Spectroscopic properties of macrocyclic oligo(phenyldiacetylenes)-II. Synthesis and theoretical study of diacetylenic dehydrobenzoannulene derivatives with weak electron-donor and -acceptor groups

Article abstract of DOI:10.1016/j.molstruc.2006.01.049

Diacetylenic dehydrobenzo[12]annulene and dehydrobenzo[18]annulene derivatives with electron-donor and -acceptor groups were synthesized (including push-pull Eglinton-Galbraith dimer derivative 1c) via an oxidative coupling reaction, and spectroscopically

Full text of DOI:10.1016/j.molstruc.2006.01.049

Journal of Molecular Structure 794 (2006) 115–124
www.elsevier.com/locate/molstruc
Spectroscopic properties of macrocyclic oligo(phenyldiacetylenes)-II.
Synthesis and theoretical study of diacetylenic dehydrobenzoannulene
derivatives with weak electron-donor and -acceptor groups*
a
Boris Zimmermann , Goran Baranovic , Zoran Stefanic , Marko Rozman
a,
b
b
ˇ
*
´
´
ˇ
a
ˇ ´
Department of Organic Chemistry and Biochemistry, Rudjer Boskovic Institute, P.O. Box 180, 10002 Zagreb, Croatia
b
ˇ ´
Department of Physical Chemistry, Rudjer Boskovic Institute, P.O. Box 180, 10002 Zagreb, Croatia
Received 30 December 2005; received in revised form 27 January 2006; accepted 30 January 2006
Available online 29 March 2006
Abstract
Diacetylenic dehydrobenzo[12]annulene and dehydrobenzo[18]annulene derivatives with electron-donor and -acceptor groups were
synthesized (including push–pull Eglinton–Galbraith dimer derivative 1c) via an oxidative coupling reaction, and spectroscopically and
structurally characterized. The solid–solid phase transition in 1b has been revealed at 45 8C by DSC measurements. Its room temperature crystal
structure has been solved by X-ray diffraction measurements. The 1H and 13C NMR chemical shifts, UV/vis and infrared absorption spectra and
Raman scattering spectra have been analyzed by using ground-state DFT calculations. The strongest absorptions in the UV/vis spectra of 1 and 2
most probably are not due to the HOMO/LUMO excitations but due to the (HOMO-1)/LUMO and HOMO/(LUMOC1) excitations. The
substitution effects on the electronic charge distribution of the all-carbon annulenic cores can be particularly well observed in the distribution of IR
intensities in the region of acetylenic stretching vibrations. IR intensities are thus useful in studying the extent of resonance interactions also in
acetylenic macrocycles.
q 2006 Elsevier B.V. All rights reserved.
Keywords: Dehydrobenzoannulenes; Diphenyldiacetylenes; Aromaticity; Donor; Acceptor; Substituent effects; Resonance interaction; Vibrational spectroscopy
1. Introduction
electron acceptor groups, as a first step in synthesis of a push–
pull dimeric macrocycle.
Dehydrobenzoannulenes have the potential of forming
expanded carbon-rich networks [1–5] potentially leading to
the theoretically predicted carbon allotropes [6,7]. Regarding
their highly conjugated structure, these compounds have
intensively been studied [8–14] as promising candidates for
innovative electronic and photonic materials [15]. The donor/
acceptor dehydrobenzo[18]annulenes attracted special interest
given that this kind of push–pull systems is good candidates for
use in non-linear optics [16,17]. At the same time, related
donor/acceptor dehydrobenzo[12]annulenes remained unex-
plored with respect to their ability of providing a communi-
cation through the conjugated bridge. With that in mind we
attempted to create in a similar way dimer derivatives with
Highly strained Eglinton–Galbraith dimer 1a (R1ZR2ZH,
Scheme 1) was synthesized more than 40 years ago [18], but
until last decade it did not draw attention, mainly because of its
high instability in the solid state [19]. However, different
derivatives of this dimer were successfully synthesized
showing much higher crystal stability [14,20,21]. Bunz and
Enkelmann concluded that the solid-state reactivity of this
dimer is not a property of the individual molecule due to the
strained nature of the bent butadiyne bridges, but is merely
guided by the packing arrangements of the diyne groups [19].
Swager’s group prepared derivatives of Eglinton–Galbraith
dimer containing electron-donating alkyl and alkoxy groups
[14] and the presence of long side-chains stabilized the crystal.
In view of the fact that solubility of dehydrobenzoannulenes is
decreasing with increasing molecule size [3], it would be of
interest to create this kind of relatively small dehydrobenzoan-
nulenes. Here, we report our results on the synthesis and
characterization of dehydrobenzo[12]annulenes and dehydro-
benzo[18]annulenes with ester and alkyl side-chains. An
attempt has been consequently undertaken to understand
* For part I see Ref. [37].
*
Corresponding author. Tel.: C385 1 468 0116; fax: C385 1 468 0195.
´
E-mail address: baranovi@irb.hr (G. Baranovic).
0022-2860/$ - see front matter q 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2006.01.049

116
B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
Scheme 1.
˚
Cu Ka radiation, wavelength 1.54180 A, using the u/2q scan
the substitutional effects on electronic properties of acetylenic
macrocycles.
technique. During data collections there were no significant
variations in intensities of the three control reflections, which
were measured every 120 min. The data were corrected for
Lorentz and polarization effects [22]. The absorption correc-
tion was based on a j-scan of seven reflections. Structures
were solved using WinGX package [23] and refined by the
package SHELXL97 [24]. Molecular geometry calculations and
illustrations of the crystal packing were prepared by PLATON98
[25]. Atomic scattering factors were those included in
SHELXL97. Crystallographic data excluding structure factors
for the structures reported in this paper have been deposited
with the Cambridge Crystallographic Data Centre as sup-
plementary publication number CCDC 216872. Copies of data
can be obtained, free of charge, on application to CCDC, 12
Union Road, Cambridge CB2 1EZ, UK (fax: C44 1223
336033 or e-mail: deposit@ccdc.cam.ac.uk).
2. Experimental
2.1. General
1H and 13C NMR spectra were recorded on a Bruker Avance
300 and Bruker Avance 600 spectrometer. The chemical shifts
of NMR are expressed in ppm from TMS as determined with
1
reference to the internal CHCl3 and CDCl3 (d 7.26 H NMR
and d 76.9 13C NMR). UV/vis spectra were taken on a Varian
Cary 50 spectrometer. Melting points were measured on an
Olympus System Microscope BX51TF with Linkam hot stage.
ˇ
´
The Central Analytical Service, Rudjer Boskovic Institute,
Zagreb, performed elemental analyses. Infrared spectra were
recorded on a Bomem MB102 spectrometer. Raman spectra
were taken on a Bruker FT-Raman 106/S spectrometer
(Faculty of Sciences, University of Zagreb). All mass spectra
were obtained using a Finigan 2001 DD (Finigan, Madison,
WI, USA) Fourier transform mass spectrometer equipped with
a Nd: YAG Quanta Ray DCR-11 laser (Spectra-Physics Inc.,
Mountain View, CA, USA) emitting at lZ1064 nm and a
nitrogen laser (VSL 337 NSD, LSI Laser Science, Newton,
USA) emitting at lZ337 nm. Following a quench pulse,
positive/negative ions were formed by a single laser pulse. All
mass spectra were recorded at 100 ms time delay.
2.3. Computational
Full geometry optimization and calculations of vibrational
frequencies and IR intensities were carried out at a B3LYP
level of DF theory with a 6-31CG* basis set. All the
calculations were performed with the GAUSSIAN03 software
package [26]. The chosen basis set when used with the chosen
functional for the calculations of geometry and vibrational
frequencies of hydrocarbons has been described as
adequate [27].
2.2. X-ray structural determination
3. Results and discussion
Crystal data for 1b: yellow crystals, C64H88O8, MZ985.34,
monoclinic, space group P21/c (no. 14), aZ13.444 (4), bZ
3.1. DSC measurements
˚
9.352 (2), cZ23.741 (9) A, aZ90.00, bZ94.15 (3), gZ
3
90.008, VZ2977.1 (16) A , ZZ2, DcZ1.099 gcmK3, TZ293
Compounds 1b–c are bright yellow solids and can be stored
for weeks at 4 8C without decomposition. With decomposition
that began above 50 8C donor/acceptor-substituted macrocycle
1c showed lowest thermal stability in comparison to alkyl
derivative 1d (decomposition around 100 8C) [14] and ester
˚
(2) K, m(Cu–Ka)Z0.554 mmK1, 6395 reflections measured,
6238 unique reflections (RintZ0.019), RwZ0.190 (all data),
R1Z0.0641 (IO2s). Intensities were measured on an Enraf
Nonius CAD4 diffractometer, with graphite monochromated

B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
117
the neighboring molecules (Fig. 2) and not to adjacent
rings contrary to the packing of Eglinton—Galbraith dimer
1a (R1ZR2ZH) [19], its alkyl [14], alkynyl [13] and fluoro
derivatives [21], where the parallel stacking of planar
annulene rings determines the crystal structure. Half of the
chains are laid out in the all-trans (planar zig–zag)
conformation, while the other half exhibit gauche defor-
mation beginning with the oxygen atom. Although the
gauche conformation is thermodynamically less favored it
enables the ester chain to be parallel to the adjacent chain
thus leading to the optimal crystal packing. The angles for
the two adjacent (CO)OC groups relative to the annulene
plane are 31 and 528 (calculated values are 22 and 268,
respectively), which is significant. As a consequence of the
deviation from planarity the (CO)OC p-molecular orbitals
will remain localized with minimal overlap with the
annulene p-orbitals. Thus, only the intramolecular inter-
action between the diyne units is possible and since the
distance between the a-type carbons (Scheme 1) is only
Fig. 1. Infrared spectra of 1b at 30 and 50 8C.
derivative 1b. Differential scanning calorimetry showed that
decomposition of 1b began above 70 8C. This is an extremely
exothermic process (enthalpy value of 426 kJ/mol) occurring
over a relatively narrow 15–20 8C range (w1/2Z6 8C). Similar
to the long chain alkyl derivative 1d, the compound 1b shows
an additional endothermic peak at 45 8C presumably due to
crystal-to-crystal phase transition. This transition is
obvious if we compare IR spectra of the compound 1b at 30
and 50 8C, particularly the carbonyl band and 1250 cmK1
regions (Fig. 1).
˚
3.232 A, the intramolecular cyclization is highly probable at
higher temperatures.
The agreement between the X-ray structure and the
B3LYP/6-31CG* optimized bond lengths is reasonable
(Table 1) although the triple bond length is systematically
˚
overestimated by 0.02 A, while the central C–C bond length is
3.2. X-ray crystallography
calculated shorter by the same value. As a measure of the ring
deformation due to the substitution that preserves C2 symmetry
of a benzene ring the quantity dZð1=2ÞðHMKJKCðHIC
LMÞ=2KðIJCKLÞ=2Þ might be useful (Scheme 1). Its non-zero
value certainly is a consequence of the HM bond elongation
due to the strained nature of the [12] annulenic ring. However,
according to calculations the value of d even for the unstrained
2a should be different from zero and close to the value
calculated for the strained 1a.
The crystal structure of 1b was solved by X-ray
diffractometry on single crystals obtained from dichloro-
methane. The butadiyne bridges show expected deviation
from linearity (Table 1) in agreement with the values
described in the literature [14]. The shortest contacts of an
annulene ring are found with the carbonyl oxygens of the
two alkyl chains lying in the plane of the ring. Any of the
annulene rings is thus parallel to the chains of
Table 1
Selected bond distances and angles for some macrocyclic oligo(phenyldiacetylene) structures
Bond/
Anglea
1a
1a
1b
1b
2a
2b
2e
dpbb
Obs.c
Obs.d
Obs.e
Calc.
Obs.
Calc.
Calc.
Calc.
Calc.
Obs.
CD
1.430
1.190
1.381
1.186
1.446
1.364
–
1.429
1.218
1.368
1.217
1.428
1.401
0.007
165.7
167.3
166.9
163.9
117.0
1.437
1.196
1.380
1.425
1.224
1.363
1.438
1.196
1.376
1.196
1.438
1.419
0.011
166.1
168.5
167.6
166.9
116.9
1.424
1.223
1.362
1.417
1.223
1.358
1.416
1.223
1.357
1.417
1.224
1.358
1.421
1.199
1.374
DE
EF
FG
GH
HM
df
1.414
0.032
165.3
167.3
1.439
0.024
1.438
0.016
164.3
167.6
167.7
164.2
118.1
1.429
0.019
1.428
0.013
1.425
0.011
–
–
CDE
DEF
EFG
FGH
GHM
166.4
167.0
117.6
117.3
117.9
a
˚
CD is a bond length (A) between atoms C and D, etc. (Scheme 1).
b
Diphenyldiacetylene crystal structure [29]. The crystal structures of unsymmetrically substituted dehydrobenzo[18]annulenes are known [16] and are unsuitable
for comparison with the calculated molecular structures.
c
This is a dehydro[12]annulene molecule with –CbC–Si(iPr)3 groups instead of hydrogens [28].
See Tovar et al. [13] for the structure of a dehydrobenzo[12]annulene molecule where all hydrogens have been replaced by the –CbC–tBu groups.
Ref. [21].
d
e
f
dZð1=2ÞðHMKJKCðHICLMÞ=2KðIJCKLÞ=2Þ (Scheme 1).

118
B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
Fig. 2. ORTEP packing diagram of 1b showing the unit cell structure.
1
3.3. H and 13C NMR spectra
molecular axis. Similarly the atomic charges in the structures
2c and 2d are sensitive to the torsional angles of the ester
groups while the observed 13C NMR shifts indicate a C2-axis
lying in the molecular plane.
The anti-aromatic nature of dehydrobenzo[12]annulene ring
as well as aromatic nature of dehydrobenzo[18]annulene ring
was confirmed by comparing their 1H NMR spectra with
spectra of acyclic diethynylbenzenes 3 (see Supplementary
material) and unsubstituted 1a [30] and 2a [31,32] (Scheme 1
and Table 2). The arene protons of dimers 1 undergo a 0.5 ppm
upfield (paratropic) shifts relative to 3, while trimeric
macrocycles 2 show 0.2 ppm downfield (diatropic) shifts. As
already speculated [32] the weakness in diatropicity and
paratropicity of these macrocycles might be due to benzannela-
tion. However, if these systems are able to sustain the circular
current only as a single loop of electrons (no current through
the bonds shared by the two types of rings) it is then more
correct to ask whether the unsubstituted octadehydrodiben-
zo[12]annulene system is aromatic or not and analogously for
the unsubstituted dodecadehydrotribenzo[18]annulene system.
In other words, the central annulenic ring should not be treated
as a separate entity when the fused benzene rings are present.
The observed 13C NMR spectra (Supplementary material)
show that all the compounds, 1b–d and 2b–e, possess in
solution at least an effective C2 symmetry. The calculated 1c
structure has no such symmetry (when, for example, Mulliken
atomic charges are viewed) although the alkyl half of the
molecule locally has the C2 symmetry axis along the long
3.4. UV/vis spectra
Before starting any discussion of the recorded UV/vis
spectra it is instructive to analyze the molecular orbital plots
particularly when charge transfer transitions are suspected
[33]. The molecular and accompanying electronic structures
have here been predicted by using DFT and the subsequent
discussion implies (a) that the Kohn–Sham molecular orbitals
are similar to the Hartree–Fock orbitals and (b) the acceptance
of the possibility to interpret excited electronic states only in
terms of singly excited configurations formed from the Kohn–
Sham orbitals. The central benzoannulenic core in 1a, 1b and
1d obviously has D2h symmetry, while in 1c this symmetry is
weakly perturbed. Therefore, the one-electron HOMO (B2g)-
LUMO (B1g) transition, A1g/B3g (the long molecular axis is
z-axis while the x-axis is perpendicular to the molecular plane),
is forbidden in the former three compounds while only weakly
allowed in the latter (Fig. 3). Similarly, the central
benzoannulenic core in 2a, 2b and 2e obviously has D3h
symmetry, while in 2c and 2d this symmetry is weakly
perturbed (Fig. 4). The HOMO and LUMO are both degenerate
E00 orbitals and by promoting single electron four different
Table 2
1H NMR data of arene protons of 1, 2 and 3 (Scheme A1 in Supplementary
material) in CDCl3
0
0
electronic states are formed, A1 CA2 CE0, where only the
transition A1 /E0 is allowed that is not necessarily the lowest.
0
The electronic absorption spectra of 1 (Fig. 5, Tables 3
and 4) exhibit the characteristic pattern found for dehydro-
benzoannulene cores (1a in [18], alkoxy derivatives of 1a in
[34] and 2a in [5]). The spectra of 1 are dominated by the band
at w310 nm which presents a shift of w40 nm in comparison
with the average location of the most prominent doublet in the
spectrum of, for example, 3f. The corresponding shift for 2 is
w65 nm. In comparison with electron-donating derivatives,
the UV/vis spectra of ester macrocycles 1b and 2b showed
small but distinct red shift (ca. 8 nm). The alkyl/ester-
substituted [12]annulene 1c showed two new absorption
bands at 307 and 326 nm (Fig. 5). Given that the compound
1c is donor/acceptor substituted p-conjugated system it is
dH0
dH00
dH000
Molecule
3c
7.29
7.83
7.06
7.28
6.78
6.76
7.68
8.02
7.49
7.96
7.44
3f
1aa
1b
1c
7.23
1d
2ab
2b
2c
7.97
7.46
8.01
7.45
2d
2e
a
Ref. [30].
b
Refs. [5,31].

B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
119
Fig. 3. LUMO plots (up) and HOMO plots (down) of 1a–1d.
Fig. 4. Plots of degenerate LUMOs (up) and HOMOs (down) of 2a, 2b, 2c and 2e (2d not shown in order to save space).
reasonable to assume that absorption band at 326 nm originates
from intramolecular charge-transfer (ICT) transition, and thus
should show larger solvatochromic shift than other bands in the
spectrum of 1c. We measured UV/vis spectra of 1c in pentane,
dichloromethane and 1-decanol (solubility problems precluded
measurements in more polar solvents), but the observed shift of
only w5 nm (Fig. 6), although in the right direction and larger
than for the other bands, is too small to be unambiguously
attributed to an ICT band. The electron-donating/accepting
abilities of the chosen substituents are known to be only weak
to moderate but nevertheless, small effects of the ICT in the
UV/vis spectra where the general dehydrobenzo[12]annulenic
pattern dominates have certainly been observed. The molecular
orbital plots (Fig. 3) are fully in accord with these experimental
findings. Strong uniform delocalization across the benzoannu-
lenic core present in both the HOMO and LUMO of 1a and 1d
confirms the minor influence of the alkyl substitutions on the
HOMO–LUMO transition. The LUMO of the alkyl/ester 1c
exhibits slight charge transfer from the alkyl to the ester
substituents with the central part unchanged.
changes go along with those found in previously published
donor/acceptor dehydrobenzo[18]annulenes where aniline and
nitro groups were used [16]. The molecular orbital plots
(Fig. 4) are consistent with these experimental observations. In
the fully alkyl substituted 2e the HOMO and LUMO are
unchanged relative to the parent 2a. Some charge rearrange-
ments are seen in 2c but again due to the forbidden character of
the lowest transitions it is not possible to associate them with
the shifts in experimental absorption maxima.
Analogously to polycyclic aromatic hydrocarbons, elec-
tronic bands of 1 and 2 are expected to show considerable
vibrational fine structure because the chromophores are planar
[28]. The bands at 293 and 313 nm (average values) in 1 are
separated by 2181 cmK1 and this value should correspond to
the triple bond stretching vibration in an electronically excited
state. Since, this is close to the values measured for the ground
state it might be an indication for the slightly changed
geometry of the all-carbon core in the excited state. Since,
the calculated lowest absorption should roughly be 100 nm
longer than the HOMO–LUMO gap, weak bands around
520 nm for 1 and 490 nm for 2 should be looked for. However,
no data is available on the ionization potentials of our
compounds to assess the value of the HOMO–LUMO gap
and thus the regions of end absorptions. The HOMO–LUMO
gap in [12]annulenes is smaller than in [18]annulenes (Tables 3
and 4) and this would cause the strong red shift of the end
absorption of [12]annulenes as compared to [18]annulenes
[28]. However, this is correct only if very weak, barely
observable absorptions at z410 and z450 nm (not listed in
Table 3) are taken as the real end absorptions in [12]annulenes.
No such very weak absorptions above z405 nm could have
been found in the [18]annulenes. Thus, the strongest
absorptions in the UV/vis spectra of 1 and 2 most probably
are not due to the HOMO/LUMO excitations but due to the
(HOMO-1)/LUMO and HOMO/(LUMOC1) excitations.
The spectra of the alkyl/ester-substituted [18]annulenes 2c
and 2d showed the same pattern for the dehydrobenzo[18]an-
nulene core with four diagnostic peaks [16] and with the end
absorption point extended about 10 nm more than 2b. With
respect to the parent compound 2a (R1ZH, R2ZH), the
spectrum of 2e is red shifted by only 8 nm, while the spectra of
2b, 2c and 2d (practically coinciding with each other) are red
shifted by 16 nm. The doublet of bands at 359 and 369 nm in
unsubstituted 2a is preserved in substituted derivatives but
changes its appearance and the two components are not always
well resolved [16]. Thus, the second component is easily
confused with the very weak band found at 392 nm in 2e and at
405 nm in 2b, 2c and 2d. This transition is observable even in
unsubstituted 2 at w390 nm [5]. Although the alkyl/ester
groups in macrocycles 2c and 2d do not significantly change
electronic absorption spectra, it seems that the observed

120
B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
is reasonable to expect similar behavior. Alkyl groups are all
weak resonance donors, while ester groups are weak acceptors
and redistribution of p-electron charges in the annulenic core
including benzene rings will certainly take place upon
substitution. The correlation with Mulliken charges might be
useful [35]. The sum of Mulliken charges on all carbon atoms
of the benzoannulenic system show that the all-carbon core has
donated electrons to alkyl and ester substituents (Tables 5
and 6). This is opposite to what has already been concluded that
the all-carbon cores in the systems without fused benzene rings
are strong electron acceptors [36]. Unfortunately the calculated
charges have not been partitioned into p- and s-electron
charges since only the former are relevant for the
resonance effects. The main differences are found for the
benzene ring atoms A and B, while other atoms, C, ., H, seem
to be well protected from the substituents by the same benzene
ring.
Assuming D2h symmetry for a [12]annulene ring, there are
four acetylenic bond stretchings, AgCB1gCB2uCB3u, i.e. only
two of them are expected to be IR active. Vibrational spectra
were calculated for molecules having –CH3 group instead of –
C10H21. The influence of such modifications in this spectral
region, i.e. on the stretching vibrations of the annulenic core, is
negligible. The agreement of not only wavenumbers but also
relative IR intensities as well with the observed spectra is rather
satisfactory (Fig. 7 and Table 7). While the B2u transition
remains equally IR active independently of the substitution, the
B3u vibration is getting weaker when the acceptor groups
replace the donor groups. This is an indication that its intensity
should be correlated with the resonance interaction. The atom
displacements in the B3u mode are described as simultaneous
elongation of the DE and QP bonds accompanied with the
contraction of the FG and ON bonds (Scheme 1). The
existence of the permanent molecular dipole moment in 1c
along the long molecular axis means that the strong absorption
for the totally symmetric mode should occur. Unfortunately the
spectra are not of high enough resolution to confirm
the prediction that the strongest band in 1c should actually be
the symmetric acetylenic stretching. They are also showing two
acetylenic IR bands at w2195 and w2125 cmK1 (Fig. 7). A
very strong band corresponding to the totally symmetric CbC
stretching vibration is present at 2191 cmK1 in the Raman
spectra of 1 (Table 7). The other Raman active CbC stretching
modes are weak as in the spectra of alkyl derivatives [37]. Both
bands are found in the infrared spectrum of Eglinton–Galbraith
dimer 1 (RZH) at 2187 and 2116 cmK1 with the same
wavenumber difference of 70 cmK1. Contrary to that, in the IR
spectra of alkyl derivatives the intensity difference between the
two acetylenic bands is getting greater with increasing the size
of alkyl substituents and in the spectrum of 1d the second band
cannot be determined [37]. Intensity of the latter band is
especially large in the spectra of ester dimers 1a and 1b. Larger
intensity of the band at 2195 cmK1 in the spectrum of donor/
acceptor 1c is clearly visible (Fig. 7). All dimeric compounds 1
show an infrared band at 480 cmK1, which already
was described as a characteristic band of the octadehydrodi-
benzo[12]annulene ring [37].
Fig. 5. Electronic absorption spectra of 1 and 2 in CHCl3.
3.5. IR and Raman spectra
As it is very well known the CbC absorption in
disubstituted acetylenes is mainly affected by the resonance
interactions of the substituents with the triple bond ([35] and
references therein). In the present case, we are dealing with
systems possessing several coupled diacetylenic bridges and it

B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
121
Table 3
Observed UV/vis spectra (lmax in nm) of 1 (CHCl3)
1aa
1f
1b
1c
1d
Average
252
269
285
z274
278.2
262.0
275.0
292.1
263G11
275G6
293G8
307
277.8
297.0
276.8
296.2
307.0
316.0
325.7
296.8
304
363
316.2
349.2
316.9
311.3
313G9
350G14
363
z350
z363
377.0
430
376.9
375.9
428
368.9
410
375G6
418G16
402b
a
1a and 2a refer to the unsubstituted dehydrobenzo[12]annulene and -[18]annulene, respectively. Observed lmax for 1a from Ref. [18].
The last row gives the HOMO–LUMO gaps (B3LYP/6-31CG*, in nm).
b
Table 4
Observed UV/vis spectra (lmax in nm) of 2 (CHCl3)
2aa
2f
2b
275.9
2c
2d
2e
Average
266
280
309
275.8
293.0
326.1
282.5
296.0
327.7
278.4
294.9
326.5
271.3
288.0
317.8
275G7
291G5
322G5
325
293.2
325.8
z325
339.5
z350
367.1
377.9
z393
360
330
349.6
349.6
349.3
344.3
344G5
350
359
369
376.8
387.1
376.1
387.1
378.7
387.1
405.8
398
376.4
384.8
405.8
396
372G6
382G5
406G10
378G20
z405
380
355b
a
Observed lmax for 2a from Ref. [3].Very similar to 2e but uniformly downshifted by z10 nm.
HOMO–LUMO gap (B3LYP/6-31CG* in nm).
b
Among the vibrations of the [18]annulene ring of D3h
leading to the changed bond polarities, i.e. enhanced IR activity
of some CbC stretching modes. As for the dimers here again
one band might be chosen that best correlates with the
resonance interaction. According to the calculated and
observed trends in IR intensities, the lower of the two
degenerate pairs seems to be most appropriate for the
characterization of resonance interactions and the reason
being the same as for the B3u vibrations of the dimers. The
atom displacements in either of the two lower E0 modes are
described as simultaneous elongation of the DE and QP bonds
accompanied with the contraction of the FG and ON bonds
symmetry there are six triple bond stretchings, A1 CA2 C2E0.
On0ly degenerate vibrations should show some IR activity while
A1 and E0 vibrations are Raman active. According to
calculations two pairs should be apart by around 70 cmK1. In
the Raman spectra of 2 (Table 8) there are two extremely
strong bands at 2216 and 2198 cmK1 corresponding to the
doubly degenerate and totally symmetric CbC stretchings,
respectively, (the former band was omitted in the previous
paper [37]). In the infrared spectra of trimeric compounds 2
(Table 7) there are three acetylenic bands at w2215, w2200
and w2150 cmK1. Since, the band at 2147 cmK1 does not
appear to be split, one component of the doublet observed at
around 2210 cmK1 could originate from the strongly Raman
0
0
0
active A1 vibration or be a consequence of crystal packing.
The doublet is present even in the spectra of symmetrically
substituted compounds. While the third band is not found in the
spectrum of 2d its intensity is much larger in the spectra of
ester derivatives 2b. As expected, intensity of the acetylenic
bands in the spectra of donor/acceptor trimers 2c and 2d is
considerably larger. The vibrational spectra have been
recorded only for the solid phase and the influence of the
intermolecular interactions on the vibrational intensities cannot
be excluded. However, since the annulenic cores are
surrounded by bulky substituents the observed intensity
changes actually prove that substituents induce significant
intramolecular charge redistribution in the annulene bridges
Fig. 6. Electronic absorption spectra of 1c in different solvents.

122
B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
Table 5
Mulliken charges of selected atoms in 1 (Scheme 1)
1a
1ba
1db
Atom
Charge
Atom
Charge
Atom
Charge
A, J
B, I
K0.073
K0.218
K0.351
K0.256
0.502
A
B
C
D
E
F
0.638
0.136
A, J
B, I
0.600
K0.178
K0.617
K0.407
0.489
C, H
D, G
E, F
K0.441
K0.416
0.536
C, H
D, G
E, F
0.567
G
H
I
K0.446
K0.468
0.223
J
K0.394
Sumc
Dd
K1.584
0
K0.130
1.454
K0.452
1.132
a
The calculated 1b structure has C2 symmetry with the axis perpendicular to the annulenic plane.
The calculated 1d structure has D2h symmetry.
b
c
d
The sum of Mulliken charges on all the carbon atoms of the benzoannulenic core.
The difference between the sums of Mulliken charges.
Table 6
Mulliken charges of selected atoms in 2 (Scheme 1)
2a
2b
2e
Atom
Charge
Atom
Charge
Atom
Charge
A, J
B, I
0.038
K0.958
0.443
A
B
C
D
E
F
0.225
0.168
A, J
B, I
0.585
K0.619
K0.292
K1.018
1.070
C, H
D, G
E, F
0.122
C, H
D,G
E, F
K0.985
1.062
K1.202
1.052
1.101
G
H
I
K1.224
0.005
0.192
J
K0.680
0.225
K
L
0.168
Suma
Db
K2.400
0
0.456
2.856
K1.644
0.756
a
The sum of Mulliken charges on all the carbon atoms of the benzoannulenic core.
The difference between the sums of Mulliken charges.
b
Fig. 7. Observed (black) IR acetylenic bands of 1 and 2 normalized relative to common alkyl band at 2920 cmK1 (KBr pellets) and calculated (gray).

B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
123
Table 7
Observed and calculated wavenumbers and IR intensities of 1 in the region of acetylenic stretchings
1a
1b
1c
1d
IRa
Calc.b
IRc
Rd
Calc.b
IRc
Rd
Calc.b
IRc
Rd
Calc.b
2210 Ag 0.0
2201 B2u 14.6 2194me
2198s
2209 A 0.0 2192s
2200 B 20.0 sh
2141 A 0.0
2191s
2208 A 83.6
2198 A 20.0
2139 A 2.2
2129 A 24.5
2198s
2208 Ag 0.0
2199 B2u 20.5
2138 B1g 0.0
2127 B3u 6.9
2187s
2189s
2140 B1g 0.0
2136w
2133w
2128w
f
2116m
2130 B3u 17.3 2127s
2131 B 52.9 2121w
–
a
Ref. [18].
b
c
d
e
f
Scaled by 0.963 [37].
IR (KBr).
Raman (powder).
s, strong in this region; m, medium; w, weak; sh, shoulder.
Not observed.
Table 8
Observed and calculated wavenumbers and IR intensities of 2 in the region of acetylenic stretchings
2a
2b
2c
2d
2e
IRa,b
Calc.c
IRa
Rd
Calc.
IRa
Rd
Calc.
IRa
Rd
Calc.
IRa
Rd
Calc.
2221 E0
2.7
2221 E0
2218 E0
3.1
2218 E0
2207s
2216w
2218m
2220 E
12.8
2216m
2219 A
15.6
2219 A
43.5
2214m
2198s
2214m
2204w
2148s
2220 E
12.8
2214m
2200m
2218 B
37.6
2214sh
2198s
2217 A
41.6
2.7
3.1
0
0
0
0
2201s
2194 A
0.0
2199s
2192 A
84.8
2192 A
120.1
2152 A
2.1
2197s
2195 A1
0.0
2192 A1
0.0
2154 A
0.0
2153 B
6.8
2153 A2
0.0
2147 E0
2151 A2
0.0
2146 E0
2148 E
35.4
2146w
2147 B
32.6
2146 A
14.0
2146w
8.3
2147 E0
8.3
1.8
2146 E0
1.8
2148 E
35.4
2147 A
8.4
2145 A
4.9
a
IR (KBr).
b
c
d
No other bands listed [5].
Scaled by 0.963 [37].
Raman (powder).
while the carbon atoms of the third acetylenic bridge are
practically motionless (Scheme 1). However, the most
sensitive again is the totally symmetric mode A10 which causes
the greatest changes of the permanent dipole moment.
The strongest Raman modes in 1c, 2c and 2d also exhibit
strong IR activity and therefore, these compounds are expected
to have large vibrational hyperpolarizibilities [15]. The same
mode is also expected to be strongly coupled with the first
allowed electronic transition. It is now obvious that the totally
symmetrical acetylenic stretching in either case is an effective
conjugation coordinate [38].
we conclude that dimeric compounds are in the solid phase
predominantly stabilized by bulky substituents whereas the
donor or acceptor nature of substituents has far less influence.
According to our knowledge dimeric macrocycle 1c is the first
example of push–pull Eglinton–Galbraith dimer derivative.
Although IR spectra of the compounds 1c, 2c and 2d show
significant intensity, i.e. charge redistribution in the benzoan-
nulenic ring, it is not sufficient to induce dramatic shifts in the
corresponding UV/vis spectra. Since, the extreme polarization
of the conjugated backbone is present in some push–pull
dehydrobenzo[18]annulenes [16], it seems that lack of its
occurrence in the case of our compounds can be attributed to
weak electron-donor and -acceptor properties of alkyl and ester
substituents. Similar level of annulenic ring polarization in the
dimer 1c and trimers 2c and 2d therefore, suggests that
substantial polarization of the conjugated backbone in the
dimeric dehydrobenzo[12]annulene could be induced by other
electron-donor and -acceptor groups (specifically, amino and
nitro groups) , thus creating good candidates for use as novel
photonic materials. Most important advantage of dehydroben-
zo[12]annulenes over trimeric dehydrobenzo[18]annulenes is
their greater solubility which is very important regarding
4. Conclusions
In summary, we have reported the preparation, character-
ization and structure determination of the Eglinton–Galbraith
dimer derivatives with electron acceptor (1b) and donor (1d)
groups, and analogous trimeric macrocycles 2b and 2e, along
with donor/acceptor substituted macrocycles 1c, 2c and 2d.
Using the intermolecular route [16] we have nevertheless,
succeeded in obtaining the product of C2v symmetry in
macroscopic amounts. Contrary to the previous research [14]

124
B. Zimmermann et al. / Journal of Molecular Structure 794 (2006) 115–124
[17] S.A. Sarkar, J.J. Pak, G.W. Rayfield, M.M. Haley, J. Mater. Chem. 11
(2001) 2943.
common poor solubility of dehydrobenzoannulenes. On the
other hand, even though these dimeric macrocycles are quite
stable at room temperature, they are decomposing more rapidly
and at much lower temperatures than their trimeric counter-
parts [14,16].
[18] O.M. Behr, G. Eglinton, A.R. Galbraith, R.A. Raphael, J. Chem. Soc.
(1960) 3614.
[19] U.H.F. Bunz, V. Enkelmann, Chem. Eur. J. 5 (1999) 263.
[20] T. Nishinaga, H. Nakayama, N. Nodera, K. Komatsu, Tetrahedron Lett.
39 (1998) 7139.
[21] T. Nishinaga, N. Nodera, Y. Miyata, K. Komatsu, J. Org. Chem. 67 (2002)
6091.
Acknowledgements
[22] K. Harms, S. Wocadlo, University of Marburg, Marburg, Germany, 1995.
[23] L.J. Farrugia, J. Appl. Crystallogr. 32 (1999) 837.
[24] G.M. Sheldrick, SHELX97: Program for the Refinement of Crystal
This work was supported by the Ministry of Education,
Science and Sport of the Republic of Croatia (Project no.
0098057). The generous supply of experimental time on the
¨
¨
Structures, Universitat of Gottingen, Germany, 1997.
[25] A.L. Spek, PLATON98: A Multipurpose Crystallographic Tool 101201
Version, University of Utrecht, Utrecht, The Netherlands, 2001.
[26] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R.
Cheeseman, J.A. Montgomery, Jr., T. Vreven, K.N. Kudin, J.C. Burant,
J.M. Millam, S.S. Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi,
G. Scalmani, N. Rega, G.A. Petersson, H. Nakatsuji, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox, H.P. Hratchian, J.B. Cross,
C. Adamo, J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J.
Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y. Ayala, K. Morokuma,
G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S. Dapprich,
A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S.
Clifford, J. Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng,
A. Nanayakkara, M. Challacombe, P.M.W. Gill, B. Johnson, W. Chen,
M.W. Wong, C. Gonzalez, J.A. Pople, GAUSSIAN 03, Revision C.02,
Gaussian, Inc., Wallingford, CT, 2004.
´
Raman spectrometer by Prof. Z. Meic (Department of
Analytical Chemistry, Faculty of Sciences, University of
Zagreb) is gratefully acknowledged.
Supplementary data
Supplementary data (synthesis and full characterization of
all compounds) associated with this article can be found, in the
online version, at doi:10.1016/j.molstruc.2006.01.049. [39]
References
[1] J.A. Marsden, M.M. Haley, J. Org. Chem. 70 (2005) 10213.
[2] W. Brad Wan, M.M. Haley, J. Org. Chem. 66 (2001) 3893.
[3] W. Brad Wan, S.C. Brand, J.J. Pak, M.M. Haley, Chem. Eur. J. 6 (2000)
2044.
[27] W. Koch, M.C. Holthausen, A Chemist’s Guide to Density Functional
Theory, second ed., Wiley–VCH, New York, 2001.
[4] U.H.F. Bunz, Y. Rubin, Y. Tobe, Chem. Soc. Rev. 28 (1999) 107.
[5] M.M. Haley, S.C. Brand, J.J. Pak, Angew. Chem. Int. Ed. 36 (1997) 836.
[6] R.H. Baughman, H. Eckhardt, M. Kertesz, J. Chem. Phys. 87 (1987) 6687.
[7] F. Diederich, Nature 369 (1994) 199.
[28] J. Anthony, C.B. Knobler, F. Diederich, Angew. Chem. Int. Ed. Engl. 32
(1993) 406.
[29] F.R. Fronczek, M.S. Erickson, J. Chem. Cryst. 25 (1995) 737.
[30] L. Guo, J.D. Bradshaw, C.A. Tessier, W.J. Youngs, J. Chem. Soc. Chem.
Commun. (1994) 243.
[8] M.M. Haley, Synth. Lett. (1998) 557.
[9] M.M. Haley, J.J. Pak, S.C. Brand, Topp. Curr. Chem. 201 (1999) 81.
[10] J.A. Marsden, G.J. Palmer, M.M. Haley, Eur. J. Org. Chem. (2003) 2355.
[11] J. Anthony, A.M. Boldi, C. Boudon, J.P. Gisselbrecht, M. Gross, P. Seiler,
C.B. Knobler, F. Diederich, Helv. Chim. Acta 78 (1995) 797.
[12] T. Nishinaga, Y. Miyata, N. Nodera, K. Komatsu, Tetrahedron 60 (2004)
3375.
[31] J. Juselius, D. Sundholm, Phys. Chem. Chem. Phys. 3 (2001) 2433.
[32] A.J. Boydston, M.M. Haley, Org. Lett. 3 (2001) 3599.
[33] J.A. Marsden, J.J. Miller, L.D. Shirtcliff, M.M. Haley, J. Am. Chem. Soc.
127 (2005) 2464.
[34] M.E. Gallagher, J.E. Anthony, Tetrahedron Lett. 42 (2001) 7533.
[35] A.R. Katritzky, R.D. Topsom, Chem. Rev. 77 (1977) 639.
[36] F. Mitzel, C. Boudon, J.-P. Gisselbrecht, M. Gross, F. Diederich, Chem.
Commun. (2002) 2318.
[13] J.D. Tovar, N. Jux, T. Jarrosson, S.I. Khan, Y. Rubin, J. Org. Chem. 63
(1998) 4856.
[14] Q. Zhou, P.J. Carroll, T.M. Swager, J. Org. Chem. 59 (1994) 1294.
[15] X. Zhou, A.-M. Reu, J.-K. Feng, X.-J. Liu, Can. J. Chem. Rev. Can. Chim.
82 (2004) 1172.
ˇ
´
[37] N. Biliskov, B. Zimmermann, G. Baranovic, J. Mol. Struct. 661 (2003) 65.
[38] M. Tommasini, C. Castiglioni, M. Del Zoppo, G. Zerbi, J. Mol. Struct.
480/481 (1999) 179.
[16] J.J. Pak, T.J.R. Weakley, M.M. Haley, J. Am. Chem. Soc. 121 (1999)
8182.
[39] Q. Zhou, T.M. Swager, J. Am. Chem. Soc. 117 (1995) 12593.