Temperature-driven structural evolution of carbon modified LiFePO₄ in air

Abstract

Lithium iron phosphate (LiFePO₄) is an appealing cathode material for lithium ion batteries. However, the degradation of LiFePO₄ in air presents an unavoidable challenge, due to the vulnerability of divalent Fe against oxygen attack. In this work, we have carried out comprehensive research on the thermal stability and temperature-driven evolution of nanocarbon modified LiFePO₄ in air. The results show that LiFePO₄ retains structural stability up to 250 °C for short periods of exposure to air. At long exposure times, structural evolution occurs at a much lower temperature, 150 °C. The structural evolution proceeds as the temperature increases, and finishes at 400 °C. The final products are monoclinic Li₃Fe₂(PO₄)₃ and α-Fe₂O₃. A quantitative evolution map has been developed through electrochemical cyclic voltammetry and galvanostatic tests. The results show that the largest changes take place between 200 and 250 °C.

---

1 Introduction

Lithium iron phosphate (LiFePO₄) is one of the most appealing cathodes for large format lithium ion batteries (LIBs) due to its low cost, high specific capacity, excellent cycling performance, and superior safety.^1,2 These merits are inherently ascribed to its olivine structure. Olivine LiFePO₄ crystallizes in an orthorhombic structure, in which oxygen atoms form a slightly distorted hexagonal close-packed arrangement, and both Li and Fe atoms take the octahedral positions. The PO₄ polyanion enhances the redox potential of Fe²⁺/Fe³⁺ to 3.4 V (vs. Li, unless otherwise stated) via an inductive effect, but also prevents a possible oxygen release under harsh conditions.³ However, the low electronic conductivity (∼10⁻⁹ S cm⁻¹) and the resulting sluggish kinetics present a significant challenge for the scalable utilization of LiFePO₄. To address this issue, numerous strategies have been proposed and shown promising prospects during the last few decades. These strategies include surface modification,⁴ nanocarbon wiring,^5,6 ion doping,^7,8 size reduction,⁹ shape tuning,¹⁰ and so forth.

Generally, reduction of particle size can efficiently mitigate the kinetic drawbacks of LiFePO₄.^11,12 However, fine LiFePO₄ materials are sensitive to air exposure, due to their large surface area and high activity.¹³ This presents a significant challenge for the storage and processing of LiFePO₄ materials, when the materials are highly at risk of being exposed to air at high temperatures. Previously, Martin et al. reported that air exposure of LiFePO₄ at 120 °C could result in the loss of 2.3% of lithium (oxidation of LiFePO₄).¹⁴ Certainly, this evolution would be more serious at higher temperatures. Therefore, it is essential to probe the thermal evolution process of LiFePO₄ at ambient temperature, and particularly at temperatures above room temperature. Nonetheless, such an issue has barely been addressed so far.

In this work, the evolution of LiFePO₄ in air as a function of the temperature has been examined. Carbon coated LiFePO₄ was chosen as the starting material, as carbon modification is currently popular for olivine materials. A series of air-exposed LiFePO₄ derivatives at different temperatures were prepared. These materials were systematically investigated using various structural, spectroscopic, and electrochemical analyses. Analysis of these results enable us to draw a clear map of the temperature-driven evolution of LiFePO₄ in air.

2 Experimental

Carbon coated LiFePO₄ materials were prepared via solid-state reaction from Li₂CO₃, Fe(II)C₂O₄·2H₂O and NH₄H₂PO₄ and sucrose. The final product was predicted to contain 1 wt% carbon. Details of the experiments can be found in the ESI.† Thermal evolution of LiFePO₄ materials was probed through two approaches. One was by collecting thermogravimetric analysis (TGA) data on a SDT 2960 apparatus (TA Instruments). The other was to expose LiFePO₄ materials to different temperatures, and then to characterize these derivatives using chemical and electrochemical analyses.

3 Results and discussion

The structure of LiFePO₄ was identified by XRD and Rietveld refinement (Fig. S1†). The XRD pattern can be fully indexed to the olivine structure (JCPDS no. 81-1173) without diffraction peaks due to impurities. Rietveld analysis on the XRD data revealed the lattice parameters to be a = 1.0316(2) nm, b = 0.6002(3) Å, c = 0.4690(3) Å, and V = 0.2904 nm³, in line with previous results.^3,15 This result implies that the olivine phase has been readily fabricated through a solid route.

Fig. 1 shows the TEM images of LiFePO₄, revealing that most grains are in the nanoscale with an average particle size of 100 nm. An amorphous nanolayer covering the particles can be clearly observed, which is identified as the carbon derived from sucrose (Fig. 1a). The thickness of the carbon nanolayer is only about 2 nm (Fig. 1b) due to low carbon loading, but it remarkably enhances the material conductivity to 2 × 10⁻² S cm⁻¹.

Fig. 1 Morphology of the prepared LiFePO₄. (a) TEM and (b) high resolution TEM images. The lattice fringe spacing of 0.392 nm shown in (b) coincides with the (210) facets of olivine LiFePO₄.

The thermal evolution of LiFePO₄ in air was examined by TGA, and the results are presented in Fig. 2. The data were collected from room temperature to 700 °C at a ramp rate of 10 °C min⁻¹. Mass changes reflecting a possible phase evolution can be readily seen from the TGA curve. There are two possible reactions involved in the process. One is the oxidation of LiFePO₄ (eqn (1)) and the other, the burning of the carbon coating (eqn (2)).¹⁶ In detail, the mass loss of 0.4 wt% below 150 °C is due to the removal of adsorbed water. From 150 to 250 °C, the mass remains constant (inset in Fig. 2), suggesting that LiFePO₄ is still stable in air at this temperature range. From 250 to 400 °C, the gradual mass augment suggests a possible oxygen uptake due to oxidation of the divalent Fe species (eqn (1)). Interestingly, this mass then remains steady in the following temperature range of 400–450 °C. However, this is probably because the mass gain and loss due to LiFePO₄ oxidation and carbon loss (eqn (2)) reach a balance. Above 450 °C, most of the amorphous carbon has burnt out and the bulk oxidation of divalent Fe occurs, leading to a net mass gain of 3.6 wt%. As the full oxidization of LiFePO₄ (eqn (1)) results in a mass gain of 5.1 wt%, the exact carbon loading in the LiFePO₄ product is 1.1 wt%.

6LiFePO₄ + 3/2O₂ → 2Li₃Fe₂(PO₄)₃ + Fe₂O₃ (1)

C + O₂ → CO₂ (2)

Fig. 2 TGA curves of LiFePO₄ in air at ramp rate of 10 °C min⁻¹. The inset shows the curve before 350 °C.

To investigate the structural evolution under long exposure to air, the LiFePO₄ samples were held at set temperatures (150–400 °C) for 3 h. These derivatives were then subject to structural, spectroscopic and electrochemical analyses to reveal the structural evolution process. XRD patterns of these LiFePO₄ products are shown in Fig. 3. It is clearly seen from Fig. 3 that degradation of the olivine phase starts at 200 °C. At this temperature, the diffraction intensity decreases and splitting of the (020) peak in the pattern is observed. When the temperature increases to 250 °C, new diffraction peaks due to monoclinic Li₃Fe₂(PO₄)₃ (ref. 17) and α-Fe₂O₃ (ref. 18) appear in the pattern of the derivative. This result indicates that bulk phase evolution due to oxygen uptake occurs. Further increasing the temperature to 300 °C results in more Li₃Fe₂(PO₄)₃ and Fe₂O₃ phases at the expense of the olivine structure. When the LiFePO₄ is treated at 400 °C, no peaks due to the olivine phase can be detected in the XRD pattern, suggesting a complete structural evolution.

Fig. 3 Evolution of XRD patterns of LiFePO₄ exposed to air at different temperatures. The LiFePO₄ samples were held at the set temperatures for 3 h.

In addition to XRD, the temperature-driven structural evolution of the LiFePO₄ was further studied by FTIR. It is known that LiFePO₄ has two types of vibrational motions in the FTIR: internal modes originating from intramolecular vibrations of PO₄ polyanion and an external mode due to lattice vibration.¹⁹ The latter, involving Li ion motion, usually occurs below 400 cm⁻¹ and it is difficult to discern, thus will not be discussed here. As shown in Fig. 4a, fresh LiFePO₄ exhibits four absorption bands at 1139, 1095, 1056, and 973 cm⁻¹, which can be assigned to the stretching vibrations of the PO₄ group. Five additional bands at 637, 578, 551, 502 and 471 cm⁻¹ can be ascribed to the bending mode of the PO₄ group.²⁰ The derivative treated at 150 °C shows quite a similar spectrum as the fresh LiFePO₄, which implies an intact microstructure, consistent with the XRD results. However, structural evolution can be clearly observed for the sample heated at 200 °C. In this spectrum, the stretching bands of the PO₄ group become broad and the band at 1095 cm⁻¹ loses its intensity to a large degree. Pronounced evolution is visible in the spectrum of the sample treated at 250 °C. The band at 1095 cm⁻¹ disappears and two new bands at 668 and 437 cm⁻¹ appear, which can be ascribed to α-Fe₂O₃.²¹ Further raising the temperature to 300 and 400 °C results in the disappearance of more absorption bands ascribed to the olivine phase. This is accompanied with the emergence of new absorption bands located at 1182, 962, 602 cm⁻¹, which are related to monoclinic Li₃Fe₂(PO₄)₃.

Fig. 4 Evolution of (a) FTIR and (b) XPS spectra of LiFePO₄ exposed to air upon heating at various temperatures.

As the oxidation of divalent Fe triggers the structural evolution of the olivine phase, XPS spectroscopy was used to probe the evolution of the chemical valence of Fe in LiFePO₄. XPS is a typical surface technique with high sensitivity, thus it should give direct evidence of structural evolution. Fig. 4b shows the XPS spectra of Fe 2p for LiFePO₄ and derivatives. The peaks at 710.5 eV and 723.6 eV for the fresh LiFePO₄ can be assigned, respectively, to Fe 2p_3/2 and Fe 2p_1/2 in the divalent state.^22,23 A shift of the Fe 2p peaks to higher binding energies is observed for the heated derivatives, indicating that divalent Fe species in the surface are readily oxidized even at 150 °C. For the derivatives exposed to a temperature of 200 °C or above, the Fe 2p_3/2 and 2p_1/2 peaks shift to higher binding energies of 711.5 eV and 725.2 eV, respectively. This shift, in combination with the occurrence of two shoulder peaks at 719.8 eV and 732.7 eV, proves that divalent Fe has been transformed into the trivalent form.²² It is worth noting that this temperature is lower than that obtained by XRD and FTIR experiments, since XPS is more sensitive to surface changes and the evolution of the Fe species probably initiates from the surface of the grains.

The XRD and FTIR results suggest that evolution of bulk LiFePO₄ initiates at 200 °C, while the XPS data suggest that surface degradation starts at a lower temperature of 150 °C.¹⁴ To understand this evolution from the viewpoint of electrochemistry, CV and galvanostatic tests on the LiFePO₄ samples were carried out. Fig. 5a displays their CV curves in the first two cycles at a scanning rate of 0.1 mV s⁻¹. The fresh LiFePO₄ discloses a distinct redox pair at 3.4 V, representing the typical Fe²⁺/Fe³⁺ redox reaction in olivine.^1,24 This reaction reveals a considerable reversibility, due to the small particle size of LiFePO₄ and uniform carbon coating.^25,26 Surprisingly, the CV curve is highly sensitive to structural evolution. A redox pair emerges at 2.7 V for LiFePO₄ after exposure at 150 °C, unambiguously reflecting the existence of structural changes at this temperature. The 2.7 V peak can be ascribed to Li insertion/extraction in monoclinic Li₃Fe₂(PO₄)₃.²⁷ The peak for this redox pair continues to grow with the increasing temperature at the expense of the original 3.4 V pair. Finally, the peak at 3.4 V disappears for the 400 °C sample, and only the 2.7 V redox peak remains, suggesting a full evolution of the LiFePO₄ phase.²⁸ This CV results correlate well with the structural and spectroscopic analyses.

Fig. 5 (a) CV graphs and (b) discharge curves of LiFePO₄ exposed to air upon heating at various temperatures.

Fig. 5b shows the galvanostatic discharge curves during the first cycle. The fresh LiFePO₄ delivers a reversible capacity of ∼150 mA h g⁻¹ with a flat potential plateau at 3.4 V, indicating the favorable electrochemical activity of the olivine material.⁸ Similarly, a slight change at the end of the discharge can be found for the LiFePO₄ exposed to air at 150 °C. This change becomes remarkable for the samples treated at 200 °C or higher temperatures, and a new potential at 2.7 V appears. The new plateau matches the CV peak at 2.7 V, and can be due to Li insertion into the monoclinic Li₃Fe₂(PO₄)₃ phase. Theoretically, Li₃Fe₂(PO₄)₃ can electrochemically accommodate two Li ions per formula unit, leading to a capacity of 128 mA h g⁻¹.²⁷ Increasing the temperature results in a reduction of the 3.4 V plateau and a concomitant increase of the 2.7 V one monotonically. The LiFePO₄ derivative held at 400 °C only exhibits one plateau at 2.7 V, indicating the entire loss of the olivine phase. Meanwhile, the overall discharge capacity decreases with the increasing temperature, probably due to the lower capacity of the monoclinic phase and the gradual loss of the conductive carbon network. Ignoring kinetic factors, we can quantitatively draw an evolution map by comparing the length of the two plateaux (Fig. S2, ESI†). The results reveal that thermal evolution starts at 150 °C and the major process takes place between 200 and 250 °C. Previously, Martin et al. reported a 2.3% evolution of LiFePO₄ at a lower temperature of 120 °C.¹⁴ This discrepancy may stem from the carbon coating layer. During the synthesis, we added sucrose to form a composite rather than carbon as Martin et al. did in their work. This in situ carbon coating process forms an uniform overlayer on the surface of the LiFePO₄ particles and thus provide good protection from oxygen attack,^29–31 which can significantly retard the structural transformation.

4 Conclusions

The temperature-driven structural evolution of carbon coated LiFePO₄ in air was investigated using various structural, spectroscopic and electrochemical techniques. The results indicate that nanocarbon coated LiFePO₄ is stable up to 250 °C over short periods of exposure to air. Under long exposure times, however, the bulk phase stability is maintained at 200 °C while the surface evolution initiates at a much lower temperature, 150 °C, as revealed by XRD, FTIR, XPS and electrochemical analyses. Structural evolution proceeds as the temperature increases, and the final products include monoclinic Li₃Fe₂(PO₄)₃ and α-Fe₂O₃. The evolution process can be quantitatively drawn through electrochemical galvanostatic tests. The results show that the largest structural changes take place between 200 and 250 °C. This research provides quantitative understanding on the temperature-driven structural evolution of olivine LiFePO₄ materials in air, and surely will benefit the related material research and production.^32–34

Acknowledgements

The support of this work by the National Natural Science Foundation of China (51302181), China Postdoctoral Science Foundation (2014M551647), and SRF for ROCS, SEM is acknowledged.

Notes and references

1. A. K. Padhi, K. S. Nanjundaswamy and J. B. Goodenough, J. Electrochem. Soc., 1997, 144, 1188–1194.
2. M. S. Whittingham, Chem. Rev., 2014, 114, 11414–11443.
3. C. Hu, Y. Wu and Y. Dai, Funct. Mater. Lett., 2014, 07, 1450016.
4. M. Molenda, A. Chojnacka, M. Bakierska and R. Dziembaj, Mater. Technol., 2014, 29, A88–A92.
5. J. W. Zhang, L. H. Zhuo, L. L. Zhang, C. Y. Wu, X. B. Zhang and L. M. Wang, J. Mater. Chem., 2011, 21, 6975–6980.
6. J. F. Ni, Y. H. Han, L. J. Gao and L. Lu, Electrochem. Commun., 2013, 31, 84–87.
7. J. F. Ni, H. H. Zhou, J. T. Chen and X. X. Zhang, Chin. J. Inorg. Chem., 2005, 21, 472–476.
8. J. Ni, Y. Zhao, J. Chen, L. Gao and L. Lu, Electrochem. Commun., 2014, 44, 4–7.
9. J. F. Ni, M. Morishita, Y. Kawabe, M. Watada, N. Takeichi and T. Sakai, J. Power Sources, 2010, 195, 2877–2882.
10. C. Nan, J. Lu, C. Chen, Q. Peng and Y. Li, J. Mater. Chem., 2011, 21, 9994.
11. L. X. Yuan, Z. H. Wang, W. X. Zhang, X. L. Hu, J. T. Chen, Y. H. Huang and J. B. Goodenough, Energy Environ. Sci., 2011, 4, 269–284.
12. C. Gong, F. Deng, C.-P. Tsui, Z. Xue, Y. S. Ye, C.-Y. Tang, X. Zhou and X. Xie, J. Mater. Chem. A, 2014, 2, 19315–19323.
13. Y. G. Wang, P. He and H. S. Zhou, Energy Environ. Sci., 2011, 4, 805–817.
14. J. F. Martin, A. Yamada, G. Kobayashi, S. Nishimura, R. Kanno, D. Guyomard and N. Dupré, Electrochem. Solid-State Lett., 2008, 11, A12–A16.
15. P. F. Xiao, B. Ding, M. O. Lai and L. Lu, J. Electrochem. Soc., 2013, 160, A918–A926.
16. I. Belharouak, C. Johnson and K. Amine, Electrochem. Commun., 2005, 7, 983–988.
17. M. Morcrette, C. Wurm and C. Masquelier, Solid State Sci., 2002, 4, 239–246.
18. X.-L. Cheng, J.-S. Jiang, M. Hu, G.-Y. Mao, Z.-W. Liu, Y. Zeng and Q.-H. Zhang, CrystEngComm, 2012, 14, 3056–3062.
19. C. M. Burba and R. Frech, J. Electrochem. Soc., 2004, 151, A1032–A1038.
20. J. F. Ni, Y. Han, J. Liu, H. Wang and L. Gao, ECS Electrochem. Lett., 2013, 2, A3–A5.
21. G. H. Jiang and J. S. Jiang, Chem. Res. Chin. Univ., 2004, 25, 405–408.
22. M. Descostes, F. Mercier, N. Thromat, C. Beaucaire and M. Gautier-Soyer, Appl. Surf. Sci., 2000, 165, 288–302.
23. K. Shimoda, H. Sugaya, M. Murakami, H. Arai, Y. Uchimoto and Z. Ogumi, J. Electrochem. Soc., 2014, 161, A1012–A1018.
24. Z.-X. Chi, W. Zhang, X.-S. Wang, F.-Q. Cheng, J.-T. Chen, A.-M. Cao and L.-J. Wan, J. Mater. Chem. A, 2014, 2, 17359–17365.
25. J. Yang, J. Wang, Y. Tang, D. Wang, B. Xiao, X. Li, R. Li, G. Liang, T.-K. Sham and X. Sun, J. Mater. Chem. A, 2013, 1, 7306–7311.
26. B. Ding, G. Ji, Y. Ma, P. Xiao, L. Lu and J. Y. Lee, J. Power Sources, 2014, 247, 273–279.
27. H. Karami and F. Taala, J. Power Sources, 2011, 196, 6400–6411.
28. D. L. Smiley, L. J. Davis and G. R. Goward, J. Phys. Chem. C, 2013, 117, 24181–24188.
29. K. Weichert, W. Sigle, P. A. van Aken, J. Jamnik, C. Zhu, R. Amin, T. Acartuerk, U. Starke and J. Maier, J. Am. Chem. Soc., 2012, 134, 2988–2992.
30. J. F. Ni, L. J. Gao and L. Lu, J. Power Sources, 2013, 221, 35–41.
31. J. Ni, L. Zhang, S. Fu, S. Savilov, S. Aldoshin and L. Lu, Carbon, 2015, 92, 15–25.
32. J. Wang, J. Yang, Y. Zhang, Y. Li, Y. Tang, M. N. Banis, X. Li, G. Liang, R. Li and X. Sun, Adv. Funct. Mater., 2013, 23, 806–814.
33. J. Wang, Y. Tang, J. Yang, R. Li, G. Liang and X. Sun, J. Power Sources, 2013, 238, 454–463.
34. J. Wang, J. Yang, Y. Tang, J. Liu, Y. Zhang, G. Liang, M. Gauthier, Y. K. Chen-Wiegart, M. Norouzi Banis, X. Li, R. Li, J. Wang, T. K. Sham and X. Sun, Nat. Commun., 2014, 5, 3415.

---

Footnote

† Electronic supplementary information (ESI) available: Materials synthesis and characterization, electrochemical evaluation, and evolution of LiFePO₄. See DOI: 10.1039/c5ra04744g

---