A facile synthetic approach to nanostructured Li 2 S cathodes for rechargeable solid-state Li – S batteries †

Li – S solid state batteries, employing Li 2 S as a pre-lithiated cathode, present a promising low cost, high capacity and safer alternative to their liquid electrolyte counterparts, where dis-solution of intermediate polysul ﬁ de species can result in loss of active material and a subsequent decrease in ionic conductivity. A nanostructured Li 2 S material would a ﬀ ord greater ﬂ exibility in optimising the cathode composite for more harmonious electrode – electrolyte interactions, yet facile routes to such nanoscale materials are limited. Here, we report a facile and scalable microwave approach to directly synthesize nanostructured Li 2 S from a glyme solution containing lithium polysul ﬁ des. As-synthesized Li 2 S presents an ideal architecture for the construction of free-standing cathodes for all-solid-state Li – S batteries. Reversible electrochemical lithiation of S to Li 2 S rechargeable – S

A facile synthetic approach to nanostructured Li 2  Reversible electrochemical lithiation of S to Li 2 S in rechargeable Li-S batteries is a promising approach to realizing lowcost, high energy density batteries. 1 Abundant, light-weight sulfur offers a potential specific capacity of 1672 mA h g −1 which, at an order of magnitude higher than that provided by conventional intercalation-type cathodes such as LiCoO 2 and LiFePO 4 , makes Li-S batteries suitable for high energy storage applications such as powering electric vehicles. 1,2 Li 2 S has been suggested as a pre-lithiated cathode, negating the necessity of metallic lithium as an anode. 2,3 With the application of a solid-electrolyte, further advantages such as higher ionic conductivities and potentially mitigating deleterious polysulfide shuttle activity are possible. [4][5][6] However, a serious bottleneck in the widespread uptake of such a pre-lithiated material has been the limited synthetic routes to Li 2 S. Preferably, Li 2 S would be composited with an electronic conductor and a Liion conducting medium on the nanoscale to promote greater electrode-electrolyte interactions, in an architecture that would accommodate the volume changes associated with cycling. The application of nanostructured Li 2 S also Recent years have seen efforts made in developing novel Li 2 S composites, for example in combination with carbon and/or solid electrolyte. 3 A significant benefit to the community would be a reliable and scalable synthetic route to nanostructured Li 2 S, which up to now has been limited and in several cases complicated and costly. Reported routes include impregnating carbon materials with ethanolic solutions of commercial Li 2 S, 7-16 chemical reaction of sulfur with lithium triethylborohydride or stabilized lithium metal powder, 4,17-23 carbon-based reduction of Li 2 SO 4 and polysulfides, [24][25][26][27][28][29][30] reactions between H 2 S and lithium naphthalenide, 31 and burning Li foils in a CS 2 vapour. 32 Thus, there remains a lack of facile routes to Li 2 S which are scalable and afford control over the final particle morphology and size. Here we report the synthesis of nanostructured Li 2 S, prepared by a simple microwave approach, together with its performance in an all solid-state Li-S battery. Our results demonstrate that nanostructured Li 2 S employed within a free-standing cathode displays exceptional cycling stability, achieving capacities of 440 mA h g −1 at cycling rates of 100 µA cm −2 up to 400 cycles. Nanostructured Li 2 S is prepared using a simple 20-minute microwave-assisted heat treatment of a tetraglyme solution containing elemental sulfur and lithium tert-butoxide, at temperatures as low as 200°C (see ESI for experimental details †). X-ray diffraction patterns of this microwave-synthesised Li 2 S powder ( Fig. 1) reveal peak broadening consistent with a reduction in particle size (with an estimated crystallite size of 10 nm from Scherrer broadening), compared to a commercial Li 2 S sample (Sigma Aldrich, 99.98%). All peaks could be indexed to the cubic lattice of the antifluorite structure of Li 2 S. Nanoparticles of Li 2 S are confirmed by scanning electron microscopy (SEM) images, were aggregates of particles are observed. Fig. 2 illustrates the importance of the S/lithium tertbutoxide concentration ratio on the resulting particle size and distribution, with three examples shown (see Fig. S1 † for further SEM images). Our experiments reveal that Li 2 Sp r epared using 0.15 M sulfur solution gave particles with the smallest size and narrowest size distribution.
Such nanostructured material provides a promising electrode architecture in terms of providing shorter pathlengths for Li-ion transport, as well as larger electrode-electrolyte contact areas, compared with bulk materials and this sample was selected for further electrochemical study.
It has been previously observed that conductive thin layers of Li 3 PS 4 can be deposited from a THF solution containing Li 2 S and P 2 S 5 . 4,33 Here, we followed a modified approach to incorporate our microwave-synthesized nanostructured Li 2 Si n a three-phase Li 3 PS 4 /nano-Li 2 S/C nanocomposite. Firstly, nanostructured carbon black is composited with P 2 S 5 using a melt-diffusion technique (Fig. S2 †) before mixing with microwave-synthesized Li 2 S in THF (molar ratio of P 2 S 5 :Li 2 Si s 1 : 20) to produce the final Li 3 PS 4 /Li 2 S/C nanocomposite. This corresponds to 40 wt% Li 2 S in the composite. Characterisation of the final composite by SEM, EDX and XRD are shown in Fig. 3. EDX analysis reveals a homogeneous phosphorus and sulfur distribution across the material, while the presence of Li 3 PS 4 is not evident from XRD nor from SEM ( pure β-Li 3 PS 4 has a distinctive morphology of micron-sized, porous brickshaped particles). These data suggest the formation of a thin film of Li 3 PS 4 , which is either amorphous or is formed on a much shorter length scale not detectable by Bragg diffraction.  consistent with previous reports. 4 This is further confirmed by performing the control experiment, in which a P 2 S 5 -rich/C composite is mixed with Li 2 S (molar ratio of P 2 S 5 :Li 2 Si s 1 : 10). In addition to formation of bulk β-Li 3 PS 4 , which is confirmed by XRD and SEM (Fig. S3 †), a homogeneous distribution of P and S is also observed across this material indicating the formation again of a thin coating of Li 3 PS 4 . An all-solid-state battery was prepared using the Li 3 PS 4 / Li 2 S/C nanocomposite as a free-standing cathode, β-Li 3 PS 4 as the solid electrolyte and metallic lithium as the anode. These components were pressed together under 2 tons of pressure, employing Al and Ni foils to support and collect the current from the cathode and the anode, respectively (Fig. 4a, insert). The cell cycled at 60°C and at current densities ranging between 20 and 100 µA cm −2 for up to 400 cycles (corresponding to ∼ C/2 and C/10). Fig. 4a shows the typical charge and discharge profiles of the battery at 20 µA cm −2 . In addition to retaining the capacity, it is notable that the voltage profiles comprise well-defined discharge and charge plateaus, displaying small potential hysteresis. We attribute this small hysteresis to improved electronic and ionic conductivities of the composite cathode and a reduced interfacial resistance at the solid-electrolyte/cathode interface, as illustrated by impedance spectroscopy measurements (see below). Increasing the discharge/charge rate from 20 to 100 µA cm −2 reduced the capacity from ∼ 840 to ∼ 440 mA h g −1 , but this capacity is retained over more than 300 additional cycles with coulombic efficiency close to 100% (Fig. 4b). Fig. S4 † displays SEM images of the Li 3 PS 4 /Li 2 S/C electrode before and after cycling revealing no obvious change of the morphology after 410 cycles at different rates. Some surface cracks are observed which can be attributed to volume changes during cycling; importantly, these cracks do not lead to a loss of contact between the electrode domains or to delamination of the elec-trode and the solid electrolyte. Impedance measurements prior to and after cycling, indicated an increase of the resistance of the cell from ∼250 Ω to ∼600 Ω after 410 cycles at different rates (Fig. S5 †). This limited growth of cell resistance with long-term cycling is consistent with capacity retention and maintaining electrode integrity and suggests good stability of the solid-electrolyte/cathode interface. For comparison, in a recent study on Li 6 PS 5 Cl-Li 2 S all-solid-state batteries, a cycled battery employing Li 2 S/C as the cathode exhibited an interfacial resistance that is one order of magnitude higher than the resistance of the fresh cell after only 37 cycles. 34

Conclusions
In summary, we demonstrate that nanostructured Li 2 S can be prepared by a simple, fast, microwave heat treatment of a glyme solution containing lithium polysulfides. The as-synthesized Li 2 S presents an ideal architecture for the construction of free-standing cathodes comprising Li 3 PS 4 /Li 2 S/C nanocomposites. Crucially, this synthetic approach is highly reproducible, having been repeated numerous times, and scalable. The performance of this nanostructured Li 2 S as a cathode in  an all-solid-state battery demonstrates excellent capacity retention at current densities of up to 100 μAc m −2 , with limited increases in cell resistance and close to 100% coulombic efficiencies, which we attribute to increased electrode-electrolyte contact area and shorter Li + diffusion pathlengths.

Conflicts of interest
There are no conflicts to declare.