Tuning second-order optical nonlinearities in push-pull benzimidazoles

Article abstract of DOI:10.1002/ejoc.200300786

The synthesis and full characterization of new chromophores with second order optical nonlinearities containing the 2-phenyl-(5,6)-nitrobenzimidazole group is reported. Starting from 2-{4-[(4-N,N-dihydroxyethylamino)phenylazo] phenyl}-5(6)-nitrobenzimidazole, a combined theoretical and experimental approach, including theoretical computations of second order nonlinear optical activity (MNDO/AM1), X-ray structural analysis and synthetic strategies, has led to a significant optimization (more than 50%) of the NLO activity of related chromophores. The results indicate that 6-nitro-substituted compounds are more active than 5-nitro-substituted ones and that a further increase of NLO activity can be achieved by insertion of a carbon-carbon double bond between the benzimidazole and 2-phenyl rings. Calculations also suggest that an additional improvement of the nonlinearity should be expected upon functionalization of the N1 atom of the 6-nitrobenzimidazole with electron-withdrawing groups. Experimental nonlinearities (EFISH technique, μβ/ 10^-48 esu, γ = 1.907 μm, DMF solution) between 940 and 1550 were measured. ( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004).

Full text of DOI:10.1002/ejoc.200300786

FULL PAPER
Tuning Second-Order Optical Nonlinearities in Push-Pull Benzimidazoles
Antonio Carella,[a] Roberto Centore,*[a] Alain Fort,[b] Andrea Peluso,[c] Augusto Sirigu,[a] and
Angela Tuzi[a]
Keywords: Azo compounds / Chromophores / Heterocycles / Nonlinear optics
The synthesis and full characterization of new chromophores
with second order optical nonlinearities containing the 2-
phenyl-(5,6)-nitrobenzimidazole group is reported. Starting
from 2-{4-[(4-N,N-dihydroxyethylamino)phenylazo]phenyl}-
ones and that a further increase of NLO activity can be
achieved by insertion of a carbon−carbon double bond be-
tween the benzimidazole and 2-phenyl rings. Calculations
also suggest that an additional improvement of the nonlin-
5(6)-nitrobenzimidazole, a combined theoretical and experi- earity should be expected upon functionalization of the N1
mental approach, including theoretical computations of se- atom of the 6-nitrobenzimidazole with electron-withdrawing
cond order nonlinear optical activity (MNDO/AM1), X-ray
structural analysis and synthetic strategies, has led to a signi-
groups. Experimental nonlinearities (EFISH technique, µβ/
10−48 esu, λ = 1.907 µm, DMF solution) between 940 and 1550
ficant optimization (more than 50%) of the NLO activity of were measured.
related chromophores. The results indicate that 6-nitro-sub-
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
stituted compounds are more active than 5-nitro-substituted Germany, 2004)
Introduction
emphasise: high NLO activity, high chemical, photochem-
ical and thermal stability, possibility of chemical func-
tionalization (e.g. as diol or methacrylate ester) for poly-
merization or crosslinking, low tendency to give structured
phases (crystalline or liquid-crystalline), low optical absorp-
tion in the 1.3 and 1.55 µm telecommunication bands (and
corresponding second harmonic wavelengths), ease and low
cost of synthesis.
In particular, chromophores bearing low aromaticity het-
erocycles along the conjugation path have been investigated
both theoretically and experimentally and are considered as
the most suitable for potential applications.[4]
Over the last few years we have systematically studied
the properties of polymers containing the 2-phenyl-6-nitro-
benzoxazole system in the chromophore moiety, where ther-
mochemical stability and a fairly good NLO activity have
been combined.[5] A closely related heterocycle system is 2-
phenyl-(5,6)-nitrobenzimidazole (Scheme 1). Similar to
benzoxazole, it shows excellent chemical and thermal sta-
bility and is relatively easy to synthesise.[6] Tautomerism is
a feature of compounds containing a benzimidazole frag-
ment due to the acid nature of the hydrogen atom bonded
to nitrogen. An interesting feature of benzimidazole is the
possibility of chemical functionalization at the nitrogen of
the pentaatomic ring (for instance by alkylation), which re-
moves the possibility of tautomerism (giving rise to two
constitutional isomers) and introduces a new potentially
useful chemical variable for the optimization of the NLO
activity of the chromophore (e.g. introduction of groups
with suitable electronic properties). The introduction of
The study of molecular hyperpolarizabilities is a field of
growing interest as it has been shown quite recently that
organic compounds with high first order hyperpolarizabili-
ties can be used as active components for electro-optical
modulators capable of solving the main problem of fiber
optic communication technology — the low voltage modu-
lation at very high frequency (THz) in fiber optic communi-
cation links.[1]
The formulation of organic or hybrid (organic-inorganic)
materials for applications in second order nonlinear optics
(NLO) generally involves two distinct steps: first the prep-
aration of an organic chromophore of adequate NLO ac-
tivity and proper chemical functionalization (the two level
model[2] provides the basic tool for designing NLO active
compounds), and secondly its covalent or non-covalent in-
sertion into a polymer or a three-dimensional network with
successive (or simultaneous) polar alignment of the NLO-
active segments.[3]
The choice of the chromophore is crucial because several
key requirements must be satisfied.[3] Among those we can
[a]
`
Dipartimento di Chimica, Universita di Napoli ‘‘Federico II’’,
Via Cinthia, 80126 Napoli, Italy
E-mail: roberto.centore@unina.it
[b]
[c]
´
IPCMS-CNRS,
Groupe
d’Optique
Nonlineaire
et
´
d’Optoelelectronique,
23 Rue du Loess, 67037 Strasbourg Cedex, France
`
Dipartimento di Chimica, Universita degli Studi di Salerno,
Via S. Allende, 84081 Baronissi, Salerno, Italy
Supporting information for this article is available on the
WWW under http://www.eurjoc.org or from the author.
2620
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejoc.200300786
Eur. J. Org. Chem. 2004, 2620Ϫ2626

Tuning Second-Order Optical Nonlinearities in Push-Pull Benzimidazoles
FULL PAPER
bulky lateral substituents on the chromophore is expected 1, for which µβ values are 80 and 580 (1.9 µm, 10Ϫ48 esu),
to positively affect the NLO performances of the material respectively.[3a]
by reducing the extent of centrosymmetric dipolar coupling
of the chromophores during poling.[3]
Table 1. Thermal properties of the chromophores
M.p. (°C)[a]
Tdec. (°C)[b]
BI1
Ϫ
221
Ϫ
296
295
292
314
BI2
BIC1
BIC2
270
Scheme 1
[a] Melting temperature. [b] Decomposition temperature taken as the
temperature corresponding to 5% weight loss in the thermogravi-
metric run (10 K/min, air).
In this paper we report the synthesis and characteriz-
ation, including experimental (EFISH) and theoretical
evaluation of second order molecular nonlinearity, of the
benzimidazole chromophores shown in Scheme 2.
Table 2. Experimental NLO properties of chromophores
µβ (10Ϫ48 esu)[a]
λmax (nm)[b]
µβ(0) (10Ϫ48 esu)[c]
The chromophores are functionalized as diols and there-
fore they can be used as monomers in polycondensation
reactions (acetylated models of BI2 and BIC2 have also
been prepared for the X-ray structural characterization).
In spite of their potential benefits, benzimidazole com-
pounds tailored for second order NLO applications have
received little consideration up to now. We have found only
two reports in the literature in which the properties of poly-
mers based on chromophores similar to BI2, but based on
the 5-nitro isomer rather than the 6-nitro isomer, were stud-
ied.[7] The 5-nitro isomer is the less NLO active isomer, as
BI1
940
950
1550
1400
472
466
482
487
674
684
1074
968
BI2
BIC1
BIC2
[a]
[b]
Measured by the EFISH technique in DMF at 1.907 µm.
[c]
Measured in DMF. Extrapolated to zero frequency by the two-
levels model.
Table 3. Calculated NLO properties of the chromophores
´
inferred from Kekule conjugation schemes. Therefore, this
µ (D) βvec (10Ϫ30 esu) µβvec(0) (10Ϫ48 esu)
paper is the first report on the molecular second order NLO
activity of 6-nitrobenzimidazole chromophores.
BI1
BI2
tautomer 5 11.9
tautomer 6 10.8
10.9
38.2
48.2
44.5
46.4
53.9
53.3
68.5
454.6
520.6
485
505.8
587.0
581.0
746.6
Results and Discussion
BIC1 tautomer 5 10.9
tautomer 6 10.9
BIC2
BICF
10.9
10.9
The present work is a combined experimental and theor-
etical approach in which synthetic strategies and calcu-
lations of first order molecular hyperpolarizabilities lead to
the design of new NLO-active organic molecules with im-
It is well-known that accuracy of computed hyperpolariz-
proved performances. The thermal properties of the chrom- abilities is only achievable with very high computational
ophores are reported in Table 1. A fair thermal stability is costs.[8] However, in this context, reliable predictions of
observed, with decomposition temperatures close to 300 °C trends is the only thing which really matters, since the se-
for all the chromophores. Measured and calculated second cond order NLO response of all the chromophores was ex-
order nonlinearities, as well as other related quantities are perimentally determined by the EFISH technique. Thus, we
reported in Table 2 and Table 3, respectively. The data in resorted to semiempirical methods because it has been
Table 2, in particular, can be compared with those of the shown by many authors[9] that low-cost semiempirical
standard chromophores p-nitroaniline and Dispersed Red methods, in particular INDO[10] and MNDO[11] with both
Scheme 2
Eur. J. Org. Chem. 2004, 2620Ϫ2626
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2621

A. Carella, R. Centore, A. Fort, A. Peluso, A. Sirigu, A. Tuzi
FULL PAPER
AM1 and PM3 parametrization, are useful in predicting re- alkylated chromophores are described, but the synthetic
liable qualitative trends and relative magnitudes, so as to procedures can be reasonably extended to more bulky
properly address synthetic modifications which improve the alkyl groups.
material’s response.
Theoretical computations on chromophore BI2 show
Theoretical computations performed on both tautomers that the advantage of having only the more active 6-nitro
of BI1 show two important structural features that could isomer is balanced by the increased dihedral angle between
significantly affect its NLO activity. First of all, the benzim- benzimidazole and the 2-phenyl rings (48.2° according to
idazole and the 2-phenyl rings are not in a coplanar ar- MNDO/AM1 full geometry optimization), due to the
rangement; the computed dihedral angle is about 32° for greater steric encumbrance of ethyl as compared to hydro-
both the 5-nitro and 6-nitro tautomers as a consequence gen. Nevertheless, the computed µβ(0) of BI2 is slightly
of the steric repulsion between the NH and ortho-phenyl higher than the Boltzmann average of the µβ(0) of the two
hydrogens. Distortions from coplanarity usually decrease tautomers of BI1 (476.9 ·10Ϫ48 esu), which is consistent
the molecular hyperpolarizabilities, as predicted by both the with the experimental results (Table 2 and 3), thus con-
two-levels model[2] and by theoretical computations.[9a,9b,12] firming the reliability of the computational procedure ad-
In the closely related case of 2-phenylbenzoxazole chrom- opted here in predicting how hyperpolarizability changes as
ophores, theoretical computations have shown a significant the chemical structure of the chromophore is changed.
decrease of β as the corresponding twist becomes greater
The X-ray molecular structure of the diacetate ester of
than about 20°.[12] Secondly, according to the µβ(0) values BI2 (ACBI2) is shown in Figure 1; selected bond lengths,
(see Table 3), the 6-nitro tautomer of BI1 has a significantly bond angles and torsion angles are given in Table 4 for
greater NLO activity than the 5-nitro one (about 10%), ACBI2 and ACBIC2, together with the computed values.
while the latter is slightly more stable than the former, the The computed MNDO/AM1 optimum geometries are in
calculated energy difference being about 0.4 kcal/mol. Thus, good agreement with the X-ray data, the bond lengths be-
˚
computations suggest two ways to improve the NLO ac- ing reproduced to within 0.02 A; only the dihedral angles
tivity of this compound: removal of the tautomerism by
considering only 6-nitro-substituted compounds and release
of the steric hindrance by increasing the distance between
the benzimidazole and 2-phenyl rings (without interrupting
the π conjugation) in order to allow a more coplanar ar-
rangement of the two rings.
With respect to the first point, we synthesized the 6-nitro-
substituted BI2 chromophore by selective alkylation of the
starting nitro-phenylenediamine. In fact, since the amino
group in the para position to the nitro group is much less
nucleophilic than the amino group in the meta position,
alkylation occurs only at the amino group in the 1-position
of 5-nitro-1,2-phenylenediamine. In this paper only N-ethyl at the 30% probability level
Figure 1. ORTEP drawing of ACBI2; thermal ellipsoids are shown
˚
Table 4. Selected bond lengths (A), bond angles (°) and torsion angles (°) for ACBI2 and ACBIC2; calculated values for the optimized
geometry of BI2, BIC2 and BICF at the MNDO/AM1 level are also reported
X-ray
MNDO/AM1
ACBI2
1.364(6)
1.397(6)
1.259(5)
1.418(6)
1.324(5)
ACBIC2
1.376(4)
1.415(4)
1.257(4)
1.421(4)
BI2
BIC2
1.396
1.429
1.232
1.437
BICF
1.396
1.429
1.232
1.437
N1ϪC9
N2ϪC12
N2ϪN3
N3ϪC15
N4ϪC21
N4ϪC23
N5ϪC21
N5ϪC23
1.396
1.429
1.232
1.437
1.323(4)
1.366(4)
1.374(5)
C8ϪN1ϪC9
C4ϪN1ϪC9
C8ϪN1ϪC4
C11ϪC12ϪN2ϪN3
C12ϪN2ϪN3ϪC15
C16ϪC15ϪN3ϪN2
C17ϪC18ϪC21ϪN5
C17ϪC18ϪC21ϪC22
C21ϪC22ϪC23ϪN5
118.9(5); 123.8(7)
121.4(4)
119.0(5)
162.6(4)
177.6(3)
160.7(4)
Ϫ147.0(4)
120.2(3)
121.8(3)
117.0(3)
Ϫ176.1(3)
Ϫ178.6(3)
Ϫ162.9(3)
121.2
121.2
121.2
117.3
174.2
179.1
154.1
131.8
117.2
175.8
179.3
160.7
117.3
175.2
179.2
157.1
164.9(4)
174.4(4)
158.2
162.7
162.5
180.0
2622
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2620Ϫ2626

Tuning Second-Order Optical Nonlinearities in Push-Pull Benzimidazoles
referring to the orientations of some rings are slightly differ-
FULL PAPER
In fact, as compared with ACBI2, the chromophore takes
ent, but this can probably be ascribed to packing effects.
up a flatter conformation in the crystals [the maximum
The observed dihedral angle between the benzimidazole atomic distance from the calculated average plane of the
˚
and 2-phenyl rings of ACBI2 [37.5(2)°; 48.2° according to conjugated part is 0.367(4) A for non-H atoms]. The benz-
the MNDO/AM1 optimization of BI2] is determined by the imidazole and vinyl groups are close to being coplanar [the
˚
intramolecular close contact C19···C28 (3.206 A). A non- dihedral angle is 6.6(4)° between the mean planes] and so
coplanar arrangement of the phenyl rings linked to the azo are the benzimidazole and the phenyl ring C9 to C14 [di-
group is also observed [dihedral angle 40.3(2)° between hedral angle of 2.2(2)°], while the dihedral angle between
mean planes] as a result of small torsions (Ͻ20°) around the benzimidazole and phenyl ring C15 to C20 is 22.4(2)°
the bonds C15ϪN3 and C12ϪN2, in close agreement with (19° in the MNDO/AM1 computed geometry of BIC2).
the results yielded by MNDO/AM1 geometry optimization The dihedral angle between the planes of the phenyl rings
(37.4°). As we have reported for the homologous benzo- attached to the azo group [22.0(2)°; 23° in the optimized
xazole chromophore,[12] small torsions around these bonds, geometry of BIC2] is smaller than in ACBI2 and it is essen-
in particular around C15ϪN3, have low energetic costs and, tially due to the torsion around the N3ϪC15 bond. Again,
at variance to the torsions around the C18ϪC21 bond, no this torsion is expected to be of low energy and without
significant effect on β.
significant influence on β.[12]
As concerns the second point suggested by compu-
In agreement with the theoretical prediction, an increase
tations — the release of steric hindrance between the benzi- (ϩ47%) is observed in the experimental value of µβ on
midazole and 2-phenyl rings — a good compromise be- going from BI2 to BIC2 (see Table 2).
tween the predicted hyperpolarizabilities and the synthetic
As concerns the alkylation of the 6-nitrobenzimidazole
procedures suggested that the insertion of a CϭC double group, the experimental results in Table 2 show that BIC1
bond between the benzimidazole and 2-phenyl rings should has a slightly higher NLO activity than BIC2. Calculations
lead to a significant increase of the second order NLO re- also indicate that the 6-nitro tautomer of BIC1 is more ac-
sponse of the chromophore. Insertion of the vinylene group tive than BIC2 (see Table 3); this effect could be due to the
was accomplished in a satisfactory manner by using cin- different electronic character of H and ethyl (ethyl is a slight
namic acid derivatives (instead of benzoic acid) in building electron donor group). Calculations performed replacing
the benzimidazole moiety. The different structures are given the N-ethyl group of BIC2 by an NϪCF3 group (compound
in Scheme 3.
BICF in Table 3 and 4) show a strong calculated β increase
(ca. 28%) and this seems to support the possibility of
further improving the second order molecular nonlinearity
by selective alkylation of 6-nitrobenzimidazole with strong
electron-withdrawing groups.
Experimental Section
Scheme 3
Synthesis: Research in the field of second order NLO active com-
pounds is, by necessity, applications oriented. It is therefore worth
noting that all the chromophores in this paper have been synthe-
sized (see Scheme 4) from low-cost commercial precursors [4-nitro-
1,2-phenylenediamine, technical grade p-acetamidobenzaldehyde
and p-aminocinnamic acid hydrochloride, bis(hydroxyethyl)aniline]
in two or three steps, with yields varying from moderate to high;
expensive manipulations under inert atmosphere, as well as hyperd-
rying of reaction solvents, are unnecessary.
The common step for the synthesis of all chromophores is the last
one in Scheme 4 — diazotization of a (nitrobenzimidazol-2-yl)-
phenylamine and subsequent coupling with commercial bis(2-
hydroxyethyl)aniline. Although all the (nitrobenzimidazol-2-yl)-
phenylamines could, in principle, be prepared by the same route,
the need to increase the final yields of chromophore and to use
low-cost commercial starting materials suggested the three different
routes shown in Scheme 4.
The presence of the CϭC double bond has two effects,
both pointing toward an enhancement of the hyperpolariz-
ability: a) the 1Ϫ6 H···C contact is replaced by a 1Ϫ6 H···H
contact, allowing for a more planar conformation of the
chromophore; b) the conjugation length is increased.
MNDO/AM1 computations confirm those expectations:
the minimum energy geometry of BIC2 reported in Table 4
exhibits a more planar arrangement and a significant in-
crease of µβ(0) relative to that of BI2 (Table 3). The X-ray
molecular structure of ACBIC2, reported in Figure 2, sup-
ports the above arguments.
2-(4-p-Acetamidophenyl)-5(6)-nitrobenzimidazole (I): p-Acetamido-
benzaldehyde (1.066 g, 6.53 mmol) and 2-amino-4-nitroaniline (1 g,
6.53 mmol) were dissolved in 50 mL of glacial acetic acid at 100
°C. After a few minutes the formation of an orange precipitate
occurred. At this point lead() tetraacetate (95%, 3.048 g,
6.53 mmol) was added in portions. After some minutes the color
of the suspension turned from orange to yellow. The system was
Figure 2. ORTEP drawing of ACBIC2; thermal ellipsoids are
shown at the 30% probability level
Eur. J. Org. Chem. 2004, 2620Ϫ2626
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2623

A. Carella, R. Centore, A. Fort, A. Peluso, A. Sirigu, A. Tuzi
FULL PAPER
Scheme 4
kept at 100 °C for 1 h and, after cooling to room temperature, the
solid was collected by filtration. Yield: 40%. M.p. 344 °C, solid-
2-(4-Acetamidophenyl)-1-ethyl-6-nitrobenzimidazole (II): A solution
of 2-ethylamino-4-nitroaniline (6 g, 33 mmol) and p-acetamidoben-
state phase transition at 183 °C. 1H NMR ([D6]DMSO): δ ϭ 2.078 zaldehyde (5.385 g, 33 mmol) in 30 mL of glacial acetic acid was
(s, 3 H), 7.692Ϫ7.795 (m, 3 H), 8.071Ϫ8.149 (m, 3 H), 8.420 (s, 1
H), 10.247 (s, 1 H), 13.451 (s, 1 H) ppm.
refluxed for 26 h. The solution was then cooled to room tempera-
ture and water was slowly added until the formation of a yellow
precipitate was observed. The solid was then collected by filtration.
Yield 44%. M.p. 281 °C. 1H NMR ([D6]DMSO): δ ϭ 1.333 (t, J ϭ
7.0 Hz, 3 H), 4.466 (q, J ϭ 7.0 Hz, 2 H), 7.815 (m, 5 H), 8.133
(dd, J1 ϭ 9, J2ϭ 2.4 Hz, 1 H), 8.648 (d, J ϭ 2.4 Hz, 1 H), 10.261
(s, 1 H) ppm.
2-(4-Aminophenyl)-5(6)-nitrobenzimidazole (AI): AI aniline was pre-
pared in the same way as AII aniline (see below).Yield 90%. M.p.
1
259 °C. H NMR ([D6]DMSO): δ ϭ 5.778 (broad, 2 H), 6.726 (d,
J ϭ 8.3 Hz, 2 H), 7.703 (d, J ϭ 9.28 Hz, 1 H), 7.892 (d, J ϭ 8.3
Hz, 2 H), 8.132 (dd, J1 ϭ 9.03, J2 ϭ 1.70 Hz, 1 H), 8.352 (s, 1
H) ppm.
2-(4-Aminophenyl)-1-ethyl-6-nitrobenzimidazole (AII): II (3 g,
9.2 mmol) was added to a solution of 7 mL of 96% H2SO4 and
55 mL of water. The resulting suspension was refluxed for 90 mi-
nutes. As the temperature increased the solid dissolved to give a
dark solution. The solution was then cooled to room temperature
and a solution of NaOH (10% by weight) was added dropwise up
to pH 5. The formation of a yellow solid was observed which was
collected by filtration. Yield 93%. M.p. 211 °C. 1H NMR
([D6]DMSO): δ ϭ 1.368 (t, J ϭ 7.4 Hz, 3 H), 4.450 (q, J ϭ 7.4 Hz,
2 H), 5.797 (s, 2 H), 6.719 (d, J ϭ 8.4 Hz, 2 H), 7.561 (d, J ϭ 8.2
Hz, 2 H), 7.736 (d, J ϭ 8.8 Hz, 1 H), 8.105 (dd, J1ϭ 9, J2 ϭ 1.2 Hz,
1 H), 8.555 (d, J ϭ 1.4 Hz, 1 H) ppm.
2-Ethylamino-4-nitroaniline: 2-Amino-4-nitroaniline (4 g, 26 mmol)
was dissolved in 15 mL of DMF. K2CO3 (7.076 g, 52 mmol) and
bromoethane (1.9 g, 26 mmol) were then added to the solution and
the mixture was stirred at room temperature for 5 days. The inor-
ganic solids were then filtered off and the remaining DMF solution
was poured into 100 mL of water. Precipitation of a red solid was
observed. The system was stirred in an ice bath for 20 minutes,
then the solid was collected by filtration and recrystallized from
EtOH/H2O. Orange needle-like crystals were obtained. Yield: 60%.
1
M.p. 130 °C. H NMR ([D6]acetone): δ ϭ 1.309 (t, J ϭ 7.1 Hz, 3
H), 3.243 (q, J ϭ 6.8 Hz, 2 H), 4.298 (s, 1 H), 5.473 (s, 2 H), 6.711
(d, J ϭ 8.8 Hz, 1 H), 7.348 (d, J ϭ 2 Hz, 2 H), 7.553 (dd, J1
8.6, J2 ϭ2 Hz, 1 H) ppm.
ϭ
(E)-2-(4-Aminostyryl)-5(6)-nitrobenzimidazole (AIII): 2-Amino-4-
nitroaniline (3.00 g, 19 mmol) and p-aminocinnamic acid hydro-
2624
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2620Ϫ2626

Tuning Second-Order Optical Nonlinearities in Push-Pull Benzimidazoles
FULL PAPER
chloride (3.339 g, 17 mmol) were mixed with a sufficient quantity
of polyphosphoric acid to give a stirrable paste. The temperature 3.565 (s, 8 H), 4.635 (q, J ϭ 6.8 Hz, 2 H), 4.840 (s, 2 H), 6.848 (d,
at 240 °C). 1H NMR ([D6]DMSO): δ ϭ 1.350 (t, J ϭ 6.8 Hz, 3 H),
was slowly raised to 200 °C. As the temperature increased a dark,
viscous solution formed. This solution was stirred at 200 °C for 4
h, then cooled to 100 °C and poured into 400 mL water of in an
ice bath. The formation of a solid occurred immediately. The sys-
tem was basified with aqueous NaOH (50% by weight) to pH 6.
The resulting solid was collected by filtration and then treated with
boiling ethanol. The insoluble precipitate was removed by filtration
and the filtrate was poured into water. The final precipitate was
collected by filtration and dried in an oven. Yield: 59%. M.p. 260
J ϭ 9.4 Hz, 2 H), 7.619 (d, J ϭ 15.6 Hz, 1 H), 7.744Ϫ7.829 (m, 5
H), 7.976Ϫ8.010 (m, 3 H), 8.012 (dd, J1 ϭ 8.8, J2 ϭ2 Hz, 1 H),
8.603 (d, J ϭ 2 Hz, 1 H) ppm. C27H28N6O4: calcd. C 64.79, H 5.64,
N 16.79; found C 64.52, H 5.69, N 16.56.
2-[4-[(4-Bis(2-acetyloxyethyl)amino]phenylazo}phenyl)-1-ethyl-6-
nitrobenzimidazole (ACBI2) and (E)-2-(4-{[4-Bis(2-acetyloxy-
ethyl)amino]phenylazo}styryl)-1-ethyl-6-nitrobenzimidazole
(AC-
BIC2): The synthesis of the diacetate esters of BI2 and BIC2 used
in the X-ray structure analysis was performed in a similar way to
BI2 and BIC2 but using bis(2-acetyloxyethyl)aniline instead of
bis(2-hydroxyethyl)aniline in the coupling reaction.
1
°C. H NMR ([D6]DMSO): δ ϭ 5.665 (s, 2 H), 6.602 (d, J ϭ 8.4
Hz, 2 H), 6.875 (d, J ϭ 16 Hz, 1 H), 7.380 (d, J ϭ 7.8 Hz, 2 H),
7.627 (d, 2 H), 8.063 (d, J ϭ 8.8 Hz, 1 H), 8.335 (s, 1 H), 13.080
(s, 1 H) ppm.
ACBI2: M.p. 145 °C (solid state phase transition at 132 °C). 1H
NMR ([D1]CDCl3): δ ϭ 1.592 (t, J ϭ 7.2 Hz, 3 H), 2.071 (s, 6 H),
3.753 (t, J ϭ 5.8 Hz, 4 H), 4.317 (t, J ϭ 5.8 Hz, 4 H), 4.460 (q,
J ϭ 7.2 Hz, 2 H), 6.875 (d, J ϭ 8.2 Hz, 2 H), 7.912Ϫ8.071 (m, 7
H), 8.298 (d, J ϭ 8.8 Hz, 1 H), 8.444 (s, 1 H) ppm.
(E)-2-(4-Aminostyryl)-1-ethyl-6-nitrobenzimidazole (AIV): This syn-
thesis was performed similarly to that for AIII, the only difference
being that the raw product did not need any purification, and so
the step with boiling ethanol could be avoided (this greatly im-
proved the yield). Yield: 95%. M.p. 260 °C. 1H NMR ([D6]DMSO):
δ ϭ 1.350 (t, J ϭ 6.8 Hz, 3 H), 4.527 (q, J ϭ 7 Hz, 2 H), 6.597 (d,
J ϭ 8.4 Hz, 2 H), 7.113 (d, J ϭ 15.8 Hz, 1 H), 7.113 (d, J ϭ 8.4
Hz, 2 H), 7.648 (d, J ϭ 8.8 Hz, 1 H), 7.839 (d, J ϭ 15.6 Hz, 1 H),
8.054 (d, J ϭ 8.8 Hz, 1 H), 8.503 (s, 1 H) ppm.
ACBIC2: M.p. 157 °C. 1H NMR (D1]CDCl3): δ ϭ 1.551 (t, J ϭ
7.8 Hz, 3 H), 2.161 (s, 6 H), 3.731 (t, J ϭ 6.4 Hz, 4 H), 4.303 (t,
J ϭ 5.8 Hz, 4 H), 4.424 (q, J ϭ 7.4 Hz, 2 H), 6.840 (d, J ϭ 8.8
Hz, 2 H), 7.118 (d, J ϭ 15.6 Hz, 1 H), 7.722Ϫ7.902 (m, 7 H), 8.177
(d, J ϭ 8.4 Hz, 1 H), 8.239 (s, 1 H), 8.308 (s, 1 H) ppm.
2-(4-{[4-Bis(2-hydroxyethyl)amino]phenylazo}phenyl)-5(6)-nitro-
benzimidazole (BI1): AI aniline (3 g, 11.8 mmol) was placed in a
flask containing 30 mL of H2O and 3 mL of 37% HCl; the suspen-
sion was cooled to 0Ϫ5 °C in an ice-water bath. The solution ob-
tained upon dissolving NaNO2 (0.899 g, 13.0 mmol) in about
10 mL of water was then added dropwise to the suspension whilst
stirring. Stirring at low temperature was continued for 30 min after
the addition of the nitrite solution. A water-ethanol solution con-
taining sodium acetate (3.3 g, 0.04 mol) and bis(2-hydroxyethyl)-
aniline (2.136 g, 11.8 mmol) was prepared separately. The suspen-
sion containing the diazonium salt was rapidly added to this solu-
tion whilst stirring. The mixture became red immediately and after
a few seconds a red-brown precipitate of the azo compound
formed. This compound was collected by filtration and then recrys-
tallized from DMF/H2O to give pure BI1 as a microcrystalline
brown-violet product. Yield 71%. 1H NMR ([D6]DMSO): δ ϭ
3.646 (s, 8 H), 4.907 (s, 2 H), 6.925 (d, J ϭ 8.79 Hz, 2 H), 7.845
(d, 3 H), 7.875 (d, J ϭ 8.24 Hz, 2 H), 8.184 (d, J ϭ 7.96 Hz, 1 H),
8.389 (d, J ϭ 8.24 Hz, 2 H), 8.593 (s, 1 H), 13.728 (s, 1 H) ppm.
C23H22N6O4: calcd. C 61.88, H 4.97, N 18.82; found C 61.69, H
5.09, N 18.95.
Physical Measurements: The thermal behavior of the compounds
was studied by differential scanning calorimetry (PerkinϪElmer
DSC-7, under nitrogen, scanning rate 10 K/min), temperature-con-
trolled polarizing microscopy (Zeiss microscope, Mettler FP5 mic-
rofurnace) and TGA-DTA analysis (TA Instruments, air, 10 K/
min). 1H NMR spectra were recorded with Varian XL 200 spec-
trometer.
X-ray Analysis: Single crystals were obtained by slow evaporation
from a toluene solution for ACBI2 and from a chloroform/heptane
solution for ACBIC2. For ACBI2 the crystal phase investigated was
the one giving the solid-state phase transition at 132 °C; for AC-
BIC2 the low melting (134 °C) crystal phase was studied. Data
collections were performed on an Enraf Nonius MACH3 auto-
mated diffractometer, using graphite-monochromated Mo-Kα radi-
˚
ation (λ ϭ 0.71069 A). No absorption correction was applied. The
structures were solved by direct methods (SIR92 program[13]) and
refined by the full-matrix least-squares method (SHELXL program
of the SHELX-97 package[14]). In the case of ACBI2, one of the
two acetyloxyethyl tails is disordered over two sites. These were
handled satisfactorily with 0.5 occupation factors and introducing
constraints for some bond lengths. The C, N and O atoms of the
statistical tails were given isotropic displacement parameters; the
remaining C, N and O atoms were refined anisotropically. H atoms
were placed in calculated positions and refined by the riding model
with Uiso equal to Ueq of the carrier atom. Some crystal, collection
and refinement data are reported in Table 5. All crystallographic
data have been deposited with the Cambridge Crystallographic
Data Centre (CCDC). CCDC-226562 (for ACBI2) and -226563
(for ACBIC2) contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge at
www.ccdc.cam.ac.uk/conts/retrieving.html [or from the Cambridge
Crystallographic Data Centre, 12, Union Road, Cambridge
The other three chromophores (BI2, BIC1, BIC2) were synthesized
in the same manner from the corresponding aniline.
1-Ethyl-2-(4-{[4-Bis(2-hydroxyethyl)amino]phenylazo}phenyl)-6-
nitrobenzimidazole (BI2): M.p. 221 °C. 1H NMR ([D6]DMSO): δ ϭ
1.355 (t, J ϭ 7.2 Hz, 3 H), 3.568 (s, 8 H), 4.510 (q, J ϭ 6.8 Hz, 2
H), 4.585 (s, 2 H), 6.855 (d, J ϭ 8.8 Hz, 2 H), 7.758Ϫ7.939 (m, 7
H), 8.142 (d, J ϭ 8.8 Hz, 1 H), 8.676 (s, 1 H) ppm. C25H26N6O4:
calcd. C 63.28, H 5.52, N 17.71; found C 63.28, H 5.56, N 17.62.
(E)-2-(4-{[4-Bis(2-hydroxyethyl)amino]phenylazo}styryl)-5(6)-
1
nitrobenzimidazole (BIC1): H NMR ([D6]DMSO): δ ϭ 3.553 (s, 8
H), 4.585 (s, 2 H), 6.820 (d, J ϭ 8.2 Hz, 2 H), 7.285 (d, J ϭ 16.2
Hz, 1 H), 7.709Ϫ7.775 (m, 8 H), 8.063 (d, J ϭ 8.8 Hz, 1 H), 8.415
(s, 1 H), 13.267 (s, 1 H) ppm. C25H24N6O4: calcd. C 63.55, H 5.12,
N 17.79; found C 63.26, H 5.18, N 17.57.
(E)-1-Ethyl-2-(4-{[4-bis(2-hydroxyethyl)amino]phenylazo}styryl)-6- CB2 1EZ, UK; Fax: (internat.) ϩ44-1223/336-033; E-mail:
nitrobenzimidazole (BIC2): M.p. 270 °C (solid state phase transition
deposit@ccdc.cam.ac.uk].
Eur. J. Org. Chem. 2004, 2620Ϫ2626
www.eurjoc.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2625

A. Carella, R. Centore, A. Fort, A. Peluso, A. Sirigu, A. Tuzi
FULL PAPER
ation of theoretical µβvec’s with respect to the experimental values
(see above).
Table 5. Crystal collection and refinement data for ACBI2 and AC-
BIC2
ACBI2
ACBIC2
Acknowledgments
`
Thanks are due to CIMCF of the Universita di Napoli ‘‘Federico
Chemical formula
Molecular mass
T (K)
C29H30N6O6
558.59
C31H32N6O6
584.63
II’’ for allowing use of the MACH3 Nonius diffractometer. This
research was supported by MIUR of Italy (PRIN 2001 and
PRIN 2002).
293
triclinic, P1
¯
Crystal system, space group
˚
a (A)
9.369(3)
9.910(3)
17.085(5)
78.58(3)
83.08(3)
62.87(3)
1382.9(7)
2, 1.341
0.096
8.739(7)
12.847(3)
13.572(4)
84.37(2)
74.31(7)
89.56(6)
1460(1)
2, 1.330
0.094
˚
[1] [1a]
b (A)
Y. Shi, C. Zhang, H. Zhang, J. H. Bechtel, L. R. Dalton,
B. H. Robinson, W. H. Steier, Science 2000, 288, 119.
Pliska, W.-R. Cho, J. Meier, A.-C. Le Duff, V. Ricci, A. Otomo,
M. Canva, G. I. Stegeman, P. Raimond, F. Kajzar, J. Opt. Soc.,
C. Zhang, L. R. Dalton, M.-C. Oh,
H. Zhang, W. H. Steier, Chem. Mater. 2001, 13, 3043.
Lee, H. E. Katz, C. Erben, D. M. Gill, P. Gopalan, J. D. Heber,
D. J. McGee, Science 2002, 298, 1401. [1e] N. Savage, IEE Spec-
trum 2003, 20.
˚
[1b]
c (A)
T.
α (°)
β (°)
γ (°)
ˇ
[1c]
Am. B 2000, 17, 1554.
3
˚
[1d]
V (A )
M.
Z, Dx (Kg/m3)
µ (mmϪ1
θ range
)
1.22° to 27.97° 1.57° to 27.97°
6659/364
[2] [2a]
Data/parameters
Goodness of fit (GOF) (all data) 1.002
7031/388
0.916
0.0519
0.2067
J. L. Oudar, D. S. Chemla, J. Chem. Phys. 1977, 66, 2664.
J. L. Oudar, J. Chem. Phys. 1977, 67, 446.
L. R. Dalton, A. Harper, A. Ren, F. Wang, G. Todorova,
[2b]
[3] [3a]
R1 on F[IϾ3σ(I)]
0.0756
0.3063
wR2 on F2 (all data)
J. Chen, C. Zhang, M. Lee, Ind. Eng. Chem. Res. 1999, 38, 8.
[3b]
L. Dalton, Adv. Polymer Sci. 2002, 158, 1.
[4] [4a]
V. P. Rao, A. K.-Y. Jen, J. Chandrasekhar, I. N. N. Nam-
[4b]
boothiri, A. Rathna, J. Am. Chem. Soc. 1996, 118, 12443.
I. D. L. Albert, T. J. Marks, M. A. Ratner, J. Am. Chem. Soc.
[4c]
1997, 119, 6575.
A. Abbotto, S. Bradamante, A. Facchetti,
[4d]
G. A. Pagani, J. Org. Chem. 1997, 62, 5755.
M. He, T. M.
Leslie, J. A. Sinicropi, Chem. Mater. 2002, 14, 4662.
[5] [5a]
P. Ambrosanio, R. Centore, S. Concilio, B. Panunzi, A. Si-
[5b]
rigu, N. Tirelli, Polymer 1999, 40, 4923.
T. Beltrani, M.
Bösch, R. Centore, S. Concilio, P. Günter, A. Sirigu, Polymer
2001, 42, 4025. [5c] T. Beltrani, M. Bösch, R. Centore, S. Conci-
NLO Measurements: The EFISH technique was used for the exper-
imental determination of the µβ nonlinearities of the chromophores
(µ is the ground state permanent dipole moment and β the quad-
ratic hyperpolarizability of the molecules). The light source was a
Q-switched Nd:YAG laser, whose emission at 1.06 µm was shifted
to 1.907 µm by stimulated Raman emission in a high-pressure hy-
drogen cell. At this wavelength the molecular optical nonlinearities
measured are to a large extent free from resonance enhancement.
lio, P. Günter, A. Sirigu, J. Polymer Sci.: Part A: Polym. Chem.
[5d]
2001, 39, 1162.
V. Bruno, A. Castaldo, R. Centore, A. Si-
rigu, F. Sarcinelli, M. Casalboni, R. Pizzoferrato, J. Polymer
Sci.: Part A: Polym. Chem. 2002, 40, 1468.
J. B. Wright, Chem. Rev. 1951, 48, 397.
Chem. Rev. 1974, 74, 279.
E. M. Cross, K. M. White, R. S. Moshrefzadeh, C. V.
Francis, Macromolecules 1995, 28, 2526.
Van den Broeck, E. Gubbelmans, W. Ballet, T. Verbiest, A.
Persoons, Optical Materials 2002, 21, 67.
[6] [6a]
[6b]
P. N. Preston,
[7] [7a]
[7b]
C. A. Samyn, K.
Measurements were calibrated relative to a quartz wedge: the ex-
perimental value of d
Quartz
ϭ 1.2·10Ϫ9 esu at 1.06 µm was extrapo-
11
[8]
[9]
A. Willetts, J. E. Rice, D. M. Burland, D. P. Shelton, J. Chem.
Phys. 1992, 97, 7590.
lated to 1.1·10Ϫ9 esu at 1.907 µm. Neglecting the purely electronic
contribution compared to the orientational part and allowing
Kleinman symmetry for rod-like molecules, the projection, β, of
the vector part of the tensor βijk along the direction of the dipole
moment was determined. A detailed description of the experimen-
tal apparatus is given elsewhere.[15] Measurements were performed
in DMF solution.
[9a] N. Matsuzawa, D. A. Dixon, J. Phys. Chem. 1992, 96, 6232.
[9b]
B. Beck, U. W. Grummt, J. Phys. Chem. B 1998, 102, 664.
J. L. Diaz, B. Villacampa, F. Lopez-Calahorra, D. Velasco,
[9c]
Chem. Mater. 2002, 14, 2240.
[10]
I. D. L. Albert, T. J. Marks, M. A. Ratner, J. Phys. Chem. 1996,
100, 9714.
M. J. S. Dewar, W. Thiel, J. Am. Chem. Soc. 1977, 99, 4899.
A. Castaldo, R. Centore, A. Peluso, A. Sirigu, A. Tuzi, Struct.
Chem. 2002, 13, 27.
A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J.
Appl. Crystallogr. 1993, 26, 343.
G. M. Sheldrick, SHELX-97, Program for Crystal Structure
Determination, University of Göttingen, Germany, 1997.
T. Thami, P. Bassoul, M. A. Petit, J. Simon, A. Fort, M. Bar-
zoukas, A. Villaeys, J. Am. Chem. Soc. 1992, 114, 915.
J. J. P. Stewart, QCPE program 455, 1993.
[11]
[12]
Computational Details: Calculations of hyperpolarizabilities were
carried out with the MOPAC package,[16] using the MNDO Hamil-
tonian[11] and the AM1 parametrization.[17] The finite-field ap-
proach,[18] employing both an energy and a dipole expansion, was
used to calculate the response of the charge density to the zero
frequency field. The scalar product µβvec was calculated from
Equation (1) :[19]
[13]
[14]
[15]
[16]
[17]
M. J. S. Dewar, E. G. Zoebisch, E. F. Healy, J. J. P. Stewart, J.
Am. Chem. Soc. 1985, 107, 3902.
A. D. Buckingam, Adv. Chem. Phys. 1967, 12, 107.
A. Kurtz, J. P. P. Stewart, K. M. Dieter, J. Comp. Chem. 1990,
(1)
[18] [18a]
[18b]
H.
All the structures were fully optimized until the gradient norm
dropped below 0.1 kcal/A. Calculations refer to the gas phase; sol-
vent effects are expected to enhance the nonlinear response and
therefore they could be responsible for the significant underestim-
11, 82.
˚
[19]
N. Matsuzawa, D. A. Dixon, Int. J. Quantum Chem. 1992, 44,
497.
Received December 17, 2003
2626
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
Eur. J. Org. Chem. 2004, 2620Ϫ2626