All-Solid-State Lithium Battery Working without an Additional Separator in a Polymeric Electrolyte

Considering the safety issues of Li ion batteries, an all-solid-state polymer electrolyte has been one of the promising solutions. Achieving a Li ion conductivity of a solid-state electrolyte comparable to that of a liquid electrolyte (>1 mS/cm) is particularly challenging. Even with characteristic ion conductivity, employment of a polyethylene oxide (PEO) solid electrolyte has not been sufficient due to high crystallinity. In this study, hybrid solid electrolyte (HSE) systems have been designed with Li1.3Al0.3Ti0.7(PO4)3 (LATP), PEO and lithium bis(trifluoromethanesulfonyl)imide (LiTFSI). A hybrid solid cathode (HSC) is also designed using LATP, PEO and lithium cobalt oxide (LiCoO2, LCO)—lithium manganese oxide (LiMn2O4, LMO). The designed HSE system has 2.0 × 10−4 S/cm (23 °C) and 1.6 × 10−3 S/cm (55 °C) with a 6.0 V electrochemical stability without an additional separator membrane introduction. In these systems, succinonitrile (SN) has been incorporated as a plasticizer to reduce crystallinity of PEO for practical all-solid Li battery system development. The designed HSC/HSE/Li metal cell in this study operates without any leakage and short-circuits even under the broken cell condition. The designed HSC/HSE/Li metal cell in this study displays an initial charge capacity of 82/62 mAh/g (23 °C) and 123.4/102.7 mAh/g (55 °C). The developed system overcomes typical disadvantages of internal resistance induced by Ti ion reduction. This study contributes to a new technology development of all-solid-state Li battery for commercial product design.


Introduction
Li ion batteries have been credited for a great revolution of the strong intermittent renewable-energy sources replacing fossil fuels. They also critically contribute to the development of communication and transportation by the rise of super-slim smart phones and electric cars in a practical range. However, since announced in 1991, heels of Li ion battery phones and laptops have been recalled because of a flame causing injury to the users [1]. The cycle life is limited because of poor cycling efficiency of Li electrode. The secondary cells are more sensitive to impurities such as water in the electrolyte and the electrode materials. In addition, the cells under running would have Li dendrites leading to occasional explosions. Nonetheless, because of radical needs for high energy density and reliability of batteries, Li technology has been ceaselessly focused on. With high energy evaluation as an intercept for 10%, 50% and 90% of the cumulative mass and D 50~6 µm is obtained for effective mixing process.

Preparation of Hybrid Solid Electrolytes (HSE) and Hybrid Solid Electrolyte Cathode (HSC)
The Li ion conducting polymer PEO is mixed with LiTFSI ([EO]: [Li] = 8:1) and previously synthesized LATP (LATP:PEO = 8:2 by weight) in acetonitrile (AN). The polymer solution is mixed using a centrifugal mixer (THINKY mixer ARM-310, Oxfordshire, UK) at 2000 rpm for 15 min, with or without 5 wt % of SN (Succinonitrile). The slurry is mixed again for 15 min. For the tests, this HSE slurry is casted on Al foil for analysis or on an electrode for coin cell test. After natural drying at room temperature for 1 h, samples are dried completely in a 50 • C vacuum oven for 12 h.
For better binding in the electrode, higher molecular weight PEO (M w~3 × 10 5 and 6 × 10 5 ) was employed. LATP, PEO (LATP:PEO = 5:5 by weight) and LiTFSI ([EO]: [Li] = 10:1) were mixed in AN using THINKY mixer for 15 min. LiMn 2 O 4 and LiCoO 2 as active materials and Super-P as conductive material were added in additional solvent AN, followed by mixing for 15 min. After the HSC slurry was casted on Al foil by Scalpel, the electrodes were dried at 80 • C in a vacuum oven for 24 h. HSE slurry was casted on the dried electrode and Li metal. The aforementioned HSE and HSC thickness is designed uniformly at 270 µm and 80 µm scale and the active material loading of electrodes is designed uniformly at a 10 mg cm −2 (±0.1 mg cm −2 ) scale. The HSC/HSE composite was dried under the aforementioned condition. Differing from the cathode side, the Li metal/HSE composite was dried at room temperature for 24 h to prevent crumbling. To evaluate the electrochemical performance, the CR2032 coin-type cell and all-solid-state pouch cell at a 5 × 10 cm 2 scale was fabricated with HSC/HSE composite and Li metal/HSE by a lamination process. The construction of the cell is described as Figure 1. All processes were conducted in a dry-room. For better binding in the electrode, higher molecular weight PEO (Mw ~3  10 5 and 6  10 5 ) was employed. LATP, PEO (LATP:PEO = 5:5 by weight) and LiTFSI ([EO]: [Li] = 10:1) were mixed in AN using THINKY mixer for 15 min. LiMn2O4 and LiCoO2 as active materials and Super-P as conductive material were added in additional solvent AN, followed by mixing for 15 min. After the HSC slurry was casted on Al foil by Scalpel, the electrodes were dried at 80 °C in a vacuum oven for 24 h. HSE slurry was casted on the dried electrode and Li metal. The aforementioned HSE and HSC thickness is designed uniformly at 270 μm and 80 μm scale and the active material loading of electrodes is designed uniformly at a 10 mg cm −2 (±0.1 mg cm −2 ) scale. The HSC/HSE composite was dried under the aforementioned condition. Differing from the cathode side, the Li metal/HSE composite was dried at room temperature for 24 h to prevent crumbling. To evaluate the electrochemical performance, the CR2032 coin-type cell and all-solid-state pouch cell at a 5 × 10 cm 2 scale was fabricated with HSC/HSE composite and Li metal/HSE by a lamination process. The construction of the cell is described as Figure 1. All processes were conducted in a dry-room. Configuration of all-solid-state pouch cell at a 5 × 10 cm 2 scale (left scheme) and its operation (right picture).

Characterizations
Synthesized LATP and the HSC/HSE composites were observed using a scanning electron microscope (SEM, SEC Mini SEM SNE-3000M, Pleasanton, CA, USA) and X-ray diffraction (XRD, Rigaku MiniFlex 600, The Woodlands, TX, USA). XRD patterns of samples were obtained over a 2θ range from 10 deg to 80 deg with Cu Kα radiation at room temperature. The scan rate was 6 deg/min. Linear sweep voltammetry (LSV) was performed for both oxidation and reduction procedures for selected samples using a potentiostat (Bio Logic SP-150) with scan rate of 20 mV/s from OCV to 1.5 V or 6.0 V. Ionic conductivity is calculated by the AC impedance method with symmetric SS/Al/HSE/Al/SS cell using a multi-channel potentiostat (Bio Logic VMP3, Seyssinet-Pariset, France). The samples were placed at 25 • C and 55 • C for 12 h, and then analyzed from 500 kHz to 1 Hz at open-circuit voltage with 5.0 mV amplitude. With a differential scanning calorimeter (DSC, DISCOVERY DSC 2500, TA Instrument, New Castle, DE, USA), the calorimetric measurement is performed from room temperature to 100 • C at the heating rate of 10 • C/min. For each C-rate (1C = 1.5 mA cm −2 ), the cell was charged using a constant current charge (0.05C & each C-rate) using a charge step from 3.0 V to 4.3 V, followed by a constant voltage charge at 4.3 V. For discharge, the constant current discharge (0.05C & each C-rate) from 4.3 V to 3.0 V was applied at room temperature or 55 • C.

Design of Hybrid Solid Electrolytes (HSE) and Hybrid Solid Cathode (HSC)
This study introduces a designed hybrid solid electrolyte (HSE) and hybrid solid cathode (HSC). The HSCs are designed by combining lithium manganese oxide (LMO, LiMn 2 O 4 ) with lithium cobalt oxide (LCO, LiCoO 2 ). In Table 1, composition and ion conductivity of the designed HSC and HSE are summarized at two temperature conditions (23 and 55 • C). Li ion-conducting PEO is mixed with lithium aluminum titanium phosphates (LATP) synthesized in this study (Experimental Section). This sticky white HSE slurry is casted on Al or Li metal foil. For better casting process, the viscosity of the slurry is modified by controlling the content of LATP and PEO as well as molecular weight of PEO. The battery efficiency is compared with lithium(trifluoromethanesulfonyl)imide (LiTFSI) and lithium hexafluorophosphate (LiPF 6 ). In addition, the function of SN is evaluated in terms of the interaction with PEO and LiTFSI. Ion conductivity is compared with pure PEO electrolyte (HSE-1) and PEO/LATP (20/80) composite electrolyte with LiPF 6 (HSE-2). With LiTFSI, the PEO/LATP (20/80) composite electrolyte without (HSE-3) and with (HSE-4) SN incorporation is compared. Two types of LMO/LCO (30/30)-based cathodes are compared containing the PEO/LATP (12.8/12.8) composite combined with LiTFSI, without (HSC-1) and with (HSC-2) SN incorporation. A representative coin cell preparation using the designed HSE-4 and HSC-2 is illustrated in Figure 1a. The slurry is coated on Li metal foil by solution casting. LATP, PEO, LiTFSI and SN are Polymers 2018, 10, 1364 6 of 16 mixed in a designed composition using a centrifugal mixer (Thinky Mixer) resulting in a sticky white HSE-4 slurry. LMO and LCO are mixed with PEO and SN, resulting in HSC-2. The HSC-2 is coated on Al foil, followed by HSE-4 slurry casting. The Li metal anode and the designed cathode plates are bonded by heating HSE sides to form a layered structure of a coin cell. In addition, this process is scaled-up for a pouch cell (5 × 10 cm 2 ) in which copper foil is additionally employed outside of the Li metal plate for physical protection (Figure 1b). The designed all-solid-state lithium pouch battery is successfully working in spite of its breakage without any leakage and short-circuits (Supporting Information, Movie S1).

Structure of the All-Solid-State Battery System
The simplified battery sample is prepared to investigate the performance of HSE and HSC, which are casted and dried on a layered structure stacked from Al foil, cathode (HSC-2) and electrolyte (HSE-4) layers without anode Li metal and protective Cu foil layers. A representative scanning electron microscopy (SEM) image of each layer is shown in a cross-sectional side-view ( Figure 2a) and in a top-view ( Figure 2b). In the cross-sectional image, the size of particulates in HSC layer (below the dotted-line in the picture) is smaller compared with that in the HSE layer; this is caused by the relatively small size of the LCO and LMO particles ( Figure 2d). The images of pure LATP powder are compared in Figure 2c. The top-view images of the designed battery system (Figure 2b) are similar to those of the LATP powder pellet (Figure 2c). For all the systems, the inter-particulate spaces are effectively filled with PEO binding. The SEM images confirm effective physical contact of ion-conducting PEO and solid LATP in the designed HSE layer without phase separation, which is crucial for effective ionic conductivity of a battery. Figure 2b depicts the reticulated PEO connects LATP particles forming ion-conducting pathways, which can decrease boundary resistance in the solid LATP electrolyte. All of these morphologies are advantageous for the increase in the bulk ionic conductivity [47][48][49].
A representative coin cell preparation using the designed HSE-4 and HSC-2 is illustrated in Figure 1a. The slurry is coated on Li metal foil by solution casting. LATP, PEO, LiTFSI and SN are mixed in a designed composition using a centrifugal mixer (Thinky Mixer) resulting in a sticky white HSE-4 slurry. LMO and LCO are mixed with PEO and SN, resulting in HSC-2. The HSC-2 is coated on Al foil, followed by HSE-4 slurry casting. The Li metal anode and the designed cathode plates are bonded by heating HSE sides to form a layered structure of a coin cell. In addition, this process is scaled-up for a pouch cell (5 × 10 cm 2 ) in which copper foil is additionally employed outside of the Li metal plate for physical protection (Figure 1b). The designed all-solid-state lithium pouch battery is successfully working in spite of its breakage without any leakage and short-circuits (Supporting Information, Movie S1).

Structure of the All-Solid-State Battery System
The simplified battery sample is prepared to investigate the performance of HSE and HSC, which are casted and dried on a layered structure stacked from Al foil, cathode (HSC-2) and electrolyte (HSE-4) layers without anode Li metal and protective Cu foil layers. A representative scanning electron microscopy (SEM) image of each layer is shown in a cross-sectional side-view ( Figure 2a) and in a top-view ( Figure 2b). In the cross-sectional image, the size of particulates in HSC layer (below the dotted-line in the picture) is smaller compared with that in the HSE layer; this is caused by the relatively small size of the LCO and LMO particles ( Figure 2d). The images of pure LATP powder are compared in Figure 2c. The top-view images of the designed battery system ( Figure  2b) are similar to those of the LATP powder pellet ( Figure 2c). For all the systems, the inter-particulate spaces are effectively filled with PEO binding. The SEM images confirm effective physical contact of ion-conducting PEO and solid LATP in the designed HSE layer without phase separation, which is crucial for effective ionic conductivity of a battery. Figure 2b depicts the reticulated PEO connects LATP particles forming ion-conducting pathways, which can decrease boundary resistance in the solid LATP electrolyte. All of these morphologies are advantageous for the increase in the bulk ionic conductivity [47][48][49].

Conductivity of the Designed Systems
The total resistance is evaluated by ionic conductivity based on the electrochemical impedance spectroscopy (EIS) ( Table 1). Using the measured resistance and following relation, ionic conductivity (σ, S/cm) of the designed HSE and HSC is evaluated, where l is the thickness (cm) of the HSE, A is the area (cm 2 ) of the sample, S is the total resistance (Ω) obtained from EIS spectra. The ionic conductivity of the designed HSE shows almost three orders higher (~10 4 S/cm) than that of pure PEO (10 6~1 0 8 S/cm) [50] at the same EO to Li (8:1). This result confirms the contribution of high ionic conductivity of LATP embedded in the ion-conducting PEO by the reduced bulk and boundary resistance. Due to characteristic temperature-dependence, ionic conductivity at 55 C is about 10 times higher than that at 23 C in every sample. The composite system consisting of LiTFSI shows similar ionic conductivity as that using LiPF6. By adding SN, ionic conductivities (2.0 × 10 4 S/cm @ 23C & 1.6 × 10 3 S/cm @ 55 C) are enhanced compared with those (1.5 × 10 4 S/cm @ 23 C & 1.4 × 10 3 S/cm @ 55C) without SN introduction by increased segmental mobility of PEO in HSE [40,51]. In particular, the increase in ionic conductivity by SN introduction is prominent at room temperature.

Solid Electrolyte Interface (SEI)
Even with several controversies the LATP and Li metal is reported to react immediately upon a physical contact and forms an unfavorable solid electrolyte interface (SEI) layer on the Li metal surface [52,53]. Through this unfavorable phenomenon Li ion movement is reduced, thus decreasing the ionic conductivity. However, this can be suppressed by surface modification using PEO coating. The role of PEO in HSE is, at the same context, surrounding each LATP particle to protect the Li metal surface. The efficacy of the designed HSE on Li metal anode is investigated by measuring the AC impedance ( Figure 3c

Conductivity of the Designed Systems
The total resistance is evaluated by ionic conductivity based on the electrochemical impedance spectroscopy (EIS) ( Table 1). Using the measured resistance and following relation, ionic conductivity (σ, S/cm) of the designed HSE and HSC is evaluated, where l is the thickness (cm) of the HSE, A is the area (cm 2 ) of the sample, S is the total resistance (Ω) obtained from EIS spectra. The ionic conductivity of the designed HSE shows almost three orders higher (~10 −4 S/cm) than that of pure PEO (10 −6~1 0 −8 S/cm) [50] at the same EO to Li (8:1). This result confirms the contribution of high ionic conductivity of LATP embedded in the ion-conducting PEO by the reduced bulk and boundary resistance. Due to characteristic temperature-dependence, ionic conductivity at 55 • C is about 10 times higher than that at 23 • C in every sample. The composite system consisting of LiTFSI shows similar ionic conductivity as that using LiPF 6 . By adding SN, ionic conductivities (2.0 × 10 −4 S/cm @ 23 • C & 1.6 × 10 −3 S/cm @ 55 • C) are enhanced compared with those (1.5 × 10 −4 S/cm @ 23 • C & 1.4 × 10 −3 S/cm @ 55 • C) without SN introduction by increased segmental mobility of PEO in HSE [40,51]. In particular, the increase in ionic conductivity by SN introduction is prominent at room temperature.

Solid Electrolyte Interface (SEI)
Even with several controversies the LATP and Li metal is reported to react immediately upon a physical contact and forms an unfavorable solid electrolyte interface (SEI) layer on the Li metal surface [52,53]. Through this unfavorable phenomenon Li ion movement is reduced, thus decreasing the ionic conductivity. However, this can be suppressed by surface modification using PEO coating. The role of PEO in HSE is, at the same context, surrounding each LATP particle to protect the Li metal surface. The efficacy of the designed HSE on Li metal anode is investigated by measuring the AC impedance (Figure 3c (Figure 3c). The result suggests that the reaction of HSE-4 on Li metal is far more effective than that of LATP. An equivalent circuit to describe the observed AC impedance spectra is shown in the inset of Figure 3c, which represents a hybrid solid electrolyte sandwiched between two blocking electrodes. In this equivalent circuit, Rct is the bulk resistance of the solid electrolyte, Cdl (constant phase element) denotes the bulk capacitance of the hybrid solid electrolyte, and Ri corresponds to the double layer capacitance at the electrode/electrolyte interface. Constant phase elements rather than capacitors were employed to describe non-idealities in AC impedance responses. The experimental data were fitted using the Randles equivalent circuit. Indeed, the AC impedance of a lithium metal symmetric cell is composed of the resistance at the electrolyte/electrode interface and the Ohmic resistance of the electrolyte itself.
In order to evaluate the impedance aforementioned result, X-ray photoelectron spectroscopy (XPS) measurements of the HSE samples are performed before and after the reaction with Li metals for 25 days (Figure 3d). For this, Ti peaks of HSE-4 pasted on Li metal after aging procedures for 25 days (blue line) are compared with HSE-4 on Al foil, with the physical substrate (red line) employed as a standard. Doublet peaks consist of Ti 2p1/2 and Ti 2p3/2: two strong Ti 4+ peaks, i.e., 2p1/2 (Ti 4+ , 464.6 eV) and 2p3/2 oxide (Ti 4+ , 458.8 eV), two Ti 3+ peaks (457 eV and 463.1 eV) [55].  (Figure 3c). The result suggests that the reaction of HSE-4 on Li metal is far more effective than that of LATP. An equivalent circuit to describe the observed AC impedance spectra is shown in the inset of Figure 3c, which represents a hybrid solid electrolyte sandwiched between two blocking electrodes. In this equivalent circuit, R ct is the bulk resistance of the solid electrolyte, C dl (constant phase element) denotes the bulk capacitance of the hybrid solid electrolyte, and Ri corresponds to the double layer capacitance at the electrode/electrolyte interface. Constant phase elements rather than capacitors were employed to describe non-idealities in AC impedance responses. The experimental data were fitted using the Randles equivalent circuit. Indeed, the AC impedance of a lithium metal symmetric cell is composed of the resistance at the electrolyte/electrode interface and the Ohmic resistance of the electrolyte itself.
In order to evaluate the impedance aforementioned result, X-ray photoelectron spectroscopy (XPS) measurements of the HSE samples are performed before and after the reaction with Li metals for  (Figure 3d). For this, Ti peaks of HSE-4 pasted on Li metal after aging procedures for 25 days (blue line) are compared with HSE-4 on Al foil, with the physical substrate (red line) employed as a standard. Doublet peaks consist of Ti 2p 1/2 and Ti 2p 3/2 : two strong Ti 4+ peaks, i.e., 2p 1/2 (Ti 4+ , 464.6 eV) and 2p 3/2 oxide (Ti 4+ , 458.8 eV), two Ti 3+ peaks (457 eV and 463.1 eV) [55].
Even with physical contact of HSE-4 with Li metal, the binding energy of two samples in this study is same. The resulting XPS has no difference where the Ti 2p 1/2 and Ti 2p 3/2 remain at the same energy position, indicating there is no Ti reduction (from Ti 4+ to Ti 3+ ) [54].

Electrochemical Stability
To investigate the electrochemical stability of each electrolyte layer, linear sweep voltammetry (LSV) was measured from 1.5 to 6 V for both reduction and oxidation procedures (Figure 4). The oxidative stability is measured from the open current voltage (OCV) by sweeping up the voltage, while the reduction is measured by sweeping down the voltage. In Figure 4, the voltage was swept from~2.2 V up to 6 V and then down again to 2.2 V. Both electrolytes HSE-3 and HSE-4 clearly show an onset of oxidation around 4 and 3.4 V, respectively. After going up to 6 V, the electrolyte will be partially or completely decomposed. During the sweeping down from that stage, the reductive stability was measured in a separate experiment: 1.2 mol L −1 LiPF 6 in a 15:50:35 (in volume ratio) mixture of ethylene carbonate (EC) and ethyl methyl carbonate (EMC) and dimethyl carbonate (DMC) as the liquid electrolyte used as a standard contrast is oxidized at 4.8 V indicated by the sudden increase in the current (blue line). HSE (HSE-3) and HSE+SN (HSE-4) are also oxidized even if the absolute current intensity is relatively lower than that of liquid electrolyte. This electrochemical investigation of the designed cells indicates protective film formation on the cathode surface in the first cycle which works as a solid electrolyte interphase (SEI) in liquid electrolytes. On the other hand, during the reduction procedure of both HSE-3 and HSE-4, the current is very low without noticeable peak in all voltage regions observed in this study. Even with physical contact of HSE-4 with Li metal, the binding energy of two samples in this study is same. The resulting XPS has no difference where the Ti 2p1/2 and Ti 2p3/2 remain at the same energy position, indicating there is no Ti reduction (from Ti 4+ to Ti 3+ ) [54].

Electrochemical Stability
To investigate the electrochemical stability of each electrolyte layer, linear sweep voltammetry (LSV) was measured from 1.5 to 6 V for both reduction and oxidation procedures (Figure 4). The oxidative stability is measured from the open current voltage (OCV) by sweeping up the voltage, while the reduction is measured by sweeping down the voltage. In Figure 4, the voltage was swept from ~2.2 V up to 6 V and then down again to 2.2 V. Both electrolytes HSE-3 and HSE-4 clearly show an onset of oxidation around 4 and 3.4 V, respectively. After going up to 6 V, the electrolyte will be partially or completely decomposed. During the sweeping down from that stage, the reductive stability was measured in a separate experiment: 1.

Property of the Designed Electrolytes
Due to inherent high boundary resistance, LATP hardly generates high ionic conductivity. A value of ca. 3.4 × 10 3 S/cm at 293 K, is among the highest conductivities reported for LATP-based systems [56][57][58]. To overcome this disadvantage, PEO is employed in this study. The structure of the PEO-incorporated system prepared on Al foil is investigated using X-ray powder diffraction (XRD) in Figure 5a. The XRD patterns of LATP-incorporated systems are almost similar. XRD patterns of HSE films exhibit lower intensity by the polymer incorporation without peak broadening. With the SN incorporation, XRD peak intensity becomes far lower, indicating the dilution of the whole system. Nonetheless the lattice structure is effectively maintained.

Property of the Designed Electrolytes
Due to inherent high boundary resistance, LATP hardly generates high ionic conductivity. A value of ca. 3.4 × 10 −3 S/cm at 293 K, is among the highest conductivities reported for LATP-based systems [56][57][58]. To overcome this disadvantage, PEO is employed in this study. The structure of the PEO-incorporated system prepared on Al foil is investigated using X-ray powder diffraction (XRD) in Figure 5a. The XRD patterns of LATP-incorporated systems are almost similar. XRD patterns of HSE films exhibit lower intensity by the polymer incorporation without peak broadening. With the SN incorporation, XRD peak intensity becomes far lower, indicating the dilution of the whole system. Nonetheless the lattice structure is effectively maintained.
The ionic conductivity and thermal characteristics are modified according to the molecular weight of the PEO, which changes from 1 × 10 5 , 3 × 10 5 to 6 × 10 5 in this study. The segmental mobility of PEO is investigated using differential scanning calorimetry (DSC) (Figure 5b). With the increase of PEO molecular weight, the peak temperature increases from 65.741 • C (PEO M w = 1 × 10 5 ), 67.663 • C (PEO M w = 3 × 10 5 ) and 68.157 • C (PEO M w = 6 × 10 5 ). In addition, the magnitude of the heat density decreases from 178.31 J/g, 172.41 J/g to 170.87 J/g. However, the designed HSE systems with LATP and SN do not generate characteristic peaks at the given condition in Figure 5c. This indicates that incorporation of LATP and SN effectively broadens the amorphous state to the wide temperature ranges, increasing the ionic conductivity under that condition. The DSC result of the system consisting of LiTFSI has been verified to become amorphous when the ratio of the [EO]/ [Li] is higher than ten [59]. In this study, the system consisted of LiTFSI with the ratio [EO]/[Li] = 8 exhibits an amorphous phase with the help of SN introduction. Based on this ionic conductivity result, the system with higher molecular weight PEO M w = 6 × 10 5 is selected to obtain the enhanced mechanical property for the design of HSE, while lower molecular weight PEO M w = 3 × 10 5 is employed for better mixing efficacy to design the cathode electrode. The ionic conductivity and thermal characteristics are modified according to the molecular weight of the PEO, which changes from 1 × 10 5 , 3 × 10 5 to 6 × 10 5 in this study. The segmental mobility of PEO is investigated using differential scanning calorimetry (DSC) (Figure 5b). With the increase of PEO molecular weight, the peak temperature increases from 65.741 °C (PEO Mw = 1 × 10 5 ), 67.663 °C (PEO Mw = 3 × 10 5 ) and 68.157 °C (PEO Mw = 6 × 10 5 ). In addition, the magnitude of the heat density decreases from 178.31 J/g, 172.41 J/g to 170.87 J/g. However, the designed HSE systems with LATP and SN do not generate characteristic peaks at the given condition in Figure 5c. This indicates that incorporation of LATP and SN effectively broadens the amorphous state to the wide temperature ranges, increasing the ionic conductivity under that condition. The DSC result of the system consisting of LiTFSI has been verified to become amorphous when the ratio of the [EO]/ [Li] is higher than ten [59]. In this study, the system consisted of LiTFSI with the ratio [EO]/[Li] = 8 exhibits an amorphous phase with the help of SN introduction. Based on this ionic conductivity result, the system with higher molecular weight PEO Mw = 6 × 10 5 is selected to obtain the enhanced mechanical property for the design of HSE, while lower molecular weight PEO Mw = 3 × 10 5 is employed for better mixing efficacy to design the cathode electrode.  The HSC-1/HSE-2/Li curve has more polarization loss compared to other curves. Further, the plateau region is quite broader for three other cases compared to the discharge curve of HSC-1/HSE-2/Li in terms of the slope nature. Although the difference is small in all the systems, the result of HSC-1/HSE-2/Li shows slight a disadvantage compared to the three other systems. Electrochemical Impedance  curves. Further, the plateau region is quite broader for three other cases compared to the discharge curve of HSC-1/HSE-2/Li in terms of the slope nature. Although the difference is small in all the systems, the result of HSC-1/HSE-2/Li shows slight a disadvantage compared to the three other systems. Electrochemical Impedance Spectroscopy (EIS) results at 55 • C (Figure 6b) and room temperature (Figure 6c) are shown. LiPF 6 is generally used for liquid electrolytes, while LiClO 4 and LiTFSI are favorably employed for solid electrolyte systems. The ionic conductivity of the designed HSE is not so different (Table 1)  Spectroscopy (EIS) results at 55 C (Figure 6b) and room temperature (Figure 6c) are shown. LiPF6 is generally used for liquid electrolytes, while LiClO4 and LiTFSI are favorably employed for solid electrolyte systems. The ionic conductivity of the designed HSE is not so different (Table 1)   However, for LiPF6 (HSC-1/HSE-2/Li for this study) and LiTFSI incorporated systems (HSC-1/HSE-4/Li for this study), the resistance of the system in actual Li ion cell state doubles at 55 C and LiPF6 (HSC-1/HSE-2/Li in this study) and in LiTFSI incorporated systems (HSC-1/HSE-3/Li for this study); the resistance of the system in the actual Li ion cell state increases four-fold compared with that at room temperature. In the Li ion cell state, the re-crystallization kinetics of PEO has been effectively slowed by the use of Li ions and bulky anions [60]. The moisture generated by the However, for LiPF 6 (HSC-1/HSE-2/Li for this study) and LiTFSI incorporated systems (HSC-1/HSE-4/Li for this study), the resistance of the system in actual Li ion cell state doubles at 55 • C and LiPF 6 (HSC-1/HSE-2/Li in this study) and in LiTFSI incorporated systems (HSC-1/HSE-3/Li for this study); the resistance of the system in the actual Li ion cell state increases four-fold compared with that at room temperature. In the Li ion cell state, the re-crystallization kinetics of PEO has been effectively slowed by the use of Li ions and bulky anions [60]. The moisture generated by the electrode-cooling, decomposes LiPF 6 into PF 5 , leading to the resistance increase by the interaction between PEO and SEI [61]. Through this effect, in addition to overcharging, the constant voltage region becomes broader during the charging process while capacity decreases during the discharging procedure. When moisture-resistant LiTFSI or LiTFSI+SN are used, aforementioned phenomena are reduced and capacity can be increased. In the system composed of LiTFSI+SN (HSE-4), SN promotes the Li ion dissolution, thus capacity increases far higher than that of LiTFSI-only (HSE-3). The charging-discharging process occurring at room temperature also shows similar results for the three systems while the HSC-1/HSE-2/Li composed of LiPF 6 exhibits a dramatic difference. This result indicates that the moisture formation after the cooling procedure turns the LiPF 6 into PF 6 gas and SEI in PEO, where SEI decreases the efficiency of the battery function significantly.

Capacity and 1st Coulomb Efficiency of the Designed Coin Cells
The Li ion transport resistance in the HSE system is determined by its through-plane resistance instead of the ionic conductivity. The actual Li ion transport resistance may increase with increasing HSE thickness [62]. The designed HSEs in this study are the same thickness. Therefore, the higher conductivity of HSE+SN(HSE-4) ensures excellent Li ion transport performance. This is due to higher ionic conductivity of HSE-4 than HSE-3 and HSE-4 (Table 1), and the reduced resistance of a designed system even with the same thickness (Figure 6c). Figure 7a displays 1st charge-discharge profiles of the LCO-LMO/electrolyte/Li metal systems of selected HSC-1/HSE-4/Li and HSC-2/HSE-4/Li (both are based on LiTFSI+SN composition) at room temperature. This indicates that SN addition boosts the battery function compared with the cells employing LiPF 6 (HSC-1/HSE-2/Li) and LiTFSI-only (HSC-1/HSE-3/Li), which are not working effectively at room temperature. The difference is clearer by the C-rate performance at 55 • C (Figure 7b). The C-rate values of each region are on the graph: 0.05 C, 0.1 C, 0.2 C, 0.5 C, and 0.05 C. At the 1st charging procedure, the capacity of each cell is 111.4 mAh/g (HSE-2), 118.5 mAh/g (HSE-3), 118.9 mAh/g (HSE-4), and 123.4 mAh/g (HSC-2), while at the discharging procedure, it is 96.2 mAh/g, 97.6 mAh/g, 99.9 mAh/g, and 102.7 mAh/g, respectively. At the initial formation, Coulombic efficiency was 86.4%, 82.3%, 84.0%, and 83.2%, respectively, based on the ratio of discharged capacity to charged capacity. In addition, their Coulombic efficiency increases continuously and is stabilized. After the initial cyclic repetition, the overall cyclic repetition over the 99% is maintained. After 20 cycles, it is also maintained by the charging-recharging at 0.05 C.
Polymers 2018, 10, x FOR PEER REVIEW 12 of 16 electrode-cooling, decomposes LiPF6 into PF5, leading to the resistance increase by the interaction between PEO and SEI [61]. Through this effect, in addition to overcharging, the constant voltage region becomes broader during the charging process while capacity decreases during the discharging procedure. When moisture-resistant LiTFSI or LiTFSI+SN are used, aforementioned phenomena are reduced and capacity can be increased. In the system composed of LiTFSI+SN (HSE-4), SN promotes the Li ion dissolution, thus capacity increases far higher than that of LiTFSI-only (HSE-3). The charging-discharging process occurring at room temperature also shows similar results for the three systems while the HSC-1/HSE-2/Li composed of LiPF6 exhibits a dramatic difference. This result indicates that the moisture formation after the cooling procedure turns the LiPF6 into PF6 gas and SEI in PEO, where SEI decreases the efficiency of the battery function significantly. The Li ion transport resistance in the HSE system is determined by its through-plane resistance instead of the ionic conductivity. The actual Li ion transport resistance may increase with increasing HSE thickness [62]. The designed HSEs in this study are the same thickness. Therefore, the higher conductivity of HSE+SN(HSE-4) ensures excellent Li ion transport performance. This is due to higher ionic conductivity of HSE-4 than HSE-3 and HSE-4 (Table 1), and the reduced resistance of a designed system even with the same thickness ( Figure 6c).  Figure  7b). The C-rate values of each region are on the graph: 0.05 C, 0.1 C, 0.2 C, 0.5 C, and 0.05 C. At the 1 st charging procedure, the capacity of each cell is 111.4 mAh/g (HSE-2), 118.5 mAh/g (HSE-3), 118.9 mAh/g (HSE-4), and 123.4 mAh/g (HSC-2), while at the discharging procedure, it is 96.2 mAh/g, 97.6 mAh/g, 99.9 mAh/g, and 102.7 mAh/g, respectively. At the initial formation, Coulombic efficiency was 86.4%, 82.3%, 84.0%, and 83.2%, respectively, based on the ratio of discharged capacity to charged capacity. In addition, their Coulombic efficiency increases continuously and is stabilized. After the initial cyclic repetition, the overall cyclic repetition over the 99% is maintained. After 20 cycles, it is also maintained by the charging-recharging at 0.05 C.

Discussion and Conclusions
In this study, hybrid solid electrolytes (HSEs) composed of LATP, PEO and SN are successfully designed without protection layer introduction. The solid electrolyte, LATP interaction with Li ion, promotes the reduction of Ti 4+ into Ti 3+ . The LATP shows ion conductivity of 3 × 10 −3 S/cm under the state of Ti 4+ but it becomes far lower under the state of Ti 3+ due to structural deformation affecting Li ion movement [11]. We observed no reduction of Ti 4+ into Ti 3 in the designed system which is investigated by Electrochemical Impedance Spectroscopy (EIS) and X-ray photoelectron spectroscopy (XPS).
In addition to the new hybrid solid electrolyte (HSE), new hybrid solid cathodes (HSCs) are designed and investigated, composed of LATP, PEO, LiTFSi and SN. Typically employed for mass-production, slurry casting methods are utilized followed by lamination to reduce contact resistance. In addition, during the dynamic charge-discharge procedure of the designed Li ion cells, the effects of internal resistance on the charge-discharge procedure and C-rate changes are investigated. The interfacial resistance of HSE-4/Li metal symmetric cell is greatly improved compared with the LATP/Li metal symmetric cell ( Figure 3). This suggests that the solution casting process and lamination introduced in this study lower the contact resistance between the Li metal and HSE-4. With good cycling stability, the designed cell also exhibits reasonable interfacial contact efficiency with electrode.
We suggest the new electrolyte system is advantageously utilized in all solid-state Li batteries. Even without any protection layer, the designed system shows no reduction of Ti. This study contributes to a new design technology and further possible mass-production of all solid-state Li batteries in more economical and effective ways.