Attempted Resolution of Citalopram Using (-)-O,O′-Di-p-toluoyl-(R,R)-tartaric Acid, and Reflections on an Alkylation Reaction; Comment on an Article by Elati et al.

A recent article by Elati et al. (Elati, C. R.; Kolla, N.; Vankawala, P. J.; Gangula, S.; Chalamala, S.; Sundaram, V.; Bhattacharya, A.; Vurimidi, H.; Mathad, V. T. Org. Process Res. Dev. 2007, 11, 289-292) describes the synthesis of escitalopram by means of a three-step process: (i) an alkylation reaction to provide dides-methylcitalopram, (ii) resolution of didesmethylcitalopram by classical resolution using (-)-O-O'-di-p-toluoyl-(RR)-tartaric acid (DTT) as the chiral acid, and (iii) dimethylation of the resolved product to give escitalopram. However, they also mention resolution of citalopram itself by classical resolution, again using DTT as the chiral acid. We have been unable to reproduce their resolution of citalopram, obtaining only racemic or nearly racemic material. In order to better understand the system, we constructed two ternary solubility diagrams from solubility data at different temperatures. The resultant isotherms show the presence of a solid solution across the majority of the diagram in the temperature range 0-25 degrees C. This finding was in agreement with data from X-ray diffractograms. In addition, the solubility of the desired (S)-citalopram center dot DTT salt was found to be a factor of 5 higher than that of the corresponding R/S double addition salt. Furthermore, kinetics studies have indicated that the formation/crystal growth of the R/S double addition salt is preferred/faster than that of the desired (S)-citalopram center dot DTT salt. Taken as a whole, our findings show that resolution is not possible in any practical sense in the system described by Elati et al. Furthermore, we believe that detailed examination of their alkylation procedures casts doubt on their reproducibility.