Enhanced Synthesis and Electrochemical Performance of Cobalt Hexacyanoferrate via Protonation‐Regulated Coprecipitation

ABSTRACT

Metal hexacyanoferrates (MHCFs) are promising cathode materials for aqueous batteries owing to their high capacity and operating voltage. However, their electrochemical performance is often compromised by the presence of structural vacancies in the hexacyanoferrate framework. Conventional strategies to mitigate these vacancy defects typically rely on slowing the synthesis kinetics, which significantly reduces production throughput and hinders large‐scale applications. Herein, we introduce a protonation‐regulated coprecipitation (PRC) method that leverages the protonation equilibria of hexacyanoferrate (II/III) in an acidic medium, enabling the rapid and scalable synthesis of cobalt hexacyanoferrate (CoHCF). This approach achieves a high yield of approximately 90% within 15 min while maintaining low defect density. Adjusting the initial molar ratio of hexacyanoferrates enables effective modulation of the nucleation behavior of CoHCF and fine control over the particle geometry. The CoHCF with optimal geometry exhibits a high capacity of 159.4 mAh g⁻¹ at a current density of 100 mA g⁻¹ during sodium ion insertion‐extraction. A corresponding H‐type cell employing a Zn metal anode delivers an outstanding energy density of 236.6 Wh kg⁻¹ with an energy efficiency of 93.0%. These results demonstrate the effectiveness of the PRC process in producing high‐performance CoHCF, providing a scalable strategy for grid‐scale aqueous energy storage technologies.

A novel synthesis strategy is proposed which leverages the distinct acid–base equilibria of two hexacyanoferrate species for the rapid, chelator‐free production of low defect cobalt hexacyanoferrate (CoHCF). Despite a processing time of ∼15 min, the resulting CoHCF prepared via the protonation‐regulated coprecipitation exhibits excellent energy density and energy efficiency surpassing those of previously reported metal hexacyanoferrate‐based aqueous battery systems.

1. Introduction

With escalating global demand for energy and the urgent need to reduce carbon emissions, the transition to renewable energy has accelerated [1, 2]. However, the intermittent nature of renewable energy sources like solar and wind power requires efficient grid‐scale energy storage systems (ESS) to maintain a stable and reliable electricity supply [3, 4]. For the seamless integration of renewable energy into the electrical grid, ESS must meet key criteria such as scalability, cost‐effectiveness, long‐term operational stability, and safety.

Among the various energy storage technologies, aqueous rechargeable batteries (ARBs) have emerged as promising candidates for grid‐scale ESS owing to their abundant resources, simple manufacturing, environmental benignity, and inherent safety [5, 6]. Unlike lithium‐ion batteries, which require strict moisture control, ARBs can be produced more economically without the need for controlled environments. Notably, the use of water‐based electrolytes reduces environmental and health risks while preventing thermal runaway. Despite these compelling advantages, the energy density of ARBs reported so far remains relatively low, which continues to constrain their deployment [7, 8, 9]. Overcoming this limitation is essential to ensure the practical viability of ARBs.

Developing high‐performance cathode materials is of critical importance to achieve higher energy density in ARBs. Due to the intrinsically high gravimetric capacity and given electrode potential of metal anodes such as zinc, aluminum, and magnesium, the overall energy density of ARBs primarily depends on the capacity and redox potential of the cathode [10]. In this context, a variety of cathode materials, including layered transition metal oxides (LTMOs) [11, 12], polyanionic compounds (PCs) [13, 14], organic materials [15, 16], and Prussian blue analogues (PBAs) [17, 18] have been extensively studied. Among them, PBAs have garnered considerable attention due to their unique structural and electrochemical characteristics. PBAs possess an open‐framework structure in which two transition metals are interconnected via cyanide ligands through coordination bonds. This open‐framework structure enables the reversible insertion and extraction of a wide range of cations, such as Na⁺, K⁺, H₃O⁺, and NH₄ ⁺ [19, 20, 21, 22]. Moreover, depending on the combination of transition metals, PBAs can be tuned to deliver either high gravimetric capacities of up to 170 mAh g⁻¹ [23, 24] or high redox potentials approaching 1.0 V vs. Ag/AgCl [25]. These characteristics make PBAs highly promising cathode candidates for ARB systems.

Metal hexacyanoferrates (MHCFs) are a subclass of PBAs in which iron ions are coordinated to cyanide ligands via carbon atoms, whereas other transition metal ions are coordinated via nitrogen atoms [26]. Their general chemical formula can be expressed as A _x MFe(CN)₆·yH₂O, where A represents an alkali metal ion, M denotes a transition metal ion, and x and y are stoichiometric parameters. MHCFs can be readily synthesized through a simple coprecipitation method involving transition metal cations and hexacyanoferrate anions. This process was generally carried out under ambient pressure and at mild temperatures below 100°C, setting MHCFs apart from LTMOs and PCs that commonly require complex and energy‐intensive fabrication routes such as solvothermal synthesis, high‐temperature calcination, or sintering [27].

Albeit endowed with these favorable attributes, harnessing the latent performance of MHCFs hinges on the precise regulation of the synthesis process. Owing to the extremely low solubility product of MHCFs [28, 29], mixing precursor solutions containing transition metal ions and hexacyanoferrate ions induces instantaneous and uncontrolled nucleation [30, 31]. This explosive nucleation hampers the intercalation of A‐site cations into the solid lattice, which is crucial for maintaining electroneutrality. The resulting cation deficiency disrupts charge balance and encourages the formation of hexacyanoferrate vacancies [32]. Consequently, the coordination sites of M‐site ions that fail to bond with the nitrogen atoms are instead occupied by water molecules, known as coordinated water [33]. These defects associated with hexacyanoferrate vacancies have been widely recognized as a major cause of the electrochemical performance degradation of MHCFs. Bulk defects distributed throughout the particle interior disrupt ionic pathways, resulting in low ion conductivity [34, 35], while the substitution of electrochemically active hexacyanoferrate ions with water molecules reduces the gravimetric capacity [36, 37]. Furthermore, surface‐exposed defects accelerate lattice dissolution at the particle‐electrolyte interface, degrading the reversibility and long‐term stability of the electrode [32].

A variety of synthetic strategies have been explored to regulate the reaction kinetics of MHCF formation and suppress the defects [38, 39, 40, 41, 42]. Of these strategies, the use of chelating agents has been widely adopted [29, 31, 43]. The addition of chelating agents during synthesis rigorously attenuates the concentration of reactive free M‐site ions by forming highly stable complexes [44]. As a result, both nucleation and growth rates are substantially slowed, leading to the formation of low defect MHCFs with outstanding electrochemical performance [45, 46, 47]. However, using chelating agents inevitably results in prolonged synthesis times, sometimes exceeding 24 h [38, 48]. Compounding this issue, chelating agents necessitate cumbersome post‐purification procedures involving organic solvents because of their strong affinity for MHCF surfaces [49]. Other synthetic routes such as acid‐mediated dissociation of hexacyanoferrates or mechanochemical ball milling have also faced drawbacks, typically low yields [42, 50] in the former and high energy consumption [51, 52] in the latter, restricting their use in large‐scale production. Therefore, discovering more efficient synthetic approaches for producing highly crystalline MHCFs remains imperative for scale‐up production.

In this work, we present a protonation‐regulated coprecipitation (PRC) strategy for the synthesis of cobalt hexacyanoferrate (CoHCF), which leverages the distinct protonation equilibria of hexacyanoferrate (II) and hexacyanoferrate (III). The higher degree of protonation of hexacyanoferrate (II) effectively mitigated the explosive nucleation of CoHCF. Subsequently, hexacyanoferrate (III), generated in situ by hydrogen peroxide, promoted rapid yet regulated particle growth owing to its lower degree of protonation. Compared with conventional approaches, this strategy achieved both a markedly shortened synthesis time of only 15 min with a high yield of approximately 90%, and the formation of well‐defined cubic particles with a low defect density. Building on this, the nucleation density at the onset of the reaction was modulated by adding a trace amount of hexacyanoferrate (III) to the initial precursor solution. Extensive characterizations, including scanning electron microscopy (SEM), transmission electron microscopy (TEM), X‐ray diffraction (XRD), thermogravimetric analysis (TGA), derivative thermogravimetry (DTG), and energy‐dispersive X‐ray spectroscopy (EDS), collectively revealed that such modulation enabled fine control over the particle size and morphology without increasing defect densities. As a result, the CoHCF with an optimized geometry exhibited excellent gravimetric capacity, rate capability, cycling stability, and resistance to self‐discharge. Finally, the energy density and efficiency of a H‐type cell employing a Zn metal counter electrode were evaluated against previously reported MHCF‐based aqueous systems, clearly demonstrating the superior electrochemical performance of the CoHCF synthesized via the PRC method.

2. Results and Discussion

2.1. Protonation‐Regulated Synthesis of CoHCF

Figure 1a illustrates the schematic synthesis process of CoHCFs via the PRC method. The detailed experimental procedure is described in the experimental section. In short, an aqueous solution containing sodium sulfate, sulfuric acid, cobalt sulfate, and hydrogen peroxide was first prepared, followed by the addition of an aqueous solution containing hexacyanoferrate (II). After brief homogenization under vigorous stirring, the precursor mixture was heated to 70°C for the designated duration to initiate coprecipitation. The fundamental strategy of the PRC method is to preclude the instantaneous nucleation of CoHCFs through the protonation of hexacyanoferrate (II) species. The pK _a value, which is the negative logarithmic form of the acid dissociation constant K _a, quantitatively correlates solution pH with the equilibrium distribution between an acid and its conjugate base. For instance, sulfate ions (SO₄ ²⁻), with a pK _a of 1.99, exist as an equimolar mixture with bisulfate ions (HSO₄ ⁻) when the solution pH equals 1.99. Meanwhile, hexacyanoferrate (II) exhibits stepwise protonation with pK _a values of 4.16, 2.65, and −1.08 for the mono‐, bi‐, and tri‐protonated species, respectively [53]. Under strongly acidic conditions (pH < 1), the majority of hexacyanoferrate (II) exists in forms with a high degree of protonation [54], while sulfate ions are also predominantly present as bisulfate. This reduction in the concentration of free hexacyanoferrate (II) is known to retard its associated reaction kinetics [55, 56, 57]. In particular, the lower reactivity of protonated hexacyanoferrate (II) toward metal cations, compared with its deprotonated counterpart [58], provides a means to mitigate the explosive nucleation during MHCF formation. Upon heating, the hexacyanoferrate (II) is rapidly oxidized to hexacyanoferrate (III) by hydrogen peroxide. This oxidation triggers spontaneous deprotonation, as the pK _a of monoprotonated hexacyanoferrate (III) is considerably lower (pK _a = −0.6) [59] than that of the hexacyanoferrate (II). This regulated generation of deprotonated hexacyanoferrate (III) provides a well‐tempered concentration of reactive species, which promotes homogeneous nucleation and fosters the growth of high‐quality crystals.

FIGURE 1.

Schematic illustrations showing the principle of the PRC method and its concentration evolution (LaMer curve). (a) In a strongly acidic medium, protonation of hexacyanoferrate (II) suppresses rapid nucleation upon contact with Co²⁺. Heating in the presence of H₂O₂ accelerates oxidation and subsequent deprotonation to generate reactive hexacyanoferrate (III), enabling regulated coprecipitation of CoHCF. (b) Corresponding LaMer curve showing a metastable stage (I), nucleation stage (II), and growth stage (III). HCF denotes hexacyanoferrate, and p‐HCF represents its protonated form.

Figure 1b depicts the three reaction stages of the PRC method using a LaMer curve to provide a comprehensive understanding of the sequential process. In the LaMer model, the concentration of precursor solute is classified into three characteristic regimes: the saturation concentration (C _sat), above which growth of existing nuclei occurs; the critical nucleation concentration (C _min), beyond which spontaneous nucleation occurs; and the critical limiting supersaturation (C _max), representing the upper limit of metastable supersaturation, at which instantaneous nucleation occurs [60, 61, 62]. In the PRC method, the coprecipitation reaction of hexacyanoferrate (II) species is strongly retarded by their high degree of protonation. Instead, the progression through the distinct reaction stages is governed by the concentration of hexacyanoferrate (III), which functions as the reactive precursor. During stage I, hexacyanoferrate (II) undergoes gradual oxidation by hydrogen peroxide as soon as it is introduced into the solution. As a result, the concentration of hexacyanoferrate (III) can increase beyond the saturation concentration (C _sat). However, as long as the concentration remains below the critical nucleation threshold (C _min), the solution maintains a metastable state without the formation of new nuclei. Thermal activation expedites the oxidation reaction, triggering a rapid surge in the hexacyanoferrate (III) concentration. Once the concentration exceeds C _min, stage II begins during which nucleation takes place. Finally, in stage III, the simultaneous formation and growth of nuclei in stage II consume hexacyanoferrate (III), lowering its concentration below C _min and terminating the formation of new nuclei. Thereafter, the hexacyanoferrate (III) species continuously generated from the ongoing oxidation are exclusively incorporated into the growing particles formed in stage II until depletion.

In conventional chelator‐assisted synthesis of MHCFs, the effective concentration of metal ions is deliberately maintained at a very low level below the C _max to prevent uncontrolled nucleation. An analogous effect is achieved through the extensive protonation of hexacyanoferrate (II) in the PRC method. While the underlying regulatory mechanism is conceptually similar to that of chelating agents, the PRC method distinguishes itself by enabling rapid release of active precursors via the oxidation of hexacyanoferrate (II). This feature allows the synthesis of CoHCFs to proceed within a brief temporal window. Moreover, the absence of organic chelating agents streamlines the post‐synthesis purification process, reducing both the environmental footprint and the overall cost associated with waste treatment [63, 64].

On the other hand, it is well established that the partial dissociation of cyanide ligands from hexacyanoferrate ions in an acidic environment can generate iron hexacyanoferrate (FeHCF) [65, 66, 67]. This reaction consumes the hexacyanoferrate species and may induce unintended heterogeneous impurities, thereby interfering with the controlled synthesis of CoHCFs. Interestingly, hydrogen peroxide was found to play a dual role in the PRC method, not only initiating the coprecipitation of CoHCFs but also effectively inhibiting the formation of FeHCF. This behavior was confirmed by a control experiment in which an acidic solution containing hexacyanoferrate (II) was heated at 70°C for 2 h. In the absence of hydrogen peroxide, the solution color gradually turned blue, indicative of FeHCF formation. In contrast, the solution containing hydrogen peroxide retained its pale yellow color (Figure S1). This difference is likely attributed to the higher solubility of the fully oxidized FeHCF compared to its reduced states [68, 69, 70]. In addition, hydrogen peroxide is known to decompose toxic cyanide ions into benign carbonate and nitrate ions, significantly enhancing process safety by neutralizing the hazardous by‐product [71, 72].

Increasing the concentration of alkali metal ions, such as Na⁺ or K⁺, in a precursor solution facilitates the maintenance of local electroneutrality during crystal growth, thereby suppressing the formation of defects in MHCFs [73, 74]. In the PRC method, sodium sulfate supplies abundant Na⁺ ions while simultaneously influencing the reaction kinetics via changes in solution pH. Specifically, the formation of bisulfate ions reduces the proton concentration, shifting the protonation equilibria in the solution toward the dissociated state. This shift increases the concentration of free hexacyanoferrate (III) species, accelerating the precipitation kinetics of CoHCF. To elucidate the influence of sodium sulfate in the PRC method, the geometrical evolution of CoHCFs synthesized at a fixed sulfuric acid concentration, with and without sodium sulfate was monitored over time (Figure S2). In the absence of additional sodium sulfate, the CoHCFs slowly grew into well‐defined cubic particles with sizes exceeding 5 µm. The potent reaction regulation capability of the PRC method was reflected in a sustained growth period exceeding 1 h, accompanied by a broad particle size distribution. This phenomenon originates from protonation equilibria strongly biased toward associated states, extending beyond hexacyanoferrate (II) to encompass hexacyanoferrate (III). As a result, the sluggish precursor consumption rate prolonged the nucleation stage, amplifying the disparity in growth times among individual particles and broadening the size distribution [75, 76]. The addition of sodium sulfate, however, induced a marked shift in the synthesis dynamics. Expedited nucleation and growth of particles led to a substantial reduction in both particle size and reaction time. Furthermore, the narrowed particle size distribution signifies that particle growth was initiated synchronously following a brief nucleation window [77, 78, 79], a condition conducive to ensuring uniform structural quality. In light of these findings, further investigations in this study focused on CoHCFs synthesized in the presence of sodium sulfate, as this condition is anticipated to curtail synthesis time while simultaneously ensuring a low defect density.

In parallel with defect engineering, the size and morphology of MHCFs are of paramount importance owing to their pronounced influence on electrochemical performance when employed as cathode materials. In general, smaller particle size and larger specific surface area are favored to enhance the accessibility of redox‐active sites, which is advantageous for achieving high gravimetric capacity and superior rate capability [80, 81, 82]. As discussed above, sodium sulfate in the PRC method serves to reduce both synthesis time and product size. Nevertheless, it was observed that even when the sodium sulfate concentration approached its saturation limit [83], the particle size remained on the order of approximately 2 µm. Therefore, alternative strategies are required to further refine the particle size to the sub‐micrometer scale. The population of nuclei generated during the nucleation stage plays a pivotal role in determining the size of final product [84, 85, 86]. Wang et al. reported effective control of FeHCF particle size by exploiting the distinct nucleation kinetics of hexacyanoferrate (II) and hexacyanoferrate (III) [87]. Drawing upon this insight, a similar strategy can be implemented in the PRC method. Specifically, incorporating a trace amount of hexacyanoferrate (III) into the initial precursor solution offered an effective avenue for modulating the nucleation process.

Figure 2a illustrates the LaMer curve when hexacyanoferrate (III) is introduced into the initial precursor. The pre‐added hexacyanoferrate (III) induces nucleation within a short time window and supplies a controlled population of nuclei. Upon heating, these preformed nuclei undergo growth by assimilating hexacyanoferrate (III) species delivered via the oxidation reaction. Because the hexacyanoferrate (III) produced during this heating phase is rapidly consumed at these pre‐existing reaction sites, the formation of additional nuclei is effectively suppressed. Consequently, the total number of particles within the solution is predominantly governed by the amount of the pre‐added hexacyanoferrate (III).

FIGURE 2.

Schematic illustrations showing the effect of pre‐added hexacyanoferrate (III) on nucleation and particle evolution in the PRC method. (a) Modified LaMer curve. Pre‐adding hexacyanoferrate (III) triggers a modulated nucleation, producing a defined population of nuclei. Upon heating, growth of the preformed nuclei proceeds while secondary nucleation is inhibited by maintaining the concentration below C _min, yielding size‐refined, monodisperse particles. (b) Schematic illustration of how particle populations and geometry vary with the hexacyanoferrate (III) fraction in the initial precursor. A higher nucleation density reduces particle size, whereas excessive loading promotes particle agglomeration.

Figure 2b illustrates the influence of the initial hexacyanoferrate (III) fraction on the morphological evolution of CoHCFs. Increasing the fraction of hexacyanoferrate (III) leads to the generation of a larger number of nuclei during the nucleation stage. Although a higher nucleation density generally promotes particle size reduction by distributing the precursor over a greater number of growth centers, an excessive initial loading can be detrimental. The congested particles cause the probability of inter‐particle interactions to surge quadratically [88, 89], leading to uncontrolled agglomeration. More importantly, the superfluous introduction of initial hexacyanoferrate (III) may undermine the rigorous reaction regulation required for effective defect suppression. To systematically investigate these trade‐offs, a comprehensive series of experiments was conducted to analyze the influence of the hexacyanoferrate (III) fraction on particle geometry, in conjunction with lattice structure, composition, and electrochemical performance.

2.2. Geometry and Structural Characterization of CoHCFs

Figure 3a–d shows the SEM images of CoHCF‐(X:Y) where X and Y denote the initial molar ratio of hexacyanoferrate (II) and hexacyanoferrate (III), respectively. The monodisperse particle size of each sample implies that the nucleation occurred in a single, brief event, since a repeated or prolonged nucleation would have caused broader size distributions. Furthermore, it is clearly shown that the mixing ratio of the hexacyanoferrate precursors transformed the particle geometry of CoHCFs. As determined from SEM images, the average particle sizes were 2.3 µm for CoHCF‐(100:0), 560 nm for CoHCF‐(98:2), 390 nm for CoHCF‐(95:5), and 260 nm for CoHCF‐(90:10). In addition to the size reduction, a marked change in particle morphology was apparent in the CoHCF‐(90:10). While other samples maintained well‐defined cubic shapes, CoHCF‐(90:10) exhibited pronounced particle agglomeration, indicative of excessive nucleation. This contrast is further corroborated by the TEM images (Figure S3), where the transition from smooth, faceted surfaces of CoHCF‐(100:0), CoHCF‐(98:2), and CoHCF‐(95:5) to the irregularly textured particles of CoHCF‐(90:10) is distinctly captured. From these observations, it is evident that the nucleation density in the PRC method was successfully modulated by adjusting the mixing ratio of the hexacyanoferrate precursors.

FIGURE 3.

Structural and compositional characterization of CoHCF synthesized with different hexacyanoferrate precursors ratios. (a–d) SEM images of CoHCF‐(100:0), CoHCF‐(98:2), CoHCF‐(95:5), and CoHCF‐(90:10). (e) XRD patterns and (f) lattice constants and the I ₂₂₀ /I ₂₀₀ intensity ratios derived from the XRD patterns. (g) TGA curves showing mass losses assigned to interstitial water and coordinated water. (h) Corresponding DTG curves. (i) Fe‐to‐Co atomic ratios determined by EDS.

To examine whether variations in the nucleation stage influence the lattice structure, XRD analysis was performed. Figure 3e shows the XRD patterns of CoHCFs and the standard XRD pattern of FeHCF (JCPDS, no. 52–1907). The sharp diffraction peaks with no impurity peaks indicate the high crystallinity and purity across the samples. The XRD patterns confirmed that all samples share a face‐centered cubic crystal structure. To further probe subtle lattice‐level effects, specific lattice metrics were quantitatively extracted from the diffraction patterns. Figure 3f shows the lattice constants and I ₂₂₀ /I ₂₀₀ ratios. The lattice constants were determined by averaging values from eight diffraction peaks (200, 220, 400, 420, 422, 440, 600 and 620) in the XRD patterns using Bragg's law. The nearly identical lattice constants (∼9.95 Å) suggest structural consistency regardless of the nucleation conditions, given that the defects and accompanying coordinated water can cause lattice distortion [90]. Meanwhile, the increased I ₂₂₀ /I ₂₀₀ ratio observed for smaller particles suggests a reduced contribution from the (200) plane, which aligns with the geometrical transition confirmed by electron microscopy [91].

Consistent with the lattice structures, compositional analysis also revealed uniformity among the CoHCF samples. Figure 3g,h presents the TGA and DTG curves to analyze the water content of the samples. Water in PBAs exists as either interstitial or coordinated water [92, 93]. The DTG data show two well‐separated peaks, each ascribed to the release of the distinct type of water. From the results, the contents of interstitial and coordinated water of the CoHCFs including a small amount of adsorbed water are quantified and annotated in the TGA data (Figure 3g). The comparable TGA profiles and water distributions support the compositional homogeneity. Moreover, the low content of coordinated water implies a suppressed defect density within the CoHCF lattice [33]. The high crystallinity of the samples was further supported by the near‐stoichiometric distribution of transition metal elements within the lattice. Figure 3i presents the Fe‐to‐Co atomic ratios obtained from EDS analysis, revealing that the hexacyanoferrate vacancies, corresponding to the deficiency of iron relative to cobalt, were less than 10% with only marginal differences among the four samples. Here, the error bars represent the standard deviation obtained from five measurement points for each sample. The lowest defect density was observed for the CoHCF‐(95:5), with a vacancy content of 6.88 ± 1.61%. The depressed water activity (0.60–0.65) of the reaction solution [94] combined with the oxidation‐limited coprecipitation kinetics may have contributed to this result. In addition, EDS mapping shows a uniform distribution of sodium ions throughout the particles, indicating their effective incorporation into the CoHCF framework (Figure S4). Such incorporation is also expected to suppress defect formation, enhancing the overall crystallinity of the CoHCF samples.

Collectively, these results demonstrate that the mixing ratio of hexacyanoferrate precursors primarily governs the shape and morphology of CoHCFs in the PRC method, without perturbing the lattice structure or stoichiometric composition. This finding provides compelling evidence for the clear separation of nucleation and crystal growth stages. While the pre‐added hexacyanoferrate (III) modulated the nucleation density, the subsequent growth environment remained invariant and produced particles of varying geometry with nearly identical quality. Such unique characteristics highlight the distinct advantage of the PRC method. Conventional chelator‐based strategies allow particle size control by tuning the binding strength between chelating agents and metal ions, yet this is often at the cost of increased lattice defects which deteriorate electrochemical performance [82, 95, 96]. Alongside defect suppression and particle size control, practical considerations such as material yield are also indispensable for large‐scale implementation, serving as fundamental determinants of manufacturing efficiency, scalability, and economic feasibility [97, 98]. In this study, the yields of the samples were evaluated after excluding the weight of crystal water (Figure S5). Even under the rapid synthesis time, the PRC method achieved a yield of up to 90%, emphasizing its potential for scalable production.

2.3. Electrochemical Performance of CoHCFs

The electrochemical performance of CoHCFs was evaluated using a three‐electrode setup comprising an Ag/AgCl reference electrode and a high‐surface‐area activated carbon counter electrode. The typical mass loading of CoHCF in the working electrode was 1–2 mg cm⁻². A 17 mol kg⁻¹ sodium perchlorate water‐in‐salt electrolyte was chosen for its expanded electrochemical stability window, high ionic conductivity, and cost‐effectiveness, making it an ideal choice for ARB systems. Specifically, the perchlorate anion is known to stabilize the MHCF structure by alleviating the dissolution of MHCFs during electrochemical charge–discharge processes, thereby ensuring long‐term structural integrity and cycling stability [99, 100]. Before the following tests, cyclic voltammograms (CVs) of the as‐prepared samples were recorded to estimate the initial oxidation state of CoHCFs (Figure S6). All CVs show two distinct redox peaks which correspond to the redox reaction of Co²⁺/Co³⁺ at 0.3–0.6 V vs Ag/AgCl and Fe²⁺/Fe³⁺ at 0.8–1.1 V vs Ag/AgCl [101, 102]. As the complete reduction curves of Co²⁺/Co³⁺ redox reactions were observed during the initial scans, all as‐prepared CoHCFs were confirmed to be in a half‐oxidized state. The valence state of Co and Fe determined from the X‐ray photoelectron spectroscopy (XPS) analysis were also consistent with this result, suggesting preferential existence of Co³⁺ and Fe²⁺ species (Figures S7 and S8).

Figure 4a–d presents the CVs of the CoHCF samples at various scan rates. The symmetric oxidation and reduction peaks in the CV plots serve as evidence of the high reversibility of sodium ion insertion and extraction in the CoHCFs. All samples exhibited well‐defined redox peaks for both Co and Fe, whose comparable prominence across different scan rates suggests balanced contributions of each redox‐active center to the total charge storage. In addition, the formal potentials remained invariant among the samples. This consistency is in line with their similar water and defect contents identified previously, which are known to influence the formal potentials of MHCFs [86, 103]. However, when comparing the current levels, an obvious difference is observed. The specific peak currents increased with the amount of pre‐added hexacyanoferrate (III), except for CoHCF(90:10) which showed diminished and broadened redox peaks. Considering the negligible difference in lattice structure and stoichiometric composition, changes in the electrochemical behavior should be attributed to the geometrical variations. Figure 4e depicts the influence of particle size and morphology on electrochemically accessible fraction in cathode particles. Although the analysis of the Co²⁺/Co³⁺ redox couple was complicated by peak splitting, which may be associated with the structural transformation from the half‐oxidized cubic state to the fully reduced rhombohedral state [24], the Fe²⁺/Fe³⁺ couple exhibited a linear relationship between the peak currents and the square root of the scan rate. This linearity validates the diffusion‐limited nature of the sodium intercalation/deintercalation kinetics in the CoHCFs (Figure S9), in accordance with the Randles–Sevcik equation [104, 105]. By normalizing the equation against the mass of active material, a relationship between specific peak current and specific surface area can be derived. Typically, a reduction in particle size increases the specific surface area, which in turn enhances the specific peak current, as predicted by the equation. However, this prediction holds for samples up to CoHCF‐(95:5), where progressive refinement of particle size was accompanied by a corresponding increase in specific peak current. In contrast, CoHCF‐(90:10) exhibited a weakened peak current despite the smallest particle size. Although its irregular morphology is expected to further enlarge the specific surface area, this did not translate into improved electrochemical performance. The presence of intergranular boundaries formed by the agglomeration of nuclei may account for this discrepancy [106, 107]. As shown in the SEM and TEM images, unlike the other samples, CoHCF‐(90:10) is predominantly composed of secondary aggregates formed by primary particles. The resulting extensive boundaries could disrupt continuous ionic and electronic transport pathways, counteracting the expected benefits of reduced particle size and extended surface area.

FIGURE 4.

Cyclic voltammograms of (a) CoHCF‐(100:0), (b) CoHCF‐(98:2), (c) CoHCF‐(95:5), (d) CoHCF‐(90:10) recorded at scan rates of 0.2–2 mV s⁻¹ in a 17 mol kg⁻¹ NaClO₄ electrolyte. (e) Schematic illustration of size‐ and morphology‐dependent electrochemical accessibility in cathode particles.

Figure 5a–d presents the galvanostatic charge–discharge voltage profiles of CoHCFs at various current densities, which further corroborate the electrochemical trends observed in the preceding analyses. The enhancement in electrochemical performance with decreasing particle size was confined to the regime in which no morphological changes occurred. The gravimetric capacities of CoHCFs measured at a current density of 100 mA g⁻¹ were 134.3, 161.8, 159.4, and 151.5 mAh g⁻¹ for CoHCF‐(100:0), CoHCF‐(98:2), CoHCF‐(95:5), and CoHCF‐(90:10), respectively. At higher current densities, the advantage of reduced particle size without agglomeration became more pronounced. This feature is evidenced by the outstanding rate capability and the lowest charge–discharge voltage hysteresis of CoHCF‐(95:5) (Figure 5e,f and Figure S10). For clarity, the hysteresis values were calculated as the quotient of the energy loss per charge–discharge cycle and the corresponding discharge capacity. To minimize the influence of coulombic efficiency on the evaluated energy loss, the hysteresis values were determined at current densities of at least 200 mA g⁻¹, where the coulombic efficiency exceeded 99.5%. Cyclability tests at 500 mA g⁻¹ (Figure S11) revealed severe degradation in CoHCF‐(90:10). This deterioration is attributed to its excessive surface area and unstable intergranular boundaries, which accelerate parasitic side reactions and induce particle fracture during cycling. In contrast, CoHCF‐(98:2) and CoHCF‐(95:5) demonstrated robust stability, retaining over 98% of their initial capacities. Notably, CoHCF‐(100:0) exhibited the highest capacity retention of 99.5%, likely stemming from restricted side reactions due to its minimized specific surface area. This observation suggests that larger particle sizes may offer advantages in applications where long‐term cyclability is prioritized over high‐rate performance. Taken together, these results indicate the existence of an optimal hexacyanoferrate precursor mixing ratio in the PRC method that effectively reduces particle size while preventing agglomeration to achieve superior electrochemical performance. On this basis, CoHCF‐(95:5) was selected for further investigation, as it exhibited an excellent balance of gravimetric capacity, rate capability, and cycling stability.

FIGURE 5.

Galvanostatic charge–discharge voltage profiles of (a) CoHCF‐(100:0) (b) CoHCF‐(98:2) (c) CoHCF‐(95:5) (d) CoHCF‐(90:10) at various current densities. (e) Rate capability and (f) voltage hysteresis extracted from the charge–discharge profiles as a function of current density.

Figure 6a,b shows the long‐term cycling performance of CoHCF‐(95:5) at a current density of 500 mA g⁻¹. The electrode delivered an average coulombic efficiency of 99.8% and retained 86.7% of its initial capacity after 500 cycles. The capacity decay rate was as low as 0.026% per cycle, indicating excellent cycling durability. Furthermore, the characteristic dual voltage plateaus were well preserved throughout the cycling, underscoring the high electrochemical stability of the material. The SEM image of CoHCF‐(95:5) after cycling revealed a well‐preserved cubic morphology, further demonstrate the high structural robustness of the particles (Figure S12). To evaluate practical viability, the self‐discharge behavior was assessed using a 6 h open‐circuit voltage (OCV) hold test (Figure 6c,d). Given the dual voltage plateaus of CoHCF, evaluation was conducted at two distinct states of charge (SOC): a fully charged state, in which both plateaus were charged, and a half charged state, in which only the lower plateau was utilized. Impressively, the CoHCF‐(95:5) retained 91.6% of its capacity in the fully charged state and an even higher 97.9% in the half charged state, which is comparable to that achieved by introducing solid‐electrolyte‐interphase layers in aqueous electrolytes [108]. In aqueous systems, parasitic water‐splitting reactions represent one of the primary contributors to self‐discharge, which are known to be facilitated by defects in MHCFs [109, 110, 111, 112]. Furthermore, water‐in‐salt electrolytes have been widely reported to restrict such side reactions by reducing the activity of free water molecules [113]. Accordingly, the low self‐discharge rate observed here can be attributed to the combined effects of a minimal surface defect concentration in CoHCF prepared via the PRC method and the reduced reactivity of water in the water‐in‐salt electrolyte. The excellent capacity retention demonstrates the suitability of CoHCF‐based aqueous batteries for addressing the intermittency of renewable energy sources.

FIGURE 6.

Cycling stability, self‐discharge behavior, and Zn–CoHCF cell performance of CoHCF‐(95:5). (a) Long‐term cycling performance and coulombic efficiency at a current density of 500 mA g⁻¹. (b) Charge–discharge voltage profiles at the first, 250th, and 500th cycles. Self‐discharge behavior evaluated by charge–discharge profiles measured at a current density of 100 mA g⁻¹ after a 6 h open‐circuit rest under (c) fully charged and (d) half charged conditions. (e) Schematic of the H‐type asymmetric Zn–CoHCF cell separated by an anion‐exchange membrane, employing NaClO₄ catholyte and NaClO₄/Zn(ClO₄)₂ anolyte. (f) Charge–discharge voltage profiles of the Zn–CoHCF cell at a current density of 100 mA g⁻¹ for the first, 25th, and 50th cycles. Comparison of representative MHCF‐based aqueous systems in terms of (g) energy density and (h) energy efficiency, calculated based on the cathode mass.

To evaluate the energy density and energy efficiency for practical energy storage, an H‐type asymmetric cell was assembled using a Zn metal anode (Figure 6e). Detailed cell assembly procedures are provided in the Experimental Section. Direct contact between MHCFs and Zn²⁺ ions poses a risk of spontaneous M‐site ion exchange, driven by the strong coordination affinity of zinc ions toward the nitrogen atoms of cyanide ligands [114, 115]. Since the incorporated zinc species are electrochemically inert within the MHCF framework, such substitution leads to a loss of electrochemically accessible redox sites. To prevent the adverse ion crossover, an anion‐exchange membrane was introduced between the Zn anode and the CoHCF cathode. Figure 6f shows the charge–discharge voltage profiles of the Zn–CoHCF cell measured at a current density of 100 mA g⁻¹ during the first, 25th, and 50th cycles. An energy density of 236.6 Wh kg⁻¹, calculated based on the cathode mass, is achieved together with an excellent energy efficiency of 93.0%. It is noteworthy that the voltage polarization associated with the resistance of the AEM was calculated to be only 11.9 µV, indicating that the majority of the energy loss originated from imperfect coulombic efficiency and polarization at the electrode interfaces and within the electrolyte. Upon continued cycling, energy densities of 231.5 and 224.6 Wh kg⁻¹ were retained after 25 and 50 cycles, respectively, while maintaining consistent energy efficiency over 93% (Figure S13). These outstanding performances originate from the synergistic combination of high gravimetric capacity and average working voltage, and low polarization of the CoHCF cathode. Meanwhile, despite the high energy density of the CoHCF, the cell‐level energy density remains limited owing to the substantial electrolyte mass required for Zn²⁺ supply (see Note S1). The development of an efficient aqueous sodium‐ion anode with a low redox potential, high gravimetric capacity, and excellent cycling stability may provide a viable pathway toward realizing the practical full‐cell systems.

To contextualize these achievements, the electrochemical performance of the present system was compared with previously reported MHCF‐based aqueous energy storage systems (Figure 6g,h and Table S1) [116, 117, 118, 119, 120, 121, 122, 123, 124, 125]. In this comparison, the CoHCF‐(95:5) surpasses most literature‐reported MHCF benchmarks, exhibiting superior energy density and energy efficiency. Of particular significance is the high energy efficiency for practical ESS implementations. By minimizing resistive energy losses that would otherwise dissipate as heat during operation [126, 127], this characteristic can reduce the overhead required for thermal management. Cumulatively, these results position the CoHCF synthesized via the PRC method as a promising candidate for advanced aqueous battery applications.

3. Conclusions

In this work, we developed a PRC method that enables the rapid, scalable synthesis of low defect CoHCF cathodes by exploiting the protonation equilibria of hexacyanoferrate species.

By selectively suppressing the reactivity of hexacyanoferrate (II) in an acidic medium, the PRC method effectively precludes the uncontrolled, explosive nucleation typical of conventional coprecipitation. Subsequently, controlled particle growth is achieved through the gradual release of hexacyanoferrate (III) via heat‐accelerated oxidation, enabling precise regulation of the coprecipitation process. This approach delivers a high synthesis yield of approximately 90% within only 15 min, demonstrating clear advantages in terms of throughput and scalability over conventional chelator‐assisted methods. Furthermore, the nucleation density of the CoHCF was modulated by adjusting the initial molar ratio of hexacyanoferrate (II) and hexacyanoferrate (III), enabling systematic control over particle size and morphology. Structural and compositional analyses revealed that this modulation reconfigures particle geometry without perturbing lattice integrity or stoichiometry, implying effective decoupling of nucleation and growth stages in the synthesis process. While reducing particle size generally enhanced electrochemical performance by increasing active surface area, excessive nucleation led to particle agglomeration and the formation of intergranular boundaries, which impeded ionic and electronic transport. These results suggest the existence of an optimal nucleation regime in which particle miniaturization is achieved without incurring performance degradation arising from excessive intergranular boundaries. Electrochemical evaluations demonstrated that the CoHCF with optimized geometry exhibits excellent gravimetric capacity, rate capability, cycling stability, and resistance to self‐discharge. Notably, the CoHCF retained 97.9% of its capacity after a 6 h rest, underscoring its suitability for energy storage applications operating under intermittent conditions. When paired with a Zn metal anode in an asymmetric aqueous cell configuration, the CoHCF delivered an exceptional performance, achieving an energy density of up to 236.6 Wh kg⁻¹ and an energy efficiency exceeding 93%, surpassing previously reported MHCF‐based aqueous battery systems. Taken together, this study establishes the PRC method as a powerful and versatile synthesis platform that reconciles rapid processing, high yield, and precise defect control. Beyond CoHCF, the underlying principle of anion protonation‐mediated reaction regulation offers mechanistic insights that could inform the rational design of high‐quality Prussian blue analogues and facilitate their scalable production, opening new opportunities for advanced aqueous energy storage technologies.

4. Experimental Section

4.1. Chemicals and Components

Potassium hexacyanoferrate (II), potassium hexacyanoferrate (III), cobalt sulfate, and zinc perchlorate were purchased from Sigma–Aldrich. Sulfuric acid and sodium perchlorate were purchased from Junsei. Sodium sulfate was purchased from Wako. All reagents were used as purchased without additional purification. For cathode preparation, CoHCF powders, carbon black nanopowder (Plasma Chem), and polyvinylidene fluoride (PVDF, Sigma–Aldrich) were mixed in the weight ratio of 7:2:1 using N‐methyl‐2‐pyrrolidone (NMP, DAEJUNG) as the solvent. The resulting slurry was cast onto a titanium mesh current collector (Nilaco Corporation) and dried at 80°C for 12 h. An anion‐exchange membrane (FAD‐55, Fumasep) and a zinc metal plate (Nilaco Corporation) were used for cell assembly.

4.2. Synthesis of CoHCFs

CoHCFs were prepared based on a controlled coprecipitation process. In a typical procedure, 0.5 mol of Na₂SO₄, 0.78 mol of H₂SO₄ and 0.214 mmol of CoSO₄ were dissolved in 130 g of deionized water. After cooling the solution to room temperature, 2.85 mmol of H₂O₂ was added to obtain solution A. Solution B was prepared by dissolving 0.178 mmol K₄Fe(CN)₆ in 10 g of deionized water. Under vigorous stirring, solution B was quickly injected into solution A. The final molalities of the mixture solution were 3.5 mol kg⁻¹ Na₂SO₄, 5.5 mol kg⁻¹ H₂SO₄, 2 × 10⁻² mol kg⁻¹ H₂O₂, 1.5 × 10⁻³ mol kg⁻¹ CoSO₄ and 1.25 × 10⁻³ mol kg⁻¹ K₄Fe(CN)₆. For comparison, CoHCF sample synthesized without sodium sulfate was prepared using identical conditions except for the omission of Na₂SO₄. To control the geometry of CoHCFs, solution B was prepared with a constant total molality of hexacyanoferrate species, while varying the molar ratio between K₄Fe(CN)₆ and K₃Fe(CN)₆, which were labeled accordingly to reflect differing synthesis conditions. After brief mixing, the resulting mixtures were heated to 70°C using temperature‐controlled bath circulators for the designated durations (2 h for the sample without sodium sulfate, 25 min for the sample with sodium sulfate (i.e., CoHCF‐(100:0)) and 15 min for the CoHCF‐(98:2), CoHCF‐(95:5), and CoHCF‐(90:10)). The purple precipitates were vacuum filtered, washed with deionized water several times, and dried at 80°C for 1 h.

4.3. Materials Characterization

The geometry of CoHCFs was examined using scanning electron microscopy (COXEM CX‐200TM) and field emission transmission electron microscope (JEOL JEM‐2100F). X‐ray diffraction patterns were collected by a PANalytical X'Pert Powder diffractometer (PANalytical) with Cu Kα radiation (λ = 1.54 Å). Temperature‐dependent weight loss was measured by a thermogravimetric analyzer (Netzsch TG 209 F3). Elemental compositions were analyzed by an energy‐dispersive X‐ray spectrometer (Oxford X‐MaxN). Elemental and oxidation state information were collected by X‐ray photoelectron spectroscopy (Thermo scientific, K‐Alpha). Electrochemical measurements, including cyclic voltammetry and galvanostatic charge–discharge tests, were conducted using computer‐controlled voltage‐current meter (Corrtest instruments CS310) and battery tester (Neware CT‐4008Q‐5V100mA‐124). To evaluate the energy density and energy efficiency of the CoHCF, an H‐type cell configuration was employed. The inner diameter of the bridge of H‐type cell was 1.5 cm. The CoHCF cathode was immersed in a 17 mol kg⁻¹ NaClO₄ aqueous electrolyte, while the Zn metal anode was immersed in a 16 mol kg⁻¹ NaClO₄ electrolyte containing 0.5 mol kg⁻¹ Zn(ClO₄)₂. The electrodes were separated by an anion‐exchange membrane. Prior to use, the membrane was sequentially treated in 1 M NaOH aqueous solution for 24 h, followed by 1 M NaClO₄ for another 24 h, and rinsed thoroughly with deionized water.

Conflicts of Interest

None of the authors have a conflict of interests to disclose.

Supporting information

Supporting File: smll74991‐sup‐0001‐SuppMat.docx.

Data Availability Statement

The data that supports the findings of this study are available in the Supplementary Material of this article.