Preparation and characterisation of conductive Langmuir-Blodgett films of a tetrabutylammonium-Ni (dmit), complex Leonid M. Goldenberg,".b*'Christopher Pearson,b Martin R. Bryce*' and Michael C. Petty*b ahtitUteof Chemical Physics in Chernogolovka, Russian Academy of Science, Chernogolovka, 142432 Moscow region, Russia bSchool of Engineering and Centre for Molecular Electronics, University of Durham, Durham, UK DH1 3LE 'Department of Chemistry and Centre for Molecular Electronics, University of Durham, Durham, UK DH1 3LE The behaviour of tetrabutylammonium-Ni(dmit),at the air-water interface has been investigated and Langmuir-Blodgett (LB) films have been built-up on a variety of substrates from the floating layer containing different concentrations of tricosanoic acid.The morphology, spectroscopic, electrical, electrochemical and spectroelectrochemical properties of the LB films are reported. In- plane dc room-temperature conductivity values of ort=lo-, S cm-' and thermal activation energies AE =0.08-0.1 eV over the temperature range 300-100 K, have been achieved by post-deposition electrochemical doping with four different anions, and by chemical doping with iodine vapour. The films doped with iodine and perchlorate retained this conductivity value upon storage for ca. 1 month. Electrochemical doping during deposition of the LB films resulted in conductivity values of ort=lov3S cm-'. The Langmuir-Blodgett (LB) technique is now recognised as an important method for organising charge-transfer complexes at the molecular level.' Many of the materials studied are amphiphilic systems containing the electron acceptor tetracy- ano-p-quinodimethane (TCNQ)2-4 or the electron donor tetra- thiafulvalene (TTF) and its close analog~es.~-~ The LB films of these materials are characterised by strong one-dimensional interactions and the most highly conducting layers possess in- plane conductivity values in the range art= 10-,-1 S cm-' after doping with iodine vapour.Over the last decade, work on crystalline molecular conductors has established that many cation salts of M(dmit), anions [M=Ni, Pd, Pt, Au; dmit= bis(4,5-dimercapto-l,3-dithiole-2-thione)] exhibit metallic behaviour, and superconductivity has been observed in a few salts.' The presence of peripheral sulfur atoms in the M(dmit), anions, which engage in non-bonded S---S interactions, facili- tates an increase in the dimensionality of the system, thereby stabilising the metallic state.In the search for LB films of charge-transfer systems with higher dimensionality, Japanese workers have prepared thin films of tetraalkylammonium M(dmit), complexes (M =Au, Ni, Pd or Pt) by attaching long hydrophobic chains to the cation; conductivity values of 10-3-50 S cm-' were obtained after chemical or electrochemi- cal doping, assuming a thickness per layer of 3 nm." Because of the unstable nature of the condensed films of these materials at the air-water interface, the complexes were usually deposited as 1 :1 mixtures with a fatty acid using the horizontal touching technique."*12 Monolayer formation of related amphiphilic tetraalkylammonium M(dmit), complexes (M =Pt, Pd upon storage for at least several months at room temperature.We have used LB films of the 1 : 1 stoichiometry complex as the basis of a thin-film transistor21 and as a charge-injection layer in an electroluminescent device.,, Recently, we have succeeded in depositing stable LB mono- layer and multilayer films of some non-amphiphilic charge- transfer materials, e.g. ethylenedithio-TTF (EDT-TTF) deriva- tives with aryl substituents using only 25% molar fatty acid as additive,23 and a TTF derivative containing an azobenzene substituent without addition of any fatty We also have been able to deposit reproducibly multilayer films of pure bis(ethy1enedithio)TTF (BEDT-TTF) with no added fatty These res~lts,~-,~ demonstrate for the first time that for some charge-transfer systems the presence of a traditional long-chain substituent is not necessary for the formation of LB films.This finding is an important development, as the attachment of a long chain can often present difficulties in synthesis and, more especially, in the purification of the material. In this context, we decided to explore the LB film behaviour of the commercially available dmit salt Bu4N-Ni(dmit), 1, which does not contain a long hydro- phobic chain. Herein we report film formation of complex 1 using the LB and Ni) has been reported by Taylor and co-~orkers.~~-~~ technique, and the electrical, electrochemical and spectroscopic We have previously reported LB films of (N-o~tadecylpyridinium)~-Ni(dmit),with in-plane conductivity values following iodine d~ping'~,~~ as high as ort=0.9 S cm-', while films of the analogous 1: 1 salt N-octadecyl-pyridinium-Ni(dmit), possess lower conductivity val~es,'~ ort=0.2S cm-'.For the latter salt, similar conductivity values were obtained by electrochemical doping.,' In the case of films of the 1 :1 salt, high conductivity values were observed only when aggregates of micrometre dimensions were formed." Films of both complexes (N-octadecylpyridinium),-Ni(dmit), salts (x=1 or 2) were obtained without the need for added fatty acid, and in contrast to most LB films of charge- transfer materials,' they retained their high conductivity values studies of conducting films obtained by electrochemical and chemical doping.Experimenta1 Complex 1 was obtained from Tokyo Kasei Organic Chemicals and used as received. Tricosanoic acid (TA) was obtained from Sigma. LB film deposition was undertaken in a class 10000 micro- electronics clean room using constant-perimeter barrier LB troughs designed and built in Durham.26 The solution of 1 in dichloromethane (0.5 g dmP3) and TA in chloroform (1 g dm -3) were used for preparation of spreading solutions. The J. Muter. Chem., 1996,6( 5), 699-704 699 solution was spread onto the surface of ultrapure water (obtained by reverse osmosis, deionisation and ultraviolet sterilisation) at pH 5 8 k0 2 at 20 f2 "C The pressure us area isotherms were measured at a compression rate of approxi- mately 4 x nm2 complex-' s-' LB films were built up on glass, indium tin oxide glass (ITO, sheet resistance 300R per square, from Balzers) and glass slides coated with Au Au electrodes on glass were obtained by vacuum deposition of 30-35nm of Au undercoated with 3-5nm of Cr to improve adhesion of Au to the glass The glass was cleaned by ultrasonic cleaning in a 5% solution of Decon 90 for 30 min, rinsing in ultrapure water and drying in a stream of nitrogen For IT0 and Au-coated glass, the above procedure was adopted first and then, to improve the hydro- philic properties, these substrates were treated prior to LB transfer with saturated Na2Cr20,-conc H2S04 solution for ca 10s and carefully washed with ultrapure water 27 Film thicknesses were measured using a surface-profiling Tencor Instruments Alpha-Step 200 (stylus force = 11& 1 mg) A layer of aluminium (thickness ca 150 nm) was evaporated over the step between the organic layer and the uncoated substrate The film morphology was investigated using a scanning elec- tron microscope operating at 20 kV Optical absorption spectra were recorded for films deposited onto glass using a Perkin- Elmer Lambda 19 UV-VIS-NIR spectrometer For electro- chemistry, IT0 slides with an area of between 20 and 30 cm2 were used for LB film transfer After film deposition, the slides were carefully cut with a diamond-tipped stylus to form several electrodes with contact areas of 0 1-0 5 cm2 We have devel- oped a technique for the production of gold electrode arrays with narrow gaps without the need for photolithography a glass slide was first coated with a film of gold by thermal evaporation, several parallel lines were then scribed in this gold film using the tip of a new single-edge industrial razor blade This method produced gaps in the gold film of width cu 30pm which could be coated subsequently with an LB film Interdigitated Au electrodes with a gap of 20 pm were prepared by photolithography An EG&G PARC model 273 potentiostat with an Advanced Bryans X Y recorder was used for electrochemical experiments Pt mesh served as the counter electrode and IT0 or Au electrodes covered with LB films served as working electrodes All potentials were recorded us an Ag/AgCl reference electrode LiClO, (Fluka, microselect), KC1 (Fluka, microselect), Bu,NBr (Aldrich), KI (Hopkin & Williams), NaC10, (Aldrich) and ultrapure water were used for the preparation of electrolyte solutions Spectroelectrochemistry was undertaken using the Lamba 19 spectrophotometer described above with a Ministat (Thomson Electrochem Ltd, Newcastle upon Tyne, UK) A spectroelectrochemical cell was based on a cuvette of thickness 1 cm, Pt wire was used as the counter electrode, while Ag wire served as a quasi-reference Chemical doping of the LB films was achieved by exposure to iodine vapour in a sealed container In-plane dc conductivity studies were undertaken using a two-probe technique The electrical contacts were either a carbon cement (Neubauer Chemicallien) on glass or, as described above, Au electrodes with gap 30 pm The conductivity normal to the film surface was measured by using evaporated Au top contact dots with diameter 0 1 cm for the films deposited on Au-coated glass slides Room-temperature measurements were undertaken in a screened sample chamber, evacuated to a pressure of ca mbar using a rotary pump Electrical conductivity measurements over the temperature range 100-300 K were made in helium using an Oxford Instruments DN704 exchange gas cryostat The voltage source in these experiments was a Time Instruments voltage calibrator and the current was monitored using a Keithley picoammeter Electrochemical doping during LB film transfer was achieved at a constant current of 3-6 pA supplied from a Farnell Instruments current 700 J Muter Chew , 1996, 6(5), 699-704 -k ',i 0 0 0 10 0 20 030 molecular area/nm* Fig.1 Pressure us area isotherms of 1 mixed with different concen- trations of TA Subphase pH 5 8 +O 2, temperature 20+2 "C, mono-layer compression rate ca 4 x 10 nm2 complex ' s ' a, 0% TA, b, 24% TA, C,66% TA source between a moving (vertical dipping) Au array electrode and another Au counter electrode placed in the subphase outside the barrier In another version of this experiment the counter electrode was attached to the working electrode and the two were dipped together For these experiments a 0 1 mol dm-3 KI or NaC10, subphase was used Results and Discussion LB film preparation and morphology Condensed pressure us area isotherms could be obtained for floating films of pure 1, or for 1 mixed with TA Fig 1 shows examples for the pure complex, and for two different concen- trations of the mixture with TA The isotherms were reproduc- ible (within 5YO)for subsequent expansions and compressions It is noteworthy that the isotherms were independent of the time that the film was left on the subphase before compression, this contrasts markedly with results reported by Taylor et a1 l4 and our~elves'~ for amphiphilic Ni(dmit), analogues Here, true monolayers of some systems were obtained only if the floating film was kept for several hours on the subphase in its uncompressed state The average molecular area per complex 1 can be calculated by extrapolating the high-pressure regions of the isotherms (>30 mN m-') in Fig 1 to zero pressure and subtracting the average area occupied by the fatty acid molecules (assuming the TA molecules are all in contact with the subphase surface and have a cross-sectional area of 0 20 nm2) Fig 2 shows the resulting dependence upon fatty acid concentration The Ni(dmit), moiety may be modelled by a box of dimensions of 161 nm x 0 623 nm x 0 366 nm', while the tetrabutylam-monium cation will be more similar to a sphere of approximate diameter 13 nm At low TA concentrations, Fig 2 indicates 0 20 0 I cuFE016/. I I 000 ' 1 0 20 40 60 80 100 concentration of TA(%) Fig. 2 Area per complex, calculated assuming the cross-section of TA as 0 2 nm2 (see text for details) us concentration of TA Subphase pH 5 8 & 0 2, temperature 20 f2 "C, monolayer compression rate ca 4x 10 ' nm2 complex ' s ' \ 16 b A 37 12 Ak A3E AAY 245 0 20 40 60 80 100 concentration of TA(%) Fig.3 Single layer thickness of films of 1 determined by Alpha-Step profiling us.concentration of TA that the organometallic complex is almost certainly in the form of a multilayer on the water surface. As the proportion of fatty acid increases, the average molecular area per complex also increases, indicating that more of the Ni(dmit), molecules and/or the counter ions are incorporated in the fatty acid monolayer. LB film deposition was undertaken at a surface pressure of 30-35 mN m-'. Pure 1 exhibited Z-type deposition.For mixtures with fatty acid, the film transfer started as Z-type deposition but this reverted to Y-type after several (1-4, depending on the exact composition) dipping cycles. The deposition ratio was 1.OkO.1 for both the Z and the Y cycles. Multilayers assembled from pure 1 appeared reasonably uni- form to the eye but their quality improved with increasing fatty acid content. A plot of the average thickness per layer, measured with the surface profiling Alpha-Step, vs. TA concentration is shown in Fig. 3. The large values of thickness obtained with low fatty acid concentrations confirm the suggestion above that the floating film is more than one molecule (complex) in thickness.With increasing content of TA, the average thickness per transferred layer decreases and approaches a constant value of 5 nm. This is considerably greater than the longest side of the Ni(dmit), anion (1.61 nm) but is consistent with a film in which the Ni(dmit), anion is mixed within the fatty acid matrix (thickness 3.0 nm) and the tetrabutylammonium cation is on the top. This would give a total thickness of ca. 3.0 nm + 1.3 nm =4.3 nm. This model results in the calculated average area per complex to equal that of the Ni(dmit), cross- section (ca. 0.23 nm2), which corresponds well with the observed value of 0.18 nm2 at high concentrations of TA (Fig.1). The LB films of 1 were studied using scanning electron microscopy (SEM). Fig. 4 shows the electron micrograph of a 26-layer LB film of 1containing 44% TA (note that at this TA concentration, the mixed LB films are still electrically conduc- tive, see Fig. 6 later). The film consists of two phases, with lighter aggregates randomly distributed in a darker matrix. The dimension of the aggregates is < 1 pm, which is consider- ably smaller than the aggregates previously observed for LB films of the amphiphilic We suggest that the LB films of 1 contain separated phases of the Ni(dmit), complex and TA well distributed in each other. As the film is conductive, the dark regions are likely to consist of the organometallic species.Optical absorption Fig. 5(u) shows the optical absorption for LB films of 1mixed with 38 mol% of TA for different numbers of layers. Absorption bands are observed at ca. 1350, 1150, 650, 600, 460, 400, 360 and 320 nm. These are associated with electronic transitions within the Ni(dmit), moietiesz8 and are similar to the bands observed for the LB films of N-octadecyl-Fig. 4 Scanning electron micrograph for a 26-layer LB film of 1 mixed with 44 mol% of TA deposited on Au, after iodine doping and vacuum treatment 3.5 3 2.5 2 1.5 1 h .-Y 0.5 t3 0 nY5 300 800 1300 1800 v 1.52l 300 800 1300 I800 wavelengthhm Fig.5 Optical absorption of LB films of 1 mixed with 38 mol% of TA deposited on glass: (a) dependence of absorption on the number of layers for as-deposited films (inset shows the plot of absorbance at 1150 nm us.the number of layers); (b) effect of iodine doping and vacuum treatment for a 38-layer LB film: -, as deposited; ---, iodine doped; ---, after vacuum treatment (high conductivity state) pyridinium-Ni(dmit),.19 The insert in Fig. 5(a)shows that the plot of absorbance at 1150 nm us. the number of layers is linear, indicating reproducible deposition of the LB layers. Fig. 5(b) shows the evolution of the optical absorption spec- trum after doping a 38-layer film with iodine vapour. Immediately after doping, the peaks at 1350 and 1150nm disappeared and a new peak was evident at 920nm.As the film was treated in vacuum to gain the maximum conductivity value (see below) the intensity of the peak at 920 nm decreased. J. Muter. Chern., 1996, 6(5),699-704 701 This absorption is associated with the formation of a mixed-valence state Electrochemistry and spectroelectrochemistry The electrochemical response for LB films of 1 built up from floating layers with 24% TA was found to be similar in different electrolytes Fig 6 shows the response in Bu,NBr and KCl electrolytes Redox potential data obtained using four electrolytes are collated in Table 1 Both NilI1+NiIV (Fig 6, couple A) and NiI1-+Ni1I1(couple B) redox transitions within the dmit ligand were observed at potentials similar to those found previously for an amphiphilic derivative2' and the results are consistent with solution studies on other Ni(dmit), com-plexes 29 We also measured the voltammetric response for multilayer films of 1 and it was found to be similar to the single-layer response Moreover, the electroactivity increased with increasing number of layers as shown in Fig 7, where the reduction peak currents are plotted us the (number of layers) x (transfer ratio) This suggests that all layers in the multilayer assembly participate in the electrochemical activity These data indicate that the redox processes in these films are quite facile and are not hindered by counter ion diffusion This result forms a good basis for studies on electrochemical doping Fig 8 shows the results of spectroelectrochemical studies in two Bu,NBr and KI electrolyte solutions An optical change was observed after applying a potential step for only a few minutes This contrasts with the several hours required before a response was obtained from more compact LB films of an A b 05 VA B Fig.6 Cyclic voltammogram of a single LB layer of 1 mixed with 24 mol% of TA deposited on an IT0 electrode (a) 3 rnol dm KCl solution, scan rate 0 05 V s ',(b) 0 5 rnol dm Bu,NBr solution, scan rate 0 05 V s Table 1 Redox potentials, determined by cyclic voltammetry, for a single layer of complex 1 mixed with 24mol% TA in different electrolyte solutions electrolyte solution (concentration/mol dm 3, E,"/V E,"/V KC1 (3) -008 +036 LiClO, (0 4) -026 +014 KI (1) -028 +016* Bu,NBr (0 5) -018 +027 ~~~ "Determined as midpoint between the cathodic and anodic peaks bReduction peak potential, as the oxidation peak was not observed due to the rising background current 702 J Muter Chew , 1996,6(5), 699-704 20 I 0 16 A%I0 12 A 4.9 A 0 3 6 9 12 15 (no of layers) x (transferratio) Fig.7 Reduction peak currents measured during cyclic voltammetry of LB layers of 1 mixed with 24 mol% of TA deposited on an IT0 electrode in 3 mol dm KC1 solution, scan rate 0 05 V s us (number of layers) x (transfer ratio) 0,first cathodic peak, A second cathodic peak 01. I I. 1, I I I I. I .'.'I I. v 350 650 950 1250 350 575 800 1025 1250 wavelengthhm Fig.8 Absorption spectra for a 19-layer LB film of 1 mixed with 40 mol% of TA deposited on IT0 Measurements at different potentials us Ag wire (a) 05 mol dm Bu,NBr solution (-, open circuit, ___ , + 1 V, ---, -0 3 V), (b) 1 mol dm KI solution (-, open circuit, ---, +06V, -, -03V) amphiphilic TTF derivative studied previously 30 In accord with the spectroscopic changes accompanying chemical doping of LB films of 1, electrochemical oxidation resulted in the disappearance of a peak at 1150 nm and the emergence of a new peak at ca 900nm On changing the potential to reduction, the spectrum reverted to that measured previously In-plane electrical conductivity The in-plane dc conductivities for the as-deposited LB films of pure 1 or for samples mixed with TA were generally <10 'S cm-' However, after doping with iodine vapour the conductivity increased almost immediately, and continued to grow after the application of vacuum Maximum conductivity values in the range art= 10-3-10-2 S cm-' were achieved after ca.20 min storage under vacuum. These values were calculated using the measured film thickness, which depended on the concentration of TA (Fig. 3). The conductivity for complex 1 is, therefore, ca. one order of magnitude lower than that recently reported for the 1 :1 N-octadecylpyridinium deriva- tive.lg Fig. 9 shows the dependence of the in-plane room-temperature dc conductivity for iodine-doped samples of 1 on the concentration of TA. This plot indicates that percolation threshold is achieved at a TA concentration of ca.50%. For electrochemical doping experiments, multilayer films were built-up on Au electrodes with a 30 pm gap from a floating layer with a different concentration of TA. The films were oxidised either at a constant potential of 0.6-0.9 V or at a constant current of ca. 1-2 PA. The process was terminated from time to time, the films were washed with water and dried in a nitrogen steam and the conductivity was measured. Conductivity values achieved in different electrolytes, and by iodine-vapour doping, are collated in Table 2. We have also measured the conductivity normal to the film surface for iodine-doped samples and samples doped electrochemically in 3 mol dm-3 KCl. This conductivity was the same order of magnitude as the in-plane conductivity.In contrast to the usual method of post-deposition doping, Tieke et aL3' suggested the technique of electrochemical oxi- dation with a 'live' electrode during LB film deposition, and this method was subsequently used by Morand et al.32 However, neither group reported conductivity data on the films obtained using this approach. We have now applied this technique to LB films of complex 1, using constant current oxidation and a 0.1 mol dm-3 KI or NaC10, subphase. The in-plane conductivity of the resulting LB films was crt= S cm-l, which is similar to the values discussed above obtained from post-deposition oxidation. To our knowledge, this is the first time that conductivity data have been measured as a result of electrochemical doping during LB film deposition.We monitored the conductivity of different samples with time in order to estimate the stability of the conducting films. 10-* I 0 1044 0 20 40 60 80 100 concentrationof TA(%) Fig. 9 Plot of in-plane dc conductivity measured at room temperature for iodine-doped samples of LB films of 1 deposited on an Au electrode with a 30 pm gap us. concentration of TA Table 2 Room-temperature dc in-plane conductivity values for the multilayer films (15-30 layers) of 1 mixed with 24-44% TA, obtained by electrochemical doping in different electrolytes, and by iodine- vapour doping (unless specified otherwise data were obtained on a gold electrode with a 30 pm gap) electrolyte conductivity/ concentration of TA (concentration/mol dmW3) S cm-l (mol%) ~ ~~ KC1 (3) (0.3-1) x 24-35 LiC10, (0.4) (0.7-1) x 10-3 24-35 Bu,NBr (0.5) (0.7-2) x 10-3 24-35 (0.5-1) x 10-3 24-35 (0.3-7) x 10-3 24-44 (0.2-8) x lop2 24-44 "For carbon cement contacts onto glass.0I 0-la 5n 0.002 0.006 0.010 0.0145-c -14 tv-18 L 0 0 1...1...1...1. .. I... 0.002 0.006 0.010 0.014 T /K Fig. 10 Current us. reciprocal temperature for a 24-layer LB film of 1 mixed with 24 mol% of TA deposited on an Au electrode with a 30 pm gap, applied voltage 1 V: (a)after electrochemical doping in 0.4 mol dm-3 LiC10, electrolyte; (b) after iodine doping. AE is the thermal activation energy; 0,decreasing temperature; 0,increasing temperature The general conclusion is that samples doped with iodine vapour were stable (either with carbon or Au contacts) and the conductivity decreased only slightly over one month.The same behaviour was observed for samples doped electrochemi- cally in LiC10, or KI solutions. On the other hand, samples doped in Bu,NBr solution showed less reproducible behaviour, although some samples showed reasonable stability of the conductivity value for one month. Less stable were samples doped electrochemically in KCl solution, for which a drop in conductivity of 1-2 orders of magnitude occurred after one month. The temperature dependences over the range 300-100 K of the in-plane electrical conductivity for the LB films doped electrochemically with perchlorate and chemically with iodine, obtained in the same deposition experiment, are shown in Fig.10. In both cases, the conductivity followed an exponential dependence upon temperature, with thermal activation energies AE =0.1 and 0.08 eV, respectively. These values are similar to those obtained for the films of other amphiphilic Ni(dmit), complexes doped with iodine (0.05-0.06 eV)" or electrochemi- cally (0.07 eV).20 The results presented in this paper (e.g. variable deposition ratio) suggest that LB films of 1 are not as ordered as those built up from amphiphilic organometallic compounds. We have, therefore, undertaken some experiments using films of 1 produced by solution casting. Layers were deposited from a chloroform solution (1 g dm-3).The thickness of the resulting films was ca. 120nm (measured by Alpha-Step) and, after iodine doping, the in-plane dc room-temperature conductivity was 1x loF2S cm-'. This result confirms the presence of considerable disorder of the Ni(dmit), moieties within the LB films. J. Muter. Chem., 1996, 6(5), 699-704 703 Conclusions H Tachibana, M Tanaka and Y Kawabata, Thin Solid Films, 1989,179,183 We have reported the preparation and characterisation of LB films of Bu,N-Ni(dmit), 1 mixed with TA LB films were reproducibly built up from the floating layer containing differ- ent concentrations of TA and they were conductive after chemical or electrochemical doping with TA concentrations of 11 12 Y F Miura, M Takenaga, A Kasai, T Nakamura, Y Nishio, M Matsumoto and Y Kawabata, Thin Solid Films, 1992, 210/211,306 Y F Miura, M Takenaga, A Kasai, T Nakamura, M Matsumoto and Y Kawabata, Jpn J Appl Phys, 1991, 30, 3503 up to 50% Cyclic voltammetry and spectroelectrochemistry for LB films of complex 1 revealed a facile response for both monolayer and multilayer films Chemical doping by iodine vapour and electrochemical doping of the LB multilayers by different anions resulted in a maximum room-temperature dc in-plane conductivity of ca 1 x lo-, S cm-I The temperature 13 14 15 16 D M Taylor, A E Underhill, S K Gupta and C E Wainwright, Makromol Chem ,Macromol Symp, 1991,46, 199 D M Taylor, S K Gupta, A E Underhill and C E Wainwright, Thin Solid Films, 1992, 210/211,287 S K Gupta, D M Taylor, P Dynarowicz, E Barlow, C E Wainwright and A E Underhill, Langmuir, 1992,8,3057 S K Gupta, D M Taylor, A E Underhill and C E Wainwright, dependence of conductivity over the range 300-100 K was found to be similar for both the electrochemically and the chemically doped films 17 18 Synth Met, 1993,58,373 A S Dhindsa, J P Badyal, C Pearson, M R Bryce and M C Petty, J Chem Soc Chem Commun, 1991,322 C Pearson, A S Dhindsa, M C Petty and M R Bryce, Thin Solid L M G thanks the University of Durham for financial support M R B thanks the University of Durham for a Sir Derman 19 Films, 1992,210/211,257 C Pearson, A S Dhindsa, L M Goldenberg, R A Singh, R Dieing, A J Moore, M R Bryce and M C Petty, J Muter Christopherson Research Fellowship We thank Professor P Delhaes for suggestions concerning electrochemical doping during film deposition 20 21 Chem, 1995,5,1610 L M Goldenberg, A P Monkman, C Pearson, J Gibson, M R Bryce and M C Petty, Thin Solid Films, 1996, in press C Pearson, J E Gibson, A J Moore, M R Bryce and M C Petty, Electron Lett, 1993,29, 1377 References 22 G Williams, A J Moore, M R Bryce and M C Petty, Thin Solid Films, 1994,244,936 1 Reviews T Nakamura and K Kawabata, Techno Japan, 1989,22, 8, B Tieke, Adv Muter, 1990,2,222,M R Bryce and M C Petty, Nature, 1995,374,771 A Barraud, A Raudel-Teixier, M Vandevyver and P Lesieur, Nouv J Chem ,1985,9,365 T Nakamura, M Matsumoto, F Takei, M Tanaka, T Sekiguchi, E Manda and Y Kawabata, Chem Lett, 1986,709 A S Dhindsa, G H Davies, M R Bryce, J Yarwood, J P Lloyd, M C Petty and Y M Lvov, J Mol Electron, 1989,5,135 J Richard, M Vandevyver, A Barraud, J P Morand, R Lapouyade, P Delhaes, J F Jacquinot and M Roulhay, J Chem Soc Chem Commun, 1988,754 C Dourthe, M Izumi, C Garrigou-Lagrange, T Buffeteau, P Desbat and P Delhaes, J Phys Chem ,1992,96,2812 A S Dhindsa, Y P Song, J P Badyal, M R Bryce, Y M Lvov, M C Petty and J Yarwood, Chem Muter, 1992,4,724 L M Goldenberg, R Andreu, M Saviron, A J Moore, J Garin, M R Bryce and M C Petty, J Mater Chem, 1995,5, 1593 Reviews M R Bryce, Chem Soc Rev, 1991, 20, 355, A E Underhill, J Mater Chem, 1992, 2, 1, P Cassoux and L Valade, in Inorganic Materials, ed D W Bruce and D O’Hare, 23 24 25 26 27 28 29 30 31 32 L M Goldenberg, J Y Becker, 0 Paz-Tal Levi, V Y Khodorkovsky, M R Bryce and M C Petty, J Chem SOC Chem Commun ,1995,475 L M Goldenberg, S Wegener, M C Petty and M R Bryce, unpublished results L M Goldenberg, C Pearson, M R Bryce and M C Petty, manuscnpt in preparation M C Petty and W A Barlow, in Langmuir-Blodgett Films, ed G G Roberts, Plenum Press, New York, 1990, p 93 Y Fu, J Ouyang and A B P Lever, J Phys Chem, 1993, 97, 12753 G Steimecke, H -J Sieler, R Kirmse and E Hoyer, Phosphorus Sulfur, 1979,7,49 J-B Tommasino, B Pomarede, D Medus, D de Mantauzon and P Cassoux, Mol Cryst Liq Cryst, 1993,237,445 L M Goldenberg, G Cooke, C Pearson, A P Monkman, M R Bryce and M C Petty, Thin Solid Films, 1994,238,280 B Tieke, A Wegmann, W Fischer, B Hilti, C W Mayer and J Pfeiffer, Thin Solid Films, 1989,179, 233 J P Morand, L Brzezinski and M C Lopez, Thin Solid Films, 1992,210/211,280 10 Wiley, Chichester, 1992,ch 1 T Nakamura, H Tanaka, K KOJima, M Matsumoto, Paper 5/07323F, Received 7th November, 1995 704 J Mater Chew , 1996, 6(5), 699-704