NEW METHOD FOR THE PREPARATION OF A LI-P-S PRODUCT AND CORRESPONDING PRODUCTS

Abstract

The present invention concerns a new method for the preparation of a Li—P—S product, as well as the products obtainable by said methods, and uses thereof especially as solid electrolytes.

Claims

1-10. (canceled)

11. A method of preparing a Li—P—S product, the method comprising at least the following steps: (a) mixing at least Li.sub.4P.sub.2S.sub.6, Li.sub.2S and sulfur to obtain a first mixture; (b) heating the first mixture in an inert atmosphere, under vacuum or under H.sub.2S flow, for a period of time and at a temperature sufficient to produce the Li—P—S product; and (c) cooling and optionally powdering the Li—P—S product.

12. The method according to claim 11, wherein the Li—P—S product is selected from the group consisting of: Li.sub.7P.sub.3S.sub.11, Li.sub.3PS.sub.4, Li.sub.7PS.sub.6, and Li.sub.96P.sub.3S.sub.12.

13. The method according to claim 11, wherein the temperature in step (b) is comprised from 150° C. to 600° C.

14. The method according to claim 11, wherein the heating in step (b) is made over a time period of from 1 hour to 200 hours.

15. The method according to claim 11, wherein Li.sub.4P.sub.2S.sub.6 is obtained from the reaction between Li.sub.2S and P.sub.2S.sub.5.

16. The Li—P—S product obtainable by the method of claim 11.

17. A product of formula Li.sub.7P.sub.3S.sub.11, comprising a crystal structure and a volume V per formula unit at room temperature, using the cell parameters acquired at standard atmosphere, comprised between 407 and 412 angstrom cube, as measured by X-Ray Diffraction, using a Bruker D8 diffractometer with Cu Kα radiation.

18. A method of using the product of 17, the method comprising using the product, alone or in combination with any crystalline or amorphous conductive Li-material, as solid electrolyte.

19. A solid electrolyte comprising at least the product of claim 17.

20. A battery comprising at least the product of claim 17.

Description

FIGURES

[0071] FIG. 1: Comparison of the XRD patterns of the simulated pattern of Li.sub.4P.sub.2S.sub.6 [1], the synthesized Li.sub.4P.sub.2S.sub.6 (Example 1), and the ball-milled product of the Reaction 1 (Example 2).

[0072] FIG. 2: Comparison of the Raman spectra of the synthesized Li.sub.4P.sub.2S.sub.6 (Example 1), the ball-milled product of the Reaction 1 (Example 2), the crystalline Li.sub.7P.sub.3S.sub.11 that is formed by annealing Example 2 (Example 3), and the crystalline Li.sub.7P.sub.3S.sub.11 that was synthesized from the conventional reagents Li.sub.2S and P.sub.2S.sub.5.

[0073] FIG. 3: Comparison of the XRD patterns of the crystalline Li.sub.7P.sub.3S.sub.11 that is synthesized via the Reaction 1 (Example 3), the crystalline Li.sub.7P.sub.3S.sub.11 that was synthesized from the conventional reagents Li.sub.2S and P.sub.2S.sub.5, and the simulated pattern of Li.sub.7P.sub.3S.sub.11 [11].

[0074] FIG. 4: Comparison of the XRD patterns of the crystalline Li.sub.7P.sub.3S.sub.11 that is synthesized via the Reaction 1 (Example 3), and the empty Be-equipped sample holder.

[0075] FIG. 5: Results of the profile fitting of Examples 4 and 3. The Bragg positions were indicated with black vertical lines, the fits were plotted with a solid black line, and the collected data were shown in black hollow spheres. The regions marked with asterisk were omitted from the fit as they involve the contribution of the sample holder as shown in FIG. 4.

[0076] FIG. 6: Deconvolution of the signals coming from the structural moieties PS.sub.4.sup.3−, P.sub.2S.sub.7.sup.4− and P.sub.2S.sub.6.sup.4− in the Raman spectrum of Example 3. The relative ratios of the peak areas are also noted on the top left of the figure.

[0077] FIG. 7: Deconvolution of the signals coming from the structural moieties PS.sub.4.sup.3−, P.sub.2S.sub.7.sup.4− and P.sub.2S.sub.6.sup.4− in the Raman spectrum of Example 4. The relative ratios of the peak areas are also noted on the top left of the figure.

[0078] FIG. 8: Comparison of the ionic conductivity values versus inverse temperature (1/T) of the ball-milled product of the Reaction 1 (Example 2), the crystalline Li.sub.7P.sub.3S.sub.11 that is synthesized via the Reaction 1 (Example 3), the crystalline Li.sub.7P.sub.3S.sub.11 that was synthesized from the conventional reagents Li.sub.2S and P.sub.2S.sub.5.

[0079] FIG. 9: Comparison of the σ×T values versus inverse temperature (1/T) of the ball-milled product of the Reaction 1 (Example 2), the crystalline Li.sub.7P.sub.3S.sub.11 that is synthesized via the Reaction 1 (Example 3), the crystalline Li.sub.7P.sub.3S.sub.11 that was synthesized from the conventional reagents Li.sub.2S and P.sub.2S.sub.5. The linear fits to calculate the slope of the lines were also shown as solid black lines.

EXAMPLES

[0080] The disclosure will now be illustrated with working examples, which is intended to illustrate the working of disclosure and not intended to take restrictively to imply any limitations on the scope of the present disclosure. Other examples are also possible which are within the scope of the present disclosure.

Example 1

[0081] Li.sub.2S and P.sub.2S.sub.5 (both produced by Sigma Aldrich) were used as starting materials. 1.5 g of total powder at a molar ratio of 2:1 were put in a 45 mL ZrO.sub.2 jar with 12 ZrO.sub.2 balls (3 g/ball, 10 mm diameter) in an Ar filled glovebox. The jar was sealed with scotch and parafilm to prevent air exposure, then was taken out of the glovebox and was placed in Fritzch Planetary Micro Mill Pulverisette 7. It was ball-milled with 510 RPM rotating speed for 38 hours while employing 15 minute breaks in every 15 minutes of milling, in order to prevent excessive heating of the jar. The jar was then moved to an Ar filled glovebox to recover the powder. Then, the resulting white powder was pelletized at 530 MPa with a 6 mm diameter die. The pellet was vacuum sealed in a carbon coated quartz tube, then the tube was heated to 350° C. with 5° C./min heating rate, and was kept at the same temperature for 36 hours. After the annealing step, the tube was slowly cooled down to RT, and it was opened in an Ar filled glovebox.

Example 2

[0082] 400 mg of Example 1, 46 mg of sulfur, and 22 mg of Li.sub.2S (produced by Sigma Aldrich) were mixed in an Ar filled glovebox to balance the reaction below:

3Li.sub.4P.sub.2S.sub.6+3S+Li.sub.2S=2Li.sub.7P.sub.3S.sub.11  Reaction 1:

[0083] The mixture was put in a 45 mL ZrO.sub.2 jar with 8 ZrO.sub.2 balls (3 g/ball, 10 mm diameter). The jar was sealed with scotch and parafilm to prevent air exposure, then was taken out of the glovebox and was placed in Retsch PM200 Planetary Ball- milling Apparatus. It was ball-milled with 510 RPM rotating speed for 76 hours while employing 15 minute breaks in every 15 minutes of milling, in order to prevent excessive heating of the jar. The jar was then moved to an Ar filled glovebox to recover the powder.

Example 3

[0084] Example 2 was pelletized at 530 MPa with a 6 mm diameter die. The pellet was vacuum sealed in a carbon coated quartz tube, then the tube was annealed at 200° C. for 84 hours. After the annealing step, the tube was slowly cooled down to RT, and it was opened in an Ar filled glovebox to recover the sample.

Example 4 (Comparative)

[0085] Li.sub.2S and P.sub.2S.sub.5 (both produced by Sigma Aldrich) were used as starting materials. 1.5 g of total powder at a molar ratio of 7:3 were put in a 45 mL ZrO.sub.2 jar with 12 ZrO.sub.2 balls (3 g/ball, 10 mm diameter) in an Ar filled glovebox. The jar was sealed with scotch and parafilm to prevent air exposure, then was taken out of the glovebox and was placed in Fritzch Planetary Micro Mill Pulverisette 7. It was ball-milled with 510 RPM rotating speed for 76 hours while employing 15 minute breaks in every 5 minutes of milling, in order to prevent excessive heating of the jar. The jar was then moved in an Ar filled glovebox to collect the powder. The resulting white powder was pelletized at 530 MPa with a 10 mm diameter die. The pellet vacuum sealed in a carbon coated quartz tube, then the tube was annealed at 200° C. for 168 hours. After the annealing step, the tube was slowly cooled down to RT, and it was opened in an Ar filled glovebox.

Characterization Tools

[0086] X-ray diffraction of the samples were collected using a Bruker D8 diffractometer with Cu Kα radiation at RT. The samples were sealed in a Be-equipped sample holder in an Ar filled glovebox prior to the experiment. The diffractions were collected in 2θ range of 10° to 100° in 13 hours. The lattice parameters were determined by fitting the diffraction profiles using Full-Prof Suite. Profile fitting procedures of Example 3 and Example 4 were limited in a shorter 2θ range (10° to 28°) to increase the accuracy of the fit because the high number of Bragg positions in P-1 space group at higher angles could mislead the fitting process. The small peaks which are attributed to the Be-equipped sample holder (see FIG. 4) were removed before the fitting process to increase the precision of the fitting.

[0087] The Raman spectra were collected using a Raman DXR Microscope (Thermo Fischer Scientific) with excitation laser beam wavelength of 532 nm and a low laser power of 0.1 mW to prevent excessive heating of the sample. The fitting processes were performed using Omnic Software of Thermo Fischer Scientific.

[0088] Before the impedance spectroscopy measurements, powder samples were cold-pressed in an Ar filled glovebox. Example 2 and Example 3 were pressed with a 6 mm diameter die with 530 MPa pressure, while Example 4 was pressed with a 10 mm diameter die with 530 MPa pressure. The pellets were then sandwiched between pre-dried carbon paper electrodes, and then loaded into air-tight sample holders. The AC impedance spectra were collected by using Biologic MTZ-35 frequency response analyser. During the measurements, the AC potential for excitation was set at 50 mV for all the samples. The frequency range of the measurement of Example 2 was 0.05 Hz to 30 MHz, whereas a range of 1 Hz to 30 MHz was applied in the measurements of Example 3 and Example 4. The spectrum of each sample was recorded at stabilized temperature values varying between −30° C. and 50° C. in steps of 10° C. The ionic conductivity values were obtained by fitting the data into equivalent circuit models using ZView software. The slopes of the σT versus 1/T plots were used to determine activation energy values.

Experimental Results

[0089]
3Li.sub.4P.sub.2S.sub.6+3S+Li.sub.2S=2Li.sub.7P.sub.3S.sub.11  Reaction 1:

[0090] The Bragg peaks observed in the XRD pattern of Example 1 show correlation to the simulated peak positions of crystalline Li.sub.4P.sub.2S.sub.6 [1], as shown in FIG. 1. In the Raman spectrum of Example 1 in FIG. 3, the only peak centered at 383 cm.sup.−1 is due to the vibration of P—S bond in P.sub.2S.sub.6.sup.4− anion [2,3]. Hence, the sample was considered phase pure and was used as a precursor in the Reaction 1.

[0091] The Reaction 1 was performed via mechanochemical synthesis route, and the reaction product was named as Example 2. As shown in FIG. 1, the product was X-ray amorphous, which indicated that the reagents were successfully amorphized. It is known in the literature that when Li.sub.2S (<75 at %) and P.sub.2S.sub.5 are used as reagents for the mechanochemical synthesis of Li—P—S compounds, the reaction leads to the formation of X-ray amorphous products, similar to the case of Example 2 [4-10]. Until today, a Li—P—S compound (i.e. Li.sub.7P.sub.3S.sub.11) has never been synthesized using a different Li—P—S compound (i.e. Li.sub.4P.sub.2S.sub.6) as precursor [6-10].

[0092] After the mechanochemical reaction, the vibrations of P—S bonds in P.sub.2S.sub.7.sup.4− and PS.sub.4.sup.3− moieties appear in the Raman spectrum of Example 2, see FIG. 2. It is inferred from the drastic decrease in the intensity of the signal of P.sub.2S.sub.6.sup.4− unit that some of the phosphorus ions in the P.sub.2S.sub.6.sup.4− moieties were oxidized from +4 to +5 during the mechanochemical reaction. It is known that P.sub.2S.sub.7.sup.4− unit can be reacted to form P.sub.2S.sub.6.sup.4− unit by reducing the phosphorus at 280-300° C. [5,7]. It was shown for the first time with the Reaction 1 that the process could be reversed by a mechanochemical reaction.

[0093] The XRD pattern of Example 3 showed correlation to the simulated pattern of crystalline Li.sub.7P.sub.3S.sub.11 [11], as evidenced by XRD in FIG. 3. Additional peaks with small intensity were also observed at 17.3°, 24.5° and 25.5°, which originate from the Be-equipped sample holder, with which the pattern of the powder was collected. The comparison of the pattern of Example 3 and the Be-equipped sample holder is presented in FIG. 4. The XRD pattern of Example 4, which was synthesized using Li.sub.2S and P.sub.2S.sub.5 as precursors was also shown in FIG. 3 for comparison. Several differences between Examples 3 and 4 were noted. Sharper peaks were present in the pattern of Example 3, which indicates higher crystallinity compared to Example 4. The lattice parameters of Examples 3 and 4 were obtained by profile fitting. The resulting fit and the Bragg positions were shown in FIG. 5. In the figure, the hollow circles indicate the collected data, and the black line indicates the resulting profile fit. Significant differences were observed in the profile parameters of Examples 3 and 4, and the reference from the literature [11], see Table 1. The unit cell of Example 3 has a drastically shorter lattice parameter a, and larger alpha and gamma angles compared to Example 4 and the material reported in the literature. Beta angle of Example 3 was observed to be larger than the literature example, but it was comparatively smaller than the one of Example 4. These results proved that the crystalline Li.sub.7P.sub.3S.sub.11 obtained by the Reaction 1 was structurally different than the materials synthesized by starting from the conventional reagents Li.sub.2S and P.sub.2S.sub.5.

TABLE-US-00001 TABLE 1 Lattice parameters, in triclinic P-1 description, for Examples 3 and 4 versus lattice parameters reported in the literature [11] a (Å) b (Å) c (Å) alpha (°) beta (°) gamma (°) V (Å.sup.3) Example 3 12.40 6.04 12.52 103.30 113.25 75.07 823 Example 4 12.46 6.05 12.53 103.14 113.33 74.70 828 Reference [11] 12.50 6.03 12.53 102.85 113.20 74.47 829

[0094] The Raman spectra of Examples 3 and 4 show the presence of bond vibrations of PS.sub.4.sup.3−, P.sub.2S.sub.7.sup.4− and P.sub.2S.sub.6.sup.4− moieties at 421 cm.sup.−1, 404 cm.sup.−1 and 383 cm.sup.−1, respectively, see FIG. 2. These results show correlation to the results published in the literature [7,8,10,12-15]. In order to make a semi-quantitative comparison of the quantities of the moieties in Examples 3 and 4, the spectra were fitted into three peaks centered at the reported positions of structural moieties [5]. The relative ratios of the peak areas were also compared. As shown in FIGS. 6 and 7, Examples 3 and 4 manifested drastic differences in their local structures.

[0095] Relative ratio of PS.sub.4.sup.3−/P.sub.2S.sub.7.sup.4− in Example 3 was calculated to be 0.28, whereas the ratio in Example 4 deviated significantly and was noted down as 0.55. It was also observed that the relative intensity of peak of P.sub.2S.sub.6.sup.4− units was 2.63% lower in the case of Example 3. These results clearly indicated that the material which was synthesized from the reagents Li.sub.2S and P.sub.2S.sub.5 possessed different structural properties than the material synthesized from a different Li—P—S compound as in the case of the Reaction 1.

[0096] The ionic conductivities of Examples 2, 3 and 4 were shown as a function of temperature (−30° C. to 20° C.) in FIG. 8. The activation energies were calculated from the σT vs 1/T plots using Equation 1, and the linear fits of the data to calculate the slope were also shown in FIG. 9:

[00001]$\begin{array}{cc}\sigma *T={\sigma }_0*e^{-\frac{\mathrm{Ea}}{\mathrm{kT}}}& \mathrm{Equation}1\end{array}$

[0097] Example 2 demonstrated 5×10.sup.−5 Scm.sup.−1 conductivity at 20° C. with an activation energy of 0.40 eV. After the crystallization of Li.sub.7P.sub.3S.sub.11 via subsequent annealing of Example 2, Example 3 showed a significant increase in the conductivity by 6×10.sup.−4 S.Math.cm.sup.−1 at 20° C. The activation energy for the conduction was also decreased to 0.37 eV, which is smaller than that of Example 4 (0.38 eV), which was synthesized from the precursors Li.sub.2S and P.sub.2S.sub.5. Example 4 possessed a slightly higher ionic conductivity of 9×10.sup.−4 Scm.sup.−1 at 20° C. These results also show correlation to the values reported in the literature [6, 14-18].

[0098] The results obtained from different characterization tools indicate that crystalline Li.sub.7P.sub.3S.sub.11 can be synthesized by using an alternative synthesis route as described in the Reaction 1.

[0099] The resulting product (Example 3) according to the invention showed significant difference in terms of lattice parameters compared to the values reported in the literature, and the ones of the crystalline Li.sub.7P.sub.3S.sub.11 that was synthesized by using Li.sub.2S and P.sub.2S.sub.5 as reagents (Example 4—comparative). Additionally, the relative ratios of the structural moieties in Examples 3 and 4 were calculated to be drastically different. It was also shown by the impedance spectra that the ionic conductivities of both examples and the materials reported in the literature are quite similar. It was also noted that a successful reaction from one member of the Li.sub.2S—P.sub.2S.sub.5 binary system to the other member has been reported for the first time in the literature. This discovery paves the way to finding other alternative reaction pathways between different members of the binary system.

REFERENCES

[0100] [1] S. Neuberger, S. P. Culver, H. Eckert, W. G. Zeier, J. Schmedt auf der Günne, Refinement of the crystal structure of Li4P2S6 using NMR crystallography, Dalt. Trans. (2018). doi:10.1039/C8DT02619J.

[0101] [2] C. Dietrich, M. Sadowski, S. Sicolo, D. A. Weber, S. J. Sedlmaier, K. S. Weldert, S. Indris, K. Albe, J. Janek, W. G. Zeier, Local Structural Investigations, Defect Formation, and Ionic Conductivity of the Lithium Ionic Conductor Li4P2S6, Chem. Mater. 28 (2016) 8764-8773. doi:10.1021/acs.chemmater.6b04175.

[0102] [3] M. Tachez, J. Malugani, R. Mercier, R. Guy, Ionic conductivity of and phase transition in Li thiophosphate Li3PS4, Solid State Ionics. 14 (1984) 181-185. doi: http://dx.doi.org/10.1016/0167-2738(84)90097-3.

[0103] [4] M. Tatsumisago, S. Hama, A. Hayashi, H. Morimoto, T. Minami, New lithium ion conducting glass-ceramics prepared from mechanochemical Li2S—P2S5 glasses, Solid State Ionics. 154 (2002) 635-640. doi:10.1016/S0167-2738(02)00509-X.

[0104] [5] C. Dietrich, D. A. Weber, S. J. Sedlmaier, S. Indris, S. P. Culver, D. Walter, J. Janek, W. G. Zeier, Lithium ion conductivity in Li2S—P2S5 glasses—building units and local structure evolution during the crystallization of superionic conductors Li3PS4, Li7P3S11 and Li4P2S7, J. Mater. Chem. A. 5 (2017) 18111-18119. doi:10.1039/C7TA06067J.

[0105] [6] Y. Seino, T. Ota, K. Takada, A. Hayashi, M. Tatsumisago, A sulphide lithium super ion conductor is superior to liquid ion conductors for use in rechargeable batteries, Energy Environ. Sci. 7 (2014) 627-631. doi: 10.1039/C3EE41655K.

[0106] [7] Y. Seino, M. Nakagawa, M. Senga, H. Higuchi, K. Takada, T. Sasaki, Analysis of the structure and degree of crystallisation of 70Li2S—30P2S5 glass ceramic, J. Mater. Chem. A. 3 (2015) 2756-2761. doi:10.1039/C4TA04332D.

[0107] [8] Y. Aoki, K. Ogawa, T. Nakagawa, Y. Hasegawa, Y. Sakiyama, T. Kojima, M. Tabuchi, Chemical and structural changes of 70Li2S—30P2S5 solid electrolyte during heat treatment, Solid State Ionics. 310 (2017) 50-55. doi: 10.1016/j.ssi.2017.08.006.

[0108] [9] K. Minami, A. Hayashi, M. Tatsumisago, Crystallization Process for Superionic Li7P3S11 Glass-Ceramic Electrolytes, J. Am. Ceram. Soc. 94 (2011) 1779-1783. doi:10.1111/j.1551-2916.2010.04335.x.

[0109] [10] M. R. Busche, D. A. Weber, Y. Schneider, C. Dietrich, S. Wenzel, T. Leichtweiss, D. Schröder, W. Zhang, H. Weigand, D. Walter, S. J. Sedlmaier, D. Houtarde, L. F. Nazar, J. Janek, In Situ Monitoring of Fast Li-Ion Conductor Li7P3S11 Crystallization Inside a Hot-Press Setup, Chem. Mater. 28 (2016) 6152-6165. doi:10.1021/acs.chemmater.6b02163.

[0110] [11] H. Yamane, M. Shibata, Y. Shimane, T. Junke, Y. Seino, S. Adams, K. Minami, A. Hayashi, M. Tatsumisago, Crystal structure of a superionic conductor, Li7P3S11, Solid State Ionics. 178 (2007) 1163-1167. doi:10.1016/j.ssi.2007.05.020.

[0111] [12] R. C. Xu, X. H. Xia, Z. J. Yao, X. L. Wang, C. D. Gu, J. P. Tu, Preparation of Li7P3S11 glass-ceramic electrolyte by dissolution-evaporation method for all-solid-state lithium ion batteries, Electrochim. Acta. 219 (2016) 235-240. doi: 10.1016/j.electacta.2016.09.155.

[0112] [13] M. Calpa, N. C. Rosero Navarro, A. Miura, K. Tadanaga, Preparation of sulfide solid electrolytes in the Li2S—P2S5 system by a liquid phase process, lnorg. Chem. Front. (2017). doi:10.1039/C7QI00737J.

[0113] [14] Y. Wang, D. Lu, M. Bowden, P. Z. El Khoury, K. S. Han, Z. D. Deng, J. Xiao, J.-G. Zhang, J. Liu, Formation mechanism of Li7P3S11 solid electrolytes through liquid phase synthesis, Chem. Mater. (2018) acs.chemmater.7b04842. doi: 10.1021/acs.chemmater.7b04842.

[0114] [15] M. Calpa, N. C. Rosero Navarro, A. Miura, K. Tadanaga, Preparation of sulfide solid electrolytes in the Li2S—P2S5 system by a liquid phase process, lnorg. Chem. Front. (2018) 501-508. doi:10.1039/C7QI00737J.

[0115] [16] I. H. Chu, H. Nguyen, S. Hy, Y. C. Lin, Z. Wang, Z. Xu, Z. Deng, Y. S. Meng, S. P. Ong, Insights into the Performance Limits of the Li7P3S11 Superionic Conductor: A Combined First-Principles and Experimental Study, ACS Appl. Mater. Interfaces. 8 (2016) 7843-7853. doi:10.1021/acsami.6b00833.

[0116] [17] S. Wenzel, D. A. Weber, T. Leichtweiss, M. R. Busche, J. Sann, J. Janek, Interphase formation and degradation of charge transfer kinetics between a lithium metal anode and highly crystalline Li7P3S11 solid electrolyte, Solid State Ionics. 286 (2016) 24-33. doi:10.1016/j.ssi.2015.11.034.

[0117] [18] S. Ito, M. Nakakita, Y. Aihara, T. Uehara, N. Machida, A synthesis of crystalline Li7P3S11 solid electrolyte from 1,2-dimethoxyethane solvent, J. Power Sources. 271 (2014) 342-345. doi:10.1016/j.jpowsour.2014.08.024.