Factors determining tautomeric equilibria in Schiff bases

Article abstract of DOI:10.1016/j.molstruc.2011.10.011

Schiff bases derived from o-hydroxyaldehydes present keto and enol tautomeric forms; the relative equilibrium between these two tautomers depending on the particular aldehyde the Schiff bases is derived from. Thus benzaldehyde produces a stable enol tauto

Full text of DOI:10.1016/j.molstruc.2011.10.011

Journal of Molecular Structure 1006 (2011) 600–605
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
Factors determining tautomeric equilibria in Schiff bases
´
Martha Flores-Leonar, Nuria Esturau-Escofet, José M. Méndez-Stivalet, Armando Marın-Becerra,
Carlos Amador-Bedolla
⇑
´
Facultad de Quımica, Universidad Nacional Autónoma de México, México DF 04510, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
Schiff bases derived from o-hydroxyaldehydes present keto and enol tautomeric forms; the relative equi-
librium between these two tautomers depending on the particular aldehyde the Schiff bases is derived
from. Thus benzaldehyde produces a stable enol tautomer, while a naphthaldehyde produces a mixture
of keto and enol tautomers. The energy difference between these tautomers is very small (ꢀ5 kJ/mol) and
therefore close to current precision limits of ab initio and DFT based quantum calculations. NMR spectros-
copy results, which allows for the determination of the stable structure when one tautomer is prevalent,
can be very difficult to interpret when both tautomers are present. We calculate energy differences
between the tautomers and demonstrate that the precision of current DFT calculations is not sufficient
to predict the most stable structure. On the other hand, DFT calculations of the NMR chemical shifts
(using the GIAO technique) can properly interpret the spectroscopy results allowing the characterization
of the experimentally present tautomers and the estimation of the relative abundance of each when both
are present.
Received 10 August 2011
Received in revised form 12 September
2011
Accepted 8 October 2011
Available online 19 October 2011
Keywords:
Schiff bases
Enol–keto tautomer equilibrium
DFT–GIAO NMR chemical shifts
Ó 2011 Elsevier B.V. All rights reserved.
1. Introduction
it for the chemical shift of the carbon atom forming an enol bond of
155 ppm, and an upper limit for the keto bond of 180 ppm [9,10].
Schiff bases are compounds in which an imine group (AC@NA)
is present. Some of them, of the general type R0C6H3(OH)(CH@NR00),
present enol–keto tautomerism. These compounds are synthesized
from carbonylic derivatives such as salicylaldehyde or o-hydrox-
yaldehydes with a monoamine in 1:1 proportion via a single step
condensation. These type of compounds have been widely used
in biochemistry, coordination chemistry, catalytic reactions and or-
ganic synthesis [1–5].
Equilibrium between the tautomeric forms has been studied
also for acidic and basic solutions. Its displacement from the
keto–amine tautomer at neutral pH to the enol–imine tautomer
at acidic solution being explained in terms of the formation of a
hydrogen bond with the acid molecule [7]. However, our 1H and
13C NMR data show that the equilibrium is not displaced to the
enol tautomer but instead tautomers are hydrolyzed back to the
aldehyde precursor.
Tautomeric forms enol–imine and keto–amine exist, both pre-
senting an intramolecular hydrogen bond [4–8]. In the solid state
both forms have been reported, whereas in solution, salicylaldi-
mines species produce preferently the enol–imine tautomer (enol)
while 2-hydroxy-1-naphthaldimines species produce mainly the
keto–amine tautomer (keto). Both tautomers have been detected
by X-ray structural analyses that show that the transformation
from enol–imine to keto–amine is accompanied by a considerable
increase in the CAN distance from 1.317 Å to 1.330 Å, as well as the
reduction in the CAO distance from 1.279 Å to 1.263 Å [7]. Based
on these results it has been proposed that the dominant tautomer
depends on the type of carbonylic precursor and not on the stereo-
chemistry of the molecule nor the substituent of the nitrogen of
the imine [6,8]. 13C NMR has been widely used for identification
of the dominant tautomer, leading to the prediction of a lower lim-
In order to further understand this behavior, quantum chemis-
try calculations were performed for a series of Schiff bases species
to compare the stability of the keto and enol forms for different R0
and R00 substituents. Additionally, four of these substances were
synthesized and their structure determined by 1H NMR and 13C
NMR. We selected 12 different compounds based either on the sal-
icylaldehyde or the naphthaldehyde precursor. According to the
usual organic chemistry argument, preservation of aromaticity ac-
counts for the stabilization of the keto tautomer instead of the enol
one. We studied this hypothesis by choosing molecules based on
the naphthaldehyde precursor both with aromatic rings (based
on 2-hydroxy-1-naphthaldehyde and 1-hydroxy-2-naphthalde-
hyde) and without aromaticity (based on 3-hydroxy-2-
naphthaldehyde).
The total of 12 different molecules for studying the enol–keto
energy difference is shown in Scheme 1. Molecules a, b, c, and d
were synthesized and characterized. Notice that molecules b, d, f,
h, j, and l preserve some aromaticity in the keto form.
⇑
Corresponding author. Tel.: +52 55 56223804; fax: +52 55 56223776.
E-mail address: carlos.amador@unam.mx (C. Amador-Bedolla).
0022-2860/$ - see front matter Ó 2011 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2011.10.011

M. Flores-Leonar et al. / Journal of Molecular Structure 1006 (2011) 600–605
601
7
8
6
4
9
6
5
5
10
7
11
6'
2'
5'
3'
6'
2'
5'
3'
4
CH
CH
N
H
CH
CH
N
H
1
1
4'
4'
N
H
N
H
3
2
1'
3
2
1'
R
R
R
R
O
O
O
O
enol
keto
enol
keto
a
c
e
g
b
d
f
R = H
R = H
R = m-Cl
R = o-Cl
R = p-Cl
R = m-Cl
R = o-Cl
R = p-Cl
h
CH
N
N
CH
N
N
CH
N
N
CH
N
N
O
H
O
H
O
H
O
H
i
i
j
j
enol
keto
enol
keto
O
N
N
l
O
N
N
H
H
O
CH
O
CH
CH
CH
H
N
H N
N
N
k
k
keto
l
enol
enol
keto
Scheme 1. The 12 different molecules considered in the study of the enol–keto equilibrium.
Table 1
Energy differences between the enol–imine and keto–amine tautomers for the 12
species considered,
E = Eketo ꢁ Eenol, in kJ/mol. The presence of a solvent is modeled
with PCM for CHCl3 and DMSO.
The rest of this paper is organized as follows. In Section 2 we
discuss our quantum chemical calculations and their results. In
Section 3 we present our experimental results from 1H NMR and
13C NMR for molecules a, b, c, and d. In Section 4 we present results
from the GIAO based NMR calculations and use them to interpret
experimental results obtained for the four molecules synthesized
and predict them for the rest of the molecules. Section 5 show
our conclusions.
D
Species
6-311++G(d,p)
PCM
Gas
CHCl3
DMSO
a
b
c
d
e
f
g
h
i
+19.5
+2.4
+21.8
+4.3
+21.3
+6.3
+20.2
+4.3
+14.2
ꢁ3.2
+32.8
ꢁ7.1
+10.9
ꢁ4.3
+13.5
ꢁ2.2
+16.0
ꢁ0.4
+13.1
ꢁ2.4
+6.9
+6.9
ꢁ7.3
+9.7
ꢁ5.2
+11.9
ꢁ3.4
+9.1
2. DFT calculations
Density functional theory calculations, using the B3LYP [11,12]
functional, were performed for 12 different species both of the
keto–amine and enol–imine tautomers. Calculations were carried
out using the Gaussian 09W [13] package and employing basis
functions of the 6-311++G(d,p) type for all atoms. Optimal geomet-
ric structures were obtained for each isolated molecule (gas phase)
and also for the molecule in solution using the Polarizable Contin-
uum Model (PCM). Solvents used for NMR spectra of a, b, c, and d,
CHCl3 and DMSO, were considered.
ꢁ5.5
+3.8
j
k
l
ꢁ8.7
+23.4
ꢁ12.0
ꢁ10.9
+19.2
ꢁ14.1
tautomer in the gas phase. The solvent effect reduce the differ-
ence between the enol and the keto form and favors stability of
the latter. In DMSO this stabilization is large averaging more than
10 kJ/mol, thus turning the slightly favorable enol tautomer into a
slightly favorable keto tautomer for all species based in naphthal-
dehyde. Still, this change in energy difference, resulting in a more
stable keto tautomer by ꢁ5.2 kJ/mol in the case of species d for
example, is in the limit of precision of DFT calculations. Whether
the observed structure corresponds to the keto form can not be
Table 1 shows the energy differences between enol and keto
tautomers.
Results show that for species based on benzaldehyde (a, c, e, g,
and i) the enol–imine form is favored strongly in gas phase, while
for species based in either 1-2 or 2-1-naphthaldehyde the energy
difference between enol–imine and keto–amine is small, justify-
ing the possibility of both tautomers being observed at room tem-
perature. Only two species (j and l) favored the keto–amine

602
M. Flores-Leonar et al. / Journal of Molecular Structure 1006 (2011) 600–605
predicted based on this energy difference. At most, these results
can be used to predict that, in DMSO, most molecules based on
benzaldehyde will be present in the enol form, with the possible
exception of molecules a and i that can present a mixture of the
two tautomers at room temperature. Correspondingly, species
based on naphthaldehyde will favor either the existence of a mix-
ture of the enol and keto tautomers for species d, f, and h or per-
haps only the keto form for species b, j, and l. In CHCl3, the
stabilization of the keto tautomer is less pronounced, averaging
7 kJ/mol. This is reflected in the experimental results on species
b and d, which show a higher proportion of the keto tautomer
in the estimated equilibrium constant for DMSO (see below
Section 4).
This predictions could be confirmed by synthesizing the actual
compounds. As mentioned earlier, we were able to synthesize spe-
cies a and c (benzaldehyde based, enol favoring) and b and d
(naphthaldehyde based, mixed tautomers favoring). NMR spectra
were obtained for these four species (see below) and their struc-
ture analyzed with the help of calculated chemical shifts for C
atoms as obtained by the use of the GIAO method [14] available
through the calculation performed by using the Gaussian 09W
[13] package.
3.3. Synthesis of (E)-2-((3-chlorophenylimino)methyl)phenol
(Scheme 1c)
To a solution of salicylaldehyde (500
lL, 4.7 mmol) in metha-
nol (50 mL), a solution of m-chloroaniline (496
l
L, 4.7 mmol) in
methanol (10 mL) was added while stirring. The reaction mixture
was stirred for additional 30 min. The volume was then reduced
in half and the solution cooled at ꢁ18 °C for a couple of hours.
A yellow crystalline solid was obtained which was filtered off un-
der vacuum and washed with 20 mL of cold ether. Yield 1.0152 g
(93.5%).
3.4. Synthesis of (E)-1-((3-chlorophenylimino)methyl)naphthalen-2-ol
and (Z)-1-((3-chlorophenylamino)methylene)naphthalen-2-one
(Scheme 1d)
To a solution of 2-hydroxy-1-naphthaldehyde (0.5 g, 2.9 mmol)
in methanol (50 mL),
a solution of m-chloroaniline (307 lL,
2.9 mmol) in methanol (10 mL) was added while stirring. The reac-
tion mixture was stirred for additional 30 min. The volume was
then reduced in vacuo to 20 mL and the resulting solution was
cooled at ꢁ18 °C for a couple of hours. On cooling, a yellow cryst-
aline solid was obtained which was filtered off, washed with 20 mL
of cold ether and dried under vacuum. Yield 0.5733 g (70.2%).
1H and 13C NMR results for a(c) show that the enol tautomer is
obtained exclusively without presence of keto tautomer. For 1H
NMR in CDCl3 the OH proton is observed at 13.3(12.9) ppm as a
broad signal, while the proton joined to the Schiff base carbon
atom (N@CHAAr) appears as a singlet at 8.6(8.6) ppm. The aro-
matic protons appear from 6.9 to 7.5(6.9 to 7.5) ppm. 1H NMR in
DMSO results show similar shifts with the OH proton appearing
at 13.1(12.7) ppm also as a broad signal, the N@CHAAr proton is
now observed at 8.9(8.9) ppm and the aromatic signals appear
from 6.9 to 7.7(6.9 to 7.7) ppm.
In addition, results from the calculations were used to ratio-
nalize how the electronic structures (HOMO, LUMO energies
and their distribution in the molecule) of enol and keto tautomers
differ, and how these differences change when the species is
based on benzaldehyde with respect to when it is based on
naphthaldehyde.
3. Experimental and NMR results
All solvents and reagents were used as received. 1H and 13C
NMR spectra were determined using a Varian VNMR-400 MHz
spectrometer (399.96 MHz for 1H and 100.58 MHz for 13C) at
299 K. Chemical shifts (d in ppm) are referenced to solvent peaks
CDCl3 7.26 for 1H and 77.0 for 13C, DMSO-d6 2.49 for 1H and 39.5
for 13C.
In contrast, when the precursor is 2-hydroxy-1-naphthalde-
hyde, structure b(d), 1H NMR studies show that tautomeric equi-
libria is present in both CDCl3 and DMSO. In CDCl3 the NH
proton is observed at 15.5(15.1) ppm as
a
doublet
Both one-dimensional (1H and 13C) and two-dimensional 2D
NMR (inverse detected HSQC and HMBC) spectra were acquired
with standard conditions and with standard pulse programs taken
from the Varian software library.
(J = 4.7(2.7) Hz), and the proton corresponding to the Schiff base
carbon (NACH@Ar) is observed also as a doublet at 9.3(9.3) ppm
(J = 4.7(2.7) Hz); these results are only possible for a keto–amine
form owing to the location of the hydrogen atom on nitrogen.
The presence of this doublet deserves special comment. In a slow
regime we would observe two signals, corresponding to both keto
and enol tautomers, a singlet for enol and a doublet for keto, the
latter due to the AHNACH@coupling. However, since the equilib-
rium between ceto and enol tautomers is a very fast interconver-
sion, we observe an averaged signal from the doublet and the
singlet, i.e. a pseudo-doublet without a well defined minimum
(see Supplementary information). This coupling has been observed
3.1. Synthesis of (E)-2-((phenylimino)methyl)phenol (Scheme 1a)
To a solution of salicylaldehyde (500
lL, 4.7 mmol) in metha-
nol (50 mL), a solution of aniline (428 L, 4.7 mmol) in methanol
l
(10 mL) was added while stirring. The reaction mixture was stir-
red for additional 30 min. The volume was then reduced in half
and the solution cooled at ꢁ18 °C for a couple of hours. A yellow
crystalline solid was obtained which was filtered off under vac-
uum and washed with 20 mL of cold ether. Yield 0.6489 g
(70.2%).
3
in similar systems with coupling constant, JHNCH, values in the
range 2–4 Hz [15]. To confirm this assertion, we also measured
1H NMR signals at lower temperatures for species b in CDCl3 and
found that, at ꢁ20 °C, the doublet shifts high field, the coupling
constant increases from 4.0 Hz to 6.0 Hz, and the minimum is
now well defined; moreover, at ꢁ50 °C, the doublet shifts to high
field even more, the coupling constant increases further to 7.2 Hz
and the minimum is even more pronounced (see Supplementary
information). Thus, at lower temperatures the equilibrium is
shifted even more towards the stable keto form, in agreement with
calculations (see Section 2).
Aromatic protons appear from 7.0 to 8.1(7.1 to 8.1) ppm. For
structure d, the minor peaks appearing as singlets at 13.1 ppm
for OH proton and at 10.8 ppm for the HA(C@O)AAr proton are
attributed to the hydrolyzed compound, which was detected by
HMBC experiment observing a correlation at 193 ppm for 13C
3.2. Synthesis of (E)-1-((phenylimino)methyl)naphthalen-2-ol and
(Z)-1-((phenylamino)methylene)naphthalen-2-one (Scheme 1b)
To a solution of 2-hydroxy-1-naphthaldehyde (0.5 g, 2.9 mmol)
in methanol (50 mL), a solution of aniline (264 lL, 2.9 mmol) in
methanol (10 mL) was added while stirring. The reaction mixture
was stirred for additional 30 min. The volume was then reduced
in vacuo to 20 mL and the resulting solution was cooled at
ꢁ18 °C for a couple of hours. On cooling, a yellow crystalline solid
was obtained which was filtered off, washed with 20 mL of cold
ether and dried under vacuum. Yield 0.6675 g (93.4%).

M. Flores-Leonar et al. / Journal of Molecular Structure 1006 (2011) 600–605
603
where rben and rx are the NMR isotropic magnetic shielding values
which corresponds to the aldehyde carbon atom displacement of
the 2-hydroxy-1-naphthaldehyde precursor.
for benzene and the corresponding nucleus, respectively, and dben is
the experimental chemical shift of benzene (128.62 ppm in ben-
zene, 128.37 ppm in CDCl3, and 128.30 ppm in DMSO [18]).
For species a and c, calculated chemical shifts for a given 13C
atom differ only slightly (less than 7 ppm) between enol and keto
When solvent is changed to DMSO, for structure b(d), signals
corresponding to NH and NACH@Ar groups are found as singlets
at 15.8(15.5) and 9.6(9.7) ppm respectively, this is due to the sol-
vent coordinating to the acidic proton involved in the equilibrium,
thus eliminating the posibility of coupling between these two pro-
tons. The aromatic protons signals appear from 7.0 to 8.5 ppm.
In 13C NMR, the principal differences between the structures
a(c) and b(d) can be observed on carbon 2 and on Schiff base car-
bon (atom 7 in a, c and 11 in b, d). For structure a(c), the chemical
shifts in CDCl3 are 161.11(161.13) ppm and 162.64(163.68) ppm
for C2 and C7 respectively, while in DMSO are 160.26(160.19)
and 163.49(164.69) ppm; i.e. C2 (enol carbon) is upfield from the
Schiff base carbon C7. When the precursor is changed to 2-hydro-
xy-1-naphthaldehyde, b(d), 13C NMR chemical shifts change, dis-
placing significantly the C2 signal downfield at 171.23(168.28)
ppm, thus inverting Schiff base carbon (now C11) which appears
at 154.15(156.43) ppm in CDCl3; a similar thing happens in DMSO
where shift values are 171.16(169.01) and 155.25(157.08) ppm.
This change implies the presence of the keto–amine tautomer,
and therefore, tautomeric equilibrium.
0
tautomers, except for atoms C2; C7; C1 ; C3 (Scheme 1 a and c). Ta-
ble 2 shows calculated chemical shifts for these atoms both for
enol and keto tautomers, as well as measured chemical shifts for
the synthesized compounds. For species a enol calculated and
experimental values have a MAD of 2.29 (2.94) ppm, while keto
calculated and experimental values have
a MAD of 13.70
(14.32) ppm in CDCl3 (DMSO); while for species c enol calculated
and experimental values have a MAD of 2.37 (2.75) ppm, while
keto calculated and experimental values have a MAD of 14.45
(15.18) ppm in CDCl3 (DMSO). The correspondence between the
experimental values and the enol tautomer confirms the experi-
mental interpretation that this is the structure present in solution
both for CDCl3 and DMSO.
For species b and d, differences between calculated chemical
shifts for the enol and keto tautomers are substantial only for the
0
same set of four carbon atoms as in species a and c, C2; C11; C1 ; C3
(Scheme 1 b and d). Table 3 shows calculated chemical shifts for
these atoms both for enol and keto tautomers, as well as measured
chemical shifts for the synthesized compounds. For species b, enol
4. GIAO calculation and interpretation
calculated and experimental values have
a MAD of 4.73
Gauge Invariant Atomic Orbitals (GIAO) basis set allows the the-
oretical prediction of chemical displacements through the calcula-
tion of magnetic shielding tensors [14]. The best ‘‘virtual’’ reference
standard needed for this calculation has been the subject of recent
studies [16,17]. According to Sarotti and Pellegrinet [17], a good
selection for sp2 13C in DFT calculations as a reference standard is
(4.44) ppm, while keto calculated and experimental values have a
MAD of 9.81 (9.92) ppm in CDCl3 (DMSO); that is, an increase in
MAD for the enol tautomer and a decrease for the keto tautomer
with respect to values for species a. Similarly, for species d, enol
calculated and experimental values have
a MAD of 3.02
benzene. The predicted chemical shift for a given nucleus ðdcxalc
Þ
(3.85) ppm, while keto calculated and experimental values have a
MAD of 11.41 (10.99) ppm in CDCl3 (DMSO); showing also an in-
crease in MAD for the enol tautomer and a decrease for the keto
tautomer with respect to values for species c. Moreover, while
is calculated as
dxcalc
¼
rben
ꢁ
rx þ dben
;
ð1Þ
Table 2
Chemical shifts (ppm) of selected atoms for species a and c. Experimental and calculated (both enol and keto tautomers) in CDCl3 and DMSO. Calculated isotropic magnetic
shielding values for benzene used in Eq. (1) are 49.55 and 49.45 for CDCl3 and DMSO, respectively.
Atom
CDCl3
DMSO
dexp
denol
dketo
denol ꢁ dketo
dexp
denol
dketo
denol ꢁ dketo
a
c
C2
C7
161.11
162.64
148.48
117.22
166.02
162.47
152.33
117.43
184.24
147.76
140.71
126.22
ꢁ18.23
+14.71
+11.62
ꢁ8.79
ꢁ17.87
+17.60
+11.83
ꢁ9.15
160.26
163.49
148.06
116.57
166.02
162.03
151.87
117.31
184.41
147.50
140.44
126.09
ꢁ18.39
+14.53
+11.43
ꢁ8.79
ꢁ17.99
+17.32
+11.69
ꢁ9.19
0
C1
C3
C2
C7
161.13
163.68
149.83
117.33
166.34
164.16
153.42
117.54
184.21
146.56
141.58
126.70
160.19
164.69
149.86
116.61
166.35
163.88
153.10
117.39
184.34
146.56
141.41
126.59
0
C1
C3
Table 3
Chemical shifts (ppm) of selected atoms for species b and d. Experimental and calculated (both enol and keto tautomers) in CDCl3 and DMSO. Calculated isotropic magnetic
shielding values for benzene used in Eq. (1) are 49.55 and 49.45 for CDCl3 and DMSO, respectively.
Atom
CDCl3
DMSO
dexp
denol
dketo
denol ꢁ dketo
dexp
denol
dketo
denol ꢁ dketo
b
d
C2
C11
171.23
154.15
144.75
122.54
167.35
158.11
152.67
119.40
185.91
139.10
141.14
128.45
ꢁ18.57
+19.00
+11.53
ꢁ9.05
ꢁ18.67
+18.70
+11.06
ꢁ9.33
171.16
155.25
143.60
122.39
167.39
157.61
152.15
119.30
186.12
139.09
140.98
128.32
ꢁ18.73
+18.52
+11.17
ꢁ9.02
ꢁ18.80
+18.32
+12. 90
ꢁ9.34
0
C1
C3
C2
C11
168.28
156.43
147.64
121.37
167.57
159.24
153.99
119.15
186.23
140.54
142.93
128.47
169.01
157.08
146.04
121.63
167.57
158.82
155.68
119.04
186.37
140.51
142.78
128.38
0
C1
C3

604
M. Flores-Leonar et al. / Journal of Molecular Structure 1006 (2011) 600–605
enol
theo
ceto
theo
a
enol
theo
ceto
theo
a
exp
b
exp
b
C2
exp
c
exp
c
exp
d
exp
d
exp
exp
C7/11
C3
190
180
170
160
150
140
130
120
110
190
180
170
160
150
140
130
120
110
Chemical shift (ppm)
Chemical shift (ppm)
(b)
(a)
Fig. 1. Chemical shifts for the four carbon atoms that differ the most between the enol and keto tautomers. Green (red) bars correspond to enol (keto) predicted chemical
shifts for species a, b, c, d. Blue lines correspond to experimental values for species a and c, while black lines to experimental values for species b and d. (a) Predicted chemical
shifts from DFT/GIAO calculations. In (b) predicted chemical shifts have been displaced by a constant to agree with the experimental value for species c. The displacement
needed is less than 7 ppm for all four carbon atoms. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
for species a and c experimental values were mostly outside the
range between enol and keto predictions, for species b and d all
experimental values are inside this range. This confirms the exper-
imental interpretation (based both on 1H NMR and 13C, see above)
that tautomeric equilibrium is present in solution for species b and
d.
Predictions from DFT/GIAO calculations are not exact, although
they are precise enough to point to the presence of tautomeric
equilibrium for species b and d, as shown in Fig. 1a. We assume
that a constant shift can be used to adjust the calculated chemical
shift for species c, i.e. enol only; and then apply this shift to all
other species. All experimental measures for species b and d lie
in the middle of the interval between the chemical shifts corre-
sponding to enol and keto tautomers, as shown in Fig. 1b. We
can thus calculate the equilibrium constant,
5. Discussion and conclusions
Enol and keto tautomers are very close in energy for Schiff bases
and thus compete closely for stability. While the keto form is dis-
favored by the destruction of the aromatic ring as compared to the
enol form, the creation of a more stable Oꢂ ꢂ ꢂHAN hydrogen bond
(as compared to OAHꢂ ꢂ ꢂN) could be argued in favor of the keto over
the enol form. According to the present results both experimental
and theoretical, the stability depends only on the existence of aro-
maticity, even though the polarity of the solvent stabilizes the keto
form.
The creation of the keto form always destroy the aromaticity of
the nearby phenyl ring. Therefore, the keto form is normally less
stable than the enol form. But if a second aromatic ring is present,
and the change from enol to keto maintains the aromaticity of this
second ring, the keto form will be the stable one. It has been known
for a long time that naphthol Schiff bases derivatives (which pre-
serve an aromatic ring in the keto form) can easily react through
its keto form, while phenol Schiff bases derivatives are more diffi-
cult to react in this way (Bucherer reaction [19]). In naphthol based
compounds an aromatic ring is preserved, stabilizing the interme-
diate species, while in phenol based ones this is not possible. This
explanation applies to the equilibria of structures a, c, e, g, and i,
phenol based, for which the enol form is always stable; and to
structures b, d, f, h, and l, naphthol based, for which the keto form
is more stable. A special case is structure k, which although naph-
thol based, cannot keep its aromaticity in the keto form an thus fa-
vors the enol form.
The definition of an universal parameter that accounts for aro-
maticity has been sought for extensively but is still lacking [20].
Amongst the numerous parameters suggested, the most used are
the nucleus independent chemical shift (NICS) [21], the harmonic
oscillator measure of aromaticity (HOMA) [22], and the aromatic
stabilization energy [23]; notwithstanding the use of the HOMO–
LUMO gap for the estimation of aromaticity [24]. For the estima-
tion of the change in aromaticity between the enol and keto
tautomers in our Schiff bases we use the HOMO–LUMO gap ob-
tained from DFT calculations. For species a, c, e, and g (phenol
based) the average reduction in the HOMO–LUMO gap is
0.0317 Ha (i.e. the reduction in aromaticity) corresponding to a
more stable enol form. Whereas for species b, d, f, and h (naphthol
based) the average reduction in the HOMO–LUMO gap is 0.0142 Ha
a
;exp
vg
ꢁ dei nol;a
j
j
X
jdi
keto
enol
1
4
Keq
¼
¼
;
ð2Þ
jdki eto;a ꢁ dai
v
g
;exp
i
a
;exp
0
where i ¼ fC2; C7=11; C3; C1 g; di
is the experimental value for spe-
cies
a
= {b,d}, and denol(keto),avg is the average of calculated values for
enol (keto) species a, b, c, d. For species b the calculated equilibrium
constant corresponds to 53% (58%) keto tautomer present in solu-
tion for CDCl3 (DMSO), while for species d, corresponds to 36%
(45%), in agreement with the total energy calculations that predict
a more stable keto tautomer for species b than d and for solution
in DMSO than in CDCl3.
As an example of the precision required for total energy differ-
ences to match this prediction of the equilibrium constant we take
species d. According to a simple Boltzmann distribution of the rel-
ative abundance of the keto and enol tautomers at room tempera-
ture, based on the energy differences calculated using DFT (Table 1,
species d), equilibrium constants should correspond to 71% (89%)
keto tautomer presence in solution for CDCl3 (DMSO), that is,
two times bigger than estimations based on DFT/GIAO adjusted
values. To make total energy calculations match DFT/GIAO estima-
tions, the energy difference between keto and enol tautomers
should change from ꢁ2.2(ꢁ5.2) kJ/mol to +1.4 (+0.5) kJ/mol for
CDCl3 (DMSO). These changes in energy differences are smaller
than current precision of DFT based calculations.

M. Flores-Leonar et al. / Journal of Molecular Structure 1006 (2011) 600–605
[3] K. Gupta, A. Sutar, Coord. Chem. Rev. 252 (2008) 1420–1450.
605
corresponding to a much smaller reduction of the aromaticity and
thus favoring the stability of the keto form. Accordingly, species i
and k show a reduction of the HOMO–LUMO gap in going from
enol to keto forms of 0.0245 Ha and 0.0344 Ha; such large reduc-
tion correlates with a more stable enol form. Finally, species j
and l show small reductions of 0.0067 Ha and 0.0096 Ha and also
a much stabler keto form.
Schiff bases present a subtle tautomeric equilibria between the
enol and keto forms. The energy difference between these tautom-
ers is very close to the precision limit of current quantum mechan-
ical calculations. Even though, theoretical results allow for the
prediction of the presence of both tautomers in solution. This pre-
diction has been confirmed by experimental analysis of the synthe-
sized compounds. The use of theoretical estimations of chemical
shifts, based on the GIAO method, allows for the interpretation of
NMR spectra.
[4] M. Yildiz, Z. Kiliç, T. Hökelek, J. Mol. Struct. 441 (1998) 1–10.
[5] T. Hökelek, Z. Kiliç, M. Isßiklan, M. Toy, J. Mol. Struct. 523 (2000) 61–69.
[6] R. Hernández-Molina, A. Mederos, Comprehensive Coordination Chemistry II,
Pergamon, pp. 411–446.
[7] H. Nazir, M. Yildiz, H. Yilmaz, M. Tahir, D. Ülkü, J. Mol. Struct. 524 (2000) 241–
250.
[8] J. Costamagna, J. Vargas, R. Latorre, A. Alvarado, G. Mena, Coord. Chem. Rev.
119 (1992) 67–88.
[9] S.H. Alarcón, A.C. Olivieri, M. González-Sierra, J. Chem. Soc., Perkin Trans. 2
(1994) 1067–1070.
[10] S.H. Alarcón, A.C. Olivieri, D. Sanz, R.M. Claramunt, J. Elguero, J. Mol. Struct. 705
(2004) 1–9.
[11] A. Becke, J. Chem. Phys. 98 (1993) 5648–5652.
[12] C. Lee, W. Yang, R. Parr, Phys. Rev. B 37 (1988) 785–789.
[13] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
G. Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M.
Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, T. Vreven, J.J.A. Montgomery, J.E. Peralta, F. Ogliaro,
M. Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, O. Farkas, J.B. Foresman,
J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.02, 2009. Gaussian,
Inc., Wallingf.
[14] R. Ditchfield, J. Chem. Phys. 56 (1972) 5688–5691.
[15] R. Claramunt, C. López, M.S. Mar´ıa, D. Sanz, J. Elguero, Prog. Nucl. Magn. Reson.
Spectrosc. 49 (2006) 169–206.
[16] R. Jain, T. Bally, P.R. Rablen, J. Org. Chem. 74 (2009) 4017–4023.
[17] A. Sarotti, S. Pellegrinet, J. Org. Chem. 74 (2009) 7254–7260.
[18] H.E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 62 (1997) 7512–7515.
[19] H.T. Bucherer, J. Prakt. Chem. 69 (1904) 49–91.
Acknowledgements
C. A.-B. is thankful to DGAPA–UNAM for financial support (PAP-
IIT IN206507) and to DGTIC–UNAM for supercomputer access. A.
M.-B. acknowledges CONACyT through Project 41933-Q.
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.molstruc.2011.10.011.
[20] M. Alonso, B. Herradón, J. Comput. Chem. 31 (2010) 917–928.
[21] P. Schleyer, C. Maerker, A. Dransfeld, H. Jiao, N. Hommes, J. Am. Chem. Soc. 118
(1996) 6317–6318.
[22] T. Krygowski, J. Chem. Inf. Comp. Sci. 33 (1993) 70–78.
[23] M. Cyranski, P. Schleyer, T. Krygowski, H. Jiao, G. Hohlneicher, Tetrahedron 59
(2003) 1657–1665.
References
[1] H. Okawa, H. Furutachi, D. Fenton, Coord. Chem. Rev. 174 (1998) 51–75.
[2] N. Youssef, E. El-Zahany, A. El-Seidy, A. Caselli, S. Fantauzzi, S. Cenini, Inorg.
Chim. Acta 362 (2009) 2006–2014.
[24] F. De Proft, P. Geerlings, Chem. Rev. 101 (2001) 1451–1464.