DE102015005836A1 - Electrolyte, in particular solid electrolyte, preferably lithium ion-conducting material - Google Patents

Abstract

The invention relates to an electrolyte, in particular solid electrolyte, preferably lithium ion-conducting material, wherein the solid electrolyte, in particular the lithium ion-conducting material is a crosslinked system in which for a given stoichiometry of the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi and the sum of all cation Oxidation numbers, in each case multiplied by the oxidation number of each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi, whose mean weighted by the relative abundance of each cation gives numbers greater than "one", preferably greater than "two", characterized in that the relative permittivity ε of the electrolyte, in particular solid electrolyte, preferably of the lithium ion-conducting material above the value of 23.9817-312 41.5 / I * + 0.0000194269 * I *, preferably above 25.0319 - 33014.0 / I * + 0.0000207389 * I *, more preferably above 26.1809 - 34971.4 / I * + 0, 0000221847 * I *, even more preferably above 27.4431 - 37144.2 / I * + 0.0000237855 * I *, even more preferably above 28.8365 - 39570.2 / I * + 0.0000255681 * I *, still more preferably above 30.3821 - 42296.2 / I * + 0.000027565 * I *, even more preferably above 32.1061 - 45381 / I * + 0.0000298173 * I *, even more preferably above 34.0444-48900 , More preferably above 36.2286 - 52953.6 / I * + 0.0000353135 * I *, even more preferably above 38.7201 - 57670.8 / I * + 0.0000387142 * I *, even more preferably above 41.583 - 63229.0 / I * + 0.0000426993 * I *, where I * is the normalized ionic strength.

Description

The invention relates to an electrolyte, in particular a solid electrolyte, preferably a lithium ion-conducting material as a solid electrolyte. The solid electrolyte is an inorganic material suitable as a solid electrolyte, especially a separator material for a lithium ion battery.

Information on the structure of such systems and the materials involved can be found in Jennifer L. Schaefer, Yingying Lu, Surya S. Moganty, Praveen Agarwal, N. Jayaprakash, Lynden A. Archer, Electrolytes for High-Energy Lithium Batteries, Appl. Nanosci. 2 (2012), pp. 91-109 as well as in the patent US 2009 / 0317724A1 , Solids electrolytes for lithium batteries are said to have a high electrical bulk conductivity based almost exclusively on the movement of lithium ions and, accordingly, a negligible electronic contribution to electrical conductivity. Furthermore, the electrolyte, in particular the solid electrolyte, should also have stability against chemical reactions with the electrodes, especially with elemental lithium and lithium alloys, and thermal expansion coefficients suitable for the electrode materials, environmental compatibility, easy handling and low costs. This will be on Philippe Knauth: Inorganic solid Li ion conductors: An overview. Solid State Ionics 180 (2009), pages 911-916 whose disclosure is fully included. Furthermore, in the case of polycrystalline ceramic materials, it is advantageous if the grain boundary resistances are negligible. However, the desired properties do not have to apply or be maintained only for a specific temperature but for a temperature range. Thus, a certain constancy of ionic conductivity over this temperature range is required. Such a consistency in the temperature range of -50 ° C to 80 ° C is required especially for military applications. It is therefore an object of the invention to deliver electrolytes, in particular solid electrolytes whose ionic conductivity in a certain temperature range show minimal or small deviations, ie substantially constant are. The exact value of the ionic conductivity is of subordinate importance over the constancy over a certain temperature range.

JP5074466B2 describes a polymer-based solid electrolyte wherein the conductivity is represented by admixing a conductive powder and the base polymer has a minimum permittivity. US2003143467A1 describes an organic liquid-based electrolyte wherein the conductivity is represented by admixing a conductive clay powder and the liquid has a high permittivity. US2007092801A1 describes an electrolyte based on a molten salt with permittivitätserhöhen components.

KR20070082402A1 from LG Chemical Ltd. describes a special electrolyte (separator) consisting of porous inorganic particles in an organic matrix and in which the inorganic particles have a relative permittivity of at least 5.

US2009317724A1 describes a glass-ceramic or polymer-ceramic electrolyte with a dielectric additive selected from the group consisting of lithium oxide, boron nitride, silicon oxide, alumina, calcium oxide, zirconium oxide, titanium oxide, lithium aluminate and silicon nitride.

The object of the invention is to overcome the disadvantages of the prior art. In particular, it is an object of the invention to provide an ionic conductor which has an ionic conductivity over a certain temperature range, in particular the temperature range from -50 ° C. to + 80 ° C., which is largely constant over this temperature range. This corresponds, as will be explained below, to a low activation enthalpy of the ion movement in the solid electrolyte.

The object of the invention is surprisingly achieved when the electrolyte, preferably the lithium ion-conducting material is designed according to claim 1. Further developments, in particular special lithium ion-conducting materials are the subject of the dependent claims.

Particular preference is given to oxidic, sulphidic or nitridic compounds (or mixed forms).

In order to determine the activation enthalpy of the ions in the solid electrolyte, the solid electrolyte is regarded as a dielectric continuum with a dispersed or "dissolved" conductive oxide, in the sense of an extension of the concept of "interstitial solid solution", by A. Tiwari, Nanostructured materials research laboratory University of Utah. In the case of a lithium ion conductor, as exemplified here, Li ₂ O is assumed as the lead oxide. The "matrix" then forms the dielectric continuum.

Alternatively, a lead sulfide or a lead nitride or a mixed form can be used. From such a consideration of the dielectric continuum, it follows that the matrix should have as high a permittivity as possible. The larger this permittivity, the lower the interaction between the two ions of the Leitoxids, so lithium and oxygen, and the more freely mobile is the lithium ion.

The counterion, in the case of a Leitoxids so the oxygen, can be immobile and built into the matrix. In this case, the positively charged ions are identical to the lithium ions and the negatively charged counterions are negative excess charge carrying defects. This will be on Alan B. Lidiard, Ionic conductivity, Handbook of Physics 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 the disclosure of which is incorporated in the present application.

It is possible to quantify what has been shown above by establishing a term scheme for the lithium ions. Assuming first that the energy level of the lithium ions is only determined by the dielectric and that interactions with other ions, especially the counterions, are negligible, then a lithium ion is at an energy level approximately from the Born's formula for the free solvation enthalpy determine. This will be on M. Born, Volume and Hydration Heat of the Ions, Zeitschrift für Physik, No. 1, pp. 45-48 (1920) directed. This state is the state of an ion which is freely mobile in the dielectric and therefore corresponds to the excited state of an ion in the solid electrolyte, ie after the activation enthalpy has been supplied. This activation enthalpy is necessary for the lithium ion to move out of the potential well formed by adjacent negative charges in which it otherwise resides.

This potential well is not a fixed bond but a possible position in which a mobile ion can be located. Changing an ion from an immobile, tightly bound state to a mobile, state requires additional energy.

To Alan B. Lidiard, Ionic conductivity, Handbuch der Physik 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 , one can calculate the activation enthalpy necessary for the exchange of a mobile ion approximately using the extended Debye-Hückel theory. However, the charge carrier concentrations customary with ion conductors are considerably larger than fits the generally stated scope of the extended Debye-Hückel theory. It should also be considered that the Debye-Hückel theory is a fluid theory.

Surprisingly, it was found for the solid electrolytes according to the invention that the activation enthalpies of known ion conductors can nevertheless be calculated very well with the aid of the extended Debye-Hückel theory, so that the extended Debye-Hückel theory together with a continuum analysis of an ion conductor describe the solid-state electrolytes according to the invention.

A successful application of the Debye-Hückel theory in the formation of polarization layers on blocking electrodes in ionic conductors is described in JC Dyre, P. Maass, B. Roling, DL Sidebottom, "Fundamental questions concerning ion conduction in disordered solids", Rep. Prog. Phys. 72 (2009), 046501 ,

Regarding the formulas for the extended Debye-Hückel theory, reference is made to R. Schäfer, Documents on the Advanced Practical Course Physikalische Chemie I, Website, TU Darmstadt, Access 23.8.2013, the disclosure of which is included in full in the application. In contrast to Alan B. Lidiard, Ionic conductivity, Handbuch der Physik 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 , in the work of Schäfer the Debye-Hückel expression is not considered as energy, but as free enthalpy, so that a conversion by means of a Maxwell relation is necessary to get from the free enthalpy to the enthalpy. This will be applied below.

wobei e die Elementarladung, ε die relative Permittivität, ε₀ die Vakuumpermittivität, N_A die Avogadrozahl, R die Gaskonstante, T die absolute Temperatur und a der Ionenradius ist. Der Ionenradius beträgt für Li 90 pm, ( R. D. Shannon, ”Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides”, Acta Cryst, 1976, S. 751–767 ). Dabei handelt es sich jeweils um den – nach Shannon – im Vergleich zu anderen Ionenradiusangaben realistischeren „Crystal Radius”. Außerdem wird eine sechsfache Koordination mit den umgebenden Sauerstoffatomen unterstellt. Diese Annahme passt, wenn man die verschiedenen möglichen Koordinationen und die zugehörigen Ionenradien ansieht, am besten zu den Paulingschen Packungsregeln, siehe L. Pauling, ”The principles determining the structure of complex ionic crystals”, J. Am. Chem. Soc. 51 (4), 1929, S. 1010–1026 . Die Ionenstärke I (in Mol pro Volumen) ist nach R. Schäfer, Unterlagen zum Fortgeschrittenen-Praktikum Physikalische Chemie I, Website, TU Darmstadt, Zugriff 23.8.2013 , definiert als:

For the deviation of the chemical potential of the ions with valence z _i from the behavior of the ideally diluted solution, one obtains with it, thus according to the extended Debye-Hückel theory:

where e is the elementary charge, ε the relative permittivity, ε ₀ the vacuum permittivity, N _A the Avogadro number, R the gas constant, T the absolute temperature and a the ionic radius. The ionic radius for Li is 90 pm, ( RD Shannon, "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides", Acta Cryst, 1976, pp. 751-767 ). These are each - according to Shannon - compared to other ionic radius data more realistic "Crystal Radius". In addition, a sixfold coordination with the surrounding oxygen atoms is assumed. This assumption, when considering the different possible coordinations and the associated ionic radii, fits best with Pauling's packing rules, see L. Pauling, "The principles of the structure of complex ionic crystals", J. Am. Chem. Soc. 51 (4), 1929, pp. 1010-1026 , The ionic strength I (in moles per volume) is after R. Schäfer, Advanced Physics Chemistry I, Website, TU Darmstadt, Access 23.8.2013 , defined as:

N _i is the number of the j-th ion species (only the lithium ions and the oxygen ions necessary for the charge compensation of the lithium ions are considered) in volume V.

Across from Alan B. Lidiard, Ionic conductivity, Handbuch der Physik 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 , the simplifying assumption is made that all lithium ions are equally mobile and all are available for charge transport.

erhält man dann für die Enthalpie pro Teilchen h_i:

With the following Maxwell relation, ( M. Ding, K. Xu, S. Zhang, TR Jow, K. Amine, G. Henriksen, "Change of Conductivity with Salt Content, Solvent Composition, and Temperature for Electrolytes of LiPF6 in Ethylene Carbonate-Ethyl Methyl Carbonate". Journal of The Electrochemical Society, 148, 2001, p. A1196-Al204 )

then one obtains for the enthalpy per particle h _i :

This activation enthalpy depends on the ionic strength I and the permittivity ε.

By permeability of the matrix is meant the background permeability, ie the permeability resulting from the sum of electronic and ionic polarizability, without the fraction resulting from the conductivity itself. This background permeability is the percentage of permeability measured at frequencies below the infrared ( Charles Kittel, Introduction to Solid State Physics, 5th Edition, R. Oldenbourg-Verlag Munich Vienna, 1980, p. 441 ), up to frequencies at which the effect of conductivity occurs ( Ekaterina I. Igorodina, Maria Forsyth, and Douglas R. MacFarlane, "On the components of the dielectric constant of ionic liquids: ionic polarization?", Phys. Chem. Chem. Phys. 11, 2452-2458 (2009) ). Orientation polarization effects due to alignment of permanent dipoles do not exist in the considered solid electrolyte.

The activation enthalpy thus determined is, as stated above, an important factor and in two ways. It has, as also listed above, an influence on the ion mobility and thus contributes in addition to the carrier density, the number of free (intermediate) lattice sites u. a. the magnitude of the ionic conductivity in the desired temperature range at, for. B. in the environment of room temperature. In addition, it is crucial for the constancy of the ionic conductivity in a temperature range, in particular in the range -50 ° C to + 80 ° C.

For this, an expression of ionic (DC) conductivity containing the enthalpy of activation is considered. In the simplest case, this is, as in N. Dietrich, preparation and characterization of β-eukryptite and modified eucryptite ceramics LiAl _1-y M _y SiO ₄ (M = Cr, Mn, Fe) for use as High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, formulated in 2007, represented by a simple Arrhenius law:

In this case, σ _{0 is} a prefactor, which includes the density of the charge carriers per volume, the mean jump distance, the jump test frequency and the density of the free (intermediate) lattice sites necessary for a jump.

Theoretically more satisfying is the following approach ( N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 ):

In equation (6) the Boltzmann constant k, the absolute temperature T and the activation energy or enthalpy E _{A are} in the exponential term. Equation of enthalpy and energy is legitimate for solids. E _A thus corresponds to h _i from equation (4).

The conversion from E _A * to E _A at a temperature T takes place in such a way that the condition is set that the relative slopes of both representations coincide at this temperature T. Then you have to calculate: E _A = E * / A + kT (7)

The highest possible constancy of the ion conductivity in a temperature range is achieved when the value for the activation enthalpy determined by the permittivity and the ionic strength is as small as possible. This is achieved by selecting components of the highest possible polarizability and appropriately measuring the ionic strength. It is particularly preferred if the material is selected so that the requirements of the extended Debye-Hückel theory are fulfilled as well as possible. This means that the compensating charges around the cation are as distributed as possible in a liquid. This is achieved particularly well in a glass as a supercooled liquid. Even partially or polycrystalline materials meet these requirements well.

The adjustment of the ionic strength I via the stoichiometry, the adjustment of the relative permittivity ε on the dielectric polarizabilities of the ions involved. In the selection of the components, it can be based on the fact that the electronic polarizability, ie the effect of the displacement of the respective electron cloud relative to the core, is proportional to the volume of the ion. This will be on Jai Shanker, MP Verma, Correlation between electronic polarities and ionic radii in alkali halides, Journal of Physics and Chemistry of Solids, Volume 37, Issue 9, 1976, Pages 883-885 and SC Agrawal, OP Sharma, Jai Shanker, "Effect of the polar potential on electronic polarities in alkaline earth chalcogenides", Pramana Journal of Physics, January 1978, Volume 10, Issue 1, pp 11-15 , referenced. Furthermore, it is pointed out that the ionic polarizability, ie the effect of the mutual displacement of ions of different charge, is also great in the case of large-volume and thus weakly bound ions, so that the latter can also be considered proportional to the volume of the respective ions. In this regard, the electronegativity table in Pauling, The Nature of the Chemical Bond, New York, Cornell University Press, 1960 directed.

For oxidic and fluoridic systems has RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , the "additivity rule" with regard to the dielectric polarizability, ie the sum of electronic polarizability and ionic polarizability, according to which this dielectric polarizability can be represented to a good approximation as a linear superposition of fixed values for the dielectric polarizabilities of the ions. The paper by RD Shannon "Dielectric polarizabilities of ions in oxides and fluorides" is fully incorporated herein by reference.

The above has fixed values for the polarizabilities RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 tabulated: Table 1: Dielectric polarizabilities according to Shannon

These values are also used for (semi) sulphidic and (semi) nitridic systems. In view of the above proportionality of the polarizability of an atom to the atomic or ionic volume, values are subsequently used for the polarizability of the trivalent nitrogen anion and the divalent sulfur anion which are calculated from the polarizability of the oxygen anion by conversion using the ionic radii. The ionic radii used become RD Shannon, "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides", Acta Cryst, 1976, pp. 751-767 taken. This is the RD Shannon aa O. more realistic compared to other ionic radius "Crystal Radius". When converting from the oxygen ion to the (much larger) sulfur ion, it refers to the coordination number "6" and obtains the dielectric polarizability of the sulfur ion to 4.93 Å ³ . When converting from the oxygen ion to the (only slightly larger) nitrogen ion, the coordination number "4" is used, giving the dielectric polarizability of the nitrogen ion as 2.42 Å ³ .

The polarizability of the hafnium ion, which is not mentioned above, becomes apparent A. Feteira, D. Sinclair, K. Rajab, M. Lanagan, "Crystal Structure and Microwave Dielectric Properties of Alkaline-Earth Hafnates, AHfO3 (A = Ba, Sr, Ca)", J. Am. Ceram. Soc., 91 [3], 2008, pp. 893-901 directed.

The relative permittivity is then calculated using this information according to the Clausius-Mosotti equation from the total molecular polarizability am. This will be on RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 The disclosure content of which is fully incorporated into the present application.

Formula (8) uses the Gaussian unit system. V _m stands for the molecule volume.

The following table lists exemplary values for the activation enthalpy h at room temperature as a function of relative permittivity ε and ionic strength according to the formulas described in detail above. The values for the ionic strength I are given for two glass-ceramics, which are considered in more detail below, namely LAS (lithium aluminum-silcium) glass-ceramics with an ionic strength I of 31804 mol / m ³ and LLZO (lithium / lanthanum / zirconium oxide, e.g. Li ₇ La ₃ Zr ₂ O ₁₂ with an ionic strength I of 63768 mol / m ³ ) and lithium beta-alumina with an ionic strength I of 11312 mol / m ³ . The relationship between activation enthalpy h, ionic strength I and relative permittivity ε follows from the formula (4). The above indexing of "h" to identify the considered ion species is omitted here. "H" therefore always refers to the lithium ion.

Tabelle 2: Beispiele für den Zusammenhang zwischen Ionenstärke I, relativer Permittivität ε und Aktivierungsenthalpie h gemäß vorbeschriebener Theorie. Die Zeilen in Tabelle 2 beziehen sich auf feste Ionenstärken, die Spalten auf feste relative Permittivitäten und die Tabelleneinträge auf die sich hieraus ergebenden Aktivierungsenthalpien h. The ionic radius of lithium to be used in this formula is assumed to be 90 μm as above.

Table 2: Examples of the relationship between ionic strength I, relative permittivity ε and activation enthalpy h according to the above-described theory. The rows in Table 2 refer to fixed ionic strengths, the columns to fixed relative permittivities and the table entries to the resulting activation enthalpies h.

By interpolation or calculation with equation (4) it is found that for the lowest of the considered ionic strengths - 11312 mol / m ³ - from a relative permittivity ε of about 21.5, an activation enthalpy of below 0.265 eV is achieved. This value has a special meaning.

Below this activation enthalpy, the conductivities at the limits -50 ° C and + 80 ° C of the temperature range to be considered differ vertices of less than a factor of 100, see equation (6). At the average ionic strength (31804 mol / m ³ ) this is only the case from a relative permittivity ε of about 23.7 and at the highest considered ionic strength of 63768 mol / m ³ only from an ε of about 24.7.

The latter can be represented by a simple mathematical formula, if the relationship between the activation enthalpy on the one hand and the ionic strength I as well as the permittivity ε on the other hand is inferred from the relationship (4). This relationship is partially inverted and considered relative to a certain enthalpy of activation and to a certain ionic strength I relative permittivity. The temperature and the ionic radius are fixed; in the present case, these are the room temperature and the ionic radius of lithium. The numerical values belonging to the parameter ranges of interest can be read off the table above. To arrive at a simple mathematical formula, one interpolates these numerical values. The interpolation used here includes the minus first to minus second order in the enthalpy of activation, the minus first to first order in the ionic strength and all the mixed terms that can be combined therefrom. We now introduce the dimensionless ionic strength I * = I / (mol / m ³ ) and obtain that the relative permittivity must be above the value 23.9817 - 31241.5 / I * + 0.0000194269 · I * again to obtain an activation enthalpy of less than 0.265 eV. The following table shows how to further graduate: Table 3: Relative permittivities ε according to the invention for achieving the desired maximum changes in ionic conductivity in the range from -50 ° C to + 80 ° C

According to the invention, ion conductors which fulfill one of these conditions, and which preferably at least partially consist of glass former for oxide glasses.

This means that in the invention the electrolytes are crosslinked systems in which, for a given stoichiometry, the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi and the sum of all cation oxidation numbers for each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi multiplied by its oxidation number , Gives numbers whose mean weighted by the relative abundance of the respective cation is greater than "one", preferably greater than "two". The relative permittivity ε of the electrolyte, in particular solid electrolyte, preferably of the lithium ion-conducting material is above the value
23.9817 - 31241.5 / I * + 0.0000194269 · I *,
preferably above
25.0331 - 33014.0 / I * + 0.0000207389 * I *, more preferably above
26,1809 - 34971,4 / I * + 0,0000221847 · I *, even more preferably above
27,4431 - 37144,2 / I * + 0,0000237855 * I *, even more preferably above
28,8365 - 39570,2 / I * + 0,0000255681 * I *, even more preferably above
30.3821 - 42296.2 / I * + 0.000027565 * I *, even more preferably above
32,1061-45381 / I * + 0,0000298173 * I *, even more preferably above
34.0414 - 48900.8 / I * + 0.0000323777 * I *, even more preferably above
36,2286 - 52953,6 / I * + 0,0000353135 · I *, even more preferably above
38,7201 - 57670,8 / I * + 0,0000387142 · I *, even more preferably above
41.583 - 63229.0 / I * + 0.0000426993 · I *
where I * is the normalized ionic strength of the electrolyte.

The numbers resulting from the previous step for each cation ("... the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb , Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi and the sum of all cation oxidation numbers, for each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi multiplied by its oxidation number ... "), correspond to the average number of oxygen bridges that each cation B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al , Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi with another cation B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi.

These mean oxygen bridge numbers derived from a network-forming cation B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr , Hf, Cr, Mn, Fe, Tl, Pb, Bi, emanate to other network-forming cations, if more than one type of network-forming cations is involved, at least "1" weighted at the relative frequency of the respective cation, preferably at least "1". 2 "amount.

In the following, the case where only one kind of network-forming cation is involved will be considered. In this case, there is only one kind of network-forming cation, for example zirconium, so that the average number of oxygen bridges leading from one zirconium atom to another zirconium atom is identical to the mean weighted by the relative abundance of the particular cation and that of the relative Frequency of the respective cation weighted average greater than "one", preferably greater than "two" is. "

If more than one network-forming cation species are involved, it should be averaged over them, with the respective number of oxygen bridges to other network-forming cations to be weighted with the relative abundance of the particular network-forming cation species within the total of the network-forming cation species.

The "relative abundance" refers to each of the aforementioned total of network-forming cations. On average, an averaging thus describes the cations of the type "network former", "imperfect network former", "structurally active elements", or "conditional glassformer".

The general reason for the latter is described below.

The relative permittivity ε is initially intended to mean the value measured on the material. Regarding the measuring method see z. B. Agilent, "Basics of Measuring the Dielectric Properties of Materials" - Application Note, June 26, 2006 ,

The procedure for the design of such a material is as follows. Initially, as indicated below, a combination of elements suitable for glass formation and a corresponding compositional range are selected. The adjustment of the ionic strength is carried out via the stoichiometry, the adjustment of the relative permittivity via the estimation of RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , The information required for the latter of the molecule volume or the density in order to calculate the molecular volume can be done for glasses with the aid of relevant nonlinear fits. Such are, including the appropriate fit parameters, as finished software packages under the trade names INTERGLAD (NEW GLASS FORUM, Japan Glass Industry Center, 3-21-16, Hyakunin-cho, Shinjyuku-ku, Tokyo 169-0073 JAPAN) or SCIGLASS ( ITC, Inc., 22 Considine Rd., Newton, Ma. 20459, USA). For crystals or the crystallites in polycrystalline materials, the density z. B. determine by means of density functional calculations. In this regard, reference is made to the codes and tutorials of the CPMD Consortium, see http://cpmd.org/ , reachable via Max Planck Institute for Solid State Research Stuttgart, Heisenbergstraße 1, 70569 Stuttgart, Germany. Compare too R. Jalem, M. Nakayama, T. Kasuga, K. Kanamura, Abstract # 577, 224th ECS Meeting, 2013 ,

Which components are suitable as glass formers can be found in the structural glass formation theories. Glass forming systems are all those systems that form loose networks, see WH Zachariasen, The Atomic Arrangement in Glass, J. Chem. Soc. 54, No. 10 (1932), pp. 3841-3851 , This requires the occurrence of special cations known as network builders. A common classification of cations as network formers or network transducers or intermediate oxides follows the field strength of the cations, see A. Dietzel, "The Cation Field Strengths and Their Relationships with Devitrification, the Bonding, and the Melting Points of Silicates," Zeitschrift fur Elektrochemie und angewandte Physikische Chemie 48 (1942), pp. 9-23 , Networkers are more covalent, network converters more ionic oxygen bonds, see TS Izumatani, Optical Glass, AIP Trans. Series, New York (1986) ,

Incidentally, here one sees a connection between the question of glass formation and ionic conductivity; This is the above general reason why the ion conductors should preferably at least partially consist of glass formers. Stretching wide-meshed networks in which lithium ions can move well requires the presence of network-forming agents, see RD Shannon, T. Berzins: Ion conductivity in low carnegieite compositions based on NaAlSiO4. Mat. Res. Bull. 14 (1979), pp. 361-367 where it is specifically mentioned that network structures containing the tetrahedron (BO ₄ ) ^5- , (PO ₄ ) ^3- or (SiO ₄ ) ^4- , have a high ion mobility.

This fits in with the fact that network designers reduce the packing density in glass and vice versa, see T. Rouxel, Elastic Properties and Short-to Medium-Range Order in Glasses, J. Am. Ceram. Soc., 90, No. 10 (2007), pp. 3019-3039 ,

The recommendation of a high proportion of network educators RD Shannon, T. Berzins: Ion conductivity in low carnegieite compositions based on NaAlSiO4. Mat. Res. Bull. 14 (1979), pp. 361-367 , but is not limited to glassy ionic conductors, since the tendency of the network formers to open covalently bound, wide-meshed networks that are well suited to ionic conductivity is not exclusively coupled to the glassy state.

Tabelle 4: Dietzelsche Feldstärken nach R. Conradt: Vorlesung Werkstoffkunde Glas, Kapitel 2: Glaschemie. Lehrstuhl für Glas und keramische Verbundwerkstoffe, Institut für Gesteinshüttenkunde, Rheinisch-Westfälische Technische Hochschule Aachen, Stand Oktober 2010 Conradt shares in an extension of Dietzel's field strength theory, see R. Conradt: Lecture Material Science Glass, Chapter 2: Glass Chemistry. Chair of Glass and Ceramic Composites, Institute of Mineralogy, RWTH Aachen University, October 2010, cations in "network formers" (B, Si, Ge, P), "imperfect network formers" (Sn, As, Sb , Se, Te, Ti, V, Nb, Ta, Mo, W), structurally active elements (Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe as a limiting case), conditional glassformers ( Tl, Pb, Bi), "network modifiers 2" (Mg, Zn, Cd, Sc, Y, La and the lanthanides, Co, Ni) and "network modifiers 1" (Li, Na, K, Rb, Cs, Ca , Sr, Ba), wherein the covalent bond character in a bond with oxygen decreases from the "network formers" to the "network modifiers 1" and the ionic character correspondingly increases. Periodic table: elements (ions) / top left valence / top right coordination / bottom Dietzel field strength / after Conradt

Table 4: Dietzel field strengths according to R. Conradt: Lecture Material Science Glass, Chapter 2: Glass Chemistry. Chair of Glass and Ceramic Composites, Institute of Mineral Engineering, RWTH Aachen University, October 2010

In order to reach the goal of a wide-meshed network, a high percentage of network creators is recommended. However, it is known that systems with a network-forming component of less than 50 mol% are capable of glass formation, see HJL Trapp, JM Stevels, "Physical properties of inverted glasses", Glastechn. Ber. 32 (1959), p. 32-52 , and thus generally for the construction of network structures (also, for example, in partially or polycrystalline materials).

According to the invention, the boundary between a (glassy or non-glassy) network-forming and a non-network-forming system is drawn so that at least as many cations of the type "network former", "imperfect network former", "structurally active elements", or (in the sense a "non-exclusive" or "conditional glassformer" is that in an oxide embodiment of the invention and random arrangement of atoms on the average, at least one of the oxygen bridges emanating from these cations leads to another of these cations. This corresponds to the minimum condition that must be set for an inverted glass so that there is even a remnant of network at all. The cooling of such systems is tricky, but feasible, see S. Misture, Viscous Glass Sealants for SOFC Applications, DOE Award Number: DE-NT0005177, Alfred University, 2012 ,

After the requirement for high relative permeability, this is another requirement according to the invention.

An approximate, approximate measure of this is obtained by first considering the sum of the oxidation numbers of all cations of the type "network former", "imperfect network former", "structurally active elements", or "conditional glassformer" as well as for the given stoichiometry the sum of all cation oxidation numbers is calculated, forms the quotient and then multiplied by its oxidation number for each type of network former, imperfect network former, structurally active elements, or conditional glassformer. These products should then be greater than "one" in the average weighted average relative to the particular cation.

As noted above, network builders tend to be more covalent oxygen bonds, see TS Izumatani, Optical Glass, AIP Trans. Series, New York (1986) , Therefore, the network of network-forming and connecting oxygen atoms can be considered as a covalent network, in which the coordination number of each atom is dictated by the bond, see Tarsilla Gerthsen, Chemistry for Mechanical Engineering: Inorganic Chemistry for Materials and Processes, KIT Scientific Publishing, Karlsruhe, 2006, p. 217 , and thus in the case of single bonds is equal to the oxidation number. Accordingly, each network-forming cation is then surrounded by as many oxygen atoms as corresponds to its oxidation number. This network-forming cation is called Knb, i. All of these oxygen atoms are then coordinated with other cations according to their oxidation number "2", either by (a) covalent bonds with one additional network-forming cation or (b) ionically with network-converting cations. Taking into account the electroneutrality of the overall system, the relative frequency with which case (a) occurs is equal to the proportion of the cumulative oxidation numbers of all network-forming cations to the accumulated oxidation numbers of all cations. This relative abundance is equal to the quotient of the sum of the oxidation numbers of all cations of the type "network former", "imperfect network former", "structurally active elements", or "conditional glassformer" and the sum of all cation oxidation numbers. If this relative frequency is multiplied by the oxidation number of K _{nb, i} , one obtains the average number of further network-forming cations to which K _{nb, i is} connected.

In general, the above assumption of purely covalent bonds is not met, and the bonds of the network-forming cations to the surrounding oxygen atoms are partition bonds. How large is the ionic portion of each bond is Linus Pauling, The Nature of the Chemical Bond, Cornell University Press, New York, 1960 , refer to.

The number of oxygen atoms surrounding a network-forming cation, ie the coordination number of this cation, then follows the packing rules of Linus Pauling, The Nature of the Chemical Bond, Cornell University Press, New York, 1960 , This coordination number of the cation is generally greater than its oxidation number, cf. Shannon "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides, Acta cryst., Pp. 751-767, 1976 , It, that is, the coordination number of the cation, is then always greater than or equal to its oxidation number, if the average coordination number of the surrounding oxygen atoms is greater than or equal to "2", ie the amount of the oxidation number of the oxygen, so if the oxygen only single bonds to his Neighbors arrive. This follows directly from the packing rules of Linus Pauling, The Nature of the Chemical Bond, Cornell University Press, New York, 1960.

You could now change the calculation above and z. B. in the multiplication of the relative abundance with the oxidation number of the cation K _{nb, i} replace this oxidation number by the (larger) coordination number, so as to take into account the stronger networking of the cation K _{nb, i} with its neighbors. But that would run counter to the goal of establishing a condition from which one can speak of a covalently spanned network. If the ionic nature of the cation K _{nb, i} binding increases with its neighbors because another element with a greater difference in electronegativity to oxygen is used, the above-formulated condition should not be mitigated.

On the contrary, the fact that the ionic nature of the bond to oxygen tends to increase from the "network formers" through the "imperfect network formers" and the "structurally active elements" to the "conditional glassformers" is accommodated by the opposite the condition formulated above, the following case is preferred. The quotient of the sum of the oxidation numbers of all cations of the type "network former", "imperfect network former" or "structurally active elements" and the sum of all cation oxidation numbers is formed at all. For each type of cation "network former", "imperfect network former "," structurally active elements "is multiplied by its oxidation number. These products should then be greater than "one" in the average weighted average relative to the particular cation.

More preferable is the following case. The quotient of the sum of the oxidation numbers of all cations of the type "network former" or "imperfect network former" and the sum of all cation oxidation numbers is formed at all. For each cation of the type "network former" or "imperfect network former" is multiplied by its oxidation number. These products should then be greater than "one" in the average weighted average relative to the particular cation.

Even more preferred is that in the previous section is two instead of an oxygen bridge, the said products are therefore all greater than "two".

The above illustration refers to oxide systems. The partial or complete substitution of the oxygen as an anion by sulfur and / or nitrogen is included in the disclosure.

The network forming system as described above contains a portion of conductive ions, preferably lithium, which is formally considered to be a component of a conductive oxide or conductive nitride or mixture form contained in the composition. This conductive oxide or conductive sulfide or conductive nitride is preferably present in an ionic strength of at least 1000 mol / m ³ .

The invention will be described below based on embodiments without limitation.

First, for comparison, a material as a comparative example, which is very far removed from an ion-conductive material of the present invention having a low activation enthalpy at room temperature, namely lithium aluminum silicate (LAS) as Comparative Example 1, is considered.

LAS exists in a crystal structure (β-eucryptite), in which an ion movement by exchange of lithium from an aluminum-near to a silicon-near place is possible. This will be on A. Lichtenstein, R. Jones, S. Gironcoli, S. Baroni, "Anisotropic thermal expansion in silicates: A density functional study of β-eucryptites". Physical Review B 62 No 17, 2000, pp. 11487-11493 directed. LAS is at 600 ° C in the modification belonging to this temperature H. Guth, "Structure Investigations on the One-Dimensional Li-Ion Guide β-Eucryptite (LiAlSiO4) Using Neutron Diffraction", Dissertation, Nuclear Research Center Karlsruhe, September 1979 , known as a good ionic conductor, but not at room temperature.

For the relative permittivity, the measurement result becomes 5.5 for a β-eukryptite ceramic of T. Ogiwara, Y. Noda, K. Shoji, O. Kimura, "Low-Temperature Sintering of High-Strength β-Eucryptite Ceramics with Low Thermal Expansion Using Li 2 O-GeO 2 as a Sintering Additive". J. Am. Ceram. Soc., 94, 2011, pp. 1427-1433 used. If it is further assumed that the ionic strength of LAS is 31804 mol / m ³ because of the stoichiometry and the density, then one can calculate the activation enthalpy. The density is taken from D. Barthelmy, website http://www.webmineral.com/data/Eucryptite.shtml , Accessed October 27, 2013, to 2.67 g / cm ^3. For the calculation of the ionic strength it is assumed that one mole of LiAlSiO _{4 contains} half a mole of the lead oxide Li ₂ O; analogously, one also proceeds in all other examples. The calculated result is compared with the measurement result from the study of N. Dietrich, production and characterization of β-eukryptite and modified eukryptite ceramics LiAl _1-y M _y SiO ₄ (M = Cr, Mn, Fe) for use as high-performance materials , Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007, specifically with the activation enthalpy of the grain conductivity of polycrystalline β-eukryptite, which was obtained from a solid-state reaction. Method or reference h _i READ Formula (4); ε = 5.8 (Ogiwara, Noda, Shoji, & Kimura) 1.16 eV Experiment / LAS from solid-state reaction (Dietrich) 1.10-1.22 eV, mean 1.17 eV Table 5: Activation enthalpy of lithium ion conductivity in LAS theoretically according to formula (4) and experimentally

As shown in Table 5, the experimentally determined value is of the same order of magnitude as the theoretical value obtained by the formula (4). Because of the large value of the activation enthalpy fulfilled This material, however, the requirement for a flat course of the conductivity-temperature curve in the range -50 ° C to + 80 ° C not. In order to lower the activation enthalpy, it would be necessary to chemically change the material in the sense of the invention in such a way that the relative permittivity increases greatly. However, because of the large discrepancy between the given and a relative permittivity according to the invention, this is not possible. Therefore, this material is unsuitable as a starting point for an embodiment of the invention.

A typical conductivity value is 10 ^-5 S / cm at 350 ° C, see N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 , With the activation enthalpy of 1.17 eV, this value is extrapolated to about 5 × 10 ^-16 / (Ω · cm) at 25 ° C.

As a next comparative example, consider a material with a slightly higher conductivity and lower activation enthalpy at room temperature, namely a lithium phosphate glass (LiPO ₃ ). For the relative permittivity, we use the measurement result 8.74 of Watanabe, T. Yamamoto, and T. Yamada, "The Chemical Structure and Electric Property of Lithium Phosphate Glass," Memoirs of College of. Engineering, Chubu University p. 71 (1984) , We still use that the ionic strength of LiPO _{3 is} 46781 mol / m ³ due to its stoichiometry and density, thus calculating the activation enthalpy. The density is taken together with the experimentally determined activation enthalpy and the conductivity F. Muñoz, A. Durán, L. Pascual, L. Montagne, B. Revel, A. Rodrigues, Increased electrical conductivity of LiPON glasses produced by ammonolysis, Solid State Ionics 179, 574-579 (2008) , Since the experimental enthalpy determined experimentally at 0.76 eV corresponds to E _A ^* , we correct for the comparison by kT, where we use for T room temperature. Method or reference h _i LiPO ₃ Formula (4); ε = 8.74 (Watanabe, Yamamoto, & Yamada) 0.77 eV Experiment / LiPO3 glass (Muñoz, Durán, Pascual, Montagne, Revel, Rodrigues) 0.78 eV Table 6: enthalpy of activation of the lithium ion conductivity in LiPO ₃ theoretically according to formula (4) and experimentally

Obviously, experimental and theoretical values are again in agreement here. A typical value for the conductivity is 1.2 × 10 ^-9 / (Ω · cm) at 25 ° C, see F. Muñoz, A. Durán, L. Pascual, L. Montagne, B. Revel, A. Rodrigues, Increased electrical conductivity of LiPON glasses produced by ammonolysis, Solid State Ionics 179, 574-579 (2008) ,

As Comparative Example 3 lithium lanthanum zirconate (Li ₇ La ₃ Zr ₂ O ₁₂ , LLZO) is considered, which is suitable as an ion conductor for room temperature. LLZO belongs to the grenades. It exists as a crystal in two structures, a tetragonal, stable without doping at room temperature and a cubic, stable at higher temperatures, but z. B. can be made room temperature stable by alumina doping. The latter is particularly interesting in view of their particularly high ionic conductivity.

A stabilization of the cubic phase at room temperature is z. B. with 1.5 weight percent Al ₂ O ₃ possible, see, for. B. Y. Zhang, F. Chen, R. Tu, Q. Shen, L. Zhang, "Field Assisted Sintering of Dense Al-Substituted Cubic Phase Li7La3 Zr2O12 Solid Electrolytes", Journal of Power Sources (2014) , in print. The conductivity then reaches 5.7 · 10 ^-4 / (Ω · cm); the associated density is 5.09 g / cm ³ and the associated enthalpy of activation 0.3 eV.

We analyze LLZO as above LAS, neglecting the 1.5 weight percent Al ₂ O ₃ . With the method of RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , we determine the relative permittivity to 22,3655. We still need the density that we have Zhang, F. Chen, R. Tu, Q. Shen, L. Zhang, "Field Assisted Sintering of Dense Al-Substituted Cubic Phase Li7La3 Zr2O12 Solid Electrolytes", Journal of Power Sources (2014) , in print, to take about 5.1 g / cm ³ , and the atomic masses we M. Winter, web site "Webelements", retrieved on 24. 10 2013 from http://www.webelements.com/ , remove. From the stoichiometry and the density, the ionic strength follows to 63768 mol / m ³ . Thus, according to formula (4), we obtain a room-temperature activation enthalpy of 0.29 eV. Method or reference h _i LLZO Formula (4); ε = 22,3655 (after Shannon) 0.29 eV Experiment / Polycrystalline LLZO (Zhang, Chen, Tu, Shen, Zhang) 0.3 eV Table 7: Activation enthalpy of the lithium ion conductivity in LLZO according to formula (4) and experimentally determined value.

LLZO is a glass-forming system. Zircon is a structurally active element. Produced in a glass composite, zirconium would crosslink according to its valence via four oxygen bridges, at the end of which could be one of the eight valences of the two zirconium atoms per molecular assembly, or one of seven non-bridging oxygen atoms attributable to lithium ions or one of nine, the total of nine valencies three Lanthanatome attributable oxygen atoms. (The number of valencies is equal to the oxidation number.) Cf. Radovan Karell, Jozef Kraxner, Maria Chromeikova, Properties of selected zirconia containing silicate glasses, Ceramics-Silikáty 50 (2) 78-82 (2006) , Lithium is a network transducer of the first, and lanthanum is a second type of network transducer. On average, each of the four valences of a zirconium atom will result in 8 / (8 + 7 + 9) other zirconium atoms; Multiplication by 4 indicates that each zirconium atom is cross-linked on average with 32/24 additional zirconium atoms.

This consideration corresponds to the above condition that the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al , Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi (here "8" for 2 zirconia atoms per molecular unit and the oxidation number +4 of the zirconium) and the sum of all cation oxidation numbers (here "24 Multiplied by the oxidation number of each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi (here only 1 such cation, namely zirconium, with the Oxidatzionszahl +4) numbers (here 32/24), their weighted with the relative abundance of each cation weighted average (again 32/24, since only one such type of cation) is greater than "one", preferably greater than "two". There are LLZO variants in which zirconium is substituted by tantalum, e.g. Li ₅ La ₃ Ta ₂ O ₁₂ (LLTa). With the method of RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , we determine the relative permittivity of a 100% dense material to 24,0086. We still need the theoretical density that we have Masashi, Kotobukin, Kiyoshi Kanamurau, Fabrication of all-solid-state battery using Li5La3Ta2O12 ceramic electrolyte, Ceramics International 39 (2013) 6481-6487 , to take 6.1 g / cm ³ , and the atomic masses, which we M. Winter, web site "Webelements", retrieved on 24. 10 2013 from http://www.webelements.com/ , remove. From the stoichiometry and the density, the ionic strength follows to 45507 mol / m ³ .

Thus, according to formula (4), we obtain a room-temperature activation enthalpy of 0.27 eV. We compare with a density functional calculation which gives an activation enthalpy of 0.26 eV for the self-diffusion coefficient, see R. Jalem, M. Nakayama, T. Kasuga, K. Kanamura, Abstract # 577, 224th ECS Meeting, 2013 , This corresponds according to the Nernst-Einstein relation the above E _A , see z. B. N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 so that we can compare directly with the result of formula (4): Method or reference h _i LLTa Formula (4); ε = 24,0086 (after Shannon) 0.27 eV Computer experiment / simulated LLTa (Jalem, Nakayama, Kasuga, Kanamura) 0.26 eV Table 8: Activation enthalpy of the lithium ion conductivity in LLTa according to formula (4) and experimentally determined value.

In Comparative Example 3, the condition of a glass-forming system, and even in one of the preferred forms, is satisfied, but not the condition of the invention for ε as a function of ionic strength I. Tantalum is an imperfect glass former. Produced in a glass composite, tantalum would be crosslinked according to its valence via five oxygen bridges, at the end of which could sit one of the ten valences of the two tantalum atoms per molecular unit, or one of five non-bridging oxygen atoms attributable to lithium ions or one of nine, the total of nine valencies three Lanthan atoms to be assigned oxygen atoms. On average, each of the five valences of a tantalum atom will result in 10 / (10 + 5 + 9) other zirconia atoms; Multiplication by 5 indicates that each tantalum atom is cross-linked on average with 50/24 other tantalum atoms.

Inventive embodiments in which the condition for ε depending on the ionic strength I are satisfied, go back to bismuth borate systems. Boron is a glass former, and bismuth is at least a conditional glass former. Bismuth borate glasses can be prepared to very high bismuth levels 86 cation% as glasses with measurable glass transition temperature, see George, Vira, Stehle, Meyer, Evers, Hogan, et al., "Physical Analysis of the Physical Properties of Bismuth and Lead Based Glasses," Physics and Chemistry of Glasses, Part B, Volume 40 of the European Journal of Glass Science and Technology , Number 6, 1999, pp. 326-332 , even in the same stoichiometry as crystalline phases, these phases may have a very high density such. For example, the cubic Bi24B2O39, see I.-I. Oprea, "Optical Properties of Borate Glass-Ceramics", Dissertation, University of Osnabrück, 2005 , Bismuth borates are readily ceramized as described in I.-I. Oprea, "Optical Properties of Borate Glass-Ceramics", Dissertation, University of Osnabrück, 2005.

A first embodiment results from said system Bi ₂₄ B ₂ O ₃₉ by replacing a bismuth charge-neutral by a calcium ion and a lithium ion. Here it is assumed that the calcium, which has (for all coordinations) about the same ion radius as the bismuth, occupies a bismuth site, and the lithium moves independently in the network. Furthermore, it is assumed that the molecular volume for the substituted crystalline system is the same as for the original one, namely 1.02288 × 10 ^-28 m ³ , which can be seen from the density of 9.189 g / cm ³ , see I.-I. Oprea, "Optical Properties of Borate Glass-Ceramics", Dissertation, University of Osnabrück, 2005 , and the molecular weight calculated 5661.142 g.

Further, assuming that ceramization is almost complete, and that after ceramization, the molecular volume of the crystalline system is present, RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 calculate the relative permittivity εe to 36.9. The ionic strength I is 2434.76 / (mol / m ³ ). Because of the low ionic strength I, for ε: ε> 41.583 - 63229.0 / 2434.76 + 0.0000426993 · 2434.76 = 15.7 is obtained. This is the harshest of conditions in Table 3.

The second embodiment is based on Bi ₁₀ B ₂ O ₁₈ , which can be prepared as a glass with a density of 8.92 g / cm ³ and the molecular volume of 4,466 · 10 - 28m ³ , see George, Vira, Stehle, Meyer, Evers, Hogan, et al., "Physical Analysis of the Physical Properties of Bismuth and Lead Based Glasses," Physics and Chemistry of Glasses, Part B, Volume 40 of the European Journal of Glass Science and Technology , Number 6, 1999, pp. 326-332 , This system has after RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , initially (unsubstituted) a relative permittivity of 32.99.

Substituting as previously described a bismuth against a calcium and a lithium ion and assuming that lithium ion does not change the molecular volume, it follows RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 , a lower estimate of the relative permittivity of 27.3. The ionic strength is 5576.35 / (mol / m ³ ). Thus, even for such a system ε> 36.2286 - 52953.6 / 5576.35 + 0.0000353135 · 5576.35 = 26.9. reached. The system thus corresponds to the third sharpest condition from Table 3 and thus to the invention.

A system Bi ₂₄ B ₂ O ₃₉ can also be considered as network-forming after the substitution of a bismuth atom by a calcium and a lithium atom. This is due to the apparent predominance of boron as a "network former" and bismuth as a "conditional network former" over the other cations. Analyzing the basic unit LiCaBi ₂₃ B ₂ O ₃₉ , the quotient results from the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo , W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi and the sum of all cation oxidation numbers to (23 x 3 + 2 x 3) / (23 x 3 + 2) · 3 + 1 + 2) = 75/78. Multiplied by the oxidation number for both bismuth and boron "3", both bismuth and boron have the numerical value 225/78 as the mean number of oxygen bridges to other network-forming ions. The averaging over the two cation types bismuth and boron is therefore trivial. The relative abundance of bismuth among the network-forming cations is 23/25, the relative abundance of boron among the network-forming cations is 2/25; the weighted average is 225/78 * 23/25 + 225/78 * 2/25 = 225/78, and as expected is greater than "1", even greater than "2". With the invention, an electrolyte, in particular a lithium ion-conducting material is specified for the first time, which has a substantial constancy of ionic conductivity over a predetermined temperature range, which preferably ranges from -50 ° C to + 80 ° C.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 2009/0317724 A1 [0002]
• JP 5074466 B2 [0003]
• US 2003143467 A1 [0003]
• US 2007092801 A1 [0003]
• KR 20070082402 A1 [0004]
• US 2009317724 A1 [0005]

Cited non-patent literature

• Jennifer L. Schaefer, Yingying Lu, Surya S. Moganty, Praveen Agarwal, N. Jayaprakash, Lynden A. Archer, Electrolytes for High-Energy Lithium Batteries, Appl. Nanosci. 2 (2012), pp. 91-109 [0002]
• Philippe Knauth: Inorganic solid Li ion conductors: An overview. Solid State Ionics 180 (2009), pages 911-916 [0002]
• Alan B. Lidiard, Ionic conductivity, Handbook of Physics 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 [0011]
• M. Born, volume and heat of hydration of the ions, Journal of Physics, No. 1, pp. 45-48 (1920) [0012]
• Alan B. Lidiard, Ionic conductivity, Handbuch der Physik 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 [0014]
• JC Dyre, P. Maass, B. Roling, DL Sidebottom, "Fundamental questions concerning ion conduction in disordered solids", Rep. Prog. Phys. 72 (2009), 046501 [0016]
• Alan B. Lidiard, Ionic conductivity, Handbuch der Physik 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 [0017]
• RD Shannon, "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides", Acta Cryst, 1976, pp. 751-767 [0018]
• L. Pauling, "The principles of the structure of complex ionic crystals", J. Am. Chem. Soc. 51 (4), 1929, p. 1010-1026 [0018]
• R. Schäfer, Advanced Physics Chemistry I, Website, TU Darmstadt, Access 23.8.2013 [0018]
• Alan B. Lidiard, Ionic Conductivity, Handbook of Physics 20, 246-349, Springer-Verlag, Berlin-Heidelberg-New York, 1957, especially p. 307 [0020]
• M. Ding, K. Xu, S. Zhang, TR Jow, K. Amine, G. Henriksen, "Change of Conductivity with Salt Content, Solvent Composition, and Temperature for Electrolytes of LiPF6 in Ethylene Carbonate-Ethyl Methyl Carbonate". Journal of The Electrochemical Society, 148, 2001, p. A1196-Al204 [0021]
• Charles Kittel, Introduction to Solid State Physics, 5th Edition, R. Oldenbourg-Verlag Munich Vienna, 1980, p. 441 [0023]
• Ekaterina I. Igorodina, Maria Forsyth, and Douglas R. MacFarlane, "On the components of the dielectric constant of ionic liquids: ionic polarization?", Phys. Chem. Chem. Phys. 11, 2452-2458 (2009) [0023]
• N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 [ 0027]
• Jai Shanker, MP Verma, "Correlation between electronic polarities and ionic radii in alkali halides", Journal of Physics and Chemistry of Solids, Volume 37, Issue 9, 1976, Pages 883-885 [0031]
• SC Agrawal, OP Sharma, Jai Shanker, "Effect of the crystal potential on electronic polarities of ions in alkaline earth chalcogenides", Pramana Journal of Physics, January 1978, Volume 10, Issue 1, pp 11-15 [0031]
• L. Pauling, "The Nature of the Chemical Bond," New York, Cornell University Press, 1960. [0031]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0032]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0033]
• RD Shannon, "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides", Acta Cryst, 1976, pp. 751-767 [0034]
• A. Feteira, D. Sinclair, K. Rajab, M. Lanagan, "Crystal Structure and Microwave Dielectric Properties of Alkaline-Earth Hafnates, AHfO3 (A = Ba, Sr, Ca)", J. Am. Ceram. Soc., 91 [3], 2008, pp. 893-901 [0035]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0036]
• Agilent, "Basics of Measuring the Dielectric Properties of Materials" - Application Note, June 26, 2006 [0051]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0052]
• http://cpmd.org/ [0052]
• R. Jalem, M. Nakayama, T. Kasuga, K. Kanamura, Abstract # 577, 224th ECS Meeting, 2013 [0052]
• WH Zachariasen, The Atomic Arrangement in Glass, J. Chem. Soc. 54, No. 10 (1932), pp. 3841-3851 [0053]
• A. Dietzel, "The Cation Field Strengths and Their Relationships with Devitrification, Linkage, and the Melting Point of Silicates," Zeitschrift fur Elektrochemie und angewandte Physikische Chemie 48 (1942), p. 9-23 [0053]
• TS Izumatani, Optical Glass, AIP Trans. Series, New York (1986) [0053]
• RD Shannon, T. Berzins: Ion conductivity in low carnegieite compositions based on NaAlSiO4. Mat. Res. Bull. 14 (1979), pp. 361-367 [0054]
• T. Rouxel, Elastic Properties and Short-to Medium-Range Order in Glasses, J. Am. Ceram. Soc., 90, No. 10 (2007), pp. 3019-3039 [0055]
• RD Shannon, T. Berzins: Ion conductivity in low carnegieite compositions based on NaAlSiO4. Mat. Res. Bull. 14 (1979), pp. 361-367 [0056]
• HJL Trapp, JM Stevels, "Physical properties of inverted glasses", Glastechn. Ber. 32 (1959), p. 32-52 [0058]
• S. Misture, Viscous Glass Sealants for SOFC Applications, DOE Award Number: DE-NT0005177, Alfred University, 2012 [0059]
• TS Izumatani, Optical Glass, AIP Trans. Series, New York (1986) [0062]
• Tarsilla Gerthsen, Chemistry for Mechanical Engineering: Inorganic Chemistry for Materials and Processes, KIT Scientific Publishing, Karlsruhe, 2006, p. 217 [0062]
• Linus Pauling, The Nature of the Chemical Bond, Cornell University Press, New York, 1960. [0063]
• Linus Pauling, The Nature of the Chemical Bond, Cornell University Press, New York, 1960. [0064]
• Shannon "Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides, Acta cryst., Pp. 751-767, 1976 [0064]
• A. Lichtenstein, R. Jones, S. Gironcoli, S. Baroni, "Anisotropic thermal expansion in silicates: A density functional study of β-eucryptites". Physical Review B 62 No 17, 2000, pp. 11487-11493 [0073]
• H. Guth, "Structure Investigations on the One-Dimensional Li-Ion Guide β-Eucryptite (LiAlSiO4) Using Neutron Diffraction", Dissertation, Nuclear Research Center Karlsruhe, September 1979 [0073]
• T. Ogiwara, Y. Noda, K. Shoji, O. Kimura, "Low-Temperature Sintering of High-Strength β-Eucryptite Ceramics with Low Thermal Expansion Using Li 2 O-GeO 2 as a Sintering Additive". J. Am. Ceram. Soc., 94, 2011, pp. 1427-1433 [0074]
• http://www.webmineral.com/data/Eucryptite.shtml [0074]
• N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 [ 0076]
• Watanabe, T. Yamamoto, and T. Yamada, "The Chemical Structure and Electric Property of Lithium Phosphate Glass," Memoirs of College of. Engineering, Chubu University p. 71 (1984) [0077]
• F. Muñoz, A. Durán, L. Pascual, L. Montagne, B. Revel, A. Rodrigues, Increased Electrical Conductivity of LiPON glasses produced by ammonolysis, Solid State Ionics 179, 574-579 (2008) [0077]
• F. Muñoz, A. Durán, L. Pascual, L. Montagne, B. Revel, A. Rodrigues, Increased Electrical Conductivity of LiPON glasses produced by ammonolysis, Solid State Ionics 179, 574-579 (2008) [0078]
• Y. Zhang, F. Chen, R. Tu, Q. Shen, L. Zhang, "Field Assisted Sintering of Dense Al-Substituted Cubic Phase Li7La3Zr2O12 Solid Electrolytes, Journal of Power Sources (2014) [0080]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0081]
• Zhang, F. Chen, R. Tu, Q. Shen, L. Zhang, "Field Assisted Sintering of Dense Al-Substituted Cubic Phase Li7La3 Zr2O12 Solid Electrolytes", Journal of Power Sources (2014) [0081]
• http://www.webelements.com/ [0081]
• Radovan Karell, Jozef Kraxner, Maria Chromeikova, Properties of selected zirconia containing silicate glasses, Ceramics-Silikáty 50 (2) 78-82 (2006) [0082]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0083]
• Masashi, Kotobukin, Kiyoshi Kanamurau, Fabrication of all-solid-state battery using Li5La3Ta2O12 ceramic electrolyte, Ceramics International 39 (2013) 6481-6487 [0083]
• http://www.webelements.com/ [0083]
• R. Jalem, M. Nakayama, T. Kasuga, K. Kanamura, Abstract # 577, 224th ECS Meeting, 2013 [0084]
• N. Dietrich, Production and Characterization of β-Eucryptite and Modified Eucryptite Ceramics LiAl1-yMySiO4 (M = Cr, Mn, Fe) for High Performance Materials, Dissertation, Saarbrücken University, Institute of Inorganic and Analytical Chemistry and Radiochemistry, 2007 [ 0084]
• George, Vira, Stehle, Meyer, Evers, Hogan, et al., "Physical Analysis of the Physical Properties of Bismuth and Lead Based Glasses," Physics and Chemistry of Glasses, Part B, Volume 40 of the European Journal of Glass Science and Technology , Number 6, 1999, pp. 326-332 [0086]
• I.-I. Oprea, "Optical Properties of Borate Glass-Ceramics", Dissertation, University of Osnabrück, 2005 [0086]
• I.-I. Oprea, "Optical Properties of Borate Glass-Ceramics", Dissertation, University of Osnabrück, 2005 [0087]
• George, Vira, Stehle, Meyer, Evers, Hogan, et al., "Physical Analysis of the Physical Properties of Bismuth and Lead Based Glasses," Physics and Chemistry of Glasses, Part B, Volume 40 of the European Journal of Glass Science and Technology , Number 6, 1999, pp. 326-332 [0089]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0089]
• RD Shannon, "Dielectric polarizabilities of ions in oxides and fluorides", J. Appl. Phys., 1993, pp. 348-366 [0090]

Claims (5)

Electrolyte, in particular solid electrolyte, preferably lithium ion-conducting material, wherein the solid electrolyte, in particular the lithium ion-conducting material is a crosslinked system, in which for a given stoichiometry of the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb , Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi and the sum of all cation oxidation numbers, for each the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe, Tl, Pb, Bi, multiplied by its oxidation number, gives numbers whose mean weighted by the relative abundance of the respective cation greater than "one", preferably greater than "two", characterized in that that the relative permittivity ε of the electrolyte, in particular Solid electrolyte, preferably of the lithium ion-conducting material, above the value 23.9817-31424 _, 5 / I * + 0.0000194269 I *, preferably above 25.0319 - 33014.0 / I * + 0.0000207389 * I *, more preferably above 26.1809 - 34971.4 / I * + 0.0000221847 * I * _, even more preferably above 27, 4431 - 37144.2 / I * + 0.0000237855 * I *, even more preferably above 28.8365 - 39570.2 / I * + 0.0000255681 * I *, even more preferably above 30.3821 - 42296.2 / I * + 0.000027565 * I *, even more preferably above 32.1061 - 45381 / I * + 0.0000298173 * I *, even more preferably above 34.0414 - 48900.8 / I * + 0.0000323777 * I *, even more preferably above 36.2286 - 52953.6 / I * + 0.0000353135 * I *, even more preferably above 38.7201 - 57670.8 / I * + 0.0000387142 * I *, even more preferably above 41.583 - 63229.0 / I * + 0.0000426993 · I *, where I * is the normalized ionic strength of the electrolyte. Electrolyte, in particular lithium ion-conducting material according to claim 1, characterized in that for a given stoichiometry the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe and the sum of all cation oxidation numbers multiplied by the oxidation number of each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W, Be, Al, Ga, In, Zr, Hf, Cr, Mn, Fe give numbers whose average weighted by the relative abundance of each cation is greater than "One", preferably greater than "two", Electrolyte, in particular lithium ion-conducting material according to claim 1, characterized in that for a given stoichiometry the quotient of the sum of the oxidation numbers of all cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W and the sum of all cation oxidation numbers multiplied by the oxidation number of each of the cations B, Si, Ge, P, Sn, As, Sb, Se, Te, Ti, V, Nb, Ta, Mo, W numbers whose average weighted by the relative abundance of the respective cation is greater than "one", preferably greater than "two", Electrolyte, in particular lithium ion-conducting material according to claim 1, characterized in that the lithium ion-conducting material is a bismuth borate system, in which one or more bismuth atoms is replaced charge neutral by a respective calcium ion and one lithium ion. Electrolyte, in particular lithium ion-conducting material according to Claim 4, characterized in that the bismuth borate system before substitution is Bi ₂₄ B ₂ O ₃₉ or Bi ₁₀ B ₂ O ₁₈ .