Tautomerism in N-(2-hydroxy-1-naphthylidene)amino acids and the search for an answer to the difficult question about where the proton belongs

Article abstract of DOI:10.1016/j.molstruc.2016.03.040

N-(2-hydroxy-1-naphthylidene)-l-valine 1, N-(2-hydroxy-1-naphthylidene)-l-phenylalanine 2, and N-(2-hydroxy-1-naphthylidene)-l-threonine 3 were prepared and characterized with spectroscopic methods, elemental analyses, and values of optical rotation. Comp

Full text of DOI:10.1016/j.molstruc.2016.03.040

Accepted Manuscript
Tautomerism in N-(2-hydroxy-1-naphthylidene)amino acids and the search for an
answer to the difficult question about where the proton belongs
Gisela Warncke, Sabine Fels, Erica Brendler, Uwe Böhme
PII:
S0022-2860(16)30235-6
10.1016/j.molstruc.2016.03.040
MOLSTR 22351
DOI:
Reference:
To appear in: Journal of Molecular Structure
Received Date: 14 December 2015
Revised Date: 14 March 2016
Accepted Date: 14 March 2016
Please cite this article as: G. Warncke, S. Fels, E. Brendler, U. Böhme, Tautomerism in N-(2-hydroxy-1-
naphthylidene)amino acids and the search for an answer to the difficult question about where the proton
belongs, Journal of Molecular Structure (2016), doi: 10.1016/j.molstruc.2016.03.040.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical Abstract
1

ACCEPTED MANUSCRIPT
Tautomerism in N-(2-hydroxy-1-naphthylidene)amino acids and
the search for an answer to the difficult question about where the
proton belongs
Gisela Warnckea, Sabine Felsa, Erica Brendlerb, Uwe Böhmea*
a TU Bergakademie Freiberg, Institut für Anorganische Chemie, Leipziger Straße 29, 09596
Freiberg. Tel: +49 3731 39 2050; Fax: +49 3731 394058.
b TU Bergakademie Freiberg, Institut für Analytische Chemie, Leipziger Straße 29, 09596
Freiberg. Tel: +49 3731 39 2266; Fax: +49 3731 393666.
* E-Mail: Uwe.Boehme@chemie.tu-freiberg.de.
Abstract
N-(2-hydroxy-1-naphthylidene)-
L-valine 1, N-(2-hydroxy-1-naphthylidene)-
L-phenylalanine
2, and N-(2-hydroxy-1-naphthylidene)-
L-threonine 3 were prepared and characterized with
spectroscopic methods, elemental analyses, and values of optical rotation. Compound 1
undergoes a solid state order-disorder phase transition at 231 K. The X-ray structures of the
high and low temperature phase of 1 have been determined. Single crystal X-ray structures of
2 and 3 have been determined as well. The tautomerism of N-(2-hydroxy-1-
naphthylidene)amino acid derivatives is discussed controversial in the literature. A bond
lengths statistical analysis shows that all three compounds exist uniformly in the keto-amine
form in the solid state. Quantum chemical calculations, NMR, and UV-Vis spectroscopy were
used to obtain further insight into the existence of phenol-imine and keto-amine structures in
this class of compounds.
Keywords: Schiff base, chiral ligand, tautomerism, X-ray structure, NMR spectroscopy
2

ACCEPTED MANUSCRIPT
1. Introduction
Schiff base ligands have proven useful in silicon and tin complex formation, with the products
showing antimicrobial [1], bacteriostatic [2], bacteriocidal [3], biocidal [4] and fungicidal [5,
6] properties as well as antitumor activity [6]. After research on Salen-type [7, 8, 9, 10,] and
tridentate O,N,O’-ligands in silicon and tin complexes [11, 12, 13, 14, 15,] we are working on
hypercoordinated silicon and tin complexes with chiral ligands [16, 17, 18]. Our recent work
is focused on 2-hydroxynaphthaldehyde derivatives shown in Figure 1 [19]. Such ligands are
already known for their complexation abilities for lanthanum and cerium [20], zirconium [21],
manganese [22, 23, 24, 25, 26], iron [27, 28, 29], cobalt [30], nickel [31, 30], copper [32, 30],
zinc [33], cadmium [33], silicon [1], and tin [19, 34].
In contrast to the interest in the coordination chemistry of N-(2-hydroxy-1-
naphtylidene)amino acids only a few papers deal with characterization and properties of the
free Schiff bases [29, 35]. This may arise from the fact that in many studies the Schiff bases
are formed in situ during the complexation reactions. Herein we report the syntheses and
characterization of three N-(2-hydroxy-1-naphthylidene)amino acids (Figure 1). The question
whether these compounds exist in phenol-imine form (PI) or keto-amine form (KA) in the
solid state is investigated on the basis of the X-ray structural data in combination with bond
lengths statistics drawn from the Cambridge Structural Database. NMR and UV-Vis
spectroscopy were used to obtain further insight into the occurrence of the tautomeric forms
in solution and quantum chemical calculations to discover the relation between these two
tautomers.
O
15
14
O
13
R
1 R =
12
OH
O
R
16
OH
12
MeOH,
rflx.
NH
2
H
N
Ph
+
H
O
11
2 R =
3 R =
- H O
H
8
5
14
2
12
12
9
1
4
7
6
2
15
14
OH
3
10
OH
Figure1: Preparation of 2-hydroxy-1-naphthylidene-amino acids 1, 2 and 3 from L-valine (R = i-
propyl), L-phenylalanine (R = CH2-Ph), and L-threonine (R = *CHOH-CH3), respectively.
3

ACCEPTED MANUSCRIPT
2. Experimental
2.1 Materials and Methods
The necessary chemicals were used as commercially available. Solvents were dried according
to standard procedures. Melting points were determined with a Polytherm A from Wagner &
Munz using samples in sealed capillaries.
Solution state NMR spectra were recorded on a Bruker DPX 400 and Avance III 500
spectrometer at 293 K [1H (400.13 MHz), 13C (100.61 MHz) and 1H (500.13 MHz), 13C
(125.76 MHz), respectively] with tetramethylsilane as internal standard. The concentration of
the samples varied from 0.4 to 40 mg/ ml depending on solubility and scope of measurement,
for routine characterization 10 mg /ml were used. 8 to 128 (1H) and 128 to 16 k (13C) scans
were accumulated, 2D spectra were recorded with 128 or 256 t1-increments. The assignment
of the signals was performed using H,H-COSY, H,H-NOESY, H,C-HSQC and HMBC
experiments and is consistent to the assignment made by Rozwadowski et al. [36].
13C CP MAS NMR measurements were carried out at 100.6 MHz on a Bruker AVIII HD 400
WB spectrometer with a DVT probe using 4 mm ZrO2 rotors at 10 kHz spinning speed. A
contact time of 2 ms with an 70% ramp on the 1H channel was applied for CP, experiment
recycle delay of 5 s, 1k scans and an acquisition time of 50 ms with tppm15 decoupling were
used. The chemical shift was referenced externally using the CH-group signal of adamantane
(38.5 ppm with respect to TMS= 0 ppm).
Elemental analyses were performed with Foss Heraeus CHN-O-Rapid, UV-Vis spectra with
Jasco V-650 spectrophotometer, and Polarimetry with Perkin Elmer Polarimeter 241.
Concentration of UV-Vis samples were 0.2 mmol/l. The stability of optical rotation values has
been tested up to 120 hours after preparing the solution. No substantial deviations were
observed within experimental error of ±0.5°.
2.2 Synthesis of N-(2-hydroxy-1-naphthylidene)-
L-valine (1)
2-Hydroxy-1-naphthaldehyde (30 mmol, 2.58 g) was added to a suspension of
L-valine (20
mmol, 1.17 g) in methanol/ethanol (20 ml/250 ml). The reaction mixture was stirred under
reflux for 6 h, while it slowly turned yellow. After cooling to room temperature, the solution
was stirred at room temperature for another 24 hours. The reaction mixture was filtered and
the yellow solution concentrated in vacuo. Diethyl ether (100 ml) was added to the yellow
solid and stirred at room temperature in order to remove 2-hydroxy-1-naphthaldehyde. The
yellow suspension was filtered. The remaining yellow solid was washed with diethyl ether
and dried in vacuo. Yield: 4.40 g (81%). m.p. = 185-189 °C (literature [35] 187 °C). 1H NMR
4

ACCEPTED MANUSCRIPT
(DMSO, ppm): δ = 14.49 (s, 1H, C2-OH), 13.37 (s, 1H, C13-OH), 9.15 (s, 1H, H11), 8.07 (d,
1H, H8, 3JH-H = 8.4 Hz), 7.77 (d, 1H, H4, 3JH-H = 9.3 Hz), 7.66 (dd, 1H, H5, 3JH-H = 7.9 Hz,
4JH-H = 1.3 Hz), 7.43 (m, 1H, H7, 3JH-H = 8.4 Hz, 7 Hz, 4JH-H = 1.3 Hz), 7.22 (m, 1H, H6, 3JH-H
= 7 Hz, 7.9 Hz, 4JH-H = 1 Hz), 6.78 (d, 1H, H3, 3JH-H = 9.3 Hz), 4.34 (d, 1H, H12, 3JH-H = 4.3
Hz), 2.34 (m, 1H, H14, 3JH-H = 4.3 Hz, 3JH-H = 6.9 Hz); 0.96 (d, 6H, H15/H16, 3JH-H = 6.9 Hz).
13C NMR (δ, ppm): δ = 176.1 (C2), 171.8 (C13), 159.7 (C11), 137.2 (C4), 134.1 (C9), 129
(C5), 128 (C7), 125.5 (C10), 124.8 (C3), 122.5 (C6), 118.6 (C8), 106.3 (C1), 69.3 (C12), 31.1
(C14), 17.1/18.9 (C15/C16). 13C CP/MAS NMR (δ, ppm): δ = 178.6 (C2), 175.0 (C13), 161.0
(C11), 137.7 (C4), 134.4 (C9), 128.2 (C5), 127.0 (C7), 125.7 (C10), 123.4 (C3), 122.3 (C6),
118.3 (C8), 106.3 (C1), 68.1 (C12), 32.7 (C14), 22.2 (C15/C16).
[α]D20 = -109° (0.1 g / 100 ml DMSO). UV-Vis in DMSO, λ [nm] (ε in l/(mol*cm)): 307
(11840), 403 (10150), 423 (10340). Anal. calcd. for C16H17NO3: C 70.83, H 6.32, N 5.16;
found: C 70.78, H 6.26, N 5.15.
2.3 Synthesis of N-(2-hydroxy-1-naphthylidene)-
2-Hydroxy-1-naphthaldehyde (22.5 mmol, 3.87 g) was added to a suspension of
-phenylalanine (15 mml, 2.48 g) in methanol/ethanol (20 ml/150 ml). The yellow suspension
L-phenylalanine (2)
L
was stirred at room temperature for 18 h, followed by 6 hours under reflux. After one hour a
yellow precipitate formed from the hot yellow solution. After cooling to room temperature,
the solution was stirred at room temperature for two more days. The solid was filtered,
washed twice with 10 ml methanol and dried in vacuo. Yield: 3.58 g (75%). m.p. = 184-186
°C (literature [35] 170 °C). . 1H NMR (DMSO, ppm): δ = 14.29 (s, 1H, C2-OH), 13.44 (s,
1H, C13-OH), 8.95 (s, 1H, H11), 7.90 (d, 1H, H8, 3JH-H = 8.4 Hz), 7.76 (d, 1H, H4, 3JH-H = 9.3
Hz), 7.66 (dd, 1H, H5, 3JH-H = 7.9 Hz, 4JH-H = 1.4 Hz), 7.43 (m, 1H, H7, 3JH-H = 7 Hz; 8.4 Hz,
4JH-H = 1.4 Hz), 7.30-7.18 (m, 6H, Ph-Har + H6), 6.78 (d, 1H, H3, 3JH-H = 9.3 Hz), 4.77 (dd,
1H, H12, 3JH-H = 5.0 Hz; 8.3 Hz), 3.37 (dd, 1H, H14b, 3JH-H = 5.0 Hz, 2JH-H = 13.9 Hz), 3.20
(dd, 1H, H14a, 3JH-H = 8.3 Hz, 2JH-H = 13.9 Hz). 13C NMR (DMSO, ppm): δ = 174.9 (C2),
171.7 (C13), 159.5 (C11), 137 (C4), 136.4 (ipso-Ph), 133.8 (C9), 129.5 (o-Ph), 128.9 (C5),
128.4 (m-Ph), 127.9 (C7), 126.8 (p-Ph), 125.6 (C10), 124.3 (C3), 122.6 (C6), 118.6 (C8),
106.2 (C1), 65.2 (C12). [α]D20 = 0° (1 g / 100 ml DMSO). UV-Vis in DMSO, λ [nm] (ε in
l/(mol*cm)): 307 (14840), 404 (11030), 425 (11180). Anal. calcd. for CHNO: C 75.22, H
5.37, N 4.39; found: C 72.67, H 5.01, N 4.25.
5

ACCEPTED MANUSCRIPT
2.4 Synthesis of N-(2-hydroxy-1-naphthylidene)-
2-Hydroxy-1-naphthaldehyde (20 mmol, 3.40g) was dissolved in methanol (200 ml).
-threonine (20 mmol, 2.40g) was added and the reaction mixture was stirred under reflux for
L-threonine (3)
L
3 h. The yellow solution was filtered and allowed to cool to room temperature. While standing
overnight, fine yellow needles crystallized and were collected for x-ray diffraction analysis.
The yellow solution was concentrated in a vacuum to on third of its volume whereby a yellow
solid precipitated. The solid was filtered off and dried in a desiccator in vacuum over CaCl2.
Yield: 3.16g (58%). m.p. = 188-190 °C. 1H NMR (DMSO, ppm): δ = 13.82 (s, 1H, C2-OH),
13.11 (s, 1H; C13-OH), 9.02 (s, 1H, H11), 7.98 (d, 1H, H8, 3JH-H = 8.4 Hz), 7.72 (d, 1H, H4,
3JH-H = 9.4 Hz), 7.61 (dd, 1H, H5, 3JH-H = 7.9 Hz, 4JH-H = 1.4 Hz), 7.41 (m, 1H, H7, 3JH-H = 7
Hz; 8.4 Hz, 4JH-H = 1.4 Hz), 7.18 (m, 1H, H6, 3JH-H = 7 Hz; 7.9 Hz, 4JH-H = 1 Hz), 6.70 (d, 1H,
H3, 3JH-H = 9.4 Hz), 4.36 (d,1H, H12), 4.31 (m, 1H, H14), 1.14 (d, 3H, H15, 3JH-H = 6.4 Hz).
13C NMR (DMSO, ppm): δ = 178.6 (C2), 171.2 (C13), 159.2 (C11), 137.7 (C4), 134.6 (C9),
129.0 (C5), 128.1 (C7), 126.1 (C3), 125.2 (C10), 122.3 (C6), 118.3 (C8), 105.8 (C1), 68.9
(C12), 66.6 (C14), 20.5 (C15). [α]D20 = -10.5° (0.2g/100 ml DMSO). UV-Vis in DMSO, λ
[nm] (ε in l/(mol*cm)): 306 (13630), 404 (13010), 424 (13950). Anal. calcd. for C15H15NO4:
C 65.91, H 5.54, N 5.13; found: C 66.18, H 5.56, N 5.10.
2.5 X-ray structure analyses
Single crystals of the compounds suitable for X-ray structure analysis were grown from
methanol solution at 276 K, mounted on a glass fiber and transferred to the nitrogen gas
stream of the diffractometer (Stoe IPDS 2T; graphite-monochromated Mo-Kα radiation, λ =
0.71073 Å). The structures were solved by direct methods (SHELXS97) [37] and refined by
full-matrix least-squares methods on F2 for all unique reflections (SHELXL2014) [38]. The
position of the hydrogen atom between the phenolic oxygen atom O1 and the nitrogen atom
N1 was identified from residual electron density in a final stage of the refinement process in
all three compounds (1-3). Refinement of this hydrogen atom position was performed without
any restraints. The position of the carboxylic hydrogen atoms in 1-3 and the hydrogen atom of
the threonine group in 3 was also localized from residual electron density. All other hydrogen
atoms were positioned geometrically and refined with a riding model. The compounds 1-lt, 1-
ht, and 3 crystallize in the chiral space group P21. The Flack-parameter for these structures is
not reliable, since these are light atom structures. Crystallographic data are summarised in
Table 1.
6

ACCEPTED MANUSCRIPT
Table 1: Crystallographic Data of Compounds 1-lt, 1-ht, 2, and 3.
3 [39]
1-lt
1-ht
2
3
Empirical formula
Formula weight
C16H17NO3
271.31
C16H17NO3
271.31
C20H17NO3
319.35
C15H15NO4
273.28
T (K)
153(2)
253(2)
294(2)
153(2)
r.t.
λ (Å)
0.71073
monoclinic
P21
0.71073
monoclinic
P21
0.71073
monoclinic
P21/c
0.71073
monoclinic
P21
Crystal system
Space group
a (Å)
16.194(1)
10.243(1)
16.703(1)
90
5.3493(5)
10.4044(1)
12.7115(1)
90
13.908(1)
10.761(1)
11.323(1)
90
5.0640(2)
11.2479(4)
11.0867(4)
90
5.109(2)
11.334(2)
11.155(3)
b (Å)
c (Å)
α (°)
β (°)
93.609(4)
90
94.690(7)
90
108.041(4)
90
91.590(2)
90
91(3)
γ (°)
Volume (Å3), Z
2765.1(3), 8
705.11(7), 2
1611.27(4), 4
631.25(4), 2
645.8(3), 2
1.405
Calculated density
(g cm-3)
1.303
1.278
1.316
1.438
Absorption
0.090
0.088
0.089
0.105
0.103
coefficient (mm-1)
F(000)
1152
288
672
288
θ Range for data
collection (°)
2.52-27.09
3.22-27.42
1.54-27.50
2.58-28.00
-20<h<20
-13<k<13
-21<l<21
-6<h<6
0<h<18
-6<h<6
Limiting indices
-13<k<13
-16<l<16
3160/33/216
-13<k<13
-14<l<13
3705/1/234
-14<k<14
-14<l<14
3042/1/194
Data/restraints/
parameters
11815/4/777
1373/-/-
Total/unique
reflections
23665/11815
6676/3160
7203/3705
15115/3042
Rint
0.0696
99.5
0.0315
99.8
0.0302
100.0
0.0206
99.9
Completeness
Refinement method
Full-matrix least squares on F²
0.990
Goodness-of-fit on
F²
1.09
1.089
1.05
Final R index
[l>2σ(l)]
R1 = 0.0506,
wR2 = 0.0899
R1 = 0.0406,
wR2 = 0.0887
R1 = 0.0398,
wR2 = 0.0987
R1 = 0.0266,
wR2 = 0.0713 wR2 = 0.1058
R1 = 0.0438,
R1 = 0.0921,
wR2 = 0.1077
R1 = 0.0599,
wR2 = 0.1010
R1 = 0.0568,
wR2 = 0.1069
R1 = 0.0272,
wR2 = 0.0719
R index [all data]
Largest difference
peak/ hole (e Å-3)
0.172/ -0.176
0.159/ -0.143
0.191/ -0.175
0.214/ -0.173
0.286/ -0.274
7

ACCEPTED MANUSCRIPT
2.6 Quantum chemical calculations
The DFT calculations were carried out using GAUSSIAN 09 [40]. Geometries were fully
optimized at the density functional theory level (DFT), using Becke’s three-parameter hybrid
exchange functional and the correlation functional of Lee, Yang and Parr (B3LYP) [41, 42].
Geometry optimizations and harmonic frequencies were calculated for all elements with the
polarized 6-31G(d,p) basis set [43, 44, 45].
Relaxed potential energy surface scans have been performed with the Opt=ModRedundant
utility in Gaussian 09 with B3LYP/6-31G(d,p). This option includes the specification of
redundant internal coordinates. In these cases, a specific torsion angle has been changed in 5
degree steps. On every step the geometry of the molecule was completely optimized,
restricting only the torsion angle to the specified value. This method allows access to a
defined section of the potential energy surface.
Solvent model was applied with the polarizable continuum model (PCM) [46]. The same
methodology as for the gas phase structures was applied here.
3. Results and discussion
The compounds 1-3 (Figure 1) were prepared similar to procedures previously described in
the literature [29, 35].
3.1 X-ray structures
ORTEP-drawings of 1-ht, 1-lt, 2, and 3 with atomic numbering schemes are shown in Figures
2, 3, 5, and 6. A summary of crystallographic data can be found in Table 1. During the X-ray
diffraction experiment on N-(2-hydroxy-naphthylidene)-L-valine (1), a reversible phase
transition was observed at 231 K. The temperature of the phase transition was determined on
the diffractometer by repeated cell determinations at different temperatures. On cooling down
the crystal below the phase transition temperature new reflections arose which lead to a
quadruple cell volume. The phase transition is fully reversible, since the cell determination
was repeated several times above and below the transition temperature. All investigations
have been performed with the same crystal. The high- and low-temperature forms of 1 have
been investigated in detail. The crystal structure of the high-temperature phase has been
determined at 253 K (1-ht) and the structure of the low-temperature phase at 153 K (1-lt).
The high-temperature form 1-ht crystallizes in the monoclinic space group P21 with one
crystallographic independent molecule in the asymmetric unit. The isopropyl group is
8

ACCEPTED MANUSCRIPT
disordered (Figure 2). An intramolecular hydrogen bond is formed between N1-H1A...O1 The
distance N1…O1 is 2.547 Å and the angle N1-H1A-O1 is 143°. A second contact might be
discussed between N1-H1A…O2 with N1…O2 = 2.651 Å and N1-H1A-O2 = 82°. The long
distance N1...O2 and the acute angle hints to a weak interaction. There is one intermolecular
hydrogen bond present in the crystal lattice, i.e. O3-H…O1 with a distance O3…O1 of 2.509 Å
and an O3-H…O1 angle of 175°. This interaction influences also the molecular conformation.
Figure 2: Molecular structure of high temperature form 1-ht showing the atomic numbering scheme.
The thermal ellipsoids of the non-hydrogen atoms are drawn at the 50% probability level.
The low-temperature form 1-lt also crystallizes in the monoclinic space group P21. The unit
cell contains four crystallographically independent molecules (Figure 3). These molecules can
be described as four different conformational isomers. According to Buerger [47], 1
undergoes an order-disorder phase transition which is described as different atoms occupying
the same crystallographic sites or the same atoms statistically occupying different
crystallographic sites. The lengths of chemically equivalent bonds of the four molecules do
not differ significantly from each other. The bond lengths are also similar to the bond lengths
in 1-ht (Table 2). Due to the flexibility of the i-propyl group the four molecules differ by
torsion angles (Table 3) of the amino acid group. Newman projections of the i-propyl group in
1-lt molecules are shown in Figure 4. Molecules a and d are represented by the left drawing
and molecules b and c by the right drawing, respectively.
Similar as in the high temperature form bifurcated intramolecular hydrogen bonds are found
in the low temperature form. The shorter interaction (major component) is always between the
keto oxygen atoms bound to the naphthylidene ring and the imine hydrogen atom. Distances
9

ACCEPTED MANUSCRIPT
vary from 2.537 Å for N2…O4 to 2.581 Å for N4…O10. The angles N-H…O are between 138
and 140°. Longer N…O distances between 2.561 and 2.753 Å and acute angles from 78 to 86°
are observed for the interaction between the imine hydrogen and one oxygen atom of the
carboxyl group (minor component). Therefore this part of the bifurcated hydrogen bonds
should be classified as very weak contacts. There exist intermolecular hydrogen bonds
between the carboxyl group of one Schiff base molecule and the oxygen atom at the
naphthylidene group of an adjacent Schiff base conformer. This is essentially the same
intermolecular interaction as in 1-ht. The oxygen-oxygen distances vary between 2.519-
2.551 Å and the oxygen-hydrogen-oxygen angles between 161 and 175°.
Table 2: Selected bond lengths [Å] and angles [°] of molecules 1-ht, 1-lt, 2, and 3.
1-lt a)
3
1-ht
2
(Özcan [39])
a
b
c
d
1.300(2)
(1.30(3))
1.428(2)
(1.43(1))
1.416(2)
(1.402(5))
1.307(2)
(1.31(3))
1.449(2)
1.540(2)
1.543(2)
O1-C2
C1-C2
1.295(4)
1.427(4)
1.413(4)
1.304(3)
1.292(5)
1.301(5)
1.309(5)
1.291(5)
1.300(2)
1.420(2)
1.410(2)
1.303(2)
1.439(5)
1.421(6)
1.311(5)
1.432(6)
1.415(6)
1.309(5)
1.429(6)
1.416(5)
1.314(5)
1.438(6)
1.415(5)
1.311(5)
C1-C11
C11-N1
N1-C12
C12-C13
C12-C14
1.460(3)
1.538(4)
1.543(4)
1.580(1)
1.515(6)
1.205(3)
1.299(4)
121.0(3)
119.3(3)
123.8(3)
126.4(2)
106.7(2)
109.5(2)
122.2(3)
113.0(2)
104.8(6)
115.4(3)
1.461(5)
1.536(5)
1.547(6)
1.459(5)
1.534(6)
1.547(6)
1.450(5)
1.548(6)
1.552(5)
1.460(5)
1.528(5)
1.553(5)
1.457(2)
1.523(2)
1.539(2)
C14-C15
1.524(6)
1.523(6)
1.532(6)
1.535(6)
1.511(2)
1.513(2)
C13-O2
1.218(4)
1.307(5)
121.2(3)
119.1(4)
123.8(4)
125.8(4)
108.3(3)
109.3(3)
122.4(4)
112.8(3)
1.213(5)
1.315(5)
121.0(3)
119.2(4)
123.3(4)
126.0(4)
107.3(3)
111.2(3)
122.4(4)
112.6(3)
1.215(5)
1.311(5)
121.2(4)
119.5(4)
123.3(4)
126.4(4)
107.5(3)
112.0(3)
121.6(4)
113.1(3)
1.220(5)
1.317(5)
120.9(4)
119.4(4)
124.6(4)
124.8(4)
107.8(3)
108.7(3)
122.7(4)
112.9(3)
1.300(2)
1.203(2)
121.0(1)
119.7(1)
124.9(1)
124.0(1)
112.1(1)
111.5(1)
114.5(1)
120.7(1)
1.300(2)
1.218(2)
120.9(1)
119.3(1)
122.2(1)
128.1(1)
110.7(1)
107.6(1)
112.0(1)
122.2(1)
C13-O3
C1-C2-O1
C11-C1-C2
C1-C11-N1
C11-N1-C12
N1-C12-C13
N1-C12-C14
C12-C13-O2
C12-C13-O3
C12-C14-C15
109.9(4)
109.8(3)
110.3(3)
110.1(3)
114.0(1)
113.6(1)
10

ACCEPTED MANUSCRIPT
a) Equivalent bonds and angles have been used for the crystallographic independent molecules
of 1-lt.
a
b
c
d
Figure 3: Molecular structures of the crystallographic independent molecules a-d of the low
temperature form 1-lt showing the atomic numbering scheme. The thermal ellipsoids of the non-
hydrogen atoms are drawn at the 50% probability level.
Table 3: Torsion angles [°] of the four crystallographically independent molecules in 1-lt and 1-ht.
1-lt a)
1-ht
molecule a molecule b molecule c
molecule d
-178.1(4)
174.6(4)
129.3(4)
-59.5(4)
C9-C1-C11-N1
C1-C11-N1-C12
-176.2(4)
172.5(4)
-176.1(4)
173.0(4)
117.1(4)
-53.8(4)
-177.2(4)
175.2(4)
112.4(4)
-65.6(4)
-176.8(3)
172.0(3)
131.8(3)
-162.2(8).
C11-N1-C12-C14 136.4(4)
N1-C12-C14-C15 -65.4(4)
11

ACCEPTED MANUSCRIPT
170.0(4)
-46.2(8).
-63.7(5)
-106.0(3)
-17.2(4)
162.3(3)
N1-C12-C14-C16 170.5(3)
C11-N1-C12-C13 -102.0(4)
70.3(4)
58.5(4)
177.8(3)
-118.9(4)
3.1(6)
-121.6(4)
16.0(6)
-107.5(4)
-15.0(5)
164.7(3)
N1-C12-C13-O2
N1-C12-C13-O3
-24.7(5)
156.4(3)
-176.7(4)
-164.5(3)
a) Equivalent torsion angles have been used for the crystallographic independent molecules of
1-lt.
Figure 4: Newman-projections of the i-propyl group in 1-lt. Molecules a and d refer to the left and b
and c to the right projection, respectively.
N-(2-hydroxy-1-naphthylidene)-L-phenylalanine (2) crystallizes in the monoclinic space
group P21/c with one crystallographically independent molecule in the asymmetric unit
(Figure 5). The lattice symmetry includes an inversion center, thus the compound is a racemic
mixture. This is also shown by the optical rotation of 0° of the bulk material. Selected bond
lengths and angles are given in Table 2. There is an intramolecular hydrogen bond between
N1-H1A...O1 with a distance N1…O1 = 2.588 Å, and an angle N1-H1…O1 = 136°. A second
contact might be discussed between N1-H1A…O2 with a distance N1…O2 = 2.627 Å and an
angle N1-H1…O2 = 82°. These are supplemented by one intermolecular interaction between
the carboxylic O2-H and the oxygen atom O1 at the naphthylidene group of a neighbouring
molecule with a distance O1…O2 of 2.514 Å and an angle O2-H2…O1 of 173°. This
interaction influences also the overall conformation of the molecule.
12

ACCEPTED MANUSCRIPT
Figure 5: Molecular structure of 2 showing the atomic numbering scheme. The thermal ellipsoids of
the non-hydrogen atoms are drawn at the 50% probability level.
N-(2-Hydroxy-naphthylidene)-
and with one crystallographically independent molecule in the asymmetric unit (Figure 6). A
crystal structure analysis of N-(2-hydroxy-1-naphthylidene)- -threonine was published
L-threonine (3) crystallizes in the monoclinic space group P21
L
already in 2003 by Özcan et al. [39]. Additionally we determined the value of optical rotation
(see above) and offer a qualitatively improved crystallographic dataset of the compound (R1 =
2.66% and R2 = 2.72% vs. R1 = 4.38% and R2 = 10.58%). There is a bifurcated intramolecular
hydrogen bond between N1-H1A...O1 (major component) and N1-H1A…O4 (minor
component) with distances N1…O1 = 2.516 Å, N1…O4 = 2.812 Å and angles N1-H1…O1 =
139°, N1-H1…O4 = 102°. There are two intermolecular hydrogen bonds which influence solid
state packing and molecule conformation. The O2-H…O1 hydrogen bond is approximately
equivalent to those in 1 and 2 and has a O2…O1 distance of 2.472 Å and an angle O2-H…O1
of 174°. Due to the additional hydroxyl group in the amino acid moiety, another
intermolecular hydrogen bond between the threonine hydroxyl group O4-H4 and O3 is
formed, with O3 referring to the double bound oxygen of the carboxylic group. The O4…O3
distance is 2.941 Å and the angle O4-H…O3 is 165°.
13

ACCEPTED MANUSCRIPT
Figure 6: Molecular structure of 3 showing the atomic numbering scheme. The thermal ellipsoids of
the non-hydrogen atoms are drawn at the 50% probability level.
Selected bond lengths and angles of compounds 1, 2, and 3 are given in Table 2. In general,
the corresponding values do not differ significantly between these three compounds. The
nature of the amino acid group does not have a substantial influence on the structural
parameters. In 3, the hydroxyl group in the amino acid unit causes more hydrogen bonds and
therefore a tighter packing in the crystal structure.
3.2 Values of optical rotation
Values of optical rotation were determined with [α]D20 = -109° (0.1 g / 100 ml DMSO) for 1,
[α]D20 = 0° for 2 (1 g / 100 ml DMSO), and [α]D20 = -10.5° (0.2 g/100 ml DMSO) for 3.
Therefore it can be assumed that 1 and 3 were prepared as chiral compounds under retention
of the stereochemistry of the amino acid. In contrast to that compound 2 suffered from
racemization. This process has already been observed for enantiomerically pure phenylalanine
esters in MeCN/H2O in presence of e.g. salicylaldehyde at pH 7, but not for the free
phenylalanine in presence of 1-hydroxy-2-naphthaldehyde in slightly basic conditions [48].
Being not as activated for racemization processes as arylglycine amino acid derivatives,
arylalanine equivalents are also able to stabilize the negative charge occurring after hydrogen
abstraction from the asymmetric carbon (C12).
3.3 1H and 13C NMR spectroscopy
In the 1H NMR spectra recorded in DMSO-d6 the hydrogen atom involved in the
intramolecular hydrogen bond has the highest chemical shift (O1…H…N1; 14.49 ppm for 1,
14

ACCEPTED MANUSCRIPT
14.29 ppm for 2, 13.82 ppm for 3, for atom numbering see Figure 1) followed by the
carboxylic hydrogen atom (13.37 for 1, 13.44 for 2, 13.11 for 3). This high shift for
O1…H…N1 hints towards a quite acidic hydrogen atom and therefore towards a hydrogen
bridge. The signal for the imine hydrogen H11 appears around 9 ppm (9.15 ppm for 1, 8.95
ppm for 2, 9.02 ppm for 3) and occurs at concentrations used for routine NMR
characterization as a singlet in all cases presented herein. 1H NMR shifts for hydrogen atoms
H3-H8 are situated between 6.6 and 8.0 ppm. Signals at 4.34 ppm for 1, 4.77 ppm for 2, 4.36
ppm for 3 belong to the hydrogen atom H12 at the asymmetric carbon C12.
In 1 the multiplet at 2.33 ppm belongs to H14 and is caused by the coupling to the two
diastereotopic methyl groups at about 0.96 ppm. The signal for H14 of 3 is at around 2.34
ppm as a multiplet due to coupling with H12 and the methyl group.
13C NMR shifts characteristic for N-(2-hydroxy-1-naphthylidene)amino acids occur at about
172 ppm for the carboxylic carbon C13 (171.8 ppm for 1; 171.7 ppm for 2; 171.2 ppm for 3),
for C11 at 159.7 ppm for 1; 159.5 ppm for 2; 159.2 ppm for 3 and for C2 at 176.1 ppm for 1;
174.9 ppm for 2; 178.6 ppm for 3. Nine signals between approximately 138 to 105 ppm
belong to the aromatic carbon atoms C1 and C3-C10. The 13C NMR chemical shift of C12 is
69.3 ppm, 65.2 ppm and 66.6 ppm for 1, 2, and 3 respectively. The two diastereotopic methyl
groups of the amino acid in 1 show slightly different chemical shifts of 17.2 and 19.0 ppm.
The chemical shift of the methyl group in 3 is found at 20.5 ppm.
These results are in accordance with literature data [49, 50].
3.4 Intramolecular hydrogen bonding and tautomerism in naphthylidene amino acid Schiff
bases 1-3
Phenol-imine (PI) and keto-amine structures (KA, Figure 7) may occur in N-(2-Hydroxy-1-
naphthylidene)amino acids and related salicylaldimine Schiff bases. This form of
intramolecular tautomerism has been discussed in numerous papers (For reviews and general
papers about this subject see [51, 52, 53, 54, 55] and literature cited therein). The electronic
absorption spectra of several N-(2-Hydroxy-1-naphthylidene)amino acids derived from
glycine, alanine, leucine, valine, and phenylalanine have been investigated in various solvents
by Ebead et. al. [35]. They stated that these compounds should exist predominately or
completely in the keto-amine form in solution. In contrast to that, Özcan et al. [39] found the
phenol-imine form in the solid state structure of N-(2-Hydroxy-1-naphthylidene)-l-threonine
(3). All hydrogen atoms were geometrical positioned in this previous structure determination
and refined with a riding model. For us it seems not appropriate to decide about the nature of
15

ACCEPTED MANUSCRIPT
a tautomer on the basis of a geometrical positioned hydrogen atom. Therefore we were
motivated to have a closer look at the solid state structures of N-(2-Hydroxy-1-
naphthylidene)amino acids.
PI
KA
Figure 7: Phenol-imine (PI) and keto-amine (KA) tautomeric form of N-(1-hydroxy-1-
nyphthylidene)amino acids.
3.4.1 Discussion of the X-ray structural data
The X-ray structure analyses of 1-3 deliver experimental facts about the preferred tautomeric
form in the solid state. The position of the hydrogen atom between the phenolic oxygen atom
O1 and the nitrogen atom N1 was identified from residual electron density during the
refinement process of all three X-ray structures. The further refinement without any restraints
indicates that the hydrogen atom is localised close to the nitrogen atom, thereby forming the
KA structure. However, one might argue that the position of hydrogen atoms cannot be
determined reliable during the X-ray experiment [56]. Therefore the bond lengths between the
surrounding non-hydrogen atoms O1-C2, C2-C1, C1-C11, and C11-N1 were included into the
discussion in order to obtain a reliable decision which tautomer is present in the solid state
structures. As can be seen in Table 2, these bond lengths do not vary significantly between the
X-ray structures of 1-ht, 1-lt, 2, and 3. Therefore, we conclude that all three Schiff bases have
the same tautomeric form in solid state.
Bond lengths statistics for both tautomeric forms were drawn from the Cambridge Structural
Database [57] in order to decide which tautomer is present in the solid state structures of 1-3.
For this purpose the fragments shown in Figure 8 have been drawn and searched for in the
16

ACCEPTED MANUSCRIPT
CSD. Please notice that these structural fragments include all N-(2-hydroxy-1-naphthylidene)-
and N-(2-hydroxy-1-salicylaldimine)-derivatives. There were 1998 hits for the phenol-imine
form (PI) and 104 hits for the keto-amine form (KA). The bond lengths shown in Figure 8
should vary considerably from PI to KA form due to the alternating bond character between
formal single and double bonds. Indeed there are significant differences between the bond
lengths in both forms (Table 4). The comparison with the mean values of the bond lengths of
1-3 shows undoubtedly the presence of the KA form. On the basis of this statistical analysis
we can conclude that the tautomeric form assigned by Özcan et al. [39] for 3 was wrong.
C12
N1
C12
N1
C11
11
C
H
H
C1
O1
C1
O1
C2
C2
PI
KA
Figure 8: Schematic drawing of phenol-imine (PI) and keto-amine (KA) form in N-(2-hydroxy-1-
naphthylidene)- and N-(2-hydroxy-1-salicylaldimine)-derivatives.
Table 4: Statistical Data from the CSD for the tautomers shown in Figure 8 and mean values for the X-ray
structures 1-ht, 2, and 3 in Å.
Tautomer
phenol-imine * 1.343 1.406 1.450
keto-amine * 1.291 1.431 1.414
mean values 1-3 1.298 1.425 1.414
O1-C2 C2-C1 C1-C11 C11-N1
1.286
1.303
1.305
*CSD search for distances, CSD version 5.36 (november 2014), search parameters: only
organics.
3.4.2 Considerations regarding 1H and 13C NMR spectroscopy in solution
In order to determine the existing tautomeric structure of Schiff bases in solution 1H and 13C
NMR spectroscopic experiments were already successfully performed on salicylaldehyde [53,
58, 59] and 2-hydroxy-1-naphthaldehyde derived Schiff bases [49, 50, 58]. In general, the
most characteristic NMR signals are that of imine hydrogen atom H11 and carbon atom C2
with regard to chemical shift and signal form. In 1H NMR spectra of the PI form the signal of
17

ACCEPTED MANUSCRIPT
imine hydrogen atom H11 appears as a singlet, whereas in the KA form a doublet is formed
due to the 3JH-H coupling to the hydrogen atom bound at N1. The signal of carbon atom C2
shifts downfield due to the change from a phenolic carbon, which is part of an aromatic
system, to a carbonyl carbon atom in the KA. From these observations in the spectra about
position and splitting of the signals it should be possible to draw a conclusion about the
predominant tautomeric form in solution.
Detailed 1H and 13C NMR spectroscopic experiments were performed using 1 as an example.
In DMSO solutions at concentrations used for routine characterization (about 10 mg/ml) H11
appears as a slightly broadened singlet, hydrogen N1-H-O1 as well as the carboxylic
hydrogen as a broad singlet.
The increased linewidth of these signals indicates inter- and/or intramolecular exchange
processes between the acidic hydrogen atoms of the compounds including the N1…H
environment. To reduce the exchange rate low temperature NMR measurements as presented
by Rozwadowski et al. [53] were intended in solvents having suitable melting points. For
comparability with the results from 5-nitrosylaldehyde derived Schiff bases [53] we first
chose chloroform as solvent. Unfortunately 1 has a very low solubility in chloroform making
the record of 13C NMR spectra impossible.
In the 1H NMR spectrum of 1 in deuterated acetone the signal of the imine hydrogen H11
appears even in highly concentrated solutions as a sharp singlet at 9.3 ppm, those of the acidic
hydrogens appear as very broad singlets. The 13C NMR signal of C2 has a shift of 172.1 ppm,
thus beeing significantly upfield compared to the chemical shift of C2 obtained in the solid
state CP MAS spectrum for the KA form with 178.6 ppm. Both indicate that the equilibrium
(see tautomeric forms in Figure 7) in acetone solution is on the side of the PI structure. For
that reason and as it was questionable if the achievable temperatures would be sufficient to
reduce the exchange rate to the slow exchange mode on the NMR time scale variable-
temperature NMR experiments did not seem promising.
Compound 1 shows a good solubility in DMSO. As the observed proton exchange reactions
can be both inter- as well as intramolecular, it was intended to reduce the less interesting
intermolecular exchange rates by diluting the solution. Therefore we performed NMR
measurements in the concentration range 0.4 – 40 mg/ml in DMSO-d6 at 400.13 MHz.
At concentrations of ≥ 20 mg / ml no splitting of the imine hydrogen H11 signal but an
increased linewidth compared to the aromatic protons is observed. Splitting starts at
10 mg / ml.
18

ACCEPTED MANUSCRIPT
Figure 9: Parts of the 1H NMR spectra of 1 in DMSO at different concentrations (from top to bottom: 40 mg,
20 mg, 10 mg, 2 mg, 0.4 mg / ml DMSO): a) phenol hydrogen O1-H; b) imine hydrogen H11; c) hydrogen atom
H12 at asymmetric carbon atom C12. Signals are scaled to comparable intensity.
At concentrations of 2 mg/ml and 0.4 mg/ml DMSO splitting of hydrogen signals H11 (s →
d), phenol hydrogen O1-H (s → t) and hydrogen atom H12 (d → t) is observed (Figure 9),
indicating that the phenolic proton is interacting with N1 giving rise to a coupling to H11 and
H12. For the lowest concentration a coupling constant of 8.4 Hz can be determined for the
H11 doublet. Additionally, the carboxyl hydrogen signal gets narrower with decreasing
concentrations. As expected in low concentrated solutions the intermolecular exchange
between the hydroxyl protons is reduced resulting in a longer dwell time of the hydrogen
atom in the intramolecular hydrogen bridge O1…H…N1, thus H-H coupling to H11 and H12
can now be made visible by H,H COSY NMR (Figure 10) and the coupling constant can be
determined.
The 13C NMR signal of the phenol carbon atom C2 did not experience significant changes by
changing concentrations. In conclusion it can be stated that the tautomeric equilibrium is not
noticeable influenced by the concentration of the solution.
19

ACCEPTED MANUSCRIPT
Figure 10: Section of the H,H COSY NMR spectrum of 1 (2 mg/ml in DMSO) showing
coupling of phenol hydrogen O1-H and the hydrogen atoms H11 (imine) and H12 (N-C*H).
Interestingly the splitting for H11, H12 and O1-H can also be induced by the use of wet
DMSO. Here the water quickly exchanges with the carboxyl groups, reducing the exchange
between the carboxyl group and the hydrogen in O1…H…N1. Thus the intramolecular
hydrogen bridge O1…H…N1 is stabilized.
A share of 62 to 64 % KA form can be estimated from the values of 3JH-H coupling at H11
(8.4 Hz compared to 13.1 Hz for molecules in pure KA form [53, 58]) and the 13C NMR shift
of carbon atom C2 with 176.2 ppm (compared to 178.6 ppm for the KA form in solid state
and 172.1 ppm for the PI form in aceton), respectively, at room temperature in DMSO. These
values fit – within the experimental error- very nicely to the results obtained by UV/VIS
spectroscopy.
By heating the solution of 1 with a concentration of 40 mg / ml DMSO an increasing proton
exchange between the carboxylic group and the intramolecular hydrogen bond is observed by
broadening of their hydrogen signals COOH and O1…H…N1. The signals of imine hydrogen
H11 and of H12 become sharper and are shifted towards higher field (Figure 11).
20

ACCEPTED MANUSCRIPT
Figure 11: Parts of the 1H NMR spectra of 1 in DMSO (40 mg/ml) at different temperatures;
a) phenol hydrogen O1…H and H of the carboxylic acid; b) imine hydrogen H11; c) H12 at
asymmetric carbon atom C12.
21

ACCEPTED MANUSCRIPT
Figure 12: Part of the 13C NMR spectra of 1 in DMSO solution (40 mg/ml) at different
temperatures; from the left: carbon atom C2; carboxyl carbon C13; imine carbon C11.
Now also in 13C NMR spectra a high field shift of all signals can be observed which is
especially strong for phenol carbon atom C2 (Figure 12, 20 °C: δ (13C) = 176.2 ppm, 80 °C: δ
(13C) = 174.1 ppm). This indicates that the PI / KA equilibrium is moving towards the
phenol-imine side (Figure 7) with increasing temperature.
3.4.3 UV-Vis spectra
Tautomeric equilibria in Schiff bases have been already investigated with UV-Vis
spectroscopy [60, 61]. It was shown in the literature that bands at ca. 300 nm indicate the
presence of the phenol-imine form and bands above 400 nm the presence of the keto-amine
form. UV-Vis spectra of 1-3 have been measured in DMSO. Therein both bands are observed
in the expected regions and indicate that the compounds 1 and 2 consist of around 60 % keto-
amine form (KA) in DMSO solution at room temperature. In compound 3 the keto-amine
form prevails with 66%.
3.4.4 Quantum chemical calculations
Extensive quantum chemical analyses of tautomeric equilibria in ortho-hydroxyaryl Schiff
bases have already been published [54, 62, 63,]. Quantum chemical calculations about the
tautomerism in naphthylidene amino acid Schiff bases have recently been published by Ebead
et al. [35]. Therein the authors made geometry optimizations at the PM6 and the
B3LYP/6-31G** level of theory. However the geometries used by the authors seemed rather
arbitrary and there were no information available about the potential energy surface and a
possible transition state between phenol-imine and keto-amine form. Therefore we performed
quantum chemical calculations with molecule 1 in order to gain further insight into the
tautomerism in naphthylidene amino acid Schiff bases. The same DFT method / basis set
combination as in [35] was used. Calculated molecules are denoted as PIa-b for the phenol-
imine and as KAa-c for the keto-amine form. Geometries have been optimized in the gas phase
and in DMSO solution.
3.4.4.1 Gas phase structures
The molecules a and b from the X-ray structure 1-lt have been taken as starting point for
optimization of the keto-amine form. These were optimized to two separate minima KAa and
22

ACCEPTED MANUSCRIPT
KAb with zero imaginary frequencies. Both structures have nearly identical energies (Table
5). The keto-amin form published by Ebead [35] was also reoptimized and is denoted here as
KAc. This structure is 42.3 kJ/mol higher in energy than KAa and KAb. KAc is indeed a local
minimum on the potential energy surface, but it is not the global minimum. Furthermore the
phenol-imine form was constructed by shifting the hydrogen atom in KAa from the nitrogen
atom to the oxygen atom. Geometry optimization and frequency calculation of this molecule
gave PIa as the global minimum. The phenol-imine form published by Ebead [35] was also
reoptimized and is denoted here as PIb. This structure is 66.0 kJ/mol higher in energy than PIa
and is also a local minimum on the potential energy surface. A schematic energy profile
diagram for the calculated stationary points in the gas phase is shown in Figure 13.
Table 5: Relative energies of optimized molecule geometries at the B3LYP/6-31G(d,p) level.
Gas phase structures ꢀG (kJ/mol) in DMSO
ꢀG (kJ/mol)
4.2
4.2
46.5
5.3
0.0
1.6
0.0
39.6
7.5
6.0
KAa
KAb
KAc
TS
PIa
PIb
KAa(DMSO)
KAb(DMSO)
KAc(DMSO)
TS(DMSO)
PIa(DMSO)
PIb(DMSO)
Keto-Amin Form
Transition state
Phenol-Imin Form
66.0
62.2
23

ACCEPTED MANUSCRIPT
Figure 13: Schematic energy profile diagram for calculated stationary points in the gas phase
(top) and in DMSO (bottom).
24

ACCEPTED MANUSCRIPT
Relaxed potential energy surface scans for the gas phase structures of the keto-amine and the
phenol-imine forms have been calculated by changing the torsion angle C2-C1-C11-N1 from
0 to 360°. Both PES are shown in Figure 14. As can be seen from the graph the lowest energy
for both forms is reached at a torsion angle of zero degree. The structure KAc has a torsion
angle C2-C1-C11-N1 of 171.93 and represents therefore a local minimum on the PES in
Figure 14. The torsion angle of PIb is 33.26°. The slope of the torsion angle starting from zero
degree is very low, therefore the frequency calculation pretends here a local minimum. The
calculation of the PES proves that the minima published by Ebead [35] are local minima
which do not represent the geometries with the lowest energy on the PES. From comparison
of both graphs it becomes evident, that the global minimum for the gas phase structure is the
phenol-amine form PIa which is 4.2 kJ/mol lower in energy than the lowest energy form of
the keto-imine form (KAa/KAb). This is in contrast to the X-ray structural data were the keto-
imine form was obtained by crystallization from solution.
Additionally the transition state TS for the tautomeric process was calculated. It has a very
low energy barrier with only 5.3 kJ/mol for the transformation from PIa to KAa (see Figure
13).
Figure 14: Potential energy surface scans for the keto-amin (KA) and the phenol-imine forms (PI) in
the gas phase.
25

ACCEPTED MANUSCRIPT
3.4.4.2 Structures in DMSO
The energy values are changing in DMSO solution. Herein the keto-amine form KAb(DMSO)
becomes the global minimum in this system (see Table 2 and Figure 13). The closely related
geometry KAa(DMSO) is 1.6 kJ/mol higher in energy. The phenol-imine form PIa(DMSO) is
6.0 kJ/mol higher in energy. This energy difference implies that there should be a ratio of
about 11:1 between keto-amine and phenol-imine form present in DMSO solution. Both
forms published by Ebead [35] are again local minima with substantial higher energies. The
transition state TS(DMSO) represents also in DMSO a low barrier for the transition from one
form to the other. The independently calculated PES in DMSO solution prove these results
(Figure 15). The lowest energy forms are observed at a torsion angle of zero degree. The keto-
amin form is more stable in solution. This is experimentally proven by the fact that the
chemical shift of C2 (δ= 176.1 ppm) in DMSO solution is very close to the value for the KA
form observed in the solid state (δ= 178.6 ppm).
Figure 15: Potential energy surface scans for the keto-amin (KA) and the phenol-imine forms (PI) in
DMSO solution.
26

ACCEPTED MANUSCRIPT
Comparison of the PES scans in the gas phase (Figure 14) and in DMSO (Figure 15) shows
that the phenol-imine form has in both media a barrier of ca. 60 kJ/mol for rotation around the
torsion angle, whereas energy barrier for the rotation of the keto-amine form is lowered in
solution to ca. 120-130 kJ/mol. Further local minima may arise from different orientations of
the isopropyl and the carboxyl group at this molecule. But these have probably only minimal
influence on the total energy of the molecule. The PES calculations presented here represent a
good model to describe energy changes of compound 1.
4. Conclusion
During our studies on Schiff base compounds derived from amino acids and
2-hydroxyaromatic aldehydes, we were able to prepare three N-(2-hydroxy-1-
naphthylidene)amino acid derivatives. The crystal structures of N-(2-hydroxy-1-
naphthylidene)-
L-valine (1), N-(2-hydroxy-1-naphthylidene)-
L-phenylalanine (2), and N-(2-
hydroxy-1-naphthylidene)-
L-threonine (3) uniformly exist in the keto-amine form in the solid
state. The solid state structures are further stabilized by intra- and intermolecular hydrogen
bonds. Equilibria between keto-amine and phenol-imine form are present in DMSO solutions
of 1-3. This was shown with 1H NMR und UV-Vis spectra. According to NMR and UV-Vis
data the keto-amine form prevails with about 60 to 66% in DMSO solution. The condition of
tautomeric equilibrium depends mainly on the solvent and temperature but not on the
concentration of the solution. With increasing temperature the equilibrium is shifted towards
the PI form. The 1H NMR spectra are affected by interfering intra- and intermolecular
exchange processes, the rate of the latter can be reduced to the slow exchange mode with
respect to the NMR frequency range by diluting the solution.
Quantum chemical calculations show that there are small energy differences between the
keto-amine and the phenol-imine form in the gas phase and in DMSO solution. Calculation of
the transition state shows, that there is only a small energy barrier below 10 kJ/mol for the
transformation of one form to the other.
Coming back to the question from the headline of this paper we can state the following: The
keto-amine form is more stable in room temperature DMSO solution and its existence was
proven by 1H NMR and UV-Vis. Therefore crystallization of the keto-amine form from
solution is feasible.
5. Acknowledgements
27

ACCEPTED MANUSCRIPT
The authors thank the Freistaat Sachsen for financial support and the computing centre of the
TU Bergakademie Freiberg for computing time at the High Performance Compute Cluster.
S.F. thanks the Freistaat Sachsen for a PhD fellowship (Sächsisches Landesstipendium).
Appendix A. Supplementary material
CCDC 1439783-1439786 contains the supplementary crystallographic data for this paper.
These data can be obtained free of charge via ttp://www.ccdc.cam.ac.uk/conts/retrieving.html
6. References
1
2
M. Nath, S. Goyal, Phosphorus, Sulfur Silicon Relat. Elem. 177 (2002) 447-463.
T. M. Aminabhavi, N. S. Biradar, S. B. Patil, V. L. Roddabasanagoudar, Inorg. Chim.
Acta 107 (1985) 231-234.
3
L. Tian, Z. Shang, X. Zheng, Y. Sun, Y. Yu, B. Qian, X. Liu, Appl. Organomet.
Chem. 20 (2006) 74-80.
4
5
M. Jain, S. Gaur, V. P. Singh, R. V. Singh, Appl. Organomet. Chem. 18 (2004) 73-82.
M. Jain, S. Gaur, S. C. Diwedi, S. C. Joshi, R. V. Singh, A. Bansal, Phosphorus, Sulfur
Silicon Relat. Elem. 179 (2004) 1517-1537.
6
7
8
9
M. Nath, R. Yadav, J. Therm. Anal. Calorim. 55 (1999) 135-142.
F. Mucha, U. Böhme, G. Roewer, Chem. Commun. 1998, 1289-1290.
F. Mucha, J. Haberecht, U. Böhme, G. Roewer, Monatsh. Chem. 130 (1999) 117-132.
J. Wagler, U. Böhme, G. Roewer, Angew. Chem. Int. Ed. 41 (2002) 1732-1734;
corresponds to J. Wagler, U. Böhme, G. Roewer, Angew. Chem. 114 (2002) 1825-
1827.
10
J. Wagler, U. Böhme, E. Brendler, B. Thomas, S. Goutal, H. Mayr, B. Kempf, G. Ya.
Remennikov, G. Roewer, Inorg. Chim. Acta 358 (2005) 4270-4286.
11
12
13
14
15
16
17
18
U. Böhme, B. Günther, Inorg. Chem. Comm. 10 (2007) 482-484.
U. Böhme, S. Jähnigen, Acta Cryst. C64 (2008) o364–o366.
U. Böhme, J. Haushälter, Inorg. Chem. Comm. 12 (2009) 35-37.
U. Böhme, S. Fels, Acta Cryst. C66 (2010) o202-o205.
A. Schwarzer, L. E. H. Paul, U. Böhme, Acta Cryst., Sect C69 (2013) 1336-1339.
U. Böhme, S. Fels, Inorg. Chim. Acta 406 (2013) 251-255.
U. Böhme, S. Wiesner, B. Günther, Inorg. Chem. Comm. 9 (2006) 806-809.
G. Warncke, U. Böhme, B. Günther, M. Kronstein, Polyhedron 47 (2012) 46-52.
28

ACCEPTED MANUSCRIPT
19
20
U. Böhme, S. Fels, Acta Cryst. C69 (2013) 44-46.
G. Y. Aly, M. K. M. Rabia, M. A. F. Al-Mohanna, Synth. React. Inorg. Met.-Org.
Chem. 34 (2004) 45-66.
21
22
Y. Fujii, Y. Kuwana, S. Takahashi, K. Shimizu, K. Hiroi, Bull. Chem. Soc. Jpn. 55
(1982) 2598-2610.
I. Sakiyan, N. Gündüz, T. Gündüz, Synth. React. Inorg. Met.-Org. Chem. 31 (2001)
1175-1187.
23
24
25
26
I. Sakiyan, H- Yilmaz, Synth. React. Inorg. Met.-Org. Chem. 33 (2003) 971-983.
A. Dogan, I. Sakiyan, E. Kilic, Journal of Solution Chem. 33 (2004) 1539-1547.
I. Sakiyan, E. Logoglu, S. Arslan, N. Sari, N. Sakiyan, BioMetals 17 (2004) 115-120.
I. Sakiyan, R. Özdemir, H. Ögütcü, Synth. React. Inorg., Met.-Org., Nano-Met. Chem.
44 (2014) 417-423.
27
28
A. M. Shaker, L. A. E. Nassr, Int. J. Chem. Kinet. 35 (2003) 592-601.
A. M. Shaker, A. M. Aida, L. A. E. Lobna, Synth. React. Inorg. Met.-Org. Chem.33
(2003) 103-117.
29
30
31
32
L. H. Abdel-Rahman, R. M. El-Khatib, L. A. E. Nassr, A. M. Abu-Dief, F. E.-D.
Lashin, Spectrochim. Acta, Part A 111 (2013) 266-276.
P. K. Sharma R. N. Handa, S. N. Dubey, Synth. React. Inorg. Met.-Org. Chem.26
(1996) 1219-1230.
M. R. Mahmoud, S. A. El-Gyar, A. A. Moustafa, A. Shaker, Polyhedron, 6 (1987)
1017-1020.
S. Y. New, X. Wu, S.-Q. Bai, L. L. Kon, T. S. A. H., F. Xue, Cryst. Eng. Comm. 13
(2011) 4288-4235.
33
34
35
M. Abdel-Moustafa, Bull. Pol. Acad. Sci., Chem. 43 (1995) 119-123.
R. Aman, G. Matela, E-Journal of Chem. 2013, 2013, 4pp.
Y. H. Ebead, H. M. A. Salman, M. A. Abdallah, Bull. Kor. Chem. Soc. 31 (2010) 850-
858.
36
Z. Rozwadowski, K. Ambroziak, M. Szypa, E. Jagodzinska, S. Spychaj, W. Schilf, B.
Kamienski J. Mol. Struct. 734 (2005) 137-142.
37
38
G. M. Sheldrick, Acta Crystallogr. A64 (2008) 112–122.
G. M. Sheldrick, Acta Crystallogr. C71 (2015) 3–8.
39
40
Y. Özcan, S. İde, İ. Şakıyan, E. Logoglu, J. Mol. Struct. 658 (2003) 207-213.
Gaussian 09, Revision B.01, 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. A.
Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N.
Kudin, V. N. Staroverov, T. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A.
Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M.
29