Photochemistry of 3,6-bis(styryl)pyridazines in solution and in neat liquid crystalline phase-optical switching and imaging techniques

Article abstract of DOI:10.1016/j.tet.2008.08.075

3,6-Bis(styryl)pyridazines 1a-f with 2-6 alkoxy groups show on irradiation in solution a stereoisomerization which leads to a photostationary state of (E,E)- and (E,Z)-isomer. Sensitizing and quenching experiments reveal that the (E,E)→(E,Z) route is a pure triplet process, whereas the (E,Z)→(E,E) route can have a minor singlet by-reaction. Hexyloxy or dodecyloxy chains on the terminal benzene rings convey the (E,E)-isomers of compounds 1a,b,d-f thermotropic liquid crystalline properties. In particular S_A, S_C, and S_F/I phases were studied with regard to their photochemical behavior. Depending on the system, photodegradation of the smectic phase to the isotropic melt (S→I) or photoinduction of the smectic phase (I→S) can be achieved by alteration of the (E,E)/(E,Z) ratio. These processes represent the basis for imaging techniques (writing and erasing of information).

Full text of DOI:10.1016/j.tet.2008.08.075

Tetrahedron 64 (2008) 10754–10760
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
Photochemistry of 3,6-bis(styryl)pyridazines in solution and in neat liquid
crystalline phasedoptical switching and imaging techniques
*
Herbert Meier , Thorsten Lifka, Peter Seus, Annette Oehlhof, Sabine Hillmann
Institute of Organic Chemistry, University of Mainz, Duesbergweg 10-14, 55099 Mainz, Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 27 May 2008
Received in revised form 20 August 2008
Accepted 25 August 2008
Available online 30 August 2008
3,6-Bis(styryl)pyridazines 1a–f with 2–6 alkoxy groups show on irradiation in solution a stereo-
isomerization which leads to a photostationary state of (E,E)- and (E,Z)-isomer. Sensitizing and quenching
experiments reveal that the (E,E)/(E,Z) route is a pure triplet process, whereas the (E,Z)/(E,E) route can
have a minor singlet by-reaction.
Hexyloxy or dodecyloxy chains on the terminal benzene rings convey the (E,E)-isomers of compounds
1a,b,d–f thermotropic liquid crystalline properties. In particular SA, SC, and SF/I phases were studied with
regard to their photochemical behavior. Depending on the system, photodegradation of the smectic
phase to the isotropic melt (S/I) or photoinduction of the smectic phase (I/S) can be achieved by
alteration of the (E,E)/(E,Z) ratio. These processes represent the basis for imaging techniques (writing and
erasing of information).
Keywords:
E/Z Isomerization
Liquid crystals
Phase behavior
Triplet quencher
Triplet sensitizer
Ó 2008 Published by Elsevier Ltd.
1. Introduction
246 nm and the n/p* transition at lmax¼313 nm (measurement in
ethanol).3 (E)-3-Styrylpyridazine has a strong red-shifted pp* band
Stilbenoid systems represent a highly interesting class of com-
pounds in materials science and photochemistry. Their chromo-
phores/fluorophores/electrophores have tunable properties for
many technical applications.1 Whereas, the number of publications
and patents on oligomers and polymers consisting of (E)-stilbene
building blocks is huge, there are comparably few studies on
analogous compounds, which contain azine rings (pyridine, pyr-
idazine, pyrimidine, pyrazine, triazine, tetrazine) instead of ben-
zene rings (Scheme 1).2,20
with lmax¼290 nm and a shoulder for the n/p* transition at
350 nm (measurement in n-heptane).4 In addition to the internal
conversion (IC) S2(n2_pp*)/S1(n_p2p*)/S0(n2_ 2) intersystem
p
crossing (ISC) to the triplet state has to be faced: S1(n_p2p*)/
T2(n2_pp*) and S2(n2_pp*)/T1(n_p2p*). According to El-Sayed’s
rule,5,6 first-order spin-orbit couplings make the ISC S1/T2 and
S2/T1 allowed, whereas the processes S1/T1 and S2/T2 are
forbidden. The triplet states have their own typical photophysics
and photochemistry. We expect efficient isomerizations E$Z on
irradiation of such systemsdeven when the chromophores have an
extended conjugation. In 1,4-distyrylbenzenes and higher
oligo(1,4-phenylenevinylene)s [OPVs] such stereoisomerizations
do not take place in the E/Z direction.7 The bond orders of the
olefinic double bonds are too high for a rotation in the excited
singlet state S1 and an ISC to the triplet state T1 does virtually not
occur.7,8
N
Scheme 1. Stilbenoid building blocks, which contain heteroaromatic six rings with 1–4
nitrogen atoms.
We prepared recently (E,E)-3,6-bis(styryl)pyridazines 1, which
represent ideal mesogens for photosensitive liquid crystals.2a The
incorporated nitrogen atoms bear free electron pairs, which enable
Reversible (E,E)$(E,Z) isomerizations should certainly be the
prevailing photoreactions of the (E,E)-3,6-bis(styryl)pyridazines 1
in solution. Moreover, the present study has the aim to find out, if
such optical switching processes are principally working in the LC
phases of 1 and how their efficiency depends on the molecular
order in these LC phases, which represent constrained media for
the involved electronically excited particles.
n/p* transitions in addition to the usual
electron excitation of stilbenoid compounds. Unsubstituted pyr-
idazine shows the allowed
p* transition at a lmax value of
p/
p* transitions in the
p/
A certain amount of (E,Z) configurations of 1 should lead to the
breakdown of the liquid crystal, which then should be restored,
when the reverse photoreaction (E,Z)-1/(E,E)-1 provokes, that the
* Corresponding author. Fax: þ49 6131 3925396.
E-mail address: hmeier@mail.uni-mainz.de (H. Meier).
0040-4020/$ – see front matter Ó 2008 Published by Elsevier Ltd.
doi:10.1016/j.tet.2008.08.075

H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
10755
portion of ‘wrong’ configuration falls below the critical percentage.9
Some (E,Z) configurations can certainly be tolerated by a not very
highly ordered liquid crystaldin particular at the domain borders.
2. Results and discussion
2.1. UV–vis absorption and formation of triplet states
Scheme 2 shows the di-, tetra-, and hexaalkoxy substituted
(E,E)-3,6-distyryl-pyridazines 1a–f, which we are discussing here.
Their preparation and their mesophases were described earlier.2a
R1
R1
R2
R2
R3
N
N
1
R3
R1
H
H
OC6H13
H
H
R2
OC6H13
OC6H13 OC6H13
OC6H13
OC12H25
OC12H25
R3
H
1
a
b
c
d
e
OC6H13
H
OC12H25
OC12H25
f
OC12H25
OC12H25
Scheme 2. Alkoxy substituted 3,6-bis(styryl)pyridazines 1a–f.
Figure 1. Selected orbitals (CNDO/CI calculation) of 10a (cisoid conformation with
olefinic double bonds, which are syn-oriented to the N atoms). The numbers refer to
the atomic pz coefficients: þC, ꢁB, and for HOMO-6 to the s, px, and py coefficients.
Compounds 1a–f have the character of donor–acceptor–donor
systems9 and exhibit a strong long-wavelength absorption between
280 and 420 nm. [3max¼(5ꢀ0.3)104]. Table 1 summarizes their lmax
values. Number and length of the side chains have a small influence
on the absorption. The absorption bands do not have a vibrational
structure at room temperature and they do not reveal the position
of the hidden n_/p* transition. A strong indication for the par-
ticipation of the latter transition is the fact, that none of the com-
(Fig. 1) has a somewhat higher energy and consists mainly (70%) of
the transition HOMO-6/LUMOþ2. The calculated lvalues listed in
Table 1 are average values for the possible conformers, which differ
among each other by almost ꢀ8 nm. The MMX/CNDO/CI calcula-
tions of the other compounds 10 gave quite similar results: the p*
p/
pounds shows
a p
fluorescence S1(pp*)/S0(
2), whereas the
(HOMO/LUMO) transitions have somewhat lower energy than the
n_/p* transitions. The number of OH groups has a very small in-
fluence on the electron transition energy.12 Additional vibrational
energy should in any case enable internal conversions (ICs) be-
tween pp* and np* singlet states and subsequent intersystem
crossings (ISCs) to T1 or T2 (Fig. 2).13
corresponding 1,4-bis(styryl)benzenes have high fluorescence
quantum yields (fFz0.8–0.9).10,11
In order to get an idea of the forbidden n_/p* transitions, we
calculated the electron excitations of compounds 10a–f, in which
the OR groups of 1a–f are replaced by OH. The olefinic double bonds
of 1/10 can have a transoid and two different cisoid arrangements.
Rotation of the terminal phenyl groups leads for 1b,e/10b,e to 10
planar rotamers, whereas 1a,c,d,f/10a,c,d,f have 3 planar con-
formers, each. Force field calculations MMX furnished the geome-
tries of these conformers, which then served as structural basis for
CNDO/CI calculations of the electron excitations. Figure 1 shows
four selected orbitals of 10a, which proved to be important for the
electron transitions. The configuration interaction reveals that the
2.2. Photochemistry in solution
An efficient intersystem crossing (ISC) to the triplet state is
a precondition for a reversible E/Z photoisomerization. Scheme 3
shows the process (E/E)-1$(E/Z)-1, which leads to a photosta-
tionary state. The (Z,Z) configuration is not included. (Detection
limit below 2% in the 1H NMR spectroscopy.) The reaction spectra of
Figure 3 illustrate the obtention of the photostationary state of 1a
allowed p/
p* transition is to 90% a HOMO/LUMO transition and
represents the energy-lowest transition. The n_/p* transition
Table 1
S2 (n-π*)
S1(π π*)
ISC
T2 (n-π*)
T
1 (π π*)
IC
IC
Long-wavelength absorption maxima of the (E,E)-3,6-bis(styryl)pyridazines 1a–c
(measured in cyclohexane) and 1d–f (measured in dichloromethane) and calculated
lmax values of the corresponding hydroxy compounds 10a–f (gas phase)
ISC
F
A
lC
A
Compound 1
lmax [nm]
p
/
p*
Compound 10
lmax [nm]
p/
p*
n/p*
a
b
c
d
e
f
365
376
372
369
380
376
a
b
c
d
e
f
344
347
344
344
347
344
324
323
323
324
323
323
S0
Figure 2. Term scheme of the (E,E)-3,6-bis(styryl)pyridazines 1/10. Absorption A,
fluorescence F, internal conversion IC, intersystem crossing ISC.

10756
H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
R1
R3
N
N
R1
R2
R2
R3
)-1
(E,E
hν
R2
R3
R1
N
R1
R3
N
R2
(E,Z)-1
hν
R2
R3
Figure 4. Reaction spectra (top to bottom) of the irradiation of a 2ꢂ10ꢁ4 M solution of
R1
(E,E)-1a in oxygen-containing benzene.
N
N
summarizes the possible cleavage products 2–7 derived from 1c;
FD MS spectroscopy proved the generation of 4-hexyloxy
benzaldehyde (2), 4-hexyloxybenzoic acid (4), 6-[2-(4-hexyloxy
phenyl)vinyl]pyridazine-3-carbaldehy-de (3) and its carboxylic acid
5. Pyridazine-3,6-dicarbaldehyde (6) in and its 3,6-dicarboxylic
acid (7) were not found, but the major reaction product is under
this conditions a polymer, which may have 6 and 7 incorporated.
Our first assumption, that the cleavage of 1 can be due to the
formation of singlet-oxygen with 1 as sensitizer according to Eq. 1,
could not be verified.
R1
R2
(Z,Z)-1
R3
H
13C6O
CHO
H13C6O
H13C6O
CHO
2
3
N
N
N
N
COOH
CHO
H13C6O
CHO
3
þ O2
S ð1Þ hn S ð1Þ
T
2;1
ð1Þ
S ð1Þ þ O
.
!
(1)
ISC
1
COOH
!
ƒ!
!
0
1;2
0
2
5
4
Thermal production of 1O2 from 9,10-diphenylanthracene-endo-
HOOC
COOH
peroxide did neither give a 1,2-dioxetane of 1a nor its cleavage
products. Therefore, a radical process has to be postulated for the
oxidative photodegradation of 1.
N
N
N
N
7
6
Scheme 3. Photoisomerization of (E,E)-1/(E,Z)-1 and possible cleavage products of 1c
The other pyridazines 1b–f behave quite similar. So we carefully
excluded O2 in the photoisomerization experiments and de-
termined the photostationary states of 1d–f by irradiation of
2ꢂ10ꢁ4 M solutions in benzene with a medium pressure mercury
on irradiation in the presence of oxygen.
lamp equipped with a Pyrex filter (
l
ꢃ290 nm). Table 2 shows the
results, which were obtained by 1H NMR measurements.
Interestingly the number of side chains has a considerable in-
fluence on the (E,E)/(E,Z) ratio. The 50:50 ratio of 1d (two dode-
cyloxy chains) is changed for 1e (four dodecyloxy chains) in favor of
the (E,E) configuration and then for 1f (six dodecyloxy chains) in
the opposite direction in favor of the (E,Z) configuration. Since the
absorption of 1d–f is almost identical, the different isomerization
tendencies may be due to a different participation of conformers.
The (E,Z)-isomers can be easily identified by their 1H NMR
spectra. In contrast to the (E,E)-isomers, the pyridazine ring rep-
resents in the (E,Z)-isomers an AB system instead of a singlet and
the 3J coupling constant of the olefinic protons on the (Z)-config-
ured double bond is 12.3ꢀ0.2 Hz instead of 15.1ꢀ0.2 Hz for the (E)-
configured double bond.
To obtain full 1H and 13C NMR characterization, we separated
(E,E)-1d/-(E,Z)-1d by fractional crystallization. The component
(E,E)-1d has a much lower solubility in benzene and can be filtered
Figure 3. Reaction spectra of the irradiation of a 2ꢂ10ꢁ4 M solution of (E,E)-1a in
oxygen-free benzene. The upper curve (a) corresponds to the start with pure (E,E)
configuration, the lower curve (b) corresponds to the photostationary state (E,E)/
(E,Z)¼54:46.
Table 2
Photostationary states of the pyridazines 1a,d–f obtained in 2ꢂ10ꢁ4 M solutions in
on irradiation with 366 nm in benzene in the absence of oxygen.
The reaction spectra of Figure 4 demonstrate the complete de-
composition of 1a on irradiation in the presence of oxygen. Oxygen
provokes a cleavage of the olefinic double bonds. Scheme 3
benzene (lꢃ290 nm)
Compound
1a
1d
1e
1f
(E/E)/(E/Z)
54:46
50:50
75:25
40:60

H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
10757
(a) Increasing photoisomerization (E,E)-1/(E,Z)-1 will destruct
the LC phase, provided that the photostationary state is lying in
the range of the isotropic melt. The temperature for the tran-
sition LC/I (isotropic phase) can be significantly lowered in
comparison to the pure mesophase.
off. The remaining (E,Z)-1d is a colorless powder, which does not
form a mesophase. A chromatographic separation on SiO2 is not
advisable because the catalytic (E,Z)/(E,E) isomerization enriches
then the (E,E)-isomers.
Although we favored from the very beginning the triplet
mechanism of the isomerization, a competitive by-road on the
energy hypersurface of the excited singlet state cannot a priori be
completely ruled out. Eq. 2 for the photostationary (E,E)/(E,Z) ratio
takes this into account.
(b) Irradiation of an isotropic phase I consisting of a mixture of
(E,E)-1 and (E,Z)-1 will lead to the opposite ‘border crossing’,
when the photostationary state is located in the LC region.
(c) The (E,Z)/(E,E) isomerization can principally work thermally
or catalytically and thus transfer the isotropic phase to an LC
layer, provided that the thermal (E,E)/(E,Z) equilibrium is in the
LC range. Combination of the processes (a) and (c) represent
another possibility of switching (or writing and erasing).
(d) The final borderline crossing process would consist of an in-
crease of the (E,Z) configuration by thermal activation, till the
LC phase breaks down. However, an LC phase at a temperature,
where the thermal equilibrium is in the range of the isotropic
melt, is principally unstable. Therefore, (d) is registered by
a dashed line in Scheme 5.
ðE; EÞ
ðE; ZÞ
3EZfsEZ
3EEfsEE
EE þ fsISCðE; ZÞ$fTEZ
EE
!
!
¼
(2)
þ
fsISCðE; EÞ$fTEE
EZ
!
EZ
!
In order to find out if the singlet quantum yield fsEE/EZ for the
isomerization (E,E)-1f/(E,Z)-1f can be neglected, we added azu-
lene as a triplet quencher. Increasing amounts of azulene decreased
dramatically the rate of (E,Z)-1f generation. A 7.2ꢂ10ꢁ3 M solution
of (E,E)-1f in benzene, doped with a 7.2ꢂ10ꢁ2 M solution of azulene
brought the isomerization to a complete stop. Therefore, fEsE/EZ
can be neglected. When (E,Z)/(E,E) mixtures were applied for this
experiment, the original isomer ratio is not preserved. A slow en-
richment of the (E,E)-isomer can be observeddeven for high azu-
lene concentrations. The most likely explanation is that fsEZ/EE is
small but not completely negligible. An alternative explanation
would involve different efficiencies in the quenching of the triplet
states of (E,E)-1f and (E,Z)-1f.
% (E,E)
% (E,Z)
LC
I
h
a)
Ph St
Th Eq
A final experiment in this context was performed with pyrene as
triplet sensitizer. A NiSO4 filter solution was used for a selective
excitation of pyrene. It filters out the radiation in the absorption
range of 1a. A 2ꢂ10ꢁ4 M solution of (E,E)-1a in benzene, which
contained 6ꢂ10ꢁ4 mol Lꢁ1 pyrene was irradiated and yielded the
(E,E)-1a/(E,Z)-1a ratio of 50:50. This ratio corresponds fairly to the
ratio 54:46 found in the absence of pyrene. Sensitized and direct
photolyses have virtually the same result, because the isomeriza-
tion is almost completely a triplet reaction. Scheme 4 summarizes
the direct, sensitized, and quenched photoprocesses, which are
valid for compounds 1.
h
b) Ph St
T
c)
d)
Th Eq
T
Scheme 5. Possible photochemical (cases (a) and (b)) and thermal (cases (c) and (d))
crossings of the borderline between LC phase and isotropic phase I depending on the
position of the photostationary state (Ph St) and the thermal equilibrium (Th Eq) on
the scale of (E,E) and (E,Z) isomers.
The dialkoxy substituted compounds 1a and 1d form thermo-
tropic smectic (S) or nematic (N) phase. According to the second
heating curve in the DSC, 1a exists in the narrow temperature range
between 161 and 170 ꢄC in an SF/I, between 170 and 283 ꢄC in an SC,
and above 283 ꢄC in an N phase.2a A similar behavior with some-
what lower transition temperatures was observed for 1d, which
contains two dodecyloxy groups (K–139 ꢄC–SF/I–168 ꢄC–SC–
260 ꢄC–I).2a All these mesophases do not allow an (E,E)/(E,Z)
photoisomerization. Irradiation into the broad absorption band,
which reaches from 250 to about 500 nm, exhibits photostability.
Neither an exposure to daylight for many days nor an irradiation
S0 (E,E)-1
S0 (E,Z) -1
ISC
h
ISC
h
T1 (pyrene)
Sens.
S0 (pyrene)
T1 (pyrene)
Sens.
S0 (pyrene)
S1,2 (E,Z) -1
S1,2 (E,E)-
1
S (pyrene)
S (pyrene)
1
1
h
h
ISC
ISC
T2,1 (E,Z
) -1
T2,1 (E,E)-1
Isom.
S0 (azulene)
Quench.
S0 (azulene)
Quench.
ISC
ISC
with
l
ꢃ320 nm changes the materials. However, an irradiation
T1 (azulene)
T1 (azulene)
(lꢃ320 nm) of an isotropic 50:50 mixture of (E,Z)-1d and (E,E)-1d
at 150 ꢄC leads very fast to the smectic phase SF/I with a content of
(E,Z)-1d below 3%. Obviously, the photostationary state is far on the
(E,E) side. A short heat shock to temperatures above 260 ꢄC trans-
forms the mesophase to the isotropic melt of pure (E,E)-1d. Com-
pound 1a shows an analogous photochemical behavior.
S0 (E,E) -1
n), sensitized (Sens.), and quenched (Quench.) photoreactions of
S0 (E,Z)-1
Scheme 4. Direct (h
3,6-bis(styryl)pyridazines 1, which effect isomerization (Isom.) and/or retention of the
two involved configurations (E,E) and (E,Z).
Compound 1b has a sharp melting point at 128 ꢄC (onset
temperature in the DSC) and does not form liquid crystals. The
same substitution pattern present in 1e, but with OC12H25 instead
of OC6H13 chains, leads on cooling to the formation of a smectic
phase: I–124 ꢄC–SA–106 ꢄC–K.2a In contrast to the highly ordered
smectic phases of 1a and 1d, compound 1e forms an SA phase,
which has a much lower order and consequently can be de-
graded photochemically. The self-organization in the SA phase
allows the (E,E)-1e/(E,Z)-1e isomerization. Figure 5 shows the
transition of the SA phase at 122 ꢄC to the isotropic phase on
2.3. Photochemistry of the LC phases
The crucial question is now: how can these photoisomerization
processes be transferred from solution experiments to meso-
phases? The (E,Z) configuration does certainly not fit into any
calamitic or phasmidic LC phase. A certain percentage of ‘wrong’
configuration may be tolerated, but soon a critical ratio (E,Z)-1/
(E,E)-1 should be reached, where the self-organization of the
mesophase becomes impossible. This borderline between LC and I
(Fig. 4) can be used in different mode for imaging and optical
switching processes:
irradiation (
lꢃ320 nm). Less than 3% of (E,Z)-1e provoke this
transformation.

10758
H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
Figure 6. Imaging technique by irradiation (
(E,E)-1f/(E,Z)-1f (black area) through a mask. Polarization microscopy shows the bi-
l
ꢃ320 nm, T¼40 ꢄC) of a isotropic mixture
refringent areas, where the SA phase was photochemically induced.
contains 20% (E,Z)-1e and 80% (E,E)-1e cannot be transformed to
the SA phase – neither by irradiation nor by annealing. The amount
of ‘wrong’ configuration is probably then too high for the ousting in
border areas of LC domains and the photostationary ratio (E,E)-1e/
(E,Z)-1e is obviously not sufficient for the photochemical genera-
tion of a mesophase.
Both drawbacks are not valid for the hexa-substituted, phas-
midic compounds 1c and 1f. They form SA phases at room tem-
perature2a and irradiation of isotropic (E,E)/(E,Z) mixtures with
a high content of (E,Z) configuration (60%) furnishes LC phases. 1H
NMR measurements revealed that the photochemical trans-
formation of a 40:60-mixture of (E,E)-1f/(E,Z)-1f at 30 ꢄC to
a 80:20-mixture is sufficient for the generation of a tight LC phase.
Figure 6 demonstrates the use of this process for information
storage. A film negative of the word ‘cher’ served as mask for the
irradiation (
l
ꢃ320 nm) at 40 ꢄC. The letters have a thickness of
about 50
mm. The resolution limit is in the range of 10 mm. A short
heat shock to 80 ꢄC provokes the erasing of the picture by forma-
tion of the isotropic phase. Pure (E,E)-1f has the clearing point of
the SA phase at 84 ꢄC.2a
Figure 5. Degradation of the SA phase of 1e at 122 ꢄC in the irradiated right part.
(polarization microscopy, scale 1:90).
The photochemical induction of the SA phase is possible, be-
cause the photostationary state of 1f is in the SA phase. That means
on the other side, that a photodegradation of the SA phase by the
isomerization (E,E)-1f/(E,Z)-1f is not possible. Long-term irradi-
ations of the isotropic phase reveal some fatigue because of the
formation of oligomers. The SA phase is photostabledeven at
temperatures relatively close to the clearing point, where the vis-
cosity is lowered.
Interestingly a short annealing of the probe restores the SA
phase. Since a thermal isomerization (E,Z)-1e/(E,E)-1e is not
possible under these conditions, we assumed either a catalytic
isomerization or the separation of the (E,Z)-1e molecules on the
domain borders.
The (E,Z) stereoisomers of 1 can be transformed in solution to
the thermodynamic equilibrium by the action of acids. Therefore,
we prepared LC phases of mixtures of (E,E)-1e and the organic acids
814–18 or 9,19–21 which form themselves mesophases (Scheme 6).
3. Conclusion
The 3,6-bis(styryl)pyridazines 1a and 1d, which bear two alkoxy
side chains, represent case (b) of Scheme 5. The photostationary
state is in the LC range, so that the mesophase can be induced by
irradiation of an isotropic (E,E)/(E,Z) mixture, but the mesophase
cannot be photodegraded. The tetrasubstituted system 1e behaves
opposite.23 Corresponding to type (a) of Scheme 5, the self-orga-
nization of the SA phase can be photochemically destroyed, but not
induced, because the photostationary state lies in the range of the
isotropic melt. Most suitable for imaging techniques are the SA
phases of 1c and 1f, because they exist at room temperature. As for
1a and 1d, the generation of a mesophase by partial photo-
isomerization (E,Z)/(E,E) is possible, since the photostationary
state is located in the LC range of the (E,E)/(E,Z) ratio. After the
‘writing’ process an ‘erasing’ is possible by a short heat shock.
Thermal or catalytic isomerization (E,E)-1$(E,Z)-1 [cases (c) and
(d) of Scheme 5] could be excluded under the conditions used here.
COOH
H13C6O
COOH
H25C12
O
8
9
K
I
153 °C SC 105 °C
I
162°C
N
156 °C SC 125 °C
K
Scheme 6. Smectic (SC) and nematic (N) phases of 4-hexyloxybenzoic acid (8) and
4-dodecyloxycinnamic acid (9).
Up to 5% 8 and 10% 9 were mixed with (E,E)-1e in the molten
state. Although 8 and 9 generate in pure form SC phases, the mix-
tures enter a self-organization of the SA type.22 The single differ-
ence between these mixed SA phases and the SA phase of neat 1e is
a minor lowering of the phase transition temperatures (D
T<4 ꢄC).
Irradiation of the mixed SA phases gave the same results as for the
pure phase of 1e. Photodegradation of the SA phases occurred,
when a small amount (ꢅ3%) of (E,Z)-1e was formed. The restoration
of the SA phase by annealing was not accelerated by the present
acids. Therefore, we assume that the major effect of the annealing is
that the ‘wrong’ (E,Z) configurations are ousted to the domain
borders. The turnover SA/I/SA can be repeated several times, be-
cause the (E,Z) configurations are ‘repaired’ in the next irradiation
period. A handicap of this writing-erasing technique is due to the
fact that temperatures of about 120 ꢄC are needed for the use of the
SA phase. Moreover, we found that the repair mechanism is not
perfect, so that the system fatigues. An isotropic mixture, which
4. Experimental
4.1. General
Melting points were determined on a Bu¨chi 510 melting point
apparatus and are uncorrected. The DSC measurements were per-
formed on a Perkin–Elmer DSC 7. IR spectra were measured with
a Beckman Acculab spectrometer. A Zeiss MCS 224/234 diode array
spectrometer was used for the UV–vis spectroscopy. The Bruker

H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
10759
ꢆ
spectrometer AC 200 and AM 400 served for the measurement of
the 1H and 13C NMR spectra. CDCl3 was used as solvent and TMS as
internal standard. The mass spectra were obtained on a Finnigan
MAT 95. Polarization microscopy was performed with a Leitz
Ortholux II microscope and a Mettler FP-52 heating system.
[Mþ ]. Anal. Calcd for C44H64N2O2 (653.0): C 80.93, H 9.89, N 4.29;
found: C 81.13, H 10.00, N 4.12.
4.3.2. Photooxidative degradation of (E,E)-3,6-bis[2-(4-
hexyloxyphenyl)vinyl]pyridazine (1a)
A solution of 1a (17.4 mg, 0.036 mmol) in 185 mL dry benzene
4.2. Synthesis of the 3,6-bis(styryl)pyridazines 1a–f and the
carboxylic acids 8 and 9
was irradiated at 30 ꢄC with a Hanovia 450 W medium pressure
lamp, equipped with a Pyrex filter (
l
ꢃ290 nm). During the irradi-
ation of 3.5 h a stream of air was purged through the solution.
UV–vis spectroscopy indicated the consumption of the starting
compound. The solvent was evaporated and the mixture subjected
to a FD MS analysis, which revealed the presence of 1a: m/z 484
The preparation of (E,E)-1a–f,2a 8,14–18 and 919–21 was described
earlier.
ꢆ
ꢆ
4.3. Irradiations in solution
[Mþ ], 4-hexyloxybenzaldehyde (2): m/z 206 [Mþ ], 6-[2-(4-hexy
ꢆ
loxyphenyl)vinyl]pyridazine-3-carbaldehyde (3): m/z 310 [Mþ ],
ꢆ
For analytical purposes, irradiations were performed in quartz
cuvettes with an Osram XBO-1000 W OFR high-pressure Xenon
lamp in an AMKO device. The monochromator was adjusted to 320
or 366 nm. That corresponds to the two maxima in the continues
emission of the lamp between 300 and 800 nm. A water filter
served for the absorption of the IR radiation, so that the solutions in
oxygen-free benzene were kept at room temperature. The UV–vis
spectra were measured in a direction perpendicular to the irradi-
ation beam.
4-hexyloxybenzoic acid (4): m/z 222 [Mþ ], and 6-[2-(4-hexyloxy
ꢆ
phenyl)vinyl]pyridazinecarboxylic acid (5): m/z 326 [Mþ ]; 2 as
a daughter ion of 4 and 3 as a daughter ion of 5 could be excluded.
The peak intensities do not represent the real amounts of the ions
were: 1a/2/3/4/5¼6:100:21:21:4. The 1H NMR spectrum of the
reaction mixture revealed that 2 was the major product (10%) of
low molecular weight. About 80% are oligomeric products.
The reaction of 1awith 9,10-diphenylanthracene-9,10-endo-peroxide25
did neither give a 1,2-dioxetane of 1a nor the cleavage products 2–5. Apart
from unreacted 1a, only oligomers of undefined structure were generated.
For preparative purposes a Hanovia 450-W medium pressure
lamp, equipped with a Pyrex filter (
l
ꢃ290 nm), was used. Solutions
of 1 (180 mL, 2.0ꢂ10ꢁ4 mol Lꢁ1 in oxygen-free benzene) were ir-
radiated in an immersion vessel under vigorous stirring in an Ar
atmosphere. A cooling system between lamp and solution was
regulated between 25 and 70 ꢄC, so that clear solutions without
precipitation were maintained.
4.4. Irradiation of the LC phases
Isotropic melts (25 mg) of the LC forming compounds were used
to generate films between two Duran glass plates (
l
ꢃ310 nm). The
irradiations were performed with an UVGL-58 lamp, whose emis-
sion lies between 320 and 400 nm with a maximum at 352 nm. A
Digsitherm DT-3434 heating system kept the temperature on the
desired level (accuracy ꢀ1 ꢄC). The following conditions were used:
The photostationary states of 1a,d–f (Table 2) were determined
by 1H NMR measurements with repeated and averaged integration.
The quenching experiments were performed with a 7.2ꢂ10ꢁ3 M
solution of (E,E)-1f in benzene to which increasing amounts of
azulene were added. When the tenfold concentration (7.2ꢂ10ꢁ2 M)
of azulene was reached, the isomerization (E,E)/(E,Z) came to
a complete stop.
(a) (E,E)-1a: 290 ꢄC (N), 200 ꢄC (SC), 165 ꢄC (SF/I).
(b) (E,E)-1d: 250 ꢄC (SC), 200 ꢄC (SC), 145 ꢄC (SF/I).
For the triplet-sensitized photoreaction, a 2.0ꢂ10ꢁ4 M solution
of (E,E)-1a in benzene was used, which contained 6.0ꢂ10ꢁ4 mol Lꢁ1
pyrene. The cooling system was filled with an aqueous solution of
NiSO4$6H2O (300 g Lꢁ1). A thickness of about 1 cm of the NiSO4
solution ensured that the light between 320 and 480 nm was fil-
tered off, so that the incident light between 290 and 320 nm was
selectively absorbed by pyrene.
During the irradiation up to 150 min, all phases were kept intact
and the 1H NMR spectra of the irradiated films (dissolved in CDCl3
or C6D6) did not contain any signals of (E,Z) configurations.
(c) (E,E)-1d/(E,Z)-1d (50:50): 150 ꢄC (I).
After 1 min irradiation, the SF/I phase was formed, and the 1H
NMR spectrum revealed that the content of (E,Z)-1d was below 3%.
A short heat shock to temperatures above 260 ꢄC led to the isotropic
melt. (E,E)-1a exhibits the same behavior.
4.3.1. (E,Z)-3,6-Bis[2-(4-dodecyloxyphenyl)-vinyl]pyridazine
(E,Z)-1d
The preparative irradiation described above yielded a 1:1-mix-
ture (23.5 mg) of (E,E)-1d and (E,Z)-1d. UV–vis spectroscopy in-
dicated after about 4 h the photostationary state. The isomer ratio
was determined by 1H NMR spectroscopy. Repeated fractional
crystallization from benzene permitted the separation of the two
stereoisomers, because the (E,Z)-isomer is much better soluble.
Pure (E,E)-1d is a yellow powder, which melts at 139 ꢄC to the
smectic phase.2a Pure (E,Z)-1d is a pale yellow powder, which starts
(d) (E,E)-1e: 122 ꢄC (SA), 140 ꢄC (I).
The smectic phase disappeared quickly, before 3% (E,Z) config-
uration was formed. Long-term irradiations (10–90 min) of SA or
irradiations of the isotropic melt (I) led to an appreciable amount of
oligomers up to 20%. Irradiation of a crystalline film for 90 min at
110 ꢄC revealed a complete photostability.
to decompose above 120 ꢄC. 1H NMR (400 MHz, CDCl3):
d 0.86 (t,
6H, CH3), 1.24 (m, 32H, CH2), 1.45 (m, 4H, CH2), 1.85 (m, 4H, CH2),
3.92 (t, 2H, OCH2), 3.99 (t, 2H, OCH2), 6.78/7.21 (AA0BB0, 4H, aromat.
H, (Z) side), 6.82/6.92 (AB, 3J¼12.4 Hz, 2H, olefin. H, (Z) configura-
tion), 6.89/7.50 (AA0BB0, 4H, aromat. H, (E) side), 7.19/7.51 (AB,
3J¼16.2 Hz, 2H, olefin. H, (E) side), 7.24/7.31 (AB, 3J¼8.9 Hz, 2H,
(e) (E,E)-1c and (E,E)-1f: 25 ꢄC (SA).
The SA phases are photostable. Long-term irradiations (60–
120 min) led to the formation of 10% dimer and oligomers.
heteroarom. H).24 13C NMR (100 MHz, CDCl3):
d 14.1 (CH3), 22.7,
26.0, 29.3, 29.4, 29.6, 29.6, 29.7, 29.7, 31.9 (CH2, partly super-
imposed), 68.1, 68.2 (OCH2), 114.5, 114.9, 122.4, 123.0, 126.3, 126.6,
128.7, 130.2, 134.3, 135.0 (aromat. and olefin. CH), 128.5, 128.5,
156.4, 157.3, 159.1, 160.0 (aromat. Cq). FD MS: m/z (%) 652 (100)
(f) (E,E)-1f/(E,Z)-1f (40:60): 20 ꢄC, 30 ꢄC, 40 ꢄC.
Irradiation led immediately to an enrichment of the (E,E)-iso-
mer. After a few minutes the SA phase was formed. 1H NMR

10760
H. Meier et al. / Tetrahedron 64 (2008) 10754–10760
3. Dorohoi, D.; Partenie, H.; Chiran, L.; Anton, C. J. Chim. Phys. Phys. Chim. Biol.
1994, 91, 419–432.
4. Perkampus, H. H.; Bluhm, T.; Knop, J. V. Spectrochim. Acta, A 1972, 28, 2163–
2177.
5. El-Sayed, M. A. J. Chem. Phys. 1962, 36, 573–574.
6. El-Sayed, M. A. J. Chem. Phys. 1964, 41, 2462–2467.
7. Meier, H. Angew. Chem. 1992, 104, 1425–1446; Angew. Chem., Int. Ed. Engl. 1992,
31, 1399–1420.
measurement revealed that a ratio of 80% (E,E)-1f and 20% (E,Z)-1f
is already sufficient for a high-quality texture. Long-term irradia-
tion showed some fatigue due to dimerization (6%). A short heat
shock to 80 ꢄC erased the texture, but did not change the (E,E)/(E,Z)
ratio.
8. Sandros, K.; Sundahl, M.; Wennerstro¨m, O.; Norinder, U. J. Am. Chem. Soc. 1990,
112, 3082–3086.
Acknowledgements
9. Meier, H. Angew. Chem. 2005, 117, 2536–2561; Angew. Chem., Int. Ed. 2005, 44,
2482–2506.
10. Meier, H.; Lifka, T.; Stalmach, U.; Oehlhof, A.; Prehl, S. Eur. J. Org. Chem. 2008,
1568–1574.
11. Oelkrug, D.; Tompert, A.; Egelhaaf, H.-J.; Hanack, M.; Steinhuber, E.; Hohloch, M.;
Meier, H.; Stalmach, U. Synth. Met. 1996, 83, 231–237.
12. The measured as well as the calculated electron excitation energies are lowest
for the tetrasubstituted compounds 1b,e/10b,e0.
We are grateful to the Deutsche Forschungsgemeinschaft, the
Fonds der Chemischen Industrie and the Centre of Materials Sci-
ence of the University of Mainz.
References and notes
13. Moreover, the energetical sequence of close lying singlet states S1/S2 or triplet
states T1/T2 can depend on the solvent (medium).
1. Selected monographs and reviews: (a) Grimsdale, A. C. In Organic Light-Emitting
Devices; Mu¨llen, K., Scherf, U., Eds.; Wiley-VCH: Weinheim, 2006; pp 215–243;
(b) Meier, H. In Carbon-Rich Compounds; Haley, M. M., Tykwinski, R. R., Eds.;
Wiley-VCH: Weinheim, 2006; (c) Hadziioannou, G.; van Hutten, P. F. Semi-
conducting Polymers; Wiley-VCH: Weinheim, 2000; (d) Segura, J. L.; Martin, N.
J. Mater. Chem. 2000, 10, 2403–2435; (e) Martin, R. E.; Diederich, F. Angew. Chem.
1999, 111, 1440–1469; Angew. Chem., Int. Ed. 1999, 38, 1350–1377; (f) Scherf, U.
Top. Curr. Chem. 1999, 201, 163–222; (g) Mu¨ llen, K.; Wegner, G. Electronic Ma-
terials: The Oligomer Approach; Wiley-VCH: Weinheim, 1998; (h) Kraft, A.;
Grimsdale, A. C.; Holmes, A. B. Angew. Chem. 1998, 110, 416–443; Angew. Chem.,
Int. Ed. 1998, 37, 403–428; (i) Tour, J. M. Chem. Rev. 1996, 96, 537–553.
2. See, for example: (a) Lifka, T.; Zerban, G.; Seus, P.; Oehlhof, A.; Meier, H. Tet-
rahedron 2008, 64, 6551–6560; (b) Stadler, A.-M.; Puntoriero, F.; Campagna, S.;
Kyritsakas, N.; Welter, R.; Lehn, J.-M. Chem.dEur. J. 2005, 11, 3997–4009; (c) Van
der Eycken, E.; Jidong, Z.; Kilonda, A.; Compernolle, F.; Toppet, S.; Moornaert, G.;
Van der Auweraer, M.; Jackers, C.; Verbouwe, W.; De Schryver, F. C. J. Chem. Soc.,
Perkin Trans. 2 2002, 928–937; (d) Eynde, J. J. V.; Pascal, L.; Van Haverbeke, Y.;
Dubois, P. Synth. Commun. 2001, 31, 3167–3174; (e) Brooker, S.; Kelley, R. J. Chem.
Soc., Dalton Trans. 1996, 2117–2122; (f) Nohara, M.; Hasegawa, M.; Hosokawa, C.;
Tokailin, H.; Kusumoto, T. Chem. Lett. 1990, 2, 189–190.
ˆ
14. Malthete, J.; Canceill, J.; Gabard, J.; Jacques, J. Tetrahedron 1981, 37, 2815–2821.
15. Coco, S.; Espinet, P.; Marcos, E. J. Mater. Chem. 2000, 10, 1297–1302.
16. Kim, B. G.; Kim, S.; Park, S. Y. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 2001, 370,
391–394.
17. Prajapati, A. K.; Sharma, H. C.; Chudgar, N. K. Mol. Cryst. Liq. Cryst. 2001, 364,
815–824.
18. Prajapati, A. K.; Pandya, H. M. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 2003,
393, 31–40.
19. Tuffin, R. P.; Goodby, J. W.; Bennemann, D.; Heppke, G.; Loetzsch, D.; Scher-
owsky, G. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A 1995, 260, 51–68.
20. Lifka, T.; Meier, H. J. Prakt. Chem./Chem.-Ztg. 1995, 337, 641–646 and references
therein.
21. Prajapati, A. K.; Thakkar, U.; Bonde, N. Mol. Cryst. Liq. Cryst. Sci. Technol., Sect. A
2003, 393, 41–48.
22. Higher contents of 8 and 9 in these mixtures lead to phase separation.
23. The other tetrasubstituted compound 1b does not form a mesophase.
24. Assignemnt by INDOR measurements.
25. Wassermann, H. H.; Scheffer, J. R.; Cooper, J. L. J. Am. Chem. Soc. 1972, 94, 4991–
4996.