A lighter solid electrolyte that still lets lithium ions move easily is one of the goals of all-solid-state battery research. A joint research team from the University of Electro-Communications and Nichia Corporation has now identified a crystal structure that may point toward it. The results were published in the peer-reviewed journal Advanced Functional Materials on September 24, 2026, and the University of Electro-Communications made an official announcement on October 6. The corresponding authors are Ryoei Ichikawa of Nichia and Professor Jun Nakamura of the University of Electro-Communications. University of Electro-Communications
It would be premature, however, to read this as the completion of a lightweight next-generation battery. What the team presented is a fundamental evaluation of a solid electrolyte's properties, combining ceramic synthesis experiments with computational science. They did not build a battery cell that actually charges and discharges. The calculations showed that lithium ions can move easily, but the ion conductivity of the synthesized samples was extremely low. That large gap is where the study's key finding lies.
All-solid-state batteries are expected to be a next-generation energy storage technology, but many obstacles to commercialization remain. One is the weight of the solid electrolyte. Adopting a nonflammable inorganic solid electrolyte to reduce fire risk can increase the weight of the whole battery, depending on the material and structure. Oxide-based solid electrolytes in particular offer excellent stability in air, but typical examples contain many heavy elements. In electric vehicles and drones, a heavier electrolyte affects driving range and payload capacity.
This study offers a clue for addressing the weight problem. The team focused on a crystal that had mainly been studied as a light-emitting material, and used atomic-level calculations to identify pathways along which lithium ions can move easily. Its academic value lies in showing a crystal structure that could combine light weight with high ion conductivity.
The study also showed that having easy ion pathways inside a crystal is not enough on its own to produce sufficient conductivity. It highlights how difficult it is to bring out a material's inherent performance in a real battery.
A new crystal structure that could cut the weight of oxide-based solid electrolytes
Oxide-based solid electrolytes are considered one of the leading candidates in all-solid-state battery research. Unlike sulfide-based solid electrolytes, they carry a small risk of producing toxic hydrogen sulfide through reactions with moisture, and they can be handled relatively stably in air.
However, typical conventional oxide-based solid electrolytes have a problem: the materials themselves are heavy. They contain elements with large atomic masses, such as lanthanum (La) and zirconium (Zr), to stabilize the crystal structure.
For example, the crystal density of the representative garnet-type solid electrolyte (LLZO) reaches about 5.12 g/cm³.
A heavier solid electrolyte may improve battery safety, but it is a disadvantage for raising energy density per unit weight. Keeping electrolyte weight down is an important challenge when extending the range of electric vehicles, or in aviation applications where lighter airframes matter especially.
The team therefore turned to , a silicate compound with the so-called -type crystal structure.
This substance has a monoclinic crystal structure, with a calculated theoretical density of 2.74 g/cm³. Compared with LLZO's 5.12 g/cm³, its crystal density is about 46% lower.
was originally known as a host material for green LED phosphors. In other words, it is not a newly synthesized compound. What distinguishes this study is that it examined a substance previously studied mainly as an optical material from a different angle: lithium-ion conduction.
Combining Nichia's accumulated knowledge of phosphor materials with computational materials science at the University of Electro-Communications revealed the potential of a crystal structure that had not been seriously considered as a solid electrolyte.
The "roughly 46% weight reduction" refers only to a comparison of crystal density. It does not mean that the weight of a finished battery cell would fall by 46%.
Actual battery cells use many other components, including positive and negative electrode active materials, conductive additives, binders, current collectors, and casing materials. Even if the solid electrolyte has a lower density, how much the weight and energy density of the whole battery improve depends on factors such as the thickness of the electrolyte layer and how it is combined with the electrodes.
What was shown is not a direct demonstration of lighter batteries, but the discovery of a promising crystal structure for developing lightweight solid electrolytes.
Synthesis experiments and two computational methods used to analyze ion movement
The team investigated the material's properties by combining ceramic synthesis experiments with computational science.
First, they synthesized polycrystalline powder of by a solid-state reaction method and prepared sintered pellet samples. They used these to examine the crystal structure and measure ion conduction properties.
To investigate in detail how lithium ions move inside the crystal, they also used two computational methods together.
One was first-principles calculation based on density functional theory (DFT), which is grounded in quantum mechanics. It can evaluate interactions between atoms and the energy required for lithium ions to move. However, it is computationally demanding and limits the number of atoms and the time scales that can be handled.
The other was molecular dynamics simulation (PFP-MD) using the neural-network-based machine learning potential "Preferred Potential (PFP)." By representing interatomic interactions with a machine learning model, it makes it easier to track atomic motion in larger systems and over longer time scales than first-principles calculation.
By combining the two methods, the team analyzed lithium-ion migration pathways, which are difficult to observe directly in experiments, and the factors that hinder that movement.
They also compared calculated and experimental results. When the activation energies for each crystal orientation obtained from the molecular dynamics simulations were averaged, they agreed well with the activation energy obtained from experiments on polycrystalline pellets.
This supports the idea that the computational model captures the characteristics of ion conduction in the real material. It is significant that the study did not examine only a theoretical structure but checked it against measurements of the synthesized ceramics.
Still, agreement in activation energy alone does not prove that all ion motion inside the crystal is reproduced accurately. The migration paths and defect behavior estimated by calculation will need further experimental verification.
The study also aimed at synthesizing the material in the laboratory and clarifying the conduction mechanism by calculation. It did not evaluate charge-discharge performance or long-term durability in a practical battery cell with electrodes attached.
A low 0.30 eV migration barrier and conduction pathways extending in three dimensions
One notable result concerns the migration pathways of lithium ions.
Looking at the crystal structure of , the paths available to lithium ions appear to be limited to one-dimensional, tunnel-like channels running along the crystal's $b$ axis.
One-dimensional ion conduction paths generally have a weakness: if impurities or crystal defects arise along the way, ion movement is easily blocked.
On a road with only one lane, a blockage stops all traffic. One-dimensional conduction paths in a crystal have a similar problem.
The PFP-MD simulations, however, gave a different result.
They showed that the channels along the $b$ axis, which look independent in the crystal structure, are in fact connected to one another and form a three-dimensional migration network for lithium ions.
When the thermal vibration of atoms is taken into account, lithium ions can move not only along one tunnel but also into adjacent tunnels. As a result, even if some paths are blocked by crystal defects, ions may be able to go around through other paths.
This is considered a structure less affected by local defects than a purely one-dimensional conductor.
First-principles calculations further showed that the energy barrier for a lithium ion at an interstitial site in the crystal to migrate is only 0.30 eV.
The migration barrier is the energy hurdle an ion must overcome to move from one position in the crystal to another. In general, the lower this value, the more easily ions move.
A value of 0.30 eV is low even compared with oxide-based solid electrolytes expected to be put to practical use.
For example, the migration barrier of garnet-type LLZO is reported to be 0.26 eV, and that of (LATP), which has a NASICON-type structure, is reported to be 0.27 eV.
データを表で見る
| イオン移動障壁 (eV) | |
|---|---|
| LLZO（基準材料） | 0.26 |
| LATP（基準材料） | 0.27 |
| RbLi(Li3SiO4)2（本研究） | 0.3 |
The team points to Coulomb repulsion between neighboring lithium ions as a factor behind this low migration barrier.
Lithium ions carry a positive charge, so they repel one another. When one ion moves in the crystal, repulsion from surrounding ions encourages other ions to move as well. This kind of "concerted motion," in which multiple ions move together, is thought to be related to the low migration barrier.
However, this is an explanation of the migration mechanism derived from calculation, not the result of directly filming the movement of individual lithium ions in an experiment.
The figures for the studied material and the representative oxide-based solid electrolytes used for comparison are summarized below.
| Material | Role in this study | Theoretical density (g/cm³) | Ion migration barrier (eV) | Measured ion conductivity | Notes |
|---|---|---|---|---|---|
| Subject of this study | 2.74 | 0.30 | S/cm (150℃) | Measured on a stoichiometric sample. Practical conductivity at room temperature not yet demonstrated | |
| (LLZO) | Existing oxide-based reference material | 5.12 | 0.26 | Not measured in this study | Contains heavy elements such as lanthanum and zirconium |
| (LATP) | Existing oxide-based reference material | No comparison value given | 0.27 | Not measured in this study | Cited as a comparison for migration barrier |
Migration barriers are values reported in this study and the cited literature; they were not all measured or calculated under identical conditions.
This comparison shows that has a lower crystal density than the leading existing materials, with a similarly small migration barrier for lithium ions.
But a low migration barrier is not the same as high ion conductivity for the material as a whole.
The conductivity of the actually synthesized remained extremely low compared with what the calculated ease of ion movement would suggest.
The explanation lies in another issue the study uncovered: the formation of crystal defects.
Why is conductivity low if the ions move easily?
Against the low migration barrier of 0.30 eV obtained from first-principles calculations, the stoichiometric pellets synthesized in the laboratory showed an ion conductivity of S/cm at 150°C.
This is extremely low.
For solid electrolytes in all-solid-state batteries, although it depends on the application and design, an ion conductivity of roughly $10^{-3}$ to $10^{-2}$ S/cm at room temperature is considered one practical target.
The measured value is about 30,000 to 300,000 times lower than that target. Moreover, this measurement was made at 150°C, not at room temperature.
Why does a material that calculations show to be easy for ions to move through fail to deliver sufficient conductivity in practice?
The team determined that the cause is not the ion movement itself but the difficulty of forming the crystal defects needed for movement.
Ion conductivity is determined mainly by the number of mobile charge carriers and how easily they move.
Even in a crystal structure through which ions move very easily, conductivity of the whole material will not be high if few lithium ions are actually able to move.
What matters in this study are lithium ions at "interstitial sites," positions different from the normal atomic positions in the crystal.
In a perfect crystal, lithium ions basically occupy fixed lattice positions. When some lithium ions move from their original positions to interstitial sites, vacancies remain in the original positions and extra lithium ions appear in the interstices.
The pairing of a vacancy and an interstitial lithium formed this way is called a "Frenkel defect pair."
The calculations found that the energy required to form this Frenkel defect pair is 1.26 eV.
This is far higher than the 0.30 eV migration barrier of interstitial lithium.
In other words, once interstitial lithium exists, it can move with relatively little energy. But generating that interstitial lithium in the first place requires a large amount of energy.
In a stoichiometric crystal, the energy needed to form these defects is high, so the number of mobile lithium ions produced thermally is limited.
As a result, although the crystal structure itself allows ions to move easily, the material as a whole shows only low conductivity.
To use a road analogy, it is like having a highway built for fast driving but almost no cars on it.
However good the road is, if few cars travel on it, the total amount of cargo carried per unit time does not increase.
For this material too, the pathways ions use and the number of ions that can move along them have to be considered separately.
The team concludes that this difficulty of defect formation is the main factor limiting ion conductivity in stoichiometric . The Advanced Portfolio
The potential of lithium-rich compositions, and remaining challenges
If the main factor limiting ion conductivity is a shortage of defects, the direction for improvement becomes clear.
Rather than waiting for defects to form naturally in the crystal, the approach is to adjust the composition so that the number of lithium ions available for movement is increased in advance.
The team proposes a "lithium-rich composition," with more lithium than the stoichiometric composition, as a concrete direction.
If extra lithium ions can be introduced into the crystal lattice, it may be possible to secure the charge carriers needed for ion conduction without going through the high-energy defect formation process.
Substitution with elements of different valence is also a common way to tune defect concentration in solid electrolyte materials design. However, which elements and compositions suit this crystal must be verified separately.
Professor Nakamura of the University of Electro-Communications evaluates the crystal structure revealed in this study as a framework well suited to lithium-ion conduction that provides clear guidance for future materials design.
It is important to note that the performance improvement expected from lithium-rich compositions has not yet been confirmed experimentally.
The study did not actually synthesize lithium-rich samples and demonstrate improved ion conductivity.
Even if extra lithium can be placed in the crystal in calculations, the intended composition and structure cannot necessarily be achieved in actual synthesis.
Increasing the lithium content could also destabilize the target crystal structure or produce other compounds. And even if extra lithium can be introduced, it must be confirmed that it exists in a state where it can actually move.
An important future task is therefore to verify whether lithium-rich materials can be synthesized stably and whether conductivity actually improves.
Challenges also remain in ceramic manufacturing, not just in composition.
In polycrystalline ceramics obtained by solid-state reaction, ion conduction is affected not only by the interior of each crystal grain but also by the "grain boundaries" between grains.
If voids or impurities remain between particles, the conductivity of the whole pellet may fall even if ions move easily inside the grains.
Techniques are therefore needed to optimize sintering conditions, densify the ceramics, and reduce resistance at grain boundaries.
The University of Electro-Communications' official announcement also lists, as future challenges, achieving higher conductivity by optimizing the lithium content and improving manufacturing processes to increase density between particles.
Much more verification is needed before the material could be built into an actual battery.
For example, researchers must examine whether the electrolyte remains chemically and electrochemically stable in contact with lithium metal anodes and high-potential cathode materials.
If the electrolyte undergoes reductive or oxidative decomposition at the interface with an electrode, reaction products that hinder ion movement may form.
There is also the question of whether volume changes in the electrodes during charging and discharging degrade contact with the solid electrolyte. To confirm that the material can withstand long-term use, actual battery cells must be built and subjected to charge-discharge cycle tests.
This study did not demonstrate interfacial stability with electrodes or charge-discharge performance as a battery cell.
Still, there is considerable significance in having found potential as a solid electrolyte in a crystal known mainly as a light-emitting material.
The result is not the completion of a high-performance all-solid-state battery, but the identification of a lightweight crystal structure through which ions move easily, along with clarity about what must be improved to bring out its performance.
Going forward, an important gauge of progress will be how far room-temperature ion conductivity can be raised by actually synthesizing lithium-rich materials.
Also in question is whether the material can be densified, its interfacial stability with electrodes improved, and the work developed into a battery cell that actually charges and discharges.
When evaluating research results on all-solid-state batteries, it is necessary to distinguish between the low ion migration barrier obtained by calculation, the conductivity actually measured and the temperature at which it was measured, and the performance track record as a battery cell.
Whether the crystal structure discovered here leads to lightweight, high-performance all-solid-state batteries depends on future material synthesis and demonstration experiments.