Charpentier Maxime D, Devogelaer Jan-Joris, Tijink Arnoud, Meekes Hugo, Tinnemans Paul, Vlieg Elias, de Gelder René, Johnston Karen, Ter Horst Joop H
EPSRC Centre for Innovative Manufacturing in Continuous Manufacturing and Crystallization (CMAC), University of Strathclyde, Technology and Innovation Centre, 99 George Street, Glasgow G1 1RD, U.K..
Institute for Molecules and Materials, Radboud University, Heyendaalseweg 135, 6525AJ Nijmegen, The Netherlands.
Cryst Growth Des. 2022 Sep 7;22(9):5511-5525. doi: 10.1021/acs.cgd.2c00615. Epub 2022 Aug 25.
Pharmaceutical cocrystals are highly interesting due to their effect on physicochemical properties and their role in separation technologies, particularly for chiral molecules. Detection of new cocrystals is a challenge, and robust screening methods are required. As numerous techniques exist that differ in their crystallization mechanisms, their efficiencies depend on the coformers investigated. The most important parameters characterizing the methods are the (a) screenable coformer fraction, (b) coformer success rate, (c) ability to give several cocrystals per successful coformer, (d) identification of new stable phases, and (e) experimental convenience. Based on these parameters, we compare and quantify the performance of three methods: liquid-assisted grinding, solvent evaporation, and saturation temperature measurements of mixtures. These methods were used to screen 30 molecules, predicted by a network-based link prediction algorithm (described in Cryst. Growth Des. (6), 3428-3437) as potential coformers for the target molecule praziquantel. The solvent evaporation method presented more drawbacks than advantages, liquid-assisted grinding emerged as the most successful and the quickest, while saturation temperature measurements provided equally good results in a slower route yielding additional solubility information relevant for future screenings, single-crystal growth, and cocrystal production processes. Seventeen cocrystals were found, with 14 showing stability and 12 structures resolved.
药物共晶体因其对物理化学性质的影响及其在分离技术中的作用而备受关注,特别是对于手性分子。新型共晶体的检测是一项挑战,需要可靠的筛选方法。由于存在许多结晶机制不同的技术,其效率取决于所研究的共形成物。表征这些方法的最重要参数包括:(a)可筛选的共形成物比例;(b)共形成物成功率;(c)每个成功的共形成物产生多种共晶体的能力;(d)新稳定相的鉴定;以及(e)实验便利性。基于这些参数,我们比较并量化了三种方法的性能:液相助磨法、溶剂蒸发法和混合物饱和温度测量法。这些方法用于筛选30种分子,这些分子由基于网络的链接预测算法(在《晶体生长与设计》(6),3428 - 3437中描述)预测为目标分子吡喹酮的潜在共形成物。溶剂蒸发法缺点多于优点,液相助磨法是最成功且最快的方法,而饱和温度测量法虽然过程较慢,但能提供同样好的结果,且能产生与未来筛选、单晶生长和共晶体生产过程相关的额外溶解度信息。共发现了17种共晶体,其中14种具有稳定性,12种晶体结构得到解析。