Abstract
Towards enhancement of the energy density of Li-ion batteries, BiF3 has recently attracted considerable attention as a compelling conversion-type cathode material due to its high theoretical capacity of 302 mAh g−1, average discharge voltage of ca. 3.0 V vs. Li+/Li, the low theoretical volume change of ca. 1.7% upon lithiation, and an intrinsically high oxidative stability. Here we report a facile and scalable synthesis of phase-pure and highly crystalline orthorhombic BiF3 via thermal decomposition of bismuth(III) trifluoroacetate at T = 300 °C under inert atmosphere. The electrochemical measurements of BiF3 in both carbonate (LiPF6-EC/DMC)- and ionic liquid-based (LiFSI-Pyr1,4TFSI) Li-ion electrolytes demonstrated that ionic liquids improve the cyclic stability of BiF3. In particular, BiF3 in 4.3 M LiFSI-Pyr1,4TFSI shows a high initial capacity of 208 mA g−1 and capacity retention of ca. 50% over at least 80 cycles at a current density of 30 mA g−1.
Introduction
At present, the replacement of intercalation-type cathodes by their conversion-type counterparts is pursued as a path for improving the energy density of Li-ion batteries (LiBs)1,2,3,4,5. Among the plethora of conversion-type cathodes investigated, metal fluorides (Fe6,7,8, Cu9,10,11, Bi12,13,14, Ni15) have come into major research spotlight as they possess compelling properties, such as intrinsically high oxidative stability, high lithiation potentials of > 2 V vs. Li+/Li as compared to the voltages of transition metal oxides or sulfides, and high theoretical capacities in the range of 302–712 mAh g–1 2,3,16,17,18. Consequently, in combination with a metallic lithium anode, fluorides may offer energy densities of up to 890–1680 Wh kg−1, which is significantly higher than that of intercalation-type cathodes such as LiFePO4 (578 Wh kg−1), LiNi1/3Mn1/3Co1/3O2 (610 Wh kg−1), and LiCoO2 (546 Wh kg−1). Overall, although the reversibility of conversion reactions of metal fluorides has been demonstrated, most of them possess poor capacity retention and limited cycle life, which is primarily caused by large volume changes and associated deterioration of the mechanical contact between the active material and conductive additive upon lithiation/delithiation.
BiF3 stands out from a range of conversion-type fluorides cathode materials because of its high theoretical capacity of 302 mAh g−1, owing to Bi3+/Bi0 redox process at 3.0 V vs. Li+/Li and low theoretical volume change of ca. 1.7% upon lithiation (Fig. 1a)2,19,20,21. The volume change upon delithiation is comparable to conventional intercalation-type cathodes such as LiFePO4 (ca. −6.8%)22, LiNi0.5Mn0.3Co0.2O2 (ca. −2.2%)23, and LiCoO2 (ca. 1.9%)23. For comparison, extensively studied FeF3 has much higher volume changes of up to 26% for three-electron operation (Fig. 1b)2. Additionally, BiF3 cathodes are characterized by low voltage hysteresis of ca. 0.4 V24, contrary to 1.5–2 V for FeF3 (Fig. 1c)25,26,27,28,29.
This study was motivated by (i) a lack of low-cost and robust synthesis methods of BiF3 and (ii) its well-recognized potential as a high-energy-density cathode material in Li-ion batteries. Traditionally, BiF3 is synthesized from various Bi3+ precursors such as Bi2O3 or BiOCl via treatment with concentrated HF20,30,31. Alternative solution-based approaches employ Bi(NO3)3 or Bi(NO3)(OH)2 and NH4F32,33,34. However, both HF and NH4F fluorine sources are hazardous and highly toxic and the employment of aqueous media often results in the formation of bismuth oxofluorides impurities through a hydrolysis reaction. Harnessing the storage potential of BiF3, at present, it is hindered by its poor capacity retention in carbonate-based Li-ion electrolytes, which is associated with the formation of Li2CO3 containing cathode-electrolyte interfaces (CEIs) upon discharge, catalyzed by metallic Bi0,12,24,35. In the context of minimization of these side reactions, herein, we sought to thoroughly test room-temperature ionic liquids (ILs) as alternative Li-ion electrolytes. Recently, Pasta et al36. demonstrated superior cycling stability for a similar conversion FeF2 cathode, caused by the formation of stable solid electrolyte interphase. Moreover, ILs have been found to provide exceptional long cycling stability for conventional intercalation-type electrodes37,38,39. However, to our knowledge, the electrochemical performance of BiF3 cathode in ILs has not yet been studied.
In this work, we present a novel facile and scalable synthesis synthetic route of BiF3 by the thermal decomposition of non-toxic, safe-to-handle, single-source molecular Bi trifluoroacetate (TFA) precursor under N2 atmosphere, yielding highly crystalline orthorhombic BiF3. The electrochemical performance of BiF3 was thoroughly assessed in both carbonate and IL Li-ion electrolytes. We determined that the BiF3 exhibits high cyclic stability as cathode material in lithium-ion batteries based on 4.3 M LiFSI-Pyr1, 4TFSI IL, delivering a high initial capacity of 208 mAh g–1. Half of this capacity was retained after prolonged operation over 80 cycles at a current density of 30 mA g−1.
Results and discussion
Synthesis and characterization of BiF3
Fig. 2a outlines the synthesis of BiF3 via thermal decomposition of Bi(TFA)3 as a single-source precursor under N2 atmosphere (12 h at 300 °C). Bi(TFA)3 was synthesized according to the procedure given in Ref. 40,41. (see Methods Section and Supplementary Fig. 1 for details). Powder X-ray diffraction (XRD) measurements along with Rietveld refinement42 confirmed the formation of chemically pure, highly crystalline orthorhombic modification of BiF3 (Pnma, space group no. 62, a = 6.5604 Å b = 7.0174 Å, c = 4.8450 Å, ICSD No. 1269, Fig. 2b, c). The R(F) factor is 7% (see Methods section for details). No significant broadening of the reflections associated with crystallite size effects is detected. The powder pattern shows a strong preferred orientation of the crystallites along the [010]-direction, which is in line with the flake morphology of BiF3 powder (mean particle size is 4.9 μm), as shown in Fig. 2d. The BiF3 product is air-stable and only shows minor hydrolysis towards the formation of cubic BiOxF3-2x phase even after several months on air (see Supplementary Fig. 2). In contrast, it was found that if the synthesis is carried out under air or if the precursor is ground under air, the synthesis yields different BiOxF3-2x phases (Supplementary Fig. 3) with 0.4 < x < 0.6 (cubic, Fm-3m, space group no. 225) and x = 1 (tetragonal, P4/nmm, space group no. 129) along with the formation of a white, amorphous sublimate on the quartz tube outside the oven (Supplementary Fig. 4).
a Schematic of the BiF3 synthesis (trifluoroacetic acid and trifluoroacetic anhydride are named as TFAA and TFAH, respectively); b A comparison of the experimental (blue) and calculated (green) powder diffraction pattern of BiF3 together with the difference (cyan) and the background (red), as obtained from the Rietveld refinement with GSAS-II43. c Orthorhombic structure of BiF3. d SEM image of the BiF3 particles.
Since BiF3 is an electrically insulating compound, the synthesized μm-sized BiF3 particles were reasoned to be too large for their employment in electrodes. Consequently, to reduce the particle size and establish sufficient contact with the conductive additives, o-BiF3 was dry ball-milled with multi-walled carbon nanotubes (CNTs) and carbon black (CB) at 800 rpm for 1.5 h under Ar atmosphere. Interestingly, ball-milling causes the transformation of the orthorhombic structure of BiF3 into the cubic one (Fm-3m, space group no. 225, Fig. 3a). These results are in contrast with previous findings by Bervas et al19., who reported the formation of the high-pressure tysonite modification (\(P\bar 3{{{{{{{\mathrm{c1}}}}}}}}\), space group no. 165)44,45 after high-energy ball-milling at 1100 rpm. Obviously, considering that we conducted ball-milling at a lower speed of 800 rpm, the generated pressure was insufficient to transform o-BiF3 into the high-pressure modification. Due to the fact that XRD patterns of ball-milled BiF3 can be assigned to either α-BiF341, or BiOxF3-2x with 0.4 < x < 0.612, two phases cannot be distinguished based on powder XRD analysis because both lattice parameters of the reference compounds and the reflection intensities in the XRD patterns are almost identical. However, EDX measurements indicated that the O content in cubic BiF3 is relatively low (see Fig. 3c, d and Supplementary Fig. 5–9), suggesting that ball-milling yields the formation of α-BiF3 phase with minor O impurities rather than a BiOxF3-2x phase. Slightly enhanced amounts of oxygen in cubic BiF3 as compared to orthorhombic BiF3 probably originate from the presence of adsorbed water on the surface of CNTs and CB conducting additives used for ball-milling. Notably, ball-milling of BiF3 induced significant broadening of XRD reflections, indicating the reduction of crystallite sizes and the presence of considerable strain within the material (Fig. 3a). As the product is not monodisperse, a meaningful quantification of strain and crystallite size was not possible. Importantly, SEM measurements revealed that the mean particle size of BiF3 after ball-milling decreases from 4.9 μm to 1.2 μm and the distribution becomes much narrower from 4.9 to 1.0 μm (Fig. 3b). Also, ball-milling caused the formation of cracks in the BiF3 particles (see Supplementary Fig. 7 and 8), thus allowing conductive additives to intermix well with the BiF3 active material (Fig. 3c). Carbon EDX mapping after ball-milling clearly shows that carbon coats the entire surface of BiF3 particles. Additionally, to further investigate the contact between the active material and conductive additive on a nanoscale level, TEM measurements of BiF3 after ball-milling were performed. Supplementary Fig. 10 shows that the obtained BiF3/C powder consist of a significant amount of BiF3 nanoparticles (ca. 20 nm), which are embedded into a CNT and CB matrix.
Electrochemical performance
For the electrochemical measurements, BiF3 electrodes were prepared by ball-milling of BiF3/C powder with (polyvinylidene difluoride) pVdF binder in (N-Methylpyrrolidone) NMP solvent, and the resulting slurries were cast onto an Al foil current collector via paint brushing, followed by drying at 80 °C for 12 h under vacuum. Coin-type cells were employed for the electrochemical measurements. The cell consisted of a lithium disk as a counter and reference electrode and a BiF3 working electrode. A glass-fiber separator was placed in between the two electrodes and impregnated with Li-ion electrolyte. All the electrochemical tests were performed in the voltage range of 2 – 4 V vs. Li+/Li at a current density of 30 mA g−1.
Firstly, we assessed the electrochemical performance of BiF3 employing a conventional electrolyte (1 M LiPF6-EC/DMC). The charge and discharge voltage profiles, as well as corresponding differential capacity plots (dQ/dV) are summarized in Fig. 4 and rescaled in Supplementary Fig. 13. During the first discharge, two reduction peaks at 2.5 V (R1) and 2.2 V (R2) vs. Li+/Li were observed, which can be attributed to the reduction of BiF314,46 and BiOxF3-2x46. The voltage difference between the two peaks of only 0.3 V indicates that the O content in BiOxF3–2x must be rather low. The O impurities might originate for two reasons: the presence of minor quantities of amorphous BiOxF3–2x in as-prepared BiF3/C powder or/and the fact that residual water from the electrolyte could also hydrolyze BiF3 to form BiOxF3–2x. The differential capacity then keeps decreasing to −750 mAh g–1 V–1 at 2 V vs. Li+/Li (Supplementary Fig. 13). This is presumably associated with the reduction of the carbonate electrolyte catalyzed by metallic Bi0. Of note, carbonate electrolytes, in particular cyclic carbonates, form Li2CO3 containing cathode-electrolyte interfaces (CEIs) upon reduction at 1.6–2 V in the presence of Bi038,52,53. The capacity for the first cycle is 280 mAh g–1, which is close to the theoretical maximum of 302 mAh g−1 for BiF3.
Charge and discharge voltage profiles and associated dQ/dV curves of BiF3 cathode measured in LiPF6 (1 M and 4.3 M)-EC/DMC and LiFSI (1 M and 4.3 M)-Pyr1,4TFSI Li-ion electrolytes for the first (a, e), second (b, f) and tenth (c, g) cycles. Combined dQ/dV plots measured upon cycling of BiF3 cathode in 1 M LiPF6-EC/DMC Li-ion electrolyte for first the 35 cycles; (d) and 4.3 M LiFSI (h) in Pyr1,4TFSI Li-ion electrolytes. The BiF3/Li half cells were cycled at a current density of 30 mA g−1 in the voltage range of 2‒4 V vs. Li+/Li. The electrochemical performance of BiF3 measured in other carbonate-based electrolytes (1 M LiPF6 in EC/DMC + 3 wt-% FEC, EMC and EMC + 3 wt-% FEC) is given in Supplementary Fig. S14 for comparison.
Upon charge, several distinct electrochemical processes occur as follows from the appearance of the oxidation peaks at 2.5 V (O1), 2.8 V (O2) and 3.2 V (O3) vs. Li+/Li. The first small oxidation O1 peak at 2.5 V vs. Li+/Li can be associated with the oxidation of Bi0 to Bi2O3 in the presence of Li2CO3 \(\left( {2Bi^0{{{{{{{\mathrm{ + }}}}}}}}3{{{{{{{\mathrm{Li}}}}}}}}_2CO_3\left( {{{{{{{{\mathrm{CEI}}}}}}}}} \right) \to {{{{{{{\mathrm{Bi}}}}}}}}_2O_3{{{{{{{\mathrm{ + }}}}}}}}3CO_2{{{{{{{\mathrm{ + }}}}}}}}6Li^ + {{{{{{{\mathrm{ + }}}}}}}}6e^ - } \right)\)35. Next, the O2 peak at 2.8 V vs. Li+/Li is caused by the oxidation of Bi forming BiOxF3-2x while the O3 peak at 3.2 V vs. Li+/Li is assigned to BiF3 formation19,21,46. A similar behavior with three distinct oxidation peaks has already been reported for BiOxF3-2x cathodes46. At voltages above 3.9 V vs. Li+/Li, the differential capacity increases slightly again, pointing to electrolyte oxidation. Interestingly, the three oxidation peaks are only present in the first cycles. From the tenth cycle onwards, they merge into two peaks with a concomitant decrease of the discharge capacity down to 152 mAh g–1. Thus, the O1 peak (Bi0/Bi2O3) shifts to higher voltages and then disappears completely, while the O2 peak (Bi0/BiOxF3-2x) also shifts to higher voltages and prevails. The O3 peak (Bi0/BiF3) at 3.2 V vs. Li+/Li moves to slightly lower voltages of 3.0‒3.1 V vs. Li+/Li after 15 cycles and then shifts again to a higher voltage of 3.2–3.3 V vs. Li+/Li onwards. The shift in two directions is possibly caused by two independent phenomena. On the one hand, a shift to higher voltages upon cycling might be associated with loss of electrical contact of BiF3 part due to conversion reaction, eventually increasing the applied current density per electrochemically active sites of BiF3, thus resulting in higher overpotentials. On the other hand, the shift of O3 peak toward lower voltages, followed by its disappearance in the following cycling, can be explained by homogenization of Bi2O3 and BiF3 domains towards the formation of a BiOxF3-2x phase \(( {xBi_2O_3 + (3 - 2x)BiF_3\mathop{\longrightarrow} \limits^{Cycling} 3BiO_xF_{3 - 2x}} )\).
Next, we measured BiF3 employing IL Li-ion electrolyte based on lithium bis(fluorosulfonyl)imide (LiFSI) in 1-butyl-1-methylpyrrolidinium bis(trifluoromethylsulfonyl)imide (Pyr1,4TFSI). As it was previously reported, imidazolium47,48,49, and pyrrolidinium (Pyr)50,51,52,53 IL electrolytes have the advantage of not forming oxygen-containing decomposition products that might help to overcome the partial formation of BiOxF3-2x phases. Contrary to the results obtained with carbonate-based electrolyte, the employment of the IL counterpart resulted in only one reduction peak at 2.4 V (R1) upon initial discharge. Further decrease of the differential capacity at ca. 2 V (Fig. 4e, h) might be associated with the formation of CEI or another side reaction. Yet, the extent of this side reaction is far less pronounced as compared to measurements performed in LiPF6 in EC/DMC electrolyte. Interestingly, upon the subsequent charge, no O1 oxidation peak at 2.5 V vs. Li+/Li (Bi0/Bi2O3) was observed. Only one broad oxidation peak was measured at 3.2 V vs. Li+/Li (O3), which is associated with the oxidation of Bi0 to BiF3. This observation indicates that the employment of oxygen-free IL electrolyte hinders the formation of Li2CO3 and, therefore, its chemical reaction with Bi, yielding Bi2O3.
Next, we assessed the impact of the salt concentration in both carbonate and IL electrolytes on the electrochemical performance of BiF3 cathode. Cells were prepared with 4.3 M LiPF6 in EC/DMC + 3 wt-% FEC and 4.3 M LiFSI in Pyr1,4TFSI electrolytes, accordingly. For effective penetration of the cathode by the highly viscous electrolytes, the prepared coin-type cells were heat-treated at 75 °C for 24 h prior to measurements.
The initial discharge of BiF3 in 4.3 M carbonate electrolyte resulted in a pronounced reduction plateau at 2.6–2.7 V vs. Li+/Li, corresponding to an initial discharge capacity of 481 mAh g–1 (Fig. 4a). This is far higher than the theoretical discharge capacity of 302 mAh g−1. Upon subsequent charge, a pronounced oxidation peak at about 3.7 V vs. Li+/Li was observed, after which the cell short-circuited at a capacity of 114 mAh g–1, caused presumably by Li dendrite formation. In contrast to carbonate media, BiF3 cathodes in highly concentrated IL electrolytes were characterized by stable electrochemical performance. Galvanostatic discharge of BiF3 in 4.3 M LiFSI in Pyr1,4TFSI revealed the formation of two small reduction peaks at 2.6 V (R1) and 2.4 V (R2) vs. Li+/Li, which can be assigned to the reduction of BiF3 and BiOxF3-2x, respectively. The differential capacity curve does not drop below –300 mAh g–1 V–1, indicating that no significant amount of CEI is formed. The total capacity of BiF3 cathode after initial discharge was ca. 208 mAh g-1. Upon charge, only one oxidation peak O1 at 3.3 V vs. Li+/Li was observed, associated with oxidation of Bi0 to BiF3. In subsequent cycles, the reduction peak R1 shifts slightly to 2.5 V vs. Li+/Li while oxidation peak O1 shifts to 3.4 V vs. Li+/Li, resulting in a slight increase of voltage hysteresis from 0.8 V to 0.9 V. At voltages above 3.9 V vs. Li+/Li, the differential capacity increases slightly again, which is probably associated with electrolyte decomposition. Overall, the differential capacity plots measured for BiF3 in the highly concentrated IL electrolyte do not show any noticeable formation of CEIs upon both reduction and oxidation.
We then compared the cycling stabilities of BiF3 employing 1 M LiPF6-EC/DMC and 4.3 M LiFSI-Pyr1,4TFSI electrolytes. As follows from Fig. 5, BiF3 with carbonate-based electrolyte exhibited a high initial discharge capacity of 280 mAh g‒1. However, this capacity fades rapidly below 100 mAh g–1 after 38 cycles, corresponding to the capacity retention of 36%. On the contrary, BiF3 measured in IL electrolytes displayed significantly higher capacity retention of 46% over 80 cycles, although its initial capacity is slightly lower (208 mAh g–1). Comparison of the electrochemical performance of BiF3 with reported systems comprising BiF312,14,19,21,24 iron fluorides7,17,54 or alkali iron fluorides55,56 can be found in the supporting information (see Supplementary Table 3). Notably, superior cycling stability of BiF3 was revealed in 1 M LiFSI-Pyr1,4TFSI electrolyte over 180 cycles, with the capacity ranging between 80 and 90 mAh g-1 (see Supplementary Fig. 15). The enhanced electrochemical performance of BiF3 in LiFSI-Pyr1,4TFSI electrolytes can be attributed to the formation of a stable CEI that effectively suppresses continuous electrolyte reduction during battery cycling, as observed for carbonate-based electrolytes.
Conclusions
In summary, we have reported a facile, low-cost, and scalable synthesis of crystalline o-BiF3 via thermal decomposition of a single-source Bi(TFA)3 precursor under N2 atmosphere. When the synthesis is performed in air, the formation of BiOxF3-2x with different amounts of oxygen (0.4 < x < 0.6) takes place. Side by side comparison of the electrochemical performance of BiF3 in carbonate- and IL-based electrolytes revealed that the use of ILs significantly improves cyclic stability of BiF3. In particular, a high initial capacity of ca. 208 mAh g−1 was obtained at a current density of 30 mA g–1 (∼C/10), and half of this capacity was retained after 80 cycles. We speculate that employment of IL overcomes the major constraint of carbonate-based electrolytes associated with accumulation of Li2CO3 CEIs at 1.6–2 V vs. Li+/Li, which hinders Li-ion percolation from the electrolytes towards the surface of BiF3 particles.
Methods
Synthesis of Bi(TFA)3
The synthesis was carried out as reported elsewhere40, with slight modifications14. Bi2O3 (10.0 g, 21.5 mmol) was mixed with trifluoroacetic acid (10 ml, 139 mmol) and trifluoroacetic anhydride (20 ml, 139 mmol) in a three-neck flask, connected to a Schlenk line, and stirred under N2 atmosphere at 60 °C for 24 h. The oxide fully dissolved, forming a viscous, brownish solution. The solvents were evaporated under vacuum, resulting in a white powder (yield: 96%).
Synthesis of BiF3
The as-synthesized Bi(TFA)3 (3.30 g, 6.03 mmol) was added inside an N2-filled glovebox (O2, H2O < 0.1 ppm) to an alumina crucible and placed into a silica quartz tube that was fused on one side and open on the other. The open end was then sealed with a valve and transferred into a tube furnace. The valve was used to release the gaseous decomposition products. The sample was heated under N2 flow from RT to 100 °C at 300 °C h−1 at 300 °C h−1 where it was kept for 2 h, and then heated at 50 °C h−1 to 300 °C where it was kept for 12 h before cooling to RT at 300 °C h−1, yielding a brown powder (1.60 g, 6.00 mmol, 100%). The yield was determined by weighing the product before and after the heat treatment. The stability of BiF3 was tested under air by storing the product in a 5 ml glass vial under air for 3 months.
Electrochemical measurements
In a typical cathode preparation, BiF3 (37.5 mg, 50 wt%) and the conductive additives carbon black (CB, 15 mg, 20 wt%) and multi-walled carbon nanotubes (CNT, 50–90 nm, 15 mg, 20 wt%) were dry-mixed under Ar in a planetary ball mill (Fritsch, Pulverisette 7) in an anatase beaker with 28 anatase balls for 1:30 h at 800 rpm. Carbon nanotubes were used to enhance electron percolation in the BiF3 cathode. Afterwards, polyvinylidene fluorine (pVdF, 7.5 mg, 10 wt-%) was added as a binder. N-methyl-2-pyrrolidone (NMP, 650 μL, 900 wt-%) were added as a solvent and ball-milled under Ar for 1:30 h at 500 rpm to form a slurry. This slurry was then applied by brushing it onto carbon-coated Al foil (12 mm Ø). The painted current collectors were dried under air at 75 °C for 1 h and then dried under vacuum at 80 °C for 18 h. The final loading of BiF3 was ca. 0.1 mg cm −2. In an Ar-filled glovebox, the cathode was incorporated into an air-tight coin-type cell. Elemental Li coins (12 mm Ø) were used as counter and reference electrodes. Typically, 200–300 μL of ion-conducting electrolyte was added onto the glass microfiber (Whatman) separator. IL-containing cells were then heated under Ar for 24 h at 75 °C to ensure good wetting of the cell. Likewise, the 4.3 M LiPF6 in EC/DMC was heated under Ar for 20 min at 75 °C to ensure good wetting. If the cell was heated longer, gas evolution caused the cell to burst. Moreover, every cell was prepared at least twice to ensure the reproducibility of the results. Galvanostatic cycling at 30 mA g−1 was performed in a voltage range between 2 and 4 V on a multichannel potentiostat/ galvanostat from Biologic (MPG2).
Transmission and Scanning electron microscopy (TEM and SEM) measurements
TEM measurements were performed on the Talos F200X (ThermoFisher Scientific, FEG, Uacc = 200 kV). SEM measurements of the as-obtained samples were done on a Quanta 200 F microscope (Thermo Fisher Scientific) operated at an acceleration voltage Vacc = 20 kV. Energy-dispersive X-ray spectroscopy (EDXS) was performed with an Octane SDD detector (EDAX, Ametec) attached to the microscope column. For spectra recording and quantification (ZAF correction), the software Gemini (EDAX) was used.
Powder X-ray diffraction measurements
The powder XRD patterns were collected at RT on a Stoe STADI P powder X-ray diffractometer (Cu Kα1 radiation, λ = 1.540598 Å, focusing germanium monochromator) equipped with a Dectris Mythen 1 K silicon strip detector.
Rietveld Refinement
The composition of the BiF3 product was analyzed based on the powder XRD pattern (Cu Kα1 radiation, λ = 1.540598 Å) by a Rietveld refinement with the GSAS II program43. For the refinement, the starting model of o-BiF3 given by Greis et. al30. was used (space group Pnma, a = 6.5614(4) Å, b = 7.0153(5) Å, c = 4.8414(3) Å, Bi on 4c with x = 0.3547(1), y = 0.25, z = 0.0349(1); F1 on 4c with x = 0.5361(17), y = 0.25, z = 0.6271(26) and F2 on 8d with x = 0.1652(14), y = 0.0577(11), z = 0.3528(15)). Except the lattice constants and the displacement parameters, none of the structural parameters was refined. The unit cell parameters obtained by the refinement were: a = 6.56037(8) Å, b = 7.01741(16) Å, c = 4.84501(7) Å. The displacement parameters of the F-atoms were constrained to be equal and to be 1.67 times larger than the one for the Bi-atom. The displacement parameters for Bi refined to Uiso = 0.0272(5) Å2 and accordingly the displacement parameters for the F-atoms are Uiso = 0.0453 A2. The Figures of Merit were: reduced χ2 = 4.46, wR = 0.14574, R(F) = 0.0744, R(F2) = 0.27704 (on 73 reflections), Nobs = 4020, Nvals = 21. The March-Dollase ratio for a preferred orientation along the [010]-direction is 1.543.
Particle Size Distributions
ImageJ was used to count the individual particles. Because of the strongly anisotropic shape, every particle was measured with two lines perpendicular to each other to get an average size for the particle. The SEM images used for counting can be seen in Supplementary Figs 11, 12. The sample sizes were N = 678 for the distribution before ball-milling and N = 2231 after ball-milling. The fitting of the particle size distribution was carried out using the reliability module in python57. Parameters were estimated using a maximum likelihood estimation. The equations for the particle size distributions used for fitting are given as:
Data availability
The data that support the plots within this paper and other finding of this study are available from the corresponding author upon reasonable request.
References
Kraytsberg, A. & Ein-Eli, Y. A critical review-promises and barriers of conversion electrodes for Li-ion batteries. J. Solid State Electrochem 21, 1907–1923 (2017).
Wu, F. & Yushin, G. Conversion cathodes for rechargeable lithium and lithium-ion batteries. Energy Environ. Sci. 10, 435–459 (2017).
Cabana, J., Monconduit, L., Larcher, D. & Palacín, M. R. Beyond intercalation-based Li-ion batteries: the state of the art and challenges of electrode materials reacting through conversion reactions. Adv. Mater. 22, E170–E192 (2010).
Turcheniuk, K., Bondarev, D., Amatucci, G. G. & Yushin, G. Battery materials for low-cost electric transportation. Mater. Today 42, 57–72 (2021).
Wang, L. et al. Li-free Cathode Materials for High Energy Density Lithium Batteries. Joule 3, 2086–2102 (2019).
Guntlin, C. P. et al. Nanocrystalline FeF3 and MF2 (M = Fe, Co, and Mn) from metal trifluoroacetates and their Li(Na)-ion storage properties. J. Mater. Chem. A 5, 7383–7393 (2017).
Guntlin, C. P., Kravchyk, K. V., Erni, R. & Kovalenko, M. V. Transition metal trifluoroacetates (M = Fe, Co, Mn) as precursors for uniform colloidal metal difluoride and phosphide nanoparticles. Sci. Rep. 9, 6613 (2019).
Huang, Q. et al. Insights into the effects of electrolyte composition on the performance and stability of FeF2 conversion-type cathodes. Adv. Energy Mater. 9, 1803323 (2019).
Hua, X. et al. Comprehensive study of the CuF2 conversion reaction mechanism in a lithium ion battery. J. Phys. Chem. C. 118, 15169–15184 (2014).
Seo, J. K. et al. Revisiting the conversion reaction voltage and the reversibility of the CuF2 electrode in Li-ion batteries. Nano Res 10, 4232–4244 (2017).
Hua, X. et al. Revisiting metal fluorides as lithium-ion battery cathodes. Nat. Mater. 20, 841–850 (2021).
Gmitter, A. J. et al. Formation, dynamics, and implication of solid electrolyte interphase in high voltage reversible conversion fluoride nanocomposites. J. Mater. Chem. 20, 4149–4161 (2010).
Hu, B. et al. Effects of amorphous AlPO4 coating on the electrochemical performance of BiF3 cathode materials for lithium-ion batteries. J. Power Sources 218, 204–211 (2012).
Oszajca, M. F. et al. Colloidal BiF3 nanocrystals: A bottom-up approach to conversion-type Li-ion cathodes. Nanoscale 7, 16601–16605 (2015).
Teng, Y. T., Wei, F. & Yazami, R. Synthesis of NixCo(1−x)F2 (x = 0, 0.25, 0.50, 0.75, 1.0) and application in lithium ion batteries. J. Alloy. Compd. 653, 434–443 (2015).
Amatucci, G. G. & Pereira, N. Fluoride based electrode materials for advanced energy storage devices. J. Fluor. Chem. 128, 243–262 (2007).
Arai, H., Okada, S., Sakurai, Y. & Yamaki, J.-I. Cathode performance and voltage estimation of metal trihalides. J. Power Sources 68, 716–719 (1997).
Zhao, X., Zhao-Karger, Z., Fichtner, M. & Shen, X. Halide-based materials and chemistry for rechargeable batteries. Angew. Chem. Int. Ed. 59, 5902–5949 (2020).
Bervas, M., Badway, F., Klein, L. C. & Amatucci, G. G. Bismuth fluoride nanocomposite as a positive electrode material for rechargeable lithium batteries. Electrochem. Solid-State Lett. 8, A179 (2005).
Fiordiponti, P., Panero, S., Pistoia, G. & Temperoni, C. Nonaqueous batteries with BiF3 cathodes. J. Electrochem. Soc. 125, 511–515 (1978).
Bervas, M. et al. Investigation of the lithiation and delithiation conversion mechanisms of bismuth fluoride nanocomposites. J. Electrochem. Soc. 153, A799 (2006).
Padhi, A. K., Nanjundaswamy, K. S. & Goodenough, J. B. Phospho‐olivines as positive‐electrode materials for rechargeable lithium batteries. J. Electrochem. Soc. 144, 1188–1194 (1997).
Dolotko, O., Senyshyn, A., Mühlbauer, M. J., Nikolowski, K. & Ehrenberg, H. Understanding structural changes in NMC Li-ion cells by in situ neutron diffraction. J. Power Sources 255, 197–203 (2014).
Gmitter, A. J., Gural, J. & Amatucci, G. G. Electrolyte development for improved cycling performance of bismuth fluoride nanocomposite positive electrodes. J. Power Sources 217, 21–28 (2012).
Badway, F., Pereira, N., Cosandey, F. & Amatucci, G. G. Carbon-metal fluoride nanocomposites. J. Electrochem. Soc. 150, A1209 (2003).
Badway, F., Cosandey, F., Pereira, N. & Amatucci, G. G. Carbon metal fluoride nanocomposites. J. Electrochem. Soc. 150, A1318 (2003).
Ma, R. et al. Large-scale fabrication of graphene-wrapped FeF3 nanocrystals as cathode materials for lithium ion batteries. Nanoscale 5, 6338–6343 (2013).
Hu, X. et al. Li-storage performance of binder-free and flexible iron fluoride@graphene cathodes. J. Mater. Chem. A 3, 23930–23935 (2015).
Fan, X. et al. In situ lithiated FeF3/C nanocomposite as high energy conversion-reaction cathode for lithium-ion batteries. J. Power Sources 307, 435–442 (2016).
Greis, O. & Martinez-Ripoll, M. Darstellung, Temperaturverhalten und Kristallstruktur von BiF3. Z. Anorg. Allg. Chem. 436, 105–112 (1977).
Zhao, J. et al. Size-controlled synthesis and morphology evolution of bismuth trifluoride nanocrystals via a novel solvent extraction route. Nanoscale 5, 518–522 (2013).
Feng, C. et al. A newly discovered BiF3 photocatalyst with a high positive valence band. J. Mol. Catal. A Chem. 401, 35–40 (2015).
Hu, B. et al. Improved electrochemical properties of BiF3/C cathode via adding amorphous AlPO4 for lithium-ion batteries. Electrochim. Acta 102, 8–18 (2013).
Devi, N., Ghosh, S., Perla, V. K., Pal, T. & Mallick, K. Laboratory based synthesis of the pure form of gananite (BiF3) nanoparticles: a potential material for electrochemical supercapacitor application. N. J. Chem. 43, 18369–18376 (2019).
Gmitter, A. J., Halajko, A., Sideris, P. J., Greenbaum, S. G. & Amatucci, G. G. Subsurface diffusion of oxide electrolyte decomposition products in metal fluoride nanocomposite electrodes. Electrochim. Acta 88, 735–744 (2013).
Cao, D. et al. Lithium dendrite in all-solid-state batteries: growth mechanisms, suppression strategies, and characterizations. Matter 3, 57–94 (2020).
Elia, G. A., Ulissi, U., Jeong, S., Passerini, S. & Hassoun, J. Exceptional long-life performance of lithium-ion batteries using ionic liquid-based electrolytes. Energy Environ. Sci. 9, 3210–3220 (2016).
Agostini, M. et al. A high-power and fast charging Li-ion battery with outstanding cycle-life. Sci. Rep. 7, 1104 (2017).
Elia, G. A. et al. A long-life lithium ion battery with enhanced electrode/electrolyte interface by using an ionic liquid solution. Chem. Eur. J. 22, 6808–6814 (2016).
Reiß, GuidoJ., Frank, Walter & Schneider, J. ürgen Synthesis and crystal structure of bismuth (III) trifluoroacetate-trifluoroacetic acid adduct, Bi(OOCCF3)3.HOOCCF3. Main. Group Met. Chem. 18, 287–294 (1995).
Hassel, O. & Nilssen, S. Der Kristallbau des BiF3. Z. Anorg. Allg. Chem. 181, 172–176 (1929).
Rietveld, H. Line profiles of neutron powder-diffraction peaks for structure refinement. Acta Crystallogr 22, 151–152 (1967).
Toby, B. H. & Von Dreele, R. B. GSAS-II: the genesis of a modern open-source all purpose crystallography software package. J. Appl. Crystallogr. 46, 544–549 (2013).
Aurivillius, B. X-ray studies on the system BiF3-Bi2O3. I. Preliminary phase analysis and a note on the structure of BiF3. Acta Chem. Scand. 9, 1206–1208 (1955).
Aurivillius, B. L., Thomas. X-ray studies of the system BiF3-Bi2O3. II. A bismuth oxide fluoride of defective Tysonite type. Acta Chem. Scand. 9 1209–1212 (1955).
Bervas, M., Klein, L. C. & Amatucci, G. G. Reversible conversion reactions with lithium in bismuth oxyfluoride nanocomposites. J. Electrochem. Soc. 153, A159 (2006).
Kim, J.-K., Matic, A., Ahn, J.-H. & Jacobsson, P. An imidazolium based ionic liquid electrolyte for lithium batteries. J. Power Sources 195, 7639–7643 (2010).
Seki, S. et al. Lithium secondary batteries using modified-imidazolium room-temperature ionic liquid. J. Phys. Chem. B 110, 10228–10230 (2006).
Seki, S. et al. Imidazolium-based room-temperature ionic liquid for lithium secondary batteries: relationships between lithium salt concentration and battery performance characteristics. ECS Electrochem. Lett. 1, A77–A79 (2012).
MacFarlane, D. R., Meakin, P., Sun, J., Amini, N. & Forsyth, M. Pyrrolidinium imides: a new family of molten salts and conductive plastic crystal phases. J. Phys. Chem. B 103, 4164–4170 (1999).
Paillard, E. et al. Electrochemical and physicochemical properties of PY14FSI-based electrolytes with LiFSI. J. Electrochem. Soc. 156, A891 (2009).
Bhatt, A. I., Best, A. S., Huang, J. & Hollenkamp, A. F. Application of the N-propyl-N-methyl-pyrrolidinium Bis(fluorosulfonyl)imide RTIL containing lithium bis(fluorosulfonyl)imide in ionic liquid based lithium batteries. J. Electrochem. Soc. 157, A66 (2010).
Mun, J. et al. The feasibility of a pyrrolidinium-based ionic liquid solvent for non-graphitic carbon electrodes. Electrochem. commun. 13, 1256–1259 (2011).
Xiao, A. W. et al. Understanding the conversion mechanism and performance of monodisperse FeF2 nanocrystal cathodes. Nat. Mater. 19, 644–654 (2020).
Kravchyk, K. V., Zünd, T., Wörle, M., Kovalenko, M. V. & Bodnarchuk, M. I. NaFeF3 nanoplates as low-cost sodium and lithium cathode materials for stationary energy storage. Chem. Mater. 30, 1825–1829 (2018).
Sun, S. et al. Enhanced charge storage of Na3FeF6 with carbon nanotubes for lithium-ion batteries. Solid State Ion. 312, 61–66 (2017).
Reid, M. R. A Python library for reliability engineering; Zenodo, 2020. https://zenodo.org/record/3938000#.YFHpndwxk2w (accessed 2021-03-17).
Acknowledgements
The authors are grateful to the research facilities of ETH Zurich (ETH Electron Microscopy Center, Department of Chemistry and Applied Biosciences) and Empa (Empa Electron Microscopy Center and Laboratory for Mechanics of Materials & Nanostructures) for access to the instruments and for technical assistance.
Author information
Authors and Affiliations
Contributions
J.F.B and K.V.K conceptualized and wrote the original draft. F.K. and M.W. performed SEM, EDX and XRD measurements/analysis, respectively. K.V.K and M.V.K. discussed and revised the manuscript.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review information
Communications Chemistry thanks the anonymous reviewers for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Baumgärtner, J.F., Krumeich, F., Wörle, M. et al. Thermal synthesis of conversion-type bismuth fluoride cathodes for high-energy-density Li-ion batteries. Commun Chem 5, 6 (2022). https://doi.org/10.1038/s42004-021-00622-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s42004-021-00622-y
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.