Synthesis and linear/nonlinear optical properties of a new class of 'RHS' NLO chromophore

Article abstract of DOI:10.1039/b315274j

Examples of a new class of zwitterionic, "right-hand side" merocyanines containing a cyanodicyanomethylidenedihydrofuran electron acceptor have been prepared. As well as allowing for the facile synthesis of these chromophores, our synthetic methodology enables considerable variation in both the donor moiety as well as the extent of conjugation between the donor and acceptor systems. As expected, all of these "right-hand side" systems are negatively solvatochromic, with the difference between λ _max (polar vs. nonpolar solvents) increasing with the extent of conjugation. In accord with expectations, hyper-Raleigh scattering (HRS) measurements confirm that molecules with the greatest conjugation pathway (for e.g. 21c, 23c) have the largest first hyperpolarisabilites, β_o. In addition, our HRS evaluation indicates that the 4-quinolinylidene donor nucleus is superior to both the 4-pyridinylidene and benzothiazolylidene systems. The figures of merit, μ_(calc). β_o(measured), that we obtain for some of these compounds, are of a similar magnitude to the best "left hand side" examples reported in the literature. In order to demonstrate the versatility of our synthetic technique, representative polymer-tetherable derivatives of these compounds have been prepared, as have the corresponding TDI-based polyurethanes.

Full text of DOI:10.1039/b315274j

View Article Online / Journal Homepage / Table of Contents for this issue
A R T I C L E
Synthesis and linear/nonlinear optical properties of a new class
of ‘RHS’ NLO chromophore{
Andrew J. Kay,a Anthony D. Woolhouse,*a Yuxia Zhaob and Koen Clays*b
aOpto-Organics Group, Industrial Research Limited, P.O.Box 31 310, Lower Hutt,
New Zealand
bLaboratory for Chemical and Biological Dynamics, Department of Chemistry, University
of Leuven, Celestinenlaan 200D, B3001 Leuven, Belgium
Received 26th November 2003, Accepted 16th February 2004
First published as an Advance Article on the web 15th March 2004
Examples of a new class of zwitterionic, ‘‘right-hand side’’ merocyanines containing a
cyanodicyanomethylidenedihydrofuran electron acceptor have been prepared. As well as allowing for the facile
synthesis of these chromophores, our synthetic methodology enables considerable variation in both the donor
moiety as well as the extent of conjugation between the donor and acceptor systems. As expected, all of these
‘‘right-hand side’’ systems are negatively solvatochromic, with the difference between lmax (polar vs. nonpolar
solvents) increasing with the extent of conjugation. In accord with expectations, hyper-Raleigh scattering (HRS)
measurements confirm that molecules with the greatest conjugation pathway (for e.g. 21c, 23c) have the largest first
hyperpolarisabilites, bo. In addition, our HRS evaluation indicates that the 4-quinolinylidene donor nucleus is
superior to both the 4-pyridinylidene and benzothiazolylidene systems. The figures of merit, m(calc).bo(measured), that
we obtain for some of these compounds, are of a similar magnitude to the best ‘‘left hand side’’ examples reported
in the literature. In order to demonstrate the versatility of our synthetic technique, representative polymer-tetherable
derivatives of these compounds have been prepared, as have the corresponding TDI-based polyurethanes.
In spite of the fact that the (larger) ground state dipole
Introduction
moments of ‘right-hand-side’ chromophores can potentially
detract from any consideration for use of such compounds, we
have maintained an interest in this class of chromophore. We
believe that it is, in principle, still possible to achieve acceptable
molecular figures of merit (m.b) by trading some b for some m
and, at the same time, optimally lessening the structural
complexity of the molecules. Furthermore, it should also be
possible to use the site-isolation concept10 to minimise
undesirable dipole–dipole interactions in any polymer system
that contains such compounds.
Our previous work11,12 has focussed upon merocyanines
whose ground states are predominated by a zwitterionic dis-
tribution, as evinced by negative solvatochromic behaviours.
These stunted chromophores furnished only modest figures of
merit, those of the azasesquifulvalene cyanines being an order
of magnitude greater than those merocyanines containing
heterocyclic electron acceptors.
Herein we report details of the syntheses and of the linear
and nonlinear optical properties of a new class of zwitterionic
nonlinear chromophore—that containing an aromatisable
donor system linked through a simple polyenic interconnect
to the powerful cyanodicyanomethylidenedihydrofuran elec-
tron acceptor, representatives of which are presently com-
manding widespread interest.4,8
In recent years there has been intense interest in molecules
displaying large and efficient nonlinear optical responses (m.b)
and in the materials that permit the effective translation of
these molecular properties into large, efficient and usable
macroscopic responses (x(2)).1–6 This interest is due to their
potential application in emerging optoelectronic and photonic
technologies.7 In attempting to design all-plastic—that is, all
organic/polymeric—materials as active components of such
devices, a number of criteria need to be addressed and certain
design parameters traded. Structural features of a molecule
that give rise to a desirable, high molecular nonlinearity might
not necessarily be as desirable from the points of view of, for
example: synthetic expediency; transparency at communica-
tions wavelengths; compatibility either when doped or
functionalised within a polymer matrix; thermal and photo-
stability and maintaining the noncentrosymmetry of the poled
chromophore array (i.e. temporal stability of the EO effect).
The super-high b chromophore-containing polymers that have
made it to high performance prototype device status3,8 are those
designed by the groups of Dalton and Jen that are based upon
‘left-hand-side’ (Marder–Perry plot9) systems in which the
donor (D) and acceptor (A) moieties flank large, conjugated
p-manifolds and which contain auxiliary substituents that
minimise potentially deleterious aggregation effects (e.g. 1).
Results and discussion
Synthesis and characterisation
In our previous synthetic work that was designed to utilise the
easily accessible and functionalisable 1,4-dihydropyridinylidene
nucleus as the ‘aromatisable’ donor in right-hand-side NLO
molecular systems, we showed that access could be gained to
merocyanines of the types depicted in structures 2–4 (e.g. R1 ~
dihydroxypropyl).11,12
In the stunted series 3 (n ~ 0 with various heterocyclic
acceptors), the syntheses were achieved by nucleophilic
{ Electronic supplementary information (ESI) available: synthesis and
characterisation data. See http://www.rsc.org/suppdata/jm/b3/b315274j/
T h i s j o u r n a l i s ß T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 4
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 1

View Article Online
substitution of 4-phenylthio-N-alkylpyridinium halides 5 by
the heterocyclic nucleophile and, in the ethylidene-bridged
systems 2 and 3 (n ~ 1 with various heterocyclic acceptors), the
syntheses were achieved by way of the reaction of N-alkyl-4-
of which is the FTC chromophore 1 above. All of these LHS
systems are characterised by possessing an N,N-dialkyl-4-
anilino moiety as the donor nucleus.
Preliminary calculations of both m(g) and bo at the
MOPAC97/AM1 level (see Table 2, later) led us to believe
that such RHS (zwitterionic) systems should have appreciably
high figures of merit, most especially those in which polyenic
bridges separated the donor and the cyanodicyanomethyl-
idenedihydrofuran acceptor. The work we describe herein is
designed to ‘prove the concept’ by building only the basic
polyenic interconnect, for example, as in generic 7. This is a
prelude to designing molecules containing not only length but
also structural rigidity and opportunity to introduce sub-
stituents (onto the interconnect, for example) that might
contribute to the minimisation of chromophore dipole–dipole
interactions.
picolinium halides
6 with electrophilic methyleneamido-
functionalised heterocycles. With the exception of the
dimethoxymerocyanine 2 (R ~ OMe), only very modest
figures of merit were obtained for these compounds using both
calculated and measured b-values (HRS) and calculated values
of dipole moment (MOPAC97/AM1). These studies also
showed that heteroaromatisable systems were generally inferior
electron acceptors when compared to quinonoid systems.
In principle, two pathways exist for the construction of the
zwitterionic systems under consideration. These are, in essence,
the ‘build-up’ and ‘build-down’ approaches (Scheme 1) and are
based upon our earlier observations11 and those of others14,15
with respect to the use of the enamino/enamido functionality,
in particular, as an effective electrophile whether it is attached
to the donor or to the acceptor. Since we had set out to address
the use of three donor nuclei—the 4-pyridinylidene (4-PY), the
4-quinolinylidene (4-Q) and the benzothiazolidinylidene
(BT)—the more expedient approach is clearly the ‘build-up’
strategy as this requires only two sets of three reactants;
namely, the three azinium/azolium donors and the three
(oligoen)amido substituted dihydrofuran acceptors. The
‘build-down’ approach would require three of each of the
three oligoenamido substituted azinium/azolium donors to be
reacted with the 4-methyl-substituted dihydrofuran acceptor.
Each of the (oligoen)amido dihydrofuran acceptors was syn-
thesised in good yield by reacting commercially-available
dialdehyde bisanils 8 (a, N,N’-diphenylformamidine; b, malon-
dialdehyde bisanil hydrochloride; c, glutaconic dialdehyde
We became interested in exploring the use of the cyanodi-
cyanomethylidenedihydrofuran heterocycle 9 as an electron
acceptor in RHS chromophore systems principally because of
the interest in its use stimulated by Wang et al.13 and, more
recently, by He et al.4 In fact, this powerful electron acceptor is
contained within a number of very high-b molecules, the likes
Scheme 1 Possible routes to target merocyanines.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 2

View Article Online
bisanil hydrochloride) with equimolar 3-cyano-4,5,5-trimethyl-
2(5H)-furanylidenepropanedinitrile 9 in solution in acetic
anhydride (containing an equivalent of sodium acetate when
the bisanil hydrochloride salt was used). The compounds 11a–c
were isolated as highly crystalline, intensely coloured sub-
stances that precipitated from solution. Furthermore, with the
application of chemistry described by He et al.4 we were able
to introduce an additional, potential tethering functionality
(4-hydroxyphenyl-) at the C5 sp3 centre of the dihydrofuran
nucleus, viz. 10. In the preparation of the acceptor 12, the
phenolic hydroxyl in 10 was acetylated and stayed as such in
condensation with the benzothiazole 17. In like manner, a more
configuration-locked enamido acceptor 13 was synthesised from
the commercially-available bisanil of 2-chlorocyclohexene-1,3-
dialdehyde.
The 4-picolinium, 4-lepidinium and benzothiazolium salts
14–17, bearing one or other of N-alkyl, N-hydroxyalkyl or
N-dihydroxyalkyl substituents, were prepared as previously
described.11 Whilst the ortho equivalents of the picoline and
quinoline donors—namely, the 2-picolinium and quinaldinium
salts 18–20—were initially outside the intended scope of this
work, we have addressed the chemistry in order to demonstrate
the applicability of the methodology to gain access to
representative examples, particularly in view of the fact that
calculated figures of merit for the 2-pyridinylidene (2-PY) and
2-quinolinylidene (2-Q) derived chromophores are not appre-
ciably different to those of their isomers (see Table 2, later).
Condensation reactions were performed most effectively by
reacting equimolar quantities of each of the donor and
acceptor components in refluxing acetic anhydride containing
an equivalent of triethylamine for 10 minutes. In virtually all
instances the product crystallised from hot solution and was
recovered by filtration upon cooling.
Attempts to cleave the acetoxy functionality on chromo-
phore 24d using standard mild alkaline hydrolysis conditions
proved unsuccessful and the material was recovered. These
conditions, which always enabled the easy recovery of the
acetylated aliphatic side-chain hydroxyls on the N-linked
tethering functionalities (e.g. see Experimental, syntheses of
22a,b), were insufficient to enable the recovery of the phenolic
hydroxyl. We will be exploring the prospect of retaining
TBDMS protection until the final chromophore nucleus is
constructed (through two sequential reactions in ‘alkaline’
acetic anhydride) and, thereafter, regenerating the hydroxyl
group. Subsequent attempts to purify many of these com-
pounds were marred by compound insolubility, most particu-
larly with all members of the benzothiazolylidene series and
with the more extended systems from all donors.
NMR spectral data have revealed the existence of some
interesting, but not unexpected, structural features in the
pyridinylidene and quinolinylidene series. In all instances, there
is evidence of the presence of two isomers present in essentially
equal proportions in solution in DMSO, and this is amply
demonstrated in the 1H NMR spectrum of compound 22a
(Fig. 1). With respect to Fig. 1, asterisks have been placed upon
resonances assigned to the olefinic interconnect protons of one
isomer only. The connectivities between the two sets of three
multiplets, each of which comprises two doublets due to
terminal olefinic protons and a doublet of doublets due to the
central olefinic proton, are clearly evident.
Furthermore, the same is true in the suite of compounds
derived from the use of the 2-picolinium salts 18 and 19, but
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 3

View Article Online
Fig. 1 1H NMR spectrum of 22a showing the presence of two rotomers. The asterisks indicate the olefinic (viz. the p-interconnect) signals due to
one rotomer, namely a doublet at 5.76 ppm, coupled to a doublet of doublets at 7.61 ppm coupled to a doublet at 6.36 ppm.
more difficultly discernable from the line-broadening in the
spectra of compounds derived from the corresponding
2-quinaldinium salts 20. With respect to the 4-pyridinylidene
series (21a–c, 22a,b) in which the interconnect ranges from two
to four double bonds, the pyridylidene ring protons appear as
two two-proton doublets, a fact which would suggest that the
(rotational) isomerism is not at the donor end of the polyene as
might have been expected from a mechanistic point of view.
Nucleophilic displacement of the anilido functionality would
be expected to occur from both faces of the terminal sp2 centre,
the effect of which would be to generate cisoid and transoid
diene moieties at the donor end of the chain. The appearance of
only a single set of resonances (d 3.2–4.4) attributable to the
dihydroxypropyl tethering functionality on the donor nitrogen
atom clearly indicates a magnetic equivalence of this group in
both isomers of 22a.
1 3 2 4
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0

View Article Online
In addition, in the spectrum of compound 22a which itself
will serve to illustrate the phenomenon, no resolution or
We note that this apparently general phenomenon in the
PY and Q series is not the same as that reported by Zhang
et al.16 in their ‘build-down’ approach to the synthesis of
the ring-locked CLD-1 chromophore 28. This compound is
isolated as an inseparable mixture of trans (85%) and cis
(15%) isomers about a non-terminal double bond after having
been synthesised by way of a conventional Knoevenagel
condensation from an enal precursor in which this ratio was
68 : 32.
In the benzothiazolylidene system, there is no evidence of a
rotational isomerism that manifests itself in the manner we have
1
described. In the H NMR spectrum of 24b, which is the least
marred by line-broadening of the olefinic proton resonances, the
appearance of a single doublet at d 5.84, in particular, clearly
indicates that there is only the one contributor.
1
simplification is achieved by recording the H NMR spectrum
at 90 uC. Coupling constants (J 12–15 Hz) are consistent with
an all-trans interconnect in each compound. Together, these
data suggest that the rotational isomerism is about the bond
that links the electron-acceptor ring to the p-interconnect, that
is, at the end opposite to that expected mechanistically. By way
of confirmation, the NOESY pulse sequence clearly shows that
in one isomer a terminal olefinic proton, Ha, (doublet) is the
nearest neighbour to the geminal methyls of the dihydrofuran
ring, i.e. 22a1, and in the other, that a non-terminal olefinic
proton, Hb, (doublet of doublets) is the nearest neighbour to
the same ring methyls, i.e. 22a2.
In view of spectral complexity and because it is outside the
scope of this paper, we make no effort to address another
structural feature associated with these chromophores, namely,
that which pertains to the direction in which the polyenic
interconnect evolves from the donor nucleus. This would
involve the assignment of syn/anti status to denote the
relationship to the peri-proton in the Q series or to the ring-
nitrogen atom in the BT series, for example.
Within the suite of the chromophore precursors 11a–c, we
observe a similar isomerisation in solution that occurs most
notably with the shortest member 11a and, to a much lesser
extent, with the next homologue 11b. With the express purpose
of perhaps gaining an understanding of the reasons for the
formation of rotamers in the PY and Q chromophores, we
studied the fate of the ‘acceptor’ 11a in solution in DMSO.
When the NMR spectrum is recorded immediately upon
preparation, only a single set of resonances is observed, this
corresponding to those expected of a trans stereochemistry
(J 14.4 Hz). These data, however, preclude any assignment of
the cisoid/transoid nature of the diene ‘moiety’ but an
examination of the 2D-NOESY correlation confirms that the
olefinic proton nearer in space to the gem-methyls appears at d
8.7 and that the other (next nearer) proton is at d 5.1. If it is
accepted that the deshielded proton (at d 8.7) is that nearer the
furan acceptor, then this would imply that the cisoid rotamer
configuration, 11a2, is that which prevails initially in solution
and thus in the solid state.
We therefore infer from these data that the interconnects are
all-trans in members of the PY and Q series, whether it be with
the 1,4- or the 1,2-system. The reason for the consistent
observation of an approximately 1 : 1 mixture of rotamers
about the p-interconnector/acceptor bond is unclear. This is
particularly so in view of the fact that the DDHf value derived
from each of the MOPAC97-calculated DHf values for
22a1 (49.2 kcal mol21) and 22a2 (51.2 kcal mol21) is only
2.0 kcal mol21, a finding that leads to an expectation of signal
coalescence in the NMR spectrum. Non-bonded interactions
between the central olefinic proton of the interconnect and
either the gem-methyls or the vinylic nitrile or between the
terminal olefinic proton and each of the same functionalities
are not expected to favour one population over the other.
Clearly there is an insignificant electronic driving force that
might have been expected to favour the all-trans stereochem-
istry from the donor ring to the exocyclic dicyanomethylidene
residue, viz. 22a1. An HPLC-MS analysis of a DMSO solution
of 25a, for example, gave rise to only a single peak [m/z MH1
317.4; (2M)H1 633.5 amu] in spite of the fact that the same
solution was shown by NMR to contain both rotamer
populations.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 5

View Article Online
In solution at ambient temperature in either DMSO or
chloroform, this compound slowly degrades, clearly by way of
an isomer in which the olefinic protons (at d 9.71 and 5.25) are
related by 9.0 Hz coupling and which sees the concomitant
appearance of an exchangeable one-proton signal. This process
is under investigation but it is presumed to be irrelevant to the
formation of the essentially 1 : 1 mixtures of final chromo-
phores, i.e. it is triggered by hydrolysis. After a week in solution
at ambient temperature, the signals in the NMR spectrum are
most clearly those of acetanilide.
As testimony to the fact that the ‘build-down’ approach also
works, chromophore 30 (compound 21/22; R ~ CH2CH2OAc,
R’ ~ H, n ~ 1) was synthesised through the brief reaction of
4-[2-anilinovinyl]-1-(2-hydroxyethyl)pyridinium iodide salt 29
with the cyanodicyanomethylidenedihydrofuran acceptor 9 in
the presence of triethylamine in acetic anhydride. Here again,
the same pair of rotameric isomers about the p-interconnect/
acceptor bond were isolated. As with other compounds
containing the N-hydroxyethyl and N-dihydroxypropyl tether-
ing substituents that are acetylated during the condensation,
the acetates can be cleaved by treatment with dilute aqueous
alkali without affecting the integrity of the chromophore nucleus.
UV-vis spectral characterisation
The electronic absorption spectra of the chromophores from all
PY, Q and BT series show intense charge-transfer absorption
maxima in the 500–850 nm range and reveal the expected red
shifts as the extent of conjugation is increased (Table 1).
Interestingly, for almost all of the 2- and 4-PY and Q systems,
which exist as mixtures of rotamers about the p-system, the
charge transfer bands are essentially symmetrical in all solvents
and show no evidence of an ‘aggregate band’ on the blue side of
the absorption. The BT chromophores, on the other hand, all
exhibit blue-shifted shoulders to their charge transfer bands.
Significant windows of transparency exist in the 350–450 nm
region of spectra for all compounds. Furthermore, all
compounds are negatively solvatochromic,17,18
a feature
which qualitatively distinguishes these zwitterionic RHS
systems from those ‘belonging’ to the LHS. The effect of
benzannelation on lmax of PY systems is quite pronounced,
there being red shifts of up to 150 nm in going from PY to Q,
for example. The BT series shows not only the effects of
benzannelation but also the effects of introducing a component
of the p-interconnect which confers a degree of rigidity or
configuration locking to the molecule. The compound contain-
ing the chlorocyclohexenylidene linkage 24e is red-shifted by
approximately 40 nm when compared to its iso-p-electronic
series member 24c.
Hyper-Raleigh scattering studies
The first hyperpolarisabilities, b, of a representative suite of
these zwitterionic chromophores have been measured using
the hyper-Raleigh scattering (HRS) technique19–24 with a
femtosecond-pulsed fundamental of 800 nm and second
harmonic at 400 nm, at which wavelength no two-photon-
induced fluorescence was detected by temporal resolution of
emitted light signals. The dynamic b values are therefore true
experimental values without an over-estimate due to fluo-
rescence. The static bo values have been deduced based upon
the two-level model.25–27 This model has been shown to give
consistent static bo values from dynamic b values acquired at
different measurement wavelengths most particularly when
chromophores such as these possess single, well-resolved
charge-transfer bands.28 Furthermore, the resonance enhance-
ment due to the fact that the charge-transfer band is in between
the fundamental and the second harmonic indicates that the
signs for the dynamic b at the measurement wavelength and for
For the sake of a series completion, by employing the
chemistry that had led successfully to the series of stunted
zwitterionic chromophores,12 we were unable to construct
compounds representative of the n ~ 0 case in structure 7
(R ~ –CH3, –CH2CH2OH) employing the 4-phenylthio-
pyridinium iodide 5 and the dihydrofuran acceptor in a
number of solvent/catalyst systems. The ‘active’ methylene
derived from 9 is clearly a poor nucleophile when compared
to the heterocycles that furnished the stunted systems such
as 3.
Table 1 Electronic absorption data for selected chromophores
Compound (series)
lmax (DMF)/nm
log10e (DMF)
lmax (methanol)/nm
lmax (pyridine)/nm
Dlmax (lMeOH 2 lpyr)
21a (4-PY)
21b (4-PY)
21c (4-PY)
22a (4-PY)
22a (4-PY)
25a (2-PY)
25b (2-PY)
25c (2-PY)
26a (2-PY)
26b (2-PY)
23a (4-Q)
23b (4-Q)
23c (4-Q)
23d (4-Q)
27a (2-Q)
27b (2-Q)
27c (2-Q)
24a (BT)
570
600
615
572
604
556
588
598
556
590
660
735
735
730
620
710
746
605
705
810
854
4.86
4.78
4.76
4.79
4.64
4.90
4.81
4.29
4.72
4.61
5.00
4.88
4.70
4.72
5.18
5.12
4.85
5.18
5.21
4.91
4.91
564
592
595
570
598
550
582
580
552
584
654
724
705
710
610
700
700
598
698
805
840
600
670
685
598
662
580
662
670
578
652
682
782
860
865
634
740
850
614
718
830
865
236
278
290
228
264
230
280
290
226
268
228
258
2155
2115
224
240
2150
216
220
225
225
24b (BT)
24c (BT)
24e (BT)
1 3 2 6
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0

View Article Online
the static bo are opposite, this being based upon assumptions
applied to the two-state model.25–27
x(2). So as to minimise the extent to which these interactions
might prevent optimal chromophore alignment in target
systems such as 21c and 23c, our future work will focus
upon additional design features in chromophores of similar
length. These will include configurational-locking and bulking
of the p-interconnect with pendant functionality (to enhance
solvent-solubility as well) and also include the introduction of
an additional site for tethering in the acceptor moiety.
The HRS-derived bo values are presented in Table 2 together
with calculated values of the ground-state dipole moment m and
bo and the corresponding figures of merit (m.bo). In spite of the
fact that 800 nm radiation triggered some (not unexpected)
photodegradation of various of these chromophores, the
results clearly follow the expected trend which shows an
increasing b as a function of increasing conjugation length.
Interestingly, the effects of benzannelation upon the PY
systems—that is, in going from 4-PY 21a–c to 4-Q 23a–c—
have a marked effect upon the size of bo. This is perhaps an
indication of the increasing donor strength of Q systems over
PY systems. Based upon the acquisition of limited data points
for the BT chromophores it would appear that the BT nucleus
is midway in strength between the PY and Q donors, as some of
our earlier findings had shown.11 We note also in passing that,
with the exception of the values presented for 21a, the
calculated bo values are underestimates of the static values
calculated from experimental b.
Polymerisation
In spite of the fact that many of these zwitterionic chro-
mophores are high melting compounds that are sparingly
soluble in most organic solvents, those possessing the N-(2,3-
dihydroxypropyl) tether (i.e. 22a,b and 26a,b) were readily
amenable to stoichiometric condensation polymerisation with
2,4-toluenediisocyanate in DMSO.11,12 The polymers were
easily recovered and purified using DMSO/methanol cycles.
This work will be reported in due course.
The figures of merit (mcalc.bo(HRS)) we obtain for the longest
of each of the 4-PY and 4-Q chromophores, 21c and 23c
respectively (Table 2) are of the same order of magnitude as
that reported for the ‘benchmark’ FTC chromophore,
members of which (LHS) class are characterised generally as
having exceptional optical nonlinearities.2,8,29 The value of
Experimental
Commercially available reagents were obtained from Aldrich
and Avocado and were used without additional purification.
1H and 13C NMR spectra were recorded on a Bruker
AVANCE 300 MHz spectrometer and proton multiplicities
are defined by the usual notations. The assignments of
resonances were made employing DEPT, COSY, HSQC and
NOESY pulse sequences.
m
calc.bo(HRS) reported2 for FTC 1 is 8255 6 10248 esu and
the values we obtain for the highest members of each of the
PY and Q suites—21c and 23c—are 8900 6 10248 and 9384 6
10248 esu, respectively. Clearly, the concept of being able to
realise large molecular figures of merit by trading ‘some b for
some m’ is an exploitable one, and most especially with these
relatively easily accessible RHS chromophores that contain
(tetherable) azine/azinium donor moieties and the powerful
cyanodicyanomethylidenedihydrofuran acceptor system. The
parameter m.bo/MW is included in Table 2 as a measure of the
potential for scaling the molecular to macroscopic optical
nonlinearity. This is done in the knowledge of the fact that
the scaling of m.bo to m.bo/MW is inaccurate as a result of
chromophore–chromophore interactions.30 However, in using
this parameter, it is evident that the large PY and Q
chromophores—21c and 23c—still offer up very large figures
of merit. Because the chromophores are highly polar,
zwitterionic systems, there is a greater potential for incurring
unfavourable intermolecular dipole–dipole interactions within
a matrix, the likes of which will strongly disfavour the
realisation of large and usable macroscopic nonlinearities
UV-vis absorption spectra were recorded on either a
Hewlett-Packard 8452A diode array spectrophotometer or a
Cary 50 B10 UV-vis spectrophotometer. Accurate mass mea-
surements were made on a PE Biosystems Mariner mass
spectrometer operating in the electrospray mode. Micro-
analyses were performed by the Campbell Microanalytical
Laboratory, University of Otago, Dunedin, New Zealand and
melting points are uncorrected. Because many of the com-
pounds retained solvent (of crystallisation) to varying degrees
even after prolonged periods under high vacuum, a greater
reliance was placed upon the acquisition of accurate mass data.
The apparatus and experimental procedures for measuring
femtosecond hyper-Raleigh scattering (HRS) were exactly as
described previously.20–24 b values were determined by using
b
an external reference. All measurements were performed in
DMSO and optical local field correction factors were applied.
800 for crystal violet chloride (338 6 10230 esu in methanol) as
Table 2 Calculated and experimental values of m and bo, and the corresponding figures of merit
ma (calc)/
m
(calc)?bo(calc)
/
m
(calc)?bo(found)
/
m(calc)?
bo(found)/MW
a
bo (calc)/
b
bo (found)/
Compound
(Series)
10218 esu
10230 esu
10230 esu
10248 esu
10248 esu
21a (4-PY)
21b (4-PY)
21c (4-PY)
25a (2-PY)
25b (2-PY)
25c (2-PY)
23a (4-Q)
23b (4-Q)
23c (4-Q)
27a (2-Q)
27b (2-Q)
27c (2-Q)
24a (BT)
24b (BT)
24c (BT)
17.7
17.9
17.8
15.6
15.9
15.9
15.5
12.7
13.8
14.7
14.7
14.6
10.4
12.9
12.8
13.0
148
268
343
125
240
319
153
229
253
142
247
309
100
173
213
125
360
500
—
—
—
440
560
680
—
—
—
2620
4749
6105
1950
3816
5072
2372
2908
3491
2087
3631
4511
1040
2232
2726
2213
6444
8900
—
—
—
6820
7112
9384
—
—
—
7.0
18.8
24.1
17.2
16.8
20.9
240
260
—
2496
3354
—
6.2
7.8
1-FTC2
635
8255
a MOPAC (CS Chem3D Pro, CambridegeSoft) AM1 level using the precise keyword. b bo is the static first hyperpolarisability estimated by
using the two-state model.25–27 This is derived from b, the dynamic first hyperpolarisability, which was measured using a femtosecond-pulsed
Ti-sapphire fundamental at 800 nm.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 7

View Article Online
1
{3-Cyano-4,5-dimethyl-5-(4-hydroxyphenyl)-2(5H)-furanylidene}-
propanedinitrile, 10
C20H16N4O2 requires MH1 m/z 345.13480; D ~ 0.6 ppm). H
NMR (d6-DMSO) d 8.72 (d, J 14.4 Hz, 1H), 7.63 (m, 3H), 7.47
(m, 2H), 5.10 (d, J 14.4 Hz, 1H), 2.05 (s, 3H), 1.66 (s, 6H).
Irrespective of concentration, the 13C NMR spectrum was
unexpectedly, but consistently, poor. Only the methyl and
methine signals were identifiable. 13C NMR (d6-DMSO) d 145.1
(CH), 130.9 (CH), 130.5 (CH), 128.6 (CH), 98.7 (CH), 26.1
(CH3), 23.5 (CH3). lmax (DMF) 430 log e 4.74.
The procedure described by He et al.,4 in which lithioethyl vinyl
ether (tBuLi/ethyl vinyl ether/THF/278 uC) is reacted (210 uC)
with phenones was employed using the TBDMS derivative of
4-hydroxyacetophenone,31 (5.0 g, 20.0 mmole). The crude
intermediate a-ketol was then treated in solution in THF with
3 equivalents of tetrabutylammonium fluoride at ambient
temperature for 2 hours after which time the solution was
quenched with an ethyl acetate/water mixture. The organic
phase was concentrated and subjected to flash chromatography
over silica, eluting with 20% ethyl acetate/hexane, to give
3-hydroxy-3-(4-hydroxyphenyl)butan-2-one (2.3 g, 64%) as
colourless needles, m.p. 101–102 uC (Found: C; 66.47, H; 6.79.
C10H12O3 requires C; 66.65, H; 6.71%). This deprotected
ketol (5.0 g, 27.8 mmole) was then reacted with a mixture
comprising malononitrile (9.1 g, 138.0 mmole), acetic acid
(0.93 g, 15.5 mmole), and ammonium acetate (0.37 g,
4.8 mmole) in pyridine (45 ml) at ambient temperature for
16 hours. After this time, the red mixture was quenched into an
ice/water slush and the resulting pink solid recovered by
filtration. The purification of this solid was best accomplished
by flash chromatography over silica eluting with 20–30%
acetone/hexane, followed by recrystallisation from ethyl
acetate/hexane after which the furanylidene-propanedinitrile
10 was obtained (4.1 g, 51%) as a colourless crystalline solid,
m.p. 225–227 uC. (Found: C; 69.35, H; 3.89, N; 15.35.
C16H11N3O2 requires C; 69.30, H; 4.00, N; 15.16%). 1H
NMR (CDCl3) d 9.37 (s, 1H,), 7.03 (d, J 8.4 Hz, 2H), 6.90 (d, J
8.4 Hz, 2H), 2.21 (s, 3H), 1.98 (s, 3H).
{4-(4-Acetanilido-trans-1,3-butadienyl)-3-cyano-5,5-dimethyl-
2(5H)-furanylidene} propanedinitrile, 11b was purified by
recrystallisation from acetone and isolated as a brick-red
crystalline solid (86%), m.p. 272–274 uC (dec). (Found: C;
71.22, H; 4.58, N; 15.01. C22H18N4O2 requires C; 71.35, H;
4.86, N; 15.14%) Found: MH1 m/z 371.15025; C22H18N4O2
requires MH1 m/z 371.15141; D ~ 3.1 ppm). H NMR (d6-
1
DMSO) d 8.54 (d, J 13.3 Hz, 1H), 7.91 (dd, J 15.3, 11.2 Hz,
1H), 7.57 (m, 3H, aromatic), 7.40 (m, 2H, aromatic), 6.40 (d, J
15.4 Hz, 1H), 5.41 (dd, J 13.3, 11.2 Hz, 1H), 1.99 (s, 3H), 1.70
(s, 6H). 13C NMR (d6-DMSO) 177.5 (CQ), 176.3 (CQ), 169.7
(CQ), 151.4 (CH), 145.3 (CH), 138.3 (CH), 130.6 (CH), 129.8
(CH), 128.8 (CH), 115.2 (CH), 113.4 (CQ), 113.0 (CH), 112.6
(CQ), 111.4 (CQ), 98.8 (CQ), 95.2 (CQ), 52.6 (CQ), 25.9 (CH3),
23.5 (CH3). lmax (DMF) 526 log e 5.00.
{4-(6-Acetanilido-trans,trans-1,3,5-hexatrienyl)-3-cyano-5,5-
dimethyl-2(5H)-furanylidene}propanedinitrile, 11c was purified
by recrystallisation from acetic anhydride as a purple crystal-
line solid (53%), m.p. 259–260 uC. (Found: MH1 m/z
397.16590; C24H20N4O2 requires MH1 m/z 397.16535;
D ~ 1.4 ppm). 1H NMR (d6-DMSO) d 8.04 (d, J 13.8 Hz,
1H), 7.7–7.5 (m, 4H), 7.60 (m, 1H), 7.37 (d, J 6.9 Hz, 2H),
7.30 (dd, J 14.4, 11.7 Hz, 1H), 6.45 (dd, J 14.1, 11.4 Hz,
1H), 6.42 (d, J 15.3 Hz, 1H), 5.18 (dd, J 13.8, 11.1 Hz,
1H), 1.92 (s, 3H), 1.69 (s, 6H). 13C NMR (d6-DMSO)
177.4 (CQ), 175.4(CQ), 169.2(CQ), 150.3(CH), 148.2(CH),
139.8(CH), 138.7(CQ), 130.6 (CH), 129.6 (CH), 128.9 (CH),
116.1 (CH), 113.4(CQ), 112.9, 112.6(CQ), 111.6(CQ), 98.8(CQ),
96.2(CQ), 53.5(CQ), 25.8 (CH3), 23.4 (CH3). lmax (DMF) 628
log e 5.02.
4-[2-Anilinovinyl]-1-(2-hydroxyethyl)pyridinium iodide, 29
A mixture of N-(2-hydroxyethyl)pyridinium iodide (24 g,
65.2 mmole) and N,N’-diphenylformamidine (18 g, 91.8 mmole)
was stirred at 120 uC for 1 h. On cooling the tarry mixture was
triturated with 2 6 30 ml of ether and then allowed to stand,
whereupon a dark brown solid formed. This solid was
recovered by filtration and recrystallised from methanol to
give the iodide salt 29 as olive-green microcrystals (17.9 g;
54%), m.p. 192–193 uC. (Found: C; 48.94, H; 4.69 N; 7.94.
{4-[(3-Acetanilidomethylene)-2-chloro-1-cyclohexen-1-yl]-E-
ethenyl-3-cyano-5,5-dimethyl-2(5H)-furanylidene}propanedini-
trile, 13 was recovered from the reaction mixture, washed with
acetic anhydride and then isopropanol and then dried to give
the title compound as a purple crystalline solid (71%), m.p.
225–227 uC. (Found: MH1 m/z 471.15925; C27H23ClN4O2
1
C15H17IN2O requires C; 48.93, H; 4.65, N; 7.61%). H NMR
(d6-DMSO) d 10.30 (s, 1H), 8.50 (d, J 13.0 Hz, 1H), 8.33 (d, J
7.1 Hz, 2H), 7.77 (d, J 6.8 Hz, 2H), 7.33 (m, 4H), 7.03 (m, 1H),
5.89 (d, J 13.0 Hz, 1H), 5.13 (t, J 5.10 Hz, 1H), 4.30 (t, J 4.9 Hz,
2H), 3.78 (m, 2H). 13C NMR (d6-DMSO) 155.4 (CQ), 142.9
(CH), 142.6 (CH), 140.9 (CQ), 129.9 (CH), 123.0 (CH), 119.0
(CH), 116.2 (CH), 99.1 (CH), 60.7 (CH2), 60.4 (CH2). lmax
(DMF) 428 log e 4.82.
requires MH1 m/z 471.15823; D ~ 2.1 ppm). H NMR (d6-
1
DMSO) d 8.42 (d, J 16.0 Hz, 1H), 7.91 (s, 1H), 7.6–7.4 (m, 5H),
6.68 (d, J 16.0 Hz, 1H), 2.53, (m, 2H), 1.90 (m, 2H), 1.83 (s,
6H), 1.60 (m, 2H). 13C NMR (d6-DMSO) 144.8 (CH), 134.8
(CH), 130.8 (CQ), 129.5 (CH), 128.9 (CH), 117.1 (CH), 99.8
(CQ), 27.5 (CH2), 26.3 (CH3), 21.9 (CH2). No other lines were
visible. lmax (DMF) 502 log e 4.64.
Synthesis of {(N-acetyl-N-phenyl-oligoenamino)-2(5H)-furanyl-
idene}propanedinitrile acceptors 11a–c, 12, 13: general condensa-
tion procedure
{4-[4-Acetanilido-trans-1,3-butadienyl]-5-(4-acetoxyphenyl)-
3-cyano-5-methyl-2(5H)-furanylidene}propanedinitrile, 12 was
synthesised using the 4-hydroxyphenyl-substituted furanyl-
idene propanedinitrile 10 and the bisanil 8b and recovered
by flash chromatography over silica (30% acetone/hexanes)
and isolated as maroon prisms (68%), m.p. 219–220 uC.
(Found: C; 71.17, H; 4.25, N; 11.66. C29H22N4O4 requires C;
71.02, H; 4.49, N; 11.43%. Found: MH1 m/z 491.17167
C29H22N4O4 requires MH1 m/z 491.17138; D ~ 0.6 ppm).
1H NMR (d6-acetone) d 8.15 (d, J 13.6 Hz, 1H), 7.6–7.2
(m, 9H), 7.48 (bd dd, 1H), 6.39 (d, J 15.3 Hz, 1H), 5.39 (dd,
J 13.6, 11.4 Hz, 1H), 2.27 (s, 6H), 1.92 (s, 3H). 13C NMR
(d6-acetone) 178.0 (CQ), 175.1 (CQ), 170.1 (CQ), 169.8(CQ),
153.5 (CQ), 151.9 (CH), 145.0 (CH), 139.7 (CQ), 135.0 (CQ),
131.7 (CH), 130.8 (CH), 129.8 (CH), 129.2 (CH),123.8
(CH), 116.5 (CH), 113.5 (CH), 112.9 (CQ), 111.9 (CQ),
100.0 (CQ), 24.5 (CH3), 23.7 (CH3), 21.4 (CH3). lmax (DMF)
530 log e 4.99.
A mixture of the bisanil monohydrochloride (5.0 mmol), 4,5,5-
trimethyl-3-cyano-2(5H)-furanylidenepropane dinitrile 932
(5.1 mmol) and anhydrous sodium acetate (5.1 mmol) in
acetic anhydride (ca. 20 ml) was refluxed for 5–10 min before
being allowed to cool and stand overnight. In the case of
N,N’-diphenylformamidine free base, no sodium acetate was
employed. Adducts were recovered by filtration as highly
crystalline, coloured solids, and were washed with acetic
anhydride (2 6 5 mL), followed by copious water and finally
isopropanol. After drying in vacuum they were suitable for use
without further purification.
{4-(2-Acetanilidoethenyl)-3-cyano-5,5-dimethyl-2(5H)-furanyl-
idene}propanedinitrile, 11a was purified by recrystallisation from
acetone and isolated as yellow plates (79%), m.p. 274–278 uC
(dec). (Found: C; 69.64, H; 4.57, N; 16.22. C20H16N4O2 requires
C; 69.77, H; 4.65, N; 16.28%); Found: MH1 m/z 345.13460;
1 3 2 8
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0

View Article Online
Synthesis of p-bridged pyridinylidene, quinolinylidene and
benzothiazolidinylidene {2(5H)-furanylidene}propanedinitrile
chromophores 21–27: general condensation procedures.
1H, higher field branch of AB quartet, -CH2OH). 13C NMR
(d6-DMSO, no quaternary resonances cited) 144.0 (py-CH),
143.5 (py-CH), 138.6 (CH), 138.5 (CH), 121.2 (py-CH), 120.4
(py-CH), 118.5 (CH), 117.6 (CH), 104.6 (CH), 104.0 (CH), 70.8
(CH), 63.3 (CH2), 62.1 (CH2), 27.6 (CH3), 27.4 (CH3). lmax
(DMF) 572 log e 4.79; (MeOH) 570; (pyridine) 598.
Method A. Equimolar quantities of the appropriate
N-methylazinium halide (14, 18, or 20), the oligoenamido
acceptor, 11a–c, and triethylamine were dissolved in acetic
anhydride (10 ml/mmole acceptor) and the solution refluxed for
5–10 min before being allowed to cool slowly. Crystalline
adducts were recovered by filtration and washed thoroughly
with fresh acetic anhydride followed by copious quantities of
water and then isopropanol and then dried. Yields were
consistently in excess of 60%.
(iii) 2-Hydroxyethyl)quinolin-4(1H)-ylidene donors—Method B
‘{4{2-[N-(2-Hydroxyethyl)quinolin-4(1H)-ylidene]ethenyl}-3-cyano-
5,5-dimethyl-2(5H)-furanylidene}’propanedinitrile,
23a
was
obtained as a green-brown powder (89%), m.p. w300 uC.
(Found: MH1 397.16615 C24H20N4O2 requires MH1 m/z
1
397.16590 D ~ 0.6 ppm). H NMR (d6-DMSO) two isomers,
Method B. As for Method A, but with methanol instead of
acetic anhydride. This method was used for reactions with the
azinium and azolium halides 16 and 17.
approximately 1 : 1 d 8.8–8.5 (m, 2.5H), 8.3–7.7 (m, 4H), 7.48
(d, J 6.3 Hz, 0.5H), 7.32 (d, J 13.3 Hz, 0.5H), 7.21 (d, J 14.0 Hz,
0.5H), 6.16 (d, J 12.7 Hz, 0.5 H), 5.94 (d, J 12.3 Hz, 0.5H), 5.10
(bd s, 1H), 4.77 (bd ‘s’, 2H), 3.85 (bd ‘s’, 2H), 1.74 (s, 3H), 1.50
(s, 3H). 13C NMR (d6-DMSO) 175.3 (CQ), 163.9 (CQ), 162.6
(CQ), 152.5 (CQ), 152.4 (CQ), 146.0 (CH), 145.1 (CH),
141.3 (CH), 138.5 (CQ), 134.1 (CH), 134.0 (CH), 128.0 (CH),
127.7 (CH), 126.2 (CH), 125.9 (CH), 125.6 (CQ), 119.0
(CH), 118.7 (CH), 117.9 (CQ), 117.1 (CQ), 116.3 (CQ), 114.5
(CH), 113.6 (CH), 112.1 (CH), 110.9 (CH), 106.8 (CH),
105.8 (CH), 93.6 (CQ), 59.3 (CH2), 57.7 (CH2), 57.4 (CH2), 27.7
(CH3), 27.2 (CH3). lmax (DMF) 660 log e 5.00; (MeOH) 654;
(pyridine) 682.
Method C. Equimolar quantities of the appropriate N-(2,3-
dihydroxypropyl)picolinium chloride 15 or 19,11 and the
oligoenamido acceptor were treated with catalytic triethyl-
amine in refluxing acetic anhydride as described above.
The cooled reaction mixtures were poured into ether (ca.
25 ml/mmole acceptor) and stirred vigorously for several
minutes. The liquors were decanted and the oily residues
washed by stirring with further portions of ether. The residual
insoluble oils were stirred vigorously with aqueous sodium
hydroxide solution (2% w/v, 25 ml/mmole) at 90 uC for 30 min
and the resulting solids recovered by filtration, washed with
water (to neutrality), isopropanol and then dried. Yields of
the chromophores recovered in this manner were again in
excess of 55%.
Representative example of the ‘build-down’ method
‘{4-{2-[N-(2-Acetoxyethyl)pyridin-4(1H)-ylidene]ethenyl}-3-
cyano-5,5-dimethyl-2(5H)furanylidene}’propanedinitrile, 30. A
mixture of the pyridinium iodide 29 (1.0.g, 2.6 mmol), acceptor
9 (0.52 g, 2.6 mmol) and triethylamine (1.0 mL) in 15 mL of
acetic anhydride was stirred at 100 uC for 10 min. After cooling
the solution was poured into ether (100 ml) and the mixture
stirred vigorously for 5 min, the liquor decanted and the
residual tar triturated with water, whereupon a dark solid
formed. This was collected by filtration and washed with water
and then isopropanol to afford a blue-purple solid (34%), m.p.
261–265 uC. (Found: MH1 m/z 389.16019; C22H20N4O3
requires MH1 m/z 389.16082 D ~ 1.6 ppm). 1H NMR (d6-
DMSO) d 8.54 (d, J 6.7 Hz, 1H), 8.47 (d, J 6.7 Hz, 1H), 8.5–8.4
(m, 0.5H), 7.87 (d, J 6.6 Hz, 1H), 7.6–7.7 (m, 0.5H), 7.53, (d, J
6.6 Hz, 1H), 6.34 (d, J 14.6 Hz, 0.5H), 6.28 (d, J 14.8 Hz, 0.5H),
5.76 (d, J 12.6 Hz, 0.5H), 5.68 (d, J 12.2 Hz, 0.5H), 4.57 (nm,
(i) N-Methylpyridin-4(1H)-ylidene donors—Method A
{4[2-(N-Methylpyridin-4(1H)-ylidene)ethenyl]-3-cyano-5,5-
dimethyl-2(5H)-furanylidene}propanedinitrile, 21a was obtained
as a grey/green microcrystalline solid (83%), m.p. w300 uC.
Found: MH1 m/z 317.13910 C19H16N4O requires MH1 m/z
1
317.13969; D ~ 1.8 ppm). H NMR (d6-DMSO) d 8.48 (d, J
6.9 Hz, 1H), 8.41 (d, J 7.0 Hz, 1H), 8.35 (dd, J 14.8, 12.7 Hz,
0.5H), 7.86 (d, J 7.0 Hz, 1H), 7.63 (dd, J 14.4,12.7 Hz, 0.5H),
7.53 (d J 7.0 Hz, 1H), 6.34 (d J 14.5 Hz, 0.5H), 6.28 (d, J
14.8 Hz, 0.5H), 5.73 (d, J 12.5 Hz, 0.5H), 5.65 (d, J 12.1 Hz,
0.5H), 4.07 (s, 3H), 1.66 (s, 3H), 1.43 (s, 3H). 13C NMR
(d6-DMSO) 160.5 (CQ), 159.4 (CQ), 153.3 CQ), 144.1 (CH),
143.6 (CH), 138.3 (CH), 138.2 (CH), 121.5 (CH), 120.7 (CH),
118.6 (CH), 117.6 (CH), 116.8 (CQ), 115.4 (CQ), 104.6
(CH), 103.9 (CH), 92.9 (CQ), 92.6 (CQ), 46.2 (CH3), 46.1
(CH3), 27.6 (CH3), 27.4 (CH3). lmax (DMF) 570 log e 4.86;
(MeOH) 564; (pyridine) 600.
2H), 4.46 (nm, 2H), 2.00 (s, 3H), 1.67 (s, 3H), 1.44 (s, 3H). 13
C
NMR (d6-DMSO) 169.4 (CQ), 151.1 (CQ), 147.0 (CQ), 142.1
(CQ), 128.8 (CH), 127.8 (CH), 126.8 (CH), 124.7 (CH), 123.7
(CH), 122.5 (CH), 116.8 (CQ), 116.0 (CH), 107.5 (CH), 106.5
(CH), 94.8 (CQ), 59.1 (CH2), 50.3 (CH2), 25.1 (CH3), 21.2
(CH3). lmax 572 (DMF) log e 4.80.
(ii) N-(2,3-Dihydroxypropyl)pyridin-4(1H)-ylidene donors—
Method C
Acknowledgements
‘{4{2-[N-(2,3-Dihydroxypropyl)pyridin-4(1H)-ylidene]ethenyl}-
3-cyano-5,5-dimethyl-2(5H)-furanylidene}’propanedinitrile, 22a
was obtained as a blue grey powder (59%), 281–283 uC. (Found:
MH1 m/z 377.16085 C21H20N4O3 requires MH1 m/z 377.16082
This work was funded in part by the New Zealand Foundation
for Research, Science and Technology, contract number
C08X0206. We wish to thank Dr Herbert Wong for his
valued assistance and patience in acquiring NMR data.
1
D ~ 0.1 ppm). H NMR (d6-DMSO), isomer 1, d 8.79 (d, J
6.6 Hz, 1H), 8.37 (dd J 14.7, 12.0 Hz, 0.5H), 7.53 (d, J 6.6 Hz,
1H), 6.29 (d, J 14.7 Hz, 0.5H), 5.67 (d, J 12.0 Hz, 0.5H), 1.44 (s,
3H); isomer 2, d 8.37 (d J 6.6 Hz, 1H), 7.86 (d J 6.6 Hz, 1H),
7.61 (dd, J 14.4, 12.9 Hz, 0.5H), 6.36 (d, J 14.4 Hz, 0.5H), 5.76
(d, J 12.3 Hz, 0.5H), 1.67 (s, 3H). Resonances attributable to
the dihydroxypropyl substituent were coincident for each
isomer at d 5.34 (dd, J 7.2,5.7 Hz, 1H, -CH2OH), 4.96 (t, J
5.4 Hz, 1H, -CHOH), 4.48 (dd, J 13.6, 3.0 Hz, 1H, lower field
branch of AB quartet, -NCH2), 4.20 (m, 1H, higher field
branch of AB quartet, -NCH2), 3.83 (bd m, 1H, -CHOH), 3.48
(m, 1H, lower field branch of AB quartet, -CH2OH), 3.33 (m,
References
1
L. R. Dalton, W. H. Steier, B. H. Robinson, C. Zhang, A. Ren,
S. Garner, A. Chen, T. Londergan, L. Irwin, B. Carlson, L. Fifield,
G. Phelan, C. Kincaid, J. Amend and A. Jen, J. Mater. Chem.,
1999, 9, 1905.
2
3
L. Dalton, A. Harper, A. Ren, F. Wang, G. Todorova, J. Chen,
C. Zhang and M. Lee, Ind. Eng. Chem. Res., 1999, 38, 8.
H. Ma, A. K.-Y. Jen and L. R. Dalton, Adv. Mater., 2002, 14, 1339
and references cited therein.
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0
1 3 2 9

View Article Online
4
5
6
M. He, T. M. Leslie and J. A. Sinicropi, Chem. Mater., 2002, 14,
2393–4662.
16 C. Zhang, L. R. Dalton, M-C. Oh, H. Zhang and W. H. Steier,
Chem. Mater, 2001, 13, 3043.
17 A. Botrel and A. L. Beuze, J. Chem. Soc., Faraday Trans., 1985,
80, 1235.
18 M. Niedbalska and I. Gruda, Can. J. Chem., 1990, 68, 691.
19 R. W. Terhune, P. D. Maker and C. M. Savage, Phys. Rev. Lett.,
1965, 14, 681.
S. S. H. Mao, Y. Ra, L. Guo, C. Zhang, L. R. Dalton, A. Chen,
S. Garner and W. H. Steier, Chem. Mater., 1998, 10, 146.
A. W. Harper, S. S. H. Mao, Y. Ra, C. Zhang, J. Zhu, L. R. Dalton,
S. Garner, A. Chen and W. H. Steier, Chem. Mater., 1999, 11,
2886.
7
8
9
O. Kaliski, GB-253 Organics for Electro-optical Applications,
BCC Inc. Report, 2002.
20 K. Clays and A. Persoons, Phys. Rev. Lett., 1991, 66, 2980.
21 K. Clays and A. Persoons, Rev. Sci. Instrum., 1992, 63, 3285; 1994,
65, 2190.
22 G. Olbrechts, R. Strobbe, K. Clays and A. Persoons, Rev. Sci.
Instrum., 1998, 69, 2233.
L. R. Dalton, H. R. Fetterman, F. Wang, W. Steier, A. W. Harper,
A. S. Ren and J. Michael, US Patent 6 067 186, 2000.
S. R. Marder, L. T. Cheng, B. G. Tiemann, A. C. Friedli,
M. Blanchard-Desce, J. W. Perry and J. Skindhoj, Science, 1994,
263, 511.
23 E. Hendrickz, K. Clays and A. Persoons, Acc. Chem. Res., 1998,
31, 675.
10 J. Luo, S. Liu, M. Haller, L. Liu, H. Ma and A. K.-Y. Jen, Adv.
Mater., 2002, 14, 1763.
24 K. Clays, K. Wostyn, K. Binnemans and A. Persoons, Rev. Sci.
Instrum., 2001, 72, 3215.
11 A. J. Kay, A. D. Woolhouse, G. J. Gainsford, T. G. Haskell,
T. H. Barnes, I. T. McKinnie and C. P. Wyss, J. Mater. Chem.,
2001, 11, 996.
12 A. J. Kay, A. D. Woolhouse, G. J. Gainsford, T. G. Haskell,
C. P. Wyss, S. M. Giffin, I. T. McKinnie and T. H. Barnes,
J. Mater. Chem., 2001, 11, 2271.
13 F. Wang, A. Ren, M. He, A. W. Harper, L. R. Dalton,
S. M. Garner, H. Zhang, A. Chen and W. H. Steier, Polym.
Mater. Sci. Eng., 1998, 78, 42.
14 E. H. Rodd and G. E. Watts, British Patent 366 964, Chem. Abstr.,
1933, 27, 2040; US Patent 2 032 502, Chem. Abstr., 1936, 30,
2768.
25 J. L. Oudar and D. S. Chemla, J. Chem. Phys., 1977, 66, 2664.
26 J. Zyss and J. L. Oudar, Phys. Rev. A, 1982, 26, 2016.
27 J. L. Bredas, F. Meyers, B. M. Pierce and J. Zyss, J. Am. Chem.
Soc., 1992, 114, 4928.
28 B. H. Robinson, L. R. Dalton, A. W. Harper, A. Ren, F. Wang,
C. Zhang, G. Todorova, M. Lee, R. Aniszfeld and S. Garner,
Chem. Phys., 1999, 245, 35.
29 H. Ma, S. Liu, J. Luo, S. Suresh, L. Liu, S. H. Kang, M. Haller,
T. Sassa, L. R. Dalton and A. K.-Y. Jen, Adv. Funct. Mater., 2002,
12, 566.
30 L. R. Dalton, A. W. Harper and B. H. Robinson, Proc. Natl.
Acad. Sci. USA, 1997, 94, 4842.
15 L. G. S. Brooker, G. H. Keyes, R. H. Sprague, R. H. Van Dyke,
E. Van Lare, G. Van Zandt and F. L. White, J. Am. Chem. Soc.,
1951, 73, 5326.
31 N. S. Wilson and B. A. Keay, Tetrahedron Lett., 1996, 37, 153.
32 G. Melikian, F. P. Rouessac and C. Alexandre, Synth. Commun.,
1995, 25, 3045.
1 3 3 0
J . M a t e r . C h e m . , 2 0 0 4 , 1 4 , 1 3 2 1 – 1 3 3 0