Two-directional dendritic macromolecules based on a 3,4-dihydrothiophene S , S -dioxide core: synthesis and thermolysis

It is known that tetranitrile 2 undergoes cheleotropic expulsion of SO 2 to yield the corresponding disubstituted butadiene. We herein expand this process to evaluating the expulsion of SO 2 from within the framework of a dendritic macromolecule. First and second tier dendrimers with the core 3,4-dihydrothiophene S , S -dioxide were synthesized and their thermolysis was examined. The results were consistent with the facile elimination of SO 2 coupled with an unexpected secondary dehydration processes at the elevated temperatures associated with this procedure.


Introduction
The core, 3,4-dihydrothiophene S,S-dioxide (1), can be readily synthesized from butadiene and sulfur dioxide but, more importantly, it is commercially available.The facile base-catalyzed addition of four equivalents of acrylonitrile to the 2 and 5 positions of 1, via literature procedures, 1,2 gave the desired tetranitrile 2, albeit in low (18-24%) overall conversion (Scheme 1).An interesting property of 2 is, however, found in its thermolysis, in that it undergoes 1 facile cheleotropic expulsion of sulfur dioxide to generate the tetrasubstituted butadiene 3; whereas, hydrolysis of 2 using concentrated hydrochloric acid afforded the corresponding tetraacid 4 in excellent yields by modification of an earlier procedure. 2Since this tetraacid can now be prepared on large scale in reasonable overall yields, it was interesting to evaluate the expulsion ISSN 1551-7004 Page 211  ARKAT USA, Inc of SO2 from within a macromolecular framework, which would afford a two-directional dendritic construction possessing an unsaturated internal functionality at a precise central locus.

Results and Discussion
The quantitative hydrolysis of 2 afforded the desired tetraacid 4, whose structure was established by the 4:1 ratio ( 1 H NMR) for the triplet at δ 2.95 (α-CH2) to that of the singlet at δ 6.59 assigned to the olefinic protons and the presence of a new peak ( 13 C NMR) at δ 173.

Scheme 1
Using the tetraacid 4, as the starting material, its amidation 3,4 with four equivalents of the 1 3 monomer (Beheras amine 5) 5 in the presence of dicyclohexylcarbodiimide (DCC) and N-hydroxybenzotriazole (HBT) provided (60%) the desired dodecaester 6.The structure of 6 was established by ( 13 C NMR) in which there appeared two peaks at δ 171.1 and δ 171.4 for the amide and ester carbonyl groups, respectively, as well as the vinyl ring protons ( 1 H NMR) at δ 6.17.Treatment of the dodecaester 6 with formic acid at ambient temperature afforded (92%) the colorless dodecaacid 7, which was characterized ( 1 H NMR) by the disappearance of the peak at δ 1.43 for the tert-butyl groups and the proper ratio of vinyl proton (δ 5.98) to NH (δ 7.3) or CO2H (δ 12.7) of 1:2 or 1:6, respectively.
Construction of the next tier was achieved in the same manner as described above.Formation of the 36-tert-butyl ester 8 from dodecaacid 7 was verified by 1 H NMR in which a tert-butyl signal appeared at δ 1.42 with an expected ratio to =CH of 162:1.Conversion of 8 to the corresponding 36-acid 9 afforded a white solid in moderate yield (56%).The identity of the 36acid 9 was confirmed by the disappearance of the tert-butyl signal (δ 1.42) and the appearance of a signal for the acidic CO2H protons at δ 12.05.The thermogravimetric analysis (TGA; Figure 1) for tetranitrile 2 showed the expected loss of SO2 at ca. 222 o C; the experimental mass loss of 18.6% was consistent with the calculated value of 18.9%.The TGA for tetraacid 4 showed a similar expulsion of SO2 at ca. 219 o C suggesting the formation of the corresponding tetrasubstituted butadiene acid (Scheme 3); the measured weight loss of 16.3% from 4 was in total accord with the calculated value of 15.7% for the desired expulsion of SO2 thus leaving the diene 10. Figure 1 depicts the TGA data for 2 and 4.There was also notable loss of water of hydration at ca. 100 o C, in that it is difficult to prepare an anhydrous sample.But more interestingly, there was, under these conditions, no evidence for anhydride formation in the case of 4, this must occur at temperatures greater than that for SO2 expulsion.It is well known that thermolysis of simple tert-butyl esters results in the corresponding carboxylic acids with the loss of isobutylene at ca. 220 o C. 6 The TGA of the exemplary, simple dendritic dodeca-tert-butyl ester (Figure 2) showed a weight loss of 37.9% at ca 220 o C.This corresponds to the expected loss of 12 isobutylene molecules along with four water molecules, suggesting the formation of a lactam via an intramolecular cyclization during the thermolysis process.It is noteworthy that the free Beheras amine (5) readily cyclizes at 60 o C to generate the corresponding lactam; 7 thus at 200 o C, the cyclization to the five membered imide appears reasonable.
Thermolysis of dodecaester 6 at 180 o C under vacuum (ca. 1 mbar) for four hours afforded an off-white material that solidified on standing.The resulting product was identified as the corresponding substituted octaacid butadiene 11.Firstly, the disappearance ( 1 H NMR) of the characteristic spike at δ1.42 for CCH3 with simultaneous appearance of the 13 C NMR signal for CO2H at δ 174 confirmed the loss of isobutylene.This observation is similar to that made during formic acid hydrolysis.Secondly, the olefinic HC=CH signal at δ 131.65 disappeared while two new olefinic signals at δ 121.5 and δ 139.0 emerged.This supports the loss of SO2 and the formation of the diene system.Thirdly, the existence of three C=O peaks at δ 172, 174 and 176 supports the presence of free CO2H moieties and the intramolecular cyclization process-giving rise to the unwanted lactam product.Although this may be a simple way to remove the amide proton from within the macromolecular construct, we were unsuccessful in retarding this unfavorable dehydration process.Simpler models are currently being utilized to see if controlled thermolysis can afford the desired conversions.