstruction in the electrolyte, resulting in a large impedance at EEI. After the first cycle, the Rct of the Zn//MnO2 full cells using NDs-DMSO-ZS is smaller than the DMSO-ZS. This phenomenon reflects the vital function of NDs surface functional groups, which promote the diffusion and charge transfer of Zn2+ at EEI37,38.Fig. 5: The electrochemical performance of Zn//MnO2 full cells in different electrolytes.a CV curves of the Zn//MnO2 full cells of NDs-DMSO-ZS with the scan rate from 0.1 to 1.0 mV s−1. b Log (i) vs log (v) plots of two peaks in CV curves. c EIS curves of the Zn//MnO2 full cells in different electrolytes. d Rate performance of the Zn//MnO2 full cells from 0.1 to 3 A g−1 in different electrolytes. e Charge-discharge profiles of rate performance in the NDs-DMSO-ZS. f Cyclic stabilities and efficiencies of the Zn//MnO2 full cells in different electrolytes at a current density of 1 A g−1.In addition, we also performed rate performance tests on the Zn//MnO2 full cells using different electrolytes (Fig. 5d). The Zn//MnO2 full cells using NDs-DMSO-ZS exhibit excellent rate performance, exhibiting high reversible specific capacities of 272.3, 206.6, 152.1, 120.4, 97.7, and 78.1 mAh g−1 at current densities of 0.1, 0.2, 0.5, 1, 2, and 3 A g−1, respectively. It can be attributed to the fast reaction kinetics and remarkable Zn2+ plating/stripping behavior, which maintained excellent rate performance. In contrast, the reversible specific capacities of the Zn//MnO2 full cells using DMSO-ZS and ZS are relatively lower, showing notably poorer rate performance at different current densities71. Additionally, it is noteworthy that during the constant current charge-discharge process shown in Fig. 5e and Fig. S8, all Zn//MnO2 full cells display two distinct sloping charge-discharge plateau curves. These plateau curves correspond to the insertion and extraction of Zn2+, showing good agreement with the CV curve results and remaining consistent with previous research8. Finally, as shown in Fig. 5f, the Zn//MnO2 full cells using NDs-DMSO-ZS demonstrate a sustained reversible capacity of 80.8 mAh g−1 over more than 2000 cycles at a current density of 1 A g−1 (with a mass loading of 2.1 mg cm−2). The outstanding performance of the cells can be primarily attributed to the synergistic effect of DMSO and NDs, working together to inhibit the growth of Zn dendrites and suppress the occurrence of other side reactions. This study also highlights the substantial potential of NDs in the long-term cycling stability of AZIBs.
A dual-functional electrolyte additive displaying hydrogen bond fusion enables highly reversible aqueous zinc ion batteries
