Unconventional Langmuir-Blodgett films: Alignment of an optically nonlinear dye where the donor and π-electron bridge are hydrophobic and the acceptor is hydrophilic

Article abstract of DOI:10.1039/b200565b

The optically nonlinear dye, 5-{4-[2-(4-dibutylaminophenyl)vinyl]benzylidene}-2-(1-ethylpropyl)-5,6,7, 8-tetrahydroisoquinolinium bromide (1a), differs from other Langmuir-Blodgett (LB) film-forming materials. The molecule is amphiphilic and, therefore, may be aligned at the air-water interface but, instead of a conventional aliphatic tail, has an extended π-electron bridge. It forms non-centrosymmetric LB films which exhibit second-harmonic generation (SHG) and have a high second-order susceptibility: χ⁽²⁾zzz = 40 pm V^-1 at 1.064 μm for a monolayer thickness of 1.8 nm and chromophore tilt angle of 55° from the substrate normal. The octadecyl sulfate salt (1b) also forms non-centrosymmetric films but with improved properties: χ⁽²⁾zzz = 90 pm V^-1 at 1.064 μm for d = 2.4 nm and φ = 33°. Both films exhibit charge-transfer bands at 360 nm but with absorbance cut-off above 560 nm for 1a and 530 nm for 1b. The latter is fully transparent at the fundamental and harmonic wavelengths and its susceptibility is the highest to date for such a film. This results from an optimised packing arrangement, a conjugation enhanced molecular hyperpolarisability of β = 4.3 × 10^-38 m⁴ V^-1, and the proximity of the absorption band, albeit non-overlapping at 532 nm. The properties are compared with those of conventional long alkyl tailed materials (2 and 3) of the same chromophore.

Full text of DOI:10.1039/b200565b

View Article Online / Journal Homepage / Table of Contents for this issue
Unconventional Langmuir–Blodgett films: alignment of an optically
nonlinear dye where the donor and p-electron bridge are hydrophobic
and the acceptor is hydrophilic
Geoffrey J. Ashwell,* Alison A. Maxwell and Andrew Green
The Nanomaterials Group, Cranfield University, Cranfield, UK MK43 0AL.
E-mail: g.j.ashwell@cranfield.ac.uk
Received 16th January 2002, Accepted 10th April 2002
First published as an Advance Article on the web 17th June 2002
The optically nonlinear dye, 5-{4-[2-(4-dibutylaminophenyl)vinyl]benzylidene}-2-(1-ethylpropyl)-5,6,7,8-
tetrahydroisoquinolinium bromide (1a), differs from other Langmuir–Blodgett (LB) film-forming materials. The
molecule is amphiphilic and, therefore, may be aligned at the air–water interface but, instead of a conventional
aliphatic tail, has an extended p-electron bridge. It forms non-centrosymmetric LB films which exhibit second-
harmonic generation (SHG) and have a high second-order susceptibility: x(2) ~ 40 pm V21 at 1.064 mm for a
zzz
monolayer thickness of 1.8 nm and chromophore tilt angle of 55u from the substrate normal. The octadecyl
zzz
sulfate salt (1b) also forms non-centrosymmetric films but with improved properties: x(2) ~ 90 pm V21
at 1.064 mm for d ~ 2.4 nm and Q ~ 33u. Both films exhibit charge-transfer bands at 360 nm but with
absorbance cut-off above 560 nm for 1a and 530 nm for 1b. The latter is fully transparent at the fundamental
and harmonic wavelengths and its susceptibility is the highest to date for such a film. This results from an
optimised packing arrangement, a conjugation enhanced molecular hyperpolarisability of b ~ 4.3 6
10238 m4 V21, and the proximity of the absorption band, albeit non-overlapping at 532 nm. The properties
are compared with those of conventional long alkyl tailed materials (2 and 3) of the same chromophore.
Experimental
Introduction
Progress in the design of materials for SHG1 necessitates a
trade-off in efficiency for transparency as well as a dilution of
the active component when supramolecular organisation is
realised by electric field poling2 or LB deposition.3,4 The
effective volume of the optically nonlinear chromophore is
limited to ca. 15% in poled films, to avoid phase separation
from the polymer matrix, and is only slightly better in LB
monolayers where the occupancy is typically 40–50% with the
remainder taken up by hydrophobic alkyl groups. These are
necessary for alignment at the air–water interface but tend to
disrupt the non-centrosymmetry by enforcing head-to-head
(hydrophilic) and tail-to-tail (hydrophobic) arrangements in
multilayer films. Interleaving the layers with inactive spacers5–8
may give the necessary alignment but reduces the chromophore
occupancy to ca. 20–25%. An alternative approach is to
substitute opposite ends of the chromophore with alkyl
groups,9–11 which results in hydrophobic surfaces, both top
and bottom, and retention of the initial alignment throughout
the multilayer structure. In such films, an effective volume
of the chromophore layer of ca. 50% arises when either the
alkyl groups of adjacent layers interdigitate (‘‘molecular
Lego’’)12 or the lengths are optimised.13
Synthesis
4-[2-(4-Dibutylaminophenyl)vinyl]benzonitrile. To a solution
of 4-dibutylaminobenzaldehyde (3.5 g, 15 mmol) in methanol
(100 cm3) was added (4-cyanobenzyl)triphenylphosphonium
chloride (6.2 g, 15 mmol) and potassium tert-butoxide (1.7 g,
15 mmol). The mixture was heated at reflux for 24 h and then
cooled. The resultant yellow precipitate was purified by column
chromatography on silica gel, eluting with chloroform, to yield
In this work, we now focus upon improving the chromo-
phore occupancy within the LB layer and demonstrate
alignment of a donor–(p-bridge)–acceptor molecule without
long alkyl groups (1a and 1b). The donor and conjugated
electron-bridge act as the hydrophobic component and the
acceptor as the hydrophilic component for alignment at the
air–water interface. The second-order susceptibility of 1b,
x(2)zzz ~ 90 pm V21 at 1.064 mm, is the highest value to date for
a transparent film and confirms optimum non-centrosymmetric
alignment within the monolayer. The high value is attributed,
in part, to the extended p-electron bridge connecting the
donor and acceptor moieties and, in part, to a greatly improved
occupancy compared with conventional amphiphilic dyes.
2192
J. Mater. Chem., 2002, 12, 2192–2196
This journal is # The Royal Society of Chemistry 2002
DOI: 10.1039/b200565b

View Article Online
4-[2-(4-dibutylaminophenyl)vinyl]benzonitrile as a 1 : 1 mixture
of the cis and trans forms: combined yield, 60%; mp 110–112 uC;
lmax (CHCl3), 403 nm. Found: C, 82.8; H, 8.2; N, 8.4%.
C23H28N2 requires: C, 83.09; H, 8.49; N, 8.43%. The two forms
were partially separated by preparative plate chromatography,
eluting with chloroform–hexane (1 : 1 v/v).
5-{4-[2-(4-Dibutylaminophenyl)vinyl]benzylidene}-2-(1-ethyl-
propyl)-5,6,7,8-tetrahydroisoquinolinium bromide. To a solu-
tion of 4-[2-(4-dibutylaminophenyl)vinyl]benzaldehyde (0.20 g,
0.6 mmol) and 2-(1-ethylpropyl)-5,6,7,8-tetrahydroisoquinoli-
nium bromide (0.17 g, 0.6 mmol) in methanol (100 cm3) was
added piperidine (0.1 cm3) and the resultant mixture heated at
reflux for 100 h. The solvent was removed in vacuo and the
crude product purified by column chromatography on silica
gel, eluting with chloroform–methanol (90 : 10 v/v), followed by
recrystallisation from chloroform–diethyl ether (1 : 50 v/v), and
then repeated by preparative plate chromatography: yield, 20%;
mp 135–137 uC (decomp.). lmax (CHCl3): 395 and 490 nm.
Found: C, 74.1; H, 8.4; N, 4.6%. C37H49N2Br requires: C,
1
trans form. H NMR (CDCl3, 250 MHz, J/Hz): dH 0.96 (t,
J 7.3, 6H, CH3); 1.37 (sextet, J 7.3, 4H, CH2); 1.59 (quintet,
J 7.5, 4H, CH2); 3.30 (t, J 7.6, 4H, CH2N); 6.62 (d, J 9.2, 2H,
Ar-H); 6.83 (d, J 16.2, 1H, CLC-H); 7.13 (d, J 16.3, 1H, CLC-H);
7.43 (d, J 8.5, 2H, Ar-H); 7.51 (d, J 8.4, 2H, Ar-H); 7.57 (d,
J 8.5, 2H, Ar-H). m/z (FAB): 332 (M1, 100%).
cis form. 1H NMR (CDCl3, 250 MHz, J/Hz): dH 0.96 (t,
J 7.3, 6H, CH3); 1.37 (sextet, J 7.3, 4H, CH2); 1.59 (quintet,
J 7.5, 4H, CH2); 3.24 (t, J 7.6, 4H, CH2N); 6.29 (d, J 12.2, 1H,
CLC-H); 6.47 (d, J 8.9, 2H, Ar-H); 6.58 (d, J 12.2, 1H, CLC-H);
7.06 (d, J 8.9, 2H, Ar-H); 7.38 (d, J 8.9, 2H, Ar-H); 7.50 (d,
J 7.5, 2H, Ar-H). m/z (FAB): 332 (M1, 100%).
1
73.86; H, 8.21; N, 4.66%. H NMR (CDCl3, 250 MHz, J/Hz):
dH 0.90 (t, J 7.3, 6H, CH3); 0.95 (t, J 7.2, 6H, CH3); 1.36 (sextet,
J 7.3, 4H, CH2); 1.59 (quintet, J 7.4, 4H, CH2); 1.92 (quintet,
J 6.5, 2H, CH2); 2.11 (quintet, J 7.2, 4H, CH2); 2.96 (t, J 5.7,
CLC-CH2); 3.12 (t, J 6.0, 2H, CLC-CH2); 3.31 (t, J 6.8, 4H,
CH2N); 4.77 (quintet, J 7.1, 1H, N1CH); 6.65 (d, J 8.2, 2H,
Ar-H); 6.89 (d, J 15.9, 1H, CLC-H); 7.12 (d, J 16.1, 1H, CLC-
H); 7.39–7.50 (m, 6H, Ar-H); 7.60 (s, 1H, CLC-H); 8.33 (d, J
6.6, 1H, Qn-H); 9.00 (d, J 6.5, 1H, Qn-H); 9.23 (s, 1H, Qn-H).
m/z (FAB): 521 (M1 2 Br2, 100%).
4-[2-(4-Dimethylaminophenyl)vinyl]benzonitrile. This was pre-
pared from 4-(dimethylamino)benzaldehyde by adapting the
procedure described above. cis and trans forms, in a ratio of ca.
1 : 1, were obtained: combined yield 60%; mp 255–256 uC (161 uC
darkens); lmax (CHCl3): 386 nm. Found: C, 82.3; H, 6.5; N,
11.1%. C17H16N2 requires: C, 82.22; H, 6.49; N, 11.28%.
trans form. 1H NMR (CDCl3, 250 MHz, J/Hz): dH 3.02
(s, 6H, N(CH3)2); 6.72 (d, J 8.7, 2H, Ar-H); 6.88 (d, J 16.2, 1H,
CLC-H); 7.16 (d, J 15.8, 1H, CLC-H); 7.44 (d, J 8.4, 2H, Ar-H);
7.52 (d, J 8.8, 2H, Ar-H); 7.59 (d, J 8.4, 2H, Ar-H). m/z (FAB):
248 (M1, 100%).
5-{4-[2-(4-Dibutylaminophenyl)vinyl]benzylidene]}-2-dodecyl-
5,6,7,8-tetrahydroisoquinolinium iodide. The dye was obtained
from 1-dodecyl-5,6,7,8-tetrahydroisoquinolinium iodide by
adapting the procedure described above: yield 12%; mp 162–
163 uC; lmax (CHCl3), 360 and 492 nm. Found: C, 70.4; H, 8.6;
1
N, 3.8%. C44H63N2I requires: C, 70.76; H, 8.50; N, 3.75%. H
cis form. 1H NMR (CDCl3, 250 MHz, J/Hz): dH 2.97 (s, 6H,
N(CH3)2); 6.36 (d, J 12.1, 1H, CLC-H); 6.60 (d, J 8.4, 2H,
Ar-H); 6.63 (d, J 12.0, 1H, CLC-H); 7.11 (d, J 8.4, 2H, Ar-H);
7.41 (d, J 8.0, 2H, Ar-H); 7.51 (d, J 8.8, 2H, Ar-H). m/z (FAB):
248 (M1, 100%).
NMR (400 MHz, CDCl3, J/Hz): dH 0.88 (t, J 6.3, 3H, CH3);
0.97 (t, J 7.1, 6H, CH3); 1.26 (br s, 18H, CH2); 1.33–1.43
(m, 4H, CH2); 1.53–1.60 (m, 4H, CH2); 1.94–2.05 (m, 4H,
CH2); 2.99 (t, 2H, J 6.3, CLC-CH2); 3.07 (t, J 6.3, 2H, CLC-
CH2); 3.31 (t, J 7.5, 4H, N(CH2)2); 4.74 (t, J 7.2, 2H, CH2N1);
6.64 (d, J 8.5, 2H, Ar-H); 6.88 (d, J 16.5, 1H, CLC-H); 7.14 (d,
J 16.5, 1H, CLC-H); 7.39–7.53 (m, 7H, Ar-H & CLC-H); 8.16
(d, J 6.7, 1H, Qn-H); 8.77 (d, J 6.0, 1H, Qn-H); 8.96 (s, 1H,
Qn-H). m/z(FAB): 620 (M1 2 I2, 100%).
trans 4-[2-(4-Dibutylaminophenyl)vinyl]benzaldehyde. Diiso-
butylaluminium hydride (1.5 M in toluene: 18 cm3, 27 mmol)
was added dropwise to a solution of 4-[2-(4-dibutylaminophe-
nyl)vinyl]benzonitrile (3.0 g, 9 mmol) in chloroform (100 cm3)
at 278 uC and the temperature maintained for 1 h. Upon
reaching room temperature, the solution was poured into
acidified water and the organic layer extracted, washed with
water (3 6 50 cm3) and dried (MgSO4). The solvent was
removed in vacuo and the crude product purified by column
chromatography on silica gel, eluting with chloroform, and
then by preparative plate chromatography, eluting with a
mixture of chloroform and hexane (1 : 1 v/v). The trans form
was obtained as a yellow solid: yield, 40%; mp 68–69 uC; lmax
(CHCl3), 409 nm. Found: C, 82.0; H, 8.5; N, 4.0%. C23H29NO
requires: C, 82.34; H, 8.71; N, 4.18%. 1H NMR (CDCl3,
250 MHz, J/Hz): dH 0.97 (t, J 7.2, 6H, CH3); 1.37 (sextet, J 7.4,
4H, CH2); 1.59 (quintet, J 7.5, 4H, CH2); 3.30 (t, J 7.5, 4H,
CH2N); 6.63 (d, J 8.9, 2H, Ar-H); 6.89 (d, J 16.2, 1H, CLC-H);
7.19 (d, J 16.2, 1H, CLC-H); 7.40 (d, J 8.9, 2H, Ar-H); 7.58 (d,
J 8.3, 2H, Ar-H); 7.82 (d, J 8.3, 2H, Ar-H); 9.95 (s, 1H, CHO).
m/z (FAB): 335 (M1, 100%).
5-{4-[2-(4-Dibutylaminophenyl)vinyl]benzylidene}-2-octadecyl-
5,6,7,8-tetrahydroisoquinolinium iodide. The dye was obtained
from 1-octadecyl-5,6,7,8-tetrahydroisoquinolinium iodide by
adapting the procedure described above: yield 20%; mp 108–
110 uC; lmax (CHCl3), 360 and 492 nm. Found: C, 71.8; H, 9.2;
1
N, 3.3%. C50H75N2I requires: C, 72.26; H, 9.10; N, 3.37%. H
NMR (400 MHz, CDCl3, J/Hz): dH 0.85 (t, J 6.8, 3H, CH3);
0.95 (t, J 7.3, 6H, CH3); 1.23 (br s, 30H, CH2); 1.35 (sextet, J
7.5, 4H, CH2); 1.57 (quintet, J 6.9, 4H, CH2); 1.89 (quintet, J
5.8, 2H, CH2); 1.98 (quintet, J 7.6, 4H, CH2); 2.91 (t, 2H, J 5.4,
CLC-CH2); 3.03 (t, J 5.9, 2H, CLC-CH2); 3.28 (t, J 7.5, 4H,
N(CH2)2); 4.68 (t, J 7.3, 2H, CH2N1); 6.62 (d, J 8.2, 2H, Ar-
H); 6.85 (d, J 16.1, 1H, CLC-H); 7.09 (d, J 16.2, 1H, CLC-H);
7.36–7.41 (m, 4H, Ar-H); 7.43 (s, 1H, CLC-H); 7.48 (d, J 8.8,
2H, Ar-H); 8.19 (d, J 6.8, 1H, Qn-H); 8.77 (d, J 6.5, 1H, Qn-H);
9.05 (s, 1H, Qn-H). m/z (FAB): 704 (M1 2 I2, 100%).
5-{4-[2-(4-Dimethylaminophenyl)vinyl]benzylidene}-2-octadecyl-
5,6,7,8-tetrahydroisoquinolinium iodide. The dye was obtained
from 1-octadecyl-5,6,7,8-tetrahydroisoquinolinium iodide and
4-[2-(4-dimethylaminophenyl)vinyl]benzaldehyde by adapting
the procedure described above: yield 28%; mp 181–183 uC; lmax
(CHCl3), 355 and 472 nm. Found: C, 70.3; H, 8.5; N, 3.6%.
C44H63N2I requires: C, 70.76; H, 8.50; N, 3.75%. 1H NMR
(CDCl3, 250 MHz, J/Hz): dH 0.87 (t, J 6.4, 3H, CH3); 1.24 (br s,
32H, CH2); 1.85–2.01 (m, 2H, CH2); 2.91 (t, J 6.0, 2H, CLC-
CH2); 3.00 (s, 6H, N(CH3)2); 3.04 (t, J 6.0, 2H, CLC-CH2); 4.68
(t, J 7.2, 2H, CH2N1); 6.71 (d, J 8.7, 2H, Ar-H); 6.89 (d, J 16.2,
trans 4-[2-(4-Dimethylaminophenyl)vinyl]benzaldehyde. This
was obtained from 4-[2-(4-dibutylaminophenyl)vinyl]benzo-
nitrile by adapting the procedure described above: yield 50%;
mp 247–250 uC; lmax (CHCl3), 401 nm. Found: C, 81.1; H, 6.8;
1
N, 5.6%. C17H17NO requires: C, 81.24; H, 6.82; N, 5.57%. H
NMR (CDCl3, 250 MHz, J/Hz): dH 3.02 (s, 6H, NCH3); 6.73
(d, J 8.2, 2H, Ar-H); 6.94 (d, J 16.2, 1H, CLC-H); 7.22 (d,
J 16.3, 1H, CLC-H); 7.45 (d, J 8.3, 2H, Ar-H); 7.60 (d, J 8.2,
2H, Ar-H); 7.84 (d, J 8.1, 2H, Ar-H); 9.96 (s, 1H, CHO).
m/z (FAB): 251 (M1, 100%).
J. Mater. Chem., 2002, 12, 2192–2196
2193

View Article Online
sodium octadecyl sulfate, in a 1 : 1 mole ratio, from chloroform
and methanol respectively. The inorganic counterions, Na1
and Br2, dissolve and leave the octadecyl sulfate salt of the
dye at the surface. When compressed as above the isotherm
exhibits areas consistent with the expected cross-sections: the
film undergoes collapse at 42 mN m21 with an onset area of
0.40 nm2 molecule21 and then shows a second transition at
51 mN m21 and 0.20 nm2 molecule21, which probably
corresponds to collapse of the resultant bilayer structure
(Fig. 1). In this study, the film was deposited on the upstroke at
a surface pressure of 35 mN m21 and a rate of 0.1 mm s21, this
giving an optimum second-harmonic intensity. The bromide
salt shows optimum behaviour when deposited at 30 mN m21
.
Thicknesses were obtained from SPR studies, which were
performed on glass/Au/monolayer structures using a Kretsch-
mann configuration14 and p-polarised monochromatic radia-
tion at the following excitatory wavelengths (Fig. 2): 1.064 mm
and 532 nm (Nd : YAG); 632.8, 611.9, 604.0, 594.1 and
543.5 nm (multi-line He : Ne). The attenuated total reflection
was measured before and after deposition and the data
analysed by one and two-layer models respectively using
Fresnel reflection formulae. Analysis provided thicknesses of
1.75 ¡ 0.06 nm and 2.43 ¡ 0.06 nm for the bromide and
octadecyl sulfate salts respectively, the values being confirmed
by atomic force microscopy. The refractive indices of both films
are almost identical and independent of wavelength throughout
the visible range: n ~ 1.69 ¡ 0.05 for 532 nm ¡ l ¡ 632.8 nm
and 1.62 ¡ 0.10 at 1.064 mm. This is expected from the
spectrum, the films having a peak absorbance at 360 nm with
cut-off above ca. 560 nm for 1a and 530 nm for 1b (Fig. 3).
The latter is transparent at both the fundamental and
harmonic wavelengths for SHG, the monolayer absorbance
being less than the detection limit of the spectrophotometer
Fig. 1 Surface pressure versus area isotherms of 1: bromide salt (broken
line); octadecyl sulfate salt (solid line). The bromide salt is partially
water soluble and this accounts for the unrealistic molecular area.
1H, CLC-H); 7.12 (d, J 16.2, 1H, CLC-H); 7.40–7.48 (m, 6H,
Ar-H); 7.51 (s, 1H, CLC-H); 8.18 (d, J 6.7, 1H, Qn-H); 8.76 (d, J
6.5, 1H, Qn-H); 9.06 (s, 1H, Qn-H). m/z(FAB): 619 (M1 2 I2,
100%).
Deposition and film characterisation. LB films were depos-
ited onto hydrophilically treated glass slides and gold-coated
glass substrates for SHG and surface plasmon resonance (SPR)
studies respectively. The specific conditions for LB deposition
and details relating to the film characterisation are provided
below. Additional information concerning these techniques can
be found in early publications5,11 and references therein.
and estimated at v1024
.
Results and discussion
The second-harmonic intensity was measured in transmis-
sion using a p-polarised Nd : YAG laser and half-wave plate
to alter the polarisation. The intensity is negligible at normal
incidence as found for most LB films and, therefore, the
polarisation dependence was investigated with the laser beam
incident at 45u. However, the dependence was very different for
the two salts with I2v(p A p)/I2v(s A p) # 1.5 for the bromide
and 15 for the octadecyl sulfate (Fig. 4). These indicate
different tilt angles and, using the method of Kajikawa
et al.,15 the data correspond to a tilt of the chromophore
charge-transfer axis of 33u from the substrate normal for the
latter. The method fails for the bromide salt but, from the ratio
of the layer thickness and the theoretically modelled molecular
Unconventional amphiphiles
The bromide salt of dye 1 was spread from dilute chloroform
solution (1 6 1024 M) onto the pure water subphase of a Nima
Technology LB trough, left for 10 min, and then compressed at
0.5 cm2 s21, which corresponds to a change of ca. 0.1% s21 of
the compartment area. The isotherm is featureless with collapse
occurring at 41 ¡ 2 mN m21 but, for this unconventional
amphiphile, the area is less than the molecular cross-section
(Fig. 1). The monolayer film is probably partially soluble in the
aqueous subphase but this does not influence its deposition.
Furthermore, it may be overcome by co-spreading the dye and
Fig. 2 SPR spectra from a glass/Au/monolayer structure of the octadecyl sulfate salt of 1 on a 40 nm thick gold overlay at the following excitatory
wavelengths: 532 and 1064 nm (Nd : YAG); 543.5, 594.1, 604.0; 611.9 and 632.8 nm (He : Ne). The spectra broaden with decreasing wavelength and
the angle for resonance increases.
2194
J. Mater. Chem., 2002, 12, 2192–2196

View Article Online
Fig. 5 Cationic dye 1 as modelled and, for clarity, omitting the
counterion.
bromide salt provides unreliable data and, in this case, an
area of 0.47 nm2 molecule21 at the deposition pressure was
calculated by a Sauerbrey analysis16 of the frequency change
following deposition onto a 10 MHz quartz crystal. This gives
a hyperpolarisability of 4.5 6 10238 m4 V21
.
The susceptibility of the octadecyl sulfate salt is the highest
to date for any transparent film and may be compared with,
for example, 16 pm V21 at 1.064 mm for LB films of the
extensively studied 2-docosylamino-5-nitropyridine17–19 which
has a peak absorbance at 374 nm and is effectively transparent
above 500 nm. In this work an optimum of x(2)zzz ~ 90 pm V21
has been realised for films where lmax ~ 360 nm with cut-off
above 530 nm.
Fig. 3 UV/visible spectrum of an LB monolayer of the octadecyl sulfate
salt of 1 showing a peak absorbance at 360 nm and cut-off at ca.
530 nm, the residual absorbance above this wavelength being noise.
The bromide salt shows similar behaviour but with absorption cut-off
above 560 nm.
Long tailed analogues of 1b with a shorter p-bridge, for
example, 5-{4-[2-(4-dibutylamino)benzylidene]}-2-octadecyl-
5,6,7,8-tetrahydroisoquinolinium octadecyl sulfate,10b,13 have
shown asymmetric current–voltage characteristics when sand-
wiched between gold electrodes, the behaviour being assigned
to molecular rectification. However, the electrical properties
are disadvantaged by the hydrophobic alkyl substituent which
acts as a barrier and separates the active donor–(p-bridge)–
acceptor component from the electrode. The novel design of
dye 1 (Fig. 5) overcomes this problem, the alignment being
achieved by utilising a conductive p-bridge rather than an
insulating aliphatic tail. Therefore, it is anticipated that its
films should rectify with improved behaviour.
Conventional amphiphiles
Fig. 4 Second-harmonic intensity from
a monolayer film of the
Dyes 2 and 3 have long aliphatic tails and spontaneously align
when their iodide salts are co-spread with sodium octadecyl
sulfate at the air–water interface. The dimethylamino analogue
(3) has been reported previously.24 Its films exhibit a quadratic
increase of the second-harmonic intensity with the number of
LB layers when alternated with poly(tert-butyl methacrylate).
octadecyl sulfate of 1 salt versus the angle of the half-wave plate, the
maximum corresponding to a p-polarised fundamental beam and the
minima to s-polarised. The laser beam is incident at 45u to the film.
length, an angle of 55u may be estimated. As both films are
effectively transparent at the harmonic wavelength it may
be assumed that Kleinman’s symmetry is valid and that the
components of the susceptibility are limited to x(2)
x(2)zxx. By comparison with the intensity from the Maker fringe
envelope of a Y-cut quartz reference plate (d11 ~ 0.5 pm V21),
the susceptibilities for l ~ 1.064 mm are 40 pm V21 and
28 pm V21 respectively for the bromide salt and 90 and
19 pm V21 for the octadecyl sulfate. The molecular
hyperpolarisability (b) is probably dominated by the compo-
nent along the molecular charge transfer axis and the relation
between x(2) and b is as given below:
The second-order susceptibility and absorbance are x(2)
~
zzz
45 pm V21 at 1.064 mm and A ~ 3.5 6 1024 bilayer21 at
532 nm respectively with cut-off above 600 nm. The
dibutylamino analogue with a dodecyl tail (2a) is an isomer
of dye 3 and its LB films exhibit similar behaviour, as do those
of the octadecyl derivative (2b). In each case, the second-
harmonic intensities from monolayer films are comparable
to those obtained from films of 1b and confirm optimum
non-centrosymmetric alignment for this unconventional LB
film-forming analogue. However, as a consequence of the
long aliphatic tails of 2 and 3, which are SHG-inactive
but contribute to the film thickness, the susceptibilities are
significantly lower.
and
zzz
x(2) ~ Nf 2v(f v)2bcos3Q
(1)
zzz
x(2) ~ ½Nf 2v(f v)2bcosQsin2Q
(2)
zxx
Conclusion
where N is the number of molecules per unit volume, f v and f 2v
are local field correction factors at v and 2v respectively,
and f ~ (n2 1 2)/3 where n is the refractive index at the
corresponding wavelength. For the octadecyl sulfate salt, the
molecular hyperpolarisability is 4.3 6 10238 m4 V21 where
the number density is calculated from the product of layer
thickness and area from the isotherm at the deposition
pressure. The pressure–area isotherm of the partially soluble
Chromophore occupancy and alignment are significant factors
when optimising the second-order susceptibility of bulk films
whereas, at the molecular level, the hyperpolarizability is
dependent upon the p-electron bridge20,21 and nature of the
substituent groups.22,23 It may be enhanced by increasing the
conjugation length and, in this work, we have successfully
improved the bulk (x(2)) and molecular (b) components by
extending the p-bridge of the stilbazolium dye and utilising it
J. Mater. Chem., 2002, 12, 2192–2196
2195

View Article Online
9
G. J. Ashwell, P. D. Jackson and W. A. Crossland, Nature, 1994,
368, 438.
as the hydrophobic component for alignment at the air–water
interface and subsequently for LB deposition. The suscept-
ibility is the highest obtained to date for any transparent film.
10 (a) G. J. Ashwell, R. Ranjan, A. J. Whittam and D. Gandolfo,
J. Mater. Chem., 2000, 10, 63; (b) G. J. Ashwell and D. Gandolfo,
J. Mater. Chem., 2001, 11, 246.
11 G. J. Ashwell, D. Gandolfo and R. Hamilton, J. Mater. Chem.,
2002, 12, 416.
Acknowledgement
12 G. J. Ashwell, R. Hamilton, B. J. Wood, I. R. Gentle and D. Zhou,
J. Mater. Chem., 2001, 11, 2966.
13 G. J. Ashwell and D. Gandolfo, J. Mater. Chem., 2002, 12, 411.
14 E. Kretschmann, Z. Phys., 1971, 241, 313.
We are grateful to Mukhtar Amiri and Anne Whittam for
technical assistance and the EPSRC (UK) for financial support
and for providing a studentship to A. A. M.
15 K. Kajikawa, K. Kigata, H. Takezoe and A. Fukuda, Mol. Cryst.
Liq. Cryst., 1990, 182, 91.
16 G. Sauerbrey, Z. Phys., 1959, 155, 206.
References
17 (a) G. Decher, B. Tieke, C. Bosshard and P. Gu¨nter, J. Chem. Soc.,
Chem. Commun., 1988, 933; (b) C. Bosshard, G. Decher, B. Tieke
and P. Gu¨nter, Proc. SPIE - Int. Soc. Opt. Eng., 1988, 1017, 141.
18 (a) M. Ku¨pfer, M. Flo¨rsheimer, C. Bosshard and P. Gu¨nter,
Electron. Lett., 1993, 29, 2033; (b) M. Ku¨pfer, M. Flo¨rsheimer,
C. Bosshard and P. Gu¨nter, Nonlinear Opt., 1995, 10, 341.
19 (a) N. Asai, H. Tamada, I. Fujiwara and J. Seto, J. Appl. Phys.,
1992, 10, 4521; (b) I. Fujiwara, N. Asai and V. Howarth, Thin
Solid Films, 1992, 221, 285.
1
C. Bosshard, K. Sutter, P. Preˆtre, J. Hulliger, M. Flo¨rsheimer,
P. Kaatz and P. Gu¨nter, Organic Nonlinear Optical Materials,
Advances in Optics, Vol. 1, ed. A. F. Garito and F. Kajzar, Gordon
and Breach, Basel, Switzerland, 1995.
K. D. Singer, S. L. Lalama, J. E. Sohn and R. D. Small, in
Nonlinear Optical Properties of Organic Molecules and Crystals,
ed. D. S. Chemla and J. Zyss, Academic Press, Orlando, 1987.
M. C. Petty, Langmuir-Blodgett Films: an Introduction, Cambridge
University Press, Cambridge, 1996.
2
3
4
5
G. J. Ashwell, J. Mater. Chem., 1999, 9, 1991.
20 A. Dulcic, C. Flytzanis, C. L. Tang, D. Pe´pin, M. Fe´tizon and
Y. Hoppilliard, J. Chem. Phys., 1981, 74, 1559.
21 R. A. Huijts and G. L. J. Hesselink, Chem. Phys. Lett., 1989, 156,
209.
22 A. Dulcic and C. Sauteret, J. Chem. Phys., 1978, 69, 3453.
23 S. R. Marder, D. N. Beratan and L.-T. Cheng, Science, 1991, 252,
103.
24 G. J. Ashwell, A. J. Whittam, M. A. Amiri, R. Hamilton, A. Green
and U. W. Grummt, J. Mater. Chem., 2001, 11, 1345.
G. J. Ashwell, E. J. C. Dawnay, A. P. Kuczynski and P. J. Martin,
Proc. SPIE – Int. Soc. Opt. Eng., 1991, 1361, 589.
M. Era, K. Nakamura, T. Tsutsui, S. Saito, H. Niino, K.
Takehara, K. Isomura and H. Tanaguchi, Jpn. J. Appl. Phys.,
1992, 210–211, 163.
P. Hodge, Z. Ali-Adib, D. West and T. A. King, Macromolecules,
1993, 26, 1789.
S. H. Ma, X. Z. Lu, J. H. Xu, W. C. Wang and Z. M. Zhang,
J. Phys. D, 1997, 30, 2651.
6
7
8
2196
J. Mater. Chem., 2002, 12, 2192–2196