An NMR, IR and theoretical investigation of the methyl effect on conformational isomerism in 3-fluoro-3-methyl-2-butanone and 1-fluoro-3,3-dimethyl-2-butanone

Article abstract of DOI:10.1002/poc.471

The solvent dependence of the ¹H and ¹³C NMR spectra of 3-fluoro-3-methyl-2-butanone (FMB) and 1-fluoro-3,3-dimethyl-2-butanone (FDMB) was examined and the ⁴J_HF, ¹J_CF and ²J_CF couplings are reported. Density functional theory (DFT) at the B3LYP/6-311 ++ G(2df, 2p) level with ZPE (zero point energy) corrections was used to obtain the conformer geometries. In both FMB and FDMB, the DFT method gave only two minima for cis (F - C - C = O, 0°) and trans (F - C - C = O, 180°) rotamers. Assuming the cis and trans forms, the observed couplings in FMB when analysed by solvation theory gave the energy difference E_cis - E_trans of 3.80 kcal mol^-1 (1 kcal = 4.184 kJ) in the vapour phase (cf. the DFT value of 3.21 kcal mol^-1), decreasing to 2.6 kcal mol^-1 in CCl₄ and to 0.27 kcal mol^-1 in DMSO. In FDMB the observed couplings when analysed similarly by solvation theory gave E_cis - E_trans = 1.80 kcal mol^-1 in the vapour phase, decreasing to 0.47 kcal mol^-1 in CCl₄ and to - 1.25 kcal mol^-1 in DMSO. The introduction of a methyl group geminal to the fluorine atom shifts the conformational equilibrium towards the trans rotamer, in contrast to no significant effect when the methyl group is introduced at the α-carbon further from the fluorine atom. Copyright

Full text of DOI:10.1002/poc.471

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY
J. Phys. Org. Chem. 2002; 15: 211–217
DOI:10.1002/poc.471
An NMR, IR and theoretical investigation of the methyl
effect on conformational isomerism in 3-¯uoro-3-methyl-
2-butanone and 1-¯uoro-3,3-dimethyl-2-butanone
ClaÂudio F. Tormena,1 Roberto Rittner1* and Raymond J. Abraham2
1Instituto de QuõÂmica, Universidade Estadual de Campinas, Caixa Postal 6154, 13083-970 Campinas, SaÄo Paulo, Brazil
2Chemistry Department, University of Liverpool, P.O. Box 147, Liverpool L69 3BX, UK
Received 22 August 2001; revised 30 October 2001; accepted 2 November 2001
ABSTRACT: The solvent dependence of the 1H and 13C NMR spectra of 3-fluoro-3-methyl-2-butanone (FMB) and
1-fluoro-3,3-dimethyl-2-butanone (FDMB) was examined and the 4JHF, 1JCF and 2JCF couplings are reported. Density
functional theory (DFT) at the B3LYP/6–311 ‡‡ G(2df,2p) level with ZPE (zero point energy) corrections was used
to obtain the conformer geometries. In both FMB and FDMB, the DFT method gave only two minima for cis
(F—C—C=O, 0°) and trans (F—C—C=O, 180°) rotamers. Assuming the cis and trans forms, the observed
couplings in FMB when analysed by solvation theory gave the energy difference Ecis À Etrans of 3.80 kcal molÀ1
(1 kcal = 4.184 kJ) in the vapour phase (cf. the DFT value of 3.21 kcal molÀ1), decreasing to 2.6 kcal molÀ1 in CCl4
and to 0.27 kcal molÀ1 in DMSO. In FDMB the observed couplings when analysed similarly by solvation theory gave
Ecis À Etrans = 1.80 kcal molÀ1 in the vapour phase, decreasing to 0.47 kcal molÀ1 in CCl4 and to À1.25 kcal molÀ1 in
DMSO. The introduction of a methyl group geminal to the fluorine atom shifts the conformational equilibrium
towards the trans rotamer, in contrast to no significant effect when the methyl group is introduced at the a-carbon
further from the fluorine atom. Copyright  2002 John Wiley & Sons, Ltd.
KEYWORDS: conformational analysis; fluoroketones; NMR; solvation; theoretical calculations
INTRODUCTION
methyl group) shifts the conformational isomerism
significantly. In fluoroacetone (FA),4 the trans rotamer
predominates over the cis rotamer only in solvents of low
polarity (CCl4, CS2 and CDCl3), whereas for 3-fluoro-2-
butanone (FB)7 the trans rotamer predominates over the
cis rotamer in all solvents (CCl4 to DMSO). This beha-
viour is presumably due to the increased steric repulsion
between the two methyl groups in the gauche position
which destabilizes the cis rotamer in 3-fluoro-2-buta-
none.
The last three decades have seen tremendous interest in
fluorinated biological analogues and their pharmacologi-
cal properties.1 In determining the mode of action of
fluoro compounds in vivo, the conformational changes
induced by the fluorine substituent are of primary impor-
tance. These are primarily electronic as the steric effect of
fluorine is very small. An example is the interaction
between fluorine and the oxygen of a carbonyl group (i.e.
F—C—C=O) which has been widely investigated. This
Here we investigated the conformational isomerism in
the related compounds 3-fluoro-3-methyl-2-butanone
(FMB), where the remaining hydrogen atom in C-3 of
FB is replaced by a methyl group [Fig. 1(a)] and 1-fluoro-
3,3-dimethyl-2-butanone (FDMB), where all the hydro-
gen atoms of C-3 in FA are replaced by methyl groups
[Fig. 1(b)].
group has been shown to have a predominantly twofold
potential in fluoroacetic acid,2 fluoroacetyl chloride3 and
fluoroacetone4–6 in all of which the conformational
isomerism was shown to be between the cis and trans
forms. In these cases the cis form is more stable in
solvents of medium and high polarity.
It has also been observed7 that a small change in
structure (the replacement of a hydrogen atom by a
1
The H and 13C NMR spectra and the IR spectra of
FMB and FDMB in different solvents were obtained.
Both the 4JHF, 1JCF and 2JCF couplings and the IR spectra
are sensitive to the F—C—C=O orientation. The use of
DFT calculations plus solvation theory8 allows us both to
define the interconverting rotamers in FMB and FDMB
and to obtain the conformer energy differences in the
vapour phase and in solution.
*Correspondence to: R. Rittner, Instituto de Qu´ımica, Universidade
Estadual de Campinas, Caixa Postal 6154, 13083-970 Campinas, Sa˜o
Paulo, Brazil.
E-mail: rittner@iqm.unicamp.br
Contract/grant sponsor: FAPESP.
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

212
C. F. TORMENA, R. RITTNER AND R. J. ABRAHAM
the solute quadrupole) and a is the solute radius. The third
term is the dipole–dipole term for very polar media. The
solute radius is obtained directly from the molar volume
(VM) of the solute (VM/N = 4pa3/3, where N is Avoga-
dro’s number). The molar volume can be obtained from
the density of pure liquid, if known, or directly in the
program from additive atomic volumes. Similarly, the
solute refractive index may be inserted if known or
calculated directly from additive contributions.
For a molecule in state B a similar equation is obtained
differing only in the values of kB and hB. Subtraction of
the two equations gives the experimentally required
quantity DES (EA À EB ), the energy difference in any
S
S
solvent S of given relative permittivity, in terms of DEV
V
(EA À EBV) and calculable or measurable parameters.
The dipole and quadrupole moments of the molecules are
calculated directly from the partial atomic charges in the
molecule using the CHARGE routine.13 This theory has
been shown to give an accurate account of the solvent
dependence of a variety of conformational equi-
libria.4,7,11,12
Figure 1. 2a) Rotamers trans and cis for FMB; 2b) rotamers
trans and cis for FDMB
THEORY
Ab initio (DFT) calculations were performed using the
Gaussian 98 program.9 The potential energy surfaces
were obtained at the recommended B3LYP/6–31G(d,p)
level to find the most stable rotamers for the studied
molecules. The optimized geometries, which would be
used in solvation theory, were obtained at a higher level,
the B3LYP/6–311 ‡‡ G(2df,2p) level, and to determine
more accurate energies zero point energy (ZPE) correc-
tions10 were performed. However, the ZPE calculations
do not lead to changes in the geometry parameters. The
solvation calculations were performed using the
MODELS program, applying the geometries from
Gaussian as input file. This theory (solvation theory)
has been described fully elsewhere,4,7,8,11,12 so only a
brief description is given here. The solvation energy of a
molecule is given by including both the dipole and
quadrupole reaction fields and also a direct dipole–dipole
term to take account of the breakdown of the Onsager
reaction-field theory in very polar media.
RESULTS AND DISCUSSION
To our knowledge, there have been no previous
theoretical study of these molecules. The Gaussian
calculations gave for both FMB and FDMB two stable
rotamers, the cis and trans forms. The optimized
geometries and energies of these rotamers are given in
Table 1. The DFT-calculated dipole moments are for
FMB 0.94 D (trans) and 4.41 D (cis) and for FDMB 1.29
D (trans) and 4.28 D (cis). Using the DFT geometries, the
CHARGE routine13 gave dipole moments for FMB of
1.05 D (trans) and 4.70 D (cis) and for FDMB dipole
moments of 1.04 D (trans) and 4.42 D (cis). The DFT and
CHARGE dipole moments are in good agreement, hence
the partial atomic charges may be used with confidence in
the solvation calculations. The values of the solvation
Table 1. Calculated geometries and energies for FMB and
FDMB
On this basis, the solvation energy of any molecule in
state A, i.e. the difference between the energy in vapour
(EAV) and in any solvent (EAS) of relative permittivity ",
is given by the equation
FMB
cis
1.203
FDMB
Parameter
trans
1.208
trans
cis
˚
r(C
=
O) (A)
1.209
1.531
1.543
1.204
1.529
1.541
V
S
˚
EA À EA ˆ kAx=ꢀ1 À lx† ‡ 3hAx=ꢀ5 À x†
‡ bf ‰1 À expꢀÀbf =16RT†Š
r(C—C) (A)
1.522
1.542
1.526
1.541
˚
r(CF—CMe) (A)
r(C—F) (A)
r(C—H) (A)
ꢀ1†
˚
˚
1.411
1.092
1.393
1.091
121.2
1.390
1.091
122.9
1.375
1.090
122.6
2
2
where x = (" À 1)/(2" ‡ 1), l = 2(nD À 1)/(nD ‡ 2), b =
4.30(a3/2/r3)(kA ‡ 0.5hA)1/2 and f = [(" À 2)/(" ‡ 1)/"]1/2
for " >2 and is zero otherwise, nD is the refractive index
and T is the temperature (K). The first term is due to the
€(C—C=O)°
118.2
€(F—C—C)°
€(H—C—C)°
f(F—C—C=O)
109.0
109.6
180.0
107.4
110.5
0.00
116.2
109.2
180.0
111.0
111.0
0.00
Erel. (kcal molÀ1
)
0.00
0.94
3.20
4.41
0.00
1.29
1.70
4.28
solute dipole (kA = mA /a3, mA is the solute dipole) and the
2
Dipole moment (D)
second term to the solute quadrupole (hA = qA /a5, qA is
2
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

METHYL EFFECT ON CONFORMATIONAL ISOMERISM OF a-FLUOROKETONES
Table 2. Parameters for reaction-®eld calculations
213
Dipole moment (D) k (kcal molÀ1
)
h (kcal molÀ1
)
nD
VM
l
FMB
cis
trans
cis
4.70
1.05
4.42
1.04
7.3249
0.3642
5.5267
0.3072
0.7371
5.1805
1.6604
2.9080
1.3739
1.3739
1.3786
1.3786
109.89 0.4566
109.89 0.4566
128.68 0.4618
128.68 0.4618
FDMB
trans
Table 3. Chemical shifts 2ppm) and coupling constantsa 2Hz) for FMB
b
c
Solvent
H-3
H-4
C-2
C-3
C-1
C-4
3JHF
4JHF
1JCF
2JCF
2JCF
CCl4
2.20
2.28
2.82
2.24
2.22
2.21
2.22
1.40
1.45
2.02
1.43
1.41
1.42
1.40
207.7
210.6
209.4
210.4
210.0
210.4
209.5
97.8
98.7
98.9
99.1
99.3
99.3
98.4
23.7
24.5
24.0
24.6
24.4
24.4
24.3
23.7
23.9
23.8
24.0
24.0
23.3
23.4
21.17
21.49
21.38
25.54
21.46
21.67
21.72
5.20
5.06
5.05
4.95
4.87
4.78
4.59
179.4
178.5
178.4
178.2
178.0
177.4
177.8
29.9
24.0
24.0
24.2
24.2
24.2
24.3
23.9
CDCl3
29.7
29.2
29.2
28.9
28.5
28.1
Pure liquid
CD2Cl2
Acetone-d6
CD3CN
DMSO-d6
a
HF couplings accurate to Æ0.04 Hz and CF couplings accurate to Æ0.1 Hz.
b
c
F—C—C
=O.
F—C—CH3.
Table 4. Chemical shifts 2ppm) and coupling constantsa 2Hz) for FDMB
Solvent
H-2
H-4
C-2
C-1
C-3
C-4
2JHF
1JCF
2JCF
CCl4
4.88
5.12
5.13
5.60
5.28
5.21
5.36
1.12
1.16
1.16
1.56
1.16
1.13
1.09
206.7
209.4
209.3
209.1
209.3
209.9
209.4
82.2
82.2
82.6
82.4
83.0
83.1
82.4
42.4
42.7
42.8
42.6
42.9
42.7
41.7
25.5
25.8
26.0
25.8
26.0
25.8
25.4
47.71
47.11
47.11
47.40
47.41
47.11
46.50
186.0
183.6
181.7
180.5
179.3
178.2
176.7
14.0
12.6
12.5
12.3
12.3
11.8
11.5
CDCl3
CD2Cl2
Pure liquid
Acetone-d6
CD3CN
DMSO-d6
a
HF couplings accurate to Æ0.04 Hz and CF couplings accurate to Æ0.1 Hz.
parameters [Eqn. (1)] are given in Table 2. The refractive
index and molar volume were also calculated by the
program.
rotamers in the vapour phase, cis and trans. The NMR
data in Table 3 can now be combined with the solvation
calculations via the equation
The 1H and 13C NMR data (chemical shifts and
coupling constants) for FMB and FDMB are given in
Tables 3 and 4. These data can now be used with the
calculations given earlier to determine the conforma-
tional equilibria in these molecules. The couplings were
Jobs ˆ ncisJcis ‡ ntransJtrans
ncis ‡ ntrans ˆ 1
ncis=ntrans ˆ eÀÁE=RT
4
first shown to be linearly related. For FMB the JHF vs
2JCF plot is linear (correlation coefficient 0.98) and for
1
2
ÁE ˆ Ecis À Etrans
ꢀ2†
FDMB the JCF vs JCF plot is linear (correlation
coefficient 0.98). Thus, as in previous investigations,4,7,11
the change in coupling constants with solvent may be
reasonably attributed to changes in the rotamer popula-
tions.
where ncis and ntrans are the mole fractions of the cis and
2
trans rotamers. The value of the JCF in the pure liquid
(29.2 Hz) gives with the data in Table 3, an interpolated
value of 7.5 for the pure liquid relative permittivity.
The solvent data in Table 3 can now be used with Eqn.
(2) to search for the best solution for both the rotamer
energy difference and the values of Jcis and Jtrans. This
3-Fluoro-3-methyl-2-butanone
The Gaussian calculations show that there are two stable
gives DEV = 3.80 kcal molÀ1 2JCF(cis) = 25.3 Hz and
,
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

214
C. F. TORMENA, R. RITTNER AND R. J. ABRAHAM
Table 5. Conformer energy differences 2kcal molÀ1) and observed and calculated
couplings for FMB and FDMB
FMB
FDMB
2JCF (Hz)
2JCF (Hz)
Solvent
Ecis À Etrans
Calc.
Obs.
Ecis À Etrans
Calc.
Obs.
CCl4
2.59
1.77
1.28
0.77
0.42
0.27
1.40
29.7
29.6
29.3
28.9
28.3
28.1
29.4
29.9
29.7
29.2
28.9
28.3
28.1
29.2
0.47
À0.22
À0.62
À1.00
À1.19
À1.25
À0.91
14.1
12.9
12.3
12.0
11.8
11.7
12.0
14.0
12.6
12.5
12.3
11.8
11.5
12.3
CDCl3
CD2Cl2
Acetone-d6
CD3CN
DMSO-d6
Pure liquid
2JCF(trans) = 29.8 Hz, and the energy differences in the
solvents and couplings given in Table 5. The values of the
remaining couplings for the two rotamers may be
obtained from the linear relationships between the
The NMR data in Table 5 show that for FDMB the
trans rotamer is more stable than the cis rotamer in CCl4
only (DE = 0.47 kcal molÀ1), whereas in the other
solvents the cis form is the most stable. These results
are in agreement with the data from the FTIR spectra
(Fig. 3) in solvents of varying polarity. In CCl4 [Fig.
3(a)], two partially resolved carbonyl absorption bands,
bearing almost the same intensity, are observed. More-
over, the higher frequency band (which has been assigned
to the cis form)14 has the greater intensity in the
remaining solvents [Fig. 3(b)–(d)], indicating the pre-
dominance of the cis over the trans rotamer in these
solvents. It has been shown15 that the molar absorptivity
is different for each rotamer, so the IR data cannot be
used as a quantitative measure of the conformational
equilibrium, but it does support the above analysis.
The results obtained for FDMB and FA4 (Fig. 4) are
very similar, with a population inversion from the trans
to the cis rotamer in more polar solvents. FMB also
shows a similar behaviour to FB7 (Fig. 4), the trans
rotamer predominating in all solvents. Hence both pairs
of compounds show that the introduction of methyl
groups at the a-carbon, further from the fluorine atom,
does not lead to any significant effect in the rotational
equilibrium. However, the introduction of a methyl group
geminal to the fluorine atom (FB in comparison with FA
and FMB in comparison with FDMB) shifts the
equilibrium towards the trans rotamer even in more
polar solvents.
These observations indicate that there are steric
interactions in the cis rotamers between the acetyl methyl
group and one a-methyl group in FB and two methyl
groups in FMB. Moreover, there is steric repulsion
between the methyl group geminal to the fluorine atom
and the fluorine lone pairs, which now interact with the
carbonyl group orbitals destabilizing the cis rotamer,
shifting the equilibrium towards the trans rotamer. This
destabilizing effect is clearly shown by the relative
energies (Ecis À Etrans), in the vapour phase, which are
2.20 and 3.70 kcal molÀ1 for FA4 and FB7 and 1.80 and
3.80 kcal molÀ1 for FDMB and FMB.
4
observed couplings (Table 3), to give for the JHF
couplings, 5.19 Hz (trans) and 3.80 Hz (cis) and for 1JCF
179.0 Hz (trans) and 174.9 Hz (cis). The energy differ-
ence in the vapour phase compares well with that
calculated by Gaussian 98 [3.21 kcal molÀ1 (1 kcal =
4.184 kJ)].
The NMR data in Table 5 show that for FMB the trans
rotamer is the most stable form in all solvents, a
conclusion which is in agreement with the results from
the FTIR spectra (Fig. 2). In non-polar and medium-
polarity solvents [Fig. 2(a)–(d)] the carbonyl absortion is
a single sharp band, whereas in a polar solvent (CH3CN)
[Fig. 2(e)] there is a shoulder at a higher frequency,
which has been assigned to the cis form.14 The
predominance of the trans over the cis form is probably
due to the steric repulsion between the methyl groups in
the cis form.
1-Fluoro-3,3-dimethyl-2-butanone
The Gaussian 98 calculations show two stable rotamers in
the vapour phase, cis and trans. The value of the 2JCF in
the pure liquid (12.3 Hz) gives, with the data in Table 4,
an interpolated value of 16.3 for the pure liquid relative
permittivity. The NMR data in Table 4 can be combined
with Eqn. (2) to search for the best solution for both the
rotamer energy difference and the values of Jcis and Jtrans
.
This gives DEV = 1.80 kcal molÀ1, 2JCF(cis) = 11.3 Hz and
2JCF(trans) = 15.3 Hz, and the energy differences and
1
couplings given in Table 5. The values of the JCF
coupling in the two rotamers may be obtained from the
linear relationship between the observed couplings, to
1
1
give JCF(cis) = 176.2 Hz and JCF(trans) = 193.4 Hz. The
energy difference in vapour phase again compares very
well with that calculated by Gaussian 98 (1.70 kcal
molÀ1).
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

METHYL EFFECT ON CONFORMATIONAL ISOMERISM OF a-FLUOROKETONES
215
Figure 2. The carbonyl absorption band in the IR spectrum of FMB in 2a) hexane, 2b) CCl4, 2c) CHCl3, 2d) CH2Cl2 and 2e) CH3CN
The results of this work demonstrate that a geminal
methyl group has a significant effect on the conforma-
tional equilibria of these a-fluoroketones.
attributed to the steric effects between the two methyl
groups, which tend to be as far apart as possible, while
there are no significant steric effects in FDMB. More-
over, the conformational equilibrium displayed by FMB,
which has a methyl group geminal to a fluorine atom,
shows a marked resemblance to that of 3,3-difluoro-
butanone.7 The latter exhibits just one stable conformer,
the one with the methyl group cis to the carbonyl group,
leaving the two methyl groups (C-1 and C-4) in a trans
(or anti) arrangement.
CONCLUSION
NMR data, combined with solvation theory, provide a
consistent analysis of conformational isomerism in FMB
and FDMB in solvents of varying polarity. In FMB the
trans rotamer predominates over the cis rotamer in all
solvents, but in the FDMB the trans form prevails over
the cis form only in CCl4, whereas in more polar solvents
the cis form is more stable than the trans form.
It is noteworthy that FDMB presents a conformational
behaviour similar to that of a-fluoroacetone,4 for which
there is no possibility of methyl group steric interactions
from both sides of the carbonyl group.
The predominance of the trans rotamer in FMB can be
It is also remarkable that no steric effect from a methyl
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

216
C. F. TORMENA, R. RITTNER AND R. J. ABRAHAM
Figure 3. The carbonyl absorption band in the IR spectrum of FDMB in 2a) CCl4, 2b) CHCl3, 2c) CH2Cl2 and 2d) CH3CN
group was observed in fluoroesters12 and fluoroamides,11
where the replacement of hydrogen by a methyl group did
not lead to significant changes in their conformational
equilibria.
cmÀ3 solutions with a probe temperature of ca 25°C.
[2H12]Cyclohexane was used as the deuterium lock signal
1
for the CCl4 solution and pure liquid. The H and 13C
spectra were all referenced to internal Me4Si. Typical
conditions were proton spectra 48 transients, spectral
width 2500 Hz with 32K data points and zero filled to
128K to give a digital resolution of 0.04 Hz. Proton-
decoupled carbon spectra were obtained with typical
conditions 1024 transients, 3 ms pulse delay, spectral
width 18000 Hz with 64K data points and zero filled to
256K for 0.1 Hz digital resolution.
The spectra were all first order and the coupling
constants and chemical shifts were taken directly from
the spectra. The NMR data is presented in Tables 3 and 4.
The IR spectra were recorded with a Bomem MB 100
FTIR spectrometer, using a sodium chloride cell with
0.5 mm spacer for dilute (ca 0.03 M) solutions, with the
solvent as background when recording the solute
spectrum.
EXPERIMENTAL
The solvents were obtained commercially, stored over
molecular sieves and used without further purification.
1H and 13C NMR spectra were obtained on a Varian
Gemini spectrometer operating at 300.06 MHz for proton
and 75.45 MHz for carbon. Spectra were of ca 20 mg
3-Bromo-3-methyl-2-butanone16,17. 3-Methyl-2-buta-
none (40.2 g, 0.467 mol) in diethyl ether (100 ml) was
placed in a 250 ml three-necked flask equipped with a
condenser, addition funnel and magnetic stirrer. Bromine
(74.7 g, 0.467 mol) was added dropwise over a period of
60 min. The reaction mixture was washed with water
(4 Â 100 ml) and dried over MgSO4. The solvent was
Figure 4. Structural formulae of ¯uoroketones
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217

METHYL EFFECT ON CONFORMATIONAL ISOMERISM OF a-FLUOROKETONES
217
removed and the desired product was vacuum distilled
through a Vigreux column to give pure 3-bromo-3-
methyl-2-butanone (b.p. 64°C/50 mmHg), yield 49.4 g
(64.5%). 1H NMR (300 MHz, CDCl3, 20°C, TMS):
ꢀ = 2.4 (s, 3H, CH3), 1.8 (s, 6H, 2CH3). 13C NMR
(75 MHz, CDCl3, 20°C, TMS): ꢀ = 203.5, 63.7, 29.5,
24.1.
(s, 9H, tBu). 13C NMR (75 MHz, CDCl3, 20°C, TMS):
ꢀ = 209.4, 82.2, 42.7, 23.9.
Acknowledgements
The authors thank FAPESP for financial support and
for a fellowship (to C.F.T.) and CNPq for a fellowship
(to R.R.). Computer facilities (Gaussian 98) from
CENAPAD-SP are also gratefully acknowledged.
3-Fluoro-3-methyl-2-butanone17,18. The synthesis was
carried out in a 250 ml three-neck flask, equipped with
addition funnel, magnetic stirrer and a distillation column
and condenser to remove the FMB immediately after
its formation. 3-Bromo-3-methyl-2-butanone (74.0 g,
0.461 mol) was reacted with potassium bifluoride
(52.8 g, 0.676 mol) in dry diethylene glicol (150 ml) at
170°C. The low-boiling product immediately distilled
over. The NMR spectrum of the distillate showed the
correct pattern for FMB and indicated the presence of
some impurity. Redistilation gave pure FMB, b.p. 88°C,
REFERENCES
1. (a) CIBA Symposium, Carbon–Fluorine Compounds, Chemistry,
Biochemistry and Biological Activities. Association of Scientific
Publishers: Amsterdam, 1972; (b) Taylor NF (ed). Fluorinated
Carbohydrates, Chemical and Biochemical Aspects. ACS Sym-
posium Series 374. American Chemical Society: Washington, DC,
1988.
2. van Eijck BP, van der Plaats G, van Roon PH. J. Mol. Struct. 1972;
11: 67–91.
3. Durig JR, Phan HV, Hardin JA, Little TS. J. Chem. Phys. 1989; 90:
6840–6851.
4. Abraham RJ, Jones AD, Warne MA, Rittner R, Tormena CF. J.
Chem. Soc., Perkin Trans. 2 1996; 533–539.
5. Saegebarth E, Krisher LC. J. Chem. Phys. 1970; 52: 3555–3562.
6. Durig JR, Hardin JA, Phan HV, Little TS. Spectrochim. Acta, Part
A 1989; 45: 1239–1255.
7. Abraham RJ, Tormena CF, Rittner R. J. Chem. Soc., Perkin Trans.
2 1999; 1663–1667.
8. Abraham RJ, Bretschneider E. In Internal Rotation in Molecules.
Orville-Thomas WJ (ed), Academic Press: London, 1974; chapt.
13.
9. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Zakrzewski VG, Montgomery JA, Stratmann RE,
Burant JC, Dapprich S, Millam JM, Daniels AD, Kudin KN, Strain
MC, Farkas O, Tomasi J, Barone V, Cossi M, Cammi R, Mennucci
B, Pomeli C, Adamo C, Clifford S, Ochterski J, Petersson GA,
Ayala PY, Cui Q, Morokuma K, Malick DK, Rabuck AD,
Raghavachari K, Foresman JB, Cioslowski J, Ortiz JV, Stefanov
BB, Liu G, Liashenko A, Piskorz P, Komaromi I, Gomperts R,
Martin RL, Fox DJ, Keith T, Al-Laham MA, Peng CY,
Nanayakkara A, Gonzalez C, Challacombe M, Gill PMW, Johnson
BG, Chen W, Wong MW, Andres JL, Head-Gordon M, Replogle
ES, Pople JA. Gaussian 98. Gaussian: Pittsburgh, PA, 1998.
10. Foresman JB, Frish A. Exploring Chemistry with Electronic
Structure Methods. Gaussian: Pittsburgh, PA, 1996.
11. Tormena CF, Rittner R, Abraham RJ, Basso EA, Pontes RM. J.
Chem. Soc., Perkin Trans. 2 2000; 2054–2059.
12. Abraham RJ, Tormena CF, Rittner R. J. Chem. Soc., Perkin Trans.
2 2001; 815–820.
13. Abraham RJ, Grant GH, Haworth IS, Smith PE. J. Comput.-Aided
Mol. Des. 1991; 5: 21–39.
14. Olivato PR, Rittner R. Rev. Heteroatom. Chem. 1996; 15: 115–
159.
15. Bodot H, Dicko DD, Gounelle Y. Bull. Soc. Chim. Fr. 1967; 870–
876.
1
yield 11 g (23%). H NMR (300 MHz, CDCl3, 20°C,
4
TMS): ꢀ = 2.3 [d, J(H,F) = 5.0 Hz, 3H, CH3], 1.4 [d,
3J(H,F) = 21.5 Hz, 6H, 2CH3]. 13C NMR (75 MHz,
CDCl3, 20°C, TMS): ꢀ = 210.6, 98.7, 24.5, 23.9.
1-Bromo-3,3-dimethyl-2-butanone16.
Pinacolone
(16.0 g, 0.16 mol) in diethyl ether (60 ml) was placed in
a 150 ml three-neck flask equipped with a condenser,
addition funnel and magnetic stirrer. Bromine (25.0 g,
0.16 mol) was added dropwise over a period of 60 min.
The reaction mixture was washed with water (4 Â 50 ml)
and dried over MgSO4. The solvent was removed and the
desired product was vacuum distilled through a Vigreux
column to give pure 1-bromo-3,3-dimethyl-2-butanone
1
(b.p. 78°C/4 mmHg), yield 18.5 g (65.0%). H NMR
(300 MHz, CDCl3, 20°C, TMS): ꢀ = 4.2 (s, 2H, CH2), 1.2
(s, 9H, tBu). 13C NMR (75 MHz, CDCl3, 20°C, TMS):
ꢀ = 206.0, 44.2, 31.7, 26.7.
1-Fluoro-3,3-dimethyl-2-butanone18,19. In a 250 ml
three-neck flask, equipped with an addition funnel,
magnetic stirrer and a distillation column and condenser
to remove the FDMB immediately after its formation,
potassium bifluoride (15.0 g, 0.252 mol) in dry diethylene
glycol (150 ml) was heated at 200°C and 1-bromo-3,3-
dimethyl-2-butanone (30.0 g, 0.168 mol) was added
dropwise. The low-boiling product immediately distilled
over. The NMR spectrum of the distillate showed the
correct pattern for FDMB and indicated the presence of
some impurity. Redistilation gave pure FDMB, b.p.
16. Cox RA, Warkentin J. Can. J. Chem. 1972; 50: 3242–3247.
17. Fry AJ, Migron Y. Tetrahedron Let. 1979; 36: 3357–3360.
18. Bergmann ED, Cohen S, Hoffman E, Rand-Meir Z. J. Chem. Soc.
1961; 3452–3457.
19. Bogolyubskii AV, Ll’chenco A Ya, Yagupol’skii LM. J. Org.
Chem. USSR 1987; 23: 2027–2030.
1
125°C, yield 8 g (40%). H NMR (300 MHz, CDCl3,
20°C, TMS): ꢀ = 4.9 [d, 2J(H,F) = 47.7 Hz, 2H, CH2], 1.1
Copyright  2002 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2002; 15: 211–217