In silico methods applied in forensic science: simulated NMR spectra of new amphetamine- and cathinone-derived psychoactive substances

New Psychoactive Substances (NSP) are synthetic compounds which mainly intend to mimic recreative effects of controlled substances and avoid their identification. There are many properties related to these drugs that are still unknown. Forensic Scientists have a great demand for information that are important for both law enforcement as for human rights guarantees. In silico methods present great potential to help to characterize these substances. The main goal of this work was use theoretical methods to foresee NMR spectra for pairs of similar amphetamines and cathinones. Structures were optimized by B3LYP/6-31G**. NMR spectra were generated by different in silico free softwares: ORCA 4.0.1 (B3LYP and PBE0, 6-31G**) and nmrdb.org. MestReLab, a commercial software, was also used. Available experimental data were compared by Principal Component Analysis (PCA). Results showed that both free and commercial softwares had good predictions. However, B3LYP/631G** ORCA calculations were more precise despite more time consuming.