丝带
造粒
微晶纤维素
硬脂酸镁
压实
材料科学
滚压成形
复合材料
颗粒(地质)
机械工程
工程类
纤维素
化学
剂型
化学工程
色谱法
作者
Seo‐Young Park,Shaun Galbraith,Yanling Liu,HaeWoo Lee,Bumjoon Cha,Zhuangrong Huang,Thomas O’Connor,Sau Lee,Seongkyu Yoon
标识
DOI:10.1016/j.powtec.2018.02.042
摘要
Roller compaction is a dry granulation process for manufacturing solid drug formulations. A flowing powder is continuously fed into the counter-rotating rolls and compacted into dense ribbons. These ribbons are subsequently broken into granules by milling. In this study, an integrated flowsheet model of the dry granulation process was developed for unit operations of powder feeding, roller compaction and milling with gSOLIDS (PSE Enterprise, UK) and tested against experimental data. The effect of roll force and roll speed on ribbon density and granule size was assessed with the flowsheet model and validated with experimental data. The experimentation was conducted with microcrystalline cellulose with magnesium stearate (1%, w/w) in a Fitzpatrick CCS220/M3B with serrated rollers. Experimental design on roll force, roll speed and different screen sizes and mill speeds were conducted to investigate their impacts on the quality attributes of granule size, ribbon density, and roll gap. Parameters in each unit were estimated and the sensitivity of the model outputs to the material property based parameter values was also investigated. The predictions of the ribbon density and roll gap at various operating conditions were assessed. The roll force has a more significant effect on ribbon density than the roll speed, while the roll speed has a more significant effect on the roll gap. Higher roll force and narrower roll gap lead to dense ribbon. The ribbon density is much less sensitive to the roll speed compared with the roll force. The roll gap has a considerable influence on ribbon quality. The developed flowsheet model for the continuous dry granulation process can be directly used for process optimization, risk assessment and control strategy determination.
科研通智能强力驱动
Strongly Powered by AbleSci AI