Search

An effective strategy for making new ordered molecular materials by design is to build them from subunits that associate with their neighbors and thereby hold them in predetermined positions. Such tectons can be made by attaching multiple peripheral sites of association to geometrically suitable cores. In tectons 3−5, for example, simple triaminotriazine groups grafted to a nominally tetrahedral tetraphenylmethyl core form intermolecular hydrogen bonds according to reliable motifs, and crystallization thereby leads to the formation of porous three-dimensional networks. Functionalization of the triaminotriazine groups with alkyl chains gives rise to amorphous tectons, and the increase of the chain length lowers the glass transition temperatures of these materials, allowing their association in the melt to be studied by rheological methods. The rheological properties depend in a characteristic way on the length of the alkyl substituents. In particular, viscosity and Tg both decrease with increasing length of the alkyl chains in the series of tectons 7−10, suggesting that intertectonic hydrogen bonding is increasingly diminished by the steric effects of the substituents. In this way, formation of extended hydrogen-bonded networks becomes gradually unfavorable, and mobility and fluidity are enhanced. For these reasons, melts derived from tectons 7−10 do not behave like conventional linear polymers or dendrimers, but rather form a novel class of materials with unique properties of their own.

Molecules that associate to form porous networks can be made by attaching multiple hydrogen-bonding sites to suitable cores. Pentaerythrityl tetraphenyl ether, a four-armed core, is the progenitor of dendritic derivatives with more arms, including dipentaerythrityl hexaphenyl ether 7. An advantage of such dendritic derivatives is that the resulting networks are held together by larger numbers of intermolecular hydrogen bonds.

Treatment of 2,2‘-(1,3-butadiyne-1,4-diyl)bis[6-(1,1-dimethylethyl)-4-methylphenol] (1) with 2 equiv of Al(i-Bu)3 converts the two hydroxyl groups into OAl(i-Bu)2 groups, thereby transforming diphenol 1 into the corresponding bis(di-i-butylaluminum phenoxide) 8, which holds two strongly Lewis acidic atoms of aluminum in a potentially convergent orientation. Bidentate Lewis acid 8 forms a crystalline 1:1 adduct with 1,2-dimethoxyethane (DME). An X-ray crystallographic study revealed that the adduct is a linear oligomer in which the two Lewis acidic sites of reagent 8 each bind a basic oxygen atom from different molecules of DME. However, low-temperature NMR studies indicated that in solution a discrete 1:1 adduct 9 is favored, in which the two Lewis acidic sites of reagent 8 each bind one of the two basic sites in a single molecule of DME. Formation of adduct 9 provides an example of the recognition and binding of a multidentate Lewis base by a complementary multidentate Lewis acid. Addition of 2 equiv of TiCl4 to diphenol 1 converts the hydroxyl groups into OTiCl3 groups and produces the corresponding bis(trichlorotitanium phenoxide) 11, which forms an unusual 1:2 complex with CH3COOC2H5. An X-ray crystallographic study of this complex established that each Lewis acidic atom of titanium binds only 1 equiv of CH3COOC2H5 to form an unprecedented pentacoordinate adduct with a square-pyramidal geometry. Formation of diverse reagents 8 and 11 from the same precursor demonstrates that the strategy of converting organic compounds with suitably oriented hydroxyl groups into the corresponding metal alkoxides is a versatile and effective way to make multidentate Lewis acids.

Treatment of oxo[μ-(3,4,5,6-tetramethyl-1,2-phenylene)]dimercury (6) with trifluoromethanesulfonic acid produces the bis(trifluoromethanesulfonate) salt of cyclic compound 7, in which two 1,2-phenylenedimercury units are bridged by hydroxyl groups. μ-Hydroxide 7 is a tetradentate Lewis acid, and it forms a novel 1:2 complex with diethylformamide. An X-ray crystallographic study established that μ-hydroxide 7 defines an approximately planar and symmetric 10-membered ring, which holds four Lewis acidic atoms of mercury in an orientation that allows them to interact simultaneously with each of the two carbonyl oxygen atoms of the bound molecules of diethylformamide. Formation of the complex causes an important decrease in the carbonyl stretching frequency of the amide, which suggests that multiple coordination may weaken the C&dbd;O bond and enhance its reactivity. Treatment of dichloro(3,4,5,6-tetramethyl-1,2-phenylene)dimercury (9) with 2 equiv of AgOSO2CF3, followed by crystallization of the product from diethylformamide, provided a 2:1 complex of diethylformamide with the corresponding 1,2-phenylenedimercury bis(trifluoromethanesulfonate) 8. An X-ray crystallographic study revealed that the complex is unusual because each Lewis acidic atom of mercury forms one of its two primary bonds to the formally neutral carbonyl oxygen atom of diethylformamide, and the anionic trifluoromethanesulfonate ligands are relegated to secondary bridging positions. Structural and spectroscopic effects suggest that binding of diethylformamide by bis(trifluoromethanesulfonate) 8 is unusually strong, even though it is monodentate.

Treatment of 2,2‘-(1,2-ethynediyl)bis[6-(1,1-dimethylethyl)-4-methylphenol] (2) with 2 equiv of Al(i-Bu)3 converts the hydroxyl groups into potentially Lewis acidic OAl(i-Bu)2 groups with a convergent orientation. Their intramolecular interaction produces the novel structure 11, which incorporates an Al2O2 ring that is markedly distorted by constraints imposed by the diphenylacetylene framework. Unlike normal Al2O2 rings, which are planar and have trigonal planar oxygen atoms, the Al2O2 ring of compound 11 is puckered, and its oxygen atoms are distinctly pyramidalized. In addition, the average Al−O bond length in compound 11 (1.913(2) Å) is significantly longer than these found in related dimers of alkylaluminum alkoxides and aryloxides (1.840−1.895 Å).

A variety of 2,7-disubstituted derivatives of 1,8-naphthalenediol (2) of potential value as ligands for the complexation of transition metals can be prepared from 1,8-naphthalenediyl bis(diethylcarbamate) (3) and 1,8-bis(methoxymethoxy)naphthalene (9) by directed ortho-metalation and subsequent functionalization. The reaction of TiCl4 with 1,8-naphthalenediol (2) in the presence of diisopropyl ketone produced an adduct of a novel chloride-bridged dimer of [1,8-naphthalenediolato(2^-)-O,O‘]dichlorotitanium (16; R = H). Its unusual structure indicates that complexes derived from 1,8-naphthalenediols can have features not normally observed in complexes derived from simple phenols.