'Naked fluoride ion' from elemental fluorine

Article abstract of DOI:10.1016/S0022-1139(99)00010-X

Tetrakis(dimethylamino)ethene 1 reacts with elemental fluorine to give a salt 3, which contains anhydrous fluoride ion as indicated by the ¹⁹F NMR chemical shift.

Full text of DOI:10.1016/S0022-1139(99)00010-X

Journal of Fluorine Chemistry 94 (1999) 213±215
`Naked ¯uoride ion' from elemental ¯uorine
Richard D. Chambers*, William K. Gray, Graham Sandford, Julian F.S. Vaughan
Department of Chemistry, University of Durham, South Road, Durham DH1 3LE, UK
Received 21 September 1998; accepted 1 February 1999
Abstract
Tetrakis(dimethylamino)ethene 1 reacts with elemental ¯uorine to give a salt 3, which contains anhydrous ¯uoride ion as indicated by the
19F NMR chemical shift. # 1999 Elsevier Science S.A. All rights reserved.
Keywords: Naked ¯uoride ion; Elemental ¯uorine; NMR
There has been considerable effort directed, over many
years, to the search for ¯uoride ion salts that are both soluble
in organic solvents and essentially anhydrous in order to
maintain the reactivity of the ¯uoride ion [1±5]. Various
approaches to this problem have been made, including the
synthesis of `anhydrous' tetramethylammonium ¯uoride,
where water was removed by preliminary displacement with
isopropyl alcohol, followed by removal of the alcohol. This
salt gave a high frequency chemical shift for ¯uoride at ꢀF
73 ppm in acetonitrile, which is the highest frequency
recorded for a ¯uoride salt [2]; chemical shifts recorded for
¯uoride are immensely variable e.g. ꢀF 73 to 146 ppm
[1] and are very sensitive to solvent and hydrogen bonding,
i.e. the chemical shift varies with the amount of moisture in
the solvent and the presence of other species, such as HF,
which may undergo rapid exchange.
Previously [6] we reported the novel salt 3, obtained by
reaction of tetrakis(dimethylamino)ethene (TDAE) 1 with
the per¯uorinated alkene 2, Scheme 1. The salt 3 can be
regarded as an analogue of hexamethylguanidinium ¯uoride
[7].
Salt 3 was partly soluble in acetonitrile and sulpholane,
and dF 110 ppm in d6-DMSO was recorded for 3. We have
now developed a direct route for the synthesis of 3 by
reaction of 1 with elemental ¯uorine in acetonitrile. A
simple and controlled reaction occurred giving a red solution
which gradually deposited 3 as a white solid, but the 19F
NMR spectrum of the acetonitrile solution showed ꢀF
130 ppm for 3, implying that the system was contaminated
with hydrogen ¯uoride. Consequently, we took precautions
to remove residual hydrogen ¯uoride from the system by
passing the ¯uorine over sodium ¯uoride heated to 1008C in
a copper tube (commercial ¯uorine typically contains up to
5% hydrogen ¯uoride (v/v)). Subsequent reaction of the
puri®ed ¯uorine with 1 gave an orange solution which
rapidly deposited 3 as a white solid, and passage of two
equivalents of ¯uorine gave quantitative conversion to 3.
Now, the NMR spectrum of the acetonitrile solution showed
ꢀF 73 ppm, i.e. identical with the value observed for
anhydrous tetramethylammonium ¯uoride [2], and there-
fore, we have a simple route to an anhydrous ¯uoride salt.
However, the sample of 3 made from puri®ed ¯uorine was
quite insoluble in aprotic solvents, although it was extremely
soluble in alcohol and water, and furthermore, the sample
was completely unreactive in various halogen exchange
reactions. This demonstrates again that there will usually
be a compromise necessary between one extreme, where the
¯uoride ion is largely unsolvated and will, consequently,
form a strong crystal lattice and another extreme where the
¯uoride ion is highly solvated forming a concentrated solu-
tion of largely unreactive ¯uoride ion e.g. as in water.
Therefore, we returned to the samples made from unpuri®ed
¯uorine (i.e. samples of 3 containing some hydrogen ¯uor-
ide) and discovered that these samples, suspended in sul-
pholane, perform effectively as a source of ¯uoride ion in a
series of transformations as shown in Scheme 2.
The salt 3 was characterised by a variety of means.
Addition of boron tri¯uoride etherate to a suspension of 3
in acetonitrile gave a clear solution of the tetra¯uoroborate
where the 19F NMR spectrum showed quite clearly the
tetra¯uoroborate ion, ꢀF 150 ppm. Furthermore, electro-
spray mass spectrometry gave relatively simple spectra with
*Corresponding author. Tel.: +44-191-374-3120; fax: +44-191-384-
4737; e-mail: r.d.chambers@durham.ac.uk
‡
m/z (ES ) at 199.98, [(Me2N)2C=C(NMe2)2]2‡, and m/z
0022-1139/99/$ ± see front matter # 1999 Elsevier Science S.A. All rights reserved.
P I I : S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 0 1 0 - X

214
R.D. Chambers et al. / Journal of Fluorine Chemistry 94 (1999) 213±215
Scheme 3.
centred at ꢀF 125.9 ppm, indicating different but close
crystallographic environments for the ¯uoride ion. It was
established previously that ꢀF for ¯uoride ion moves to
lower frequency in the solid state [8].
Elsewhere [9±11], we have described reactions where
elemental ¯uorine acts as an electrophile to form carbon±
¯uorine bonds, e.g. with diketones [9,10] and aromatic
compounds [11] and one of the interesting aspects of the
present result is the signi®cance for mechanism. TDAE 1 is
Scheme 1.
a very strong electron donor with a ®rst ionisation energy of
1
591 kJ mol
,
which is lower than that of zinc
(906 kJ mol 1) and approaches the values for lithium
(521 kJ mol 1) and sodium (497 kJ mol 1) [12]. It is, there-
fore, remarkable that a controlled reaction occurs between 1
and elemental ¯uorine without any special precautions
(apart from the usual dilution of ¯uorine in nitrogen and
the use of a largely inert solvent). In principle, we could
write alternative mechanistic schemes for reaction of ¯uor-
ine with a nucleophile/donor as shown in Scheme 3, where
the important initial step would involve either two-electron
(1) or one-electron (2) transfers, followed by (3) or (4), as
indicated and these possibilities seem to be extremely
dif®cult to differentiate.
However, if TDAE 1 reacted via single electron transfer
(2), then the ¯uorine atom would need to have a ®nite
lifetime for the difference between (1) and (2) to be mean-
ingful to the experimentalist. Then we would anticipate the
further reaction of ¯uorine atoms to be relatively indiscri-
minate, producing at least some hydrogen ¯uoride by
reaction with either the solvent or methyl groups in 1.
However, the high frequency chemical shift position of
the ¯uoride ion produced in the reaction of 3 with puri®ed
¯uorine essentially precludes this route because of the
absence of hydrogen ¯uoride. Therefore, we conclude that
the mechanism is that represented by Scheme 4.
Scheme 2.
(ES ) 86.83, BF4 . A solution of 3 in D2O clearly showed
the di-cation derived from 3, but the ¯uoride ion chemical
shift ꢀF ca. 130 ppm depended on concentration. However,
the solid-state 19F NMR spectra were more complicated and
depended on the method of preparation, although the 13C
NMR spectra were essentially identical (resonances at 41.0
and 42.2 ppm corresponding to the two magnetically inequi-
valent methyl groups, and at 155.9 ppm, the sp2 carbon
atoms were observed). For a sample of 3 derived from
reduction of ¯uorinated alkene 2, Scheme 2, several very
broad peaks were present, centred at 105 and 186 ppm,
indicating the presence of ¯uoride ion and hydrogen ¯uor-
ide. In contrast, however, the spectrum of 3 derived from
puri®ed ¯uorine had a relatively sharp group of peaks, 1:1:2,
Scheme 4.

R.D. Chambers et al. / Journal of Fluorine Chemistry 94 (1999) 213±215
215
[5] C. Carmona, G. Eaton, M.C.R. Symons, J. Chem. Soc., Chem.
Commun. (1987) 873.
Acknowledgements
[6] M.W. Briscoe, R.D. Chambers, S.J. Mullins, T. Nakamura, J.F.S.
Vaughan, F.G. Drakesmith, J. Chem. Soc., Perkin Trans. 1 (1994)
3115.
We thank Dr. S. Korn (Zeneca) for very helpful discus-
sions, Zeneca (studentship to WKG), EPSRC for a ROPA
award (to JFSV) and the Royal Society (University
Research Fellowship to GS) for funding.
[7] S.M. Igumnov, N.I. Delyagina, I.L. Knunyants, Bull. Chem. Soc.
USSR (1986) 1191.
[8] J.H. Clark, E.M. Goodman, D.K. Smith, S.J. Brown, J.M. Miller,
J. Chem. Soc., Chem. Commun. (1986) 657.
[9] R.D. Chambers, M.P. Greenhall, J. Hutchinson, Tetrahedron 52
(1996) 1.
References
[10] R.D. Chambers, J. Hutchinson, J. Thomson, J. Fluorine Chem. 78
(1996) 165.
[1] K.O. Christe, W.W. Wilson, J. Fluorine Chem. 46 (1990) 339.
[2] K.O. Christe, W.W. Wilson, R.D. Wilson, R. Bau, J. Feng, J. Am.
Chem. Soc. 112 (1990) 7619.
[11] R.D. Chambers, C.J. Skinner, J. Hutchinson, J. Thomson, J. Chem.
Soc., Perkin Trans. 1 (1996) 605.
[12] H. Bock, H. Borrmann, Z. Havlas, H. Oberhammer, K. Ruppert,
A. Simon, Angew. Chem., Int. Ed. Eng. 30 (1991) 1678.
[3] K.O. Christe, W.W. Wilson, J. Fluorine Chem. 47 (1990) 117.
[4] J.M. Miller, J.H. Clark, J. Chem. Soc., Chem. Commun. 22 (1982)
1318.