Cyclic Ion Mobility of Isomeric New Psychoactive Substances Employing Characteristic Arrival Time Distribution Profiles and Adduct Separation

Analysis of new psychoactive substances (NPS), which is essential for toxicological and forensic reasons, can be made complicated by the presence of isomers. Ion mobility has been used as a standalone technique or coupled to mass spectrometry to detect and identify NPS. However, isomer separation has so far chiefly relied on chromatography. Here we report on the determination of isomeric ratios using cyclic ion mobility-mass spectrometry without any chromatographic separation. Isomers were distinguished by mobility separation of lithium adducts. Alternatively, we used arrival time distribution (ATD) profiles that were characteristic of individual isomers and were acquired for protonated molecules or fragment ions. Both approaches provided comparable results. Calculations were used to determine the structures and collision cross sections of both protonated and lithiated isomers that accurately characterized their ion mobility properties. The applicability of ATD profiles to isomer differentiation was demonstrated using direct infusion and flow injection analysis with electrospray of solutions, as well as desorption electrospray of solid samples. Data processing was performed by applying multiple linear regression to the ATD profiles. Using the proposed ATD profile-based approach, the relationships between the determined and given content of isomers showed good linearity with coefficients of determination typically greater than 0.99. Flow injection analysis using an autosampler allowed us to rapidly determine isomeric ratios in a sample containing two isomeric pairs with a minor isomer of 10% (determined 9.3% of 3-MMC and 11.0% of 3-FMC in a mixture with buphedrone and 4-FMC). The proposed approach is not only useful for NPS, but also may be applicable to small isomeric molecules analyzed by ion mobility when complete separation of isomers is not achieved.


S.1 Instrumental settings
Table S1.Cyclic TWIMS parameters for single pass separation for all isomeric pairs and calibration Table S2.Separation time in a multi-pass experiment CCS calibration Table S3.Calibration mixture  Table S4.Relative Gibbs energies of protonated 3-MMC and buphedrone.Table S5.Relative Gibbs energies of 3-MMC and buphedrone Li + adducts.

CCS calibration
The mixture of small molecules (Table S3) at 2 µg/mL each in acetonitrile/water (1/1, v/v, 0.1 % of formic acid) was infused directly (5µL/min) into the ESI source using the default setting.Data were acquired in 1 TOF push per bin and were collected for single and multi-pass experiments.DriftScope 2.9 software (Waters Corp., Wilmslow, UK) detected peak maxima.Bin numbers were converted to the millisecond scale.S4.

Figure S4 .
Figure S4.Determination of 3-MMC in the mixture with buphedrone.Multiple linear regression (MLR) applied to ATD profiles of protonated molecules, single pass separation.

Figure S2 .
Figure S2.M06-2X/6-31+G(d,p) optimized structures of 3-MMC and buphedrone [M+Li] + ions with calculated CCS in nitrogen.C, H, N, O color coding is as in Figure S2.Hydrogen bonds are shown with ochre double-headed arrows.All bond lengths are in Ångstrøms.

Figure S9 .
Figure S9.+ESI mass spectra after 1 pass separation: A) methedrone; B) BDB.The inset shows the zoomed detail of the signal of the fragment ions (m/z 135).

Table S1 .
Cyclic TWIMS parameters for single pass separation for all isomeric pairs and calibration

Table S2 .
Separation time in a multi-pass experiment Arrival times did not include injection time.Calibrant arrival times (t a ) were obtained from one and two pass experiments (98 and 196 cm path length).
a Description as in Table