steroids 73 ( 2 0 0 8 ) 574–577 available at www.sciencedirect.com journal homepage: www.elsevier.com/locate/steroids Rapid communication A highly selective method for 11-OH protection of steroidal 11, 17-diols Paul A. Glossop, Marcel Hoogenraad, Torren M. Peakman, Dannielle F. Roberts * Pfizer Global R & D, Ramsgate Road, Sandwich, Kent, CT13 9NJ, United Kingdom article info Article history: Received 8 November 2007 Received in revised form 21 December 2007 Accepted 15 January 2008 Published on line 20 January 2008 Keywords: Mono-protection of diols Regioselective 3 ◦ desilylation Steroids Reaction optimisation abstract A highly selective method to protect the 11-OH position of steroid (1) has been devel- oped. This is achieved via double silyl protection of the 11, 17-diol, followed by selective desilylation of the 17-OH under basic conditions without the need for a fluoride source. © 2008 Elsevier Inc. All rights reserved. 1. Introduction Steroid (1) is a useful intermediate easily obtained in two steps by oxidative cleavage of prednisolone and subsequent esterification [1]. It was of interest to us to identify a method of mono-protecting the 11-hydroxy position in order to fur- ther explore substitution at the 17-hydroxyl of this template (Fig. 1). It has been previously shown that 21-Ac-prednisolone (2) can be selectively mono-silylated at the 11-OH position using a 2-step protection/deprotection sequence (Scheme 1) [2]. In our hands this protocol was found to be entirely reliable and robust up to a multigram scale. It therefore seemed not an unreasonable assumption that this method would be trans- ferable to substrate (1). ∗ Corresponding author. Tel.: +44 1304 648387. E-mail address: Dannielle.roberts@pfizer.com (D.F. Roberts). Trimethylsilylation of diol (1) proceeded smoothly to give doubly protected compound (5), however application of the existing deprotection conditions at this stage gave a com- plex mixture including regenerated starting material (1) and remaining (5) in addition to the desired product (4). Although a moderate yield of required intermediate (4) was eventually obtained, this was only after several tedious chromatographic purifications (Scheme 2). 2. Results and discussion In order to have a process that would be amenable to scale-up we considered reaction optimisation to be prudent. We thus set about exploring a number of variables in the deprotec- 0039-128X/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.steroids.2008.01.016