A Review of Lithium and Non-lithium Based Solid State Batteries

 Journal of Power Sources 282 (2015) 299e 299 e322

Contents lists available at  ScienceDirect

 Journal of Power Sources j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m/ m/ l o c a t e / j p o w s o u r

Review

A review of lithium and non-lithium based solid state batteries  Joo Gon Kim Kim a 1, Byungrak Son a  ** 1, Santanu Mukherjee  b 1, Nicho Nicholas Schuppe Schuppert  b , Alex Bates b, Osung Kwon c, Moon Jong Choi a, Hyun Yeol Chung  d , Sam Park  b  * ,

,

,

,

,

a

Wellness Convergence Research Center, DGIST, 333 Techno Jungang-daero, Hyeongpung-Myeon, Dalseong-Gun, Daegu 711-873, Republic of Korea Department of Mechanical Engineering, University of Louisville, KY 40292, USA c College of Liberal Education, Keimyung University, Daegu 704-701, Republic of Korea d Department of Information and Communication Engineering, Yeungnam University, Gyeongbuk 712-749, Republic of Korea b

h i g h l i g h t s

 A comprehensiv comprehensive e review of all aspects aspects of solid state batteries: batteries: design, materials.   Tabular representations to underscore the characteristics of solid state batteries.  Solid state electrolytes to overcome the safety issues of liquid electrolytes.

a r t i c l e

i n f o

 Article history: Received 13 October 2014 Received in revised form 17 January 2015 Accepted 9 February 2015 Available online 16 February 2015 Keywords: Solid state batteries Solid electrolytes Lithium based batteries Lithiation/delithiation Solid electrodes

a b s t r a c t

Conventional Conventional lithium-ion lithium-ion liquid-electro liquid-electrolyte lyte batteries are widely used in portable portable electronic electronic equipment equipment such as laptop computers, cell phones, and electric vehicles; however, they have several drawbacks, including expensive sealing agents and inherent hazards of  �re and leakages. All solid state batteries utilize solid state electrolytes to overcome the safety issues of liquid electrolytes. Drawbacks for all-solid state lithium-ion lithium-ion batteries batteries include include high resistance at ambient temperatures temperatures and design intricacies. intricacies. This paper is a comprehensive review of all aspects of solid state batteries: their design, the materials used, and a detailed literature review of various important advances made in research. The paper exhaustively studies lithium based solid state batteries, as they are the most prevalent, but also considers non-lithium based systems. Non-lithium based solid state batteries are attaining widespread commercial applications, as are also lithium based polymeric solid state electrolytes. Tabular representations and schematic diagrams are provided to underscore the unique characteristics of solid state batteries and their capacity to occupy a niche in the alternative energy sector. ©  2015 Elsevier B.V. All rights reserved.

1. Introduction and background

The advent of solid state batteries must be understood in the context of the challenges faced by modern storage systems, especially cially Li-ion Li-ion batteries batteries.. Existing Existing Li-ion batteries batteries,, apart from the storage and active components, contain considerable quantities of  auxiliar auxiliary y materials materials and cooling cooling equipmen equipmentt [1] [1].. Loss Loss of batter battery y quality due to continuous charging and discharging cycles, �ammability, dissolution of the electrolyte, and from vehicle to grid utiliz utilizati ation on has been been anothe anotherr import important ant concer concern. n. Solid Solid state state

*  Corresponding

author. author. E-mail addresses:  [email protected]  (B. Son), [email protected] Son),  [email protected]  (S. Park). These authors contributed equally to this work.

**  Corresponding

1

http://dx.doi.org/10.1016/j.jpowsour.2015.02.054 0378-7753/ ©  2015 Elsevier B.V. All rights reserved.

batteries are being extensively studied and researched with a view to solving these problems [1,2] problems  [1,2].. Conventional batteries, e.g., the Liion battery, battery, usually usually consist consist of a liquid liquid electrolyte electrolyte,, which which helps helps transport Li ions to and from the cathode and anode [3 anode  [3 e5] 5].. However, this increases the chance of leakage of the electrolyte if any holes are present; this is one of the main drawbacks of the conventional Li-ion battery. Another problem inherent in the liquid electr electroly olyte te batte battery ry is the format formation ion of dendri dendrites tes of Li, which which make make it prone to explosion  [5,6]  [5,6].. In order to surmount these problems, a solid electrolyte can be positioned between the electrodes. This is the principl principle e underlyi underlying ng the solid solid state battery [3,4] [3,4].. The advantag advantage e of this this typeof typeof batt batter ery y isa redu reduct ctio ion n in the the net net weig weight ht and and volu volume me of  the battery, greater energy output, and easy transfer of Li ions, which which afford affordss better better ef �ciency   [3,5]. [3,5]. Solid Solid state state batter batteries ies also also exhibit some advantages over the other commonly known energy

300

 J.G. Kim et al. / Journal of Power Sources 282 (2015) (2015) 299e 299e 322

storage devices, capacitors  capacitors   [7,8]. [7,8]. The advantages lie in the very small self-discharge of the solid state batteries, minimal wear and tear, and yield of a more uniform output voltage  [7,8]  [7,8].. In recent decades, decades, solid solid state batteries, batteries, especial especially ly solid solid state state lithium lithium ion batteries batteries,, have been widely widely used [9e13] 13].. Ideall Ideally, y, a solid solid state state electrolyte should have high cation conductivity, with good mechanical chanical properties properties and good chemical chemical stability stability that cannot cannot be easily reduced reduced by the metal itself  [9,14].  [9,14]. Moreover, Moreover, owing to rapidly growing microelectronics and integrated optoelectronics circuits, there is an increasing demand for new, lightweight batteries with high-c high-cyc ycle le life life and high high energy energy densi density ty.. In a commer commercia ciall lithi lithium um ion ion battery, the liquid electrolyte carries the risk of explosion and  � re; moreover, the separators and packaging limit the size of the batteries. All these factors have contributed to the development of allsolid solid state state batteries batteries [14e16] 16].. However, solid state batteries also present challenges, such as their relatively lower power density, high ionic ionic resistanc resistance e at room temperature temperatures, s, and manufactur manufacturing ing cost   [3,15]. [3,15]. The atomic atomic layer layer deposi depositio tion n (AL (ALD) D) techni techniqu que e is an important method applied in the manufacture of solid state thin �lm electrolytes [17 electrolytes  [17e20] 20],, but it is rather costly and indeed forms the bulk of the cost of the battery battery  [3,21]  [3,21].. 2. Solid state state batteries batteries

A solid state battery is similar to a liquid electrolyte battery except in that it primarily employs a solid electrolyte. The parts of  the solid state Li ion battery include the anode, cathode and the solid electrolyte electrolyte [22,23] [22,23].. The anode anode is attach attached ed to copper copper foil, foil, which which helps improve the electrical conductivity of the battery.  [22]  [22].. During the charging cycle, there is movement of the Li ions of the LiCoO2  crystal toward the electrolyte interface [22,24] interface  [22,24].. As a result, the Li ions cross over to the carbon layers in the anode through the electrolyte. During the discharging cycle, the reverse process takes place, and the Li ions travel via the electrolyte toward the LiCoO 2 particles [22,23,25] particles [22,23,25].. Solid state batteries can overcome some of the inherent problems of liquid electrolyte batteries, being less hazardous and having a less �ammable electrolyte-electrode system and better storage capacity. In the  � eld of power supply for cardiac pacemakers with low-power requirements, all solid state batteries are well well establish established ed because because of safety, safety, lifetime, lifetime, and achievab achievable le energy density  [26,27]  [26,27].. As mentioned in a book, all solid state battery is one of new type of batteries with excellent safety and high energy energy density density [28] [28].. Substi Substitut tution ion of liquidelect liquidelectro roly lyte te by a solid solid allows simpli �cation of the cell structure, and many restrictions in terms of architecture and safety are eliminated  [29,30]  [29,30].. Solid state electrol electrolytes ytes are being intensively intensively researched researched as the key which which present present safety advantages advantages over present liquid liquid Li-ion Li-ion technolo technology gy [31e33] 33].. The non-�ammability of their solid electrolytes offers a fundamental solution to safety concerns and remarkable environmental compatibility compatibility [34  [34e38] 38].. Also the solid state electrolytes tend to last longer, as they undergo less wear and tear during operation, are more proof against shocks and vibrations, and can operate within a larger temperature range, up to about 200  C. However, they have several disadvantages as well [6] well  [6].. Solid state electrolytes, and consequently batteries, ries, are are not suitable suitable for use in low low and ambien ambientt temper temperatu ature re conditio conditions, ns, and the power power and current current output output is generall generally y less [39,40].. This is because of the large resistance of the solid oxide at [39,40] room temperature temperature to ionic ionic conducti conductivity vity,, whereas whereas this does not occur at elevated temperatures. In addition, at ambient or room temperature, the stress created at the electrode-electrolyte interface due to continuous contact with the solid electrolyte tends to reduce the longevity of the battery  [6,41]  [6,41]..  Fig. 1 is 1  is a schematic diagram of a lithium based solid state battery. The curved arrows indicate the movement of the Li ions during the charging and the

discharging process, respectively. The electrons produced due to the reaction are used to drive a load in the external circuit. The set of cathode and anode materials and their corresponding suitable electr electrol olyte ytess are also also given given in Fig.1 Fig.1,, marked marked with with matchi matching ng colors colors (in the web version).  2.1.  2.1. Structure of solid state batteries  2.1.1  2.1.1.. Cathode The cathode in the solid state battery is important, as it supplies the batter battery y with with the necess necessary ary ions ions during during the chargi charging ng proce process ss and vice versa during the discharging process. The cathode must be structurally structurally stable during this process. process. It is important that the ionic condu conducti ctivit vity y of the cathod cathode e be good, good, as the chargi charging ng and dischargi charging ng proce process ss invol involves ves the transf transfer erenc ence e of ions ions across across it [6,24,42].. Commo [6,24,42] Commonly nly used used cathod cathode e mater material ialss forlithium forlithium based based solid solid state batteries are lithium metal oxides, as they exhibit most of the above necessary properties. Lithium cobalt oxide (LCO), which has the stoichiometric stoichiometric structure LiCoO 2, is a widely used lithium metal based oxide. LCO exhibits a layered structure that is suitable for the lithiation/delithiation process, and it has a relatively high speci �c energy of about 150 mAh g1, which makes it a preferred cathode material [6,24,43] material [6,24,43].. LCO exhibits an octahedral arrangement with a layer of lithium atoms between oxygen and cobalt  [44,45]  [44,45].. However, ever, it is relative relatively ly costly costly to to manufactur manufacture, e, especial especially ly with with the use of  of  cobalt [45] cobalt  [45].. Lithium manganese oxide (LiMn 2O4) is another material used in the cathode of solid state batteries  [46,47]  [46,47].. This compound produces very little resistance to the passage of lithium ions during the lithiation and delithiation process, thanks to its spinel based based struct structur ure, e, which which makes makes it suita suitable ble for use [6] [6].. LiMn LiMn2O4 has has its its drawbacks too, notably phase change during the ion transfer process, cess, which which hinder hinderss stabil stability ity,, and a lower lower capaci capacity ty than than LCO. LCO. LiFePO 4, another lithium based phosphate, has the advantage of  being less hazardous and less expensive to produce than the other lithium lithium based oxide oxide materials materials   [6,48]. [6,48]. Moreover, Moreover, LiFePO4   has an olivine based structure (a one-dimensional chain of lithium ions), which greatly assists the transfer of ions and provides less resistance to the path of ion transfer [6,24] transfer  [6,24].. On the other hand, phosphorus has a high self-discharge rate, which reduces the longevity of this material. Apart from these lithium based oxides, vanadium based oxides have have also also been been teste tested, d, as they they exhib exhibit it simila similarr layere layered d struc structur tures es that that help during during the lithiati lithiation/d on/delit elithiati hiation on process process.. However However,, they produce low output voltages and sometimes they lack longevity, which has limited their use as cathode cathode materials materials in solid state batteries [6,49,50] batteries [6,49,50].. In fact, the overall performance of a solid state battery is still limited by the performance of cathode materials, as its speci speci�c capacity capacity is generall generally y lower lower   [14]. [14]. The applic applicati ation on of  nanoparticles to the typical cathode, such as LiCoO 2, can produce better properties, but they will react more strongly with the electrolyte at high temperature and lead to more safety issues than using such materials in the micrometer range  [14,51]  [14,51].. Coating the nanoparticles with a stabilizing layer can reduce this problem, but it also reduces the lithiation/delithiation rate of the cathode.  2.1.2.  2.1.2. Anode Materials Materials that can store Li/Li Li/Liþ to great capacity capacity are usually usually potentia potentially lly good anode anode material materials. s. This is because because the anode anode is where the lithiation takes place during the charging process. Pure lithium metal has been tried as an anode material. Unemoto et al. have used lithium anodes in lithium sulfur battery systems with room room temperatu temperature re ionic ionic lithium lithium (RTIL) (RTIL) liquid liquid fused with silica nanoparticle solid state electrolyte and they were able to achieve a discharge capacity capacity of 690 mAh g1 after 45 cycles of operation operation [52]  [52].. Cai et al. have used lithium metal anode with a LiMn 2O4  cathode

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

301

Fig. 1.  A schematic representation of a representative lithium based solid state battery, showing the direction of ion movement and some of the possible anode, electrolyte, and cathode combinations.

system and have obtained a 91.39% capacity retention after 1000 cycles of use at room temperature conditions  [53]. Taib et al. have incorporated lithium anodes in solid state cells having biopolymer (chitosan) based solid state electrolyte and have obtained a stable discharge capacity of 160 mAh g1 upto 50 cycles of operation [54]. However pure lithium metal anodes have their drawbacks. An important problem associated with lithium anodes is their inability to be used in high temperature operations, because of the lithium metal's very high sensitivity to elevated temperature conditions [55,56]. Accelerating rate calorimeter (ARC) results have shown that, as the number of cell cycles of operation rises, there is a considerable increase in the self-heating of the Li anodes because of  the increase in the surface area of the anodes due to the lithiation process [56]. According to Sacken et al., the primary reason for this failure is that the structural integrity and morphology of Li cannot be manipulated properly [56]. Therefore, several safety issues need to be kept in mind when using Li anodes. Another drawback of the dendrite formation effect is the development of non-uniform current density acrossthe cell, which in turn leads to local temperature gradients throughout the cell surface [57]. This makes it dif �cult to integrate the cell, especially in miniature devices, as the temperature gradient becomes hazardous [57]. Also, this dendrite formation steadily depletes the amount of lithium necessary for the cell cycling process and may even result in the short circuit of the cell [57]. Apart from uncombined metallic lithium, lithium based metal oxides such as lithium titanate (LTO) have also been used. The stoichiometry of lithium titanate is Li4T5O12. The main advantage of  LTO is its octahedral structure, which can easily integrate lithium ions within it during the lithiation process  [58e60]. Also, LTO does not undergo much structural change during the lithiation/delithiation process and is relatively stable, which makes it suitable for use. However, LTO has a rather low speci�c energy of 175 mAh g1, which sometimes limits its usage to areas where low power output is required   [41,59]. Carbon and carbon based materials are commonly used anode materials in solid state batteries  [61,62]. Graphite too is quite widely used as an anode material in solid state batteries, yielding several advantages, such as having a layered structure that can incorporate the lithium ions during the

lithiation/delithiation process, its ability to withstand large numbers of charging and discharging cycles, and relative ease of  manufacture [62,63]. Graphite can also easily be doped with other materials to improve its capacity. The major problem with graphite is its rather low capacity. Soft carbon is another material used. It bene �ts from having a disordered structure that is favorable toward incorporating Li ions [58,63]. Therefore, at present, graphitic carbons are commonly used for the anode of the commercial lithium battery because they have a higher speci�c charge and more negative redox potential than most metal oxides or polymers; further, they demonstrate better cycling performance than lithium alloy, thanks to high structural stability and less volumetric expansion in the lithiation reaction [14]. Fig. 2 is a schematic representation of the different solid state battery systems with their applications and respective capacities. Metallic alloy-based anodic systems have also been studied, notably Sn, Pb, Sb, Al, and Zn along with their alloy systems [4]. The principle behind their use is the ability of the alloy atoms to store multiple lithium atoms in the host crystal lattice. One of the bestknown tin-based anodes is Li 4.4Sn, which is the �nal phase after lithiation. A theoretical maximum gravimetric capacity of  959.5 mAh g1 has been calculated forthis system [5]. However, the primary problem with these tin-based anodes is that they suffer from large volume changes (and consequently large stresses) during the lithiation process, fracture takes place quickly, and hence their longevity is rather short  [5]. Silicon- and carbon-based composite anodes have been tested. Carbon is used to reduce the large volume expansion experienced by the Si anode during lithiation [5]. Wang et al. reported a gravimetric capacity of 794 mAh g 1 of  Si/C composite anode systems after about 20 cycles of operation [5,64]. Anodes in solid state battery systems tend to be subjected to large quantities of stress, which take its toll on their longevity. Several groups have studied the development of stress and its distribution in solid state battery anodes during operation. Mukhopadhyay and Sheldon have shown that stress in the Li-based anode due to the lithiation/delithiation process arises mainly from a change in lattice dimensions  [65]. Unalloyed metallic Li being relatively soft, these problems are magni �ed compared to anodic alloys. Some of the processes that lead to disintegration are plastic

302

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

Fig. 2.  The capacities of different types of batteries and their respective applications [41,159]. The diagram shows that thin  � lm Li ion based solid state batteries provide some of the best properties in terms of both energy density and capacity.

deformation, fragmentation, and fracturing [62,65]. Another factor is that the lithiation/delithiation results in a reduction in elastic modulus along the basal planes and there is a consequent increase in elastic modulus perpendicular to the basal planes. These unequal expansions and reductions of the Li-based anodes further make their integration dif �cult [66]. As a result of the stresses and mechanical changes, the energy necessary for the phase transformation is reduced, as is the ability of the Li anode to operate at elevated temperatures [65]. This leads to a number of losses during the running of the cell, especially hysteresis losses during the lithiation/delithiation process, which serve to decrease the overall output of the cell [65,67]. Huang et al. have developed a comprehensive mechanics model for understanding the stress generated during lithiation in solid state lithium ion battery anodes [68]. They determined that the important factors are plastic deformation of  the electrode and also stress due to Li rich and poor phases formed during the lithiation process [68]. During the delithiation process, however, tensile hoop stress is created, which results in contraction of the sample.  Fig. 3  illustrates the stress induced in a Si nanoparticle and its consequent cracking due to the lithiation process taking place. Silicon nanowires have been developed as lithium ion battery anodes, as they tend to contain less stress due to the

lithiation/delithiation process [69]. Sethuraman et al. have studied the development of stress in silicon anodes in-situ due to the electric potential in the cell during its performance  [67]. They reported that the Si anodes undergo cyclically varying tensile and compressive stresses of magnitudes sometimes as high as 1.5 GPa [67]. Based on the thermodynamic considerations obtained from the Larche-Cahn chemical potential theory, the authors demonstrated the existence of electrical potential-based stress in the anodic silicon and found the ratio of potential change to stress to be 62 mV GPa1 [67].  2.1.3. Con � guration of electrodes Various electrode and electrolytic con�gurations have been tested for solid state battery systems. For example, one of the  � rst new designs from the usual coin cell strategy was plastic-based LiPON electrolyte or PLiON, which provided a great deal of  �exibility and ease of use  [70]. Some other con �gurations that have been tried are the cylindrical, prismatic, and  � at. These have been considered to be the most versatile con �gurations, leading to optimum surface area contact between the anode/electrolyte and the cathode/electrolyte [70]. Another important factor is the dendrite formation in the Li anodes, owing to its close proximity to the

Fig. 3. Part (a) shows a TEM image of lithiation and consequent cracking of the Si nanoparticle. As can be seen from the i mage, the crystalline Si nanoparticle (c-Si) shares a common interface with the tungsten electrode on side and a Li counter electrode. During the lithiation process, as seen in part (b) the Si particle has cracked and the Li envelopes it quickly and even enters the particle forming a core shell of pure Si with an external amorphous LixSi. Thus it can be understood how lithiation causes cracks due to induced stress [355,356].

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

electrolyte and the consequent electrical reaction. Scrosati et al. were able to solve this problem by doping the Li anodes with other metals (Co, Ta); this con �guration was called the rocking-chair technology   [71]. Short circuiting due to dendrite formation is a practical problem which has been described by Nishi   [72]. According to him, dendrite formation causes the problem of short circuiting by penetrating into the electrodes and causing poor cycle performance  [72]. According to Song et al., the presence of long interconnect and channels allow paths for the dendrites to grow and proceed [73]. These, consequently lead to the reduction of cell ef �ciency and ultimately internal short circuiting. Therefore they have recommended the application of non-porous polymer membranes to prevent the growth of these dendrites  [73]. The choice of  the electrode/electrolyte con�guration is, therefore, very important to the proper functioning of the cell, and its optimization is absolutely essential. Further, thick  � lm technology  has been studied by several groups. This basically involves the deposition of layers of  composite materials from their precursor solvents; these composite materials are further ground and compacted with a polymeric binder to produce the necessary electrode. Kim et al. have used thick �lm technology to develop electrodes for microbatteries, which were deposited on metallic current collectors  [74]. A polymeric gel electrolyte consisting of PVDF-LiPF 6-propylene carbonate (PC)-ethylene carbonate (EC) was used to complete the cell  [74]. Vapor deposition technology to deposit thin �lm electrodes has been studied by Fleischauer et al.  [75]. They were able to deposit porous silicon thin �lms having a very high aspect ratio, which were approximately 500 nm in height. This resulted in a form of 3D microstructure for the thin  � lm batteries [75]. Teixidor et al. have demonstrated the formation of carbon pillars as microelectrodes for thin �lm lithium ion batteries   [76]. For this purpose, lithographic patterning coupled with thermal decomposition of crosslinked photoresists was used. The photoresist acted as the solvent and the curing of the dispersion was done by UV light  [76]. Interdigitated con�guration cells were studied and modeled by Zadine et al. [77]. They used �nite element analysis (FEA) for modeling and were able to study the rate constraints of the electrochemical processes, especially those due to collector resistance with this technique [77]. Composite ceramic electrodes, an important class of solid state electrodes, have been developed and studied by a number of  groups. Mizuno et al. have prepared composite ceramic electrodes by thorough grinding of the precursor materials LiCoO2 and Li2S.P2S5. [78]. The �nal step in the design of the composite ceramic electrode involved the pressing together of the Li based positive electrode, the In foil negative electrode and the electrolyte which were then held together by stainless steel current collectors. The authors have reported a discharge capacity of 75 mAh g 1 at a cut off voltage of 1.5 V [78]. Wei and coworkers have developed a metal oxide electrode for application in all solid state sodium rechargeable batteries [79]. A 22 mm thick P2Na2/3[Fe1/2Mn1/2]O2  cathode and 52 mm thick Na2Ti3O7$La0.8Sr0.2MnO3   composite anode prepared by solid state reactions was used, yielding a capacity of  152 mAh g1 at 350   C [79]. Delaizir et al. have studied the process of spark plasma sintering to develop ceramic electrodes for all solid state batteries [80]. They have developed monoclinic Li 3V 2(PO4)3  as both the anode and the cathode. Non uniform sized precursor materials were used to perform the sintering so as to achieve the highest degree of  compactness. The authors have mentioned that the compactness of  the electrolyte is important in controlling the electrical conductivity; an optimal ionic conductivity of 2.8 *  104 S cm1 was obtained   [80]. They have demonstrated rather promising surface capacity of 2.2 mAh cm2 for a cut off potential of 2.45 V  [80]. Electrode architecture is another important factor while “

“

”

”

303

designing the solid state battery. To have a better interface between electrodes and solid electrolyte, Soo et al. have developed rubbery block copolymer type electrode systems [81]. They have used these electrodes in Li based solid state battery systems and good cycling properties have been noticed under room temperature conditions [81]. Zhang et al. have developed columnar nanostructured tin oxide electrodes for Li ion rechargeable batteries  [82]. These tin oxides are essentially columnar grains with a coating of nanoparticles, 20 nm in diameter   [82]. A reversible capacity of  460 mAh g1 is obtained under ambient conditions at a current density of 0.3 mA cm 2 [82]. Li et al. have studied mesoporous anode systems which were made of Co3O4 [83]. These mesoporous systems demonstrated a capacity of 700 mAh g1 after 20 cycles of  operation which has been considered to be a reasonable result [83]. To facilitate electrochemical reactions, several approaches have been researched to have a good solidesolid contacts between electrode and electrolyte [35,36,84e90]. The achievement of close contacts and the increase contact areas are important aspects to have an effective chargeetransfer reaction as explained by Tasumisago et al. [84]. Using nanocomposites by a ball milling process, the surface coating of active material particles with thin �lms associated with a pulsed laser deposition (PLD) techniques, and using supercooled liquid of glass electrolyte have been established to have a favorable contact at the interface between electrode and electrolyte which leads to a good performance of all solid state batteries [36,84,85,89e92].  2.1.4. Electrolyte As the performance of a solid state battery depends on the diffusion of ions within the electrolyte, solid electrolytes are required to have high ionic conductivity and very low electronic conductivity and should exhibit a high degree of chemical stability [14,93]. Crystalline materials such as lithium halides, lithium nitride, oxy-salts, and sul �des have been found to be good as solid electrolytes. The most favorable features of the solid electrolyte are that there is no corrosive or explosive leakage and the chance of  internal short circuit is less, and hence it is safer   [14,94]. Solid electrolytes should have a suf �cient number of mobile ions to enable conduction to proceed smoothly. Solid state electrolytes should therefore have enough vacancies in their crystal lattice to permit the ions to move, and the overall activation energy must be low [94,95]. Different types of solid state electrolytes have been employed, based on their con �gurations and electrode/electrolyte materials setups. Therefore, solid state electrolytes are broadly classi�ed into two types, bulk solid state electrolytes and thin �lm solid state electrolytes. The primary distinction is on the degreeof thickness of  the electrolyte; thickness of bulk solid state electrolytes are usually in the range of several hundred micrometers, whereas that of thin �lm solid state electrolytes are in the range of hundreds of nanometers to several microns. Another primary area of difference between the two is the way they are fabricated: bulk solid state electrolytes are usually prepared by techniques such as mechanical milling, sintering and compaction, annealing and heat treatment whereas thin  � lm solid state electrolytes are fabricated by pulsed laser deposition, spark plasma sintering, CVD etc.

a) Bulk solid state electrolytes (electrolytes at the macroscale) Among the most commonly used solid state electrolytes are the solid polymeric electrolytes (SPE)  [96]. The main aim while developing SPEs is to obtain ionic conductivitiesas high as those of liquid electrolytes  [96]. If the polymer is obtained by complexing it with an inorganic salt, then it has a low lattice energy, which makes the resultant SPE more stable [96]. Frequently used polymereinorganic

304

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

salt electrolyte complexes include PEO eLiCF3SO3  and PEO-LiBF4 [96e98]. One of the important types of inorganic solid state electrolytes is the NASICON type electrolyte having the general formula LiM2(PO4)3,   where M is usually Ti, Ge, or Hf   [99]. However, an important drawback of this type of electrolyte is its steady degradation due to grain boundary effects that creep in Refs.  [99,100]. Polymer gel electrolytes have also been tried by several groups [101]. The primary advantage of this type of electrolyte is that the host of the polymer compound forms a relatively strong and stable bond with considerable amounts of organic liquid. This considerably enhances the cell performance [101]. Composite polymers are another class of polymeric systems that have been utilized  [102]. They consist of a distribution of nanosized granular materials as �llers in a larger SPE host. However some problems still do exist with polymeric systems like  � ammability and sometimes a lack of  signi�cant room temperature conductivity. Current research looks at overcoming these problems [102,103]. Solid state electrolytes with the garnet structure (general formula A3B2C3O12) are also used frequently   [100]. They include Li5La3B0 2O12, where B0 is usually Bi, Sb, Na, or Ta. These garnet type solid state electrolytes have shown ionic conductivities around 4  * 104 S cm1 [100]. LiPON-based electrolytes show the greatest promise because of their inherent safety, ability to transport charges quickly, and thermal stability. Vanadium-based electrolytes have also been studied [6,104e106]. These xerogel vanadium pentoxides contain sheets of  two vanadium oxide layers, with all the vanadyl bonds on the outside, leading to a distorted octahedral coordination around the lithium instead of the square pyramidal coordination found in crystalline V 2O5   [6,106]. MnO6   crystals are connected to one another in three dimensions by sharing of the edges. LiO4  tetrahedral shares each of its four corners with another MnO6 unit but is actually structurally unique  [107]. This leads to the formation of a three-dimensional network of octahedral and tetrahedral sites, whereby the Li ions can travel through the pores of the electrolyte. The different types of building units are responsible for the wide variety of properties in MDOs   [108,109].   MDOs exhibit excellent electrochemical properties of several manganese-oxide phases and are attracting much attention as positive electrodes in lithium batteries. Apart from the above mentioned inorganic solid state electrolytes, organic solid state electrolytes have also been studied. Kim et al. have developed polymer solid state electrolytes based on Polyethylene glycol (PEG) coupled with organic hybrids and plasticizers [110]. The authors have reported ionic conductivities of 4.5 5 * 10 S cm1 at room temperature conditions   [110]. Xuan and coworkers have developed a polymer electrolyte from ionic liquid (IL) and PEO and have reported a reversible cycle capacity of  140 mAh g1 for their solid state systems   [111]. Sulfur-based lithium solid state electrolytes have been tried as well. The most prominent is Li3.25Ge  0.25P0.75S  4 , which demonstrated considerable electrochemical stability and very good ionic conductivity of 2.2 * 103 S cm1 at room temperature [119,120]. Sul�des materials with extremely high lithium ionic conductivity of 12 mS cm 1 at room temperature was reported by Kamaya et al.   [112]. Y. Seion et al. reported that a heat-treated Li2SeP2S5  has an extremely high ionic conductivity of 1.7 *102 S cm1 and the lowest conduction activation energy of 17 kJ mol1 at room temperature [113]. The proper functioning of the solid state battery depends not only on its chemistry but also on its proper integration in the cell and its stability with respect to the cathode/electrolyte and anode/electrolyte interface. b) Thin  � lm solid state electrolytes (electrolytes at the nanoscale)

The most commonly used thin  � lm solid state electrolyte is the LiPON (lithiumephosphorus-oxy-nitride) electrolyte  [114e116]. It is usually a glassy electrolyte and is obtained by the RF magnetron sputtering process [114]. The N versus O ratio is the most important factor in the fabrication of this electrolyte and, with optimum N/O ratios, ionic conductivities as high as 1 e2  m S cm1 have been achieved  [114]. Current studies have also shown that polyethylene oxide (PEO) coupled with LiXF 6, where X is P, As, or Sb (PEO:LiXF 6), is a good ionic conductor  [117,118]. Conductivity and mechanical strength were improved further by the addition of inorganic  � llers such as Al2O3  or TiO2 [119]. The LIPON thin  � lms were obtained by RF magnetron sputtering of the Li 3PO4   target, using nitrogen gas [120]. The exact composition of the electrolyte obtained was Li3.3PO3.8N0.22, along with optimum conductivities of 10 4 S cm1 [120].  2.2. Charge transfer mechanism for a solid state battery

The charge transfer mechanism of solid state electrolytes is distinct from that of liquid electrolyte-based cells and batteries. Solid state electrolytes depend on diffusion of ions as the driving force behind the charge transfer mechanism. In a Li-based solid state electrolyte, the diffusion process results in Li ions from the electrolyte entering the cathode and anode by the lithiation process, as �rst reported by Whittingham and Yazami   [121,122]. Similarly, delithiation is the loss of Li ions from the particular compound (electrode, in this case). The charging involves delithiation from the cathode and lithiation in the anode via diffusion through the electrolyte. Likewise, during the discharging process, the opposite reaction takes place, involving the diffusion of the ions via the solid state electrolyte [122,123]. This lithiation process demands that the host species be able to accommodate the Li ions rather easily, and the cathode and anode should thus be selected accordingly. The higher the number of Li ions �owing into the electrodes, the greater the current that is produced, and the cell voltage is proportional to the Fermi potential difference between the anode and cathode [3,9,121]. This potential difference decreases with the working of the cell [9]. Conductivity of ions acrossthe solid electrolyte takes place usually by migration of mobile ions across vacancies or interstitials in the electrolyte  [124]. The presence of  defects in the solid electrolyte crystal lattice (Frenkel defects) further facilitates the ion hopping mechanism across the solid state electrolyte [124]. Other prevalent mechanisms include the interstitialcy or the knock-on   mechanism, in which a metal cation does not move unless it is able to move its neighboring ions and thereby create a path for itself  [124]. A theory that has been proposed is the random walk , in which there is occasional ion hopping that is totally random   [124]. The Li ion travels into an interstitial point in the host (the electrode, in this case) lattice and moves from one interstitial to another by the same mechanisms as explained above [125,126]. Li ions usually diffuse via this mechanism because of their small size and the relatively less energy needed to move from point to point  [3]. There must be holes or vacancies in the host (electrode) lattice for diffusion of Li ions to take place through them. A schematic representation of the diffusion process for the Li ions across the grain boundary is shown in Fig. 4. This demonstrates that a concentration gradient is created at the grain boundary, which helps in the grain boundary diffusion process. This method of lithiation, however, requires much larger activation energy than other mechanisms of diffusion [126,127]. Charge transfer at the interface of the solid state electrolyte and electrode is also another important phenomena that needs to be considered. An important way in which the interface can be distinguished is by the equilibrium of the charged species across the interface   [128]. Interfaces across which the charged species “

“

”

”

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

305

Fig. 4.  A formation of the concentration gradient of Li ions at the grain boundary with time as the cell undergoes its charging and discharging cycles. This concentration gradient acts as the driving force for the diffusion of Li ions at the grain boundary interface  [127,357]. The initial concentration gradient is denoted by C 0 and the thickness of the grain by  d . A 2D structure is shown here with x and y directions as marked.

tend to equilibriate and these are called non-blocking  interfaces and those across which the charged species do not equilibriate are blocking interfaces   [128]. An example of the former is the Ag/ Ag4RbI5  interface, whereas C/Ag4RbI5  constitutes a blocking interface. In a non-blocking interface subject to an applied potential difference,the rate of cation transfer from electrode to electrolyte is equal to the rate of metal deposition on the electrode In the blocking interfaces, this equilibrium rate is not reached  [128]. The �ow of charge across the interface is a function of the difference between the thermodynamic potentials at the electrolyte and the electrode surfaces respectively. The movement of the charged species may occur by convection or diffusion as the case may be [128,129]. When other external impulses, such as applied potential does not exist, the charge transfer takes place simply by a chemical potential gradient [129]. An important parameter to consider here is the Debye length (r d).The spread or distribution of the charged species across the interface is determined by the size of r d  and a uniform spread of the charge occurs only when r d  is greater than the ionic radii  [128]. Solid electrolyte interfaces are an important aspect of the charge transfer process. They usually start to form on the surface of the electrode due to the gradual decomposition of the electrolyte. The SEI layer acts as a passivating agent preventing any further degradation of the electrolyte and the electrode. The SEI also helps in improving the cyclic performance of the electrodes. Andersson et al. have shown the formation of lithium based SEIs having the general formula LixPFyOz  and Hu and coworkers have shown the formation of SEIs in Cr 2O3  based systems [130,131]. “

”

 2.3. Design considerations in selection of a solid state electrolyte

The electrolyte is one of the most important parts of the solid state battery.Solidstate electrolytes are known mainly for their fast and ef �cient ion transport properties. It has also recently been shown by Gadjourava et al. that solid state electrolytes should usually have a crystalline structure for a high ionic conduction rate [132,133]. The following are some of the important general parameters that must be considered in selection of the electrolyte. (i) Chemical and thermal stability:

The solid state electrolyte is usually held between the anode and the cathode, and unwanted reactions must be avoided at the interfaces of the electrodes with the electrolyte  [96]. Also, the electrolytes must be able to work perfectly in an electrochemical regime from 0 V to about 4 e5 V   [96,134]. Therefore, chemical and electrochemical stability is an essential property in a solid state electrolyte. Versatility of the solid state battery dictates that it must be able to work in a broad temperature range. Usually, lithium based rechargeable batteries used for military purposes are sub jected to a temperature range of 50  Cto80  C, and the electrolyte must be able to function perfectly in this temperature range without showing any thermal degradation [96]. Thermal stability is thus another important factor that must be considered when choosing the electrolyte. (ii) Ion transference number: The ion transference or transport number indicates the contribution of the different ions toward the total electric current carried by the electrolyte [96,135]. The iontransference numberof the solid state electrolyte should be close to unity or even, ideally, exactly unity [96,136]. This is because the mainpurpose of the electrolyte is to allow the �ow of the ions (cations) and prevent any electrons from traveling. However, in reality, the ion transference number is usually about 0.5, indicating only half of the necessary ionic species moving through the electrolyte  [96,137,138]. Therefore, the electrolyte selected should have an ion transference number as close to unity as possible, leading to ef �cient transport of the cations, reduction in the concentration of polarization, and consequently higher power density [96]. A general comparison can be made here, for example, a solid polymer-based lithium salt such as (PEO) 20LiClO4  has a lithium transference number of 0.31, whereas that for PEO-based LiAsF6 exhibits an ion transference number of 0.44  [139]. However, suitably prepared gel polymer �lms have shown ion transference numbers greater than 0.95, as demonstrated by Zainol et al.  [140]. Ghosh and coworkers have prepared a solid lithium-based polymer electrolyte with the high ion transference number of 0.9 at room temperature [141]. Ion transport numbers of liquid electrolytes are slightly higher, that of molten sodium chloride being

306

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

approximately 0.5 [142]. (iii) Mechanical strength and stability: The solid state electrolyte will have to withstand clamping forces when it is  � xed between the anode and the cathode. It will also have to bear the wear and tear of regular usage. It must work not only in a smaller laboratory setup but must also be able to reproduce its functionality when scaled up for industrial usage [96,143]. It must therefore have high mechanical strength and stability.  2.4. Processing routes for solid state electrolytes

Engineering the microstructure of the solid state electrolyte is vital, as this gives optimum properties and also helps in maintaining the properties of the electrode/electrolyte interface without steady degradation or corrosion.

obtained under ambient conditions [151]. A lithium based solid state electrolyte created by the process of  hot press sintering at 1230  C and having the composition Li7La3Zr2O12 (LLZ) has been developed and studied by Kotobukiand coworkers [152]. The structure of the resulting solid state electrolyte resembles a garnet. The researchers have reported an optimal electrical conductivity of 1.8 *  104 S cm1 at room temperature conditions [152]. Sul�de and oxide glass electrolytes have been used by rapid melt quenching and mechanochemical techniques  [153e156]. A. Hayashi et al. showed the successful preparation from a mixture of  crystalline Li2S and P2S5, using a mechanical milling technique, and also compared with melt quenching in 2001 [154]. An oxide glassceramic electrolyte of Li2.9B0.9S0.1O3.1 with high Liþ ion conductivity and low melting property was developed by mechanical milling and subsequent heat treatment at 290   C by M. Tatsumisago et al. [153]. b) Processing routes for thin  � lm solid state electrolytes:

a) Processing routes for bulk solid state electrolytes: Garnet type solid state electrolytes such as Li 7La3Zr2O12   have been fabricated at extremely high temperatures (1500 K) by Murugan et al. [144]. However, this very high temperature fabrication is a practical drawback. Sol egel synthesis techniques have been tried with varying degrees of effectiveness   [144]. Laser annealing is another important fabrication technique, which has been found to have the advantageous ability to be localized and thoroughly manipulated [145]. Inorganic solid state electrolytes such as LLZO have been prepared by the laser annealing technique. Precursors used are Li 2CO3, La2O3, and Zr(NO3)2.6H2O immersed in HNO3  to form a solution [146]. Consequent  � ltration followed by subjecting to intense laser pulses results in the formation of the required LLZO solid state electrolyte powder   [146]. The spark plasma sintering technique has been used by Chang et al. to develop Al substituted LiHf 2(PO4)3-based solid state electrolytes   [147]. The precursors used were Li2CO3, HfO2, NH4H2PO4, and Al(NO3)2.9H2O. XRD patterns have indicated a crystalline structure similar to NASICON, and the SEM microstructure shows a predominantly porous microstructure  [147]. Electrical tests have shown that spark plasma sintered samples have an optimum conductivity of 2.6  *  105 S cm 1, as against 6.5 * 106 S cm1 for conventionally sintered samples, which proves the advantage of the process [147]. Polymeric solid state electrolytes have also been studied in detail and various mechanisms have been employed to prepare them. Sequira and Hooper have prepared lithium-based polymeric electrolytes having the empirical formula (PEO)x-LiCF 3SO3, using a sol egel technique, and they have studied its properties in the range of 100   Ce170   C [148]. Plasticized polymers based on PEO eLiN(CF3SO2)2  have been prepared by condensation with tetraglyme and the resultant polymeric solid state electrolyte has conductivity of 4.74 * 104 S/ cm at room temperature conditions [149]. Cold compaction method to prepare the solid state electrolyte is another important technique that has been tried by several groups. Feng et al. developed a Li1.5Al0.5Ge1.5(PO4)3   (LAGP) solid state electrolyte by the method of mechanical milling, calcination and cold compaction [150]. LiPF6 was used as the electrolyte along with Li metal as the counter and reference electrodes. The researchers have shown electrical conductivity of the as prepared LAGP sample up to 3.5  * 106 S cm 1 and stability up to 7 V  [150]. Jak et al. have developed a ceramic based solid state lithium electrolyte using static as well as dynamic compaction [151]. The authors have noted a difference in Li ion conductivity; the dynamically compacted system gave a higher conductivity of up to three orders of magnitude. Overall, an optimal ionic conductivity 2 *  104 S cm1 was

The pulsed laser deposition (PLD) technique has been tested by Nakagawa et al. to develop a solid state electrolyte, especially thin �lm lithium-based solid electrolytes   [157]. An ArF excimer laser was used for the purpose and thin �lms were obtained, which gave optimal electrical conductivities of approximately 4.1  * 107 S cm1 at room temperatures [157]. Kim et al. have investigated the use of  magnetron sputtering to fabricate thin  � lm solid state Li eBeOeN electrolyte systems  [158]. Glassy amorphous thin �lms were obtained and characterization results showed that the amount of N in the �lms decreased with the addition of more Li in the composition [158]. As can be seen from the above review, different fabrication processes are possible for varying types of solid state electrolytes, and they will discussed in more detail in the subsequent sections. 3. Thin  � lm solid state batteries

Nanomaterials and thin  � lms are capable of suitable utilization, thanks to their enhanced properties. Nanoscale electrolytes prepared in the form of thin  � lms offer high energy density (research on active electrode and electrolytic materials currently puts the theoretical energy densities at about 200 Wh kg1), a longer lifetime,  � exibility, and extreme lightness, which tends to reduce the overall weight of the system   [159]. Thin �lm battery systems, especially Li based systems, are already in applications in zero emission vehicles, in the aerospace sector, in military facilities, and in medical instrumentation [159,160].  3.1. Background

One of the earliest thin �lm solid state batteries to have been studied was that based on the Li electrolyte, produced by Hitachi Co., Japan [161]. The battery was called the all-solid state thin  � lm battery  and incorporated a TiS 2 cathode, a metallic lithium anode, and a Li3.6Si0.6P0.4O4   thin �lm electrolyte prepared by the RF sputtering technique   [161]. However, it was not immediately deployed commercially, as it was not suf �cient for the larger electronic devices prevalent at that time [161]. NTT Co., Japan, achieved further advances in thin  � lm solid state batteries by developing a Li3.4V 0.6Si0.4O4 glassy electrolyte based solid state battery by the RF sputtering technique [162,163]. Eveready Battery Co. and Bellcore Co. have developed solid state batteries using sul�de glasses (Li4P2S7 and Li3PO4eP2S5) as electrolytes [164]. More recently, Baba et al. developed the rocking chair  type solid state battery using LiPON electrolyte, LixV 2O5   anode, and LiMn2O4   cathode, by RF “

”

“

”

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

sputtering   [165]. They have also developed solid state batteries without using a Li anode (for example by using a V 2O5 anode) and have been able to obtain capacities of approximately 10  m Ah cm2 [165,166].  3.2. Thin �lm solid state battery electrodes

Electrodes of thin �lm batteries play an important role in the proper performance of the battery; hence, it is essential to choose the right electrode materials for solid state batteries   [159]. The most important factors to be kept in mind when selecting the materials are their capacity for energy storage, a considerable contact surface area with the electrolyte for the electrochemical reactions to take place effectively, and the ability to support lithiation/delithiation, especially if the battery to be designed will be Li based [159]. For anodes, materials that have good Li storage propertiestend to be favored.Alloys of Al,Sn, and Si are suitable choices, as they provide good speci�c capacity (a LieSi alloy, Li4.4Si, provides a theoretical speci�c capacity of 420 0 mAh g1, versus 3600 mAh g1 for Li metal) and have reversible electrochemical reactions with lithium [167e170]. Initially, thin  � lm Li based solid state battery systems used transition metal oxides or chalcogenides as anodes, but these materials usually had reduced redox potential values. These materials included WO 2, TiS2, and MoO2  [71,171,172]. Since then, several other materials have been tested to develop suitable anodes for solid state batteries. Amorphous carbon has been used, as its processing temperature is lower and it provides better storage capacity than graphite   [173,174]. It has also been hypothesized that the addition of aluminum and gallium to graphite will improve capacity, as they form solid solutions. Alloying of carbonaceous anodes with transition elements such as vanadium and nickel are also being studied for the purpose [175,176]. A good cathode material is expected to have light weight, superior energy density, very limited self-discharge, and good cyclical capacity [159]. Commonly used cathode materials include LiCoO 2, LiMnO2, olivines such as LiFePO 4  and metal chalcogenides such as TiS2  and V 2O5  [177,178]. Transition metal dioxides of the general empirical formula LiMO 2 (where M ¼  V, Cr, Fe, Co, and Ni) are very common. Structurally, they resemble NaCl; however, the structure appears a bit distorted and the parent structure is the a-NaFeO2 type [179e184]. The lithium and the transition metal atoms are positioned at the octahedral and the interstitial sites, resulting in an overall layered structure. It is this structure that provides thermodynamic stability and good electrical capacity, making these suitable as cathode materials. Doping of LiCO 2 has also been tried with elements such as Mg, Al, Cu, Ni, or Sn to improve its reversibility and lower the processing cost, with good results [185,186]. Li based phosphor olivines (e.g., LiFePO 4) have been used as cathodes in solid state thin  � lm battery systems. The advantages of LiFePO4 are its high energy density, thermal stability, relatively cheap processing cost, and high theoretical capacity (170 mAh g 1) [187,188]. However, it has relatively low electrical conductivity (~10 9 e ~1010 S cm1), and efforts are therefore underway to improve its properties by further processing and alloying techniques [188]. The effect of external/manufacturing pressures also play an important role in the fabrication of thin  � lm batteries. As stated by Patil et al., lower manufacturing pressures result in a good electrode porosity, however, it causes rather poor connection between the particles [159]. On the other hand, high pressures provide for low porosity and better contacts between the particles. It has been noted that thin �lm electrolytes demonstrating low porosity show greater stresses due to Li ion insertion/deinsertion   [159]. Therefore the application of an optimum working pressure is necessary to for the best cell working  [189]. Sometimes, mathematical and analytical

307

models have been utilized to optimize the external pressure for the best cell output [189]. A schematic representation of a cross-section across a thin  � lm solid state battery is given in  Fig. 5, which shows the arrangement of the cathode, anode, and the respective current collector layers and how they are aligned with respect to the substrate.  3.3. Thin �lm systems

Lithiumeoxygen and nitrogen based polymer thin �lm electrolytes have been widely studied for their use in solid state batteries  [158,190,191]. Notable in this regard is the work on Li ion conducting thin  � lm solid state polymeric electrolytes by Kim et al. [158]. The solid state electrolyte obtained was a glassy Li eBeOeN thin �lm. The sputtering target was made by sintering and had compositions of Li2OeB2O3   (1LBO), 3Li2OeB2O3   (3LBO), and 5Li2OeB2O3  (5LBO). The sputtering was done on Al substrates in nitrogen atmospheres [158]. Electrochemical studies have shown that the 3LBO based  � lm has maximum ionic conductivity of 2.3  * 106 S cm1 at room temperature [158]. A Si eV anode based thin �lm solid state battery was fabricated by Lee et al. [192]. It had the con�guration Si0.7V 0.3/LIPON/LiCoO2 and has demonstrated a capacity of 50  mAh/cm within a range of 2 V and 3.9 V  [192]. Wu et al. developed Li based thin  � lm electrolyte for solid state battery applications   [193]. The target consisted of a LieTieSiePeO �lm, which was fabricated by cold pressing. The precursors of the target were Li2CO3, SiO2, TiO2, and NH4H2PO4   in the molar ratios of  1.3:0.6:4:5.4   [193]. RF magnetron sputtering was used for the fabrication process of the thin �lm solid state electrolyte  [193]. Characterization has shown that the thin �lm had a smooth, �awless surface and a thickness of 2.6  m m [193]. EDX analysis gave the approximate formula of the thin �lmasLixTi2Si0.32P3.6O7.16N2.52. XRD data showed the mostly amorphous nature of the electrolyte. A room temperatureconductivity of 9.2 * 106 S cm1 was obtained [193]. Among the non-Li based solid state thin �lm batteries, Aberouette et al. studied the performance of Ag deposited on Ag-doped germanium chalcogenide thin �lm solid state electrolyte system [194]. Si was used as the substrate and Si 3N4  was deposited on it using CVD to a thickness of 1800 A. After this, the surface was thoroughly cleaned and a further 300 A thickness of chalcogenide glass and 150 A thickness of silver were deposited on the substrate. The chalcogenide obtained had different compositions: Ge 20Se80, Ge30Se70, Ge33Se67, and Ge40Se60   [194]. The entire system was irradiated by a light of intensity 4.5 mW cm2 to photo-dissolve the Ag in the chalcogenide �lm [194]. Structural characterization results showed the presence of Ag deposits on the �lm surface and uniform dendritic morphology. Also, electrical tests indicated a signi�cant difference in the local resistivity in the Ag rich and the Ag poor regions. The authors suggested this to be a simple method of developing thin �lm solid electrolytes for solid state battery applications   [194]. Barium cerate/zirconate perovskite (BCZY) based solid state thin �lm cells have been studied by various groups. A primary reason for the increased interest in this class of  materials as solid state electrolytes is their relative electrochemical stability and considerable ionic conductivity [195,196]. Co-pressing and co-�ring methods and spray methods to develop the cells have been extensively researched and are in common use  [197e199]. Barison et al. studied the performance and analysis of doped barium cerate thin �lms having the formula BaCe 0.65Zr0.2Y 0.15O3d  (BCZY) as solid state electrolytes  [200]. The sputtering was done on a commercially obtained BCZY target in an argon and/or oxygen atmosphere with working pressure between 0.5  * 102 mbar and 2 2 * 10 mbar. Deposition rates of 0.7  mm h1 were obtained, and the thin �lms obtained had a thickness of between 2 mm and 10 mm

308

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

Fig. 5.   Cross-section of a thin  � lm solid state battery showing the different components. The substrate acts as a support on which the thin

[200]. The substrate was chosen to be Ni-BCZY and NiO-BCZY. SEM studies have shown that thin �lms deposited at higher pressures (2.1 * 102 mbar) tend to show more crystallinity than those deposited at lower pressures (5 *  103 mbar). XRD studies also indicate the crystallinity of the sputter deposited �lms, although higher temperature deposited �lms (400  C) tend to show a greater degree of crystallinity, as seen from the sharpnessof the XRDpeaks, which can be attributed to better phase formation at higher temperatures  [200]. The authors successfully demonstrated sputter deposition of BCZY solid state electrolytes on porous substrates, which can have solid state battery applications  [200]. Zhou et al. reported progress in research on nanostructured thin �lm electrodes for lithium batteries. The thickness of a typical nanostructured thin  � lm is less than 200 nm [201]. Thin  � lm electrodes do not require additives to enhance conductivity or structure. The resistance of the �lm is signi �cantly reduced by the reduction in thickness [201]. The dense structure of a thin �lm contributes to its durability. Thin �lm electrodes can be produced by magnetron sputtering, pulsed laser deposition (PLD), electron beam evaporation, chemical-vapor deposition, electrostatic-spray evaporation, chemical-vapor deposition, electrostatic-spray deposition (ESD), or solegel fabrication [201]. Chiu et al. used the sputtering technique to develop solid state batteries having LiNiO2 and LiNi 0.8Co0.2O2 polycrystalline electrolytes [202,203]. Nickel as a component in the thin  � lm electrolytes was also tested by Xia et al., who developed LiNi0.5Mn0.5O2   electrolytes by the PLD technique   [204]. Similar work on micron sized solid state batteries has also been done by Kushida et al.  [205], showing that a number of micron sized batteries placed on a Si substrate have a uniform output capacity of  approximately 9.2 mAh cm2 for 100 cycles at 3.6 V  [205]. A lithium silicate based solid state battery was developed by Nakagawa et al., who employed the pulsed laser deposition (PLD) technique to fabricate the thin �lm solid state electrolyte. They reported an ionic conductivity of 2.5 * 108 S cm1 at room temperature for the Li2SiO3  thin  � lm [157]. The electrolyte was characterized as amorphous in nature. A Li anode and LiCoO 2  cathode were used for the study [157]. Other techniques used to fabricate the electrolytes for thin  � lm solid state batteries include the electrostatic-spray deposition technique and the DSM-Soul�ll process   [206]. The electrostaticspray deposition process has been employed to prepare nanocrystalline ceramic powders. The Soul�ll process involves the integration of ceramic in a very high molecular weight polymer [206]. This matrix is then extruded to produce the thin �lm. The process has been used to produce BPO 4$yLi2O thin  � lm solid state electrolytes  [206]. The latest improvement in the thin  � lm battery solid state battery technique was achieved by Kuwata et al. This battery involved having a LiCoO2  cathode, Li 6.1V 0.61Si0.39O5.36  solid state glassy electrolyte, and SnO based anode  [207]. The entire fabrication was done simply by the sequential PLD technique [207,208].

 � lm

is usually deposited [159].

A novel technique is the use of MoO 3  nanobelts to enhance the performance of lithium based solid state batteries   [209]. The rationale for the choice of MoO 3   is its high structural anisotropy [209], which helps the lithiation process. The authors showed a capacity retention as high as 92% after 15 cycles of operation, visa-vis the 60% retention in non-MoO3  lithiated solid state batteries [209]. LiMnPO4 nanoplates were grown on the cathodes to enhance the performance of Li based solid state batteries   [210]. The high theoretical energy density of LiMnPO4   (701 Wh kg1) was the motivation for this study [210]. The LiMnPO4 was grown via a solid state reaction using oleic acid surfactant, yielding a uniform potential of 4.1 V and a speci�c capacity of 168 mAh g 1 [210]. The feasibility of Cu in solid state battery systems has been studied by several groups. An important piece of work in this regard was conducted by Dahm and coworkers  [211]. They used a complex sulfonium iodide coupled with cuprous iodide as the solid state electrolyte. Maximum conductivity of 1.3 S cm 1 was obtained for the electrolyte having the formula 4-methyl-1,4-oxathianium iodide [211]. Another improvement in the area of solid state batteries involving copper components, especially at the nanoscale, is the development of CuSbS2   nanobricks by Zhang et al.  [212]. The nanobrick electrolyte was prepared by the hot injection method using an oleylamine surfactant   [212]. Discharging and charging capacities of 1090 mAh g1 and 761.6 mAh g1 were obtained, respectively, but they then deteriorated with the lifecycle of the battery, and the capacity after 50 cycles of operation was only about 85.7 mAh g1 [212]. Sodium based thin �lm solid state batteries have also been studied. Early work in this regard was done by West et al. and Munshi et al., who developed sodium based anodes such asNaxCr3O8 [213,214]. However, even though these cells performed their operations, the output voltages obtained were not satisfactory. More recently, cells utilizing P(EO) 8NaCF3SO3  as solid state polymer electrolytes and Na15Pb4  as anodes have achieved some success [215,216]. However, issues remain, such as the rapid drop in capacity after relatively few cycles and the irreversible phase change of the electrodes. These still need to be overcome  [217]. 

 3.4. Challenges and current status

A lot of research and developmental work is still going on in the of Li ion batteries to improve its performance. A cyclical performance of 10,000 cycles without much drop in cell output and potential has been noted, indicating good performance [218,219]. It has also been proposed that using stacks of cells instead of single cell can help improve the capacity of the cells. However thin  � lm systems present certain drawbacks related to the electrolyte thickness and integration which prevent them from achieving their full potential. One important drawback of the thin  � lm battery system is that its performance may be constrained by the nature of its geometry [159]. The current that can be drawn from the thin  � lm battery is �eld

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

309

 Table 1 The properties of Li based solid state battery systems, along with their open circuit voltages and electrical outputs.

Anode Li alloy Indium foil Li metal Li metal  Anode: Li metal Molten Li Indium Li metal

Cathode Li4Ti5O12-carbon  � ber composite Composite consisting of LiCoO2, acetylene powder and electrolyte powder LiCoO2 Li metal Solid state composite air cathode Cu0.1V 2O5 painted on the electrolyte surface LiCoO2 Li foil

Electrolyte

OCV

Electrical output 1

Ref.

70Li2Se29P2S5eP2S3 glasses Li2SeP2S5

1.8 V 2V

140 mAh g 90 mAh g1

[240] [39]

Li5La3Ta2O12 Lithium borosilicate glass Lithium based glass ceramic Al doped Li7La3Zr2O12 (100-x) (0.6Li 2S. 0.4SiS2). xLi4SiO4 Polyethylene oxide based lithium (LiFePO4)

3.5 V 3V 2.6e3.6 V 2e3.5 V 3.1 V 3.45 V

83 mAh g 1 N/A   11.34 mAh 176 mAh g1 90 mAh g 1 140 mAh g1

[247] [248] [249] [255] [257] [312]

highly dependent on the geometry and interfacial contact at the electrolyte/cathode and the electrolyte/anode interface   [159,220]. A thin �lm electrolyte presents a rather �at surface to both the anode as well as the cathode side, which constitutes an apparent drawback compared to conventional bulk electrodes, where the surface roughness of the solid electrolytes enhances the net surface area available, thereby increasing output current  [159]. Another important factor that affects the output and integration of thin �lm systems is the lack of thickness of the electrolyte and the interfacial resistance that occurs at the electrode/electrolyte interface  [221]. Tarascon and Armand have stated that too thin electrolytes tend to have large amount of volumetric expansion which has a negative effecton the working of the battery [70]. Also minor mistakes in the correct positioning of the electrodes and electrolytes result in a large drop in cell potential and contaminations of both the electrodes and the electrolyte [222]. An interlayer has been developed to help prevent this from occurring  [221,223]. Understanding the type of secondary reaction taking place at the interface, which is most common at high temperatures (thereby limiting cell performance), is another priority, as this may detract from the cell's functionality at high temperatures  [159]. One of the major integration problems encountered by thin �lm solid state batteries is the loss of charge retention capacity. This is because, after the  � rst charging/discharging cycle, the solid electrolyte interface is formed, which reduces the charge retention capacity  [159]. To overcome this problem, a proportional mixture of cathodic and anodic material is required. Another problem during manufacturing and integration is low pressure during manufacturing, as high pressures cannot be applied, owing to the inherent thin nature of the solid state electrolytes [159]. This results in weak contacts between the various components of the cell, even though the electrode porosity is reasonably good. However, this problem can be ameliorated by selecting a proper pressure during the fabrication process  [159]. Other challenges of thin  � lm solid state batteries are a considerable volumetric change in the electrodes takes place during the charging-discharging process, leading to loss of material. If this can be avoided by changing the cell chemistry or better engineering design, then the output of the solid state cells can be improved further. Solid state thin �lm battery fabrication relies to a large extent on sputtering techniques, which are costly and hard to scale up during production times [224]. Also, it has been dif �cult to apply everything that has been obtained during the R &D stage to actual industrial outputs  [224]. However, further progress is expected in this  � eld soon. A target has been set of achieving 200 Wh kg 1 as capacity and 500 Wh l1 or more within the next few years [225]. It is expected that solid state cells, especially lithium based cells, will have several advanced applications such as biometric scanners and smart cards [159,225].

4. Lithium based solid state batteries

Much research has concentrated on the study of lithium and

 

lithium ion based solid state batteries, mostly devoted to the study of different lithium ion based electrolyte systems and how the solid state batteries perform with these systems. Lithium based solid state electrolytic systems fall into the following categories.  Table 1 presents a summary of the types of anode,cathode,and electrolytes used in solid state batteries and their respective open circuit voltage and electrical outputs. 4.1. Glassy electrolyte systems 4.1.1. Background Glassy electrolyte systems have received a lot of interest in recent years, starting with the chance discovery by Kunze of the AgIeAg2SeO4  glassy solid state electrolyte system  [226]. Conductivities of approximately 102 S cm1 were obtained at room temperature conditions  [226]. Further development in this area was accomplished by Malugani et al., who developed solid state glassy electrolyte systems of the type Li2SeP2S5 doped with LiI, and obtained ionic conductivities in the range of 10 3 S cm1 under room temperature conditions [227]. However, the solid state electrolyte was very hygroscopic in nature, and this affected the proper functioning of the cell   [227]. Subsequently, Akridge and Vourlis succeeded in introducing thiosilicate glasses into their solid state systems [228,229]. Levasseur achieved another important advance by developing thin �lm lithium borate based glassy electrolytes with amorphous titanium oxysul�de as the cathode   [230]. Considerable further developments have occurred in this area, especially by utilizing Radio Frequency (RF) sputtering to develop oxynitride electrolytes   [231]. On similar lines, Jones and Akridge have more recently developed solid state oxo-thio glassy solid state systems, which have provided ionic conductivities of approximately 2  *  105 S cm 1 [232]. 4.1.2. Phenomenon of ion transport  Anderson and Stuart proposed the theory of ion transport in their seminal work in 1954 [233]. Further interpretation and study of this model has been undertaken by several researchers, notably Martin and Angell [234], who suggested that the ionic transport mechanism consists of a simple ion hopping from one site to another based on the energy gradient  [234,235]. The two most important energy terms involved in this equation are the electrostatic binding energy (Eb) and the electrostatic strain energy (E s) [235]. The Anderson eStuart model has been highly successful in explaining ion transport in glassy solid state electrolytes by taking into account some very important parameters, viz. the elastic modulus (G) of the system, the Madelung constant ( b), which is a measure of the separation between the ions, and the covalency parameter (Y), which is a measure of the charge neutralization between the solid state ions and their immediately adjacent neighbors [235]. The most notable application of the Anderson eStuart model has been the successful explanation of iontransport in glassy Li based iodides such as LiI  [235].

310

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

However, some signi�cant drawbacks remain. The model does not explain the presence of vacant cationic sites in the electrolytes. Neither does it fully explain the preference of the ions to occupy particular sites in the glass lattice [235]. These anomalies have been resolved by the model proposed by Bunde et al., which is known as the dynamic structure model  [236]. This model proposes that cations have their own particular pathways and thereby make their own sites in the glassy lattice. Relaxation of the lattice structure results in modi �cation of the position of the empty sites, so that theycan be taken upby the next batch of moving cations[236]. This dynamic structure model has proven its usefulness by helping explain the ion exchange in silicate and alumino-silicate glasses [236]. Further advances in this area have been achieved by Funke et al., who proposed the uni �ed site relaxation model   [237,238]. This takes into account the entire spectrum of ion hopping from dc to infrared frequencies [238,239]. Sulfur and phosphorus based glassy solid state electrolytes have been tested for applications in lithium based solid state batteries   [240e242]. Tatsumisago and Hayashi studied the performance of lithium, phosphorus and sulfur based glassy electrolytes in solid state batteries  [240], and were able to obtain glassy electrolytes with increased ionic concentrations by mechanical milling and twin roller rapid quenching methods. Another technique adopted by researchers to improve ionic conductivity has been to add small quantities of lithium orthooxosalts (LixMOy)  [240]. Conductivity values of 5 *  103 S cm1 have been obtained for these glass based electrolytes at room temperatures under open circuit conditions, equating to about half  the room temperature conductivity of liquid electrolytes   [240]. Also, during the solid state battery operation (i.e., at a constant current density of 12.7 mA cm 2 and 100  C), these electrolytes have been able to show energy densities of 140 mAh g1 and an ability to work for 700 cycles without suffering any reduction in output  [240]. Citing some of the advantages of the glass based electrolytes, the authors pointed out that these batteries hold great promise for future useas solidstate batteries in several applications [240]. Work on similar lines was performed by Trevey and coworkers   [241], who used differing compositions of a Li2SeLi2OeP2S5  solid electrolyte in their research. The electrolytes were prepared by mechanical milling techniques using planetary ball mills for the purpose  [241]. The researchers prepared three different systems: the -20P2S5, -25P2S5, and -30P2S5 systems [241]. Results indicated that conductivity decreased steadily with an increase in the amount of P2S5 in the system, from 3  * 104 S cm1 to approximately 1.6 * 104 S cm1 [241]. Overall, the samples exhibited a general amorphous nature, even though sharp peaks were observed at 26.8 and 34 . Crystalline Li3PS4   and Li4P2S6 phases are seen from the Raman data  [241]. Electrical studies indicated that the best performance was exhibited by the electrolyte with the -20P2S5  composition, which had an electrical output of 90 mAh g1, whereas the -30P2S5  had the poorest performance [241]. The authors indicated that the performance dropped steeply after about 25 cycles and therefore these electrolytes do not yet present a viable commercial alternative [241]. Nagao, Hayashi, and Tatsumisago studied lithium and sul�de based solid electrolytes for solid state battery applications  [242]. The main active component of the electrolyte was the Li 2SeP2S5 part, and the composition that was chosen primarily was the 80Li2Se20P2S5. The solid state electrolyte was fabricated properly using planetary ball mills, along with a small quantity of lithium titanate  [242]. Electrical results have shown the charge and discharge voltages to be around 2 V and 1.1 V, respectively, and a discharge capacity of 120 mAh g1 [242]. A sudden change in the potential toward the end of the charging process was attributed to internal short-circuiting in the lithium batteries [242]. A reversible capacity of 1350 mAh g1 was observed

for the cells, at a current density of 0.013 mA cm 2. Results showed that, even though the output capacity was high, further studies are still required before they can put the �ndings to practical use [242]. Work on Li2SeP2S5   systems has also been performed by other groups, who have tried adding different compounds to stabilize the electrolyte system, but capacity fading of these sul �de based electrolyte systems has remained the major issue [243e246]. Similarly, considerable research has been done in the study of glassy phosphorus based lithium ion solid state batteries  [39,247e249]. Integrated solid state batteries were studied by Goncalves et al.  [250]. The study was based on lithium phosphorus oxy-nitride (LiPON) �lm electrolytes. Annealing of the cathode material LiCoO 2 �lms exhibited a polycrystalline structure with orientation in the (104) planes and fewer at (003) and (101) planes [250]. The LiPON 1  m m �lms were deposited on top of Al disc plates, with the surface already covered with 100 nm of Pt to avoid an unwanted reaction between the LiPON and the Al plate. The LiPON �lm was covered with a 100 nm layer of Pt and 500 nm of Al coatings for protection [250]. Results indicated that, when LiPON was coupled with LiCoO 2 as the cathode material, there was a net improvement in the charging and discharging ability of the cell, and the electrochemical ef �ciency simply increases   [250]. These factors further serve to establish the superiority of LiPON  � lms in solid state battery utilization.   Table 2   lists the solid state electrolytes commonly used commercially and their room temperature conductivities, as well as their crystalline nature. 4.2. Ceramic based electrolyte systems

Ohta et al. coated the surface of the LiCoO 2   with Li4Ti5O12by spray-coating  [251]. They used seven different samples with varying thicknesses. Characterization was done by electrochemical impedance spectroscopy (EIS), which showed the reduction of the interfacial resistance by the interposition of Li 4Ti5O12   [251]. The spectrum consisted of high as well as low frequency regions and looked similar to Warburg impedance. High frequency limits were almost constant, which the authors attributed to a change in the resistance of the electrolytic layers. Different thicknesses gave different interfacial resistances [251]. Glass ceramic based electrolytes tended to improve the ef �ciency of the cell, thanks to better charge distribution characteristics  [251]. Kotobuki and Kanamura worked on lithium, lanthanum, and tantalum based ceramic electrolytes for lithium based solid state batteries [247]. The electrolyte was prepared by the sintering method and the resultant crystal structure of the electrolyte was like that of a garnet. The anode consisted of Li metal, whereas the cathode was LiCoO 2, both of  which were in contact with the Li5La3Ta2O12   (LLTa) electrolyte [247]. Very clear redox couple of the LiCoO 2   was obtained, even after prolonged storage of the battery for about a year. XRD results proved its garnet like structure, whereas cyclic voltammetry and other electrochemical testing showed that this solid state battery had oxidation peaks at 3.75 V and 3.95 V, which corresponded to the oxidation peaks of LiCoO 2  [247]. Discharge capacity obtained was about 83 mAh g 1, approximately half the theoretical capacity [247]. The authors have suggested that a three-dimensional design of the battery will be able to enhance its functional capability [247].  Jak et al. developed an all-solid state lithium battery to operate at room temperature up to 150  C [252]. The ceramic electrolyte is made from Li-doped BPO 4,  a crystalline material with a primary particle size of about 50 nm. At room temperature, the SE conductivity was 2  106 S/cm. Conductivity increased to 2 * 104 S cm1 for explosively compacted samples. An alternative electrolyte is Li 1.3Al0.3Ti1.7(PO4)3 (LATP), with a bulk conductivity of  2 * 103 S cm1 at room temperature [252]. Thin foils are used to reduce internal resistance and increase ionic conductivity. The

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

311

 Table 2 Li based solid state electrolytes, their stoichiometries, crystal structure, and repeat units; and optimum conductivity values.

Electrolyte

Crystal nature

Stoichiometry

Ionic conductivity at room temperature (S cm1)

Ref.

Lithium lanthanum titanate (LLTO) Lithium based superionic conductor (LISICON) Sulfur doped lithium based superionic conductor (thioLISICON) Lithium lanthanum barium tantalum oxide Li ion conducting mesoporous oxide Sul�de glass LiPON

Crystalline Crystalline Crystalline

Li 3xLa2/3    x TiO3 Li14ZnGe4O16 Li3.4Si0.4P0.6S0.4

103 106 6.4  *  10 4

[1,14,339,340] [14,96,318,339,341] [14,260,339,341]

Crystalline Composite Amorphous Amorphous

Li 6La2BaTa2O12 LiIeAl2O3 Li2S  þ  LiI  þ  GeS 2  þ  Ga 2S3 Li2.88PO3.73N0.14

4  *  10 5 2.6  *  10 4 103 3.3  *  10 6

[14,165,339,342] [14,200,339,343] [14,278,339] [14,160,210,339]

cathode material is LiMn2O4   (LMO) and the anode material is Li4Ti5O12  (LTO). In a liquid electrolyte, the liquid penetrates the electrodes, giving rise to high ionic conductivity. In an all-solid state battery, this interaction is absent. To increase the ionic conductivity of the SE, an amount of electrolyte material is mixed with electrode material and carbon black [252]. A second method to increase ionic conductivity in the SE is to use a Li-ion conducting polymer in place of the polymeric binder. Softening the electrode-electrolyte interface by adding an inert liquid, ethylene carbonate (EC), has been shown to increase ionic conductivity. Magnetic pulse compaction was used, after all components were assembled, to improve densities and interfacial contacts. A capacity of 160 mAh g 1 was reported [252]. Birke et al. developed a monolithic, fully inorganic solid state lithium battery   [253]. The goals were to use a ceramic lithium electrolyte with high ionic conductivity and a large stability window, and to use a second ion conductor for a sintering additive to prevent high preparation temperatures. The sintering agent used was 0.44LiBO2$0.56LiF with an ionic conductivity of 10 6 S cm1 [253]. The negative electrode was Li 4Ti5O12, the solid electrolyte was Li1..3Al0.3Ti1.7(PO4)3, and the positive electrode was LiMn 2O4. The solid electrolyte had a lithium rich stability limit of 2.4 V vs. lithium and presented a total conductivity of 3  * 104 S cm1 [253]. Kobayashi et al. developed a lithium all-solid state battery using spark-plasma sintering (SPS) [254]. The electrolyte at the positive electrode was made from a lithium lanthanum titanate (Li,La)TiO 3 [Li0.41La0.47TiO2.91] with an ionic conductivity of 10 3 S cm1 at 22  C and an activation energy of 30.1 kJ mol 1 [254]. Using electrostatic spray deposition (ESD), LiMn 2O4, a positive electrode, was deposited on the electrolyte pellets. The electrolyte on the negative side was a solid polymer electrolyte (SPE), ethylene oxide co-2-(2methoxyethoxy) ethyl ether, and the negative electrode was Li metal. The resulting total conductivity was 2    105 S cm 1 [254]. Meikhail et al. studied the usage of lithium borosilicate glass electrolyte in lithium based solid state batteries. An important advantage of using glass is that glass is isotropic and has no grain boundaries, which can enhance conductivity  [248]. The lithium based borosilicate was prepared by melt quenching in a mold at a temperature of about 1273e1450 K from H3BO3, Li2Co3, and LiCl precursors  [248]. The researchers showed that the activation energy of lithium borate based glass is linearly dependent, while that of corresponding glasses that have halides in them follow a polynomial pattern. The authors attributed this to the presence of  halide ions, which cause this type of variation  [248]. Also, lithium borosilicate has shown better conductivity than lithium chloride because of the presence of lithium oxide instead of lithium chloride [248]. Kumar et al. studied the working of a solid state rechargeable lithium ion battery having an oxygen/air component [249]. Structurally, their cell was made up of a lithium anode, lithium ion conducting glass ceramic (GC) as the cathode, and a high surface area air cathode made up of carbon and conducting to lithium ions. They were able to demonstrate very good thermal stability and

performance of the cell in the temperature range of 30e105  C at a current density of 0.05e0.25 mAcm2 [249]. The OCV obtained was in the range of 2.6e3.6 V, depending on the temperature and other operating conditions. Charging and discharging capacities of 4.87 and 5 mAh, respectively, were obtained [249]. An energy density of  about 750 Wh kg1 was reported for this cell. The results, at these particular temperature conditions, indicate that the cell can be used for different purposes that require high energyand power densities [249]. Jin and McGinn's research focused on an Al doped lithium based solid state battery. The electrolyte composition used in their work was Li7La3Zr2O12 [255]. The anode and cathode were Li metal and Cu0.1V 2O5 based slurry painted on the electrolyte, respectively. Anoutputof 53mAhg 1 was obtained at a constant current density of 5  m A cm2 [255]. The best performance of the battery was obtained at a temperature of around 50   C, at which the output was 176 mAh g1 when a current density of 10 mA cm2 was applied [255]. Impedance spectroscopy has shown that the interfacial resistance of both the anode and the cathode drops down steeply at about 50  C; hence the better performance of the cell around this temperature  [255]. Emery, Brousse, and coworkers tested a lithium-based solid state electrolyte system by the principle of sintering   [256]. The precursor materials were Li 3N and elemental Mn, and they were subject to thorough mechanical milling followed by sintering under a nitrogen stream at 750   C for 7 h. XRD patterns have shown that two phases, Li7Mn4  and Li6.2Mn4, tend to coexist  [256]. Having an empirical formula of Li7   x Mn4, itis seenthat there isa gradual loss in crystallinity with increasing x  [256].

4.3. Amorphous electrolyte based systems

Amorphous electrolytes have been studied by various groups for use in solid state batteries   [86,257]. Sakurai et al. worked on amorphous lithium based �lms [86]. Liquid electrolytes carry the inherent dangers of leakage and �ammability, to counter which the researchers proposed a solid amorphous electrolyte  [86]. Komiya and colleagues used amorphous lithium based electrolytes in their research [257]. The composition for their electrolyte was (100    x) (0.6Li2S. 0.4SiS2). xLi4SiO4. LiCoO2  was used as the cathode and indium as the anode. XRD was employed to study the crystallinity of the electrolyte prepared. It was observed that the XRDresponded to various crystalline peaks in the sample, namely those corresponding to Li2S, SiS2, and Li4SiO4   [257]. Charge and discharge characteristics showed that the cells with x  ¼  5 and 20 gave output voltages of approximately 3.1 V and 2.7 V, respectively  [257]. The small drop in this discharge voltage is because of the smaller contact area between the LiCoO2   and the electrolyte surface. The electrical capacity of the cells is high, but they go on decreasing with the number of cycles they operate, starting from 90 mAh g 1 in the  � rst cycle to about 70 mAh g 1 after ten cycles. The output becomes a constant at approximately 60 mAh g1 [257]. The coulombic ef �ciencies approach 100% in the initial cycles but

312

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

 Table 3 The various lithium based solid state battery systems (anode, cathode, and electrolyte systems), along with their fabrication techniques and optimum conductivity values under working conditions.

Anode

Electrolyte

Cathode

Electrolyte fabrication technique

Conductivity

Cobalt oxide doped polyethylene oxide (PEO) electrolyte Graphite based Polystyrene doped with varying quantities of Al2O3

LiMn2O4/electrolyte

N/A

Potassium

Polyethylene oxide (PEO)/KBrO3

Carbon paper doped with iodine

Metathesis polymerization reaction Solegel reaction of polystyrene, n-butyl acetate, LiCF 3SO3, and Al2O3  precursors Hot pressing of PEO and KBrO 3 powders.

Li foil

Li doped polyacrylonitrile (PAN) electrolyte

Carbon composite electrode

LiMn2O4

Li foil

Mg

Li

Au

Mg ribbons

Graphite

LiCoO2

Prepared from heat treated and solutions of PAN, EC, PC, and LiPF6  precursors Polyurethane acrylate (PUA) based Li0.33MnO2/PEG/LiTFSI/carbon Solvent free casting with electrolyte with SiO2  content black 2-hydroxyethyl acrylate, Li0.33MnO2/PEG/LiTFSI/carbon black urethane acrylate, and isophorone di-isocyanate precursors Mg(CH3COO)2  doped PVA-PEG (I2  þ (I2  þ  C  þ electrolyte) Solution casting from 50: C  þ  electrolyte) 50 wt% of PEA and PVG followed by addition of  Mg(CH3COO)2  with constant stirring Tri-ethyl sulfonium bisimide, lithium Li Formation of an ionic liquid TFSI, polyethylene oxide and (IL) with S2TFSI, TFSI. and PEO tetrahydrofuran and solution casting with tetrahydrofuran Poly(epichlorohydrin-co ethylene Au Prepared by solegel technique oxide-co allyl glycidyl ether) [p(EEO-AGE)] from the coupled with lithium perchlorate and P(EEO-AGE) precursor lithium bis(tri�uoromethanesulfonyl)imide PVA(1-x)  (MgBr2)x/2(PWA)x/2 70:30 TiO2:graphite Prepared by solution casting of the x  ¼  0.0, 0.1, 0.2, 0.3, 0.4  &  0.5 gm. precursors, coupled with continuous stirring xLi2S.(100    x)P 2S5 LiCoO2 Prepared by hot and cold pressing of  where x  ¼  50e80 Li2SeP2S5  precursors

Ref.  

[309]

5.83  *  10 5 S cm 1 for 10 wt% Al2O3

[316]

7.74  *  10 8 S cm 1 at 303 K a composition of  70PEO: 30KBrO3 N/A  

[317]

[318]

5.5*104 S cm  1

[319]

�ltered

decrease continuously thereafter [257]. The authors concluded that the cell shows good electrical stability and characteristics for practical use, although its output decreases; further work is necessary in that direction [257]. Table 3 presents a summary of the types of anode, cathode, and electrolytes used in solid state batteries and their respective open circuit voltage and electrical outputs. 5. Non-lithium based solid state batteries

Non-lithium based solid state electrolytes have been used extensively to study the functioning and performance of solid state batteries. Ceramics are some of the most common non-Li based

3.23*105 S cm1 for 30 [320] wt% Mg(CH3COO)  2  at 100   C

10 mS cm1 at 45   C

 

[321]

about 4.2  *  10 5 S cm1 at 55   C

[322]

105 S cm  1 at 20   C for 10 wt% SN

[326]

3.1  *  104 S cm 1 for the 75Li2S:25P2S5 composition

[328]

solid state electrolytes   [258]. These solid electrolytes usually conduct ions by the translation of point defects, and energy is required in the process. This makes them good conductors at higher temperatures, as the high temperature provides the necessary energy for the process  [99,258]. Phosphates such as NASICON are considered to be good sodium ion conductors. Similarly, sul �de based solid state electrolytes are used in many solid state applications [99]. Non-Li based solid state electrolytes are described in this section in greater detail.   Table 4  presents, in greater detail, the different non-Li based solid state battery systems and their fabrication methods.

 Table 4 The various non Li based solid state electrolyte systems used, their fabrication techniques and their optimum conductivity values under working conditions.

Anode

Electrolyte

Cathode

Electrolyte fabrication technique

Conductivity

Glass and silver powder in weight ratio 67: 33

AgI doped borate glasses having the composition xAgI-(1 e x)[0.67Ag2O e 0.33B2O3]

Carbon and Iodine in weight ratio 30: 70

6.73  *  10 4

Silver

Ag4P2O7  doped with varying concentrations of SbI 3. (Ag2OeP2O5eLiCl) containing silver components

Iodine coupled with carbon  � ber Graphite coupled with iodine

Solid state reaction involving silver iodide, silver oxide and boron oxide as the precursors. Melt quenching

Dy2O3eWO3eBi2O3

Gold paste mixed with isopropanol

Silver

Gold paste mixed with isopropanol

Melt quenching of  LiCl, NH4H2PO4, AgNO3 and alumina powders. Mechanical milling of  Bi2O3, Dy 2O3  and WO3 followed by calcination at 800   C for 16 h

Ref. 1

U

cm1

[259]

4.1  *  10 4 S cm 1 at 60 mol% SbI3

 

[264]

2.88  *  10 2 S cm 1 for 25 wt% Al2O3   [266]

1.4  *  10 4 S cm 1 at 200   C

 

[280]

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

5.1. Silver based systems

Masoud et al. studied the potential applicability of borate glasses having silver constituents in solid state battery systems [259]. Borate glasses, especially AgI doped borate glasses, are presumed to have fast ionic conductivity. The borates used for experiment had the formula xAgI-(1  x)[0.67Ag2O  0.33B2O3], where x was chosen to be 0.4, 0.5, 0.6, 0.7, and 0.8  [259]. The cathode consisted of carbon and iodine in the weight ratio 30: 70, whereas the anode was composed of glass and silver powder in the weight ratio of 67: 33 [259]. The chargeedischarge characteristics indicated the highest performance when x ¼  0.6 in the formula of the silver based borate glass. It presented a DC conductivity of 6.73 * 104 U1 cm1 at ambient temperature conditions  [259]. The authors found that silver doped borate glasses, especially with x ¼ 0.6 in the formula composition, can be a suitable candidate for solid state batteries [259]. The presence of AgI improved conductivity, as it supplied mobile Ag  þ  ions and produced a weak ionic bond with easy availability of Agþ ions [259]. Silver based glassy electrolytes were also studied by Agrawal et al. The stoichiometry of the solid state electrolyte in this case was x[0.75AgI: 0.25AgCl]: (1  x) [Ag2O: WO3] [260]. Silver metal was used as the anode and Carbon along with iodine (C þ I2) as the cathode. The electrolyte was prepared by the technique of melt quenching from its salt precursors. Room temperature conductivity of 4 *  103 S cm1 was obtained for the stoichiometry 0.7[0.75AgI: 0.25AgCl]: 0.3[Ag 2O: WO3]. Characterization results have proven the existence of the glassy electrolyte, which the authors claim produces satisfactory results under working conditions  [260]. In addition, there was a steady increase in the room temperature conductivity of the silveretungsten glass system with increasing x up to x ¼  0.7, after which there was a noticeable decrease  [260]. The authors attributed the increase to be the optimum concentration and to have the largest number of mobile ions. This is shown graphically in  Fig. 6. Das et al. studied the working of solid state batteries using different zinc cadmium halide doped silver phosphate glass electrolytes [261]. The halide dopedglass electrolytes were preparedby the melt quenching technique of zinc or cadmium doped silver nitrate and ammonium dihydrogen phosphate in equimolar ratio in a platinum dish. Along with this electrolyte, silver metal was used

Fig. 6.  A graphical representation of dependence of conductivity at room temperature versus different stoichiometries of the silveretungsten glass system [260].

313

as the anode and graphite combined with iodine was the cathode. Characterization showed that the glasses so prepared were usually amorphous and crystallinity was almost absent   [261]. Electrochemical characterization results indicated a high ionic transfer number in the range of 98.5% for the cadmium chloride doped glasses, with a conductivity of approximately 4.64 *  105 S cm1 [261]. Solid state electrolyte based on silver thiocyanate complexed with polyethylene oxide (PEO-AgSCN) was studied by Sekhon et al. [262]. The electrolyte was prepared by solidi�cation from the precursor solutions and thorough drying. The results show that complexing between the polymer and the silver salt takes place in the amorphous stage, because no new peaks are noticed in the XRD data, and this is con�rmed by the FTIR spectra. A maximum conductivity of 1.3 *  106 S cm1 is obtained for electrolyte samples that have an O: Ag ratio of 8.8: 1  [262]. Progressive increase in Ag ion concentration tends to lower conductivity. Conductivity is also seen to increase at the melting point of the polymer, owing to a phase change from the quasi-crystalline phase to an amorphous phase [262]. It has also been observed that, on the exposure of  these electrolytes to light, white spots are formed on their surface; this has been attributed to the photolysis of the silver salt. The authors suggested that solid state electrolytes of this type are mainly anionic conductors and can be used for solid state battery applications [262]. AgI doped silver borate based superionic glasses as solid electrolytes were also studied by Bhattacharya and Ghosh [263]. AgI doped borate based glasses of the composition xAgI  (1  x) (Ag2Oe2B2O3) were prepared by the melt quenching technique, the anode being silver metal foil and graphite along with iodine being the cathode. The amount of AgI can be manipulated fromx ¼ 0 t o x ¼ 0.5. Asymmetric BeO bonds in the borateion were observed from the strong bands at 1350 cm1 and 1000 cm1 in the FTIR spectra [263]. Results showed that, with the increase of AgI up to x  ¼  0.5 in the formula composition, there was a change in the borate structure as it expanded to accommodate more Ag ions. The authors explained that, with the increased doping of the borate glass with AgI, more Ag ions took part in the ion transport, which greatly increased the ionic conductivity of the electrolyte system [263]. Suthanthiraraj and Sarumathi studied silver phosphate (Ag4P2O7) based solid electrolytes having varying concentrations of  antimony iodide (SbI3) [264]. The electrolyte was prepared by the melt quenching technique. XRD was used to con�rm the formation of the silver pyrophosphate phase. The anode and cathode were composed respectively of silver and iodine coupled with carbon �ber and phenothiazine. Impedance spectroscopic analysis was done in the range of 20 Hz to 1 MHz and a maximum ionic conductivity of 4.1 * 104 S cm1 was obtained at a concentration of  60 mol% SbI3 [264]. Addition of SbI 3 of more than 60 mol % resulted in a decrease in conductivity, owing to the formation of cross links in the pyrophosphate network and hindrance to the movement of  silver ions [264]. Also, the authors noticed nano-sized AgI particles in the electrolyte from FESEM analysis, formed from an endothermic reaction at 420 K; this improved ionic conductivity. They reported the unique nature of these multiphase superionic solids, which assist ionic conductivity   [264]. Fast ion conducting silver based solid state electrolytes having a tellurite component were studied by Lefterova et al. [265]. The base composition of the solid state electrolyte was AgIeAg2SO4eTeO2, and the electrolyte was prepared by melt quenching from its precursors. Observations of  the performance of the electrolyte showed that the best ionic conductivity of 2.3 *  102 S cm1 at room temperature was obtained for the electrolyte having composition 60 AgI  e  24Ag2SO4 e 16 TeO2 [265]. However, the authors noted that AgI had to be added to obtain this conductivity [265]. They concluded that the addition of tellurites and halide systems to a glassy silver based electrolyte greatly improved its ionic conductivity  [265].

314

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

Phosphate glasses and phosphate based solid state electrolytes (Ag2OeP2O5eLiCl) containing silver components were tested for applications in batteries by Das and coworkers  [266]. The silver phosphate glass samples were prepared by forming a mortar and then melting and quenching [266,267]. The anode was silver metal and the cathode consisted of graphite combined with iodine. The maximum conductivity reported was approximately 8.91 * 105 S cm1 under ambient conditions and 4.16 *  103 S cm1 at 200  C for 15 wt% LiCl. With the addition of varying quantities of  Al2O3, a maximum conductivity of 3.32 * 104 Scm1 under ambient conditions and 2.88 * 102 Scm1 was obtained for 25 wt% Al2O3. Conductivity increased with an increase in temperature, and similarly with Al2O3   concentration. Analysis of the chargeedischarge characteristics showed that undoped Ag2OeP2O5 glasses (OCV of 0.298 V) do not fare as well as those with LiCl doping (OCV of 0.668 V) [266]. Further, the electrolyte when doped with Al2O3  (up to 25%) was found to perform better than an electrolyte with no Al 2O3  doping. and therefore has potential use in solid state batteries   [266]. Venkateswarlu and Satyanarayana developed and tested a silver based all-solid state primary battery [268]. The electrolyte was a glassy compound with the composition AgIeAg2O-(SeO2 þ V 2O5); silver selenovanadate (SSV). The solid electrolyte had a conductivity of 2.63    102 S cm1 at 30  C and activation energy of 0.22 eV   [268]. Performance was tested with several cathode materials. Silver based solid electrolytes have a high ionic conductivity and stability under temperature variations. The anode was Ag/SE and the cathode was [(I  þ  C)  þ  SE]  þ  T AAI, where Ag 3 was silver powder, SE the SSV solid electrolyte, I iodide, C graphite, and T AAI tetra-alkyl ammonium iodide ( A ¼   methyl, ethyl, and butyl). SE was added to the anode to improve interfacial contacts. The cathode was varied by changing the weight ratios of  the individual components. The anode mixture was pressed at 5000 kg cm2, producing a 10 mm diameter and 1.5 e2.0 mm thick pellet [268]. The cathode pellet was 1 mm thick. The outer battery  jacket was made of an ebonite material and sealed with epoxy resin. The OCV reached thermodynamically calculated values of  687 mV. An energy density of 3.21 Wh kg1 was reported [268]. Delaizir et al. developed an all-solid state silver battery assembled in one step using spark plasma sintering (SPS)   [269]. At room temperature, silver and copper ion conductors tended to exhibit high ionic conductivity; 0.05 S cm1 for glassy Ag6I4WO4 and 0.34 S cm1 for crystalline Rb4Cu16I7Cu14. They used composite electrodes for their battery system, obtained by mechanical milling of Ag0.7V 2O5   and Ag6I4WO4   (1:1 weight ratio) for approximately 2 h. The ionic conductivity of the glass-ceramic Li2SeP2S5 was 104 S cm1 [269]. The cell studied by Delaizir et al. was a symmetric silver cell: Ag0.7V 2O5//Ag6I4WO4//Ag0.7V 2O5. It is a glassceramic and stable in open air  [269]. 5.2. Perovskite based systems

ABO3   type crystals that exhibit the perovskite structure are commonly used as ionic conductors  [270]. Lanthanum gallate has very good ionic conductivity and is used in solid state batteries and electrochemical cells in the medium temperature range of  approximately 700e1000 K [271]. Lanthanum gallate based solid electrolytes exhibit comparably low thermal expansions, like zirconia based solid state electrolytes. They can also be made to conduct oxygen and oxide ions by doping them with alkaline earth metals. However, these electrolytes also have certain disadvantages, which include the volatility of gallium oxide, the cost of  procuring gallium, and high reactivity of the perovskites under oxidizing and reducing conditions  [272]. Ceria based solid state electrolytes have also been considered by a number of researchers. These include electrolytes of the family of Ce 1    x MxO2    d  (M  ¼  Gd

or Sm and x ¼   0.1 or 0.2)   [273,274]. Ceria based electrolytes demonstrate better ionic conductivities than comparable ZrO2 electrolytes, especially at low temperatures. However, ceria based electrolytes suffer from the drawback that Ce 4þ ions are partially reduced to Ce3þ ions, which hinders oxygen ion transport [275,276]. Bi2O3 have been considered for solid state electrolytes by several researchers. A comprehensive study of the FCC Bi 2O3  ( d-Bi2O3) was undertaken by Takahashi et al. [277]. They reported that  d -Bi2O3  has the highest oxide ion conductivity when doped with aliovalent metal ions. It has been shown that d-Bi2O3 exhibits oxide ion conductivity approximately 1e2 orders of magnitude greater than zirconia for the same temperatures [277]. The best conductivities are shown by the  d  phase with a  � uorite defect; however, this is a high temperature phase and exists at 1000e1100 K. Bismuth based mixed conductors and bismuth molybdates such as Bi2MoO6 and Bi2MoO9 have also been studied [278,279]. Jiang et al. conducted a detailed study of a bismuth based solid electrolyte for application in solid state battery systems [280]. They based their choice of electrolyte material on the superior ionic conductivity demonstrated by the  d Bi2O3  system. The electrolyte was made by the solid state reaction mechanism. The authors mechanically milled the stock powder precursors (Bi2O3, Dy2O3, and WO3) and then calcined at high temperatures (800  C for 16 h), followed by sintering  [280]. Electrical conductivity values were found to have the best values for dopant compositions of about 11 e13 %, and the highest conductivity value obtained was 1.4  * 104 S cm 1 at 200   C [280]. This is considerably higher than that of erbia, which is known to have the highest conductivity (2.8  * 105 S cm1 at 200  C) among materials exhibiting a similar structure. However, higher dopant concentrations greater than 17% were found to reduce conductivity  [280]. This was because the high amount of doping disturbs the stability of the structure, which is best maintained at 11 e13 % doping. The authors showed the working of an electrolyte with better erbia based electrolytes, which is suitable for solid state applications [280]. Barium cerate based perovskites were studied by Iwahara and colleagues [277]. Barium doped ceria usually exhibitsvery good oxide ion conductivity, as well as n-type semiconductor properties. These perovskites normally show a cubic or octahedral structure [281]. Zuo et al. worked on developing a new solid electrolyte that is barium based  [282], fabricating a new composition, namely Ba(Zr0.1Ce0.7Y 0.2)O3   d or BZCY7 [282]. The sample was prepared by solid state reaction techniques, and Ba(Ce 0.8Y 0.2)O3  or BCY20 was used as a control to compare the properties  [282]. A comparative study of the ionic conductivities showed that BZCY7 had higher ionic conductivities than the usual solid electrolytes such as YSZ and LSGM. The solid state system used Ni-BZCY7 as the anode and Ba(Ce0.4Pr0.4Y 0.2)O3  (BCPY4) as the cathode. The BZCY7 electrolyte demonstrated an ionic conductivity of 9 *  103 S cm1 at 500  C, whereas those for YSZand LSGMat the same temperatures were 5 * 104 S cm1 and 5 * 103 S cm1, respectively [282]. Open circuit voltages obtained at 500  C and 600  C were 1.05 V and 1.01 V, respectively. A maximum power density of 270 mW cm 2 was obtained at 700  C. Moreover, polarization studies indicated that at 500  C the polarization resistance of the electrodes and the electrolytes was 3.2  U cm2 and 2  U cm2, respectively [282]. The authors succeeded in demonstrating that a barium based electrolyte has better electrical properties than standard solid state electrolytes such as YSZ or LSGM. However, the major challenges are the development of suitable cathodes for this electrolyte and obtaining good conductivity values above 550   C [282]. 5.3. Sodium based systems

Sodium and sulfur based solid state electrolytic systems were

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

studied by Hayashi et al. [283], who fabricated a sul �de based glass ceramic electrolyte obtained from the high temperature Na3PS4 cubic phase. Na eSn alloy was used as the counter and reference electrode, and TiS 2  was the working electrode [283]. During electrochemical measurements, room temperature conductivity of  approximately 104 S cm1 was obtained, along with an output voltage of 1.6 V [283]. The authors pointed out that thiswas the �rst time that high temperature stabilized cubic Na 3PS4  had been used as a solid state electrolyte system  [283]. In addition, a variety of  other sodium based solid state electrolyte and battery systems have been tested. The most common are the sodiumion conducting solid state electrolyte system having the general formula Na 1þxZr2SixP3xO12, commonly referred to as NASICON based systems [284]. These electrolytes usually have a three-dimensional structure and a network of channels and pores through which the Na  þ  ions move. The most common NASICON structures usually contain two SiO 4 tetrahedra for every PO4   tetrahedra present   [285e290]. Further improvements to NASICON based systems have been achieved by the replacement of zirconium with titanium   [291].   Similarly, replacement of zirconium with hafnium has also yielded improved conductivity   [292]. However, other elements such as indium, scandium, germanium, magnesium, gallium, and �uorine have also been tested doped with the NASICON electrolyte, producing lower conductivities  [286,287,293e299]. A sintering process has been used by Fuentes et al. to prepare NASICON-based solid state electrolytes from the precursor materials SiO2, Na3PO4.12H2O, and ZrO2   [300]. They were thoroughly pulverized in a mechanical mill with intermittent heating. Sintering was done at a temperature of 1200  C for 10 h exposed to the atmosphere. Further  � ring of the pellets took place in vacuum before a �nal sintering in air at approximately 1300  C for 16 h   [300]. Subsequently, different NASICON samples were prepared at increasing or decreasing sintering temperatures. XRD characterization data have shown that the NASICON is formed only if calcination has taken place at 1100  C or higher and the precursors remain uncombined at lower temperatures [300]. Electrical studies have shown that the activation energy of the NASICON pellets ranges from 0.36 eV to 0.39 eV for those sintered at high temperatures and 0.31 eV e0.32 eV for those sintered at higher temperatures [300]. Another sodium based electrolyte that has been the subject of  research is beta alumina having the formula Na2O.Al2O3. It basically has a layered structure that consists of very dense areas of  aluminum oxide interspersed with less dense layers of sodium oxide [301]. This results in a structure that has a densely packed arrangement of oxygen atoms with some aluminum atoms at the octahedral and tetrahedral sites [302]. The conduction mechanism has been attributed to the movement of the sodium ions among the lattice and the interstitial sites in the crystal structure  [303,304]. The potential drawbacks of the Li based solid state systems have prompted research into sodium based solid state electrolyte systems. Sodium based systems exhibit good conductivity and can act as replacements for Li based systems, especially when incorporated with polymer gel electrolytes [305]. Therefore, non-Li based solid state batteries have been found to possess good ionic conductivity, resistance to leakage, and absence of side reactions. Extensive research has been done in this �eld and many newer solid state electrolytes are still being fabricated. However, some challenges remain, such as the cost of the raw materials (especially transition metals), miniaturization, scalability, and ef �cient performance at lower temperatures. A detailed  � owchart of the general process of the steps involved in the fabrication of a solid state battery (anode/electrolyte/cathode) system is shown in  Fig. 7,   with the help of images to further elucidate the steps involved in the process.

315

6. Polymer electrolyte based solid state batteries

Several researchers have conducted studies of polymer electrolyte based solid state batteries. The history of polymer based electrolytes starts with the reporting of conductivities of polyethylene oxide (PEO) alkali metal salt complexes by Wright et al. [306]. PEO based polymers have been widely used since then as solid state electrolytes [307]. The most commonly used solid state electrolyte system is PEO and polypropylene oxide complexed with Li salts. The prevalence of PEO based systems has been attributed to the ability of the ethylene group to form complexes easily with lithium based salts and also the ease with which the polymer side chains dissolve Li in them   [307,308]. Polymer based electrolytes possess several advantages, such as enhanced safety and stability against easy discharge [309e311]. A general schematic of ion conduction through polymer electrolytes is shown in   Fig. 8. A comprehensive list of different polymers, their repeating units, and the temperatures at which they give optimum conductivity is presented in Table 5. 6.1. Inorganic polymer composite based electrolyte systems

a) PEO based inorganic electrolyte systems: Appetecchi et al. studied the performance of hot-pressed nanocomposites containing Li based solid state electrolytes. Polyethyleneoxide (PEO) based lithium electrolytes have the advantage of good ionic conductivity, which can be utilized in solid state batteries [312]. The electrolytes were prepared by ball milling followed by hot pressing at a temperature of 80 e100  C. Characterization of the electrolyte was done by differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). The thermal data revealed a sharp peak at 64.8  C and 66  C, which indicates the melting of PEO. Electrical measurements were carried out within a decreasing temperature range of 100   Ce76  C [312]. Open circuit voltages of around 3.45 V were obtained. The most striking result shown by the cell was that it exhibited almost no change in voltage during the whole discharge process. An increase in the ohmic losses of the cell was observed when the transition was made from charging to discharging. A maximum electrical capacity of about 140 mAh g1 was obtained, corresponding to a temperature of 100  C, and the effective output dropped to 120 mAh g1 at 83  C [312]. The results indicate that these electrolytes can be used in Li/LiFePO 4 based solid state batteries. Some amount of reduction in capacity is noticed, owing to a reduction in the conductivity of the electrolyte  [312]. Bullock and Ko �nas synthesized a cobalt oxide based polymer system, by a metathesis polymerization reaction mechanism, to act as the electrolyte in an all solid state battery [309]. The solid state battery had a LiMn 2O4 anode and cathode, the electrolyte being unsaturated polyethylene oxide (PEO) containing cobalt oxide nanoparticles. These nanoparticles can be used to store Li ions during the electrochemical reaction, and the unsaturated PEO improves ionic conductivity [309]. NMR spectra and gel permeation chromatography (GPC) have conclusively shown the formation of the polymer. The researchers theorized that this novel type of polymeric electrolyte system will prevent leakages and help in the development of solid state batteries with enhanced ionic conductivity and electrical output [309]. Similar research based on the PEO based electrolyte was conducted by Hassoun et al. [313], who prepared the electrolyte by the hot-pressing technique; energy density of 2250 Wh g 1 was obtained for an output voltage of 2.5 V   [313].   The output voltage remained fairly stable; this therefore represents an achievable solid state battery that can be used for future applications   [313]. Shin et al. studied the application of a PEO based

316

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

Fig. 7. A schematic representation of the fabrication process of a solid state battery system. (a) The fabrication of the cathode is done by mixing it with binder (PTFE  &  C). (b) Fabrication of the cathode on the stainless steel substrate which acts as support. (c) Thoroughly purging the cathode as well as the electrolyte in vacuum at elevated temperature to remove any traces of water vapor or impurities. (d) Keeping it in a furnace to obtain remove any trace of alcohol. (e) Pasting the cathode with the electrolyte. (e) Testing of the SSB.

producing a favorable outcome [314]. Jeon et al. similarly studied a PEO based solid state electrolyte matrix interspersed with LiClO 4 dissolved in it  [315]. However, in this case the working capacity of  the cell decreased with the number of cycles because of nonuniform sulfur positioning along the electrolyte  [315]. Lim et al. developed an Al 2O3  based solid polymer based electrolyte by the solegel reaction mechanism for a lithium based solid state battery [316]. The precursors for the polymer were Al2O3, LiCF3SO3, n-butyl acetate, and polystyrene. Characterization of the electrolyte after preparation was done by an impedance analyzer and SEM   [316]. The cathode consisted of LiCoO2 and the anode was graphite based. Electrochemical results showed that ionic conductivity tended to increase with Al2O3   concentration in the electrolyte and then decrease. The highest ionic conductivity obtained was 5.83 * 105 S cm1 for 10 wt% Al2O3   [316]. SEM images showed well dispersed Al 2O3 �llers and no clumping formation up to 10 wt% Al2O3  concentration. The authors attributed this to greater concentrations of Al2O3  forming larger aggregates and thereby hindering the ion transfer process. Solid inorganic polymer electrolytes were studied by Chandra et al., who fabricated a polyethylene oxide (PEO)/potassium bromate (KBrO3) based electrolyte by the hot press technique [317]. The anode consisted of potassium, whereas the cathode had a layer of carbon  � lm doped with iodine. An increase in ionic conductivity of about 100 times was noticed with the progressive addition of the

Fig. 8.  Ion conduction through a polymer electrolyte. The reactions that take place at the interface are clearly shown. Lithium ions move into and out of the electrodes via the intercalation process, and the electrolyte allows the  � ow of the ions.

LiCF3SO3  solid state electrolyte doped with titania  [314], and were able to show that the addition of titania resulted in a decrease of  overall crystallinity of the electrolyte and also reduced the contact resistance between the electrode and electrolyte, thereby

 Table 5 Li based polymer solid state electrolytes used in various solid state systems and their repeating units, along with the output room temperature conductivities obtained.

Polymer system Polyvinyl alcohol [PVAc] Polymethyl-methacrylate [PMMA] Polyethylene oxide [PEO] Polyethylene oxide [PEO] Polyoxymethylene [POM] Polypropylene oxide [PPO] Polysiloxane/poly-dimethyl-siloxane [DMS] P EO linked poly siloxane [PGP S] Polypropylene oxide [PPO]

Repeating unit (C2H4O)n e[CH2C(eCH3) (eCOOCH3)]ne e[CH2CH2O]ne e[CH2CH2O]ne   e[CH2O]ne e[CH3CH2CH2O]ne   e[(CH3)2SiO]ne PE O-H(OSiH2)nOH   e[CH3CH2CH2O]ne

Electrolyte used  

(PVAc)-LiCF3SO3 PMMA-LiClO 4 PEO-LiBF4 PEO8-LiClO4 POM-LiClO 4 PPO8eLiClO4 DMS-LiClO 4 PGPS-LiClO4 PPO-LiCl4-LiBr-AlCl3

Room temp. conductivity (S cm 1) 9

10 4  *  10 3 106 108 108 108 105 104 2*102

Ref. [96,344] [96,211,345,346] [96] [307,347] [114,307,348,349] [307] [307,350] [307,351] [307,352e354]

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

KBrO3  up to 50 wt% salt concentration, and the optimum conductivity shown by the electrolyte was 7.74 *  108 S cm1 at room temperatures at a composition of 70PEO: 30KBrO3 [317]. Highersalt concentration introduced brittleness into the electrolyte �lm, which hindered its further performance. SEM studies too showed a gradual reduction in crystallinity with the addition of further potassium salt. The OCV obtained was 2.58 V  [317]. This new electrolyte has performed satisfactorily under low current conditions and relatively large potassium salt conditions (~50% salt concentration) [317]. The graphical representation of the increase in ionic conductivity with increasing temperature is shown in Fig. 9. Here it may also be noticed that there is a major jump in ionic conductivity values at 65  C for all compositions, which the authors attributed to a semicrystalline phase transition process taking place in the polymeric PEO system. b) Non PEO based inorganic electrolyte systems: A Li/polyacrylonitrile (PAN) based solid electrolyte was developed by Abraham and Jiang for lithium based solid state batteries, by the solegel technique [318]. The anode consisted of sol egel thin Li metal foil (50  m m thickness), and the cathode was a thin layer of  carbon composite  � ber. The entire setup was kept in a heat sealed, closed plastic bag, in which pores were made to allow oxygen to reach the cathode. An OCV of 3.05 V was obtained for atmospheric air and 2.85 V for pure dry oxygen  [318]. The OCV was found to remain almost constant during the discharging process from 2.85 V  to 2.75 V   [318]. A strong absorption peak at 795 cm1 and electrochemical calculation showed that the main product formed during the discharging process was Li 2O2. The cell output obtained was 800 mAh g1 and 600 mAh g1 at 1 mA cm 2 and 2 mA cm2, respectively  [318]. It has been suggested that having the cell in the form of strips or stacks improves battery output  [318]. Jiang et al. developed a new solid polyurethane acrylate (PUA) based ceramic polymer for applications in lithium based solid state batteries [319]. The fabrication of this electrolytewas done by a solvent free casting method, and the cathode and anode materialswere Li0.33MnO2/PEG (polyethylene glycol) and lithium foil, respectively [319]. Duringthe charging/discharging cycles, the cell demonstrated an initial discharge capacity of 180 mAh g1 and a reversible capacity of  140 mAh g1 after 100 cycles of running  [319]. There was a small

Fig. 9.  A graphical representation of the relationship between conductivity and temperature of the polymer electrolyte �lms having the stoichiometry (1  x)PEO:xKBrO3. A linear increase in conductivity was observed up to 65 C, at which there is a major spike in conductivity for the different compositions  [317].  



317

decrease in the capacitance with an increasing number of cycles, attributed to increased interfacial resistance between the electrolyte and cathode with the number of cycles   [319]. The cell demonstrated stable voltages of 3.5 V at the start, before dropping to 3.25 V after approximately �ve days. A self-discharge rate of  0.05% per day at 60   C was noted, which is better than comparable liquid electrolyte Li-ion batteries [319]. Polu et al. have fabricated a polyvinyl alcohol (PVA)/polytethylene glycol (PEG) coupled with magnesium acetate solid electrolyte by a solution cast technique for use in solid state batteries [320]. The solid state battery consisted of  a magnesium based anode and a cathode having carbon �bers doped with iodine. They found maximum conductivity to be 3.23  * 105 S cm1 at 30 wt% of Mg(CH3COO)2  at 100  C operating temperature  [320]. The authors observed that conductivity increased with increasing salt concentration and then decreased [320]. This was attributed to an increase in the number of active charge carriers with increasing salt concentration up to the maximum and then formation of ion pairs and ion triplets, which hinder further conductivity  [320]. Fisher et al. studied sulfur based hybrid solid polymer electrolytes [321]. The hybrid electrolyte consisted of triethyl sulfonium bis(tri �uorosulfonyl)imide (S 2TFSI), lithium TFSI, and polyethylene oxide (PEO), and was fabricated by the solution casting method. Lithium based electrodes were used to complete the circuit   [321]. The experiment was performed by applying a current density of 0.1 mA cm2 and reversing the current direction hourly. Ionic conductivities of 0.117 mS cm 1 and 1.20 mS cm1 were observed at 0  C and 25  C, respectively   [321]. When the temperature was increased to 45  C, the authors noticed an even greater increase in the conductivity value, to 10 mS cm 1. The authors succeeded in demonstrating the high stability of this electrolyte, even at greater than 4.5 V at the Li/Li þ cathodic interface [321]. Rodrigues et al. studied the properties of polymer based solid state electrolytes [322]. The basic monomer of the polymer electrolyte was the terpolymer: poly(epichlorohydrin-co ethylene oxide-co allyl glycidyl ether) [322]. The �nal polymer was obtained by adding other precursors such as lithium perchlorate and lithium bis(tri�uoromethanesulfonyl)imide by the solegel technique. Ion blocking gold electrodes of 10 mm diameter were used as both anode and cathode. The experiments were performed in the temperature regime of 25  Ce100  C, and readings were taken at 7degree intervals  [322]. A maximum of 4.2 *  105 Scm1 was obtained at a temperature of 55   C for this electrolyte [322]. The glass transition temperature Tg of the electrolyte was obtained at  48   C [322]. The authors claimed that the presence of oxygen ions in the terpolymer was bene�cial for ion transport, and they exhibited good thermal stability at high temperatures [322]. Guobao and coworkers investigated polymer based secondary lithium batteries [323]. The X-ray diffraction peaks corresponding to PTFE became weak when the carbon electrode accommodated more lithium. The peak disappeared when the lithium consumed exceededa criticalvalue of about 1200 mAhper gram of PTFE under these experimental conditions   [323]. The small discrepancy between the experimental value (1200 mAh g1) and the theoretical value (1070 mAh g1) indicated that the decomposition of the electrolyte on the surface of the carbon electrode also contributed, to a small extent, to capacity loss  [323]. Hence, the PTFE indeed reacted with electrochemically formedlithium on the surface of the carbon electrode during the discharge process of the Li/C cell [323]. Electronically conducting polymers such as phosphor-olivines were studied by Chung and coworkers   [324]. Electrical conductivity measurements were made on samples sectioned from pellets densi�ed at 700e850  C. All doped compositions Li 1  xM xFePO4 (M ¼ Mg, Al, Ti, Nb or W) showed room-temperature conductivities in excess of 10 3 S cm 1. The temperature dependence of conductivity in the doped samples varied somewhat with �ring

318

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

temperature. Samples �red at 700   C were not dense enough for us to rule out a signi �cant contribution to the DC resistivity from the grain boundaries, and samples  � red at 850   C contained a signi �cant fraction of Fe2P phase  [324]. It has been assumed that the conductivity is predominantly electronic at these high values. Samples containing 1 atom% each of Mg 2þ, Zr4þ, and Nb5þ have conductivities within a factor of three of one another, and show similar activation energies in the range 60e80 meV. The identically processed undoped sample has a much higher activation energy, close to 500 meV  [324]. The activation energies may include terms for both defect formation and migration (for example for a polaronic mechanism) [324]. 6.2. Polymer based thin �lm electrolyte systems

a) PEO based thin  � lm electrolyte systems A study on polyethylene oxide (PEO) thin  � lm solid electrolytes was reported by Ibrahim and Johan  [325]. The study involved varying the composition of the electrolyte with compounds such as lithium hexa�uoride (LiF 6), ethylene carbonate (EC), and amorphous carbon nanotubes (aCNT) [325]. The electrolyte was fabricated by the solution casting method, and thermal characterizations were done by differential scanning calorimetry. The maximum electrical conductivity obtained was 10 3 S cm 1 at 5 wt % aCNT [325]. The authors showed that the electrical conductivity improved with the addition of  aCNTs, as the nanotubes prevented the recrystallization of the PEO based electrolyte and established routes for the easier transport of Li þ ions  [325]. The work of Abdel-Samiea et al. concentrated on developing and studying solid polymer electrolytes interspersed with small crystals to be used for magnesium based rechargeable cells   [326]. The electrolyte was fabricated from polyvinyl alcohol (PVA), hydroquinone, magnesium bromide, phosphotungstic acid (PWA), and varying amounts of succinonitrile (SN) precursors, by the solution cast technique [326]. The cathode and anode were made of 70:30 TiO2:graphite and Mg pellets, respectively. The initial conductivity values were low, in the region of about 107 S cm1, but then increased to about 106 S cm1 when SN was slowly added. The highest conductivity of the electrolyte obtained was about 105 S cm 1 at 20   C for 10 wt% SN [326]. The authors' explanation for this increase was that SN reduces the crystallinity of the electrolyte and the increased polarity of SN itself  [326]. A maximum capacity of 17.5 mAh g1 has been reported for this cell  [326]. b) Non PEO based thin  � lm systems: Rhodes and coworkers made a detailed study of solid polymer electrolytes of nanoscale dimensions  [327]. The electrolytes were obtained by electro-deposition of phenol on acetonitrile and using indiumetin oxide substrates [327]. AFM was used to characterize the thickness of the �lm, which was found to be approximately 50nm [327]. The optimum conductivity obtained was about 7 ± 4 * 1010 S cm 1. This provides a route to the application of nanoscale thin �lms for miniaturized solid state devices  [327]. Sakuda et al. developed a sul�de based solid electrolyte for applications in Li based solid state batteries  [328]. They used glassy sul �des such as Li2SeP2S5 as the main constituents in their solid state sul �de based electrolytes   [328], which were prepared by both hot and cold pressing from their initial powder stage. Ionic conductivity was found to have greatly increased when the amount of molding pressure was increased. A maximum ionic conductivity of  104 S cm 1 was obtained for a molding pressure of 70 MPa, and a further increase of the pressure resulted in decreasing conductivity [328]. When used in a battery with a LiCoO2  cathode and graphite

anode, it gave a discharge capacity of 133 mAh g 1 and reasonable stability under a number of cycles  [328]. The authors believed that these glassy solid state electrolytes would be useful, as they have considerable microstructural free volume, which enables easy ion diffusion [328]. Polysiloxane polymer based solid state electrolytes were developed and studied by Hooper et al. [329]. The electrolyte they used for this purpose was poly (bis[2-(2-methoxy-ethoxy) ethoxy]propylsiloxane. These solid state polymer electrolytes were produced by hydrosilation reaction from their respective precursor allyl ethers, followed by hydrogenation with hexane  [329]. The researchers also doped these polymers with varying quantities of Li to study changes in electrical and conducting properties [329]. They reported the best conductivity values as 3.9  *  104 S cm 1 at room temperature for oxygen to Li ratios approximately 40:1  [329]. Li doping has the advantage of improving room temperature conductivity values and providing more stability   [329]. Many other groups have conducted research on solid polymer electrolytes for use in solid state batteries. Bannister et al. obtained optimum ionic conductivities of about 105 S cm1 from an acrylate based polymer electrolyte   [330]. The conductivity measurement was done at 100  C [330]. Sanchez et al. studied the properties of polyelectrolyte lithium salt having a per �uorosulfonated side chain  [331]. However, they obtained low conductivities of about 105 S cm1 at 30   C, which they attributed to the lithium ion's strong interaction with the sulfonyl group  [331]. Ito and colleagues studied lithium salt based polyelectrolytes having PEO oligomers and sulfonated side chains [332]. They obtained room temperature conductivity of  4.45 *  106 S cm1 and a Liþ transference number of 0.75 [332]. Angel et al.studied a novel polymer in salt based system, in which the electrolyte was a AlCl 3 e   LiBr e   LiClO4 e   PPO based system [333], and were able to obtain room temperature conductivities as high as 0.02 S cm 1 [333]. Polymers obtained from organic matter have been used to prepare solid state batteries too. The advantage of using organic and natural polymers is their abundance, affordability, and biodegradability  [334]. Yulianti and coworkers used chitosan (a naturally occurring polymer) with different doping agents such as Cu and Ag to develop a polymer based solid state battery system [334]. A schematic for the fabrication process and testing of a similar polymer based system is presented in  Fig. 10. The dopings were done by the ion implantation method. The authors concluded that this could provide a new way of developing more environmentally friendly solid state batteries [334]. Nystrom et al. developed an ultrafast all-polymer based solid state battery [335]. The electrolyte was fabricated from cellulosic �bers obtained from algae, and a coating of polypyrrole was provided, approximately 50 nm thick [335]. The battery exhibited only 6% loss in capacity over about 100 cycles of operation and showed good promise for commercial applications   [335].   Recently, increasing ionic conductivity, two approaches have been investigated, (i) suppressing crystal generation in the polymer by adding  � ller and (ii) creating a comb-shaped polymer framework   [336,337]. The ionic conduction in poly (ethylene carbonate) (PEC) based rubbery electrolyte including lithium salts was reported by Tominaga et al. [338]. The Liþ transfer number of PEC e LiBF4 (44.4 wt%) electrolyte was measured and the value was estimated to be ~0.5 at 100  C which [338]. Okumura and Nishimura reported that the lithium ion transfer number of a PEC based polymer electrolyte is 0.4 and the highest ionic conductivity is 0.47 mS cm1 at20  C [336]. PEC based polymer electrolytes have a moderate conductivity with a high lithium ion transfer number. “

”

7. Conclusion

There has been a tremendous recent increase in the amount of  research and understanding of solid state batteries, especially in the

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

319

thinner cells. Electrolytes of about 10  m m thickness have recently been developed, which have been able to produce power densities comparable with those of existing electrolytes. An increase in the application of polymer based solid state electrolytes has been very bene�cial, greatly improving the �exibility of the electrolyte systems and at the same time giving good ionic conductivity values. Considerable work needs to be done in this area before solid state batteries, especially lithium based systems, become ubiquitous. Elucidating the lithiation/delithiation mechanism, developing a sound model, acquiring a thorough understanding of the chemistries and reactions going on at the electrode-electrolyte interface, and preventing capacity from fading by engineering the electrode microstructure are among the important aspects to which attention should be paid. Solid state batteries have shown very good promise and potential during the time they have been under research, and the current focus should be on slowly bringing them to a state where they can be deployed in medium- and then large-scale operations.  Acknowledgment

This work was supported by the DGIST R &D Program of the Ministry of Education, Science, and Technology of Korea (15-BD01). References [1] M.T. review, in: Technology Review, 2011, pp. 48e50. [2]   S.B. Peterson, J. Apt, J.F. Whitacre, J. Power Sources 195 (2010) 2385 e2392. [3]   J. Davison, J. Boyce, Chemical Engineering, Worcester Polytechnic Institute, Worcester, 2012, p. 126. [4]   P.V. Braun, J. Cho, J.H. Pikul, W.P. King, H. Zhang, Curr. Opin. Solid State Mater. Sci. 16 (2012) 186e198. [5]   J. Chen, Materials 6 (2013) 156 e183. [6]   M.S. Whittingham, Chem. Rev. 104 (2004) 4271e4301. [7] C. Corporation, in, Cymbet Corporation, 2012, pp. 1e6. [8] P. Wray, Am. Ceram. Soc. Bull. 91 (2012) 1 e58. [9]  C. Delmas, M. M enetrier, L. Croguennec, S. Levasseur, J.P. P eres, C. Pouillerie, G. Prado, L. Fourn es, F. Weill, Int. J. Inorg. Mater. 1 (1999) 11 e19. [10]   R. Jasinski, J. Electroanal. Chem. 26 (1970) 189 e194. [11]  R. Jasinski, J. Electroanal. Chem. Interfacial Electrochem. 15 (1967) 89 e91. [12]   J.B. Goodenough, Solid State Ionics 94 (1997) 17 e25. [13]   V.V. Kharton, F.M.B. Marques, A. Atkinson, Solid State Ionics 174 (2004) 135e149. [14] C.C. Ho, in: Department of Physics and Materials Science, City University of  Hong Kong, Hong Kong, 2010, pp. 1 e42. [15]   C.A. Vincent, Solid State Ionics 134 (2000) 159 e167. [16]  N. Terada, T. Yanagi, S. Arai, M. Yoshikawa, K. Ohta, N. Nakajima, A. Yanai, N. Arai, J. Power Sources 100 (2001) 80 e92. [17] J. Liu, M.N. Banis, X. Li, A. Lushington, M. Cai, R. Li, T.-K. Sham, X. Sun, J. Phys. Chem. C 117 (2013) 20260e20267. [18]   S.M. George, Chem. Rev. 110 (2010) 111e131. [19]   R.L. Puurunen, J. Appl. Phys. 97 (2005) . [20] J. Hamalainen, J. Holopainen, F. Munnik, T. Hatanp aa, M. Heikkila, M. Ritala, M. Leskela, J. Electrochem. Soc. 159 (2012) A259eA263. [21]   M. Yano, A. Tomita, M. Sano, T. Hibino, Solid State Ionics 177 (2007) 3351e3359. [22]   P.G. Bruce, Chem. Commun. 19 (1997) 1817e1824. [23] Y. Yoon, C. Park, J. Kim, D. Shin, J. Power Sources 226 (2013) 186e190. [24] J. Fergus, J. Power Sources 195 (2010) 939e954. [25]   A.R. Armstrong, C. Lyness, P.M. Panchmatia, M.S. Islam, P.G. Bruce, Nat. Mater. 10 (2011) 223e229. [26]  W. Weppner, Secondary batteries  e lithium rechargeable systems  j  All-solid state battery, in: J. Garche (Ed.), Encyclopedia of Electrochemical Power Sources, Elsevier, Amsterdam, 2009, pp. 162 e168. [27]   B.B. Owens, B. Scrosati, P. Reale, Primary batteries  e  nonaqueous systems  j Solid-state: silvereiodine, in: J. Garche (Ed.), Encyclopedia of Electrochemical Power Sources, Elsevier, Amsterdam, 2009, pp. 120 e128. [28]   M. Kotobuki, H. Munakata, K. Kanamura, Chapter 4.2-All-solid-state Li Battery for future energy technology, in: S. Somiya (Ed.), Handbook of Advanced Ceramics, second ed., Academic Press, Oxford, 2013, pp. 343 e351. [29]   F.M. Gray, M.J. Smith, Secondary batteries  e  lithium rechargeable systems  j Lithium polymer batteries, in: J. Garche (Ed.), Encyclopedia of Electrochemical Power Sources, Elsevier, Amsterdam, 2009, pp. 169 e176. [30] K. Takada, T. Inada, A. Kajiyama, M. Kouguchi, H. Sasaki, S. Kondo, Y. Michiue, S. Nakano, M. Tabuchi, M. Watanabe, Solid State Ionics 172 (2004) 25 e30. 









Fig. 10. A �owchart showing the fabrication and development technique of a Li polymer based solid state electrolyte and graphite anode. LIPON is known to remain stable throughout long cycles of performance [334].

last two decades. Most of the work has focused on developing solid state battery systems with enhanced energy and power densities. Cathodes and anodes with good microstructural properties are being developed to provide good electrochemical and ionic conduction pathways. However, some fundamental challenges remain. A primary challenge is the high cost of production, particularly of  the lithium which makes it dif �cult to use in large-scale applications. Currently, most of the lithium based solid state batteries have been used for mobile and small-scale applications. Secondly, the output power densities and longeterm cycle life are still not satisfactory. One of the major reasons for this is the fast degradation of the lithium based cathodes owing to microstructural change or dendrite formation. Proper understanding of the electrodeelectrolyte interface and tailoring the microstructure of the cathode accordingly can help mitigate this problem. In-situ testing probes can help us understand some of the reaction mechanisms going on at the interface. At present also, the weight of the active material is about 60% of the total battery, which is also a hindrance to optimum performance, and research is going on to develop

€

€

€

€

€€

€

320

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

[31]  S. Boulineau, J.-M. Tarascon, J.-B. Leriche, V. Viallet, Solid State Ionics 242 (2013) 45e48. [32]  A. Ueda, M. Nagao, A. Inoue, A. Hayashi, Y. Seino, T. Ota, M. Tatsumisago,  J. Power Sources 244 (2013) 597e600. [33] Z. Liu, Y. Tang, X. Lü, G. Ren, F. Huang, Ceram. Int. 40 (2014) 15497e15501. [34]   K. Takada, Acta Mater. 61 (2013) 759 e770. [35] Y. Nishio, H. Kitaura, A. Hayashi, M. Tatsumisago, J. Power Sources 189 (2009) 629e632. [36]   N. Ohta, K. Takada, I. Sakaguchi, L. Zhang, R. Ma, K. Fukuda, M. Osada, T. Sasaki, Electrochem. Commun. 9 (2007) 1486 e1490. [37]   J. Hassoun, M. Pfanzelt, P. Kubiak, M. Wohlfahrt-Mehrens, B. Scrosati,  J. Power Sources 217 (2012) 459e463. [38]   K. Yamamoto, Y. Iriyama, T. Asaka, T. Hirayama, H. Fujita, K. Nonaka, K. Miyahara, Y. Sugita, Z. Ogumi, Electrochem. Commun. 20 (2012) 113 e116. [39]   F. Mizuno, A. Hayashi, K. Tadanaga, T. Minami, M. Tatsumisago, Solid State Ionics 175 (2004) 699e702. [40]   F. Rosciano, P.P. Pescarmona, K. Houthoofd, A. Persoons, P. Bottked, M. Wilkening, Phys. Chem. Chem. Phys. 15 (2013) 6107 e6112. [41] M.V. Tyufekchiev, S. Hur, in: Chemical Engineering, Worcester Polytechnic Institute, 2013. [42] F.R. Gamble, J.H. Osiecki, M. Cais, R. Pishardy, F.J. Disalvo, T.H. Geballe, Science 174 (1971) 493e497. [43] K. Mitzushima, P.C. Jones, P.J. Wiseman, J.B. Goodenough, Mater. Res. Bull. 15 (1980) 783e789. [44] G.G. Amatucci, J.M. Tarascon, L.C. Klein, Solid State Ionics 83 (1996) 167 e173. [45] E. Rossen, J.N. Reimers, J.R. Dahn, Solid State Ionics 62 (1993) 53 e60. [46] M. Yonemura, A. Yamada, H. Kobayashi, M. Tabuchi, T. Kamiyama, Y. Kawamoto, R.J. Kanno, J. Mater. Chem. 14 (2004) 1948 e1958. [47] J.M. Tarascon, F. Coowar, G. Amatuci, F.K. Shokoohi, D.G. Guyomard, J. Power Sources 54 (1991) 103e108. [48]  S. Yang, Y. Song, K. Ngala, P.Y. Zavalij, M.S. Whittingham, J. Power Sources 119 (2003) 239e246. [49] D.W. Murphy, P.A. Christian, F.J. Disalvo, J.N. Carides, J. Electrochem. Soc. 126 (1979) 497e499. [50]   R. Schollhorn, F. Klein-Reesink, R. Reimold, J. Chem. Soc. Chem. Comm. 9 (1979) 398e399. [51]  Y.V. Mikhaylik, J.R. Akridge, J. Electrochem. Soc. 150 (2003) A306 eA311. [52]  A. Unemoto, H. Ogawa, Y. Gambe, I. Honma, Electrochim. Acta 125 (2014) 386e394. [53] Y. Cai, Y. Huang, X. Wang, D. Jia, Xincun Tang, Ceram. Int. 4 (2014) 1 e5. [54]  N. UmiraTaib, N.H. Idris, J. Membr. Sci. 468 (2014) 149 e154. [55]   Z.-i. Takehara, J. Power Sources 68 (1997) 82 e86. [56]  U.v. Sacken, E. Nodwell, A. Sundher, J.R. Dahn, J. Power Sources 54 (1995) 240e245. [57]   D. Fauteux, R. Koksbang, J. Appl. Electrochem. 23 (1993) 1 e10. [58] Y. Wu, E. Rahm, R. Holze, J. Power Sources 114 (2003) 228e236. [59] H. Geng, A. Mei, Y. Lin, C. Nan, Mater. Sci. Eng. B 164 (2009) 91e95. [60]   J.-K. Ahn, S.-G. Yoon, Electrochem. Solid-State Lett. 8 (2005) A75 eA78. [61]   M. Kotobuki, Open Electrochem. J. 4 (2012) 28 e35. [62] U. Kasavajjula, C. Wang, A.J. Appleby, J. Power Sources 163 (2007) 1003e1039. [63] H. Azuma, H. Imoto, S. Yamada, K. Sekai, J. Power Sources 81 (1999) 1 e7. [64]   C.S. Wang, G.T. Wu, X.B. Zhang, Z.F. Qi, W.Z. Li, J. Electrochem. Soc. 145 (1998) 2751e2758. [65]  A. Mukhopadhyay, B.W. Sheldon, Prog. Mater. Sci. 63 (2014) 58 e116. [66]   J. Christensen, J. Newman, J. Electrochem. Soc. 153 (2006) A1019 eA1030. [67]  V.A. Sethuraman, V. Srinivasan, A.F. Bower, P.R. Guduru, J. Electrochem. Soc. 157 (2010) A1253eA1261. [68] S. Huang, F. Fan, J. Li, S. Zhang, T. Zhu, Acta Mater. 61 (2013) 4354 e4364. [69]   C.K. Chan, R. Ruffo, S.S. Hong, R.A. Huggins, Y. Cui, J. Power Sources 189 (2009) 34e39. [70]  J.M. Tarascon, M. Armand, Nature 414 (2001) 359 e367. [71]   B. Scrosati, J. Electrochem. Soc. 136 (1992) 2774 e2782. [72] Y. Nishi, J. Power Sources 100 (2001) 101e106. [73] J.Y. Song, Y.Y. Wang, C.C. Wan, J. Power Sources 77 (1999) 183 e197. [74] H. Kim, R. Auyeung, A. Pique, J. Power Sources 165 (2007) 413e419. [75]   M.D. Fleischauer, J. Li, M.J. Brett, J. Electrochem. Soc. 156 (2009) A33 eA36. [76]  G.T. Teixidor, R.B. Zaouk, B.Y. Park, M.J. Madou, J. Power Sources 183 (2008) 730e740. [77]  V. Zadin, D. Brandell, Electrochim. Acta 57 (2011) 237 e243. [78]  F. Mizuno, A. Hayashi, K. Tadanaga, M. Tatsumisago, J. Power Sources 146 (2005) 711e714. [79] T. Wei, Y. Gong, X. Zhao, K. Huang, Adv. Funct. Mater. 130 (2014) 1 e5. [80]   G. Delaizir, V. Viallet, A. Aboulaich, R. Bouchet, L. Tortet, V. Seznec, M. Morcrette, J.-M. Tarascon, P. Rozier, M. Dolle, Adv. Funct. Mater. 22 (2012) 1993e2000. [81]  Philip P. Soo, Biying Huang, Young Il Jang, Yet Ming Chiang, D.R. Sadoway, A.M. Mayes, J. Electrochem. Soc. 146 (1999) 32 e37. [82] Y. Zhang, Y. Liu, M. Liu, Chem. Mater. 18 (2006) 4643 e4646. [83]  Y. Li, B. Tan, Y. Wu, Nano Lett. 8 (2008) 265 e270. [84]  M. Tatsumisago, M. Nagao, A. Hayashi, J. Asian Ceram. Soc. 1 (2013) 17 e25. [85] A. Sakuda, A. Hayashi, T. Ohtomo, S. Hama, M. Tatsumisago, J. Power Sources 196 (2011) 6735e6741. [86]   Y. Sakurai, A. Sakuda, A. Hayashi, M. Tatsumisago, Solid State Ionics 182 (2011) 59e63.

[87]  A. Hayashi, K. Noi, A. Sakuda, M. Tatsumisago, Nat. Commun. 856 (2012) . [88]   A. Sakuda, A. Hayashi, M. Tatsumisago, Sci. Rep. 3 (2013) . [89] A. Hayashi, Y. Nishio, H. Kitaura, M. Tatsumisago, Electrochem. Commun. 10 (2008) 1860e1863. [90]  H. Kitaura, A. Hayashi, T. Ohtomo, S. Hama, M. Tatsumisago, J. Mater. Chem. 21 (2011) 118e124. [91]   M. Nagao, A. Hayashi, M. Tatsumisago, J. Mater. Chem. 22 (2012) 10015e10020. [92] N. Ohta, K. Takada, L. Zhang, R. Ma, M. Osada, T. Sasaki, Adv. Mater. 18 (2006) 2226e2229. [93]  V. Thangadurai, W. Weppner, Ionics 12 (2006) 81 e92. [94] J.B. Goodenough, in: L. Wang (Ed.), University of Texas at Austin, Austin, 2012, pp. 1e18. [95]   N.A. Anurova, V.A. Blatov, G.D. Ilyushin, O.A. Blatova, A.K. Ivanov-Schitz, L.N. Dem'yanets, Solid State Ionics 179 (2008) 2248 e2254. [96] R.C. Agrawal, G.P. Pandey, J. Phys. D: Appl. Phys. 41 (2008) 1 e18. [97]  P. Mustarelli, E. Quartarone, C. Tomasi, A. Magistris, Solid State Ionics 135 (2000) 1241e1247. [98]   M. Scagliotti, F. Parmigiani, G. Chiodelli, A. Magistris, G. Samoggia, G. Lanzi, Solid State Ionics 28e30 (1988) 1766e1769. [99] P. Knauth, Solid State Ionics 180 (2009) 911e916. [100] S. Teng, J. Tan, A. Tiwari, Curr. Opin. Solid State Mater. Sci. 18 (2014) 29 e38. [101] J.M. Tarascon, A.S. Gozdz, C. Schmutz, F. Shokoohi, P.C. Warren, Solid State Ionics 86e88 (1996). [102] W. Wieczorek, K. Such, J. Przyluski, Z. Florianczyk, Synth. Met. 45 (1991) 373 . [103]   J.E. Weston, B.C.H. Steele, Solid State Ionics 7 (1982) . [104]  M.Y. Saidi, J. Barker, H. Huang, L. Swoyer, G. Adamson, Electrochem. Solid State Lett. 5 (2002) A149eA151. [105] J. Gaubicher, G. Goward, C. Wurm, C. Masquelier, L.F. Nazar, Chem. Mater. 12 (2000) 3240e3242. [106]   S.C. Yin, H. Grondley, P. Strobel, L.F. Nazar, J. Am. Chem. Soc. 125 (2003) 326e327. [107]   J.B. Goodenough, Solid State Ionics 69 (1994) 184 e198. [108]   S. Bach, M. Henry, N. Baf �er, J.J. Livage, J. Solid State Chem. 88 (1990) 325e333. [109]  S. Bach, J.P. Pereira-Ramos, C. Cachet, M. Bode, L.T. Yu, Electrochim. Acta 40 (1995) 785e789. [110] D.-G. Kim, J. Shim, J.H. Lee, S.-J. Kwon, J.-H. Baik, J.-C. Lee, Polymer 54 (2013) 5812e5820. [111] L. Xuan, Y. An, W. Fang, L. Liao, Y. Ma, Z. Ren, G. Yin, Int. J. Electrochem. Sci. 6 (2011) 6590e6598. [112] N. Kamaya, K. Homma, Y. Yamakawa, M. Hirayama, R. Kanno, M. Yonemura, T. Kamiyama, Y. Kato, S. Hama, K. Kawamoto, A. Mitsui, Nat. Mater. 10 (2011) 682e686. [113] Y. Seino, T. Ota, K. Takada, A. Hayashi, M. Tatsumisago, Energy Environ. Sci. 7 (2014) 627e631. [114]   N.J. Dudney, Electrochem. Soc. Interface (2008) 44e48. [115] H. Nakahara, S.Y. Yoon, S. Nutt, J. Power Sources 158 (2006) 600e607. [116]   T. Kawamura, S. Okada, J.I. Yamaki, J. Power Sources 156 (2006) 547 e554. [117]  Z. Gadjourova, D. Martín y Marero, K.H. Andersen, Y.G. Andreev, P.G. Bruce, Chem. Mater. 13 (2001) 1282e1285. [118]  Z. Stoeva, I. Martin-Litas, E. Staunton, Y.G. Andreev, P.G. Bruce, J. Am. Chem. Soc. 125 (2003) 4619e4626. [119]  S.D. Jones, J.R. Akridge, F.K. Shokoohi, Solid State Ionics 69 (1994) 357 e368. [120]   W.C. West, J.F. Whitacre, J.R. Lim, J. Power Sources 126 (2004) 134 e138. [121]   M.S. Whittingham, Science 192 (1976) 1126e1127. [122] R. Yazami, P. Touzain, J. Power Sources 9 (1983) 365e371. [123] J.P. Peres, A. Demourgues, C. Delmas, Solid State Ionics 111 (1998) 135e144. [124]   P.G. Bruce, Solid State Electrochemistry, Cambridge University Press, St. Andrews, 1995. [125]   W. Li, J.N. Reimers, J.R. Dahn, Solid State Ionics 67 (1993) 123 e130. [126] J.P. Peres, F. Weill, C. Delnias, Solid State Ionics 116 (1999) 19 e27. [127]   M. Park, X. Zhang, M. Chung, G.B. Less, A.M. Sastry, J. Power Sources 195 (2010) 7904e7929. [128]   P.G. Bruce, Solid State Electrochemistry, in: P.G. Bruce (Ed.), Solid State Electrochemistry, Cambridge University Press, Cambridge, 1995. [129]  E. Gileadi, Electrode Kinetics for Chemists, Engineers and Materials Scientists, VCH Publishers, Inc., Tel Aviv, 1993 . [130]   A.M. Andersson, D.P. Abraham, R. Haasch, S. MacLaren, J. Liu, K. Amine,  J. Electrochem. Soc. 149 (2002) A1358eA1369. [131] J. Hu, H. Li, X. Huang, L. Chen, Solid State Ionics 177 (2006) 2791e2799. [132]   Z. Gadjourova, Y.G. Andreev, D.P. Tunstall, P.G. Bruce, Nature 412 (2001) 520e523. [133]   H. Shirakawa, E.J. Louis, A.G. MacDiarmid, C.K. Chiang, A.J. Heeger, J. Chem. Soc. Chem. Commun. 16 (1977) 578e580. [134]  H. Park, T. Yoon, J. Mun, J.H. Ryu, J.J. Kim, S.M. Oh, J. Electrochem. Soc. 160 (2013) A1539eA1543. [135]   J. Evans, C.A. Vincent, P.G. Bruce, Polymer 28 (1987) 2324 e2328. [136]   E. Quartarone, P. Mustarelli, Chem. Soc. Rev. 40 (2011) 2525 e2540. [137]   P.G. Bruce, M.T. Hardgrave, C.A. Vincent, Solid State Ionics 53e56 (1992) 1087e1094. [138]   P.G. Bruce, T.M. Hardgrave, C.A. Vincent, Solid State Ionics 53e56 (1992) 1087e1094. [139] Y. Kato, S. Yokoyama, T. Yabe, H. Ikuta, Y. Uchimoto, M. Wakihara, Electrochim. Acta 50 (2004) 281e284.

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

[140]  N.H. Zainol, S.M. Samin, L. Othman, K.B.M. Isa, W.G. Chong, Z. Osman, Int. J. Electrochem. Sci. 8 (2013) 3602e3614. [141]  A. Ghosh, C. Wang, P. Ko �nas, J. Electrochem. Soc. 157 (2010) A846 eA849. [142]   G.G. Tenchov, Anal. Lab. 3 (1994) 1 e9. [143] Z.Q. Wang, M.S. Wu, G. Liu, X.L. Lei, B. Xu, C.Y. Ouyang, Int. J. Electrochem. Sci. 9 (2014) 562e568. [144]  R. Murugan, V. Thangadurai, W. Weppner, Angew. Chem. Int. Ed. 46 (2007) 7776e7781. [145]  J. Narayan, J. Fletcher, C.W. White, W.H. Christie, J. Appl. Phys. 52 (1981) 7121e7128. [146]   J. Tan, A. Tiwari, Electrochem. Solid-State Lett. 15 (2012) A37 eA39. [147] C.-M. Chang, S.-H. Hong, H.-M. Park, Solid State Ionics 176 (2005) 2583e2587. [148]   C.A.C. Sequira, A. Hooper, Solid State Ionics 9 e10 (1983) 1131e1135. [149]  E.H. Cha, D.R. Macfarlane, M. Forsyth, C.W. Lee, Electrochim. Acta 50 (2004) 334e337. [150] J.K. Feng, L. Lu, M.O. Lai, J. Alloys Compd. 501 (2010) 255 e258. [151]   M.J.G. Jak, E.M. Kelder, J. Schoonman, N.M.V.D. Pers, A. Weisenburger,  J. Electroceram. 2 (1998) 127e134. [152]  M. Kotobuki, H. Munakata, K. Kanamura, Y. Sato, T. Yoshida, J. Electrochem. Soc. 157 (2010) A1076eA1079. [153]  M. Tatsumisago, R. Takano, K. Tadanaga, A. Hayashi, J. Power Sources 270 (2014) 603e607. [154] A. Hayashi, S. Hama, H. Morimoto, M. Tatsumisago, T. Minami, J. Am. Ceram. Soc. 84 (2001) 477e479. [155]   A. Hayashi, D. Furusawa, Y. Takahashi, K. Minami, M. Tatsumisago, Phys. Chem. Glasses e  Eur. J. Glass Sci. Technol. Part B  e Eur. J. Glass Sci. Technol. Part B 54 (2013) 109e114. [156]   A. Sakuda, A. Hayashi, M. Tatsumisago, Sci. Rep. 3 (2013) 2261 . [157] A. Nakagawa, N. Kuwata, Y. Matsuda, J. Kawamura, J. Phys. Soc. Jpn. 79 (2010) 98e101. [158] J.M. Kim, G.B. Park, K.C. Lee, H.Y. Park, S.C. Nam, S.W. Song, J. Power Sources 189 (2009) 211e216. [159] A. Patil, V. Patil, D.W. Shin, J.-W. Choi, D.-S. Paik, S.-J. Yoon, Mater. Res. Bull. 43 (2008) 1913e1942. [160] J.P. Carmo, R.P. Rocha, A.F. Silva, L.M. Goncalves, J.H. Correia, in: Proceedings of the Eurosensors XXIII Conference, vol. 1, 2009, pp. 453 e456. [161]  K. Kanehori, Y. Ito, F. Kirino, K. Miyauchi, T. Kudo, Solid State Ionics 18/19 (1986) 818e822. [162] H. Ohtsuka, J. Yamaki, Jpn. J. Appl. Phys. 28 (1989) 2264e2267. [163] H. Ohtsuka, S. Okada, J. Yamaki, Solid State Ionics 40 (1990) 964e966. [164]   J.R. Akridge, H. Vourlis, Solid State Ionics 28 e30 (1988) 841e846. [165]  M. Baba, N. Kumagai, H. Kobayashi, O. Nakano, K. Nishidate, Electrochem. Solid-State Lett. 2 (1999) 320e322. [166] M. Baba, N. Kumagai, H. Fujita, K. Ohta, K. Nishidate, S. Komaba, B. Kaplan, H. Groult, D. Devilliers, J. Power Sources 119 (2003) 914 e917. [167]  M. Winter, J.O. Besenhard, Electrochim. Acta 45 (1999) 31 e50. [168]   M.S. Park, S. Rajendran, Y.M. Kang, K.S. Han, Y.S. Han, J.Y. Lee, J. Power Sources 158 (2006) 650e653. [169]   R.A. Huggins, Handbook of Battery Materials, in: J.O. Besenhard (Ed.), Handbook of Battery Materials, Wiley-VCH, Weinheim, 1999. [170] M. Yoshio, S. Kugino, N. Dimov, J. Power Sources 153 (2006) 375e379. [171]  E.J. Plichta, W.K. Behl, J. Electrochem. Soc. 140 (1993) 46 e49. [172]  J.J. Auborn, Y.L. Barberio, J. Electrochem. Soc. 134 (1987) 638 e641. [173]   K. Sato, M. Noguchi, A. Demachi, N. Oki, M. Endo, Science 264 (1994) 556e558. [174] Y.P. Wu, C. Wan, C. Jiang, S.B. Fang, Introduction, Principles and Advances of  Lithium Secondary Batteries, Tsinghua University Press, Beijing, 2002 . [175] Y.P. Wu, W.G. Ju, S.B. Fang, Y.Y. Jiang, J. Power Sources 70 (1998) 114e117. [176]   Y.P. Wu, S.B. Fang, Y.Y. Jiang, J. Power Sources 75 (1998) 167 e170. [177]   Y. Wen, L. Zeng, Z. Tong, L. Nong, W. Wei, J. Alloys Compd. 416 (2006) 206e208. [178] C.H. Mi, X.G. Zhang, X.B. Zhao, H.L. Li, J. Alloys Compd. 424 (2006) 327 e333. [179]  K. Kobayashi, K. Kosuge, S. Kachi, Mater. Res. Bull. 4 (1969) 95 e106. [180] W. Rüdorff, G. Walter, H. Becker, Z. Anorg. Allg. Chem. 285 (1956) 287e296. [181]  C.L. Liao, M.T. Wu, J.H. Yen, I.C. Leu, K.Z. Fung, J. Alloys Compd. 414 (2006) 302e309. [182] S.C. Han, H.S. Kim, M.S. Song, P.S. Lee, J.Y. Lee, H.J. Ahn, J. Alloys Compd. 351 (2003) 273e278. [183] J. Yamaki, M. Makidera, T. Kawamura, M. Egashira, S. Okada, J. Power Sources 153 (2006) 245e250. [184]  P. Liu, S.H. Lee, Y. Yan, C.E. Tracy, J.A. Turner, J. Power Sources 158 (2006) 659e662. [185] S. Yonezawa, T. Okayama, H. Tsuda, M. Takashima, J. Fluor. Chem. 87 (1998) 141e143. [186] H. Huang, G.V. Rao, B. Chowdari, J. Power Sources 81e82 (1999) 690e695. [187]  A.K. Pahdi, K.S. Najundaswamy, J.B. Goodenough, J. Electrochem. Soc. 144 (1997) 1188e1194. [188]   H. Huang, S.C. Yin, L.F. Nazar, Electrochem. Solid State Lett. 4 (2001) A170eA172. [189] S. Atlung, K. West, J. Power Sources 26 (1989). [190] F. Munoz, A. Duran, L. Pascual, L. Montagne, B. Revel, A.U.M. Rodrigues, Solid State Ionics 179 (2008) 574e579. [191]   J.B. Bates, N.J. Dudney, G.R. Gruzalski, R.A. Zuhr, A. Choudhury, C.F. Luck,  J.D. Robertson, J. Power Sources 43 (1993) 103e110.

321

[192]   S.J. Lee, H.K. Baik, S.M. Lee, Electrochem. Commun. 5 (2003) 32 e35. [193]   F. Wu, Y. Liu, R. Chen, S. Chen, G. Wang, J. Power Sources 189 (2009) 467e470. [194]   M.N. Kozicki, M. Mitkova, J.P. Aberouette, Phys. E 19 (2003) 161 e166. [195] K.H. Ryu, S.M. Haile, Solid State Ionics 125 (1999) 355e367. [196]  S. Barison, M. Battagliarin, T. Cavallin, L. Doubova, M. Fabrizio, C. Mortal o, S. Boldrini, L. Malavasi, R. Gerbas, J. Mater. Chem. 18 (2008) 5120 e5128. [197] Z. Tao, L. Bi, Z. Zhu, W. Liu, J. Power Sources 194 (2009) 801e804. [198]   F. He, D. Song, R. Peng, G. Meng, S. Yang, J. Power Sources 195 (2010) 3359e3364. [199] K. Xie, R. Yan, X. Xu, X. Liu, G. Meng, Mater. Res. Bull. 44 (2009) 1474 e1480. [200] S. Barison, L. Doubova, A. Sanson, in: 9th European Solid Oxide fuel Cell Forum, Lucerna, 2010, pp. 5/146-145/151. [201]   Z. Yongning, F. Zhengwen, Prog. Chem. 23 (2011) 336 e348. [202]   K.F. Chiu, J. Electrochem. Soc. 151 (2004) A1865 eA1869. [203]   K.F. Chiu, F.C. Hsu, G.S. Chen, M.K. Wu, J. Electrochem. Soc. 150 (2003) A503eA507. [204] H. Xia, Y.S. Meng, M.O. Lai, L. Lu, J. Electrochem. Soc. 157 (2010) A348 eA354. [205]  K. Kushida, K. Kuriyama, T. Nozaki, Appl. Phys. Lett. 81 (2002) 5066 e5068. [206]   J. Schoonman, E.M. Kelder, J. Power Sources 68 (1997) 65 e68. [207]   N. Kuwata, J. Kawamura, K. Toribami, T. Hattori, N. Sata, Electrochem. Commun. 6 (2004) 417e421. [208]   F. Kirino, Y. Ito, K. Miyauchi, T. Kudo, Nippon. Kagakukaishi 3 (1986) 445e450. [209] L. Mai, B. Hu, W. Chen, Y. Qi, C. Lao, R. Yang, Ying Dai, Z.L. Wang, Adv. Mater. 19 (2007) 3712e3716. [210] D. Choi, D. Wang, I.-T. Bae, J. Xiao, Z. Nie, W. Wang, V.V. Viswanathan, Y.J. Lee,  J.-G. Zhang, G.L. Graff, Z. Yang, J. Liu, Nano Lett. 10 (2010) 2799e2805. [211]   R.H. Dahm, S. Hackwood, R.G. Linford, J.M. Pollock, Nature 272 (1978) 522e524. [212] Z. Zhang, C. Zhou, Y. Liu, J. Li, Y. Lai, M. Jia, Int. J. Electrochem. Sci. 8 (2013) 10059e10067. [213] K. West, B. Zachau-Christiansen, T. Jacobsen, S. Atlung, J. Electrochem. Soc. 30 (1985) 1205e1208. [214]   M.Z.A. Munshi, W.H. Smyrl, Solid State Ionics 45 (1991) 183 e189. [215]  T.R. Jow, L.W. Shacklette, M. Max �eld, D. Vernick, J. Electrochem. Soc. 134 (1987) 1730e1733. [216]   Y. Ma, M. Doyle, T.F. Fuller, M.M. Doeff, L.C.D. Jonghe, J. Newman,  J. Electrochem. Soc. 142 (6) (1995) 1859e1868. [217] M.M. Doeff, S.J. Visco, Y. Ma, M. Peng, L. Ding, L.C.D. Jonghe, Electrochim. Acta 40 (1995) 2205e2210. [218]   B. Wang, J.B. Bates, F.X. Hart, B.C. Sales, R.A. Zuhr, J.D. Robertson,  J. Electrochem. Soc. 143 (1996) 3203e3213. [219]   B.J. Neudecker, N.J. Dudney, J.B. Bates, J. Electrochem. Soc. 147 (2000) 517e523. [220]   Y.-N. Zhou, M.-Z. Xue, Z.-W. Fu, J. Power Sources 234 (2013) 310 e332. [221] V. Petrovsky, T. Suzuki, P. Jasinski, Tatiana Petrovsky, H.U. Anderson, Electrochem. Solid-State Lett. 7 (2004) A138eA139. [222]   S.B. Adler, J. Electrochem. Soc. 149 (2002) E166 eE172. [223]   H. Uchida, S. Arisaka, M. Watanabe, Electrochem. Solid-State Lett. 2 (1999) 428e430. [224] NanoMarkets, in, NanoMarkets, LC, Glen Allen, VA, 2009. [225]   M. Broussely, G. Archdale, J. Power Sources 136 (2004) 386 e394. [226] D. Kunze, in: W.V. Gool (Ed.), Fast Ion Transport, North Holland, Amsterdam, 1973, p. 495. [227]   J.-P. Malugani, B. Fahys, R. Mercier, G. Robert, J.P. Duchange, S. Baudry, M. Broussely, J.P. Gabano, Solid State Ionics 9 e10 (1983) 659e665. [228]   J. Akridge, H. Vourlis, Solid State Ionics 18 e19 (1986) 1082e1087. [229]   J. Akridge, H. Vourlis, Solid State Ionics 28 e30 (1988) 841e846. [230]   A. Levasseur, M. Menetrier, R. Dormoy, G. Meunier, Mater. Sci. Eng. B 3 (1989) 5e12. [231] J.B. Bates, N.J. Dudney, D.C. Lubben, G.R. Gruzalski, B.S. Kwak, X. Yu, R.A. Zuhr,  J. Power Sources 54 (1995) 58e62. [232] S.D. Jones, J.R. Akridge, J. Power Sources 43 e44 (1993) 505e513. [233] O.L. Anderson, D.A. Stuart, J. Am. Ceram. Soc. 37 (1954) 573e580. [234]  S.W. Martin, C.A. Angell, J. Non-Cryst. Solids 83 (1986) 185 e207. [235]   M.D. Ingram, A.H.J. Robertson, Solid State Ionics 94 (1997) 49 e54. [236]  A. Bunde, M.D. Ingram, P. Maass, K.L. Ngai, J. Non-Crystal. Solids 131 e133 (1991) 1109e1112. [237] K. Funke, Prog. Solid State Chem. 22 (1993) 111e195. [238]   K. Funke, J. Non-Crystal. Solids 172 e174 (1994) 1215e1221. [239]   A. Bunde, K. Funke, M.D. Ingram, Solid State Ionics 86e88 (1996) 1311e1317. [240]   M. Tatsumisago, A. Hayashi, Solid State Ionics 225 (2012) 342 e345. [241]   J.E. Trevey, J.R. Gilsdorf, S.W. Miller, S.-H. Lee, Solid State Ionics 214 (2012) 25e30. [242]   M. Nagao, A. Hayashi, M. Tatsumisago, Electrochem. Commun. 22 (2012) 177e180. [243]   A. Hayashi, R. Ohtsubo, T. Ohtomo, F. Mizuno, M. Tatsumisago, J. Power Sources 183 (2008) 422e426. [244]  T. Kobayashi, Y. Imade, D. Shishihara, K. Homma, M. Nagao, R. Watanabe, T. Yokoi, T. Yamada, R. Kanno, T. Tatsumi, J. Power Sources 182 (2008) 621e625. [245] Y. Kim, S. Martin, Solid State Ionics 177 (2006) 2881e2887. [246]   A. Atkinson, Solid State Ionics 28e30 (1988) 1377e1387. 

322

 J.G. Kim et al. / Journal of Power Sources 282 (2015) 299e 322

[247]  M. Kotobuki, K. Kanamura, Ceram. Int. 39 (2013) 6481 e6487. [248]   M.S. Meikhail, I.A. Gohar, A.A. Megahed, J. Appl. Phys. D 26 (1993) 1125e1129. [249]   B. Kumar, J. Kumar, R. Leese, J.P. Fellner, S.J. Rodrigues, K.M. Abrahamc,  J. Electrochem. Soc. 157 (2010) A50eA54. [250] L.M. Goncalves, J.F. Ribeiro, M.F. Silva, M.M. Silva, J.H. Correia, Procedia Eng. 5 (2010) 778e781. [251] N. Ohta, K. Takada, L. Zhang, R. Ma, M. Osada, T. Sasak, Adv. Mater. 18 (2006) 2226e2229. [252] M.J.G. Jak, E.M. Kelder, J. Schoonman, J. Solid State Chem. 142 (1999) 74 e79. [253]   P. Birke, F. Salam, S. D oring, W. Weppner, Solid State Ionics 118 (1999) 149e157. [254]   Y. Kobayashi, H. Miyashiro, T. Takeuchi, H. Shigemura, N. Balakrishnan, M. Tabuchi, H. Kageyama, T. Iwahori, Solid State Ionics 152e153 (2002) 137e142. [255]   Y. Jin, P.J. McGinn, Electrochim. Acta 89 (2013) 407 e412. [256]   N. Emery, E. Panabi ere, O. Crosnier, S. Bach, T. Brousse, P. Willmann, J.P. Pereira-Ramos, J. Power Sources 247 (2014) 402e405. [257]  R. Komiya, A. Hayashi, H. Morimoto, M. Tatsumisago, T. Minami, Solid State Ionics 140 (2001) 83e87. [258] J.W. Fergus, J. Power Sources 195 (2010) 4554e4569. [259] E.M. Masoud, M. Khairy, M.A. Mousa, J. Alloys Compd. 569 (2013) 150 e155. [260] R.C. Agrawal, M.L. Verma, R.K. Gupta, Solid State Ionics 171 (2004) 199e204. [261]   S.S. Das, B.P. Baranwal, C.P. Gupta, P. Singh, J. Power Sources 114 (2003) 346e351. [262] S.S. Sekhon, G. Singh, S.A. Agnihotry, S. Chandra, Solid State Ionics 80 (1995) 37e44. [263]   S. Bhattacharya, A. Ghosh, Chem. Phys. Lett. 424 (2006) 295 . [264]  S.A. Suthanthiraraj, R. Sarumathi, Appl. Nanosci. 3 (2012) 501 e508. [265] E. Lefterova, S. Bliznakov, P. Angelov, S. Vassilev, Y. Dimitriev, in: International Workshop on Portable and Emergency Energy Sources  e  From materials to Systems , 2005, pp. 197e199. [266]   S.S. Das, P.K. Srivastava, N.B. Singh, J. Non-Crystal. Solids 358 (2012) 2841e2846. [267] V. Licina, A. Mogus-Milankovic, S.T. Reis, D.E. Day, J. Non-Crystal. Solids 353 (2007) 4395e4399. [268]  M. Venkateswarlu, N. Satyanarayana, Mater. Sci. Eng. B 54 (1998) 189 e195. [269]  G. Delaizir, N. Mana �, G. Jouan, P. Rozier, M. Dolle, Solid State Ionics 207 (2012) 57e63. [270]  T. Ishihara, H. Matsuda, Y. Takita, J. Am. Chem. Soc. 116 (1994) 3801 e3803. [271] M. Feng, J.B. Goodenough, Eur. J. Solid State Inorg. Chem. 31 (1994) 663e672. [272]  K. Yamaji, T. Horita, M. Ishikawa, N. Sakai, H. Yokokawa, Solid State Ionics 121 (1999) 217e224. [273]   B.C.H. Steele, J. Mater. Sci. 36 (2001) 1053 e1068. [274] H. Inaba, H. Tagawa, Solid State Ionics 83 (1996) 1e16. [275] B.C.H. Steele, Solid State Ionics 129 (2000) 95e110. [276]   M. Goedickemeier, L.J. Gauckler, J. Electrochem. Soc. 145 (1998) 414 e421. [277]  T. Takahashi, H. Iwahara, Y. Nagai, J. Appl. Electrochem. 2 (1972) 97 e104. [278] L. Boon, R. Metselaar, Solid State Ionics 16 (1985) 201e209. [279]   P.J. Gellings, H.J.M. Bouwmeester, Catal. Today 12 (1992) 1 e105. [280]  N. Jiang, E.D. Wachsman, S.-H. Jung, Solid State Ionics 150 (2002) 347 e353. [281]   N. Bonanos, K. Knight, B. Ellis, Solid State Ionics 79 (1995) 161 e170. [282]   C. Zuo, S. Zha, M. Liu, M. Hatano, M. Uchiyama, Adv. Mater. 18 (2006) 3318e3320. [283]  A. Hayashi, K. Noi, A. Sakuda, M. Tatsumisago, Nat. Commun. 3 (2012) 1 e5. [284] N. Anantharamulu, K.K. Rao, G. Rambabu, B.V. Kumar, V. Radha, J. Mater. Sci. 46 (2011) 2821e2837. [285]   F. Funabiki, T. Yano, S. Shibata, M. Yamane, Solid State Ionics 160 (2003) 281e288. [286] Y. Miyajima, T. Miyoshi, J. Tamaki, M. Matsuoka, Y. Yamamoto, C. Masquelier, M. Tabuchi, Y. Saito, H. Kageyama, Solid State Ionics 124 (1999) 201 e211. [287]  J.M. Winand, A. Rulmont, P. Tarte, J. Mater. Sci. 25 (1990) 4008 e4013. [288] S. Yamanaka, K. Yoshioka, M. Hattori, Solid State Ionics 40e41 (1990) 43e47. [289]   A.K. Ivanov-Schitz, A.B. Bykov, Solid State Ionics 100 (1997) 153 e155. [290]   M. Barj, H. Perthuis, P. Colomban, Solid State Ionics 9 e10 (1983) 845e850. [291]  K. Shimazu, Y. Yamamoto, Y. Saito, O. Nakamura, Solid State Ionics 86 e88 (1996) 106e110. [292] R.J. Cava, E.M. Vogel, D.W.J. Jr., J. Am. Ceram. Soc. 65 (1982) C157eC159. [293]   S. Susman, C.J. Delbecq, T.O. Brun, Solid State Ionics 9 e10 (1983) 839e844. [294]   Q. Zhang, Z. Wen, Y. Liu, S. Song, X. Wu, J. Alloys Compd. 479 (2009) 494e499. [295] H. Fujimoto, K. Ohara, Y. Zou, R. Yamamura, N. Inoue, Phys. Lett. A 305 (2002) 298e303. [296] N. Inoue, K. Oiwa, T. Hayashi, Ionics 6 (2000) 107e111. [297]   Y. Saito, K. Ado, T. Asai, H. Kageyama, O. Nakamura, Solid State Ionics 58 (1992) 327e331. [298] S. Yang, G. Li, L. You, J. Tao, C.K. Loong, S. Tian, F. Liao, J. Lin, Chem. Mater. 19 (2007). [299]   J.H. Kim, T.S. Oh, M.S. Lee, J.G. Park, Y.H. Kim, J. Mater. Sci. 28 (1993) €



“

”

1573e1577. [300] R.O. Fuentes, F.M.B. Marques, J.I. Franco, Bol. Soc. Esp. Ceram. Vidr. 38 (1999) 631e634. [301] X. Lu, G. Xia, J.P. Lemmon, Z. Yang, J. Power Sources 195 (2010) 2431 e2442. [302]   K. Edstrom, J.O. Thomas, G.C. Farrington, Acta Crystallogr. B 47 (1991) 210e216. [303]   M.S. Whittingham, R.A. Huggins, J. Chem. Phys. 54 (2003) 414 e416. [304]   J.F. Bocquet, G. Lucazeau, Solid State Ionics 24 (1987) 235 e247. [305] J. Fergus, Solid State Ionics 227 (2012) 102e112. [306] P.V. Wright, Br. Polym. J. 7 (1975) 319e327. [307] F.B. Dias, L. Plomp, J.B.J. Veldhuis, J. Power Sources 88 (2000) 169e191. [308]   F.M. Gray, Solid Polymer Electrolytes, VCH, New York, 1991. [309]  S.E. Bullock, P. Ko �nas, Mat. Res. Soc. (2002) 15.11 e15.14. [310]  G.S. Macglashan, Y.G. Andreev, P.G. Bruce, Nature 398 (1999) 792 e794. [311] F.B. Dias, L. Plomp, J. Veldhuis, J. Power Sources 88 (2000) 169 e191. [312]   G.B. Appetecchi, J. Hassoun, B. Scrosati, F. Croce, F. Cassel, M. Salomon,  J. Power Sources 124 (2003) 246e253. [313] J. Hassoun, H.-G. Jung, Y.-K. Sun, B. Scrosati, in: 219th ECS Meeting, 2011. [314] J.H. Shin, K.W. Kim, H.J. Ahn, J.H. Ahn, Mater. Sci. Eng. B 95 (2002) 148 e156. [315] B.H. Jeon, J.H. Yeon, K.M. Kim, I.J. Chung, J. Power Sources 109 (2002) 89e97. [316] Y.-J. Lim, Y.-H. An, N.-J. Jo, Nanoscale Res. Lett. 7 (2012) 1 e6. [317] A. Chandra, A. Chandra, S. Thakur, Int. J. Chem. Res. 3 (2011) 81 e85. [318]   K.M. Abraham, Z. Jiang, J. Electrochem. Soc. 143 (1996) 1 e5. [319] G. Jiang, S. Maeda, Y. Saito, S. Tanase, T. Sakai, J. Electrochem. Soc. 152 (2005) A767eA773. [320]   A.R. Polu, R. Kumar, K.V. Kumar, Adv. Mater. Lett. 3 (2012) 406 e409. [321] A.S. Fisher, M.B. Khalid, M. Widstrom, P. Ko �nas, J. Power Sources 196 (2011) 9767e9773. [322]   L.C. Rodrigues, P.C. Barbosa, M.M. Silva, M.J. Smith, Electrochim. Acta 53 (2007) 1427e1431. [323] G. Li, R. Xue, L. Chen, Solid State Ionics 90 (1996) 221e225. [324] Sung-Yoon Chung, J.T. Bloking, Y.-M. Chiang, Nat. Mater. 1 (2002) 123 e128. [325]   S. Ibrahim, M.R. Johan, Int. J. Electrochem. Sci. 7 (2012) 2596 e2615. [326] B.M. Abdel-Samiea, R. Gamal, E. Sheha, Cleantech, (2012) 164 e167. [327]  C.P. Rhodes, J.W. Long, M.S. Doescher, J.J. Fontanella, D.R. Rolison, J. Phys. Chem. B 108 (2004) 13079e13087. [328] A. Sakuda, A. Hayashi, M. Tatsumisago, Sci. Rep. 3 (2013) 1e5. [329]   R. Hooper, L.J. Lyons, D.A. Moline, R. West, Am. Chem. Soc. Commun. 18 (1999) 3249e3251. [330]   D.J. Bannister, G.R. Davies, I.M. Ward, J.E. McIntyre, Polymer 25 (1984) 1600e1602. [331] J. Sanchez, F. Alloin, D. Benrabah, R. Arnaud, Int. Conf. Solid State Ionics 68 (1997) 43e51. [332]   K. Ito, Y. Tominaga, H. Ohno, Electrochim. Acta 42 (1997) 1561 e1570. [333]   C.A. Angell, C. Liu, E. Sanchez, Nature 362 (1993) 137 e139. [334]   E. Yulianti, A.K. Karo, L. Susita, Sudaryanto, Procedia Chem. 4 (2012) 202e207. [335]  G. Nystrom, A. Razaq, M. Strømme, L. Nyholm, A. Mihranyan, Nano Lett. 9 (2009) 3635e3639. [336]   T. Okumura, S. Nishimura, Solid State Ionics 267 (2014) 68 e73. [337]  F. Croce, F. Capuano, A. Selvaggi, B. Scrosati, G. Scibona, J. Power Sources 32 (1990) 381e388. [338]   Y. Tominaga, V. Nanthana, D. Tohyama, Polym. J. 44 (2012) 1155 e1158. [339] J. Wolfenstine, J.L. Allen, J. Sumner, J. Sakamoto, Solid State Ionics 180 (2009) 961e967. [340]   T.S.A.S.N. Murthy, J. Power Sources 27 (1989) 119 e126. [341] I.R.E. Agency, in, 2013. [342]   S.I.B. Fang, Y. Wei, T. Arai, M. Kumagai, Electrochim. Acta 47 (2002) 3971e3976. [343]   J. Benziger, J. Nehlsen, D. Blackwell, T. Brennan, J. Itescu, J. Membr. Sci. 261 (2005) 98e106. [344]   H. Craighead, Nature 442 (2006) 387e393. [345] A. Dalvi, K. Shahi, Solid State Ionics 148 (2002) 431e436. [346] M.K. Debe, A.K. Schmoeckel, G.D. Vernstrom, R. Atanasoski, J. Power Sources 161 (2006) 1002e1011. [347] M. Doyle, J. Newman, J. Power Sources 54 (1995) 46e51. [348]  D.R.P. Egan, C.T.J. Low, F.C. Walsh, J. Power Sources 196 (2011) 5725 e5730. [349] U.D.o. Energy, Energy Ef �ciency and Renewable Energy, (2011). [350]  G. Acres, J. Frost, G. Hards, R. Potter, T. ralph, D. Thompsett, Catal. Today 38 (1997) 393e400. [351]   K.K. Hansen, M. Mogensen, ECS Trans. (2008) 153e160. [352]  C.-T. Hsieh, Jia-Yi Lin, J.-L. Wei, Int. J. Hydrog. Energy 34 (2009) 685 e693. [353]   Y. Inaguma, M. Nakashima, J. Power Sources 228 (2013) 250 e255. [354] J. Lahiri, Y. Lin, P. Bozkurt, I. Oleynik, M. Batzill, Nat. Nano 5 (2010) 326 e329. [355] X.H. Liu, L. Zhong, S. Huang, S.X. Mao, T. Zhu, J.Y. Huang, ACS Nano 6 (2012) 1522e1531. [356]   X.H. Liu, J.Y. Huang, Energy Environ. Sci. 4 (2011) 3844 e3860. [357]   J.C. Fisher, Acta Metall. 3 (1955) 109 e110. €