Variable-temperature NMR study of the enol forms of benzoylacetones

Article abstract of DOI:10.1002/mrc.1673

The enol forms of the β-diketones, benzoylacetones, have been studied using long-range carbon-hydrogen couplings involving the chelate OH proton, O¹H chemical shifts, ¹³C chemical shifts and deuterium isotope effects on ¹³C chemical shifts. Studies were done in the temperature range from 268 to 181 K. The compounds are shown to be tautomeric, in opposition to a more symmetrical, delocalised, close to one-potential well structure as found in the solid at very low temperature. The same is true for dibenzoylmethane. The isotope effects on chemical shifts are very sensitive gauges of structure in these almost symmetrical systems. Equilibrium constants are determined and related to other similar compounds. Copyright

Full text of DOI:10.1002/mrc.1673

MAGNETIC RESONANCE IN CHEMISTRY
Magn. Reson. Chem. 2005; 43: 992–998
Published online 5 September 2005 in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/mrc.1673
Variable-temperature NMR study of the enol forms of
benzoylacetones
Evgueni V. Borisov,1 Evgueni V. Skorodumov,1 Valentina M. Pachevskaya2 and
Poul Erik Hansen3∗
1
Institute of Bio-organic Chemistry, National Academy of Sciences of Belarus, ul. Kuprevicha, 5/2, 220141, Minsk, Belarus
Institute of Organo-Element Compounds, Russian Academy of Sciences, Vavilov str. 28, 117813, Moscow, Russia
Department of Life Sciences and Chemistry, Roskilde University, P.O. Box 260, DK-4000, Roskilde, Denmark
2
3
Received 21 March 2005; Revised 22 June 2005; Accepted 26 June 2005
The enol forms of the b-diketones, benzoylacetones, have been studied using long-range carbon–hydrogen
couplings involving the chelate OH proton, O1H chemical shifts, 13C chemical shifts and deuterium isotope
effects on 13C chemical shifts. Studies were done in the temperature range from 268 to 181 K. The compounds
are shown to be tautomeric, in opposition to a more symmetrical, delocalised, close to one-potential well
structure as found in the solid at very low temperature. The same is true for dibenzoylmethane. The
isotope effects on chemical shifts are very sensitive gauges of structure in these almost symmetrical
systems. Equilibrium constants are determined and related to other similar compounds. Copyright  2005
John Wiley & Sons, Ltd.
KEYWORDS: NMR; 1H; 13C; deuterium isotope effects on chemical shifts; 13C; O1H coupling constants; ˇ-diketones; enol
forms; potential wells
Ð Ð Ð
of lengthening of O–H distances in O–H O hydrogen
INTRODUCTION
bonds in ˇ-diketones showed its essential distinction from
Hydrogen bonding of the enol forms of ˇ-diketones and the
related intramolecular enol–enol tautomerism have fasci-
nated investigators for many years. The key controversy has
been the two ways of describing enolic ˇ-diketones, either
as a system with a one-potential well, also described as a
‘symmetrical’, strongly delocalised system (Fig. 1(a)) or as a
tautomeric system with a two-potential well (Fig. 1(b)). Early
on, Chan et al.1 suggested an equilibrium between a symmet-
rical and a normal tautomeric structure. Later, Bratan and
Strohbusch2 suggested a ‘symmetrical’, strongly delocalised
system, whereas Shapet’ko et al.3 suggested ‘a continuous
set of different nuclear configurations’. Baugham et al., using
microwave spectra,4 Emsley,5,6 Hansen and Bolvig, using
deuterium isotope effects on chemical shifts,7 and Borisov
et al., using long-range coupling constants,8 found evidence
for a tautomeric equilibrium. Gilli et al.9–11 have carried
out a large number of X-ray structure-based studies of
these compounds and formulated the resonance-assisted
hydrogen bond theory. Carbon–hydrogen couplings involv-
ing the chelate proton have also been studied in a wide
range of substances to elucidate the structure.12 Recently, an
X-ray and neutron diffraction study at very low temperature
of benzoylacetone (BA) has shown an almost symmetrical
structure.13 Theoretical studies showed this to be a transition-
state-like situation.14 Quite recently, a comparative study
15
Ð Ð Ð
the behaviour of usual O–H O bonds.
For symmetrical compounds, isotopic perturbation of
equilibrium is a fruitful way of demonstrating the presence
of a two-well potential.16–18 This has been done for malon-
aldehydes by replacing one of the aldehydic protons with
deuterium.16,17
Gilli et al.19 have recently suggested that acetylacetones
(ACAC) and BAs in the solid show a two-potential well,
whereas dibenzoylmethane (DBM) is supposed to have its
proton in a one-potential well.
In the present combined study, a series of substituted
BAs, including the parent compound (Scheme 1), and two
model compounds (DBM and ACAC) are investigated
at various low temperatures by measuring nJ(13C,O1H)
couplings (in the following, nJ(C,OH)), isotope effects on
13C chemical shifts and 13C and O1H chemical shifts in an
attempt to throw more light on the structure in the liquid
phase and to study the influence of substituents.
RESULTS
NMR parameters were measured at various temperatures
from 268 to 181 K both to minimise possible intermolecular
exchange and to monitor changes due to temperature.
The temperature ranges for different compounds were
slightly different, depending on the solubility of the specific
substance in the solvent. The temperature needed to avoid
exchange on the NMR scale is roughly 227 K (CD2Cl2,
ŁCorrespondence to: Poul Erik Hansen, Department of Life Sciences
and Chemistry, Roskilde University, P.O. Box 260, DK-4000,
Roskilde, Denmark. E-mail: poulerik@ruc.dk
Contract/grant sponsor: INTAS; Contract/grant number: 96-1021.
Copyright  2005 John Wiley & Sons, Ltd.

NMR study of the enol forms of benzoylacetones 993
6.00
5.50
5.00
4.50
4.00
3.50
3.00
2.50
C-1
C-3
C-4
C-2
C-10
(a)
(b)
J
Figure 1. (a) One-potential ‘symmetrical’ presentation and
(b) two-potential (tautomeric) presentation of enolic forms of
BAs.
200.0 210.0 220.0 230.0 240.0 250.0
T, K
Figure 2. Temperature dependences of the long-range
13C,O1H coupling constants in p-methoxy BA.
remote position from the substituents and its good S/N
ratio (Table 2). In addition, the real 3J(C,OH)trans coupling
constants depend very little on O1H chemical shifts for a
broad range of substances.12 This feature makes it possible
to use the 3J(C,OH)trans coupling constant of acetylacetone as
a reference.
Scheme 1. Substituted benzoylacetones studied.
THF-d8) for all substances. The parameters are listed for
this temperature unless indicated otherwise.
J(13C–O1H)
The equilibrium constant KT of the enol–enol equilibrium
(Scheme 2) can therefore be estimated as the ratio of
3J(C-10,OH) and twice the coupling constant of the CH3
carbon of the reference, acetylacetone. This is done for 6–12
temperatures with a 5-degree spacing, in the range 18–268 K.
Plotting ln KT vs 1/T, the thermodynamic parameters can be
determined. This leads for the p-methoxy derivative, for
which the temperature slope (defined by eight points) is the
largest, to H D ꢀ1.3 kJ/mole and S D ꢀ4.2 J/mole/K.
The equilibrium constants for the other substances studied
are shown in Table 2.
Carbon–hydrogen couplings are shown in Table 1 and are
measured both in CD2Cl2 and in THF-d8. The accuracy of
the values obtained in the SELJRES experiments depends
very much on the signal-to-noise (S/N) ratio, which is
different for proton-carrying signals and quaternary carbons.
For methyl signals, the accuracy was determined, in the
best experiments, as 0.01 Hz. Carbon–hydrogen couplings
are found to have small temperature dependences (Fig. 2).
Likewise, solvents have a rather small effect.
For quantitative purposes (evaluation of equilibrium
constants), signal of the CH3 group carbons (C-10) in
SELJRES spectra was found the most useful because of its
It should also be mentioned that the 2J(C-1,OH) and
2J(C-3,OH) values are quite different within the compounds,
Table 1. Measured long-range carbon–proton coupling constants of substituted benzoyl acetones involving the OH proton
a
at 227 K in CD2Cl2 and THF-d8
(2J(C-1,OH)
C J(C-3,OH))/2
2
Compoundc
2J(C-1,OH)
2J(C-3,OH)
3J(C-4,OH)b
3J(C-10,OH)b
3J(C-2,OH)
p-OCH3
p-CH3
H
p-F
p-Cl
m-F
DBM
ACAC
3.72 (3.92)
3.51 (3.73)
3.40 (3.47)
3.62 (3.65)
3.40 (3.59)
3.42
3.08 (2.99)
3.36 (3.32)
3.46 (3.40)
3.08 (3.17)
3.27 (3.30)
3.34
3.40
3.44
3.43
3.35
3.34
3.38
3.31d
3.31d
3.02 (2.82)
3.42 (3.28)
3.49 (3.39)
3.15 (2.97)
3.37 (3.42)
3.45
3.44 (3.59)
3.06 (3.22)
2.92 (2.96)
3.35 (3.45)
2.97 (3.09)
2.97
5.66 (5.73)
5.64 (5.72)
5.65 (5.65)
5.64 (5.68)
5.56 (5.69)
5.54
3.16 (3.42)
3.45(3.49)
3.16 (3.42)
3.45(3.49)
3.28 (3.40)
3.15(3.12)
3.28 (3.40)
3.15(3.12)
5.80 (5.94)
5.42(5.44)
a For the temperature variation, see Table 4. Values in brackets are in THF-d8.
b For acetylacetone and dibenzoylmethane, values are shown for coupling constants of OH with carbons in ˛-position to
carbonyl carbons.
c See Scheme 1.
d Average value of DBM and ACAC.
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998

994
E. V. Borisov et al.
Table 2. Equilibrium constants for tautomeric equilibria of
substituted benzoylacetones (CH2Cl2) and in
The remaining carbon resonances change to some extent. C-4
normally shifted in the opposite direction to that of C-10.
1-aryl-3-(4-F-phenyl)triazenes (THF) at 183 Ka
Isotope effects on 13C chemical shifts
KT triazenesb
Isotope effects on 13C chemical shifts (C D υ13C(OH)ꢀ
υ13C(OD)), with deuteration at the chelate proton, were mea-
sured in one-tube experiments at the same low temperatures
as long-range 13C–O1H couplings and 13C chemical shifts. A
number of interesting features are seen (Table 5). The sum of
2C-1(OD) and 2C-3(OD) is similar for all the compounds,
although the values vary within a compound. The differ-
X
KT (227 K)
KT (183 K)
p-CH3O
p-CH3
H
p-F
p-Cl
m-F
1.39 (1.33)c
0.96 (1.0)
0.89 (0.89)
1.07 (1.24)
0.93 (1.08)
0.82
1.20
0.95
0.87
1.10
0.91
0.89
4.95
0.41
0.25
1.00
0.36
0.14
2
2
ence between C-1(OD) and C-3(OD) is largest for the
p-OCH3 and p-F derivatives. The temperature coefficients
are quite large (Table 4). This is also true for the symmetri-
cal compounds, DBM and ACAC. For the non-symmetrical
a Data for substituted BAs were obtained by linear interpolation
or extrapolation of experimental data.
b Data for substituted diaryltriazenes (Scheme 2) taken from
Ref. 20.
2
compounds, the sum of 2C-1(OD) and C-3(OD) show a
c Values in brackets are based on two-bond isotope effects on
chemical shifts (227 K).
similar trend as υOH. This is also the case for the symmetrical
compounds. The change in the isotope effects on chemical
shifts with temperature, C-1(OD), is approximately the
same for DBM and ACAC, ¾1.5 ð 10ꢀ3 ppm per degree,
whereas for C-1 and C-3 of the BAs they vary considerably
but still in a symmetrical fashion, so that the mean value of
C-1(OD) and C-3(OD) is equal to that found for DBM
and ACAC (Table 5).
DISCUSSION
One of the main problems related to enolic ˇ-diketones is
to distinguish between one-potential wells (Fig. 1(a)) and
Table 4. Changes in parameters per degree, values ð10ꢀ3a
Scheme 2. Enol–enol equilibrium in benzoylacetones and
structure of 1-aryl-3-(4-F-phenyl)triazenes 1.
υ
13C-1
υ
13C-3 υO1H C-1(OD) C-3(OD)
Compound (ppm) (ppm) (ppm)
(ppm)
(ppm)
but the average for a series of investigated substances is
very close to the average of those of acetylacetone and DBM
(Table 1).
p-OCH3
p-CH3
H
p-F
p-Cl
m-F
DBM
ACAC
7.5
0.5
ꢀ3.5
2.5
ꢀ3
ꢀ4
0
ꢀ7
ꢀ3
ꢀ3.25
ꢀ3.6
ꢀ3.6
ꢀ3.3
ꢀ3.4
ꢀ3.3
ꢀ5.3
ꢀ2.6
ꢀ4
1.0
ꢀ0.7
ꢀ1.4
0
ꢀ1.0
ꢀ0.75
ꢀ1.4
ꢀ1.25
ꢀ2.4
ꢀ1.2
ꢀ2.7
ꢀ1.2
ꢀ1.25
ꢀ1.4
ꢀ1.25
2.5
ꢀ4
2
0
0
ꢀ2
13C chemical shifts
The C-1 and C-3 chemical shifts of the BAs are seen to
fall approximately midway between those of DBM and
ACAC (Table 3). The changes with temperature are very
small but are different in different directions, depending on
the compound and the carbon in question. The chemical
shifts of C-2 practically do not change with temperature.
ꢀ2
a Value at high temperature minus value at low temperature
divided by difference of temperature.
Table 3. 13C chemical shifts of benzoylacetones at 227 Ka in CD2Cl2
Compound
C-1
C-3
C-7
C-4
C-5,9
C-6,8
C-2
C-10
Other
p-OCH3
p-CH3
H
p-F
p-Cl
m-Fb
DBM
ACAC
191.73
193.63
194.60
192.86
194.49
194.95
185.93
192.08
184.83
184.06
183.49
182.84
182.45
181.99
185.93
192.08
163.28
143.82
132.86
164.90
138.60
114.03
133.09
127.28
131.94
134.69
130.65
133.29
137.07
135.22
24.94
129.43
129.71
129.03
129.26
129.19
114.07
127.27
127.22
115.54
128.69
95.91
96.46
96.90
96.03
96.86
97.21
93.16
100.82
25.37
25.87
26.14
25.15
26.06
26.15
55.90
21.86
119.65,123.04
127.43
162.00,130.77
129.10
24.94
a The temperature coefficients are given in Table 4.
b The chemical shifts are given in the order indicated in the top row. J13C–19F are: C-3 D 2.6 Hz, C-4 D 7.1 Hz, C-5 D 21.2 Hz,
C-6 D 245.6 Hz, C-7 D 32.1 Hz, C-8 D 8.2 Hz, C-9 D 3.3 Hz.
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998

NMR study of the enol forms of benzoylacetones 995
Table 5. Deuterium isotope effects on 13C chemical shifts of
y = 6.7x − 0.6
R2 = 0.998
1.5
2
benzoylacetones (ppm) and O1H chemical shiftsa at 227 K
Compound C-1 C-3 C-2 C-7 C-4 C-10 υ(O1H)
1
0.5
0
p-OCH3
p-CH3
H
1.01 0.53 ꢀ0.05 0.04c ꢀ0.10 n.o.
16.70
0.75 0.74 ꢀ0.05 0.07 ꢀ0.04 ꢀ0.07 16.60
0.68 0.79 ꢀ0.05 0.05 ꢀ0.03 ꢀ0.10 16.50
−0.3 −0.2 −0.1
−0.5
0
0.1
0.2
0.3
0.4
p-F
0.92 0.56
–
0.04
0.02
–
16.48
−1
−1.5
−2
p-Cl
m-F
DBM
ACAC
0.79 0.69 ꢀ0.04 0.04 ꢀ0.04 ꢀ0.06 16.41
0.77 0.70 ꢀ0.03 0.04
–
ꢀ0.07 16.35
17.31
0.71
0.75
–
–
ꢀ0.09
–
–
ꢀ0.05
–
0.03b
ꢀ0.05 ꢀ0.05 15.77
−2.5
lnK(benzoylacetones)
a The temperature coefficients are given in Table 4.
b Observed at 207 K.
c Observed at 217 K.
Figure 3. Correlation of ln Ks of substituted benzoylacetones
and structurally similar diaryltriazenes (see Scheme 2 for
formula) (data from Ref. 17 in THF at 183 K).
two-potential wells (Fig. 1b). One way of investigating this
problem for symmetrical compounds is to use isotopic per-
turbation of equilibrium.16–18 For non-symmetrical systems,
the equilibrium can be perturbed by deuteration at the enolic
proton position.3,5,7
Furthermore, for two-potential well systems, properties
will depend on mole fractions as exemplified by chemical
shifts (in our case 13C):
or near-symmetrical set of isotope effects on chemical shifts.
In the latter, the isotope effects on chemical shifts will be
like in other equilibrium systems.7,21–23 For the equilibrium
system, observable isotope effects on chemical shifts can be
regarded in a simple fashion as consisting of an intrinsic and
an equilibrium contribution according, for example for C-1,
to equation:
υ13C-1obs D XA υ13C-1A C ꢀ1 ꢀ XAꢁ υ13C-1B
υ13C-3obs D ꢀ1 ꢀ XAꢁ υ13C-3B C XA υ13C-3A
ꢀ1ꢁ
ꢀ2ꢁ
2
4
C-1obs D XA ð C-1ꢀODꢁint C ꢀ1 ꢀ XAꢁ ð C-1ꢀODꢁint
C Xꢀυ13C-1A ꢀ υ13C-1Bꢁ.
ꢀ3ꢁ
XA being the mole fraction of tautomer A.
where, XA is as pointed out above, X being the change in
the mole fraction upon deuteration and υ13C-1A, the carbonyl
chemical shift of tautomer A. For symmetrical systems like
ACAC and DBM, no equilibrium part exists. The observable
effect at, for example C-1, is hence an average of the intrinsic
effects, 2C-1(OD)A and 4C-1(OD)B.7 For non-symmetrical
systems in which the chemical-shift difference is similar for
the involved carbons (Table 6), the equilibrium part will
more or less cancel out and the average of 2C-1(OD)obs and
2C-3(OD)obs gives values close to the intrinsic isotope effect.
For the studied BAs, we find that the average values for all
compounds are similar to those observed for ACAC and
DBM. Furthermore, 2C-1(OD) is larger than 2C-3(OD) for
compounds with K > 1 and the reverse for compounds with
K < 1 (Table 2). In addition, the finding that the differences
increase even more at lower temperature suggests strongly
that we are dealing with tautomerism, but not with a one-well
‘symmetrical situation’. In the latter case, as the temperature
is lowered, the chelate proton should be centred even more,
leading to nearly identical isotope effects on chemical shifts.
This finding is also supported by studies of primary
isotope effects on chemical shifts. In that case, both the
values for BAs, ACAC and DBM are large and positive
pointing towards a two-potential well system.21
As the sums of the υC O and υCO–H chemical shifts are
very similar for both the tautomers (Table 6), this will lead
to a temperature-independent average, e.g. (υ13C-1obs C υC-
3
obsꢁ/2. Furthermore, Eqns 1 and 2 show that whereas the
value X moves away from 0.5, the values of υ13C-1obs
and υ13C-3obs will shift in opposite directions. Lowering
the temperature will enhance this trend. Similar arguments
can be used for coupling constants and, in a slightly more
elaborate fashion, for isotope effects (see the following).
A correlation between ln KT of the substituted BAs in
CH2Cl2 extrapolated to 183 K (calculated from 3J(C-10, OH)
couplings) and data for substituted diaryltriazenes (Table 2)
from Ref. 20 was investigated. The plot is shown in Fig. 3.
The triazenes are proven to be tautomeric.20 The correlation
therefore reflects the latent tautomeric nature of BAs.
Isotope effects on 13C chemical shifts
The isotope effect pattern is supposed to be quite different
in the two scenarios, the one-potential well and the two-
potential well. In the former case, we expect a symmetrical
Table 6. Calculated (DFT)a nuclear shieldings (ppm) and
coupling constants for BA
2J(C-1, 2J(C-3, 3J(C-2,
Having established an equilibrium situation, the isotope
effects on chemical shifts can be used to determine the mole
fractions and the equilibrium constants. From a series of
ˇ-diketones and other compounds, an S-shaped function
Tautomer
ꢂ
13C-1
ꢂ
13C-3 ꢂO1H
OH)
OH)
OH)
A
B
ꢀ0.3
21.5
9.6
14.3
14.5
ꢀ0.3
ꢀ2.3
ꢀ2.0
ꢀ0.2
5.3
5.1
14.7
2
between mole fraction and C(OD) was found.18,19 Using
a B3LYP/6-31G(d,p) type.
this dependence, the equilibrium constants, given in brackets
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998

996
E. V. Borisov et al.
Table 7. Deuterium isotope effects on 13C chemical shifts of
A
1.5
1
substituted dibenzoylmethanes (ppm)a
R/R0 H/H CH3/CH3 OCH3/OCH3 H/CH3 H/OCH3
0.5
0
C-1
C-2
C-3
C-p
C-p’
0.60
0
0.60
0
0.61
0
0.61
0.57
0.02
0.57
0
0.64
ꢀ0.06
0.56
0
0.74
0
0.46
0
0
0.2
0.4
0.6
0.8
1
−0.5
Molefraction,X
ꢀ0.04
0
ꢀ0.04
0
0
0.03
2
1.5
1
B
a Measured in CDCl3 at ambient temperature.
0.5
0
0
0.2
0.4
0.6
0.8
1
−0.5
Mole fraction, X
Figure 4. Dependence of 2C-1,3(OD) versus mole fraction, X.
Data for establishing the graph are taken from Ref. 7. (A) Data
points from the present work are shown by grey squares. For
C-1, the data are fitted by the following function: 2C-1(OD) D
24.979x4 ꢀ 33.944x3 C 7.270x2 C 1.924x C 0.556 (B) For C-3,
2C-3(OD) D ꢀ16.817x4 C 17.264x3 C 1.659x2 ꢀ 1.750x
C0.069.
into the common trend as well for two-bond, as for cis three-
bond couplings. Thus in both cases, no evidence of essential
differences between ACAC, DBM and the BAs is found.
13C chemical shifts
in Table 2, can be obtained (Fig. 4). These are found to be
similar to those obtained from coupling constants.
For an equilibrium system, the 13C chemical shifts will be
the weighted average of the chemical shifts of the nuclei
of the two tautomers, as discussed previously. Therefore,
they can be used for estimation of the order of equilibrium
constants (Table 2). Using C-1 chemical shifts, the order is
p-OCH3 > p-F > p-CH3 > p-Cl > H > m-F (this conclusion
is based on the well-known feature that carbonyl carbons
are deshielded relative to enol carbons; see also Table 6).
This series qualitatively coincides with the order based on
3J(C-10,OH) coupling constants (Table 2).
For C-4 and C-10, the situation is similar to that of C-1 and
C-3. The sum of nC-4(OD) and nC-10(OD) is ¾ꢀ0.045 ppm
for all compounds (Table 5) and similar to those observed
for DBM and ACAC. For the BAs the following pairs are
n
observed, C-4(OD), nC-10(OD): p-OCH3, ꢀ0.09, 0; p-Cl,
ꢀ0.03, ꢀ0.05; m-F, 0, ꢀ0.065; p-CH3, ꢀ0.03, ꢀ0.07; H, ꢀ0.03,
ꢀ0.07 ppm. Again, the average value is constant. The two
isotope effects on chemical shifts show a spread and one
finds the trend from equilibrium constants.
As we lower the temperature, the C-1 chemical shift
changes in the direction so that the dominant form becomes
more highly populated. Of course, the 13C chemical shift
must be corrected for the change in the chemical shift as
such. This can be judged for C-1 from ACAC and DBM
(Table 4). Strangely enough, the data for ACAC and DBM
move in opposite directions. This can possibly be related to
reorientation of the phenyl rings with temperature in DBM.
The change of C-2 carbon resonance with temperature is very
small.
The isotope effects on chemical shifts obtained for BAs
may be compared with those of DBMs (Table 7). A pattern
similar to that of DBMs is observed. Furthermore, a large
positive primary isotope effect is found for DBM strongly
supporting a two-well potential.21,24
Coupling constants
For the pair of two-bond couplings to C-1 and C-3, 2J(C-
1,OH) and 2J(C-3,OH), it can be seen that the average value
has very little dependence on υO1H, but the individual values
are quite different, depending on the substituents (Table 1).
The p-OCH3 and p-F substituents give the largest variations.
This feature again supports that the system is tautomeric and
shifted most strongly away from K D 1 for mesomerically
acting substituents such as fluorine or the methoxy group. A
similar pattern can be seen for the pair of trans three-bond
couplings to C-4 and to C-10. The value for 3J(C-2,OH)cis can
be seen to increase slightly with hydrogen bond strength as
given by υO1H (Fig. 5). This feature was also found for a
series of benzene derivatives and enolic ˇ-diketones.12 The
average values for ACAC and DBM (by averaging, we model
chemical shifts and coupling constants for BA) do fit well
5.9
5.8
5.7
5.6
5.5
J
5.4
5.3
16.30
16.40
16.50
d(O1H), ppm
16.60
16.70
Figure 5. 3J(C-2,OH)cis versus υO1H; the triangle represents an
average of the data for ACAC and DBM.
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998

NMR study of the enol forms of benzoylacetones 997
basis set was used. The nuclear shieldings were calcu-
lated using the gauge including atomic orbitals (GIAO)
approach.31,32
CONCLUSIONS
The type of potential for the enolic form of ˇ-diketones has
been reinvestigated and a two-potential well, tautomeric
system is the favoured one down to 181 K for BAs.
Mesomerically, electron-donating substituents like methoxy
and fluoro in para-position favours the B-tautomer (Scheme
2). And as a whole, the influence of substituents on
the position of the equilibrium is identical to that for
diaryltriazenes. On the basis of both deuterium isotope
effects on 13C chemical shifts, the long-range 13C–O1H
coupling constants–O1H chemical shifts correlations and
primary deuterium isotope effects on chemical shifts the
same pattern is also found for DBM. This is contrary to
the recent study by X ray in which it was suggested
that DBM shows a one-potential well.16 Lowering the
temperature increases the hydrogen bond strength as judged
both from isotope effects on chemical shifts and OH
chemical shifts. However, these changes are moderate.
In the solid at 20 K, an almost symmetrical structure
of BA was found.13 We suggest a reduction in twist
angle to be the driving force leading to more steric
Acknowledgements
The present study was supported by INTAS grant 96-1021, which
is gratefully acknowledged. Professor Hans Fritz is thanked
for providing the isotope effect data of dibenzoylmethanes
(Table 6).
REFERENCES
1. Chan SI, Lin L, Clutter D, Dea P. Proc. Natl. Acad. Sci. U. S. A.
1970; 65: 816.
2. Bratan S, Strohbusch F. Chem. Ber. 1972; 105: 2284.
3. Shapet’ko NN, Bogachev YuS, Khatipov SA. Zh. Fiz. Khim. 1988;
62: 2632.
4. Baugham SL, Smith Z, Wilson WB, Duerst RW. J. Am. Chem. Soc.
1984; 106: 2260.
5. Emsley J. Structure and Bonding, vol. 57. Springer-Verlag: Berlin,
1984; 147.
6. Emsley J, Freeman NJ, Bates PA, Hursthouse MB. J. Chem. Soc.,
Perkin Trans. 1 1988; 297.
7. Bolvig S, Hansen PE. Magn. Reson. Chem. 1996; 34: 467.
8. Borisov EV, Khlebnikova TS. Dokl. Acad. Nauk, BSSR 1987; 31:
359.
9. Gilli G, Belluci F, Ferretti V, Bertolasi V. J. Am. Chem. Soc. 1989;
111: 1023.
10. Bertolasi V, Gilli P, Ferretti V, Gilli G. J. Am. Chem. Soc. 1991; 113:
4917.
compression25,26 and a shorter O O distance that ultimately
leads to a longer O–H distance and to a more symmetrical
structure.
Ð Ð Ð
11. Gilli G, Bertolasi V. In The Chemistry of Enols, Rappoport Z (ed).
John Wiley & Sons: New York, 1990; 713.
EXPERIMENTAL
12. Borisov EV, Zhang W, Bolvig S, Hansen PE. Magn. Reson. Chem.
1998; 36: S104.
13. Madsen GKH, Iversen BB, Larsen FK, Kapon M, Reisner GM,
Herbstein FH. J. Am. Chem. Soc. 1998; 120: 10 040.
14. Schiott B, Madsen GKH, Bruice TC. J. Am. Chem. Soc. 1998; 120:
12 117.
Compounds
The compounds were prepared according to Ref. 27. For
m-fluorobenzoylacetone: (mp 42–44 C); elemental analysis:
calculated: C-66.66%; H-5.00%; found: C-66.59%; I-4.97%.
°
15. Borisov EV. Russ. Chem. Bull. 2000; 49: 758.
16. Perrin CL, Nielson JB, Kim Y-J. Ber. Bunsen-Ges. Phys. Chem. 1998;
102: 403.
17. Perrin CL, Kim Y-J. J. Am. Chem. Soc. 1998; 120: 12 641.
18. Perrin CL, Nielson JB. Annu. Rev. Phys. Chem. 1997; 48: 511.
19. Gilli P, Bertolasi V, Pretto L, Ferretti V, Gilli G. J. Am. Chem. Soc.
2004; 126: 3845.
20. Borisov EV, Peregudov AS, Postovoi SA, Fedin EI, Kravtsov DN.
Bull. Acad. Sci. USSR, Div. Chem. Sci. 1986; 449.
21. Bolvig S, Hansen PE. Curr. Org. Chem. 2000; 4: 19.
22. Hansen PE, Bolvig S. Magn. Reson. Chem. 1997; 35: 520.
23. Hansen PE. Magn. Reson. Chem. 1986; 24: 903.
24. Bolvig S, Hansen PE, Morimoto H, Wemmer D, Williams P.
Magn. Reson. Chem. 2000; 38: 525.
NMR
NMR spectra were recorded at 50.32 MHz on a Bruker
AC-P 200 spectrometer at temperatures between 268 and
181 K. Compounds were dissolved in CD2Cl2, THF-d8 in 1 M
concentration whenever possible.
The SELJRES experiments28 were performed using the
Bruker library program with the following acquisition
parameters spectral width 10.02 Hz ꢀF1ꢁ ð 12 195.12 Hz ꢀF2ꢁ,
NS D 64, NE D 32, data matrix 64 W ð 4K, relaxation
delay 15 s, selection pulse width 8.45 µs, CPD decoupling.
Processing parameters were data matrix 1 K ð 4 K, digital
resolution 0.01 Hz/point, filters were GM(ꢀ0.025; 0.01) F1
and sine (shifted ꢃ/2) F2.
The dispersion for the Me signal in acetylacetone at 207 K
(15%v/v) was 0.02 Hz (95%)(five independent experiments).
This data can be used as an estimation of the upper level of
accuracy of the experimental data. The temperatures were
calibrated using an MeOH sample (Tesla). The temperature
standard deviation was estimated as 0.5 K (all range of
temperatures).
25. Abildgaard J, Bolvig S, Hansen PE. J. Am. Chem. Soc. 1998; 120:
9063.
26. Filarowski A, Koll A, Glowiak T. Monatsch. Chem. 1999; 130:
1097.
27. Chen LS, Cummings SC. Inorg. Chem. 1978; 17: 2358.
28. Bax A, Freeman R. J. Am. Chem. Soc. 1982; 104: 1099.
29. Frisch MJ, Trucks GW, Schlegel HB, Scuseria GE, Robb MA,
Cheeseman JR, Montgomery JA Jr, Vreven T, Kudin KN,
Burant JC, Millam JM, Iyengar SS, Tomasi J, Barone V,
Mennucci B, Cossi M, Scalmani G, Rega N, Petersson GA,
Nakatsuji H, Hada M, Ehara M, Toyota K, Fukuda R,
Hasegawa J, Ishida M, Nakajima T, Honda Y, Kitao O,
Nakai H, Klene M, Li X, Knox JE, Hratchian HP, Cross JB,
Bakken V, Adamo C, Jaramillo J, Gomperts R, Stratmann RE,
Yazyev O, Austin AJ, Cammi R, Pomelli C, Ochterski JW,
Ayala PY, Morokuma K, Voth GA, Salvador P, Dannenberg JJ,
Zakrzewski VG, Dapprich S, Daniels AD, Strain MC, Farkas O,
Malick DK, Rabuck AD, Raghavachari K, Foresman JB, Ortiz JV,
DFT calculations
The molecular geometries were optimised using the Gaus-
sian 03 suite of programs29 and B3LYP Density Func-
tional Theory (DFT) (Beckes exchange30 and a mix of
the built-in Gaussian-type basis sets). The 6-31G(d,p)
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998

998
E. V. Borisov et al.
Cui Q, Baboul AG, Clifford S, Cioslowski J, Stefanov BB,
Liu G, Liashenko A, Piskorz P, Komaromi I, Martin RL,
Fox DJ, Keith T, Al-Laham MA, Peng CY, Nanayakkara A,
Challacombe M, Gill PMW, Johnson B, Chen W, Wong MW,
Gonzalez C, Pople JA. Gaussian 03, Revision B.03. Gaussian
Incorporated: Wallingford, 2004.
30. Becke D. Phys. Rev. A. 1988; 38: 3098.
31. Ditchfield R. Mol. Phys. 1974; 27: 789.
32. Wolinski K, Hilton FF, Pulay P. J. Am. Chem. Soc. 1990; 112: 8251.
Copyright  2005 John Wiley & Sons, Ltd.
Magn. Reson. Chem. 2005; 43: 992–998