The true configuration of the benzilosazone isomers Marı ´a V. Mirı ´fico, a,b, * Jose ´ A. Caram a and Enrique J. Vasini a a Instituto de Investigaciones Fisicoquı ´micas Teo ´ ricas y Aplicadas (INIFTA), Facultad de Ciencias Exactas, Departamento de Quı ´mica, Universidad Nacional de La Plata. C.C. 16, Suc. 4, (1900) La Plata, Argentina b Fac. de Ingenierı ´a, Universidad Nacional de La Plata, Argentina Received 17 May 2006; revised 20 June 2006; accepted 21 June 2006 Available online 7 August 2006 Abstract—The 1 H NMR-based argument previously used to assign configurations to the three stereoisomers of benzilosazone is briefly reviewed. The configuration of the stable isomer is shown to be Z,Z by single crystal X-ray diffraction analysis, instead of E,E as previously reported. This assignment, together with physical measurements and spectroscopic (NMR and UV) data, allows the establishment of the configuration of all isomers. Computational methods are employed to clarify the relation between the configuration of the isomers and their 1 H NMR, thus explaining the origin of the previous erroneous assignment. Ó 2006 Elsevier Ltd. All rights reserved. The three possible diastereoisomers of benzilosazone, which differ in the spatial arrangement around the two C@N double bonds, are shown in Scheme 1. These isomers were obtained by Spassov, 1 who labeled the stable isomer, melting at 234 °C, as b-osazone. He named the two other isomers in accordance with their physical properties as c g (from gelb = yellow, mp 175 °C), and c r (from r = rhombic, mp 196 °C). Both c-osazones revert to the b-osazone on melting and resolidification. Woodward and Wintner 2 verified the work of Spassov and clarified earlier papers by showing that some reported configurational isomers of benzilosazone were instead chain isomers or mixtures of isomorphous compounds. In this context, they measured the X-ray diffraction of b-osazone crystal powder. Of course, pow- der X-ray diffraction does not provide information on molecular structure. Commenting on the assignment of a configuration to the isomers, they cautiously stated ‘in the absence of evidence to the contrary, it seemed not unreasonable to assume an E,E configuration for the b-osazone.’ The c g isomer was assigned the asym- metric E,Z configuration on the basis of a doublet in the N–H band of its IR spectrum. Spassov and Christova claimed to have established the configurations of the isomers, 3 based on their 1 H NMR results in DMSO-d 6 . They observed that the two osazones, c r and b, presented only one N–H proton resonance (c r = d: 9.15, 2H; b = d: 9.45, 2H). The c g - osazone showed two N–H proton resonances, at d 11.20 and 9.45 ppm, both equal in intensity (1H). The low field signal at d 11.20 ppm was assigned to the ‘che- lated’ proton of the E,Z isomer (Scheme 1), which nec- essarily belongs to the Z-configured half of the molecule. The remaining signal, at d 9.45, should correspond to the N–H proton of the E-configured half. Thus, they argued that the b isomer, presenting one 2H signal at d 9.45 ppm, should be the E,E isomer. This left for the protons in the Z,Z configuration the 9.15 ppm chemical shift, corresponding to the c r isomer. The authors stated that their results confirmed the X-ray structural assignment of Woodward and Wintner. How- ever, as mentioned above, such assignment had not been made. The present work revises these assignments on the basis of the first measurement of the single crystal X-ray diffraction of a benzilosazone isomer. A comparison of 0040-4039/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.tetlet.2006.06.109 * Corresponding author. Tel.: +54 221 425 7430; fax: +54 221 425 4642; e-mail: mirifi@inifta.unlp.edu.ar N N H N HN Ph Ph Ph Ph N NH N HN Ph Ph Ph Ph N NH N NH Ph Ph Ph Ph E, E E, Z Z, Z Scheme 1. Tetrahedron Letters 47 (2006) 6919–6922