Abstract Raman spectroscopy and multivariate data evaluation were used to verify the chemical nature and the content of the active pharmaceutical ingredients (API) (acetylsalicylic acid, ibuprofen and paracetamol) in painkillers. A class-modelling approach (SIMCA) of spectral data was used to verify that the correct API was indeed present in the pharmaceutical preparation and to prove the selectivity of the developed method towards other commercial APIs; PLS regression was used for the verification of the API amount. The root mean square error (RMSE) of the PLS models for the quantitation of the APIs were 11.3 % for paracetamol, 13.2 % for acetylsalicylic acid, and 6.2 % for ibuprofen in drug preparations containing the API at levels between 7.1 % and 92.3 %. This level of accuracy appears to be acceptable for a rapid screening method, which makes it fit-for-purpose for deployment in customs and forensic laboratories involved in the surveillance of the legal and illicit drug market.

中文翻译：

---

用于质量控制和检测不合格止痛药的拉曼光谱

摘要 采用拉曼光谱和多元数据评估方法验证止痛药中活性药物成分（API）（乙酰水杨酸、布洛芬和扑热息痛）的化学性质和含量。光谱数据的类建模方法 (SIMCA) 用于验证药物制剂中确实存在正确的 API，并证明所开发方法对其他商业 API 的选择性；PLS 回归用于 API 量的验证。在 API 含量介于 7.1% 和 92.3% 之间的药物制剂中，用于 API 定量的 PLS 模型的均方根误差 (RMSE) 对扑热息痛为 11.3%，乙酰水杨酸为 13.2%，布洛芬为 6.2%。这种准确度水平对于快速筛选方法来说似乎是可以接受的，