Vol. 3

INORGANIC POLYMERS

29

INORGANIC POLYMERS Introduction Inorganic polymer science represents an area that has been held back by the synthetic problem of constructing macromolecular chains. However, advances in the last two decades or so of the twentieth century have led to the preparation of a variety of new polymers that contain main group elements, transition-metals, and even lanthanides. It is plausible that some of these new materials, with properties that are difficult or impossible to achieve with existing organic materials, may fulfill the requirements of specialized markets; such developments remain an interesting future challenge. Polysiloxanes (silicones) represent the most well-developed class of inorganic polymers and these materials are discussed in a separate article (see SILICONES). In addition, although surveyed briefly in this article, more detailed information on POLYPHOSPHAZENES and POLYSILANES and POLYCARBOSILANES can be found elsewhere in the Encyclopedia. Here we focus on other main classes of inorganic polymers.

Inorganic Polymers Based on Main Group Elements Polyphosphazenes. Polyphosphazenes (1) have a polymeric backbone composed of alternating phosphorus and nitrogen atoms. The side groups, R, can be alkoxy, aryloxy, amino, alkyl, aryl, inorganic, or organometallic groups. This large range of accessible structural variations is accompanied by a wide range of polymer properties that are highly dependent upon the nature of the side groups (see Table 1). Encyclopedia of Polymer Science and Technology. Copyright John Wiley & Sons, Inc. All rights reserved.

30

INORGANIC POLYMERS

Vol. 3

Table 1. Properties of Selected Polyphosphazenesa Formula

T g ,◦ C

T m ,◦ C

[NP(O C4 H9 )2 ]n [NP(On C3 H7 )2 ]n (NPF2 )n

−105 −100 −96

−68, −40

[NP(OC2 H5 )2 ]n [NP(OCH2 CH2 OCH2 CH2 OCH3 )2 ]n [NP(OCH3 )2 ]n (NPCl2 )n

−84 −84 −76 −66

−7.2 (39)b

[NP(OCH2 CF3 )2 ]n

−66

242

[N3 P3 (OCH2 CF3 )5 (CH3 )]n [N3 P3 (OCH2 CF3 )x (CH2 Si(CH3 )3 )]n [NP(OCH2 CF3 )(OCH2 (CF2 )x CF2 H)]n [NP(OC9 H19 )2 ]n [NP(CH3 )(alkyl)]n [NP(CH3 )2 ]n [NP(n C6 H13 )2 ]n (NPBr2 )n

−63 −61 −60c −56 −50d −46 −29 −15

[NP(OC6 H5 )(OC6 H4 C2 H5 )]n [NP(OC6 H5 )2 ]n

−10e −8

390

[NP(OC6 H4 COOH)2 ]n [NP(OC6 H4 COOC2 H5 )2 ]n

−5 8

127

n

[NP(OC6 H4 CH3 )(OC6 H4 CHO)]n [NP(NHCH3 )2 ]n [NP(OC6 H5 )(OC6 H4 C6 H5 -o)]n [NP(NHC2 H5 )2 ]n [NP(CH3 )(C6 H5 )]n [NP(OC6 H5 )(OC6 H4 C6 H5 -p)]n [NP(CH3 )(alkyl)]n [N3 P3 (OCH2 CF3 )4 (C5 H4 FeC5 H4 )]n g [NP(NHC6 H5 )2 ]n [NP(OC6 H4 C6 H5 -p)2 ]n a Refs.

f

11 14 24 30 37 43 ∼50g 61 91 93

1,2–3. the stretched polymer. c Varies with values x and ratio of side groups. d Broad, poorly defined transitions. e Varies with ratio of side groups. f Complex melting phenomena. g Ferrocenyl polymer. b For

143 129

>350

Properties Elastomer Elastomer Hydrolytically unstable elastomer Elastomer Water-soluble elastomer Elastomer Hydrolytically unstable elastomer Microcrystalline thermoplastic (films) Elastomer Elastomer Elastomer Elastomer Amorphous gums or waxes Microcrystalline powder Wax-like solid Hydrolytically unstable, leathery material Elastomer Microcrystalline thermoplastic (films, fibers) Glass, soluble in aqueous base Microcrystalline thermoplastic (films) Thermoplastic Water-soluble glass and film former Glass Glass, soluble in aqueous acid Glass Amorphous gums or waxes Amber-colored glass Glassy thermoplastic Microcrystalline thermoplastic (high η)

Vol. 3

INORGANIC POLYMERS

31

In addition, the phosphorus–nitrogen backbone inherently possesses a unique range of unusual properties. For example, it is extremely flexible which in turn can give rise to low glass-transition temperatures, particularly in the case of poly(alkoxyphosphazenes) such as the n-butoxy derivative (T g = −105◦ C) (1,4). Furthermore, the backbone is thermally and oxidatively stable, as well as optically transparent from 220 nm to the near infrared region, which makes it resistant to breakdown in many harsh environments, as evidenced by the flame-retardant properties of many polyphosphazenes. Ring-Opening Polymerization (ROP). The first polyphosphazene synthesized, poly(dichlorophosphazene) (2), was prepared in cross-linked form by Stokes at the end of the nineteenth century by the thermal ROP of the cyclic trimer [Cl2 PN]3 (3) (1). This material, referred to as “inorganic rubber,” remained a chemical curiosity because of its intractability and hydrolytic instability until the mid-1960s when it was shown that if the ROP of pure [Cl2 PN]3 is carried out carefully, uncross-linked poly(dichlorophosphazene) (2), which is soluble in organic solvents, is formed (5). Subsequent reaction of this highly reactive polymeric species with nucleophiles has been shown to yield a wide range of hydrolytically stable poly(organophosphazenes) (eq. 1) (1,6–8).

(1)

This macromolecular substitution route, used primarily with alkoxides, aryloxides, or primary amines, is largely responsible for the immense structural diversity of poly(organophosphazenes) and allows the tuning of certain properties and the introduction of others through the choice of nucleophile. Highlights concerning the use of the macromolecular substitution route with 2 and related materials involve the introduction of side groups, which lead to liquid crystallinity (see polymer 4) (9), photochromism (10), photocross-linkability (11– 13), and the preparation of novel polymers such as (5) with short-chain branching (14,15).

32

INORGANIC POLYMERS

Vol. 3

The thermal ROP route requires the synthesis and careful purification of the cyclic trimer [Cl2 PN]3 and the use of elevated temperatures where control of molecular weight is very difficult and cross-linking can take place at high conversion, which can limit the yield. Condensation Polymerization. Although several polyphosphazenes have been commercialized, much work has focused on the development of cheaper and more convenient methods for making these materials. Also, the reaction of poly(dichlorophosphazene) with organometallic reagents such as Grignard or organolithium reagents generally leads to chain cleavage as well as substitution, and thus macromolecular substitution does not provide a satisfactory route to polymers with only alkyl and aryl side groups bound by direct P C bonds. To these ends several promising condensation routes have been developed. In the early 1980s, a condensation route to polyphosphazenes from phosphoranimines was discovered (eq. 2) (16).

(2) The polymerization is in fact a chain-growth reaction and allows access to high molecular weight polyphosphazenes such as poly(dimethylphosphazene) and poly(methylphenylphosphazene) (2). Methyl deprotonation/substitution of these polymers as well as electrophilic aromatic substitution of the phenyl substituents in poly(methylphenylphosphazene) have been developed as versatile strategies for the derivatization of both of these polymers (eq. 3) (3).

(3)

Vol. 3

INORGANIC POLYMERS

33

An alternative, direct route to fluoroalkoxy phosphazene polymers and aryl derivatives which also permits access to block copolymers has been developed (eqs. 4 and 5) (17,18).

(4)

(5) The development of condensation routes to poly(dichlorophosphazene) have also been reported. One promising route operates at 200◦ C (eq. 6) (19).

(6) In 1995, details of a synthesis of poly(dichlorophosphazene) which operates at room temperature and allows for good molecular weight control were reported (eq. 7) (20). It involves the condensation of phosphoranimines in the presence of cationic initiators such as a PCl5 . The polymerization has been shown to proceed via a cationic chain-growth mechanism that shows “living” characteristics (21). Two equivalents of the initiator react with one equivalent of a chlorinated phosphoranimine to form a reactive ion pair (eg, [Cl3 P N PCl3 ]+ PCl6 − ), which further reacts with monomer to propagate chain growth.

(7) This synthetic method has been extended to the direct synthesis of poly(organophosphazenes) as well as the development of star and block copolymers. For example, triarmed star-branched polyphosphazenes (eg, 6) can be synthesized through the initiation of trifunctional phosphoranimines (22). It has also been shown that the presence of “living” active sites at the termini of the polymer chains allows for addition of a second monomer and the formation of block copolymers (23), such as (7) which is formed through the initiation of a difunctional linear phosphoranimine and the subsequent introduction of two different monomers (24). These developments offer the prospect of improved routes

34

INORGANIC POLYMERS

Vol. 3

to phosphazene polymers, which may well facilitate more rapid and extensive commercialization.

Monomer Synthesis. Both the ROP and condensation routes to poly(dichlorophosphazene) require high purity monomers. For the traditional ROP process, ultrahigh purity [NPCl2 ]3 is required. The most common method for its preparation is through the reaction of PCl5 with NH4 Cl in a high boiling halogenated solvent, such as chlorobenzene or tetrachloroethane, at 150◦ C. The yields can approach 70–80% under specialized conditions, but 50% is more typical (25). The phosphoranimine most commonly used in the aforementioned condensation route to poly(dichlorophosphazene) is Cl3 P NSi(CH3 )3 . The synthesis of this compound has been described (21) and is based on a modification of an earlier procedure (26) involving the reaction of PCl5 with LiN(Si(CH3 )3 )2 at −78◦ C in hexane. However, the purification required by this method is a challenge and leads to an overall yield of 105 . Since 1980, there has been a remarkable growth in interest concerning these polymers and they have been found to possess a variety of fascinating properties. The backbone of silicon atoms gives rise to unique electronic and optical properties.

One of the most remarkable features of the all-silicon backbone is that it leads to the delocalization of σ electrons, a phenomenon which is virtually unknown in carbon chemistry (30). This can be understood in terms of the nature of the molecular orbitals associated with the Si Si σ bonds. These are more diffuse than those associated with C C σ bonds as they are constructed from higher energy 3s and 3p atomic orbitals. This leads to significant interactions between the adjacent Si Si σ bonds along a polysilane chain, a situation analogous to that for the π bonds in π-delocalized polymers such as polyacetylene. Thus, a band model is more appropriate than a localized model (1,28). As a consequence of the delocalization of σ electrons, the σ σ ∗ transition, which occurs at 220 nm in (CH3 )3 Si Si(CH3 )3 , moves to lower energy as the number of silicon atoms in the chain increases. In the high polymers, the σ σ ∗ band-gap transitions occur in the near uv region at ca 300–400 nm (Table 2). The lowest absorption bands of polysilanes have also recently been assigned to excitonic rather than interband transitions (32). The electron delocalization also leads to appreciable electrical conductivity following doping. For example, conductivities of up to 0.5 S/cm have been reported for polysilastyrene (10), after doping with AsF5 (1). In addition, many of the polymers are thermochromic as the conformations adopted by the polymer change with temperature, which alters the degree of σ -delocalization along the main chain. Because of their low energy σ σ ∗ transitions, polysilanes

36

INORGANIC POLYMERS

Vol. 3

Table 2. Ultraviolet–visible Absorption Data for Typical Polysilanesa,b R2

λmax , nm

ε/SiSi

C3 H7 C4 H9 n C6 H13 n C12 H25 C2 H4 -C6 H5 c C6 H13 n C3 H7 n C4 H9 n C6 H13 n C14 H29 n C6 H13 n C8 H17 C6 H5 p-C6 H4 CH3 p-C6 H4 OCH3 p-C6 H4 On C6 H13 p-C6 H4 C6 H5 β-naphthyl p-C6 H4 C2 H5 p-C6 H4 n C4 H9 p-C6 H4 s C5 H11 p-C6 H4 t C4 H9 p-C6 H4 n C6 H13 m-C6 H4 n C6 H13 p-C6 H4 n C6 H13

306 304 306 309 303 326/320 310 314 316/317 318 322 318 341 337/338 344 354 352 350 390 395 390 376 397 400 394

5600

R1 CH3 CH3 CH3 CH3 CH3 CH3 n C3 H7 n C4 H9 n C6 H13 n C14 H29 n C3 H7 n C6 H13 CH3 CH3 CH3 CH3 CH3 CH3 p-C6 H4 C2 H5 p-C6 H4 n C4 H9 p-C6 H4 s C5 H11 p-C6 H4 t C4 H9 p-C6 H4 n C6 H13 m-C6 H4 n C6 H13 p-C6 H4 n C6 H13 a In

n n

5100 5000 9950 7390 8400 9700 8400 10600 8785 9300 8600 8180 5400 4000 2800 10200 26600 16200 3400 23300 21300 18600

solution. 28 and 31.

b Refs.

are photosensitive and have attracted considerable attention as photoresist materials in microlithography (1,27,28).

Synthesis. The first report of a soluble polysilane appeared in 1978 and the material was prepared by the treatment of a mixture of organodichlorosilanes with sodium metal (33). Instead of only the expected cyclic oligomers, a polymeric product, termed polysilastyrene (10), was formed. Poly(dimethylsilane) had been previously prepared as a highly crystalline insoluble material (1,27,28). The introduction of phenyl groups in the random copolymer reduces the crystallinity and allows the material to be soluble and processible.

Vol. 3

INORGANIC POLYMERS

37

The main method used to synthesize polysilanes involves the thermally-induced Wurtz coupling reaction of organodichlorosilanes with alkali metals (eq. 8). Although improvements in this process have been reported (eg, the use of ultrasound), the harsh conditions for this reaction tend to limit the side groups that can be successfully introduced to nonfunctionalized alkyl and aryl units and makes scale-up unattractive (1).

(8) Because of these limitations, considerable effort has been focused on the development of new synthetic routes to polysilanes. Transition-metal-catalyzed dehydrogenative coupling, discovered in 1985 (eq. 9) (34), is potentially very attractive; however, the molecular weights of the polysilanes formed to date are generally quite low (M n < 8,000). The catalysts used for these coupling reactions are usually titanocene or zirconocene derivatives (34,35).

(9) The catalytic dehydrogenation route yields novel polysilanes with Si H functionalities, which are of interest as ceramic precursors (36). In addition, it has been shown that a variety of new side groups can be introduced using a derivatization approach (eq. 10) (37).

(10) In 1991, a novel ROP route to polysilanes was reported (eq. (11)) (38). The key to this approach is to take readily accessible octaphenylcyclotetrasilane, which is too sterically crowded to undergo ROP, and to replace some of the phenyl groups by smaller methyl substituents (via a two-step process) to make the ring polymerizable.

(11)

38

INORGANIC POLYMERS

Vol. 3

Another route to polysilanes that involves the anionic polymerization of disilabicyclooctadienes, which function as sources of masked disilenes (eq. 12), has been described (39,40). Amphiphilic block copolymers formed by this anionic route, such as poly(1,1-dimethyl-2,2-dihexyldisilene)-b-(2-hydroxyethyl methacrylate), undergo self-assembly to form micelles (41).

(12)

Uses. The delocalization of σ electrons in polysilanes gives rise to unique electronic and optical properties. Also, several polysilanes have been found to function as useful thermal precursors to silicon carbide fibers and these materials have attracted attention with respect to microlithographic applications and as polymerization initiators (1,27,28). The use of these materials as hole transport layers in electroluminescent devices has also been explored (42). Indeed, the photoconductivity of poly(methylphenylsilane) doped with C60 has been studied and has been found to be comparable with the best materials available (43).

Polygermanes and Polystannanes Properties. The remarkable properties of polysilanes has led to significant interest in the development of polymer chains based on the heavier Group 14 elements, germanium and tin. Studies of polygermanes indicate that the σ delocalization is even more extensive than for polysilanes and that the σ σ ∗ band-gap transition for the high polymers is significantly red-shifted by ca 20 nm in comparison to the silicon analogues (44,45) (Table 3). Other studies have shown that these materials possess semiconductive behavior upon oxidative doping (49) as well as significant nonlinear optical behavior (50) and thermochromicity (45). High molecular weight polystannanes possess σ electrons that are extensively delocalized as illustrated by the band-gap transition, which occurs at 384–388 nm for poly(dialkylstannanes) (in THF) and at even higher wavelengths for some poly(diarylstannanes) (Table 3) (48,51). In addition, exposure of thin films of the polymers to the oxidant AsF5 leads to significant electronic conductivities of ca 0.01–0.3 S/cm (52). Polystannanes are highly photosensitive and exhibit photobleaching behavior, and on uv-irradiation depolymerize to yield cyclic oligomers. The materials are thermally stable to 200–270◦ C in air and, at more elevated temperatures, function as interesting precursors to SnO2 (52). Synthesis. Polygermanes (11) were prepared in the mid-1980s by Wurtz coupling techniques similar to those used to prepare the silicon analogues (eq. 13) (1). A variety of alkyl and aryl derivatives can be prepared by this method but the harsh reaction conditions are not tolerant of many functional groups (53).

Vol. 3

INORGANIC POLYMERS

39

Table 3. Ultraviolet–visible Absorption Data for Some Typical Polygermanes and Polystannanes [ER1 R2 ]n a E

R1

R2

λmax , nm

Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Ge Sn Sn Sn Sn Sn Sn

CH3 CH3 CH3 CH3 CH3 CH3 C2 H5 n C3 H7 n C4 H9 n C5 H11 n C6 H13 C6 H5 p-C6 H4 t C4 H9 p-C6 H4 n C6 H13 o-C6 H4 C2 H5 p-C6 H4 On C4 H9 o-C2 H5 -p-On C4 H9 C6 H4 p-C6 H4 N(Si(CH3 )3 )2

C6 H5 p-C6 H4 F p-C6 H4 CF3 p-C6 H4 CH3 m-C6 H4 (CH3 )2 p-C6 H4 OCH3 C2 H5 n C3 H7 n C4 H9 n C5 H11 n C6 H13 n C4 H9 p-C6 H4 t C4 H9 p-C6 H4 n C6 H13 o-C6 H4 C2 H5 p-C6 H4 On C4 H9 o-C2 H5 -p-n C4 H9 -C6 H4 p-C6 H4 N(Si(CH3 )3 )2

332b /327c 336b 332b 326b 330b 338b 293c 312c 325c 327c 325c 337c 432d 436d 468d ,e 448d 506d 450d

a All

values were measured in THF unless otherwise noted. via demethanative coupling (46). c Synthesized via electrochemical polymerization (47). d Synthesized via catalytic polymerization (48). e Value obtained on a thin film of the polymer. b Synthesized

(13) Dehydrocoupling has been investigated, but has proven relatively unsuccessful (54). Electrochemical reduction of halogermanes has proven somewhat successful and has provided a route to poly(germane–germane) and poly(germane–silane) copolymers (47,55). The ruthenium catalyst, Ru(P(CH3 )3 )4 (CH3 )2 , can be employed in the demethanative coupling of trimethylgermane, which gives relatively high molecular weight polygermanes under mild conditions (25◦ C) (eq. 14) (46,56).

(14) Early attempts to generate polystannanes by Wurtz coupling of organodichlorostannanes have yielded only low molecular weight oligomers and reduction products. The first high molecular weight materials were made using

40

INORGANIC POLYMERS

Vol. 3

transition-metal-catalyzed dehydrogenative coupling of secondary stannanes R2 SnH2 (eq. 15) (51). Yellow polystannanes (12) (R = n-butyl, n-hexyl, or n-octyl) of substantial molecular weight (up to M w = ca 96,000, M n = ca 22,000) were prepared using various zirconium catalysts. This method is applicable to the preparation of poly(diarylstannanes) possessing solubilizing substituents (48,57). Interestingly, a change in catalyst from zirconocene-based systems to HRh(CO)(P(C6 H5 )3 )3 has been shown to lead to highly branched polystannane structures (58).

(15)

Uses. Both polygermanes and polystannanes may find applications because of their unique optical, electronic, and chemical properties. Some of these potential uses include photoresist layers (44,59,60), third-order nonlinear optical materials (50), charge transport polymers (61,62), photoconductors, microlithographic materials (63), and photoinitiators (59). Boron-Containing Polymers Boron-containing polymers are of considerable intrinsic interest, as possible reactive intermediates and as precursors to boron-based ceramics (64–69). The synthesis of polyborazines (13) (M w up to ca 7600, M n up to ca 3400) via thermally induced dehydropolymerization of borazines (eq. 16) has been reported (68). The polymers were isolated as white solids and characterization suggested the presence of a significantly branched structure. Pyrolysis at 1200◦ C yielded white turbostratic boron nitride in 85–93% yield.

(16) A wide range of novel polymers with boron in the backbone have been prepared by means of boration polymerizations (65,70–82). Diynes can be polymerized by hydroboration (70,71), phenylboration (65), and haloboration (72) to yield polymers (14),(15), and (16) (eq. 17). When an appropriate aromatic or heteroaromatic diyne is used, the resulting polymers have been shown to have extended π-conjugation through the vacant p-orbital of the boron atom (73,74). In fact, several have been shown to exhibit blue fluorescence emission.

Vol. 3

INORGANIC POLYMERS

41

(17) Diisocyanates have been shown to undergo haloboration–phenylboration polymerization to give halosubstituted polymers (17) (75). They also undergo alkoxyboration in the presence of mesityldimethoxyborane to produce poly(boronic carbamates) (18) (eq. 18) (76).

(18) Dicyano compounds undergo hydroboration to produce poly(cyclodiborazanes) (19), which have proven relatively stable towards air and thermal oxidation (eq. 19) (77–79). Some examples have been prepared where the dicyano compound allows for incorporation of a charge transferred structure (80). These exhibit extended π-conjugation through the cycloborazane unit.

(19)

42

INORGANIC POLYMERS

Vol. 3

Finally, the polycondensation reaction between bifunctional Grignard reagents and aryldimethoxyboranes, carried out under a nitrogen atmosphere, in THF, gives rise to poly(phenylene-boranes) (20) (eq. 20) (83). The conjugated polymers may have potential applications in electronic devices such as LEDs.

(20) Transition-metal-catalyzed dehydrocoupling of phosphine–borane adducts (21) has been shown to give rise to high molecular weight polyphosphinoboranes (22) (84). In the 1950s and 1960s, several claims of the synthesis of polyphosphinoboranes (22) were made (eq. 21). The main route studied was the thermal dehydrocoupling of R2 PH·BH3 adducts at 200◦ C and above; however, structural characterization of the polymers was minimal and reported yields and molecular weights were very low (85,86).

(21) The dehydrocoupling of various adducts (21) has now been studied in the presence of various catalysts, such as RhCl3 , [Rh(µ Cl)(1,5-cod)]2 , and [Rh(1,5-cod)2 ]+ (87). The bulky adduct (21) (R1 = R2 = C6 H5 ) was shown to undergo dehydrogenative coupling to form only a linear dimer or a mixture of the cyclic trimer and tetramer, depending upon the temperature used. However, primary phosphine–borane adducts, such as C6 H5 PH2 ·BH3 and iC4 H9 PH2 ·BH3 , were found to undergo catalytic dehydrogenative polymerization under similar conditions to yield soluble polyphosphinoboranes (22) (eq. 22). When the polymerization is carried out in solution, the resulting polymers are low in molecular weight (eg, M w ≈ 5600 for R = C6 H5 ) whereas the neat polymerization affords high molecular weight phosphorus–boron polymers (eg, M w ≈ 31,000 for (22) R = C6 H5 ). These polymers are air and moisture stable in the solid state, and detailed studies of the physical properties have yet to be conducted but the novel phosphorus–boron backbone allows for interesting possibilities such as low temperature flexibility, flame retardancy, and ceramic formation.

Vol. 3

INORGANIC POLYMERS

43

(22)

Polycarbophosphazenes Polycarbophosphazenes possess a backbone of phosphorus, nitrogen, and carbon atoms and can be regarded as derivatives of “classical polyphosphazenes” (1) in which every third phosphorus atom is replaced by carbon. The first examples of these materials were discovered in 1989 (88). Thermal ROP of a cyclic carbophosphazene was used to prepare the chlorinated polymeric species (23), which undergoes halogen replacement reactions with nucleophiles such as aryloxides and aniline to yield hydrolytically stable poly(aryloxycarbophosphazenes) (24) (M w = ca 105 , M n = 104 ) (eq. 23) (88–91). The polymer backbone in these materials was found to be less flexible than in classical polyphosphazenes. For example, the halogenated polymer (23) possesses a T g of −21◦ C compared to a value of −66◦ C for poly(dichlorophosphazene) (2).

(23) The reaction of (23) with alkylamines has also been studied (91). The resulting poly(alkylaminocarbophosphazenes) are sensitive to hydrolysis. However, arylamino derivatives are moisture stable and, in addition, a novel, regioselectively substituted polymer (25) was successfully prepared via the sequential reaction with NH(C6 H5 )2 and trifluoroethoxide anions (eq. 24).

(24)

44

INORGANIC POLYMERS

Vol. 3

Sulfur–Nitrogen–Phosphorus Polymers Sulfur–nitrogen–phosphorus polymers possess backbones that can be regarded as compositional hybrids of those present in sulfur–nitrogen polymers, such as the solid-state polymer poly(sulfur nitride) [SN]x or polyoxothiazenes [RS(O) N]n and classical polyphosphazenes, [R2 P N]n (1) (92). Poly(sulfur nitride), [SN]x , possesses remarkable properties such as electrical conductivity at room temperature and superconductivity below 0.3 K (93). [SN]x is insoluble and has a polymeric structure in the solid state with interchain S···S interactions. As these interactions are crucial to the properties of the material, [SN]x is best regarded as a solid-state polymer rather than a polymeric material with discrete macromolecular chains of the type discussed in this article. The first well-characterized examples of sulfur–nitrogen–phosphorus materials, polythiophosphazenes, were reported in 1990 (94). These polymers were prepared via the thermal ROP of a cyclothiophosphazene (eq. 25). This yielded the hydrolytically sensitive polythiophosphazene (26) with a backbone of three-coordinate sulfur(IV), nitrogen, and phosphorus atoms. Although reaction of (26) with nucleophiles such as aryloxides yielded materials (27) with improved hydrolytic stability, degradation was still rapid except where very bulky substituents such as o-phenylphenoxy were present.

(25) Although the backbone of polythiophosphazenes appears to be quite fragile, a particularly interesting feature of the substitution reactions of (26) is that the S Cl bond is much more reactive than the P Cl bonds. Regioselective substitution at the sulfur center is possible and yields macromolecules (29) with different aryloxy substituents at sulfur and phosphorus (eq. 26) (94,95).

(26) In 1991, another class of sulfur–nitrogen–phosphorus polymers, polythionylphosphazenes, were reported (96,97). These materials, which possess four-coordinate sulfur(VI) atoms in the backbone, possess improved stability and were prepared by a thermal ROP of cyclic thionylphosphazenes (30), with either chlorine or fluorine at the sulfur(VI) center, at 165–180◦ C (eq. 27) (98). An

Vol. 3

INORGANIC POLYMERS

45

ambient temperature synthesis involving initiation by Lewis Acids such as GaCl3 has been subsequently developed (99).

(27) The halogenated polythionylphosphazenes (31) that are formed in these ROP reactions (together with small quantities of macrocyclic byproducts) are quite sensitive to hydrolysis but a variety of moisture-stable derivatives have been prepared by reaction of this species with aryloxides or amines (100–102). Mixed substituent aryloxy–alkoxy polymers have also been prepared (92). Interestingly, with aryloxides, regioselective substitution at phosphorus is observed and in the resulting polymers (32) the sulfur(VI)-halogen bond remains intact. Remarkably, this regioselectivity is the exact reverse of that detected for the polythiophosphazenes described earlier where the sulfur(IV)-halogen bond is more reactive. In contrast to the reactions with aryloxides, reaction with primary or secondary amine nucleophiles leads to substitution at both the phosphorus and the sulfur(VI) centers to give poly(aminothionylphosphazenes) (33) (102,103).

Ab initio calculations on isotactic polythionylphosphazene (31) (X = Cl or F) indicate a localized electronic structure for the polymer backbone and predict that a cis–trans helical conformation is the most energetically favorable for isolated macromolecules (104). This is in contrast to the analogous “classical” polyphosphazene (1) where a trans-planar conformation is preferred. Studies of the properties of the polythionylphosphazenes also reveal significant differences in thermal transition behavior and polymer morphology compared to classical polyphosphazenes. For example, the polymer [NSOF{NP(OC6 H5 )2 }2 ]n is an amorphous elastomer (T g = −15◦ C), whereas the analogous classical polyphosphazene [NP(OC6 H5 )2 ]n is a microcrystalline thermoplastic (T m = 390◦ C, T g = −6◦ C). The T g s of the fluorinated polythionylphosphazenes are lower than those of the analogues with chlorine at sulfur. For example, for (31) (X = F) T g = −56◦ C whereas for (31) (X = Cl) T g = −46◦ C (101). Poly(aminothionylphosphazenes) possess high gas permeability and have found utility as matrices for phosphorescent dyes for oxygen sensing applications (105). Also noteworthy is an interesting condensation route reported which leads to polymers with backbones of alternating S(O) N and P N units (106).

46

INORGANIC POLYMERS

Vol. 3

Preliminary reports of the first polythiazylphosphazenes (34), which possess three-coordinate sulfur(III) atoms, have appeared (eq. 28) (107). These materials would represent true hybrids of poly(sulfur nitride) and polyphosphazenes and further developments in this area should prove to be particularly interesting.

(28)

Polyoxothiazenes Partially characterized polyoxothiazenes were briefly reported in the early 1960s (108). However, in 1992 the first well-characterized examples (35) with alkyl or aryl substituents at sulfur were described (109). These polymers, which possessed estimated molecular weights of M w = ca 105 and M n = ca 104 , were synthesized via the condensation polymerization of N-silylsulfonimidates at 120–170◦ C over 2–8 days (eq. 29). These reactions are catalyzed by added Lewis acids (eg, BF3 ·O(C2 H5 )2 ) and bases (eg, fluoride). By using a mixture of different sulfonimidates, random copolymers such as [CH3 S(O) N]m [C6 H5 S(O) N]n were also successfully prepared. Free sulfonimidates were also found to thermally condense to yield poly(organooxothiazenes) at lower temperatures than their N-silyl analogues (eq. 29) (109,110).

(29)

Polyoxothiazenes appear to be highly polar. For example, [CH3 S(O) N]n is soluble in DMF, DMSO, hot water, and concentrated H2 SO4 . Studies of the thermal transition behavior of these materials have indicated that they are amorphous, which is consistent with an atactic structure. Interestingly, the T g of [CH3 S(O) N]n is ca. 60◦ C, which is dramatically higher than for [(CH3 )2 P N]n (T g = −46◦ C). This suggests a much less flexible backbone for polyoxothiazenes compared to polyphosphazenes, as might be expected from

Vol. 3

INORGANIC POLYMERS

47

studies on polythionylphosphazenes (vide supra). Thermogravimetric analysis (TGA) showed that the polymers are stable to weight loss up to ca 270◦ C (at a heating rate of 10◦ C/min). Theoretical studies on [CH3 S(O) N]n have indicated that a cis–trans helical conformation is the most stable for this polymer (110).

Inorganic Polymers Based on Transition Elements Ferrocene-Based Polymers. The excellent thermal stability and interesting physical (eg, redox) properties associated with the ferrocene moiety have led to extensive efforts aimed at the incorporation of this unit into polymer structures. The inclusion of this moiety in the side-group structure of polymers has been very successful and requires only minor modifications of previously established synthetic methodologies. For example, poly(vinylferrocene) (36) can be prepared via the free radical addition polymerization of vinylferrocene (111). The incorporation of ferrocene moieties into the main chains of polymers where the metal atoms are separated by a considerable distance has also been achieved. The extensive organic chemistry of the metallocene nucleus has allowed for the preparation of well-defined difunctional ferrocenes that have been used in controlled polycondensation reactions to yield well-defined products of appreciable molecular weight. Examples of products derived from such reactions involve the poly(arylene–siloxane–ferrocenes) (eg, 37) (112), ferrocene-containing polyesters (113), and novel “accordian” type polymers (114,115).

The versatile chemistry of the ferrocenyl moiety has also allowed for the preparation of a large number of dendrimeric structures with this group at the periphery (116–118). A series of dendrimers based upon polypropylenimine cores have been reported (119,120) and star-shaped macromolecules with ferrocene units at the periphery have been produced (121), as well as amido-ferrocene dendrimers (eg, 38) (122). A convergent approach has allowed for the synthesis of dendrimers (39) for which the ferrocene units at the periphery display electronic interaction (123).

48

INORGANIC POLYMERS

Vol. 3

Vol. 3

INORGANIC POLYMERS

49

In contrast, very few examples of polymers with ferrocene units in close proximity, which could take advantage of interactions between the metal atoms, have been prepared. The development of materials with interesting physical (eg, electronic and/or magnetic) properties might be anticipated based on the properties of molecular species in which two ferrocene units are linked close together. Thus, in such systems the iron atoms can interact and in some cases yield delocalized, mixed valent species upon one-electron oxidation, even when the metal atoms are up to 0.7 nm apart (124,125). Work in this field has largely yielded low molecular weight (M n  10,000) and often poorly defined materials (126). For the synthesis of well-characterized polymers (M n < ca 11,000) with main chains of ferrocene groups and vinylene, divinylene, or oligovinylene units see Reference 127.

Polyferrocenylenes A typical early route to polyferrocenylenes (40) with M n < 5000 involved polycondensation processes such as the recombination of ferrocene radicals generated via the thermolysis of ferrocene in the presence of peroxides. However, these materials have been found to possess other fragments such as CH2 and O in the main chain (128,129). More structurally well-defined polyferrocenylenes (40) (M n < 4000) have been prepared (130) via the condensation reaction of 1,1-dilithioferrocene·TMEDA (tetramethylethylenediamine) with 1,1-diiodoferrocene and, significantly, the reaction of 1,1 -dihaloferrocenes with magnesium (eq. 30) has been shown to afford low molecular weight (M n = 4600 for soluble fractions) materials with appreciable crystallinity (131). In the latter case, oxidation with 7,7,8,8-tetracyanoquinodimethane (TCNQ) afforded doped polymers that were delocalized on the M¨ossbauer time scale (ca 10 − 7 s) at room temperature and which possessed electrical conductivities of up to 10 − 2 S/cm.

(30)

Polyferrocenylsilanes Properties. Polyferrocenylsilanes possess a backbone of alternating ferrocene and organosilane units. Since the early 1990s when the first high molecular weight, well-characterized examples were prepared by a ROP approach, considerable effort has been directed towards understanding the properties of polyferrocenylsilane materials, the vast majority of which are soluble in common organic solvents (132,133). It was noted early on that electrochemistry of the high polymers such as (41) (R = R = CH3 ) possess two reversible oxidation waves in a 1:1 ratio (132,134). This provided clear evidence for the existence of interactions between the iron atoms, and led to the proposal that initial oxidation

50

INORGANIC POLYMERS

Vol. 3

occurred at alternating iron sites along the main chain. Work on model oligomers with between two and nine ferrocene units has provided clear evidence in support of this postulate (135). Similar electrochemical behavior has subsequently been detected for a range of other polyferrocenylsilanes (134). Oxidative doping of poly(ferrocenyldimethylsilane) with I2 has been shown to yield semiconducting materials (σ = ca 10 − 4 S/cm) whereas the pristine materials are insulating (σ = ca 10 − 14 S/cm) (133). A report indicates that several tetracyanoethylene (TCNE)-oxidized low molecular weight polyferrocenylsilanes (M w = ca 1500) show electron delocalization on the M¨ossbauer time scale (ca 10 − 7 s) and also ferromagnetic ordering at low temperatures (136). Studies of high molecular weight M w > 105 analogues have not reproduced this behavior (137). Thin films of the homopolymers and block copolymers with organic or inorganic co-blocks are attracting attention for numerous applications, such as chemomechanical sensors, electrochromic materials, electrode mediators, variable refractive index materials, hole transport layers, charge dissipation coatings for dielectrics, lithographic resists, and photonic band-gap materials (133,138–144).

The polymers also exhibit interesting morphology and several of the symmetrically substituted derivatives will crystallize. For example, the dimethyl derivative (41) (R = R = CH3 ) is an amber, film-forming thermoplastic and possesses a T m at 120–145◦ C depending on crystallite size and a T g at 33◦ C whereas, in contrast, the n-hexyl analogue (41) (R = R = n-hexyl) is an amber, gummy amorphous material with a T g of −26◦ C (133). The packing in the crystalline regions of polymer (41) (R = R = CH3 ) has been shown to be analogous to that of a linear pentamer (135,141,142). In addition, as the iron atom in ferrocene acts as a “molecular ball-bearing,” this gives these polymers a large degree of conformational flexibility and consequently T g s are lower than might be expected for polymers with such a bulky unit in the main chain (133) (Table 4). Several polyferrocenylsilanes can be fabricated in the melt (eg, R = R = CH3 above 150◦ C) (Fig. 1). Polyferrocenylsilanes have been found to exhibit excellent thermal stability to weight loss (up to 350–400◦ C) and have been shown to yield interesting composites containing Fe nanoparticles at 500–1000◦ C (145– 147). Controlled cross-linking of the polyferrocenylsilanes can be used to make magnetic ceramic films and monoliths with the same shape as the polymer precursor as well as solvent-swellable, redox-active gels (148,149). The solution properties of polyferrocenylsilanes such as the dimethyl derivative (41) (R = R = CH3 ) have been well-characterized by light-scattering experiments and viscometry and Mark–Houwink parameters have been established (150).

Vol. 3

INORGANIC POLYMERS

51

Table 4. Thermal Transition and gpc Molecular Weight Data for Selected Polyferrocenylsilanesa R/OR H CH3 C2 H5 n C3 H7 n C4 H9 n C5 H11 n C6 H13 CH3 CH3 CH3 CH3 CH3 CH3 CH3 OCH3 OC2 H5 OCH2 CF3 On C4 H9 On C6 H13 On C18 H37 OC6 H5 (OCH2 CH2 )OCH3 CH3

R /OR

T g (T m )b , ◦ C

Mn c

PDId

H 16 (165) –e –e 5 CH3 33 (122–143) 3.4 × 10 1.5 C2 H5 22 (108) 4.8 × 105 1.6 n C3 H7 24 (98) 8.5 × 104 2.7 n C4 H9 3 (116,129) 3.4 × 105 2.6 n C5 H11 −11 (80–105) 3.0 × 105 1.6 n C6 H13 −26 7.6 × 104 1.5 H 9 (87, 102) 4.2 × 105 2.0 CH2 CH2 CF3 59 8.1 × 105 3.3 CH CH2 28 7.7 × 104 2.1 n C18 H37 1 (16) 5.6 × 105 2.5 C6 H5 90 1.5 × 105 2.0 Fc f 99 7.1 × 104 2.3 CH2 CH2 CH2 Cl 27 2.7 × 105 1.5 OCH3 19 1.5 × 105 1.9 OC2 H5 0 3.8 × 105 2.1 OCH2 CF3 16 2.2 × 105 1.2 On C4 H9 −43 3.9 × 105 2.1 On C6 H13 −51 0.9 × 105 2.4 On C18 H37 (32) 2.3 × 105 2.2 OC6 H5 54 2.3 × 105 2.0 (OCH2 CH2 )OCH3 −53 1.0 × 105 2.6 (OCH2 CH2 )x OCH3 g −69, −72h 5.6 × 104 , 1.9 × 105 2.3, 2.2

a Refs.

133 and 143. data collected at a heating rate of 10◦ C min − 1 . c GPC data and molecular weight values are relative to polystyrene standards. Although in this case gpc provides only molecular weight estimates, absolute determinations of M w by static light scattering for several polymers have indicated that gpc underestimates the real values by a factor of 2 (144). d PDI = M /M . w n e Insoluble polymer. f Fc = (η-C H )Fe(η-C H ). 5 4 5 5 g x ∼ 8. hTwo different molecular weight samples. b DSC

Water-soluble polyferrocenylsilanes have also been prepared and these possess, for example, oligoethoxy or ionic side chains (151,152). These materials can be used in layer-by-layer assembly processes to form superlattices with a range of potential applications (153). Novel random copolymers (42) with oligosilane spacers have also been prepared by using a thermal copolymerization process (154,155). These polymers possess interesting photophysical and charge transport properties. Indeed, the skeletons of the polysilane segments can be selectively cleaved using uv light because of the photosensitive nature of the Si-Si bond. Block copolymers containing polyferrocenylsilane blocks (vide infra) have demonstrated interesting self-assembly behavior and are of interest for nanostructure applications (156). For example, cylindrical worm-like micelles with a polyferrocenylsilane core

52

INORGANIC POLYMERS

Vol. 3

Fig. 1. Samples of melt processed polyferrocenylsilanes.

and a polysiloxane corona can be fabricated and are of potential use as semiconducting nanowires and as etching resists in nanolithography (157).

Synthesis. Ring-Opening Polymerization. Early attempts to prepare macromolecules in which the ferrocene units are separated via an organosilane spacer group focused on the use of polycondensation reactions. Partially characterized, impure polyferrocenylsilanes were prepared via the reaction of dilithioferrocene with organodichlorosilanes. The molecular weights of 1400–7000 reported for these materials are characteristic of polycondensation processes where exact reaction stoichiometries are virtually impossible to achieve because one reactant, in this case dilithioferrocene, cannot be readily prepared in pure form (116). In 1992, the first synthesis of high molecular weight polyferrocenylsilanes (41) (M w = 105 –106 , M n > 105 ) via a thermal ROP route was reported (132,158). This process involved heating silicon-bridged [1]ferrocenophanes (43) in the melt at 130–220◦ C (eq. 31). The presence of a single-atom bridging the ferrocene unit in the monomer leads to a strained structure in which the planes of the cyclopentadienyl rings are tilted with respect to one another by an angle of ca 21◦ . In contrast, in ferrocene the cyclopentadienyl rings are parallel. The presence of strain in the ferrocenophane, which has been measured to be ca 80 kJ/mol for (43) (R = R = CH3 ), is believed to provide the driving force for the ROP process (132,158).

Vol. 3

INORGANIC POLYMERS

53

(31) Since this initial discovery, a wide range of silicon-bridged [1]ferrocenophanes with either symmetrically or unsymmetrically substituted silicon atoms have been prepared and similarly polymerized (Table 4). Polymerization of a mixture of different silicon-bridged [1]ferrocenophanes has also been shown to yield random copolymers (159). Silicon-bridged [1]ferrocenophanes undergo living anionic ROP using initiators such as n-C4 H9 Li in THF (160,161). This has permitted the synthesis of polyferrocenylsilanes with controlled molecular weights and narrow polydispersities and has also allowed the preparation of the first block copolymers containing skeletal transition-metal atoms (160,161). Block copolymers such as (44) have been prepared with other monomers that undergo anionic polymerization such as cyclic siloxanes (see eq. 32) or organic monomers such as polystyrene and isoprene (161,162). The resulting block copolymers undergo phase separation in the solid state and, in solution, micellar aggregates are formed (156,163).

(32) Also, the transition-metal-catalyzed ROP of silicon-bridged [1]ferrocenophanes in the presence of various transition-metal complexes (eg, PtII , Pt0 , RhI , PdII ) has been developed (164). This route, which takes place in solution at room temperature, is much milder than, and doesn’t have the same stringent monomer purity requirements as, anionic ROP. Furthermore, molecular weight control is possible through the use of Si H containing capping agents such as (C2 H5 )3 SiH, and access to block and graft copolymers and star polymers is possible (164,165). Monomer Synthesis. Sila[1]ferrocenophane monomers such as (43) are readily available on a substantial laboratory scale (>100 g) from the reaction of dilithioferrocene tetramethylethylenediamine (fcLi2 ·TMEDA) with the appropriate dichloroorganosilane (166). Spirocyclic sila[1]ferrocenophanes such as (45) and (46) are also easily synthesized (167). Sila[1]ferrocenophane monomers with alkoxy, aryloxy, and amino substituents at silicon are readily accessible through reaction of dichlorosila[1]ferrocenophane with the appropriate alcohol, phenol or amine in the presence of base (168).

54

INORGANIC POLYMERS

Vol. 3

Other Polyferrocenes The ROP route has been extended to the synthesis of other polymers from [1]ferrocenophane precursors. Polyferrocenylgermanes (47) were first reported in 1993 and have been well-characterized and possess quite similar thermal transition behavior, morphology, and electrochemical behavior to the analogous polyferrocenylsilanes (169). Poly(ferrocenylsilane-ferrocenylgermane) random copolymers (48) have also been prepared via the thermal polymerization of mixtures of the respective monomers (169).

Polyferrocenylphosphines (49) (and the corresponding phosphine sulfides) are also accessible via the thermal ROP of phosphorus-bridged [1]ferrocenophanes (170). Polymers of this type have been previously prepared by condensation routes and the catalytic potential of some of their transition-metal derivatives has already been noted. In addition, the first sulfur-bridged [1]ferrocenophanes have been prepared and polymerized to give polyferrocenylsulfides, (eg, 50) (171).

Hydrocarbon-bridged [2]-ferrocenophanes (51) possess strained ring-tilted structures (tilt-angles = ca 21◦ ) and these species have been found to yield polyferrocenylethylenes (52) via ROP at 250–300◦ C (eq. 33) (172). As a consequence of the presence of a more insulating bridge, these polymers show much smaller interactions between the iron atoms compared to polyferrocenylsilanes.

Vol. 3

INORGANIC POLYMERS

55

(33) Analogous hydrocarbon-bridged [2]ruthenocenophanes (tilt angles = ca 29–30◦ ) undergo thermal ROP to yield poly(ruthenocenylethylenes) (172). These materials exhibit significantly different electrochemistry from their iron analogues. In early 1992, it was reported that [3]trithiaferrocenophanes, which are essentially unstrained, function as precursors to poly(ferrocenylene persulfides) via a novel atom abstraction polymerization route (eq. 34) (173). Thus, reaction of [3]-trithiaferrocenophanes (53) with P(C4 H9 )3 led to the formation of the phosphine sulfide S P(C4 H9 )3 and the polymers (54).

(34) The presence of a butyl substituent on the cyclopentadienyl ring is necessary for the polymer to be soluble. The molecular weight (M w ) of (54) (R = n-C4 H9 ) was determined to be 40,000 by gel permeation chromatography (gpc). The [3]ferrocenophanes (53) (R = H) and (53) (R = n-C4 H9 ) can be copolymerized to give soluble copolymers with M w = 25,000. Poly(ferrocenylene persulfides) possess a range of novel properties (173– 175). They are photosensitive and the S S bonds can be reversibly reductively cleaved with Li[B(C2 H5 )3 H] and then regenerated upon oxidation with I2 . Their electrochemical behavior is similar to that detected for polyferrocenylsilanes except that the interaction between the iron sites appears to be even greater. The atom abstraction route using P(t C4 H9 )3 as a desulfurization agent has also been extended to the preparation of other poly(ferrocenylene persulfides) with t-butyl substituents and also high molecular weight (M w = 50,000–1,000,000) network polymers by the use of [3]ferrocenophanes with two trisulfido bridges as monomers (174).

Face-to-Face Metallocene Polymers The development of rigid-rod metallocene polymers with a multistacked structure using condensation routes has been reported (176–180). This involved treatment of the ferrocene monomers (55) with FeCl2 and Na[N(Si(CH3 )3 )2 )] and this yielded purple polymers (56) with molecular weights up to M n = 18,000, although higher

56

INORGANIC POLYMERS

Vol. 3

molecular components were also present (eq. 35). Mixed metal copolymers containing Ni and Fe whose soluble fractions were of low molecular weight (M n < 3000) were also reported by using Ni(acac)2 instead of FeCl2 (176). The electrical and magnetic properties of these novel polymers and copolymers are clearly worthy of investigation. Interestingly, structural work on well-defined oligomers suggests that the stacked metallocene units in the polymer form a helical structure (180).

(35)

Coordination Polymers In the past, attempts to prepare coordination polymers have been hindered by the insolubility and consequent intractability of the products. These problems arise from the inherent skeletal rigidity of these materials, and the introduction of solubilizing or flexibilizing groups either in the polymer backbone or side-group structure is necessary for useful products to be obtained. Such modification has yielded a range of interesting and well-characterized materials with intriguing properties. For example, novel liquid crystalline polymers (57) containing paramagnetic CuII centers have been prepared (181) and soluble, luminescent silver-containing polymers (58) have been reported (182).

In addition, well-characterized lanthanide containing polymers (59), which possess polyelectrolyte behavior and exhibit interesting photophysical properties, have been reported (183). The tetradentate Schiff-Base ligands greatly stabilize the lanthanide ions in solution and allow for efficient energy transfer to the

Vol. 3

INORGANIC POLYMERS

57

lanthanide centers. Several of the polymers are soluble in polar organic solvents such as DMSO, and molecular weights (M n ) up to 1.8 × 104 have been established. The polymer (59) (Ln = Eu or Eu/Y) is an interesting candidate for luminescent and lasing applications (184). Similar polymers containing cerium and zirconium in the main chain have also been prepared and these possess M n values up to 3 × 104 (185). Films of these polymers have conductivities of ∼ 10 − 7 S/cm which increase to ∼10 − 3 S/cm upon I2 doping (186).

Phthalocyanine-based polymers, such as the “shish-kebab” polymers (60), are also of considerable interest and significant electrical conductivities of up to ca 0.1 S/cm have been detected for chemically or electrochemically doped materials (187,188). If flexible organic substituents are present on the periphery of the phthalocyanine ring, these materials can also be soluble (at least low molecular weight fractions).

Well-defined and readily soluble ruthenium coordination polymers (61) have been synthesized through the reaction of a bisbidentate ligand and a metal center that already possesses one bidentate ligand (189–191). In these complexes, the random stereospecificity at the ruthenium centers results in a ribbon-like conformation of the polymers with the extension of the chain dependent upon the substituent R. The RuII centers appear to behave independently and the polymers appear to be stable against heat and uv irradiation. They have molecular weights on the order of M w = 40,000–50,000.

58

INORGANIC POLYMERS

Vol. 3

A similar strategy applied to the reaction of a bistridentate monomer with an appropriate ruthenium compound gives rise to the rod-like polymers (62), which are soluble due to the n-hexyl substituents (192,193). These polymers display a pronounced polyelectrolytic effect in salt-free dimethylacetamide.

Polythiophene-metal complex hybrid polymers such as (63) and (64) have been explored (194,195). The polythiophene-cobalt salen hybrid (63) participates in the electrocatalytic reduction of oxygen and is highly conducting in nature. Polymers such as (64) have also been shown to be conducting. A variety of related structures have been prepared and similar strategies have also resulted in the preparation of polymeric metallorotaxanes (196).

Vol. 3

INORGANIC POLYMERS

59

The generation of dendrimeric coordination polymers has also been an area of significant activity. Imaginative routes to novel Ru- or Pt-polypyridyl systems (eg, 65) (197) have been reported and many other ligand systems have been exploited (117,118,198–201). These materials are of interest with respect to their photophysical and electrochemical properties and possibly their catalytic behavior.

60

INORGANIC POLYMERS

Vol. 3

Rigid-Rod Organometallic Polymers Macromolecules with backbones that possess conjugated C C units and transition-metal atoms, termed polymetallaynes, represent some of the best-characterized examples of transition-metal-based polymers prepared to date (202). The first polymetallaynes contained nickel, palladium, or platinum atoms in the main chain and were isolated in 1977 as yellow, film forming materials (203–205). These were prepared by efficient copper halide-catalyzed coupling processes (eg, eq. 36) and possessed estimated molecular weights (M w ) from 13,000 to 120,000.

(36) There has been an important expansion in this area which has yielded a range of new rigid-rod materials via the creative use of a variety of new and well-defined polycondensation strategies. For example, a new route to the polyplatinynes (66) that involved the reaction of trans-PtCl2 (PR3 )2 complexes with bis(trimethylstannyl)diynes (eq. 37) has been reported (206). These rigid-rod polymers possessed estimated weight-average molecular weights up to ca 100,000 according to gpc measurements.

(37) This synthetic procedure can be extended to allow the incorporation of other transition elements into the polymer main chain such as iron (to give 67) by using FeCl2 [(C2 H5 )2 PCH2 CH2 P(C2 H5 )2 ]2 as the transition-metal-containing reactant (207). In addition, condensation routes to organonickel polymers (68), have been devised (208,209) and interesting organocobalt (210) and organozirconium (211) polymers containing metallacyclopentadiene moieties in the main chain have been reported.

Vol. 3

INORGANIC POLYMERS

61

In another interesting synthetic development, it was shown that a range of rhodium-containing polymetallynes (69) are accessible via the reaction of the unsubstituted diynes with Rh(PR3 )4 CH3 in a reaction that involves reductive elimination of methane and the loss of a phosphine ligand (eq. 38) (212). In the case where trimethylphosphine ligands are attached to rhodium, the polymers are insoluble but the tri(n-butyl)phosphine analogues are soluble and yield free-standing solvent-cast films from THF.

(38) Over the past 15 years the physical properties of polymetallaynes have received continued attention because of their novel rigid-rod structures and their conjugated backbones (213–222). Thus, polyplatinynes form ordered, liquid crystalline mesophases in solvents in which they are soluble such as trichloroethylene (213), and these materials also possess novel, third-order nonlinear optical properties (214) that are of interest for electrooptic device applications. Optical ¨ absorption and photoluminescence spectroscopic studies and extended Huckel calculations have shown that polymetallaynes possess a delocalized polymer backbone whose electronic structure is modified by the nature of the transition metal, coligands, and the unsaturated hydrocarbon spacer (215–217). For example, optical band gaps for a series of polyplatinynes with platinum centers joined by σ -conjugated acetylide-arene linkages of varying length have been measured to be in the range of 2.5–3.1 eV, which is lower than for model complexes and is consistent with conjugation through the metal centers (216). Other developments in the area of rigid-rod transition-metal-based polymers include the synthesis of a range of thermotropic liquid crystalline organocobalt polymers (eg, 70) in which the metal is bound to skeletal cyclobutadiene units (eq. 39) (223).

(39) In addition, the preparation and characterization of novel lyotropic liquid crystalline aramides (71) with complexed chromium(tricarbonyl) units have been

62

INORGANIC POLYMERS

Vol. 3

reported (224). These materials are soluble in organic solvents, which leads to potential processing advantages for the uncomplexed organic polymer as the Cr(CO)3 groups can be easily added and removed and poly(p-phenyleneterephthalamide) is only soluble in concentrated sulfuric acid (224,225). Work on the coordination of transition-metal fragments to π -hydrocarbon units can be found in Reference 226.

BIBLIOGRAPHY “Inorganic Polymers” in EPST 1st ed., Vol. 8, pp. 664–691, by D. L. Venezky, U.S. Naval Research Laboratory; “Inorganic Polymers” in EPSE 2nd ed., Vol. 8, pp. 138–147, by A. Rheingold, University of Delaware. 1. J. E. Mark, H. R. Allcock, and R. West, Inorganic Polymers, Prentice Hall, Englewood Cliffs, N.J., 1992; I. Manners, Angew. Chem. Int. Ed. Engl. 35, 1603 (1996); R. D. Archer, Inorganic and Organometallic Polymers, Wiley-VCH, New York, 2001. 2. R. H. Neilson and co-workers, ACS Symp. Ser. 572, 232 (1994), and references therein. 3. P. Wisian-Neilson, ACS Symp. Ser. 572, 246 (1994). 4. R. de Jaeger and M. Gleria, Prog. Polym. Sci. 23, 179 (1998), and references therein. 5. H. R. Allcock and R. L. Kugel, J. Am. Chem. Soc. 87, 4216 (1965). 6. H. R. Allcock, J. Inorg. Organomet. Polym. 2, 197 (1992). 7. H. R. Allcock, Chem. Mater. 6, 1476 (1994). 8. H. R. Allcock, Adv. Mater. 6, 106 (1994). 9. H. R. Allcock and C. Kim, Macromolecules 22, 2596 (1989). 10. H. R. Allcock and C. Kim, Macromolecules 24, 2846 (1991). 11. H. R. Allcock, A. A. Dembek, and E. H. Klingenberg, Macromolecules 24, 5208 (1991). 12. H. R. Allcock and C. G. Cameron, Macromolecules 27, 3131 (1994). 13. G. Facchin and co-workers, J. Inorg. Organomet. Polym. 1, 389 (1991). 14. D. C. Ngo, J. S. Rutt, and H. R. Allcock, J. Am. Chem. Soc. 113, 5075 (1991). 15. H. R. Allcock and co-workers, Macromolecules 27, 7556 (1994). 16. R. H. Neilson and P. Wisian-Neilson, Chem. Rev. 88, 541 (1988); C. H. Walker, J. V. St. John, and P. Wisian-Neilson, J. Am. Chem. Soc. 123, 3846 (2001). 17. R. A. Montague and K. Matyjaszewski, J. Am. Chem. Soc. 112, 6721 (1990). 18. K. Matyjaszewski, J. Inorg. Organomet. Polym. 2, 5 (1992); K. Matyjaszewski and co-workers, J. Polym. Sci., Part A: Polym. Chem. 30, 813 (1992); K. Matyjaszewski and co-workers, Polymer 35, 5005 (1994); U. Franz, O. Nuyken, and K. Matyjaszewski, Macromolecules 26, 3723 (1993). 19. G. D’Halluin and co-workers, Macromolecules 25, 1254 (1992). 20. C. H. Honeyman and co-workers, J. Am. Chem. Soc. 117, 7035 (1995). 21. H. R. Allcock and co-workers, Macromolecules 29, 7740 (1996). 22. J. M. Nelson and H. R. Allcock, Macromolecules 30, 1854 (1997). 23. H. R. Allcock and co-workers, Macromolecules 30, 2213 (1997). 24. H. R. Allcock and co-workers, Macromolecules 33, 3999 (2000).

Vol. 3 25. 26. 27. 28. 29. 30.

31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69.

INORGANIC POLYMERS

63

H. R. Allcock and co-workers, Inorg. Chem. 38, 280 (1999). E. Niecke and W. Bitter, Inorg. Nucl. Chem. Lett. 9, 127 (1973). R. West, J. Organomet. Chem. 300, 327 (1986). R. D. Miller and J. Michl, Chem. Rev. 89, 1359 (1989). K. Matyjaszewski, J. Inorg. Organomet. Polym. 1, 463 (1991). H. Teramae and K. Takeda, J. Am. Chem. Soc. 111, 1281 (1989); J. T. Nelson and W. J. Pietro, J. Phys. Chem. 92, 1365 (1988); A. Savin and co-workers, Angew. Chem. Int. Ed. Engl. 31, 187 (1992). P. Trefonas, Encycl. Polym. Sci. Eng. 13, 162 (1987). H. Tachibana and co-workers, Phys, Rev. B 47, 4363 (1993). K. S. Madiyasni, R. West, and L. D. David, J. Am. Ceram. Soc. 61, 504 (1978). C. Aitken, J. F. Harrod, and E. Samuel, J. Organomet. Chem. 279, C11 (1985); C. T. Aitken, J. F. Harrod, and E. Samuel, J. Am. Chem. Soc. 108, 4059 (1986). T. D. Tilley, Acc. Chem. Res. 26, 22 (1993), and references therein. M. Scarlete and co-workers, Chem. Mater. 6, 977 (1994). Y.-L. Hsiao and R. M. Waymouth, J. Am. Chem. Soc. 116, 9779 (1994). M. Cypryk, Y. Gupta, and K. Matyjaszewski, J. Am. Chem. Soc. 113, 1046 (1991); E. Fossum and K. Matyjaszewski, Macromolecules 28, 1618 (1995). K. Sakamoto and co-workers, J. Am. Chem. Soc. 111, 7641 (1989). H. Sakurai and co-workers, ACS Symp. Ser. 572, 8 (1994). T. Sanji, F. Kitayama, and H. Sakurai, Macromolecules 32, 5718 (1999). H. Suzuki and co-workers, Adv. Mater. 5, 743 (1993). Y. Wang, R. West, and C. H. Yuan, J. Am. Chem. Soc. 115, 3844 (1993). P. Trefonas and R. West, J. Polym. Sci., Polym. Chem. 23, 2099 (1985). R. D. Miller and R. Sooriyakumaran, J. Polym. Sci., Part A: Polym. Chem. 25, 111 (1987). S. M. Katz, J. A. Reichl, and D. H. Berry, J. Am. Chem. Soc. 120, 9844 (1998). M. Okano and co-workers, Electrochim. Acta 44, 659 (1998). V. Y. Lu and T. D. Tilley, Macromolecules 33, 2403 (2000). T. Hayashi and co-workers, Chem. Lett. 647 (1992). T. Kodaira and co-workers, Adv. Mater. 7, 917 (1995); J.-C. Baumert and co-workers, Appl. Phys. Lett. 53, 1147 (1988). T. Imori and T. D. Tilley, Chem. Commun. 1607 (1993). T. Imori and co-workers, J. Am. Chem. Soc. 117, 9931 (1995). K. Mochida and H. Chiba, J. Organomet. Chem. 473, 45 (1994). C. Aitken and co-workers, J. Organomet. Chem. 349, 285 (1988). M. Okano, T. Toriumi, and H. Hamano, Electrochim. Acta 44, 3475 (1999). J. A. Reichl and co-workers, J. Am. Chem. Soc. 118, 9430 (1996). V. Lu and T. D. Tilley, Macromolecules 29, 5763 (1996). J. R. Babcock and L. R. Sita, J. Am. Chem. Soc. 118, 12481 (1996). K. Mochida and co-workers, J. Organomet. Chem. 542, 75 (1997). H. Ban, K. Deguchi, and A. Tanaka, J. Appl. Polym. Sci. 37, 1589 (1989). M. Abkowitz and M. Stolka, Solid State Commun. 78, 269 (1991). M. Abkowitz and M. Stolka, J. Non-Cryst. Solids 114, 342 (1989). K. Mochida and co-workers, Appl. Organomet. Chem. 11, 949 (1997). Y. Chujo, N. Takizawa, and T. Sakurai, Chem. Commun. 227 (1994). Y. Chujo, ACS Symp. Ser. 572, 398 (1994). R. T. Paine and C. K. Narula, Chem. Rev. 90, 73 (1990); R. T. Paine, J. Inorg. Organomet. Polym. 2, 183 (1992). R. T. Paine and L. G. Sneddon, ACS Symp. Ser. 572, 358 (1994). P. J. Fazen and co-workers, Chem. Mater. 2, 96 (1990). P. J. Fazen and co-workers, Chem. Mater. 7, 1942 (1995).

64 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107.

108. 109. 110. 111. 112. 113. 114. 115. 116. 117.

INORGANIC POLYMERS

Vol. 3

Y. Chujo and co-workers, Macromolecules 24, 345 (1991). Y. Chujo and co-workers, Macromolecules 25, 33 (1992). Y. Chujo, I. Tomita, and T. Saegusa, Macromolecules 23, 687 (1990). N. Matsumi, K. Naka, and Y. Chujo, J. Am. Chem. Soc. 120, 5112 (1998). N. Matsumi, M. Miyata, and Y. Chujo, Macromolecules 32, 4467 (1999). Y. Sasaki and Y. Chujo, Polym. Prepr. Jpn. 45, 214 (1996). N. Matsumi and Y. Chujo, Macromolecules 31, 3802 (1998). Y. Chujo and co-workers, Macromolecules 25, 27 (1992). Y. Chujo, I. Tomita, and T. Saegusa, Macromolecules 27, 6714 (1994). N. Matsumi, K. Naka, and Y. Chujo, Macromolecules 31, 8047 (1998). N. Matsumi, T. Umeyama, and Y. Chujo, Macromolecules 33, 3956 (2000). N. Matsumi and co-workers, Macromolecules 31, 3155 (1998). N. Matsumi, K. Kotera, and Y. Chujo, Macromolecules 33, 2801 (2000). N. Matsumi, K. Naka, and Y. Chujo, J. Am. Chem. Soc. 120, 10776 (1998). H. Dorn and co-workers, Angew. Chem. Int. Ed. 38, 3321 (1999). R. I. Wagner and F. F. Caserio Jr., J. Inorg. Nucl. Chem. 11, 259 (1959). A. B. Burg, J. Inorg. Nucl. Chem. 11, 258 (1959). H. Dorn and co-workers, J. Am. Chem. Soc. 122, 6669 (2000). I. Manners and co-workers, J. Am. Chem. Soc. 111, 5478 (1989). H. R. Allcock and co-workers, Macromolecules 24, 2024 (1991). H. R. Allcock and co-workers, Inorg. Chem. 32, 5088 (1993). H. R. Allcock, S. M. Coley, and C. T. Morrissey, Macromolecules 27, 2904 (1994). I. Manners, Coord. Chem. Rev. 137, 109 (1994). M. M. Labes, P. Love, and L. F. Nichols, Chem. Rev. 79, 1 (1979). J. A. Dodge and co-workers, J. Am. Chem. Soc. 112, 1268 (1990). H. R. Allcock, J. A. Dodge, and I. Manners, Macromolecules 26, 11 (1993). M. Liang and I. Manners, J. Am. Chem. Soc. 113, 4044 (1991). M. Liang and I. Manners, Makromol. Chem., Rapid Commun. 12, 613 (1991). J. C. van de Grampel, Rev. Inorg. Chem. 3, 1 (1981); J. C. van de Grampel, Coord. Chem. Rev. 112, 247 (1992). A. R. McWilliams and co-workers, J. Am. Chem. Soc. 122, 8848 (2000). Y. Ni and co-workers, Angew. Chem. Int. Ed. Engl. 34, 998 (1995). Y. Ni and co-workers, Macromolecules 25, 7119 (1992). Y. Ni and co-workers, Macromolecules 29, 3401 (1996). M. N. Nobis and co-workers, Macromolecules 33, 7707 (2000). R. Jaeger and co-workers, Macromolecules 28, 539 (1995). Z. Pang and co-workers, Adv. Mater. 8, 768 (1996). V. Chunechom and co-workers, Angew. Chem. Int. Ed. 37, 1928 (1998). T. Chivers and co-workers, in R. Steudel, ed., Studies in Inorganic Chemistry, Vol. 14: The Chemistry of Inorganic Ring Systems, Elsevier, Amsterdam, Netherlands, 1992, p. 271. F. Seel and G. Simon, Angew. Chem. 72, 709 (1960); G. W. Parshall, R. Cramer, and R. E. Foster, Inorg. Chem. 1, 677 (1962). A. K. Roy, J. Am. Chem. Soc. 114, 1530 (1992). A. K. Roy and co-workers, J. Am. Chem. Soc. 115, 2604 (1993). C. U. Pittman Jr. and co-workers, Macromolecules 3, 746 (1970). W. J. Patterson, S. P. McManus, and C. U. Pittman Jr., J. Polym. Sci. Part A-1 12, 837 (1974). K. Gonsalves, L. Zhan-ru, and M. D. Rausch, J. Am. Chem. Soc. 106, 3862 (1984). M. E. Wright, and M. S. Sigman, Macromolecules 25, 6055 (1992). M. E. Wright and co-workers, Macromolecules 27, 3016 (1994). P. Nguyen, P. G´omez-Elipe, and I. Manners, Chem. Rev. 99, 1515 (1999). M. A. Hearshaw and J. R. Moss, Chem. Commun. 1 (1999).

Vol. 3 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 165. 166. 167.

INORGANIC POLYMERS

65

C. Gorman, Adv. Mater. 10, 295 (1998). B. Alonso and co-workers, Chem. Commun. 2575 (1994). I. Cuadrado and co-workers, Organometallics 15, 5278 (1996). J.-L. Fillaut, J. Linares, and D. Astruc, Angew. Chem. Int. Ed. Engl. 33, 2460 (1994). C. Val´erio and co-workers, J. Am. Chem. Soc. 119, 2588 (1997). I. Cuadrado and co-workers, J. Am. Chem. Soc. 119, 7613 (1997). U. T. Mueller-Westerhoff, Angew. Chem. Int. Ed. Engl. 25, 702 (1986). W. H. Morrison Jr. and D. N. Hendrickson, Inorg. Chem. 14, 2331 (1975); J. A. Kramer and D. N. Hendrickson, Inorg. Chem. 19, 3330 (1980). E. W. Neuse and H. Rosenberg, J. Macromol. Sci., C: Rev. Macromol. Chem. 4, 1 (1970). R. Bayer, T. P¨ohlmann, and O. Nuyken, Makromol. Chem., Rapid Commun. 14, 359 (1993); C. E. Stanton and co-workers, Macromolecules 28, 8713 (1995). C. U. Pittman Jr. and Y. Sasaki, Chem. Lett. 383 (1975). N. Bilow, A. L. Landis, and H. Rosenberg, J. Polym. Sci. Part A-1 7, 2719 (1969). E. W. Neuse and L. Bednarik, Macromolecules 12, 187 (1979). T. Yamamoto and co-workers, Inorg. Chim. Acta 73, 75 (1983). D. A. Foucher, B.-Z. Tang, and I. Manners, J. Am. Chem. Soc. 114, 6246 (1992). I. Manners, Chem. Commun. 857 (1999). D. A. Foucher and co-workers, Angew. Chem. Int. Ed. Engl. 32, 1709 (1993). R. Rulkens and co-workers, J. Am. Chem. Soc. 118, 12683 (1996). M. Hmyene and co-workers, Adv. Mater. 6, 564 (1994). J. K. Pudelski and co-workers, Macromolecules 29, 1894 (1996). L. I. Espada and co-workers, J. Inorg. Organomet. Polym. 10, 169 (2000). R. Resendes and co-workers, Adv. Mater. 12, 327 (2000). I. Manners, Pure Appl. Chem. 71, 1471 (1999). J. Rasburn and co-workers, Chem. Mater. 7, 871 (1995); S. Barlow and co-workers, J. Am. Chem. Soc. 118, 7578 (1996). V. S. Papkov and co-workers, Macromolecules 33, 7107 (2000). K. Kulbaba and I. Manners, Macromol. Rapid Commun., 22, 711 (2001). I. Manners, Adv. Organomet. Chem. 37, 131 (1995). R. Petersen and co-workers, Chem. Mater. 7, 2045 (1995). M. J. MacLachlan and co-workers, Adv. Mater. 10, 144 (1998). M. J. MacLachlan and co-workers, J. Am. Chem. Soc. 122, 3878 (2000). M. J. MacLachlan and co-workers, Science 287, 1460 (2000). M. J. MacLachlan, A. J. Lough, and I. Manners, Macromolecules 29, 8562 (1996). J. A. Massey and co-workers, J. Polym. Sci. Part B: Polym. Phys. 38, 3032 (2000). K. N. Power-Billard and I. Manners, Macromolecules 33, 26 (2000). ¨ F. Jakle, Z. Wang, and I. Manners, Macromol. Rapid Commun. 21, 1291 (2000). M. Ginzburg and co-workers, Langmuir 16, 9609 (2000). E. Fossum and co-workers, Macromolecules 28, 401 (1995). R. Rulkens and co-workers, Macromolecules 30, 8165 (1997). J. A. Massey and co-workers, Adv. Mater. 10, 1559 (1998). J. A. Massey and co-workers, J. Am. Chem. Soc. 123, 3147 (2001). I. Manners, Can. J. Chem. 76, 371 (1998). J. K. Pudelski and co-workers, Macromolecules 28, 7301 (1995). R. Rulkens, Y. Ni, and I. Manners, J. Am. Chem. Soc. 116, 12121 (1994). Y. Ni, R. Rulkens, and I. Manners, J. Am. Chem. Soc. 118, 4102 (1996). J. A. Massey and co-workers, J. Am. Chem. Soc. 122, 11577 (2000). J. A. Massey and co-workers, J. Am. Chem. Soc. 120, 9533 (1998). P. G´omez-Elipe and co-workers, J. Am. Chem. Soc. 120, 8348 (1998). R. Resendes and co-workers, Macromolecules 33, 8 (2000). Y. Ni and co-workers, J. Chem. Ed. 75, 766 (1998). M. J. MacLachlan and co-workers, Organometallics 17, 1873 (1998).

66 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215.

INORGANIC POLYMERS

Vol. 3

P. Nguyen and co-workers, Macromolecules 31, 5977 (1998). T. J. Peckham and co-workers, Macromolecules 29, 2396 (1996). T. J. Peckham and co-workers, Macromolecules 32, 2830 (1999). R. Rulkens and co-workers, J. Am. Chem. Soc. 119, 10976 (1997). J. M. Nelson and co-workers, Chem. Eur. J. 3, 573 (1997). P. F. Brandt and T. B. Rauchfuss, J. Am. Chem. Soc. 114, 1926 (1992). C. P. Galloway and T. B. Rauchfuss, Angew. Chem. Int. Ed. Engl. 32, 1319 (1993). D. L. Compton and T. B. Rauchfuss, Organometallics 13, 4367 (1994); D. L. Compton and co-workers, Chem. Mater. 7, 2342 (1995). H. M. Nugent, M. Rosenblum, and P. Klemarczyk, J. Am. Chem. Soc. 115, 3848 (1993). M. Rosenblum, Adv. Mater. 6, 159 (1994). M. Rosenblum and co-workers, Macromolecules 28, 6330 (1995). R. Arnold, S. A. Matchett, and M. Rosenblum, Organometallics 7, 2261 (1988); B. M. Foxman and co-workers, Organometallics 12, 4805 (1993). B. M. Foxman, D. A. Gronbeck, and M. Rosenblum, J. Organomet. Chem. 413, 287 (1991). M. Marcos, L. Oriol, and J. L. Serrano, Macromolecules 25, 5362 (1992). D. Perreault and co-workers, Inorg. Chem. 31, 3688 (1992). H. Chen and R. D. Archer, Macromolecules 28, 1609 (1995). H. Chen and R. D. Archer, Macromolecules 29, 1957 (1996). H. Chen, J. A. Cronin, and R. D. Archer, Macromolecules 27, 2174 (1994). J. A. Cronin, S. M. Palmer, and R. D. Archer, Inorg. Chim. Acta 251, 81 (1996). M. Hanack and M. Lang, Adv. Mater. 6, 819 (1994). T. J. Marks, Angew. Chem. Int. Ed. Engl. 29, 857 (1990). R. Knapp, A. Schott, and M. Rehahn, Macromolecules 29, 478 (1996). S. Kelch and M. Rehahn, Macromolecules 30, 6185 (1997). S. Kelch and M. Rehahn, Macromolecules 31, 4102 (1998). S. Kelch and M. Rehahn, Chem. Commun. 1123 (1999). S. Kelch and M. Rehahn, Macromolecules 32, 5818 (1999). R. P. Kingsborough and T. M. Swager, Chem. Mater. 12, 872 (2000). S. S. Zhu, R. P. Kingsborough, and T. M. Swager, J. Mater. Chem. 9, 2123 (1999). J. Buey and T. M. Swager, Angew. Chem. Int. Ed. 39, 608 (2000). E. C. Constable and P. Harverson, Inorg. Chim. Acta 252, 9 (1996). S. Achar and R. J. Puddephatt, Angew. Chem. Int. Ed. Engl. 33, 847 (1994). S. Achar and R. J. Puddephatt, Chem. Commun. 1895 (1994). S. Serroni and co-workers, Angew. Chem. Int. Ed. Engl. 31, 1493 (1992). G. R. Newkome and co-workers, Chem. Commun. 925 (1993). M. H. Chisholm, Angew. Chem. Int. Ed. Engl. 30, 673 (1991). K. Sonogashira, S. Takahashi, and N. Hagihara, Macromolecules 10, 879 (1977). S. Takahashi and co-workers, Macromolecules 11, 1063 (1978). N. Hagihara, K. Sonogashira, and S. Takahashi, Adv. Polym. Sci. 41, 149 (1981). S. J. Davies and co-workers, Chem. Commun. 187 (1991). B. F. G. Johnson and co-workers, J. Organomet. Chem. 409, C12 (1991). K. C. Sturge and co-workers, Organometallics 11, 3056 (1992). X. A. Guo and co-workers, Macromolecules 27, 7825 (1994). H. Nishihara and co-workers, Adv. Mater. 5, 752 (1993); I. Tomita, A. Nishio, and T. Endo, Macromolecules 27, 7009 (1994). S. S. H. Mao and T. D. Tilley, J. Am. Chem. Soc. 117, 5365 (1995). H. B. Fyfe and co-workers, Chem. Commun. 188 (1991). A. Abe, N. Kimura, and S. Tabata, Macromolecules 24, 6238 (1991). C. C. Frazier and co-workers, Polymer 28, 553 (1987). J. Lewis and co-workers, J. Organomet. Chem. 425, 165 (1992).

Vol. 3

INORGANIC POLYMERS

67

216. 217. 218. 219. 220. 221. 222. 223. 224. 225.

M. S. Khan and co-workers, J. Mater. Chem. 4, 1227 (1994). G. Frapper and M. Kertesz, Inorg. Chem. 32, 732 (1993). W. J. Blau and co-workers, J. Mater. Chem. 1, 245 (1991). A. E. Dray and co-workers, Synthetic Metals 41–43, 871 (1991). C. W. Faulkner and co-workers, J. Organomet. Chem. 482, 139 (1994). K. A. Bunten and A. K. Kakkar, J. Mater. Chem. 5, 2041 (1995). G. Jia and co-workers, Organometallics 12, 263 (1993). M. Altmann and co-workers, Adv. Mater. 7, 726 (1995). A. A. Dembek, R. R. Burch, and A. E. Feiring, J. Am. Chem. Soc. 115, 2087 (1993). D. Tanner, J. A. Fitzgerald, and B. R. Phillips, Angew. Chem. Int. Ed. Engl. Adv. Mater. 28, 649 (1989). 226. A. A. Dembek, P. J. Fagan, and M. Marsi, Macromolecules 26, 2992 (1993); A. S. Abd-El-Aziz and C. R. de Denus, Chem. Commun. 663 (1994).

SARA C. BOURKE IAN MANNERS University of Toronto

INTERFACIAL PROPERTIES.

See SURFACE PROPERTIES.