J. MATER. CHEM., 1992, 2(1), 37-41 Co-ordination Compounds on the Surface of Laponite: Tri-2-pyridylamine Complexes Stephen P. Bond," Carl E. Hall,bCraig J. McNerlin," William R. McWhinnie" and David J. Waltonb a Department of Chemical Engineering & Applied Chemistry, Aston University, Aston Triangle, Birmingham B4 7ET, UK Department of Applied Physical Sciences, Coventry Polytechnic, Priory Street, Coventry CVI 5FB, UK Freshly prepared [Co(tripyam),](ClO,), (tripyam=tri-2-pyridylamine) contains some low-spin isomer (,E ground state) 'frozen' into the solid; this reverts to the 4T form over 3 months. Ion exchange of orange [Co(tripyam),]*+ from aqueous solution onto laponite gives a pink clay, the exchanged cation being [C~(tripyam),(H,O),]~+ with bident ate p y rid y Iamin e Iig a nd s.Isomeric [C u (t r ipya m ),I (C I04),(trident ate Iig a nds) and [C u (t ripya m ),( CI04),] (bidentate ligands) are absorbed onto laponite as [C~(tripyam),(H,O),]~+ with bidentate ligands; however, the copper clay is thermochromic undergoing a reversible change from green to blue at 100°C. The temperature of the colour change may be increased by employing copper(I1) complexes of substituted tri-2-pyridylamines. The thermochromism is a function of the variable denticity of the ligands. Orange [Co(tripyam),12+ (terdentate ligands) may be exchanged using a novel microwave method which accelerates the ion-exchange reaction, but not the aquation reaction; however, over 1 year the sorbed complex aquates to [C~(tripyam),(H,O),]~+. Although low-spin [Fe(tripyam)J2+ (terdentate ligands) is stable on laponite, in general the clay surface has greater affinity for the bis(bidentate) complex species.Cyclic voltammetric studies of acetonitrile solutions show an order of ease of oxidation of cobalt(1r) complexes: [Co(bipy),12+ > [Co(6-dmdpa),12+ >[Co(tripyam),12+ where bipy =2,2'-bipyridyl and 6-dmdpa =(6-methyl-2-pyridy1)-di-(2-pyridy1)amine. Clay(1aponite)-modified platinum electrodes dipped into acetonitrile solutions of [Co(bipy)J3+ or [Co(tripyam),12+ give very similar E,,, values to those obtained for the free complexes in acetonitrile solution. However, electrodes prepared from laponite pre-exchanged with pink [C~(tripyam),(H,O),]~+ or [Co(bipy),I3+ were rigorously electro-inactive.Keywords: Tri-2-pyridylamine; Laponite ; Cobalt complex; Copper complex; Thermochromic clay Davison' has demonstrated that tri~-2,2'-bipyridylcobalt(111) tris-2,2'-bipyridyl and bis(tri-2-pyridylamine) complexes, but ions exchanged onto hectorite may be reduced chemically with pyridyl conjunction in the former but not the latter (NaBH,) and electrochemically to the cobalt(1) complex. systems, it was of interest to study the clay-exchanged He~torite/[Co(bipy)~]~and NaBH, together reduce nitro- bis(tri-2-pyridylamine) complexes. This paper reports some+ benzene to aniline in excellent yield, the clay-supported cata- new co-ordination chemistry of metal(11)-tri-2-pyridylamine lyst offering the twin advantage of ease of separation and complexes on laponite, a synthetic lithium-magnesium reusability.Large complex cations exchanged into the inter- trioctahedral smectite clay, the closest natural analogue of lamellar regions of the clay effectively pillar the material and which is hectorite. Some electrochemical studies of CMEs +it has been speculated that this fact might facilitate catalyst- based on [C~(tripyam)~]' are also presented briefly. substrate interaction in the above reaction. Others2 have demonstrated the affinity of [Co(bipy),]' + for clay colloids, and recently3 a quantitative examination of a [Co(bipy),12 +/bentonite clay-modified electrode (CME) sys- Experimental and Results tem was reported. The electrochemically produced Co' species Laponite was obtained from Laporte Industries Limited.was shown to be an effective catalyst for the dehalogenation Metal salts were commercial specimens and were used as of both 4,4'-dibromodiphenyl and chloroalkanes present in obtained. organised assemblies achieved in the presence of 1 -hexadecyl trimethylammonium bromide. In order to obtain more information on the sorbed complex Tri-2-pyridylamine. This was synthesised by a literature species and to extend the range of complexes considered, we (Found: C, 72.6; H, 4.93; N, 22.6%. C15H12N4 meth~d.~ have resurrected some complexes of tri-2-p~ridylamine.~ The requires C, 72.6; H, 4.87; N, 22.6%). M.p. 129 "C (literatureg ligand tri-2-pyridylamine (tripyam) co-ordinates either as a m.p. 130 "C.) bidentate or as a vicinal (tripodo-) terdentate base., In the bis-terdentate complex, [C~(tripyam),](ClO,)~, which exhibits spin cross-over beha~iour,~the 295 K structure6 (pseudo- (6-Methyl-Zpyridyl)di-(2-pyridyl)arnine.This oil was syn-'high spin') shows Co-N bond lengths in the range 2.100(2)- thesised as reported previously." (Found: c, 73.3; H, 5.51; N, 2.152(2)8, and N-Co-N angles from 84.86(7) to 86.09(8)". 21.4%.C16H14N4 requires C, 73.3; H, 5.34; N, 21.4%.) By contrast, the low-spin Fe" complex deviates only trivially from true octahedral ~ymmetry.~ The behaviour of the corre- sponding copper(I1) system, Cu(tripyam),(ClO,), is quite com- (4-MethyI-2-pyridyl)di-(2-pyridyl)amine.This semicrystalline plex, with interchangeable bi- and ter-dentate solid was also prepared as previously." (Found: C, 73.2; H, Given that an M(pyridyl-)6 environment pertains to both 5.40; N, 21.4%.) Synthesis of Complexes Bis(tri-2-pyridyEamine)cobaEt(11)Diperchlorate. This was pre-pared as reported previ~usly.~ (Found: C, 47.4; H, 3.04; N, 14.8%.C30H24C12C~N808 requires C, 47.7; H, 3.20; N, 14.9%.) This orange material was dissolved in water in a flask which was stoppered and placed on a mechanical shaker for 1 week, giving a pink solution, Amax/nm 500, cf. 502 for an aqueous solution of [C0(py)~(H~0)~]~ + where py =pyridine. An orange solid, [C~(tripyam),](ClO,)~ 3H20 was isolated from the aqueous solution. (Found C, 44.8; H, 3.33; N, 14.0%. C30H30C12CON8011requires C, 44.6; H, 3.71; N, 13.9%.) Conductivity measurements in acetonitrile confirmed that the perchlorate groups were ionised in that solvent.Bis[(6-methyE-2-pyridyl)di-(2-pyridyl)amine]cobaIt(11) Diper-chlorate Monohydrate. This was prepared using ethanol rather than triethylorthoformate' ' as a solvent. Thus (6-methyl-2- pyridyl)di-(2-pyridyl)amine (10.5 g) in 10 cm3 ethanol was added to Co(C104),*6H20 (0.32g) in 10cm3 ethanol and refluxed for 30 min to give a brown-orange precipitate. The solid was separated and washed with ethanol to give an orange solution and a green residue. On concentration, the orange solution gave an orange solid which was separated, washed with ethanol, and dried. The green residue and orange product gave identical C, H, and N analyses and both had A (molar conductivity) =335 s mol-'dm3 in CH3CN.Infrared (IR) spectra were similar and indicated ionic perchlorate in each case. The orange material had peff=4.91pg. {Found (green): C, 47.8; H, 4.06; N, 14.0%; (orange): C, 47.7; H, 4.08; N, 14.0%. [C0(6-mpdpa)~](ClO.+)~ H20, C32H30C1&ON805 requires C, 47.9; H, 4.00; N, 14.0%.) Bis(tri-2-pyridylamine)iron(11)Diperchlorate Monohydrate. This was prepared by reaction of 10.35 g Fe(C104)2*6H20 in triethylorthoformate (10 cm3) with tri-2-pyridylamine (0.7 g) in ethanol (10 cm3) with passage of dinitrogen. A red-orange solid precipitated immediately and was separated and washed with ethanol. (Found: C, 46.9; H, 3.22; N, 14.6%. C30H28FeN809 requires C, 46.9; H, 3.38; N, 14.6%).The previously reported complex was anhydrous4 but the Mossbauer parameters of the new material are identical [6 = 0.62 & 0.05 mm s-' us. Na2Fe(CN)5(NO)-2H20] with those for [Fe(tripyam), ](ClO,), (6 =0.63f 0.05 mm s-') reported in an earlier paper.I2 However, the Mossbauer spectrum also showed a resonance (5% of total intensity) for which 6= 1.34f0.05 mm s-', A=2.53+0.05 mm s-', corresponding to high-spin iron(r1). Tris(di-2-pyridyl)iron(11)diperchlorate is known to be high spin and has 6= 1.27k0.05, A= 2.51+_0.05mm s-', hence the impurity is likely to be the tris(bidentate tri-2-pyridy1)amine iron@) species. In support of this, if the above experiment is repeated without the passage of dinitrogen, a pale-green complex is found, Fe(tripyam), (C104)2*3H20 with bidentate ligands.(Found C, 52.0; H, 4.06; N, 16.0%. C45H42C12FeN12011 requires C, 51.2; H, 4.00; N, 15.9Yo.) Synthesis of Copper(@ Complexes. Blue [Cu(tripyam),] (C104)2 and yellow-green [Cu(tripyam),(ClO,),] were pre- pared as reported previ~usly.~ Reactions were also carried out between 4- and 6-methyl-di-(2-pyridyl)aminesand cop- per@) perchlorate. Although an earlier method was followed, the products isolated here were hydroxo-bridged dimers of the type [CuL(OH)],(ClO,), where L is 6-methyl-di-(2-pyridyl)-amine. {Found (4-methyl): C, 43.3; H, 3.12; N, 12.2%. C32H30C12CU2N,@10 requires C, 43.4; H, 3.40; N, 12.6%. Found (6-methyl): C, 42.2; H, 3.22; N, 11.9%. C32H32C12CU2N@ll (i.e.monohydrate) requires C, 41.7; H, J. MATER. CHEM., 1992, VOL. 2 3.69; N, 12.1 YO.)Conductivity measurements in acetonitrile confirmed that the complexes were 2 :1 electrolytes based on the dimeric formula, thus [CU,L,(OH)~](C~O~)~, molA dm-3)=278 S mol-' dm3 cf. [COL~](C~O~)~, molA dm-3)=280 S mol-' dm3. Exchange of Complexes onto Laponite Laponite RD was used in the sodium form. Method 1. Laponite (5 g) was added to deionised water (100 cm3) and treated with [C~(tripyam)~](ClO,), (0.75 g) dissolved in deionised water (150 cm3). The mixture was placed in a conical flask which was sealed and set on a mechanical shaker for 1 week. The clay assumed a pink colouration, the supernatant liquid, initially orange, was colourless.The clay was separated, washed with de-ionised water, and air dried. Other metal complexes were exchanged onto laponite using a similar procedure. Method 2. Laponite (1 g) was added to absolute ethanol (10 cm3) and 0.12 g [C~(tripyam)~](ClO,), was added. The mixture was placed in a Teflon reaction vessel and sealed tightly. The vessel was placed in a Sharp Carousel I1 R-84- 801 650 W microwave oven and was subjected to five bursts of 1 min of microwave energy (maximum power) with a delay of 1 min between bursts. On cooling, the vessel was opened and the orange clay was separated and washed with ethanol, then air dried. Other complex ions were exchanged using the same procedure. Some observations on the products are summarised in Table 1.Physical Measurements Spectra. Diffuse reflectance spectra of clays were measured with a Pye-Unicam SP.800 instrument using MgO as a reference. Some EPR spectra were measured with a JEOL PE-1X type spectrometer. Measurements at variable tempera- ture were provided by Dr. K. D. Sales of Queen Mary and Westfield College. IR spectra were recorded for KBr discs on Nujol mulls using a Perkin-Elmer 17 10 FTIR instrument. X-Ray Powder Diflraction. X-Ray powder diffraction (XRD) patterns were recorded with a Philips X-ray diffractometer using Co-Kcr radiation. Some data for the basal spacing [d(OO l)] of complex-ion-exchanged laponite are incorporated in Table 1. Electrochemical Measurements. Cyclic voltammetry measure- ments were made on Princeton Research Model 273 or 362 scanning potentiostats linked to a Rikadenki XY recorder. Potentials are cited relative to SCE.In all cases the supporting electrolyte was 0.1 mol dm- tetrabutylammonium perchlor- ate (TBAP) in dry CH3CN solvent, degassed with N2 prior to scanning. For all voltammetry, the supporting electrolyte was 0.1 mol dm -TBAP in CH3CN. Clay-modified electrodes were made by sonication of laponite (10 g) in 100 cm3 deion- ised water for 30 min. The resultant gel was centrifuged at 4000 rpm for 3 h to remove large particles. Platinum dichlor- ide (0.05 g) and polyvinylalcohol (0.66 g) were added to a 40 :60 ethanol-de-ionised water mixture (50 cm3) and refluxed for ca. 4 h to form a homogeneous mixture to which 5 g of the clay gel was added. Reflux was continued for 1 h.A platinum plate (1 cm xl cm) which had been cleaned by sonication was treated on one side with a few drops of the clay-PVA-Pt mixture which was allowed to dry before the reverse surface was treated similarly. The electrode was then J. MATER. CHEM., 1992, VOL. 2 exchanged complex CCu(tripy am)2l(Clo4)2 (blue isomer) CCu(tripyam),l(CIO,),(green isomer) [Cu(4-mpdpa)(OH)](C104) & 0.1 A. [Co(tripyam),](C10,), Table 1 Some properties of complex cation-exchanged laponite colour diffuse room higher reflectance temperature temperature ( "C) d(O01)/k maximum/nm ~~~~~ green blue (100) 15.1 ca. 420sh reversible green blue (100) 15.1 ca. 420sh reversible green blue (145) ca.15.5 reversible green-brown orange-brown (21 5) ca. 15.5 irreversible pink 15.1 ca. 500 orange 15.1 ca. 440 vbr orange 15.1 red-orange 15.1 ca. 420 sh, ca. 500 sh 13.5 crystals, A,,, =460 nm from ref. 4. dipped into a 0.1 mol dm-3 solution of the desired complex for 1 h, washed well with distilled water, and air dried. Some numerical data are given in Table2. To examine possible differences between the electrochemical response of CMEs dipped into solutions of redox active cations and those prepared from clay specimens previously ion exchanged, two electrodes were prepared with clay that had initially been exchanged with complex cation i.e. lap~nite/[Co(tripyam)~]~-+ (prepared by Method 1) and laponite/[Co(bipy),] (prepared+ by Method 1, complex synthesised following Burstall and Nyholm' 3).Some typical voltammograms are illustrated in Fig. 1, 2. Miscehzneous. Magnetic measurements were taken at room temperature by the Gouy method. Conductivity measurements were made on lop3mol dm-3 solutions using a Mullard bridge in conjunction with a bright platinum-dip electrode. New 57Fe Mossbauer data were provided by Dr. F. J. Berry (University of Birmingham). Chemical isomer shift data cited in the text are relative to Na,[Fe(CN),(NO)] *2H20to facili- tate comparison with earlier data; to convert to iron metal standard, subtract 0.257 mm s-Discussion The spin cross-over ("TH~E) behaviour of [Co(tripyam),] (C104), has been do~umented.~~" When freshly prepared, the value of peff is 4.84~~,~a value somewhat lower than values normally observed for relatively symmetrical cobalt(I1) octa- Table 2 Electrochemical data for cobalt complexes in solution and as components of CMEs (laponite) comp1ex CCo(tripy am)2l(C104)2 solution (Me CN) CME (laponite): dip-coated CCo(6-mPdPa),I(C104)zsolution (Me CN) CME (laponite): dip coated CCO(biPY)31(c104)3solution (Me CN) CME (laponite): dip coated intercalated species [C~(bipy)~]~+ (CME/laponite) [Co( tr~pyam),(H,O),]~ (CME/laponi te) + El ,,/V (us. S.C.E.) 0.50 0.49 0.39 no satisfactory CV 0.2 1 0.205 electroinactive electroinactive -0.5 0 +a5 +in +Fig.1 [Co(tripyam),]' immobilised on a laponite-modified elec- trode, prepared by dipping the electrode into an acetonitrile solution of the complex.(a) 500; (b) 200; (c) 100; (d) 50 pV s-' hedral complexes (i.e. >5 pB). This fact, together with the observation of weak IR spectral bands similar to those of low-spin [Fe(tripyam)2](C104)2 in addition to a more domi- nant set similar to high-spin [Ni(tripyam)2](C104)2, led Kulas- ingamI4 to suggest that some low-spin isomer was trapped in the initially prepared specimen. However, subsequent vari- able-temperature measurements5*" showed Curie-Weiss behaviour down to 200 K, a transformation to the low-spin form occurred between 190 and 143K." The freshly prepared [C~(tripyarn),](ClO,)~ used in this work was subjected to EPR analysis. A weak anisotropic signal was seen at room temperature centred on g=2, but after 1 month the signal had become broad and isotropic.After 3 months the complex was EPR silent at room temperature but after it had been J. MATER. CHEM., 1992, VOL. 2 (b1 Fig. 2 Voltammograms from a laponite-modified electrode, laponite pre-exchanged with [Co(bipy),13+. (a) 20 pA cm-', 100 mV s-'; (b)50 pA cm-' 200 mV s-' cooled a broad resonance was seen to arise at 190 K (full-width at half-maximum, FWHM =493 Hz). As cooling con- tinued, the intensity of the signal increased and the linewidth narrowed, Maximum intensity was reached at 145 K. At 105 K, g=2.1285 and FWHM =239 Hz. Thus, it is confirmed that the correct material had been prepared and Kulasingam's astute observation is also confirmed: it appears that some low-spin isomer is 'frozen' into the freshly prepared solid but, on passage of time, a true room-temperature equilibrium favouring the state is reached. Attempts to exchange [Co(tripyam),](ClO,), onto laponite raised further problems.A conventional solution-contact method (Method 1) gave a pink clay with a reflectance spectrum (Table 1) distinct from that of the orange [Co(tripyam),12 +. The difficulty of oxidising this cobalt(1r) complex has been mentioned previously;" thus, a change of terdentate to bidentate ligands was a more likely explanation. [Co(tripyam),](ClO,), in water gave, on shaking for 1 week, a pink solution. Comparison of the visible spectrum for this solution with that of [C~(pyridine),(H,O),]~ and the diffuse + reflectance spectrum of the pink clay (experimental section) leaves little doubt that the exchanged cation is diaquo{bis[tri- 2-pyridylaminecobalt(r1)]>with bidentate amine ligands.It appears that the clay surface has a greater affinity for this ion than for the bis(terdentate) species. The ease of aquation reflects the fact that two Co-N bonds of [Co(tripyam),] (ClO,), are significantly longer at 2.152(2)A than the other four at 2.100(2) A.6 The copper(I1) perchlorate complexes of tri-2-pyridylamine may be prepared in two isomeric forms, i.e. blue [Cu(tri- pyam),](ClO,), with terdentate ligands and, yellow-green [C~(tripyam),(ClO,),]~ with bidentate ligands, but aqueous solutions of both contain [Cu(tripyam),(H,O),]' with+ bidentate ligands and this is the species which exchanges onto laponite.The observed basal spacings, d(O0l), are very similar +for [M(tripyam)2(H20)2]2 (where M =Co, Cu) exchanged clays (Table 1). This raised the question of whether the bis(terd- entate) form of these complexes could be accommodated by the clay. Accordingly the low-spin [Fe(tripyam),](C10,)2 was prepared. The contamination with a hitherto unrecognised tris[tri-2-pyridylamine]iron(11)complex with bidentate ligands has been detailed. The complex with terdentate ligands exchanged onto laponite unchanged but gave a basal spacing similar to those observed for [M(tripyam)2(H20)2]2 + in which bidentate ligands are present. It is possible that the terdentate co-ordinated ligand complexes lock into the pseudo-hexagonal rings of the silicate layers.When the copper-complex-exchanged clays were heated to 100 "C a greenjblue change occurred corresponding to a bidentatej terdentate change. This change is completely reversible. The temperature at which the colour change occurs may be altered by turning to other ligands, e.g. 4- and 6-methyl-di-(2-pyridyl)amine.Some new complexes of cop- per@) were serendipitously prepared ([CuL(OH)]i +) and exchanged onto laponite. A reversible colour change occurred at 145 "C for the 4-methyl ligand, and an irreversible change occurred at 215 "C for the 6-methyl ligand. Thus, the point is illustrated that thermochromic clays may be prepared by exchange with metal complexes of ligands with variable denticity, and the temperature of the colour change may be tuned.The change is not instantaneous, but the robustness of the materials suggests that they could find application as hazard warnings of excess temperature in or on appropriate equipment. The use of microwave heating in chemistry laboratories can produce remarkable increases in the rates of organicI5 and organometallic'6 reactions and very remarkable heating of some inorganic substrates may OCCU~.'~ It was demonstrated recently that the process of ion exchange onto clay may be greatly accelerated. For example, lithium-exchanged laponite may be produced in 5 min in the microwave oven1* compared with several weeks using the classic method of Posner and Quirk.lg Furthermore, the rapidity of the microwave method may reveal chemistry which could be missed during the passage of time required by conventional exchange methods.l8 In this paper it has been shown that microwave heating (method 2) can greatly accelerate the exchange process. Thus, when laponite is treated with ethanolic [Co(tripyam),](ClO,), in the microwave oven for 5 min, an orange clay is produced, the diffuse reflectance spectrum of which is consistent with the presence of [Co(tripyam),12 [bis(terdentate) ligands]. + However, on storing this material for 12 months the aquation reaction to give pink [C~(tripyam),(H,O),]~ [bis(bidentate)+ ligands] occurs slowly on the clay surface.This strengthens the view that the clay has a greater affinity for the bis(bident- ate) form, [M(tripyam),(H20),l2 +. The microwave experi- ment has apparently greatly accelerated the exchange reaction, but not the aquation reaction. In comparison with [Co(bipy)J2 +,the corresponding com- plex of tri-2-pyridylamine, [Co(tripyam),I2 +,was much more resistant to oxidation, indeed the synthesis of cobalt(w) tri-2- pyridylamine complexes has not yet been achieved. Yet, by contrast the preparation of cobalt(m) complexes of substituted tri-Zpyridylamines was relatively easy.' The opportunity was taken to compare the cyclic voltammetric behaviour of these complexes for acetonitrile solutions of the cobalt(@ complexes both in free solution and as CME containing the compounds. The solvent for all electrochemistry was acetonitrile and solutions of the cobalt(I1) complexes in acetonitrile were carefully monitored by visible absorption spectroscopy to ensure that no change of ligand denticity occurred. The preparation of the CME followed a procedure developed by Ghosh and co-workers20*21 which involved a preliminary deposition of the clay film on a platinum electrode followed J.MATER. CHEM., 1992, VOL. 2 by dipping the prepared electrode in an acetonitrile solution of the complex. Using an acetonitrile solution of [Co(tripyam),12 (terdentate ligands), the electrode assumed + the characteristic orange colour of the bis(terdentate) complex. In order to compare the electrochemical behaviour of the pink bis(bidentate) complex, it was necessary to prepare the electrode from the pre-exchanged and rigorously washed clay.Electrodes were also prepared from bis[6-methyl-di-(2-pyridyl)amine]cobalt(r~) ([Co(6-dmdpa),l2 '1, dipped in acetonitrile solution, and tris(2,2'-bipyridyl)cobalt(111) {[C~(bipy)~]~both dipped and pre-exchanged. The El /2+>, values obtained from the cyclic voltammograms of the solu- tions are detailed in Table 2. The Co2 redox process was + investigated and the data confirm that the ease of oxidation is in the order: + +[Co(bipy),] + >[C0(6-dmdpa)~]' >[Co(tripyam),] , thus the difficulty of synthesising [Co(tripyarn),l3 is+ expressed more quantitatively. The greater basicity expected for the 2-pyridyl group bearing the 6-methyl substituent should lead to greater CT bond strength compared to that for the co-ordinated unsubstituted 2-pyridyl groups.However, the difference may be more significant for Co"' than for Co". The dipped CME, prepared from [Co(tripyam),12 and+ [Co(bipy),13+ gave good-quality cyclic voltammograms with EIl2 data in close agreement with the results obtained from solution measurements. Despite a number of electrode preparations, the cyclic voltammograms obtained from [C0(6-dmdpa)~]~+/CME were unsatisfactory. The attempt to compare the redox behaviour of the bis-(terdentate) [Co(tripyam),12' and the bis(bidentate) [C~(tripyarn),(H,O),]~+ was of necessity confined to CME studies; also the bidentate species was only available pre- exchanged on laponite.The resulting CME was electroinactive (Table 2). King et have demonstrated that cations, bound electro- statically to any exchange site on a smectite clay, are rigorously electro-inactive. The electroactivity which characterises CMEs was attributed to cations which are surface bound, in excess of the clay cation-exchange capacity, by an ion-pair mechan- ism. A recent helpful review23 concludes that the electroactiv- ity of a CME will depend on factors such as those detailed by King et and also on the method of film preparation. A CME was therefore prepared using the method detailed in the experimental section with laponite which had been pre- exchanged with [Co(bipy), J3 +;this electrode was also rigor- ously electro-inactive. We therefore conclude that, using our conditions of electrode preparation, the electroactivity is dependent on sorbed ion pairs from dipping the CME in acetonitrile solutions of the complex.Unfortunately therefore, it is not possible using this methodology to study the redox behaviour of any complex which can exist on the clay exchange sites, but not in acetonitrile solution. S.P.B. thanks SERC for a studentship. We thank Dr. K. D. Sales (Queen Mary and Westfield College, London) and Dr. F. J. Berry (University of Birmingham) for variable-tempera- ture EPR data and Mossbauer data, respectively. References I N. Davison and W. R. McWhinnie, Znorg. Chim. Acta, 1987,131, 9.2 W. E. Rudzinski and A. J. Bard, J. Electroanal. Chem., 1986,199, 323. 3 C. Shi, J. F. Rusling, Z. Wang, W. S. Willis, A. M. Winiecki and S. L. Suib, Langmuir, 1989, 5, 650. 4 W. R. McWhinnie, G. C. Kulasingam and J. C. Draper, J. Chem. SOC.A, 1966, 1199. 5 P. F. B. Barnard, A. T. Chamberlain, G. C. Kulasingam, W. R. McWhinnie and R. C. Dosser, J. Chem. SOC., Chem. Commun., 1970, 520. 6 E. S. Kucharski, W. R. McWhinnie and A. H. White, Aust. J. Chem., 1978,31, 2647. 7 E. S. Kucharski, W. R. McWhinnie and A. H. White, Aust. J. Chem., 1978, 31, 53. 8 J. C. Lancaster and W. R. McWhinnie, Znorg. Chim. Acta, 1971, 5, 515. 9 J. P. Wibaut and G. L. C. La Bastide, Res. Trau. Chim., 1933, 52, 493. 10 J. C. Lancaster and W. R. McWhinnie, J. Chem. SOC. (C), 1970, 2435. 11 D. F. B. Barnard, J. C. Lancaster, M. E. Fernandopulle and W. R. McWhinnie, J. Chem. SOC.,Dalton Trans., 1973,2172. 12 W. R. McWhinnie, R. C. Poller and M. Thevarasa, J. Chem. SOC. (A), 1967, 1671. 13 F. H. Burstall and R. S. Nyholm, J. Chem. SOC.,1952, 3570. 14 G. C. Kulasingam, Ph.D. Thesis, University of London, 1967. 15 R. N. Gedye, F. E. Smith and K. C. Westaway, Can. J. Chem., 1988,66, 17. 16 M. Ali, S.P. Bond, S. A. Mbogo, W. R. McWhinnie and P. W. Watts, J. Organomet. Chem., 1989, 371, 11. 17 D. R. Baghurst and D. M. P. Mingos, J. Chem. SOC., Chem. Commun., 1988, 829. 18 S. P. Bond, A. Gelder, J. Homer, W. R. McWhinnie and M. C. Perry, J. Mater. Chem., 1991, 1, 327. 19 A. M. Posner and J. P. Quirk, Proc. R. SOC. London, A, 1964, 278, 35. 20 P. K. Ghosh and A. J. Bard, J. Am. Chem. SOC.,1983,105,5691. 21 P. K. Ghosh, A. W.-H. Man and A. J. Bard, J. Electroanal. Chem., 1984, 169, 3 15. 22 R. D. King, D. G. Nocera and T. J. Pinnavaia, J. Electroanal. Chem., 1987,236,43. 23 A. Fitch, Clays Clay Miner., 1990, 38,391. Paper 1/02894D, Received 17th June, 1991