Metathetical reactions in the system Na(K)PF6 - LiBF4 - Aprotic media

Article abstract of DOI:10.1002/zaac.200500111

¹⁹F NMR spectroscopy, X-ray powder diffraction, elemental analysis, and ab initio quantum chemical calculations were used to study metathetical reactions between potassium or sodium hexafluorophosphate and lithium tetrafluoroborate in a mixture of propylene carbonate (PC) - dimethyl carbonate (DMC). It was shown that the increase in size of the cations in the second coordination sphere from Na⁺ to K⁺ results in an increase of the equilibrium conversion. This is in agreement with the influence of the cation size on the solubility of tetrafluoroborates in the media investigated.

Full text of DOI:10.1002/zaac.200500111

Metathetical Reactions in the System Na(K)PF6 ؊ LiBF4 ؊ Aprotic Media
Andriy V. Plakhotnyka,*, Lidiya D. Tarasovaa, Ludger Ernstb, and Reinhard Schmutzlerb
a Dniepropetrovsk/Ukraine, Department of Chemistry and Environmental Engineering, Dniepropetrovsk National University of Railway
Transport
b Braunschweig/Germany, Institut für Anorganische und Analytische Chemie der Technischen Universität
Received March 9th, 2005.
Dedicated to Professor Erhard Kemnitz
Abstract. 19F NMR spectroscopy, X-ray powder diffraction, ele-
mental analysis, and ab initio quantum chemical calculations were
used to study metathetical reactions between potassium or sodium
hexafluorophosphate and lithium tetrafluoroborate in a mixture of
propylene carbonate (PC) Ϫ dimethyl carbonate (DMC). It was
shown that the increase in size of the cations in the second coordi-
nation sphere from Naϩ to Kϩ results in an increase of the equilib-
rium conversion. This is in agreement with the influence of
the cation size on the solubility of tetrafluoroborates in the media
investigated.
Keywords: Metathetical reactions; Tetrafluoroborates; Hexafluoro-
phosphates; 19F NMR spectroscopy; Ab initio calculations
Table 1 Thermodynamic parameters for reactions (1) and (2).
Introduction
Lithium batteries and super-capacitors are nowadays more
and more widely used [1, 2]. This requires the development
and introduction of new advanced methods of producing
their components. Solutions of lithium hexafluorophos-
phate in mixtures of carbonate esters are used most widely
as electrolytes for rechargeable lithium batteries [3Ϫ5]. At
the same time, the production of battery grade LiPF6 is
complicated by difficulties arising from its high hygro-
scopicity and low thermal stability [6Ϫ8]. Traditional meth-
ods of producing the electrolytes involve dissolving LiPF6
in dehydrated aprotic organic solvents. Metathetical reac-
tions between lithium tetrafluoroborate (LiBF4) and alkali
metal (Na or K) hexafluorophosphates in aprotic media
could be used as alternative methods of preparation.
The thermal stability of the three fluoro complex salts
used as starting materials is higher than that of LiPF6
[9Ϫ12]. In this account, exchange reactions (metathetical
reactions) between KPF6 (or NaPF6) and LiBF4 in aprotic
solvents are described. The reaction mixtures were homo-
geneous at the beginning of the process. Equilibrium con-
versions obtained experimentally are compared with the
calculated data.
ΔH0298/(kcal/mol)
ΔG0298/(kcal/mol)
Dielectric
constant (ε)
Medium
KPF6
NaPF6
KPF6
NaPF6
Gas phase
PC
DMC
PCϩDMC
(1:1)
1
64.4
3.8
6.84
Ϫ5.76
Ϫ1.41
Ϫ5.52
4.39
Ϫ2.51
Ϫ0.44
Ϫ2.39
5.90
Ϫ5.97
Ϫ1.60
Ϫ5.72
4.02
Ϫ2.64
Ϫ0.58
Ϫ2.52
34.1
equilibria are shifted to the side of the initial reactant or to
the final product. The nature of the reaction media and, in
particular, their dielectric constants (ε), are essential. En-
thalpies and Gibbs energies for the reactions (1) and (2)
were calculated (see Table 1).
KPF6 ϩ LiBF4 ϭ KBF4ȇ ϩ LiPF6
NaPF6 ϩ LiBF4 ϭ NaBF4ȇ ϩ LiPF6
(1)
(2)
Table 1 presents the thermodynamic parameters of reac-
tions, calculated for the gas phase and for the aprotic me-
dia, PC, DMC, and their mixtures (1 : 1 by volume).
Ab initio quantum chemical calculations of the thermo-
chemical parameters (particularly enthalpy, ΔH0
and
298
Results and Discussion
Gibbs free energy, ΔG0298) have been performed, based on
the restricted Hartree-Fock method. A 6-31g basis set [13]
with the Gaussian 2003 (G03W) program package [14], for
the gas and liquid phases, were employed. The calculations
were performed for the presence of a solvent (a continuum
model was used), based on an Onsager model [15, 16]. In
the calculations, values of dielectric constants from [17]
were used. The dielectric constant of the mixture of solvents
was calculated, according to the additivity rule,
It was of interest to calculate thermodynamic parameters
of the metathetical reactions discussed to find out if the
* Dr. A. V. Plakhotnyk
Department of Chemistry and Environmental Engineering
Dniepropetrovsk National University of Railway Transport
Academician Lazaryan Str., 2
49010 Dniepropetrovsk/Ukraine
Fax: ϩϩ 38 Ϫ (0)562 Ϫ 47-19-65
e-mail: andriyϪpl@mail.ru
ε(PCϩDMC) ϭ (εPC ϩ εDMC)/2.
1840
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
DOI: 10.1002/zaac.200500111
Z. Anorg. Allg. Chem. 2005, 631, 1840Ϫ1842

Metathetical Reactions in the System Na(K)PF6 Ϫ LiBF4 Ϫ Aprotic Media
Figure 1 376 MHz 19F NMR spectrum of a stoichiometric mixture
of NaPF6 and LiBF4 in PC/DMC (1:1 by volume). Salt concentra-
tions 0.4 mol lϪ1. Signals a, b, and c are assigned to [PF6]Ϫ,
[PO2F2]Ϫ and [BF4]Ϫ, respectively.
The data given in Table 1 prove that the equilibria of both
reaction (1) and (2) in the gas phase are shifted towards
the side of the initial reactants, and that the equilibrium
conversions are 0.5Ϫ3.0 %. In aprotic media the situation
is opposite, and the equilibria are shifted towards the prod-
uct side. The increase in size of the cations in the second
coordination sphere, from Naϩ to Kϩ results in an in-
creased conversion.
The precipitates started to form a few seconds after the
homogeneous solutions of the hexafluorophosphates in PC
and lithium tetrafluoroborate in DMC were mixed. The
precipitates were identified as NaBF4 or KBF4 by chemical
analysis, and X-ray powder diffraction. For instance, at
room temperature the equilibrium was reached within ap-
proximately 5 min, and at 0 °C in about 20 min. The com-
position of the systems was, however, analyzed 1Ϫ2 hours
after mixing to ensure that the equilibrium had been re-
ached.
Figure 2 Dependence of the conversion on the initial molar ratio
of reagents (%): I Ϫ conversion calculated from thermodynamic
parameters (Table I), II Ϫ conversion calculated from solubility
product data; III Ϫ experimental data (19F NMR). (a) KPF6
LiBF4 system; (b) NaPF6 ϩ LiBF4 system.
ϩ
(θ) of potassium (sodium) hexafluorophosphate and lithium
tetrafluoroborate. Curves I (Fig. 2a and Fig. 2b) were calcu-
lated from the Gibbs energies ΔG0 in Table 1, according
298
to the following equations:
ΔG0 ϭ ϪRTlnK and K ϭ X2 / (1-X)·(θ-X)
298
In Figure 1 is presented the 19F NMR spectrum of the
system at equilibrium, comprising a stoichiometric mixture
of NaPF6 and LiBF4 in a PCϪDMC mixture (1:1 by vol-
ume). Salt concentrations were 0.4 mol lϪ1. The most in-
tense doublet (a) with δF ϭ Ϫ72.6 [J(P,F) ϭ 708 Hz] be-
longs to [PF6]Ϫ. The weak doublet (b) with δF ϭ Ϫ84.0
[J(P,F) ϭ 980 Hz] was assigned to an impurity of [PO2F2]Ϫ
[18]. It is difficult to avoid the formation of this product
during the synthesis of the hexafluorophosphate. The di-
fluorophosphate content is 0.3Ϫ0.4 %, which is consistent
with the result of the chemical analysis of the sodium hexa-
fluorophosphate sample. The signal at δF ϭ Ϫ154.5 (c) is
expanded in the center of Figure 1. It is reliably assigned to
[BF4]Ϫ through its chemical shift, and the occurrence of two
signals in a 1:4 intensity ratio (10B/11B isotope effect upon
the 19F chemical shift) [19Ϫ21].
where K is the equilibrium constant. Curves II were calcu-
lated from the solubility products (SP) of potassium and
sodium hexafluorophosphate determined by us (complete
dissociation of the salts in the PCϪDMC mixture was as-
sumed). For stoichiometric mixtures of the reagents conver-
sions (X) were calculated as,
X (%) ϭ {1 Ϫ (√SP)/C}·100.
With excess hexafluorophosphate, conversions were calcu-
lated according to the formula,
X (%) ϭ {1ϪSP/(C2(θ-1))}·100.
where C is the initial concentration of the reactants. The
deviation between the data calculated by different methods
and the experimental data is a few % for the sodium salt,
and approximately 1 % for the potassium salt. Taking into
account the use of different methods, the agreement be-
Figure 2 shows the calculated (curves I and II) and the
experimental dependence, by 19F NMR spectroscopy (curve
III), of the metathetical conversions (X) on the molar ratio
Z. Anorg. Allg. Chem. 2005, 631, 1840Ϫ1842
zaac.wiley-vch.de
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
1841

A. V. Plakhotnyk, L. D. Tarasova, L. Ernst, R. Schmutzler
[5] A. V. Plakhotnyk, L. Ernst, V. N. Plakhotnyk, R. Schmutzler,
in: Symposium on Battery and Fuel Cell Materials; Graz, Aus-
tria, 2004, Abstracts p. 163.
[6] S. E. Sloop, J. K. Pugh, S. Wang, J. B. Kerr, K. Kinoshita,
Electrochem. Solid-State Lett. 2001, 4, A42; C. A. 2001, 135,
35149.
[7] K. S. Gavrichev, G. A. Sharpataya, L. N. Golushina, V. N.
Plakhotnyk, I. V. Goncharova, Zh. Neorg. Khim. 2002, 47,
1048; C. A. 2002, 137, 316582.
tween theoretical and experimental data is quite satisfac-
tory. The weaker influence of θ on the conversions found
from the NMR integrals (curves III), compared to the data
calculated from SP (curves II) is caused, most likely,
through partial cationϪanion association of [BF4]Ϫ and Kϩ
and, especially, Naϩ. High conversions in the reactions (1)
and (2), especially in the case of KPF6, make it possible to
obtain electrolytes by metathetical reactions in aprotic me-
dia.
[8] Y. Matsuda, N. Nakashima, M. Morita, J. Electrochem. Soc.
1981, 131, 2552.
Besides the described experiments, special experiments
with excess LiBF4 were carried out to obtain electrolytes
with a low concentration of Kϩ.
[9] A. S. Kana’an, J. M. Kanamueller, High Temp. Science 1979,
11, 23; C. A. 1979, 91, 163952.
[10] V. N. Plakhotnyk, A. A. Evtukh, K. S. Gavrichev, V. V. Rossi-
khin, in: Proceedings of The First International Siberian
Workshop on Advanced Inorganic Fluorides; Novosibirsk,
Russia, 2003, 195.
Experimental Section
[11] T. C. Ehlert, M. M. Hsia, J. Chem. Eng. Data 1972, 17, 17.
[12] V. N. Plakhotnyk, V. B. Tulchinskiy, Zh. Neorg. Khim. 1980,
25, 346; C. A. 1980, 92, 136301.
The fluoro complex salts used were of 99.2Ϫ99.5 % purity. KPF6
or NaPF6 were dissolved in PC, and LiBF4 was dissolved in DMC.
Appropriate volumes of the solutions were mixed. 19F NMR spec-
tra were recorded on a Bruker Avance DRX-400 spectrometer at
376 MHz. They were referenced to the frequency determined for
CCl3F in PC/DMC (1:1). The solutions to be studied were placed
in 5 mm o. d. tubes and acetone-d6 in a concentric capillary was
used for the field/frequency deuterium lock. Precipitates of KBF4
and NaBF4 formed during the process were filtered off through
finely meshed filters and were washed with diethyl ether. They were
identified by titrimetric (fluorine [22] and boron [23]), and X-ray
powder diffraction analyses. The solubility of sodium and potas-
sium fluoroborate in a PCϪDMC mixture (1:1 by volume) was de-
termined on samples containing 99.5Ϫ99.9 % of KBF4 or NaBF4,
respectively. The salts were added to the mixture of solvents in
hermetically sealed quartz flasks. The flasks were placed in a
thermostat TKϪ1 („Electronpribor“, Lviv, Ukraine), installed in-
side a dry box, and the mixtures were continuously stirred magneti-
cally during the reaction. PC (Aldrich, 99 %) and DMC (Aldrich,
99ϩ %) were used without purification. The water content was
checked by Karl Fischer titration. The determination of silicon,
believed to be present as fluorosilicate or fluorohydroxysilicate [24],
proved that there was no reaction between quartz and the fluoro
complex solutions in the aprotic media employed during the experi-
ments.
[13] R. Ditchfield, W. J. Hehre, J. A. Pople, J. Chem. Phys. 1971,
54, 724.
[14] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M.
A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tom-
asi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega,
G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toy-
ota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H.
P. Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gom-
perts, R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C.
Pomelli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A.
Voth, P. Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dap-
prich, A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick,
A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz,
Q. Cui, A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov,
G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin,
D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayak-
kara, M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen,
M. W. Wong, C. Gonzalez, J. A. Pople. Gaussian 03, Revision
B.03, Gaussian, Inc., Pittsburgh PA, 2003.
[15] L. Onsager, J. Amer. Chem. Soc. 1936, 58, 1486.
[16] M. W. Wong, K. B. Wiberg, and M. J. Frisch, J. Chem. Phys.
1991, 95, 8991.
Acknowledgements. Andriy V. Plakhotnyk is grateful to the Deut-
sche Forschungsgemeinschaft (DFG) for a grant during the tenure
of which this research was conducted. We also thank Professor M.
Bahadir (Braunschweig) for the analyses of sodium and potassium,
and for a helpful discussion, and Professor V. Rossikhin (Dniepro-
petrovsk) for the quantum chemical calculations.
[17] C. Laurence, P. Nicolet, M. T. Dalati, J.-L. M. Abboud, R.
Notario, J. Phys. Chem. 1994, 98, 5807.
[18] A. V. Plakhotnyk, L. Ernst, R. Schmutzler, J. Fluorine Chem.
2005, 126, 27.
[19] V. N. Plakhotnik, L. Ernst, P. Sakhaii, N. F. Tovmash, R.
Schmutzler, J. Fluorine Chem. 1999, 98, 133.
[20] V. N. Plakhotnyk, R. Schmutzler, L. Ernst, Yu. V. Kovtun, A.
V. Plakhotnyk, J. Fluorine Chem. 2002, 116, 41.
[21] Yu. A. Buslaev, V. V. Evsikov, M. N. Buslaeva, R. K. Mazitov,
V. N. Plakhotnik, Dokl. Akad. Nauk SSSR 1973, 213, 1349;
C. A. 1974, 80, 125426.
References
[1] I. A. Kedrinskiy, V. G. Yakovlev, Litiy-Ionnye Accumulatory
(Lithium Ion Accumulators), Platina, Krasnoyarsk, Russia,
2002, p. 233.
[22] I. G. Ryss, Chemistry of Fluorine and its Inorganic Derivatives,
Goskhimizdat, Moscow 1956, p. 591.
[2] J.-P. Wiaux, in: International Congress for Battery Recycling,
Abstracts of papers, Vienna, Austria, 2002, 1.
[3] M. Arrabito, N. Penazzi, S. Panero, P. Reale, J. Power Sources
2001, 94, 225; C. A. 2001, 134, 313565.
[23] I. M. Kolthoff, R. Belcher, V. A. Stenger, G. Matsuyama, Volu-
metric Analysis, vol. 2, Interscience Publishers, New York
1957, p. 142.
´
[24] G. Charlot, Les methodes de la chimie analytique: analyse
´
[4] M. K. Tasaki, K. Kanda, S. Nakamura, M. Ue, J. Electro-
chem. Soc. 2003, 150, A1628.
quantitative minerale, 4th ed., vol. II, Masson, Paris 1961, p.
844.
1842
© 2005 WILEY-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim
zaac.wiley-vch.de
Z. Anorg. Allg. Chem. 2005, 631, 1840Ϫ1842