CN110120250B

CN110120250B - High-flux prediction method of coating material compatible and stable with solid electrolyte

Info

Publication number: CN110120250B
Application number: CN201910292851.0A
Authority: CN
Inventors: 吴凡; 李泓
Original assignee: Yangtze River Delta Physics Research Center Co ltd; Institute of Physics of CAS; Tianmu Lake Institute of Advanced Energy Storage Technologies Co Ltd
Current assignee: Yangtze River Delta Physics Research Center Co ltd; Institute of Physics of CAS; Tianmu Lake Institute of Advanced Energy Storage Technologies Co Ltd
Priority date: 2019-04-11
Filing date: 2019-04-11
Publication date: 2023-06-20
Anticipated expiration: 2039-04-11
Also published as: CN110120250A

Abstract

The invention provides a high-flux prediction method of a coating material compatible and stable with solid electrolyte, which adopts a new high-flux analysis mode, is effectively applied to a very large database, successfully searches more than 67,062 materials, and finds a coating material capable of optimally stabilizing the interface between sulfide solid electrolyte and typical electrode (cathode and anode) materials.

Description

High-flux prediction method of coating material compatible and stable with solid electrolyte

Technical Field

The invention belongs to the field of solid-state electrolytes and solid-state batteries, and particularly relates to a high-flux prediction method of a coating material compatible and stable with the solid-state electrolytes.

Background

All-solid-state lithium ion batteries are one of the most promising directions in the next-generation energy storage field ^1–3 . Most notably, ceramic solid electrolytes having non-flammable characteristics can address many of the safety concerns associated with the use of highly flammable organic liquid electrolytes. Furthermore, unlike liquid electrolytes, solid electrolytes do not suffer from the problem of viscosity increase at low temperatures, which limits their operation, a critical operating condition required in the rapidly evolving electric vehicle market ^4–6 . In some cases, the solid electrolyte may even have a higher ionic conductivity than the liquid electrolyte ^6,7 . In addition, solid electrolytes have the potential to construct batteries using higher energy density electrode materials than liquid electrolytes ^8–10 . For example, lithium metal has the lowest lithium chemical potential relative to the negative electrode, and thus represents a theoretical limit for possible negative electrode materials in lithium ion batteries. However, the problem of lithium dendrites limits the practical use of lithium metal and other high capacity materials in liquid lithium ion batteriesApplication. While ceramic solid state electrolytes offer the possibility of physically suppressing lithium dendrites ^10,11 。

As with liquid electrolytes, the key performance indicators of solid electrolytes are their stability and ionic conductivity. For lithium batteries, two very potential classes of solid state electrolytes are garnet-type oxides, respectively ^12–16 And ceramic sulfides ^6,17–20 . Among these two classes of typical high performance electrolytes are Li-La-Zr-O (LLZO) oxide and Li-X-P-S (LXPS, x=si, ge, etc.) sulfide, respectively. The oxide can maintain good stability even in a wide voltage range ¹⁷ But it generally has a low ionic conductivity [ ]<1mS cm ^-1 ). In contrast, sulfides have excellent ionic conductivity up to 25mS cm ^-1 ^6,20 The disadvantage is poor electrochemical stability and decomposition in the operating voltage range of the battery ^17–19 。

The instability of the solid electrolyte may result from inherent material decomposition itself (bulk decomposition) or interfacial reactions upon contact with other materials. In terms of the material itself, solid electrolytes tend to be chemically stable (i.e., minimal spontaneous decomposition), but have electrochemical reaction sensitivity in cells with sufficient lithium ion reserves, which can consume or generate lithium ions, and undergo reduction or oxidation reactions. The voltage stability window defines the range of lithium chemical potentials within which the solid electrolyte does not electrochemically decompose. The lower limit of the voltage window indicates the onset of reduction, or consumption of lithium ions and corresponding electrons, while the upper limit indicates the onset of oxidation, or production of lithium ions and electrons. Since the applied voltage extends through the entire solid electrolyte layer, the voltage stability window will affect all solid electrolyte particles. At the contact point ^8,11,22 An interfacial reaction occurs between the solid electrolyte and the "coating" material. These reactions may be two-body chemical reactions, with only the solid electrolyte and coating material participating in the reaction; or a three-body electrochemical reaction in which both the solid electrolyte, the coating material and lithium ions participate. Wherein the two-body chemical reaction is not related to the degree of charge (or voltage), while the three-body electrochemical reaction is related, depending on whether lithium ions are involved or notAnd reacting.

Previous studies have shown that the most common lithium ion electrode materials, such as LiCoO ₂ (LCO) and LiFePO ₄ (LFPO) with most solid electrolytes, especially high performance ceramic sulfides ^21-23 An unstable interface is formed. Thus, successful use of ceramic sulfides in solid state batteries would require a suitable coating material to reduce the instability of these interfaces. These coating materials need to have both electrochemical stability per se and interfacial electrochemical stability with ceramic sulfides over the full voltage operating range. In addition, if different solid state electrolytes are used in different battery components (e.g., positive, negative, electrolyte) to achieve maximum material stability per se, the choice of coating material must also be changed for different components to maintain a chemically stable interface.

Disclosure of Invention

The invention aims to provide a high-flux prediction method of a coating material compatible and stable with a solid electrolyte, which can minimize the calculation cost, efficiently search a large data volume database and obtain a potential coating material (namely the coating material) meeting the requirements.

The high-throughput prediction method of the coating material compatible and stable with the solid electrolyte comprises the following steps:

s1, determining a set containing the minimum number of element sets of all materials to be selected in an analysis object, wherein the element sets are union sets of element combinations formed by elements of the materials to be selected in a group of materials to be selected and elements of a solid electrolyte material respectively, at least one element combination in the element combinations is the same as the union set, and other element combinations are the same as the union set or are subsets of the union set;

s2: constructing a convex hull of a fraction (x) consumed by one of the material to be selected and the solid electrolyte material to be decomposed at the interface from gibbs free energy of the reaction product, and constructing a total decomposition energy function (G) of the reaction from an energy difference of the reaction product and the reaction raw material _hull (x) Wherein each material to be selected covered in each element set shares a convex hull constructed from the elemental composition of that element set;

s3, determining the value (x) of the fraction consumed by the decomposition of one of the materials to be selected and the solid electrolyte material at the interface in the most dynamically driven reaction by adopting a pseudo-dichotomy calculation mode _m )；

S4: total decomposition energy function (G) _hull (x) From the intrinsic chemical instability decomposition energy (G) of the reaction raw materials ⁰ _hull (x) (G ') and interfacial instability decomposition energy (G' _hull (x) (ii) the composition according to requirement (i) the average inherent electrochemical dissociation energy per atom of the respective material to be selected at a given voltage is less than the thermal disturbance energy (|G) _hull (x＝1)|≤k _B T) and (ii) the interfacial unstable decomposition energy at a given voltage is lower than the thermal disturbance energy (|G ')' _hull (x _m )|≤k _B T), screening out materials meeting the functional stability requirement under a given voltage.

Wherein the set of the minimum number of element sets spanning all the candidate materials in the analysis object is obtained by:

s1.1: combining each material to be selected in the analysis object with the element composition of the solid electrolyte material to obtain a series of element combinations;

s1.2: and (3) sorting a series of element combinations based on the length of the element composition, iterating the element combinations according to the length from small to large, removing the element combinations which are the same as or are subsets of the element combinations with relatively large length, wherein each element combination finally obtained is an element set, and the number of the obtained element sets is minimum.

Wherein the solid electrolyte material is a sulfide solid electrolyte, preferably M-X-P-S-Y, x=si, ge, sn, M is an alkali metal, including Li, na, K, rb, cs, and Y is a doping element, including O, N, se, F, cl, br, I.

Wherein the given voltage range is an anode (negative working voltage) range of 0-5V, preferably 0-1.5V, and a cathode (positive working voltage) range of 2-4V.

Wherein, the program calculation of the convex hull adopts a Python material genome database to construct the composition and energy data of the total decomposition energy function from the material project (Materials Project).

Wherein, the energy change of volume and entropy is ignored in the convex hull, and alkali metal is not considered as independent change dimension when the electrochemical stability calculates the convex hull, because the element molar quantity of the alkali metal can be not conserved and freely moved in the electrochemical system.

Wherein the fraction of the interfacial decomposition consumption of the respective candidate material and the solid electrolyte material is a value (x _m ) The determination of (1) comprises the steps of:

s3.1: mapping atomic proportions to vector elements using vector notation to represent a given material composition in an interfacial reaction, obtains the total decomposition energy (G _hull (x) A derivative function of the fraction (x) of the degradation consumption at the interface with respect to one of the material to be selected and the solid electrolyte material;

s3.2: in the case where the most dynamic driven value (x _m ) Range x of (2) _range And determining an initial guess x ₀ Finding an initial guess x from the derived derivative function ₀ And adjust x by slope _range Repeating the process until x _range The upper and lower limits of the range of (2) differ by less than a prescribed threshold value, resulting in a most kinetically driven value (x _m ). The prescribed threshold is generally not less than 0.01%.

The novel high-throughput analysis mode adopted by the invention can effectively apply the method to a very large database, so that more than 67,062 materials can be searched, and a coating material capable of optimally stabilizing the interface between sulfide solid electrolyte and typical anode and cathode materials is found. And adopts an example Li ₁₀ SiP ₂ S ₁₂ More than 1000 negative side coating materials and more than 2,000 positive side coating materials, which have the required chemical and electrochemical stability, i.e. functional stability, are successfully predicted, with good potential application value. The importance of different anionic species in stability in contact with LSPS at different cell voltages is also disclosed. According to the high-flux screening result and the statistical rule of the anionic substances, the principle of selecting the tendencies of the coating materials is provided, and the battery is provided at the later timeSuch selection criteria may be employed in the design to preferentially match and select materials from a bill of materials for high throughput screening results.

Drawings

The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention. In the drawings:

fig. 1: (A) The number of convex hulls required to evaluate 67k material stability, light color for calculation iterations in material units and dark color for calculation iterations in element combinations. (B) Schematic of pseudo-binary algorithm for interface stability between LSPS and arbitrary material a. G ⁰ _hull Indicating the energy of decomposition of the material itself in the absence of an interface, G' _hull Representing an increased instability breakdown energy due to the presence of the interface (i.e., LSPS in contact with material a). The most kinetically driven reaction occurs at x=x _m 。D _A And D _LSPS The decomposition products of the cladding material a and LSPS are shown, respectively, without an interface (i.e., material a is not in contact with LSPS, e.g., x=0, 1). (C) The ratio of each element in the periodic table and the newly added chemical interface instability decomposition energy (G' _hull (x _m ) Is used) and the correlation is determined. The elements in the straight line box represent an increase in concentration that enables G' _hull Decreasing and improving interfacial stability, the elements in the dashed box represent increasing concentrations of the elements to enable G' _hull The darker color of the increased and decreased interface stability represents a greater degree of influence; only elements present in less than 50 crystal structures are not boxed due to the lack of high volume data.

Fig. 2: (A-C) at 0,2 and 4V voltages, the elemental species fraction and interfacial electrochemical instability decomposition energy (G' _hull (x _m ) A) relationship. The element whose value is closer to-1 (straight line box) represents that its concentration increases to enable G' _hull Elements with values closer to +1 (dashed box) represent increasing concentrations that enable G 'to decrease and improve interfacial stability' _hull Raise and lower interfacial stability; due to the lack of high volume data, only elements present in less than 50 crystal structuresAre not boxed.

Fig. 3: (A) Convex hull energies of the LSPS at different voltages (relative to lithium metal) (convex hull energies refer to the energy difference between the original LSPS and the most stable/lowest energy decomposition products, i.e. the energy released by the most likely LSPS decomposition reaction, as seen thermodynamically). The solid line box/dashed line box portion shows the voltage range of the oxidation/reduction decomposition reaction. The shaded area indicates that the LSPS does not consume or generate lithium (e.g., lithium neutral) in this voltage range. The oxidation/reduction decomposition reaction region is characterized in that the convex hull can be increased/decreased with an increase in voltage. (B) And (C) are convex hull energies of different anionic species compounds (e.g., oxygen-containing compounds and sulfur-containing compounds, etc.) at respective boundary voltages within the negative and positive electrode voltage ranges, respectively. The data points above/below the neutral decomposition line represent that the particular material is purely oxidation/reduction in the negative/positive voltage range. Those compounds located directly above the neutral decay line do not consume or generate lithium ions upon decomposition. (D) The material itself electrochemically decomposes the relationship between the average convex hull energy and the voltage.

Fig. 4: comparison of LSPS interfacial stability for compounds containing different classes of anions. (A) Maximum kinetic driving energy (i.e. total energy released by chemical reaction) of LSPS and each compound family (containing anions of different species) _hull (x _m ) Contribution of interfacial instability (G' _hull (x _m )). (B) The total electrochemical labile reaction energy (G) of each anionic compound family at a given voltage _hull (x _m )). (C) Interfacial instability at constant voltage (G' _kull (x _m ) Average contribution to the total electrochemical instability of each anionic compound group.

Fig. 5: the results of the computational screening of each compound family containing different anions for functional stability (i.e., the material itself is stable and electrochemically stable at the interface with the LSPS solid electrolyte). (A) And (B) are the total number (lines) and percentages (bars) of functionally stable materials contained by the various groups of anionic compounds within the negative and positive voltage ranges, respectively, the bottom bar (light) representing the percentage of functionally stable material within the respective group and the top bar (dark) representing the percentage of potentially functionally stable material (depending on whether its lithiation/delithiation reaction is reversible).

Fig. 6: LCO, snO ₂ LTO and SiO ₂ XRD contrast pattern of structural decay at the interface after contact with the solid electrolyte material (no voltage applied). In ((a)), the guide,

●,■,/>

respectively represent LCO, LSPS, siO ₂ ，Li ₃ PO ₄ Cubic Co ₄ S ₃ Monoclinic Co ₄ S ₃ . In ((b)), ++>

●,■,/>

Respectively represent SnO ₂ ，LSPS，SiO ₂ ，P ₂ S ₅ And Li (lithium) ₂ S, S. In ((c)), the index of the code,

respectively represent LTO, LSPS and Li _1.95 Ti _2.05 S ₄ . In ((d)),) is marked>

Respectively represent SiO ₂ And LSPS. The shaded areas in ((a) - (d)) represent areas where the crystalline phase changes significantly after the mixture of materials and solid electrolyte is heated to 500 degrees celsius. From the graph, the interfacial chemical compatibility increases gradually from ((a) to ((d)), corresponding to theoretical predictions of solid state electrolyte and LCO, snO ₂ LTO and SiO ₂ The interfacial reaction energy of (2) was gradually decreased, 200,97,75 and 0 meV/atom, respectively. ((e), (f)) Li ₂ S and SnO ₂ The dashed box, the straight line box and the non-border shaded area indicate whether the material is oxidized, reduced or not decomposed in these voltage ranges.

Fig. 7: liCoO ₂ ，LSPS，Li ₄ Ti ₅ O ₁₂ ，SnO ₂ And SiO ₂ XRD patterns at different temperatures (room temperature and 500 ℃) were compared. The results show that no significant change in each material was observed between room temperature and 500 ℃.

Fig. 8: powder mixtures (i.e. LiCoO) at different temperatures (room temperature, 300 ℃,400 ℃ and 500 ℃) ₂ +LSPS，SnO ₂ +LSPS，Li ₄ T _i5 O ₁₂ +LSPS and SiO ₂ +lsps). For LiCoO ₂ +LSPS，SnO ₂ +LSPS and Li ₄ Ti ₅ O ₁₂ +LSPS, initial reaction temperatures of 500 ℃,400 ℃ and 500 ℃ respectively were observed. For SiO ₂ +LSPS, no reaction was observed even at temperatures up to 500 ℃.

Detailed Description

For a better understanding of the present invention, the present invention will be described in detail below with reference to specific examples and drawings.

1. Predictive screening method

The invention introduces a new calculation mode, and can more effectively perform interface analysis, thereby effectively searching for a proper coating material which is compatible with solid electrolyte and electrochemically stable in the working voltage range. The data used in the calculation is the result of a calculation based on the density functional theory as a material item (Materials Project) ^24,27 And uses a material Application Programming Interface (API) ²⁸ And (5) butting. Using Python Material genomics (pymatgen) ²⁹ Libraries, combined with literature ²² ,30 ^,31 The method mentioned in (a) computes a convex hull.

As an application example, the invention searches among 69,640 material items in MP, li ₁₀ SiP ₂ S ₁₂ (LSPS) find suitable positive and negative electrode coating (cladding) materials. In the present invention, a coating material that is stable on the material level itself and forms a stable interface with LSPS in a prescribed voltage range is referred to as a "functionally stable (or compatible stable)" material (coating/cladding)。

To establish the standard, we have mainly sought a standard of 0-1.5V and 2-4V (vs Li ⁺ Li) negative and positive electrode coating materials that are functionally stable over the voltage window. These voltage ranges are selected based on the cycling ranges common in today's lithium ion batteries. Within the negative operating voltage range, we are particularly interested in finding a voltage that stabilizes at 0V (vs Li ⁺ Li) because it allows metallic lithium as the negative electrode material.

To effectively evaluate the stability of the interface between these material entries and the LSPS in MP, our method involved two new calculation modes.

The first need minimizes the amount of computation. Traditional pseudo binary (pseudo dichotomy) calculation method ^21,22 The stability of a given interface can be solved approximately, but its computational cost is expensive and difficult to develop on a large scale, and the main bottleneck in high-throughput interface stability analysis is the cost of constructing and evaluating many high-dimensional convex hulls. In the case of phase stabilization of the material, the dimension of the convex hull (hull) is determined by the number of elements. For example, calculating the interfacial chemical stability of LSPS and LCO would require a combination of elemental compositions of both to form a 6-dimensional elemental composition of { Li, si, P, S, co, O }, and when calculating the electrochemical stability, the system would be open to lithium (i.e., the lithium element may not be conserved before and after the reaction), so the limiting factor of lithium could be removed, and lithium removed from the elemental composition, such that the desired elemental composition would be reduced to 5 dimensions ({ Si, P, S, co, O }).

For simplifying the calculation model, we do not consider the combination of elements with dimensions greater than 8 for the interface coating material of LSPS, in other words, LSPS contains four elements { Li, si, P, S }, and we consider only those materials with a maximum of 4 additional elements to be selected; of the 69,640 material entries in total in the MP database, 67,062 materials meet the above requirement of no more than 8 dimensions, which is the object of our determination.

The element combinations of some materials may be a subset of the element combinations of other materials, for example, a convex hull of 6-dimensional element combinations { Li, si, P, S, fe, O } that requires solution for calculating the interfacial stability between LSPS and ferric sulfate is the same as a convex hull of 6-dimensional element combinations { Li, si, P, S, fe, O } that requires solution for calculating the interfacial stability between LSPS and lithium iron phosphate, and the convex hull further includes 5-dimensional convex hulls { Li, si, P, S, fe } that require solution between LSPS and ferrous sulfide as a subset; after the calculation data of the 6-dimensional convex hull { Li, si, P, S, fe, O } is completed, the calculation data of the 6-dimensional convex hull { Li, si, P, S, fe } is directly read without repeated calculation. Thus, this can be exploited by first determining a set of the minimum number of element sets that contains all the candidate materials in the analysis object. For each element set, a group of materials to be selected is covered, and each element combination formed by each material to be selected in the group and the element composition of the solid electrolyte material is the same as the element set or is a subset of the element set. Thus, only one convex hull constructed by taking each element in the element set as a coordinate axis is needed to be calculated, and the convex hull can be used for all other materials to be selected in the group. This mode avoids computing convex hulls one by one for 67,062 materials, but can reduce the total number of convex hulls from 67,062 (one for each material) to 11,935 (one for each set of elements). As shown in a in fig. 1, the number of convex hulls below 7 is small, and those low-dimensional convex hulls are incorporated into higher-dimensional convex hulls and addressed at the same time. And, the element combination with the same element composition is also solved by the same convex hulls, so that the number of the convex hulls needing to be calculated is greatly reduced.

To determine the set of the minimum number of element sets spanning 67,062 of the materials to be selected, each of the 67,062 materials to be selected is combined with an element of the LSPS, respectively, producing a series of element combinations, ordering the element combinations by decreasing length (e.g., ordering in the decreasing dimension of the desired convex hull), then iterating the element combinations, the iterative process removing any element combinations equal to or a subset of the previous element combinations, the final result being a minimum number of element combinations, each element combination being eventually an element set, each element set corresponding to all the materials to be selected to which it is directed or a subset thereof.

The second mode for minimizing computational cost is pseudo binary ^21，22 Algorithm, after obtaining the convex hull data, pseudo binary calculation can be performed, as shown in B in FIG. 1. Since the decomposition at the interface between the two materials can consume an arbitrary amount of each material, the fraction (x) of the decomposition consumption of one of the material (a) to be selected and the solid electrolyte material (LSPS) at the interface can vary between 0 and 1.

(1-x)LSPS+xA→∑d _i D _i (1)

By pseudo-binary calculation, the value (x) of the fraction of the interfacial decomposition consumption of each of the candidate materials and the solid electrolyte material in the most kinetically driven reaction can be determined _m ) Which represents the situation where the decomposition reaction releases the most energy and the decomposition is most severe. The right part (D) of formula 1 represents the fraction of each thermodynamically favored decay product (D), with subscript i being the distinction of the different products. Based on the Gibbs energy of the reaction product (Hull (x) = Σd _i (x)G _i ) Constructing a convex hull of given x, the total decomposition energy function of equation 1 (G _hull (x) Is) is:

G _hull (x)＝∑d _i (x)G _i -(1-x)G _LGPS -xG _A (2)

the most kinetically driven reaction between the LSPS and the material to be selected is the reaction that maximizes the magnitude of equation 2 (i.e., the most negative), which defines the parameters:

max|G _hull (x)|≡|G _hull (x _m )| (3)

this maximum decomposition energy is a result of two factors, as shown by B in fig. 1. The first factor is G ⁰ _hull (x) Is the inherent chemical instability decomposition energy caused by the inherent instability of two reaction raw materials, namely LSPS (D) _LSPS ) And a coating material (D _A ) Corresponding to the reaction (1-x) LSPS+xA.fwdarw.1-x) D _LSPs +D _A Is a decomposition energy of the (c). The second factor is interfacial instability (G' _hull (x) By subtracting the instability of the material itself from the total convex hull energy, as in equation 4:

physically, G ⁰ _hull (x) Indicating the instability of the materials themselves when they are present independently of one another, G' _hull (x) Indicating an increase in instability caused by interfacial interactions upon material contact.

By requiring (i) the electrochemical dissociation energy inherent to each atom of each candidate material at a given voltage to be less than the thermal disturbance energy (|G) _hull (x＝1)|≤k _B T) and (ii) the interfacial instability decomposition energy at a given voltage is less than the thermal disturbance energy (|G ')' _hull (x _m )|≤k _B T), the functional stability of each of the 67,062 materials at a given voltage was determined. Under these conditions, the only instability in the system is that of the LSPS material itself, which can be stabilized by increasing the strain energy in the system ²⁵ . Of the 67k materials, 1,053 materials that were functionally stable over the negative operating voltage range were found (0-1.5V vs Li ⁺ /Li), and 2,669 materials functionally stable in the positive electrode range (2-4V vs Li ⁺ /Li). In addition, 152 materials in the negative electrode operating voltage range and 142 materials in the positive electrode operating voltage range were found to violate condition (i), but their decomposition was achieved only by lithiation/delithiation, so that the practical application of these materials as LSPS coating materials was dependent on the reversibility of the lithiation/delithiation process, as well as having potential functional stability. All of these materials are listed in list 1 and indexed by the corresponding material item (MP) ID number.

In the above procedure, pseudo-binary calculations attempt to find the optimal ratio of LSPS to coating material for chemical reaction, so that the energy released by the decomposition reaction is maximal and thus kinetic driven most, where x=x _m . This problem is simplified by using vector symbols to map the atomic duty ratio into a vector to represent a given material composition in the interfacial decomposition reaction. For example, liCoO2→ (112) based on (Li Co O) means that 1 lithium, 1 cobalt and 2 oxygen are present in the unit formula. Using this representation, the decomposition in equation 1 can be written in vector form:

using

Representing vectors>

The expression matrix, equation 5 becomes:

the relative composition derivative of each decay product may be determined by reversing the formula 6

Find.

Equation 7 allows calculation of the derivative of the convex hull energy with respect to the fractional parameter x:

by using equation 7, and the fact that the convex hull is a convex function, a pseudo-binary search can be performed to find G _hull Maximum value and corresponding X _m Values. The process includes first defining a dual element vector defining that x is known to exist _m Range x of (2) _range = (0, 1) and initial guess value x ₀ =0.5. Evaluating the convex hull at initial guess yields decomposition products { Di } and corresponding energies { G } _Di

Equations

7 and 8 can then be used to find the slope of the convex hull energy. If the convex hull energy slope is positive, x _range →(x ₀ 1) if it is negative, x _range →(0,x ₀ ). The process is repeated until x _range The upper and lower limits differ by less than 0.01%Will always be done in 14 steps (2 ^-14 ≈0.006％)。

In the calculation process, the change in volume and entropy is ignored (ΔG≡ΔE) ^18,22,30,31 . Formulas 5-8 are defined for the purpose of calculating chemical stability. In calculating electrochemical (lithium open system) stability, the free energy needs to be subtracted by one term μ _Li N _Li To establish (DeltaPhi. Apprxeq. DeltaE-mu) _Li △N _Li ) Wherein mu _Li Is the chemical potential of Li, and N _Li Is the amount of lithium ions in the structure; in addition, the lithium composition is not included in the composition vector of formula 6 to allow the number of lithium atoms to be changed before and after the reaction.

2. Experiment and detection

Chemical compatibility between various coating materials and LSPS was tested experimentally by manually milling a powder mixture of LSPS and coating materials (with/without high temperature heating) followed by X-ray diffraction (XRD) measurements at room temperature. Any chemical reaction between the powders causes a change in the composition and structure of the original phase, which can be detected by a change in peak position and intensity in the XRD pattern. Notably, even if interfacial reactions are predicted to occur based on thermodynamic calculations, some amount of energy may be required to overcome the kinetic barriers faced by these reactions to occur ⁸ . Thus, the mixed powder was heated at high temperature (300 ℃,400 ℃,500 ℃) to determine the onset temperature of the interfacial reaction and the reaction products, and the effect of kinetics was further evaluated by comparing these results with the thermodynamic reaction products calculated by DFT.

Latent coating materials (LCO, snO) ₂ ，SiO ₂ LTO) and solid electrolyte were studied by XRD at room temperature. XRD samples were prepared in an Ar-filled glove box and LSPS powder and latent coating material (weight ratio=55:30) were separately hand milled. To test the onset temperature of the reaction of the latent coating material and LSPS solid electrolyte, the powder mixture was spread sufficiently on a hot plate to heat to different nominal temperatures (300, 400 and 500 degrees celsius) and then XRD characterization was performed. XRD testing employed a Rigaku Miniflex 600 diffractometer equipped with a 2-theta range of 10-80 degInternal cukα radiation. All XRD sample racks were sealed with Kapton film in an Ar filled glove box to avoid exposure to air during testing.

The latent coating material (Li ₂ S and SiO ₂ ) Carbon black and poly (tetrafluoroethylene) (PTFE) at 90:5:5, and manually milled in an Ar filled glove box. The powder mixture was then manually rolled into a film in sequence from which a disk (5/16 inch diameter, load about 1-2 mg) was punched to form a working electrode for Cyclic Voltammetry (CV) testing. These electrodes were assembled with Li metal (negative electrode) as counter electrode, two sheets of glass fiber separator and commercial electrolyte (1M LiPF ₆ At 1:1 (volume ratio) ethylene carbonate/dimethyl carbonate (EC/DMC) solvent). CV test was performed by Solartron 1455A, and a voltage sweep rate of 0.1mV/s at room temperature of 0-5V was used to investigate potential coating materials (Li ₂ S and SiO ₂ ) Is provided.

3. Results and analysis

C in fig. 1 and fig. 2 describe the correlation between the atomic fraction of each element and the interface stability. C in FIG. 1 depicts each element in the chemical reaction with G' _hull (x _m ) While A, B, C in FIG. 2 describe G' _hull (x _m ) Correlation to electrochemical reactions at 0,2 and 4V, respectively, relative to lithium metal. The negative correlation between the elemental compositions means that increasing the content of the element improves the interface stability. C in FIG. 1 shows that chemical stability is optimal for compounds containing a large amount of anions such as sulfur, selenium and iodine. A C in FIG. 2 shows the elemental species and G 'at low and high voltages' _hull (x _m ) The correlation between the two is reduced, respectively, which indicates that at these extreme voltages, interfacial decomposition is mainly subject to reduction/oxidation by the intrinsic material itself

Rather than interfacial effects (G' _hull ) Is a dominant factor. At a voltage of 2V relative to metallic lithium (B in FIG. 2), in addition to chalcogen and halide groups exhibiting negative correlation, are largeMost elements show positive correlation (higher instability).

Data set analysis in terms of anion composition was performed in view of the high correlation contrast of anionic species with respect to interfacial stability. To eliminate overlap between data points, only compounds with one of the monoanions { N, P, O, S, se, F, I } or oxygen-containing oxyanions plus { N, S, P } are considered. The 45,580MP entries meeting these criteria are listed in table 1, as well as the percentages of each anion species that find electrochemically stable at the material level.

TABLE 1

The total amount of monoanionic and oxyanionic compounds, and the percentage of electrochemically stable compounds in each of the negative electrode operating voltage range (0-1.5V) and the positive electrode operating voltage range (2-4V) in total. For example, F represents all compounds containing F in the formula, a total of 2902, wherein the ratio of the electrochemically stable F-containing compound over the negative electrode operating voltage range is 0.6%, and O+N represents all compounds containing O and N in the formula

A in fig. 3 shows the effect of an applied voltage on the convex hull energy of a material, the solid electrolyte being LSPS. When the slope of the convex hull energy with respect to voltage is negative, its decomposition corresponds to a reduction reaction, and if the slope is positive, it is an oxidation reaction. There is a region of zero slope in the middle, which means that there is no interaction with the lithium ion reservoir (i.e., the reaction is neutral with respect to lithium, no reaction is produced nor consumed). In view of this, B and C in fig. 3 plot the characteristic redox behavior of each different anionic salt compound over the negative and positive operating voltage ranges, respectively. The 45 ° line of "neutral decay" represents those compounds that have the same convex hull energy at both extreme voltages and therefore do not react with lithium ions, and the data points above/below this line increase/decrease in convex hull energy with respect to voltage and are therefore oxidized/reduced over the plotted voltage range.

B in fig. 3 shows that, consistent with expectations, most of the compounds were reduced in the negative electrode operating voltage range of 0-1.5V relative to lithium metal. A large amount of nitrogen-containing compound was observed to occupy the y-axis, indicating a higher level of stability when it was in direct contact with lithium metal. This is consistent with previous calculations, indicating that binary and ternary nitrides are more stable to lithium metal than sulfides or oxides ²⁶ . However, in the positive operating voltage range (C in fig. 3), more variability in the anion species can be seen: the oxygen-containing anions and the fluorine-containing compounds are mainly reducing, while the phosphorus-containing, sulfide and selenium-containing compounds are oxidizing; the presence of oxygen-containing compounds on both sides of the neutral decay line means that the oxides may lithiate/delithiate in the range of 2-4V.

The average convex hull energy for each anionic species is given in 0.5V steps in the interval 0-5V in D in fig. 3. The nitrogen-containing compounds proved to be the most stable family of compounds at 0V, whereas the iodine and phosphorus containing compounds had relative stability. The average stability of the phosphorus and iodine containing compounds exceeds nitrogen at voltages above 0.5V and 1.0V, respectively. At high voltages (> 4V), it can be seen that fluorine-and iodine-containing compounds are stable, while nitrogen-containing compounds are the least stable.

For each anionic salt, the total dissociation energy (G _hull (xm)) and interfacial instability decomposition energy (G' _hull (x _m ) The resulting scores are shown in fig. 4. A in fig. 4 shows the average instability caused by the chemical reaction between the anionic salt and the LSPS, the sulfur and selenium containing compounds forming on average the most chemically inert interface with the LSPS, in contrast to the fluorine and oxygen containing compounds being the most reactive. As a general trend, those are generally more unstable (higher G _hull (x _m ) With respect to the contribution of the class of compounds to the material itself

Also maintains a higher interfacial contribution (G' _hull (x _m )). This means that the differences in the inherent chemical stability of each class are compared to the interfacial reactivity of LSPS when determining the chemical stability of the interfaceThe effect of (3) is smaller.

B in fig. 4 shows the average total electrochemical breakdown energy of the interface in 0.5V steps from the range of 0-5V. In general, each anionic species follows a path that appears to be governed by the electrochemical stability of the LSPS material itself (A in FIG. 3), at low voltage (< 1V) and high voltage [ (]>This is especially true in the range of 4V), where the electrochemical effect is most pronounced. The maximum deviation of interfacial stability from the stability of the LSPS itself occurs in the region of 1-3V. The compounds with the lowest chemical decomposition energy (S, se, I, P containing compounds) deviate the least from LSPS in this "intermediate" voltage range, whereas the compounds with large decomposition energy (N, F, O, o+ containing compounds) deviate more, which trend suggests that the low and high voltage ranges are dominated by electrochemical reduction and oxidation of the material itself, respectively, while the intermediate range is dominated by interfacial chemical reactions. For example, at 0V, al is expected ₂ O ₃ And LSPS will decay to { Li } ₉ Al ₄ ，Li ₂ O，Li ₃ P，Li ₂ S，Li ₂₁ Si ₅ And the same decay products as each material decomposed independently at 0V. Thus, the presence of the interface has no energy effect.

The average interfacial contribution of electrochemical decomposition is shown as C in fig. 4. All anionic species tend to G 'at 0V' _hull (x _m ) =0, which means that the material tends to fully reduce to lithium binary at 0V, in which case the interface effect is negligible compared to the instability of the material itself. Significant interfacial instability occurs in the intermediate voltage range and then decreases again at high voltages. Again, this means that interfacial chemical effects dominate in the intermediate voltage range, while material itself reduces/oxidizes to dominate at low/high voltages. At high voltages, the interface contribution to instability approaches the reaction energy of the most oxidized material and LSPS. Thus, for any voltage above 4V, the interface will increase the energy instability equivalent to this chemical reaction. This is consistent with the high voltage asymptotic behavior, while the low voltage behavior always tends to 0eV atom ^-1 . For example, for any voltage above 4V, LFO will decompose into { Li, fePO ₄ While LSPS willIs decomposed into { Li, P ₂ S ₅ ，SiS ₂ S. The introduction of an interface allows these oxidation products to chemically react and form FeS ₂ And SiO ₂ 。

The total number of each anionic salt group-containing material identified as functionally stable or potentially functionally stable is given in fig. 5a (negative voltage operating range) and B (positive voltage operating range), which are both inherently stable in their own right and form stable interfaces with the LSPS over the specified voltage range. For the negative voltage operating range, the compounds containing nitrogen, phosphorus and iodine have the highest percentage of stable compounds (2-4%), while all other categories are below 1%, the positive voltage operating range shows a higher percentage, the sulfur-containing compounds reach 35%, and both iodine and selenium are above 10%.

The XRD patterns of the powder mixtures heated at room temperature and 500 ℃ are compared in (a) - (d) in fig. 6. Several candidate coating materials (i.e., snO ₂ ，Li ₄ Ti ₅ O ₁₂ ，SiO ₂ ) Mixed with LSPS (fig. 6 (b) - (d)), whereas lco+lsps mixed powder was used for comparison (fig. 6 (a)). At room temperature and 500℃the phases (i.e. SnO ₂ ，Li ₄ Ti ₅ O ₁₂ ，SiO ₂ ，LiCoO ₂ And LSPS) was used as a reference (fig. 7). By comparing these XRD patterns, it is evident that at room temperature, no coating material reacts with LSPS, as the XRD pattern shows only peaks of the original phase. However, after heating at 500 ℃ for 6 hours, the different materials showed completely different reactivity with LSPS. Severe reaction of LCO with LSPS was observed because the peak intensity and position of XRD pattern of the mixed powder were completely changed throughout the 2-theta range of 10-80 degrees ((a) in fig. 6). The original LCO and LSPS peaks disappear or decrease while additional peaks belonging to new reaction products (e.g. SiO ₂ ，Li ₃ PO ₄ Cubic Co ₄ S ₃ Monoclinic Co ₄ S ₃ ) LCO was shown to be incompatible with LSPS. In sharp contrast, siO ₂ The peak intensity and position of the XRD pattern of the +lsps mixture were never changed, showing only the original peaks before and after heating at 500 ℃. These direct evidence indicate that, although a large scale is providedExternal energy in an amount of, but when SiO ₂ No interfacial reaction occurs upon contact with LSPS. SnO (SnO) ₂ And LTO also show incompatibility with LSPS because new peaks belonging to the reaction product appear in the XRD pattern of its 500 ℃ heated sample, however, the peaks of the reaction product are much weaker than in the case of lco+lsps. The shaded areas in fig. 6 ((a) - (d)) highlight the 2-theta range of peak positions and intensity variations for the 4 materials as one characterization of the incompatibility of the different materials with LSPS. It can be seen that this incompatibility order is LCO>SnO ₂ >LTO>SiO ₂ This is consistent with our theoretical predictions based on thermodynamic calculations. The onset temperatures of the interfacial reactions of the various materials with LSPS are shown in fig. 8.

The electrochemical stability of a typical coating material is characterized by Cyclic Voltammetry (CV), wherein the decomposition of the coating material being tested can be manifested by current peaks at certain voltages associated with lithium. Li (Li) ₂ S and SnO ₂ Two typical coating materials are used as examples to show a good correspondence between our theoretical predictions and experimental observations. Li (Li) ₂ The CV test of S (FIG. 6 (e)) shows a relevant flat region between 0-1.5V, while the large oxidation peak dominates the 2-4V region, in contrast to SiO ₂ The CV test of (f) in fig. 6) shows a net decrease in the region of 0-1.5V and a neutral region with little decomposition between 2 and 4V. These results again demonstrate that our theoretical predictions based on thermodynamic calculations are correct.

High-throughput pseudo-binary analysis of material item DFT data shows that the interface with LSPS decays in the range of 1.5 to 3.5V by the dominant chemistry and is electrochemically reduced/oxidized at lower/higher voltages. As the voltage approaches 0V, the fraction of the decomposition energy due to the interface effect disappears. The results indicate that all material classes tend to decay to the maximally lithiated Li binary and elemental compounds at low voltages, in which case the presence of the interface has no effect.

In terms of anion content, it is important to know that the operating conditions of the coating material are properly matched. For example, sulfur and selenium containing compounds exhibit very high functional stability in the 2-4V cathode range (25% in all sulfides and selenides). However, in the range of 0-1.5V negative electrode operating voltage, less than 1% of the functionally stable coating materials are formed in these same materials, with iodine, phosphorus and nitrogen having the highest properties. The oxygenate has a large number of functionally stable phases in two voltage regions, but the percentage is low due to the larger number of oxygenate data points. Our results demonstrate the powerful function of our new computing platform for chemical, electrochemical and functional stability analysis of material databases with large data, while our specific examples on LSPS sulfide solid electrolytes predict many valuable coating materials on both cathode and anode sides. Thus, our work will accelerate the design of next generation solid state batteries with excellent interface stability.

The foregoing description of the preferred embodiments of the invention is not intended to be limiting, but rather is intended to cover all modifications, equivalents, alternatives, and improvements that fall within the spirit and scope of the invention.

Reference to the literature

1.Takada,K.Progress and prospective of solid-state lithium batteries.Acta Mater.61,759–770(2013).

2.Zhang,B.et al.Mechanisms and properties of ion-transport in inorganic solid electrolytes.Energy Storage Materials10,139–159(2018).

3.Manthiram,A.,Yu,X.&Wang,S.Lithium battery chemistries enabled by solid-state electrolytes.Nat.Rev.Mater.2,1–16(2017).

4.Lin,H.P.et al.Low-Temperature Behavior of Li-Ion Cells.Electrochem.Solid-State Lett.4,A71(2001).

5.Tarascon,J.M.&Armand,M.Issues and challenges facing rechargeable lithium batteries.Nature414,359–367(2001).

6.Kato,Y.et al.High-power all-solid-state batteries using sulfide superionic conductors.Nat.Energy1,16030(2016).

7.Kamaya,N.et al.A lithium superionic conductor.Nat.Mater.10,682–6(2011).

8.Tian,Y.et al.Compatibility issues between electrodes and electrolytes in solid-state batteries.Energy En-viron.Sci.10,1150–1166(2017).

9.Li,J.,Ma,C.,Chi,M.,Liang,C.&Dudney,N.J.Solid electrolyte:The key for high-voltage lithium batter-ies.Adv.Energy Mater.5,1–6(2015).

10.Hayashi,A.,Ohtomo,T.,Mizuno,F.,Tadanaga,K.&Tatsumisago,M.All-solid-state Li/S batteries with highly conductive glass-ceramic electrolytes.Electrochem.commun.5,701–705(2003).

11.Nagao,M.,Hayashi,A.&Tatsumisago,M.High-capacity Li2S–nanocarbon composite electrode for all-solid-state rechargeable lithium batteries.J.Mater.Chem.22,10015(2012).

12.El-Shinawi,H.,Cussen,E.J.&Corr,S.A.Enhancement of the lithium ion conductivity of Ta-doped Li ₇ La ₃ Zr ₂ O ₁₂ by incorporation of calcium.Dalt.Trans.1,9415–9419(2017).

13.Murugan,R.,Thangadurai,V.&Weppner,W.Fast lithium ion conduction in garnet-type Li7La 3Zr2O12.Angew.Chemie-Int.Ed.46,7778–7781(2007).

14.Deviannapoorani,C.,Dhivya,L.,Ramakumar,S.&Murugan,R.Lithium ion transport properties of high conductive tellurium substituted Li7La3Zr2O12cubic lithium garnets.J.Power Sources240,18–25(2013).

15.Geiger,C.A.et al.Crystal chemistry and stability of‘Li7La 3Zr2O12’garnet:A fast lithium-ion conductor.Inorg.Chem.50,1089–1097(2011).

16.Ren,Y.et al.Oxide Electrolytes for Lithium Batteries.J.Am.Ceram.Soc.98,3603–3623(2015).

17.Han,F.,Zhu,Y.,He,X.,Mo,Y.&Wang,C.Electrochemical Stability of Li10GeP2S12 and Li7La3Zr2O12 Solid Electrolytes.Adv.Energy Mater.6,(2016).

18.Mo,Y.,Ong,S.P.&Ceder,G.First Principles Study of the Li ₁₀ GeP ₂ S ₁₂ Lithium Super Ionic Conductor Material.Chem.Mater.24,15–17(2012).

19.Ong,S.P.et al.Phase stability,electrochemical stability and ionic conductivity of the Li _10+-1 MP ₂ X ₁₂ (M＝Ge,Si,Sn,Al or P,and X＝O,S or Se)family of superionic conductors.Energy Environ.Sci.12,148–156(2012).

20.Kuhn,A.,Duppel,V.&Lotsch,B.V.Tetragonal Li10GeP2S12 and Li7GePS8–exploring the Li ion dy-namics in LGPS Li electrolytes.Energy Environ.Sci.6,3548–3552(2013).

21.Richards,W.D.,Miara,L.J.,Wang,Y.,Kim,J.C.&Ceder,G.Interface Stability in Solid-State Batteries.Chem.Mater.28,266–273(2016).

22.Zhu,Y.,He,X.&Mo,Y.First principles study on electrochemical and chemical stability of solid electro-lyte–electrode interfaces in all-solid-state Li-ion batteries.J.Mater.Chem.A4,3253–3266(2016).

23.Wang,Y.,Richards,W.D.,Bo,S.H.,Miara,L.J.&Ceder,G.Computational Prediction and Evaluation of Solid-State Sodium Superionic Conductors Na7P3X11(X＝O,S,Se).Chem.Mater.29,7475–7482(2017).

24.Jain,A.et al.Commentary:The materials project:A materials genome approach to accelerating materials innovation.APL Mater.1,(2013).

25.Wu,F.,Fitzhugh,W.,Ye,L.,Ning,J.&Li,X.Advanced sulfide solid electrolyte by core-shell structural design.Nat.Commun.9,1–11(2018).

26.Zhu,Y.,He,X.&Mo,Y.Strategies Based on Nitride Materials Chemistry to Stabilize Li Metal Anode.Adv.Sci.4,1–11(2017).

27.Jain,A.et al.A high-throughput infrastructure for density functional theory calculations.Comput.Mater. Sci.50,2295–2310(2011).

28.Ong,S.P.et al.The Materials Application Programming Interface(API):A simple,flexible and efficient API for materials data based on REpresentational State Transfer(REST)principles.Comput.Mater.Sci.97,209–215(2015).

29.Ong,S.P.et al.Python Materials Genomics(pymatgen):A robust,open-source python library for materials analysis.Comput.Mater.Sci.68,314–319(2013).

30.Ong,S.P.,Wang,L.,Kang,B.&Ceder,G.Li-Fe-P-O2 phase diagram from first principles calculations.Chem.Mater.20,1798–1807(2008).

31.Zhu,Y.,He,X.&Mo,Y.Origin of Outstanding Stability in the Lithium Solid Electrolyte Materials:In-sights from Thermodynamic Analyses Based on First-Principles Calculations.ACS Appl.Mater.Interfaces7,23685–23693(2015).

List 1

FUNCTIONALLY STABLE ANODE COATINGS

Functionally stable anodic coating

/>

/>

/>

/>

/>

/>

POTENTIALLY FUNCTIONALLY STABLE ANODE COATINGS

Possible/potential functionally stable anode coatings

/>

FUNCTIONALLY STABLE CATHODE COATINGS

Functionally stable cathode coating

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

/>

POTENTIALLY FUNCTIONALLY STABLE CATHODE COATINGS

Possible/potential functionally stable cathode coatings

/>

Claims

1. A method for high throughput prediction of a coating material compatible with a solid state electrolyte, comprising the steps of:

s2: constructing a convex hull of fraction x consumed by one of the materials to be selected and the solid electrolyte material at the interface according to the Gibbs free energy of the reaction product, and constructing a total decomposition energy function G of the reaction according to the energy difference between the reaction product and the reaction raw material _hull x, wherein each material to be selected covered in each element set shares a convex hull constructed by the element components of the element set;

s3, determining the value x of the fraction consumed by the decomposition of one of the materials to be selected and the solid electrolyte material at the interface in the most dynamically driven reaction by adopting a pseudo-dichotomy calculation mode _m ；

S4: total decomposition energy function G _hull x is the energy of decomposition G by the inherent chemical instability of the reaction starting material ⁰ _hull (x) And interfacial instability decomposition energy G' _hull x is formed, and average inherent electrochemical decomposition energy of each atom of each material to be selected under a given voltage is smaller than thermal disturbance energy |G according to requirements i _hull (x＝1)|≤k _B T and ii have interfacial unstable decomposition energy lower than thermal disturbance energy |G 'at given voltage' _hull (x _m )|≤k _B T, screening out materials meeting the functional stability requirement under a given voltage;

in S3, the value x of the fraction consumed by the separation of one of the candidate materials and the solid electrolyte material at the interface in the most kinetically driven reaction _m The determination of (1) comprises the steps of:

s3.1: mapping atomic proportions into vector elements using vector symbols to represent a given material composition in an interfacial reaction, resulting inTo obtain the total decomposition energy G _hull A derivative function of x with respect to a fraction x of the degradation consumption of one of the candidate material and the solid electrolyte material at the interface;

s3.2: at a value x where the most dynamic driving is known to exist _m Range x of (2) _range And determining an initial guess x ₀ Finding an initial guess x from the derived derivative function ₀ And adjust x by slope _range Repeating the process until x _range The upper and lower limits of the range of (2) differ by less than a prescribed threshold value to obtain the most kinetically driven value x _m 。

2. The high-throughput prediction method according to claim 1, wherein said set of minimum number of element sets comprising all candidate materials in the analysis object is obtained by:

s1.2: and (3) sorting a series of element combinations based on the length of the element composition, iterating the element combinations according to the length from small to large, removing the element combinations which are the same as or are subsets of the element combinations with the length, and finally obtaining each element combination as an element set.

3. The high throughput prediction method of claim 1, wherein said solid electrolyte material is a sulfide solid electrolyte.

4. A high throughput prediction method according to claim 3, wherein said sulfide solid electrolyte is M-X-P-S-Y, X = Si, ge, sn, M is an alkali metal, including Li, na, K, rb, cs, Y is a doping element, including O, N, se, F, cl, br, I, P is a phosphorus element, and S is a sulfur element.

5. The high throughput prediction method of claim 1, wherein said given voltage range is 0-5V.

6. The high-throughput prediction method of claim 5, wherein said given voltage range is a negative operating voltage range of 0-1.5V and a positive operating voltage range of 2-4V.

7. The high-throughput prediction method of claim 1, wherein the program calculation of the convex hull uses a Python materials genome database to construct the composition and energy data of the total decomposition energy function from materials item Materials Project.

8. The high-throughput prediction method of claim 1, wherein energy variations in volume and entropy are ignored in the convex hull, and electrochemical stability calculation of the convex hull does not take alkali metals into consideration as independent variation dimensions.

9. The high throughput prediction method of claim 1, wherein said threshold is not less than 0.01%.

10. A stabilized coated bill of materials compatible with solid state electrolytes, characterized in that it is obtained by the high throughput predictive method according to any one of claims 1 to 9.