Reconstructed NiCo Alloy Enables High-Rate Ni-Zn Microbattery with High Capacity

Abstract

Miniaturized powering devices with both sufficient capacity as well as fast charging capability are anticipated to support microelectronics with multi-functions. However, most reported miniaturized energy storage devices only display limited performances around capacity or rate performance, and it remains challenging to develop high-rate microdevices with large capacities. Herein, a reconstructed NiCo alloy is proposed as a promising microcathode for a Ni-Zn microbattery with a high-rate performance and large capacity. With the reconstructed layer compactly adhered on the metal substrate, the activated NiCo alloy demonstrates an excellent conductivity close to metals. Meanwhile, the abundant alloying defect contributes to a relatively higher reconstruction depth up to 20 nm. Both the superior electron transport and the higher reaction depth facilitate the simultaneous excellent performance in the reaction rate and capacity. As a consequence, the microcathode achieves a large capacity up to 1.51 mAh cm⁻², as well as an excellent rate performance with a capacity retention of 82.9% when the current density is expanded to 100 mA cm⁻². More surprisingly, such excellent performance can shift towards the full Ni-Zn microbattery, and the fast-charging capability based on large capacity can stably maintain 7000 cycles. This unique strategy of reconstructed NiCo alloy microcathode provides a new direction for the construction of high-performance output units.

1. Introduction

With the boom of the Internet of Things (IoT) in the fields of integrated systems and flexible electronics, there forms a growing demand for high-performance miniaturized energy storage devices, including microbatteries (MBs) and microsupercapacitors (MSCs) [1,2,3,4,5]. Typically, MSCs are featured with having the merits of a long cycling durability and a high-rate capability, while also having the drawbacks of self-discharge issues and a low energy supply, which severely restrict their practical applications. To the contrary, MBs with high energy densities could well maintain a stable voltage output for a long period in powering advanced miniaturized devices [6,7,8,9,10]. Among the ever-developed MBs, aqueous Ni-Zn MBs hold the most application anticipation for their relatively high voltage output (~1.8 V), fast reaction kinetics, and abundant resources. There have been reports about Ni-Zn batteries with high energy densities through dense Ni cathodes [11] and 3D Zn anodes [12]. However, the development of Ni-Zn MBs is neither sufficient in the capacity supply nor in the reaction rate.

To address the above issues of Ni-Zn MBs associated with capacity supply and reaction rate, many reports have proposed powder techniques with various structures and composition designs, or deposition techniques with active materials on current collectors [13,14,15]. However, these traditional techniques suffer from the limited reaction transport when a thick electrode is applied [3]. Generally, when more active materials are constructed in a given electrode, it is hard to guarantee that the reactivity is always maintained at a high level. This effectiveness decline could be attributed to the accompanied higher reaction impedance and longer diffusion distance, thus resulting in a worse power performance when more energy is provided. Additionally, even though the design of the thin-layer reaction with a fine structure is effective at reactivity [16], the contact resistance of active materials and current collectors cannot be ignored if a high-rate current is applied. Recently, there have been reports about the concept of reconstruction based on the nickel element in catalysis, which could effectively provide a high performance of catalyzers. Generally, as an activation strategy, reconstruction refers to the in situ transformation of pre-fabricated materials through oxidization/reduction reactions under alkaline electrochemical environments [17,18,19]. This approach is essentially a rearrangement of the atoms on the atomic layer of the crystal surface, leading to a change in the two-dimensional structure of the surface layer, and the utilizing depth of the substrate is very shallow (~10 nm). A reconstruction strategy can also be used as a reference for material activation in the field of energy storage. For example, in our previous work, we firstly adopted the strategy of reconstruction to obtain in situ active Ni(OH)₂/NiOOH on a nanoporous Ni substrate for high-rate Ni-Zn MB [20]. The reconstructed nickel microcathode was featured as having an ultrahigh reactivity as well as an excellent rate performance. However, an inadequate Ni deposition and a shallow reconstruction layer correspondingly led to low production of active materials, which cannot provide satisfactory capacity. To this end, how to construct a high-rate Ni-Zn MB with a large capacity remains challenging.

Herein, a NiCo alloy was proposed as the pre-fabricated material for reconstruction, and the constructed Ni-Zn MB delivered both a high-rate performance and large capacity. Initially, we co-deposited the NiCo alloy hierarchical porous structure by virtue of a bubble template. The NiCo alloy skeleton is larger and more porous than that of pure nickel, demonstrating an enhanced surface area for reconstruction activation. Then, during the reconstruction process in aqueous alkali, the abundant alloying defect contributes to a relatively higher reconstruction depth of up to 20 nm. Therefore, more Ni(OH)₂/NiOOH components are produced while the superior electron transport is still well maintained. Meanwhile, benefiting from the naturally formed double hydroxide in the reconstructed layer, the structure deformation during the proton insertion/extraction is greatly eased, resulting in an enhanced cycling durability. As a consequence, both high-rate performance and large capacity are ensured for the reconstructed microcathode on the premise of electrochemical stability. The reaction capacity is realized to 1.51 mAh cm⁻², and the current density can be expanded up to 100 mA cm⁻². Lastly, the cycling retention is more than 80% in 7000 cycles. The final fabricated Ni-Zn MB inherits this excellent electrochemical performance with a slightly decreased capacity of 1.39 mAh cm⁻², an enhanced rate current of 200 mA cm⁻², and an almost maintained cycling durability in 7000 cycles.

2. Materials and Methods

2.1. Preparation of Reconstructed NiCo Alloy Microcathode

The CHI 760E electrochemical workstation (CH Instruments, Shanghai, China) was utilized for the electrochemical operation. In brief, the customized miniaturized nickel plate (effective area: 0.2 cm × 0.5 cm) served as the working electrode with a saturated calomel electrode as the reference electrode, and a Pt plate as the counter electrode. The deposition process for the NiCo alloy was conducted at a cathodic potential of 6 V in the solution of 2 M NaCl, 2 M NH₄Cl, 0.1 M NiCl_2, and x M CoCl₂ (the variable x refers to 0, 0.015, 0.03, and 0.045). After washing the microelectrodes with the deionized water a few times, the mentioned reconstruction processes were switched to an alkaline solution (1 M KOH) with a Hg/HgO electrode as the reference electrode instead. The typical reconstruction method of CV refers to a cyclic sweep within the potential range of 0.2–0.6 V at a scan rate of 10 mV s⁻¹. Furthermore, according to the previous report, the saturated reconstruction cycles are within 750 cycles, and, herein, we followed the procedures.

Safety Statement: During the electrodeposition process, the chlorine and hydrogen will be released. Finally, the experiment is suggested to be conducted in a fume hood.

2.2. Preparation of Zinc Microanode

The zinc microelectrode was obtained via the electrochemical process as well. The same customized miniaturized nickel plate was selected as the working electrode and an ordinary zinc plate worked as the counter electrode. The electrodeposition was conducted at a constant potential of −1.5 V for 30 min in a solution of 6 M ZnO-saturated solution of KOH.

2.3. Assembly of Ni-Zn MB

Firstly, 500 mg of sodium polyacrylate was slowly added into a 20 mL solution of 6 M ZnO-saturated solution of KOH. Then, after stirring for 30 min, the transparent gel electrolyte was formed. Two films tailored from plastic bag were selected as the package materials. Firstly, three edges of the two films were sealed via a sealer machine. Afterward, the reconstructed NiCo alloy microcathode and Zn microanode were put inside the micro pocket and separated via two thin cardboards. After transferring the gel electrolyte into the middle and removing the separated cardboards, the final edge was sealed and the packaged Ni-Zn MB was obtained.

2.4. Structure and Composition Characterizations

The morphologies and microstructural and component characteristics of the samples were measured using FE-SEM (ZEISS Gemini 300, Carl Zeiss AG, Oberkochen, Germany), transmission electron microscopy with a voltage of 200 kV (TEM, F200X, Thermo Fisher Scientific, Waltham, MA, USA). The chemical states and atomic structure information were investigated by XPS (Thermo Scientific K-Alpha, Thermo Fisher Scientific, Waltham, MA, USA).

2.5. Electrochemical Measurements

Cyclic voltammetry (CV) and galvanostatic charge/discharge (GCD) tests were all conducted on a CHI 760E. The single microelectrode was tested in a three-electrode system and the Ni-Zn MB was packed for the test. All the tests related with single NiCo alloy microelectrodes were conducted in 1 M KOH solutions while the Ni-Zn packed MB was in the gel electrolyte of the 6 M ZnO-saturated solution of KOH. The specific capacity (C) of the single microelectrode or assembled MB was calculated by the formulas C = It/A, where I is the discharge current (mA cm⁻²) and A is the geometric area of the microelectrode for the operation.

3. Results and Discussion

We prepared four microelectrodes (denoted as Ni, NiCo0.15, NiCo0.3, and NiCo0.45) by varying the amount of Co²⁺ in the deposition process. As shown in Figure 1, the deposited pure Ni is densely featured with uniform small micro pores (mostly around 5 μm) as a result of the accompanied release of gas. When the Co²⁺ was added, it could be clearly observed that the skeleton of the deposited alloy and the pores became larger, and this trend tended to expand with the increase in added Co²⁺. This is possibly due to the provided alloy deposition environment and the initial formed alloy substrate with a high reactivity [21], which lowered the reaction barrier for both metal depositions and hydrogen evolution. As a result, the deposition of metals as well as the gas release became severe at the same given potential, and a large metallic skeleton with micro pores and cracks was finally obtained. Meanwhile, we also combined energy dispersive spectrometer to analyze the 4 SEM images, and the resulted Ni/Co ratios were 1:0, 1:0.18, 1:0.35, and 1:0.51, respectively. This ratio result reflects that the deposition potential of Co is slightly lower than that of Ni and the ratio of deposited alloy is close to that of the original added ions.

To analyze the composition of the reconstructed layer, the TEM and high-resolution TEM (HRTEM) were investigated. As could be observed in Figure 2a, the contrast in the TEM images revealed that the depth of the activated layer was up to 20 nm, and the subsequent redox reactions were maintained since compact connection with the metallic substrate. The d spacings about the reconstructed lattice fringes were 0.244 and 0.232 nm, corresponding to both (101) planes of Co(OH)₂ (JCPDS No. 001-0357) and Ni(OH)₂ (JCPDS No. 003-0177), indicating that the deposited alloy was partly reconstructed. For the inner layer, the d spacings of 0.177 and 0.204 nm clearly referred to the (200) plane of Ni (JCPDS No. 001-1258) and the (111) plane of Co (JCPDS No. 001-1259), respectively (Figure 2b). We then carried out the XPS measurement to clarify the chemical environment of the reconstructed alloy microelectrode. As shown in Figure 2c, the signals of Ni(OH)₂ (855.75 and 873.3 eV) and NiOOH (856.8 and 874.45 eV) existed in the reconstructed layer of pure Ni [22], indicating that the final reconstructed products contained the nickel hydroxide and its oxidized derivative. Furthermore, affected by the alloying effect, the electron cloud migration left a slight negative shift (approximately 0.4 eV) for the reconstructed NiCo alloy [23]. As shown in Figure 2d, when it comes to the Co 2p signals, both Co(OH)₂ (781 and 796.36 eV) and CoOOH (782.3 and 797.9 eV) could be identified [24], similar to that of nickel. All of the results confirmed the hydroxides product after the reconstruction treatment.

Around the electrochemical performance associated with the enhanced capacity, we compared the discharge curves of four reconstructed microelectrode whose deposition times were set to be 60 s. As shown in Figure 3a, with the increase in the Co content, the related alloy microelectrode displayed a higher capacity. Compared with pure nickel hydroxide, the slight addition of cobalt hydroxide means more active sites and an optimized diffusion polarization, thus increasing the discharge plateau even though more active materials were provided. However, when the Co content further increased to 0.3 or 0.45, the ohmic resistance arising from the increased load of active materials may affect the reaction polarization [25,26], generating a negative effect on the discharge plateau. Here, it should be noted that the increased capacity was not only attributed to the enhanced reconstruction depth, but also from the increased deposition when Co ions were involved in the reduction reactions. Finally, when more metallic structures were loaded via extending the deposition time, the optimal mixture ratio was not NiCo0.45, as its structure would become unstable when deposition time reached 120 s. Instead, the NiCo0.3 ensured a stable deposition time of up to 180 s. The related discharge curves under various current densities and the cycling performance of such microelectrode are shown in Figure 3b,c, respectively. The discharge capacities were 1.51, 1.49, 1.45, 1.37, and 1.25 mAh cm⁻² at the current densities of 5, 10, 20, 50, and 100 mA cm⁻², respectively. To evaluate the stability of the high-rate performance, a cycling test at the current density of 100 mA cm⁻² was conducted for both microelectrodes of NiCo0.3 and pure Ni (180 s). Having benefited from the stabilizing effect of double hydroxides [4,27], the NiCo0.3 microelectrode displayed a capacity retention of 83.36% after 7000 cycles, while that of pure Ni microelectrode dropped to 48.83% after 3000 cycles. These results confirmed that the reconstructed alloy (NiCo0.3) is a promising cathode with a high capacity, high rate, and high cycling durability.

The assembly process of Ni-Zn MB is shown in Figure 4a. To further demonstrate the practical performance of this microcathode in assembled Ni-Zn MBs, CV curves, galvanostatic discharge profiles, and cycling test were conducted. As shown in Figure 4b, there was no obvious change in the CV shape when the scan rate increased from 1 to 10 mV s⁻¹, indicating a good electrochemical reversibility. According to previous research about double hydroxides, the redox peak can be ascribed to the following electrochemical reaction: Zn + CoOOH + NiOOH + 2 KOH + 2 H₂O ⇌ K₂[Zn(OH)₄] + Co(OH)₂ + Ni(OH)₂ [28,29]. For the rate performance of this Ni-Zn MB, the increased electrolyte concentration (6 M) further expanded the rate current to 200 mA cm⁻². The discharge capacities were 1.39, 1.37, 1.36, 1.32, and 1.23 mAh cm⁻² at the current densities of 10, 20, 50, 100, and 200 mA cm⁻², respectively (Figure 4c). Similar to the exploration of the mentioned stability of the fast charge/discharge capability, the long-term cycling durability of the Ni-Zn MB was investigated at 200 mA cm⁻². As observed in Figure 4d, the assembled MB delivered an excellent cycling performance with a capacity retention of 92.0% in 7000 cycles along with a nearly 100% Coulombic efficiency. All these performances clearly indicate the successful assembly of the Ni-Zn MB with the reconstructed NiCo0.3 microcathode, and that the high-rate performance with a large capacity and the cycling durability is advanced in the ever-reported MBs, which is promising for practical applications in microelectronics [8,15,30].

4. Conclusions

In summary, a high-rate Ni-Zn microbattery with a large capacity was realized through the reconstruction approach for the NiCo alloy. By virtue of the bubble template, a highly porous metallic structure was obtained with various Ni-Co ratios in the preparation. When the Co content in the NiCo alloy increased, the higher alloy content facilitated the deposition reactions, leading to an enhanced metal loading and surface area. Additionally, benefiting from the highly active NiCo alloy structure, a deeper reconstruction of up to 20 nm was achieved, resulting in a higher capacity when compared with pure nickel. Meanwhile, the formed reconstructed layer contained double hydroxides which possessed enhanced electrostatic forces between the hydroxides when compared with the pure nickel hydroxide, hence realizing a more stable cycling performance. As a result, the final assembled Ni-Zn MB with reconstructed alloy microelectrode achieved a high capacity of up to 1.39 mAh cm⁻², an ultrahigh rate performance with a capacity retention of 88.5% when the current density increased to 200 mA cm⁻², and an excellent cycling stability with over 90% retention after 7000 fast-charging cycles. This optimization design from the perspective of substrate metal provides an effective strategy for developing MBs with large capacities and high-rate performances simultaneously.