Drug Product Science and Technology, Bristol-Myers Squibb, 1 Squibb Drive, New Brunswick, NJ 08901, USA.
Drug Product Science and Technology, Bristol-Myers Squibb, 1 Squibb Drive, New Brunswick, NJ 08901, USA.
Int J Pharm. 2012 Apr 15;426(1-2):116-131. doi: 10.1016/j.ijpharm.2012.01.032. Epub 2012 Jan 25.
Instrumented roll technology on Alexanderwerk WP120 roller compactor was developed and utilized successfully for the measurement of normal stress on ribbon during the process. The effects of process parameters such as roll speed (4-12 rpm), feed screw speed (19-53 rpm), and hydraulic roll pressure (40-70 bar) on normal stress and ribbon density were studied using placebo and active pre-blends. The placebo blend consisted of 1:1 ratio of microcrystalline cellulose PH102 and anhydrous lactose with sodium croscarmellose, colloidal silicon dioxide, and magnesium stearate. The active pre-blends were prepared using various combinations of one active ingredient (3-17%, w/w) and lubricant (0.1-0.9%, w/w) levels with remaining excipients same as placebo. Three force transducers (load cells) were installed linearly along the width of the roll, equidistant from each other with one transducer located in the center. Normal stress values recorded by side sensors and were lower than normal stress values recorded by middle sensor and showed greater variability than middle sensor. Normal stress was found to be directly proportional to hydraulic pressure and inversely to screw to roll speed ratio. For active pre-blends, normal stress was also a function of compressibility. For placebo pre-blends, ribbon density increased as normal stress increased. For active pre-blends, in addition to normal stress, ribbon density was also a function of gap. Models developed using placebo were found to predict ribbon densities of active blends with good accuracy and the prediction error decreased as the drug concentration of active blend decreased. Effective angle of internal friction and compressibility properties of active pre blend may be used as key indicators for predicting ribbon densities of active blend using placebo ribbon density model. Feasibility of on-line prediction of ribbon density during roller compaction was demonstrated using porosity-pressure data of pre-blend and normal stress measurements. Effect of vacuum to de-aerate pre blend prior to entering the nip zone was studied. Varying levels of vacuum for de-aeration of placebo pre blend did not affect the normal stress values. However, turning off vacuum completely caused an increase in normal stress with subsequent decrease in gap. Use of instrumented roll demonstrated potential to reduce the number of DOE runs by enhancing fundamental understanding of relationship between normal stress on ribbon and process parameters.
亚历山大 WP120 滚轮压片机上的仪器化压辊技术已成功开发并应用于测量过程中带状物料的法向应力。使用安慰剂和活性预混物研究了工艺参数（如辊速（4-12 rpm）、送料螺杆速度（19-53 rpm）和液压辊压力（40-70 bar））对法向应力和带状物料密度的影响。安慰剂混合物由微晶纤维素 PH102 和无水乳糖以 1:1 的比例组成，还含有交联羧甲基纤维素钠、胶体二氧化硅和硬脂酸镁。活性预混物是使用各种活性成分（3-17%，w/w）和润滑剂（0.1-0.9%，w/w）的组合制备的，其余辅料与安慰剂相同。三个力传感器（称重传感器）沿压辊宽度线性安装，彼此等距，一个传感器位于中心。边部传感器记录的法向应力值低于中部传感器记录的法向应力值，且波动性大于中部传感器。结果表明，法向应力与液压压力成正比，与螺杆与压辊速度比成反比。对于活性预混物，法向应力也是可压缩性的函数。对于安慰剂预混物，随着法向应力的增加，带状物料的密度也随之增加。对于活性预混物，除了法向应力外，带状物料的密度也是间隙的函数。使用安慰剂建立的模型发现可以很好地预测活性混合物的带状物料密度，并且随着活性混合物中药物浓度的降低，预测误差减小。活性预混物的有效内摩擦角和可压缩性特性可以用作使用安慰剂带状物料密度模型预测活性混合物带状物料密度的关键指标。通过使用预混物的孔隙率-压力数据和法向应力测量，证明了在滚轮压缩过程中在线预测带状物料密度的可行性。研究了在进入压区之前对预混物进行真空脱气的效果。不同程度的真空脱气对安慰剂预混物的法向应力值没有影响。然而，完全关闭真空会导致法向应力增加，随后间隙减小。仪器化压辊的使用展示了通过增强对带状物料上法向应力与工艺参数之间关系的基本理解，减少 DOE 运行次数的潜力。
