In In In In

Fig. 12.2 Illustration of the different repulsive interactions driving the all-anti imine functions in the polyisocyanide chain out of planarity.

Fig. 12.3 Two proposed most stable conformations for polyisocyanides.

Fig. 12.3 Two proposed most stable conformations for polyisocyanides.

creases to R = C(CH3)3. For intermediate steric bulk (e.g. R = CH3), the authors found two helical minima with different degrees of helicity. The influence of steric bulk of the substituent on the conformation of polyisocyanides was experimentally corroborated by Yamada et al. who found that polyisocyanides derived from phenylalanine readily lose their initial helical conformation in solution when the carboxyl group is protected as an ethyl ester. Increasing the bulk of the ester to a t-butyl ester significantly increased the stability of the helical conformation of the polyisocyanide [40].

An alternative conformation for polyisocyanides was postulated on the basis of calculations by Clericuzio et al. [41]. The suggested repulsion between the N lone pairs in the planar all-anti conformation of polyisocyanides is absent in the so-called ''syndio'' conformation in which dimeric sections are alternatively (E,E) and (Z,Z) (Fig. 12.3). On the basis of both ab initio and molecular mechanics (MM) calculations, this conformation was found to be by far the most stable among a number of different possible geometries for polyisocyanides. The syndio conformation is non-helical but highly symmetrical and has a regularly alternating configuration of side chains on C=N double bonds, and an alternating 180° + 90° conformation of the backbone N=C-C=N angles (Fig. 12.3). The driving force for this conformation seems to lie partly in the large preference for E,E trans-planar diiminic units: the rotation around the N=C-C=N central bond in the E,E ethane diimine (as calculated by ab initio) shows a (s-cis)-(s-trans) energy difference of @34 kJ mol-1 in the flexible rotor approximation.

The proposed syndio-conformation was postulated to be very likely for oligo-(phenylisocyanide)s. NMR spectroscopic investigations demonstrated that the 1,3-diphenyl-1,2,3-tri(phenylimino)propane has a planar diimine unit with the third imine located at a 90° dihedral angle from the planar s-trans portion [42]. For polymers of this compound, i.e. poly(phenylisocyanide), the authors observed that in halocarbon solvents or tetrahydrofuran solution the conformation slowly changes from a helix to a random coil [43].

In addition to the computational approaches which suggest possible alternative regular conformations for polyisocyanides different from the helical one, Green et al. pointed out that polyisocyanides, especially with small pendant groups, actually adopt a quite irregular conformation. They emphasized the difference in the Mark-Houwink exponent a between several polyisocyanides, obtained by viscos ity measurements [44]. The Mark-Houwink exponent gives an indication about the rigidity of molecules [45]: if a < 1, the polymer has a random-coil character and if a > 1, the polymers have a rigid-rod character. The Mark-Houwink exponents for poly(a-phenylethyl isocyanide) (10) in toluene, b-phenylethyl isocyanide in tetrahydrofuran and racemic 2-octyl isocyanide in toluene were found to be a = 1.36 [46], 0.68 [47] and 1.75 [48, 49], respectively. These results suggest that the structure of the pendant group strongly affects the chain dimensions. Using light scattering experiments Green et al. showed that even the relatively stiff poly(a-phenylethyl isocyanide) 10 has only a limited persistence length of @3 nm (polyisocyanopeptides by contrast, possess much longer persistence lengths; see Section 12.4). In addition, a large chemical shift dispersion for all carbon atoms, including the backbone carbon, was found in the 13C NMR spectrum of this polymer. This dispersion was even stronger for polyisocyanides lacking an a-substitu-ent, such as 8 and 9 (Fig. 12.4). Since a broad chemical shift dispersion was also observed for polymers from achiral monomer units, Green et al. suggested that this stereo-irregularity is associated with syn-anti isomerism about the carbon-nitrogen double bond (Fig. 12.4D).

Fig. 12.4 Signal for the imine carbon atom in the polymer backbone in the 13C NMR spectrum for polyisocyanides lacking an «-substituent (A and B) and with an «-substituent (C). (D) Stereoisomeric possibilities for a triad in a polyisocyanide. (Adapted with permission from Ref. 44 Copyright 1988.)

Fig. 12.5 Schematic drawing of the different polymer conformations for polyisocyanides as illustrated by Millich and Baker. Blocks 1-4 correspond to structures with different imine conformations and different helicity (see text).

Fig. 12.5 Schematic drawing of the different polymer conformations for polyisocyanides as illustrated by Millich and Baker. Blocks 1-4 correspond to structures with different imine conformations and different helicity (see text).

The carbon nuclei may be sensitive to sequences longer than the triads depicted in Fig. 12.4D. These results indicate that the conformational data for polyisocyanides proposed from interpretation of CD spectroscopy [50-59] are difficult to reconcile with the structural disorder revealed by the high-field 13C NMR spectroscopic data. Millich and Baker [33] already suggested the possibility of blocks with different syn-anti isomerism of the imino group together with the possibility of helix reversals as illustrated in Fig. 12.5. Block 1 is obtained by rotation of block 2 around the short axis (similar for 3 and 4), which corresponds to syn-anti isomerism of the imino group, while blocks 1 and 3 and blocks 2 and 4 have opposite helix senses. In the case of achiral monomers, 1 and 4, and 2 and 3, are mirror images, whereas for chiral monomer units they are diastereomers.

Takahashi et al. showed that polymerization conditions can have an influence on the stereoregularity of the resulting polymer [60]. Polyisocyanide 12 prepared by polymerizing 11 with NiCl2 in methanol at room temperature revealed a lower specific rotation (= +354) and CD (Ae364 = +3.9) than polymer 13 (Md = +1070 and Ae364 = +13.0), which was prepared by Pd-Pt catalyst 5a in re-fluxing THF (Fig. 12.6). By annealing polymer 12 in refluxing THF for 15 h the

Fig. 12.6 Reaction scheme for the polymerization of 11 by NiCl2 6H2O and Pd-Pt catalyst 5a.

specific rotation and CD intensity increased to values of [a]D = +1038 and De364 = +11.6, whereas no increase was observed for polymer 13. More information on the conformational changes was obtained by 13C-NMR spectroscopy; a rather broad signal for the imino carbons of the backbone (width at half height: W1/2h = 208 Hz) was observed for polymer 12 in contrast to the much sharper signal for polymer 13 (W1/2h = 133 Hz). After annealing polymer 12 in deuter-ated toluene at 80 ° C for 15 h, a similar sharp signal as for 13 was found. Comparable results were obtained for other polymers, including achiral polymers, thus the 13 C NMR signal distribution for the imino backbone carbon cannot be explained by stereochemical means. The authors therefore proposed, in line with the work of Green et al., that the initial stereo-irregularity in the polymers formed by the nickel-catalyzed polymerization is associated with the existence of both syn-and anti-isomers of the imino groups in the backbone. The irregular conformation can be transformed into the thermodynamically stable stereoregular form by syn-anti isomerization of the imino group at high temperatures. The polymerization at high temperature with Pd-Pt catalyst 5a immediately leads to the stereoregular conformation.

Yashima et al. further observed that polyisocyanides are not necessarily always present in a stable locked structure, but that some of them can have more dynamic conformations. They provided evidence for a reversible transition between two conformational states of poly(4-carboxyphenylisocyanide) 14 under influence of optically active amines and amino alcohols such as 15 and 16 (Scheme 12.3) [61, 62].

Based on their observations they suggested, that apart from the aforementioned imino syn-anti isomerism, these changes are caused by backbone (s-trans)-(s-cis) isomerism (Fig. 12.7). Initially a 41-helix is formed that rapidly looses its regular helix structure to form a irregular structure with s-trans, s-cis and (s-cis)-(s-trans) domains (step 1). This structure will slowly transform into the stable all s-trans structure at 30 °C resulting in sharper peaks in the 1HNMR spectrum (step 2). It is proposed that the latter structure can give a helical arrangement upon bind-

Fig. 12.7 Proposed conformational changes and helix induction in 14. (Adapted with permission from Ref. 61, Copyright 2002, American Chemical Society.)

ing chiral amines by directing the random twist around the C-C bonds in a single direction (step 3). In water a helix could also be induced, which after removal of the optically active amines maintained its helical conformation at ambient temperatures, but at elevated temperatures the helix unfolded readily. It was postulated that a combination of hydrophobic and chiral ionic interactions in water is responsible for the helix formation and the memory effect because induced helices in DMSO were unable to maintain their helical structure (see also Chapter 11 and Fig. 11.16) [62, 63].

Summarizing, we may conclude that two important structures for polyisocyanides have been proposed: (i) a helix structure that is close to the 41-helix initially suggested by Millich et al., which is most likely for polymers with bulky sidearms and (ii) the syndio structure as was calculated by Salvadori et al., which seems to be most favorable for polyisocyanides with small side-groups. In addition to these two distinct structures, Yashima et al. calculated a 125-helix for their induced helices [61] and Young et al. both a 41-helix and a 31-helix [64].

Irregularities that have been observed in polyisocyanides are mainly explained by syn-anti isomerism of the imino side-groups and by (s-cis)-(s-trans) isomerism of the carbon backbone. This latter isomerism leads to structures which are intermediates between a 41-helix and the syndio structure. Other possible explanations for observed irregularities are helix inversions as discussed by Millich and the existence of two types of helical pitches coexisting within one polymer as was calculated by Kollmar and Hoffmann for polyisocyanides with intermediate bulk.

Was this article helpful?

0 0

Post a comment