The synthesis of new 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) styryl derivatives as mediators for the living free-radical polymerization is described. Two of the alpha-methyl groups at the 2- and 6-position of the parent TEMPO styryl alkoxyamine have been replaced by hydroxymethyl and silyloxymethyl groups. To further increase the steric hindrance around the alkoxyamine oxygen atom, the remaining two methyl groups have been substituted with larger ethyl groups. Styrene polymerizations using hydroxy-substituted TEMPO derivatives are fast, but are not well-controlled. As previously shown for other OH-substituted alkoxyamines, intramolecular H-bonding leads to an acceleration of the C-O bond homolysis and, hence, to an acceleration of the polymerization process. However, the OH groups also increase the alkoxyamine decomposition rate constant. The kinetics of the C-O bond homolysis have been determined using EPR spectroscopy. Decomposition studies have been conducted with the aid of 1H NMR spectroscopy. In contrast to the OH-substituted alkoxyamines, highly hindered silyloxy-substituted TEMPO alkoxyamines turned out to be excellent mediator/initiators for the controlled styrene polymerization. Polystyrene with M(n) of up to 80 000 g/mol and narrow polydispersities (PDI) has been prepared using the new alkoxyamines. Reactions have been conducted at 105 degrees C; however, even at 90 degrees C controlled but slow polymerizations can be achieved. Furthermore, and more importantly, poly(n-butyl acrylates) with narrow PDIs (<1.15) have been prepared at 105 degrees C with the new alkoxyamines. Controlled acrylate polymerization can be conducted at temperatures as low as 90 degrees C. The silylated alkoxyamines presented belong to the most efficient initiator/mediators for the controlled acrylate polymerization known to date. The effect of the addition of free nitroxide on the acrylate polymerization is discussed. Moreover, the synthesis of diblock copolymers with narrow PDIs is described.

The synthesis of new six- and seven-membered cyclic alkoxyamines bearing ethyl groups at the alpha-N position of the alkoxyamines is described. The key step in the synthesis of the sterically hindered six-membered cyclic alkoxyamines is a Wadsworth-Horner-Emmons olefination with bisphosphonate 1. The seven-membered cyclic alkoxyamines were prepared from the corresponding six-membered keto alkoxyamines by ring-enlargement with trimethylsilyl(TMS)-diazomethane. The use of the new alkoxyamines as regulators/initiators for radical polymerization is discussed. Efficient controlled and living polymerization of styrene and n-butyl acrylate was obtained with the six-membered tetraethyl alkoxyamine 13. Controlled polymerizations can be conducted even at 90 degrees C. In addition, alkoxyamine 13 can be used for the preparation of AB diblock and ABA triblock copolymers with narrow polydispersities. The influence of the replacement of methyl groups in the alpha-position of the N atom in cyclic alkoxyamines by larger ethyl groups on the styrene polymerization (reaction time, PDI, kinetics of the C-O bond homolysis) is discussed. In addition, thermal decomposition of the new alkoxyamines was studied. Furthermore, the synthesis of N,N-bissilylated alkoxyamines is described. The silylated alkoxyamines are not suitable as regulators/initiators for the controlled/living radical polymerization. The C-O bonds in silylated alkoxyamines are stronger than the C-O bonds in analogous N,N-dialkylated alkoxyamines. The experimental results are verified by calculations with Gaussian 98 (A. 9).

The synthesis of three new sterically hindered Hawker−Braslau type alkoxyamines for the nitroxide-mediated living radical polymerization is described. Efficient polymerizations can be performed at 105 °C. With alkoxyamine 11 controlled polymerization of n-butyl acrylate and styrene is possible even at 90 °C. AA and AB diblock copolymers can be prepared in a controlled manner using alkoxyamine 11. Kinetic EPR experiments for the determination of the C−O bond activation energies of the new alkoxyamines are described. Furthermore, 1H NMR alkoxyamine decomposition studies are presented. These kinetic experiments are used for the discussion of the polymerization results. Furthermore, the X-ray structure of a Hawker-Braslau type nitroxide is presented. X-ray data and ESR data are used to explain the high stability of these nitroxides.

The effect of the variation of the alkoxyamine concentration on the conversion and polydispersity of the nitroxide‐mediated living free‐radical polymerization of styrene is discussed. Four different alkoxyamines (1–4) have been used for these studies. For an alkoxyamine with a small equilibrium rate constant (K), such as styryl–TEMPO 2, the conversion is governed by the autopolymerization of styrene. For efficient alkoxyamines 1, 3, and 4, the conversion at high alkoxyamine concentrations is higher than the conversion obtained by autopolymerization. At high alkoxyamine concentrations, the conversions vary to a small extent for all the alkoxyamines studied. As long as the conversion remains high, the polydispersity index is small. In addition, simulations of polymerizations with a program for modeling nonlinear dynamics are discussed. Polymerizations with efficient alkoxyamines at high alkoxyamine concentrations are well described by the kinetic scheme applied. K for alkoxyamines 1 and 4 has been estimated with the simulations. © 2004 Wiley Periodicals, Inc. J Polym Sci Part A: Polym Chem 42: 3342–3351, 2004

A new method for the in situ generation of various Ti(III) complexes is presented. Cyclohexadienyl-Ti(IV) derivatives, which are readily prepared from the corresponding lithiated cyclohexadienes, afford the corresponding Ti(III) complexes upon thermal C À Ti-bond homolysis. The Ti(III) compounds generated using this novel method have successfully been used in the reductive dimerization of benzaldehyde. In particular, TiBr 3 , TiCp 2 Cl and Ti(O-i-Pr) 3 have been generated via this approach. Moreover, the method also offers an entry to new chiral Ti(III) complexes as documented by the preparation of Ti(III)CpTADDOLate.Keywords: radical chemistry; reductive cyclization; stereoselective reductions; synthetic methods; titanium Radical chemistry has gained increasing importance during the last 30 years.[1] In contrast to ionic chemistry, many functional groups are tolerated under radical conditions. Several methods for the clean generation of radicals have been reported to date. However, most of the radical reactions are conducted using toxic trialkyltin hydrides. To circumvent the use of toxic tin compounds, many research groups are currently looking at environmentally benign radical processes.[2] Electron-transfer reagents have successfully been used in this context. In particular, samarium diiodide (SmI 2 ) has been found to be a highly efficient electron-transfer reagent for conducting various radical processes.[3] However, SmI 2 is rather expensive and is readily oxidized on air.As an alternative to SmI 2 , Ti(III) complexes have been studied by various groups. [4 -11] Nugent and RajanBabu showed that titanocene(III) chloride can be applied to reductively open epoxides.[5] The b-titanoxy radicals thus generated can be used in radical cyclizations and intermolecular addition reactions. Gansäuer [6,7] later showed that the epoxide openings can be performed using catalytic [8] amounts of titanocene(III) chloride.Reduction of activated alkyl halides have been achieved using Ti(III).[9] Moreover, Ti(III) complexes have successfully been used in pinacol reactions. [10,11] Even enantioselective couplings have recently been reported. [12] Most of the Ti(III) complexes studied have been generated from the corresponding Ti(IV) compounds and a coreducing reagent. [13] Recently, we published our results on the use of cyclohexadienyl-Ti compounds in ionic allylations (Scheme 1). [14] We assumed that these compounds may also offer a clean entry to Ti(III) complexes upon simple thermal Ti-carbon bond homolysis. In contrast to established methods for the generation of Ti(III) derivatives, a coreducing reagent is not necessary in our approach. [15] Since the starting cyclohexadienyl-Ti(IV) compounds are readily prepared by transmetalation using the appropriate Ti(IV) complex, new unexplored Ti(III) reagents should be available via this route. In this communication we report first results on the use of cyclohexadienyl-Ti(IV) derivatives as precursors for Ti(III) complexes. Moreover, these reducing reagents will be appli...