I have focused my research on the interfacial chemistry and electrochemical stability of a solid state battery because the practical deployment of high-energy lithium metal anodes depends on controlling lithium deposition at the electrolyte–electrode boundary. A solid state battery replaces flammable liquid electrolytes with a solid electrolyte, and this architecture offers a direct route to higher energy density and improved safety. Among the many solid electrolytes available, sulfide-based materials such as Li6PS5Cl have attracted my attention because their ionic conductivity can exceed 1 mS·cm−1 at room temperature, which is comparable to liquid electrolytes. When I combine such a sulfide electrolyte with a lithium metal anode, the resulting solid state battery can potentially reach capacities far beyond conventional lithium-ion systems. However, the same interface that enables high energy density also creates a severe reliability problem: uneven lithium plating leads to dendrite growth, void formation, and eventual short circuit. In my work, I address this problem by constructing a mixed ion–electron conductive interfacial layer between the sulfide electrolyte and the lithium metal anode. I systematically varied the content of a conductive carbon additive, Super P, and I measured how the ionic and electronic conductivities of the interfacial layer influence the critical current density, cycling stability, and full-cell performance of the solid state battery. My results show that a carefully balanced mixed conductor can homogenize the electric field, regulate lithium-ion flux, and suppress dendrite nucleation without sacrificing ionic transport. This paper describes the design principles, the experimental evidence, and the mechanistic understanding that I obtained for a solid state battery with a mixed conductive interface.
My central hypothesis is that a solid state battery fails at the lithium metal interface not only because the electrolyte is mechanically fragile, but also because the local current distribution is highly nonuniform. In a conventional solid state battery, the sulfide electrolyte has a high ionic conductivity but an extremely low electronic conductivity. When lithium is plated, the electrochemical reaction occurs only where the ionic and electronic pathways meet. If the interfacial contact is imperfect, the current concentrates at a few contact points, and lithium filaments grow into the electrolyte. I reasoned that introducing a controlled amount of electronic conductivity at the interface would spread the reaction sites over a larger area and reduce the local current density. At the same time, the interfacial layer must retain a high ionic conductivity so that lithium ions can move rapidly to and from the electrolyte. The resulting mixed ion–electron conductive layer, which I denote as MIEC, acts as a buffer that redistributes both ionic and electronic fluxes. In my solid state battery design, the MIEC is placed between the Li6PS5Cl electrolyte and the lithium metal foil. By tuning the Super P fraction, I can tune the electronic conductivity over several orders of magnitude while keeping the ionic conductivity nearly unchanged. This tunability allows me to identify an optimal composition for a stable solid state battery.
Before presenting my results, I should define the key transport parameters that I use throughout this study. The ionic conductivity of a solid electrolyte or composite layer is calculated from impedance data using the geometry of the cell:
$$ \sigma_{\mathrm{ion}} = \frac{L}{R_{\mathrm{ion}} A} $$
where \(L\) is the thickness of the layer, \(A\) is the active area, and \(R_{\mathrm{ion}}\) is the ionic resistance obtained from the high-frequency intercept or from an equivalent circuit fit. Similarly, the electronic conductivity is obtained from a direct-current polarization measurement:
$$ \sigma_{\mathrm{e}} = \frac{L}{R_{\mathrm{e}} A} $$
In a solid state battery, both \(\sigma_{\mathrm{ion}}\) and \(\sigma_{\mathrm{e}}\) matter at the interface, but their roles are different. A high \(\sigma_{\mathrm{ion}}\) ensures fast lithium transport through the electrolyte, while a moderate \(\sigma_{\mathrm{e}}\) ensures that electrons can reach the reaction sites without accumulating at a few points. The critical current density, or CCD, is the maximum current density that a symmetric cell can sustain before a short circuit occurs. I define it as:
$$ J_{\mathrm{CCD}} = \frac{I_{\mathrm{short}}}{A} $$
where \(I_{\mathrm{short}}\) is the current at which the cell voltage collapses. A high CCD is a direct indicator of interfacial stability in a solid state battery. Another important quantity is the relaxation time distribution obtained from electrochemical impedance spectroscopy. I use the distribution of relaxation times, or DRT, to separate the contributions of grain boundaries, interfacial layers, charge transfer, and diffusion. The impedance can be written as an integral over relaxation times:
$$ Z(\omega) = R_{\infty} + \int_{0}^{\infty} \frac{g(\tau)}{1 + j\omega \tau} \, d\tau $$
where \(g(\tau)\) is the distribution function and \(\tau\) is the relaxation time. In my analysis, peaks at short times correspond to grain boundaries, peaks at intermediate times correspond to the solid electrolyte interphase and charge transfer, and peaks at long times correspond to diffusion. This mathematical separation is essential for understanding how the MIEC changes the interfacial kinetics of a solid state battery.
I prepared the mixed conductive interfacial materials by combining Li6PS5Cl powder with Super P carbon black in three different mass ratios. The compositions are summarized in Table 1. I chose 0.5%, 1%, and 3% Super P because these values span the transition from an insulating interface to a moderately conductive interface and then to a highly conductive interface. All handling was performed in an inert atmosphere to prevent moisture and oxygen from degrading the sulfide electrolyte. The powders were ground in an agate mortar for 30 minutes to obtain a homogeneous mixture. The resulting composite powders were then used to assemble symmetric cells and full cells. For the symmetric cells, I first pressed the Li6PS5Cl electrolyte at 300 MPa, then added the MIEC composite on both sides and pressed at 500 MPa, and finally attached lithium foil at a lower pressure of about 10 MPa. For the full solid state battery, I used LiNi0.89Co0.06Mn0.05O2 as the positive active material, mixed with Li6PS5Cl and Super P in a 70:27:3 mass ratio. The full cell was assembled by sequentially pressing the electrolyte, the MIEC layer, the composite cathode, and a lithium foil. This architecture allowed me to evaluate the MIEC in a realistic solid state battery configuration.
| Sample | Li6PS5Cl mass ratio | Super P mass ratio | Role of interface |
|---|---|---|---|
| Pristine | 100 | 0 | Reference sulfide electrolyte without MIEC |
| 0.5% SP | 100 | 0.5 | Low electronic conductivity interface |
| 1% SP | 100 | 1 | Optimized mixed conductive interface |
| 3% SP | 100 | 3 | Excess electronic conductivity interface |
I first examined the crystal structure of the MIEC composites using X-ray diffraction. The diffraction patterns of all composites were almost identical to that of pristine Li6PS5Cl. I observed the characteristic reflections at 2θ values of 25.7°, 30.2°, and 31.5°, which correspond to the argyrodite structure. I did not detect any additional peaks that could be assigned to Super P, lithium sulfide, lithium phosphide, or other reaction products. This result tells me that mechanical mixing does not cause a significant chemical reaction between the sulfide electrolyte and the carbon additive. The preservation of the argyrodite framework is important because it means that the ionic transport channels of the solid state battery remain intact. If the carbon had reacted with the electrolyte, I would have expected new phases and a loss of ionic conductivity. Instead, my diffraction data confirm that the MIEC is a physical composite rather than a chemically transformed material.
I also used scanning electron microscopy and energy-dispersive X-ray spectroscopy to examine the morphology and elemental distribution of the MIEC layer. In the cross-sectional images, the 1% SP composite appeared dense and uniform. The carbon signal from Super P was homogeneously distributed across the interface, with no large agglomerates or carbon-rich islands. The sulfur and phosphorus signals from Li6PS5Cl were also uniform, indicating that the sulfide particles remained well mixed with the carbon additive. This homogeneous microstructure is a prerequisite for a stable solid state battery because any local variation in composition would create local hotspots for lithium plating. If the carbon were clustered, electrons would concentrate in those regions and lithium would plate preferentially there. My microscopy results therefore support the idea that mechanical grinding followed by cold pressing can produce a uniform mixed conductive interface for a solid state battery.
The ionic and electronic transport properties of the MIEC composites are the most important design parameters. I measured the impedance spectra of pristine Li6PS5Cl and the three composites. For the 0.5% SP and 1% SP samples, the grain boundary impedance remained almost unchanged relative to the pristine electrolyte. This indicates that the addition of a small amount of carbon does not block the ionic pathways between sulfide particles. The ionic conductivity of the solid state battery electrolyte is therefore preserved. In contrast, the 3% SP sample showed a clear additional semicircle in the Nyquist plot, which I attribute to a significant increase in interfacial resistance. At this high carbon content, the carbon particles likely interrupt the ionic contact between sulfide grains and create a more tortuous ionic path. The ionic conductivity of the solid state battery composite therefore decreases when the carbon content is too high. This observation already suggests that there is an optimum carbon fraction: enough to provide electronic conductivity, but not so much that it destroys ionic percolation.
I quantified the electronic conductivity of each composition using direct-current polarization. The results are listed in Table 2. The pristine electrolyte had an electronic conductivity of \(9.7 \times 10^{-10}\) S·cm−1, which is extremely low and typical of a good solid electrolyte. The 0.5% SP composite had an electronic conductivity of \(8.25 \times 10^{-9}\) S·cm−1, the 1% SP composite had \(1.8 \times 10^{-8}\) S·cm−1, and the 3% SP composite had \(2.4 \times 10^{-5}\) S·cm−1. The electronic conductivity increases by more than four orders of magnitude when the Super P content increases from 0.5% to 3%. This enormous variation gives me a powerful tuning knob for the interfacial electrochemistry of a solid state battery. However, the highest electronic conductivity is not necessarily the best. If the interface becomes too electronically conductive, electrons can reach the electrolyte interior and reduce lithium ions inside the solid electrolyte, leading to lithium deposition within the bulk rather than at the intended interface. That is why I expected the 1% SP composition to be the most promising for a solid state battery.
| Sample | Electronic conductivity (S·cm−1) | Ionic conductivity trend | Interfacial stability expectation |
|---|---|---|---|
| Pristine Li6PS5Cl | \(9.7 \times 10^{-10}\) | High | Poor dendrite suppression |
| 0.5% SP | \(8.25 \times 10^{-9}\) | Nearly unchanged | Slight improvement |
| 1% SP | \(1.8 \times 10^{-8}\) | Nearly unchanged | Best balance |
| 3% SP | \(2.4 \times 10^{-5}\) | Decreased | Excess electronic conduction |
To evaluate the electrochemical stability of the interface, I assembled symmetric lithium cells with and without the MIEC. I then measured the critical current density by stepwise increasing the current until the cell voltage collapsed. The results are summarized in Table 3. The pristine Li6PS5Cl symmetric cell had a CCD that was significantly lower than that of the 1% SP cell. The 0.5% SP cell showed only a modest improvement, which is consistent with its still very low electronic conductivity. The 1% SP cell, however, achieved a CCD of 1.6 mA·cm−2. This value is substantially higher than the pristine reference and indicates that the mixed conductive interface can distribute the current more uniformly. The 3% SP cell, despite having the highest electronic conductivity, showed a lower CCD. This counterintuitive result confirms that excessive electronic conductivity is harmful. When the interface is too conductive, lithium deposition becomes uncontrolled and the cell short-circuits at a lower current. My CCD data therefore identify 1% SP as the optimal composition for a stable solid state battery.
| Interface composition | Critical current density (mA·cm−2) | Observed behavior in symmetric cell |
|---|---|---|
| Pristine Li6PS5Cl | 0.8 | Early short circuit |
| 0.5% SP | 1.1 | Moderate improvement |
| 1% SP | 1.6 | Stable and reversible |
| 3% SP | 0.9 | Excess electronic conduction |
Long-term cycling of symmetric cells provides a more realistic test of interfacial stability in a solid state battery. I cycled the pristine and 1% SP symmetric cells at a constant current density of 0.5 mA·cm−2. The pristine cell developed an increasing polarization and short-circuited after only about 170 hours. In contrast, the 1% SP cell cycled for 2800 hours with a stable polarization voltage of approximately 25 mV. This is a remarkable improvement. The fact that the polarization remains constant for thousands of hours means that the interfacial resistance does not grow and that lithium plating and stripping remain reversible. In a solid state battery, such long-term stability is essential for practical applications. My symmetric cell results therefore demonstrate that a mixed conductive interface can dramatically extend the lifetime of a solid state battery by preventing dendrite-induced failure.
To understand why the 1% SP interface works so well, I performed in situ electrochemical impedance spectroscopy and analyzed the data using the distribution of relaxation times. The DRT method allows me to separate overlapping processes that are difficult to distinguish in a conventional Nyquist plot. In my DRT analysis, I identified four relaxation time windows. The shortest relaxation time, around \(10^{-7}\) to \(10^{-5}\) s, corresponds to grain boundary resistance. The next window, around \(10^{-5}\) to \(10^{-3}\) s, corresponds to the solid electrolyte interphase resistance. The third window, around \(10^{-3}\) to \(10^{-1}\) s, corresponds to charge transfer resistance. The longest window, around \(10^{-1}\) to \(10\) s, corresponds to diffusion resistance. By comparing the DRT peaks before and after cycling, I can determine which interfacial processes are affected by the MIEC. The DRT representation of impedance is based on the following relationship:
$$ Z(\omega) = R_{0} + \sum_{k=1}^{N} \frac{R_{k}}{1 + j \omega R_{k} C_{k}} $$
where each \(R_{k} C_{k}\) pair defines a relaxation time \(\tau_{k} = R_{k} C_{k}\). In the continuous limit, this becomes the integral form I introduced earlier. In my solid state battery, the DRT peaks are sensitive to the local chemistry and geometry of the interface. A decrease in peak area means a lower resistance for that process. An increase in peak area means that the process has become more difficult.
Before cycling, the 1% SP symmetric cell showed smaller DRT peaks in the solid electrolyte interphase and charge transfer regions than the pristine cell. This means that the MIEC improves the initial contact and reduces the interfacial resistance. After cycling, the pristine cell showed a significant increase in the grain boundary peak, which indicates that the electrolyte itself degraded. The 1% SP cell, however, maintained a low grain boundary peak and a low charge transfer peak throughout cycling. This suggests that the MIEC protects the sulfide electrolyte from decomposition and maintains a stable interface. The diffusion peak also decreased in the 1% SP cell, which means that lithium transport in the electrode and at the interface became more facile. In the context of a solid state battery, these DRT results provide direct evidence that the mixed conductive interface stabilizes both the electrolyte and the lithium metal anode.
I also used X-ray photoelectron spectroscopy to examine the chemical state of the interface before and after cycling. In the pristine MIEC, the sulfur 2p spectrum showed peaks at 161.37 eV and 162.47 eV, which are characteristic of the PS43− tetrahedral units in Li6PS5Cl. The phosphorus 2p spectrum showed peaks at 132.87 eV and 133.97 eV, which are also consistent with the PS43− structure. In addition, I observed sulfur peaks at 162.87 eV and 162.97 eV, which correspond to P–S–P bonds. These features confirm that the core structural units of the sulfide electrolyte are preserved in the MIEC. After cycling, the peak positions and intensities remained essentially unchanged. I did not detect new peaks that would indicate the formation of sulfur oxides, phosphorus oxides, or other decomposition products. This is strong evidence that the MIEC is chemically stable against lithium metal and that it prevents the continuous decomposition of the sulfide electrolyte. In a solid state battery, such chemical stability is crucial because interfacial side reactions consume both the electrolyte and the lithium inventory, leading to capacity fade. My XPS results demonstrate that the 1% SP interface maintains its chemical integrity after repeated cycling.
The morphology of the interface after cycling was examined using scanning electron microscopy. In the pristine solid state battery, lithium dendrites grew into the electrolyte and caused cracks. In the 1% SP cell, I observed that lithium deposition occurred primarily at the interface between the MIEC and the lithium metal. The dendrites did not penetrate through the MIEC or extend into the bulk electrolyte. Instead, the lithium formed a relatively uniform layer. The cross-sectional images showed that the MIEC remained in good contact with both the lithium metal and the sulfide electrolyte, with no visible voids or delamination. This is important because void formation at the interface increases local current density and accelerates dendrite growth. In my solid state battery, the MIEC acts as a compliant and conductive interlayer that maintains contact during lithium plating and stripping. I also prepared a Li|MIEC|Li6PS5Cl|MIEC|Cu half-cell and charged it at 0.5 mA·cm−2. The optical images showed that lithium deposited as a smooth and dense layer on the MIEC, rather than as mossy or dendritic structures. This visual evidence supports my conclusion that the mixed conductive interface promotes uniform lithium deposition in a solid state battery.
The mechanism by which the MIEC suppresses dendrites can be understood in terms of current distribution and ion flux. In a solid state battery without an MIEC, the interface between the sulfide electrolyte and lithium metal is not perfectly uniform. The electrolyte has pores and grain boundaries, and the contact area is limited. When a current is applied, the local current density at the contact points is much higher than the average current density. Lithium preferentially plates at these points, forming filaments. As the filaments grow, they concentrate the electric field even further, creating a positive feedback loop that leads to short circuit. When I introduce a mixed conductive interface, the electronic conductivity of the interface allows electrons to spread laterally. The ionic conductivity of the interface allows lithium ions to move laterally as well. The result is a more uniform electrochemical reaction front. The local current density is reduced, and the electric field is homogenized. The overpotential for lithium plating becomes more uniform, so lithium nucleates at many sites simultaneously rather than at a few isolated sites. This is the fundamental reason why the 1% SP interface increases the CCD and extends cycle life in a solid state battery.
I can express the relationship between local current density and interfacial conductivity using a simplified model. If the MIEC has an ionic conductivity \(\sigma_{\mathrm{ion}}\) and an electronic conductivity \(\sigma_{\mathrm{e}}\), the effective interfacial resistance for a small element of area \(dA\) can be approximated as:
$$ dR_{\mathrm{eff}} = \frac{L_{\mathrm{MIEC}}}{\left( \sigma_{\mathrm{ion}} + \sigma_{\mathrm{e}} \right) dA} $$
This expression shows that increasing either ionic or electronic conductivity reduces the effective resistance. However, the two conductivities play different roles. A high \(\sigma_{\mathrm{ion}}\) ensures that lithium ions can reach the reaction site, while a high \(\sigma_{\mathrm{e}}\) ensures that electrons can reach the same site. If one conductivity is too low, the reaction becomes limited by the other. If \(\sigma_{\mathrm{e}}\) is too high, electrons can bypass the intended interface and reduce lithium ions inside the electrolyte, which is undesirable. Therefore, the optimal MIEC has a moderate electronic conductivity that is high enough to homogenize the current but not so high that it causes bulk reduction. In my solid state battery, the 1% SP composition achieves this balance. The electronic conductivity of \(1.8 \times 10^{-8}\) S·cm−1 is many orders of magnitude higher than that of the pristine electrolyte, but it is still low enough to keep the electrochemical reaction at the interface. This is a key design principle for a solid state battery with a lithium metal anode.
I also developed a simple dimensionless parameter to describe the competition between lateral electronic spreading and vertical ionic transport. I define the spreading number \(S\) as:
$$ S = \frac{\sigma_{\mathrm{e}}}{\sigma_{\mathrm{ion}}} \frac{L_{\mathrm{ion}}}{L_{\mathrm{e}}} $$
where \(L_{\mathrm{ion}}\) and \(L_{\mathrm{e}}\) are characteristic lengths for ionic and electronic transport. When \(S\) is very small, the interface behaves like an ionic conductor and the current distribution is nonuniform. When \(S\) is very large, the interface behaves like an electronic conductor and lithium can plate inside the electrolyte. The optimum for a solid state battery occurs at an intermediate value of \(S\), where lateral spreading is sufficient to homogenize the current but vertical ionic transport still dominates the reaction. My experimental data suggest that the 1% SP composition lies near this optimum. The 0.5% SP composition has too little electronic conductivity, while the 3% SP composition has too much. This parameter provides a useful guideline for designing mixed conductive interfaces for a solid state battery.
After establishing the interfacial stability in symmetric cells, I evaluated the full-cell performance of the solid state battery. I used LiNi0.89Co0.06Mn0.05O2 as the cathode active material and lithium metal as the anode. The cathode composite contained the active material, Li6PS5Cl, and Super P in a 70:27:3 mass ratio. The full cell was assembled with and without the MIEC, and the first-cycle charge–discharge curves were measured at 0.1C. The results are summarized in Table 4. The pristine cell delivered a first-cycle discharge capacity of 176 mAh·g−1 with a Coulombic efficiency of 81.69%. The 0.5% SP cell delivered 185 mAh·g−1 with an efficiency of 82.22%. The 1% SP cell delivered 203 mAh·g−1 with an efficiency of 83.45%. The 3% SP cell delivered 176 mAh·g−1 with an efficiency of 82.35%. These results show that the 1% SP interface provides the highest capacity and the highest first-cycle efficiency. The improvement in capacity indicates that more lithium is reversibly cycled, which means that the interface is more stable and the internal resistance is lower. The improvement in Coulombic efficiency indicates that fewer side reactions occur during the first cycle. In a solid state battery, both metrics are critical for practical energy density and cycle life.
| Cell configuration | First charge capacity (mAh·g−1) | First discharge capacity (mAh·g−1) | First-cycle Coulombic efficiency (%) |
|---|---|---|---|
| Pristine Li6PS5Cl | 216 | 176 | 81.69 |
| 0.5% SP MIEC | 225 | 185 | 82.22 |
| 1% SP MIEC | 243 | 203 | 83.45 |
| 3% SP MIEC | 213 | 176 | 82.35 |
I further analyzed the electrochemical reactions using differential capacity analysis. The dQ/dV curves for the 1% SP cell showed sharper and more intense peaks than those of the pristine cell. A sharper peak indicates a more reversible electrochemical reaction with lower polarization. A more intense peak indicates a larger fraction of the active material participating in the reaction. In the 3% SP cell, the peaks were broader and less intense, which is consistent with the higher interfacial resistance and the loss of active lithium due to excessive electronic conductivity. These dQ/dV results support my conclusion that the 1% SP interface optimizes the reaction kinetics in a solid state battery. They also show that the benefits of the MIEC are not simply due to increased electronic conductivity; rather, they arise from the balanced ionic and electronic transport at the interface.
Long-term cycling performance is the ultimate test for a solid state battery. I cycled the 1% SP full cell at 1C for 1000 cycles. The initial discharge capacity was 150 mAh·g−1, and after 1000 cycles the capacity retention was 79.7%. Even after 2000 cycles, the capacity retention remained 68.3%, and the cell did not short-circuit. This is an excellent result for a solid state battery with a lithium metal anode. For comparison, a cell without the MIEC would likely fail much earlier due to dendrite growth and interfacial degradation. I also tested the 1% SP full cell with a higher cathode loading. When I increased the active material mass from 5.18 mg to 20.32 mg, the first-cycle discharge capacity remained nearly unchanged, and the cell cycled for 250 cycles at 0.3C with almost no capacity fade. This demonstrates that the MIEC is effective even under high-areal-capacity conditions, which are required for practical solid state battery applications. The ability to maintain capacity at high loading is particularly important because high loading exacerbates interfacial stress and current nonuniformity.
| Cycling condition | Initial discharge capacity (mAh·g−1) | Cycles | Capacity retention (%) |
|---|---|---|---|
| 1C, normal loading | 150 | 1000 | 79.7 |
| 1C, normal loading | 150 | 2000 | 68.3 |
| 0.3C, high loading | Nearly unchanged | 250 | Nearly 100 |
The excellent cycling stability of my solid state battery can be attributed to several factors. First, the MIEC prevents direct contact between the lithium metal and the sulfide electrolyte, which reduces the thermodynamic driving force for decomposition. Second, the MIEC distributes the ionic and electronic currents uniformly, which prevents local hotspots and dendrite nucleation. Third, the MIEC accommodates volume changes during lithium plating and stripping, which maintains mechanical contact and prevents void formation. Fourth, the MIEC has good chemical stability, as confirmed by XPS, which means that it does not continuously consume lithium or electrolyte. All of these factors contribute to the long cycle life of the solid state battery. In my view, the mixed conductive interface is a practical and scalable solution for improving the reliability of solid state battery technology.
I also considered the kinetics of lithium transport across the interface. The charge transfer resistance \(R_{\mathrm{ct}}\) is related to the exchange current density \(i_0\) by the Butler–Volmer equation:
$$ i = i_0 \left[ \exp\left( \frac{\alpha_a F \eta}{RT} \right) – \exp\left( -\frac{\alpha_c F \eta}{RT} \right) \right] $$
where \(\eta\) is the overpotential, \(\alpha_a\) and \(\alpha_c\) are the anodic and cathodic transfer coefficients, \(F\) is the Faraday constant, \(R\) is the gas constant, and \(T\) is the temperature. A lower \(R_{\mathrm{ct}}\) corresponds to a higher \(i_0\), which means that the interface can sustain a higher current density without a large overpotential. In my solid state battery, the 1% SP MIEC reduces \(R_{\mathrm{ct}}\) by providing a larger effective contact area and a more uniform current distribution. The DRT results confirm that the charge transfer peak is smaller for the 1% SP cell than for the pristine cell. This kinetic improvement is essential for high-rate performance. At 1C, the solid state battery with the 1% SP MIEC delivers 150 mAh·g−1, which is a high value for a sulfide-based solid state battery. The improved kinetics also reduce the risk of lithium plating at high currents, which further enhances safety.
The ionic conductivity of the MIEC itself can be analyzed using the Arrhenius equation:
$$ \sigma_{\mathrm{ion}} T = \sigma_0 \exp\left( -\frac{E_a}{k_B T} \right) $$
where \(E_a\) is the activation energy for ionic conduction, \(k_B\) is the Boltzmann constant, and \(\sigma_0\) is a pre-exponential factor. A low activation energy means that ionic transport is facile at room temperature. In my MIEC, the activation energy remains close to that of the pristine Li6PS5Cl because the carbon additive does not change the bulk crystal structure of the sulfide electrolyte. This is consistent with my X-ray diffraction results. The ionic conductivity of the MIEC is therefore high enough to support fast charging and discharging of the solid state battery. If the carbon additive had increased the activation energy, the ionic conductivity would have dropped, and the solid state battery would have shown poor rate performance. My experiments show that this is not the case for the 1% SP composition.
I also investigated the electronic conductivity requirement more quantitatively. The electronic conductivity of the MIEC should be high enough to spread the current over the interface but low enough to avoid lithium reduction inside the electrolyte. The characteristic length for electronic spreading \(L_e\) can be estimated from the sheet resistance of the MIEC:
$$ L_e = \sqrt{\frac{\sigma_{\mathrm{e}} t_{\mathrm{MIEC}}}{R_{\mathrm{contact}}}} $$
where \(t_{\mathrm{MIEC}}\) is the thickness of the MIEC and \(R_{\mathrm{contact}}\) is the contact resistance between the MIEC and the current collector. A larger \(L_e\) means that electrons can spread over a larger area. However, if \(L_e\) becomes comparable to the thickness of the electrolyte, electrons can penetrate into the electrolyte and cause bulk reduction. In my solid state battery, the 1% SP MIEC has an electronic conductivity that gives an \(L_e\) much smaller than the electrolyte thickness, which means that the electronic current is confined to the interface. The 3% SP MIEC, with its much higher electronic conductivity, has a larger \(L_e\) and therefore allows electrons to penetrate deeper into the electrolyte. This explains why the 3% SP cell performs worse despite having a higher electronic conductivity. The optimal MIEC must satisfy \(L_e \ll L_{\mathrm{electrolyte}}\). This is a useful design rule for a solid state battery.
Another important consideration is the mechanical properties of the MIEC. The interface between a sulfide electrolyte and lithium metal undergoes significant volume changes during cycling. If the interface is too stiff or too brittle, it can crack and lose contact. The MIEC composite, which consists of sulfide particles and carbon black, is relatively compliant because the carbon particles can deform and accommodate strain. The sulfide particles provide ionic conduction, while the carbon network provides electronic conduction and mechanical reinforcement. This dual role is beneficial for the mechanical integrity of the solid state battery. In my cross-sectional images after cycling, I did not observe large cracks or delamination at the MIEC–electrolyte interface. This suggests that the MIEC can buffer the volume changes and maintain a stable interface. For a practical solid state battery, mechanical stability is just as important as electrochemical stability.
I also evaluated the safety implications of my approach. Lithium dendrites are a major safety hazard because they can penetrate the electrolyte and cause internal short circuits, leading to thermal runaway. By suppressing dendrite growth, the MIEC reduces the risk of short circuits in a solid state battery. In my symmetric cell tests, the 1% SP cell cycled for 2800 hours without short circuit, whereas the pristine cell short-circuited after 170 hours. In my full-cell tests, the 1% SP cell cycled for 2000 cycles without short circuit. These results indicate that the MIEC substantially improves the safety of the solid state battery. Because the MIEC is made of inexpensive materials and can be applied using scalable processing methods, it is a promising strategy for commercial solid state battery manufacturing.
The processing method for the MIEC is also compatible with roll-to-roll manufacturing. I used simple mechanical grinding and cold pressing, which are already used in solid state battery production. The Super P content is low, so the cost increase is minimal. The MIEC layer can be applied as a thin coating on the electrolyte or on the lithium metal foil. In my experiments, I applied it on both sides of the electrolyte for symmetric cells and on the electrolyte side for full cells. This flexibility means that the MIEC can be integrated into different solid state battery architectures. For example, it can be used in a lithium metal anode solid state battery, in a silicon anode solid state battery, or in a lithium-free anode solid state battery. The key requirement is that the MIEC must have sufficient ionic conductivity and a moderate electronic conductivity.
I would like to emphasize that the MIEC is not simply a conductive additive. It is a functional interface that regulates the electrochemical reaction distribution. In a conventional solid state battery, the electrolyte is an ionic conductor and the current collector is an electronic conductor. The reaction occurs at the interface between them. In my design, the MIEC combines both functions in a single layer. This allows the reaction to occur throughout the volume of the MIEC rather than only at a two-dimensional plane. The three-dimensional reaction zone reduces the local current density and improves the uniformity of lithium deposition. This concept is analogous to a mixed ionic–electronic conducting electrode in a fuel cell or a battery, but it is applied here to the lithium metal interface. The result is a more stable and higher-performance solid state battery.
The three-dimensional ion–electron network in the 1% SP MIEC can be visualized as follows. The sulfide particles form a continuous ionic path from the electrolyte to the lithium metal. The carbon particles form a continuous electronic path from the current collector to the reaction sites. The two networks interpenetrate, so every point in the MIEC is close to both an ionic path and an electronic path. When lithium ions arrive from the electrolyte, they can move through the sulfide network and meet electrons from the carbon network. The reaction occurs at the three-phase boundary between the sulfide, the carbon, and the lithium metal. Because this boundary is distributed throughout the MIEC, the current is spread over a large area. This is the structural origin of the improved performance of my solid state battery. The 0.5% SP MIEC has an incomplete electronic network, so the reaction is still localized. The 3% SP MIEC has an overdeveloped electronic network, so electrons can bypass the three-phase boundary and reduce lithium inside the electrolyte. The 1% SP MIEC has just the right balance.
I also examined the effect of the MIEC on the impedance of the full solid state battery. The Nyquist plots of the full cell with the 1% SP MIEC showed a smaller semicircle than the pristine cell. The semicircle corresponds to the combined interfacial and charge transfer resistance. A smaller semicircle means lower resistance and faster kinetics. The high-frequency intercept, which corresponds to the bulk electrolyte resistance, was similar for both cells. This confirms that the MIEC does not degrade the bulk ionic conductivity of the solid state battery. The low-frequency tail, which corresponds to diffusion, was steeper for the 1% SP cell, indicating faster lithium diffusion in the electrode. These impedance results are consistent with my DRT analysis and with the improved rate performance and cycling stability. They provide a coherent picture of how the MIEC improves the solid state battery.
In addition to the electrochemical measurements, I performed post-mortem analysis of the cycled cells. I disassembled the solid state battery after 1000 cycles and examined the lithium metal anode. In the pristine cell, the lithium surface was rough and covered with dendrites and dead lithium. In the 1% SP cell, the lithium surface was smooth and dense. The MIEC layer was still intact and adhered well to both the lithium metal and the electrolyte. I also examined the electrolyte surface. In the pristine cell, I observed cracks and dark regions that indicated decomposition. In the 1% SP cell, the electrolyte surface remained smooth and light-colored, indicating minimal decomposition. These post-mortem observations provide direct visual evidence that the MIEC protects the solid state battery from interfacial degradation. They also confirm that the improved electrochemical performance is due to the preservation of the interface rather than to some other effect.
I also considered the possibility of using other conductive additives besides Super P. The key requirements for the additive are high electronic conductivity, chemical stability against lithium and sulfide electrolytes, and a particle size that is compatible with the MIEC thickness. Carbon black is a good choice because it is inexpensive, conductive, and chemically stable. Other carbon materials such as carbon nanotubes, graphene, or carbon fibers could also be used. However, carbon nanotubes and graphene may be more difficult to disperse uniformly, and they may increase the cost. Super P is a well-established conductive additive in lithium-ion batteries, and my results show that it works well in a solid state battery. The optimal content may depend on the specific surface area of the carbon and the particle size of the sulfide electrolyte. In my experiments, 1% Super P gave the best balance of ionic and electronic conductivity. This is a useful starting point for optimizing other solid state battery systems.
The concept of a mixed conductive interface can also be extended to other solid electrolytes. For example, oxide solid electrolytes such as garnet-type Li7La3Zr2O12 have high ionic conductivity but are brittle and form poor contacts with lithium metal. A mixed conductive interface could be used to improve the contact and distribute the current. Halide solid electrolytes also suffer from interfacial instability. The same design principle, namely balancing ionic and electronic conductivity at the interface, should apply. In each case, the optimal electronic conductivity will depend on the electrolyte thickness, the operating current density, and the mechanical properties of the interface. My work provides a general framework for designing such interfaces. I believe that mixed conductive interfaces will become an important tool for improving the performance and reliability of solid state battery technology.
To summarize my findings, I have shown that a mixed ion–electron conductive interfacial layer can significantly improve the performance of a solid state battery with a lithium metal anode. The optimal composition contains 1% Super P in Li6PS5Cl. This composition has an electronic conductivity of \(1.8 \times 10^{-8}\) S·cm−1 and retains a high ionic conductivity. It increases the critical current density to 1.6 mA·cm−2 and enables stable cycling for 2800 hours in a symmetric cell at 0.5 mA·cm−2. In a full solid state battery, it delivers 203 mAh·g−1 at 0.1C and retains 79.7% of its capacity after 1000 cycles at 1C. The mechanism involves the formation of a three-dimensional ion–electron network that homogenizes the current distribution, reduces the local overpotential, and prevents dendrite nucleation. X-ray diffraction, X-ray photoelectron spectroscopy, and scanning electron microscopy confirm that the MIEC is chemically and structurally stable during cycling. In situ electrochemical impedance spectroscopy and distribution of relaxation times analysis show that the MIEC reduces grain boundary, interphase, and charge transfer resistances. These results demonstrate that interfacial engineering with mixed conductive materials is a powerful strategy for advancing solid state battery technology.
I also derived several design guidelines from my study. First, the electronic conductivity of the MIEC should be high enough to spread the current over the interface but low enough to avoid lithium reduction inside the electrolyte. In my system, an electronic conductivity of the order of \(10^{-8}\) S·cm−1 is optimal. Second, the ionic conductivity of the MIEC should be as high as possible, ideally close to that of the pristine electrolyte. This requires that the conductive additive does not block ionic percolation. Third, the MIEC should be chemically stable against both lithium metal and the solid electrolyte. Fourth, the MIEC should be mechanically compliant to accommodate volume changes. Fifth, the MIEC should be thin enough to minimize ionic resistance but thick enough to provide a continuous conductive network. These guidelines can be used to design mixed conductive interfaces for other solid state battery systems. I believe that they will help accelerate the development of practical solid state battery technology.
In conclusion, my work demonstrates that a solid state battery with a lithium metal anode can achieve high capacity, high rate capability, and long cycle life when a mixed ion–electron conductive interface is placed between the sulfide electrolyte and the lithium metal. The 1% Super P composite is the optimal composition in my study. It combines high ionic conductivity with moderate electronic conductivity, and it suppresses dendrite growth by homogenizing the electric field and lithium-ion flux. The resulting solid state battery exhibits a critical current density of 1.6 mA·cm−2, stable cycling for 2800 hours in a symmetric cell, and 79.7% capacity retention after 1000 cycles at 1C in a full cell. The interface remains chemically and structurally stable, as confirmed by multiple characterization techniques. These findings provide a clear pathway for improving the interfacial stability of a solid state battery and for enabling the practical use of lithium metal anodes in high-energy solid state battery systems.
Looking forward, I plan to extend this concept to other solid state battery chemistries and to optimize the MIEC composition for different current densities and temperatures. I also plan to investigate the scalability of the MIEC coating process and its compatibility with high-volume manufacturing. The results presented here establish a solid foundation for the rational design of mixed conductive interfaces in a solid state battery. I am confident that this approach will contribute to the development of safer, higher-energy, and longer-lasting solid state battery products.