The Thio-LISICON family, a sulfur derivative of LISICON, was initially introduced by Kanno et al. [21]. They have replaced O²⁻ ions with S²⁻ ions, which increased the mobility of Li⁺ ions due to their larger size and more polarizable character. It has been shown that this substitution allows for an increase in the ion conductivity by two orders of magnitude (e.g., 2 × 10⁻⁶ S cm⁻¹ for a Li_3.6Si_0.6P_0.4O₄ and 10⁻⁴ S cm⁻¹ for Li₂S-SiS₂-LiI) [21].

In recent years, one electrolyte and its derivatives have increasingly emerged as the first suitable candidate. This is, for example, LGPS with the empirical formula Li₁₀GeP₂S₁₂. Here, even conductivities in the particle of more than 10⁻² S cm⁻¹ are reported ^[21][22][21,22]; a practical achievement was ~10⁻³ S cm⁻¹ in a suitably prepared whole electrode ^[23][24][23,24]. It is expected that a powder of this electrolyte can be compacted into films with an inert auxiliary binder so that satisfactory contact of the individual crystals with a sufficient density can be achieved in order to suppress dendrites.

Moreover, research results indicate that Li₁₀GeP₂S₁₂ is compatible with NMC (8:1:1) cathodes, LiNi_0.8Co_0.1Mn_0.1O₂, which is considered an up-and-coming candidate for Li-ion cells. Such a cathode would then contain the solid electrolyte in the gaps of the densest NMC sphere (cubic close pack), for example, about 10–25 µm thick NCM cathode particles. The solid electrolyte must have particle sizes of at least one order of magnitude less for this to occur, an important design feature for the solid electrolyte powder. Thus, particles in the nanometer range are required.

However, there are still some challenges concerning their stability with the electrode interface and affordability. Regarding the electrochemical stability of LGPS, it is relatively limited due to the reduction of Ge⁴⁺ in Ge⁰ below 1.0 V vs. Li, and the oxidation of S²⁻ above 2.8 V vs. Li. Its strong reactivity at the interface with the Li metal forms products, which lead to the decomposition of LGPS, producing undesirable interphase products composed of Li₃P, Li₂S, and Li-Ge alloy [25]. Actually, the decomposition products Li₃P and Li₂S are expected to be sufficiently good solid ionic conductors to enable ionic conductivity, but Li_xGe alloy causes unfavorable repeated volume surges during the formation and delithiation of the alloy. That is why the stability to so-called 4 and even 5 V electrodes of sulfidic compounds needs further study.

In order to overcome these stability issues, several surface modification methods have been proposed ^[15][24][25][15,24,25]. For example, Zhang et al. proposed the use of a protective layer between LGPS SE and the Li metal anode, namely LiH₂PO₄ [24]. They have coated the Li metal with different concentrations of H₃PO₄ tetrahydrofuran and created an in-situ LiH₂PO₄ surface. In this way, they achieved 113.7 mAh g⁻¹ of a discharge capacity up to 500 cycles (at 0.1C with 80 wt% H₃PO₄). Nevertheless, the surface treatments increased the total impedance of the Li anode (more than 2.5 times). Additionally, it has been reported that long-term cycling causes a volume changing effect.

Another proposition to increase the stability of the LGPS solid electrolyte towards the Li metal anode is to prepare the cells with double-layer (bilayer) electrolytes ^{[26][27][28][29][30]}[26,27,28,29,30]. Within this approach, a Li-compatible SE-layer (mostly based on Li₂S-P₂S₅) is inserted between LGPS and Li metal. The cell characteristics prepared with bilayer electrolytes are summarized in Table 1.

As can be seen in Table 1, the cells prepared with the bilayer InSEs showed better cycle and capacity performance compared to that of single-layer LGPS electrolyte. Relatively higher cycle numbers for the electrolytes prepared with Li₂S-P₂S₅ and Li_9.6P₃S₁₂ can be attributed to their dendrite-resistant glass-ceramic structures (higher Shear Modulus). In case of electrolytes prepared with Li₂S-P₂O₅, the cycle number is significantly higher than that of any other cells. Many studies reported that electrolytes with oxygen atoms in their structure showed remarkable cycle stability because oxygen ions are able to suppress the side reactions between sulfide electrolyte and lithium metal ^[27][28][27,28].

Another decisive factor for the commercialization of LGPS electrolytes is their price. Li₂S, P₂S₅, GeS₂ are the starting materials and the synthesis of LGPS solid electrolyte takes place between 500 and 600 °C in an inert atmosphere [31]. As germanium is an expensive element (1300 US $/kg) [32], the substitution of the critical element Ge has been achieved by isovalent elements such as Si or Sn (starting materials SiS₂ and SnS₂), which are much more affordable while only moderately affecting only moderately the conductivity performances [33]. A detailed price analysis is provided in Section 3.

As mentioned in the introduction, another interesting sulfide electrolyte is based on argyrodite-type compounds Li₆PS₅X (X = Cl, Br, I), shortly named as LPS-X solid electrolyte. Their preparation generally includes mechanical milling followed by heating at 500–600 °C in order to obtain the argyrodite phase. The starting materials are Li₂S, P₂S_5, and LiX [34]. Deiseroth et al. presented a detailed study on the synthesis of different LPS-X electrolytes and examined their crystal structures [34]. The highest Li-ion mobility was observed for Li₆PS₅Br electrolyte ^[14][34][14,34]. On the other hand, the compounds prepared with LiI showed a conductivity of only 4.6 × 10⁻⁷ S cm⁻¹ [14]. Even though Li₆PS₅I has an almost identical lattice structure compared to its relatives, it has not been clearly understood so far where the difference in the ionic conductivity values comes from. Recently, Hanghofer et al. studied substitutional disorder effects of LPS-X electrolytes by broadband impedance spectroscopy and ⁷Li NMR relaxation measurements [35]. They found that the anion disorder in LPS-Cl and LPS-Br supports faster Li-ion transport. Nazar et al. also showed that the same kind of phenomena can be observed in mixed halide argyrodites LPS-X (X = Cl_0.75Br_0.25, Cl_0.5Br_0.5, and Cl_0.25Br_0.75) [36]. For example, the electrolytes containing Cl_0.75Br_0.25 and Cl_0.5Br_0.5 showed an ionic conductivity of 3.2 × 10⁻³ and 3.9 × 10⁻³ S cm⁻¹, respectively, where this value was obtained as 3.4 × 10³ S cm⁻¹ for an electrolyte containing Cl_0.25Br_0.75. This behavior was explained by the low concentrations of Br, which resulted in more conductive pathways in the material structure. In contrast, excessive concentrations effectively blocked these channels and reduced the degree of percolation and ionic conductivity. Another study, which was performed by Viallet et. al, also showed that the conductivity of LPS-Cl electrolyte can be increased by optimization of the milling time [37]. They increased the ionic conductivity value from 2 × 10⁻⁴ S cm⁻¹ to 1.33 × 10⁻³ S cm⁻¹ by increasing the milling time from 1 to 10 h.

The cells prepared with a LGPS solid electrolyte show the same dendrite problems as cells prepared with argyrodite-type electrolytes. Although the LPS-Cl compound shows good stability to the phenomenon of lithium dissolution and deposition, it has been shown that contact losses at the interface with lithium occur when high currents are applied. High currents result in the formation of dendrites and the appearance of other degradation products (Li₂S, P₂S_x and polysulfides) ^[38][39][38,39]. One of the solutions to overcome this problem is the use of LiCoO₂/L_i4Ti₅O₁₂ cells [40]. Nevertheless, a breakthrough battery design has been presented by researchers from the Samsung Advanced Institute of Technology (SAIT) and the Samsung R&D Institute Japan (SRJ) [41]. They used a silver-carbon (Ag-C) composite layer at the anode, and there was no need to handle the Li metal sheet during the cell assembly process. The cathode acted as the sole source of the lithium and while charging, Li metal deposited on the stainless steel current collector. These cells showed an energy density greater than 900 Wh l⁻¹ and long cycle life (1000 times). However, taking into account that silver is an expensive element and the change of Ag-distribution in the Ag-C nanocomposite layer (after 100 cycles) make the commercialization of this cell difficult. The cost/performance ratio continues to remain a challenge.