Aluminum and lithium sulfur batteries: a review of recent progress and future directions

A Li–S battery consists of two main components, such as an anode (made of Li) and cathode (made of S) electrodes, which are separated by a separator and conductive electrolyte that can be either solid or liquid (for schematic illustration see figure 2(a)).

Zoom In Zoom Out Reset image size

Sulfur is in the form of polyatomic molecules with various structures. The octasulfur with chemical formula S₈ is an inorganic chemical and the most stable allotrope of sulfur at room temperature. During the charge–discharge process of the Li–S batteries, insoluble S₈ undergoes a morphological and structural change and turns to insoluble lithium polysulfides such as Li₂S₂/Li₂S and soluble Li₂S_x (8 ⩽ x ⩽ 3) in a liquid electrolyte. However, these dissolved polysulfides make cycling travel between the anode and cathode while the charging process is in turn. The main problem of this shuttling is the polysulfides also involve the reoxidation reactions at the cathode and side reduction reactions with a lithium anode, which results in a poor cycle life that low battery efficiency has been observed. To deal with this problem using an ion-selective membrane can be a method to get rid of polysulfides shuttling for improving the electrochemical performance of the battery. Recently, Wang et al reported that using Li₂S with a dual-phase electrolyte can hinder polysulfide migration and protect the lithium metal from direct contact with the electrolyte [49].

One of the main reasons of nonlinear charging/discharging of the sulfur-based batteries is the poor volumetric packing of sulfur, which gives rise to quick degradation of the battery arising from the induced mechanical stress. To solve this problem, researchers recommend to develop carbon based cathode composites and improving the conductivity. For this, increasing the sulfur loading of the cathode can be a solution. Manthiram et al, used carbon-cotton cathode in order to demonstrate stable Li–S batteries [50]. Carbon-cotton has many micro-/macroporous which allow to load high amount of active material due to having high surface area. Therefore, the carbon-cotton cathode has been demonstrated to exhibit enhanced cycle stability and improved cell-storage ability and even a high static capacity [50]. Recent studies showed that the effectiveness of the cathode can be related to the sulfur molecule size. For example, Xin et al has proved that using small sulfur allotropes such as S₂, S₃ and S₄, avoided the polysulfide formation [51]. In addition, it was reported that LiS batteries can be important potential candidates with their high specific capacities, efficient cycling stabilities, and superior rate capabilities.

There are several studies for carbon based Li–S battery applications. Researchers aim to increase initial discharge capacity of the LiS batteries and also their cycling performance. Zhang et al, reported the synthesis of a new amorphous tantalum oxide, in which oxygen vacancies are embedded inside a microporous carbon matrix, as an electrocatalyst for the LiS batteries [52]. The oxygen vacancy in the tantalum oxide (Ta₂O_5−x ) changes the coordination number and electron band structure which increases the intrinsic conductivity and function as catalytic centers to accelerate LPS conversion. With this created oxygen vacancies and used microporous carbon matrix excellent Li–S performance such as long-term cyclability over 1000 cycles with an ultra-low capacity fading rate of 0.0295% per cycle has been obtained [52].

Zhan et al used carbon nanotube bucky paper and they deposited sulfur and Co_x S electrocatalyst on it by a facile electrodeposition method to obtain a binder-free cathode material for LiS batteries [53]. Remarkably, this bucky paper gave high initial discharge capacity of 1650 mAh g⁻¹ at 0.1 C. And for at current rate of 0.5 C, this synthesized cathode material has a capacity of 1420 mAh g⁻¹ for the first cycle and reduces to 715 mAh g⁻¹ after 500 cycles.

Gueon et al, showed the widely used sulfur/CS₂ solution does not easily penetrate the hollow porous carbon sphere, due to the high interfacial energy of CS₂. However, using a mixed solution of N-methyl-2-pyrrolidone (NMP) or isopropyl alcohol (IPA) and CS₂ significantly improves the infiltration of the solution into the pores and resulting in complete sulfur encapsulation [54]. In addition, they found the control of sulfur loading such as the location of loading greatly affects the rate and reversibility of the sulfur transformation reaction and so the performance of the LiS battery.

Recently, Kim et al, prepared tungsten oxide/zirconia (WO₃/ZrO₂) and incorporated into a 2D compacted Li₂S-graphene matrix through pelletization. By this way, direct conversion of Li₂S to S₈ observed due to high electrical conductivity of graphene and by the way tungsten oxide/zirconia particles behave as a polysulfide mediator to avoid the outflow of soluble polysulfide into the electrolyte solution [55].

Ultrathin polycrystalline bismuth nanosheets which have approximately of 4 nm thickness is developed as an electrocatalyst for polysulfide conversion by Xu et al [56]. This ultrathin bismuth LiS battery shows high reversible capacity of 408 mAh g⁻¹ after 500 cycles at 10 C.

Successfully synthesized carbon fibers/Ti₃C₂T_x can be another potential candidate for commercial LiS battery devices. The interlayer of this material has three dimensional cross-linked conductive networks providing fast electron transfer channels and, thus, increase sulfur usage and save 'dead' sulfur. Jin et al, obtained high initial discharge capacity of 1380 mAh g⁻¹ at a current density of 0.1 C and also carbon fibers/Ti₃C₂T_x has very low capacity loss of 0.044% per cycle within 1000 cycles at 1 C [57]. In addition, it presents a low self-discharge property which has proved by open circuit voltage (OCV) values. Jin et al presents that OCV retention is maintained as high as 97.8% after 200 h.

Wang et al demonstrated titanium included LiS battery using the following process [58]; as illustrated in figure 3(a), firstly the sulfonated hollow polystyrene spheres (SPS) were prepared and then by using sol–gel method SPS was covered with amorphous TiO₂ layers. Following the annealing in a NH₃ atmosphere, the formation of carbon mesoporous TiN dual-shell nanospheres (CTiN) were observed. Finally, sulfur melt diffusion method was used in order to obtain CTiN–S composites. It was mentioned in the report that CTiN dual-shell hollow nanospheres can improve the chemical adsorption, the physical confinement, and catalysis for sulfur species conversion. It was also reported that CTiN–S electrodes give rise to a high capacity of 820 mAh g⁻¹ over 150 cycles at 0.2 C.

Zoom In Zoom Out Reset image size

Two-dimensional MXenes materials (see figure 3(b)) have high conductivity and their layered structure reduces the surface volume ratio which is good for adsorption for adatom or molecule, and they have rich surface terminations. These properties make them ideal material for confining the polysulfide shuttling effects and allow one to use them for modifying separators and enhancing the electrochemical performance of LiS batteries. Song et al, improved the performance of LiS batteries by coating Ti₃C₂T_x MXene nanosheets (T is for O, OH, F atoms which are used for surface termination) on commercial 'Celgard' membrane. They reported that after coating the LiS battery with Ti₃C₂T_x MXene electrical conductivity is increased, so Li–S batteries including MXene-functionalized separators exhibit high discharge capacity of 1046.9 mAh g⁻¹ at a low rate of 0.2 C. However, increasing of current density does not decrease the discharge capacity too much and exhibit discharge capacity of 743.7 mAh g⁻¹ at a low rate of 1 C [60]. Other type of MXenes such as Mo₂CT_x or W₂C are also used to improve LiS battery performance.

The electrode of Mo₂CT_x on carbon nanotube exhibits good electrochemical performances with regards to their high capacity (1314 mAh g⁻¹), efficient rate capability and high initial reversible capacity for at 1.8 mg cm⁻² sulfur loading and the capacity decreases to 959 mAh g⁻¹ at 5.6 mg cm⁻² sulfur loading [61]. Similarly, W₂C clusters on the nitrogen–phosphorous co-doped carbon matrix (W₂C/N/P-rGO) delivered high reversible capacities under different sulfur loadings. For example, discharge capacity of W₂C/N/P-rGO at 3.1 mg cm⁻² sulfur loading changes from ∼1200 mAh g⁻¹ to ∼900 mAh g⁻¹ with the increasing of cycling [62]. Detailed investigation and review about the two-dimensional MXenes for LiS batteries has been done by Zhang et al [59].

The research on energy storage is mainly based on constructing abundant material-based batteries of low-cost exhibiting high energy density. Thus, recent research efforts have been focused to search for alternative materials such as Na, K, Ca, AL, and Mg for the construction of batteries. Among the abundant materials, aluminum is one of the most abundant element apart from oxygen and silicon and it has been used as an anode material in alkaline aqueous batteries for more than six decades [63]. Al metal has 3+ valence electrons that it can exchange three electrons during the electrochemical process in a battery system, which can result approximately in a higher volumetric capacity (8046 mAh cm⁻³ which is three times larger than that of lithium battery systems, 2062 mAh cm⁻³) [64]. Al is relatively stable in the atmospheric environment and easy to process, so it can improve the safety of chemical battery system and reduce the process cost.

To fulfill its potential, aluminum battery systems are creditable for accomplishing of either plating or stripping Al in ionic liquid electrolytes. Before the introduction of sulfur as a cathode in Al-based batteries, graphene, metal-oxides, and conductive polymers were used as cathode materials in rechargeable batteries. Besides a geographically well-distributed and streamlined production, sulfur with a high predicted specific capacity (1675 mAh g⁻¹) [65], and volumetric capacity (3340 mAh cm⁻³) over a two-electron reduction process per atom and low cost has become an important research object of cathode materials for secondary batteries [29, 32, 39, 41]. The charge storage of S through a conversion reaction avoids the intercalation of bulk chloroaluminate ions, which are present in Al battery non-aqueous electrolytes. As a result, coupling of an Al anode with an S cathode in a battery provides a significant predicted energy density of 1371 Wh kg⁻¹.

The schematic illustration of the aluminum–sulfur battery is shown in the left part of the figure 4(a), the mechanism of Al–S battery is based on conversion mechanism of S to Al₂S₃-like reduction process in series of polysulfide intermediate species. The reaction mechanism, during discharge/charge (see figure 4(b)), can be formulated using the equations on the anode (equation (1)), and the cathode (equation (2)) [66, 67].

$$\begin{equation}\mathrm{2}\mathrm{A}\mathrm{l}+{\mathrm{14}\mathrm{A}\mathrm{l}\mathrm{C}\mathrm{l}}_{4}^{-}\to {\mathrm{8}\mathrm{A}\mathrm{l}}_{\mathrm{2}}{\mathrm{C}\mathrm{l}}_{7}^{-}+6{\mathrm{e}}^{-}\end{equation} \tag{ 1 }$$

$$\begin{equation}{\mathrm{8}\mathrm{A}\mathrm{l}}_{\mathrm{2}}{{\mathrm{C}\mathrm{l}}_{7}^{-}+6{\mathrm{e}}^{-}+3\mathrm{S}\to \mathrm{A}\mathrm{l}}_{\mathrm{2}}{\mathrm{S}}_{3}+14{\mathrm{A}\mathrm{l}\mathrm{C}\mathrm{l}}_{4}^{-}.\end{equation} \tag{ 2 }$$

Apparently, the weak interaction of Al and S results in a high kinetic barrier and requires chloroaluminate ion bond breaking in order to let the aluminum ion to be free [68]. The right part of the figure 4(a) shows the charge–discharge curve of the aluminum–sulfur secondary battery. To prepare this battery, Wang et al [69] used nitrogen-doped three-dimensional hierarchical porous carbon/sulfur (N–C/S) composite as a positive electrode (cathode). According to their results, the performance of the aluminum sulfur battery which is based on N–C/S composite cathode material is very good. For instance, the first-lap discharge specific capacity is 1800 mAh g⁻¹ and became almost half of this value after 10th cycles. However, this reducing becomes slow with the increasing of the cycles and as can be seen there is very small difference for the specific capacity values between the 15th and 20th cycles. At the same time, the specific capacity is still higher than 700 mAh g⁻¹ after 20 cycles, which is high enough.

Zoom In Zoom Out Reset image size

In figure 4(c), illustration of the assembled pouch cell-type AlS batteries is shown where BN was used to support S cathode material [71]. Zhang et al, showed that two-dimensional materials such as BN, MoS₂, and WS₂ can fix S and sulfide compounds as the repeated charge/discharge processes proceeds. Note that, among the considered 2D materials, BN was demonstrated to display the highest capacity of 532 mAh g⁻¹.

The first rechargeable Al–S battery that comprises EMIC-AlCl₃ (1-ethyl-3-methylimidazolium chloride) ionic liquid electrolyte exhibited a high discharge capacity of 1500 mAh g⁻¹ [30]. The demonstrated cell was shown to operate for only 20 cycles, therefore, its capacity was revealed to rapidly decay as result of S low electronic conductivity (5 × 10⁻²⁸ Sm⁻¹), which also results in low kinetics of the battery electrochemical reaction. Following, a rechargeable Al–S battery in an ionic-liquid electrolyte, in which an activated carbon cloth/sulfur was used as composite cathode, was successfully realized [67]. However, difficulties in the oxidation stage of AlS_x led to poor reversibility in Al/S chemistry. By confining the S species inside the microporous carbon pores of size less than 2 nm, it was shown to improve the electron access of S, prevent oxidation, increase the size of interfacial reaction area, and even reduce the diffusion length of Al³⁺ ions. Such process was reported to result in solid-state conversion reaction of S. By considering NBMPBr/AlCl₃ (N-butyl-N-methylpiperidinium bromide) ionic liquid electrolyte, due to the lower dissociation of Al₂Cl₆Br⁻ anions as compared to that of Al₃ ${\text{Cl}}_{7}^{-}$, the weak kinetics of Al–S batteries can be improved [68]. However, even this approach could not get over the low cycling problem of the battery (see figure 4(d)). Very recently, composite structures of sulfur and single-walled carbon nanotubes having small diameter were constructed as a cathode building block for AlCl₃: [EMIM]-based aluminum batteries in order to solve the poor stability and sluggish charge-storage problems [40]. Wu et al demonstrated rechargeable Al–selenium batteries (ASeBs) by using an eutectic solvent, thiourea-AlCl₃, as an electrolyte. In their experiment, by a lowtemperature process, Se nanowires were grown directly on a flexible carbon cloth substrate (Se NWs@CC) and used as a cathode. The high electronic conductivity and low ionization potential of selenium (Se) was considered as its advantages for the improvement of the battery kinetics on electrochemical reaction and on the polarization reduction. In the final product of Se NWs@CC cathode, it was observed that the build material possesses a specific capacity of 260 mAh g⁻¹ at 50 mAg⁻¹, a sufficiently long cycling life (100 cycles) as well as a high Coulombic efficiency 93% at 100 mAg⁻¹ [72].

Due to their low cost, predicted high capacity and energy density [73–75], Li–S batteries have gained great interest in various applications such as efficient energy storage technologies including portable electronic devices, hybrid/full electric vehicles, and large scale power grids [23, 76–78]. Several major technical obstacles, including dissolution of lithium polysulfides into liquid electrolyte, volume expansion of sulfur, and deposition of lithium sulfide, have been identified. For the past decade, investigation of an ideal cathode material for Li–S battery construction and enhancement of Li–S batteries for high-rate performance have become an important aim of the researchers. Accordingly, various nanostructured cathode materials have been developed including nanoporous carbon–sulfur composites, graphene–sulfur composites, one dimensional (1D) carbon–sulfur composites, conductive polymer–sulfur composites, porous oxide additives, and nanostructured Li₂S cathodes [25]. These nanostructured sulfur cathodes have higher resistance against pulverization and exhibit quite fast reaction kinetics and a good trapping of soluble polysulfides. It has been reported that they have cycling capacity in the range of 568–1000 mAh g⁻¹ for about 100–150 cycles of charge–discharge [79–84].

Li₂S₈, Li₂S₄, Li₂S₂, and Li₂S that intrinsically have intermediate phase are significant products for the charge–discharge processes of Li–S batteries. Among these structures, Pan and Guan [85] investigated the four Li₂S₂ species with similar structural configurations (two different hexagonal structures with space groups P63/mmc and $P6\bar{2}m$, two different orthorhombic structures with space groups Cmca, and Immm). They have obtained the circuit voltages of orthorhombic (Immm), orthorhombic (Cmca), and hexagonal (P63/mmc) as 3.88, 3.95, and 3.91 respectively.

Owing to their high surface to volume ratio and novel electronic properties, 2D materials present great promise as host materials to anchor lithium polysulfides (Li₂S_x ). Zhao et al explore the adsorption and diffusion of various Li₂S_x (x = 1, 2, 3, 4, 6, 8) species on a phosphorene monolayer [86]. It was investigated that the diffusion mobility of Li₂S_x species having long chains (Li₂S₃, Li₂S₄, Li₂S₆, and Li₂S₈) is orientation-dependent and is much faster along the zigzag orientation as compared to armchair one. Among the Li₂S_x species, Li₂S₆ exhibit the smallest binding energy on phosphorene surface, a promising material for high-efficiency Li–S battery cathodes. In addition, Li et al [87] studied the effect of crystal structures of phosphorus polymorphs by considering different phosphorus allotropes (α-P, β-P, γ-P, δ-P, -P, ζ-P, θ-P, and η-P). Their results showed that the crystal structures of phosphorus allotropes are very important for LiS batteries. Except for the -P, the remaining all other phosphorus allotropes can be suitable for the separations and modifications in Li–S batteries. In addition, α-P, β-P, γ-P, δ-P allotropes can be more convenient substrates for Li atom and Li–S diffusion in the sense that they result in lower barriers (please see figures 5(a)–(d)).

Zoom In Zoom Out Reset image size

It has been reported in various studies that MXenes present a strong anchoring effect for soluble lithium polysulfide due to their smaller lattice constants and superior electrochemical performance. By combined experimental and theoretical studies, OH-functionalized MXene material and its defected species are explored as advantageous for entrapment of S in Li–S batteries [89]. Right after this study, Rao et al [90] obtained the interactions of lithium polysulfides (Li₂S_x ) on Ti-based bare MXenes (Ti_n X_(n−1); X = C, N; n = 2, 3, 4) and functionalized Ti₂C with -F, -O, and -OH groups (Ti₂C(OH)₂, Ti₂CO₂, Ti₂CF₂). It has been reported that the strong bonding between Ti–S and also H–S dominate the interactions between MXenes and Li₂S_x . It has also been investigated that the bare and OH-functionalized MXenes exhibits great electronic conductivity, which makes them feasible cathode materials to use in Li–S batteries with high performance. In addition to that, Lin et al [91] pointed out that O- and F-functionalized Ti₂N present metallic character, which is advantageous for the redox reaction of Li₂S_x species and the functionalized Ti₂N possess good electrical conductivity that results in high rate performance of Li–S batteries. Afterwards, Li and et al [92] studied the binding energies of lithium (poly)sulfides on Ti₂CO₂, and M₃C₂O₂ (M = Ti, V, Cr, Zr, Nb, Hf) MXenes, and also the behaviors of O-functionalized MXenes. Among these, Cr₃C₂O₂ are found to exhibit the highest binding energy and the strongest anchoring effect. It was also reported that the lattice parameters of M₃C₂O₂ MXenes and the binding energies are monotonically correlated. More explicitly, the binding energy value becomes higher as the lattice constant is decreased, and this results in an increase in anchoring effect towards lithium polysulfide species. Most recently, Wang et al [93] reported the adsorption energy of Li₂S_n (n = 1–8) species on VS₂ in the range of 1.13 eV–4.44 eV and the diffusion barrier energy of Li atom on VS₂ as 0.25 eV, which implies the fast diffusion process.

Furthermore, black-phosphorene-like MX (M = Ge, Sn; X = S, Se) monolayers (group IV monochalcogenides) and graphene@MX heterostructures have been studied by Lv et al [88] as a functional host for Li–S batteries (please see figure 5(b)). They reported that MX-based heterostructures have enhanced electrical conductivity, great anchoring performance and fast charge/discharge rate for Li–S batteries. It has also been revealed that graphene@GeS heterostructure is a highly promising anchoring material for high-energy Li–S batteries. Chen et al systematically investigated the first-row transition metal sulfides (from Sc to Zn) as polar hosts for sulfur cathodes in LiS batteries to understand the general rationale for the host design of sulfur [95]. Their predictions revealed that diffusion barrier energy for lithium ion on vanadium sulfide (VS) is very low and VS has the strongest anchoring effects on Li₂S immobilization among the other considered metal sulfides.

Yin et al [96] systematically explored the interaction mechanisms between N-doped graphene and lithium polysulfides by considering various doping configurations, including different types of N-dopants such as graphitic, pyridinic, pyrrolic and isolated/clustered amino. They unveiled that cluster of pyridinic N-dopants have stronger binding energies (1.23–3.57 eV) than electrolyte solvent molecules (DME and DOL) (0.87–0.98 eV) while isolated amino, graphitic, pyrrolic, or pyridinic N-dopant possess smaller or comparable binding energies (0.56–1.18 eV). Also, they reported that all the isolated N-dopants, except for graphitic N, show strong binding with Li polysulfides species comparing to piristine-G. An other nitrogen doped graphene is used by Yi et al [97], to investigate the adsorption energy of lithium polysulfides. They firstly adsorbed a Li atom on N-doped graphene and called this new structure as Li-trapped N-doped graphene and their results showed that the calculated binding energies of the lithium polysulfides on Li–N-graphene are in a range from 1.13 to 1.40 eV (please see figure 6(a)). The high binding energy which is more than 2.0 eV, is known to cause decomposition of the lithium polysulfides species. So, these obtained binding energies are lower than this critical point. Li₂S_n (n = 1, 2, 4, 6, 8) lithium sulfides adsorbed on ideal graphene (IG) and N-doped graphenes which are N-substituted graphene (NG), pyridinic-like graphene (PIG), and pyrrolic-like graphene (PRG) by Rao et al [98]. Their results showed that the binding energies for Li₂S_n changes from 0.94 eV to 3.17 eV and only IG + Li₂S_n and NG + Li₂S_n structures have lower binding energy values than 1.50 eV, all the other structures (PIG + Li₂S_n and PRG + Li₂S_n ) have higher than 2.50 eV binding energy. Single vacancy, divacancy or Stone–Wales (SW) type defected graphenes also investigated to understand can these structures be suitable for LiS battery applications [99, 100]. DFT calculations revealed that carbon vacancy site catches one S atom from the Li₂S_n leading to the formation of a Li₂S_n−1 molecule, which is weakly adsorbed on the created S-doped graphene. However, Li₂S_n interaction with bare and SW defected graphene is almost similar [100].

Zoom In Zoom Out Reset image size

The most favorable research strategy to find a suitable material for LiS battery is using graphene and improving its performance by external effects. Gong et al, found that heteroatom-modified graphene, h-G, (where h = B, N, O, P, or S) is effective on anchoring lithium polysulfides and on weakening the shuttle effect [105]. Wasalathilake et al, performed DFT calculations to investigate the interactions between the S₈ molecule and long chain lithium polysulfides (Li₂S₄ and Li₂S₈) and graphene and graphene with decorated hydroxyl, epoxy and carboxyl groups [106]. Their calculations revealed that microporous graphene which is decorated with oxygen containing functional groups can successfully anchor S₈, Li₂S₈ and Li₂S₄ moderate inter molecules through actions and this leads to the improvement of charge transfer and conductivity in the cathode side of Li–S batteries. Several studies showed that metal-N₄ doped graphene sheets can be an efficient anchoring material for Li–S batteries. For this graphene sheet, the researchers firstly created porphyrin-like graphene by extracting one C–C bond and replacing the four C atoms facing to the divacancy site with four N atoms. At the end one metal atom is adding in the middle of N atoms. According to Zhang et al [101], DFT calculations for Li₂S_n (n = 1, 2, 4, 6, 8) on the MN₄@graphene (M = Cu, Ni, Co, Fe, Mn, Cr, V) showed that the binding energy values of Li₂S_n on FeN₄@graphene and CrN₄@graphene is in an suitable interval (please see figure 6(b)) which indicates that these materials can be an anchor material. While the binding energies of Li₂S_n for other co-doped systems such as MnN₄@-graphene, CoN₄@graphene, CuN₄@graphene, and NiN₄@graphene are less than 1 eV for the long chain lithium polysulfides which means these four MN₄@graphene might not be an appropriate anchor material for the Li–S battery. Similar results (MN₄@graphene + Li₂S_n ) are obtained by other researchers [107, 108].

Theoretical studies were also focused on many carbon nitride or non-metallic monolayer materials to understand the interaction process of lithium polysulfides with these materials (please see figures 6(c)–(f)). Li et al studied the C₄N₄ monolayer and C₄N₄/graphyne heterostructure for LiS battery investigation. Theoretical calculations revealed that the adsorption energies of polysulfides on C₄N₄ in the range of 1.931 to 3.119 eV (please see figure 6(c)) and they showed that electrical conductivity of C₄N₄ can improve by creating C₄N₄/graphyne heterostructure [102]. He et al performed DFT calculations and obtained that oxygenated g-C₃N₄ give rise to a better electrical conductivity and adsorption energy of the Li₂S_n species are higher than the pristine g-C₃N₄. Furthermore, they also confirmed these results experimentally and found that the O-g-C₃N₄ composite cathode possesses great electrochemical performance leading to a high reversible discharge capacity of 1030 mAh g⁻¹ after 100 cycles at 0.2 C in Li–S batteries [103].

A comprehensive study is done by Zheng et al to understand the anchoring effect of non-metallic monolayers for lithium polysulfides in LiS batteries (please see figure 6(f)) [104]. They considered graphene, C₂N, C₃N₄ and BN monolayers and found that C₂N and C₃N₄ can provide suitable adsorption energies and enhance the redox kinetics. Other carbon included monolayers such as BC₂N [109], C₃B and C₃N [110] are also investigated and found that except of C₃N other two materials exhibit an excellent performance to anchor the lithium polysulfides cause of the strong chemical interaction (table 1).

Not only the limited number of experimental studies available, but also there were only a few reports describing theoretical studies on nonaqueous Al–S batteries [112]. The main problem could be the multivalence number of this metal ion. These types of multivalence metal ions such as Al, Ca, Mg are incompatible when used in LiS batteries with the typical organic liquid electrolytes. Therefore, the important thing for this kind of multivalence AlS batteries is finding proper electrolytes to effectively transport the aluminum ions and attain reversible plating/stripping of the anode metal species. Moreover, studies showed that the cathode volume expansion of AlS batteries after full discharging is too high (up to 272%) [34]. This exhibits that the designing of the cathode structure needs special attention. Here, theoretical studies will play a key role in order to understand complex electrochemical reactions through charge and discharge processes or finding suitable electrodes for AlS batteries. The characterization of surface and reduction mechanism can highlight by molecular dynamics (MD) simulation beyond DFT calculations in Al–S batteries. There are very few theoretical studies reported as of now. Very recently, Bhaurital et al observed a detailed charging and discharging processes in Al–S batteries by analysis of interfacial systems S₈(001)/[EMIM]AlClS₄ and Al₂S₃(001)/[EMIM]AlClS₄-electrode, respectively [113] (please see figures 7(a)–(c)). The discharging process can occur through the continuous reduction of S to Al₂S₃-like products via a series of polysulfide intermedia species and involves the formation of various cationic and anionic intermediate species are observed. Hereby, they opened up a way to understand the complex electrochemical processes occurred in Al–S batteries. Chu et al showed that the energy barriers are key parameters for understanding the electrochemical reaction pathways on AlCl₃/acetamide for reversible room-temperature Al–S battery which potentially presents a promising possibility for low-cost and high-performance energy storage system in future [70]. They reported that an initial capacity above 1500 mAh g⁻¹ and good rate performance and [AlCl₂(acetamide)₂]⁺ ions led to an energetically favorable reaction pathway, which reveal the superior performance in comparison to the conventional AlCl₄-based electrolyte.

Zoom In Zoom Out Reset image size

As mentioned in AlS experimental section, Yang et al used Al₂Cl₆Br⁻ instead of Al₂ ${\text{Cl}}_{7}^{-}$ as the dissociation reaction reagent. Calculations based on DFT showed that the reaction rate of Al₂Cl₆Br⁻ dissociation is 15 times faster than that of Al₂ ${\text{Cl}}_{7}^{-}$, which is also verified by experimentally [68]. We believe that theoretical studies will be very important to improve the efficiency of Li/Al–S batteries.

Yu et al, performed ab initio molecular dynamics simulations based on DFT calculations for the amorphous Li₃AlS₃ structure [114]. The inset of figure 7(d) shows a snapshot of the amorphous Li₃AlS₃ structure obtained by the melt-quenching method while the plot for radial distribution functions between Al, Li and S atoms. As can be seen, the first peak position for Al–S pairs at 2.25 Å is close to the Al–S bond length (2.28) in crystalline Al₂S₃. The peak for Li–S pairs is located at 2.5, and it is very close to Li–S bond length in crystalline Li₂S. However, the first peak for S–S pairs indicates that there are no polysulfides in the amorphous Li₃AlS₃ structure.