Volatile single-source molecular precursor for the lithium ion battery cathode

Article abstract of DOI:10.1021/ja301112q

The first single-source molecular precursor for a lithium-manganese cathode material is reported. Heterometallic β-diketonate LiMn₂(thd) ₅ (1, thd = 2,2,6,6-tetramethyl-3,5-heptanedionate) was obtained in high yield by simple one-step solid-state reactions employing commercially available reagents. Substantial scale-up preparation of 1 was achieved using a solution approach. The crystal structure of the precursor contains discrete Li:Mn = 1:2 trinuclear molecules held together by bridging diketonate ligands. The complex is relatively stable in open air, highly volatile, and soluble in all common solvents. It was confirmed to retain its heterometallic structure in solutions of non-coordinating solvents. The heterometallic diketonate 1 was shown to exhibit clean, low-temperature decomposition in air/oxygen that results in nanosized particles of spinel-type oxide LiMn₂O₄, one of the leading cathode materials for lithium ion batteries.

Full text of DOI:10.1021/ja301112q

Communication
pubs.acs.org/JACS
Volatile Single-Source Molecular Precursor for the Lithium Ion
Battery Cathode
Anantharamulu Navulla, Lan Huynh, Zheng Wei, Alexander S. Filatov, and Evgeny V. Dikarev*
Department of Chemistry, State University of New York at Albany, Albany, New York 12222, United States
S
* Supporting Information
easily hydrolytically or thermally decomposable substances
usually referred to as molecular precursors.9 The interest in
ABSTRACT: The first single-source molecular precursor
for a lithium−manganese cathode material is reported.
Heterometallic β-diketonate LiMn2(thd)5 (1, thd =
2,2,6,6-tetramethyl-3,5-heptanedionate) was obtained in
high yield by simple one-step solid-state reactions
employing commercially available reagents. Substantial
scale-up preparation of 1 was achieved using a solution
approach. The crystal structure of the precursor contains
discrete Li:Mn = 1:2 trinuclear molecules held together by
bridging diketonate ligands. The complex is relatively
stable in open air, highly volatile, and soluble in all
common solvents. It was confirmed to retain its
heterometallic structure in solutions of non-coordinating
solvents. The heterometallic diketonate 1 was shown to
exhibit clean, low-temperature decomposition in air/
oxygen that results in nanosized particles of spinel-type
oxide LiMn2O4, one of the leading cathode materials for
lithium ion batteries.
materials that incorporate more than one type of metal atoms
has brought a great need for single-source precursors (SSPs)
molecules containing all the necessary elements in the proper
ratio and decomposable in a controllable manner under mild
conditions to yield phase-pure compounds.10 Once developed,
these heterometallic precursors may be exploited for thin-film
growth as well as for other applications such as preparation of
nanocrystals with an appropriate size that exhibit fundamental
modifications in electrochemical behavior of oxide materials.
The SSP approach to lithium−transition metal oxides is
relatively undeveloped. Only a handful of heterometallic
compounds have been reported recently,11,12 starting with the
pioneering work by Boyle.11 Notably, all of the previous reports
deal with nonvolatile Li−Co(Ni) compounds. In fact, most of
the known precursors are insoluble coordination polymers
suitable for bulk decomposition only. Since stable Li-containing
compounds with transition metals tend to crystallize as
coordination polymers or ionic frameworks, the preparation
of discrete heterometallic molecules with an appropriate Li:M
ratio and ligands that can be removed cleanly under relatively
mild conditions is a difficult task. Even more challenging is to
synthesize lithium-containing compounds that are volatile and
retain their heterometallic structures upon sublimation. To the
best of our knowledge, there are no reports on molecular
precursors for the LiMn2O4 spinel cathode material. While
several dozen Li−Mn heterometallic compounds are known,
none of those can be used as a SSP.
mong the various available storage technologies,1
A
rechargeable lithium ion batteries are the fastest growing
and the most promising systems, offering high energy density,
flexible lightweight design, and longer lifespan than comparable
battery technologies.2 One of the most important active
materials for the positive electrode of lithium ion batteries3 is
lithium−manganese oxide, LiMn2O4.4 This cathode material
has a spinel-type structure that can release Li+ ions upon
oxidation of its transition metal centers. Abundant Mn
resources, low toxicity, high electronic and lithium ion
conductivity, as well as an excellent rate capacity have made
LiMn2O4 a leading candidate for vehicle battery applications.5
Development of lithium ion batteries is largely attributed to
the growing demand for smaller, more sophisticated, and
portable devices.6 Thin-film batteries,7 composed of solid-state
materials that are only nanometers or micrometers thick,
represent an improvement of the common rechargeable lithium
ion battery technology in terms of safety, temperature stability,
flexibility, and charge density. Because of their small size and
unique features, thin-film batteries have a wide range of
applications as power sources for consumer products as well as
in medicine, semiconductor industry, oil drilling, and space
exploration.
The goal of our research was to create a Li−Mn
heterometallic compound that meets certain requirements for
an ideal SSP attractive to industrial applications:
(1) to contain discrete molecules with a proper 1:2 lithium to
manganese metal ratio;
(2) to be a volatile compound that sublimes congruently
(composition in the gas phase corresponds to that in the
solid state);
(3) to be soluble in common solvents and to retain the
heterometallic structure in solution;
(4) to be readily available via a simple reaction procedure on
a large scale with high yield;
(5) to have an overall low cost of preparation, preferably
through one-pot synthesis using commercially available
starting reagents or compounds that can be conveniently
obtained from commercial sources;
For thin-film batteries, a number of methods can be
employed to deposit a thin-film cathode material onto the
current collector.8 The majority of these techniques involve the
application of metal complexes with organic ligands, which are
Received: February 3, 2012
Published: March 23, 2012
© 2012 American Chemical Society
5762
dx.doi.org/10.1021/ja301112q | J. Am. Chem. Soc. 2012, 134, 5762−5765

Journal of the American Chemical Society
Communication
(6) to exhibit low moisture- and air-sensitivity, so it can be
handled outside the glovebox;
(7) to demonstrate clean, low-temperature decomposition
resublimed to yield microcrystalline material that is identical to
LiMn2(thd)5 (1) obtained by the solid-state/gas-phase
technique (Figure 1).
Heterometallic diketonate 1 in crystalline form was found to
be relatively stable in moist air and can be handled outside the
glovebox in a course of decomposition studies. Air-sensitivity of
the complex was tested by continuous recording of its powder
diffraction pattern (SI, Figure S1). The latter revealed that the
complex remains intact for at least 24 h, after which time some
structural changes started to occur. Heterometallic diketonate is
highly volatile at temperatures of ≥100 °C, while the
sublimation process is likely of a congruent character. Indirect
proof for the latter comes from two observations: (i) the
compound can be resublimed quantitatively, and (ii) the mass-
transfer of 1 in vapor phase can be achieved at temperatures as
low as 90 °C, at which one of its possible dissociation products,
Li(thd), is not volatile.
leading to the phase-pure LiMn2O4 oxide.
Herein we report preparation, characterization, and decom-
position study of the first volatile heterometallic molecular
precursor that has a proper Li:Mn ratio for the corresponding
lithium ion battery cathode material. We also discuss in detail
how the title compound fits the criteria for an ideal SSP.
Heterometallic β-diketonate LiMn2(thd)5 (1) (thd = 2,2,6,6-
tetramethyl-3,5-heptanedionate) has been obtained in a sealed
evacuated ampule at 110 °C by stoichiometric solid-state redox
reaction of reagents that are both commercially available:
2Li + 2Mn(thd) → LiMn (thd) + Li(thd)
(1)
3
2
5
Yellow crystals of 1 can be conveniently collected with ca.
70% yield after 5 days in the cold end of the ampule, where the
temperature was kept ca. 5 °C lower. The second product,
Li(thd), is volatile only after 160 °C and thus remains in the
hot end of the container during the course of reaction and
mass-transfer. Heterometallic diketonate can be additionally
recrystallized by sublimation at 90−100 °C, if needed. Purity of
the bulk product 1 has been confirmed by elemental analysis as
well as by X-ray powder diffraction through comparison of the
experimental spectrum with the theoretical pattern calculated
from the single-crystal data (Figure 1).
Single-crystal X-ray analysis (SI, Table S1) revealed that
heterometallic diketonate 1 contains discrete trinuclear
molecules LiMn2(thd)5 (Figure 2a) with a proper 1:2 metal
Figure 2. Molecular structure of LiMn2(thd)5 (1). (a) View of
trinuclear heterometallic diketonate complex drawn with thermal
ellipsoids at the 40% probability level. Metal−oxygen bonds to
diketonate ligands involved in bridging interactions are shown in blue.
Hydrogen atoms and tert-butyl groups are omitted for clarity. Selected
bond distances and angles are listed in SI, Table S2. (b) Space-filling
representation. The view is similar to that depicted in part (a), with all
tert-butyl groups added.
ratio for the desired decomposition product. The compound
crystallizes in the centrosymmetric triclinic unit cell with two
crystallographically independent molecules. One of those is
highly disordered, but it can be perfectly refined (even with
anisotropic displacement parameters for the core atoms) as a
superposition of two enantiomers (SI, Figure S2). The lithium
atom exhibits distorted tetrahedral coordination of four
diketonate oxygens. The manganese atoms maintain distorted
octahedral geometry that was previously observed13−16 in all
Mn-containing homo- and heterometallic β-diketonate com-
plexes. The molecular formula suggests that Li atom is in +1
and Mn atoms are in +2 oxidation states. Two of the five
diketonate groups are chelating to Mn atoms, the third group is
chelating to Li/bridging to Mn, and the remaining two ligands
are acting in an unusual (though not entirely unknown17)
manner of bridging metal atoms (none of those are located in
diketonate plane) through both of the oxygens (SI, Figure S3).
Sterically bulky thd ligands provide coordinative saturation of
metal atoms in complex 1 as well as protection of the trinuclear
unit, as can be seen from the space-filling diagram of
LiMn2(thd)5 (Figure 2b). The latter ensures the absence of
metal-based intermolecular contacts and prevents polymer-
ization of the metal units so typical for lithium compounds with
small monoanionic oxo-donor ligands. It also explains the
Figure 1. X-ray powder diffraction patterns of LiMn2(thd)5 (1) bulk
products synthesized by solution (top) and solid-state (middle)
methods and their comparison with the simulated pattern (bottom)
calculated from the single-crystal data.
An even better yield of 1 (85%) was obtained in another
solid-state reaction between lithium diketonate and anhydrous
manganese(II) chloride:
5Li(thd) + 2MnCl → LiMn (thd) + 4LiCl
(2)
2
2
5
While this reaction is a bit slower than reaction 1, it does not
involve the use of metallic lithium as a reactant and employs a
much cheaper Mn source. The second product, lithium
chloride, is not volatile and does not interfere with
heterometallic product 1 which is collected in the cold zone
of the ampule. The process also employs commercially available
reactants, though we used Li(thd) that was freshly prepared
from lithium hydroxide.
We confirmed that reactions 1 and 2 can also be run in
solution to provide a fast preparation of heterometallic
precursor 1 on a gram scale. In both procedures, the yellow
amorphous powder initially obtained can be either annealed or
5763
dx.doi.org/10.1021/ja301112q | J. Am. Chem. Soc. 2012, 134, 5762−5765

Journal of the American Chemical Society
Communication
relatively low air- and moisture-sensitivity of the heterometallic
complex.
In conclusion, the first molecular precursor for a lithium−
manganese cathode material reported in this work exhibits
several characteristics of an ideal single-source precursor. The
isolation of heterometallic diketonate LiMn2(thd)5 was a result
of a long, systematic search for appropriate ligands and starting
materials that allow one to obtain the target product with
discrete molecular structure and a proper metal ratio. The
complex is highly volatile, and as such it can be broadly used to
obtain thin films, nanocrystals, or single crystals of LiMn2O4
cathode material. The retention of heterometallic structure in
solution of non-coordinating solvents opens unique oppor-
tunities for application of the title precursor in direct liquid
injection chemical vapor deposition techniques for thin-film
growth. Synthetic approaches developed in this work for the
preparation of compound 1 should help to design a new
generation of volatile molecular precursors for the lithium
(sodium)−transition metal oxide, fluoride, and silicate cathode
materials.
Heterometallic diketonate 1 is readily soluble in all common
solvents, including hexanes. Solutions of 1 in non-coordinating
7
solvents are NMR (1H, 13C, and Li) silent, as we have already
noted for other manganese(II)-containing diketonates
MMn2(β-dik)6 (M = Mn,13 Pb,15 Cd,16 and Hg216) that also
exhibit discrete molecular structures. Importantly, the absence
of NMR signals in non-coordinating solvents (hexanes, toluene,
chloroform) indicates that the heterometallic molecules
LiMn2(thd)5 remain intact in these solutions. In contrast, in
coordinating solvents (DMSO, THF) the molecule dissociates
into Mn(thd)2(sol)2 (NMR silent) and Li(thd)(sol)2 adducts,
1
7
for which H and Li signals instantly appear in NMR spectra
(SI, Figures S4−S7). Additional information on the solution
structure can be obtained by comparison of the solid-state and
solution IR spectra of 1 that are identical for non-coordinating
solvents (SI, Figure S8) but different for coordinating ones.
According to the thermogravimetric analysis data, hetero-
metallic diketonate 1 exhibits clean, low-temperature decom-
position that occurs in a single step between 145 and 220 °C.
X-ray powder diffraction analysis of decomposition traces
confirmed the presence of a LiMn2O4 oxide phase. The weight
of the decomposition residue (18.0%) corresponds well with
the theoretical value (17.5%) that was calculated on the basis of
metal content, assuming it is pure LiMn2O4 oxide (SI, Figures
S9 and S10).
ASSOCIATED CONTENT
* Supporting Information
■
S
Full synthetic and characterization details, TGA plots and traces
of thermal decomposition, IR and NMR spectra, powder X-ray
diffraction patterns, additional details on interatomic distances
and angles in the structure of 1, and X-ray crystallographic file
in CIF format. This material is available free of charge via the
Internet at http://pubs.acs.org.
Thermal decomposition of LiMn2(thd)5 (1) precursor
obtained by both solid-state and solution methods has been
carried out at 600 °C in an oxygen atmosphere oven using high-
alumina crucibles. Analysis of decomposition products by X-ray
powder diffraction revealed the presence of a phase-pure
LiMn2O4 (Figure 3). Elemental analysis indicates no
AUTHOR INFORMATION
Corresponding Author
edikarev@albany.edu
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
Financial support from the National Science Foundation is
gratefully acknowledged.
■
REFERENCES
■
(1) (a) Liu, C.; Li, F.; Ma, L.-P.; Cheng, H.-M. Adv. Mater. 2010, 22,
E28−E62. (b) Arico, A. S.; Bruce, P.; Scrosati, B.; Tarascon, J.-M.; van
Schalkwijk, W. Nat. Mater. 2005, 4, 366−377.
(2) (a) Goodenough, J. B.; Kim, Y. Chem. Mater. 2010, 22, 587−603.
(b) Manthiram, A. J. Phys. Chem. Lett. 2011, 2, 176−184. (c) Kang, B.;
Ceder, G. Nature 2009, 458, 190−193. (d) High Energy Density
Lithium Batteries: Materials, Engineering, Applications; Wiley-VCH
Verlag GmbH & Co. KGaA: Berlin, 2010. (e) Palacin, R. M. Chem.
Soc. Rev. 2009, 38, 2565−2575.
(3) (a) Fergus, J. W. J. Power Sources 2010, 195, 939−954. (b) Ellis,
B. L.; Lee, K. T.; Nazar, L. F. Chem. Mater. 2010, 22, 691−714.
(4) (a) Kim, D. K.; Muralidharan, P.; Lee, H.-W.; Ruffo, R.; Yang, Y.;
Chan, C. K.; Peng, H.; Huggins, R. A.; Cui, Y. Nano Lett. 2008, 8,
3948−3952. (b) Lee, H.-W.; Muralidharan, P.; Ruffo, R.; Mari, C. M.;
Cui, Y.; Kim, D. K. Nano Lett. 2010, 10, 3852−3856. (c) Hosono, E.;
Kudo, T.; Honma, I.; Matsuda, H.; Zhou, H. Nano Lett. 2009, 9,
1045−1051. (d) Ding, Y.-L.; Zhao, X.-B.; Xie, J.; Cao, G.-S.; Zhu, T.-J.;
Yu, H.-M.; Sun, C.-Y. J. Mater. Chem. 2011, 21, 9475−9479. (e) Luo,
J.-Y.; Xiong, H.-M.; Xia, Y.-Y. J. Phys. Chem. C 2008, 112, 12051−
12057.
(5) (a) Eisler, M. Science Progress, May 23, 2011, http://
scienceprogress.org/2011/05/innovation-case-study-nanotechnology-
and-clean-energy/. (b) Bullis, K. Technology Review, Jan 12, 2011,
http://www.technologyreview.com/energy/27049/. (c) Energy Effi-
ciency and Renewable Energy, U.S. Department of Energy. June 2010,
Figure 3. X-ray powder diffraction patterns of LiMn2O4 spinel
obtained by decomposition at 600 °C of LiMn2(thd)5 (1) samples
prepared by solution (blue) and solid-state methods (annealing times
of 2 h (red) and 24 h (black)). The powder diffraction patterns
correspond to LiMn2O4 that is shown as a peak diagram.
appreciable carbon content in the oxide material. The
appearance of spinel phase can be detected after just a few
minutes. Initially, the diffraction peaks are noticeably broad, but
the crystallinity of the oxide can be significantly increased by
annealing the sample for an additional time. SEM images (SI,
Figure S11) confirmed that the oxide phase appears as
nanosized particles measured at 20−30 nm that grow to 80−
90 nm upon annealing.
5764
dx.doi.org/10.1021/ja301112q | J. Am. Chem. Soc. 2012, 134, 5762−5765

Journal of the American Chemical Society
Communication
http://www1.eere.energy.gov/vehiclesandfuels/pdfs/hybrid_elec_
sys_goals.pdf.
(6) Bruce, P. G.; Scrosati, B.; Tarascon, J.-M. Angew. Chem., Int. Ed.
2008, 47, 2930−2946.
(7) (a) Naoi, K.; Morita, M. Electrochem. Soc. Interface 2008, 17, 44−
48. (b) Patil, A.; Patil, V.; Shin, D. W.; Choi, J.-W.; Paik, D.-S.; Yoon,
S.-J. Mater. Res. Bull. 2008, 43, 1913−1942. (c) Hu, L.; Wu, H.;
Mantia, F. L.; Yang, Y.; Cui, Y. ACS Nano 2010, 4, 5843−5848.
(8) (a) Hwang, B.-J.; Wang, C.-Y.; Cheng, M.-Y.; Santhanam, R. J.
Phys. Chem. C 2009, 113, 11373−11380. (b) Chiu, K.-F.; Chen, C.-L.
Thin Solid Films 2011, 519, 4705−4708.
(9) (a) Chew, S. Y.; Patey, T. J.; Waser, O.; Ng, S. H.; Buchel, R.;
̈
́
Tricoli, A.; Krumeich, F.; Wang, J.; Liu, H. K.; Pratsinis, S. E.; Novak,
P. J. Power Sources 2009, 189, 449−453. (b) Park, B. G.; Kim, S.; Kim,
I.-D.; Park, Y. J. J. Mater. Sci. 2010, 45, 3947−3953.
(10) (a) Malik, M. A.; O’Brien, P. Chem. Vap. Deposition 2009, 15,
207−271. (b) Veith, M.; Bender, M.; Lehnert, T.; Zimmer, M.; Jakob,
A. Dalton Trans. 2011, 40, 1175−1182. (c) Condorelli, G. G.;
̀
Malandrino, G.; Fragala, I. L. Coord. Chem. Rev. 2007, 251, 1931−
1950. (d) Kessler, V. G. Chem. Commun. 2003, 39, 1213−1222.
(11) Boyle, T. J.; Rodriguez, M. A.; Ingersoll, D.; Headley, T. J.;
Bunge, S. D.; Pedrotty, D. M.; De’Angeli, S. M.; Vick, S. C.; Fan, H. A.
Chem. Mater. 2003, 15, 3903−3912.
(12) (a) Tey, S. L.; Reddy, M. V.; Subba Rao, V. G.; Chowdari, B. V.
R.; Yi, J.; Ding, J.; Vittal, J. J. Chem. Mater. 2006, 18, 1587−1594.
(b) Li, L.; Meyer, W. H.; Wegner, G.; Wohlfahrt-Mehrens, M. Adv.
Mater. 2005, 17, 984−988. (c) Bykov, M.; Emelina, A.; Kiskin, M.;
Sidorov, A.; Aleksandrov, G.; Bogomyakov, A.; Dobrokhotova, Z.;
Novotortsev, V.; Eremenko, I. Polyhedron 2009, 28, 3628−3634.
(d) Dobrokhotova, Z.; Emelina, A.; Sidorov, A.; Aleksandrov, G.;
Kiskin, M.; Koroteev, P.; Bykov, M.; Fazylbekov, M.; Bogomyakov, A.;
Novotortsev, V.; Eremenko, I. Polyhedron 2011, 30, 132−141.
(e) Khanderi, J.; Schneider, J. J. Eur. J. Inorg. Chem. 2010, 49,
4591−4594.
(13) Zhang, H.; Li, B.; Sun, J.; Clerac, R.; Dikarev, E. V. Inorg. Chem.
2008, 47, 10046−10052.
(14) (a) Shibata, S.; Onuma, S.; Inoue, H. Inorg. Chem. 1985, 24,
1723−1725. (b) Dikarev, E. V.; Zhang, H.; Li, B. J. Am. Chem. Soc.
2005, 127, 6156−6157.
(15) Zhang, H.; Yang, J.-H.; Shpanchenko, R. V.; Abakumov, A. M.;
Hadermann, J.; Clerac, R.; Dikarev, E. V. Inorg. Chem. 2009, 48, 8480−
8488.
(16) Zhang, H.; Li, B.; Dikarev, E. V. J. Cluster Sci. 2008, 19, 311−
321.
(17) (a) Seisenbaeva, G. A.; Gohil, S.; Kessler, V. G. J. Mater. Chem.
2004, 14, 3177−3190. (b) Darr, J. A.; Poliakoff, M.; Blake, A. J.; Li,
W.-S. Inorg. Chem. 1998, 37, 5491−5496. (c) Doppelt, P.; Baum, T.
H.; Ricard, L. Inorg. Chem. 1998, 37, 5491−5496. (d) Datta, S.; Baskar,
V.; Li, H.; Roesky, P. W. Eur. J. Inorg. Chem. 2007, 26, 4216−4220.
5765
dx.doi.org/10.1021/ja301112q | J. Am. Chem. Soc. 2012, 134, 5762−5765