Yuxuan
Ji
a,
Jian
Wei
*a,
Di
Liang
a,
Bing
Chen
a,
Xueting
Li
a,
Hao
Zhang
a and
Zongyou
Yin
*b
aCollege of Materials Science and Engineering, Xi'an University of Architecture and Technology, Xi'an 710055, China. E-mail: weijian@xauat.edu.cn
bResearch School of Chemistry, The Australian National University, Canberra, Australian Capital Territory 2601, Australia. E-mail: zongyou.yin@anu.edu.au
First published on 28th June 2023
While researchers often adopt a higher operating voltage to further enlarge the actual specific capacity of LCO to expand its application scope and market share, this triggers some more intractable issues in that the capacity decays obviously and causes the attendant problem of safety. Li3PO4 shows the advantage of increasing the energy density of lithium-ion batteries due to its characteristic ionic conduction when coated onto an LCO cathode. Enhancing the conductivity of cathode materials is the key factor in the success of raising their operating voltage to meet emerging market demands. Here, we report a direct facile coprecipitation method for coating crystalized Li3PO4 onto an LCO surface that enables balancing the ionic conductivity and chemical stability. LCO@ Li3PO4 crystalline lithium phosphate can generate superior electrical contact with the cathode material for high capacity and effectively stabilize the cathode surface by reducing the formation of SEI/CEI to prolong the cycle life. The optimized LP-3 cathode can deliver a high initial discharge capacity of 181 mA h g−1 at 0.5C, with a capacity retention of 75% after 200 cycles. This study introduces a competitive strategy to produce a high-voltage LCO cathode via the most viable and economical method.
Various strategies have been explored to circumvent these drawbacks, such as elemental doping,9 surface coating,10–12 morphology design,13 and employing separators,14 electrolytes,15 and the corresponding additives.16 Although elemental doping shows a surprising result, the improvement is limited, especially at a higher voltage (>4.45 V) and under more extreme operating conditions. Doping is only effective in stabilizing the structure, eliminating the phase transition.4 To date, no study has proven that doping can modify the surface structure to control surface reactions.17 In this regard, surface coating is competitive in solving this intractable problem. The coating, just as an effective physical obstacle, impedes direct contact and hinders reactions between LCO and the electrolyte.18 Researchers have explored, in recent years, all kinds of coating materials, including oxides,12,19–21 fluorides,22–25 and phosphates.26,27 Owing to the superior strength of PO bonds, phosphates have been regarded as a potential coating material to deliver higher electrochemical stability,17,28 such as AlPO4,29 FePO4,27 Li3PO4,30 LiCoPO4,49 CePO4, and LiMgPO4.31 Although they have outstanding electrochemical stability and improved cycle life, most of them introduce inactive elements, and their conductivity is poor,32 which impedes ion and electron transfer, especially at high current densities.33 On the other hand, other researchers have investigated, in recent years, fast ion conductor materials31,32,34,51 and conductive polymers35,36 to accelerate the transmission of ions or electrons, improving the electrochemical performance, such as PAN,36 Li4Ti5O12,37 LiVO2,32 and Li2ZrO3.38 Unfortunately, the cycling stability of most fast-ion conductor materials and conductive polymers is lower than that of phosphates.
Li3PO4 is a representative choice, with higher ionic conductivity (∼6 × 10−8 S cm−1) than most oxides or fluorides39 while retaining the advantages of phosphates. For example, researchers have found that a slight coating of Li3PO4 is favorable for enlarging the capacity of the cathode and increasing the redox electrochemical potentials, owing to the inductive effect of PO43− ions and reduced loss of Co3+.40 Sun et al.41 analyzed the mechanism by which a Li3PO4 coating produced by a fire and quench method can improve the rate performance, as Li3PO4 benefited the intercalation of Li+, thus leading to a better rate capacity than before. However, the coating formed by this method is not uniform enough. Zhou et al.30 coated amorphous Li3PO4 on LCO by magnetron sputtering, whether at room temperature (RT) or high temperature, and achieved outstanding cycling stability, which offers concrete support for the application potential of the Li3PO4 coating. Y. Wang et al. developed a strategy through annealing a surface layer to form a high-voltage-stable surface coating layer in situ, which was demonstrated to be highly effective in improving the high-voltage performance of LiCoO2.48 Although the coating formed by this method is more uniform, the preparation cost is high. Most researchers have found that crystalline coatings can achieve long stable cycling,32,37 and for chemical stability, crystalline Li3PO4 is superior to amorphous Li3PO4.33,42 Therefore, the Li3PO4 coating with a certain degree of crystallinity endows the electrode material with both good capacity and stability. Therefore, it is urgent to develop a more stable Li3PO4 coating with a lower fabrication cost, which is more favorable for industrial applications.
In this paper, we used a scalable coprecipitation method with appropriate annealing, a mainly focused industrial method,43 to develop phase tunable Li3PO4 coating layers on the surface of LCO to prevent the deterioration of the cathode while accelerating the transmission of Li+. By further adjusting the annealing temperature and coating thickness, we found that LCO@Li3PO4 can achieve an outstanding initial capacity of 181 mA h g−1 with a retention of 90% after cycling at 0.5C and has a high reversible capacity of 132 mA h g−1 at 3C, which is superior to many reported phosphates and fast ion conductor coatings, implying that using the wet chemical method to form the crystalline Li3PO4 coating is a promising technique for the application of high-voltage LCO cathodes.
3LiOH + NH4H2PO4 = Li3PO4 + NH3 + 3H2O | (1) |
Fig. 1 Top: schematic illustration of the preparation process of Li3PO4 coated LiCoO2. Bottom: (a and b) the SEM images of BLCO and (c and d) LP-3 and (e–g) the corresponding elemental maps. |
In Fig. 1a–d, from the FESEM images, the particles of commercial LCO are irregular; the size is approximately 5–10 nm, and the surface is rough and porous, facilitating the formation of the coating. After coating, there are some unknown particles on the clean surface, and the difference in terms of the size and morphologies of the BLCO and LP powders is not obvious, demonstrating that the coating does not influence the main phase of the LCO. From the corresponding elemental maps of Co, O, and P in Fig. 1e–g, the coating obtained by the coprecipitation method is uniform. The images in Fig. S2† indicate that the size of nanoLi3PO4 increased with the holding time.
The XRD patterns in Fig. 2a indicate that BLCO and all the LP electrodes exhibit the same peaks. This indicates that the Li3PO4 coating does not influence the main phase and structure of bulk LCO, which is consistent with the conclusion from SEM, and the lack of additional peaks in accord with Li3PO4 may be due to its comparatively low content or crystallinity.46
As the XRD results cannot directly confirm the existence of Li3PO4, Raman spectra and XPS spectra are used to explain the conditions of the BLCO and LP-3 surface. In Fig. 2b, after the coating, the spectra at 481 cm−1 and 592 cm−1 corresponding to the Eg (O–Co–O) and A1g (Co–O) of LCO are not changed, and no additional peaks appear. This indicates that the main structure is still layered LCO, the coating is not doped in the layered structure, and the strength of the peaks is reduced, which also reflects some unknown substances on the surface. In Fig. 2c–f, the characteristic peak of P is only found in the LP-3 sample, from the full spectra in Fig. 2c, indicating that phosphorus does exist on the surface. Fig. 2d shows the Co 2p spectrum collected for LCO and LP-3. There are two main peaks in the spectra collected at Co 2p at nearly 795 eV and 780 eV for both samples corresponding to Co 2p1/2 and Co 2p 2/3, respectively,33 which indicates that the coating does not influence the main components of LCO. Moreover, in the O 1s spectra, it is notable that, after coating, we can find three kinds of characteristic peaks. The peak located at 529.34 eV corresponds to the slightly enhanced Co–O, which confirms that excessive NH4H2PO4 can produce a cleaner surface, and the slightly minimized peak at 531.28 eV corresponds to the surface absorbed oxides,42 indicating that some surface oxides may react with NH4H2PO4 to form Li3PO4. In O 1s, we can find a new peak located at 530.84 eV, which represents PO43− on the surface. In the P 2p spectra, we see a distinct peak at 133.17 eV that can be ascribed to phosphate. The above evidence directly proves that Li3PO4 was successfully coated on the LCO surface by this method.
To demonstrate the effect of different annealing times on the crystallinity of the coatings, the products with the shortest and longest annealing times, LP-1 and LP-3, were characterized by TEM. Enlarged LCO images and the corresponding FFT images are given in Fig. 3a–c, showing that the layered structure is still preserved well, with coatings on the surface. From the enlarged coating images (Fig. 3c), we can see the fuzzy lattice, and from the FFT images (Fig. 3e and h), it can be seen that the LP-3 coating is not amorphous, and the crystallinity is not high; it has slightly crystalline coating; However, in Fig. 3g and j, we cannot see the lattice and from the FFT images, we can conclude that the coating is amorphous. By comparison, it is clear that the longer the annealing time is, the higher the crystallinity of the coatings. In Fig. 3b and c, the Li3PO4 coating is visible, with a thickness of approximately 20 nm. When Li3PO4 is used as a coating material, higher crystallinity can reduce its ionic conductivity,47 while certain crystallinity can resist the corrosion of H2O, HF and O2; low crystallinity causes Li3PO4 to have certain ion conduction, accelerating the diffusion of Li+,47 which can increase the initial capacity while improving the cycle life. Consequently, we conclude that a Li3PO4 coating layer with certain crystallinity could be achieved by changing the annealing conditions.48
Fig. 3 TEM images of (a–c) LP-3 and (d) LP-1 and (e–j) enlarged images with the corresponding FFT images of LP-1 (g and j) and LP-3 (e, f and h–i). |
In Fig. 4, we compared the electrochemical performance of BLCO and LP to study the effect of this modification and confirmed the optimal preparation method for Li3PO4, which is crucial to achieve the best cycling and rate performance of the cells. The initial discharge capacities (Fig. S3a†) of BLCO and LP at 0.1C are 184, 176, 171, 186 and 160 mA h g−1. The initial capacities of BLCO and LP are similar, indicating that the coating contributes little to the initial capacity, and improper crystallinity causes a large amount of highly crystalline Li3PO4 to accumulate on the surface of LCO, which affects the transmission of electrons and reduces the initial capacity.
In Fig. 4a, we show the cycling performance of BLCO and LP areand the values are listed in Table S1.† While the gap in the initial capacity is not obvious between BLCO and LP, the enhancement of cycling stability is noticeable for LP cathodes. The capacity of LCO decreased rapidly from 176.7 to 88.2 mA h g−1 after 100 cycles. For LP cathodes, capacity retention was 78% (from 172 to 134 mA h g−1), 89% (from 168 to152 mA h g−1), 90% (from 181 to 163 mA h g−1) and 52% (from160 to 84 mA h g−1), while LCO lost half of its initial capacity from 176.7 to 88.2 mA h g−1 after 100 cycles, indicating that proper crystallinity is beneficial for the cycling performance. Moreover, the outstanding ion conductivity can significantly alleviate the defect of the efficiency of low initial capacity by insulating coatings. The rate performances of BLCO and LP are shown in Fig. 4b, and the values are listed in Table S2.† As the rate increased, for all the LP samples, the reversible capacity decreased slightly compared to that of BLCO. During the high rate of cycling, the capacity fading of LP-3 is obviously alleviated, with the highest reversible capacity reaching 152 mA h g−1 at 3C. When the rate returns to 0.5C, the capacity retention of LP-3 is over 97%. All these results indicate that the Li3PO4 coating can promote the transportation of Li+ and prevent the corrosion of electrolytes. In Fig. 4c, LP-3 maintained an ultrahigh capacity of 137 mA h g−1 after 200 cycles with a capacity retention of 76%. In Fig. 4d–e and Fig. S3c,† we show the cycling performance of LP-3 at 2C, 3C and 4C. After 100 cycles, the discharge platform is still well preserved, indicating that Li3PO4 with proper crystallinity can protect the layered structure from the corrosion of the electrolyte during the cycling process. The capacity retention can reach more than 90% after cycling at 2C and 3C. Moreover, the capacity can reach 140 mA h g−1 at 4C.
We show the CV of BLCO and LP-3 in Fig. S4a and b† to further explore the protective effect of the Li3PO4 coating on the cathode structure. For the initial charge/discharge process, both BLCO and LP-3 show four reduction peaks,7 and the most intense peak is at 3.81 V, which reflects the phase transitions between two hexagonal phases, H1 and H2. The oxidation peaks at 4.08 and 4.17 V are due to the order–disorder phase transitions.24 After coating, the location of the redox peaks is not changed, and the strength is higher, which indicates that the reversible redox reactions that occur during the charging and discharging of the LCO cathodes are still present after coating, which enables the LCO to exhibit its optimal discharge capacity.
In Fig. 5a and b, we show the EIS profiles of BLCO and LP before and after the cycling test to explain the role of the Li3PO4 coating, and after fitting, the specific value is shown in Table S3.† Crystalline Li3PO4 is an insulating substance, while amorphous lithium phosphate has excellent ionic conductivity.39,41 Studies have shown that ionic conductive coatings can promote the transport of electrons and effectively improve the electrochemical properties of materials.48,50 However, the amorphous lithium phosphate coating has poor stability, which is not conducive to the improvement of electrochemical stability. In terms of chemical stability, the crystalline Li3PO4 is superior to the amorphous one.33 The characteristics of lithium phosphate in different states influence the internal resistance. The resistance of LP-4, which has the highest crystallinity, is higher than that of any other sample. However, based on our subsequent electrochemical test results, low crystallinity has poor inert chemical properties. Therefore, we should explore a proper crystallinity that can balance the ionic conductivity and inert chemical properties, to achieve the best electrochemical performance.
To further demonstrate the protective effect of the coating, aging tests were carried out on LP-3, which has the best electrochemical properties. In Fig. 5a and b, after cycling, whether Rf or Rct, the value of LP-3 is lower than that of BLCO, which indicates that the Li3PO4 coating can stabilize the surface, reducing the formation of SEI/CEI and other substances that hamper the transfer of electric charge and increase the surface resistance. Fig. 5c–f show the TEM morphology of the BLCO and LP-3 electrodes after 100 cycles at 0.5C. Both of them generate an amorphous CEI film on the surface during the cycles, while the delicate film of LP-3 is thinner than that of BLCO. The enlarged images further disclose the surface situation of the two cycled cathodes. The part circled in yellow is the CEI film, which consists of harmful substances produced by the reaction between LCO and the electrolyte during the cycling process. The thickness of BLCO is over 65 nm and that of LP-3 is approximately 30 nm. The byproduct on LP-3 is thinner and more compact, which indicates that the coating effectively protects the cathode material and is not significantly different from the fresh cathode structure.
Based on previous analysis, we used the model in Fig. 6a to explain the role of the Li3PO4 coating. It is well known that crystalline Li3PO4 has a distinct structure with strong covalent bonding of PO, which can maintain stability in HF, O2 and H2O, allowing Li3PO4 to act as a physical obstacle to mitigate the loss of Co3+ and suppress harmful side reactions, leading to outstanding cycling stability at high voltage.39Although amorphous Li3PO4 is an outstanding ionic conductive material, it has less chemical inertness than crystalline Li3PO4, which dilutes its outstanding chemical inertness.42 We used the coprecipitation method with the optimal annealing technique to form a coating, with certain crystallinity, to combine the advantages of both ion conductivity and inert chemical properties. It has superior inert chemical stability to most fast ion conductors. Simultaneously, as the crystallinity is not high, Li3PO4 can possess better ion conductivity than most oxides and fluorides, accelerate Li+ diffusion and benefit the rate performance less than other oxides,47 achieving the best balance between stability and transmission. Moreover, it does not introduce other additional metal elements, which is beneficial for reversible capacity. It is notable that this method is cheap and simple for mass production, which makes it more competitive in the high-voltage LCO market.
In Fig. 6b, we compare the initial capacity, capacity and retention after 100 cycles. We can see that the capacity and retention of crystalline Li3PO4 are better than those of amorphous Li3PO4. Although certain crystallinity sacrifices a small amount of ionic conductivity and influences the initial capacity, it has better chemical stability, leading to higher capacity and retention after cycling. Table S4† lists some phosphates and fast ion conductor coatings. Different charge voltages and current densities provide obvious differences in capacity. Therefore, we chose some that worked at the cutoff voltage of 4.5 V and listed their different current densities to compare the capacity retention and the specific capacity before and after the cycling, as shown in Fig. 6c and d. As Li3PO4 with a certain crystallinity can combine the advantages of phosphates and fast ion conductor materials, it has higher chemical stability than some fast ion conductor materials and phosphates. By the wet chemical method, crystalline Li3PO4 can be successfully coated on the LCO surface and a relatively superior electrochemical performance can be achieved.
Footnote |
† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3nr01251d |
This journal is © The Royal Society of Chemistry 2023 |