Semiconducting Langmuir-Blodgett films of ethylenedithiotetrathiafulvalene (EDT-TTF) derivatives bearing charged and uncharged aromatic substituents

Article abstract of DOI:10.1039/a606831f

Ethylenedithiotetrathiafulvalene (EDT-TTF) derivatives 1-4 functionalised with a single aromatic ring have been synthesised and their Langmuir-Blodgett (LB) films have been assembled utilising only 25% molar ratio of fatty acid. For compounds 1, 3 and 4, predominantly Y-type deposition onto solid supports was observed with a transfer ratio close to unity. After doping with iodine vapour, the maximum in-plane conductivity values obtained were σ_rt= 10^-3 S cm^-1 for 1 and 3, and 10^-5 S cm^-1 for 4. LB deposition of 2 was not uniform and the conductivity value after doping was low. UV-VIS spectra of the LB films reveal the appearance of a charge-transfer (CT) band at λ_max = ca. 900 nm for 1 and 4 after iodine doping. A solution of compound 3 exhibited a weak absorption band at ca. 665 nm which is assigned to an intramolecular CT band; the intensity of this band increases on exposure of the solution to light. This band is not observed in LB films of 3, neither as-deposited nor after doping. Molecular orbital calculations indicate that in the minimum energy conformation of 3, the pyridinium moiety is practically orthogonal to the TTF unit and this conformation may be obtained in solution, enabling charge-transfer to occur. A more linear conformation of 3 in the LB films may prevent intramolecular charge-transfer from occurring. Monolayers of 1,3 and 4 were characterised by cyclic voltammetry which revealed two redox steps consistent with the formation of the EDT-TTF cation radical and dication, respectively.

Full text of DOI:10.1039/a606831f

Semiconducting Langmuir–Blodgett films of ethylenedithiotetrathiafulvalene
EDT–TTF) derivatives bearing charged and uncharged aromatic substituents
(
Leonid M. Goldenberg,a,b,c James Y. Becker,*d Ofra Paz-Tal Levi,d Vladimir Yu. Khodorkovsky,*d Lev M. Shapiro,d
Martin R. Bryce,*b John P. Cresswellc and Michael C. Petty*c
aInstitute of Chemical Physics in Chernogolovka, Russian Academy of Science, Chernogolovka, 142432 Moscow region,
Russia
bDepartment of Chemistry and Centre for Molecular Electronics, University of Durham, Durham, UK DH1 3L E
cSchool of Engineering and Centre for Molecular Electronics, University of Durham, Durham, UK DH1 3L E
dDepartment of Chemistry, Ben-Gurion University of the Negev, Beer-Sheva 84105, Israel
Ethylenedithiotetrathiafulvalene (EDT–TTF) derivatives 1–4 functionalised with a single aromatic ring have been synthesised and
their Langmuir–Blodgett (LB) films have been assembled utilising only 25% molar ratio of fatty acid. For compounds 1, 3 and 4,
predominantly Y-type deposition onto solid supports was observed with a transfer ratio close to unity. After doping with iodine
vapour, the maximum in-plane conductivity values obtained were s =10−3 S cm−1 for 1 and 3, and 10−5 S cm−1 for 4. LB
rt
deposition of 2 was not uniform and the conductivity value after doping was low. UV–VIS spectra of the LB films reveal the
appearance of a charge-transfer (CT) band at l =ca. 900 nm for 1 and 4 after iodine doping. A solution of compound 3
max
exhibited a weak absorption band at ca. 665 nm which is assigned to an intramolecular CT band; the intensity of this band
increases on exposure of the solution to light. This band is not observed in LB films of 3, neither as-deposited nor after doping.
Molecular orbital calculations indicate that in the minimum energy conformation of 3, the pyridinium moiety is practically
orthogonal to the TTF unit and this conformation may be obtained in solution, enabling charge-transfer to occur. A more linear
conformation of 3 in the LB films may prevent intramolecular charge-transfer from occurring. Monolayers of 1, 3 and 4 were
characterised by cyclic voltammetry which revealed two redox steps consistent with the formation of the EDT–TTF cation radical
and dication, respectively.
Within the field of molecular conductors,1 bis(ethylenedithio)-
tetrathiafulvalene (BEDT–TTF) derivatives command a central
position as many of their crystalline cation–radical salts are
metals or superconductors.2 Amphiphilic analogues which
form Langmuir–Blodgett (LB) films have been studied by a
number of groups3–19 and some multilayer LB films display
conductivity values as high as 1 S cm−1 after doping with
iodine vapour.18 Amphiphilic derivatives of the closely related
donor ethylenedithiotetrathiafulvalene (EDT–TTF) substituted
with one hydrophobic CH OC(O)C H chain were recently
reported to form LB films was tetrakis(benzylthio)–TTF
[(PhCH S) TTF]. This compound did not produce a stable
2
4
monolayer at the air–water interface unless at least a 50%
molar ratio of a fatty acid was added.22 Subsequently a non-
amphiphilic bis(EDT–TTF) derivative was shown to make a
semiconducting LB film after iodine doping, without added
fatty acid.23 Unsubstituted BEDT–TTF, which is non-amphi-
philic, forms conducting aggregates if mixed with 50% C or
6
0
stearic acid.24 Furthermore, it is noteworthy, that compound
3 contains both donor (TTF) and acceptor (pyridinium) moiet-
ies covalently linked via a non-conjugated spacer: as both these
moieties should be almost orthogonal (see Fig. 6) no substantial
intramolecular donor–acceptor (DSA) electronic interaction
would be expected. Compound 3 is analogous to the
amphiphilic D-s-A and D-p-A systems, based on TCNQ and
aromatic donors, studied by Metzger and co-workers,25,26 and
Ashwell and co-workers,27–29 respectively. This compound
may, therefore, be relevant to studies on molecular rectification
and/or second-order nonlinear optics.
2
17 35
shown to form semiconducting LB films.20
Experimental
Syntheses
Compounds 1–4 were obtained as racemic mixtures. J Values
in Hz.
In the present work, we report the study of LB films of three
novel non-amphiphilic derivatives of EDT–TTF, compounds
4,5-Bis(methylthio)-4∞,5∞-(4-pyridylethylenedithio)tetrathia-
fulvalene.† Compound 1 was prepared by cross-coupling of
4,5-bis(methylthio)-1,3-dithiole-2-thione30 (3 equiv.) and 4,5-
(4-pyridyl)ethylenedithio-1,3-dithiole-2-one (1 equiv.) (see
1
, 3 and 4, and the amphiphilic derivative 2. All of these
molecules contain a single aromatic substituent, which could
serve as the ‘traditional’ hydrophobic tail in the case of
compounds 1 and 4, or as a hydrophilic head group in the
case of compound 3. Compounds 1 and 3 have been discussed
recently by us in a preliminary communication.21 Prior to this
work, the only non-amphiphilic TTF derivative that was
†
IUPAC
name:
2-[4,5-bis(methylthio)-1,3-dithiol-2-ylidene]-5,6-
dihydro-5-(4-pyridyl)-1,3-dithiolo[4,5-b][1,4]-dithiine. Related com-
pounds 2–4 can be named similarly.
J. Mater. Chem., 1997, 7(6), 901–907
901

below), in neat trimethyl phosphite at reflux under nitrogen.
After chromatographic separation from self-coupled products,
compound 1 was isolated as yellow crystals in 22% yield, mp
chloroform solution of TA to obtain ca. 25% molar of TA.
The optimal dipping pressure for this mixture was found to
be 35–40 mN m−1.
1
58–159 °C (from acetonitrile). d (CDCl ) 2.42 (s, 6H), 3.41
LB films were deposited onto glass slides, quartz windows,
single crystal silicon, conducting indium tin oxide (ITO, sheet
resistance 300 ohms per square, from Balzers) glass slides and
Ag-coated glass slides by the conventional vertical dipping
technique. Unless specified otherwise, a dipping speed of 10 mm
min−1 was employed and the first monolayer was deposited
on the upstroke with the slide immersed in the subphase before
compression of the monolayer. To improve the hydrophilic
properties of ITO, the slides were treated with saturated
Na Cr O –concentrated H SO solution for ca. 10 s and care-
H
3
(
d, J 5.4, 2H), 4.72 (t, J 5.4, 1H), 8.29 (d, J 5.7, 2H), 8.61 (d, J
5
3
.7, 2H). Calc. C H NS : C, 38.85; H, 2.83%. Found: C,
1
5 13
8
9.02; H, 2.72%.
4
,5-(4-Pyridylethylenedithio)-1,3-dithiole-2-thione. This was
prepared according to a modified literature procedure31 using
4
-vinylpyridine as the alkene, and was obtained as yellow
crystals in 60% yield, mp 125–126 °C (from acetonitrile). Calc.
C H NS : C, 39.84; H, 2.34%. Found: C, 40.57; H, 2.41%.
1
0 7
5
2
2 7
2
4
fully washed with ultrapure water.36 Substrates with areas of
between 20 and 30 cm2 were used for LB transfer. After LB
film deposition, these were carefully cut with a diamond tipped
stylus to form several electrodes with contact areas between
0.1 and 0.5 cm2.
4
,5-(4-Pyridylethylenedithio)-1,3-dithiole-2-one. This was
synthesised by reaction of its respective 2-thione derivative
with mercury() acetate, in chloroform–acetic acid under
reflux, according to the known procedure31 and isolated
as yellow crystals in 68% yield, mp 129–130 °C. Calc.
C H NOS : C, 42.08; H, 2.47%. Found: C, 41.86; H, 2.17%.
Doping of the LB films was carried out by exposure to
iodine vapour for a given time in a sealed vessel. UV–VIS
absorption measurements were performed using a Perkin-
Elmer Lambda 19 spectrophotometer for films deposited on
glass slides and quartz windows. The film thickness was
measured using both a surface profiling Tencor Instruments
1
0 7
4
4
,5-Bis(hexadecylthio)-4∞,5∞-(4-pyridylethylene)dithiotetra-
thiafulvalene 2. Compound 2 was prepared from 4,5-bis(hex-
adecylthio)-1,3-dithiole-2-thione30,32 by an analogous pro-
cedure used for the synthesis of derivative 1 and isolated as a
red solid in 28% yield, mp 58–59 °C. Calc. C H NS : C,
Alpha-Step 200 (stylus force=11±1 mg) and
a
Rudolf
Research Auto EL IV ellipsometer (operating wavelength=
45 73
8
6
33 nm). LB films were deposited onto single crystal silicon
6
1.10; H, 8.32%. Found: C, 61.36; H, 8.41%.
for these experiments. For the Alpha-Step measurements, a
layer of aluminium (thickness ca. 200 nm) was evaporated over
the step between the organic layer and the uncoated substrate
and between two organic layers of diÂerent thicknesses. DC
conductivity measurements were made in air by a standard
two-contact method using silver paste contacts. The conduc-
tivity was calculated on the basis of the measured single layer
thickness of ca. 5 nm. Current–voltage characteristics were
ohmic, and by varying the distance between the electrodes, it
was established that the contact resistance was negligible.
Single crystal DC conductivity was measured by a four-probe
method. Cyclic voltammetry was performed using a EG&G
PARC, model 273 potentiostat with an Advanced Bryans XY
recorder. Pt mesh served as the counter electrode. All potentials
were measured versus a Ag/AgCl reference electrode and were
tested, if necessary, with a ferrocene/ferrocinium couple as
internal reference. Cyclic voltammetry in solution was per-
formed in 0.2  Bu NPF /CH Cl solution on a Pt disk [0.13
4
,5-Bis(methylthio)-4∞,5∞-(4-methylpyridinioethylenedithio)-
tetrathiafulvalene iodide 3. Compound 3 was formed by
refluxing compound 1 with methyl iodide (excess) in acetone
in the dark for 0.5 h, and isolated in 96% yield as red crystals,
mp 160–170 °C (slow decomposition). Calc. C H INS : C,
16 16
8
3
1.73; H, 2.66%. Found: C, 32.00; H, 2.48%.
4
,5-Bis(methylthio)-4∞,5∞-(4-phenylethylenedithio)tetrathia-
fulvalene 4. Compound 4 was synthesised according to the
general procedure33 starting from 4,5-(phenylethylenedithio)-
1
,3-dithiole-2-thione and 4,5-bis(methylthio)-1,3-dithiole-2-
thione, and isolated in 60% yield as orange crystals, mp
1
4
3
09.7–110.2 °C. Calc. C H S : C, 41.52; H, 3.05%. Found: C,
1
6 14 8
1.39; H, 3.12%. d (CDCl ) 2.44 (s, 6H), 3.32 (dd, ABX, J
H
3
ax
.6, J 12, 1H), 3.42 (dd, ABX, J 8, J 12, 1H), 4.62 (dd,
ab
bx
ab
ABX, J 3.6, J 8, 1H), 7.31 (s, 5H); m/z (CI) 464.2 (M+).
ax
bx
4
6
2
2
or 0.5 mm diameter (both homemade) or 1.6 mm diameter
4
,5-(Phenylethylenedithio)-1,3-dithiole-2-thione. This was
(
Bioanalytical System Inc.)] working electrode, employing IR
prepared as follows. A mixture of oligomeric 1,3-dithiole-2,4,5-
trithione34 (0.59 g, 3 mmol) and styrene (0.31 g, 3 mmol) in
toluene (30 ml) was refluxed for 40 min, evaporated, and the
residue was recrystallised from acetonitrile with addition of
activated carbon to yield the product as shiny yellow crystals
compensation. LiClO (Fluka, microselect), Bu NPF (Fluka,
4
4
6
electrochemical grade), KCl (Fluka, microselect), HClO
4
(
Aldrich, ACS reagent), CH Cl (Aldrich, HPLC, distilled from
2
2
P O ) and ultrapure water were used for the preparation of
2
5
electrolyte solutions. The Pockels electrooptic eÂect was meas-
ured to test for second-order optical nonlinearity (by monitor-
ing reflectivity) in monolayers deposited on Ag-coated glass
using the technique of surface plasmon resonance at a
wavelength of 633 nm.37
(
0.51g, 60%), mp 77–78 °C. d (CDCl ) 3.40 (dd, ABX, J 3.0,
H
3
ax
J
13.2, 1H), 3.60 (dd, ABX, J 9.6, J 13.2, 1H), 4.7 (dd,
ab
bx
ab
ABX, J 3.0, J 9.6, 1H), 7.36 (s, 5H). Calc. C H S : C, 43.97;
ax
bx
11 8 5
H, 2.68%. Found: C, 44.18; H, 2.59%.
LB Film formation and characterisation
Tricosanoic acid [TA, Me(CH ) CO H] (Sigma, 99%) was
Results and Discussion
2
21
2
used as received. The Durham LB troughs were housed in a
class 10 000 microelectronics clean room and have been
described previously.35 Compounds 1–4 were spread on the
surface of ultrapure water (obtained by reverse osmosis, deion-
isation and ultraviolet sterilisation) from CH Cl or tetrahydro-
Synthesis of EDT–TTF derivatives
The synthesis of the EDT–TTF systems 1, 2 and 4 was
accomplished by standard cross-coupling23 of the two 1,3-
dithiol-2-one (or 1,3-dithole-2-thione) half-units in the presence
of trimethyl phosphite. N-Methylation of compound 1 was
readily accomplished with methyl iodide in refluxing acetone
to form compound 3. However, attempted methylation of
analogue 2 was not a clean reaction, possibly due to the
influence of the two long alkyl chains: more forcing conditions
were required for reaction to occur, and a complicated mixture
2
2
furan (THF) solutions (0.5 g l−1) or acetone–CH Cl mixtures
2
2
(
352 or 253, v/v). The surface pressure versus molecular area
isotherms were measured at 20±2°C; pH=5.8±0.2, and com-
pression rate ca. 4×10−3 nm2 molecule−1 s−1. A mixture of 1,
3
and 4 with TA was also used. The optimal mixture was made
from a THF solution of compounds 1, 3 and 4, and a
9
02
J. Mater. Chem., 1997, 7(6), 901–907

of decomposition products was obtained, possibly arising from
radical reactions. We did not pursue this alkylation further, as
the focus of this study was the LB films of derivatives without
long alkyl chains.
film) spread from THF solution, it was possible to observe a
condensed isotherm with a limiting average molecular area of
ca. 0.40 nm2. The compressed monolayer was stable under
these conditions.
Amphiphilic analogue 2 showed an isotherm of surface
pressure versus molecular area consisting of two condensed
regions (Table 1). As this compound did not show good
deposition characteristics and the LB film conductivity was
low (Table 2) we did not study it any further. In contrast to 1,
compound 3 tended to aggregate on the water surface when
spread without TA, aÂording a small molecular area (Table 1).
However, using the same conditions as for 1, with added fatty
acid, a stable floating monolayer with a limiting average
molecular area of 0.23 nm2 was obtained. Compound 3 was
also spread from acetone–CH Cl mixture (352 or 253, v/v).
Monolayer behaviour at the air–water interface and LB film
transfer
Compound 1 spread on a water surface from dichloromethane
or tetrahydrofuran (THF) solution exhibited an isotherm of
surface pressure versus molecular area consisting of two con-
densed regions [Fig. 1(a)]. The extrapolated limiting molecular
areas for the first and second region were ca. 0.30 and 0.14 nm2,
respectively. It is notable that the molecular area in the first
condensed region is nearly double that in the second region,
which may suggest formation of a bilayer, or molecular reor-
ganisation may occur in the latter region. Fig. 1 also shows
the change in the isotherm with time, in the expanded state.
The final curve [Fig. 1(c)], obtained 2 h after spreading, was
steep with a limiting molecular area of ca. 0.22 nm2. A striking
feature of the isotherms in Fig. 1 is their intersection through
one ‘isosbestic’ point, a feature which has not, to our knowl-
edge, been reported before for monolayers. The shape of the
isotherms depended upon the quantity of material spread on
the surface; the limiting molecular area decreased with increas-
ing amount of material, which is indicative of aggregation
2
2
Both pure compound 3 and its mixture with 25% of TA
showed condensed isotherms (Table 1) with extrapolated aver-
age molecular areas of ca. 0.15 and 0.20 nm2, respectively. In
contrast to compounds 1 and 3, compound 4 spread either
from CH Cl or THF (with addition of 25% of TA) exhibited
2
2
small molecular areas (Table 1) probably indicating multilayer
formation on the subphase surface.
The transfer of compounds 1, 3 and 4 (mixed with 25% TA)
on to solid substrates was achieved by controlling the
Langmuir film at 35–40 mNm−1. Predominantly Y-type
deposition was observed with a transfer ratio of 1±0.05. The
reproducibility of LB transfer is shown for compound 1 in
Fig. 2. A linear plot of the optical density of the absorption
band at 400 nm versus the number of layers was observed.
Film thickness determination by ellipsometry and Alpha-Step
[
Fig. 1(d)]. The films of compound 1 were transferred onto
glass slides predominantly by Z-type deposition, with a transfer
ratio of 0.7±0.1. The LB films were not uniform; they tended
to cover only a narrow band near the upper edge of the
substrate.
The LB behaviour of many poor amphiphiles can be
improved by the addition of fatty acid. For LB films of 1, TA
was used, and with only 25% molar of TA (the experimentally
determined minimum amount required to produce a stable
Table 2 In-plane DC, room temp. conductivity of LB films of
compounds 1–4
compound
1
2
3
4
as deposited/S cm−1
10−6
10−3
10−8
10−6
10−6
10−3
10−6
10−5
after I2 doping/S cm−1
Fig. 1 Surface pressure versus area per molecule isotherm for com-
pound 1 measured after leaving the film on the subphase surface for
diÂerent lengths of time in its uncompressed state (0.12 mg from
CH Cl solution): (a) 10 min after spreading, (b) 1 h after spreading, (c)
Fig. 2 Plot of optical density of the adsorbance band at 400 nm versus
number of layers for as-deposited LB films of compound 1 with 25%
of TA (predominantly Y-type deposition)
2
2
2
h after spreading and (d) 0.24 mg from CH Cl solution, 10 min
2
2
after spreading
Table 1 Extrapolated average molecular areas (nm2)a for compounds 1–4, pure and mixed with 25 mol% TA
compound 1
0.14, 0.30b
compound 2
0.25, 0.50
compound 3
compound 4
pure compound from CH Cl
2
2
from CH Cl with 25% TA
0.09
0.10
2
2
pure compound from THF
0.10, 0.20
0.05
0.23
0.15
0.20
from THF with 25% TA
0.40
pure compound from acetone–CH Cl (253, v/v)
2
2
from acetone–CH Cl (253, v/v) with 25% TA
2
2
aIn the case of films mixed with TA, the area given is that of the active compound+TA (i.e. average area). bfor the change of isotherm with time
see Fig. 1.
J. Mater. Chem., 1997, 7(6), 901–907
903

measurements (Fig. 3) also confirmed reproducible deposition.
However, the single layer thickness of ca. 5 nm obtained from
this plot (Fig. 3) is greater than would be expected: an esti-
mation of the length of molecule 1 by HYPERCHEM gave a
value of 1.35–1.5 nm, depending on the conformation, whereas
the length of TA is ca. 3 nm. This indicates that the floating
film is probably more than one molecule in thickness.
The transfer of monolayers of compound 2 onto glass was
achieved by controlling the floating film at 25–30 mNm−1.
However, as noted above, deposition was not uniform and the
conductivity value was low (Table 2). We suggest, therefore,
that within this family of materials the presence of two alkyl
chains inhibits the formation of conducting films.
LB Film characterisation
Conductivity measurements. Lateral conductivity was meas-
ured for films freshly deposited and after doping with iodine
vapour, and the values obtained are presented in Table 2. The
conductivity of the LB films of compounds 1 and 3 increased
upon doping and remained stable within the range s =
rt
1
0−3–10−4 S cm−1 for at least one month after doping. DC
conductivity for compound 3 was also measured on single
crystals and showed values of 1×10−8 S cm−1 in the dark and
5
×10−8 S cm−1 in the light. It has been observed that com-
pound 3 is photochemically unstable (see the discussion below).
Therefore, we tested the influence of this photochemical reac-
tion on the formation and properties of LB films of 3.
Compound 3 was synthesised in the dark and then all manipu-
lations for LB transfer were undertaken in the dark. For
comparison, either the compressed monolayer on the air–water
interface, or the LB film deposited on a glass slide, or the LB
film after iodine doping, were illuminated with visible light for
Fig. 4 UV–VIS Spectra of LB films of (a) compound 1 (42 layers),
(
(
b) compound 3 (22 layers) and (c) compound 4 (38 layers) mixed
iii) 1 h after doping
1
0 h. Illumination of the spreading solution with visible light
with 25% of TA: (i) as-deposited; (ii) immediately after iodine doping;
for 10 h caused the solution of compound 3 in THF to turn
from red–brown to brown; however, this process did not
significantly change the limiting molecular area. We have built
up LB films using this illuminated solution. The most signifi-
cant changes in conductivity (a reduction of up to two orders
of magnitude) were found for doped films built-up from an
illuminated monolayer, or using the illuminated spreading
solution. The influence of light on films that had been deposited
in the dark was not found to be significant (for either
as-deposited or doped layers).
band decreased in air for compound 4 [Fig. 4(c), curve (iii)].
These spectra are qualitatively similar to those of LB films
amphiphilic TTF,19 EDT–TTF20 and BEDT–TTF18 deriva-
tives studied previously. Surprisingly, a similar band could not
be clearly detected for the LB films of compound 3 [Fig. 4(b)]
nor was it visible in a doped, cast film of 3 on glass with a
thickness corresponding to 15–20 layers of LB film.
It is known that the pyridinium cation is an extremely
strong acceptor with an electron aÃnity of ca. 5 eV38 and that
tetrathiafulvalene (TTF) interacts with acceptors which possess
an electron aÃnity higher than 2.5 eV.39 We have performed
AM1 geometry optimisation by HYPERCHEM software for
compound 3. The molecular conformation obtained by
calculation is shown in Plate 1. The results suggest that the
UV–VIS Spectroscopy. Doping of the LB films with iodine
vapour was monitored by visible spectroscopy, which showed
the appearance of a charge-transfer (CT) band at ca. 900 nm
for compounds 1 and 4 [Figs. 4(a) and 4(c), respectively]. This
band remained after storing the films for several hours in air
for compound 1 [Fig. 4(a), curve (iii)]. The intensity of the CT
Fig. 3 Plot of film thickness obtained by ellipsometry (+) and Alpha-
Step (#) for LB films of compound 1 mixed with 25% of TA versus
number of layers
Plate 1 Geometry optimisation for compound 3 obtained by MNDO
(AM1) methods using HYPERCHEM programs (blue=carbon,
yellow=sulfur, red=nitrogen, white=hydrogen)
9
04
J. Mater. Chem., 1997, 7(6), 901–907

pyridinium moiety is practically orthogonal to the TTF unit
and, therefore, the substituent should not influence the oxi-
dation potential that is observed by solution electrochemistry
(discussed below). In solution, simple pyridinium salts do not
react with TTF because of strong solvation eÂects. Therefore,
when equimolar amounts of N-methylpyridinium iodide and
TTF were mixed in acetone solution, a CT band was not
observed in the visible spectrum. However the situation is
diÂerent with compound 3, where the pyridinium moiety is
covalently linked to the TTF unit, which, due to the close
proximity of the electroactive groups, makes charge transfer
possible.‡ Indeed, a solution of derivative 3 in dry acetone,
freshly prepared in the dark, exhibited a weak absorption at
ca. 665 nm (Fig. 5), the extinction coeÃcient of which was
independent of concentration (within experimental error). This
band can, therefore, be attributed to an intramolecular charge
transfer band. This solution is quite stable in the absence of
light, but even the weak spectrophotometer beam during
scanning between 800 and 350 nm triggered the electron trans-
fer from the electron donating moiety (TTF) to the electron
accepting moiety (N-methylpyridinium) as evidenced by the
appearance of the strong band at ca. 675 nm, which is the
characteristic signature of the TTF cation radical40 (Fig. 5).
Formally, this process should aÂord a molecule consisting of
the cation radical located on the TTF moiety, and the unstable
N-methylpyridinium radical moiety. When a solution which
had been prepared in the dark and used for recording the
UV–VIS spectra was stored in the dark, the band at 675 nm
reached its maximum extinction coeÃcient after 8–10 h (the
solution gave a strong, poorly resolved EPR signal at this
stage) and this absorption band almost completely disappeared
within 2 d. This process proceeds considerably faster in light
Fig. 5 Visible spectra for compound 3 in acetone (0.08 g l−1): (a)
freshly prepared solution, and consecutive runs after (b) 1, (c) 2 and
(d) 3 h, respectively
of the long wave transition (which is essentially a charge-
transfer D–A interaction) is close (but not equal) to zero,
owing to the symmetry mismatch. This point is also supported
by calculations. After the electron transfer has occurred, the
strong absorption typical of a TTF radical cation should be
observed. The triggering process for the electron transfer is
not clear at present and requires further investigation. It
should be noted that the isolation of a system comprising both
donor and acceptor moieties with the ionisation potential
of the donor (ca. 6.8 eV) higher than the electron aÃnity
of the acceptor (ca. 5.4 eV) is unprecedented to the best of our
knowledge.
The calculated minimum energy conformation of 3 should
be attained in the gas phase or solution where the molecules
are free to adopt their optimum conformation, but this may
not be the case in the LB films where they are forced to adopt
a more linear conformation. In a linear conformation of
compound 3, the positive charge on the pyridinium cation will
be at a greater distance from the conjugated TTF system,
which could prevent charge transfer in the LB films. It is not
completely clear whether the absence of the absorption at 675
in the spectra of LB films of derivative 3 is the result of: (i)
the decomposition of the product of electron transfer; (ii) the
fact that electron transfer cannot occur in an LB film due to
the specific packing of the molecules, or (iii) the eÂective
solvation of the pyridinium moiety.
or in more polar solvents (e.g. Me SO, or in the presence of
2
water). The solution, after disappearence of the 675 nm band,
contained a complex mixture of unidentified products (TLC
evidence) presumably resulting from further reactions of the
N-methylpyridinium radical. Therefore, we prepared a solution
of compound 3 in acetone, observed the charge-transfer band
in the solution spectrum and then built up LB films using a
mixture of this solution plus TA in CH Cl . As above, we did
2
2
not observe a CT band in these films either as-deposited or
after iodine doping. The limiting molecular areas observed for
compound 3 (Table 1) are close to the cross-section of an alkyl
chain and indicate a very compact structure. It is possible that
the molecules of compound 3 in a Langmuir layer adopt a
linear conformation to maximise intermolecular packing forces.
The calculated energy of the HOMO for compound 3,
located predominantly on the TTF moiety and the four
attached sulfur atoms, is −8.99 eV. This value is considerably
higher then the experimental ionisation potential of TTF
Electrochemistry. The electrochemistry of these materials in
solution and in LB films has been studied by cyclic voltamme-
try, and the data are collated in Table 3. All the compounds
displayed essentially the same redox properties in solution, viz.
two reversible single-electron oxidation waves at potentials
typical of the EDT–TTF ring system,20 which indicates that
the aromatic substituents, even in the case of pyridinium
compound 3 (in agreement with MNDO calculations), do not
interact with the EDT–TTF redox system. Monolayer LB films
were characterised by CV when deposited onto ITO glass
electrodes (Fig. 6). Two redox steps have been observed for
the monolayer films of compounds 1, 3 and 4 (in each case
mixed with TA). However, the redox peaks for compound 3
were very broad and ill-developed [Fig. 6(b)]. Changing the
(
6.70–6.92 eV)41 but all MNDO-type methods, according to
our experience, underestimate the donating ability of sulfur-
containing heterocycles. The calculated energy of the LUMO
of 3, located mostly on the pyridinium moiety (and to a lesser
extent on both sulfur atoms of the ethylenedithio group) is
−
5.33 eV, in good agreement with the reported data.38 Before
intramolecular electron transfer occurs, the oscillator strength
‡
Note added at proof: After submission of our manuscript, other
workers reported the synthesis of covalently linked TTF–bipyridinium
electrolyte (KCl or HClO ) did not improve the CV response,
derivatives which display intramolecular charge-transfer in solution
4
(
K. B. Simonsen, K. Zong, R. D. Rogers, M. P. Cava and J. Becher,
and Fig. 6(b) presents the best response obtained. The peak
J. Org. Chem., 1997, 62, 679).
currents decreased in subsequent scans as shown in Fig. 6.
Table 3 Redox potentials for compounds 1–4 (versus Ag/AgCl, V)
compound 1
compound 2
compound 3
compound 4
solution
0.45; 0.8
0.44; 0.82
0.45; 0.8
0.45; 0.8
LB monolayer (mixed with TA)
0.45; 0.55
0.55; 0.8
0.4; 0.7
J. Mater. Chem., 1997, 7(6), 901–907
905

University of Durham. J.Y.B. is thankful to the Israel Academy
of Science and the Royal Society for a Fellowship for senior
scientists. M.R.B. thanks the University of Durham for a Sir
Derman Christopherson Research Fellowship. V.Yu.K. thanks
the RASHI Foundation (Israel) for support. J.P.C. thanks the
EPSRC for financial support. We are grateful to Dr C. Pearson
for the Alpha-Step measurements.
References
1
2
3
J. Mater. Chem., Special Issue on Molecular Conductors, 1995, 5
(
10), 1469.
T. Ishiguro and K. Yamaji, Organic Superconductors, Springer,
Berlin, 1990.
J. Richard, M. Vandevyver, A. Barraud, J. P. Morand,
R. Lapouyade, P. Delhaes, J. F. Jacquinot and M. Roulliay,
J. Chem. Soc., Chem. Commun., 1988, 754.
Fig. 6 Cyclic voltammograms of LB monolayers of a mixture com-
pounds 1, 3 and 4 with 25% of TA on an ITO electrode, 0.4  LiClO
4
5
J. P. Morand, R. Lapouyade, P. Delhaes, M. Vandevyver,
J. Richard and A. Barraud, Synth. Met., 1988, 27, B569.
A. Otsuka, G. Saito, T. Nakamura, M. Matsumoto, Y. Kawabata,
K. Honda, M. Goto and M. Kurahashi, Synth. Met., 1988, 27,
B575.
4
solution, versus Ag/AgCl: (a) compound 1, scan rate 100 mV s−1; (b)
compound 3, scan rate 200 mV s−1, (c) compound 4, scan rate
5
0 mV s−1. Solid line, first cycle; dashed line, second cycle.
6
C. Lalanne, P. Delhaes, E. Dupart, C. Garrigou-Lagrange,
J. Amiell, J. P. Morand and B. Desbat, T hin Solid Films, 1989,
Table 4 Pockels eÂect measurements
1
79, 171.
7
8
9
C. Dourthe, M. Izumi, C. Garrigou-Lagrange, T. BuÂeteau,
B. Desbat and P. Delhaes, J. Phys. Chem., 1992, 96, 2812.
M. Vandevyver, T hin Solid Films, 1992, 210–11, 240.
V. I. Troitsky, V. Yu. Khodorkovsky and O. Ya. Neilands, T hin
Solid Films, 1992, 210–11, 317.
thickness/
nm
x(2)(−v;v,0)/
pm V−1a
r/pm
V−1b
compound
1
3
(with TA)
1.2
3.4
1.0
(with TA)
4.3
0.49
0.16
1
1
0
1
G. V. Tormos, M. P. Cava, X-L. Wu, A. J. McKinley and
R. M. Metzger, Synth. Met., 1992, 52, 131.
ax(2) is the second-order nonlinear susceptibility.
T. S. Berzina, V. I. Troitsky, E. Stussi, M. Mule and D. De Rossi,
Synth. Met., 1993, 60, 111.
br is the electrooptic coeÃcient37.
1
2
Y. Xiao, Z. Yao and D. Jin, T hin Solid Films, 1993, 224, 237.
Y. Xiao, Z. Yao and D. Jin, J. Phys. Chem., 1993, 97, 8519.
1
3
Nonlinear optical measurements. The Pockels electrooptic
eÂect was measured for monolayers of compounds 1 and 3
14 Y. Xiao, Z. Yao and D. Jin, L angmuir, 1994, 10, 1848.
15 E. Dupart, B. Agricole, S. Ravaine, C. Mingotaud, O. Fichet,
P. Delhaes, H. Ohnuki, G. Munger and R. M. Leblanc, Thin Solid
Films, 1994, 243, 575.
(
both mixed with TA). The surface plasmon resonance (SPR)
method also allowed estimates of the film thickness to be made
1
6
T. S. Berzina, S. A. Shikin, P. S. Sotnikov, V. I. Troitsky, V. Yu.
Khodorkovsky, O. Ya. Neilands and G. G. Pukitis, Top. Mol. Org.
Eng., 1991, 7, 99.
by assuming a value of the permittivity (e =2.5 was used in
r
this case). The results are given in Table 4. In each case the
nonlinear optical r coeÃcient is relatively small.37,42 This is
most likely due to poor alignment of the molecules, as the
chromophores themselves should have large values of hyper-
polarisability. The thickness obtained for a monolayer of 1
17 S. A. Shikin, T. S. Berzina, P. S. Sotnikov, V. I. Troitsky, O. Ya.
Neilands, G. G. Pukitis and V. Yu. Khodorkovsky, Biological
Membranes, 1991, 4, 1793.
1
8
L. M. Goldenberg, V. Yu. Khodorkovsky, J. Y. Becker, P. J. Lukes,
M. R. Bryce, M. C. Petty and J. Yarwood, Chem. Mater., 1994,
(
1.2 nm) is consistent with the length of the molecule (see
6
, 1426.
earlier discussion) and suggests that the film is one molecule
in thickness. This result is diÂerent from that obtained from
ellipsometry and the Alpha-Step (ca. 5 nm). However, these
experiments were both undertaken with thicker films (Fig. 3).
It is also quite likely that the structure of a monolayer film
depsosited on Ag-coated glass (used for the SPR work) diÂers
from that on silicon.
19 Review: M. R. Bryce and M. C. Petty, Nature, 1995, 374, 771.
20 L. M. Goldenberg, R. Andreu, M. Saviron, A. J. Moore, J. Gar ´ı n,
M. R. Bryce and M. C. Petty, J. Mater. Chem., 1995, 5, 1593.
2
2
1
2
L. M. Goldenberg, J. Y. Becker, O. Paz-Tal Levi, V. Yu.
Khodorkovsky, M. R. Bryce and M. C. Petty, J. Chem. Soc., Chem.
Commun., 1995, 475.
Y. Xiao, Z. Yao and D. Jin, L angmuir, 1994, 10, 1848.
23 R. P. Parg, J. D. Kilburn, M. C. Petty, C. Pearson and T. G. Ryan,
J. Mater. Chem., 1995, 5, 1609.
2
4
Y. Xiao, Z. Yao and D. Jin, T hin Solid Films, 1994, 251, 94.
R. M. Metzger, C. A. Panetta, N. E. Heimer, A. M. Bhatti,
E. Torres, G. F. Blackburn, S. K. Tripathy and L. A. Samuelson,
J. Mol. Elect., 1986, 2, 119.
Conclusions
2
5
This work has demonstrated that semi-conducting LB films
can be formed from non-amphiphilic TTF-derived compounds
that contain an aromatic ring substituent, utilising only 25%
molar ratio of fatty acid. The LB films have been characterised
by UV–VIS spectroscopy, conductivity measurements and
cyclic voltammetry. In spite of diÂerences in the electronic
properties of compounds 1 and 3 (the former is neutral, the
latter is ionic) their corresponding LB films show similar
conductivities, those of 1 being more stable. This work paves
the way for the study of conducting LB films of a wide range
of neutral and ionic electron donor molecules, where it may
not be necessary to attach the traditional long alkyl chains
that render the systems amphiphilic.
2
6
H. Nadizadeh, D. L. Mattern, J. Digleton, X-L. Wu and
R. M. Metzger, Chem. Mater., 1994, 6, 268.
27 N. A. Bell, R. A. Broughton, J. S. Brooks, T. A. Jones, S. C. Thorpe
and G. J. Ashwell, J. Chem. Soc., Chem. Commun., 1990, 325.
2
2
8
9
G. J. Ashwell, J. R. Sambles, A. S. Martin, W. G. Parker and
M. Szablewski, J. Chem. Soc., Chem. Commun., 1990, 1374.
X -L. Wu, M. Shamsuzzona, R. M. Metzger and G. J. Ashwell,
Synth. Met., 1993, 57, 3836.
30 P. Wu, G. Saito, K. Imaeda, Z. Shi, T. Mori, T. Enoki and
31 K. Hartke, T. Kissel, J. Quante and R. Matusch, Chem. Ber., 1980,
H. Inokuchi, Chem. L ett., 1986, 441.
1
13, 1898.
3
2
A. J. Moore and M. R. Bryce, T etrahedron L ett., 1992, 33, 1373.
3
3
O. Neilands, Y. Kacens and Y. Kreizberga, Zh. Org. Khim. 1989,
2
5, 658.
L.M.G. thanks the Royal Society, the EPSRC and the
34 V. Yu. Khodorkovsky, G. V. Tormos, O. Ya. Neilands, N. V.
University of Durham for financial support to visit the
Kolotilo and A. Ya. Il’chenko, Tetrahedron L ett., 1992, 33, 973.
9
06
J. Mater. Chem., 1997, 7(6), 901–907

3
3
3
5
6
7
C. A. Jones, M. C. Petty, G. G. Roberts, G. Davies, J. Yarwood,
N. M. RatcliÂe and J. W. Barton, T hin Solid Films, 1987, 155, 187.
Y. Fu, J. Ouyang and A. B. P. Lever, J. Phys. Chem., 1993, 97,
40 J. B. Torrance, B. A. Scott, B. Welber, F. B. Kaufman and
P. E. Seiden, Phys. Rev. B, 1979, 19, 730.
41 D. L. Lichtenberg, R. L. Johnson, K. Hinkelmann, T. Suzuki and
F.Wudl, J. Am. Chem. Soc., 1990, 112, 3302.
1
3 753.
J. P. Cresswell, M. C. Petty, I. Ferguson, M. Hutchings, S. Allen,
T. G. Ryan, C. H. Wang and S. S. Wherrett, Adv. Mater. Optics
Electron., 1996, 6, 33.
42 J. P. Cresswell, M. C. Petty, J. E. Shearman, S. Allen, T. G. Ryan
and I. Ferguson, T hin Solid Films, 1994, 244, 1067.
38
9
V. E. Kampar, Russ. Chem. Rev., 1982, 51, 107.
3
V. E. Kampar and O. Neiland, Russ. Chem. Rev., 1986, 55, 334.
Paper 6/06831F; Received 2nd January, 1997
J. Mater. Chem., 1997, 7(6), 901–907
907