Lithium Phosphonate Functionalized Polymer Coating for High‐Energy Li[Ni0.8Co0.1Mn0.1]O2 with Superior Performance at Ambient and Elevated Temperatures

High‐energy Ni‐rich lithium transition metal oxides such as Li[Ni0.8Co0.1Mn0.1]O2 (NCM811) are appealing positive electrode materials for next‐generation lithium batteries. However, the high sensitivity toward moist air during storage and the high reactivity with common organic electrolytes, especially at elevated temperatures, are hindering their commercial use. Herein, an effective strategy is reported to overcome these issues by coating the NCM811 particles with a lithium phosphonate functionalized poly(aryl ether sulfone). The application of this coating allows for a substantial reduction of lithium‐based surface impurities (e.g., LiOH, Li2CO3) and, generally, the suppression of detrimental side reactions upon both storage and cycling. As a result, the coated NCM811‐based cathodes reveal superior Coulombic efficiency and cycling stability at ambient and, particularly, at elevated temperatures up to 60 °C (a temperature at which the non‐coated NCM811 electrodes rapidly fail) owing to the formation of a stable cathode electrolyte interphase with enhanced Li+ transport kinetics and the well‐retained layered crystal structure. These results render the herein presented coating strategy generally applicable for high‐performance lithium battery cathodes.


Introduction
Ni-rich LiNi 1−x−y Co x Mn y O 2 (NCM (1−x−y)xy , with 1−x−y ≥ 0.6) cathode materials are considered the most practical candidates for high-energy lithium-ion and lithium-metal batteries owing to their high capacities and good rate capability. [1] However, there are several challenges that remain to be overcome for a widespread use in commercial battery cells: [2] 1) the difficulty in synthesizing stoichiometric NCM due to the resulting from the layered to spinel and rock-salt structural transition), [9] and thus eventually have a detrimental impact on the cycling stability and capacity retention.
To address these challenges, various approaches have been adopted, including advanced material and particle designs such as the realization of core-shell structures [10] and transition metal concentration gradients, [11] the incorporation of functional dopants, [12] and the application of inorganic or organic surface coatings, including metal oxides, [13] metal fluorides, [14] metal phosphates, [3,15] and polymers. [15b,16] The latter approach appears particularly straightforward, as such protective coatings can physically prevent the contact with the ambient atmosphere, while also stabilizing the interface with the electrolyte in the battery cell -especially with regard to the essentially unlimited choice of suitable coating compositions, allowing for well-tailored properties. Nonetheless, the use of inorganic coatings commonly requires additional calcination steps at several hundred degrees Celsius, which leads to significantly higher cost and complicates recycling. In addition, lithium-free coating layers (inorganic or organic) hinder Li + transport across the interface with the electrolyte, particularly when surpassing a certain thickness, which negatively affects the cell performance at elevated current densities. [17] Herein, we report lithium phosphonate functionalized poly(arylene ether sulfone) (PP10-Li; see Figure 1a) as suitable coating layer for high-energy NCM 811 (ONCM 811 ) to overcome the aforementioned challenges. The beneficial interaction with the NCM 811 particles allows for a highly homogeneous coating at room temperature, while the lithium phosphonate function facilitates charge transfer and the poly(arylene)-type backbone enhances the thermal stability and thus safety of the cathode material. Advantageously, the application of the PP10-Li coating moreover leads to a substantially reduced amount of detrimental surface impurities such as LiOH and Li 2 CO 3 and, instead, an enrichment in LiF at the active material surface. As a result, ONCM 811 shows superior rate capability and enhanced cycling stability, especially at elevated temperatures up to 60 °C compared to the non-coated NCM 811 .

Physicochemical Characterization
In a first step, we investigated the morphology of the PP10-Li coating. Figure 1b shows a scanning electron microscopy (SEM) image of the NCM 811 active material, containing secondary particles with a diameter of either ≈2 µm or ≈15 µm to increase the electrode packing density and thus the energy density eventually. This particle morphology is well preserved after applying the PP10-Li coating, as depicted in Figure 1c. The only difference at this magnification appears to be the observation of some darker spots, which are assigned to the electronically insulating nature of PP10-Li, uniformly covering the NCM 811 particle surface as apparent at higher magnification ( Figure 1d). Transmission electron microscopy (TEM) images of NCM 811 confirm the secondary particle structure (Figure 1e) and highresolution TEM analysis of the particles (Figure 1f) reveals clear lattice fringes assigned to the (003) plane of the layered crystal structure, indicating that the material is highly crystalline (see also the fast Fourier transform (FFT) pattern in the inset), which is in good agreement with results obtained by X-ray diffraction (XRD), presented in Figure S1 (Supporting Information). In fact, in both cases -NCM 811 and ONCM 811 -the same reflections are observed, which can be well assigned to the layered α-NaFeO 2 structure (R-3m) and there is no indication of any phase impurity for neither of the two materials. The TEM analysis of the ONCM 811 reveals that the PP10-Li coating layer has a rather homogeneous thickness of 7-12 nm and uniformly covers the particle surface (Figure 1g), which is well in line with the SEM data ( Figure 1d). To further confirm the uniformity of the coating, we conducted energy-dispersive X-ray (EDX) spectroscopy mappings of a larger secondary ONCM 811 particle ( Figure 1h) and an extendedly larger region of pristine ONCM 811 -based electrodes ( Figure S2, Supporting Information); and, indeed, the mappings of the different elements ( Supporting Information), which are present only in PP10-Lireveals a very homogeneous distribution at the particle surface.
In a next step, we performed a comparative investigation of pristine NCM 811 and ONCM 811 electrodes via X-ray photoelectron spectroscopy (XPS) to study the impact of applying the PP10-Li coating on the surface chemistry. A summary of the findings for the oxygen-containing species is provided in Figure 2a (based on the fitting of the O 1s spectra shown in Figure S3, Supporting Information). The intensity of the M-O contribution (in purple) is noticeably lower in the case of ONCM 811 compared to NCM 811 as a result of the PP10-Li coating. The presence of the coating is also reflected in the contributions from S = O (in dark purple) and O = POLi (in light purple) as well as the pronounced contribution of -COC-(in orange). The presence of the latter contribution for the NCM 811 spectrum, but to a substantially lower extent, is assigned to oxygen species at the surface of the conductive carbon. The minor contribution of R-O-(C = O)-O-as additional surface species, observed for both pristine NCM 811 and ONCM 811 electrodes, originates from the reaction of the electrode material with the atmosphere. [18] Although all the material handling and processing as well as the electrode preparation were conducted in the dry room, the formation of surface contaminations such as LiOH and Li 2 CO 3 owing to the reaction with the ambient atmosphere (especially, H 2 O and CO 2 ) could not be fully suppressed. The contribution of LiOH is very low in the case of NCM 811 (≈0.22 at%) owing to the careful handling in the glove box and dry room, while it is essentially undetectable for ONCM 811 . Similarly, the contribution of Li 2 CO 3 is much more pronounced for NCM 811 (4.92 at%) compared to ONCM 811 (1.17 at%). The reduced contribution of LiOH and Li 2 CO 3 observed for ONCM 811 highlights the substantially decreased reactivity with the ambient atmosphere, as a result of the homogeneous PP10-Li coating which effectively prevents the direct exposure to H 2 O and CO 2 . [19] This finding is of particular relevance with regard to the commercial use of Ni-rich NCM, since the formation of LiOH and Li 2 CO 3 does not only cause lower specific capacities, but moreover leads to a higher interfacial resistance, pronounced side reactions with the electrolyte, and gas formation during cycling. [20] www.afm-journal.de www.advancedsciencenews.com The F 1s core-level spectra recorded for the NCM 811 and ONCM 811 electrodes, taken at different sputtering depths, are presented in Figure 2b,c, respectively. Generally, both show two distinct peaks: one with a higher intensity at 688.2 eV, which is assigned to -CF 2 -and originates from poly(vinylidene difluoride) (PVdF) that has been used as binder (while in the case of ONCM 811 presumably also including some contribution from the PP10-Li coating), [21] and one with a lower www.afm-journal.de www.advancedsciencenews.com intensity at 658.5 eV, which is ascribed to the presence of LiF. [22] The appearance of LiF in the pristine electrodes has been reported to be a consequence of the minor dehydrofluorination reaction of PVdF, catalyzed by the alkaline species (such as LiOH and Li 2 CO 3 ) present in the slurry during the electrode coating process, resulting in the formation of HF that reacts at the NCM surface and eventually forms LiF. [21a,23] Prior to the sputtering, the relative contribution is similar for NCM 811 and ONCM 811 (i.e., 0.26 and 0.22 at%, respectively -see also Figure 2d,e). However, the relative contribution increases during the Ar + sputtering treatment, implying that the LiF formation occurs largely at the NCM surface, which is in line with the earlier reported mechanism. Notably, the relative contribution of LiF after sputtering for 12 min (equivalent to an estimated sputtering depth of 9.6 nm) is higher for ONCM 811 (3.53 at% vs 2.60 at% for NCM 811 ). In combination with the lower concentration of LiOH and Li 2 CO 3 that are (largely) responsible for the formation of LiF, [21a,23] this finding indicates that PP10-Li is strongly bound to the NCM 811 surface and presumably partially reacting. Such enrichment in LiF (in addition to the PP10-Li coating layer) is expected to have a beneficial effect on the Li + transport kinetics, the suppression of transition metal dissolution, and the stability of the interface with the electrolyte. [24]

Reactivity with Moist Air
Facile storage and handling of cathode active materials (and materials in general) are critical for the battery industry. Thus, we comparatively studied also the evolution of NCM 811 and ONCM 811 upon storage under a controlled ambient atmosphere. For this purpose, we exposed the two materials to a controlled relative humidity of 65% for six days. The comparison of the resulting gain in mass reveals that NCM 811 is more sensitive to such storage conditions, as it is significantly greater compared to ONCM 811 , i.e., 0.55% versus 0.32% (Figure 3a), further supporting the beneficial impact of the PP10-Li coating as a physical barrier toward the environment. Subsequently, the gain in mass was investigated in more detail, revealing that the reversible water absorption was 0.15% and 0.16% for NCM 811 and ONCM 811 respectively, as determined by thermogravimetric vapor sorption analysis ( Figure 3b). The slightly higher reversible water absorption of ONCM 811 might be explained by the hygroscopic nature of the PP10-Li coating, absorbing more water than the NCM 811 particles -notably without reacting with the ambient moisture, as it is subsequently released again. In fact, the relatively higher reversible water absorption in combination with the relatively lower gain in mass and the lower contribution of common reaction products such as LiOH and Li 2 CO 3 (50% of the overall gained mass, Figure 3c) indicates that the PP10-Li coating suppresses the reactivity with H 2 O and CO 2 . This is also reflected in the relatively lower increase of the pH value when dispersing the two materials in deionized water ( Figure 3d). For NCM 811 , the pH value jumps to 10.73 after about 3 min and subsequently increases slowly to 10.82 after 60 min, which is lower than for ONCM 811 with 10.64 and 10.81, respectively. This is in agreement with the lower LiOH and Li 2 CO 3 content and the mitigated surface reactivity in the case of ONCM 811 , illustrating a facilitated handling and storage of the PP10-Li-coated material.

Electrochemical Characterization
As any surface coating naturally impacts the de-/lithiation kinetics, we performed a comparative analysis of NCM 811 and ONCM 811 electrodes via cyclic voltammetry ( Figure S4, Supporting Information). While both cells generally show high reversibility and the characteristic redox couples, related to the three common phase transitions occurring upon de-/lithiation (see the corresponding discussion in the Supporting Information), the determination of the apparent lithium-ion diffusion coefficient reveals slightly higher values for ONCM 811 (i.e., 9.7 × 10 −8 and 3.9 × 10 −8 cm 2 s −1 for the delithiation and lithiation, respectively, compared to 8.3 × 10 −8 and 3.4 × 10 −8 cm 2 s −1 for NCM 811 ), highlighting the positive impact of the PP10-Li coating layer on the de-/lithiation kinetics.
In order to investigate how these findings are reflected in the dis-/charge behavior under galvanostatic conditions, half-cells comprising NCM 811 and ONCM 811 -based positive electrodes were subjected to varying C rates (Figure 4). Figure 4a,b provides a comparison of representative dis-/charge profiles at varying C rates from 0.1C to 10C. Generally, the reversible specific capacities are slightly lower for ONCM 811 compared to NCM 811 (i.e., 201 vs 203 mAh g −1 (0.1C); 195 vs 197 mAh g −1 (0.2C); 186 vs 188 mAh g −1 (0.5C); 178 vs 180 mAh g −1 (1C); 170 vs 172 mAh g −1 (2C); 154 vs 156 mAh g −1 (5C); 147 vs 148 mAh g −1 (7C); and 137 vs 137 mAh g −1 (10C)), while the difference is diminishing for higher dis-/charge rates. In fact, the slightly lower capacity at low and medium C rates is assigned to the mass contribution of the PP10-Li coating, which has been included in the calculation of the specific capacity. Accordingly, when normalizing the specific capacity values with regard to the reversible capacity obtained at 0.1C, the values are essentially the same for the low and medium C rates and higher for ONCM 811 at 10C (Figure 4c), revealing a better rate capability owing to the PP10-Li coating. As a matter of fact, the initial overpotential for the delithiation at elevated dis-/charge rates (2C and higher) is significantly higher for NCM 811 compared to ONCM 811 (Figure 4a,b), which underlines the better kinetics in the case of ONCM 811 and is in agreement with the results of the cyclic voltammetry experiments. To investigate the impact of the PP10-Li coating on the long-term cycling performance, both cells were subsequently subjected to 500 cycles at 1C (Figure 4d). Initially, the capacity for the NCM 811 cell is slightly higher, as expected from the rate capability test, but it rapidly fades after about 460 cycles, resulting in a capacity retention of only 77.6%. This fading is accompanied by substantial decrease in Coulombic efficiency (CE) down to 99.09%, lowering the average CE to 99.66% upon 500 cycles. Accordingly, the rapid capacity fading toward the end of the www.afm-journal.de www.advancedsciencenews.com cycling test is attributed to detrimental side reactions at the electrode|electrolyte interface (including newly formed interfaces due to microcracks evolving in the NCM 811 particles) and their detrimental impact on the NCM 811 active material and its crystal structure. [6,25] Differently, the cycling performance of ONCM 811 is very stable for 500 cycles with a capacity retention of 81.1% and no pronounced fading after 500 cycles at 1C. This superior performance is also reflected in the very high average CE of 99.96%, indicating that the application of the PP10-Li coating layer leads to suppressed side reactions with the electrolyte and preserves a stable electrode|electrolyte interface. Subsequently, we studied also the long-term constant current cycling at elevated C rates of 3C ( Figure 4e) and 5C (Figure 4f). Also under these conditions, the ONCM 811comprising cells show a substantially enhanced cycling stability (and essentially the same capacity). The cut-off criterion of 80% capacity retention is reached after 459 cycles (3C) and 377 cycles (5C), compared to 387 cycles (3C) and 358 cycles (5C) for the NCM 811 -based cells.
Subsequently, we studied the cycling performance at elevated temperatures of 40 and 60 °C (Figure 5) to stress the beneficial impact of the PP10-Li coating. At 40 °C (Figure 5a), both NCM 811 and ONCM 811 provide higher capacities than at 20 °C owing to the increased Li + diffusion and transport kinetics. However, during the constant current cycling at 3C, the NCM 811 -based cell shows an increased capacity fading after about 130 cycles and the 80% capacity retention criterion is reached after only 176 cycles, whereas the ONCM 811 -based cell hits this limit after 221 cycles. The superior cycling stability is even more evident when increasing the ambient temperature to 60 °C (Figure 5b), i.e., a temperature that marks the onset of electrolyte decomposition and triggers substantial side reactions at the electrode|electrolyte interface. [26] While the rate capability is generally very high for both NCM 811 and ONCM 811 and better than at 20 °C with, e.g., more than 180 mAh g −1 at 10C owing to the greater kinetics, the CE is much higher for ONCM 811 and the cut-off criterion of 80% capacity retention is reached after 180 cycles at 1C (following to the rate capability test), while the NCM 811 -based cell fades rapidly and reaches this criterion after only 15 cycles. These results further highlight the highly advantageous effect of the PP10-Li coating on the stabilization of the interface with the electrolyte, successfully suppressing severe side reactions and active material decay. www.afm-journal.de www.advancedsciencenews.com

Ex Situ Investigation of the (O)NCM 811 Surface
To unveil how the application of the PP10-Li coating enhances the electrochemical performance, the electrodes cycled according to the testing protocol shown in Figure 4d were eventually analyzed via SEM/EDX. The SEM images of pristine ONCM 811 and NCM 811 electrodes are presented in Figure S5 (Supporting Information) and the ex situ SEM data are shown in Figure S6 (Supporting Information). Alike the SEM and TEM data discussed earlier (Figure 1), the surface of the pristine ONCM 811 particles appears smoother and the edges of the primary nanograins are less apparent (Figure S5a,b, Supporting Information) compared to the SEM images taken for NCM 811 (Figure S5c,d, Supporting Information) due to the presence of the PP10-Li coating. After cycling ( Figure S6, Supporting Information), however, the surface of the NCM 811 electrodes appears much smoother than prior to cycling, indicating the formation of a rather thick cathode electrolyte interphase (CEI), while the surface morphology of the ONCM 811 electrodes is largely retained. In fact, the ex situ SEM/ EDX analysis of the cycled ONCM 811 electrodes ( Figure S7, Supporting Information), in particular the homogenous elemental distribution of S, for which PP10-Li is the only source (while the other elements are found also for NCM 811 , Figure S8, Supporting Information), confirms that the coating layer is well preserved, even after such extended cycling for more than 500 cycles.
For a closer look at the particle surface, we performed also ex situ TEM measurements on these cycled electrodes (Figure 6).
While it is generally difficult to distinguish between the CEI Figure 5. Electrochemical characterization of half-cells comprising NCM 811 and ONCM 811 as the active material for the positive electrode a) at 40°C, applying a constant dis-/charge rate of 3C after three formation cycles at 0.1C, with an indication of the cycle number at which a capacity retention of 80% is reached, and b) at 60°C, applying varying dis-/charge rates, followed by constant current cycling at 1C with an indication after how many cycles a capacity retention of 80% is reached. layer and the PP10-Li coating due to the amorphous nature of both phases, the amorphous surface layer is clearly thinner in the case of ONCM 811 (Figure 6a) compared to NCM 811 (Figure 6c) -despite the contribution of the PP10-Li coating in the former case, indicating a pronounced electrolyte decomposition and CEI formation for NCM 811 . Moreover, when analyzing the crystal structure of the ONCM 811 particles in the bulk and the near-surface region (Figure 6b), the resulting FFT patterns confirm that the layered (R-3m) structure is retained. [9c,27] Differently, for the cycled NCM 811 particles (Figure 6d), the FFT patterns reveal the transition from the layered (R-3m) structure to the rock-salt (Fm-3m) structure [28] in the near-surface region (red frame) and a mix of the two crystalline phases in the bulk (orange frame). It is noteworthy that the transition from the layered to the electrochemically inactive rock-salt structure [9c] appears complete for several tens of nanometers from the particle surface toward the particle core, providing an explanation for the pronounced capacity fading in the case of NCM 811 .
Complementarily, we conducted also an ex situ XPS depthprofiling analysis of the cycled NCM 811 and ONCM 811 electrodes to gain additional insights into the thickness and composition of the CEI layer (Figure 7). The O 1s and F 1s spectra for NCM 811 are depicted in Figure 7a,b and the O 1s and F 1s spectra for ONCM 811 are presented in Figure 7c,d (including in both cases the XPS data for the pristine electrodes at the top for comparison). Generally, for both NCM 811 and ONCM 811 , In all cases, the spectra shown at the top have been obtained for pristine electrodes and serve as reference. The spectra obtained for cycled electrodes prior to any sputtering are shown in the row below and in the subsequent rows the sputtering time has been stepwise increased (from top to bottom, with an indication of the estimated sputtering depth). The color coding for the fitted contributions is shown below the XPS spectra. It should be noted that the scale for photoelectron yield at the y axis has been kept constant for all O 1s (6400) and F 1s spectra (9000). The O 1s spectra of the pristine NCM 811 and ONCM 811 electrodes are the same as those presented in Figure S3 (Supporting Information), but shown here once again for the sake of convenience when comparing the XPS spectra before and after cycling. www.advancedsciencenews.com the spectra recorded prior to any sputtering reveal contributions from organic (e.g., RCH 2 OCOOLi and -COC-) and inorganic (e.g., LiOH, Li 2 CO 3 , LiF, and Li x PF z /Li x PO y F z ) electrolyte decomposition products. In line with the XPS data obtained for the pristine electrodes, however, the ONCM 811 electrodes show substantially less Li 2 CO 3 and essentially no LiOH. Furthermore, the ONCM 811 electrodes show generally less decomposition products and a greater intensity for the M-O contribution, indicating that the CEI is thinner overall. Besides, contributions from -S = O and -OP = O are observed for cycled ONCM 811 electrodes, further corroborating that the PP10-Li layer is preserved on the particle surface. Upon sputtering the M-O contribution increases in both cases, confirming that the CEI layer is increasingly removed. Simultaneously, the relative contribution of the organic electrolyte decomposition products is gradually decreasing, though more pronounced for ONCM 811 , while the contribution of LiF is increasing in both cases, which is more pronounced for NCM 811 , however. This observation in combination with the initially higher LiF concentration for the pristine ONCM 811 electrodes further supports the conclusion that there is more electrolyte decomposition in the case of NCM 811 . Additionally, the contribution of Li 2 CO 3 is increasing particularly for NCM 811 , and the contribution of the PP10-Li-related peaks (i.e., -S = O and -OP = O) is increasing for ONCM 811 . These findings globally indicate that the outer CEI layer is more organic in nature, while the inner layer is rather composed of inorganic products, which is in line with earlier CEI studies. [29] Moreover, they show that the surface chemistry of the pristine electrodes is somehow retained, which includes the preservation of the PP10-Li coating layer at the intimate particle surface.

Evaluation of the Thermal Stability
Finally, we studied also the impact of the PP10-Li coating on the thermal behavior of the delithiated NCM 811 and ONCM 811 electrodes (charged to 4.3 V) in presence of the electrolyte by differential scanning calorimetry (DSC, Figure 8). Generally, the onset temperature for the heat evolution is significantly lower in both cases compared to the base electrolyte only (in grey), revealing the characteristic decomposition profile for DEC, [30] which underlines the greater reactivity of the latter in contact with delithiated NCM. [31] Besides, it is fairly similar for both active materials, while the overall heat evolution is substantially lower for ONCM 811 (1139.0 J g −1 ) compared to NCM 811 (1321.7 J g −1 ). In fact, the DSC curve recorded for PP10-Li only (in red) does not show any exothermic features up to 300 °C, only a glass transition and an endothermic peak at ≈230 and ≈274 °C, respectively, further highlighting the beneficial impact of the coating -not only for the electrochemical performance, but moreover regarding safety.

Conclusions
The PP10-Li coating addresses the most challenging issues of Nirich NCM, i.e., the poor cycling stability and the low thermal stability, by suppressing the undesired surface reactions with moist air (i.e., H 2 O and CO 2 ) upon storage and processing as well as with the electrolyte upon cycling, especially at elevated temperatures. As a result, PP10-Li-coated NCM 811 (ONCM 811 ) provides a substantially enhanced electrochemical performance in terms of rate capability, capacity retention, and Coulombic efficiency (99.96% vs 99.66%) owing to the favorable de-/lithiation kinetics as well as the stabilized electrode|electrolyte interface and interphase. Most remarkable is certainly the improvement in capacity retention at 60 °C from 15 cycles (NCM 811 ) to 180 cycles (ONCM 811 ). In combination with the substantially reduced heat evolution, this finding particularly highlights the beneficial impact of the coating also concerning safety. We may anticipate that the use of well-designed lithiated polymer coatings, as reported herein for the first time, is an effective strategy for today's and tomorrow's cathode materials.
Synthesis of PP10-Li: The complete synthesis of PP10-Li is shown in Figure S9 (Supporting Information) and described in the following in more detail.
Synthesis of HFB End-Capped PES10: HFB end-capped PES10 was synthesized by modifying a one-pot two-reaction procedure reported elsewhere. [32] In a typical protocol ( Figure S9, Supporting Information)), a 100-mL three-neck round-bottom flask, equipped with a mechanical stirrer, a condenser, an argon inlet/outlet, and a Dean-Stark trap, was charged with DFDPS (4.000 g, 15.732 mmol) and BP (3.052 g, 16.388 mmol). DMAc (28 mL) was added to obtain a solid concentration of 25% (wt/v). When all solid components were completely dissolved, K 2 CO 3 (6.79 g, 0.0492 mol) Figure 8. DSC profiles of (partially) delithiated NCM 811 (in yellow) and ONCM 811 (in green) electrodes in presence of the electrolyte. The delithiation cut-off was set to 4.3 V. For comparison, also the DSC profiles obtained for the pure electrolyte (in gray) and pure PP10-Li (in red) are shown. www.advancedsciencenews.com and 14 mL of toluene, serving as azeotroping agent, were added. The DMAc to toluene ratio was 2:1 v/v. The reaction bath was heated to 150 °C and kept at this temperature for 4 h to dehydrate the system. Subsequently, the bath temperature was slowly raised to 160 °C to remove the toluene. Afterward, the temperature of the reaction bath was decreased to 120 °C and the polymerization was allowed to proceed for 24 h. In a next step, the reaction temperature was set to 70 °C and 2.44 g (13.11 mmol) of HFB were added, while the argon purge was stopped at this point owing to the low boiling point of HFB. The reaction was allowed to proceed for 12 h. Then, the reaction mixture was precipitated into 2 L of 1 m HCl aqueous solution with magnetic agitation for 8 h, filtered, and rinsed with distilled water until reaching a neutral pH value. The resulting white powder was dried under vacuum at 80 °C for 24 h to obtain the final product (M n = 10 600 g mol −1 , M w = 21 500 g mol −1 , polydispersity index (M w /M n ) = 2.0). 1 H and 19 F nuclear magnetic resonance (NMR) spectra of HFB end-capped PES10 are shown in Figure S10  The bromination of PES10 (yielding BPES10) was conducted at room temperature using elemental bromine as the brominating agent in the presence of acetic acid ( Figure S9, Supporting Information)). In a typical protocol, 5.000 g of PES10 (12.10 mmol of the biphenyl unit) was loaded into a 250 mL three-neck round-bottom flask, equipped with a mechanical stirrer, a condenser, an argon inlet, and a funnel. First, 90 mL of DCM, distilled from CaH 2 , and 9 mL of acetic acid (10% v/v to DCM) were added. After the polymer was completely dissolved, 9.36 mL of bromine (Br 2 ; 0.182 mol) was added dropwise to the reaction mixture, followed by strong stirring, and the reaction was allowed to proceed at room temperature for 16 h. Subsequently, the reaction mixture was precipitated into 1000 mL of methanol, washed three times with methanol to remove excessive bromine, and kept under continuous agitation for 16 h. Afterward, the polymer was filtered and rinsed again with methanol until any residual bromine was completely removed. The final product, i.e., BPES10, was dried under vacuum at 80 °C for 24 h. 1 H and 19 F NMR spectra of BPES10 are shown in Figure S11  Synthesis of Lithium Phosphonate Functionalized PES10 (PP10-Li): The phosphonate functions were tethered to the brominated BPES10 using a palladium-catalyzed coupling reaction ( Figure S9, Supporting Information)). In a typical procedure, BPES10 (2.00 g, 7.19 mmol Br), Pd(OAc) 2 (80.7 mg, 0.36 mmol), TPP (0.283 g, 1.08 mmol), and anhydrous DMAc (40 mL) were loaded into a 100-mL two-neck roundbottom flask equipped with an argon inlet/outlet and a magnetic bar. The reaction mixture was heated to 95 °C under strong stirring to obtain a homogeneous solution. Subsequently, DEP (4.6 mL, 36.0 mmol) and DCHMA (4.6 mL, 21.6 mmol) were added, and the reaction was allowed to proceed at this temperature for 48 h. The reaction mixture was precipitated into ethanol, washed several times with ethanol, and filtered to obtain a white solid, which was dried under vacuum at 80 °C for 24 h, eventually yielding the ethyl phosphonate form of PES10 (i.e., PP10-Et, 95% yield). PP10-Et (1.00 g, 2.98 mmol of the -PO(OC 2 H 5 ) 2 function) was boiled in a 36.5% HCl aqueous solution for 4 h and washed several times with distilled water to remove any residual HCl. The resulting product, i.e., PES10 functionalized with phosphonic acid (i.e., PP10-H) was dispersed in a 0.5 m aqueous solution of LiOH, washed several times with distilled water, filtered, and dried under vacuum at 80 °C for 24 h to finally obtain the targeted PP10-Li (93% yield). 1 H and 19 F NMR spectra of PP10-Et as intermediate species are shown in Figure S12 (Supporting Information) and the 1 H NMR spectrum of the acidic form PES10 is presented in Figure S13 (Supporting Information). The comparison of the two spectra reveals the absence of peak 6 and 7, observed in Figure S12 (Supporting Information) and assigned to the ethyl groups, in Figure S13 (Supporting Information), indicating the complete conversion to the acidic form. The ion-exchange capacity (IEC) of PP10-Li as the final product was determined via 1 H NMR spectroscopy and the results are provided in Table S1 (Supporting Information).  10.0 mL of NMP were added and the mixture was heated to 120 °C. Subsequently, 4.0 mL of deionized water (DIW) were added dropwise until PP10-Li was completely dissolved. Afterward, the water was removed via azeotropic distillation. For the drying step, 10 mL of toluene were added and the solution was placed in a Dean Stark system. The mixture was heated up to 160 °C and kept at this temperature under reflux overnight. Finally, any residual water and the toluene were removed to yield PP10-Li dissolved in NMP only.
1 g of NCM 811 (a pre-commercial material that was generally stored under argon in a glass vial and the glass vial was stored in a dry room with the dew point being always below −60 °C at 20 °C ambient temperature) was added into 2 mL of the PP10-Li in NMP solution (corresponding to 2 wt% of PP10-Li with regard to the mass of NCM 811 ). The resulting dispersion was stirred slowly and heated up to 80 °C in order to slowly evaporate the NMP, finally yielding the PP10-Li-coated NCM 811 , herein referred to as ONCM 811 .
Physicochemical Characterization: 1 H and 19 F NMR spectroscopy was performed using a Bruker Ascend 400 apparatus. Size-exclusion chromatography (SEC) with a multi angle light scattering detector (MALS) was conducted at room temperature on a Waters 590 GPC equipped with a Waters 410 differential refractometer and a Waters 745 Data Module. A 1 m solution of NaNO 3 in dimethylformamide (DMF) was used as solvent, applying a flow rate of 1 mL min −1 . The 1 wt% solution of PES10 was filtered using a 0.45 mm Millipore polytetrafluoroethylene filter and three Ultrastyragel columns of 5 × 10 2 , 10 3 , and 10 4 Å. The calibration was performed using polystyrene standards.
The crystal structure of NCM 811 and ONCM 811 was studied via X-ray diffraction (XRD) -on both powders and electrodes -using a Bruker D8 Advance with Cu-Kα radiation of λ = 0.15 406 nm. The morphology of NCM 811 and ONCM 811 was investigated by scanning electron microscopy (SEM; ZEISS EVO MA 10), equipped with an energy-dispersive X-ray (EDX) spectrometer (Oxford Instruments X-MaxN, 50 mm 2 , 15 kV), and high-resolution transmission electron microscopy (HRTEM at 300 kV; FEI Titan 80-300 with image Cs-corrector). X-ray photoelectron spectroscopy (XPS) was conducted using a Phoibos 150 XPS spectrometer (SPECS) equipped with a micro-channel plate and Delay Line Detector (DLD) and a monochromatic Al-Kα (hν = 1,487 eV) X-ray source. The scans were acquired in a Fixed Analyzer Transmission mode with an X-ray source power of 200 W (15 kV), 30 eV pass energy and 0.1 eV energy steps. The depth profiling was performed using a focused 5 keV Ar + ion gun and sputtering rate of 0.8 nm min −1 . Fitting of the spectra was carried out with the CasaXPS software, using a nonlinear Shirley-type background and 70% Gaussian and 30% Lorentzian profile functions. The spectra were calibrated by means of the C 1s peak of the adventitious carbon, centered at 284.8 eV, and the intensity was normalized. The determination of the water content was carried out via thermogravimetric vapor sorption analysis (Q5000SA, TA Instruments). The samples were kept at 25 °C for 10 min for equilibration, before being heated to 60 °C with a heating rate of 1 °C min −1 . Then, the samples were kept at a constant temperature for 300 min to evaporate all the water.
Electrode Preparation and Cell Assembly: The ONCM 811 and NCM 811 electrodes were prepared by dispersing the active material (92 wt%), C-NERGY Super C65 (TIMCAL, 4 wt%), and polyvinylidene fluoride (PVdF 6020, Solvay, 4 wt%) in NMP (Aldrich) with a solid:liquid ratio of 1:1 (wt/wt). The intimately mixed slurries were cast on Al foil (battery grade) using a laboratory-scale doctor-blade. The wet electrodes were initially pre-dried at 60 °C to remove the NMP. Subsequently, discshaped electrodes were punched and subjected to a further drying under vacuum at 100 °C for 12 h (Büchi). The average active material mass www.afm-journal.de www.advancedsciencenews.com loading was 2.2 ±0.2 mg cm −2 . All electrode preparation steps were performed in a dry room with the dew point always being below −60 °C at an ambient temperature of 20 °C.
The electrochemical characterization was conducted in CR2032 coin cells (Hohsen). Metallic lithium (Honjo, battery grade, 500 µm thickness) was used as the counter electrode, 1 m LiPF 6 dissolved in a 1:1 (by weight) mixture of ethylene carbonate (EC) and diethyl carbonate (DEC) (1:1 wt/wt) with 1 wt% of vinylene carbonate (VC), supplied by BASF, was used as the electrolyte, and a single-layer polyethylene membrane (ASAHI KASEI, Hipore SV718) was used as separator. All coin cells were assembled in an argon-filled glove box (with O 2 <0.1 ppm and H 2 O < 0.1 ppm).
Electrochemical Characterization: All galvanostatic tests were conducted in a voltage range from 3.0 to 4.3 V using a Maccor battery tester (Maccor 4000). The ambient temperature (i.e., 20, 40, and 60 °C) was controlled by a climatic chamber (Binder). A dis-/charge rate of 1C corresponds to a specific current of 200 mA g −1 . Cyclic voltammetry (CV) was performed using a multi-channel potentiostat (VMP Biologic). The voltage range was set to 3.0-4.3 V and the temperature was 20 °C. For the initial five cyclic sweeps, a lower scan rate of 0.1 mV s −1 was applied, followed by varying scan rates from 0.1 to 1.5 mV s −1 .
Ex Situ Characterization: For the ex situ SEM, XPS, and TEM analysis, the cycled cells (540 cycles at varying C rates and 20 °C -see Figure 4d) were opened in an argon-filled glove box (with O 2 <0.1 ppm and H 2 O <0.1 ppm). The electrodes were washed several times with dimethyl carbonate (DMC) to remove any residual electrolyte and then dried in a glove box. For the sample transfer to the corresponding instrument, specific sample holders were used which prohibit the contact with the ambient atmosphere. For the ex situ SEM, XPS, and TEM analysis, the same experimental parameters were used as for the pristine samples. The thermal stability of NCM 811 and ONCM 811 was determined by differential scanning calorimetry (DSC; Discovery TA instruments). The electrodes were initially cycled galvanostatically in half-cells at 0.1C for three cycles, and then charged to 4.3 V with a subsequent constant voltage step with a cut-off current equivalent to 0.01C. The cells were disassembled in a glove box and rinsed with DMC to remove any residual electrolyte. About 7-8 mg of the electrode coating were scratched from the electrodes and hermetically sealed in 30-µL high-pressure gold-plated copper stainless steel vessels, together with some fresh electrolyte. Prior to the heating the samples up to 300 °C at a heating rate of 3 °C min −1 , an equilibration step of 10 min at 40 °C was applied.

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.