MULTI-LAYER ELECTROLYTE ASSEMBLY FOR LITHIUM BATTERIES

Abstract

The invention relates to an electrolyte arrangement for a cell having at least one anode (1) and at least one cathode (3) comprising at least three superposed layers (2.1, 2.2, 2.3), wherein the middle layer (2.2) comprises a porous electrically nonconductive structure, and wherein a layer of a polymer-based electrolyte (2.1, 2.3) is arranged on both opposite sides of the porous electrically nonconductive structure, wherein at least one of the superposed layers (2.1, 2.2, 2.3) contains a ceramic material, wherein the ceramic material of the middle layer (2.2) is selected from metal ion-conductive ceramic material, a ceramic material which does not conduct metal ions, and/or mixtures thereof, and the ceramic material of the polymer-based electrolyte layer(s) (2.1, 2.3) is a metal ion-conductive ceramic material.

Claims

1. An electrolyte arrangement for a cell having at least one anode (1) and at least one cathode (3) comprising at least three superposed layers (2.1, 2.2, 2.3), wherein the middle layer (2.2) comprises a porous electrically nonconductive structure, and wherein a layer of a polymer-based electrolyte (2.1, 2.3) is arranged on both opposite sides of the porous electrically nonconductive structure, characterized in that at least one of the superposed layers (2.1, 2.2, 2.3) contains a ceramic material, wherein the ceramic material of the middle layer (2.2) is selected from metal ion-conductive ceramic material, a ceramic material which does not conduct metal ions, and/or mixtures thereof, and the ceramic material of the polymer-based electrolyte layer(s) (2.1, 2.3) is a metal ion-conductive ceramic material.

2. The electrolyte arrangement according to claim 1, characterized in that a layer of a polymer-based electrolyte (2.1, 2.3) is arranged on both opposite sides of a ceramic-free middle layer (2.2), and at least one of the polymer-based electrolyte layers (2.1, 2.3) contains a lithium ion-conductive ceramic material.

3. The electrolyte arrangement according to claim 1, characterized in that in the range from ≥50% by volume to ≤100% by volume, with preference in the range from ≥70% by volume to ≤100% by volume, preferably in the range from ≥80% by volume to ≤100% by volume, of the pore volume of the porous electrically nonconductive structure is filled with a polymer-based electrolyte.

4. The electrolyte arrangement according to claim 1, characterized in that the total thickness of the electrolyte arrangement (2) is in the range from ≥5μm to ≤300 μm, with preference in the range from ≥10 .sub.μm to ≤200 μm, preferably in the range from ≥15 μm to ≤100 μm.

5. The electrolyte arrangement according to claim 1, characterized in that the polymer-based electrolyte layer(s) comprise a polymer selected from the group comprising polyethylene oxide, polyethylene glycol, polypropylene carbonate, polyethylene carbonate, poly(bis((methoxyethoxy)ethoxy)phosphazene), polypropylene oxide, polysiloxane having an average molecular weight of from 300 g/mol to 10 000 g/mol, polyolefins, poly(vinylidene difluoride-co-hexafluoropropylene), polyethylene terephthalate, polyvinylidene fluoride, polyimide, carboxymethyl cellulose and/or polytetrafluoroethylene and/or mixtures and copolymers thereof.

6. The electrolyte arrangement according to claim 1, characterized in that the polymer-based electrolytes include at least one polymer and metal salt and optionally plasticizer and/or crosslinker, wherein the metal salt is preferably an organic or inorganic salt of lithium, sodium, magnesium, aluminium or zinc.

7. The electrolyte arrangement according to claim 1, characterized in that the polymer-based electrolytes include a polymer, a lithium salt and optionally plasticizer and/or crosslinker, wherein the molar ratio of polymer to lithium salt to plasticizer and to crosslinker is in the range from ≥0.5 to ≤20:1:≥0 to ≤10:≥0 to ≤1.

8. The electrolyte arrangement according to claim 1, characterized in that the proportion of ceramic material in a polymer-based electrolyte layer is in the range from ≥5% by weight to ≤80% by weight, preferably in the range from ≥35% by weight to ≤65% by weight, based on a total weight of the polymer-based electrolyte layer of 100% by weight.

9. An electrolyte electrode composite, comprising an anode (1), a cathode (3) and an electrolyte arrangement (2) according to claim 1 arranged between the anode and cathode.

10. A primary or secondary energy store, in particular lithium metal battery, solid-state battery, solid-state accumulator, lithium-air, lithium-oxygen or lithium-sulfur battery or accumulator or supercapacitor, comprising an electrolyte arrangement (2) according to claim 1.

Description

[0043] In the figures:

[0044] FIG. 1 shows a schematic illustration of a lithium metal battery which contains a three-layer electrolyte arrangement according to one embodiment of the invention as electrolyte.

[0045] FIG. 2 shows a schematic illustration of a further lithium metal battery which contains a three-layer electrolyte arrangement according to one embodiment of the invention.

[0046] FIG. 3 shows a schematic illustration of a further lithium metal battery which contains a three-layer electrolyte arrangement according to a further embodiment of the invention.

[0047] FIG. 4 shows a cross-sectional SEM image of a three-layer electrolyte arrangement according to one embodiment of the invention, containing a glass-ceramic in layer 2.3.

[0048] FIG. 5 shows a cross-sectional SEM image of a three-layer comparative arrangement in which the electrolyte in layer 2.3 does not contain glass-ceramic.

[0049] FIG. 6 shows in FIG. 6a) an SEM surface image of layer 2.3 of FIG. 4 and in FIG. 6b) an SEM surface image of layer 2.3 of FIG. 5.

[0050] FIG. 7 shows the calculated specific conductivity of the three-layer electrolyte arrangements according to Example 1 and the comparative arrangement at different temperatures.

[0051] FIG. 8 shows the evolution of the cell voltage over time during the galvanostatic polarization of a cell using the three-layer electrolyte arrangement according to Example 1 and the comparative arrangement.

[0052] FIG. 9 shows the cycling behaviour of an Li metal cell with LFP as cathode in the potential range from 4.2 V-2.0 V and with a rate of 0.1C using the three-layer electrolyte arrangements according to Example 1 and the comparative arrangement.

[0053] FIG. 10 shows the voltage profile of an Li metal cell with NMC811 as cathode with a voltage range from 4.3 V-2.0 V and a rate of 0.1 C using the three-layer electrolyte arrangements according to Example 1 and the comparative arrangement.

[0054] FIG. 11 shows the voltage profile of an LNMO lithium metal cell at 60° C. using the three-layer electrolyte arrangement according to Example 2.

[0055] FIG. 12 shows the evolution of the cell voltage over time during galvanostatic polarization using the three-layer electrolyte arrangement according to Example 3.

[0056] FIG. 13 shows the cycling behaviour of an Li metal cell with LFP as cathode in the potential range from 4.2 V-2.0 V and with a rate of 0.1C using the three-layer electrolyte arrangements according to Example 3 and a comparative arrangement.

[0057] FIG. 14 shows the voltage profile of an Li-O.sub.2 battery at room temperature using a three-layer electrolyte arrangement according to Example 3.

[0058] FIG. 15 shows the voltage profile of a lithium metal-sulfur cell at 60° C. according to

[0059] Example 4 and of a comparative arrangement.

[0060] FIG. 16 shows the voltage profile of an LFP lithium metal cell at 20° C. and at 60° C. using a three-layer electrolyte arrangement according to Comparative Example 5.

[0061] FIG. 1 shows a schematic view of a lithium metal battery which contains a three-layer electrolyte arrangement 2 according to one embodiment of the invention. The battery has a lithium metal anode 1 as negative electrode and a cathode 3 as positive electrode. The electrolyte arrangement 2 comprising three superposed layers 2.1, 2.2 and 2.3 is arranged between anode 1 and cathode 3. The middle layer 2.2 comprises a porous, electrically nonconducting membrane on both sides of which are arranged layers 2.1 and 2.3 of a polymer-based electrolyte. At least 50% of the pore volume of the porous membrane of layer 2.2 is filled with the polymer-based electrolyte which forms layers 2.1. The polymer-based electrolyte layer 2.3 contains a glass-ceramic material. The polymer used to form the polymer-based electrolyte layers 2.1 and 2.3 can be the same or different. In the illustration of FIG. 1, the glass-ceramic material-containing polymer-based electrolyte layer 2.3 is arranged on the cathode 3.

[0062] FIG. 2 shows a schematic view of a further lithium metal battery having a three-layer electrolyte arrangement 2 arranged between a lithium metal anode 1 and a cathode 3. The layer 2.2 comprises a porous, electrically nonconducting membrane on both sides of which are arranged layers 2.1 and 2.3 of a polymer-based electrolyte. In the illustration of FIG. 2, the glass-ceramic material-containing polymer-based electrolyte layer 2.3 is arranged on the anode 1.

[0063] FIG. 3 shows a schematic view of a lithium metal battery having a three-layer electrolyte arrangement 2 according to a further embodiment of the invention arranged between a lithium metal anode 1 and a cathode 3. In the embodiment according to FIG. 3, the middle layer 2.2 contains a ceramic material which does not conduct lithium ions. Such a material of layer 2.2 can be a porous ceramic Al.sub.2O.sub.3 membrane, for example a SEPARION® membrane. In this embodiment, the polymer-based electrolyte layers 1.2 and 2.3 do not contain lithium ion-conductive glass-ceramic material.

EXAMPLE 1

1.1 Production of a Three-Layer Electrolyte Arrangement

[0064] Electrolyte I: Electrolyte I was produced by dissolving polyethylene oxide (PEO, molecular weight=4 000 000, Dow Chemical), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI, Solvionic, 99.9%), 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (Pyr.sub.1,4TFSI, Solvionic, 99.9%) and benzophenone (ReagentPlus®, 99%, Merck) in acetonitrile (Merck, 99.8%) at a molar ratio of 10:1:2:0.121. The lithium salt was in this case set to 1. The molecular weight of polyethylene oxide was calculated on the basis of the repeating units. The solution was stirred for 24 hours at 60° C. in order to obtain a viscous dispersion for the coating. The dispersion of electrolyte I was applied with a wet film thickness of 400 μm to one side of a porous polymeric polypropylene membrane (Celgard® 2500, thickness 25 μm, porosity 55%, pore diameter 64 nm according to the manufacturer's information) using a doctor blade (doctor blade coating).

[0065] Glass-ceramic-containing electrolyte II: The dispersion of electrolyte I was also used for the non-ceramic part of electrolyte II. A NASICON-type phosphor glass-ceramic (Li.sub.1+xAl.sub.xTi.sub.2-x(PO.sub.4).sub.3; Schott AG, particle size approx. 0.2 μm) was to this end added to the non-ceramic electrolyte I in a ratio of 64% by weight:36% by weight. The electrolyte dispersion II was applied with a wet film thickness of 100 μm to the other side of the porous polypropylene membrane by means of doctor blade coating.

[0066] A comparative arrangement was produced by applying electrolyte dispersion I to both sides of the membrane. The arrangements were each dried under vacuum at a temperature of 60° C. for 24 h. The membranes were then irradiated with UV light for 10 minutes in order to initiate crosslinking of the polymer and to obtain crosslinked three-layer electrolyte arrangements.

1.2 Characterization of the Three-Layer Electrolyte Arrangements

1.2.1 Analysis of the Morphology

[0067] The SEM images of the membranes were generated using an Auriga® field emission scanning electron microscope (FE-SEM) with CrossBeam® Workstation with a Schottky field emission cathode (Carl Zeiss Microscopy GmbH). The accelerating voltage was 3 kV and an in-lens detector was used. The electrolyte arrangements were each applied to a carbon adhesive pad for the recording of the SEM images.

[0068] FIG. 4 shows a cross-sectional SEM image of the three-layer electrolyte arrangement containing a glass-ceramic-containing polymer electrolyte layer identified with 2.3. Layer 2.2 comprises the Celgard® membrane and layer 2.1 is formed from the polymer electrolyte I. As can be seen from FIG. 4, the total thickness of the three-layer electrolyte arrangement was 28 μm, with layer 2.1 having a thickness of 2 μm, layer 2.2 having a thickness of 21 μm and layer 2.3 having a thickness of 5 μm. It is assumed that electrolyte dispersion I penetrates into the pores of the porous membrane during the drying process.

[0069] FIG. 5 shows a cross-sectional SEM image of the three-layer comparative arrangement. In this case, layer 2.2 also comprises a Celgard® membrane, but both layers 2.1 and 2.3 have been formed from the polymer electrolyte I without glass-ceramic. As can be seen from FIG. 5, the total thickness of the comparative arrangement was 46 μm, with layer 2.1 having a thickness of 7 μm, layer 2.2 having a thickness of 26 μm and layer 2.3 having a thickness of 13 μm. Polymer electrolyte I had also penetrated into the pore volume of the membrane in the comparative arrangement.

[0070] FIG. 6a) shows an SEM surface image of the three-layer electrolyte arrangement containing a glass-ceramic in layer 2.3. As can be seen from FIG. 6a), the ceramic was distributed homogeneously in the layer. FIG. 6b) shows an SEM surface image of the comparative arrangement without glass-ceramic in layer 2.3. This exhibited a smooth polymeric surface.

1.2.2 Determination of the Proportion of the Polymer Electrolyte in the Pores of the Porous Membrane

[0071] The fill fraction (FF) of the polymer electrolyte in the cavities of the separator was calculated according to the following method. The porosity is defined as the ratio of cavity volume to apparent geometric volume. For the Celgard®2500 PP single-layer separator, the porosity is calculated as follows:

[00001]$\begin{array}{cc}\varphi \left(\%\right)=\left(1-\frac{W_{\mathrm{separator}}/{\rho }_{PP}}{V_{\mathrm{separator}}}\right)\times 100\%& \left[\mathrm{A1}\right]\end{array}$

where: [0072] W.sub.separator is the weight of the dry separator, [0073] ρ.sub.pp is the density of the semicrystalline polypropylene and [0074] V.sub.separator is the apparent geometric volume of a 25 μm thick separator. [0075] ρ.sub.pp is estimated here according to the indicated degree of crystallization (X.sub.c, %) of this separator type (X.sub.c=35) to:

[00002]$\begin{array}{cc}{\rho }_{PP}=\frac{100}{X_c/{\rho }_c+\left(100-X_c\right)/{\rho }_a}& \left[\mathrm{A2}\right]\end{array}$

where the terms ρ.sub.c and ρ.sub.a relate to the density of the crystalline phase (0.936-0.946 g/cm.sup.3) and of the amorphous phase (0.850-0.855 g/cm.sup.3). The value of ρ.sub.pp is accordingly estimated to be ≈0.88 g/cm.sup.3.

[0076] Accordingly, the fill fraction (FF) of the polymer-based electrolyte in the separator was calculated as:

[00003]$\begin{array}{cc}\mathrm{FF}=\frac{W_{\mathrm{TSPE}}/{\rho }_{\mathrm{TSPE}}}{\varphi V_{\mathrm{separator}}}\times 100\%& \left[\mathrm{A3}\right]\end{array}$

where W.sub.TSPE is the weight of the polymer-based electrolyte within the cavities of the separator, and ρ.sub.TSPE is the measured density of the polymer-based electrolyte with a value of 1.40 g/cm.sup.3.

[0077] As a result of the measurement of the average thickness of the polymer-based electrolyte-filled Celgard®2500 separator (˜47 μm) and of the average weight of the polymer-based electrolyte-filled Celgard®2500 separator and of the pure Celgard®2500 separator with a diameter of 1.6 cm (11.94 mg and 1.98 mg, respectively), the value of ϕ.sub.Celgard®2500 is 55% and the filling of the pore volume of the porous electrically nonconducting membrane with the polymer-based electrolyte was estimated to be ≈97%.

1.2.3 Determination of the Ionic Conductivity

[0078] The ionic conductivity of the three-layer electrolyte arrangements produced in Example 1.1 was determined by measuring the electrochemical impedance using a Novocontrol Technologies impedance measuring appliance with temperature control. The measurement cell consisted of two stainless-steel electrodes with an area of 2 cm.sup.2. The respective electrolyte arrangement was placed between these. The measurements were performed at temperatures in the range from 0° C. to 80° C.

[0079] FIG. 7 respectively shows the calculated specific conductivity of the three-layer electrolyte arrangement with glass-ceramic and of the comparative arrangement without ceramic at different temperatures. As can be gathered from FIG. 7, the impedance measurements showed that the addition of the ceramic into layer 2.3 improved the conductivity at all measured temperatures, for example by 0.67 mS/cm at 60° C.

1.2.4 Determination of the Thermal Shrinkage

[0080] The thermal shrinkage test can be used to ascertain the extent to which the membrane shrinks at a fixed temperature at a specific time. To this end, the electrolyte arrangements produced according to Example 1.1 and an uncoated Celgard® membrane were held at 120° C. for 30 minutes under vacuum. 3 experiments were carried out in each case.

[0081] The result of the analysis is shown in Table 1. This shrinkage is given as a ratio of the areas before and after the thermal treatment in per cent.

TABLE-US-00001 TABLE 1 Thermal shrinkage after 30 minutes at 120° C. Electrolyte Shrinkage Fel three-layer electrolyte arrangement <1 with glass-ceramic in layer 2.3 three-layer electrolyte arrangement  6 ± 1 without glass-ceramic Celgard ® 2500 membrane 17 ± 1

[0082] The result shows that the glass-ceramic in layer 2.3 can prevent shrinkage. This increases the safety of the batteries.

1.2.5 Determination of the Transference Number for Li.SUP.+

[0083] The transference number denotes the fraction of the total electrical current which is transported by a given ionic species such as Li.sup.+. The transference number (t.sub.Li+) of the three-layer electrolyte arrangements produced in Example 1.1 was determined using a symmetric Li/Li cell (CR2032 button cell) according to the method of Evans, Vincent and Bruce, as described in: Evans, J., Vincent, C. A., & Bruce, P. G. (1987). Electrochemical measurement of transference numbers in polymer electrolytes. Polymer, 28(13), 2324-2328.

[0084] The result is shown in Table 2 below. As can be gathered from this table, the addition of a NASICON glass-ceramic to electrolyte II increased the transference number:

TABLE-US-00002 TABLE 2 Transference number at 60° C. Electrolyte t.sub.Li+ three-layer electrolyte arrangement with 0.22 glass-ceramic in layer 2.3 three-layer electrolyte arrangement 0.08 without glass-ceramic

1.2.6 Analysis of the Dendrite Penetration

[0085] The effect of the Celgard® 2500 membrane on dendrite formation was analysed in a symmetric lithium/lithium cell (CR2032 button cell). Lithium metal foil served as the reference electrode. Galvanostatic polarization was effected at 0.1 mA cm.sup.−2 at 60° C. Analysis was performed on the three-layer electrolyte arrangement without glass-ceramic produced according to Example 1.1 and a monolayer polymer electrolyte made from electrolyte I having a thickness of 40-60 μm.

[0086] FIG. 8 shows the evolution of the cell voltage over time during the galvanostatic polarization. The cell voltage is given vs. Li/Li.sup.+. As can be gathered from FIG. 8, the electrolyte I in the symmetric Li—Li cell exhibited a drop in the cell voltage at a current strength of 3 mAh/cm.sup.2. The drop in the cell voltage indicates a short circuit which is caused by the lithium dendrites formed puncturing through the electrolyte. The electrolyte which comprises a Celgard® membrane in the middle layer (2.2) was able to survive at currents up to 50 mAh/cm.sup.2 without a short-circuit occurring as a result of penetrating dendrites. This shows that the porous, electrically nonconducting membrane in the middle layer can prevent penetration of the electrolyte by dendrites.

1.2.7 Analysis of the Three-Layer Electrolyte Arrangement in an Li Metal Full Cell

[0087] The three-layer electrolyte arrangements with glass-ceramic produced according to Example 1.1 and the comparative arrangement without ceramic were electrochemically analysed in lithium metal full cells of button cell design against different cathodes. A monolayer polymer electrolyte having a thickness of 40-60 μm made from electrolyte I was analysed as a further reference.

[0088] Lithium iron phosphate (LFP) and lithium nickel manganese cobalt mixed oxide (NMC811) cathodes were produced by mixing a slurry of LFP or NMC811, electrolyte I and carbon black in a weight ratio of 8:1:1 (w/w/w) and applying this with a wet film thickness of 150 μm to an aluminium foil by means of a doctor blade. The electrodes were dried at 110° C. for 24 hours under vacuum prior to use. Round electrodes were stamped out with a diameter of 12 mm and a surface loading of approximately 2 mg cm.sup.−2 for LFP and 5 mg cm.sup.−2 for NMC811.

[0089] The button cells were constructed as follows: the positive electrode was either the LFP or NMC811 electrode and the negative electrode was Li metal. In the cell structure the glass-ceramic-containing layer 2.3 contacted the positive electrode. Constant current cycling at 60° C. and with a rate of 0.1 C was performed with LFP as cathode in the potential range from 4.2 V to 2.0 V and with NMC811 in a range from 4.3 V to 2.0 V.

[0090] FIG. 9 shows the results of the cycling measurements with LFP as cathode over 50 cycles.

[0091] As can be gathered from FIG. 9, cycling stability was increased by the glass-ceramic in layer 2.3.

[0092] FIG. 10 shows the voltage profile of the Li metal cell with NMC811 as cathode for the three-layer electrolyte arrangement with glass-ceramic. As can be gathered from FIG. 10, NMC811 can also be used as a positive electrode for this electrolyte arrangement.

EXAMPLE 2

2.1 Production of a Three-Layer Electrolyte Arrangement

[0093] Electrolyte I: Electrolyte I was produced by dissolving polyethylene oxide (PEO, molecular weight=4 000 000), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (Pyr.sub.1,4TFSI) and benzophenone in acetonitrile at a molar ratio of 10:1:2:0.121. The lithium salt was in this case set to 1. The molecular weight of polyethylene oxide was calculated on the basis of the repeating units. The solution was stirred for 24 hours at 60° C. in order to obtain a viscous dispersion for the coating. The dispersion of electrolyte I was applied with a wet film thickness of 400 μm to one side of a porous polymeric polypropylene membrane (Celgard® 2500) using a doctor blade.

[0094] Glass-ceramic-containing electrolyte II: the non-ceramic part was produced by mixing polyvinylidene fluoride (PVDF), LiTFSI and Pyr.sub.1,4TFSI at a molar ratio of 7.2:1:2 in acetone and the solution was stirred at 50° C. LLZO (garnet type, Schott AG, particle size approx. 1 μm) was then added as ion-conducting ceramic to the non-ceramic part of the electrolyte in a ratio of 65% by weight to 35% by weight. A homogeneous dispersion was obtained after 2 hours of stirring at 50° C. This electrolyte dispersion II was applied with a wet film thickness of 250 μm to the other side of the polypropylene membrane.

[0095] The coated membrane was dried under vacuum at a temperature of 60° C. for 24 h. During this drying process, electrolyte dispersion I diffused into the porous polypropylene membrane and occupied approx. 80-100% of the pore volume. The membrane was then irradiated with UV light for 10 minutes in order to initiate crosslinking of the polymer and to obtain crosslinked three-layer electrolyte arrangements.

2.2 Electrochemical Characterization in an LNMO Full Cell

[0096] The three-layer electrolyte arrangement with glass-ceramic produced according to Example 2.1 was electrochemically analysed in lithium metal full cells of button cell design against a high-voltage lithium nickel manganese oxide (LNMO) cathode.

[0097] The positive electrode was produced by mixing the active material LNMO, the conductivity additive carbon black and PVDF as binder at a weight ratio of 8:1:1. After coating onto aluminium foil and drying, the electrode was completely wetted with a mixture of LiTFSI in Pyr.sub.1,4TFSI at a ratio of 1:2 and subsequently dried for 24 hours under vacuum at a temperature of 60° C. In the cell structure of the button cell, the glass-ceramic-containing layer was arranged on the positive electrode. The negative electrode was lithium metal. Constant current cycling at 60° C. and with a C rate of 0.1 C was performed in the potential range from 4.85 V to 3.5 V.

[0098] FIG. 11 shows the voltage profile of the Li metal cell with LNMO cathode and the three-layer electrolyte arrangement with glass-ceramic. As can be gathered from FIG. 11, the voltage profile showed that the cell could be cycled. Thus, a rechargeable solid-state electrolyte cell with a 5 V cathode could be provided.

EXAMPLE 3

3.1 Production of a Three-Layer Electrolyte Arrangement

[0099] Electrolyte I: Electrolyte I was produced by dissolving polyethylene oxide (PEO, molecular weight=4 000 000), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (Pyr.sub.1,4TFSI) and benzophenone in acetonitrile at a molar ratio of 10:1:2:0.121. The solution was stirred for 24 hours at 60° C. in order to obtain a dispersion for the coating. The dispersion of electrolyte I was applied with a wet film thickness of 400 μm to one side of a porous polymeric polypropylene membrane (Celgard® 2500) using a doctor blade.

[0100] Glass-ceramic-containing electrolyte II: The dispersion of electrolyte I was also used for the non-ceramic part of electrolyte II. For this purpose, a lithium ion-conducting ceramic LLZO (garnet type, Schott AG, particle size approx. 1 μm) was added to the non-ceramic electrolyte I in a ratio of 64% by weight:36% by weight. The electrolyte dispersion II was applied with a wet film thickness of 250 μm to the other side of the porous polypropylene membrane by means of doctor blade coating.

[0101] The coated membrane was dried under vacuum at a temperature of 60° C. for 24 hours. During this drying process, electrolyte dispersion I diffused into the porous polypropylene membrane and occupied approx. 80-100% of the pore volume. The membrane was then crosslinked under UV light for 10 minutes.

3.2 Analysis of the Dendrite Penetration

[0102] The dendrite formation when using the three-layer electrolyte arrangement produced according to Example 3.1 was analysed in a symmetric lithium/lithium cell (CR2032 button cell). Lithium metal foil served as the reference electrode. Galvanostatic polarization was effected at 0.1 mA cm.sup.−2 at 60° C.

[0103] FIG. 12 shows the evolution of the cell voltage over time during the galvanostatic polarization. The cell voltage is given vs. Li/Li.sup.+. As can be gathered from FIG. 12, the three-layer electrolyte arrangement with LLZO ceramic could survive at currents of up to 50 mAh/cm.sup.2. This shows that this electrolyte arrangement is usable under the practical demands placed on current electrolytes.

[0104] Evolution of the cell voltage over time during galvanostatic polarization using the three-layer electrolyte arrangement according to Example 3.

3.3 Analysis of the Three-Layer Electrolyte Arrangement in an LFP Full Cell

[0105] The three-layer electrolyte arrangement with LLZO ceramic produced according to Example 3.1 was analysed in full cells with a negative electrode made from lithium metal and a positive electrode made from lithium iron phosphate (LFP).

[0106] The LFP cathodes were produced by mixing a slurry of LFP, electrolyte I and carbon black in a weight ratio of 8:1:1 (w/w/w) and coating this with a wet film thickness of 150 μm onto an aluminium foil by means of a doctor blade. The electrode was dried at 110° C. for 24 hours under vacuum. Round electrodes were stamped out with a diameter of 12 mm and a surface loading of approximately 2 mg cm.sup.−2. The cell was installed in a button cell, the LLZO ceramic-containing electrolyte layer facing towards the negative electrode. Constant current measurement was performed in the potential range from 4.2-2.0 V and at a C rate of 0.1 and at 60° C.

[0107] FIG. 13 shows the voltage profile of the LFP-lithium metal cell at 60° C. As can be gathered from FIG. 13, the LLZO glass-ceramic-containing electrolyte could prevent a loss of capacity and increase cycling stability. It is assumed that this is due to the fact that fewer lithium dendrites were formed.

3.4 Electrochemical Characterization in an Li/Air Cell

[0108] The electrolyte arrangement was also analysed in a lithium/air cell with an air cathode (ECC-Air cell, EL-CELL GmbH). The cathode used was a commercial Co.sub.3O.sub.4/C coated on a nickel mesh with PTFE as binder (gas diffusion cathode (GDE), MEET Co., Ltd.). The catholyte used was 2 M LiTFSI in DMSO. Constant current cycling (CCC) was performed at a current strength of 0.2 mA/cm.sup.2 and a cut-off current strength of 3 mAh/cm.sup.2 under constant oxygen stream at room temperature (20° C.±2° C.).

[0109] FIG. 14 shows the voltage profile of the lithium/air cell at room temperature. As can be gathered from FIG. 14, the cell displayed a rechargeable behaviour. The result shows that the three-layer electrolyte arrangement, where the LLZO ceramic-containing electrolyte layer faces towards the negative electrode, provides an ion-conducting path. At the same time, the electrolyte can function as a protective layer for the side reactions between the lithium metal and DMSO and also oxygen in the diffusion cell.

EXAMPLE 4

Production and Electrochemical Characterization of a Three-Layer Electrolyte Arrangement in a Lithium-Sulfur Full Cell

[0110] To produce a middle layer 2.2, electrolyte I was first produced by dissolving polyethylene oxide (PEO, molecular weight=4 000 000), lithium bis(trifluoromethanesulfonyl)imide (LiTFSI), 1-butyl-1-methylpyrrolidinium bis(trifluoromethanesulfonyl)imide (Pyr.sub.1,4TFSI) and benzophenone in acetonitrile at a molar ratio of 10:1:2:0.121. The solution was stirred for 24 hours at 60° C. in order to obtain a dispersion for the coating. The dispersion of electrolyte I was applied with a wet film thickness of 300 μm to one side of a porous polymeric polypropylene membrane (Celgard® 2500) using a doctor blade.

[0111] The dispersion of electrolyte I was also used for the non-ceramic part of electrolyte II. For the electrolyte II, a lithium ion-conducting ceramic LLZO (garnet type, Schott AG, particle size approx. 1 μm) was added to the non-ceramic electrolyte I in a ratio of 65% by weight:35% by weight. To produce the polymer-based electrolyte layer 2.1, the electrolyte dispersion II was applied with a wet film thickness of 250 μm to the other side of the porous polypropylene membrane by means of doctor blade coating. The coated membrane was dried under vacuum at a temperature of 60° C. for 24 hours. During this drying process, electrolyte dispersion I diffused into the porous polypropylene membrane and occupied approx. 80-100% of the pore volume. The polymer was then crosslinked under UV light for 10 minutes.

[0112] A second lithium aluminium germanium phosphate ceramic (LAGP)-containing polymer-based electrolyte layer 2.2 was formed on the cathode. For this purpose, first a cathode was produced by mixing a slurry of sulfur active material (Merck), conductivity additive TIMCAL C-NERGY™ SUPER C65 Carbon Black and PEO (molecular weight=4 000 000) in a weight ratio of 60:23:17 (w/w/w) and coating this with a wet film thickness of 200 μm onto aluminium foil by means of a doctor blade. The electrode was dried at 40° C. for 12 hours under vacuum. An LAGP ceramic (Toshima Manufacturing Co., Ltd., Li.sub.1.5Al.sub.0.5Ge.sub.1.5P.sub.3O.sub.12, median particle size d.sub.50=1 μm) was added to the non-ceramic electrolyte I in a weight ratio of 70% by weight:30% by weight. This dispersion was then applied with a film thickness of 10 μm to the cathode, dried and crosslinked under UV light. A round electrode was stamped out with a diameter of 12 mm and a surface loading of approximately 2 mg cm.sup.−2.

[0113] The arrangement of LLZO-containing polymer-based electrolyte layer on the polypropylene membrane was arranged on the LAGP-containing polymer-based electrolyte layer of the coated cathode, the LLZO-containing electrolyte layer facing towards the negative electrode.

[0114] The three-layer electrolyte arrangement with cathode was installed in a button cell with a negative electrode made from lithium metal and analysed. As reference, a corresponding full cell without LAGP-containing layer was analysed. The constant current measurement was performed in the potential range from 3.0-1.5 V and at a C rate of 0.1 and a temperature of 60° C.

[0115] FIG. 15 shows the voltage profile of the lithium metal-sulfur cells using the electrolyte arrangement (“without LAGP-containing layer”) and the three-layer electrolyte arrangement (“with LAGP-containing layer”) at 60° C. As can be gathered from FIG. 15, the full cell with LAGP-containing layer exhibited a higher capacity. It is assumed that this can be attributed to the LAGP-containing layer preventing a “polysulfide shuttle effect”.

EXAMPLE 5

[0116] Three-Layer Electrolyte Arrangement with Ceramic-Containing Layer 2.2
5.1 Production of a Three-Layer Arrangement with Ceramic-Containing Layer 2.2

[0117] The electrolyte was produced by dissolving polypropylene carbonate (PCC, molecular weight 50 000), LiTFSI, Pyr.sub.1,4TFSI and benzophenone at a molar ratio of 5:1:2:0.12. The solution was stirred at 60° C. for 24 h. The electrolyte was applied with a wet film thickness of 400 μm to a porous ceramic Al.sub.2O.sub.3 membrane (SEPARION®) using a doctor blade. This electrolyte was also applied with a wet film thickness of 250 μm to the other side of the porous ceramic membrane using the doctor blade method. The coated membrane was dried under vacuum at a temperature of 60° C. for 24 hours. As a result of this process, electrolyte I penetrated into the pores of the membrane and occupied approximately 80-100% of the pore volume. The membrane was then crosslinked by irradiation with UV light for 10 minutes.

[0118] The positive electrode was produced by mixing the active material LFP, the conductivity additive carbon black and PVDF as binder at a weight ratio of 90:5:5. After coating onto aluminium foil, the electrode was completely wetted with the electrolyte and then dried under vacuum for 24 hours.

5.2 Electrochemical Characterization in the LFP Full Cell

[0119] The cell was constructed as a button cell using the LFP electrode as positive electrode and a lithium metal anode. The constant current measurements were performed in the range from 3.8-2.5 V at a C rate of 0.1 and a temperature of 20° C. and of 60° C.

[0120] FIG. 16 shows the voltage profile of the LFP-lithium metal cells at 20° C. and 60° C. As can be gathered from FIG. 16, the polymer-based electrolyte layer without glass-ceramic material was compatible with an LFP cathode with a discharge plateau at 3.3 V at 20° C. and at 3.4 V at 60° C.

[0121] This shows that a three-layer electrolyte arrangement, in which the middle layer contains a ceramic material which does not conduct lithium ions and is arranged between two layers of polymer-based electrolyte without glass-ceramic material, also displays good functionality.

[0122] The invention forming the basis for this patent application originated in a project funded by the BMBF (German Federal Ministry of Education and Research) under the funding code