Materials and Processing of Lithium-Ion Battery Cathodes

: Lithium-ion batteries (LIBs) dominate the market of rechargeable power sources. To meet the increasing market demands, technology updates focus on advanced battery materials, especially cathodes, the most important component in LIBs. In this review, we provide an overview of the development of materials and processing technologies for cathodes from both academic and industrial perspectives. We brieﬂy compared the fundamentals of cathode materials based on intercalation and conversion chemistries. We then discussed the processing of cathodes, with speciﬁc focuses on the mechanisms of a drying process and the role of the binders. Several key parameters for the development of thick electrodes were critically assessed, which may offer insights into the design of next-generation batteries.


Introduction
The challenge of climate change is pushing countries all around the world to realize an energy transition from traditional fossil fuels to clean energy resources [1,2]. To realize this transition, various energy storage systems were developed in recent decades. Among them, a lithium (Li)-ion battery (LIB) is one of the most successful systems and it promoted the revolution of electronics, wearables, transportation, and grid energy storage [3][4][5]. With the development of electric transportation from road to sea and air (Figure 1a), the future will clearly be electric. This transition is not only driven by the progress in energy-and electronic-related technologies, but also by the rapidly falling prices of LIBs [6,7]. In 1991, the cell price of LIBs was high as ca. 5000 $ kWh −1 , but in 2020, it was 101 $ kWh −1 , as shown in Figure 1b [7]. As predicted by us, the price of LIBs will continue falling all the way to ca. 30-40 $/kWh by around 2040-2050 [8], which, in addition to expected cycle stability gains, is expected to provide more opportunities for the development of LIBs towards broader potential applications. Nanoenergy Adv. 2023, 3, FOR PEER REVIEW 2 (Co) minerals bring additional challenges for large-scale applications. Conversion-type cathodes such as sulfur (S), metal (e.g., iron, Fe) fluorides (FeF3, FeF2), and oxyfluorides could be alternatives because of their higher specific energy and lower cost [10][11][12]. In the past two decades, great efforts were devoted to different cathode materials. However, both conventional and much more so conversion-type cathodes still suffer from various issues [10]. In addition to cathode materials themselves, improvements in electrode processing technologies are also crucial to cell performance and cost [13][14][15]. The development of advanced processing technologies calls for critical considerations of the fabrication procedures, such as mixing, drying, pressing, and lamination [13]. Additionally, the use of advanced binder systems is important for the fabrication of high-performance cathodes, with considerable impacts on their mechanical and electrochemical properties. To design of a cathode electrode with optimal performance, basic parameters such as the defects and crystallinity of cathode particles, particle size and distribution, electrode architecture, and porosity and tortuosity should be taken into consideration [16][17][18]. This review aims to provide a comprehensive overview of materials and processing technologies currently utilized in LIB cathodes. We discuss the main features and issues of cathode materials of both intercalation and conversion types. We then delve into the processing technologies and binder systems for cathode electrodes, and analyze the key parameters that determine the electrodes' performance. The discussion also covers the (b) was reproduced from Ref. [7] (the observed BNEF data were from BloombergNEF).
A LIB cell mainly consists of two electrolyte-filled electrodes (cathode and anode), a porous separator membrane also filled with the same liquid electrolyte, and a sealing case. Among these components, the cathode usually has the highest weight fraction and often plays a dominant role in attainable cost and many cell performance characteristics, especially those that are related to battery weight, safety, and calendar life. Conventional intercalation cathodes such as lithium iron phosphate (LiFePO 4 , LFP), lithium cobalt oxide (LiCoO 2 , LCO), lithium manganese oxide (LiMn 2 O 4 , LMO), and lithium nickel cobalt manganese (or aluminum) oxide (NCM or NCA) are widely used in current LIBs [9]. However, the relatively low energy density and specific energy of these materials could be a barrier for some applications. Additionally, the increasing prices, the limited resources, various challenges with environmentally friendly mining and the toxicity of nickel (Ni) and cobalt (Co) minerals bring additional challenges for large-scale applications. Conversion-type cathodes such as sulfur (S), metal (e.g., iron, Fe) fluorides (FeF 3 , FeF 2 ), and oxyfluorides could be alternatives because of their higher specific energy and lower cost [10][11][12]. In the past two decades, great efforts were devoted to different cathode materials. However, both conventional and much more so conversion-type cathodes still suffer from various issues [10]. In addition to cathode materials themselves, improvements in electrode processing technologies are also crucial to cell performance and cost [13][14][15]. The development of advanced processing technologies calls for critical considerations of the fabrication procedures, such as mixing, drying, pressing, and lamination [13]. Additionally, the use of advanced binder systems is important for the fabrication of high-performance cathodes, with considerable impacts on their mechanical and electrochemical properties. To design of a cathode electrode with optimal performance, basic parameters such as the defects and crystallinity of cathode particles, particle size and distribution, electrode architecture, and porosity and tortuosity should be taken into consideration [16][17][18].
This review aims to provide a comprehensive overview of materials and processing technologies currently utilized in LIB cathodes. We discuss the main features and issues of cathode materials of both intercalation and conversion types. We then delve into the processing technologies and binder systems for cathode electrodes, and analyze the key parameters that determine the electrodes' performance. The discussion also covers the mechanical properties of cathodes from particle to electrode levels. We present these discussions from both academic and industrial perspectives, which will enable students, researchers, engineers, and others to gain a thorough understanding of the fundamentals involved in LIB cathodes.

Cathode Materials
Current LIBs are mainly built on intercalation chemistry, which enables the intercalation/extraction of Li ions in/from bulk electrode materials for thousands of cycles. Conventional intercalation cathodes mainly include LFP, LCO, LMO, NCM, and NCA. Among them, LCO, LFP, and NCM were particularly widely used in the battery market for electronics (mostly LCO) and electric vehicle (EVs) (mostly LFP, NCM and NCA). However, these materials are still limited by the attainable energy density, specific energy, and cost. Conversion cathode materials could be the solution, considering that conversion-type anodes (e.g., silicon (Si)) were successfully commercialized in recent years and their production and market penetration are growing rapidly. The replacement of conventional cathodes by conversion-type cathodes, such as sulfur, metal fluorides, and oxyfluorides can largely increase the specific energy and reduce the cost of the LIB cells. However, there are still significant limitations needed to be overcome before the scale-up and broad adoption of conversion-type cathodes takes place. Figure 2 illustrates the crystal structures of typical intercalation and conversion cathode materials.  [19] MDPI, ref [20] Elsevier, and ref [21] RSC.

Intercalation Cathodes
Currently, intercalation cathodes are widely used in LIBs due to their high potentials (versus Li/Li + ), long cycle life, and good structural stability in air. Layered LCO, spinel LMO, and olive LFP were widely studied in the past few decades. The first-generation commercial LIBs were built on LCO in 1990s. However, the limited energy density and high cost of Co are the major issues that hinder their further applications. In addition, there are major issues with Co supply chain [22,23]. The use of LFP or LMO could overcome supply chain limitations and reduce the cost of cathode materials, but cell energy would be reduced as well. In addition, LFP (and to a lesser extent LMO) cathodes are not economical to recycle [24]. Yet, in many cases, cost, supply chain and safety are still the major concerns. As such, LFP was recently more and more used in EV batteries despite its low energy density. Ni-based layered transition metal oxides such as NCM and NCA could provide higher energy density than LFP and LMO, but cost less than LCO, which enables them to be used in state-of-the-art EVs with a longer driving range. In recent years, researchers explored ultrahigh-Ni layered NCM/NCA, Li-and/or Mn-rich NCM, and phosphate polyanion cathodes and pushed their energy density and cycle stability to satisfy the requirements for practical applications [25][26][27]. In general, intercalation cathodes could undergo a capacity degradation due to surface phase transition, the dissolution of transition metals, gas release, or microcracking [12,28,29]. To address these challenges, researchers proposed various strategies, including the use of single crystalline materials (which do not crack), doping, protective surface layer formation, and advancing applications of electrolyte additives that better passivate the cathode surface, among others [12]. For example, using single-crystal LiNi0.6Co0.2Mn0.2O2 (NCM622) for LIB pouch cells can achieve significantly better cycle performance with a capacity retention of 83% after 3000 cycles, compared to polycrystalline NCM622 [30]. The introduction of various dopants, such as 1 mol.% molybdenum (Mo), for example, to Ni-rich layered oxide cathodes enabled a higher voltage and higher energy density, such as 880 Wh kg −1 cathode for 1000 stable cycles in layered Co-free NCM901 [31]. Despite these advances, it may be difficult to further improve their energy density of these materials by a significant degree, considering the theoretical limitations from intercalation chemistry.

Conversion Cathodes
As alternatives to current intercalation cathodes, conversion-type cathodes featuring sulfur (S) and metal fluorides can make use of conversion reactions during charging/discharging and achieve multiple electron transfers, which enables higher specific capacity and energy to be attained. Figure 3 compares the specific energy for a cell using different cathodes coupled with a conventional graphite (C) anode (Figure 3a), next-generation Si anode (Figure 3b), or Li metal anode (Figure 3c). Compared to conventional intercalation cathodes, the conversion-type cathodes such as Li2S, CuF2, and FeF3 higher specific energies, especially when coupled with Si ( Figure 3b) or Li metal anodes (Figure 3c), such as

Intercalation Cathodes
Currently, intercalation cathodes are widely used in LIBs due to their high potentials (versus Li/Li + ), long cycle life, and good structural stability in air. Layered LCO, spinel LMO, and olive LFP were widely studied in the past few decades. The first-generation commercial LIBs were built on LCO in 1990s. However, the limited energy density and high cost of Co are the major issues that hinder their further applications. In addition, there are major issues with Co supply chain [22,23]. The use of LFP or LMO could overcome supply chain limitations and reduce the cost of cathode materials, but cell energy would be reduced as well. In addition, LFP (and to a lesser extent LMO) cathodes are not economical to recycle [24]. Yet, in many cases, cost, supply chain and safety are still the major concerns. As such, LFP was recently more and more used in EV batteries despite its low energy density. Ni-based layered transition metal oxides such as NCM and NCA could provide higher energy density than LFP and LMO, but cost less than LCO, which enables them to be used in state-of-the-art EVs with a longer driving range. In recent years, researchers explored ultrahigh-Ni layered NCM/NCA, Li-and/or Mn-rich NCM, and phosphate polyanion cathodes and pushed their energy density and cycle stability to satisfy the requirements for practical applications [25][26][27]. In general, intercalation cathodes could undergo a capacity degradation due to surface phase transition, the dissolution of transition metals, gas release, or microcracking [12,28,29]. To address these challenges, researchers proposed various strategies, including the use of single crystalline materials (which do not crack), doping, protective surface layer formation, and advancing applications of electrolyte additives that better passivate the cathode surface, among others [12]. For example, using single-crystal LiNi 0.6 Co 0.2 Mn 0.2 O 2 (NCM622) for LIB pouch cells can achieve significantly better cycle performance with a capacity retention of 83% after 3000 cycles, compared to polycrystalline NCM622 [30]. The introduction of various dopants, such as 1 mol.% molybdenum (Mo), for example, to Ni-rich layered oxide cathodes enabled a higher voltage and higher energy density, such as 880 Wh kg −1 cathode for 1000 stable cycles in layered Co-free NCM901 [31]. Despite these advances, it may be difficult to further improve their energy density of these materials by a significant degree, considering the theoretical limitations from intercalation chemistry.

Conversion Cathodes
As alternatives to current intercalation cathodes, conversion-type cathodes featuring sulfur (S) and metal fluorides can make use of conversion reactions during charging/discharging and achieve multiple electron transfers, which enables higher specific capacity and energy to be attained. Figure 3 compares the specific energy for a cell using different cathodes coupled with a conventional graphite (C) anode (Figure 3a), next-generation Si anode (Figure 3b), or Li metal anode (Figure 3c). Compared to conventional intercalation cathodes, the conversion-type cathodes such as Li 2 S, CuF 2 , and FeF 3 higher specific energies, especially when coupled with Si ( Figure 3b) or Li metal anodes (Figure 3c), such as Li 2 S-Si, CuF 2 -Si, Li 2 S-Li, and CuF 2 -Li systems. In previous reviews, researchers critically discussed the pros and cons of different conversion cathodes [10,32,33]. Nanoenergy Adv. 2023, 3, FOR PEER REVIEW 4 Li2S-Si, CuF2-Si, Li2S-Li, and CuF2-Li systems. In previous reviews, researchers critically discussed the pros and cons of different conversion cathodes [10,32,33]. Despite their high energy densities, conversion cathodes currently suffer from a much faster capacity degradation during cycling compared to intercalation cathodes. In many cases, conversion cathodes can be only charged/discharged for few hundreds of cycles or less, especially in high areal capacity loadings or at elevated temperatures. The decreased cycle stability could be attributed to several issues, such as the dissolution of active species, the large structural changes, and side reactions [12]. For example, in a Li-S cell, the dissolution of polysulfide intermediates could result in a shuttle effect, which seriously affect the cell performance [34]. Additionally, the lack of stable interphase on cathodes could also limit their electrochemical stability, in contrast to a SEI layer formed on anodes during first few cycles [35][36][37]. As such, the stabilization of cathodes and interphases is essentially important to the development of next-generation LIBs. Effective strategies could cover the development of cathode coatings, the addition of advanced electrolytes or additives, the functionalization of separators/interlayers, or the use of multifunctional binders. We critically discussed these strategies in an article [12].

Electrode Processing
Conventionally, the manufacturing of cathode electrodes is based on a slurry-based process, which starts from mixing active and inactive materials (binders, conductive additives) with a suitable solvent to form a uniform slurry, then coating the slurry onto a current collector (Al for cathodes) foil and drying, then calendaring (densifying) the electrode to attain high packing density and high volumetric capacity, then electrode cutting suitable shapes, assembling the electrodes and membranes into a cell, and then finally conducting a formation cycle (initial charge and discharge, often at elevated temperatures to accelerate the process), degassing and testing for capacity retention and quality, as shown in Figure 4 [38]. In a slurry-based process, the drying process (removing a slurry solvent) is the key that often defines electrode behavior and a significant factor to total electrode fabrication costs. The drying step often involves multiple transport mechanisms (such as diffusion, convection, and migration) and is dependent on the involved volumetric sources/sinks (such as phase change, chemical bonding, gravity) and driving forces (such as local temperature, concentration, and pressure) [39]. In the following subsections, we mainly discuss the conventional slurry-based process, the role of the polymer binders, and the recently introduced solvent-free dry electrode process. Despite their high energy densities, conversion cathodes currently suffer from a much faster capacity degradation during cycling compared to intercalation cathodes. In many cases, conversion cathodes can be only charged/discharged for few hundreds of cycles or less, especially in high areal capacity loadings or at elevated temperatures. The decreased cycle stability could be attributed to several issues, such as the dissolution of active species, the large structural changes, and side reactions [12]. For example, in a Li-S cell, the dissolution of polysulfide intermediates could result in a shuttle effect, which seriously affect the cell performance [34]. Additionally, the lack of stable interphase on cathodes could also limit their electrochemical stability, in contrast to a SEI layer formed on anodes during first few cycles [35][36][37]. As such, the stabilization of cathodes and interphases is essentially important to the development of next-generation LIBs. Effective strategies could cover the development of cathode coatings, the addition of advanced electrolytes or additives, the functionalization of separators/interlayers, or the use of multifunctional binders. We critically discussed these strategies in an article [12].

Electrode Processing
Conventionally, the manufacturing of cathode electrodes is based on a slurry-based process, which starts from mixing active and inactive materials (binders, conductive additives) with a suitable solvent to form a uniform slurry, then coating the slurry onto a current collector (Al for cathodes) foil and drying, then calendaring (densifying) the electrode to attain high packing density and high volumetric capacity, then electrode cutting suitable shapes, assembling the electrodes and membranes into a cell, and then finally conducting a formation cycle (initial charge and discharge, often at elevated temperatures to accelerate the process), degassing and testing for capacity retention and quality, as shown in Figure 4 [38]. In a slurry-based process, the drying process (removing a slurry solvent) is the key that often defines electrode behavior and a significant factor to total electrode fabrication costs. The drying step often involves multiple transport mechanisms (such as diffusion, convection, and migration) and is dependent on the involved volumetric sources/sinks (such as phase change, chemical bonding, gravity) and driving forces (such as local temperature, concentration, and pressure) [39]. In the following subsections, we mainly discuss the conventional slurry-based process, the role of the polymer binders, and the recently introduced solvent-free dry electrode process.

Drying Process
In the slurry-based process, the drying process is one of the complicated and decisive procedures, which influences the porous structure formation and the distribution of cathode particles, binders, and conductive additives in the electrodes. A typical drying process can be typically classified to two stages, including surface solvent evaporation and pore-

Drying Process
In the slurry-based process, the drying process is one of the complicated and decisive procedures, which influences the porous structure formation and the distribution of cathode particles, binders, and conductive additives in the electrodes. A typical drying process can be typically classified to two stages, including surface solvent evaporation and poreemptying evaporation [40]. As shown in Figure 5, the surface solvent evaporation is a dominant and fast solvent removal process [41], which could enable over the removal of over 90% solvent with a half of time period in the total drying process [42].

Drying Process
In the slurry-based process, the drying process is one of the complicated and decisive procedures, which influences the porous structure formation and the distribution of cathode particles, binders, and conductive additives in the electrodes. A typical drying process can be typically classified to two stages, including surface solvent evaporation and poreemptying evaporation [40]. As shown in Figure 5, the surface solvent evaporation is a dominant and fast solvent removal process [41], which could enable over the removal of over 90% solvent with a half of time period in the total drying process [42]. In the beginning, the solution is dilute and different components suspend in slurry, resulting a sufficiently large and constant temperature enclosure. In ambient environment, the partial pressure of the solvent remains constant and contributes to constant evaporation rate [43]. Most of the solvent evaporation happens from the liquid surface of the solid-solvent slurry and the film shrinkage further processes until the active material consolidation finishes. Through a gravity force and due to a high evaporation rate, heavy active material particles approach each other and finally sediment to form a compact active material skeleton with tiny pores, in a process called film consolidation [41], where the thickness of consolidated film then remain fixed through the remaining solvent evaporation. In this case, the relatively light mass of conductive additives (e.g., carbon black) make it barely influenced by gravity and so the additives typically remain relatively uniformly distributed in slurry through the solvent diffusion.
In the second stage, the pore-emptying evaporation, the remaining solvent residue inside the pores is fully evaporated, finally forming a completely dry electrode coating on In the beginning, the solution is dilute and different components suspend in slurry, resulting a sufficiently large and constant temperature enclosure. In ambient environment, the partial pressure of the solvent remains constant and contributes to constant evaporation rate [43]. Most of the solvent evaporation happens from the liquid surface of the solid-solvent slurry and the film shrinkage further processes until the active material consolidation finishes. Through a gravity force and due to a high evaporation rate, heavy active material particles approach each other and finally sediment to form a compact active material skeleton with tiny pores, in a process called film consolidation [41], where the thickness of consolidated film then remain fixed through the remaining solvent evaporation. In this case, the relatively light mass of conductive additives (e.g., carbon black) make it barely influenced by gravity and so the additives typically remain relatively uniformly distributed in slurry through the solvent diffusion.
In the second stage, the pore-emptying evaporation, the remaining solvent residue inside the pores is fully evaporated, finally forming a completely dry electrode coating on the current collector. However, because the evaporation rate significantly drops from previous constant evaporation rate to a much lower number, the removal of last 10 percent of solvent could take nearly the same time as the removal of first 90 percent of the solvent in the first stage [42,43]. The mechanism of the pore-emptying evaporation is more complex. In this case, the consolidation of active materials, the thickness of the coating, the dimensions, and the porous structure of coating are already fixed. The microstructure of the coating is only tweaked by the re-distribution of conductive additives and the binder polymer. Subject to a settled porous structure made by the backbone of active material, the evaporation triggered diffusion often induces slight (local) movement of the binder mixed with conductive additive and smaller active material particles and typically deposit them at narrow pathways between bulky (largest) active material particles. The long-distance migration of binder and conductive additives are no longer available due to a relatively small evaporation rate and greatly increased friction forces.

The Role of Binder
Despite being an inactive component and occupying a small portion of the electrodes, the polymer binder typically plays a vital role in the electrode fabrication and performance. In conventional electrode design, the selection of a binder and its solvent could determine the required fabrication procedures. Overall, polymer binders make use of physical or chemical bonding interactions to build a porous and mechanically stable structure for electrodes, which need to withstand various stresses involved in battery fabrication and charging and discharging cycling [44][45][46]. The binder also helps to support the uniform dispersion of active materials. It is crucial to understand the adhesion mechanisms of binders to optimize its performance in various cell form factors and designs. In general, the adhesion mechanisms could cover macroscopic theory and molecular level theory. In both theories, the pre-binding procedure are similar. In a slurry stirring (or a dry powder blending), the polymer binder particles experience dissolution, diffusion, and penetration contacting with active materials and conductive additives (typically carbon black, CB) [47]. The binding process, "hardening", is commonly triggered by diverse physical reaction mechanisms (such as the solvent evaporation after electrode casting or the cooling process after melting binder in dry electrode process) causing actual physical or inter-molecular adhesion between various particles and the current collector [47]. In a macroscopic theory, the mechanical interlocking, electrostatic force is accounted to strong adhesion force between two objects, which are ubiquitous in our daily life. The molecular level theory of binding mechanism could be more intricate and difficult to observe, but more convincing for scientists. For example, the chemical intermolecular interaction (such as the hydrogen bonding and Van der Waals forces), the interdiffusion of polymer chain, the chemical adsorption all occur on a molecular level [48,49]. Nevertheless, the chemical intramolecular interaction (such as the covalent bonding, the metallic bonding, and the ionic bonding) between the binder and the particles are exceptional. An integration of a whole electrode system only within the scope of the intramolecular level interactions may be unrealistic, but it, nonetheless, inspires researchers to search novel-electrode designs.
Today, polyvinylidene fluoride (PVDF) is widely used as a binder in LIB electrodes due to its strong binding via a Van der Waals force and large commercial production. However, to prepare cathode slurry, some organic solvents, typically featuring toxic and expensive N-Methyl-2-pyrrolidone (NMP) must be used. The use of the PVDF/NMP binder system suffers from several drawbacks, including the poor electrical/ionic conductivity of the formed electrodes, costly drying process, the evaporation of toxic solvent, degradable structure, poor mechanical properties (especially when swollen in a binder), and weak protection for active material under high temperature or high voltage [50]. In recent years, researchers extensively explored novel binder systems to overcome those issues. Table 1 presents selected representatives of these binders and the cycle performance of their-based cathodes. Mixing PVDF with other NMP-tolerable polymers was explored to improve the dispersion of slurry components [51][52][53][54]. Recently, Yoo et al. reported a novel organosilicon binder which does not suffer from binder self-aggregation, which can reduce electrode tortuosity and enhance the electrode integrity by forming a binder web with both covalent and hydrogen bonding [55]. At a reasonably high mass loading (~8 mg/cm 2 ; approximately half of commercial loadings), the prepared LCO electrodes can retain around 92% capacity after 100 cycles, which is insufficient for commercial applications, but promising, nonetheless. Aqueous solvent-based binders recently gained particularly strong interest, as using water in slurries can reduce the cost for electrode manufacturing and recycling, and additionally enable formation of a more homogenous slurry with active materials. Unfortunately, aqueous solvent-based binders typically also initiate new challenges, such as the dissolution of lithium and transitional metal ions from active materials in contact with water [69], corrosion of the Al current collector, cracking during electrode handling and drying (as aqueous binders may be more brittle), bubble formation, and delamination due to rapid solvent evaporation and significant binder shrinkage upon electrode drying [70][71][72]. Some possible solutions were suggested to address these challenges during aqueous processing, such as adding a buffer solution to mitigate a high pH at the slurry, using a carbon coated current collector, and controlling the solvent evaporation condition, among others [73]. The most common aqueous solvent-based binders are biomass polymer, which means their structures are more designable through chemical reaction to serve their multifunctional nature. Most recent studies on biomass-based binders such as cellulose nanofibers [48], lignin [61], guar gum [64], chitosan [67], carboxymethyl cellulose (CMC)/acrylic emulsion polymer [68] are listed in Table 1.
Binders not only provide a compact skeleton for active material but also can enhance electrochemical performance by mitigating some unpleasant side reaction in electrochemical reaction. In conversion cathodes (such as S, Se, metal sulfides, and fluorides), such approaches attracted increasing attention [12]. Yet, as previously mentioned, the successful utilization of conversion cathode in industrialization is hindered by multiple existing challenges, such as the shuttle effect by dissolution of lithium polysulfide, the blockage of conversion reactions by side products, among others [10]. Novel functional binders may be able to overcome those discussed drawbacks of conversion cathodes, as was demonstrated in the case of sodium alginate [66] and chitosan [67].

Dry Electrode Process
The PVDF/NMP binder system is not only expensive for high volume manufacturing but is also environmentally harmful. Although studies showed that water-based binder systems may potentially replace PVDF/NMP, they only became successfully commercialized for anode materials [47,49,74,75]. Their use still remains challenging for cathodes, because of the current collector corrosion, active particle deterioration, gas formation at elevated temperatures and voltages, and cracking of the electrodes (mostly thick electrodes) during drying process, to name a few [72,76]. In addition to the conventional slurry-based process, processing electrode through a solvent-free dry method may become very lucrative [77], because it can skip multiple costly procedures in battery manufacturing, such as the slow and expensive solvent evaporation [78][79][80]. As illustrated in Figure 6, the dry process enables a "powder-binder" electrode manufacturing to avoid the use of NMP or other solvents. In some designs, the dry process may begin with mixing and either dry spraying of active material, binder, and conductive additive onto a current collector [80], which allows binder particles and conductive particles to be decorating the surface of active material without causing significant agglomerates. The hot-roller pressing process is applied to melt the binder particles and allows the binder to wet the surface of active material and agglomerate with conductive additives [78,80]. The voids between active materials are partially filled with conductive framework of the melted binder mixed with conductive additives. Other investigated methodologies such as extrusion and 3D printing (with hot pressing) could also work for dry processing of electrodes [77,81,82], although agglomerates of hard cathode particles may induce excessive wear and tear to extrusion equipment and, thus, should be avoided at all costs. As well as fabrication of conventional LIBs, recent studies indicate that dry electrode process have great potential for the manufacturing of all-solid-state batteries (ASSBs) [83][84][85][86][87].
challenges, such as the shuttle effect by dissolution of lithium polysulfide, the blockage of conversion reactions by side products, among others [10]. Novel functional binders may be able to overcome those discussed drawbacks of conversion cathodes, as was demonstrated in the case of sodium alginate [66] and chitosan [67].

Dry Electrode Process
The PVDF/NMP binder system is not only expensive for high volume manufacturing but is also environmentally harmful. Although studies showed that water-based binder systems may potentially replace PVDF/NMP, they only became successfully commercialized for anode materials [47,49,74,75]. Their use still remains challenging for cathodes, because of the current collector corrosion, active particle deterioration, gas formation at elevated temperatures and voltages, and cracking of the electrodes (mostly thick electrodes) during drying process, to name a few [72,76]. In addition to the conventional slurry-based process, processing electrode through a solvent-free dry method may become very lucrative [77], because it can skip multiple costly procedures in battery manufacturing, such as the slow and expensive solvent evaporation [78][79][80]. As illustrated in Figure 6, the dry process enables a "powder-binder" electrode manufacturing to avoid the use of NMP or other solvents. In some designs, the dry process may begin with mixing and either dry spraying of active material, binder, and conductive additive onto a current collector [80], which allows binder particles and conductive particles to be decorating the surface of active material without causing significant agglomerates. The hot-roller pressing process is applied to melt the binder particles and allows the binder to wet the surface of active material and agglomerate with conductive additives [78,80]. The voids between active materials are partially filled with conductive framework of the melted binder mixed with conductive additives. Other investigated methodologies such as extrusion and 3D printing (with hot pressing) could also work for dry processing of electrodes [77,81,82], although agglomerates of hard cathode particles may induce excessive wear and tear to extrusion equipment and, thus, should be avoided at all costs. As well as fabrication of conventional LIBs, recent studies indicate that dry electrode process have great potential for the manufacturing of all-solid-state batteries (ASSBs) [83][84][85][86][87].

Key Parameters for Cathode Electrodes
Increasing the thickness of cathode electrodes is a promising way to enhance pack energy. However, today, the design of thick electrodes over 80 μm is still challenging. This is because thicker and denser electrodes may slow down ion transport, reduce the utiliza-

Key Parameters for Cathode Electrodes
Increasing the thickness of cathode electrodes is a promising way to enhance pack energy. However, today, the design of thick electrodes over 80 µm is still challenging. This is because thicker and denser electrodes may slow down ion transport, reduce the utilization of battery materials, limit electrolyte infiltration and lower electrochemical performance, particularly rate capability [88][89][90][91]. Dense structures with highly compressed electrodes, thick coatings, and high concentrations of small particles can limit ionic conductivity due to the high tortuosity of pore networks and longer ion diffusion pathways [92][93][94][95][96]. Using thick electrodes can also lead to larger Li+ concentration gradients, increasing resistance and reducing performance. To achieve optimal performance in energy/power densities, thick-ness, and loadings, it is desirable to consider some key parameters (Figure 7), including the defect and crystallinity, particle size and distribution, tortuosity, and electrode architecture. tion of battery materials, limit electrolyte infiltration and lower electrochemical performance, particularly rate capability [88][89][90][91]. Dense structures with highly compressed electrodes, thick coatings, and high concentrations of small particles can limit ionic conductivity due to the high tortuosity of pore networks and longer ion diffusion pathways [92][93][94][95][96]. Using thick electrodes can also lead to larger Li+ concentration gradients, increasing resistance and reducing performance. To achieve optimal performance in energy/power densities, thickness, and loadings, it is desirable to consider some key parameters ( Figure  7), including the defect and crystallinity, particle size and distribution, tortuosity, and electrode architecture. Figure 7. Key parameters for cathode electrodes, including defect and crystallinity, particle size and distribution, tortuosity and electrode architecture.

Defect and Crystalinity
Crystallographic defects, such as vacancies, interstitials, dislocations, grain boundaries, and others, widely exist in cathode crystals and could seriously affect the physical and (electro)chemistry performance of the cathodes [97]. Usually, these defects could result in accelerated structural and electrochemical degradations. Conventional cathodes such as NCM are composed of nanoparticles-aggregated secondary microspheres, which are known to suffer from fast capacity degradation and thermal instability due to the fast dissolution of grain boundary regions (the internal surfaces between the nanoparticles), further accelerated by concentrated stresses originating from different volume changes along different directions of these nanoparticles, the processes somewhat analogous to a stress-induced corrosion [98,99]. The cathode performance can be significantly improved by defect engineering, such as introducing twin boundaries into spinel cathodes [100] or pumping surface defects into bulk lattice [101].
Recent studies showed that single-crystal cathode materials (with grain sizes significantly larger than those in polycrystalline microspheres) have a noticeably better stability. Single-crystalline cathodes are able to work at higher voltages and offer a longer cycle life compared to polycrystalline cathodes [30,102,103]. When comparing polycrystalline and single-crystalline NCM cathodes, the differences in the degree of particle cracking, active surface areas, and Li diffusion rates become clearly visible [104]. Single-crystalline NCM Figure 7. Key parameters for cathode electrodes, including defect and crystallinity, particle size and distribution, tortuosity and electrode architecture.

Defect and Crystalinity
Crystallographic defects, such as vacancies, interstitials, dislocations, grain boundaries, and others, widely exist in cathode crystals and could seriously affect the physical and (electro)chemistry performance of the cathodes [97]. Usually, these defects could result in accelerated structural and electrochemical degradations. Conventional cathodes such as NCM are composed of nanoparticles-aggregated secondary microspheres, which are known to suffer from fast capacity degradation and thermal instability due to the fast dissolution of grain boundary regions (the internal surfaces between the nanoparticles), further accelerated by concentrated stresses originating from different volume changes along different directions of these nanoparticles, the processes somewhat analogous to a stress-induced corrosion [98,99]. The cathode performance can be significantly improved by defect engineering, such as introducing twin boundaries into spinel cathodes [100] or pumping surface defects into bulk lattice [101].
Recent studies showed that single-crystal cathode materials (with grain sizes significantly larger than those in polycrystalline microspheres) have a noticeably better stability. Single-crystalline cathodes are able to work at higher voltages and offer a longer cycle life compared to polycrystalline cathodes [30,102,103]. When comparing polycrystalline and single-crystalline NCM cathodes, the differences in the degree of particle cracking, active surface areas, and Li diffusion rates become clearly visible [104]. Single-crystalline NCM could avoid cracks after cycling, the observed cracking of secondary particles in polycrystalline NCM, which could result in liquid electrolyte infiltration in the active cathode materials, lowering the charge-transfer resistance and increasing the apparent diffusion coefficient [104]. However, single-crystal cathodes could still undergo capacity fading due to the spatially inhomogeneous Li concentrations in single-crystal cathodes during cycling and the nonuniform stress induced by the coexistence of two phases [105].

Particle Size and Distribution
To achieve optimal performance, it is essential to optimize the architectural parameters in electrode design, particularly particle size and distribution, which play a vital role in determining the solid-state diffusion length of Li ions, electrode surface area, and porosity [106][107][108][109][110][111][112]. In general, smaller particles have a shorter solid-state diffusion length, which can lead to a lower overpotential and faster C-rate operation. On the other hand, the larger surface area of smaller particles results in a higher proportion of passivation layers (e.g., CEI) which could cause capacity loss over time. In contrast, a larger size of particles results in a longer solid-state diffusion path, which is regarded as one of the primary limiting factors for LIBs based on thin electrodes [109,113], in spite of their higher tap density and higher volumetric energy.
Primary particles mainly affect the solid-state diffusion of Li ions while secondary particle size distribution (PSD) affect the attainable electrode density and the electrolyte diffusion through the electrodes [111,114]. The polydispersity index (PDI) and PSD are critical factors to electrode performance [96,110,115,116]. Avoiding too large and too small particles and attaining a medium-narrow PSD (with some smaller particles filling some of the voids between the larger ones, but not blocking such voids) enables better packing density and sufficiently fast improved electrolyte access, resulting in higher capacity and more consistent performance. In contrast, too broad PSD with excessively coarse particles could lower the attainable electrode smoothness and, thus, effective packing density, while a large portion of very small particles may block the pores and hinder electrolyte access, resulting in a lower capacity, poor rate, and less consistent performance. Primary particles are prone to heterogeneity in reaction kinetics due to variations in size and shape distribution, and secondary particles are more prone to cracking and detachment from the carbon/binder domain (CBD), leading to increased heterogeneity in the electrode microstructure [117][118][119][120][121][122][123][124]. Achieving a relatively uniform distribution of particles with controlled size is challenging and can result in heterogeneous reactions within the electrode. In practical electrode assemblies, active materials are embedded in the CBD, which can further complicate the charge distribution scenario [125,126]. CBD detachment and nonuniform voids distribution can induce a mismatch between ionic conductivity and electronic conductivity, compromising the effective utility of active particles. Heterogeneity at the CBD level can lead to a non-uniform percolated pore network and ionic transport resistance, which becomes worse after calendaring and can cause high current flux locally [119]. These issues can lead to reduced battery performance and durability.
A uniform size distribution and particle network and optimization of the spacecontrolled composition and porosity within the composite electrode (e.g., near a current collector vs. A separator) is expected to enhance electrochemical performance, reduce polarization, and result in better cycle stability [115,118,127,128]. Electrodes with vertically oriented particles or vertically oriented pores can improve rate performance without sacrificing gravimetric energy density [129][130][131]. The planar shape coupled with horizontal orientation and wide size distribution of active particles, on the other hand, not only may slow down Li-ion transport, but also result in a heterogeneous distribution of current and nonuniform lithiation between particles and along the through-thickness direction [132]. A mixture of small and large particles would be beneficial for attaining higher packing density and, thus, higher volumetric energy density in stationary energy storage. However, there are trade-offs to consider when optimizing particle size and distribution [133]. For instance, increasing either overall electrode density (by using particle size mixtures) or increasing areal capacity loading can generally degrade rate performance, but these could be counter-balanced by reducing electrode tortuosity, thus enhancing electrolyte wetting and minimizing effective through-electrode ion diffusion paths. Therefore, particle size, size distribution, particle shape, particle orientation within the electrode as well as the electrode pore size, pore orientation, and pore tortuosity must all be carefully considered during electrode design to ensure optimal performance.

Tortuosity
Tortuosity refers to the extent to which the path of the pore is convoluted and twisted, thus impeding the diffusion of ions within the electrode. Tortuosity is another important factor that impacts the ionic conductivity of an electrode, because the effective ionic conductivity (D eff ) is usually influenced by porosity (ε), tortuosity (τ), and intrinsic ion conductivity (D) of the electrolyte, that is D eff = D(ε/τ) [134]. A higher porosity can enhance the mass transport but leads to lower volumetric capacity and, additionally, could lead to a lower mechanical strength, electrode cracking, or delamination. In the electrode design, minimizing tortuosity is essential to improve ion transport in thick electrodes because effective ionic conductivity is inversely proportional to tortuosity [90,126,135,136]. The transmission line model analysis (TLM) is used to measure ionic resistance and determine McMullin number and tortuosity through electrochemical measurements [135]. Decreasing these values in dense and thick electrodes can decrease ion diffusion time, reduce overall impedance Z, and improve rate performance under high current densities. Low-tortuosity pore design is an effective approach to enhance ion transfer kinetics in thick electrodes. Achieving both fast ion and electron transport pathways is essential for optimal thick electrode design since thick electrodes with high areal capacity may cause mass transport limitations at elevated C-rates. However, there is a trade-off between porosity and tortuosity, and achieving a balance between the two is crucial for optimal battery performance.

Electrode Architecture
High-performance thick electrodes require the design of the overall architecture of cathode electrodes by considering the active material crystallinity, particle size, shape and surface area, the distribution, size, and shape of the conductive additives and the binder, and the resulting electrode porosity and tortuosity. For example, the difference in the porosity distribution in top and bottom layers of the electrode could affect the key electrode features such as physical, mechanical, chemical, and electrochemical properties. Pore engineering is a promising approach to improve the performance of battery electrodes by designing a microstructure with vertically oriented pore arrays [94,137]. Techniques such as high-temperature sintering, laser patterning, and density gradient electrodes were explored to create low-tortuosity pore structures that balance porosity and tortuosity [138][139][140][141][142][143][144]. However, manufacturing of thick or even ultra-thick electrodes (>100 um) with (ultra)high mass loading is still difficult [91,145,146]. There are challenges to overcome, such as low volumetric capacity and lack of scalability in some of the explored approaches, and balancing ion transport and energy density, which can be affected by the potential decrease in energy density associated with high porosity. Therefore, careful control of porosity and tortuosity variation from the surface closer to separator to the contact closer to the current collector through electrode design and fabrication methods is crucial to facilitate ion transport, electrolyte infiltration, and ultimately improve battery performance [147,148]. In addition, heterogeneity can cause localized overcharging and Li plating, leading to rapid deterioration of the battery's microstructure and long-term performance. To address these issues, various techniques such as designing pore structures, using binder-free electrodes, attaining patterned electrodes, and 3D conductive skeletons could be further explored to help reduce heterogeneity, porosity, and tortuosity, thus improving battery performance and durability.

Conclusions
The development of next-generation LIBs calls for more considerations on cathode materials and electrode-processing technologies. In this review, we discussed the advances and challenges in current cathode materials, binders, and processing technologies. We compared conventional intercalation cathodes and conversion cathodes to demonstrate their current issues and future potentials. To meet the practical demands, it is derisible to design thicker cathode electrodes with high packing energy and fast charging/discharging rates. We discussed several key performance parameters from crystal/particle to electrode levels.
From a practical fabrication prospective, however, one should additionally consider the mechanics of cathode materials and fully formed electrodes with mechanical tests conducted, such as tensile, compression, bending, and nail-penetration (or blunt-rod) tests, which will be important for the evaluation of both high-yield mass production and the battery safety. Additionally, commutating methods such as finite element analysis (FEA) calculations and simulations, and/or machine learning/artificial intelligence (AI) should be also taken into consideration [149][150][151]. Further technology developments on both advanced cathode materials and cathode fabrication processes will be critical to promote the rapid development of battery industry.