Abstract N-benzylphenethylamines derivatives, such as NBOMe and NBOH series, are potent hallucinogen drugs that are usually sold in the illicit market as blotter papers containing “legal” LSD alternatives. The identification of these drugs is mainly performed by liquid or gas chromatography coupled with mass spectrometry, but there is a lack of a rapid screening methods to identify samples containing or not drugs of these series. For this purpose, this work offers a fast and non-destructive method applying a handheld NIR spectrometer for discrimination of drugs absorbed in blotter papers using PLS-DA and SIMCA. The method was developed in a two-stage approach: Model A for identification of samples containing or not drugs and model B for identification of samples containing NBOMe or NBOH drugs. PLS-DA models have provided efficiency rates higher 97% in both training and validation phases. Robustness of model B was evaluated with a random permutation of samples among training and validation phases, given mean efficiency rates higher than 89%. SIMCA models presented equivalent efficiency for model A, but lower efficiency for model B comparing to PLS-DA results. However, SIMCA showed a higher efficiency to deal with samples containing drugs not included in the training phase. Results support use of this method for in loco forensic analysis and as a screening method prior the chromatographic and mass spectrometry analysis.

中文翻译：

---

使用手持式 NIR 光谱仪和化学计量学方法鉴定吸墨​​纸中的 NBOMe 和 NBOH

摘要 N-苄基苯乙胺衍生物，如 NBOMe 和 NBOH 系列，是强效的致幻剂药物，通常在非法市场上作为含有“合法” LSD 替代品的吸墨纸出售。这些药物的鉴别主要采用液相色谱或气相色谱联用质谱法进行，但缺乏快速筛选方法来鉴别样品中是否含有该系列药物。为此，这项工作提供了一种快速且无损的方法，应用手持式 NIR 光谱仪使用 PLS-DA 和 SIMCA 来区分吸纸中吸收的药物。该方法采用两阶段方法开发：模型 A 用于识别含有或不含药物的样品，模型 B 用于识别含有 NBOMe 或 NBOH 药物的样品。PLS-DA 模型在训练和验证阶段都提供了高达 97% 的效率。模型 B 的稳健性是通过在训练和验证阶段之间随机排列样本来评估的，平均效率高于 89%。SIMCA 模型与模型 A 的效率相当，但与 PLS-DA 结果相比，模型 B 的效率较低。然而，SIMCA 在处理包含未包含在训练阶段的药物的样本方面表现出更高的效率。结果支持将该方法用于现场法医分析和作为色谱和质谱分析之前的筛选方法。SIMCA 模型与模型 A 的效率相当，但与 PLS-DA 结果相比，模型 B 的效率较低。然而，SIMCA 在处理包含未包含在训练阶段的药物的样本方面表现出更高的效率。结果支持将该方法用于现场法医分析和作为色谱和质谱分析之前的筛选方法。SIMCA 模型与模型 A 的效率相当，但与 PLS-DA 结果相比，模型 B 的效率较低。然而，SIMCA 在处理包含未包含在训练阶段的药物的样本方面表现出更高的效率。结果支持将该方法用于现场法医分析和作为色谱和质谱分析之前的筛选方法。