Three-Dimensionally Porous Li-Ion and Li-S Battery Cathodes: A Mini Review for Preparation Methods and Energy-Storage Performance

Among many types of batteries, Li-ion and Li-S batteries have been of great interest because of their high energy density, low self-discharge, and non-memory effect, among other aspects. Emerging applications require batteries with higher performance factors, such as capacity and cycling life, which have motivated many research efforts on constructing high-performance anode and cathode materials. Herein, recent research about cathode materials are particularly focused on. Low electron and ion conductivities and poor electrode stability remain great challenges. Three-dimensional (3D) porous nanostructures commonly exhibit unique properties, such as good Li+ ion diffusion, short electron transfer pathway, robust mechanical strength, and sufficient space for volume change accommodation during charge/discharge, which make them promising for high-performance cathodes in batteries. A comprehensive summary about some cutting-edge investigations of Li-ion and Li-S battery cathodes is presented. As demonstrative examples, LiCoO2, LiMn2O4, LiFePO4, V2O5, and LiNi1−x−yCoxMnyO2 in pristine and modified forms with a 3D porous structure for Li-ion batteries are introduced, with a particular focus on their preparation methods. Additionally, S loaded on 3D scaffolds for Li-S batteries is discussed. In addition, the main challenges and potential directions for next generation cathodes have been indicated, which would be beneficial to researchers and engineers developing high-performance electrodes for advanced secondary batteries.


Introduction
Rechargeable batteries are widely used in emergency power backup, electric vehicles, solar power storage, portable equipment, and wearable electronics [1,2]. Among them, Li-ion and Li-S batteries have competitive advantages to the others [3][4][5][6][7][8][9]. The principle of Li-ion batteries is the insertion and extraction of Li ions in the electrodes during charge and discharge [10][11][12][13], as illustrated in Figure 1. The electricity outside the battery is carried by electrons around the external circuit in the opposite direction to Li ions. For Li-S batteries, as displayed in Figure 2, S 8 reacts with Li ions to form high-order lithium polysulfides Li 2 S x (4 < x < 8), and then lower-order lithium polysulfides Li 2 S x (2 < x < 4) are formed as the Li ions insert continuously [14][15][16]. Currently, cathode materials are one of the key points for the development of high-performance Li-ion batteries and Li-S batteries [17,18]. Many cathode materials, almost with porous characteristics, have the advantages of enlarged surface area, increased specific capacity, and are able to overcome volume variation. Among various structures, three-dimensional (3D) porous structures are considered ideal and promising.
It is necessary to improve the kinetics of the extraction/insertion process of Li ion in Li-ion batteries and the reaction between Li ions and S in Li-S batteries, which can be facilitated by constructing a 3D porous morphology. In addition, 3D porous structures have several void spaces, which are able to improve the contact with electrolytes. However, a part of the electrolyte is needed to wet the active materials during the charge/discharge process, therefore the 3D structure would consume more electrolytes [19][20][21]. In addition, the reduced material loading within the 3D electrode also needs to be improved for practical applications. Both the advantages and drawbacks of the 3D porous electrodes make it an attractive field for intensive study. Herein, we focus on the 3D porous cathode materials used in both Li-ion and Li-S batteries and their preparation methods. Nanomaterials 2019, 9, x 2 of 20 volume variation. Among various structures, three-dimensional (3D) porous structures are considered ideal and promising. It is necessary to improve the kinetics of the extraction/insertion process of Li ion in Li-ion batteries and the reaction between Li ions and S in Li-S batteries, which can be facilitated by constructing a 3D porous morphology. In addition, 3D porous structures have several void spaces, which are able to improve the contact with electrolytes. However, a part of the electrolyte is needed to wet the active materials during the charge/discharge process, therefore the 3D structure would consume more electrolytes [19][20][21]. In addition, the reduced material loading within the 3D electrode also needs to be improved for practical applications. Both the advantages and drawbacks of the 3D porous electrodes make it an attractive field for intensive study. Herein, we focus on the 3D porous cathode materials used in both Li-ion and Li-S batteries and their preparation methods.  Reproduced with permission from [14]. The Royal Society of Chemistry, 2015.

Challenges of Li-Ion and Li-S Battery Cathodes
Performances of Li-ion batteries are partly dependent on the intrinsic property of the materials used in the cathodes. For cathode materials, high specific capacity, high potential, large ionic and electronic transport rate, and long life are preferred. Nowadays, most of the cathode materials are Libased intercalation compounds. The available compounds, such as LiNiO2, LiCoO2, and LiMnO2, are investigated for cathode applications. To achieve a high energy-density, some compounds composed It is necessary to improve the kinetics of the extraction/insertion process of Li ion in Li-ion batteries and the reaction between Li ions and S in Li-S batteries, which can be facilitated by constructing a 3D porous morphology. In addition, 3D porous structures have several void spaces, which are able to improve the contact with electrolytes. However, a part of the electrolyte is needed to wet the active materials during the charge/discharge process, therefore the 3D structure would consume more electrolytes [19][20][21]. In addition, the reduced material loading within the 3D electrode also needs to be improved for practical applications. Both the advantages and drawbacks of the 3D porous electrodes make it an attractive field for intensive study. Herein, we focus on the 3D porous cathode materials used in both Li-ion and Li-S batteries and their preparation methods.  Reproduced with permission from [14]. The Royal Society of Chemistry, 2015.

Challenges of Li-Ion and Li-S Battery Cathodes
Performances of Li-ion batteries are partly dependent on the intrinsic property of the materials used in the cathodes. For cathode materials, high specific capacity, high potential, large ionic and electronic transport rate, and long life are preferred. Nowadays, most of the cathode materials are Libased intercalation compounds. The available compounds, such as LiNiO2, LiCoO2, and LiMnO2, are investigated for cathode applications. To achieve a high energy-density, some compounds composed Reproduced with permission from [14]. The Royal Society of Chemistry, 2015.

Challenges of Li-Ion and Li-S Battery Cathodes
Performances of Li-ion batteries are partly dependent on the intrinsic property of the materials used in the cathodes. For cathode materials, high specific capacity, high potential, large ionic and Nanomaterials 2019, 9,441 3 of 20 electronic transport rate, and long life are preferred. Nowadays, most of the cathode materials are Li-based intercalation compounds. The available compounds, such as LiNiO 2 , LiCoO 2 , and LiMnO 2 , are investigated for cathode applications. To achieve a high energy-density, some compounds composed of multiple metal atoms, such as LiNi 1/3 Co 1/3 Mn 1/3 O 2 (NCM), are expected to replace part of the single metal compounds [22]. Compounds with MO 6 octahedra (where M is Fe, Mg, or V) and XO 4 n-tetrahedral anions (where X is Si or P) are of interest because these structures can tune transition metal redox potentials, such as Fe 3+ /Fe 2+ . However, these kinds of compounds have a common and fundamental problem in that the conductivity is poor, which reduces the electrochemical performance.
Compared to Li-ion batteries, Li-S batteries have a much higher theoretical energy density (2600 Wh kg −1 ) and better capacity (1672 mAh g −1 ) [23]. However, severe issues of sulfur need to be addressed. At first, elemental sulfur is a natural insulator (5 × 10 −30 S cm −1 ), which absolutely restricts the movements of electrons. The long-chain lithium polysulfide formed during the discharge process tends to dissolve in the organic electrolytes. When these polysulfides diffuse to the anode through the electrolytes, and react with the lithium anode, the reactions are named as shuttle mechanisms. During the transition between long-chain and short-chain lithium polysulfides, partial electrical energy is consumed and this unexpected phenomenon restricts the electrical efficiency. Meanwhile, an unstable film will be generated because of such uncontrolled reactions. As a result, the instability of solid electrolyte interface (SEI) film has a negative effect on the cycling stability of Li-S batteries. It is mentioned that the mass density of sulfur (2.03 g cm −3 ) is about 20% larger than that of Li 2 S (1.67 g cm −3 ), which causes the non-ignorable shrink and expansion during charge and discharge. Last but not least, the morphology of the lithium electrode is changeable due to the deposition of lithium ions from electrolytes. Dendrites of lithium growing on the surface of the electrode should be removed to ensure battery safety.

Hydrothermal Synthesis
Hydrothermal methods include the various techniques for crystallizing materials from high-temperature aqueous solutions at high vapor pressures. Compared with several other types of crystal growth, the hydrothermal approach is able to create crystalline phases that are not stable at the melting point. In addition, materials that have a high vapor pressure near their melting points can be grown by the hydrothermal method. These advantages make the hydrothermal method suitable for constructing many 3D electrodes for batteries. This section will mainly focus on the hydrothermal methods.
The conventional LiFePO 4 cathode has a poor electronic conductivity (~10 −9 S cm −1 ) and slow lithium ion diffusion, which restricts its wide application [24]. Several methods have been performed to enhance the electric conductivity and the rate of ion transport, including reducing the size of LiFePO 4 to the nanometer scale, coating particles with conductive materials, or expanding surface areas through structure manufacture. A 3D current collector of carbon layer coated LiFePO 4 was constructed, in order to improve the high-rate discharge capacity compared to conventional cells using the foil current collector. Furthermore, the carbon layer on LiFePO 4 nanoparticles with a diameter between 70 nm and 100 nm protects these particles and also contributes to conductivity enhancement [25]. In addition, the porous substrates made of nickel and chromium not only improve electric conductivity but also produce a large surface area.
Du et al. reported a facile in situ one-pot hydrothermal method for preparing a conductive graphene (G)/LiFePO 4 composite [26]. When LiFePO 4 nanoparticles load on the 3D graphene network, rapid electronic transmission and a short lithium ion transfer pathway can be achieved. The 3D G/LiFePO 4 composite exhibited a high capacity of 160 mAh g −1 at 0.2 C (94.12% of its theoretical capacity of 170 mAh g −1 ), a good rate performance of 115 mAh g −1 at 10 C (71.9% of its initial capacity), and good cycling performance of 94.2% capacity retention after 100 cycles, all of which imply their potential application in high rate Li-ion batteries.
Fu et al. [27] presented a novel hollow hierarchical structured composite of 0.5Li 2 MnO 3 ·0.5LiMn 0.4 Co 0.3 Ni 0.3 O 2 with a flower-like morphology. The presence of the internal cavity in the nanoplates, which consist of the complete flower-like composites, indicates that the hollow shell can provide channels for electrolytes and ions. The discharge capacities are 296.5, 270.6, 243.6, 207.8, and 187.4 mAh g −1 at rates of 0.2, 0.5, 1, 3, and 5 C, respectively. The capacity retention is over 87% after 100 cycles at 0.5 C with a capacity fading rate of 0.13% per cycle.
Li 3 V 2 (PO 4 ) 3 , with the crystal structure of monocline, is promising for the cathode material in Li-ion batteries because of its inherent characteristics of a high working potential of 4.8 V, theoretical capacity of 197 mAh g −1 , and higher energy density of 800 Wh kg −1 [28]. Li 3 V 2 (PO 4 ) 3 phase consists of a 3D framework of slightly distorted VO 6 octahedra-and PO 4 tetrahedra-sharing oxygen vertexes, which host Li ions in relatively large interstitial sites, leading to fast ionic transport. Cui et al. prepared carbon-coated Li 3 V 2 (PO 4 ) 3 nanocrystals that were modified by graphene nanosheets and carbon nanotubes through a hydrothermal method [29], as shown in Figure 3. The graphene nanosheets and the carbon nanotubes are interconnected to form a 3D conductive network, which also support Li 3 V 2 (PO 4 ) 3 particles. When using graphene sheets, the aggregated Li 3 V 2 (PO 4 ) 3 /C nanoparticles load on graphene sheets, forming a 3D layered structure. When the carbon nanotubes were also used, the interconnection between each graphene layer was improved. The carbon nanotubes enhance the structure strength and provide pathways for electron transfer. Considering the shaggy structure of the composites (Figure 4), the composite exhibits a remarkably high rate capability and long cycle stability. An initial discharge capacity of 147.5 mAh g −1 at 20 C at a potential range of 3.0-4.8 V is obtained. The capacity retention is 82.7% after 2000 cycles. shell can provide channels for electrolytes and ions. The discharge capacities are 296.5, 270.6, 243.6, 207.8, and 187.4 mAh g −1 at rates of 0.2, 0.5, 1, 3, and 5 C, respectively. The capacity retention is over 87% after 100 cycles at 0.5 C with a capacity fading rate of 0.13% per cycle. Li3V2(PO4)3, with the crystal structure of monocline, is promising for the cathode material in Liion batteries because of its inherent characteristics of a high working potential of 4.8 V, theoretical capacity of 197 mAh g −1 , and higher energy density of 800 Wh kg −1 [28]. Li3V2(PO4)3 phase consists of a 3D framework of slightly distorted VO6 octahedra-and PO4 tetrahedra-sharing oxygen vertexes, which host Li ions in relatively large interstitial sites, leading to fast ionic transport. Cui et al. prepared carbon-coated Li3V2(PO4)3 nanocrystals that were modified by graphene nanosheets and carbon nanotubes through a hydrothermal method [29], as shown in Figure 3. The graphene nanosheets and the carbon nanotubes are interconnected to form a 3D conductive network, which also support Li3V2(PO4)3 particles. When using graphene sheets, the aggregated Li3V2(PO4)3/C nanoparticles load on graphene sheets, forming a 3D layered structure. When the carbon nanotubes were also used, the interconnection between each graphene layer was improved. The carbon nanotubes enhance the structure strength and provide pathways for electron transfer. Considering the shaggy structure of the composites (Figure 4), the composite exhibits a remarkably high rate capability and long cycle stability. An initial discharge capacity of 147.5 mAh g −1 at 20 C at a potential range of 3.0-4.8 V is obtained. The capacity retention is 82.7% after 2000 cycles.  Schematic illustration of preparing nanosized Li 3 V 2 (PO 4 ) 3 /C (C-LVP), carbon nanotube modified Li 3 V 2 (PO 4 ) 3 /C (C-LVP/GNS), graphene nanosheet, and carbon nanotube co-modified Li 3 V 2 (PO 4 ) 3 /C (C-LVP/(GNS+CNT)) composites via a hydrothermal-assisted sol-gel route. Reproduced with permission from [29]. The Royal Society of Chemistry, 2016.
Vanadium pentoxide (V 2 O 5 ) is a typical intercalation compound. It possesses a layered structure that can accept more than one electron and lithium ion through the reaction V 2 O 5 + xLi + + xe − ↔ Li x V 2 O 5 . However, V 2 O 5 has shortcomings, with a moderate electrical conductivity of 10 −4 -10 −5 S cm −1 and low Li + diffusion coefficient of 10 −12 cm 2 s −1 [30]. As is known, binder-free technology loads active materials on the current collector without any binders and conductive adhesives, enabling a simplified electrode for batteries. Many free-standing 3D porous V 2 O 5 electrodes have been reported. For example, Gao et al. reported a one-step hydrothermal method to confine vanadium pentoxide particles in 3D N-doped graphene as a free-standing cathode [31]. The good conductivity and elastic properties of N-doped graphene are utilized efficiently. The composite cathode shows a high capacity of 283 mAh g −1 at 100 mA g −1 (96.3% of its theoretical capacity of 294 mAh g −1 ) and a good rate performance of 134 mA g −1 at 1 A g −1 (78% of its initial capacity). Figure 3. Schematic illustration of preparing nanosized Li3V2(PO4)3/C (C-LVP), carbon nanotube modified Li3V2(PO4)3/C (C-LVP/GNS), graphene nanosheet, and carbon nanotube co-modified Li3V2(PO4)3/C (C-LVP/(GNS+CNT)) composites via a hydrothermal-assisted sol-gel route. Reproduced with permission from [29]. The Royal Society of Chemistry, 2016. In addition, 3D hierarchical nanostructures are widely applied in electrodes because such structures suppress the agglomeration and improve the diffusion of lithium ions. Pan et al. found that the precursor of vanadyl oxalate (VOC 2 O 4 ) at different concentrations impacts on the morphology of the products [32]. As seen in Figure 5, they reported a facile solvothermal method to produce hierarchical nano and microstructures of vanadium oxide by adjusting the concentration of VOC 2 O 4 and the duration time. Urchin-like VO 2 microflowers were first synthesized without surfactants through self-assembly, which were followed by calcination to transform VO 2 into urchin-like V 2 O 5 microstructures. The V 2 O 5 microstructures were composed of nanobelts and nanorods, in which a highly porous texture was confirmed. The obtained V 2 O 5 had a specific area of 33.64 m 2 g −1 , which is ascribed to the 3D hierarchical porous structure and the large surface area of the building blocks. In addition, when evaluated as a cathode material, the V 2 O 5 cathode delivers a high specific discharge capacity of 267 mAh g −1 at a current density of 300 mAh g −1 .
Guo et al. reported a simple hydrothermal method for the synthesis of a fine V 2 O 5 -SnO 2 /CNTs composite, in which the nanosized V 2 O 5 coating and the carbon nanotubes (CNTs) were connected through SnO 2 nanoparticles [33]. After heat treatment, the composites were verified to be porous due to the stack of CNTs. The external surface of the CNTs was uniformly covered with SnO 2 and V 2 O 5 , which improved the cyclic capacity and the rate capacity as a result of the excellent conductivity. The V 2 O 5 -SnO 2 /CNTs composite cathode exhibited an initial discharge capacity of 250 mAh g −1 from 2.05 to 4.0 V. Furthermore, a high coulombic efficiency (over 99%) for all cycles was achieved, which showed an excellent cycling stability.
Our group has presented a facile method for manufacturing 3D structural electrodes by integrating active materials and graphene [34]. As displayed in Figure 6, a sandwich-like structure of V 2 O 5 @graphene@V 2 O 5 was fabricated for the cathode based on the 3D inverse opal template. At first, a Ni inverse opal was electrodeposited on a polystyrene opal template, on which graphene was then grown via chemical vapor deposition. After a Ni etching process, a layer of V 2 O 5 was grown onto the external surface of graphene. Then, another layer of V 2 O 5 was grown into the inner layer of graphene, forming the 3D V 2 O 5 @graphene@V 2 O 5 composite. The cathode provided a full electrode basis capacity of about 230 mAh g −1 at 5 C after 200 cycles, and 203 mAh g −1 after 2000 cycles, along with a Coulombic efficiency of 99.7%.
produce hierarchical nano and microstructures of vanadium oxide by adjusting the concentration of VOC2O4 and the duration time. Urchin-like VO2 microflowers were first synthesized without surfactants through self-assembly, which were followed by calcination to transform VO2 into urchinlike V2O5 microstructures. The V2O5 microstructures were composed of nanobelts and nanorods, in which a highly porous texture was confirmed. The obtained V2O5 had a specific area of 33.64 m 2 g −1 , which is ascribed to the 3D hierarchical porous structure and the large surface area of the building blocks. In addition, when evaluated as a cathode material, the V2O5 cathode delivers a high specific discharge capacity of 267 mAh g −1 at a current density of 300 mAh g −1 .  Guo et al. reported a simple hydrothermal method for the synthesis of a fine V2O5-SnO2/CNTs composite, in which the nanosized V2O5 coating and the carbon nanotubes (CNTs) were connected through SnO2 nanoparticles [33]. After heat treatment, the composites were verified to be porous due to the stack of CNTs. The external surface of the CNTs was uniformly covered with SnO2 and V2O5, which improved the cyclic capacity and the rate capacity as a result of the excellent conductivity. The V2O5-SnO2/CNTs composite cathode exhibited an initial discharge capacity of 250 mAh g −1 from 2.05 to 4.0 V. Furthermore, a high coulombic efficiency (over 99%) for all cycles was achieved, which showed an excellent cycling stability.
Our group has presented a facile method for manufacturing 3D structural electrodes by integrating active materials and graphene [34]. As displayed in Figure 6, a sandwich-like structure of V2O5@graphene@V2O5 was fabricated for the cathode based on the 3D inverse opal template. At first, a Ni inverse opal was electrodeposited on a polystyrene opal template, on which graphene was then grown via chemical vapor deposition. After a Ni etching process, a layer of V2O5 was grown onto the external surface of graphene. Then, another layer of V2O5 was grown into the inner layer of graphene, forming the 3D V2O5@graphene@V2O5 composite. The cathode provided a full electrode basis capacity of about 230 mAh g −1 at 5 C after 200 cycles, and 203 mAh g −1 after 2000 cycles, along with a Coulombic efficiency of 99.7%.

Sol-Gel Method
Another widely used method for preparing 3D electrodes is sol-gel. For example, Park et al. reported a sol-gel method with polystyrene beads as templates to prepare 3D LiMn2O4 thin films. A monolayer of polystyrene microspheres was first deposited on the surface of the substrates, which was the basis for manufacturing an open volume without overlaps in the perpendicular direction [35]. Subsequently, the prepared solvents containing precursor materials were dipped onto the substrates. After the calcination of deposited films, the polystyrene template was removed by the thermal decomposition. The precursor was converted to LiMn2O4. At last, a 3D spherical porous structure with an inverse-opal was formed. It was found that discharge capacity was associated with

Sol-Gel Method
Another widely used method for preparing 3D electrodes is sol-gel. For example, Park et al. reported a sol-gel method with polystyrene beads as templates to prepare 3D LiMn 2 O 4 thin films. A monolayer of polystyrene microspheres was first deposited on the surface of the substrates, which was the basis for manufacturing an open volume without overlaps in the perpendicular direction [35]. Subsequently, the prepared solvents containing precursor materials were dipped onto the substrates. After the calcination of deposited films, the polystyrene template was removed by the thermal decomposition. The precursor was converted to LiMn 2 O 4 . At last, a 3D spherical porous structure with an inverse-opal was formed. It was found that discharge capacity was associated with the annealing temperature, which has strong effects on the stable structure. Zhang et al. successfully encapsulated Li 3 V 2 (PO 4 ) 3 nanoparticles in amorphous carbon via a facile sol-gel method [36]. The presence of thin carbon layer and carbon network benefit Li + transport and electron conductivity. Applied as the cathode, this material exhibits a capacity of 85 mAh g −1 at a high rate of 30 C.
Li et al. reported a sol-gel method to construct the 3D LiAlO 2 -LiMnPO 4 /C for a Li-ion battery [37]. The LiAiO 2 template was first constructed from an anodic aluminum oxide (AAO) template. Then, LiMnPO 4 /C filled in the pores of the prepared template, and finally formed a honeycomb-like structure. When used as the cathode of a Li-ion battery, this material shows a high capability of 105 mAh g −1 and 98.4% retention after 100 cycles at 10 C. The electrochemical performance has been improved due to the special core-shell architectures.
Ragupathi et al. reported a sol-gel method to synthesize spherical LiCoBO 3 for use as the cathode material of a Li-ion battery [38]. The as-prepared material presents a capacity of 98 mAh g −1 at 10 C after 52 cycles. The electrochemical performance of this material is superior to others derived from different methods, which may be ascribed to the formation of its spherical shape, which facilitates electron movement and Li-ion diffusion.
The sol-gel method is commonly low-cost and simple, which make it appropriate for preparing electrodes with a large yield. Particularly, their industrial manufacture always contains a sol-gel preparation.

Solid-Based Approcah
Despite the wet-chemical strategies introduced above, some solid-based approaches have also attracted intensive attention. Wang et al. successfully utilized a topochemical reaction route to synthesize porous LiMn 2 O 4 spheres assembled by nanocrystalline, which took advantage of the porosity of Mn 2 O 3 spheres obtained from the thermal decomposition of MnCO 3 [39]. TEM and N 2 adsorption-desorption measurements showed that the whole LiMn 2 O 4 microspheres have an average size of about 45 nm. It was observed that the interior of LiMn 2 O 4 spheres, which showed a good performance in terms of capacity and cycling, were filled of abundant pores. The discharge capacity is 83 mAh g −1 at a rate of 20 C, which may be ascribed to the porous channels that are suitable for Li ion diffusion.
Huang et al. [40] used graphene and carbon nanotube conductive liquid (GNL) as a template to synthesize porous Li 1.2 Mn 0.534 Ni 0.133 Co 0.133 O 2 , as displayed in Figure 7. As for the GNL, CNTs are uniformly distributed on the layer of the graphene sheets. The manufacturing process of the composite is dependent on the preparation of Li 1.2 Mn 0.534 Ni 0.133 Co 0.133 O 2 through a high-temperature calcination of spherical precursors of transition metal carbonate. For the porous GNL-LMNCO with the spherical particles, the secondary particles are composed of the primary particles with sizes of 150-180 nm. The spheres of the secondary particles almost possess the same size of 2.0 mm. A great number of pores can be obviously detected from the stacking of the primary particles. The porous Li-rich cathode delivers a discharge capacity of 235.5 mAh g −1 , even after 100 cycles of charge/discharge, which suggests a high capacity retention of 86.2% compared to the initial capacity.
calcination of spherical precursors of transition metal carbonate. For the porous GNL-LMNCO with the spherical particles, the secondary particles are composed of the primary particles with sizes of 150-180 nm. The spheres of the secondary particles almost possess the same size of 2.0 mm. A great number of pores can be obviously detected from the stacking of the primary particles. The porous Lirich cathode delivers a discharge capacity of 235.5 mAh g −1 , even after 100 cycles of charge/discharge, which suggests a high capacity retention of 86.2% compared to the initial capacity.  In one study, de Biasi et al. synthesized a trigonal colquiriite-type crystal structure LiCaFeF6 via solid-state reaction approach [41]. This material's potential may be thanks to its safety and cycle life. When used as a cathode material, it exhibits a discharge capacity of 112 mAh g −1 and a maximum discharge current of 2.86 V. This was assigned to the reduction of Fe 3+ to Fe 2+ .
Baster et al. synthesized a series of LiNi 0.6 Co 0.4−Z Ti Z O 2 (z = 0.1, 0.2, 0.3) for the cathode materials of Li-ion batteries and compared their structural, transport, and electrochemical properties [42]. Of three materials, LiNi 0.6 Co 0.3 Ti 0.1 O 2 exhibits a discharge capacity of 100 mAh g −1 at 20 C.

Other Methods
Differing from the methods shown above, there are some other preparation approaches for making 3D cathodes for Li-ion batteries, including freeze-drying, chemical vapor deposition, vacuum filtration, and others. For instance, Doherty et al. synthesized beads of poly methyl methacrylate with different diameters as colloidal crystal templates to produce LiFePO 4 with featured pores [43]. The use of colloidal crystal templates allowed an examination of the effects of pore size in terms of the electrochemical properties. The materials with the largest pores, around 100 nm diameter, exhibited the best discharge capacities of 160 mAh g −1 at 0.1 C and 115 mA h g −1 at 5 C, respectively.
Tu et al. synthesized the 3D porous LiFePO 4 materials modified with the uniformly dispersed nitrogen-doped carbon nanotubes (N-CNTs) by a freeze-drying method [44]. It included three steps: rapid freezing, vacuum freeze-drying, and a heating treatment. The synthesized porous LiFePO 4 /NCNTs composite demonstrates a specific capacity of 159 mAh g −1 at a rate of 0.1 C, a rate performance with a capacity of 72 mAh g −1 at a high rate of 10 C, and a high capacity retention of 96.7% after 500 charge/discharge cycles.
Zhao et al. synthesized an orthorhombic LiMnO 2 nanorod by in-situ carbothermal reduction [45]. This approach included two steps: preparation of MnO 2 nanorod and carbothermal reduction to obtain LiMnO 2 . When used as a cathode in a Li-ion battery, this material exhibits outstanding electrochemical performance of 165.3 mAh g −1 at 0.1 C and 92.6% retention after 40 cycles. As for manganates, Duan et al. synthesized the LiMn 2 O 4 hollow nanofibers with a porous structure by modified electrospinning techniques on the fluorine-doped tin oxide glass [46]. The cathode made of these hollow materials delivered a specific capacity of 125.9 mAh g −1 and a cycling performance of 105.2 mAh g −1 after 400 cycles at 0.1 C, which exhibited a good battery performance.
A 3D mace-like Li 3 V 2 (PO 4 ) 3 /C nanowire and nanofiber membrane was prepared by using nanoparticles as a catalyst [47]. The fabrication process combines a modified electrospinning method with a hot-press treatment. The surface of the fibers is covered with nanowires that grow in situ, forming a mace-like morphology. Based on the mechanical property of the composite, the long-rangenetworking membrane is directly available for self-standing cathodes. The obtained cathode exhibits a good rate performance and cyclic stability in the voltage range of 3.0-4.8 V. Discharge capacities of 115.3 mAh g −1 and 108.6 mAh g −1 at 5 C and 10 C are achieved, respectively. As for the capacity retention, it is 81.4% after 500 cycles, and 78.8% after 1000 cycles, showing quite good stability. Li et al. successfully synthesized a Li 3 V 2 (PO 4 ) 3 /C composite with 3D foam-core structure via a facile immersion method [48]. The uniform Li 3 V 2 (PO 4 ) 3 /C bubbles were observed. Li 3 V 2 (PO 4 ) 3 nanoparticles were coated uniformly by an amorphous carbon layer. The cathode exhibits a good rate performance with a specific capacity of 100 mAh g −1 at 15 C in the potential range of 3.0-4.3 V, which is attributed to the 3D porous frog egg-like structure. Zhu et al. reported the synthesis of 3D microporous graphene-Li 2 FeSiO 4 /C nanocomposites [49]. As displayed in Figure 8, the 3D porous graphene framework was obtained through a templated-assembly method. During the preparation process, the surface grafted and positively charged SiO 2 spheres were assembled with negatively charged graphene oxide (GO) sheets by electrostatic interactions. The SiO 2 spheres were then removed, forming a 3D porous graphene framework. Finally, a Li 2 FeSiO 4 precursor gel was coated and annealed to get the 3D G/Li 2 FeSiO 4 /C composite. These composite have the advantage of the porosity of a 3D structure and the conductivity of graphene, with the discharge capacities reaching 255 mAh g −1 at 1 C. In addition, a flexible, free standing, and light-weight LiMnTiO4/MWCNT was synthesized by vacuum filtration method [50]. The prepared material possesses a 3D structure, where CNT networks are well wrapped in LiMnTiO4 particles. For superior performance aspects, such as flexibility, low weight, and high mechanical property, this material was used as a free-standing cathode and delivers a capacity of 161 mAh g −1 and 86.4% retention after 50 cycles at 0.5 C. This may be due to the promotion of Li + and electron transformation through the porous, MWCNTs. More details about the comparison with some 3D Li-ion battery cathodes are listed in Table 1.  In addition, a flexible, free standing, and light-weight LiMnTiO 4 /MWCNT was synthesized by vacuum filtration method [50]. The prepared material possesses a 3D structure, where CNT networks are well wrapped in LiMnTiO 4 particles. For superior performance aspects, such as flexibility, low weight, and high mechanical property, this material was used as a free-standing cathode and delivers a capacity of 161 mAh g −1 and 86.4% retention after 50 cycles at 0.5 C. This may be due to the promotion of Li + and electron transformation through the porous, MWCNTs. More details about the comparison with some 3D Li-ion battery cathodes are listed in Table 1.

3D Porous S Cathodes for Li-S Batteries
There are several reasons that the elemental sulfur itself cannot be fabricated as the cathode in Li-S batteries. One is that the sulfur is insulative and does not have a firm crystal to form a solid structure. In this case, after sulfur reacts with lithium ions, the resultant Li 2 S cannot be fixed in situ. The alternative method is to combine sulfur with other conductive materials. A frame strategy is promising and viable for solving the issues of elemental sulfur by coating sulfur onto a frame that has good conductivity, a large surface, and a porous structure [77][78][79]. Until now, carbon materials, conductive polymers, and metal or metal oxides have been studied for sulfur-based cathode materials.
The energy stored in Li-S batteries is closely related to the mass of the sulfur. As such, 3D porous structures are desired to increase the mass loading.
Various kinds of carbon matrices, including carbon nanotubes, graphene, carbon nanofibers, carbon spheres, and mesoporous carbon, have been chosen to act as a frame for Li-S batteries. In addition, the fabrication of conceptually new structures is also attractive. For example, Lu et al. reported an approach to assemble a sulfur-graphene sponge with sulfur uniformly distributed into the pores of a graphene sponge [80]. The graphene sponge framework can not only improve the electronic conductivity but also accommodate the volume change in the charge and discharge process. The electrochemical measurement showed a high areal specific capacity of 4.53 mAh cm −2 after 300 cycles and a slow decay rate of 0.08% per cycle after 300 cycles at 0.1 C. To some extent, the interlaced graphene layer can absorb the polysulfides intermediate product. More performance comparison about 3D S cathodes are listed in Table 2.
Li et al. reported a novel 3D hierarchical polypyrrole loading sulfur [81]. Firstly, silica colloid crystal was used as a template to synthesize 3D polypyrrole, which possesses a large specific area. Subsequently, sulfur was loaded on the surface of polypyrrole via a solution-diffusion route followed by low-temperature treatment. When used as cathode for Li-S batteries, this material shows a high specific capacity of 751 mAh g −1 after 100 cycles at 0.1 C.
A wood-inspired multi-channel tubular graphene architecture prepared by chemical vapor deposition was also studied [82]. This unique structure provides a large interior space that accommodates sulfur and fixes the generated polysulfide. Meanwhile, the 3D multi-channel network also promotes mass transport and ion diffusion. The benefit from these properties is that the MCTG/S cathode for Li-S batteries delivers a discharge capacity of 1390 mAh g −1 at 0.1 C.
Two composites of ZnO@S/CNT and Ni(OH) 2 @S/CNT were reported and applied in Li-S batteries [83]. ZnO@S/CNT illustrates higher initial capacity of 1663 mAh g −1 and 56.6% retention after 70 cycles. It can be attributed to the strong interaction between ZnO and S 2− x based on density functional theory (DFT) calculation.
Moreover, Li et al. synthesized Ni 3 S 2 and coated it onto the surface of Ni foam by the hydrothermal method, followed by physically coating elemental S, and finally obtained 3D Ni/Ni 3 S 2 /S [84]. As displayed in Figures 9 and 10, the reaction between Ni metal and sulfur generates Ni 3 S 2 in situ on the surface of the Ni foam. Within the hybrid, the inner Ni metal network could serve as an electron transport pathway to facilitate the electrochemical reactions. In addition, the sufficient pore spaces enable a high sulfur loading and endure the volume change caused by the sulfur. When the loading amount is 4 mg cm −2 , this material delivers a discharge capacity of 441 mAh g −1 after 100 cycles. situ on the surface of the Ni foam. Within the hybrid, the inner Ni metal network could serve as an electron transport pathway to facilitate the electrochemical reactions. In addition, the sufficient pore spaces enable a high sulfur loading and endure the volume change caused by the sulfur. When the loading amount is 4 mg cm −2 , this material delivers a discharge capacity of 441 mAh g −1 after 100 cycles. Figure 9. Schematic illustration of the Ni/Ni3S2/S hybrid cathode. Reproduced with permission from [84]. American Chemical Society, 2017.   Reproduced with permission from [84]. American Chemical Society, 2017.
The in situ construction method was also used for preparing oxide-based composites. For example, Song et al. reported VO2-VN as an excellent sulfur host for Li-S batteries [85]. The S@VO2-VN cathode displayed a remarkable rate capacity of 935 mAh g −1 and 56.1% retention after 800 cycles at 2 C. As illustrated in Figure 11, the good performance ascribed to the improved interfacial contact between VO2 and VN built-up by the in situ construction route guarantees fast diffusion of LiPSs from VO2 to VN, resulting in an improved conversion and accelerated redox kinetics towards Li2S precipitation. The in situ construction method was also used for preparing oxide-based composites. For example, Song et al. reported VO 2 -VN as an excellent sulfur host for Li-S batteries [85]. The S@VO 2 -VN cathode displayed a remarkable rate capacity of 935 mAh g −1 and 56.1% retention after 800 cycles at 2 C. As illustrated in Figure 11, the good performance ascribed to the improved interfacial contact between VO 2 and VN built-up by the in situ construction route guarantees fast diffusion of LiPSs from VO 2 to VN, resulting in an improved conversion and accelerated redox kinetics towards Li 2 S precipitation. Nanomaterials 2019, 9, x 12 of 20 Figure 11. Illustration of LiPS anchoring-diffusion-conversion processes on VO2, VN, and VO2-VN binary host surfaces. Reproduced with permission from [85]. The Royal Society of Chemistry, 2018.
Several oxide-based cathodes were reported to possess the adsorptive capability towards polysulfides, including Li2S2, Li2S4, and Li2S6. It is necessary to reduce the shuttle effect of a Li-S battery, leading to improved stability. In contrast, the sulfur-based cathodes commonly suffer from loss of active materials because of the dissolving and transferring of polysulfides. The relatively high loading is more attractive for sulfur-based cathodes, as the oxide hosts are usually of large weight and volume compared to the active component sulfur within the electrodes. As for the preparation of Li-S battery cathode, many techniques used for making Li-ion battery cathodes are also applicable. Some methods are summarized in Table 2. However, what should be noted is that the low melting point of sulfur makes the high-temperature treatment unsuitable.  Several oxide-based cathodes were reported to possess the adsorptive capability towards polysulfides, including Li 2 S 2 , Li 2 S 4 , and Li 2 S 6 . It is necessary to reduce the shuttle effect of a Li-S battery, leading to improved stability. In contrast, the sulfur-based cathodes commonly suffer from loss of active materials because of the dissolving and transferring of polysulfides. The relatively high loading is more attractive for sulfur-based cathodes, as the oxide hosts are usually of large weight and volume compared to the active component sulfur within the electrodes. As for the preparation of Li-S battery cathode, many techniques used for making Li-ion battery cathodes are also applicable. Some methods are summarized in Table 2. However, what should be noted is that the low melting point of sulfur makes the high-temperature treatment unsuitable.

Brief Comparison between Li-Ion and Li-S Batteries
As for the Li-ion batteries, they are advantageous as they show good safety, high working voltage, good specific capacity at low cycling rate, long cyclic performance, and low self-discharge. In parallel, for the Li-S batteries, capacities are commonly from 700 to 1500 mAh g −1 , which are competitive with some other batteries. Nevertheless, there are some challenges for Li-S batteries, such as the dissolution of polysulfide in electrolyte, poor conductivity of sulfur, and the damage to the active materials caused from volume change. For example, the conductivity of S means it needs more conductive carbon and polymer binder in the cathode, which decreases the mass content of active materials within the electrode. Therefore, improving the capacity of the Li-ion battery and increasing the loading of sulfur in the Li-S battery are necessary to achieve a good energy density in practice.

Summary
Some important investigations on the 3D porous-structured electrodes for both Li-ion and Li-S batteries have been summarized. Compared to conventional thick-film electrodes, 3D porous electrodes enable fast electron and Li ion transfer, as well as a robust mechanical strength, and sufficient space for volume change accommodation during charge/discharge. However, 3D porous electrodes usually consume more electrolytes, which should be addressed in future studies. Potentially, constructing light-weight 3D frames and optimizing the porous structure could be promising. For such design and optimization, some artificial intelligence (AI) methods, such as genetic programming and automated neural network, are available [113]. They are able to establish models for optimizing battery pack enclosures for electric vehicles, and analyzing the capacities for electrodes [114,115], which is necessary for achieving an optimal 3D-structured electrode [116]. As demonstrative examples, LiCoO 2 , LiMn 2 O 4 , LiFePO 4 , V 2 O 5 , LiNi 1−x−y Co x Mn y O 2 in pristine and modified forms with a 3D porous structure for Li-ion batteries have been introduced, while S loaded on 3D nanoframes for Li-S batteries are discussed as well. The preparation approaches, including template-directed chemical vapor deposition, hydrothermal synthesis, atomic layer deposition, and the related electrochemical properties, are focused on. In view of the limited length of the mini review, some of the methods have not been discussed in depth. There are still some electrode materials out of the range of this review. However, we hope the review presented here could potentially be significant for researchers and engineers who are working in related fields. In addition, it is believed that 3D porous-structured materials for next generation cathodes are promising, which would be of benefit for developing high-performance advanced secondary batteries.

Conflicts of Interest:
The authors declare no conflict of interest.