Monolithic heterojunction quasi-solid-state battery electrolytes based on thermodynamically immiscible dual phases

Sung-Ju Cho a, Gwan Yeong Jung a, Su Hwan Kim a, Minchul Jang b, Doo-Kyung Yang b, Sang Kyu Kwak *a and Sang-Young Lee *a
aDepartment of Energy Engineering, School of Energy and Chemical Engineering, Ulsan National Institute of Science and Technology (UNIST), Ulsan 44919, Korea. E-mail: skkwak@unist.ac.kr; syleek@unist.ac.kr; Tel: +82-52-217-2948
bFuture Technology Research Center, LG Chem., Seoul 07796, Korea

Received 23rd May 2018 , Accepted 17th September 2018

First published on 17th September 2018


Traditional single-phase electrolytes, which are widely used in current state-of-the-art rechargeable batteries, have difficulties simultaneously fulfilling different chemical/electrochemical requirements of anodes and cathodes. Here, we demonstrate a new class of monolithic heterojunction quasi-solid-state electrolytes (MH-QEs) based on thermodynamically immiscible dual phases. As a proof-of-concept of the MH-QEs, their application to lithium–sulfur batteries is explored. Driven by combined effects of structural uniqueness and thermodynamic immiscibility, the electrode-customized MH-QEs provide exceptional electrochemical performance that lies far beyond those accessible with conventional battery electrolytes.



Broader context

Ideal battery electrolytes should be electrochemically stable at both anodes and cathodes while ensuring facile ion transport between the electrodes. However, traditional single-phase electrolytes, which are widely used in current state-of-the-art rechargeable batteries, have difficulties simultaneously fulfilling the different requirements of anodes and cathodes. Here, we present a new class of monolithic heterojunction quasi-solid-state electrolyte (MH-QE) based on thermodynamically immiscible dual phases and explore its application to lithium–sulfur batteries as a proof-of-concept. A salient feature of the MH-QE is thermodynamic immiscibility between the two quasi-solid-state electrolyte (QE) layers that are respectively tailored for anodes/cathodes and seamlessly connected in series. Benefiting from combined effects of the monolithic heterojunction structure and thermodynamic immiscibility, the electrode-customized multifunctional MH-QEs provide exceptional cycling performance that lies beyond those accessible with conventional battery electrolytes.

The increasing demand for advanced rechargeable power sources, including high-performance lithium–ion batteries, lithium–sulfur, lithium–metal, metal–air, and flow batteries, compels us to search for new electrochemical materials and structures.1,2 In principle, anodes and cathodes of the battery systems respectively require different types of electrochemical reactions and operating environments (such as low/high potentials, aqueous/organic media and so on) to achieve reliable/sustainable cell performance.3–5

To this end, electrolytes should be chemically/electrochemically stable at both the anode and cathode while ensuring facile ionic transport between the electrodes. However, traditional single-phase electrolytes, which are widely used in current state-of-the-art rechargeable batteries, have considerable difficulties simultaneously fulfilling the different requirements of the anodes and cathodes described above. Among the various approaches to address these electrolyte issues, so-called hybrid electrolytes,3–6 which are composed of different types of electrolytes that are respectively suitable for anodes and cathodes, have recently garnered attention as an appealing electrolyte system. For example, in metal (e.g., lithium or sodium)–air batteries,3,6 air cathodes and metal anodes are assembled with aqueous and organic liquid electrolytes, respectively.

The hybrid electrolytes typically show the consecutively connected structure of two (liquid/solid, solid/solid) or three (solid/liquid/solid, liquid/solid/liquid) electrolytes. The most common solid electrolyte adopted in the hybrid electrolytes is a thick and stiff glass film based on lithium superionic conductors (LISICONs) or sodium superionic conductors (NASICONs).5 The use of the glass film in the hybrid electrolytes, however, often gives rise to unwanted increase of internal cell resistance and loss of specific energy density of cells. In addition, bulky cell dimension and complicated cell design/fabrication are difficult to avoid.

Here, we present a new class of monolithic heterojunction quasi-solid-state electrolytes (MH-QEs) based on thermodynamically immiscible dual phases as a new electrode-customized multifunctional electrolyte beyond conventional single-phase and hybrid electrolytes. As a proof-of-concept of the MH-QEs, their potential application to lithium–sulfur (Li–S) batteries is explored with particular attention to the shuttle effect. A key challenge facing Li–S batteries is preventing the shuttle effect (i.e., migration of dissolved polysulfides from sulfur cathodes toward lithium metal anodes) while allowing electrochemical conversion reactions in the sulfur cathodes.6,7 The two thermodynamically immiscible quasi-solid-state electrolyte (QE) layers are respectively tailored for anodes/cathodes and seamlessly connected in-series (i.e., monolithic heterojunction structure) by ultraviolet (UV) curing-assisted printing process, eventually leading to the formation of MH-QE. Each QE layer in the MH-QE consists of an ion-conducting medium (lithium bis(trifluoromethylsulphonyl) imide [LiTFSI] in polar solvents), alumina (Al2O3) nanoparticles as a mechanical spacer and UV-cured ethoxylated trimethylolpropane triacrylate (ETPTA) polymer as a flexible skeleton. Judicious control of ion–dipole interaction between Li+ ions and solvent molecules leads to change of solubility parameter (specifically, electrostatic force) of the electrolyte mixtures, resulting in the creation of the immiscible QE layers without concerns of electrolyte interdiffusion.

Ethyl methyl sulfone (EMS, dielectric constant (ε) = 95.0) and tetraethylene glycol dimethyl ether (TEGDME, ε = 7.7) were chosen as model solvents on the basis of their dielectric constant that affect the dissolution behavior of polysulfides.8,9 The two immiscible QE layers in the MH-QE work for sulfur cathodes and lithium metal anodes, respectively. The EMS-QE layer is oriented toward the sulfur cathode to facilitate sulfur utilization while the TEGDME-QE layer, which faces the lithium metal anode, impedes the migration of dissolved polysulfides (Li2Sx, x ≤ 8) and, thereby, suppresses the shuttle effect. Driven by the monolithic heterojunction structure and immiscibility, the electrode-customized multifunctional MH-QEs enable significant improvements in electrochemical performance which is difficult to achieve with conventional battery electrolytes.

Fig. 1a shows that the electrolyte mixture (1.0 M LiTFSI in EMS/2.3 M LiTFSI in TEGDME (denoted as EMS-EL/TEGDME-EL) = 5/5 (v/v)) is immiscible, whereas the other electrolyte mixture (1.0 M LiTFSI in EMS/1.0 M LiTFSI in TEGDME = 5/5 (v/v)) and solvent mixture (EMS/TEGDME = 5/5 (v/v)) are miscible. The 2.3 M LiTFSI in TEGDME, corresponding to LiTFSI/TEGDME = 1/1 (mol/mol), is known to form stable Li+-TEGDME complexes without free TEGDME molecules.10 This indicates the presence of strong ion–dipole interaction (contributing to increase of electrostatic force), which could play a viable role in the immiscibility of the resulting electrolyte mixture. This distinct immiscibility of the electrolyte mixture (EMS-EL/TEGDME-EL) was verified by 1H nuclear magnetic resonance (NMR) spectra (Fig. S1, ESI). The characteristic 1H NMR peaks assigned to EMS and TEGDME were clearly attributed to the top and bottom layers, respectively. Additionally, no mixed peaks were found in either layer. This NMR result demonstrates the complete immiscibility between the top and bottom layers in the electrolyte mixture (top layer = EMS-EL, bottom = TEGDME-EL).


image file: c8ee01503a-f1.tif
Fig. 1 Thermodynamic elucidation of immisciblity of electrolyte mixture. (a) Photographs of the solvent mixture (EMS/TEGDME = 5/5 (v/v)), electrolyte mixture (1.0 M LiTFSI in EMS/1.0 M LiTFSI in TEGDME = 5/5 (v/v)) and electrolyte mixture (1.0 M LiTFSI in EMS/2.3 M LiTFSI in TEGDME = 5/5 (v/v)), after storage for 24 h. (b) Calculated δ values of the solvents and electrolytes, which were deconvoluted into δvdW and δES. Δδ denotes the difference in the solubility parameter between the two components (i.e., Δδ (solvents) = δ(TEGDME) − δ(EMS), Δδ (electrolytes) = δ(TEGDME-EL) − δ(EMS-EL)). (c) MD simulation results: miscible phase of the solvent mixture (EMS/TEGDME, left image) and immiscible phase of the electrolyte mixture (EMS-EL/TEGDME-EL = 5/5 (v/v), right image). EMS, TEGDME, [TFSI], [Li(EMS)4]+, and [Li(TEGDME)]+ are shown in pink, purple, gray, red, and blue, respectively. (d) Time evolution of the χdemix and ΔEint of the solvent mixture and electrolyte mixtures with different composition ratios (EMS-EL/TEGDME-EL = 3/7, 5/5, and 7/3 (v/v), respectively).

The aforementioned immiscibility of the electrolyte mixture was theoretically elucidated using computational methods (see “Theoretical Calculations” in the ESI). First, we performed density functional theory (DFT) calculations to find the most stable molecular structures of LiTFSI, the solvents (EMS and TEGDME), and the electrolytes (EMS-EL and TEGDME-EL) (Fig. S2a and b, ESI).11,12 Based on the DFT calculation results, molecular dynamics (MD) simulations were conducted to estimate the theoretical miscibility of solvent and electrolyte mixtures in terms of the solubility parameter (δ)13,14 with a focus on vdW forces (δvdW) and electrostatic forces (δES) (Fig. S2c–f, see “Investigation of the miscibility based on calculated solubility parameters” in the ESI). The basic information of the model systems used for this theoretical calculation is provided in Table S1 (ESI).

Fig. 1b shows that EMS and TEGDME have similar δ values (Δδ (= δ(TEGDME) − δ(EMS)) ∼ 2.5 MPa1/2). Meanwhile, the addition of LiTFSI into the solvents led to a significant change in the δ values (Δδ = δ(TEGDME-EL) − δ(EMS-EL) ∼ 17.0 MPa1/2), which is mainly attributed to the large difference in the electrostatic force (ΔδES ∼ 19.5 MPa1/2) induced by ion–dipole interactions between Li+ ions and solvent molecules. Additionally, the Δδ values were also well matched with the previous classification of miscibility reported by Greenhalgh et al.15 (i.e., Δδ < 7.0 MPa1/2 for miscible compounds and Δδ > 10.0 MPa1/2 for immiscible compounds).

To better understand this unusual immiscibility behavior, we observed the structural evolution of preformed interfaces of solvent and electrolyte mixtures with different composition ratios during MD simulations (Fig. 1c, Fig. S3 and Table S2, ESI). Additionally, the time evolution of the phase separation was monitored in terms of the demixing index (χdemix)16 and interaction energies (ΔEint) among the components in the electrolyte mixtures (see “Structural analysis based on the demixing index and interaction energy” in the ESI). The solvent mixture showed a gradual increase in χdemix and a decrease in ΔEint (Fig. 1d), indicating good miscibility between EMS and TEGDME. This result was verified by MD simulation snapshots and concentration profiles of the solvent mixture as a function of the simulation time (Fig. S4, ESI). In addition, interfacial tensions (γ)17 of the solvent and electrolyte mixtures were theoretically examined (Table S3, see “Interfacial tension calculation” in the ESI). The electrolyte mixtures showed higher γ values than the solvent mixture, which is additional evidence proving the well-developed phase separation in the electrolyte mixtures. Notably, the electrolyte mixture of EMS-EL/TEGDME-EL = 5/5 (v/v) exhibited the highest γ value among the electrolyte mixtures, which is consistent with the aforementioned results of χdemix and ΔEint.

The dissolution of the polysulfides in the electrolytes (EMS-EL and TEGDME-EL) was examined. Fig. 2a shows the dissolution behavior of the polysulfide intermediate (0.1 M Li2S8) in the EMS-EL and TEGDME-EL as a function of time. In addition to the color change of the electrolyte solutions, their UV-visible spectra were monitored to quantitatively analyze the polysulfide dissolution. The EMS-EL dissolved the polysulfide within 1 h, mainly because of its high dielectric constant, whereas limited solubility (i.e., longer than 5 days) was observed in the TEGDME-EL. The charge/discharge performance of the Li–S cells containing the EMS-EL and TEGDME-EL was investigated, respectively (Fig. 2b and Fig. S5, ESI). The EMS-EL showed higher initial capacities than the TEGDME-EL but poor cycling retention, revealing a serious shuttle effect. Meanwhile, the limited polysulfide solubility of the TEGDME-EL resulted in the low capacities although the stable cycling retention was attained.


image file: c8ee01503a-f2.tif
Fig. 2 Fabrication and structural/electrochemical characterization of MH-QEs. (a) Dissolution behavior of 0.1 M Li2S8 in the EMS-EL and TEGDME-EL as a function of time. (left) Photographs and (right) UV-visible spectra. (b) Cycling performance (at a charge/discharge current density = 0.1 C/0.1 C) of the Li–S cells containing the EMS-EL and TEGDME-EL. (c) Schematic illustration depicting the fabrication of the MH-QE through the UV curing-assisted printing process. (d) Cross-sectional SEM image of the MH-QE. (e) Temperature-dependent ionic conductivity of the MH-QE and control QE (that was fabricated by stacking the EMS-QE film with the TEGDME-QE film).

These results inspired us to develop a new electrolyte concept (i.e., MH-QEs). To effectively utilize the electrochemical features of the EMS-EL and TEGDME-EL the electrolytes should exist separately and be connected in series to work for sulfur cathodes and lithium anodes, respectively. To achieve this goal, ETPTA polymer networks were incorporated in the ELs as a flexible mechanical skeleton, which led to the fabrication of QEs.18 Notably, to ensure a seamless connection (i.e., monolithic heterojunction structure) while minimizing the interfacial resistance between the EMS-QE and TEGDME-QE layers,19 we exploited the UV curing-assisted printing technique,20,21 which is schematically illustrated in Fig. 2c. During the printing process, the EMS-based QE precursor can infiltrate into interstitial voids of the sulfur cathode, along with formation of a cast layer. Subsequently, the ETPTA monomers in the EMS-based QE precursor were crosslinked to generate the ETPTA polymer networks after exposure to the UV curing, yielding the solidified EMS-QEs. Details regarding the fabrication of the MH-QEs are described in the experimental section. The formation of the UV-cured ETPTA polymer skeleton was verified by observing the change in the characteristic Fourier transform infrared spectroscopy (FT-IR) peaks attributed to acrylic C[double bond, length as m-dash]C bonds (Fig. S6, ESI).21 A cross-sectional scanning electron microscopy (SEM) image of the MH-QE revealed that the EMS-QE and TEGDME-QE layers were seamlessly connected in series on the sulfur cathode (Fig. 2d).

Facilitation of conversion reactions between sulfur active materials and polysulfides are required to ensure excellent cell performance. However, dissolved polysulfides often cause the unwanted shuttle effect.22,23 In this study, to maximize the electrochemical functions of the EMS-QE (i.e., to facilitate sulfur utilization) and TEGDME-QE (to suppress the shuttle effect) in the MH-QE, the EMS-QE and TEGDME-QE layers were oriented toward the sulfur cathode and lithium anode, respectively. Additionally, the TEGDME-QE showed reliable electrochemical stability toward the lithium metal anodes (Fig. S7, ESI). To highlight the advantageous effects of the MH-QE, a control QE was fabricated by simply stacking the EMS-QE thin film with the TEGDME-QE film (Fig. S8a, ESI). Compared with the MH-QE, interfacial cracks between the two QE films were clearly observed in the control QE (Fig. S8b, ESI). Due to the presence of interfacial defects, the control QE showed a lower ionic conductivity than the MH-QE (Fig. 2e and Fig. S8c, ESI). This result demonstrates that the UV curing-assisted printing technique contributed to the fabrication of a MH-QE with decent ionic conductivity.

The Li–S cell (assembled with the MH-QE) showed typical 1st charge/discharge profiles (Fig. S9, ESI). The two typical plateaus at 2.3 and 2.1 V in the discharge profiles correspond to the electrochemical conversions of S8 to Li2Sx (4 ≤ x ≤ 8) and Li2S4 to Li2S2/Li2S, respectively.24 The cycling performance of the Li–S cells was investigated at a slow discharge/charge current density of 0.1 C/0.1 C (Fig. 3a). The cell containing the MH-QE showed good capacity retention (discharge capacity = 670 mA h gsulfur−1 and coulombic efficiency = 98.3% after 250 cycles), even without the use of any electrolyte additives. In comparison, the cells with the EMS- and TEGDME-based electrolytes showed the poor cycling performance (Fig. 2b). This cycling performance of the MH-QE is higher than those of control electrolytes that are commonly used in Li–S batteries (1 M LiTFSI in 1,3-dioxalane (DOL)/1,2-dimethoxyethane (DME), hereinafter, denoted as DOL/DME. DOL/DME with 0.2 M LiNO3 additive).25 Due to the unwanted shuttle effect, the DOL/DME showed very poor cycling performance with low coulombic efficiency, which appeared similar to those of previous studies.26,27 The LiNO3 additive in Li–S cells is known to suppress the polysulfide migration-induced shuttle effect. However, the additive is continuously consumed and eventually depleted in the cells, resulting in the decline of cycling retention as shown in Fig. 3a. To highlight the superiority of the MH-QE, its cycling performance was compared with those of previously reported electrolyte-based approaches for Li–S cells (Fig. 3a and Table S4, ESI).25,28–32 This advantageous effect of the MH-QE on the cycling performance was also observed at a high-mass loading (=4.1 mg cm−2) of the sulfur cathode (Fig. S10, ESI).


image file: c8ee01503a-f3.tif
Fig. 3 Electrochemical performance of Li–S cells containing MH-QEs. (a) Cycling performance MH-QE vs. control electrolytes (DOL/DME, DOL/DME with 0.2 M LiNO3 additive) at a charge/discharge current density = 0.1 C/0.1 C. Additionally, the results of previously reported Li–S electrolytes25,28–32 were included for comparison. (b) SEM image of the lithium metal anode surface after 250 cycles (for MH-QE). The inset is the result of the DOL/DME. (c) Shuttle factor (f) of the MH-QE and DOL/DME, DOL/DME with 0.2 M LiNO3 additive.

To further verify the beneficial effect of the MH-QE, the surface of the lithium metal anodes was analyzed after the cycling test (250 cycles). The cells were disassembled at a discharge voltage of 1.5 V, and then, the lithium metal anodes were washed with dimethyl ether to remove the electrolyte and soluble long-chain polysulfide residuals. The lithium metal surface (for the MH-QE) was rarely contaminated with insoluble polysulfides (Fig. 3b and Fig. S11, ESI), as compared to the result of the DOL/DME. As supplementary data, the lithium metal surface (for the MH-QE) before the cycling test was characterized (Fig. S12, ESI). No significant difference in the surface structure was observed before/after the cycling test. This result demonstrates that the deposition of electrically inert insoluble polysulfides on lithium metal anodes, which is known to result from migration of polysulfides via electrolytes,33,34 was effectively suppressed by the MH-QE.

These electrochemical results demonstrated that the EMS-QE layer in the MH-QE promotes polysulfide dissolution, facilitating the electrochemical conversion of the sulfur cathodes. Simultaneously, the migration of the dissolved polysulfides was suppressed by the adjacent TEGDME-QE layer. Specifically, the dissolved polysulfides were confined inside the EMS-QE layer the during charge/discharge reactions, which eventually contributed to the better utilization of sulfur active materials and to reversible conversion reactions without the occurrence of the shuttle effect.

To quantitatively describe the shuttle effect, a shuttle factor (f = Ks·QH[Stotal]/I).35 was proposed, in which the shuttle factor (f) is correlated with the coulombic efficiency (Ceff) as follows:36

image file: c8ee01503a-t1.tif

In Fig. 3c, the f values of the DOL/DME were estimated to be higher than 0.9 during the cycling, revealing a serious shuttle effect. Meanwhile, the DOL/DME with 0.2 M LiNO3 additive showed low f values prior to the 100th cycle, however, since then, the f values sharply increased with cycling and reached 0.78 at the 250th cycle, indicating the depletion of the additive due to its continuous consumption. In contrast, the MH-QE exhibited remarkably low f values over the whole cycling. This consideration of the f values is well consistent with the cycling results shown in Fig. 3a. Additionally, the shuttle factor of the MH-QE remained almost unchanged over the entire range of cycle number. Another shuttle effect-triggered challenge is self-discharge of Li–S cells. In most cases, approximately 30–40% of the cell capacity is lost within several hours due to self-discharge.37 The MH-QE effectively mitigated the self-discharge problem of Li–S cells (Fig. S13, ESI). The open-circuit voltage (OCV) of the Li–S cell containing the DOL/DME decayed sharply, which revealed the electrochemical reduce of soluble high-order polysulfides (Li2S8/Li2S6) to semi-solid low-order polysulfides (Li2S4) (i.e., S82− + 2e → 2S42−, S62−(S3*) + e → 3/2S42−).38 In comparison, the OCV of the cell with the MH-QE remained almost constant near 2.3 V after 72 h. The OCV results were further verified by comparing the charge/discharge profiles before and after an elapsed time of 72 h. The cell with the MH-QE showed high capacity retention (∼95%) after 72 h, whereas a significant capacity loss (capacity retention ∼ 22%) was observed at the cell containing the DOL/DME. These results demonstrate that, driven by the combined effect of the thermodynamic immiscibility and structural uniqueness (i.e., monolithic heterojunction structure), the MH-QE effectively suppressed the self-discharge problem of Li–S cells.

A post-mortem analysis of the Li–S cells was conducted to better understand the advantageous effect of the MH-QE on the shuttle effect. After the cycle test, the cells were disassembled at a discharge voltage of 1.5 V, and then, the lithium metal anodes were retrieved without being washed. The ex situ X-ray photoelectron spectroscopy (XPS) analysis (Fig. 4a) showed that the characteristic S 2p3/2 peaks at 160.6, 162.3, and 164.0 eV,39–41 which were assigned to insoluble Li2S2/Li2S, low-order polysulfides (Li2Sx (I)), and high-order polysulfides or elemental sulfur (Li2Sx (II)), respectively, were observed on the lithium anode surface, along with peaks over 169.9 eV corresponding to LiCF3SO3.42 The stronger peak intensity of the insoluble Li2S2/Li2S layer (at 160.6 eV) was observed at the DOL/DME. In contrast, the corresponding S 2p3/2 peaks almost disappeared at the MH-QE. The XPS spectra of the MH-QE before the cycling test were also provided for comparison (Fig. S14a, ESI).


image file: c8ee01503a-f4.tif
Fig. 4 In-depth post-mortem analysis of Li–S cells containing MH-QEs. (a) S 2p XPS spectra of the lithium metal anodes (MH-QE vs. DOL/DME) after 250 cycles, in which the Li–S cells were discharged to 1.5 V (i.e., 100% DOD). (b) Raman spectra of the MH-QE (taken at a cell voltage of 2.1 V) in the through-thickness direction (EMS-QE layer vs. TEGDME-QE layer) after 250 cycles.

The liquid-state polysulfide distribution inside the MH-QE, which was taken at a cell voltage of 2.1 V, was investigated by examining the ex situ Raman spectra in the through-thickness direction (Fig. 4b). The peaks assigned to the S–S bending and stretching vibrations of polysulfides42,43 were observed between 100 and 800 cm−1. Notably, the peak at 482 cm−1, which is ascribed to the bending mode of S42− for long polysulfides (such as Li2S4),44 was clearly observed at the EMS-QE layer but not in the TEGDME-QE layer. As supplementary data, the ex situ Raman spectra of the MH-QE before the cycling test were provided (Fig. S14b, ESI). This result demonstrates that the TEGDME-QE layer of the MH-QE plays an important role in suppressing polysulfide migration. Furthermore, the long-term thermodynamic immiscibility of the MH-QE was confirmed by analyzing the 1H NMR spectra of the EMS-QE and TEGDME-QE layers after the cycle test (Fig. S15, ESI). The characteristic 1H NMR peaks corresponding to EMS and TEGDME remained unchanged after 250 cycles, confirming that the phase-separated state between the two QE layers in the MH-QE was retained during charge/discharge cycling.

In summary, we presented the MH-QE as a new electrode-customized multifunctional electrolyte approach for high-performance Li–S batteries. The MH-QE consisted of two immiscible EMS-QE and TEGDME-QE layers that were respectively tailored for anodes/cathodes and seamlessly connected in-series by the UV-curing-assisted printing process. The difference in the electrostatic force (ΔδES) between the electrolyte mixtures played a viable role in creating the thermodynamic immiscibility. The underlying thermodynamics of the electrolyte immiscibility was systematically elucidated through the theoretical considerations and experimental results. The two immiscible QE layers in the MH-QE respectively worked for sulfur cathodes and lithium metal anodes without concerns of electrolyte interdiffusion. The EMS-QE layer was oriented toward sulfur cathodes to allow sulfur utilization, while the TEGDME-QE layer, which faced lithium anodes, prevented migration of dissolved polysulfides. Benefiting from the combined effects (i.e., monolithic heterojunction structure and thermodynamic immiscibility), the MH-QE significantly suppressed the shuttle effect while facilitating sulfur utilization, eventually providing exceptional cycling performance which is difficult to achieve with conventional battery electrolytes. We envision that the electrode-customized, multifunctional MH-QE strategy is effective and versatile, and thus holds great promise as a new electrolyte platform for next-generation advanced power sources.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Basic Science Research Program (2017M1A2A2087810, 2017M1A2A2044501, 2018R1A2A1A05019733) and Wearable Platform Materials Technology Center (2016R1A5A1009926) through the National Research Foundation of Korea (NRF) funded by the Ministry of Science, ICT and future Planning. This work was also financially supported by Corporate R&D of LG Chem. S. K. K. acknowledges the financial support from the NRF of Korea (NRF-2014R1A5A1009799) and computational resources from UNIST-HPC.

Notes and references

  1. J.-M. Tarascon and M. Armand, Nature, 2001, 414, 359–367 CrossRef CAS PubMed .
  2. P. G. Bruce, S. A. Freunberger, L. J. Hardwick and J.-M. Tarascon, Nat. Mater., 2012, 11, 19–29 CrossRef CAS PubMed .
  3. R. Black, B. Adams and L. F. Nazar, Adv. Energy Mater., 2012, 2, 801–815 CrossRef CAS .
  4. L. Wang, Y. Wang and Y. Xia, Energy Environ. Sci., 2015, 8, 1551–1558 RSC .
  5. H. Zhou, Y. Wang, H. Li and P. He, ChemSusChem, 2010, 3, 1009–1019 CrossRef CAS PubMed .
  6. P. He, T. Zhang, J. Jiang and H. Zhaou, J. Phys. Chem. Lett., 2016, 7, 1267–1280 CrossRef CAS PubMed .
  7. Z. Guo, Y. Wang, Y. Song, C. Li, X. Su, Y. Wang, W. Cai and Y. Xia, ACS Energy Lett., 2017, 2, 36–44 CrossRef CAS .
  8. K. Xu and C. A. Angell, J. Electrochem. Soc., 1998, 145, L70–L72 CrossRef CAS .
  9. D. Brouillette, G. Perron and J. E. Desnoyers, J. Solution Chem., 1998, 27, 151–182 CrossRef CAS .
  10. S. Tsuzuki, W. Shinoda, M. Matsugami, Y. Umebayashi, K. Uenom, T. Mandai, S. Seki, K. Dokko and M. Watanabe, Phys. Chem. Chem. Phys., 2015, 17, 126–129 RSC .
  11. K. Xu, Chem. Rev., 2004, 104, 4303–4417 CrossRef CAS PubMed .
  12. K. Ueno, R. Tatara, S. Tsuzuki, S. Saito, H. Doi, K. Yoshida, T. Mandai, M. Matsugami, Y. Umebayashi, K. Dokko and M. Watanabe, Phys. Chem. Chem. Phys., 2015, 17, 8248–8257 RSC .
  13. J. H. Hildebrand and R. L. Scott, The solubility of nonelectrolytes, Reinhold Pub. Corp., USA, 1950, p. 253 Search PubMed .
  14. C. M. Hansen, J. Paint Technol., 1967, 39, 104–117 CAS .
  15. D. J. Greenhalgh, A. C. Willams, P. Timmins and P. York, J. Pharm. Sci., 1999, 88, 1182–1190 CrossRef CAS PubMed .
  16. N. Muzet, E. Engler and G. Wipff, J. Phys. Chem. B, 1998, 102, 10722–10788 CrossRef .
  17. J. S. Rowlinson and B. Widom, Molecular theory of capillarity, Courier Corp., 1982, p. 352 Search PubMed .
  18. L. Yue, J. Ma, J. Zhang, J. Zhao, S. Dong, Z. Liu, G. Cui and L. Chen, Energy Storage Mater., 2016, 5, 139–164 CrossRef .
  19. H. Qu, J. Zhang, A. Du, B. Chen, J. Chai, N. Xue, L. Wang, L. Qiao, C. Wang, X. Zang, J. Yang, X. Wang and G. Cui, Adv. Sci., 2018, 5, 1700503 CrossRef PubMed .
  20. S. H. Kim, K. H. Choi, S. J. Cho, S. Choi, S. Park and S. Y. Lee, Nano Lett., 2015, 15, 5168–5177 CrossRef CAS PubMed .
  21. E. H. Kil, K. H. Choi, H. J. Ha, S. Xu., J. A. Rogers, M. R. Kim, Y. G. Lee, K. M. Kim, K. Y. Cho and S. Y. Lee, Adv. Mater., 2013, 25, 1395–1400 CrossRef CAS PubMed .
  22. R. Xu, J. C. M. Li, J. Lu, K. Amine and I. Belharouak, J. Mater. Chem. A, 2015, 3, 4170–4179 RSC .
  23. X. B. Cheng, J. Q. Huang, H. J. Peng, J. Q. Nie, X. Y. Liu, Q. Zhang and F. Wei, J. Power Sources, 2014, 253, 263–268 CrossRef CAS .
  24. K. Xu, Chem. Rev., 2014, 114, 11503–11618 CrossRef CAS PubMed .
  25. H. S. Kim, T.-G. Jeong, N.-S. Choi and Y.-T. Kim, Ionics, 2013, 19, 1795–1802 CrossRef CAS .
  26. L. Ma, H. Zhuang, Y. Lu, S. S. Moganty, R. G. Henning and L. A. Archer, Adv. Energy Mater., 2014, 4, 1400390 CrossRef .
  27. N. Azimi, W. Weng, C. Takoudis and Z. Zhang, Electrochem. Commun., 2013, 37, 93–99 CrossRef .
  28. C. Barchasz, J.-C. Lepretre, S. Patoux and F. Alloin, J. Electrochem. Soc., 2013, 160, A430–A436 CrossRef CAS .
  29. L. Wang and H. R. Byon, J. Power Sources, 2013, 236, 207–214 CrossRef CAS .
  30. Z. Lin, Z. Liu, W. Fu, N. J. Dudney and C. Liang, Adv. Funct. Mater., 2013, 23, 1064–1069 CrossRef CAS .
  31. N. Tachikawa, K. Yamauchi, E. Takashima, J.-W. Park, K. Dokko and M. Watanabe, Chem. Commun., 2011, 47, 8157–8159 RSC .
  32. K. Suo, Y.-S. Hu, H. Li, M. Armand and L. Chen, Nat. Commun., 2013, 4, 1481 CrossRef PubMed .
  33. L. X. Yuan, X. P. Qiu, L. Q. Chen and W. T. Zhu, J. Power Sources, 2009, 189, 127–132 CrossRef CAS .
  34. J. M. Zheng, M. Gu, C. Wang, P. Zuo, P. K. Koech, J. G. Zhang, J. Liu and J. Xiao, J. Electrochem. Soc., 2013, 160, A1992–A1996 CrossRef CAS .
  35. Y. V. Mikhaylik and J. R. Akridge, J. Electrochem. Soc., 2004, 151, A1969–A1976 CrossRef CAS .
  36. J. Q. Huang, Q. Zhang, S. M. Zhang, X. F. Liu, W. Zhu, W. Z. Qian and F. Wei, Carbon, 2013, 58, 99–106 CrossRef CAS .
  37. M. L. Gordin, F. Dai, S. Chen, T. Xu, J. Song, D. Tang, N. Azimi, Z. Zhang and D. Wang, ACS Appl. Mater. Interfaces, 2014, 6, 8006–8010 CrossRef CAS PubMed .
  38. Q. Wang, J. Zheng, E. Waler, H. Pan, D. Lv, P. Zuo, H. Chen, Z. D. Deng, B. Y. Liaw and X. Yu, J. Electrochem. Soc., 2015, 162, A474–A478 CrossRef CAS .
  39. Y. Diao, K. Xie, S. Z. Xiong and X. B. Hong, J. Electrochem. Soc., 2012, 159, A1816–A1821 CrossRef CAS .
  40. Y. S. Su, Y. Fu, T. Cochell and A. Manthiram, Nat. Commun., 2013, 4, 2985 CrossRef PubMed .
  41. Y. Fu, Y. S. Su and A. Manthiram, Angew. Chem., Int. Ed., 2013, 52, 6930–6935 CrossRef CAS PubMed .
  42. J. A. Nimon, V. D. Nefe, R. E. Cantley and R. O. Buttlar, J. Mol. Spectrosc., 1967, 22, 105–108 CrossRef .
  43. M. Hagen, P. Schiffels, M. Hammer, S. Dorfler, J. Tubke, M. J. Hoffmann, H. Althues and S. Kaskel, J. Electrochem. Soc., 2013, 160, A1205–A1214 CrossRef CAS .
  44. T. Chivers and C. Lau, Inorg. Chem., 1982, 21, 453–455 CrossRef CAS .

Footnotes

Electronic supplementary information (ESI) available: Fig. S1–S15 and Tables S1–S4. See DOI: 10.1039/c8ee01503a
These authors contributed equally to this work.

This journal is © The Royal Society of Chemistry 2019