• PDF / 400,400 Bytes
• 6 Pages / 420.48 x 639 pts Page_size
• 67 Downloads / 247 Views

DOWNLOAD

REPORT

---

ORGANIC POLYRADICAL MODELS FOR ORGANIC MAGNETIC MATERIALS DAVID A. MODARELLI, FRANK C. ROSSITrO, MASAKI MINATO, AND PAUL M. LAHTI* University of Massachusetts, Department of Chemistry, Amherst, MA 01003 ABSTRACT A variety of model organic high spin molecules based upon coupled phenoxy and phenylnitrene systems has been examined by semiempirical MO-Cl theory, Experimental methodology using photochemical and thermal methods in solution and solid phases to generate and study phenoxy groups for investigation of these models is presented. INTRODUCTION Recent interest in organic polyradicals has been part of a general interest in new magnetic molecules of potential use as information storage materials. Various types of organic or organometallic magnetic materials have been recently described.[ i1 We have been interested primarily in materials based upon through-bond and through-space high spin coupling of many radicals (or other high spin centers such as nitrenes) to produce very high spin polyradicals. Experimental work such as lwamura's[2] has shown that the strategy of coupling carbenes through conjugating groups -- in accordance with qualitative rules[3-5] predicting high spin coupling -- can lead to very high spin organic molecules, albeit under conditions of high dilution in crystalline or frozen solution matrix at cryogenic conditions. We wish to couple phenoxy based radicals together to form such very high spin systems. Although one gets only one spin per center -- as opposed to two per center for carbenes and nitrenes -- the greater intrinsic stability of phenoxy based systems by comparison to the more reactive high spin centers makes their synthesis attractive. Recent efforts to prepare phenoxy based polymer systems by solution oxidation of hindered phenolic systems have not yielded high spin counts.[ 1.61 but the reason for this failure at present is not entirely clear. In this paper, we qualitatively summarize the application of our previously described[7-81 semiempirical computational algorithms to selected phenoxy based diradicals, and the relationship of these computations to connectivity-based models described by other workers.[3] Experimentally, we show how use of our previously developed photochemical and thermal methods[9] for solid state generation of stable phenoxyl radical centers may be extended to production of unhindered phenoxy systems under conditions where they may be readily detected and observed. Meta and para-stilbeneoxyl radicals have both been readily generated in matrix and in neat precursor matrix, studied by electron spin resonance (ESR) and ultraviolet-visible (UV-vis) spectroscopy, and their decomposition behavior in polymeric matrices elucidated. Similar experiments have been carried out for other non-tert-butylated phenoxy derivatives. By these and similar experiments, we hope to understand what types of phenoxy based systems are fairly robust under matrix isolation conditions (polymer matrices at room temperature and above), how efficiently our methods allow production of phenoxyl