Elsevier

Solid State Ionics

Volume 270, February 2015, Pages 33-38
Solid State Ionics

Evidence for Al doping in lithium sublattice of LiFePO4

https://doi.org/10.1016/j.ssi.2014.12.004Get rights and content

Highlights

  • Al is incorporated into Li site, however in lower amount than expected.

  • Effective route for preparation of highly lithium-deficient olivine is presented.

  • Improved electrochemical activity of Al-doped, Li-deficient material was observed.

Abstract

In this work, we applied various measurement techniques in order to evaluate the possibility of the introduction of Al and its actual level at Li site in LiFePO4 positive electrode material for Li-ion batteries. Rietveld refinement of neutron diffraction data indicated that Al is incorporated into Li site, however, in lower amount than expected, and that lithium content is substantially depressed. The observed trends were qualitatively supported by results of ICP OES analysis. In addition, the oxygen content in Li1–3xAlxFePO4 was found to decrease with Al doping. The introduction of relatively small amount of dopant could be an effective route to produce single-phase, highly lithium-deficient lithium iron phosphate.

Spectroscopic methods (EXAFS and XANES) do not allow to unambiguously confirm the substitution; nevertheless, all the obtained results are compatible with the substitution taking place. Interestingly, electrochemical studies indicated improvement of the performance of a LiFePO4 cathode material upon partial substitution of lithium by aluminum, possibly as the effect of increase of concentration of lithium vacancies.

Introduction

Since the introduction of LiFePO4 as a cathode material for Li-ion batteries by Goodenough in 1997 [1], this material has been extensively studied. This amounted to an enormous number of published papers and resulted in commercial application in reversible cells [2]. The interest in phosphoolivine stems from its high reversible capacity (ca. 170 mAh g 1), high chemical stability and suitable voltage vs. lithium anode (ca. 3.5 V). In addition, it is an environmentally benign material with low cost precursors [3], [4], [5]. However, it turned out that electrochemically active material should be nanosized [6], [7], [8], preferably with an appropriate carbon coating [9] or other coating of the grains [10]. Also, it was shown by numerical simulations [11], [12] and by experiments [13] that the diffusion in LiFePO4 occurs preferentially along the [010] direction, while for other crystallographic directions diffusion is strongly limited. As a consequence, of such 1-dimensional diffusion, it can be understood that material with platelet-like grain shape with well-developed (010) surface should exhibit enhanced electrochemical performance [8]. Nevertheless, the electronic component of electrical conductivity of phosphoolivine (ca. 10 9 S cm 1 at room temperature [14]) remains the limiting factor, which must be overcome in order to obtain enhanced characteristics of a Li-ion battery utilizing LiFePO4-based cathode. While this issue can be solved by transition to nanoscale and incorporation of various additives to enhance macroscopic conductivity of the cathode, another approach was also proposed, i.e., doping of isovalent and aliovalent elements into lithium [15], [16], iron [17], [18], [19], [20], phosphorus [21] and oxygen [22] sublattices were undertaken. Such chemical modification, if successful in terms of improvement of electrochemical performance, could lower cost of manufacturing, as compared to that of nanostructured LiFePO4.

In the case of substitution of iron by other 3D metals, the results unambiguously reveal wide range of possible substitution [17], [18], [19], [20]. However, data obtained for Li site-substituted materials are not decisive, and the possibility of doping in this sublattice remains an open question [23], [24], [25], [26], [27], [28], [29], [30], [31], [32]. It is worth noting that data proving metallic-like properties of pure and (assuming) Li site-substituted LiFePO4 were published [24], [25]; however, these results were questioned, and the observed metallic-like conductivity of the samples was rationalized as originating from the presence of metallic phosphides and/or carbo-phosphides on the surface of LiFePO4 grains [33], [34]. As described in our earlier work [35], there is still an ongoing discussion about possibility and, assuming so, the range of substitution of lithium by other cations and its expected influence on the properties of the material. This includes also the so-called anti-site defect, where a nominal Li+ site is occupied by Fe2 + ion [12]. This possibility adds to the entire complexity of actual crystal structure of the LiFePO4 parent compounds and its derivatives.

In this particular study, we look at the Al doping at Li site effect. An experimental evidence supporting the possibility of the partial introduction of aluminum into Li positions in LiFePO4 structure is given. The presented discussion is based on ICP OES chemical analysis, neutron powder diffraction (NPD), EXAFS and XANES studies. In addition, electrochemical performance of Li/Li+/doped phosphoolivine cells is evaluated and compared to undoped material as well as nanostructured derivative.

Section snippets

Experimental

Compounds with nominal and assumed Li1–3xAlxFePO4 (x: 0, 0.001, 0.005, 0.01, 0.05 and 0.1) composition were prepared by a standard, high-temperature ceramic synthesis method. Due to the electroneutrality condition, the relative amount of lithium and aluminum was fixed as Li1–3xAlx, maintaining the total charge of the Li sublattice as + 1. This in turn should ensure that all iron cations remain at + 2 oxidation state in Li1–3xAlxFePO4. For the synthesis, respective amounts of Li2CO3, NH4H2PO4, FeC2

Results and discussion

X-ray diffraction patterns for the obtained samples pointed to single-phase materials; however, the high level of background resulting from iron fluorescence did not allow for conclusive evidence for determination of location of aluminum ions. In order to evaluate the actual chemical composition of Li1–3xAlxFePO4 samples, ICP OES measurements were performed. The content of Li in Li1–3xAlxFePO4 (x = 0, 0.005, 0.01, 0.05 and 0.1) (Table 1) roughly follows the assumed lithium deficiency in the

Conclusions

Neutron diffraction experiment provided the evidence of Al being incorporated into Li sites in Li1–3xAlxFePO4 phosphoolivine. It should be noticed that the introduction of relatively small amount of dopant could be an effective route to produce single-phase, highly lithium-deficient lithium iron phosphate. These results are compatible with EXAFS and XANES spectroscopy measurements; however, aluminum has almost indistinguishable effect on the spectra. Therefore, these results are not definitive.

Acknowledgments

This work was supported by the Polish-Swiss Research Programme under grant no. 080/2010 LiBeV (Positive Electrode Materials for Li-ion Batteries for Electric Vehicles).

This work was partly supported by the European Union, within the European Regional Development Fund, through the Innovative Economy grant (POIG.01.01.02-00-108/09).

The measurements performed at synchrotron have received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement n°

References (47)

  • A. Kulka et al.

    J. Power Sources

    (2013)
  • R. Amin et al.

    Solid State Ionics

    (2008)
  • A. Kulka et al.

    Solid State Ionics

    (2012)
  • A. Yamada et al.

    J. Power Sources

    (2003)
  • C.-W. Hu et al.

    Sources

    (2014)
  • M.R. Roberts et al.

    J. Power Sources

    (2008)
  • J. Hong et al.

    J. Power Sources

    (2006)
  • C. Delacourt et al.

    Solid State Ionics

    (2006)
  • P. Fischer et al.

    Phys. B Condens. Matter

    (2000)
  • G.X. Wang et al.

    J. Power Sources

    (2006)
  • M.-H. Tuilier et al.

    Thin Solid Films

    (2012)
  • A.K. Padhi et al.

    J. Electrochem. Soc.

    (1997)
  • B.L. Ellis et al.

    Chem. Mater.

    (2010)
  • J.B. Goodenough et al.

    J. Am. Chem. Soc.

    (2013)
  • J. Molenda

    Funct. Mater. Lett.

    (2011)
  • L.X. Yuan et al.

    Energy Environ. Sci.

    (2010)
  • B.M. Koltypin et al.

    J. Power Sources

    (2007)
  • W. Zajac et al.

    Funct. Mater. Lett.

    (2011)
  • H. Huang et al.

    Electrochem. Solid-State Lett.

    (2001)
  • B. Kang et al.

    Nature

    (2009)
  • D. Morgan et al.

    Electrochem. Solid-State Lett.

    (2004)
  • M.S. Islam et al.

    Chem. Mater.

    (2005)
  • R. Amin et al.

    Electrochem. Solid-State Lett.

    (2007)
  • Cited by (37)

    View all citing articles on Scopus

    This work was presented during the 4th Polish Forum Smart Energy Conversion and Storage, Krynica, Poland 1-4.10.2013.

    View full text