Graphene-Oxide-Coated CoP 2 @C Anode Enables High Capacity of Lithium-Ion Batteries

: Cobalt diphosphides (CoP 2 ) show a high theoretical capacity and hold great promise as anode materials for lithium-ion batteries (LIBs). However, the large variation in the volume and structure of CoP 2 caused during lithium-ion insertion and extraction results in electrode fragmentation and a compromised solid electrolyte interface, ultimately leading to poor cycling performance. Herein, a composite of CoP 2 nanoparticles encapsulated in carbon matrix has been successfully synthesized by carbonization of Co-MOF-based zeolitic imidazolate frameworks (ZIF-67) and sequential phosphorization and further wrapped in graphene oxide (CoP 2 @C@GO). The formation of CoP 2 was conﬁrmed by X-ray diffraction, high-resolution transmission electron microscopy and X-ray photoelectron spectroscopy. The morphology of CoP 2 @C with and without GO wrapping was examined by scanning electron microscopy and transmission electron spectroscopy. It was demonstrated that the decoration of GO signiﬁcantly reduces the polarization of CoP 2 @C electrodes, enhancing their charge capacity and cycling stability as an anode material for LIBs. After 200 cycles, they deliver a capacity of 450 mAh · g − 1 .


Introduction
Lithium-ion batteries (LIBs) are widely used in portable electronics and electric vehicles.Currently, commercial graphite is the primary anode material used in LIBs.However, the slow kinetics of Li + intercalation and low theoretical specific capacity result in mileage anxiety for electric vehicles [1,2].Thus, there is an increasing demand for advanced anode materials in LIBs with long cycle lifespan, high capacity and superior rate performance.Hard carbon, owing to a large number of microporous structures, layered structures with a larger spacing than graphite layers, enables fast lithium-ion inversion and extraction and holds great promise in rate performance, thus it is considered another type of anode material for LIBs [3].
Transition metal phosphides are considered alternative anode materials for LIBs due to their high gravimetric and volumetric capacities (500−1800 mAh•g −1 ).In the early studies of transition metal phosphides as anode materials, their performance was relatively poor, and research primarily focused on exploring the reaction mechanisms, which strongly depend on the type of transition metal within the metal phosphides.For instance, a teardrop-shaped SnP 0.94 was prepared by reaction of tin acetate in a solution of trioctyl phosphine and triocylphosphine oxide and exhibited a reversible capacity of 740 mAh•g −1 [4] in the potential range of 0 to 1.2 V.The structural transformation of SnP 0.94 during the charging/discharging process indicates a fully reversible and purely lithium intercalation reaction process, without any metallization reaction.The electrochemical performance of such nanosized SnP 0.94 surpasses that of Sn 4 P 3 prepared using the traditional solid-state reaction method based on tin and red phosphorus.
In another study in 2009, the reaction mechanism of lithiation reaction of crystalline FeP, FeP 2 and FeP 4 as anode materials of LIBs was explored by Mossbauer spectroscopy, magnetic characterization and nuclear magnetic resonance (NMR) spectroscopy [5].Interestingly, it was found that electrochemical reaction of FeP and FeP 2 with Li ions leads to full or partial conversion to Li 3 P and Fe, accompanied with a reaction intermediate phase of FeP and LiFeP.In contrast, FeP 4 only exhibits an insertion mechanism as identified for transition metal phosphides of titanium, vanadium, manganese and tin [4], and no conversion reaction products of Li 3 P and Fe were identified at all.Usually, it is considered that the conversion of Li 3 P to LiP at potentials higher than 0.65 V may result in rapid capacity decay due to the much poorer electronic conductivity of LiP than Li 3 P [4].Thus, the formation of Li 3 P in lithiation of FeP favors the longevity of FeP anode materials in LIBs.In the following study, an amorphous iron diphosphide (FeP 2 ) [6] was obtained by reaction of Fe(N(SiMe 3 ) 2 ) 3 with phosphine (PH 3 ) in tetrahydrofuran (THF) at a low temperature of 100 • C, which exhibited extremely high discharge and charge capacity of 1258 and 766 mAh•g −1 as anode materials of LIBs.However, the cycling is limited to only 10 cycles.
Many other transition metal phosphides, such as ZnP 2 [7], Ni 3 P [8], Ni 2 P [9,10] and CuP 2 [11], have been reported.As to phosphorus-rich CuP 2 embedded in carbon matrix [11], irreversible reactions between CuP 2 and lithium were observed.Cu 3 P and amorphous phosphorus were reversibly formed in the first discharging process and subsequent cycles.It was demonstrated that the carbon content is also a significant factor in regulating the capacity of LIBs in terms of offering high electronic conductivity, fast reaction kinetics and structural integrity of the anode.
Cobalt phosphides are important anode materials, which have equally been well explored.For instance, CoP and Co 2 P binary compounds were prepared by a simple solid-state reaction method and the lithiation mechanism was explored by a combination of X-ray diffraction (XRD) and extended X-ray absorption fine structure [12].It was claimed that lithiation of CoP is possible but Co 2 P does not react with lithium or contribute to the capacity of the LIBs.Many investigations of anodes of CoP can be found in the literature, such as CoP/C modified with MCNTs [13], hierarchical porous CoP/C nanoboxes [14], CoP thin film [15], CoP nanoparticles encapsulated in coralloidal carbon [16], CoP nanoparticles wrapped in N-doped CNTs [17] and peapod-like CoP hollow nanorods [18].
Porous CoP 3 microcubes [19] were derived from the reaction of cubic Co 3 [Co(CN) 6 ] 2 particles by a modified co-precipitation method with red phosphorus and subsequently coated with polypyrrole (PPy).The conducting PPy layer is considered a buffer layer, which alleviates the large volumetric variation and promotes charge transfer kinetics during the lithiation/delithiation process.CoP [20] exhibits relatively low charge and discharge voltage plateaus, which are 0.9 and 0.6 V, respectively, rendering it a suitable anode material for secondary batteries.However, there is limited literature regarding the synthesis, structural engineering and application of CoP 2 in energy storage for LIBs [21].
It is concluded that Li-ion storage in transition metal phosphides can be mainly divided into two types [22,23]: (i) insertion mechanism and (ii) conversion mechanism, as described in Equations ( 1) and (2), respectively.M x P y + zLi → Li z M x P y (1) During the insertion and extraction reaction of Li ions in electrode materials, the large volumetric variation may result in mechanical strength, morphology evolution [24] and even fragmentation of the electrode materials and re-construction of the solid-electrolyte interface layer [25], which causes fast decay of the capacity of the batteries.Additionally, the poor electronic conductivity of Li 3 P and LiP formed during the discharging process is also an obstacle for the high capacity and long cycling performance of LIBs [4].Thus, it is desirable to develop strategies to stabilize the nanostructure of the electrode materials, constrain the solid-electrolyte interface layer and maintain the high electronic conductivity of the anode materials of metal phosphides [26].
Numerous strategies have been developed to improve the electrochemical performance of the anode materials of LIBs.For instance, preparing porous carbon-supported transition metal nanoparticles [27], sulfides and selenides [28][29][30] by using a metal organic framework as a template and self-supported phosphide nanorod arrays [31] shows great potential in obtaining high-performance anode materials of LIBs, sodium ion batteries and even lithium sulfur batteries.Graphene or graphene oxide, as a single-layered 2D carbon sheet, shows high conductivity and chemical and mechanical stability and has been widely applied in proving energy storage performances [32].Combining active electrode materials with graphene has also achieved great success in improving the electrode performance for batteries [33].Nanostructure engineering, such as making yolk-shell structures [34], hollow structures and open frameworks [35,36], shows great promise in improving the charge capacity and cycling performance of electrode materials for LIBs.
Inspired by previous work on designing high-performance nanomaterials for electrochemical energy storage performance, in this manuscript, CoP 2 @C derived from a Co-MOF-based zeolitic imidazolate framework (ZIF-67) was wrapped with graphene oxide (GO).SEM and TEM analysis revealed that the cubic CoP 2 @C nanoparticles were completely enveloped by GO.The decoration of GO on CoP 2 @C in CoP 2 @C@GO enables a reversible capacity of 452 mAh•g −1 after 200 charge/discharge cycles at the current rate of 1 A•g −1 .

Results
Figure S1A,B are SEM images of ZIF-67.After the carbonization and phosphorization of ZIF-67, CoP 2 @C nanoparticles were obtained.The morphology of CoP 2 @C and CoP 2 @C@GO samples was examined by SEM and TEM measurements, as shown in Figure 1.It is observed that CoP 2 @C nanoparticles distribute uniformly, with a smooth surface and diameter of about 50 nm (Figure 1A,D).TEM images of CoP 2 @C@GO (Figure 1B,E) indicate that CoP 2 @C nanoparticles were covered by GO sheets.The lattice fringes with an interspacing distance of 2.50 Å corresponded to (200) planes of orthorhombic CoP 2 (Figure 1C,F).
Electrochem 2023, 4, FOR PEER REVIEW 3 of 12 also an obstacle for the high capacity and long cycling performance of LIBs [4].Thus, it is desirable to develop strategies to stabilize the nanostructure of the electrode materials, constrain the solid-electrolyte interface layer and maintain the high electronic conductivity of the anode materials of metal phosphides [26].
Numerous strategies have been developed to improve the electrochemical performance of the anode materials of LIBs.For instance, preparing porous carbon-supported transition metal nanoparticles [27], sulfides and selenides [28][29][30] by using a metal organic framework as a template and self-supported phosphide nanorod arrays [31] shows great potential in obtaining high-performance anode materials of LIBs, sodium ion batteries and even lithium sulfur batteries.Graphene or graphene oxide, as a single-layered 2D carbon sheet, shows high conductivity and chemical and mechanical stability and has been widely applied in proving energy storage performances [32].Combining active electrode materials with graphene has also achieved great success in improving the electrode performance for batteries [33].Nanostructure engineering, such as making yolk-shell structures [34], hollow structures and open frameworks [35,36], shows great promise in improving the charge capacity and cycling performance of electrode materials for LIBs.
Inspired by previous work on designing high-performance nanomaterials for electrochemical energy storage performance, in this manuscript, CoP2@C derived from a Co-MOF-based zeolitic imidazolate framework (ZIF-67) was wrapped with graphene oxide (GO).SEM and TEM analysis revealed that the cubic CoP2@C nanoparticles were completely enveloped by GO.The decoration of GO on CoP2@C in CoP2@C@GO enables a reversible capacity of 452 mAh•g −1 after 200 charge/discharge cycles at the current rate of 1 A•g −1 .

Results
Figure S1A,B are SEM images of ZIF-67.After the carbonization and phosphorization of ZIF-67, CoP2@C nanoparticles were obtained.The morphology of CoP2@C and CoP2@C@GO samples was examined by SEM and TEM measurements, as shown in Figure 1.It is observed that CoP2@C nanoparticles distribute uniformly, with a smooth surface and diameter of about 50 nm (Figure 1A,D).TEM images of CoP2@C@GO (Figure 1B,E  The XRD pattern of the synthesized ZIF-67 is shown in Figure 2A, which is consistent with the simulated spectrum, indicating the high purity of the synthesized ZIF-67 [26]. Figure 2B shows the XRD pattern of CoP2@C after carbonization of ZIF-67, which is in The XRD pattern of the synthesized ZIF-67 is shown in Figure 2A, which is consistent with the simulated spectrum, indicating the high purity of the synthesized ZIF-67 [26]. Figure 2B shows the XRD pattern of CoP 2 @C after carbonization of ZIF-67, which is in agreement with the standard JCPDS card No. 26-0481, suggesting the successful preparation of pyrite CoP 2 .According to the diffraction peak at 2θ = 35.6 • , the interspacing of the (200) crystalline plane of CoP 2 in CoP 2 @C was determined to be 2.51 Å, in accordance with that obtained from HRTEM and shown in Figure 1C,F.Figures 3A and S2A display survey XPS of CoP2@C and CoP2@C@GO, where Co, P, C, O and N elements were observed.The high-resolution N 1s XPS was deconvoluted into three peaks, which were indexed to graphitic N (401.2eV), pyrrolic N (400.1 eV) and pyridinic N (399.7 eV) [37], respectively, as shown in Figure 3B. Figure 3C shows the highresolution Co 2p XPS of CoP2@C@GO.The peaks at 778.8 and 795.4 eV are assigned to Co 2p 3/2 and 2p 1/2 in CoP2 [38], which are well consistent with the result of XRD patterns in Figure 2B.Figures 3A and S2A display survey XPS of CoP 2 @C and CoP 2 @C@GO, where Co, P, C, O and N elements were observed.The high-resolution N 1s XPS was deconvoluted into three peaks, which were indexed to graphitic N (401.2eV), pyrrolic N (400.1 eV) and pyridinic N (399.7 eV) [37], respectively, as shown in Figure 3B. Figure 3C shows the highresolution Co 2p XPS of CoP 2 @C@GO.The peaks at 778.8 and 795.4 eV are assigned to Co 2p 3/2 and 2p 1/2 in CoP 2 [38], which are well consistent with the result of XRD patterns in Figure 2B.Figures 3A and S2A display survey XPS of CoP2@C and CoP2@C@GO, where Co, P, C, O and N elements were observed.The high-resolution N 1s XPS was deconvoluted into three peaks, which were indexed to graphitic N (401.2eV), pyrrolic N (400.1 eV) and pyridinic N (399.7 eV) [37], respectively, as shown in Figure 3B. Figure 3C shows the highresolution Co 2p XPS of CoP2@C@GO.The peaks at 778.8 and 795.4 eV are assigned to Co 2p 3/2 and 2p 1/2 in CoP2 [38], which are well consistent with the result of XRD patterns in Figure 2B.The XPS results of P 2p are shown in Figure 3D, which were deconvoluted into two peaks.The peak at 127.1 eV is ascribed to the P 2p of CoP 2 , indicating the successful phosphorization of cobalt, while the other peak at 132.8 eV was indexed to the partial oxidation of CoP 2 upon exposure to the air [39].It should be noted that the binding energy of Co 2p 3/2 and 2p 1/2 in CoP 2 @C is observed at 781.2 and 798.5 eV and the binding energy of P 2p in CoP 2 @C is 129.6 eV (Figure S2) [40], which are much higher than that of CoP 2 @C@GO.This indicates that decoration of GO on CoP 2 @C induces a significant red-shift of the binding energy for both P and Co of CoP 2 in CoP 2 @C@GO.The lowered binding energy and the weakened intensity of Co 2p and P 2p XPS peaks in CoP 2 @C@GO relative to those of CoP 2 @C are ascribed to the van der Waals interaction between CoP 2 and GO and the electronic screening effect of GO [41].
The electrochemical performance of the CoP 2 @C and CoP 2 @C@GO composites as anodes of rechargeable LIBs was explored in 2016-type coin cells.The initial three CV curves were acquired for anodes of CoP 2 @C and CoP 2 @C@GO in the potential window of 0.0 to 3.0 V vs. Li + /Li at a scan rate of 0.1 mV s −1 , as shown in Figure 4A.The first reduction peak at 0.523 V is related to Li + insertion into the CoP 2 @C forming Li x CoP 2 and further to Co and Li 3 P, which was shifted to 0.618 V for CoP 2 @C@GO.The second reduction peak around 0.01 V is attributed to the conversion of Li ions into carbon materials [42], which is increased to 0.173 V.The oxidization peak around 1.217 V for CoP 2 @C is associated with Co and Li 3 P oxidation to CoP 2 [43], which is almost identical for both CoP 2 @C and CoP 2 @C@GO.This indicates much reduced polarization for the CoP 2 @C anode upon GO decoration due to the enhanced electronic conductivity and the electronic screening effect.
The XPS results of P 2p are shown in Figure 3D, which were deconvoluted into two peaks.The peak at 127.1 eV is ascribed to the P 2p of CoP2, indicating the successful phosphorization of cobalt, while the other peak at 132.8 eV was indexed to the partial oxidation of CoP2 upon exposure to the air [39].It should be noted that the binding energy of Co 2p 3/2 and 2p 1/2 in CoP2@C is observed at 781.2 and 798.5 eV and the binding energy of P 2p in CoP2@C is 129.6 eV (Figure S2) [40], which are much higher than that of CoP2@C@GO.This indicates that decoration of GO on CoP2@C induces a significant red-shift of the binding energy for both P and Co of CoP2 in CoP2@C@GO.The lowered binding energy and the weakened intensity of Co 2p and P 2p XPS peaks in CoP2@C@GO relative to those of CoP2@C are ascribed to the van der Waals interaction between CoP2 and GO and the electronic screening effect of GO [41].
The electrochemical performance of the CoP2@C and CoP2@C@GO composites as anodes of rechargeable LIBs was explored in 2016-type coin cells.The initial three CV curves were acquired for anodes of CoP2@C and CoP2@C@GO in the potential window of 0.0 to 3.0 V vs. Li + /Li at a scan rate of 0.1 mV s −1 , as shown in Figure 4A.The first reduction peak at 0.523 V is related to Li + insertion into the CoP2@C forming LixCoP2 and further to Co and Li3P, which was shifted to 0.618 V for CoP2@C@GO.The second reduction peak around 0.01 V is attributed to the conversion of Li ions into carbon materials [42], which is increased to 0.173 V.The oxidization peak around 1.217 V for CoP2@C is associated with Co and Li3P oxidation to CoP2 [43], which is almost identical for both CoP2@C and CoP2@C@GO.This indicates much reduced polarization for the CoP2@C anode upon GO decoration due to the enhanced electronic conductivity and the electronic screening effect.Figure 4B displays the voltage-capacity curves of CoP2 at 1 A•g −1 in the first five cycles.The voltage-capacity curves in the initial charge/discharge cycle show two plateaus at around 1.22 and 0.618 V, respectively, in accordance with the redox peaks in the 1st cyclic voltammetry (CV) curve in Figure 4A.It is observed that the initial discharge capacity of LIBs reaches about 1450 mAh•g −1 for CoP2@C@GO and 1347 mAh•g −1 for CoP2@C@GO and a high reversible charge capacity of 1026 and 964 mAh•g −1 for CoP2@C@GO and CoP2@C, respectively.The delivered initial Coulombic efficiency is 71% for CoP2@C@GO and 72% for CoP2@C, which are close to each other.The last four discharge curves overlap with each other very well for CoP2@C@GO, indicating much-enhanced cycling performance upon GO decoration.Figure 4B displays the voltage-capacity curves of CoP 2 at 1 A•g −1 in the first five cycles.The voltage-capacity curves in the initial charge/discharge cycle show two plateaus at around 1.22 and 0.618 V, respectively, in accordance with the redox peaks in the 1st cyclic voltammetry (CV) curve in Figure 4A.It is observed that the initial discharge capacity of LIBs reaches about 1450 mAh•g −1 for CoP 2 @C@GO and 1347 mAh•g −1 for CoP 2 @C@GO and a high reversible charge capacity of 1026 and 964 mAh•g −1 for CoP 2 @C@GO and CoP 2 @C, respectively.The delivered initial Coulombic efficiency is 71% for CoP 2 @C@GO and 72% for CoP 2 @C, which are close to each other.The last four discharge curves overlap with each other very well for CoP 2 @C@GO, indicating much-enhanced cycling performance upon GO decoration.
As shown in Figure 5A, CoP 2 @C@GO exhibits a much-enhanced rate capability compared to the as-prepared CoP 2 @C.At a rate of 0.05 A g −1 , the initial discharge capacity is up to 1210 mAh•g −1 , which is superior to that of the CoP nanorod array counterpart [30] and free-standing CoP 2 nanosheets coated with hierarchical carbon [36].The subsequent decline of specific capacity is ascribed to the activation process [44].The specific capacity gradually degenerates as the current rate increases from 0.05 to 2 A g −1 .The reversible capacity of 705, 638, 564 and 460 mAh•g −1 at a current density of 0.1, 0.5, 1 and 2 A g −1 , respectively, was retained [20].This is equally superior to that of the CoP [30,45] and free-standing CoP 2 nanosheets [36] reported in the literature.When the current density is returned to 0.05 A g −1 , most of its specific capacity was recovered and retained a value of 705 mAh•g −1 .
pared to the as-prepared CoP2@C.At a rate of 0.05 A g −1 , the initial discharge ca up to 1210 mAh•g −1 , which is superior to that of the CoP nanorod array counter and free-standing COP2 nanosheets coated with hierarchical carbon [36].The sub decline of specific capacity is ascribed to the activation process [44].The specific gradually degenerates as the current rate increases from 0.05 to 2 A g −1 .The re capacity of 705, 638, 564 and 460 mAh•g −1 at a current density of 0.1, 0.5, 1 and respectively, was retained [20].This is equally superior to that of the CoP [30,45] a standing COP2 nanosheets [36] reported in the literature.When the current dens turned to 0.05 A g −1 , most of its specific capacity was recovered and retained a valu mAh•g −1 .

Figure 5. (A)
Rate performance of the CoP2@C and CoP2@C@GO at increasing current den 50 mA g −1 to 2 A g −1 .(B) The corresponding charge/discharge profiles at various rates.(C performance of CoP2@C and CoP2@C@GO arrays at 1 A g −1 . In sharp contrast, CoP2@C delivers a much lower capacity (226 mAh•g CoP2@C@GO.In particular, at a current density of 2 A•g −1 , it only delivers half the capacity of CoP2@C@GO.The improved specific capacity of CoP2@C by GO w could possibly be ascribed to the increased electronic conductivity of the anode m by GO wrapping.As indicated in earlier studies [19], a coating of a conducting P alleviates the large volumetric variation and promotes charge transfer kinetics du lithiation/delithiation process, resulting in enhanced rate performance and cycling mance.The corresponding charge/discharge curves are shown in Figure 5B, whe charge and charge plateau at 0.55 V and 1.22 V, respectively, is in accordance wit charging/discharging curves in Figure 4B.It is also observed that both the charge and discharge voltage plateaus remain constant, indicating little variation of pola  In sharp contrast, CoP 2 @C delivers a much lower capacity (226 mAh•g −1 ) than CoP 2 @C@GO.In particular, at a current density of 2 A•g −1 , it only delivers half the specific capacity of CoP 2 @C@GO.The improved specific capacity of CoP 2 @C by GO wrapping could possibly be ascribed to the increased electronic conductivity of the anode materials by GO wrapping.As indicated in earlier studies [19], a coating of a conducting PPy layer alleviates the large volumetric variation and promotes charge transfer kinetics during the lithiation/delithiation process, resulting in enhanced rate performance and cycling performance.The corresponding charge/discharge curves are shown in Figure 5B, where a discharge and charge plateau at 0.55 V and 1.22 V, respectively, is in accordance with that in charging/discharging curves in Figure 4B.It is also observed that both the charge voltage and discharge voltage plateaus remain constant, indicating little variation of polarization with the current density and superior rate capability of CoP 2 @C enabled by GO decoration.
The cycling stability of CoP 2 @C and CoP 2 @C@GO as anodes of LIBs was further evaluated by galvanostatic charge/discharge cycling at 1 A•g −1 in a voltage range of 0.01-3.0V (Figure 5C).CoP 2 @C@GO nanorod arrays show a specific capacity of about 452 mAh•g −1 after 200 cycles and a high Coulombic efficiency near 100% during all 200 cycles.However, the CoP 2 @C performs poorly in terms of stability, with only 43 mAh•g −1 after 200 cycles.The CoP 2 @C@GO exhibits superior reversible capacity and capacity retention to other anode materials for LIBs reported in the literature (Table S1) [46][47][48][49][50][51][52][53].As demonstrated in a prior report [11], increasing the carbon content offers high electronic conductivity, fast reaction kinetics, structural integrity and thus enhanced capacity of LIBs in cycling.
Figure 6 shows the SEM images of CoP 2 @C and CoP 2 @C@GO after 200 cycles.The morphology of CoP 2 @C@GO remains in shape.However, CoP 2 @C shows a lot of agglomeration and collapse of the structure.This indicates that the structure of CoP 2 @C was greatly damaged during the insertion and extraction of lithium ions, but the presence of GO can greatly increase the structural integrity of CoP 2 @C, thereby improving the cycle stability of the material.The changes in particle size for both CoP 2 @C and CoP 2 @C@GO were assessed before and after cycling using a counting method (Figure S3).The results indicated that prior to the reaction, CoP 2 @C had a particle size of approximately 65 nm ± 8 nm.However, after the reaction during cycling, significant agglomeration occurred, resulting in a larger particle size of about 150 nm ± 60 nm.In contrast, the particle size of CoP 2 @C@GO remained relatively consistent at around 60 nm (from 60 ± 8 nm to 62 ± 8 nm).This observation underscores the effectiveness of GO encapsulation in mitigating volume expansion of the anode material during battery charge and discharge cycles and substantial improvement in cycle stability.with the current density and superior rate capability of CoP2@C enabled by GO decoration.
The cycling stability of CoP2@C and CoP2@C@GO as anodes of LIBs was further evaluated by galvanostatic charge/discharge cycling at 1 A•g −1 in a voltage range of 0.01-3.0V (Figure 5C).CoP2@C@GO nanorod arrays show a specific capacity of about 452 mAh•g −1 after 200 cycles and a high Coulombic efficiency near 100% during all 200 cycles.However, the CoP2@C performs poorly in terms of stability, with only 43 mAh•g −1 after 200 cycles.The CoP2@C@GO exhibits superior reversible capacity and capacity retention to other anode materials for LIBs reported in the literature (Table S1) [46][47][48][49][50][51][52][53].As demonstrated in a prior report [11], increasing the carbon content offers high electronic conductivity, fast reaction kinetics, structural integrity and thus enhanced capacity of LIBs in cycling.
Figure 6 shows the SEM images of CoP2@C and CoP2@C@GO after 200 cycles.The morphology of CoP2@C@GO remains in shape.However, CoP2@C shows a lot of agglomeration and collapse of the structure.This indicates that the structure of CoP2@C was greatly damaged during the insertion and extraction of lithium ions, but the presence of GO can greatly increase the structural integrity of CoP2@C, thereby improving the cycle stability of the material.The changes in particle size for both CoP2@C and CoP2@C@GO were assessed before and after cycling using a counting method (Figure S3).The results indicated that prior to the reaction, CoP2@C had a particle size of approximately 65 nm ± 8 nm.However, after the reaction during cycling, significant agglomeration occurred, resulting in a larger particle size of about 150 nm ± 60 nm.In contrast, the particle size of CoP2@C@GO remained relatively consistent at around 60 nm (from 60 ± 8 nm to 62 ± 8 nm).This observation underscores the effectiveness of GO encapsulation in mitigating volume expansion of the anode material during battery charge and discharge cycles and substantial improvement in cycle stability.To gain further insights into the improved cycling stability of CoP2@C by GO wrapping, electrochemical impedance spectroscopy (EIS) was utilized as a valuable tool for investigating changes in the conductivity of the prepared electrodes, as illustrated in Figure 7.Each Nyquist plot comprises a distinctive depressed semicircle in the mid-to highfrequency domain, along with a sloping line in the low-frequency region.The diameter of this semicircle is closely associated with the charge transfer resistance (Rct).Notably, the CoP2@C@GO electrode exhibits a notably smaller Rct than that of the CoP2@C electrode.This observation highlights the superior charge transfer capability of CoP2@C@GO electrodes endowed with GO wrapping.This improvement can be attributed to the high electronic conductivity of graphene, which effectively reduces the charge transfer resistance [49].The enhanced charge transfer kinetics undoubtedly plays a pivotal role in improving To gain further insights into the improved cycling stability of CoP 2 @C by GO wrapping, electrochemical impedance spectroscopy (EIS) was utilized as a valuable tool for investigating changes in the conductivity of the prepared electrodes, as illustrated in Figure 7.Each Nyquist plot comprises a distinctive depressed semicircle in the mid-to high-frequency domain, along with a sloping line in the low-frequency region.The diameter of this semicircle is closely associated with the charge transfer resistance (Rct).Notably, the CoP 2 @C@GO electrode exhibits a notably smaller Rct than that of the CoP 2 @C electrode.This observation highlights the superior charge transfer capability of CoP 2 @C@GO electrodes endowed with GO wrapping.This improvement can be attributed to the high electronic conductivity of graphene, which effectively reduces the charge transfer resistance [49].The enhanced charge transfer kinetics undoubtedly plays a pivotal role in improving the electrochemical performance of CoP 2 @C@GO when employed as an anode material of LIBs.
the electrochemical performance of CoP2@C@GO when employed as an anode material of LIBs.
Figure 7. Nyquist plots for CoP2@C and CoP2@C@GO anodes in the frequency range of 100 kHz-100 mHz.

Synthesis of ZIF-67
All the chemicals utilized in this study were of analytical grade and were used without the need for further purification.A typical synthesis procedure for ZIF-67 is as follows: (1) Begin by dispersing 20 mmol of cobaltous nitrate hexahydrate (Co(NO3)2•6H2O equivalent to 5.820 g) in 250 mL of methyl alcohol.This is carried out under vigorous stirring for approximately 5 min, resulting in solution A. (2) In a separate container, dissolve 80 mmol of 2-methylimidazole (equivalent to 6.560 g) in 250 mL of methyl alcohol, yielding solution B. (3) Vigorously shake solution B for several minutes and then carefully pour it into solution A, all while maintaining continuous stirring.(4) As these solutions combine, an immediate color change from pink to a deep purple hue occurs.Allow the combined solution to stir for approximately 1 h.Subsequently, leave the entire solution undisturbed at room temperature for a period of 24 h.(5) The resulting precipitates can be collected through centrifugation.Afterward, wash these precipitates thoroughly with methanol, repeating the process three times.(6) Finally, transfer the washed precipitates to an oven and dry them at 70 °C overnight.This will yield the desired purple ZIF-67 material.

Synthesis of CoP2@C
The commercial red phosphorus and the ZIF-67 crystal prepared above were completely ground with a mass ratio of 1:1 in a quartz mortar for 1 h.Then, the mixed powder was transferred to a tube furnace, heated to 600 °C with a ramp rate of 2 °C min −1 and annealed under an argon atmosphere at 600 °C for 2 h for carbonization of ZIF-67 and phosphorization of Co ions.The obtained products were designated as CoP2@C.CoP 2 @C@GO CoP 2 @C Figure 7. Nyquist plots for CoP 2 @C and CoP 2 @C@GO anodes in the frequency range of 100 kHz-100 mHz.

Synthesis of ZIF-67
All the chemicals utilized in this study were of analytical grade and were used without the need for further purification.A typical synthesis procedure for ZIF-67 is as follows: (1) Begin by dispersing 20 mmol of cobaltous nitrate hexahydrate (Co(NO 3 ) 2 •6H 2 O equivalent to 5.820 g) in 250 mL of methyl alcohol.This is carried out under vigorous stirring for approximately 5 min, resulting in solution A. (2) In a separate container, dissolve 80 mmol of 2-methylimidazole (equivalent to 6.560 g) in 250 mL of methyl alcohol, yielding solution B. (3) Vigorously shake solution B for several minutes and then carefully pour it into solution A, all while maintaining continuous stirring.(4) As these solutions combine, an immediate color change from pink to a deep purple hue occurs.Allow the combined solution to stir for approximately 1 h.Subsequently, leave the entire solution undisturbed at room temperature for a period of 24 h.(5) The resulting precipitates can be collected through centrifugation.Afterward, wash these precipitates thoroughly with methanol, repeating the process three times.(6) Finally, transfer the washed precipitates to an oven and dry them at 70 • C overnight.This will yield the desired purple ZIF-67 material.

Synthesis of CoP 2 @C
The commercial red phosphorus and the ZIF-67 crystal prepared above were completely ground with a mass ratio of 1:1 in a quartz mortar for 1 h.Then, the mixed powder was transferred to a tube furnace, heated to 600 • C with a ramp rate of 2 • C min −1 and annealed under an argon atmosphere at 600 • C for 2 h for carbonization of ZIF-67 and phosphorization of Co ions.The obtained products were designated as CoP 2 @C.

Synthesis of GO
The synthesis of graphene oxide (GO) was carried out through a modified Hummers method, with the following steps: (1) Begin by combining a mixture of concentrated H 2 SO4 and H 3 PO4 in a 400 mL volume, maintaining a ratio of 9:1.(2) Add this mixture to a blend of graphite (3.0 g) and KMnO 4 (18.0 g).Keep the reaction at a constant temperature of 50 • C for a duration of 12 h.(3) After this period, allow it to cool to room temperature.(4) Gradually introduce 230 mL of H 2 O into the reaction container.Heat the solution to 98 • C and continue stirring for at least 30 min.(5) Add 500 mL of H 2 O into the resulting solution to create a diluted solution.Incorporate 5 mL of H 2 O 2 (30%) into this diluted solution.The resulting solution should exhibit a bright yellow color.(6) Wash this solution by subjecting it to three cycles of centrifugation with 5% HCl, followed by rinsing with water.These steps will lead to the production of graphene oxide (GO).
3.5.Synthesis of CoP 2 @C@GO CoP 2 @C@GO used in this study was prepared as follows.GO and CoP 2 @C were first dispersed in 1 M HCl solution, then sonicated for 10 min and stirred for 1 h.The precipitate was collected and washed with water and ethanol using a centrifuge.The product was then dried at 60 • C in air for 12 h.The obtained sample is denoted as CoP 2 @C@GO.

Materials Characterization
(1) X-ray diffraction (XRD) measurements were meticulously conducted using state-ofthe-art equipment, namely the Bruker D8 Advance instrument, which hails from Germany.This instrument harnessed the power of Cu-Kα radiation with an exceedingly precise wavelength (λ) of 1.5406 Å. (2) The comprehensive analysis of X-ray photoelectron spectra (XPS) was executed with great precision, employing the cutting-edge Phi X-tool instrument.
(3) In the pursuit of a thorough understanding of the samples' morphology, we turned to the capabilities of field-emission scanning electron microscopy (SEM), masterfully performed using the renowned Hitachi S-4800 instrument.(4) For a closer look at the nanoscale structure, high-resolution transmission electron microscopy (HRTEM) was undertaken, with no compromise in quality, thanks to the JEOL JEM-2010 microscope.

Coin Cell Assembly
A slurry-coating procedure was adopted for the preparation of the working electrode as follows.Firstly, a homogeneous slurry was prepared by mixing CoP 2 @C (or CoP 2 @C@GO), carbon black (Super P, Timcal) and sodium alginate binder at a mass ratio of 8:1:1, evenly pasted onto a copper foil by applying a film applicator and dried at 70 • C in an electric oven overnight in order to remove the solvent.The mass loading of CoP 2 @C (or CoP 2 @C@GO) was about 1.1-1.5 mg•cm −2 on each electrode.
CR2032-type coin cells were placed in an argon-filled glove box (Vigor-LG2400/750TS, Ltd., Suzhou, China), in which the oxygen and water contents were less than 1 ppm.A Li metal foil, with dimensions of 15.6 mm × 0.45 mm (diameter × thickness) was applied as both counter and reference electrode.A Celgard-2400 film was used as a separator.The recipe of the commercial electrolyte is 1.0 M LiPF 6 in a mixed solvent of ethylene carbonate (EC), ethyl methyl carbonate (EMC) and dimethyl carbonate (DMC) at a volume ratio of 1:1:1.

Electrochemical Measurements
Cyclic voltammetry (CV) and electrochemical impedance spectroscopy (EIS) measurements were conducted on a CHI 660 electrochemical workstation (Shanghai CH Instrument Co., Ltd., Shanghai, China) at room temperature.The galvanostatic charge/discharge measurements were performed on a multichannel battery testing system (CT2001A, LAND).

Conclusions
In summary, we successfully prepared CoP 2 @C nanomaterials by sequential carbonization and phosphorylation of ZIF-67 and wrapping of GO, which was confirmed by XRD, XPS and TEM.The energy storage capacity of CoP 2 @GO as an anode of LIBs was remarkably enhanced by GO wrapping, delivering a capacity of 450 mAh•g −1 after 200 cycles FigureS1A,B are SEM images of ZIF-67.After the carbonization and phosphorization of ZIF-67, CoP2@C nanoparticles were obtained.The morphology of CoP2@C and CoP2@C@GO samples was examined by SEM and TEM measurements, as shown in Figure1.It is observed that CoP2@C nanoparticles distribute uniformly, with a smooth surface and diameter of about 50 nm (Figure1A,D).TEM images of CoP2@C@GO (Figure1B,E) indicate that CoP2@C nanoparticles were covered by GO sheets.The lattice fringes with an interspacing distance of 2.50 Å corresponded to (200) planes of orthorhombic CoP2 (Figure1C,F).

Electrochem 2023, 4 ,
FOR PEER REVIEW 4 of 12 agreement with the standard JCPDS card No. 26-0481, suggesting the successful preparation of pyrite CoP2.According to the diffraction peak at 2θ = 35.6°, the interspacing of the (200) crystalline plane of CoP2 in CoP2@C was determined to be 2.51 Å, in accordance with that obtained from HRTEM and shown in Figure1C,F.

Electrochem 2023, 4 ,
FOR PEER REVIEW 4 of 12 agreement with the standard JCPDS card No. 26-0481, suggesting the successful preparation of pyrite CoP2.According to the diffraction peak at 2θ = 35.6°, the interspacing of the (200) crystalline plane of CoP2 in CoP2@C was determined to be 2.51 Å, in accordance with that obtained from HRTEM and shown in Figure1C,F.

Figure 4 .
Figure 4. (A)CV curves at a scan rate of 0.1 mV s −1 in the voltage range of 3.0-0 V vs. Li/Li + and (B) galvanostatic discharge and charge curves at 0.5 C for CoP2@C and CoP2@C@GO, respectively.

Figure 4 .
Figure 4. (A)CV curves at a scan rate of 0.1 mV s −1 in the voltage range of 3.0-0 V vs. Li/Li + and (B) galvanostatic discharge and charge curves at 0.5 C for CoP 2 @C and CoP 2 @C@GO, respectively.

Figure 5 .
Figure 5. (A)Rate performance of the CoP 2 @C and CoP 2 @C@GO at increasing current density from 50 mA g −1 to 2 A g −1 .(B) The corresponding charge/discharge profiles at various rates.(C) Cycling performance of CoP 2 @C and CoP 2 @C@GO arrays at 1 A g −1 .