1: \documentclass[twocolumn,prb,floatfix,amsmath]{revtex4}
2: \usepackage{graphicx}
3:
4: \begin{document}
5:
6: \title{ Lithiation of InSb and Cu$_2$Sb : A Theoretical Investigation }
7:
8: \author{S. Sharma}
9: \email{sangeeta.sharma@uni-graz.at}
10: \author{J. K. Dewhurst}
11: \author{C. Ambrosch-Draxl}
12: \affiliation{Institute for Theoretical Physics, Karl--Franzens--Universit\"at Graz,
13: Universit\"atsplatz 5, A--8010 Graz, Austria.}
14:
15: \date{\today}
16:
17: \begin{abstract}
18: In this work, the mechanism of Li insertion/intercalation in the anode materials InSb and Cu$_2$Sb is
19: investigated by means of the first-principles total-energy calculations. The electron localization
20: function for the lithiated products of InSb are presented. Based on these results the change in the
21: bonding character on lithiation is discussed. Further, the isomer shift for InSb and Cu$_2$Sb and thier
22: various lithiated products is reported. The average insertion/intercalation voltage and the volume
23: expansion for transitions from InSb to Li$_3$Sb and Cu$_2$Sb to Li$_3$Sb are calculated
24: and found to be in good agreement with the experimental values. These findings help to resolve the
25: uncertainity regarding the lithiation mechanism in InSb.
26: \end{abstract}
27:
28: \maketitle
29:
30:
31: \section{INTRODUCTION}
32:
33: Intermetallic compounds present an attractive alternative to graphite as anode materials in Li ion
34: insertion batteries due in particular to the high capacity, an acceptable rate capability and operating
35: potentials well above the potential of metallic lithium.
36: The work in this direction gained tempo after the announcement that tin-based compounds
37: \cite{idota97,mao99} could operate effectively as alloying hosts for Li ions. In search of new anode
38: materials one of the desirable qualities is that the structural and volumetric
39: distortion of the host anode material on lithiation are small. This could be achieved if there existed
40: a strong structural relation\cite{vaughey00,fransson01,kropf01} between the host and its lithiated products.
41: Some of the compounds which show such structural relations are Cu$_6$Sn$_5$
42: \cite{nordstrom01,benedek02,sharma03}, InSb\cite{vaughey00,johnson00,kropf01,hewitt01,vaughey01,tostmann02},
43: Cu$_2$Sb \cite{fransson01}, MnSb, Mn$_2$Sb\cite{fransson03}, SnSb \cite{li99}, GaSb\cite{vaughey01} etc.
44: These compounds can be divided into three main structure types, namely
45: NiAs (Cu$_6$Sn$_5$, MnSb), ZnS (InSb, GaSb) and Cu$_2$Sb (Mn$_2$Sb).
46: In these materials, one of the components (Cu, In, Mn, Ga) is less active and is extruded on lithiation,
47: while the other component provides the structural skeleton for further lithiation.
48: In a previous work,\cite{sharma03} the lithiation of $\eta'$-Cu$_6$Sn$_5$ was studied. In the present work,
49: we investigate one representative compound from each of the other two structure types. These are InSb
50: which belongs to the zinc blende structure and Cu$_2$Sb which is the prototype for the third class of
51: materials.
52:
53: Since it is desirable for an ideal anode material to reversibly insert/intercalate Li ions while
54: maintaining the structural stability, the knowledge of the exact amount of Li accommodated within the
55: structure before the extrusion of the less active component is extremely important. In this regard,
56: recently the electrochemistry of the Li insertion in InSb was studied experimentally, and contradictory
57: results were published:
58: 1) Vaughey {\it et al.}\cite{vaughey00} initially proposed that 2 Li atoms could be incorporated
59: within the InSb structure with very small volume expansion (5.6\%) before In extrusion, resulting in
60: excellent capacity
61: %
62: 2) Johnson {\it et al.}, \cite{johnson00} subsequently reported that In extrusion occurred at an earlier
63: stage according to a general reaction:
64: $(x+y){\rm Li + InSb } \rightarrow {\rm Li}_{x+y}{\rm In}_{1-y}{\rm Sb}
65: (0 \le x \le 2; 0 \le y \le 1)$ in which a
66: Li$_y$In$_{1-y}$Sb zinc-blende framework played an important role in the Li insertion/In extrusion
67: process, but they could not determine the precise amount of Li that could be inserted before the onset
68: of In extrusion;
69: %
70: 3) On the other hand, Hewitt {\it et al.} \cite{hewitt01} claimed, that $x$ reached a maximum value
71: of 0.27 before In atoms were extruded to immediately form Li$_3$Sb in a two-phase reaction, and that
72: Li$_2$Sb could be generated on subsequent electrochemical cycling;
73: %
74: 4) From in situ X-ray absorption measurements, Kropf {\it et al.} \cite{kropf01} and
75: Tostmann {\it et al.}\cite{tostmann02} provided supporting evidence that Li$_{x+y}$In$_{1-y}$Sb
76: electrode compositions were generated during the initial discharge of Li/InSb cells and that 40\% of
77: the extruded In was not incorporated back into the Sb lattice during the subsequent charging of the cells.
78: In the light of the uncertainty of the reaction process, there is a need to clarify the exact lithiation
79: path. To this extent we perform total-energy calculations for studying the mechanism of Li
80: insertion/intercalation in InSb. A similar pathway is also provided for Cu$_2$Sb. We also present the
81: theoretically determined isomer shift (IS) for these materials and various lithiated products.
82:
83: The change in the bonding character on lithiation is another important information that can
84: be extracted from the first-principles calculations. Such analysis regarding battery materials,
85: has been done in the past using the total\cite{sharma03} or difference\cite{ven98,aydinol97}
86: charge density. But it was noted by Bader {\it et al.}\cite{bader84} and Becke {\it et al.}
87: \cite{becke90} that the information regarding the bonding in the molecular systems is not fully
88: contained in the density ($\rho$) but rather in its Laplacian ($\nabla^2 \rho$) and in the kinetic
89: energy density. These quantities are used to calculate the electron-localization-function (ELF),
90: \cite{becke90} which varies between 0 and 1, with 1 corresponding to perfect localization and 1/2
91: corresponding to electron gas like behaviour. Thus ELF in a way quantifies the bonding between various
92: atoms in a solid and hence is ideal for studying the change in the bonding nature as new bonds are formed
93: and (or) old bonds are weakened or broken by the insertion of Li atoms. We present the ELF for some of
94: the lithiated products of InSb in the present work.
95:
96:
97: % arrangement
98: The paper is arranged in the following manner. In Section II, we present the details of the calculations.
99: Section IIIA deals with the energetics of lithiating InSb and Cu$_2$Sb. To facilitate the study of the
100: change in
101: the bonding character on the lithiation of these compounds the IS and the ELF are given in Section IIIB.
102: Section IV provides a summary of our work.
103:
104:
105: \section{METHODOLOGY}
106:
107: Total energy calculations are performed using the full potential linearized augmented planewave (FPLAPW)
108: method. In this method, the unit cell is partitioned into non-overlapping muffin-tin spheres
109: around the atomic sites and an interstitial region. In these two types of regions different basis sets
110: are used. The scalar relativistic Kohn-Sham equations are solved in a self-consistent scheme.
111: For the exchange-correlation potential we use the local density approximation (LDA). All the calculations
112: are converged in terms of basis functions as well as in the size of the $k$-point mesh representing the
113: Brillouin zone.
114:
115: % relax, wien:
116: In order to compare the total energies, calculated from first principles, for any two structures one has
117: to optimize the structural parameters. These parameters are the atomic positions and the unit cell volume,
118: which are varied until the global energy minimum is found. Thereby the experimental crystalline data is
119: used as a starting point. For all the energy differences reported in the present work these optimized
120: crystal structures are used. For these calculations the existing {\sf WIEN2k} code\cite{WIEN} is used.
121:
122: %Is:
123: The details for the calculation of the IS of $^{121}$Sb have been presented before. \cite{sharma02}
124: In the present work, the same value for the nuclear specific parameter ($\alpha = -0.846 \pm 0,002 a_0^3$
125: mm/s) is used to convert the contact charge densities to the IS. The effect of the muffin-tin sphere
126: size on the IS has also been studied previously.\cite{sharma02} It was found that the
127: IS is not affected by the choice of the muffin-tin radius. A value of 2.0a$_0$ has been used in the
128: present work for all the atoms.
129:
130: %ELF
131: The details for the ELF can be found in Ref. \onlinecite{becke90}. These calculations are performed
132: using the newly developed {\sf EXCITING } code. Calculation of ELF in a mixed basis like LAPW requires
133: the kinetic energy density to be made smooth across the muffin-tin sphere. Thus these calculations are
134: performed using super-LAPW, \cite{singh91} which requires the matching of the wave function and its
135: first and second derivatives at the muffin-tin sphere boundary making calculations computationally
136: more demanding but affordable because the size of the unit cells is relatively small.
137:
138: \section {RESULTS AND DISCUSSION}
139:
140: \subsection{Lithiation of InSb and Cu$_2$Sb}
141:
142: InSb exists in the zinc blende structure with In atoms at (0,0,0) and the Sb atoms occupying four
143: tetrahedral sites: (0.25,0.25,0.25),
144: (0.75,0.25,0.75), (0.75,0.75,0.25) and (0.25,0.75,0.75). This kind of structural arrangement leaves
145: four tetrahedral sites vacant namely (0.75,0.25,0.25), (0.25,0.25,0.75), (0.25,0.75,0.25) and
146: (0.75,0.75,0.75). Along with this there is another void in the structure at (0.5,0.5,0.5).
147: On lithiation of InSb (or other zinc blende compounds), there are two distinct probable processes:
148: 1) The incoming Li ions go to one of these voids.
149: 2) A substitutional reaction takes place, i.e., In (or the less active component) is extruded and
150: is replaced by the incoming Li ions. In light of the existing controversy regarding these processes in
151: InSb, we studied various possible lithiation paths. Using first-principles total-energy
152: calculations we determine the formation energies for various lithiated products of InSb using
153: the following formula:
154: \begin{equation}
155: \rm{yLi + Li_xInSb \rightarrow Li_{x+y}In_{1-y}Sb + yIn}
156: \end{equation}
157: The formation energy for the compound Li$_{\rm{x+y}}$In$_{\rm{1-y}}$Sb is given as
158: \begin{equation}
159: \Delta \rm{ E = E_{Li_{x+y}In_{1-y}Sb} + y E_{In}^{met} - [ E_{Li_xInSb}+ y E_{Li}^{met} ]}.
160: \end{equation}
161: Here E$_C$ is the total energy of the
162: compound $C$ and E$_A^{\rm{met}}$ is the energy per atom for metallic $A$. The formation energy defined
163: this way reflects the relative stability of the compound Li$_{\rm{x+y}}$In$_{\rm{1-y}}$Sb
164: with respect to the reactants Li and Li$_{\rm y}$InSb.\cite{ven98}
165:
166: \begin{figure}[ht]
167: \centerline{\includegraphics[width=0.85\columnwidth,angle=-90]{Form_InSb.ps}}
168: \caption {The formation energies in eV per formula unit. The compounds formed from the
169: insertion/intercalation of Li atom in the previous compounds are represented by circles and compounds
170: formed by the substitutional reaction are represented by triangles. }
171: \end{figure}
172: %
173:
174: \begin{figure}[ht]
175: \centerline{\includegraphics[width=0.9\columnwidth,angle=-90]{Triangle_InSb.ps}}
176: \caption {(a) A part of Li-In-Sb phase diagram. (b) Close up of the energetically favourable InSb-Li3Sb
177: part of the phase diagram}
178: \end{figure}
179:
180: These formation energies for InSb and its lithiated products are presented in Fig. 1.
181: The compounds formed via the insertion/intercalation reaction of Li with the previous compound are shown
182: as circles, while the compounds formed via the substitutional reaction are marked by triangles.
183: As can be seen the formation energies of certain compounds are negative, {\it i.e.} these compounds
184: are stable with respect to the reactants. A part of the Li-In-Sb phase diagram is shown in Fig. 2.
185: Fig. 2(a) consists of the Gibb's trangle which consists of the energetically stable compounds
186: in transition from InSb to Li$_3$Sb, which are
187: shown in detail in Fig 2(b). We find that insertion of 1/2 Li
188: atoms per InSb formula unit leads to a volume expansion of 5.3\%. In atoms are extruded beyond this
189: concentration of Li, leading to very little further volume expansion. These results show best agreement
190: with the experimental conclusions of Tostmann {\it et al.}\cite{tostmann02} The maximum number of
191: intercalated Li atoms before In extrusion as reported by Hewitt {\it et al.}\cite{hewitt01} is almost
192: half the value obtained in the present work.
193: According to the energetics, the compound Li$_2$InSb is unstable, but if formed leads to an enormous
194: volume expansion of 15.4\%. The total theoretical volume expansion on going from InSb to Li$_3$Sb is
195: 5.6\% which is slightly more than the experimentally reported value of 4.4\%. It can be noted from
196: Fig. 1 that, like Hewitt{\it et al.},\cite{hewitt01} we find the formation of
197: Li$_2$Sb during the first cycle to be energetically hindered. But we also note that the cubic phase of
198: this compound is lower in energy than the hexagonal phase proposed earlier.\cite{hewitt01}
199:
200: The initial incoming Li atoms have the freedom to occupy either
201: the site (0.75,0.25,0.25) or (0.5,0.5,0.5) in the InSb framework to form Li$_{0.25}$InSb.
202: We find that the probability for Li insertion at the position (0.75,0.25,0.25) is higher than at
203: (0.5,0.5,0.5). The difference in energy between the two structures, Li$_{0.25}$InSb
204: with Li at (0.75,0.25,0.25) and Li$_{0.25}$InSb with Li at (0.5,0.5,0.5), is 0.05 eV per formula unit.
205: Our calculated average intercalation voltage (AIV) for a transition from InSb to Li$_3$Sb is 0.83V,
206: which is higher than the experimental value of the plateau in the voltage profile
207: (0.6V \cite{tostmann02} and 0.7V \cite{hewitt01}).
208:
209: Another important thing to be noted from Fig. 1 is the difference in the formation energies
210: between the two consecutive compounds. After the formation of Li$_{5/4}$In$_{3/4}$Sb it is energetically
211: less demanding for Li atoms to come in and (or) for In atoms to extrude from the structure.
212: Also in the case of the reverse reaction it is energetically more demanding for In atoms to reincorporate
213: in the structure beyond a concentration of 75\%. This is in accordance with the
214: experimental findings of Tostmann {\it et al.},\cite{tostmann02} that only about 60\% of the In atoms
215: go back into the structure on charging and that Li atoms are extracted more easily and rapidly from this
216: In deficient phase of the kind $\rm{Li_{0.6+x}In_{0.4}Sb}$ leading to a good cycling capacity.
217:
218: \begin{figure}[ht]
219: \centerline{\includegraphics[width=0.85\columnwidth,angle=-90]{Form_Cu2Sb.ps}}
220: \caption {The formation energies in eV per formula unit. The compounds formed from the
221: insertion/intercalation of Li atom in the previous compounds are represented by circles, and compounds
222: formed by the substitutional reaction are represented by triangles. }
223: \end{figure}
224:
225: \begin{figure}[ht]
226: \centerline{\includegraphics[width=0.9\columnwidth,angle=-90]{Triangle_Cu2Sb.ps}}
227: \caption {(a) A part of Li-Cu$_2$-Sb phase diagram. (b) Close up of the energetically favourable
228: Cu$_2$Sb-Li3Sb part of the phase diagram}
229: \end{figure}
230:
231:
232: Cu$_2$Sb exists in a tetragonal structure with space group P4/nmm and
233: 2 formulae units per unit cell. The Cu atoms occupy the sites (0.75,0.25,0), (0.25,0.75,0.0),
234: (0.25,0.25,0.27) and (0.75,0.75,0.73) while the Sb atoms are situated at (0.25,0.25,0.70) and
235: (0.75,0.75,0.3). There exist voids in the structure at (0.5,0.5,0.5), (0.25,0.75,0.5) and (0.75,0.25,0.5).
236: The two reaction mechanisms investigated in the present are,
237: 1) the insertion/intercalation of Li atoms, and
238: 2) the substitution of Cu by Li atoms.
239: The formation energies for these reaction paths are presented in Fig. 3. Similar to the previous case,
240: the Cu atoms are extruded from the structure after insertion of 1/2 Li atoms per Cu$_2$Sb formula unit.
241: The first incoming Li atoms occupies the site (0.25,0.75,0.5) in the unit cell. Further lithiation leads
242: to replacement of the Cu atom at site (0.75,0.25,0.0) by a Li atom to form LiCu$_{3/2}$Sb.
243: The optimized structural parameters for LiCu$_{3/2}$Sb show a volume expansion of 18\% per formula unit
244: compared to Cu$_2$Sb. More incoming Li atoms leads to the formation of the Li$_2$CuSb and a free Cu atom.
245: Li$_2$CuSb exists in a cubic structure with space group F$\overline {4}$3m. In this unit cell the Sb atoms are present
246: at the origin, the face centers and four of the tetrahedral sites i.e. (0.75,0.75,0.75), (0.25,0.25,0.75),
247: (0.75,0.25,0.25) and (0.25,0.75,0.25) occupied by the Cu atoms. The Li atoms are located at the other
248: four tetrahedral sites and at (0.5,0.5,0.5).
249: The theoretically optimized structure of Li$_2$CuSb shows a volume expansion of 26.4\% per formula unit
250: compared to the initial compound Cu$_2$Sb. This is in good agreement with the experimental value of
251: 25.2\%.\cite{fransson01}
252:
253: Energetically favourable Cu$_2$Sb to Li$_3$Sb part of the phase diagram is shown in Fig. 4.
254: The calculated value of the AIV for going from Cu$_2$Sb
255: to Li$_3$Sb is 0.97 V, which is slightly higher than the experimental value of 0.82 V.\cite{fransson01}
256:
257: \subsection{Isomer shift and electron localization function}
258:
259:
260: \begin{figure}[ht]
261: \centerline{\includegraphics[width=0.85\columnwidth,angle=-90]{IS_InSb.ps}}
262: \caption {Isomer shift in mm/s for the compounds on the energetically favourable lithiation reaction paths
263: for InSb.}
264: \end{figure}
265:
266:
267: An excellent method of studying the change in the local environment and symmetry is the M\"ossbauer IS.
268: We have calculated the $^{121}$Sb IS of various lithiated products of InSb and the results are presented
269: in Fig. 5. The calculated value for InSb is in good agreement with the experimental data of -8.6 mm/s.
270: \cite{pruitt70} Insertion reaction of the Li atoms with the host makes the IS steadily more negative with
271: a sudden increase when the In is extruded from the structure. The IS for the energetically hindered
272: compounds LiInSb and Li$_2$InSb are -10.54 mm/s and -11.54 mm/s respectively. These values are largely
273: negative as compared to the other lithiated products of InSb making them easily identifiable by M\"ossbauer
274: spectroscopy.
275:
276: \begin{figure}[ht]
277: \centerline{\includegraphics[width=0.85\columnwidth,angle=-90]{IS_Cu2Sb.ps}}
278: \caption {Isomer shift in mm/s for the compounds on the energetically favourable lithiation reaction paths
279: for Cu$_2$Sb.}
280: \end{figure}
281:
282: It can be seen from Fig. 6, the IS for Cu$_2$Sb, like InSb, shows a decrease on lithiation via the
283: insertion reaction and the continuous increase when the Cu extrusion starts. The IS has been shown to
284: depend linearly on the partial number of $s$ electrons (N$_s$)\cite{sharma02} and the variation of the
285: IS with the Li concentration can be explained very well using this fact. As more Li atoms
286: insert/intercalate in the host material, a part of the charge lost by them is gained by the Sb atoms.
287: This causes an increase in N$_s$ and hence a more negative IS. On the other hand, on lithiation via
288: the substitution reaction there is no excess charge.
289: This causes the decrease in N$_s$ and hence the IS becomes less negative. Thus the change in the IS with
290: the Li atom concentration is a good signature of the reaction mechanism and hence we suggest such
291: experiments to be performed to shed more light on the reaction path and for the unambiguous clarification
292: of whether compounds like LiInSb and Li$_2$InSb are formed or not.
293:
294: \begin{figure}[ht]
295: \centerline{\includegraphics[width=0.85\columnwidth,angle=-90]{ELF.ps}}
296: \caption {Electron localization function for InSb, Li$_{0.5}$Sb and Li$_3$Sb.}
297: \end{figure}
298:
299: The change in the bonding character on lithiation of the anode material can be studied using the
300: charge density plots which have been used for Cu$_6$Sn$_5$ \cite{sharma03} and Cu$_2$Sb
301: \cite{sharmaPP}. In the present work we use ELF for such an investigation of InSb. Fig. 8 shows the
302: ELF for InSb, Li$_{0.5}$InSb and Li$_3$Sb along the (111) direction. A covalent character of bonding
303: between In atom at (0,0,0) and Sb atom at (1/4,1/4,1/4) can be seen as a strong localization of
304: charge between the two atoms. Since only the valence electrons participate in the bond formation,
305: the core
306: charge has been ignored for all the atoms. The high value of ELF at the Li atom site is due the core
307: states being taken into consideration as local orbitals. Lithiation of InSb to form Li$_{0.5}$InSb
308: leads to weakening of the bonding between In and Sb. This manifests itself by a lowering of the
309: ELF from 0.9 to 0.79 between the two atoms. The bond between Li and Sb atoms is ionic in nature.
310: This ionocity is retained all the way upto the formation of Li$_3$Sb. The presence of delocalized
311: electrons between atoms is indicative of the metallic nature of the compound.
312:
313:
314: \section{Summary}
315:
316: % energetics:
317:
318: To summarize, the two prototype structures, InSb and Cu$_2$Sb, for the use as anode materials in
319: Li-ion batteries are investigated in the present work. From first-principles total-energy calculations
320: we conclude the following:
321: \begin{itemize}
322: %
323: \item On lithiation of InSb during discharge, the In atoms are extruded from the structure after
324: insertion/intercalation of half a Li atom per InSb formula unit. The formation of Li$_{1/2}$InSb leads to
325: a volume expansion of 5.4\%
326: %
327: \item It is much easier for In and Li atoms to come out of and go into the structure after formation of
328: Li$_{5/4}$In$_{3/4}$Sb which supports the good cycling capacity.
329: %
330: \item The formation of the compounds LiInSb, Li$_2$InSb and Li$_2$Sb is energetically hindered.
331: In particular, the formation of LiInSb or Li$_2$InSb would lead to enormous volume expansions of 14.4\%
332: and 15.4\%, respectively, which is much more than any experimental observation.
333: %
334: \item Our calculated AIV for the transition from InSb to Li$_3$Sb is in agreement with the
335: experimental value of the plateau in the measured voltage profile. These findings make our results
336: closest to that of Tostmann {\it et al.}\cite{tostmann02}
337: %
338: \item In the case of Cu$_2$Sb, the Cu atoms are extruded after insertion of half a Li atom per Cu$_2$Sb
339: formula unit.
340: %
341: \item The trends in the IS with increasing Li concentration is an excellent signature of the
342: lithiation reaction mechanism. The formation of compounds like LiInSb and Li$_2$InSb can be unambiguously
343: determined using M\"ossbauer spectroscopy, since these compounds have large negative IS values compared
344: to any other lithiated product of InSb.
345: %
346: \item The ELF shows that lithiation of InSb results in the weakening of bonds between
347: In and Sb atoms which would lead, upon further lithiation, to an extrusion of In atoms. The bonding
348: changes from covalent for InSb to ionic in nature for Li$_3$Sb.
349:
350: \end{itemize}
351:
352:
353:
354: \noindent \textbf{Acknowledgements}
355:
356: We thank the Austrian Science Fund (project P16227) and the EXCITING network funded by the EU (Contract HPRN-CT-2002-00317)
357: for financial support.
358: SS would also like to thank K. Edstr\"om and M. M. Thackeray for useful discussions and suggestions during the course of this work.
359:
360:
361: %\bibliography{XSb}
362:
363: \begin{thebibliography}{24}
364: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
365: \expandafter\ifx\csname bibnamefont\endcsname\relax
366: \def\bibnamefont#1{#1}\fi
367: \expandafter\ifx\csname bibfnamefont\endcsname\relax
368: \def\bibfnamefont#1{#1}\fi
369: \expandafter\ifx\csname citenamefont\endcsname\relax
370: \def\citenamefont#1{#1}\fi
371: \expandafter\ifx\csname url\endcsname\relax
372: \def\url#1{\texttt{#1}}\fi
373: \expandafter\ifx\csname urlprefix\endcsname\relax\def\urlprefix{URL }\fi
374: \providecommand{\bibinfo}[2]{#2}
375: \providecommand{\eprint}[2][]{\url{#2}}
376:
377: \bibitem[{\citenamefont{Idota et~al.}(1997)\citenamefont{Idota, Kubota,
378: Matsufuji, Maekawa, and Miyasaka}}]{idota97}
379: \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Idota}},
380: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Kubota}},
381: \bibinfo{author}{\bibfnamefont{A.}~\bibnamefont{Matsufuji}},
382: \bibinfo{author}{\bibfnamefont{Y.}~\bibnamefont{Maekawa}}, \bibnamefont{and}
383: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Miyasaka}},
384: \bibinfo{journal}{Science} \textbf{\bibinfo{volume}{276}},
385: \bibinfo{pages}{570} (\bibinfo{year}{1997}).
386:
387: \bibitem[{\citenamefont{Mao et~al.}(1999)\citenamefont{Mao, Dunlap, and
388: Dahn}}]{mao99}
389: \bibinfo{author}{\bibfnamefont{O.}~\bibnamefont{Mao}},
390: \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Dunlap}}, \bibnamefont{and}
391: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Dahn}}, \bibinfo{journal}{J.
392: Electrochem. Soc.} \textbf{\bibinfo{volume}{146}}, \bibinfo{pages}{405}
393: (\bibinfo{year}{1999}).
394:
395: \bibitem[{\citenamefont{Vaughey et~al.}(2000)\citenamefont{Vaughey, O'Hara, and
396: Thackeray}}]{vaughey00}
397: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
398: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{O'Hara}}, \bibnamefont{and}
399: \bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Thackeray}},
400: \bibinfo{journal}{Electrochem. Solid-State Lett.}
401: \textbf{\bibinfo{volume}{3}}, \bibinfo{pages}{13} (\bibinfo{year}{2000}).
402:
403: \bibitem[{\citenamefont{Fransson et~al.}(2001)\citenamefont{Fransson, Vaughey,
404: Benedek, Edstr{\"o}m, Thomas, and Thackeray}}]{fransson01}
405: \bibinfo{author}{\bibfnamefont{L.~M.~L.} \bibnamefont{Fransson}},
406: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
407: \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Benedek}},
408: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edstr{\"o}m}},
409: \bibinfo{author}{\bibfnamefont{J.~O.} \bibnamefont{Thomas}},
410: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{M.~M.}
411: \bibnamefont{Thackeray}}, \bibinfo{journal}{Electrochem. Commun.}
412: \textbf{\bibinfo{volume}{3}}, \bibinfo{pages}{317} (\bibinfo{year}{2001}).
413:
414: \bibitem[{\citenamefont{Kropf et~al.}(2001)\citenamefont{Kropf, Tostmann,
415: Johnson, Vaughey, and Thackeray}}]{kropf01}
416: \bibinfo{author}{\bibfnamefont{A.~J.} \bibnamefont{Kropf}},
417: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Tostmann}},
418: \bibinfo{author}{\bibfnamefont{C.~S.} \bibnamefont{Johnson}},
419: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
420: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{M.~M.}
421: \bibnamefont{Thackeray}}, \bibinfo{journal}{Electrochem. Commun.}
422: \textbf{\bibinfo{volume}{3}}, \bibinfo{pages}{244} (\bibinfo{year}{2001}).
423:
424: \bibitem[{\citenamefont{Nordstr{\"o}m et~al.}(2001)\citenamefont{Nordstr{\"o}m,
425: Sharma, Sj{\"o}stedt, Fransson, H{\"a}ggstr{\"o}m, and
426: Edstr{\"o}m}}]{nordstrom01}
427: \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Nordstr{\"o}m}},
428: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Sharma}},
429: \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Sj{\"o}stedt}},
430: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Fransson}},
431: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{H{\"a}ggstr{\"o}m}},
432: \bibnamefont{and}
433: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edstr{\"o}m}},
434: \bibinfo{journal}{Hyperfine Interact.} \textbf{\bibinfo{volume}{136}},
435: \bibinfo{pages}{555} (\bibinfo{year}{2001}).
436:
437: \bibitem[{\citenamefont{Benedek and Thackeray}(2002)}]{benedek02}
438: \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Benedek}} \bibnamefont{and}
439: \bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Thackeray}},
440: \bibinfo{journal}{J. Power Sources} \textbf{\bibinfo{volume}{110}},
441: \bibinfo{pages}{406} (\bibinfo{year}{2002}).
442:
443: \bibitem[{\citenamefont{Sharma et~al.}(2003)\citenamefont{Sharma, Fransson,
444: Sj{\"o}stedt, Nordstr{\"o}sm, Johansson, and Edstr{\"o}m.}}]{sharma03}
445: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Sharma}},
446: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Fransson}},
447: \bibinfo{author}{\bibfnamefont{E.}~\bibnamefont{Sj{\"o}stedt}},
448: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Nordstr{\"o}sm}},
449: \bibinfo{author}{\bibfnamefont{B.}~\bibnamefont{Johansson}},
450: \bibnamefont{and}
451: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edstr{\"o}m.}},
452: \bibinfo{journal}{J. Electrochem. Soc. A} \textbf{\bibinfo{volume}{150}},
453: \bibinfo{pages}{330} (\bibinfo{year}{2003}).
454:
455: \bibitem[{\citenamefont{Johnson et~al.}(2000)\citenamefont{Johnson, Vaughey,
456: Thackeray, Sarakonsri, Hackney, Fransson, Edtr{\"o}m, and
457: Thomas}}]{johnson00}
458: \bibinfo{author}{\bibfnamefont{C.~S.} \bibnamefont{Johnson}},
459: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
460: \bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Thackeray}},
461: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Sarakonsri}},
462: \bibinfo{author}{\bibfnamefont{S.~A.} \bibnamefont{Hackney}},
463: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Fransson}},
464: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edtr{\"o}m}},
465: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{J.~O.}
466: \bibnamefont{Thomas}}, \bibinfo{journal}{Electrochem. Commun.}
467: \textbf{\bibinfo{volume}{2}}, \bibinfo{pages}{595} (\bibinfo{year}{2000}).
468:
469: \bibitem[{\citenamefont{Hewitt et~al.}(2001)\citenamefont{Hewitt, Beaulieu, and
470: Dahn}}]{hewitt01}
471: \bibinfo{author}{\bibfnamefont{K.~C.} \bibnamefont{Hewitt}},
472: \bibinfo{author}{\bibfnamefont{L.~Y.} \bibnamefont{Beaulieu}},
473: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{J.~R.} \bibnamefont{Dahn}},
474: \bibinfo{journal}{J. Electrochem. Soc. A} \textbf{\bibinfo{volume}{148}},
475: \bibinfo{pages}{402} (\bibinfo{year}{2001}).
476:
477: \bibitem[{\citenamefont{Vaughey et~al.}(2001)\citenamefont{Vaughey, Johnson,
478: Krof, Benedek, Thackeray, Tostmann, Sarakonsri, Hackney, Fransson,
479: Edstr{\"o}m et~al.}}]{vaughey01}
480: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
481: \bibinfo{author}{\bibfnamefont{C.~S.} \bibnamefont{Johnson}},
482: \bibinfo{author}{\bibfnamefont{A.~J.} \bibnamefont{Krof}},
483: \bibinfo{author}{\bibfnamefont{R.}~\bibnamefont{Benedek}},
484: \bibinfo{author}{\bibfnamefont{M.~M.} \bibnamefont{Thackeray}},
485: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Tostmann}},
486: \bibinfo{author}{\bibfnamefont{T.}~\bibnamefont{Sarakonsri}},
487: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Hackney}},
488: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Fransson}},
489: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edstr{\"o}m}},
490: \bibnamefont{et~al.}, \bibinfo{journal}{J. Power Sources}
491: \textbf{\bibinfo{volume}{97}}, \bibinfo{pages}{194} (\bibinfo{year}{2001}).
492:
493: \bibitem[{\citenamefont{Tostmann et~al.}(2002)\citenamefont{Tostmann, Kropf,
494: Johnson, Vaughey, and Thackeray}}]{tostmann02}
495: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Tostmann}},
496: \bibinfo{author}{\bibfnamefont{A.~J.} \bibnamefont{Kropf}},
497: \bibinfo{author}{\bibfnamefont{C.~S.} \bibnamefont{Johnson}},
498: \bibinfo{author}{\bibfnamefont{J.~T.} \bibnamefont{Vaughey}},
499: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{M.~M.}
500: \bibnamefont{Thackeray}}, \bibinfo{journal}{Phys. Rev. B}
501: \textbf{\bibinfo{volume}{66}}, \bibinfo{pages}{014106}
502: (\bibinfo{year}{2002}).
503:
504: \bibitem[{\citenamefont{Fransson et~al.}(2003)\citenamefont{Fransson, Vaughey,
505: Edstr{\"o}m, and Thackeray}}]{fransson03}
506: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Fransson}},
507: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Vaughey}},
508: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Edstr{\"o}m}},
509: \bibnamefont{and}
510: \bibinfo{author}{\bibfnamefont{M.}~\bibnamefont{Thackeray}},
511: \bibinfo{journal}{J. Electrochem. Soc. A} \textbf{\bibinfo{volume}{150}},
512: \bibinfo{pages}{86} (\bibinfo{year}{2003}).
513:
514: \bibitem[{\citenamefont{Li et~al.}(1999)\citenamefont{Li, Shi, Lu, Huang, and
515: Shen}}]{li99}
516: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Li}},
517: \bibinfo{author}{\bibfnamefont{L.~H.} \bibnamefont{Shi}},
518: \bibinfo{author}{\bibfnamefont{W.}~\bibnamefont{Lu}},
519: \bibinfo{author}{\bibfnamefont{X.~J.} \bibnamefont{Huang}}, \bibnamefont{and}
520: \bibinfo{author}{\bibfnamefont{L.~Q.} \bibnamefont{Shen}},
521: \bibinfo{journal}{J. Electrochem. Soc.} \textbf{\bibinfo{volume}{148}},
522: \bibinfo{pages}{915} (\bibinfo{year}{1999}).
523:
524: \bibitem[{\citenamefont{der Ven et~al.}(1998)\citenamefont{der Ven, Aydinol,
525: Ceder, Kresse, and Hafner}}]{ven98}
526: \bibinfo{author}{\bibfnamefont{A.~V.} \bibnamefont{der Ven}},
527: \bibinfo{author}{\bibfnamefont{M.~K.} \bibnamefont{Aydinol}},
528: \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Ceder}},
529: \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Kresse}}, \bibnamefont{and}
530: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Hafner}},
531: \bibinfo{journal}{Phys. Rev. B} \textbf{\bibinfo{volume}{58}},
532: \bibinfo{pages}{2975} (\bibinfo{year}{1998}).
533:
534: \bibitem[{\citenamefont{Aydinol et~al.}(1997)\citenamefont{Aydinol, Kohan,
535: Ceder, Cho, and Joannopoulos}}]{aydinol97}
536: \bibinfo{author}{\bibfnamefont{M.~K.} \bibnamefont{Aydinol}},
537: \bibinfo{author}{\bibfnamefont{A.~F.} \bibnamefont{Kohan}},
538: \bibinfo{author}{\bibfnamefont{G.}~\bibnamefont{Ceder}},
539: \bibinfo{author}{\bibfnamefont{K.}~\bibnamefont{Cho}}, \bibnamefont{and}
540: \bibinfo{author}{\bibfnamefont{J.}~\bibnamefont{Joannopoulos}},
541: \bibinfo{journal}{Phys. Rev. B} \textbf{\bibinfo{volume}{56}},
542: \bibinfo{pages}{1354} (\bibinfo{year}{1997}).
543:
544: \bibitem[{\citenamefont{Bader and Essen}(1984)}]{bader84}
545: \bibinfo{author}{\bibfnamefont{R.~F.~W.} \bibnamefont{Bader}} \bibnamefont{and}
546: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Essen}}, \bibinfo{journal}{J.
547: Chem. Phys.} \textbf{\bibinfo{volume}{80}}, \bibinfo{pages}{1943}
548: (\bibinfo{year}{1984}).
549:
550: \bibitem[{\citenamefont{Becke and Edgecombe}(1990)}]{becke90}
551: \bibinfo{author}{\bibfnamefont{A.~D.} \bibnamefont{Becke}} \bibnamefont{and}
552: \bibinfo{author}{\bibfnamefont{K.~E.} \bibnamefont{Edgecombe}},
553: \bibinfo{journal}{J. Chem. Phys.} \textbf{\bibinfo{volume}{92}},
554: \bibinfo{pages}{5397} (\bibinfo{year}{1990}).
555:
556: \bibitem[{WIE()}]{WIEN}
557: \bibinfo{note}{P. Blaha, K. Schwarz, G. K. H. Madsen, D. Kvasnicka and J.
558: Luitz, WIEN2k, Vienna University of Technology An Augmented Plane Wave +
559: Local Orbitals Program for Calculating Crystal Properties revised edition
560: 2001}.
561:
562: \bibitem[{\citenamefont{Sharma et~al.}(2002)\citenamefont{Sharma, Dewhurst,
563: Nordstr{\"o}m, and Johansson}}]{sharma02}
564: \bibinfo{author}{\bibfnamefont{S.}~\bibnamefont{Sharma}},
565: \bibinfo{author}{\bibfnamefont{J.~K.} \bibnamefont{Dewhurst}},
566: \bibinfo{author}{\bibfnamefont{L.}~\bibnamefont{Nordstr{\"o}m}},
567: \bibnamefont{and}
568: \bibinfo{author}{\bibfnamefont{B.}~\bibnamefont{Johansson}},
569: \bibinfo{journal}{J. Phys. Condens. Matter} \textbf{\bibinfo{volume}{14}},
570: \bibinfo{pages}{3537} (\bibinfo{year}{2002}).
571:
572: \bibitem[{\citenamefont{Singh and Krakauer}(1991)}]{singh91}
573: \bibinfo{author}{\bibfnamefont{D.}~\bibnamefont{Singh}} \bibnamefont{and}
574: \bibinfo{author}{\bibfnamefont{H.}~\bibnamefont{Krakauer}},
575: \bibinfo{journal}{Phys. Rev. B} \textbf{\bibinfo{volume}{43}},
576: \bibinfo{pages}{1441} (\bibinfo{year}{1991}).
577:
578: \bibitem[{hub()}]{huber79}
579: \bibinfo{note}{K. P. Huber and G. L. Herzberg, {\it Molecular Spectra and
580: Molecular Structure. IV. Constants of Diatomic Molecules} (Van Nostrand
581: Reinhold, New York, 1979)}.
582:
583: \bibitem[{\citenamefont{Pruitt et~al.}(1970)\citenamefont{Pruitt, Marshall, and
584: O'Donnell}}]{pruitt70}
585: \bibinfo{author}{\bibfnamefont{R.~A.} \bibnamefont{Pruitt}},
586: \bibinfo{author}{\bibfnamefont{S.~W.} \bibnamefont{Marshall}},
587: \bibnamefont{and} \bibinfo{author}{\bibfnamefont{C.~M.}
588: \bibnamefont{O'Donnell}}, \bibinfo{journal}{Phys. Rev. B}
589: \textbf{\bibinfo{volume}{2}}, \bibinfo{pages}{2383} (\bibinfo{year}{1970}).
590:
591: \bibitem[{sha()}]{sharmaPP}
592: \bibinfo{note}{L. Fransson, S. Sharma, F. Wagner, L. H{\"a}ggstr{\"o}m and K.
593: Edstr{\"o}m, (in Preprint)}.
594:
595: \end{thebibliography}
596: \end{document}
597:
598:
599:
600:
601:
602:
603: