Second harmonic generation from Langmuir-Blodgett multilayers assembled from 'active' non-polymeric amphiphiles and 'inactive' polymeric amphiphiles

Article abstract of DOI:10.1039/a708854j

Alternating Langmuir-Blodgett (LB) multilayers have been prepared from a non-polymeric hemicyanine and a poly(4-vinylpyridine) 66% quaternised with docosyl bromide, and from a non-polymeric azobenzene and the same pyridinium polymer. With both types of mu

Full text of DOI:10.1039/a708854j

J O U R N A L O F
Second harmonic generation from Langmuir–Blodgett multilayers
assembled from ‘active’ non-polymeric amphiphiles and ‘inactive’
polymeric amphiphiles
Damien Dunne,a Philip Hodge,*a Ziad Ali-Adib,a Neil B. McKeowna and David Westb
C H E M I S T R Y
aDepartment of Chemistry, University of Manchester, Oxford Road, Manchester, UK M13 9PL
bDepartment of Physics, University of Manchester, Oxford Road, Manchester, UK M13 9PL
Alternating Langmuir–Blodgett (LB) multilayers have been prepared from a non-polymeric hemicyanine and a poly(4-
vinylpyridine) 66% quaternised with docosyl bromide, and from a non-polymeric azobenzene and the same pyridinium polymer.
With both types of multilayer the intensity of the SHG signal increased as the square of the number of layers up to a thickness of
ca. 60 bilayers. The slopes of the plots correspond to second order susceptibilities of 17.5 and 9 pm V−1 respectively. Although
these properties and the stabilities of the films over time compare favourably with many other LB films described in the literature,
these films prepared from ‘active’ non-polymeric amphiphiles and a ‘passive’ polymeric amphiphile are not as satisfactory as the
closely related all-polymeric films described previously.
In recent years there has been great interest in materials for
CH CH2
CH CH2
second harmonic generation (SHG), i.e. materials which on
irradiation with light of a given frequency produce a significant
intensity of light of twice the incident frequency. Such materials
have many potential applications in optoelectronics.1–3 To
achieve SHG, materials generally need to contain good SHG
chromophores macroscopically arranged in a non-centrosym-
metric manner. One way4 of achieving the latter requirement
is to prepare Y-type Langmuir–Blodgett (LB) multilayers in
which every alternate layer is ‘active’, i.e. has an SHG chromo-
phore, and every alternate layer is ‘inactive’: see Fig. 1. Good
SHG chromophores generally have a p-electron system substi-
tuted at one end with an electron-donor group and at the
other end with an electron-withdrawing group. Examples are
the hemicyanine chromophore, present in polymer 1 and
compounds 3 and 4, and the azobenzene chromophore, present
in compounds 5 and 6.
Br–
N+
n-C22H45
2
H3C
CH3
n-C22H45
CH3
n-C22H45
CH3
N
N
N
N
CH
CH
N
CH
N+
CH
N+
X
Br–
Br–
CH CH2
CH CH2
NO2
n-C22H45
C2H5
C
O
C
O O
5 : X = H
6 : X = CO2H
O
3
4
CH2
CH2
CH2
CH2
of alternating LB multilayers from ‘active’ polymer 1 and
‘inactive’ polymer 2.8 With films containing up to a total of 600
layers the intensity of the SHG light, I , increased as the square
N
CH3
N
CH3
2v
of the number of layers, as predicted by theory.4 The macro-
scopic second order susceptibility, x(2), of these films was 12 pm
V−1. The films were stable at 20 °C for at least 8 months and
at 60 °C for at least 96 h.8 A Fabry–Perot device was successfully
fabricated using the same active materials.9
We now describe the preparation of LB multilayers from
‘active’ non-polymeric amphiphiles and the ‘inactive’ polymer
2. The main objective was to determine whether readily
available non-polymeric SHG amphiphiles could be used just
as successfully as less readily available ‘active’ polymeric amphi-
philes such as 1, i.e. to determine if it was really necessary to
have both amphiphiles polymeric in order to obtain satisfac-
torily stable films.
CH
CH
N+
Br–
n-C22H45
1
LB multilayers prepared from non-polymeric amphiphiles,
although often highly organised initially, are in general physi-
cally fragile and prone, over an extended period of time, to
molecularly reorganise.5–7 LB multilayers prepared from amphi-
philic polymers, although less well organised initially, are gener-
ally physically more stable and can be expected to be less prone
to reorganisation. We have described recently the preparation
Experimental
Melting points were determined using an Electrothermal melt-
ing point apparatus and are uncorrected. 1H NMR spectra
J. Mater. Chem., 1998, 8(6), 1391–1397
1391

were recorded on a Varian Gemini 200 MHz instrument using
CDCl solutions. Ultraviolet (UV) spectra were recorded on a
Shimadzu UV-260 spectrophotometer. Elemental analyses
were obtained using a Carlo-Erba CHNS-O EA instrument.
perature kept at 5 °C for 30 min. The mixture was then heated
at 90 °C for 3 h. After cooling, the mixture was added to
crushed ice (400 g). Adjustment of the pH to 5 using sodium
acetate precipitated the desired product, which was collected
3
and dried (7.23 g, 90%). It had mp 55 °C, d 9.71 (s, 1H,
ArCHO), 7.73 (m, 2H, aromatic), 6.72 (m, 2H, aromatic), 3.45
H
Source of amphiphiles
(t, 2H, CH N), 3.01 (s, 3H, CH N), 1.33 (m, 40H, 20 CH ) and
0.94 (s, 3H, CH ) (Found: C 80.9, H 12.4, N 3.3%. Calc. for:
C H NO, C 81.3, H 12.0, and N 3.2%).
2
3
2
Polymer 2. This polymer was available from the earlier
studies.10 It was prepared by reacting a poly(4-vinylpyridine)
3
30 53
of M 25 000 with 1-docosyl bromide: 66% of the pyridine
9
residues in the product were quaternised.
v
(c) N-Ethylpicolinium bromide. Equimolar amounts of pico-
line and bromoethane were heated at reflux temperature for
4 h. The product was precipitated into diethyl ether, collected
and dried. This gave the desired product, mp 122–123 °C
(lit.,12 121 °C).
Compound 3. This compound, prepared as described pre-
viously,11 had mp 237–239 °C (lit.,11 240 °C).
Compound 4
(a) N-Docosyl-N-methylaniline. A mixture of freshly distilled
N-methylaniline (19.58 g, 0.18 mmol), 1-docosyl bromide
(74.0 g,
0.19 mmol),
potassium
carbonate
(25.53 g,
(d) Compound 4. A mixture of N-docosyl-N-methylaniline
(4.73 g, 11.4 mmol), N-ethylpicolinium bromide (2.31 g,
11.4 mmol), piperidine (1 ml) and methanol (50 ml) was heated
under reflux for 18 h. The solvent was then evaporated off, the
residue dissolved in a minimum volume of dichloromethane
(~10 ml) and the product isolated by precipitation into hexane
(400 ml). The product 4 was collected and dried (49% yield).
It had mp 192–194 °C (Found: C 72.3, H 9.7, N 4.4, Br 12.4%.
Calc. for: C H BrN , C 72.7, H 10.0, N 4.5 and Br 12.8%).
0.185 mmol), potassium iodide (0.76 g, 5.0 mmol) and n-buta-
nol (150 ml) was heated at reflux temperature under nitrogen
and vigorously stirred for 4 d. The mixture was then cooled
and filtered. The solvent was evaporated off under vacuum
and the residue treated with aqueous potassium hydroxide
(150 ml of 50%) at 20 °C. After 2 h the mixture was extracted
with diethyl ether (4×75 ml). The extracts were combined and
dried over sodium sulfate. Removal of the solvent left N-
docosyl-N-methylaniline (71% yield) as a white waxy solid,
38 63
2
mp 41–42 °C. It had d 7.3–6.6 (m, 5H, aromatic), 3.32 (m, 2H,
H
MCH N), 2.90 (s, 3H, CH N), 1.34 (m, 40H, 20 CH ), 0.92 (t,
3H, CH ) (Found: C 83.8, H 12.6, and N 2.9%. Calc. for:
C H N, C 83.8, H 12.8, and N 3.4%).
2
3
2
Compound 5. Using the general procedure given by Vogel,13
N-docosyl-N-methylaniline was diazotized with the diazonium
salt prepared from 4-nitroaniline. This gave 4∞-N-docosyl-N-
methylamino-4-nitroazobenzene 5, mp 105–106 °C, in 57%
yield. It had d 8.3–7.7 (m, 8H, aromatic), 3.50 (m, 2H, CH N),
3
29 53
(b) 4-(N-Docosyl-N-methylamino)benzaldehyde. Phosphorus
oxychloride (5.37 g, 35 mmol) was added dropwise to anhy-
drous N,N-dimethylformamide (12.77 g, 175 mmol) at 5 °C
under nitrogen and the mixture was vigorously stirred for
30 min. N-Docosyl-N-methylaniline (7.47 g, 18 mmol) in N,N-
dimethylformamide (25 ml) was added dropwise and the tem-
H
2
3.31 (s, 3H, CH N), 1.32 (m, 40H, 20 CH ) and 0.93 (t, 3H,
CH ) (Found: C 74.0, H 9.8, N 9.6%. Calc. for: C H N O ,
C 74.3, H 10.1 and N 9.9%).
3
2
3
35 57
4 2
Compound 6. Using the general procedure given by Vogel,13
N-docosyl-N-methylaniline was diazotized with the diazonium
salt prepared from commercial 2-nitro-5-aminobenzoic acid.
This gave 5-[4-(N-docosyl-N-methylamino)phenylazo]-2-
nitrobenzoic acid 6, mp 140–142 °C in 73% yield. It had d
H
8.5–7.7 (m, 7H, aromatic), 3.33 (m, 2H, CH N), 3.02 (s, 3H,
2
CH N), 1.34 (m, 40H, 20 CH ) and 0.94 (t, 3H, CH ) (Found:
3
2
3
C 70.6, H 9.0, N 9.1%. Calc. for: C H N O , C 70.9, H 9.4
and N 9.2%).
36 57
4
4
Measurement of isotherms, preparation of films and
measurement of their properties
Details of the Langmuir troughs used,14 and the general
procedures for the measurement of isotherms14 and for the
preparation of LB multilayers14 have been described before.
In the present work all amphiphiles were spread from solutions
in chloroform. The subphase was doubly-distilled deionised
water with no additives, pH 5.2–5.6, at 20 °C. The LB films
were deposited at a dipping speed of 8 mm min−1 onto silan-
ized silicon wafers or, for SHG materials, Pyrex glass micro-
scope slides at a surface pressure of 30 mN m−1. Deposition
ratios were 1.00 0.05 in all experiments. The alternating LB
films were prepared by depositing the ‘active’ amphiphile on
the upstroke and polymer 2 on the downstroke.
X-Ray reflectivity experiments14 and SHG measurements
(incident laser light of wavelength 1.064 mm) were carried out
as described before.15–17 All the LB films were stored in the
dark at 20 °C because previous studies have shown that
hemicyanines are sensitive to light.18,19
Fig. 1 Scheme showing arrangement of ‘active’ and ‘inactive’ amphi-
philes in an idealised Y-type Langmuir–Blodgett film which results in
the film being non-centrosymmetric
1392
J. Mater. Chem., 1998, 8(6), 1391–1397

Results and Discussion
Synthesis and monolayer and multilayer properties of some
non-polymeric amphiphiles containing SHG chromophores
The initial objective was to identify non-polymeric materials
suitable for use in alternating LB films for SHG. These non-
polymeric materials would ideally both (i) form well-ordered
LB films, and (ii) contain excellent SHG chromophores. Four
compounds were investigated: hemicyanines 3 and 4 and
azobenzene derivatives 5 and 6. All four were prepared using
well-known types of synthetic reactions: see Experimental. For
each compound, monolayers were spread at the air–water
interface using chloroform solutions. The water was at 20 °C
and pH 5.2–5.6. The isotherm was recorded. Attempts were
then made to deposit the monolayers at a surface pressure of
30 mN m−1 onto hydrophobic silicon wafers and/or other
supports. X-Ray reflection studies were carried out on the LB
multilayers obtained. The main data obtained from the iso-
therms and X-ray studies are summarised in Table 1.
Fig. 3 Isotherms of azobenzenes 5 (———) and 6 (K) measured
at 20 °C on an aqueous subphase at pH 5.2–5.6. For further details
see Experimental section.
The amphiphile 3 contains an excellent SHG chromophore,
is readily available and as a consequence has been studied
extensively.11,20–29 The isotherm obtained in the present study,
see Fig. 2, is essentially the same as that described in earlier
studies.23 It is evident that as the area available per repeat
unit is decreased, beginning at quite a large surface area (ca.
approximately to the area of the hemicyanine chromophore.23
This suggests, therefore, that the hemicyanine moiety initially
lies flat on the water surface with the long alkyl chain out of
the water. The fact that this hemicyanine does not form well
ordered films prompted the study of hemicyanine 4.
The molecule of hemicyanine 4 is amphiphilic and in this
case the hydrophilic moiety is close to one end. The isotherm
of this compound, shown in Fig. 2, is good, i.e. as the area
available to the monolayer is reduced the isotherm becomes
steep. At this stage, however, the area per repeat unit is
approximately twice that of the cross-sectional area of the
amphiphile, suggesting it may form bilayers. The collapse
pressure is relatively high (>35 mN m−1). The isotherm lacks
the shoulder present in the isotherm of hemicyanine 3. Despite
many attempts, hemicyanine 4 could not be deposited as a
multilayer. The reasons for this are not clear.
˚
160 A2 per repeat unit), the surface pressure rises steadily and
eventually the isotherm becomes relatively steep. The area per
˚
repeat unit at a surface pressure of 30 mN m−1 was 34 A2 and
this corresponds well with the expected cross-sectional area of
the molecule. Overall these results suggest that this amphiphile
does not pack particularly well at the air–water interface. The
monolayers transfer readily onto silicon or Pyrex glass on the
upstrokes but not on the downstrokes. The multilayers
obtained are therefore expected to be Z-type. In X-ray reflection
studies of the multilayers one Bragg reflection was detected
˚
corresponding to a d-spacing of 47 A. This value is very close
˚
to the length of the fully extended molecule, 45 A, consistent
with the multilayer being Z-type. Given that multilayers of,
say, cadmium stearate often display more than ten orders of
sharp Bragg reflections it is clear that LB multilayers of
hemicyanine 3 have only modest order. It is likely that this,
and the modest order in the compressed monolayer, is a
consequence of the most hydrophilic part of the amphiphile
being towards the centre of the molecule. It is interesting to
note here that the isotherm of hemicyanine 3 has a shoulder
Azobenzene derivative 5 contains the 4-N,N-dialkylamino-
4∞-nitroazobenzene SHG chromophore and similar compounds
have been studied before.30,31 Compound 5 gave a good
isotherm, see Fig. 3, and the area per repeat unit at 30 mN m−1
˚
of 22 A2 is close to the cross-sectional area expected for this
chromophore. Unfortunately, the monolayers deposited on
both the upstrokes and downstrokes with deposition ratios of
only 0.70. This is probably a consequence of the modest
hydrophilicity of the nitro group and it prompted the synthesis
of compound 6 in which the hydrophilicity of the nitro group
is reinforced by a carboxylic acid group. It is interesting to
note that the UV spectra of compounds 5 and 6 in chloroform
have maxima at 488 and 489 nm respectively indicating that
for steric reasons the carboxy moiety in compound 6 is not
planar with the aromatic ring. Azobenzene derivative 6 also
gave a good isotherm: see Fig. 3. The monolayers transferred
well on both the upstrokes and downstrokes (deposition ratios
1.00 0.05). In an X-ray reflection study, the multilayers gave
two orders of Bragg reflections corresponding to a d-spacing
˚
at an area per repeat unit of ca. 95 A2 which corresponds
˚
of 48.5 A. The UV spectra of the LB films of compounds 5
and 6 display l at 468 nm, suggesting that in both systems
max
there is some aggregation of the chromophores.
On the basis of the above studies hemicyanine 3 and
azobenzene derivative 6 were selected for the SHG studies
though in both cases the multilayers are not especially well
ordered.
Preparation and properties of alternating multilayers
Fig. 2 Isotherms of hemicyanines 3 (———) and 4 (K) meas-
ured at 20 °C on an aqueous subphase at pH 5.2–5.6. For further
details see Experimental section.
Polymer 2 was used to form the ‘inactive’ layers in the
alternating LB multilayers. This polymer has an excellent
J. Mater. Chem., 1998, 8(6), 1391–1397
1393

1394
J. Mater. Chem., 1998, 8(6), 1391–1397

see Fig. 5, was a straight line up to thicknesses of ca. 60
bilayers, which is equivalent to ca. 0.33 mm. It is evident that
beyond that thickness deposition was less regular. The all-
polymeric SHG films studied previously deposited regularly
up to at least 300 bilayers,8 so the present films appear to be
less satisfactory in this respect.
SHG studies
SHG measurements were taken on the above alternating
multilayers precisely as described in our previous papers.15–17
The incident and detected light waves were linearly polarised
in the plane of incidence and reflection, and a wavelength of
1.064 mm was used: I values were referenced to the signal
from a quartz crystal and are considered accurate to 5%.
2v
The results are shown in Fig. 6 from which it is apparent that
in each system I increases quadratically with the number of
bilayers as predicted by theory up to at least 60 bilayers. The
2v
extent of scatter in the results is a measure of the reproducibility
between samples. Interestingly, the best-fit straight line through
both plots does not pass through zero. This has been found
to be the case with a number of thick multilayers we have
investigated,8,32 and it suggests that the first 10 to 20 layers
deposited pack differently to the subsequent layers.
Fig. 4 Isotherm of polymer 2 measured at 20 °C on an aqueous
subphase at pH 5.2–5.6. For further details see Experimental section.
isotherm, see Fig. 4, and it deposits onto hydrophobic silicon
wafers to give multilayers which display four or five orders of
Bragg peaks.10 As determined by X-ray reflection experiments
Since the SHG signals are close to the longest wavelength
maxima in the UV spectra of the chromophores, I may be
partly resonance enhanced. This and the lack of knowledge of
˚
the bilayer spacing is 45.4 A. This value and FTIR measure-
2v
ments indicate that the films have a Y-type structure with the
alkyl side chains interdigitated,10 i.e. in the general arrangement
shown in Fig. 1 but with both amphiphiles polymeric.
properties such as the local field effects within the layers makes
it difficult to assess accurately the molecular chromophore
hyperpolarisability response b. A study of the angle of incidence
variation of the SHG response can allow an estimate of
Alternating multilayers were prepared using each of com-
pounds 3 and 6 with polymer 2. The ‘active’ compounds were
deposited on the upstrokes and the polymer on the down-
strokes. For each system several separate films containing up
to 70 bilayers were prepared. X-Ray reflectivity studies were
made on examples of each type. In each case one order of
Bragg reflections was observed. Thus, both types of alternating
multilayers appear to be less ordered than those of polymer 2
alone. For the multilayers prepared using compound 3 and
chromophore tilt assuming
a one-dimensional model of
chromophore non-linear response.33 The conventional model
for an LB film is that the chromophore axes are tilted from
the perpendicular to the film with a very narrow angular
distribution. A numerical fit to the incidence angle dependence
of the SHG from these films indicates that using this model
for the alternating multilayers prepared using hemicyanine 3
the chromophores are tilted at an angle of ca. 50° to the
perpendicular and that in the multilayers prepared using
azobenzene derivative 6 the tilt is 30°.
˚
polymer 2 the Bragg peak corresponds to a d-spacing of 46.7 A:
for the multilayers prepared using compound 6 and polymer
˚
2 it corresponds to 46.0 A. These values are only slightly more
The slopes of the plots shown in Fig. 5 correspond to
than that of polymer 2 alone and suggest that in the alternating
films the long side chains are either interdigitated and/or are
greatly tilted from the normal.
effective non-linear susceptibilities x (2) of 17.5 and 9 pm V−1
pp
for the hemicyanine and azobenzene films respectively. The
Further alternating multilayers were deposited on fused
quartz and their UV absorption measured. In each case the
plot of absorbance at the wavelength of the main maximum,
local modification to the optical field in these materials is
difficult to quantify, but is often considered to be of the order
of unity;34 the local field effect might be discernible as an
Fig. 5 Plots of UV-absorbance versus the number of bilayers for Langmuir–Blodgett films prepared from: A, hemicyanine 3 and polymer 2, and
B, azobenzene 6 and polymer 2. For further details see Experimental section.
J. Mater. Chem., 1998, 8(6), 1391–1397
1395

Fig. 6 Plots of the square root of the intensity of the second harmonic light versus the number of bilayers for Langmuir–Blodgett films prepared
from: A, hemicyanine 3 and polymer 2, and B, azobenzene 6 and polymer 2. For further details see Experimental section.
apparent change in the non-linear response of a chromophore.
Without explicit allowance for the effects of local fields, the
results indicate a unidirectional chromophore hyperpolaris-
Conclusions
Alternating LB multilayers have been prepared from hemicyan-
ine 3 and polymer 2 and from azobenzene derivative 6 and
ability b for hemicyanine 3 of 88×10−50 C3 m3 J−2 correspond-
polymer 2. With both types of multilayer the intensity of the
ing to a peak conversion efficiency to the second harmonic of
SHG signal, I , increases essentially with the square of the
number of layers up to a thickness of ca. 60 bilayers: see Fig. 6.
5×10−5 per hemicyanine layer in the film, expressed relative
to the conversion efficiency from the quartz. Estimation of a
hyperpolarisability in this way is considered only accurate to
about 30%, but our estimate corresponds closely to previous
values quoted (~83×10−50 C3 m3 J−2) for hemicyanine mol-
ecules in LB films.20 The corresponding hyperpolarisability for
the azobenzene 6 is estimated to be b=30×10−50 C3 m3 J−2.
This compares with a value of 17×10−50 C3 m3 J−2 deter-
mined35 for Disperse Red I, which has the same chromophore
as compound 5. Thus the presence of the carboxy group in
compound 6 appears to result in a modest increase in the b-
value over that of compound 5.
2v
The slopes of the plots correspond to x(2) values of 17.5 and
9 pm V−1 respectively. Although these properties and their
stabilities over time compare favourably with many other LB
films described in the literature, these films prepared from
‘active’ non-polymeric amphiphiles and a ‘passive’ polymeric
amphiphile are not as satisfactory as the all-polymeric films
described previously.8 Thus, whilst readily available non-poly-
meric amphiphiles are more convenient to use, the derived LB
films are less stable.
We thank EPSRC for financial support.
Stabilities of alternating SHG multilayers
As a check on the stability of the films prepared from hemicyan-
ine 3 and polymer 2 an LB film consisting of 30 bilayers was
kept in the dark18,19 at 20 °C for 12 months. The intensity of
References
1
(a) Organic Materials for Non-L inear Optics, ed. R. A. Hann and
D. Bloor, Royal Society of Chemistry, London, 1989; (b) Organic
Materials for Non-L inear Optics II, ed. R. A. Hann and D. Bloor,
Royal Society of Chemistry, London 1990; (c) Organic Materials
for Non-L inear Optics III, ed. G. Ashwell and D. Bloor, Royal
Society of Chemistry, London, 1993.
the SHG light was then remeasured. I was found to be only
2v
50% of the value when it was freshly prepared. Heating the
film at 50 °C for 72 h caused I to fall to 20% of the original
value. The stability of the film prepared from azobenzene
2v
2
3
4
5
6
Nonlinear Optical Properties of Organic Molecules and Crystals, ed.
D. S. Chemla and J. Zyss, Academic Press, New York, 1987.
P. N. Prasad and D. J. Williams, Introduction to Non-linear Optical
EVects in Molecules and Polymers, John Wiley, New York, 1991.
P. Hodge and N. B. McKeown, in Nonlinear Optical Materials, ed.
R. W. Munn and C. N. Ironside, Blackie, London, 1992, p. 226.
M. J. Grundy, M. R. Musgrove, R. M. Richardson, S. J. Roser and
J. Penfold, L angmuir, 1990, 6, 519.
derivative 6 and polymer 2 was checked similarly. After 12
months at 20 °C it fell to 50% of the original value, and after
heating at 50 °C for 72 h, I fell to zero. In our earlier work8
the multilayers prepared from polymers 1 and 2 showed no
2v
detectable loss of the SHG signal either after being stored at
20 °C for 8 months or after being kept at 60 °C for 96 h. It is
clear that the films studied in the present work are of similar
stability to each other, but that both are significantly less
stable than the all-polymeric films described previously.8 Whilst
this is not entirely surprising, since in a tightly packed molecu-
lar arrangement polymer molecules, being large, will move less
freely than small molecules, the magnitude of the difference is,
in the present context, disappointing.
P. Stroeve, D. D. Saperstein and J. F. Rabolt, J. Chem. Phys., 1990,
92, 6958.
7
8
M. Shimomura, K. Song and J. F. Rabolt, L angmuir, 1992, 8, 887.
P. Hodge, Z. Ali-Adib, D. West and T. King, T hin Solid Films,
1994, 244, 1007.
9
J. P. Cresswell, M. C. Petty, C. H. Wang, B. S. Wherrett, Z. Ali-
Adib, P. Hodge, T. G. Ryan and S. Allen, Opt. Commun., 1995,
115, 271.
It should be noted here that electrically poled polymer films
have been described recently with comparable non-linear
coefficients, of the order of ~10 pm V−1, capable of a stable
response throughout at least 1000 h at 100 °C.36
10 F. Davis, P. Hodge, X. H. Liu and Z. Ali-Adib, Macromolecules,
1994, 27, 1957.
11 I. R. Girling, N. A. Cade, P. V. Kolinksy, J. D. Earls, G. M. Cross
and I. R. Peterson, T hin Solid Films, 1985, 132, 101.
1396
J. Mater. Chem., 1998, 8(6), 1391–1397

12 T. Takahashi and K. Sato, Yakagaku Zasshi, 1958, 78, 467.
13 Vogel’s T extbook of Practical Organic Chemistry, Longmans,
London, 1978, 4th edn., p. 687–720.
14 F. Davis, P. Hodge, C. R. Towns and Z. Ali-Adib, Macromolecules,
1991, 24, 5695.
15 P. Hodge, Z. Ali-Adib, D. West and T. King, Macromolecules, 1993,
26, 1789.
16 R. H. Tredgold, S. D. Evans, P. Hodge and A. Hoorfar, T hin Solid
Films, 1988, 160, 99.
17 M. Bishop, J. H. R. Clarke, L. E. Davis, T. A. King, F. R. Mayers,
A. Mohebati, R. W. Munn, M. M. Shabat, D. West and
J. O. Williams, T hin Solid Films, 1992, 210/211, 185.
18 F. H. Quina and D. G. Whitten, J. Am. Chem. Soc., 1975, 97, 1602.
19 P. Hodge, Z. Ali-Adib, D. West, A. Mohebati and T. King, in
ref. 1(c), pp. 61–67.
20 I. R. Girling, N. A. Cade, P. V. Kolinsky, R. J. Jones, I. R. Peterson,
M. M. Ahmad, D. B. Neal, M. C. Petty, G. G. Roberts and
W. J. Feast, J. Opt. Soc. Am. B., 1987, 4, 950.
21 J. S. Schildkraut, T. L. Penner, C. S. Willand and A. Ulman, Opt.
L ett., 1988, 13, 134.
22 P. Stroeve, D. D. Saperstein and J. F. Rabolt, T hin Solid Films,
1989, 179, 529.
25 P. Stroeve, D. D. Saperstein and J. F. Rabolt, J. Chem. Phys., 1990,
92, 6958.
26 J. Y. Fang, S. J. Xiao, Z. H. Lu, Y. Wei, Z. M. Sun and P. Stroeve,
Solid State Commun., 1991, 79, 985.
27 X. Liu, L. Liu, Z. Chen, X. Lu, J. Zheng and W. Wang, T hin Solid
Films, 1992, 219, 221.
28 W. F. A. Su, T. Kurata, H. Nobutoki and H. Koezuka, L angmuir,
1992, 8, 915.
29 M. A. Carpenter, C. S. Willand, T. L. Penner, D. J. Williams and
S. Mukamel, J. Phys. Chem., 1992, 96, 2801.
30 D. W. Kalina and S. G. Grubbs, T hin Solid Films , 1988, 160,
373.
31 I. Ledoux, D. Josse, P. Vidakovic, J. Zyss, R. A. Hann,
P. F. Gordon, B. D. Bothwell, S. K. Gupta, S. Allen, P. Robin,
E. Chastaing and J. C. Dubois, Europhys. L ett., 1987, 3, 803.
32 D. West, D. Dunne, P. Hodge, N. B. McKeown and Z. Ali-Adib,
T hin Solid Films, in press.
33 S. Allen, Inst. Phys. Conf. Ser., 1989, 103, 163.
34 N. Wiser, Phys. Rev., 1963, 179, 62.
35 K. D. Singer and J. H. Andrews, in Molecular Nonlinear Optics:
Materials, Phys. and Devices, ed. J. Zyss, Academic Press, London,
1994, ch. 6, p. 252.
23 M. C. J. Young, R. Jones, R. H. Tredgold, W. X. Lu, Z. Ali-Adib,
P. Hodge and F. Abbasi, T hin Solid Films, 1989, 182, 319.
24 H. Ancelin, G. Briody, J. Yarwood, J. P. Lloyd, M. C. Petty,
M. M. Ahmad and W. J. Feast, L angmuir, 1990, 6, 172.
36 Y. Shi, W. Wang, W. Lin, D. J. Olsen and J. H. Bechtel, Appl. Phys.
L ett., 1997, 70, 1342.
Paper 7/08854J; Received 9th December 1997
J. Mater. Chem., 1998, 8(6), 1391–1397
1397