Introduction

Rechargeable batteries are one of the widespread electrical energy storage technologies and have various applications in portable electronic devices, electric vehicles, and grid stabilization [1, 2]. Among a majority of batteries, lithium-ion batteries (LIBs) belong to the most advanced system due to their superior electrochemical performance such as good energy density, high voltage, good cycle life, and low self-discharge [3–6]. However, traditional LIBs, based on the graphite-lithium cobalt oxide interaction chemistry and flammable organic electrolytes, always have safety issue and are expensive.

The development of zinc-based rechargeable batteries, such as zinc-air, zinc-silver, and zinc-bromine batteries, has attracted much attention in recent years due to their low cost, safety, and ease of handling. In the recent decades, considerable efforts have been devoted to alkaline electrolytes, e.g. in zinc-manganese oxide (MnO2) batteries [7–9]. These batteries have shown good electrochemical performance but have a negative impact on the environment in the case of leakage. In addition, the problems including hydrogen evolution, shape change, and dendritic growth of zinc in aqueous solutions are still major challenges. Recently, zinc-ion batteries with neutral electrolytes [10, 11] and with organic electrolytes [12] have been reported. Nowadays, attention is focused on the exploration of ionic liquid electrolytes [13–15]. The problems associated with hydrogen evolution can be avoided and the dendritic growth of zinc can, in principle, be suppressed in ionic liquids. In addition, ionic liquid-water and ionic liquid-organic solvent mixtures have also been demonstrated to be promising electrolytes for battery applications [16–18]. By the addition of water or organic solvents, the viscosity is lowered and the ionic conductivity is enhanced, which largely improves the performance of the battery. Furthermore, it was also shown that the morphology of zinc deposits obtained by electrodeposition from the ionic liquid 1-ethyl-3-methylimidazolium dicyanamide [emim][dca] containing 3 wt% water was smooth [19]. We previously studied the electrodeposition of zinc from ionic liquids with a water content of 5 to 80 vol% [20] and also found that the structure of the electrolyte transformed from an ionic liquid-like solution to an aqueous-like solution upon the addition of water to more than 40 vol% [21]. Song et al. reported the suppression of zinc dendrites in an aqueous solution with ionic liquids as additive [22]. In comparison with conventional alkaline electrolytes, ionic liquids or ionic liquid molecular solvent mixtures can be regarded as promising electrolytes in battery applications.

Manganese dioxide (MnO2) has been widely used as cathode material in zinc-ion batteries due to its abundance, low cost, and low toxicity. However, Zn-MnO2 batteries have some drawbacks such as the formation of by-products, low coulombic efficiency, and poor reversibility during cycling. Therefore, novel cathode materials for zinc-ion batteries were explored, including vanadium oxide bronze (Zn0.25V2O5⋅nH2O) [23], NASICON-structured Na3V2(PO4)3 [24, 25], cation-deficient spinel ZnMn2O4 [26], and prussian blue analogues (PBAs) [27–29]. Among these materials, PBAs are nontoxic, inexpensive, scalable, and easy to synthesize. PBAs as cathode materials in metal-ion batteries have demonstrated long cycle life, high-rate capability, and high coulombic efficiency [30, 31].

In most cases, research attention is focused on new battery technology with high energy density and high power density but hardly on the environmental aspects. Each year, thousands of tons of batteries end up without recycling, containing heavy metals that can harm the environment. Therefore, the development of batteries which are bio-degradable or made from sustainable materials is in our opinion highly desirable. In addition, bio-batteries could have various applications in smartphones, bio-medical devices, and wearable electronics.

Recently, we have shown reversible Zn2+ intercalation into PBA nanocubes in bio-ionic liquid-water mixtures [27]. The electrochemical performance is quite good with a flooded half-cell test. In the present work, we have made a “green” battery with zinc powder as anode, bio-degradable ionic liquid choline acetate-water mixtures containing zinc acetate as electrolyte, and PBA as cathode material. The components used in the battery meet the environmental compatibility criteria and safety considerations. Prussian blue is nontoxic as it can be used as an antidote for certain kinds of heavy metal poisoning in medicine. Hexacyanoferrate in PBA does not liberate cyanide compounds until a temperature of 580 °C. The ionic liquid choline acetate is bio-degradable because it is based on vitamin B4. Zinc acetate is a commonly used additive for food supplements. Thus, the battery in combination with all components fully satisfies the desired aspect of environmental compatibility. Most importantly, the first version of this bio-degradable zinc-ion battery exhibits quite a promising electrochemical performance with a cell voltage of ∼1.6 V and a capacity of ∼54 mAh g−1.

Experimental section

Ionic liquid-based electrolyte

The ionic liquid-based electrolyte was prepared by dissolving 1 mol/L zinc acetate (Zn(OAc)2, Sigma-Aldrich, 99%) in the mixtures of choline acetate ([Ch]OAc, IO-LI-TEC, 98%) and water. The amount of [Ch]OAc is 70 wt%. The solution has a viscosity of ∼85 mPa s at room temperature, measured by a Bohlin Gemini 150 rheometer (Malvern Instruments, Ltd. UK).

Synthesis of PBA

Iron (III) hexacyanoferrate (III) nanoparticles were prepared by a solution precipitation method. First, 500 mL of an aqueous solution of 0.1 mol/L K3[Fe(CN)6] was prepared. Then, under continuous stirring, 500 mL of 0.1 mol/L FeCl3 containing 1 mL 32% HCl was added gradually to the above solution. After 30 min, the entire 500-mL FeCl3 solution was added and the suspension gets a brown colour in the beaker. The mixture is then heated to a temperature of 100 °C and maintained for 1 h. The colour of the suspension then changes from brown to dark green. After cooling to ambient temperature, the excess solvent was removed by natural sedimentation. The viscous slurry was washed three times with water to remove residual hydrogen, potassium, and chlorine ions. After washing three times, the suspension has a pH value of 6. The slurry is dried in an oven at a temperature of 100 °C to constant mass, giving a dark green powder.

The following reaction occurs:

$$ {\mathrm{K}}_3\left[\mathrm{Fe}{\left(\mathrm{CN}\right)}_6\right]+{\mathrm{FeCl}}_3\to \mathrm{Fe}\left[\mathrm{Fe}{\left(\mathrm{CN}\right)}_6\right]\downarrow +3\mathrm{KCl} $$

Preparation of PBA cathode

The PBA cathode was prepared by mixing a slurry of 80 wt% PBA, 10 wt% conductive carbon black (C-Nergy Super C65, Imery, Belgium Ltd.), 2 wt% graphite (Timrex SFG 6, Imery, Switzerland Ltd.), and 8 wt% polyvinylidene fluoride (PVDF, Sigma-Aldrich) in N-methylpyrrolidone (NMP, Sigma-Aldrich). The suspension was stirred overnight at room temperature to a homogeneous paste.

Preparation of the zinc anode

The zinc powder gel-like anode was prepared by mixing 80 wt% zinc powder (300 mesh, Grillo zinc oxide, Germany) and 20 wt% electrolyte. The slurry was stirred for 10 min, resulting in a homogeneous paste.

Assembly of the cell

The cathode paste is applied, under weight control, directly on carbon cloth (SGL Carbon) with a weight of 95 g m−2 and a thickness of 0.15 mm. Then, the cathode was dried at a temperature of 100 °C for 30 min. The calculated mass loading of PBA was ∼20 mg cm−2. A glass fiber separator (Hollingsworth & Vose GmbH, Germany), wetted with several drops of the electrolyte, was placed between the cathode and the anode. The button cell (CR2032) was assembled by means of a crimping machine (MTI, type MSK-110) with a pressure of 50 kg cm−2.

Characterization

The electrochemical measurement was performed with a VersaStat III (Princeton Applied Research) potentiostat. PBA was used as the working electrode and a zinc anode as counter and reference electrodes, respectively. The structure and morphology of the synthesized PBA were investigated by scanning electron microscopy (SEM, JSM 7610, JEOL) and transmission electron microscopy (TEM, JEM-2100, JEOL). Thermogravimetric measurements (TG) were performed on a TGA 951 analyzer (TA Instruments). The sample was heated from 25 to 500 °C at a heating rate of 10 K/min in a nitrogen atmosphere. The crystal structure was examined by X-ray diffraction (XRD, Empyrean, PANalytical). The content of the metallic elements (K and Fe) was determined by inductively coupled plasma-optical emission spectroscopy (ICP-OES, Vista MPX, Varian), and C, H, O, and N were determined by elemental analysis (Vario EL, Elementar).

Results and discussion

According to the chemical reaction, the formula of the synthesized PBA should be FeFe(CN)6. Elemental analysis (C, H, O N) and ICP-OES results show a K/Fe/CN ratio of 0.01:2.18:6, suggesting that Fe3+ was partially reduced to Fe2+ during the co-precipitation. The synthesized PBA could be mixtures of Fe(III)[Fe(III)(CN)6], Fe(III)4[Fe(II)(CN)6]3, and Fe(II)3[Fe(III)(CN)6]2. We simplified the formula of synthesized PBA as FeFe(CN)6 for short. The TGA curve in Fig. 1a shows a mass loss of 13.3% at temperatures below 150 °C, which corresponds to 2.5 water molecules per mol FeFe(CN)6. Therefore, the formula can be expressed as Fe[Fe(CN)6]·2.5 H2O. The XRD pattern of the PBA is presented in Fig. 1b, showing that the synthesized PBA is crystalline, and all diffraction peaks are consistent with a face-centered cubic structure (JCPDS no. 73-0687). SEM and TEM images of the synthesized FeFe(CN)6 particle show a uniform cubic-shaped structure with an average size of ∼300 nm (Fig. 1c–f).

Fig. 1
figure 1

Characterization of the as-prepared PBA by a TGA; b XRD; c, d SEM; and e, f TEM

Full size image

The standard potential of Fe(III)[Fe(III)(CN)6] is 1.168 V vs. SHE in an aqueous solution [32]. The zinc anode has an electrode potential of −0.76 V under standard conditions. Thus, the maximum theoretical cell voltage is 1.93 V under standard conditions. Our Zn/PBA cell has an average open circuit potential (OCP) of 1.62 V in ionic liquid-water mixtures and can provide sufficient energy to drive a digital clock (Fig. 2a, b). The voltage profile of the cathode during galvanostatic discharging at a current density of 0.1 mA cm−2 (0.3 mA, standard discharge current is usually 0.2 mA for a button cell) is shown in Fig. 2c. It seems that the two processes occur during discharging. The first ends at about 1 V, and the second starts. The two Fe(III) ions in FeFe(CN)6 have different electroreactivities. The high-spin Fe(III) coordinated with N atoms of C≡N ligands is more reactive than the low-spin Fe(III) coordinated with C atoms of C≡N ligands. The two steps could be related to the reduction of the two Fe(III), respectively. The initial specific capacity is ∼54 mAh g−1 (PBA) and the specific energy density is 50 Wh kg−1 (PBA). The galvanostatic discharge and charge voltage profiles of the PBA at a current density of 0.1 mA cm2 between +1.8 and +0.5 V for 50 cycles are displayed in Fig. 2d. The cut-off charge voltage of the Zn/PBA cell was set at +1.8 V, above which side reactions (e.g. oxygen evolution) could occur, resulting in reduced coulombic efficiencies. The capacity of each cycle is ∼50 mAh g−1 and there is almost no capacity fading after 50 cycles. A coulombic efficiency of 99.5% is achieved (Fig. 2e), except for the first cycles. During storage, the loss of voltage and capacity of batteries due to self-discharge are inevitable. The synthesized FeFe(CN)6 in its discharged state could react with the electrolyte or self-decompose, resulting in a drop of the capacity. Therefore, we investigated changes in cell voltage and capacity over a period of 8 weeks at room temperature (Fig. 2f). The OCP of the Zn/PBA cell drops by 0.1 V in 5 weeks and remains quite stable after 6 weeks. The measured capacity fades by 5% in 8 weeks. The results indicate that the PBA might react with the electrolyte at the electrode/electrolyte interface and slightly reduce the voltage and capacity.

Fig. 2
figure 2

a, b Photographs of the freshly prepared Zn/PBA button cell and its application in a digital clock. c Voltage profile during discharging at a current density of 0.1 mA cm−2. d Galvanostatic charge and discharge curves of the cell between +1.8 and +0.5 V at a current density of 0.1 mA cm−2. e Coulombic efficiency of the cell. f Self-discharge test by measuring the potential and specific capacity of the cell over 8 weeks

Full size image

The button cell was disassembled after fully discharged at a current density of 0.1 mA cm−2 into its components, which visually do not show an obvious damage (Fig. 3). The cathode material is well impregnated with the electrolyte. The dark blue-coloured PBA cathode shows that the electrochemical reduction reaction has taken place as the oxidized Fe(III)[Fe(III)(CN)6] is dark green. The zinc gel-like anode is smooth and has no indication of shape change and dendritic growth. The cathode material was peeled off for analysis and washed with acetone for three times. The SEM image and energy dispersive X-ray (EDX) analysis were shown in Fig. 4. In comparison to the SEM image in Fig. 1c, the particle sizes in Fig. 4 exhibit low volume expansion upon intercalating of the Zn ion. The PB cathode materials have open-framework structure with large channels, which are able to accommodate volume expansion during cycling processes. Therefore, they have demonstrated enhanced electrochemical performance with good life cycle and high-rate capability as cathode materials for battery applications [30, 31]. The EDX analysis shows the presence of the elements fluorine, aluminum, silicon, and chlorine in the cathode material in addition to the previously determined elements in the PBA. The presence of fluorine is due to the use of PVDF as a binder, while the existence of chlorine indicates a residue from the PBA production. The existence of silicon is due to contamination with glass fibers, which takes place during the removal of the cathode material from the separator despite of the greatest care. The presence of aluminum is due to the use of aluminum foil as sample packaging. The most interesting value is the molar ratio of Fe/Zn = 4.4 (an average result of three rectangular areas (∼2 × 2 μm) as shown in Fig. 4), allowing the conclusion that 0.45 mol zinc is intercalated in 1 mol PBA. A molar ratio of Fe/Zn = 2 would be expected if the two Fe(III) are fully reduced. The mechanism of the partial intercalation of zinc in the PBA can result from the following: (i) only the high-spin Fe(III) coordinated with N atoms of C≡N ligands can be electroreduced and (ii) the PBA nanoparticles containing spherical agglomerates inhibit the intercalation of the Zn(II) ions. Therefore, it is conceivable to pursue a new approach in the PBA production, which has more favorable PBA nanostructures for the intercalation.

Fig. 3
figure 3

a The Zn/PBA cell with zinc as anode, glass fiber as separator, PBA as cathode, and carbon cloth as current collector before assembly. b Disassembled cell after 50 charge and discharge cycles

Full size image
Fig. 4
figure 4

SEM-EDX of fully discharged PBA cathode at a current density of 0.1 mA cm−2

Full size image

Conclusion

We have presented the first results on a bio-degradable zinc-ion battery with prussian blue analogue as cathode material, glass fiber as separator, a bio-degradable ionic liquid-water mixture as electrolyte, and zinc as anode. The entire battery fulfills the aspect of environmental compatibility. The Zn/PBA battery has an open circuit voltage of 1.6 V and a specific capacity of 54 mAh g−1. The battery shows stable charge-discharge profiles with a coulombic efficiency of 99.5%. The present battery can afford an energy density of 50 Wh kg−1 comparable to lead–acid and Ni–MH batteries. Thus, bio-degradable zinc-ion batteries are potentially interesting in many respects and might be an exciting topic for the future.