Impurities are an extremely critical issue in the pharmaceutical industry especially since the strict regulations have been introduced by ICH (International Conference on Harmonisation, http://www.ich.org). Typically process-related impurities are unwanted chemicals that remain with the active pharmaceutical ingredients (APIs) and could be generated at any of the synthetic steps. For example, a major hurdle is to separate organic synthesis reaction intermediates from a mixture containing multicomponent impurities, such as inorganic salts, polymers, isomers, and colored byproducts [1]. While the first two types of compounds can be separated by quench and charcoal treatment respectively the last two are found to be the most challenging task in purification. Another example is two API impurities, i.e. C38H41N5O9S, (r.m.m. 775) and C39H44N4O10S2, (r.m.m. 792), have been isolated by solid-phase extraction and preparative HPLC and characterized, by liquid chromatography mass spectrometry and/or NMR spectrometry after a crystallization process of the drug substance, 1-(1,2,3,5,6,7-hexahydro-s-indacen-4-yl)-3-[4-(1-hydroxyl-1-methyl-ethyl)-furan-2-sulphonylurea, C39H44N4O10S2, (r.m.m. 404) [2]. The two impurities can not be easily removed by conventional processes because they are pseudo-dimers of the API and have high structural similarities to each other.

Although Process Analytical Technologies (PAT) are suggested to identify the quantity and quality of impurities in the early APIs developing stage, a more efficient process intensification is necessary to control these process-related impurities at each step of manufacturing. The development of a new generation nanofiltration membranes stable in organic solvents emerges as an important step toward solving this problem. Although just recently commercially available, organic solvent nanofiltration (OSN) has been applied successfully to various fine chemical and pharmaceutical processes, such as chiral separation, solvent exchange, catalysts separation and recycling (e.g. organometallic, phase-transfer and transition metal catalysts), lube oil recovery, and edible oil deacidification. [3]

This work is aiming to prove the feasibility and capability of a nanofiltration membrane cascade to remove impurities from APIs production streams. A multi-stage OSN membrane cascade, consisting of a series of flow-through stirred cells, is used as an experimental set-up. Two solutes with molecular weights close to the case illustrated above (i.e. Brilliant Blue R, m.w. 826 g•mol-1 and Martius Yellow, 2,4-Dinitro-1-naphthol sodium salt, m.w. 274.16 g•mol-1), representing API and colored by-product respectively are selected as a model system. Mathematical model simulations were carried out prior the experimental work in order to identify the best cascade configuration and operating conditions. The results have shown that OSN membrane cascade is a promising robust solution for purification of the APIs production streams.

In addition, a trial will be made to describe the three-component mixture separation on the basis of the solution-diffusion model for membrane transport and the film theory for liquid mass-transfer effects.

1.Organic Process Research & Development 8 (3): 376-380 May-Jun 2004.

2.Journal of Pharmaceutical Sciences 93 (9): 2296-2309 Sep 2004

3.Annals of the New York Academy of Sciences, 984: 123-141. 2003