Chemical Inhibition Method to Synthesize Highly ... - ACS Publications

Oct 31, 2016 - KEYWORDS: sodium ion batteries, Prussian blue analogues, cathode materials, ... Na-ion batteries are particularly well suited for large...
0 downloads 0 Views 8MB Size
Download PDF
Research Article www.acsami.org

Chemical Inhibition Method to Synthesize Highly Crystalline Prussian Blue Analogs for Sodium-Ion Battery Cathodes Renjie Chen,*,†,‡ Yongxin Huang,† Man Xie,† Ziheng Wang,† Yusheng Ye,† Li Li,†,‡ and Feng Wu†,‡ †

School of Materials Science & Engineering, Beijing Key Laboratory of Environmental Science and Engineering, Beijing Institute of Technology, Beijing 100081, PR China ‡ Collaborative Innovation Center of Electric Vehicles in Beijing, Beijing 100081, PR China

Downloaded via UNIV OF TEXAS AT EL PASO on November 2, 2018 at 12:40:06 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: The nucleation rate plays a critical role in the synthesis of Prussian blue analogs. Rapid precipitation may lead to a large number of vacancies and a large amount of interstitial water in the material, resulting in poor electrochemical performance in batteries. Hence, sodium citrate is used to compete with [Fe(CN)6]4− to slow down the coordination rates of Ni2+ and Mn2+ ions with ferrous cyanide ions. The feasibility of the experiment is also confirmed by theoretical analysis. Benefiting from stable crystal structure and the removal of interstitial water, the as-prepared Na2NixMnyFe(CN)6 sample exhibits a high reversible capacity of 150 mA h g−1. In addition, a high rate performance of 77 mA h g−1 is achieved at a current density of 1600 mA g−1. Most noteworthy, the Coulombic efficiency and specific capacity gradually increase in the first few cycles, which can be ascribed to the formation of a passivation layer on the surface of the electrode. Continuous testing in an electrolyte solution of 1 M NaPF6 dissolved in sulfone reveals that the presence of a passivation film is very important for the stability of the electrode. KEYWORDS: sodium ion batteries, Prussian blue analogues, cathode materials, nucleation rate, electrolyte, kinetic activation

1. INTRODUCTION Advanced energy storage systems are currently being developed to mitigate the energy crisis.1,2 Owing to the discontinuous supply of clean energy sources, which include solar, wind, and tidal energy, the energy harvested must be stored for practical application of these technologies.3 Because of the advantages in terms of price and accessibility of the raw materials required, Na-ion batteries are particularly well suited for large-scale applications.4,5 However, despite these factors and other favorable properties, low theoretical capacity and high standard electrode potential limit their use. In terms of anode materials and electrolytes used in Na-ion batteries, there are several recent reports of suitable candidates for industrial-scale application. For example, hard carbon6 and Na2Ti3O77 are suitable anode materials; and 1 mol L−1 NaPF6 in ethylene carbonate (EC) and diethyl carbonate (DEC)8 in a 1:1 volume ratio, and 1 mol L−1 NaClO4 in propylene carbonate7 can be used as the electrolyte. It is worth mentioning that the low cost pyrolyzed anthracite anode provides good prospects for practical application of sodium ion batteries.9 However, appropriate cathode materials that have high capacity and cycling stability for Na-ion batteries are lacking. Recently, Mu et al. developed an O3-type layered metal oxide cathode, which exhibited high stability in air and outstanding electrochemical properties.10 The fuel cells assembled by the carbon anodes and oxide cathodes delivered practical energy density of 100 Wh kg−1. © 2016 American Chemical Society

Metal−organic frameworks (MOFs) have been intensively investigated for energy applications.11 As a cathode material in Na-ion batteries, Prussian blue (Fe4[Fe(CN)6]3), which was first synthesized in 1978, is particularly promising.12 Prussian blue has an open three-dimensional framework with a capacity of 95 mA h g−1 and a high Coulombic efficiency of nearly 100%.13 However, poor cycling performance and low specific capacity limit its practical application. Accordingly, analogs of Prussian blue (NaxMFe(CN)6; M refers to transition metal elements in the fourth period) have been widely explored as high-capacity Na-storage cathodes. Among the different analogs investigated, sodium manganese(II) hexacyanoferrates (II) (MnHCF) have attracted the most attention. A MnHCF thin film reported by Matsuda et al.14 exhibited a high discharge capacity (109 mA h g−1), high average discharge voltage (3.4 V), good cyclability (90% after 100 cycles), and high Coulombic efficiency (above 95%) at 0.5 °C. However, a main shortcoming of MnHCF thin film is the small yield, which restricts the practical application. Although the simple solution precipitation reaction can produce a large amount of MnHCF materials, vacancies and interstitial water will be introduced into the materials, resulting in collapse of the crystal structure.15 Recently, Song et al.16 developed a method to remove the Received: August 29, 2016 Accepted: October 31, 2016 Published: October 31, 2016 31669

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces

To avoid the light-induced redox action of [Fe(CN)6]4−/3−, synthesis was conducted in a dark environment. 2.1.2. Synthesis of m-PBMN. In this method, instead of direct mixing, the reagents were added in a dropwise manner. A mixture of 20 mL of 0.034 M NiCl2·6H2O and 80 mL of 0.034 M MnCl2·4H2O was added dropwise into 100 mL of 0.04 M Na4Fe(CN)6·10H2O at a rate of 1 mL min−1 under continuous stirring. 2.1.3. Synthesis of s-PBMN. An additional 0.876 and 3.504 g of sodium citrate as a reaction rate inhibitor was added to the precursor solutions of NiCl2·6H2O and MnCl2·4H2O, respectively. Mixing was performed in a dropwise manner as previously described. 2.2. Structural Characterizations. Structural characterization of the as-prepared materials was carried out by powder X-ray diffraction (Hitachi Rigaku-D/Max-2550 PC) with monochromatic Cu Kα radiation. The micrometer-scale morphology and size of the particles were investigated by scanning electron microscopy (Hitachi, S-4800) with an accelerating voltage of 5.0 kV. Elemental mapping images were obtained using energy dispersive X-ray spectroscopy (Thermo Fisher, NS7EDS). The detailed microstructure was investigated using transmission electron microscopy (JEOL, JEM-2100F) with an accelerating voltage of 20 kV. The chemical compositions of the samples were measured using inductively coupled plasma atomic emission spectroscopy (SVISTA-MPX) for metallic elements and elemental analysis (Vario EL Cube) for nonmetallic elements. Thermogravimetric analysis and differential scanning calorimetry were conducted on an Netzsch STA449F3 instrument at a heating rate of 5 °C min−1 under an Ar environment. Fourier transform infrared spectra (ATR-Ge) were recorded in the range of 400−4000 cm−1 to detect cyanide ligands. X-ray photoelectron spectroscopy (PHI Quantera-II) was performed with a monochromatic Al Kα radiation source. Raman spectra were recorded on a Raman spectrometer (JY Labram HR 800). 2.3. Electrochemical Measurements. Electrodes were prepared by slurring 70 wt % active substrate (r-PBMN, d-PBMN, or s-PBMN), 20 wt % acetylene black, and 10 wt % polyvinylidene fluoride in Nmethyl-2-pyrrolidone, and then coating the mixture onto aluminum foil. The mass loading of active materials on electrode is approximately to 2 mg cm−2. After vacuum drying at 80 °C for about 12 h, poleshaped pieces were punched and weighed. The electrochemical performance of the sample was tested in 2032-type cells assembled in an Ar-filled glovebox with water and oxygen contents lower than 1 ppm. The Na-ion half-cell consisted of a Na-metal anode, a separator (glass fiber, Whatman), and the PBMN cathode. Then, 1.0 M NaPF6 dissolved in EC and DEC at a 1:1 volume ratio was used as the electrolyte. Another electrolyte was made with 1.0 M NaPF6 dissolved in tetramethylene sulfone (TMS) and p-toluenesulfonyl isocyanate (PTSI) at a 95:5 volume ratio. Cyclic voltammetry (CV) and Nyquist plots were carried out on a CHI 660e electrochemical workstation (Shanghai Chenhua Instrument Co. Ltd., Shanghai, China). Galvanostatic charge−discharge tests were performed using a battery test system (LAND cycler; Wuhan Kingnuo Electronics Co. Ltd., Wuhan, China). The potentials quoted throughout this paper are in reference to the Na/Na+ couple. All electrochemical experiments were carried out at room temperature. 2.4. Theoretical Simulation. Binding energy calculations were obtained using the Gaussian 09w package. The model chemistry of B3LYP and the basis set of 6-311+G (d) were adopted to achieve an accurate simulation of stable energy and optimal structure.22,23 The binding energies (Be) of ligand ions and metal ions can be calculated according to the following equation:

coordinating water in MnHCF. The obtained precipitate was dried at 100 °C in vacuum for 30 h. When the dehydrated sample was fully discharged, the material exhibited a new rhombohedral phase that showed only one pair of redox peaks at 3.44 and 3.52 V and had a high reversible discharge capacity of 150 mA h g−1. Moreover, many other efforts have been undertaken to improve the electrochemical performance of MnHCF. For instance, this has been attempted via doping two different kinds of transition metal (TM) ions in hexacyanoferrate or via retarding the nucleation rate.17,18 The formation of a solid solution can be interpreted as the replacement of the ion position without a change in the crystal structure. In Prussian blue materials, some Fe3+ ions can be replaced by two or more TM ions in the fourth period. These TM ions have different electrochemical properties from each other, and can enhance the capacity and stabilize the structure. Among them, nickel ferricyanide is regarded as a zero-strain insertion material, which can provide a firm skeleton.19 Therefore, nickel and manganese can be codoped into hexacyanoferrate precursor to obtain high capacity and cycling stability. By controlling the coprecipitation reaction rate, TM ions can be uniformly doped into Prussian blue compounds.15,20−22 The solid solution of doped Prussian blue analogs exhibits a well-defined, facecentered-cubic structure and good electrochemical performance. In terms of Na content, a slow nucleation rate of Prussian blue and its analogs is necessary to achieve a high Na content.18 However, reducing the nucleation rate remains very challenging. In this work, the novel Prussian blue analog with two different TM ion dopants was synthesized by a coprecipitation method with slow nucleation rate. A fraction of the Mn2+ ions was substituted by electrochemically inert Ni2+ ions, which further act to support the framework,17,21 to form the new analog (denoted PBMN) with a stable cubic crystal. The morphology of the particles becomes regularly shaped upon addition of sodium citrate because the competing reaction between citrate ions and ferrous cyanide ions slows the nucleation rate. The as-prepared sample obtained by slow precipitation showed a high discharge capacity (120 mA h g−1) and good cycling stability (after 400 cycles it retained 91% of the initial capacity), which can be attributed to the high content of Na+ ions (as high as 1.87 per formula unit) and low content of coordinating water in the material.

2. EXPERIMENTAL SECTION 2.1. Materials Synthesis. All the materials were purchased from Beijing Chemical Reagent Factory and were used without further purification. The sample prepared by the direct mixing of precursor solutions reacted rapidly and is designated as r-PBMN. Dropwise mixing was used to control the nucleation rate and the samples formed via this method are designated as m-PBMN. In addition to this physical method, the reaction rate was further controlled using chemical additives and the corresponding samples are designated as s-PBMN. 2.1.1. Synthesis of r-PBMN. PBMN was synthesized by a simple precipitation method. Briefly, the precursors, which were dissolved in aqueous media, were directly mixed in 200 mL of deionized water under vigorous stirring. The reaction was then carried out at room temperature for 6 h. The precursor solutions used were 20 mL of 0.034 M NiCl2·6H2O, 80 mL of 0.034 M MnCl2·4H2O, and 100 mL of 0.04 M Na4Fe(CN)6·10H2O. After aging for 24 h, the as-prepared light green precipitates were separated by centrifugation, washed with deionized water and ethanol, and then dried under vacuum at 100 °C.

Be = E − E0 − EA n +

(1)

where E, E0, and EAn+ represent the single-point energies of the complex, the organic ligands, and the metal ion in aqueous solution, respectively. The ligands are either the citric acid (CA) radical or ferrous cyanide (FC), and the metal ions are either Ni2+ or Mn2+. The ligand configuration is that of the steady state with metal ions rather than in isolation. 31670

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces

Na1.87Ni0.05Mn0.95[Fe(CN)6]0.98□0.02·4.06H2O, respectively. Hence, the color difference is mainly caused by the different Ni and Mn contents in each sample. The vacancy numbers of the three samples become very low with the decreasing nucleation rate. In particular, the Na content of s-PBMN reaches 1.87 per formula unit. In the Raman spectra in Figure S2, the Na-ion content is higher for a slower nucleation rate. A schematic of the growth processes of s-PBMN particles is shown in Figure 1b. Submicrometer-sized cubes form via a three-step reaction that involves the reaction of ferrous cyanide groups with Ni2+ and Mn2+ ions, formation of submicrometersized cubic nuclei, and the growth and stacking of crystal nuclei. The first step is the rate-determining step and [Fe(CN)6] vacancies and coordinating water also form during the first step owing to the rapid nucleation rate. Differences between the three samples can also be observed from their morphology. As displayed in Figure 2a−c, when the nucleation rate was slow, the morphology of the particles was more regular. In the transmission electron microscopy (TEM) images (Figure 2d−f), the r-PBMN sample comprises fine irregular particles that have undergone agglomeration. These particles are relatively large, ranging in size from 50 to 200 nm. Although the m-PBMN particles with a size of 100−150 nm are more regular in shape, half-formed cubes can still be observed. In highly crystalline s-PBMN, standard cubic particles with sharp edges and flat surfaces can be observed. Owing to the growth of small particles promoted by the slow nucleation rate, the size of s-PBMN particles was large (i.e., 350−450 nm). In this case, no zeolitic or crystalline water was present in the structure. The uniform distribution of Mn, Fe, Ni, and Na in the samples was demonstrated by energy dispersive X-ray spectroscopy (EDX) mapping (Figure 2g). In addition to the differences in morphology, there are differences in other physical properties between the samples. First, the crystal structures of the three samples were discussed, in particular, that of s-PBMN. As shown in Figure 3a, powder X-ray diffraction (XRD) patterns for the s-PBMN, m-PBMN,

3. RESULTS AND DISCUSSION 3.1. Physical and Chemical Properties. Differences between the samples can be observed with the naked eye. Photographs of r-PBMN, m-PBMN, and s-PBMN nanoparticles, which were formed under different nucleation rates, are shown in Figure 1a. The color gradually becomes shallow

Figure 1. (a) Photographs of r-PBMN, m-PBMN, and s-PBMN. (b) Schematic of the slow nucleation and nanocube growth process.

with the increasing nucleation rate. It can be speculated that the nucleation rate determines the elemental composition of the material. According to the chemical composition and thermogravimetric analysis curves for each product (Table S1 and Figure S1), the chemical formulas of r-PBMN, m-PBMN, and s-PBMN samples can be calculated as Na1.51Ni0.07Mn0.93[Fe(CN)6]0.89□0.11·4.98H2O (□ = Fe(CN)6 vacancy), Na1.68Ni0.06Mn0.94[Fe(CN)6]0.93□0.07·4.35H2O, and

Figure 2. (a−c) Scanning electron microscopy images of r-PBMN, m-PBMN, and s-PBMN, respectively. (d−f) Transmission electron microscopy images of r-PBMN, m-PBMN, and s-PBMN, respectively. (g) EDS maps of s-PBMN particles, demonstrating an even distribution of Mn, Fe, Ni, and Na elements. 31671

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces

Figure 3. (a) X-ray diffraction patterns of r-PBMN, m-PBMN, and s-PBMN. (b) Schematic of the crystal structure of the Prussian blue analogs.

Figure 4. Electrochemical performance of r-PBMN, m-PBMN, and s-PBMN in the voltage range of 2.0−4.0 V. (a) Specific capacities and (b) Coulombic efficiencies at a current density of 50 mA g−1. (c) Rate performance. (d,e) Cycling performance and galvanostatic charge and discharge curves of the s-PBMN electrode at a current density of 100 mA g−1, respectively. (f) Schematic of the structural stability of s-PBMN and r-PBMN upon insertion of Na+ ions.

and r-PBMN samples are characteristic of the face-centered cubic structure of Prussian blue (space group Fm3m). The lattice parameters are calculated to increase from 10.42 Å for sPBMN to 10.44 Å for m-PBMN and 10.72 Å for r-PBMN, reflecting the differences in nucleation rate between the M atoms (Ni and Mn atoms) and hexacyanoferrate. Because of the greater number of Ni2+ ions (ionic radius = 69 pm) than Mn2+ ions doped in the material (ionic radius = 67 pm) during the rapid nucleation process, larger lattice parameters were obtained for r-PBMN. In contrast, s-PBMN exhibited the minimum lattice parameter with lower content of nickel. However, the s-PBMN sample had the highest crystallinity with the sharpest diffraction peaks, with three characteristic peaks at 2θ = 23.8° (yellow), 38.1° (blue), and 48.6° (gray), corresponding to the (220), (420), and (440) planes of the cubic crystalline, respectively. As shown in the schematic of PBMN in Figure 3b, Fe2+ ions and TM ions occupy alternate corners of a cube of cornershared octahedra bridged by linear (CN) groups; thus, PBMN can be regarded as a class of MOF. High-spin TM ions bond with N atoms and low-spin Fe2+ ions bond with C atoms, forming open faces and large channels allowing insertion and extraction of Na+ ions. The Fourier transform infrared (FT-IR) spectrum (Figure S3) is also consistent with the MOF structure; however, it also confirms the presence of water in the vacancies.

The valences of the Fe, Ni, and Mn ions in s-PBMN were determined by X-ray photoelectron spectroscopy (XPS) analysis. The peak distributions displayed in Figure S4a between 707.2 and 720.0 eV are attributed to Fe2+ 2p3/2 and Fe2+ 2p1/2 in [Fe(CN)6]4−, respectively. Ni in the s-PBMN material also was detected to be bivalent (Figure S4b). Mn ions in PBMN were previously reported to be multivalent on the basis of overlapping binding energy ranges. However, the XPS in Figure S4c for s-PBMN demonstrates that Mn is nearly entirely bivalent.24 According to the inductively coupled plasma (ICP) analysis, the average valence of Mn ions is 1.99 in sPBMN; thus, s-PBMN can provide more active sites with entirely bivalent Mn. 3.2. Electrochemical Properties. The electrochemical properties of the samples were measured at a current density of 50 mA g−1 in the voltage range of 2.0−4.0 V. As shown in Figure 4a, the initial discharging capacities of r-PBMN, mPBMN, and s-PBMN were 101, 110, and 120 mA h g−1, respectively. The initial Coulombic efficiencies of the three samples were 89.7%, 94.4%, and 98.1%, respectively (Figure 4b). These differences are mainly caused by the different elemental compositions and vacancy contents among the samples. Vacancies not only reduce the number of active sites for Na+ ions storage, but also capture Na+ ions, resulting in an irreversible reaction. The formation of a passivation layer on the electrode surface may also consume Na+ ions, leading to a low initial Coulombic efficiency. Because of its smaller particle 31672

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces size and larger specific surface area, more side reactions may occur in the r-PBMN sample, resulting in its low initial discharging capacity. The Coulombic efficiencies for the three samples are greater than 95% after 100 cycles. Benefiting from the high Na-ion content and low vacancy content, s-PBMN exhibited the best performance among the three samples. After 100 cycles, the s-PBMN electrode retained a high capacity of 116 mA h g−1 and a Coulombic efficiency of more than 98%. It is noteworthy that the specific capacity of s-PBMN slightly increased during the first few cycles; this phenomenon should be explained in future work. The rate performances of the three samples are presented in Figure 4c. The s-PBMN electrode delivered specific capacities of about 120, 117, 111, 104, and 97 mA h g−1 at current densities of 50, 100, 200, 400, and 800 mA g−1, respectively. Even at a high current density of 1600 mA g−1, the electrode still delivered a discharging capacity of 78 mA h g−1. Although both the r-PBMN and m-PBMN electrodes exhibited similar stabilities as the s-PBMN electrode at low current densities, their discharging capacities decreased considerably at high current densities. This behavior reflects the stable structure and the small number of vacancies in s-PBMN. The cycle lifetime is an important evaluation index for Naion batteries. The cycling stability of the s-PBMN electrode at a current density of 100 mA g−1 is shown in Figure 4d. The discharging capacity was initially about 120 mA h g−1 and remained at 110 mA h g−1 after 400 cycles. This electrode also had a high Coulombic efficiency of about 99.5% after 50 cycles. Moreover, capacity fading of the electrode mainly occurred within the first 50 cycles (inset in Figure 4d). Therefore, it can be inferred that structure evolution and adverse side reactions occurred during initial cycles. The outstanding cycling stability can be attributed to the stable polarization voltage during the charging and discharging processes, as shown in Figure 4e. Moreover, the long and flat voltage plateaus in the vicinity of 3.2 V indicated that the main contribution to the capacity is from the Fe2+/3+ redox couple. Although the Mn2+/3+ redox couple also presents electrochemical activity for Na+ ion storage, the corresponding process mainly occurs at high voltage, which provides limited capacity in the voltage range of 2.0−4.0 V. In order to obtain higher capacity, the upper-limit voltage was raised to 4.2 V, which included the redox process of the Mn2+/3+ couple. The charge−discharge curves exhibited obvious platforms at high voltage as shown in Figure S5. However, the enhancing specific capacity of s-PBMN is negligible compared to the electrode operated at narrow voltage range. Meanwhile, the enhancing upper-limit voltage may lead to a negative effect on the stability of the structure. The Na-ion insertion processes for r-PBMN and s-PBMN are depicted in Figure 4f. Because of vacancies and interstitial water in the crystal structure, the insertion of Na+ ions, which have a larger ionic radius, may lead to structural collapse. This, in turn, leads to low specific capacity and poor cycling stability. By contrast, when Na+ ions are inserted into the s-PBMN electrode, the crystal structure remains stable and the active sites can be effectively used. The electrochemical behavior of Na+ ions during insertion and extraction was characterized by cyclic voltammetry (CV), as shown in Figure 5. Two pairs of redox peaks can be observed at 3.07/3.58 V, and 3.49/3.83 V, which correspond to twoelectron transfer: Ni2+0.05Mn2+0.06Mn3+0.89[Fe3+(CN)6]0.98 + 1.87Na+ ↔ Na1.87Ni2+0.05Mn2+0.95[Fe2+(CN)6]0.98. The higher and lower plateaus are ascribed to the Mn3+/Mn2+ and

Figure 5. Cyclic voltammograms of s-PBMN in the voltage range of 2.0−4.0 V.

[Fe3+(CN)6]3−/[Fe2+(CN)6]4− redox couples, respectively. In the first cycle, the CV curves exhibited large polarization voltages between the cathodic and anodic peaks. During the subsequent cycles, the polarization resistance was gradually reduced to a lower level. For example, the polarization voltage of the higher redox couple decreased from 0.38 to 0.28 V, and that of the lower redox couple decreased from 0.51 to 0.41 V. This relates to the activation process of the electrode. There was good reproducibility among the CV curves of repeated scans, demonstrating the excellent structural and electrochemical reversibility of this material during Na+ insertion and extraction. The increase in the discharge capacity in the first few cycles is related to the kinetic activation process. To further investigate this process, Nyquist plots were performed for the s-PBMN electrode before and after the first cycle, and after five cycles. The Nyquist plots curves shown in Figure 6a are composed of a flat semicircle in the high-frequency region and a sloping line in the low-frequency region, which correspond to a chargetransfer process and a semi-infinite Warburg diffusion process, respectively. Before cycling, the charge-transfer resistance of the s-PBMN electrode was about 576.1 Ω. Because of the formation of a passivation film on the electrode surface, the resistance after the first cycle increased considerably to 1055.7 Ω. After five cycles, the interface between the electrode surface and the electrolyte became stable, resulting in a smaller resistance of 663.5 Ω. The capacity of the s-PBMN electrode gradually decreased to a stable value after five cycles. During this activation process, a large number of Na+ ions may be consumed to provide high irreversible capacity, leading to low efficiency.25 However, the passivation layer also has a positive effect. Mn3+ ions in a crystal structure may lead to a severe Jahn−Teller effect, resulting in the dissolution of M2+ ions.24 Hence, the passivation layer effectively prevents the dissolution of Mn2+ ions and improves the cycling stability. The initial specific capacity was 118 mA h g−1. After the second cycle, the specific capacity increased slightly to 120 mA h g−1 and remained stable at this value during subsequent cycles (Figure 6b), which can be considered good electrochemical performance. Prussian blue analog electrodes were previously found to have good film-forming properties that enhance the specific capacity.26 However, some Na+ ions will be consumed at the beginning of cycling, resulting in capacity increase. In addition, the influence of polarization cannot be ignored. To analyze the polarization process in detail, differential capacity (dQ/dV) curves (Figure 6d−f), which provide deeper insight into the electrochemical process, were measured with 31673

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces

Figure 6. Polarization process of the s-PBMN electrode measured in the voltage range of 2.0−4.0 V. (a) Nyquist plots measured from 1 × 105 to 0.01 Hz. (b) Charge−discharge profiles at a current density of 100 mA g−1. (c) Polarization voltages between the [Fe(CN)6]4−/[Fe(CN)6]3− and Mn2+/Mn3+ couples. (d−f) Differential capacity plots of the second, third, and fifth cycles at a current density of 100 mA h g−1.

Figure 7. (a) Cycling performance and (b) charge−discharge profiles of the s-PBMN electrode tested in 1.0 M NaPF6 in TMS/PTSI (95:5 v/v) in the voltage range of 2.0−4.2 V.

Figure 8. Calculated binding energies of ligand ions ([Fe(CN)6]4− and C5H5O5COO3−) and metal ions (Ni2+ and Mn2+). Bonds and angles are marked in blue and orange text, respectively.

the first cycle. The changes mainly occurred in the peaks at 3.07 and 3.58 V, which correspond to redox couple 1. It needs to be clarified whether the polarization decrease relates to changes in the electrode before or after cycling. To investigate the effect of the solid-electrolyte interphase (SEI) on battery performance, a novel electrolyte, 1 M NaPF6 in TMS/PTSI (95:5 v/v), was used instead of EC/DEC (1:1 v/ v).27 Because the SEI on the cathode surface is easier to

the s-PBMN electrode. In accordance with the CV curves, the dQ/dV plots have two pairs of redox peaks, indicating a multistep transfer process of Na+ ions. The polarization voltages between the low-spin [Fe3+(CN)6]3−/[Fe2+(CN)6]4− couple (couple 1) and high-spin Mn3+/Mn2+ couple (couple 2) can be obtained by simple calculation and are shown in Figure 6c. The polarization voltages of the two redox couples in the fifth cycle decreased by about 100 mV compared with those in 31674

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces

Table 1. Binding Energies of Ions (Ni2+ and Mn2+) to Organic Ligands ([Fe(CN)6]4− and C5H5O5COO3−) in Aqueous Solution [Fe(CN)6] −6Ni [Fe(CN)6]4−−6Mn2+ C5H5O5COO3−−Ni2+ C5H5O5COO3−−Mn2+ 4

a

2+

E (hartreea)

E0 (hartree)

EAn+ (hartree)

Be (hartree)

−10868.3792 −8725.0656 −2266.6346 −1963.4203

−1821.0832 −1821.0832 −758.5490 −812.6212

−1507.8134 −1150.5486 −1507.8134 −1150.5486

−0.4156 −0.6908 −0.2722 −0.2505

Hartree is the atomic unit of energy. 1 hartree = 27.21 eV = 2625.5 kJ mol−1.

performance. After an initial kinetic activation process, the sPBMN electrode exhibited a high specific capacity of 120 mA h g−1 and retained a capacity of 110 mA h g−1 after 400 cycles under a current density of 100 mA g−1. It was inferred that a passive film forms on the surface of this Prussian blue analog, which improves the cycling stability of the electrode by preventing the dissolution of Mn2+ ions. Calculations revealed the different binding affinities of different transition metal ions with citric acid, explaining the different rates of nucleation. Therefore, nucleation rate inhibitors should be developed to improve the crystallinity of Prussian blue analogs.

produce under high voltage, the cutoff voltage was increased to 4.2 V. The s-PBMN electrodes exhibited high initial capacity in both the new and original electrolyte. However, the Coulombic efficiency of the s-PBMN electrode was very low in the first 15 cycles, as shown in Figure 7a. Because of the presence of a welldefined SEI film in the sulfone electrolyte, some Na+ ions will be consumed during cycling, which may reduce the charge and discharge efficiency of the electrode. Meanwhile, the initial charge capacity of cell can reach 175 mA h g−1. After 15 cycles, the Coulombic efficiency remained above 90%, and after 25 cycles, the specific capacity was 90% of the initial capacity. We note that although the increased cutoff charge voltage of 4.2 V can cause a small gain in capacity, it also reduces the cycling stability of the material. As shown in Figure 7b, when the voltage range was extended to 2.0−4.2 V, three pairs of redox plateaus were present. By contrast, when the electrode operated at 2.0−4.0 V, only two pairs of redox plateaus were observed. The third oxidative plateau became weaker with the increasing number of cycles. Therefore, the additional Na+ ions being inserted might damage the crystal structure of the material. In addition, the third reductive plateau might have been caused by the side reactions in novel sulfone-based electrolyte, which also led to the poor cycling stability. 3.3. Theoretical Analysis. The stable configurations of the complexes and their corresponding energies are shown in Figure 8.28 The binding energies for single Ni2+−[Fe(CN)6]4−, Ni 2+ −C 5 H 5 O 5 COO 3− , Mn 2+ −[Fe(CN) 6 ] 4− , and Mn 2+ − C 5 H 5 O 5 COO 3− were calculated as −0.4156, −0.2722, −0.6908, and −0.2505 hartree (Table 1), respectively. Because the lower binding energy facilitates product formation, complexes of ferrous cyanide are more stable than citrate complexes. Although some Ni2+ and Mn2+ ions may have preferentially bonded with citrate, they can subsequently dissociate from the citrate and recombine with ferrous cyanide. Therefore, this reduces the rate of precipitation. The increased number of Mn2+ ions that combined with ferrous cyanide can be explained as follows. When metal ions preferentially combine with carboxyl groups on the citric acid radical, the bond length of the formed −O−Mn2+−O− group (2.00 Å) is longer than that of the −O−Ni2+−O− group (1.83 Å), which indicates that the Mn2+ ion has a stronger ability to recombine with ferrous cyanide.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.6b10884. ICP data, TG and DTG curves, Raman spectroscopy, FT-IR spectra, and XPS spectrum for r-PBMN, mPBMN, and s-PBMN samples (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Tel./Fax: +86-10-68451429. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the National Key Research and Development Program (2016YFB0901501), the National Natural Science Foundation of China (21373028), the Joint Funds of the National Natural Science Foundation of China (U1564206), Major achievements Transformation Project for Central University in Beijing and Beijing Science and Technology Project (D151100003015001).

■

REFERENCES

(1) Dunn, B.; Kamath, H.; Tarascon, J. M. Electrical Energy Storage for the Grid: A Battery of Choices. Science 2011, 334, 928−935. (2) Thackeray, M. M.; Wolverton, C.; Isaacs, E. D. Electrical Energy Storage for TransportationApproaching the Limits of, and Going Beyond, Lithium-Ion Batteries. Energy Environ. Sci. 2012, 5, 7854− 7863. (3) Wessells, C. D.; McDowell, M. T.; Peddada, S. V.; Pasta, M.; Huggins, R. A.; Cui, Y. Tunable Reaction Potentials in Open Framework Nanoparticle Battery Electrodes for Grid-Scale Energy Storage. ACS Nano 2012, 6, 1688−1694. (4) Pan, H.; Hu, Y.; Chen, L. Room-Temperature Stationary SodiumIon Batteries for Large-Scale Electric Energy Storage. Energy Environ. Sci. 2013, 6, 2338−2360. (5) Slater, M. D.; Kim, D.; Lee, E.; Johnson, C. S. Sodium-Ion Batteries. Adv. Funct. Mater. 2013, 23, 947−958. (6) Komaba, S.; Murata, W.; Ishikawa, T.; Yabuuchi, N.; Ozeki, T.; Nakayama, T.; Ogata, A.; Gotoh, K.; Fujiwara, K. Electrochemical Na Insertion and Solid Electrolyte Interphase for Hard-Carbon Electrodes

4. CONCLUSIONS The quality of Prussian blue and its analogs is determined by their nucleation rate. Therefore, in the present study, we controlled the rate of nucleation in experiments and confirmed the validity of the results through theoretical calculations. Because citric acid can combine with TM ions, which decreases the nucleation rate of ferrous cyanide and TM ions, the competitive relationship between these two ligands in solution reduces the number of vacancies and the amount of interstitial water in the product. The s-PBMN sample has high crystallinity, regular morphology, and good electrochemical 31675

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676

Research Article

ACS Applied Materials & Interfaces and Application to Na-Ion Batteries. Adv. Funct. Mater. 2011, 21, 3859−3867. (7) Rudola, A.; Saravanan, K.; Mason, C. W.; Balaya, P. Na2Ti3O7: an Intercalation Based Anode for Sodium-Ion Battery Applications. J. Mater. Chem. A 2013, 1, 2653−2662. (8) Ponrouch, A.; Marchante, E.; Courty, M.; Tarascon, J. M.; Palacin, M. R. In Search of an Optimized Electrolyte for Na-Ion Batteries. Energy Environ. Sci. 2012, 5, 8572−8583. (9) Li, Y.; Hu, Y.; Qi, X.; Rong, X.; Li, H.; Huang, X.; Chen, L. Advanced Sodium-Ion Batteries Using Superior Low Cost Pyrolyzed Anthracite Anode: Towards Practical Applications. Energy Storage Mater. 2016, 5, 191−197. (10) Mu, L.; Xu, S.; Li, Y.; Hu, Y.; Li, H.; Chen, L.; Huang, X. Prototype Sodium-Ion Batteries Using an Air-Stable and Co/Ni-Free O3-Layered Metal Oxide Cathode. Adv. Mater. 2015, 27, 6928−6933. (11) Furukawa, S.; Reboul, J.; Diring, S.; Sumida, K.; Kitagawa, S. Structuring of Metal−Organic Frameworks at the Mesoscopic/ Macroscopic Scale. Chem. Soc. Rev. 2014, 43, 5700−5734. (12) Neff, V. D. Electrochemical Oxidation and Reduction of Thin Films of Prussian Blue. J. Electrochem. Soc. 1978, 125, 886−887. (13) Shen, L.; Wang, Z.; Chen, L. Prussian Blues as a Cathode Material for Lithium Ion Batteries. Chem. - Eur. J. 2014, 20, 12559− 12562. (14) Matsuda, T.; Takachi, M.; Moritomo, Y. A Sodium Manganese Ferrocyanide Thin Film for Na-Ion Batteries. Chem. Commun. 2013, 49, 2750−2752. (15) You, Y.; Wu, X.; Yin, Y.; Guo, Y. High-Quality Prussian Blue Crystals as Superior Cathode Materials for Room-Temperature Sodium-Ion Batteries. Energy Environ. Sci. 2014, 7, 1643−1647. (16) Song, J.; Wang, L.; Lu, Y.; Liu, J.; Guo, B.; Xiao, P.; Lee, J. J.; Yang, X.; Henkelman, G.; Goodenough, J. B. Removal of Interstitial H2O in Hexacyanometallates for a Superior Cathode of a Sodium-Ion Battery. J. Am. Chem. Soc. 2015, 137, 2658−2664. (17) Yang, D.; Xu, J.; Liao, X.; He, Y.; Liu, H.; Ma, Z. Structure Optimization of Prussian Blue Analogue Cathode Materials for Advanced Sodium Ion Batteries. Chem. Commun. 2014, 50, 13377− 13380. (18) You, Y.; Yu, X.; Yin, Y.; Nam, K. W.; Guo, Y. Sodium Iron Hexacyanoferrate with High Na Content as a Na-Rich Cathode Material for Na-Ion Batteries. Nano Res. 2015, 8, 117−128. (19) Yu, S.; Li, Y.; Lu, Y.; Xu, B.; Wang, Q.; Yan, M.; Jiang, Y. A Promising Cathode Material of Sodium Iron-Nickel Hexacyanoferrate for Sodium Ion Batteries. J. Power Sources 2015, 275, 45−49. (20) Zhao, F.; Wang, Y.; Xu, X.; Liu, Y.; Song, R.; Lu, G.; Li, Y. Cobalt Hexacyanoferrate Nanoparticles as a High-Rate and UltraStable Supercapacitor Electrode Material. ACS Appl. Mater. Interfaces 2014, 6, 11007−11012. (21) You, Y.; Wu, X.; Yin, Y.; Guo, Y. A Zero-Strain Insertion Cathode Material of Nickel Ferricyanide for Sodium-Ion Batteries. J. Mater. Chem. A 2013, 1, 14061−14065. (22) Engel, N.; Bokarev, S. I.; Suljoti, E.; Garcia-Diez, R.; Lange, K. M.; Atak, K.; Golnak, R.; Kothe, A.; Dantz, M.; Kühn, O.; Aziz, E. F. Chemical Bonding in Aqueous Ferrocyanide: Experimental and Theoretical X-ray Spectroscopic Study. J. Phys. Chem. B 2014, 118, 1555−1563. (23) Park, S. K.; Lee, C. K.; Lee, S. H.; Lee, N. S. Vibrational Analysis of Ferrocyanide Complex Ion Based on Density Functional Force Field. Bull. Korean Chem. Soc. 2002, 23, 253−261. (24) Li, W.; Chou, S.; Wang, J.; Wang, J.; Gu, Q.; Liu, H.; Dou, S. Multifunctional Conducing Polymer Coated Na 1+x MnFe(CN) 6 Cathode for Sodium-Ion Batteries with Superior Performance via a Facile and One-Step Chemistry Approach. Nano Energy 2015, 13, 200−207. (25) Jia, Z.; Wang, J.; Wang, Y. Electrochemical Sodium Storage of Copper Hexacyanoferrate with a Well-Defined Open Framework for Sodium Ion Batteries. RSC Adv. 2014, 4, 22768−22774. (26) Lee, H.; Kim, Y. I.; Park, J. K.; Choi, J. K. Sodium Zinc Hexacyanoferrate with a Well-Defined Open Framework as a Positive

Electrode for Sodium Ion Batteries. Chem. Commun. 2012, 48, 8416− 8418. (27) Wang, C. H.; Yeh, Y. W.; Wongittharom, N.; Wang, Y. C.; Tseng, C. J.; Lee, S. W.; Chang, W. S.; Chang, J. K. Rechargeable Na/ Na0.44MnO2 Cells with Ionic Liquid Electrolytes Containing Various Sodium Solutes. J. Power Sources 2015, 274, 1016−1023. (28) Kareis, C. M.; Lapidus, S. H.; Her, J. H.; Stephens, P. W.; Miller, J. S. Non-Prussian Blue Structures and Magnetic Ordering of Na2MnII[MnII(CN)6] and Na2MnII[MnII(CN)6]·2H2O. J. Am. Chem. Soc. 2012, 134, 2246−2254.

31676

DOI: 10.1021/acsami.6b10884 ACS Appl. Mater. Interfaces 2016, 8, 31669−31676