Lamellar polymer–LixMoO3 nanocomposites via encapsulative precipitation Lei Wang,a Jon Schindler,b Carl R. Kannewurfb and Mercouri G. Kanatzidis*a aDepartment of Chemistry and the Center for FundamentalMaterials Research,Michigan State University, East L ansing,Michigan 48824, USA bDepartment of Electrical Engineering and Computer Science, Northwestern University, Evanston, Illinois 60208, USA With LixMoO3 (x=0.31–0.40) as a host material, a new family of polymer–molybdenum bronze nanocomposites has been synthesized using an exfoliation/encapsulation methodology.Nanocomposites with poly(ethylene oxide), poly(ethylene glycol), poly(propylene glycol), poly(vinylpyrrolidinone), methyl cellulose, polyacrylamide, and nylon-6 were prepared and characterized by thermal gravimetric analysis (TGA), dierential scanning calorimetry (DSC), powder X-ray diraction, FTIR spectroscopy, UV–VIS spectroscopy, variable-temperature 7Li and 13C solid-state NMR spectroscopy and magnetic susceptibility measurements.The electrical conductivity of these materials lies in the range from 10-2 to 10-7 S cm-1, and decreases as the interlayer separation increases.The intercalated polymer imparts both mechanical strength and ease of processing to these materials. The water-soluble polymer–LixMoO3 nanocomposites can be cast into films and other shapes, which may provide opportunities for applications. Factors aecting the intercalation reaction and the structure of nanocomposites, such as variations in the preparation procedure, the polymer molecular mass and the annealing behaviour of the products are discussed.The investigation of polymer–inorganic nanophase composites able-temperature solid-state 7Li and 13C NMR spectroscopy, magnetic susceptibility measurements and electrical conduc- is motivated by many reasons, including the need for novel tivity measurements. electronic anisotropic materials, better performing battery cathode materials, functionalized structural materials with superior mechanical properties, hierarchical materials, and systems in Experimental which to study polymer orientation, epi- and endo-taxy and polymer/inorganic surface interactions.1–28 Polymer-based Materials and methods nanocomposites have been reported with layered silicates Reagents.PEO (5000000), PEO (100000), PEG (10000), (e.g.montmorillonite, hectorite, etc.),1–9 FeOCl,10 V2O5,11–13 PEG (2000), PEG (400), PPG (1000), PVP (10000), MCel MoO3,14,15,26,27 layered metal phosphates,16–20 MS2 (M= (63000), PAM (5000000), PA-6 (10000) and LiBH4 were Mo,21–24,26 Ti24), layered metal phosphorus chalcogenides purchased from Aldrich Chemical Company, Inc. After the (MPS3)25–27 and layered double hydroxides.28 The most polymers were dissolved, the polymer solutions were filtered to common methods of preparing polymer–inorganic nanocomremove insoluble polymer residues.MoO3 (99.95%) was pur- posites are (a) by monomer intercalation followed by polymer- chasedfrom JohnsonMatthey.Anhydrousdiethyl ether(99.0%), ization, (b) by in situ intercalative polymerization, (c) by direct 2,2,2-trifluoroethanol (99%), acetonitrile (99.5%) isopropyl insertion and (d) by encapsulative precipitation from solutions alcohol (99.9%) and absolute ethanol were from Columbus of exfoliated lamellar solids.The last two methods give poly- Chemical Industries Inc., Lancaster Synthesis Inc., EM Science mer–inorganic nanocomposites in which the molecular masses Inc., Mallinckrodt Chemical Inc.and Quantum Chemical and nature of the polymers can be decided before intercalation. Company respectively. No further purification was applied to Encapsulative precipitation has been applied with V2O5,11a–c,12 the chemicals above. Water used in the reactions was distilled MoO3,14a–c,26 MoS2,22b,c,23,24,26 WS2 ,29 and TiS2,24 in combi- water provided by the Department of Chemistry of MSU, and nation with various polymers. was degassed by bubbling nitrogen for 30 min before use.MoO3 is one of the layered metal oxides which shows reversible Li ion insertion properties which are relevant to rechargeable Li batteries.30,31 In intercalative electrodes of Synthesis of LixMoO3 ( 0.30<x<0.40). Commercial MoO3, rechargeable batteries, ion conductivity is very important.used as the starting material, was fired in an open quartz vial Other solid-state ionic applications such as electrochromic in the air at 600 °C for 36 h during which the crystallite size devices also need good ion conductivity. Polymer–inorganic of MoO3 remarkably increased. This MoO3 was used to react nanocomposites should exhibit fast ion conduction8 and intro- with LiBH4 to prepare the lithium molybdenum bronze.32 In duction of a polymer with anity for Li ions between the a typical reaction, 0.1 mol of MoO3 was reacted with 0.04 mol sheets of MoO3 could improve its performance as an inter- of LiBH4 in 80 ml diethyl ether for 24 h, under a nitrogen calative electrode.atmosphere. The product was collected by filtration, washed Polymer insertion into MoO3 has been reported previously, with ether and dried in vacuo.The yield was >98%. The namely, with a polymeric ionomer,14a,b with PEO14c,26 and lithium molybdenum bronze was thereafter stored in a nitrogen with polyaniline.14d,15,27 This research develops further the dry-box. The X-ray powder diraction pattern can be indexed polymer intercalation chemistry associated with MoO3 and on the basis of an orthorhombic cell similar to that of MoO3 introduces a new family of polymer–LixMoO3 nanocomposites.but expanded along the stacking (a-axis) direction, with a= Nanocomposites with poly(ethylene oxide) (PEO), poly(ethyl- 16.528 A° , b=3.775 A° and c=3.969 A° . The dhkl-spacings (A° ) are: ene glycol) (PEG), poly(propylene glycol) (PPG), poly(vinyl- 8.27200 (vs), 4.132400 (s), 3.578201 (m), 3.434210 (m), 2.755600 (w), pyrrolidinone) (PVP), methylcellulose (MCel), polyacrylamide 2.453311 (m) and 2.372411 (m).(PAM) and nylon-6 (PA6) are reported here. These nanocom- The amount of lithium in the bronze was analysed by TGA posites were characterized by thermal gravimetric analysis under oxygen flow, heating up to 650°C, and by elemental (TGA), dierential scanning calorimetry (DSC), powder X-ray analysis which was accomplished by Oneida Research Services, Inc., Whitesboro, New York.Elemental analysis of Li in the diraction, FTIR and solid-state UV–VIS spectroscopy, vari- J. Mater. Chem., 1997, 7(7), 1277–1283 1277molybdenum bronze was done by ion chromatography, while with a Quantum Design MPMS2 SQUID magnetometer.Samples were sealed in low-density polyethylene (LDPE) bags Mo was measured by X-ray fluorescence. under a nitrogen atmosphere. Room-temperature conductivity measurements were performed on pressed sample pellets with Preparation of polymer–LixMoO3 nanocomposites. The LixMoO3 was exfoliated in degassed water by 5 min of sonic- a four-probe detector connected to a Keithley-236 source measuring unit.Variable-temperature dc electrical conductivity ation, to form a suspension with a concentration of 5 g l-1. This suspension was added dropwise into a stirred polymer measurements were performed on compacted powders in pellet form with 60 and 25 mm gold wires used for the current and solution of the same volume, which contained five times excess of polymer (per repeat unit) to MoO3 and the mixture was voltage electrodes, respectively.Measurements of the pellet cross-sectional area and voltage probe separation were made stirred for 2 days under a nitrogen atmosphere. The nanocomposites formed were isolated in dierent ways according to with a calibrated binocular microscope. Electrical conductivity data were obtained with a computer-automated system their behaviour in solution.Nanocomposites of methylcellulose, polyacrylamide and nylon-6 precipitated and were col- described elsewhere.33 lected by filtration. Those containing MCel and PAM were washed with water, while the nanocomposites of PA6 were Results and Discussion washed with trifluoroethanol. Nanocomposites with poly- (ethylene oxide), poly(ethylene glycol), poly(propylene glycol) Synthesis and characterization of LixMoO3 and poly(vinylpyrrolidinone) remained in colloidal form and The lithium molybdenum bronze exfoliates readily in water to were isolated by pumping o the water to dryness.The dried form stable colloidal solutions, and this makes it an appealing material was stirred in an appropriate solvent for several hours candidate for polymer intercalation. The LiBH4 route to to dissolve the extraneous polymer.The solid product was LixMoO3 is the best one to date in providing the material filtered and was washed again with the solvent. The solvents conveniently and in high yield. The previously reported method used to process the products are listed in Table 1.The products for Lix(H2O)yMoO334 involves one step to prepare were pumped to dryness and stored in a nitrogen atmosphere. Nax(H2O)yMoO3 and two steps to accomplish ion exchange and gives only 26% yield.† An additional advantage of the Instrumentation LiBH4 method is that it is conducted in diethyl ether and so provides the anhydrous form of the molybdenum bronze. The X-Ray powder diraction patterns were obtained on a Rigaku Ru-200B X-ray diractometer equipped with graphite-mon- LixMoO3 product prepared in this fashion, though still crystalline, shows broader diraction maxima than the precursor ochromated Cu-Ka radiation.A scanning speed of 1° min-1 was chosen. TGA was performed with a Shimadzu TGA-50 MoO3. The Li insertion into MoO3 is topotactic as evidenced by our ability to fully index the X-ray powder diraction instrument under a nitrogen or oxygen flow at a rate of 46 ml min-1 and the heating rate was 10°C min-1.DSC was carried pattern of the product. The amount of Li in LixMoO3 was determined both by out on a Shimadzu DSC-50 instrument under a nitrogen flow at a rate of 20 ml min-1. The heating and cooling rates were TGA and by direct elemental analysis.When the material is heated in an oxygen atmosphere to 650 °C, it gains 1.71 mass%, 5°C min-1. Sample cells were made of aluminium and were annealed at 450 °C in nitrogen after they were cleaned. Samples owing to a change from LixMoO3 to Li2O and MoO3. This gain of mass is reproducible and corresponds to an x value of were sealed in cells under a nitrogen atmosphere before measurement.IR spectra were collected on a Nicolet IR/42 0.31(3). On the other hand, the elemental analysis showed that the molybdenum bronze consisted of 1.64% Li and 56.39% FTIR spectrometer at a resolution of 2 cm-1. Generally 64 scans were obtained for samples as KBr pellets. Electronic Mo. This gives a ratio of Li to Mo of 0.451 corresponding to the molar ratio of LiBH4 used in the lithiation reaction.It is transmission spectra were recorded with a Shimadzu UV- 3101PC UV–VIS–NIR scanning spectrophotometer. Samples of course possible that x varies slightly from sample to sample in the range 0.3<x<0.4. were dried as thin films on quartz slides. Variable-temperature solid-state 7Li and 13C NMR measure- The lithium molybdenum bronze is metastable and undergoes an intense, irreversible and exothermic phase change ments were taken on a 400 MHz Varian NMR instrument.Samples were loaded in a glove box under a nitrogen atmosphere. Magnetic susceptibility measurements were performed † Yield obtained by reproducing the experiment in ref. 34. Table 1 Some chemical and structural characteristics of polymer–LixMoO3 nanocomposites washing expansion of coherence nanocomposite polymer MWa solubility solvent d-spacing/A° gallery/A° length/A° PEO–LixMoO3 5000 000 yes MeCN 16.6 9.7 64 PEO–LixMoO3 100 000 yes MeCN 16.6 9.7 121 PEG–LixMoO3 10 000 yes MeCN 16.8 9.9 108 PEG–LixMoO3 2000 yes MeCN 13.8 6.9 72 12.6–14.7 5.7–7.8 —b PEG–LixMoO3 400 yes MeCN 13.5 6.6 111 12.6 5.7 97 PPG–LixMoO3 1000 yes EtOH 17.2 10.3 113 11.8 4.9 108 18.0, 11.5 — —b MCel–LixMoO3 63 000 no H2O 27.6 20.7 92 PVP–LixMoO3 10 000 yes PriOH 38.6 31.7 154 PAM–LixMoO3 5 000 000 no H2O 38.0 31.1 69 33.8 26.9 80 41.2 34.2 —b PA6–LixMoO3 10 000 no CF3CH2OH 22.1 15.2 39 16.8 9.9 26 aPolymer dissolved in water, except for PA6-10 000, dissolved in 2,2,2-trifluoroethanol.bPeaks too broad to obtain estimate. 1278 J. Mater. Chem., 1997, 7(7), 1277–1283at 356 °C, as detected by DSC (Fig. 1) and X-ray powder diraction. The product appears to be a mixture of at least two unknown phases. As this mixture is heated to higher temperatures it undergoes additional phase changes yielding other new phases. The details of this reaction are currently under investigation. Exfoliation and polymer encapsulation chemistry The lithium molybdenum bronze described above can be readily exfoliated in water after several minutes of sonication.The exfoliated product forms supramolecular complexes with most water-soluble polymers. The complexes form solutions or precipitates in water, depending on the type and molecular mass of the polymers.We have encapsulated PEO, PEG, PPG, PAM, MCel and PVP inside the lithium molybdenum bronze to obtain lamellar nanocomposite materials.Poly(ethyleneimine) (PEI) could not be successfully intercalated. Instead, the blue Lix(H2O)yMoO3 monolayer suspension decolorized and totally dissolved in the aqueous PEI solution. The same phenomenon occurred when ammonia was introduced in the Lix(H2O)yMoO3 suspension, suggesting that the PEI solution is too basic and attacks the MoO3 lattice. Waterinsoluble polymers were also tried, however of these, only PA6 was successfully intercalated.Details of the reactions are given in Table 1. The existence of polymer chains between the layers of the lithium molybdenum bronze was verified by IR spectroscopy Fig. 2 IR spectra of (a) poly(vinylpyrrolidinone), (b) Lix(H2O)y(PVP- and X-ray powder diraction.Fig. 2 shows a comparison of 10000)zMoO3 and (c) hydrated lithium molybdenum bronze the IR spectra of a nanocomposite Lix(H2O)y(PVP)zMoO3 and its components; from this, it is obvious that the vibrational spectrum of Lix(H2O)y(PVP)zMoO3 is a combination of the vibrational spectrum of PVP and that of Lix(H2O)yMoO3. The positions of the vibrationalpeaks arising from the encapsulated PVP are close to those of pure PVP, while the positions of the peak due to the MoNO stretch [for Lix(H2O)yMoO3] is shifted to higher wavenumbers, suggesting that the MoO3 layers in the nanocomposite are slightly more oxidized.The optical transmission absorption spectra of these macromolecular intercalates were examined. The dark-blue colour of these systems arises from the intense intervalence transitions associated with the Mo5+–Mo6+ couple.These electronic transitions are broad and range from the IR region to the visible (Fig. 3) and are responsible for the electrical conductivity of these materials. The absorption at 286 nm arises from excitations across the band-gap from the O2- p band to the Mo6+ d band and is present in all compounds including pristine MoO3 and Lix(H2O)yMoO3 .This is consistent with the expectation that host metal oxide structure is not disturbed upon intercalation. The encapsulation of polymers inside the interlayer galleries Fig. 3 Solid-state optical absorption spectra of the polymer–LixMoO3 nanocomposites: (a) MoO3, (b) Lix(H2O)yMoO3, (c) Lix(H2O)y(PEO- 100000)zMoO3, (d) Lix(H2O)y(PAM-5 000000)zMoO3 of MoO3 is also demonstrated by X-ray powder diraction, which shows an expansion of the gallery space.Fig. 4 and 5 show typical XRD patterns of some of the nanocomposites. The sharp and intense (001) reflections indicate that the MoO3 layers are well stacked. X-Ray scattering coherence lengths, which are calculated from the Scherrer formula L hkl= Kl/bcosh,35 and the gallery spacings are given in Table 1.The basal spacing of some nanocomposites depends on the preparation procedure. For example, Lix(H2O)y(PVP)zMoO3, which is water soluble, showed a d-spacing of 59.0 A° before Fig. 1 DSC diagram of LixMoO3 washing with isopropyl alcohol and 38.3 A° after washing. J. Mater. Chem., 1997, 7(7), 1277–1283 1279to control, as for example in PEO(5000 000), PEG(2000) and PPG(1000).The Lix(H2O)y(PEO-5 000 000)zMoO3 had a consistent d-spacing of ca. 16.6 A° , but the peaks were broad. Lix(H2O)y(PEG-2000)zMoO3 showed broad peaks in the range 12.6–14.7 A° . Lix(H2O)y(PPG-1000)zMoO3 sometimes showed a peak in the range 11.8–17.2 A° , while other samples showed two peaks in this range suggesting a mixture of phases.Evidently, for low molecular masses the polymers are mobile enough in the galleries to form several dierent arrangements leading to multiple phases. The eect of polymer molecular mass on product formation was examined, particularly with PEO and PEG, and was found to be significant. The high molecular mass PEO(5000 000) immediately formed a precipitate with lithium molybdenum bronze in water upon mixing while this phenomenon did not occur with PEO of lower molecular mass.The molecular mass also aects the structure of the nanocomposites. Table 1 shows that the Lix(H2O)y(PEO- 5000 000)zMoO3 sample has a lower coherence scattering length than its lower molecular mass analogues, which is attributed to the fact that it is kinetically unfavourable for extremely long polymer chains to align in an ordered structure.When the molecular mass is extremely low, i.e. in oligomer range, the gallery expansion of the intercalate is lower, almost one half of that of the long polymers. For PEO and PEG, a molecular mass of 2000 is about the upper limit of this Fig. 4 Typical XRD patterns of nanocomposites with poly(ethylene situation.The data listed in Table 1 show that the glycol) and poly(ethylene oxide) of dierent molecular masses. The Lix(H2O)y(PEG-2000)zMoO3 sample has a much shorter polymer and its molecular mass are indicated on each spectrum. coherence length than its analogues with higher or lower molecular masses. An analogous Lix(H2O)y(PEG-2000)zMoO3 sample, prepared under similar conditions, had a broad X-ray basal peak which corresponded to d-spacings varying from 12.6 to 14.7 A° , suggesting that the local conformation of the polymer is important.Annealing the Lix(H2O)y(PEG)zMoO3 and the Lix- (H2O)y(PEO)zMoO3 samples at 150°C and then gradually cooling them to room temperature tends to improve their lamellar order, especially when the starting coherence length is short.For example, after annealing, the Lix(H2O)y(PEG- 2000)zMoO3 sample whose XRD pattern had a broad peak centred at 13.4 A° gave a pattern with a sharp peak at 12.7 A° , (Fig. 6). The water-insoluble nanocomposite Lix(H2O)y(PAM- 5000 000)zMoO3 gave samples with d-spacings of 38.0, 33.8 Fig. 5 Typical XRD patterns of the various polymer–LixMoO3 nanocomposites. The polymer and its molecular mass are indicated on each spectrum.Washing these materials may not only remove extra-lamellar polymer, but could lead to polymer loss from the galleries changing their composition. This behaviour makes it dicult to determine at what polymer loading we begin to saturate the intralamellar space. Similar phenomena were described in polymer–V2O5 xerogel systems.11a Nevertheless, the observed d-spacings were consistent to within ±1 A° , as long as the preparation procedure was not altered significantly from batch to batch.The d-spacings and the degrees of lamellar stacking Fig. 6 XRD patterns of Lix(H2O)y(PEO-2000)zMoO3 showing the of products which contained polymers at the very extremes of eect of annealing on the stacking regularity of the layered structure of a nanocomposite the molecular mass range (very high or very low) were hard 1280 J.Mater. Chem., 1997, 7(7), 1277–1283and 41.2 A° . The Lix(H2O)y(PA6-10 000)zMoO3 showed d-spac- of the solid LiCl standard. At -80 °C the lineshapes in the two spectra dier, with that of Lix(H2O)y(PEO-100 000)zMoO3 ings of 22.1 and 16.8 A° . Occasionally, in the intercalation of PA6 a mixed-phase material with basal spacings of 12.8 and being slightly more asymmetric, Fig. 7. This suggests that the presence of PEO causes a distribution of Li ions over several, 9.8 A° was obtained. This shows the diculty in controlling the reaction when quick precipitation is used to obtain a specific slightly dierent sites in the gallery. Some of the sites may involve coordination of water molecules while others are phase.To prepare nanocomposites of this type, a very dilute Lix(H2O)yMoO3 suspension of <0.5 mass% is recommended. associated with the ether-like oxygen atoms of PEO or even those in the MoO3 layers. The linewidth (width at half maxi- The polymer compositions of the nanocomposites were determined by TGA in an oxygen atmosphere and are listed mum) of the resonance peak is greater in the PEO intercalated sample than in the host LixMoO3 material and this too is in Table 2.The water in the nanocomposites was estimated by the mass loss step observed below 230 °C, and the amount of consistent with a distribution of the Li ions over several sites in the former. The data in the low-temperature region suggest polymer was determined by the mass loss steps observed at higher temperatures.Despite drying under vacuum, the a more well defined coordination environment for the Li ion in LixMoO3 as would be expected in a crystalline solid. nanocomposites retain some water in the galleries. The water-soluble nanocomposites, Lix(H2O)y(PEO)zMoO3, The linewidth of 2300 Hz in LixMoO3 at 23°C is much narrower than that of ca. 12000 Hz observed for Li2Mo2O4 Lix(H2O)y(PEG)zMoO3, Lix(H2O)y(PPG)zMoO3 and Lix- (H2O)y(PVP)zMoO3, usually contain 2–4 mass% water which indicating a substantial degree of ion motion in the lattice of LixMoO3 relative to that of Li2Mo2O4.38 This is rationalized is very dicult to remove. This water is thought to be coordinated to Li+ ions. The water-insoluble nanocomposites by the fact that in the latter the Li ions fully occupy well defined crystallographic positions in the lattice39 while the Lix(H2O)y(MCel)zMoO3 and Lix(H2O)y(PAM)zMoO3, however, contain much less water.non-stoichiometric nature of LixMoO3 gives rise to Li mobility via vacant crystallographic sites. Compared to the corresponding polymer–MoS2 intercalates, 22b most polymer–LixMoO3 intercalates have much higher The resonance peak in both samples narrows dramatically as the temperature is increased from -80 to 100 °C, Fig. 8. polymer contents and larger gallery spacings. As in the polymer –MoS2 case, PVP and MCel give the largest expansions. This is attributed to rapid motion of Li ions between the MoO3 layers. At 100 °C the peak linewidth in the spectra of A marked dierence is found in PAMwhich gives an expansion as large as 34.3 A° , while PAM–MoS2, a hybrid prepared by Lix(H2O)y(PEO-100 000)zMoO3 is comparable to that of Li0.5(H2O)1.3Mo2O4.38 The onset temperature of the transition us, has only an expansion of 9.1 A° .36 This confirms that multiple layers of this polymer can enter the accessible space from a wide to a narrow peak in Lix(H2O)y(PEO- 100000)zMoO3 and LixMoO3 is similar. The spectra of both of LixMoO3.The water-soluble polymer–LixMoO3 nanocomposites can be cast into films and other shapes, which may provide opportunities for applications. The nanocomposites with high molecular mass polymers are strong, though their mechanical properties depend on the polymer. For example, the nanocomposite of PEO(5000000) is tough, while that of PAM(5000 000) is hard.Lix(H2O)y(PEO-5000 000)zMoO3 can be swollen by acetonitrile and becomes resilient and plastic with the consistency of unsulfurized rubber. When the Lix(H2O)y(PAM-5 000 000)zMoO3 is swollen by water, it is not as plastic, but is stronger and tougher. Precise mechanical measurements have not been taken yet.37 Solid-state NMR spectroscopy In order to probe the eect of the polymer on the behaviour of the lithium ions in the gallery, variable-temperature 7Li solid-state NMR static spectra were measured for LixMoO3 Fig. 7 Static 7LiNMR spectra of (a)LixMoO3 and (b) Lix(H2O)y(PEO- and Lix(H2O)y(PEO-100 000)zMoO3. In both cases a broad 100000)zMoO3 at -80°C. The broader asymmetric line in the spectrum of the nanocomposite is evident in the upfield region.peak was observed with a chemical shift very similar to that Table 2 Composition and physicochemical properties of the polymer–LixMoO3 nanocomposites limit of limit of thermal thermal electronic polymer d-spacing/ composition stability stability conductivity/ nanocomposite MW A° (according to TGA) in N2/°C inO2/°C Scm-1 pure LixMoO3 8.27 1.3×10-2 pure MoO3 6.93 3.3×10-5 PEO–LixMoO3 5×106 16.6 Li0.25(H2O)0.20(PEO)0.83MoO3 260 220 2.4×10-5 PEO–LixMoO3 100 000 16.6 Li0.25(H2O)0.32(PEO)1.04MoO3 260 220 2.9×10-5 PEG–LixMoO3 10 000 16.8 Li0.25(H2O)0.29(PEG)0.75MoO3 260 220 5.2×10-5 PEG–LixMoO3 2000 13.8 Li0.25(H2O)0.28(PEG)0.57MoO3 260 220 2.2×10-4 PEG–LixMoO3 400 13.5 Li0.25(H2O)0.38(PEG)0.33MoO3 260 220 3.1×10-4 PPG–LixMoO3 1000 17.2 Li0.25(H2O)0.18(PPG)0.99MoO3 220 200 — 11.8 Li0.25(H2O)0.52(PPG)0.14MoO3 5.3×10-4 MCel–LixMoO3 63 000 27.6 Li0.25(MCel)0.70MoO3 180 170 2.0×10-6 PVP–LixMoO3 10 000 38.6 Li0.25(H2O)0.43(PVP)1.17MoO3 220 220 1.8×10-7 PAM–LixMoO3 5×106 38.0 Li0.25(PAM)3.2MoO3 150 150 6.3×10-7 PA6–LixMoO3 10 000 22.1 Li0.25(H2O)0.44(PA6)0.32MoO3 280 270 — 16.8 Li0.25(H2O)0.48(PA6)0.32MoO3 2.1×10-4 J.Mater. Chem., 1997, 7(7), 1277–1283 1281of LixMoO3 are more similar to those of Li0.9Mo6O1740 than of other molybdenum bronzes such as the blue bronze A0.3MoO3 (A=K, Tl) and purple bronzes A0.9Mo6O17 (A=K, Na, Tl).41–44 In the latter bronzes the temperature dependence of the magnetic susceptibility shows transitions at low temperature associated with charge density waves. Such phenomena were not observed in the LixMoO3 samples reported here.Electrical conductivity Electrical conductivity values for all the nanocomposites are listed in Table 2. The room-temperature electrical conductivity of LixMoO3 was 0.013 S cm-1 which is slightly lower than values reported in related materials.45,46 As shown in Table 2, the conductivity of the nanocomposites is significantly lower than that of LixMoO3 and decreases with increasing layer Fig. 8 Temperature dependence of the linewidth of the resonance peak expansion. For materials with extremely large gallery spacing, in the solid-state 7Li NMR spectra of (a) LixMoO3 and the electrical conductivity is lower than that of MoO3 itself. (b) Lix(H2O)y(PEO-100000)zMoO3 When the measurements are performed under vacuum the conductivity decreases concomitantly with the loss of water materials end up with an equally narrow peak at 100°C from the galleries.This suggests that water is contributing to indicating comparable rates of hopping of Li ions between charge transport in these materials probably via proton dierent positions in the interior of the two materials.mobility. Fig. 10 shows the temperature dependence of the Preliminary solid-state, cross-polarization, magic angle spin- electrical conductivity of LixMoO3 and Lix(H2O)y(PEO- ning 13C NMR spectra, as a function of temperature, suggest 100000)zMoO3, which is thermally activated. Mixed ionic/ a substantial degree of motion in the polymer backbone as electronic conducting nanocomposites are of current interest.12 well. At room temperature a single broad resonance is observed for the MCH2M group at d 71.6, with a linewidth of 1380 Hz.Conclusion As the temperature rises, the resonance does not shift but it narrows dramatically and at 80°C it is only 200 Hz, consistent Anew family of polymer–molybdenum bronze nanocomposites with rapid thermal motion of the polymer chains.This motion has been synthesized. The host material, LixMoO3 , was should facilitate the hopping of Li ions in the gallery and synthesized via a LiBH4 route which is dierent from the contribute to a high ionic conductivity for the material. conventional approach. This material exfoliates in water and has anity for a large variety of polymers. Polymers such as Magnetism PEG, PEO, PPG, PVP, MCel, PAM and PA6 give well stacked lamellar nanocomposites.Depending on the nature of Because the materials are formally mixed-valence compounds and exhibit intense intervalence Mo5+–Mo6+ optical transitions, we expect unpaired electrons to be delocalized over the d orbitals of the Mo atoms. Magnetic susceptibility measurements were carried out as a function of temperature for LixMoO3 and Lix(H2O)y(PEO-100 000)zMoO3 and the data are displayed in Fig. 9. Surprisingly, the susceptibility of both these compounds is rather small with substantial contributions from temperature independent paramagnetism (xTIP). Correcting for the latter, 1/(xmolar-xTIP) vs. T plots for LixMoO3 and Lix(H2O)y(PEO-100000)zMoO3 are linear in the temperature ranges 5–120 K and 2–300 K, respectively.The Curie constants estimated from the slope of the plots yield meff of 0.15 and 0.09 respectively, however, the weak susceptibilities and the large diamagnetic and xTIP corrections make meff values unreliable. It is interesting that the magnetic properties Fig. 10 Variable-temperature electrical conductivity measurements for Fig. 9 Temperature dependence of magnetic susceptibility of LixMoO3 (#) and Lix(H2O)y(PEO-100000)zMoO3 (').The measurements were pressed pellet samples of (a) LixMoO3 and (b) Lix(H2O)y(PEO- 100000)zMoO3 conducted at 1000 G with powder samples. 1282 J. Mater. Chem., 1997, 7(7), 1277–1283M. G. Kanatzidis, Chem. Mater., 1996, 8, 525; (d) C.-G. Wu, the polymer and its molecular mass, some of the nanocom- M.G. Kanatzidis, H. O. Marcy, D. C. DeGroot and posites are soluble and can be processed into films. Electronic C. R. Kannewurf, in L ower-Dimensional Systems and Molecular transmission spectra show the broad transition associated with Electronics, ed. R. M. Metzger, et al., Plenum, New York, 1991, the Mo5+–Mo6+ couple in these nanocomposites ranging from p. 427; (e) Y.-J. Liu, D.C. DeGroot, J. L. Schindler, C. R. Kannewurf the IR to the visible region. Solid-state 7Li NMR spectroscopy and M. G. Kanatzidis, J. Chem. Soc., Chem. Commun., 1993, 593; indicates a more versatile chemical environment in the nano- (f) C.-G. Wu, D. C. DeGroot, H. O. Marcy, J. L. Schindler, C. R. Kannewurf, Y.-J. Liu, W. Hirpo and M. G. Kanatzidis, Chem. composites than in the host which may lead to high ionic Mater., 1996, 8, 1992.conductivities in these lamellar systems. The electrical conduc- 12 G. M. Kloster, J. A. Thomas, P. W. Brazis, C. R. Kannewurf and tivity of these materials is thermally activated and ranges from D. F. Shriver, Chem.Mater., 1996, 8, 2418. 10-2 to 10-7 S cm-1. 13 F. Leroux, B. E.Koene and L.F. Nazar, J. Electrochem.Soc., 1996, 143, L181. 14 (a) L. F. Nazar, X. T. Yin, D. Zinkweg, Z. Zhang and S. Liblong, Financial support from the National Science Foundation Mater.Res. Soc. Symp. Proc.,1991, 210, 417;(b)L.F.Nazar,Z.Zhang (DMR-93-06385) is gratefully acknowledged. This work made and D. Zinkweg, J. Am. Chem. Soc., 1992, 114, 6239; (c) L. F. Nazar, use of the NMR facilities of the Department of Chemistry and H.Wu and W. P. Power, J. Mater. Chem., 1995, 5, 1985; magnetic measurements facilities of the Department of Physics (d) T. A. Kerr,H.Wu and L. F. Nazar, Chem.Mater., 1996, 8, 2005. 15 R. Bissessur, D. C. DeGroot, J. L. Schindler, C. R. Kannewurf and at Michigan State University. At Northwestern University this M. G. Kanatzidis, J. Chem. Soc., Chem. Commun., 1993, 687.work made use of Central Facilities supported by NSF through 16 G. Cao and T. E. Mallouk, J. Solid State Chem., 1991, 94, 59. the Materials Research Center (DMR-91-20521). We thank 17 J. E. Pillion and M. E. Thompson, Chem.Mater., 1991, 3, 777. Kermit Johnson for assistance with the Li NMR spectroscopy. 18 A. Clearfield and C. Y. Ortiz-Avila, in ACS Symposium 499: Supramolecular Architecture, ed.T. Bein, American Chemical Society, 1992, p. 178. 19 Y.-J. Liu and M. G. Kanatzidis, Inorg. Chem., 1993, 32, 2989. References 20 Y. Ding, D. J. Jones, P. Maireles-Torres and J. Rozie`re, Chem. 1 (a) Y. Fukushima and S. Inagaki, J. Inclusion Phenom., 1987, 5, 473; Mater., 1995, 7, 562. (b) Y. Fukushima, A. Okada, M. Kawasumi, T. Kurauchi and 21 W. M. R.Divigalpitiya, R. F. Frindt and S. R. Morrison, J.Mater. O. Kamigaito, ClayMiner., 1988, 23, 27; (c)A. Usuki, M. Kawasumi, Res., 1991, 6, 1103. Y. Kojima,A.Okada,T.Kurauchi andO. Kamigaito, J.Mater. Res., 22 (a) M. G. Kanatzidis, R. Bissessur, D. C. DeGroot, J. L. Schindler 1993, 8, 1174; (d) A. Usuki, Y. Kojima, M. Kawasumi, A. Okada, and C. R. Kannewurf, Chem. Mater., 1993, 5, 595; (b) R.Bissessur, Y. Fukushima, T. Kurauchi and O. Kamigaito, J.Mater. Res., 1993, M. G. Kanatzidis, J. L. Schindler and C. R. Kannewurf, J. Chem. 8, 1179; (e) Y. Kojima, A. Usuki, M. Kawasumi, A. Okada, Soc., Chem. Commun., 1993, 1582; (c) R. Bissessur, J. L. Schindler, Y. Fukushima, T. Kurauchi and O. Kamigaito, J.Mater. Res., 1993, C. R. Kannewurf and M. G. Kanatzidis, Mol.Cryst. L iq. Cryst., 8, 1185. 1993, 245, 249; (d) L. Wang, J. L. Schindler, J. A. Thomas, 2 (a) E. P. Giannelis, Adv. Mater., 1996, 8, 29; (b) R. Krishnamoorti, C.R.Kannewurf andM.G. Kanatzidis, Chem.Mater., 1995, 7, 1753. R. A. Vaia and E. P. Giannelis, Chem. Mater., 1996, 8, 1728; 23 J. P. Lemmon and M. M. Lerner, Chem.Mater., 1994, 6, 207. (c) E. P. Giannelis, V. Mehrotra, O.Tse, R. A. Vaia and T.-C. Sung, 24 E. Ruiz-Hitzky, R. Jimenez, B. Casal, V. Manriquez, A. S. Ana and Mater. Res. Soc. Symp. Proc., 1992, 267, 969; (d) R. A. Vaia, H. Ishii G. Gonzalez, Adv. Mater., 1993, 5, 738. and E. P. Giannelis, Chem. Mater., 1993, 5, 1694; (e) R. A. Vaia, 25 I. Lagadic, A. Le�austic and R. Cle�ment, J. Chem. Soc., Chem. S. Vasudevan, W. Krawiec, L. G. Scanlon and E.P. Giannelis, Adv. Commun., 1992, 1396. Mater., 1995, 7, 154; ( f ) S. D. Burnside and E. P. Giannelis, Chem. 26 C. O. Oriakhi and M. M. Lerner, Chem.Mater., 1996, 8, 2016. Mater., 1995, 7, 1597; (g) R. A. Vaia, K. D. Jandt, E. J. Kramer 27 P. G. Hill, P. J. S. Foot and R. Davis, Synth. Met., 1996, 76, 289. and E. P. Giannelis, Macromolecules, 1995, 28, 8080; 28 P. B. Messersmith and S.I. Stupp, Chem.Mater., 1995, 7, 454. (h) P. B. Messersmith and E. P. Giannelis, J. Polym. Sci., Part A: 29 H.-L. Tsai, J. Heising, J. L. Schindler, C. R. Kannewurf and Polym. Chem., 1995, 33, 1047; (i) P. B. Messersmith and M. G. Kanatzidis, to be published. E. P. Giannelis, Chem.Mater., 1993, 5, 1064; (j) 1994, 6, 1719. 30 N. Margalit, J. Electrochem. Soc., 1974, 121, 1460. 3 (a) M. S. Wang and T. J. Pinnavaia, Chem. Mater., 1994, 6, 468; 31 J. Desilvestro and O. Haas, J. Electrochem. Soc., 1990, 137, 5C. (b) T. Lan, P. D. Kaviratna and T. J. Pinnavaia, Chem.Mater., 1994, 32 M. G. Kanatzidis and T. J. Marks, Inorg. Chem., 1987, 26, 783. 6, 573; (c) T. Lan and T. J. Pinnavaia, Chem. Mater., 1994, 6, 2216; 33 (a) B. N. Diel, T. Inabe, J. W. Lyding, K.F. Schock Jr., (d) H. Shi, T. Lan and T. J. Pinnavaia, Chem. Mater., 1996, 8, 1584; C. R. Kannewurf and T. J. Marks, J. Am. Chem. Soc., 1983, 105, (e) Z.Wang, T.Lan and T. J. Pinnavaia, Chem.Mater., 1996, 8, 2200; 1551; (b) J. W. Lyding, H. O. Marcy, T. J. Mark and (f ) T. Lan, D. Kaviratna and T. J. Pinnavaia, J. Phys. Chem. Solids, C. R. Kannewurf, IEEE T rans. Instrum.Meas. 1988, 37, 76. 1996, 57, 1005. 34 D. M. Thomas and E. M. McCarron III, Mater. Res. Bull., 1986, 4 (a) E. Ruiz-Hitzky and P. Aranda, Adv. Mater., 1990, 2, 545; 21, 945. (b)P. Aranda and E. Ruiz-Hitzky, Chem. Mater., 1992, 4, 1395. 35 D. M. Moore and R. C. Reynolds Jr., X-Ray Diraction and the 5 M. Iwai, H. Shoji, S. Shimazu and T. Uematsu, Chem. L ett., 1993, Identification and Analysis of Clay Minerals, Oxford University 403.Press, Oxford and New York, 1989, p. 83. 6 T. Kyotani, T. Mori andA.Tomita, Chem.Mater., 1994, 6, 2138. 36 L.Wang and M. G. Kanatzidis, unpublished work. 7 J.J. Tunney and C. Detellier, Chem.Mater., 1996, 8, 927. 37 L.Wang and M. G. Kanatzidis, work in progress. 8 J. C. Hutchison, R. Bissessur and D. F. Shriver, Chem.Mater., 1996, 38 S. Colson, J.M. Tarascon, S. Szu and L. C. Klein, Mater. Res. Soc. 8, 1597. Symp. Proc., 1991, 210, 405. 9 Y. Kurokawa, H. Yasuda and A. Oya, J. Mater. Sci. L ett., 1996, 39 J. M. Tarascon and S. Colson, Mater. Res. Soc. Symp. Proc., 1989, 15, 1481. 135, 421. 10 (a) M. G. Kanatzidis, L. M. Tonge, T. J. Marks, H. O. Marcy and 40 Y. Matsuda, M. Sato, M. Onoda and K. Nakao, J. Phys. C: Solid C. R. Kannewurf, J. Am. Chem. Soc., 1987, 109, 3797; State Phys., 1986, 19, 6039. (b) M. G. Kanatzidis, H. O. Marcy, W. J. McCarthy, 41 M. Greenblatt, in L ow-Dimensional Electronic Properties of C. R. Kannewurf and T. J. Marks, Solid State Ionics, 1989, 32/33, Molybdenum Bronzes and Oxides, ed. C. Schlenker, Kluwer 594; (c) M. G. Kanatzidis, C.-G. Wu, H. O. Marcy, D. C. DeGroot, Academic, Dordrecht, Boston, London, 1989. C. R. Kannewurf, A. Kostikas and V. Papaefthymiou, Adv. Mater., 42 L. F. Schneemeyer, F. J. DiSalvo, R. M. Fleming and J. V. 1990, 2, 364; (d) C.-G. Wu, H. O. Marcy, D. C. DeGroot, J. L. Schindler, C. R. Kannewurf, W.-Y. Leung, M. Benz, E. LeGo Waszczak, J. Solid State Chem., 1984, 54, 358. and M. G. Kanatzidis, Synth. Met., 1991, 41–43, 797; (e) 43 G. H. Bouchard Jr., J. Perlstein and M. J. Sienko, Inorg. Chem., M. G. Kanatzidis, C.-G. Wu, D. C. DeGroot, J. L. Schindler, 1967, 6, 1682. M. Benz, E. LeGo and C. R. Kannewurf, in Chemical Physics of 44 D. C. Johnston, Phys. Rev. L ett., 1984, 52, 2049. Intercalation II, ed. J.Fisher, NATO ASI Ser., Plenum, 1993, p. 63. 45 C. Julien and G. A. Nazri, Solid State Ionics, 1994, 68, 111. 11 (a) Y.-J. Liu, D. C. DeGroot, J. L. Schindler, C. R. Kannewurf and 46 J. O. Besenhard, J. Heydecke, E.Wudy and H. P. Fritz, Solid State M. G. Kanatzidis, Chem. Mater., 1991, 3, 992; (b) Y.-J. Liu, Ionics, 1983, 8, 61. D. C. DeGroot, J. L. Schindler, C. R. Kannewurf and M. G. Kanatzidis, Adv. Mater., 1993, 5, 369; (c) Y.-J. Liu, J. L. Schindler, D. C. DeGroot, C. R. Kannewurf, W. Hirpo and Paper 7/00202E; Received 8th January, 1997 J. Mater. Chem., 1997, 7(7), 1277–1283