Structural dependence of isotope effects in 1H and 13C nuclear magnetic resonance spectra of the trans-N-benzylideneaniline imino group

Article abstract of DOI:10.1016/S1386-1425(97)00241-2

Deutrium- and ¹⁵N-induced isotope effects on ¹H and ¹³C chemical shifts of the imino moiety of 16 trans-N-benzylideneaniline (tBA) isotopomers were determined and analysed. These effects appear to be a sensitive probe of m

Full text of DOI:10.1016/S1386-1425(97)00241-2

Spectrochimica Acta Part A 54 (1998) 327–333
1
13
Structural dependence of isotope effects in H and C nuclear
magnetic resonance spectra of the trans-N-benzylideneaniline
imino group
a,
a
Predrag Novak a, Zlatko Meic´ *, Draz
'
en Vikic´-Topic´ , Vilko Smrec
ki a,
'
Janez Plavec b
a Laboratory of Molecular Spectroscopy, Department of Chemistry, Ru/er Bosko6ic´ Institute, P.O. Box 1016, HR-10001,
Zagreb, Croatia
b National Institute of Chemistry, SI-1000, Ljubljana, Slo6enia
Received 25 July 1997; received in revised form 29 September 1997; accepted 8 October 1997
Abstract
1
Deuterium- and 15N-induced isotope effects on H and 13C chemical shifts of the imino moiety of 16 trans-N-ben-
zylideneaniline (tBA) isotopomers were determined and analysed. These effects appear to be a sensitive probe of
molecular conformation and its changes. All determined deuterium isotope effects on the imino-proton are small and
2
negative and are spread in the range from −0.88 ppb in 4-2H-tBA to −4.97 ppb in H10-tBA. The effects on the
imino-carbon are either positive or negative and have much larger values. Negative sign of isotope effects is primarily
governed by lone-pairs, while magnitude and non-additivity depend on the site of isotopic substitution, molecular
1
conformation and distribution of rotamer populations. All 15N isotope effects on both H and 13C chemical shifts are
positive. A comparison of isotope effects in tBA with those in related isoelectronic compounds is made. © 1998
Elsevier Science B.V. All rights reserved.
Keywords: NMR; Isotope effects; Conformation; Trans-N-benzylideneaniline
1. Introduction
result of a balance between steric interactions
(involving ortho-h hydrogens and lone-pair elec-
trons) and conjugation effects, leading to a non-
planar atom arrangement (Scheme 1).
Vibrational spectra of 1 and its deuterium la-
belled isotopomers have shown significant
changes of the imino group vibrational modes,
indicating their high sensitivity to isotopic substi-
tution [3c]. We have recently shown [4] that the
deuterium substitution in molecules of the Ph-Z-
Schiff bases belong to an important class of
biologically active molecules, e.g. they play a
main role in the process of vision [1,2]. Trans-N-
benzylideneaniline (tBA, 1) represents a model
aromatic Schiff base. Its conformation [3] is a
* Corresponding author. Tel.: +385 1 4680116; fax: +385
1 4680195; e-mail: vujanic@olimp.irb.hr;zmeic@olimp.irb.hr
1386-1425/98/$19.00 © 1998 Elsevier Science B.V. All rights reserved.
PII S1386-1425(97)00241-2

328
P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
2
2
2
Ph type, where Ph refers to phenyl and Z to a
bridging group (CꢀC, CꢀO, CꢀN, NꢀN, etc.),
affects NMR parameters throughout the
molecule. Thus, deuterium causes perturbations in
13C nuclear shielding up to ten bonds away [4].
Besides, some isotope effects are interesting be-
cause of their dependence on structural parame-
ters, such as torsional angles [5] and lone-pair
interaction [4e]. For a series of isotopically la-
belled 1 investigated here (Scheme 1), it is shown
that long range deuterium isotope effects and
2H1), 5 (97% H1), 6 (95% H1), 7 (91% H1), 8
2
2
2
(97% H1), 9 (79% H2), 10 (97% H5), 11 (96%
2
2
2
2H5), 12 (92% H6), 13 (85% H6), 14 (92% H10),
15 (89% H11) and 16 (99% 15N).
2
2.2. NMR measurements
NMR spectra were recorded with Varian Gem-
ini 300, Bruker AM-360 and Varian Unity Inova
600 spectrometers, operating at 75.4, 90.6 and
150.9 MHz for the 13C resonance, respectively.
Samples were measured at 294 K in 5 mm tubes.
Variable temperature measurements were per-
formed in the temperature range of 274–324 K.
Sample concentrations were 0.15–0.20 M. Deu-
terium from the solvent was used as the lock
signal and TMS as the internal standard. Narrow
region spectra with spectral widths of 500–2000
Hz were zero-filled to 64 K, thus obtaining a
digital resolution better than 0.03 Hz per point
(i.e. 90.2 ppb for 13C at 150.9 MHz) after
Fourier transformation.
Isotope effects were determined as a difference
between chemical shifts of the light and heavy
isotopomer (D=lL-lH), expressed in parts per
billion (ppb) units. Standard deviations are given
in Table 1. For non-additivity estimations total
errors were obtained by summing up the individ-
ual mean square errors of the corresponding ef-
fects.
1
their non-additivity on H and 13C chemical shifts
of the imino group are a sensitive probe of molec-
ular geometry.
2. Experimental
2.1. Preparation of isotopomers
Trans-N-benzylideneaniline isotopomers (tBA)
(2–16) were prepared by the addition of specifi-
cally deuteriated anilines to corresponding ben-
zaldehydes [6]. The isotopomers were purified by
recrystallization from 85% ethanol. The melting
point of all synthesised compounds was 50–52°.
Benzaldehyde (Kemika), aniline (Fluka), h-2H-
2
benzaldehyde (Merck), H5-benzaldehyde (Merck)
and 2H5-aniline (Campro Scientific) were commer-
cial products. 2-, 3- and 4-2H-aniline were synthe-
sised by catalytic hydrogenolysis of 2-, 3- and
4-bromoaniline, respectively, with 5% palladium
on charcoal and deuterium gas in tetrahydrofuran
[7]. 2-, 3- and 4-2H-benzaldehydes were prepared
by the following reactions: 2-, 3- and 4-2H-tolue-
nes were mixed with N-bromosuccinimide and
dibenzoylperoxide to give 2-, 3- and 4-2H-benzyl-
bromides, respectively. Benzylbromides were then
allowed to react with hexamethylenetetramine in a
mixture of acetic acid and water [8]. 2-, 3- and
4-2H-toluenes were synthesised from respective
bromotoluenes via the Grignard reaction with
magnesium activated using 1,2-dibromoethane as
entrainer [9]. The reaction mixture was quenched
3. Results and discussion
3.1. Deuterium isotope effects
1H and 13C NMR spectra (Figs. 1 and 2) of the
imino group (–ChHhꢀNh–) in deuteriated isoto-
pomers of 1 reveal a number of interesting isotope
effects (Table 1). All observed deuterium effects at
H-h are negative, i.e. deshielding, while the effects
at C-h are either positive (shielding) or negative
(Table 1). The effects at H-h are much smaller
than those at C-h, primarily due to the relatively
narrow 1H chemical shift range. From this reason,
intrinsic isotope effects on 1H chemical shifts have
rarely been reported in the literature [10]. In de-
termination of such effects (Fig. 1), high precision
measurements are required.
2
with H2O.
The deuterium contents, determined with an
Extrel FTMS 2001 DD mass spectrometer, were
2
2
the following: 2 (90% H1), 3 (90% H1), 4 (90%

P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
329

330
P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
significant line-broadening was detected in 1H
Deuterium isotope effects at H-h in tBA do not
NMR spectrum of the mixture of 3 and 1. In
isotopomers 5 and 7, having attached a deuterium
atom at o%- and m%-positions, respectively, no long
range deuterium effect on H-h chemical shift was
observed. However, shielding effects were found
at the imino-carbon C-h (Fig. 2).
exceed −5.0 ppb in magnitude. In isoelectronic
cis- and trans-stilbenes these effects are up to 3.3
and 8.57 ppb, respectively [11]. Since deuterium
isotope effects at H-h in cis- and trans-stilbenes
are positive [11], one can rationalise the negative
sign in tBA by the inductive and steric influence
of the nitrogen lone pair electrons. The shortest
possible deuterium isotope effect at H-h, is the
1
Long range effects on H shifts, i.e. those trans-
mitted over more than three bonds from the
isotopic site, were previously detected only in a
few instances: in some rigid molecules [13] and in
biomolecules such as protoporphyrins [14] and
macrolide antibiotics [15]. In the case of rigid
systems, the existence of long range deuterium
4
one over four-bonds, D, observed in isotopomer
4, amounting to −1.12 ppb (Table 1). The mag-
nitude of isotope effect depends primarily on the
distance between the isotopic and the observed
5
nucleus [12]. Correspondingly, D in isotopomer 6
1
effects on H shifts was attributed to a through-
is −1.00 ppb and 6D in isotopomer 2 is only
−0.88 ppb, the latter being also the longest range
isotope effect observed in tBA at H-h (Fig. 1b). 7D
in isotopomer 3 (4%-2H-tBA) was not revealed, but
space interaction between 1H and 2H, while in
macrolides these effects were assumed to be a
consequence of changes in hydrogen-bonding
upon isotopic substitution. Generally, isotope ef-
fects on NMR parameters are of a rotational–vi-
brational origin, i.e. they arise from different
averaged rotational–vibrational geometries of iso-
topomers [16].
Deuterium isotope effects in isotopomers 9–15
are cumulative or total since more than one deu-
terium is present in the molecule. Comparing
pentadeuterio-isotopomers 10 and 11 (Table 1),
one can see that the total deuterium isotope effect
on H-h chemical shifts is much larger in the
former (−4.31 ppb) than in the latter (−1.16
ppb), which reflects different conformations of
aniline and benzylidene moiety relative to the
azomethine group plane (Scheme 1). As a matter
of fact, all effects from deuterium in the aniline
ring of isotopomers 3, 5, 7, 9 and 11 are smaller
than those from deuterium in the benzylidene ring
of isotopomers 2, 4, 6, 8 and 10. Although the
effects in the former set of isotopomers are trans-
mitted through a longer pathway (one bond
more) than those in the latter series, previous
investigations in related molecules showed the
predominant role of conformation in transmission
of isotope effects. Our ab initio calculations pre-
dict a torsional C2ꢀC1–NꢀCh angle in tBA of
44.6°, while C2ꢀC1–ChꢀN is close to zero [5].
Therefore, the decrease in magnitude of isotope
effects transmitted from the aniline ring should
also be related to larger torsional angle.
Scheme 1.

P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
331
Fig. 1. Isotope effects at H-h in isotopomers of trans-N-benzylideneaniline. (a) H-h region of the 360 MHz proton spectrum of
isotopomer 2 (4-2H); (b) the same spectrum with the addition of 1; and (c) H-h region of the 300 MHz proton spectrum of
isotopomers 16 (15N) and 1.
3.2. 15N isotope effects
deviation is 0.07 ppb) was determined in 14
(2H10). This deviation of 0.5090.21 ppb is small
but significant. Non-additivity indicates that be-
side intrinsic isotope effects, equilibrium isotope
effects are also present, which is supported by
variable temperature measurements. A decrease in
magnitude of isotope effect in the 50° temperature
range is observed with decreasing the tempera-
ture.
The non-additivity of deuterium isotope effects
on 13C chemical shifts in tBA have been reported
recently [4e]. With two new isotopomers discussed
here, o%-2H- (5) and m%-2H-tBA (7), one can recog-
nise an appreciable deviation from additivity rule
at C-h, amounting to 29.196.5 ppb. This value
was calculated as follows. By summing up individ-
ual contributions of mono-deuteriated isoto-
pomers 2, 3, 4, 5, 6 and 7, the value of 316.195.5
ppb was obtained. However, a much smaller ef-
fect of 287.091.0 ppb was observed in 15 (2H11).
The difference between these two give the above
estimated non-additivity. As previously pointed
out [4e,f] this should be due to slight conforma-
tional changes arising from shorter C–2H than
C–1H bond lengths, which in the case of the
deuteriated phenyl rings, releases a strain between
the ortho and h positions, thus affecting rotational
15N isotope effects at both H-h and C-h are
positive, but the former is much smaller than the
latter (Table 1). This is due to the two main
reasons. Firstly, the effect at H-h is transmitted
over two bonds, while that observed at C-h is
transmitted over one bond. Secondly, 1H chemical
shift range is significantly smaller than that of 13C.
Both 15N effects are much smaller than deuterium
effects (Table 1), owing to the lower 15N/14N mass
ratio.
3.3. Additi6ity
One of the main characteristics of isotope ef-
fects in NMR spectra is their additivity. In flexible
molecules such as those containing phenyl rotors,
some apparent deviations have recently been re-
ported [4e,12]. The deviation from the additivity
1
rule of deuterium isotope effects on H shifts in
tBA isotopomers appears to be more significant
than that in isotopomers of isoelectronic
molecules of cis- and trans-stilbene [11]. The sum
of the effects measured in pentadeuterio-isoto-
pomers 10 and 11 is −5.47 ppb (standard devia-
tion is 0.14 ppb), while −4.97 ppb (standard

332
P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
4. Conclusions
Isotope effects in the imino group of trans-N-
benzylideneaniline reflect conformational changes
upon isotopic substitution and show a sensitivity
to steric and lone-pair interactions. The non-addi-
tivity in polydeuteriated isotopomers arises from a
redistribution of rotamer populations due to dif-
ferent rotational properties of undeuteriated and
perdeuteriated phenyl rotors, as was previously
found for benzophenones [12] and biphenyls [17].
Therefore, isotope effects in the imino moiety
could serve as a probe for studying Schiff bases,
both experimentally and theoretically.
Acknowledgements
This work is a part of Project No. 00980802 of
the Ministry of Science and Technology of the
Republic of Croatia. We thank Professor Heinz
Sterk (Graz), in whose laboratory some of the
NMR measurements were carried out. We are
also indebted to Dr D. Kovac' ek for ab initio
calculations.
References
[1] (a) R.A. Mathies, S.O. Smith, I. Palings, Biol. Appl.
Raman Spectrosc. 2 (1987) 59; (b) R.A. Mathies, S.W.
Lin, J.B. Ames, W.T. Pollard, Annu. Rev. Biophys.
Chem. 20 (1991) 491.
[2] R.R. Birge, L.P. Murray, R. Zidovetzki, H.M. Knapp,
J. Am. Chem. Soc. 109 (1987) 2090.
[3] (a) P. Jacques, O. Chalret, J. Mol. Struct. 68 (1980) 215;
Fig. 2. Isotope effect over five bonds at C-h in isotopomer 7
(3%-2H-tBA) measured at 150.9 MHz. (a) 2:1 mixture of isoto-
pomers 7 and 1; and (b) 1:1 mixture of the same isotopomers.
(b) L.N. Patnaik, S. Das, Int. J. Quantum Chem. 27
&
(1985) 135; (c) Z. Meic´, G. Baranovic´, T. Suste, J. Mol.
Struct. 296 (1993) 163; (d) H.B. Bu¨rgi, J.D. Dunitz,
Helv. Chim. Acta 54 (1971) 1255.
&
[4] (a) P. Vujanic´, E. Gacs-Baitz, Z. Meic´, T. Suste, V.
Smrec
'
ki, Magn. Reson. Chem. 33 (1995) 426; (b) V.
ki, N. Mu¨ller, D. Vikic´-Topic´, P. Vujanic´, Z.
Smrec
'
Meic´, J. Mol. Struct. 348 (1995) 69; (c) P. Novak, Z.
Meic´, H. Sterk, J. Chem. Soc. Perkin 2, (1996) 2531;
(d) Z. Meic´, P. Novak, D. Vikic´-Topic´, V. Smrec
Magn. Reson. Chem. 34 (1996) 36; (e) V. Smrecki, D.
Vikic´-Topic´, Z. Meic´, P. Novak, Croat. Chem. Acta 69
(1996) 1501; (f) D. Vikic´-Topic´, M. Hodos ek, A. Grao-
vac, E.D. Becker, Croat. Chem. Acta 68 (1995) 193.
' ki,
properties of benzylidene (B) and aniline (A)
rings. This brings about a redistribution of ro-
tamer populations, which affects nuclear shield-
ings.
'
' c'

P. No6ak et al. / Spectrochimica Acta Part A 54 (1998) 327–333
333
[5] P. Novak, Z. Meic´, D. Vikic´-Topic´, D. Kovac
'
ek, sub-
Magn. Reson. Chem. 34 (1996) 610.
[13] F.A.L. Anet, A.H. Dekmezian, J. Am. Chem. Soc. 101
(1979) 5449.
[14] C.J. Medforth, F.Y. Shiau, G.N. La Mar, K.M. Smith,
J. Chem. Soc. Chem. Commun. (1991) 590.
[15] J.R. Everett, J. Chem. Soc. Chem. Commun. (1987)
1978.
mitted.
[6] L.A. Bigelow, H. Eatough, Org. Synth. Coll. Vol. I, 80.
[7] D.M. Jerina, J. Dely, W.B. Witkop, Biochem. 10 (1971)
366.
[8] R.G. Coombe, D.B. Poulton, Aust. J. Chem. 31 (1978)
451.
[9] Y.-H. Lai, Synthesis (1981) 585.
[10] P.E. Hansen, Prog. Nucl. Magn. Reson. Spectrosc. 30
(1988) 207.
[11] P. Novak, Z. Meic´, D. Vikic´-Topic´, H. Sterk, J. Mol.
Struct. 410/411 (1997) 9.
[12] P. Novak, D. Vikic´-Topic´, E. Gacs-Baitz, Z. Meic´,
[16] C.J. Jameson, The dynamic and electronic factors in
isotope effects on NMR parameters, in: E. Bunciel, J.R.
Jones (Eds.), Isotopes in the Physical and Biomedical
Science, vol. 2, Elsevier, Amsterdam, 1991, p. 1.
[17] A. Almenningen, O. Bastiansen, L. Fernhalt, J. Mol.
Struct. 128 (1985) 59.
.