本研究採用超臨界快速膨脹法來製備微米級龍腦(或稱2-茨醇)粒子，所使用的超臨界流體為CO2，首先使用逐一因子實驗，探討孔口噴嘴大小、萃取溫度、萃取壓力、預膨脹溫度，以及膨脹室溫度等因素對粒子生成之影響分別作討論，由結果發現上述這些變因對粒子大小皆有不同程度的影響；表面型態與粒徑分布方面，藉由光學顯微鏡得知隨粒子越來越小，其表面型態由近似胚胎狀變為近似圓球狀，粒徑分布亦越趨狹窄。此外由XRD、TGA與DSC分析結果，經由RESS處理前後之龍腦粒子，具有相同的結晶結構與相似的熔融溫度，並無變質或分解現象發生，惟處理過後的龍腦微粒結晶度變差，對熱敏感性略為提升。而由實驗結果顯示，本研究已成功的將龍腦粒子微小化，從原始平均粒徑303.2 μm，縮小到在3.6-10.6 μm的範圍。 同時藉由Box-Behnken實驗設計法來規劃實驗，並以應答曲面法找出一個適當的二階非線性模式，瞭解粒徑大小和製備參數之間的關係，並經由變異數分析來決定各參數的顯著性以及曲面的適切性，吾人可藉由該模式來尋求以RESS方法製粒之近似最適操作條件。 This study adopted the rapid expansion of supercritical solution (RESS) process using CO2 to prepare borneol (or 2-camphanol) micro-particles. In the one-factor-at-a-time experiments, the effects of orifice nozzle size, extraction temperature, extraction pressure, pre-expansion temperature, and post-expansion temperature on particle size were individually discussed. Our results showed that the particle size was affected by these factor to various degrees. As the size of the particles decreased, the particles size distribution became narrower and embryo-shaped crystals were transformed into quasi-spherical ones through optical microscope (OM) observations. In addition, based on the analysis of X-ray diffraction (XRD), thermo-gravimetric analysis (TGA), and differential scanning calorimetry (DSC), it was found the RESS-processed and unprocessed borneol particles had the same crystal structure and similar melting temperatures. Also, no significant property changes and degradation for the RESS-processed borneol particles were observed. However, the degree of crystallinity of the processed borneol was lower than that of the unprocessed borneol. Also, the processed particles were found to be more heat sensitive than unprocessed ones. Micronization of borneol particle was successfully demonstrated in our study. By employing the RESS technique, the size of the borneol particles (avg. 303.2 μm) was reduced to the range of 3.6-10.6 μm. Box-Behnken design was used in the response surface methodology (RSM) to find a second-order nonlinear model relating particle size to process parameters considered. Significant process parameters and predicted response surface were determined and checked by analysis of variance (ANOVA). Optimum conditions can be approximately identified by the fitted model.
