Hybrid organic-inorganic materials a land of multidisciplinarity Patrick Judeinstein*' and ClCment Sanchez 'Laboratoire de Chimie Structurale Organique (URA 1384) Bdt. 0 Universitt! Paris Sud Orsay France bLaboratoire de Chimie de la Matiire Condensie (URA 1466) Universiti P. et M. Curie 4 Place Jussieu 75252 Paris Cedex France Organic-inorganic hybrids appear as a creative alternative for obtaining new materials with unusual features. This is related to their diphasic structures leading to multifunctional materials. The low-temperature processes which are used to synthesize such structures provide a wide versatility in the design of the compounds. The potentiality of the chemistry is to play on the structure of these mixtures and dissociate the various contributions in tailoring both phases and controlling the interfaces.In this paper a review of some chemistry pathways to hybrid materials is presented. The nature of the bonds between organic and inorganic phases is used to divide them in two major families class I corresponds to materials with weak interphase bonding while class I1 corresponds to materials where both phases are chemically grafted. Applications of these materials in the fields of optics iono- electronics mechanics biology and others are expected. Some applications are reviewed with respect to the versatility of the synthetic procedure. Most of the properties of these new high-technology materials are dependent on their structural and chemical composition as well as on the dynamical properties inside the blends.The possibility of combining the properties of organic and inorganic compounds in a unique material is an old challenge that started with the beginning of the industrial era. Some of the oldest and most famous organic-inorganic hybrids come from the paint industries where inorganic pigments (TiO etc.) are suspended in organic mixtures (solvents surfactants etc.). While the concept of 'hybrid' materials was not mentioned at that time the wide increase of work on organic-inorganic structures continued with the development of the polymer industry. Inorganic fillers (minerals clays talcs etc.) were added to polymers in order to improve some of the properties of the compounds. In fact the concept of 'hybrid organic- inorganic' materials emerged only very recently when the research shifted to more sophisticated materials with a higher added value.',2 Recently the study of organic-inorganic nano-composites networks and gels became an expanding field of in~estigation.~,~These new materials considered as innovative advanced materials promise new applications in many fields such as optics electronics ionics mechanics and biology.At first glance these materials are considered as biphasic mate- rials where the organic and inorganic phases are mixed at thenm to sub-pm scales. Nonetheless it is obvious that the properties of these materials are not just the sum of the individual contributions from both phases; the role of the inner interfaces could be predominant.The nature of the interface has been used recently to divide these materials into two distinct classes.' In class I organic and inorganic compounds are embedded and only weak bonds (hydrogen van der Waals or ionic bonds) give the cohesion to the whole structure. In class I1 materials the two phases are linked together through strong chemical bonds (covalent or iono-covalent bonds). Obviously within many class I1 hybrid materials organic and inorganic components can also interact via the same kind of weak bonds that define the class I hybrids. Generally the organics are 'fragile' and their thermal stability is limited to less than 250°C meaning that high temperatures are prohibited during the hybrid formation process. The sol-gel process overcomes this limitation.6 Inorganic polymerizations are performed around ambient temperature starting with metallo-organic precursors (salts or alkoxides).Metal-oxo based macromolecular networks are formed after hydrolysis- condensation reactions. The choice of reaction conditions leads to the control of the material structure and properties.' The basic background for the sol-gel chemistry will be described first. Then general methods for the production of class I and class I1 hybrid materials will be presented. Finally some applications of these hybrid materials will be outlined emphasizing the flexibility of the process which adjust the properties of the final materials. Chemistry Synthesis of Hybrid Materials Sol-gel processes Sol-gel processes are a method of forming dispersed inorganic materials in solvents through the growth of metal-oxo poly- mer~.~'~The chemistry is based on inorganic polymerization reactions.Metal alkoxides [M(OR) where M=Si Sn Ti Zr Al Mo V W Ce etc.; OR an alkoxy group OC,H2,+ J are used as molecular precursors which lead to metal-oxo polymers through hydrolysis and condensation reactions. The first step in sol-gel synthesis is the hydroxylation of the metal alkoxide which occurs upon hydrolysis of the alkoxy groups as follows M-OR +H20+M-OH +ROH Reactive hydroxy groups are generated first. They then undergo polycondensation reactions via two competing mechanisms (a) Oxolation the formation of an oxygen bridge M -OH +M -OX+M-0-M +XOH (X=H or alkyl group) (b)Olation the formation of a hydroxo bridge M-OH +HO- M -+M(OH),M (X =H or alkyl group) Metal-oxo based oligomers and polymers capped by residual hydroxo and alkoxy groups are the result of these two equilibrated reactions.The structure and morphology of the resulting metal-oxo macromolecular networks are dependent on the respective J. Muter. Chem. 1996 6(4) 511-525 rates of the different reactions Rearrangement reactions then occur leading preferentially to weakly branched polymers When these structures reach macroscopic size a gel in which solvent and free polymer are entrapped is obtained The gel state is not the only possibility Other final forms such as colloidal solutions or precipitates can be obtained.The forma- tion of gels or colloidal species reflects different growing processes and different polymer-solvent interactions Control of the nature of the intermediate species through the reaction conditions is essential to tailor the final structures. The reactiv- ity of the metal alkoxide' (nature of M and R) the hydrolysis ratio (H,O M) the solvent the reaction temperature the use of complexing agents or catalysts are the main parameters used to achieve control over the size and morphology of the resulting materials Starting from molecular precursors more and more condensed species are obtained leading to colloidal 'sols' and then to 'gels' A 'xerogel' can be made upon drying the gel under ambient conditions Evidently the possibility of obtaining solutions with controllable viscosity is of funda- mental interest to process these materials' as thin films (spin or dip coating techniques) fibres powders monoliths or particles of various sizes and shapes The chemical reactivity of the metal alkoxide in the hydro- lysis step is determined by both the nature of the metal M and the steric hindrance of the alkoxy groups *.The major parameters'' appear to be the electrophilic character of the metal atom (measured by electronegativity x) and its ability to increase its coordination number N. The unsaturation degree of the metal coordination can be simply expressed by the difference N-2 where N is the coordination number usually found in the oxide and is the oxidation state.Table reports these values for different tetravalent metal alkoxides The difference of reactivity between silicon alkoxides and other transition-metal (TM) alkoxides is demonstrated by the vari- ation of the (N-Z) value. The following sequence of reactivity is usually found Si(OR) <<<Sn(OR) =Ti(OR) < Zr(OR) = Ce(OR) Silicon alkoxides are not very reactive (low electro- philicity N -2= 0). The gelation therefore takes several weeks when neutral water is added and the reaction rate must be increased by using catalysts. The reaction rate depends strongly upon the catalysts (acidic basic or nucleophilic) Molecular precursors R',Si(OR),- ,(R' = alkyl phenyl H etc ) are also used to produce hybrid materials of class 11 their chemistry will be detailed later.Transition-metal alkoxides (Sn Ti Zr etc ) are generally very reactive* and their hydrolysis leads immediately to the precipitation of the 0x0 polymers The reactivity must be controlled by using chemical additives l2 Inorganic acids13 have been described as inhibitors but strong complexing ligands (polyhydroxylated ligands organic acids P-hydroxyacids /I-diketones and allied derivatives) are much Table Properties of some tetravalent metal isopropoxides (2= 4) x" Nh N-2' Si(OPr') 100 Sn(0Pr' ) Ti(OPr') Zr(OPr') Ce (OPr' ) " = electronegativity N =coordination number of the metal atom N -2 =degree of unsaturation. Fig. TM alkoxide complexation by P-acetylacetonates derivatives 512 J Mater Chem 1996 6(4),511-525 more efficient' for controlling the reactivity of TM alkoxides (Fig 1) Chelating agents react with TM alkoxides in the following manner.M(OR) + xHacac -+ M (OR) -Jacac) + xROH For M = Ce Ti Zr etc the strong decrease of reactivity from the pure alkoxide to the modified alkoxide M(OR),-,(acac) is proven lo l2 l4. Turbid or transparent gels are obtained for low modification ratios stable sols or molecular clusters are obtained for higher modification ratios l5 From an extensive study of the phase diagram it becomes evident that the acac groups are not fully removed from the metallic centres even for high hydrolysis ratios. These groups are still present in the final materials or at the surface of the clusters Such complexing ligands can be used to protect the surface of nanosized particles and prevent their aggregation l6.The use of functionalized complexing ligands is important in the synthesis of class IT materials and will be extensively described later Hybrid materials are made by mixing organic and inorganic components Control of the properties of the final material is achieved by controlling the chemical nature of the organic and inorganic phases the size and morphology of these domains (nm to sub-pm scale) and the nature of the interphase inter- action. The first step in the procedure is obviously to find some common suitable solvents and compatible reactants. To illustrate different approaches described in the literature some representative examples for synthesizing hybrid materials are presented Hybrid organic-inorganic materials class I In class I materials organic and inorganic components are linked together through weak bonds (van der Waals ionic or hydrogen bonds hydrophobic-hydrophihc balance).They result from much research work emerging from sol-gel and polymer chemists and these materials will present a large diversity in their structures and final properties Organic dyes embedded in sol-gel matrices. Small organic molecules entrapped in an inorganic network is certainly the most simple representation of a hybrid material It corresponds to the doping of sol-gel matrices by organic dyes inorganic ions or molecules resulting in fluorescence photochromic or non-linear optical (NLO)properties l7 Organic molecules such as rhodamines pyranines coumarins porphyrins phthalocy- anines and spiropyrans as NLO dyes have been entrapped in inorganic networks such as silica aluminosilicate or transition- metal oxide based gels (ZrO T10,) (Fig 2).The choice of the composition of the inorganic matrix is an elegant way to change the refractive index and to modulate the mechanical properties of the final materials Basically the inorganic mol- ecular precursor (alkoxide) the dye and the catalyst are mixed in a common solvent Water is then added to the mixture to begin the polycondensation and the dye molecules are uni- formly trapped in the growing polymer Another alternative is to dip an inorganic xerogel into the dye solution Capillarity could lead to a homogeneous distribution of the dye entities In fact weak interactions between the dye and the inorganic matrix (hydrogen bonds van der Waals forces etc) account for the dispersion of the dye within the structure," and the final properties of the materials (photoresponse reversibility stability etc ) Organic monomers embedded in sol-gel matrices.oSol-gel inorganic matrices are often very porous structures ( A < pore size< nm). The pores of the structure may be filled with molecules by immersing the bulk in a solution containing polymerizable organic monomer (methylmethacrylate buta- diene and derivatives etc ) and a catalyst 2o In a second step organic polymerization is started either by UV irradiation or by heating and the polymer (PMMA PBu etc) is formed (Fig 3).Transparent monoliths of large size with a tunable key to symbols used in all figures:-alkoxide alkoxide Inorganic precursor functionalbed by a inorganic cluster ,$J functionalized by labile '(alkoxide molecule) non-labile polymerizable polymerizable group groups bis(organical1y & bis( silicon end- organic molecule organic molecule modified)silicon A w (dye ...) ((surfactanle ...) alkoxide J organic monomer &% organic monomer polymer chain polymer chain Fig. Organic dyes embedded in sol-gel matrices and key to the symbols used in the figures hv -or heating Fig. Organic monomers embedded in sol-gel matrices followed by polymerization refractive index can be obtained for optical applications depending on the size and geometry of the holes the difference in the refractive index between the two phases and the organic inorganic ratios.An obvious problem is related to the difference of density between the monomer and the polymer leading to strong mechanical stresses in the materials and the formation of many optical defects. Organic functional mole- cules can be also mixed with an organic monomer. Perylene dyes and also enzymes and porphyrins have been incorporated into these hybrid materials leading to sensors or fluorescent silica microspheres composites with lasing proper tie^."^ Inorganic particles embedded in a polymer. For a long time the mechanical properties of polymeric blends have been adjusted by incorporating inorganic fillers (Fig.4). The conven- tional process is to mix together the polymer (or a prepolymer) and the inorganic particles. Nonetheless the high viscosity of the mixture leads to the agglomeration of particles. The resulting inhomogeneity within the materials decreases the polymer-filler interactions. Using a suitable solvent could prevent the lack of homogeneity but further drying steps must be planned. These techniques are also used to prepare multi- component ceramic pastes which can be cast in rno~lds.~ Powders of MgO A1,0 and SiO are mixed with a water- soluble polymer and the viscosity of the gel is adjusted by changing the concentration of the solute. Ceramic hodies of cordierite ( Mg2A1,Si,0,,) are then obtained after firing.Obviously this process is easy and fast but the chemical homogeneity required to obtain complex ceramics (cordierite mullite YBaCuO etc.) is difficult to achieve. Polymers filled with in situ generated inorganic particles. A possibility to overcome the inhomogeneity of materials obtained by embeddment of inorganic particles in polymers (see previous section) is to build the inorganic clusters inside the polymeric structure (Fig. 5)." A typical method consists of mixing together the polymer and the metal alkoxide in a suitable solvent (alcohol or THF). In a second step catalyst and water are added to the mixture and the polycondensation is performed in situ. The best homogeneity is achieved when the weak interactions developed between both phases are sufficient to force both networks to interpenetrate mutually at J.Mater. Chem. 1996. 6(4) 511-525 -stirring Fig. Inorganic particles dissolved in gels or polymers & H2° -catalyst Fig. Polymers filled with in situ grown organic particles the molecular level. The sol-gel process is therefore generating hydroxy groups M-OH; some of those could remain on the surface of the growing 0x0-polymers exhibiting Brarnsted acidity. Carbonyl groups are present in many organic polymers (such as polyamides) and are well known as strong acceptors of acidic hydrogen to form hydrogen bonds. Silica nanocom- posites have therefore been obtained with polymers such as poly (2-methyl 2-0xazoline),~~ poly (N-vinyl pyrrolidone)22 or p~ly(N,N-dimethylacrylamide).~~These materials have good optical properties which can be modulated by the silica or- ganic ratio. The presence of hydrogen bonding between the two phases has been confirmed by 13C NMR and IR spectros~opies.~~~~~ Simultaneous formation of interpenetrating organic-inorganic networks.The design and preparation of monolithic composite materials with good optical properties is a challenge. The homogeneity of the materials the size and shape of the nanodomains must be controlled. A major problem is the shrinkage of the materials during the evaporation of the solvent.24 Generally sol-gel processes are performed in dilute solutions (alcohol THF excess water etc.) and a large amount of solvent must be removed after the formation of the network.Furthermore the change of solvent composition during the process causes phase demixing and inhomogeneity of the composite. Alternative routes to avoid these problems have recently been proposed for the production of homogeneous nanocomposites based on silica with minimal shrinkage a wide choice of polymer structure and a huge range of silica-polymer compositions. The use of modified silicon alkoxides as starting units has been pr~posed~~,~~ (Fig. 6). These alkoxide molecules possess two distinct reactivities the first is due to the metal atom and causes the formation of the oxide network SiO through the condensation mechanism and the second results from the polymerizable group.Cyclic alkenols are polymerized via ring-opening metathesis polymerization initiated by a redox catalyst (Ru3+ salts) whereas alkoxy 514 J. Muter. Chem. 1996 6(4) 511-525 groups carrying methacrylate functions are polymerized by a free-radical mechanism initiated by either UV-light or heating in the presence of a catalyst. The functionalized alkoxy group is initially bonded to the metal atom oiu an M-OC bond. This bond is hydrolysed during the hydrolysis-condensation step thus releasing the polymerizable organic groups. Two networks are then formed independently without mutual chemical bonds. No solvents are released during the reactions and if the density of the monomeric species and the polymer are similar the shrinkage will be negligible.This process can lead to monolithic pieces of large size and with good optical properties. Two interpenetrating networks can be obtained by adjusting the relative kinetics of organic and inorganic reac- tions. The organic reaction must be controlled by the choice of the initiator its concentration with the use of an efficient nucleophilic catalyst (such as NaF) which leads to a fast silica polyc~ndensation.~~ Obtention of ordered organic-inorganic structures. The obten- tion of ordered structures has certainly been one of the outstand- ing goals for materials scientists during recent years. More and more physical concepts are dealing with the anisotropic proper- ties of the material and technological realizations for such advanced materials are expected in the fields of 1D and 2D conductivities membranes anisotropic optical properties non- linear optics etc.Several approaches to the production of such structures have already been described. The first method con- cerns the insertion of organic molecules or polymers into an anisotropic inorganic network [Fig. 7(u)]. The insertion reac- tions of organic molecules have been extensively described in host structures such as clays silicates metal phosphates and phosphonates metal oxide halides chalcogenides etc.28 An interesting inorganic matrix is the vanadium pentoxide (V205) gels. They are obtained through sol-gel processes by acidifi- cation of metavanadate or hydrolysis-condensation of vanadium oxo-alkoxides.29~30 Tactoidal gels of formula V2O5.nH2Oare then obtained.They are composed of negatively methylmethacryl derivative Si02 SiO cycloalkenol derivative H20 + heat or hv -NaF + R?or R" Q Fig. Simultaneous formation of interpenetrated organic and inorganic networks charged flat ribbons. Films exhibiting a layered structure are obtained when the sol is deposited on a glass substrate. The redox properties of the vanadium oxide gels their mixed electronic-ionic conductivity properties and the incorporation of organic molecules inside these structures have been widely studied.29 Insertion of organic molecules is obtained by dipping the inorganic materials into a solution containing the organic component. The driving forces for the insertion process may involve redox reactions acid-base chemistry solvent or ion exchange.Molecules such as alcohols alkylamines metallocen- ium ions viologens and sulf~xides~~~~~ have also been inserted into the host structure. Moreover the structural analysis of the final materials shows that inserted molecules are oriented along specific directions in the interlayer space. The preferred direction results from a competition between van der Waals and hydrogen bonding steric hindrance electrostatic repulsions and lattice energy; upon that depends the nature of the guest molecules. Vanadium pentoxide gels can also intercalate polymers such as poly(ethy1ene oxide) or poly(viny1 pyrr~lidone).~~ Polymeric chains are intercalated between the oxide sheet and the stacking of the layered structure is preserved.Another alternative is to insert the monomers first and then to promote their polymenz- J. Muter. Chem. 1996 6(4) 511-525 soaking+ A W AA AA AVA Fig. Ordered structures obtained by (a) insertion reactions and (b)anisotropic inorganic structure building ation in sit^^^ (oxidative polymerization of aniline pyrrole and thiophene derivatives). Recently electronically conductive poly- mer nanostructures [polypyrrole poly( 3-methylthiophene) or polyaniline] have been obtained by oxidative polymerization of the corresponding monomers entrapped in a template nano- tubular alumina matrix.35 Fibrils with diameters ranging from 20 to nm are obtained.A strong increase of conductivity is measured for the thinner fibres. This effect is related to the stronger anisotropy and molecular organization of the polymer. Strong interactions between the host template structure and the guest molecules account for this effect which may be enhanced by modifying the nature of the surface (grafting of organically modified silanes on the pore walls). A different approach is to build anisotropic inorganic par- ticles using organic molecules and self-assembled aggregates as structure-directing agents [Fig. 7(b)]. Hollow silica cylinders of sub-pm diameter are obtained by depositing silica onto phospholipid Silicate-surfactant mesophases have been obtained by mixing tetraethoxysilane (TEOS) and basic aqueous solutions of surfactants such as cetyltrimethylammon- ium hydroxide (CTAOH) and cetyltrimethylammonium bromide (CTAB).37 Depending on the respective ratios the surfactant-inorganic hybrids exhibit cubic hexagonal or lamellar structures. The cooperative organization between the inorganic species and the surfactants have been studied by 'H NMR spectroscopy X-ray scattering and electron microscopy which have demonstrated the strong anisotropy of these new com- posite materials.Similar experiments have been performed on (a1umino)silicates and the calcination of the mesophases leads to mesoporous zeolites and oxide ceramics. Lyotropic liquid crystals and quaternary ammonium surfactants have also been used as promising tools in tailoring the porous structure of inorganic gels and their textural ~rdering.~' 516 J.Muter. Chem. 1996 6(4) 511-525 Hybrid organic-inorganic materials class I1 Class I1 materials are hybrid structures in which organic and inorganic components are grafted together through strong covalent or iono-covalent chemical bonds. The molecules used as starting building blocks for class I1 hybrids possess at least two distinct functionalities alkoxy groups ( R-OM bonds) which should experience hydrolysis-condensation reactions in the presence of water and lead to an 0x0-polymer framework and metal-to-carbon links which are stable in the hydrolysis reactions. The nature of the stable metal-to-carbon link depends on the nature of the metallic cation.Organometallic links (M-C bonds) are stable towards hydrolysis when M is silicon tin mercury lead or phosphorus. On the contrary M-C bonds are not stable when M is a transition-metal cation. Those M-0-C bonds which are stable upon hydroly- sis could be the links between the organic and inorganic parts. Complexation by polyhydroxylated ligands organic acids. p-hydroxyacids p-diketones and allied derivatives are also u~ed.~.'~ Hybrids obtained from organically modified silicon alkoxides. The chemistry of organic silicon derivatives is well developed. The synthesis and properties of molecules with formula Si(OR),-xR,(x=l or 2) with non-labile Si-R' groups (R'= alkyl aromatic substituent etc.) and hydrolysable Si- OR bonds have been widely described and many of these molecules are also commercially available.If the R' group bears a reactive function (vinyl amines isocyanates etc.),then organic chemis- try could be easily performed on this branch. Two methods have been developed for the synthesis of silicon oxide based hybrid materials. Sequential synthesis. In this method both networks are obtained sequentially in a two-step reactions (Fig. 8). Firstly the inorganic network is created by the polycondensation of the silicon alkoxide which leads to the formation of an oxo- polymer surrounded by organic groups. In the second step an organic reaction is performed with the organic radical. Organic polymerizations of methylmethacryl vinyl ally1 and epoxy R‘ groups have been extensively studied and described.They lead to hybrid materials where both components are network former^.^.',^^^. The choice of reaction conditions for the two successive reactions as well as the possibility to vary the 0rganic:inorganic ratio (adding TEOS or monomer molecules) allows materials with a wide range of properties to be obtained. These materials are usually called ORMOSILs (organically modified silanes) or ORMOCERs (organically modified cer- amic~),~~and they possess great potential because the polymer brings new properties to the inorganic netw~rk~~.~.~~ (flexibility hydrophobicity refractive index modification etc.). The choice of an organic group R’ which is no longer reactive has also been investigated.These network modifiers (Si-CH,. Si-C,Hj etc.) are studied extensively for surface modification of films or particles in the fields of corrosion protection surface treatment membranes and chromatography. R’ groups with a special functionality (dye,40 crown ether,41 persistent radi~al,~’ etc.) are also widely studied. NLO-ph~re,~~ A better stability to leakage than class I materials is expected. The mixing with dialkoxysilanes R’R”Si(OR) has also been studied.44 Poljfunctional alkoxysilanes. Polyfunctional alkoxysilanes are organic units (R’) to which two or more %(OR) groups are bonded through Si-C bonds. When two trialkoxysilane groups are bonded to the R’ unit the generic formula is ( R0)3Si- R’- %(OR),. The trialkoxy groups must be further condensed in the presence of water-catalyst mixtures to obtain hybrid materials (Fig.9). Molecular or macromolecular organic units R’ are described in the literat~re.~~.~~ For molecular units precursors with many different geo- metries (aryl rigid rod spacers acyclic flexible spacers) are used to control precisely the parameters that govern the structures of the final materials.46347 The hydrolysis condensation of all these precursors leads to microporous materials with surface area from less than m2 g-’ up to m2 g-’ depending on both the nature of the precursor and the nature of the catalyst or the solvent. Removing these organics by calcination or plasma gives interesting materials in the field of nanomembranes.Alkoxjsilanes functionalized by polymers. Many studies are devoted to the use of polymer based polyfunctional alkoxy silanes in order to get hybrid organic-inorganic polymeric network^.'^,^^ Pioneering works dealt with materials obtained by coreaction of metallo-organic precursors (such as silicon or titanium) and natural polymers such as polysa~charides,~~ catalyst (HCI or NaOH or NH4F) cellulosic materials or vegetable oil derivatives. These func- tional polymers have hydroxy groups which could react with the metal-oxo polymers which are formed in situ. These cross- linkers are used to control the viscosity of the solution and enable better processing. Recently reactive alkoxysilane groups [-,%(OR),] have been grafted to many kinds of oligomers and polymers (Fig.Many chemistry pathways have been explored to produce such silicon end-capped precursors such processes include direct synthesis through organometallic methods or the coupling of a reactive macromer with a reactive trialkoxysil- ane (e.g. R‘ =isocyanate amine carboxy alkyl halide etc.). Some of the processes are illustrated in Table 2. These hybrids are very interesting because of the wide range of properties which can be achieved.” The final features of the hybrid depend on the properties of the organic polymer the degree of phase dispersion and the homogeneity at thenm scale.9d. They depend on the precursors and the hydrolysis- condensation reactions including the chemical nature and the molecular mass of the functionalized macromers the density of reactive Si(OR) groups the solvent the hydrolysis ratio and the nature of the catalyst.Highly functionalized high-technology polymeric materials are then finally obtained. They present interesting thermal mechanical optical and ionic properties. The choice of the starting materials is strongly influenced by polymer science. Hybrids based on transition-metal oxide (TMO) networks. The strong reactivity of the transition-metal-carbon bond towards hydrolysis is well known but the chemical modifi- cation reactions of TM alkoxides have been developed to overcome this problem. Grafting a silica layer on the TIMO.~~The first and \implest approach to graft organics to TMO is to use an intermediate silica layer (Fig.11). The surfaces of TM-0x0 polymers or colloids have a high density of hydroxy groups hl-OH which can react with silicon alkoxides or organically modified silicon alkoxides thus creating M-0-Si bonds. This process is used to bond vinyl epoxy or methacryl modified silane to zirconia or titania. Afterwards this organic layer can react with organic monomers and forms an organic layer around the metal-oxo polymer. A one-pot procedure is aed to treat dialkyl siloxane R’Si(OR)2 with titanium alkoxitle in an acidic rnedi~m.~’ Short linear polydialkylsiloxane chains HO [R’SiO],R or rings are growing simultaneously to form polyoxotitanate-oxo polymer [TiO,(OH),(OR”),-,-,],. When this solution evaporates some of the reactive titania sites react with those of the PDMS units.The two networks are then crosslinked through the formation of Ti -0-Si bonds which can be demonstrated by 29Si and ”0 NMR spectroscopy.58 From the high mixing level of the two phases. transparent coatings are obtained. hv or heat catalyst -(AIBN POB) Fig. Synthesis of hybrid materials from organically modified silicon atoms J. Muter. Chem. 1996 6(4) 511-525 8g H20 -catalyst (HCI or NaOH or NH4F) Y Fig. Synthesis of hybrid materials from polyfunctional alkoxysilanes H20 -catalyst (HCI or NaOH or NH4F) c“4 Fig. Synthesis of hybrid materials from alkoxysilanes functionalized by polymers Table Synthesis of modified silicon precursors“ Synthesis Ref.Comments Si(OEt)4 + HO-R -(EtO)sSi-OR 479 trans-esterifcation Si(OEt)4 + X-R --t (EtOhSi-R X= CI Br ... R = alkyl phenyl ... &ignardreaction (Et0)3Si-H + LR-(EtO)3SiPR hydrosilation (Et0)3SiAN=C=O + HO-R -(EtO)3SiANH-$-O-R acylation0 (low yield) (Et0)3SiAN=C=0 + HzN-R +(EtO)@iANH-$-NH-R acylation 0 acylation in basic -(Et0)3SiANH-$-R (good yield) (Et0)sSiANHz + HoyR media->qO formation 0 ->condensation of SiO (EtO)3siANH-R’ + X a -(EtOI3Si”E;Ia Hofmann synthesis of R‘ x = CI S02CI amines R is a monomer or a polymer; the use of a symmetric R group leads to bis(end-capped) molecules. Complexation of TM alkoxide. The use of complexing ligands ligands. For controlled hydrolysis ratios clusters of defined appears to be a key point for the synthesis of organic-inorganic sizes and structures are synthesized.Their core is formed from structures involving transition metals for which ionic M -C a compact oxide structure while the surface is protected by bonds are easily cleaved by water. In this case strong com- the organic ligands. plexing agents are used (8-diketonates acid derivatives etc.). The synthesis of organic-inorganic structures requires the When these complexes are hydrolysed the alkoxy groups are use of complexing ligands with an organic reactive function removed thus causing the formation of metal-oxo polymers which may be polymerized further (vinyl allyl methacryl etc.). while some of the ligands are still bonded to the metallic. The behaviour of some of these chelates has been studied centre.The strong decrease of the alkoxy reactivity towards recently in titania or zirconia based materials. Generally the hydrolysis is proven and stable sols colloids or gels can be synthesis scheme is similar to those described for silica based obtained while precipitates are obtained without complexing materials (Fig. 8) synthesis of the heterofunctional precursor 518 J. Muter. Chem. 1996 6(4) 511-525 -H20 hv or heat catalyst + catalyst (HCI or NaOH or NH4F) (AIBN POB) Fig. Class I1 hybrid materials anchoring TMO to organics through a silica layer (addition of the polymerizable ligand to the TM alkoxide) controlled hydrolysis to obtain TM oxide clusters or polymers and then organic polymerization of the reacting group in the presence of a catalyst light and finally heat treatment.The stability of carboxylic acid derivatives towards hydrolysis is poor and the organic ligands are more or less removed from the metal atoms during the synthesis steps. The low reactivity of vinyl derivatives or ally1 acetonate grafted to Ti02 systems has been depi~ted.~’ Organic polymerization yields are always very low. Much better results have been obtained with meth- acryl derivatives ligands such as methacrylamido salicylate (MASA) and acetoacetoxy ethyl methacrylate (AAEM) (Fig. 12). These strong chelating ligands are added to the metallic precursor (Zr-alkoxide) with various complexing ratios6’ (AAEM :Zr 0.25-0.75). In a second step both organic and inorganic polymerizations can be managed sequentially or even simultaneously if water and a radical initiator are added to the precursor solution. Milky sols are then obtained.Spectroscopic investigations of the colloids (X-ray absorption HO methacrylamido salicylate (MASA) 0 acetoacetoxy ethyl methacrylate (AAEM) Fig. Complexation agents for TM alkoxides methacrylamido salicylate (MASA) and acetoacetoxy ethylmethacrylate (AAEM) hv or heat n catalyst (AIBN POB) IR NMR) prove that a good yield of organic polymerization is achieved and Zr complexation remains. X-Ray diffraction light scattering and TEM show that branched scatterers are formed. Their structure and texture seems to be mainly gov- erned by the hydrolysis and complexation ratios.Hybrid organic-inorganic polymers are intimately interpenetrated from the nm scale (AAEM :Zr =0.25) to the sub-pm scale (AAEM Zr =0.75).For higher complexation ratios zirconium- 0x0 clusters are connected through long polymethacrylate chains while for lower complexation ratios short organic chains crosslink larger 0x0 particles. Hybrids based on template building blocks. The use of template building blocks as starting units for obtaining hybrid organic-inorganic structures is an approach which is followed by organometallic chemists for various systems (Fig. 13). The synthesis of well defined silicon 0x0-clusters provide some examples. The most studied building blocks are based on cubic functionalized silicic acid clusters61 of formula {[(CH3)2RSi]SSiS020 with R=H (QsMaH) CH=CH2 (Q&”)}.Hydrosilylation reactions are performed to couple these blocks. Tetramethyldisiloxane tetramethylcyclotetrasi-loxane and poly(methylhydrogenosi1oxane) have also been used as reactive spacers in order to produce materials with controlled porosity. The same procedure is followed in order to product. tin-oxo based hybrid species.62 Clusters with the formula [( R’j&(p3-O)14(p2-OH)6](OH)2(HOP?) (R =butyl or buteiiyl) are obtained from the controlled hydrolysis of BuSri(OPr’) . Further organic polymerization of the butenyl derivative in the gresence of AIBN leads to entities with diameters around 60 A proving that the tin-oxo clusters are attached through Fig.Hybrids based on template building blocks J. Mater. Chew. 1996 6(4) 511-525 polybutane chains. In the case of TM-0x0 clusters the M-C bonds are broken in the presence of water. Negatively charged macromolecules based on polyoxometal-lates (POM) have also been ~tudied.6~ The POM entities are organically functionalized through the M-0-Si-C links. [SiWI1O4,J SiR),I4- units carrying two reactive organic groups (R =vinyl allyl methacryl styryl) are further poly- merized in the presence of a radical initiator to yield hybrid polymers in which POMs are linked by polymethacrylate or polystyrene chains. The synthesis of transition-metal-0x0 clusters capped with polymerizable complexing ligands has already been discussed. These clusters may also be linked together when an organic polymerization is initiated leading to an assembly of nano- building blocks.Ordered hybrid materials of class 11. The development of class I1 ordered hybrid materials (Fig. 14) is a challenge for materials chemists in order to provide new materials with improved physical applications. One method used is the building of disordered isotropic structures which may be processed further (poling mechanical deformation etc.) in order to present anisotropic character. Outstanding ordered materials exist near Langmuir-Blodgett films. Oriented films are produced by the self-assembly of molecular units. In the simplest approach trialkoxy (or tri- chloro) silanes [R'Si(OR) where R is a long aliphatic chain with more than ten carbons] react with the surface silanol (Si-OH) groups of the activated silica surface.64 An oriented monolayer of aliphatic chains is formed at the oxide surface and variable compactions and molecular orientations are achieved depending on the experimental conditions.The func- tionalization of the R' group such as the introduction of an alkene a phenoxy or a sulfone group leads to layers with different structures and the possibility of further reactions. The limitation in producing monolayers is overcome by the use of carboxy derivatives such as H,CO2C(CH,),,SiC1 and multi- step self-assembly processes. More than layers can be successively deposited OF a substrate leading to anisotropic layers of more than A in thickness.More complex organic groups such as conjugated aromatic units have also been used to build such multilayer structures. The same idea is developed in building organic-phosphonate hybrids.65 In that case the I+TMO su dace + TMO surface substrate is successively dipped in Zr salt solution and organic phosphonate solution. The building of a self-assembled multi- layer may reach pm thicknesses. Phosphonates functionalized with NL066 or fluorescent dyes,67 poly( phenylvinylene) or polythiophene68 oligomers are being widely studied. Organometallic polymers. Organometallic polymers obtained by grafting functionalized organometallic molecules (such as metallocene derivatives) are also considered at the frontier of hybrid materials.Ferrocene and its derivatives have been grafted to organic or siloxane frameuorks. leading to materials with reversible redox proper tie^.^^ Ferrocene-PMMA copolymers have recently been described as organome- tallic materials with NLO properties and large xz values have been measured." Molecular assemblies of silicon phthalocyan- ine derivatives have also been reported.-' The stacking of elemental planar units leads to polymers possessing redox properties emphasizing the strong electronic interactions between the sites. Some Applications Reaching Applications through the Versatility of the Preparation Process The wide range of synthetic procedures for obtaining organic- inorganic structures leads to the vast range of properties which these materials may possess.Obviously the final materials are not just the sum of the primary components and a large synergy effect is expected from the close coexistence of the two phases through the size-domain effects and the nature of the interfaces. Searching for a material with a given property could then appear as an endless task. Guidelines can be drawn out from the basics of materials and polymer sciences. Generally. the major features of each phase are preserved in the hybrid materials (stability thermal behaviour special properties) and generally only some shifts of the properties of each phase are to be expected. Organic us. inorganic materials properties size effects For centuries the properties of inorganic (metals.ceramics. glasses) and organic (polymers) bulk materials have been investigated with regard to their applications promoting the evolution of civilizations. During the last decades and with "20 catalyst H H H 2O H ___)catalyst H ti *o H H H Fig. Ordered hybrid materials of class I1 (a) monolayers (b)multilayers 520 J. Mater. Chem. 1996 6(4). 511-525 the help of analysis techniques the relationship between the structure and properties of these materials has become clearer Some of these general data are summarized in Table The choice of the polymer is usually guided by its mechanical and thermal behaviour However other properties such as the hydrophobic/hydrophilic balance the chemical stability the biocompatibility the optical (visible and/or IR and/or UV) and/or the electronic properties and chemical functionalities (such as the ones used to solvate molecules or ions) must be considered in the choice of the organic component. The nature of the oxide is determined by the redox properties the density the refractive index.The chemistry of the metal can also guide the chemist in his choice Obviously the nature of the bonds engaged in the organic or inorganic matter are completely different thus explaining the very different behaviour of these two families of materials Another crucial effect is seen when the sizes of the phases change from large bulks to smaller and smaller objects (typi- cally in the nm range). This size effect modulates the properties of nanophased materials,72 some interesting examples of this are (1) thermal behaviour melting points or Tgcan be shifted for tens of degrees or cancelled (2) mechanical properties changes of critical yields or strains (3) rheological and stability of solutions and dispersions the viscosity of polymeric solu- tions is strongly dependent on the molecular mass M,.The stability of colloidal solutions is dependent on the mass of the particles (4)Dynamics of molecules dissolved in liquids filling porous glasses or xerogels rotational diffusion and self- diffusion coefficients decrease markedly when the size of the liquid domain is reduced (5) electronic properties are easily changed by the size of the domains especially when the objects are in thenm range Both inorganic and organic materials exhibit such effects Polyoxometallates are TMO clusters of formula M,O,k-(M=W Mo V 6dnd40).Their structure is based on stacking of MO octahedra and they present redox reversible properties (photochromism electrochromism etc ) The redox potentials and the colours of the redox states can be changed by the number (n) of the metal atoms74 At the two size extremes are the single cation M”+ and the metallic oxide A similar trend is observed in organic conductors where the redox properties of oligomers [poly(viny1idene) or poly- (phenylene)] are adapted by changing the size of the oligo- mers 75. The studies of quantum dots (selenides and sulfides) lead to similar conclusions Optical properties are modulated by the size of the nanoclusters Some similar trends are also observed when measuring the variations of properties from 3D to 2D structures for example studying the properties of a given compound when going from bulk and nanoparticles to surfaces covered by layers There are many effects which account for these size-domain effects.The strongest effect is governed by the surface and interface effects the smaller the particles the higher the devel- oped surface. These surfaces have to be considered as defects of the bulks. The high reactivity of oxide surfaces comes from the high density of hydroxy groups or dangling bonds. They can react further to give strong bonds (through olation oxolation esterification) or weak bonds (hydrogen bonds) with the surrounding medium Molecules near a ‘solid’ surface then experience interface effects as measured from molecular dynamics 7273.The strong hindrance of mobility has been proved by many techniques in porous glasses gels composites (such as by forced Rayleigh scattering by polarized light scattering by NMR or by EPR) and the deactivation of the oxide surface by alkylsilanes decrease has been proven Many studies dealing with the properties of organic dyes embedded in an inorganic matrix imply the same effect. The surface effect can be so strong that the sol-gel matrix prevents the formation of rhodamine dimers Such a result suggests that the dye-dye interactions are weaker than the dye-matrix interactions Rhodamine molecules are absorbed at the surface of the xerogel micropores and high concentrations of dye molecules (up to mol dmP3) can be reached without dimer formation.The interactions of the surface with the dissolved molecules can be checked by controlling the inner surface of the xerogels Such surface effects have been measured in silica based materials doped with photochromic spiropy- ran. The matrix is obtained uza polymerization of Si(OCH,) or EtSi(OCH,) precursors and a ‘reverse photochromic’ effect is observed. This is related to the stronger interactions of the coloured species (zwitterionic form) with the acidic silanol Si-OH groups present at the surface of the inorganic gel through hydrogen bonding The size effect also must be considered.The most trivial way of controlling the size of the nanodomains is to form transparent materials. This requires matching of the refractive index and particles with characteristic sizes in the range of the light wavelength 77. The tremendous effect of the particle size on many physical properties for nanosized objects is measured for electronic conductivities and optical and redox properties Recent theories explain these size effects but only for perfect materials which is certainly far from reality. There again an effect of the size on the quality of the samples and the densities of defects is generally observed. Table Comparison of the propreties of organic and inorganic materials property organics (polymers) inorganics (SiO and TMO) nature of bonds T (glass transition) temperature stability density refractive index covalent [C-C] (+weaker van der Waals or hydrogen bonding) low (-100 to 200°C) low (< “C) 09-1 12-1 ionic [M-01 high (> “C) high (>> 100“C) 0-4 4-2 mechanical properties hydrophobicity permeability electronic properties processabili t y elasticity plasticity rubbery- like (depending on T,) hydrophilic hydrophobic permeable to gases insulating to conducting redox properties high molding casting machining thin films from solution viscosity control hardness strength fragility h ydrophilic low permeability to gases insulating to semiconducting (SiO,/TMO) redox properties (TMO) low for powders needs to be mixed with polymers or dispersed in solutions high for sol-gel materials (similar to polymers).J Muter Chem 1996 6(4),511-525 Tuning the properties by the chemical nature of starting materials Obviously the choice of the starting components is essential to the properties of the materials Such a great deal of work has been devoted to the individual properties of polymers and oxides (silica TMO) that reviewing these features is beyond the scope of this paper. The originality of synthesizing hybrid materials through the sol-gel process offers the possibility of mixing many organic and inorganic precursors to adapt the material properties The mutual role of organic and inorganic components is considered on the basis of their ability to provide individual networks If only one of them develops a structural network the other is considered as a network modifier.The surface modification of oxides gels or glasses by organic-inorganic molecules obviously enters this description Chromatography and derived techniques use derivatization agents with func- tional groups grafted to the silica support Alkanes acids amines chiral groups enzymes liquid-crystal molecules etc are then bonded to achieve effective separations. The formation of ordered organic monolayers on substrates proceeds from the same idea Organic glasses (plexiglass etc) have been protected against scratches by the formation of thin layers of SiO or Zr0279 on the surface Organic and inorganic com- pounds can also be mixed homogeneously in the bulk in order to modify the properties of the materials for example continu- ous changes of porosity and the density and optical properties are measured when variable amounts of (OEt)3SiCH3 are mixed with Si(OEt) in order to form methyl modified silica The tricky choice of the starting components and the need to add more and more precursors can be illustrated by some developments for multifunctional materials (a) increasing the refraction index by adding. TMO to a UV polymerizable silica- methacrylate mixture for integrated (b) adding per- fluoroalkyl chains [03SiCH,CH,(CF2),CF,] to silica or zir- conia to obtain hydrophobic coatings7'.The ratio of CH2CH2(CF2),CF3 groups leads to control of the solvent contact angle which can reach a value for water similar to PTFE ( 105"),798o (c) V,O,/modified silica structures Materials with electronic properties are obtained by mixing DEDMS (diethoxydimethylsilane) and vanadium alkoxide [VO(OAmt),] When hydrolysed in acidic media the thin films are green transparent films are obtained under neutral conditions. These differences are related to the changes in the chemical processes which result in different coordination shells of the V atoms Easily reduced vanadium-oxo species are formed at low pH (five-fold coordination) while vanadium in four-fold coordination is obtained at pH =6 5.The organic- inorganic structural PDMS network stabilizes these different species (d) Contact lens materials". The effect of adding different starting molecules is illustrated in Table4 Such a high-technology material should present good optical proper- ties (refractive index no optical defect) as well as adapted mechanical properties biocompatibility hydrophilicity and O2 permeability) To optimize the properties of these materials more than six different compounds must be added in the right ratios and different parallel or successive reactions must be performed All experimental conditions must be controlled such as the solvents the order of the reactions the catalysts and the bath temperature Evidently producing the required materials is difficult and only detailed knowledge of the different reactions involved can lead to a suitable commercial hybrid organic- inorganic nanocomposite Tuning the properties by the dynamic behaviour Most of the properties of advanced hi-technology materials are strongly related to the chemical nature and structures of the components and obviously to their dynamic behaviour The search for materials with suitable mechanical properties (plasticity elasticity strain yield etc) is governed by the molecular properties and the mobility of the polymeric chains At low temperatures reduced motion of the polymer chains is observed leading to fragility while above q,cooperative movement of the chains with larger amplitude are found to occur this leads to more elastic or rubbery compounds Controlling the glass transition temperature T is of fundamental importance for many functionalized hybrids Striking examples are emerging from the search for materials with NLO properties Initially materials with isotropic properties are synthesized and further processing is needed to achieve anisotropic properties In the first step a film is obtained from a suitable solution by spin or dip coating.The NLO dye can be grafted onto the silica networks or embedded in the matrix. The second step is the poling of the material Usually the material is heated near or above its Tg in order to increase the molecular mobility and a strong electric field is applied (kV cm-'). The Corona effect can align the dipolar dye molecules The material is then cooled down and the dye molecules are trapped with a preferred direction. The choice of a suitable matrix and the grafting of the dye units to the silica network retains the anisotropic properties Hybrid materials with ionic conductivity properties have also been described Lithium salts are first dissolved in a suitable phase an organically modified silica (amines sulfonic acids sulfonamide~)~~ or silica-polymer networks [poly(ethy1-ene glycol) poly(propy1ene glycol) etc ] Transport phen- omena of cations in polymeric structures have been widely studied and demonstrate that reasonable values of the Li' mobility are reached if the surrounding solvating media also experiences a high degree of m~bility,'~ such as those measured in polymers above Tp In PEG-silica structures the dependence of the ionic conductivity (oL1+) on temperature demonstrate polymer-like behaviour in agreement with thermal analysis and the NMR experiments which present a value of Tg in the range -60 to +20°C depending on the exact material composition The high solvating character of PEG has been studied to form PEG-SiO hybrids in which very different kind of entities such as organic dyes and molecules or salts can be dissolved Polyoxometallates ([PW12040]3- [SiW,2040]4-j have already been dissolved in such structures 86.The materials exhibit electrochemical and photochromic properties that have been investigated as supports for optical data storage It appears however that both properties could not be found in the same material. The first requires a high diffusional mobility of the POM clusters87 to increase the electronic transfer rate whereas the second needs to trap the clusters and to avoid any electronic exchange.The choice of the matrix composition and structure leads to a control of the balance favouring one of the properties Recently we investigated the dynamic- structure relationships in PEG-SiO materials55 of class I and class I1. The materials were made by condensation of silica in PEG (class I) or condensation of silica end-capped PEG (class 11) In both cases transparent materials in which small silica clusters are wrapped in the PEG phase are obtained (Fig 15) Longer chain lengths and grafting of the polymer lead to a decrease in Tg A more subtle point is certainly the lack of homogeneity of thermal behaviour in the organic phase (only a part of the PEG phase experiences the glass transition) The dynamic behaviour has been mapped with EPR (nitroxide probes) and NMR (liquid and CP-MAS) experiments which demonstrate a strong hindrance of molecular mobility near the silica surfaces Prospects and Conclusion Because of the huge versatility of the synthetic processes and the nearly infinite choice of possible combinations organic- 522 J Mater Chem 1996 6(4) 511-525 added molecule properties of the layers enhancement of properties OR RO-$i-OR OR“3 RO -Si OR Si(OR)4 or/and Ti(OR)4 0 RO-Yr-0 + MMA9 OR Replacing MMA by HEMA 02permeability ; but hydrophobic.02permeability ; but OH is lost upon condensation hydrophilicity too. 02permeability and hydrophilicity ; but porous materials with poor mechanical properties.02permeability and hydrophilicity denser materials ; but brittle materials not machinable. 02 permeability and hydrophilicity denser materials reduced brittleness increased flexibility machinable ; but low wettabilitv. 02 permeability and hydrophilicity denser materials reduced brittleness increased flexibility machinable wettable scratch resistant. Fig. Idealized structures of class I and class I1 nanocomposites SO,-PEG inorganic nanocomposites could form a broad range of advanced materials.88 Some outstanding issues of hybrids must be explored at the frontiers of science technology and knowl- edge. During early studies the cooking of complex mixtures led to materials with unexpected features; many applications of these new sophisticated materials have been described.After these initial studies a more critical point of view has been adopted and more systematic studies were carried out. A wide investigation of these materials began with the help of most of the available techniques (such as water titration elementary analysis chromatography rheology light and X-ray scattering X-ray diffraction optical techniques) and spectroscopies (such as IR UV-VIS fluorescence electrochemistry mass spec-trometry multinuclear liquid and solid-state NMR X-ray spectroscopies XANES and EXAFS EPR). They demonstrated that both structural and ‘smart’ hybrid materials may be achieved in the near future. These trends are supported by the research works published during the last few.The following examples are major fields of investigation. ( 1). The search for materials with suitable mechanical properties. This is one of the foremost areas for hybrid materials and these studies are guided by the strong similarities which could be found with structural composite^.^^. The effect of mixing and/or grafting polymers to glass fibres or woven fabrics to reinforce their structures is a current issue. When the size of both phases is reduced from the pm scale down to thenm range strong modifications of the properties is expected. The effect of changing the polymers the inorganic phase the size and shape of the domains is under investigation. The study of smart hi-technology materials with useful mechan- ical properties is beginning by mimicking the recent investi- gations of polymer science.These compounds must respond to external stimuli such as solvent composition pH light electric field or temperature. Some examples arise from photo- active polymers B9 (reversible change of size under illumination) or hydrogels (abrupt swelling to a critical point accompanied by strong changes of mechanical moduli). (2) Coating materials and membranes are very active research. The goal of these materials are evidently opposite but both deal with the control of liquid and gas diffusions. The control of porosity (active surface area size of cavities) and interactions with solvents and molecules ( hydrophobic-hydrophilic balance) are the key points of these properties.Large porosity characterizes membranes and many attempts to use polyfunctional alkoxysilanes (RO),Si-R’-Si(OR) show very promising results. The pos- sibility of enhancing selective transport inside the membranes by specific carrier groups (ligands with specific interaction crown ethers cryptands etc.) is very attractive. In contrast barrier layers and anti-scratch hard coatings have been devel- oped to protect or improve polymers or glasses. The control of the refractive index of these coatings by adding TMO particules has been developed for optical fibres and light waveguide^.^' Corrosion-inhibiting coatings for metal surfaces J. Muter. Chern. 1996,6(4) 511-525 and hydrophilic protective coatings (anti-fogging) are being studied.The presence of organic entities leads to a low porosity while the inorganics are essential for the grafting of the protective layer to the surface (3) Optics was certainly one of the first applications of hybrid materials 17b. The doping of materials with dyes fluo-rescent dyes photochromic dyes or NLO dyes etc is a very active field of research and good results are often reported Up-to-date technologies have a great need for materials with advanced properties. They may be produced from hybrids photochromic hybrid layers for optical data storage with high spatial resolution optical waveguides couplers gratings and lenses for microoptical applications stable NLO materials and devices powders doped with sensitive fluorescent molecules for sensors or gel-glass dispersed liquid crystals for electro- (nematic phases entrapped in modified silica network) etc.The study of electrochromics for smart windows has also found some success using hybrid materials. The electroactive layers and the electrolytic properties can be both modulated Most of these materials must be deposited as layers of variable thicknesses on various substrates Control of the viscosity of the solution gives a noticeable versatility to the process. The possibility of forming ordered structures must be further studied because structures with anisotropic optical properties are key materials (refractive index NLO fluorescence etc ) (4) More and more compounds presenting electronic properties are also under study Redox targets have been entrapped in hybrid matrices leading to photo- or electro- active materialsg1.These properties have also been used to study the change of structure during the sol-gel transition via the determination of self-diffusion coefficients 94. These proper- ties are also expected for redox sensors and biosensors Because of the flexibility of the chemistry the redox properties of materials can be tuned Examples arise from V,O,-DEDMS and POM-Si0,-PEG systems Very promising results arise from electronic conductors such as silica-polypyrrole silica- aniline interpenetrated networks and V,O,-polypyrrole lay-ered structures. The search for anisotropic conductivity proper- ties is clearly evidenced by this last example Recently semiconductor research has also found interesting isolating materials that can be used in organic transistors and chips95 Self-assembled monolayers of alkyltrichlorosilanes exhibit a resistivity similar to a polymer layer of nm (5) Evidently hybrid structures provide a lot of opportunity for biomaterials Mother Nature designed a lot of these archi- tectures and evidently chemists aspire to improve them!.This field covers many very different investigations. The synthesis of new composite micr~structures~~ and biomimetic assem-opens the area of ordered and anisotropic networks A variety of shapes and compositions have been elaborated from mixtures of lipids and iron while silicate-surfactant mesophases are under development 98.They are also potentially useful for mimicking biogenic structures and the growth of minerals in living cells and bodies. This approach provides new tools for understanding the interactions and reactivities of molecules in confined media which can be used in the field of catalysis A different use of hybrid materials has been described with biosensors and biomolecular materials in which proteins are encap~ulated~~ with regard to the delicate nature of proteins and enzymes (narrow pH range) intricate con-ditions for the material synthesis are required A change of colour of protein doped compounds is observed when changing the medium demonstrating the porosity of the materials and the sensitivity of the units Good yields of enzymatic activity (oxalate oxidase glucose oxidase etc ) and biocatalysis by lipases'" have been measured in transparent hybnd compounds Only a few properties and potential applications of hybrid materials have been described in this paper An exhaustive list of compounds would need a much bigger volume Furthermore 524 J Muter Chem 1996 6(4) 511-525 the quick expansion of this young field of research prevents such a possibility where many papers are published every month Development of the hybrids always demands more and more chemists to build these structures more and more physicists to exploit their potentialities.Their complexity is a challenge both for most of the up-to-date analyses techniques and for the attempt to use theoretical models to forecast the properties of new compounds Obviously the study of hybrid organic-inorganic materials requires the association of scientists from many disciplines The major potential of these materials is certainly in producing more and more ordered and anisotropic structures.The control of the specific interactions between both phases is a key in this direction It should also promote better homogeneity and better properties of these materials in the future References [In regard to the immense number of papers in each field only the most representative papers for each theme are cited in this bibliography 1-_-1 Better Ceramics. Through Chemistry VI eds A K Cheetham C J Brinker M L Mecartney and C Sanchez Muter Res Soc Svmp Proc ,1994,346 2 Seventh International Workshop on Glasses and Ceramics from Gels Special issue of J Sol Gel Sci Tech 1994,2 3 Proceedings of the First European Workshop on Hybrid Organic Inorganic Materials (Synthesis Properties Applications) eds C Sanchez and F Ribot Bierville-France 4 (a) Hybrid Organic-Inorganic Materials ed C Sanchez and F Ribot Special issue of New J Chem 1994 18 (b) Hybrid Organic-fnorganic Materials ed L L Klein and C Sanchez Special Issue of J Sol-Gel Sci Tech 1995 5 U Schubert N Husing and A Lorenz Chem Mater ,1995,5 5 C Sanchez and F Ribot New J Chem 1994,18,1007 6 C J Brinker and G Scherrer in Sol-Gel Science,.The Physics und Chemistry of Sol-Gel Processing Academic Press San Diego 1989 7 J Livage M Henry and C Sanchez Prog Solid State Chem 1988,18,259 8 D C Bradley R C Mehrotra and D P Gaur in Metal Alkoxides Academic Press London 9 (a) Sol-Gel technology for thin films fiber preforms electronics and especialty shapes ed L C Klein Noyes Park Ridge NJ 1988 (b)H Schmidt in Chemistry Spectroscopy and Applications of Sol Gel Glasses ed R Reisfeld and C K Jmgensen Springer- Verlag Berlin 1991 (c) H Schmidt in Ultrastructure Processing of Advanced Materials ed D R Uhlmann and D R Ulrich Wiley New York 1992 ch 38 (d) H Schmidt in Chemical Processing of Advanced Materials ed L L Hench and J K West Wiley New York 1992 ch 10 C Sanchez F Ribot and S Doeuff in Inorganic and Organometcillrc Polymers with Special Properties ed R M Laine NATO AS1 series 1992,206,257 11 R K Iler in The Chemistry ofSrlica Wiley New York 1979.12 C Sanchez J Livage M Henry and F Babonneau J Non-Cryst Solids 1988 100 13 B E Yoldas J Muter Sci ,1986,21,1086 14 R C Mehrotra R Bohra and D P Gaur in Metal b-diketonates and Allied Derivatives Academic Press London 15 (a) C Sanchez M In P Toledano and P Griesmar Muter Res Soc Symp Proc 1992 271 669 (b) F Ribot P Toledano and C Sanchez Chem Mater ,1991,3,759 16 M Chatry M In M Henry C Sanchez and J Livage J Sol-Gel Sci Tech 1994,1,233 17 (a)R Reisfeld and Ch K Jsrgensen in Chemistry Spectroscopy and Applications of Sol-Gel Glasses ed R Reisfeld and C K Jsrgensen Springer-Verlag Berlin 1 p 207 (b) Sol-Gel Optics ed J D MacKenzie and D R Ulrich Proc SPIE 1990 1328,1992,1758,1994,2288 18 D Levy S Einhorn and D Avnir J Non-Cryst Solids 1989 113,137 19 M Canva P Georges A Brun F Chaput F Devreux and J P Boilot in Sol-Gel Optics ZI,ed J D Mackenzie Proc SPIE Washington 1758 20 (a) R Reisfeld D Brusilovsky M Eyal E Miron Z Burshtein and J Ivri Chem Phys Lett 1989 160 43 (b) E J A Pope A Asami and J D Mackenzie J Muter Res 1989,4,1018.(a) T Saegusa and Y Chujo Polym Prep 1989 1 39 (b) A Morikawa Y Iyoku M Kakimoto and Y Imai J Muter Chem 1992,2,679 M Toki T Y Chow T Ohnaka H Samura and T Saegusa Polym Bull 1992,29,653 S Kure H Matsuki R Jordan Y Chujo and T Saegusa Polym Prep Jpn 1990,39,1684 B M Novak Ado Muter 1993,5,422 B M Novak and C Davies Macromolecules 1991,24,5481 M W Ellesworth and B M Novak Chem Mater 1993,5,839 (a) R J P Corriu D Leclercq A Vioux M Pauthe and J Phalippou in Ultrastructure Processing of Advanced Ceramics ed J D Mackenzie and D R Ulrich Wiley New York 1988 113 (h) R J P Corriu C Guerin and J E E Moreau Top Stereochem 1984,43 (a) D O’Hare in inorganic Materials ed D W Bruce and D O’Hare Wiley Chichester 1992 p 165 (b)E Ruiz-Hitchky Adc Muter 1993,5 J Livage Chem Muter 1991,3 M Nabavi C Sanchez and J Livage Eur J Solid State inorg Chem 1991,28,1173 (a) P Aldebert N Baffier N Gharbi and J Livage Muter Res Bull 1981,16,669 (b)A Bouhaouss and P Aldebert Mater Res Bull 1983,18,2247 B Casal E Ruiz-Hitzky M Crespin D Tinet and J C Galvan J Chem Soc ,Faraday Trans 1989,85,4167 Y J Liu D C Degroot J L Schindler C R Kannewurf and M G Kanatzidis Adv Muter 1993,5 (a)M G Kanatzidis and C G Wu J Am Chem Soc 1989 111 4139 (h)M G Kanatzidis C G Wu H Marcy D C Degroot and C R Kannewurf Chem Muter 1990,2,22 C R Martin Acc Chem Res 1995,28,61 S Baral and P Schoen Chem Mater ,1993,5,145 A Firouzi D Kumar L M Bull T Besier P Sieger Q Huo S A Walker J A Zasadzinski C Ghnka J Nicol D Margolese G D Stucky and B F Chmelka Science 1995,267,1138 T Ddbadie A Ayral C Guizard L Cot J C Robert and Poncelet Muter Res Soc Symp Proc 1994,346 I Gautier Luneau A Mosset J Galy and H Schmidt J Muter Scz ,1990,89,3739 65 73 81 C Sanchez J Livage M Henry and F Babonneau J Non-Cryst Solids 1988 100 C Sanchez and M In J Non-Cryst Solids 1992,1471148 D Hoebbel K Endres T Reinert and H Schmidt Muter Res Soc Symp Proc 1994,346,863 F Ribot F Banse and C Sanchez Muter Res Soc Symp Proc 1994,346,121 P Judeinstein Chem Muter 1992,4,4 A Ulman Adv Mater 1990,2,573 Guang Cao Hin-Gi Hong and T E Mallouk Ace Chem Res 1992,25,420 (a)H E Katz W L Wilson and G Scheller J Am Chem SOL 1994,116,6636 (b)N J Long Angew Chem ,Int Ed Engl 1995 34,21 K Motesharei and D C Myles J Am Chem Soc 1994 116 H E Katz S F Bent W L Wilson M L Schilling and S B Ungashe J Am Chem Soc 1994,116,6631 T Inagaki H S Lee,T A SkotheimandY Okamoto,J Chem Soc Chem Commun 1989,1181 M E Wright E G Toplikar R F Kubin and M D Seltzer Macromolecules 1992,25 D C Gale and J D Gaudiello J Am Chew SOC 1991,113,1610 Molecular Dynamics in Restricted Geometries ed J Klafter and J M Drake Wiley New York J M Drake and J Klafter Phys Today May 1990 p M T Pope and A Muller Angew Chem Int Ed Engl 1991 30,34 R Schenk H Gregorius K Meerholz J Heinze and K Mullen J Am Chem Soc 1991,113,2634 D Avnir D Levy and R Reisfeld J Phys Chem 1984,88,5956 H Krug and H Schmidt New J Chem 1994,18 J Pesek and T Cash Chromutogruphia 1989,27 R Kasemann and H Schmidt New J Chem 1994,18,1117 K Izumi H Tanaka M Murakami T Deguchi A Morita N Tohge and T Minami J Non-Cryst Solids 1990,121 C Sanchez B Alonso F Chapusot F Ribot and P Audebert J Sol-Gel Sci Technol 1994,2 (a) H Schmidt and B Seiferhng Mater Res Soc Symp Proc 1986 73 739 (b) H K Schmidt Muter Res Soc Symp Proc 1990,180,961 40 F Chaput D Riehl Y Levy and J P Boilot Chem Muter 1993 (a)E Toussaere J Zyss P Griesmar and C Sanchez Nun-Linear 5,589 Optics 1991 1 349 (b)C Sanchez P Griesmar E Toussaere 41 B K Coltrain,C J T Landry J M O’Reily,A M Chamberlain G Puccetti I Ledoux and J Zyss Non-Linear Optich 1992 4 G A Rakes J S S Sedita L W Kelts M R Landry and V K Long Chem Muter 1993,5,1445 J Y Sanchez A Denoyelle and C Poinsignon Polvm Adv 42 50 M E Brik J P Bayle and P Judeinstein to be published (a) P Griesmar C Sanchez G Pucetti I Ledoux and J Zyss Mu1 Eng 1991 1 205 (b) B Lebeau J Maquet C Sanchez R Hierle E Toussaere and J Zyss J Mater Chem 1994,4,1855 R H Glaser G L Wilkes and C E Bronnimann J Non-Cryst Solidr 1989 113 K J Shea D A Loy and Webster J Am Chem Soc 1992 114,6700 R J P Corriu J J E Moreau P Thepot and M Wong Chi Man Chem Muter 1992,4,1217 G M Jamison D A Loy and K J Shea Chem Muter 1993 5 D E Rodriguez A B Brennan C Betrabet B Wang and G L Wilkes Chem Mater 1992,4 J Krdmer R K Prud’homme and P Wiltzius J Colloid interface Sci 1987,118,294 F Devreux J P Boilot F Chaput and A Lecomte Phvs Rev A 1990,41,6901 P Judeinstein J Titman M Stamm and H Schmidt Chem Mater 1994,6.(a)J Cazes C R Acad Sci 1988 247 1874 (b) L W Breed R L Elliott W J Haggerty Jr and F Baiocchi J Org Chem 26 J L Speier Adv Organornet Chem 1979,17,407 D P N Satchell and R S Satchell Chem SOC Rev 1975,4,231 P Judeinstein M E Brik J P Bayle J Courtieu and J Rault Muter Res SOC Symp Proc 1994,346,537 H W Oviatt Jr K J Shea S Kalluri Y Shi W H Steier and L R Dalton Chem Muter 1995,7,493 91 99 Technol 1993,4,99 C A Vincent in Electrochemical Science and Technology of Polymers-2 ed R G Linford Elsevier Amsterdam 1990 p P Judeinstein and H Schmidt J Sol-Gel Sci Technol 1994 3 P Judeinstein P W Ohveira J P Bayle and J Courtieu J Chim Phys 1994,91,1583 Hybrid Orgunic-Inorganic Composites ed J E Mark C Y-C Lee and P A Bianconi ACS Symp Ser ,Washington DC 1995,585 Functional Monomers and Polymers ed K Takemoto Y Inaki and R M Ottenbrite Marcel Dekker New York R Yoshida K Uchida Y Kaneko K Sakai A Kikuch Y Sakurai and T Okano Nature 1995,374,240 C Guizard and P Lacan New J Chem ,1994,18,1097 D Levy J M S Pena C J Serna and J M Oton J Non-Cryst Solids 1992,147/148,646 (a) P Judeinstein J Livage A Zarudiansky and R Rose Solid Stute ionzcs 1988 28/30 1722 (b) M A Maddo and M A Aegerter J Sol-Gel Sci Technol 1994,2,667 P Audebert P Griesmar P Hapiot and C Sanchez J Muter Chem 1992,2,1293 D Vuillaume and F Rondelez La Recherche 1995,275,461 H J Watzke and C Dieschbourg Adv Colloid Interface Sci 1994,50 Cited in Chem Eng News August 1993 p M E Davis Cong-Yan Chen S L Burkett and R F Lobo Mater Res Soc Symp Proc ,1994,346,831 (a)J I Zink J S Valentine and B Dunn New J Chem 1994,18 1109 (b)D Avnir Arc Chem Res 1995,28,328 M T Reez A Zonta and J Simpelkamp Angew Chem Int Ed 57 S Dire F Babonneau J Livage and C Sanchez J Muter Chem Engl 1995,34,301 1992 2,239 58 F Babonneau Mater Res Soc Symp Proc 1994,346,949 Paper 5/03272E Received 22nd May J Muter Chem 1996 6(4) 511-525