Received: 29 December 2020 Revised: 1 February 2021 Accepted: 6 February 2021

DOI: 10.1002/rcm.9067

RESEARCH ARTICLE

Application of a simultaneous screening method for the

detection of new psychoactive substances in various matrix
samples using liquid chromatography/electrospray ionization
tandem mass spectrometry and liquid chromatography/
quadrupole time-of-flight mass spectrometry

Nam Sook Kim | Na Young Lim | Hwan Seong Choi | Ji Hyun Lee |
Hyungil Kim | Sun Young Baek

Center for Advanced Analysis, National

Institute of Food and Drug Safety Evaluation, Rationale: Recently, new psychoactive substances (NPS) have emerged as a public
Ministry of Food and Drug Safety, Osong
health risk. Particularly, their chemical structures are modified to avoid detection.
Health Technology Administration Complex,
187 Osongsaengmyeong2-ro, Osongeup, Synthetic NPS with effects similar to those of illegal drugs have been recently
Heungdeok-gu, Cheongju-si,
detected and synthesized worldwide, including MDMB-FUBINACA and APINAC,
Chungcheongbuk-do, 363-700, Republic of
Korea making it essential to rapidly and accurately detect NPS.
Methods: Fourteen NPS with similar structures were selected and their structures
Correspondence
1 13
S. Y. Baek, Center for Advanced Analysis, identified using H and C NMR spectroscopy. Additionally, we proposed the
National Institute of Food and Drug Safety
fragmentation pattern of each compound using liquid chromatography/quadrupole
Evaluation, Ministry of Food and Drug Safety,
Osong Health Technology Administration time-of-flight mass spectrometry (LC/QTOF-MS). A simultaneous analytical method
Complex, 187 Osongsaengmyeong2-ro,
using liquid chromatography/electrospray ionization tandem mass spectrometry
Osongeup, Heungdeok-gu, Cheongju-si,
Chungcheongbuk-do 363-700, Republic of (LC/ESI-MS/MS) was also developed and applied to real samples to detect the
Korea.
14 NPS. The method was validated based on the specificity, linearity, limit of
Email: [email protected]
detection (LOD), limit of quantification (LOQ), precision, accuracy, matrix effect, and
Funding information
stability according to international validation guidelines.
National Institute of Food and Drug Safety
Evaluation, the Ministry of Food and Drug Results: The established method was used to screen 65 different matrix samples
Safety, Grant/Award Number:
using LC/ESI-MS/MS. By comparing the calculated product ion ratios with those of
18181MFDS421
standards, 2C-B in one of the real samples and 5F-MDMB-PICA in 20 samples were
identified. For re-confirmation of detected compounds, the fragmentation pattern of
each compound was compared with that of each standard using LC/QTOF-MS.
Conclusions: In this study, LC/QTOF-MS data were used to elucidate the structures
and fragmentation patterns of 14 NPS. A simultaneous method was developed using
LC/ESI-MS/MS, which was applied to 65 real samples. The presented method and
results can assist in ensuring the safety of public health from illegal adulteration.

Nam Sook Kim and Na Young Lim contributed equally to this work.

Rapid Commun Mass Spectrom. 2021;35:e9067. wileyonlinelibrary.com/journal/rcm © 2021 John Wiley & Sons Ltd 1 of 11
https://doi.org/10.1002/rcm.9067
2 of 11 KIM ET AL.

1 | I N T RO DU CT I O N 2 | EX PE RI MENT AL

In recent years, the number of new psychoactive substances (NPS) 2.1 | Chemicals and standard solutions
has rapidly increased and is an emerging global concern. NPS are
defined by the European Monitoring Centre for Drugs and Drug The 14 NPS (each purity ≥98%), JWH-193, JWH-007, FUBICA, 5F-
Addiction as substances that do not fall within the purview of MDMB-PICA, DMBA-CHIMINACA, MDMB-FUBINACA, APINAC,
international drug controls but can pose a public health threat. 1
α-PVP, MDPV, 3,4-MD-α-PHP, methylone, 4-chloroethcathinone,
NPS are synthesized by altering the chemical structures of illicit 2C-B, and 25B-NBOH, were provided by the Ministry of Food and
drugs or by discovering drugs with similar effects, such as stimulants Drug Safety (MFDS), South Korea. HPLC-grade acetonitrile (AcN) and
and hallucinogens, to evade existing drug laws.2–4 Among them, methanol were purchased from Merck (Darmstadt, Germany). Formic
synthetic cathinones, which are widespread throughout the illegal acid (FA) was purchased from Thermo Fisher Scientific (North
market, are distributed as alternatives for cocaine, ecstasy, and Waltham, MA, USA). Deionized water (DW, 18.1 mΩ) was prepared
amphetamine.5,6 using a Milli-Q purification system (Millipore, Bedford, MA, USA). To
Synthetic cannabinoids (SCs) are one of the fastest growing groups obtain a concentration of approximately 1000 μg/mL, the 14 standard
among NPS monitored by the Early Warning System (EWS) of the compounds were dissolved in methanol and solutions were prepared
European Drug and Drug Addiction Monitoring Centre (EMCDDA).1 daily by diluting the mixture with methanol to achieve the appropriate
Since it was first detected in herbal mixtures in 2008, the intentionally concentrations for the standard working solutions. The diluted
modified JWH-018 series have been continuously detected.7–9 mixtures were stored at approximately 4 C.
Synthetic cathinones, including methcathinone, which have similar
psychoactive effects to stimulant drugs such as amphetamine,
cocaine, and 3,4-methylamphetamine (MDMA), are the second largest 2.2 | LC/QTOF-MS instrumentation and
group of NPS and more than 130 compounds have been conditions
detected worldwide.1,10–12 In addition, although 2C-B, belonging to
phenethylamine, was first synthesized approximately 50 years LC/QTOF-MS analysis was performed using a high-performance
ago, variations with similar structures,13–17 such as 25B-NBOH, liquid chromatography (HPLC) system coupled to a 6545XT
25C-NBOH, or 25I-NBOMe, have recently been distributed as an AdvanceBio LC/QTOF system equipped with electrospray ionization
alternative to ecstasy. (ESI) Jetstream technology (Agilent Technologies, Waldbronn,
Analyses of NPS, such as synthetic cannabinoids, in food and Germany). An X-Bridge C18 column (150 × 2.0 mm, i.d. 3.5 μm,
biological samples are actively conducted worldwide.18,19 For Waters, MA, USA) maintained at 40 C was used for the analysis. The
example, the New Zealand Medicines and Medical Devices Safety mobile phase was composed of 0.1% FA in DW (A) and AcN (B). The
Authority (Medsafe) warned against the intake of coffee containing gradient program was as follows: initial (20% B)–3.0 min, 20% B; 3.0–
phenethylamine in 2020.20 In addition, the Swiss Agency for 13.0 min, 60% B; 13.0–16.0 min, 60% B; 16.0–18.0 min, 100% B;
Therapeutic Products (Swissmedic) announced that 13 compounds, 18.0–21.0 min, 100% B; 21.0–22.0 min, 20% B; and 22.0–25.0 min
including 5F-MDMB-PICA and MMB-CHMINA, were designated as for equilibration. The flow rate was 0.3 mL/min with an injection
NPS in 2019. 21
Detailed structural elucidations or fragmentation volume of 3 μL. The MS parameters of the Jetstream ESI source in the
patterns of each compound are still lacking to confirm the structural positive ion mode were as follows: capillary voltage, 3500 V; gas
similarities of adulterants to illegal drugs.22–24 As synthetic temperature, 300 C; nebulizer, 45 psi gauge; drying gas flow, 10 L/
cannabinoids may be adulterated in real samples, it is necessary to min; fragmentor voltage, 175 V; and skimmer, 65 V. By acquiring MS
continuously develop an optimum simultaneous analytical method spectra, the precursor ions and retention times for each compound
to detect NPS in a variety of food matrices. In this study, in order to were obtained. The MS/MS analysis was conducted in targeted
proactively block illegal health products, we selected 14 NPS MS/MS mode according to various values of collision energy ranging
considering some modified structures that have similar effects to from 5 to 40 eV. The MS/MS parameters of the Jetstream ESI source
1
illegal drugs. The selected compounds were characterized by H and in the positive ion mode were as follows: capillary voltage, 3500 V;
13
C NMR spectroscopy (Table S1, supporting information) as it was gas temperature, 300 C; nebulizer, 45 psi gauge; drying gas flow,
difficult to purchase the expensive commercial standards. In addition, 10 L/min; and fragmentor voltage, 175 V; skimmer, 65 V; charge
we present the LC/QTOF-MS fragmentation patterns of each state(Z), 1; delta retention time, 1; isolation width, narrow (1.3 m/z).
compound. Furthermore, we developed and validated a simultaneous Practically, the collision energy was optimized as JWH-193, 30 eV;
screening method for 14 NPS using LC/ESI-MS/MS. The validated JWH-007, 30 eV; FUBICA, 20 eV; 5F-MDMB-PICA, 40 eV; DMBA-
methods were applied to screen 65 different matrix samples CHIMINACA, 20 eV; MDMB-FUBINACA, 20 eV; APINAC, 15 eV;
(27 capsules, 13 tablets, 1 powder, 3 pills, and 21 different samples of α-PVP, 20 eV; MDPV, 3,4-MD-α-PHP, 30 eV; methylone, 20 eV;
tea leaves). The results of this study will contribute to the pre- 4-chloroethcathinone, 15 eV; 2C-B, 20 eV; 25B-NBOH, 20 eV. The
emptive prevention of unauthorized compounds adulterated in illegal resolving power of our system ranged from 12,000 to 29,000 in the
health products. mass scan range of m/z 50–1700 using an ESI-L Low concentration
KIM ET AL. 3 of 11

tuning mix solution. Mass accuracy was expressed as mass error (ppm) medium, and high (close to the LOQ, 10, and 20 times the LOQ,
by comparison of the measured and theoretical masses. The exact respectively). The precision and accuracy tests were assessed using
mass for each peak was measured in positive mode using an the relative standard deviation (RSD%) values and the percentage
automated calibration delivery system to inject the calibration recovery (%) of the theoretical concentration of each drug substance
solution with a reference standard of purine (m/z 121.0602) and HP- (n = 3). The recovery tests were performed by accurately adding three
921 (m/z 922.0090). The data were acquired and qualitatively different concentrations (low, medium, and high) of the standard
analyzed using the LC/QTOF-MS Mass Hunter workstation software mixture to the solid and liquid blank samples and comparing the peak
(version B.09.00; Agilent Technologies, Waldbronn, Germany). areas with the standard areas. The recovery was determined as the
average recovery (%) and standard deviation (SD) of each drug
substance (n = 3). During each validation procedure, the samples
2.3 | LC/MS/MS instrumentation and conditions spiked with the standard mixture were extracted using the sample
preparation conditions. The stability of the 14 drug substances was
A 1200 series HPLC system (Agilent Technologies) coupled to a 5500 determined by storing the samples for 24 h and 48 h in an
Q-Trap mass spectrometer (AB SCIEX, Darmstadt, Germany) was autosampler at 8 C and storing three different concentrations (low,
used. A Capcell pak MG II column (2.0 × 100 mm, 3 μm) maintained at medium, and high) at approximately 8 C for 6 h (n = 3). Matrix effects

40 C was used for the chromatographic separation of the 14 analytes. (%) were also evaluated by comparing two slopes obtained from the
Deionized water containing 0.1% FA (A) and AcN containing 0.1% FA calibration curves of the reference standards dissolved in methanol
(B) were used as mobile phases. The injection volume was 2.0 μL at a and of standards in spiked blank samples.27
flow rate of 0.3 mL/min. The following gradient program was applied:
initial (10% B)–4.0 min, 40% B; 4.0–6.0 min, 70% B; 6.0–8.0 min, 95%
B; 8.0–10.0 min, 95% B; 10.0–12.0 min, 10% B; 12.0–16.0 min for 2.5 | Sample preparation and application of the
equilibration. Ionization was performed in the positive ESI mode method to real samples
under the following conditions: curtain gas: 25 psi; collision gas: 5 psi;
ion source gases 1 and 2: 50 psi; source temperature: 500 C; ion To apply the established method to real samples containing various
spray voltage: 5500 V (positive). Analyst software (version 1.5.2; AB matrix types, real samples were investigated and selected that
SCIEX) was used to collect the LC/ESI-MS/MS data. To optimize the exaggerate the efficacy of treatment for obesity, depressant, and
multiple reaction monitoring (MRM) transitions of the LC/ESI-MS/MS anxiety. A total of 65 real samples (including 27 capsules, 13 tablets,
method, all the standards were directly injected into the MS 1 powder, 3 pills, and 21 different tea leaves) were purchased via online
equipment so that the diagnostic precursor and product ions from markets or seized from the criminal investigation public office of South
each full scan mass spectrum could be selected using the ESI source Korea. For sample preparation, the tablets, pills, and leaves were ground
of the MS/MS system. Ionization of the 14 analytes was performed to a powder using a model MM 301 mixer mill (Retsch, Haan, Germany).
by protonation of [M + H]+ in the positive ESI mode. The contents were removed from the capsules. Each homogenized
sample (1 g) was added to a 50 mL volumetric flask and extracted with
30 mL 70% methanol for 30 min at room temperature using an
2.4 | Method validation by LC/MS/MS ultrasonic bath. The extracted sample solution was made up to 50 mL
with 70% methanol, and vigorously mixed. Finally, the sample solution
To validate the proposed method, the specificity, linearity, limit of was filtered using a 0.22 μm polytetrafluoroethylene filter (Millipore,
detection (LOD), limit of quantitation (LOQ), method detection limit Milford, USA) and then injected into the LC/ESI-MS/MS apparatus. The
(MDL), method quantitation limit (MQL), precision, accuracy, recovery, sample preparation conditions were based on literature reports.27,28
and stability were evaluated according to the International
Conference on Harmonization (ICH) of Technical Requirements for
Registration of Pharmaceuticals for Human Use guidelines (ICH, 3 | RESULTS AND DISCUSSION
25
2005). The specificity was confirmed by comparing a blank matrix
sample with a sample spiked with the 14 drug substances under the 3.1 | Fragmentation of compounds using
LC/MS/MS conditions. The LOD and LOQ were established for the LC/QTOF-MS
solid and liquid samples and were determined as the concentrations at
which the signal-to-noise ratios were approximately 3:1 and 10:1, 3.1.1 | JWH moiety drugs
respectively. The MDL and MQL were estimated from the LOD and
LOQ values using 1 g of sample with 50 mL of the final extraction The fragmentation patterns of the JWH moiety substances (including
solvent.26 The linearity was evaluated using six different JWH-193, JWH-007, FUBICA, 5F-MDMB-PICA, DMBA-CHMINACA,
concentrations (1, 2, 4, 8, 10, and 20 times the LOQ) of the 14 drug MDMB-FUBINACA, and APINAC) were studied by analyzing their
substances (n = 3). The intra- and inter-day precision and accuracy MS/MS spectra. The proposed fragmentation pathway is indicated in
tests were conducted using three different concentrations: low, Figure 1.
4 of 11 KIM ET AL.

F I G U R E 1 Proposed fragmentation
pathway of JWH moiety drugs (JWH-193,
JWH-007, FUBICA, 5F-MDMB-PICA, DMBA-
CHMINACA, MDMB-FUBINACA, APINAC)

The structures of the JWH series are theoretically similar to each the molecular formula of protonated C21H23FN3O2+ (calcd.
other, as they consist of a naphthalene bonded to an indole ring via a 368.1769), with the product ions appearing at m/z 252.0819
carbonyl bridging group. In fact, similar fragmentation patterns were (C16H11FNO+) and 109.0448 (C7H6F+). For 5F-MDMB-PICA, the ion
observed, with the differences caused by the substituent groups peak, corresponding to the molecular formula of protonated
present at different positions of the indole or naphthalene C21H30FN2O3+ (calcd. 377.2235), appeared at m/z 377.2236, with
derivatives. The ion peak of JWH-193 appeared at m/z 399.2075, the product ions detected at m/z 232.1132 (C14H15FNO+), 144.0444
and corresponded to the molecular formula of protonated (C9H6NO+), and 116.0495 (C8H6N+). The ion peaks of DMBA-
C26H27N2O2+ (calcd. 399.2067). The presence of the JWH moiety CHMINACA, MDMB-FUBINACA, and APINAC appeared at m/z
was indicated by the product ions which appeared at m/z 169.0648 372.2294, 398.1871, and 367.2405, respectively, corresponding to
(C12H9O+), 141.0699 (C11H9+), and 114.0913 (C6H12NO+). The ion the molecular formulas of protonated C21H30N3O3+ (calcd.
peak of JWH-007, corresponding to the molecular formula of 372.2282), C22H25FN3O3+ (calcd. 398.1874), and C23H31N2O2+
protonated C25H26NO+ (calcd. 356.2009), was detected at m/z (calcd. 367.2380), respectively. The product ions of DMBA-
356.2021 and its product ions appeared at m/z 228.1383 CHMINACA appeared at m/z 326.2227 (C20H28N3O+), 241.1335
+ +
(C15H18NO ), 155.0491 (C11H7O ), and 127.0542 (C10H7+). The ion (C15H17N2O+), and 145.0396 (C8H5N2O+), while those of MDMB-
peak of FUBICA was detected at m/z 368.1766 and corresponded to FUBINACA appeared at m/z 338.1663 (C20H21FN3O+), 253.0772
KIM ET AL. 5 of 11

(C15H10FN2O+), and 109.0448 (C7H6F+). The product ions of C17H24NO3+ (calcd. 290.1751). The product ions appeared at m/z
APINAC appeared at m/z 215.1179 (C13H15N2O+), 135.1168 219.1016 (C13H15O3+), 140.1434 (C9H18N+), and 135.0441
(C10H15+), and 93.0699 (C7H10+). The mass errors between the +
(C8H7O2 ). The structure of 3,4-MD-α-PHP contains an additional
theoretical and experimental masses of precursor ions for the JWH methyl group in the butanylpyrrolidine functionality compared to
moiety drugs were between −0.81 and 6.81 ppm (Table 1). MDPV, as indicated by the m/z 126.1277 peak of 3,4-MD-a-PHP
compared to the m/z 140.1434 peak of MDPV. In addition,
the common fragment ion (m/z 135.0441) between 3,4-MD-α-PHP
3.1.2 | Methcathinone moiety drugs and MDPV was confirmed to originate from the methylenedioxy
group.
Methcathinones are well-known synthetic cathinone derivatives. Among The ion peak of methylone appeared at m/z 208.0961 and
them, we selected five methcathinone moiety drugs which were corresponded to the molecular formula of protonated C11H13NO3+
29–31
recently abused worldwide ; methylone, MDPV, 3,4-MD-α-PHP, (calcd. 208.0968), with product ions at m/z 190.0499 (C10H8NO3+),
α-PVP, and 4-chloroethcatinone. As shown in the compound structures, 160.0757 (C10H10NO+), and 132.0808 (C9H10N+), which indicated the
the selected compounds containing functional groups at the benzyl ring presence of a methcathinone moiety. The ion peak of
or an aliphatic agonist at the amine functionality were introduced. Based 4-chloroethcathinone, corresponding to the molecular formula of
on the fragmentation patterns acquired by LC/QTOF-MS, we confirmed protonated C11H15ClNO3+ (calcd. 212.0837), was detected at m/z
the structural similarities among the five compounds. The proposed 212.0829. The product ions appeared at m/z 194.0367
fragmentation pathway is indicated in Figure 2. (C10H9ClNO+), 159.0679 (C10H9NO•+), and 144.0444 (C9H6NO+),
The ion peaks of α-PVP and MDPV were detected at m/z indicating the presence of a methcathinone moiety. The mass errors
232.1695 and 276.1585, respectively, corresponding to the molecular between the theoretical and experimental masses of precursor
formulas of protonated C15H22NO+ (calcd. 260.0281) and C16H21NO3 ions for methcathinone moiety drugs were between −3.77 and
(calcd. 276.1594), respectively. The product ions for α-PVP appeared − 0.43 ppm (Table 1).
at m/z 161.0961 (C11H13O+), 126.1277 (C8H16N+), and 91.0542
(C7H7+), while those of MDPV appeared at m/z 175.0754 (C11H11O2+),
135.0441 (C8H7O2+), and 126.1277 (C8H16N+). By comparing the 3.1.3 | Phenethylamine moiety drugs
α-PVP m/z 91.0542 peak with the MDPV m/z 135.0441 peak, it was
confirmed that the MDPV structure contained a methylenedioxy group Phenethylamine moiety drugs including 2C-B and 25B-NBOH were
which was not present in α-PVP. In addition, the common fragment ion analyzed through their MS/MS spectra, and the proposed
(m/z 126.1277) between α-PVP and MDPV was confirmed to originate fragmentation pathway is indicated in Figure 3.
from the butanylpyrrolidine functionality (Figure 2). The ion peaks of 2C-B and 25B-NBOH appeared at m/z
The ion peak of 3,4-MD-α-PHP was detected at m/z 290.1747 260.0279 and 366.0702, respectively, corresponding to the
and corresponded to the molecular formula of protonated molecular formulas of protonated C10H15BrNO2+ (calcd. 260.0281)

TABLE 1 Precursor ions and mass errors of the 14 new psychoactive substances using LC/QTOF-MS.

Precursor ion

Classification Compounds Theoretical m/z [M + H]+ Measured m/z [M + H]+ Mass errora (ppm)
JWH moiety drugs JWH-193 399.2067 399.2075 2.00
JWH-007 356.2009 356.2021 3.37
FUBICA 368.1769 368.1766 −0.81
5F-MDMB-PICA 377.2235 377.2236 0.27
DMBA-CHIMINACA 372.2282 372.2294 3.22
MDMB-FUBINACA 398.1874 398.1871 −0.75
APINAC 367.2380 367.2405 6.81
Methcathinone moiety drugs α-PVP 232.1696 232.1695 −0.43
MDPV 276.1594 276.1585 −3.26
3,4-MD-α-PHP 290.1751 290.1747 −1.38
Methylone 208.0968 208.0961 −3.36
4-Chloroethcathinone 212.0837 212.0829 −3.77
Phenethylamine moiety drugs 2C-B 260.0281 260.0279 −0.77
25B-NBOH 366.0699 366.0702 0.82
 
measured − theory 6:
a
Mass error (ppm) = theory × 10
6 of 11 KIM ET AL.

F I G U R E 2 Proposed
fragmentation pathway of
methcathinone moiety drugs (α-PVP,
MDPV, 3,4-MD-α-PHP, methylone,
4-chloroethcathinone)

FIGURE 3 Proposed fragmentation pathway of phenethylamine moiety drugs (2C-B, 25B-NBOH)

KIM ET AL. 7 of 11

and C17H21BrNO3+ (calcd. 366.0699), respectively. The product (C10H12BrO2+), and 107.0491 (C7H7O+), indicating the presence of
ions of 2C-B appeared at m/z 243.0015 (C10H12BrO2+), 227.9655 a phenethylamine moiety similar to that in 2C-B. The mass errors
(C8H7BrNO2+), and 212.9546 (C8H6BrO2+), while those of 25B- of 2C-B and 25B-NBOH were −0.77 and 0.82, respectively
NBOH appeared at m/z 260.0281 (C10H15BrNO2+), 243.0015 (Table 1).

TABLE 2 MRM conditions for the LC/ESI-MS/MS analysis of the 14 new psychoactive substances

Compound Ion mode Precursor ion (m/z) Product ion (m/z) DP (V) CE (eV) CXP (V)
Methylone + 208.040 160.0 96 23 14
190.0 96 15 14
132.0 96 37 12
4-Chloroethcathinone + 212.037 194.0 91 17 16
159.0 91 25 12
144.0 91 37 12
α-PVP + 232.100 90.9 121 33 14
126.1 121 33 10
161.0 121 23 14
2C-B + 260.017 243.0 76 17 20
228.0 76 29 18
212.9 76 43 26
MDPV + 276.090 126.1 156 35 12
135.0 156 35 12
175.1 156 29 10
3,4-MD-α-PHP + 290.089 140.1 161 35 14
135.0 161 33 10
219.2 161 25 12
25B-NBOH + 365.987 242.9 166 29 18
260.0 166 21 14
107.0 166 31 14
JWH-193 + 399.119 169.0 56 29 16
114.0 56 33 14
141.0 56 61 12
FUBICA + 368.33 252.0 36 27 20
109.1 36 49 14
5F-MDMB-PICA + 377.131 232.0 96 21 20
144.0 96 57 14
116.0 96 73 8
DMBA-CHIMINACA + 372.211 241.1 86 31 16
326.2 86 21 14
144.9 86 49 12
MDMB-FUBINACA + 398.148 253.0 86 33 20
338.0 86 21 20
109.0 86 51 16
JWH-007 + 356.129 155.0 106 33 12
127.0 106 51 20
228.1 106 31 18
APINAC + 367.112 135.0 111 27 12
93.0 111 75 16
215.0 111 36 26
8 of 11 KIM ET AL.

3.2 | LC/ESI-MS/MS method validation information). The LODs and LOQs for the solid blank sample ranged
from 0.01 to 0.50 ng/mL and 0.03 to 1.50 ng/mL, respectively, and
The optimized declustering potentials (DP) and collision energies the LODs and LOQs for the liquid blank sample ranged from 0.01 to
(CE) for the MRM transitions are listed in Table 2. Among the product 0.40 ng/mL and 0.03 to 1.20 ng/mL, respectively (Table 3 and
ions, the highest-abundance ion was used for quantitation, while the Table S3, supporting information). The MDLs and MQLs for the solid
other ions were used for qualification. blank sample ranged from 0.5 to 25.0 ng/g and 1.5 to 75.0 ng/g,
As listed in Table 2 and shown in Figure 4, the proposed method respectively, and the MDLs and MQLs for the liquid blank sample
provides good separation between the characteristic MRM ions of ranged from 0.5 to 20.0 ng/g and 1.5 to 60.0 ng/g, respectively
each analyte without interference from the matrices. The specificity (Table 3 and Table S4, supporting information). The R2 values were
of the proposed LC/ESI-MS/MS method was confirmed by greater than 0.99 for the calibration curves of the six concentrations
comparison of solid and liquid blank samples with the solid and liquid and showed good linearity (Table 3 and Table S5, supporting
samples spiked with the 14 drug substances. The stability of the information). The matrix effects for the solid and liquid blank samples
standard mixture was evaluated and determined at 0, 24, and 48 h. were 83.22–113.99% and 85.05–114.43%, respectively (Table 3 and
The standard mixture containing 14 NPS was stable throughout the Table S6, supporting information). Therefore, this method is
stability test. Accordingly, the stability at low, medium, and high sufficiently accurate to analyze real samples in different forms
concentrations was below 14.91% of the RSD after storing the without significant matrix effects. The accuracies of the
mixtures for 48 h in a refrigerator (Table 3 and Table S2, supporting measurements ranged from 86.62% to 115.44% for intra-day

F I G U R E 4 Total ion chromatograms of A, the blank sample and B, the mixture containing the standard mixture containing the 14 new
psychoactive substances analyzed using LC/ESI-MS/MS in the positive mode (1. Methylone, 2. 4-chloroethcathinone, 3. α-PVP, 4. 2C-B,
5. MDPV, 6. 3,4-MD-α-PHP, 7. 25B-NBOH, 8. JWH-193, 9. FUBICA, 10. 5F-MDMB-PICA, 11. DMBA-CHMINACA, 12. MDMB-FUBINACA, 13.
JWH-007, 14. APINAC)
KIM ET AL. 9 of 11

measurements and 91.44% to 110.30% for inter-day measurements 3.3 | Application to real samples
(Table 3 and Table S7, supporting information). The precision of the
method, expressed as the RSD, was under 13.15% for the intra-day To evaluate the effectiveness of the established method, 65 real
measurements and under 12.56% for the inter-day measurements samples were analyzed, including 27 types of capsules, 13 types of
(Table 3 and Table S7, supporting information). Recoveries of 80.38– tablets, 1 powder, 3 types of pills, and 21 types of tea leaves. These
113.20% (solid) and 87.84–113.49% (liquids) were achieved with an samples were purchased online or were seized from the criminal
SD under 15.0% for each dosage form (Table 3 and Table S8, investigation public office of South Korea. All the samples were
supporting information). The LC/ESI-MS/MS method proposed from prepared according to the optimized preparation conditions, analyzed
the comprehensive effectiveness test results satisfies the using the validated LC/ESI-MS/MS method, and the measurements
internationally accepted standards. Thus, this is the first verified using LC/QTOF-MS for higher accuracy.
demonstration of the simultaneous and efficient separation of 14 NPS In order to confirm the detected unknown compounds in the
using LC/ESI-MS/MS. Notably, this method is highly sensitive for the applied samples, the criteria of LC/QTOF-MS were used for
detection of 14 potential NPS in real samples. identification as follows. The precursor ion must be detected, and
then all selected fragment ions must be detected in LC/QTOF-MS.
Additionally, the mass error obtained by comparing the measured m/z
value with the theoretical m/z value must be within 10 ppm.32
T A B L E 3 Validation results for the 14 new psychoactive Furthermore, the unknown compounds were identified through MRM
substances (n = 3)
transitions and ion ratios between precursor and product ions using
Performance characteristic Solid sample Liquid sample LC/MS/MS. The criteria used for identification by LC/MS/MS were
LOD 0.01–0.50 ng/mL 0.01–0.40 ng/mL as follows. The precursor ion must be detected, and then all selected

LOQ 0.03–1.50 ng/mL 0.03–1.20 ng/mL product ions must be detected in LC/MS/MS. Additionally, the
identities of the target compounds must be confirmed by comparing
MDL 0.5–25.0 ng/g 0.5–20.0 ng/g
the calculated product ion ratios with the standard ion ratios
MQL 1.5–75.0 ng/g 1.5–60.0 ng/g
established by LC/MS/MS; the results have to be within 20% of the
Matrix effect 83.22–113.99% 85.05–114.43%
standard ion ratios.33
Recovery 80.38–113.20% 87.84–113.49%
The adulterants were confirmed to be 2C-B and 5F-MDMB-PICA
Stability RSD ≤14.91%
(Table 4). As shown in Figure 5, the fragmentation patterns of
Intra-day Inter-day the detected 2C-B (m/z 116.0491, 144.0439, 232.1136) and 5F-
Accuracy 86.62 115.44% 91.44110.30% MDMF-PICA (m/z 212.9543, 227.9776, 243.0008) were determined
Precision RSD ≤13.15% RSD ≤ 12.56% by LC/QTOF-MS. As this demonstrates, the proposed method can be
a
For more detail, of each drug substance, see Tables S2S8 (supporting reliable and accurate for the determination of 14 NPS potentially
information). adulterated in illegal products.

TABLE 4 Application of the LC/ESI-MS/MS method to various real sample matrices

Sample information Detection results

Product iona

Classification Matrix type Number of evaluations Number of detections Detected compound m/z Ion ratio
b
Online purchased sample Capsule 27 n.d - - -
Tablet 2 n.d - - -
Powder 1 n.d - - -
Seized sample Pill 3 n.d - - -
Tablet 11 1 2C-B 243.0 -
228.0 ≤ 29.9%(±20%)
212.9 ≤ 12.8% (±20%)
Tea leaves 21 20 5F-MDMB-PICA 232.0 -
144.0 ≤ 17.8% (±20%)
116.0 ≤ 5.3% (±20%)
a
underline: quantitative ion/no underline: qualitative ion.
b
not detected.
10 of 11 KIM ET AL.

F I G U R E 5 LC/TOF-MS/MS spectra
and fragmentation patterns of A, 5F-
MDMB-PICA and B, 2C-B

4 | C O N CL U S I O N S PE ER RE VIEW
The peer review history for this article is available at https://publons.
In this study, 14 NPS standards of potentially adulterated compounds com/publon/10.1002/rcm.9067.
were selected and the fragmentation patterns of the compounds were
OR CID
elucidated using LC/QTOF-MS. In addition, a simultaneous analytical
method for the detection of 14 NPS was developed and validated Nam Sook Kim https://orcid.org/0000-0002-5247-3411
Na Young Lim https://orcid.org/0000-0001-7233-4083
using LC/ESI-MS/MS. Furthermore, the optimized LC/ESI-MS/MS
method was applied to 65 real samples, including herbal products.
2C-B and 5F-MDMB-PICA were detected in an illegal drug and in tea RE FE RE NCE S
1. European Monitoring Centre for Drugs and Drug Addiction. European
leaves, respectively. The suspected compounds were determined by
drug report (2018): Trends and developments. Publications Office of
comparing the retention times and precursor/product ion ratios, and the European Union, Luxembourg, 2018. http://www.emcdda.
re-confirming the fragmentation patterns with those of standards europa.eu/publications/edr/trends-developments/2018_en.
through LC/ESI-MS/MS and LC/QTOF-MS. So far, there has been Accessed November 25, 2020.
much research on NPS for biological matrix samples, but there is still a 2. United Nations Office of Drugs and Crime. Global synthetic drugs
assessment, amphetamine-type stimulants and new psychoactive
lack of analytical research of NPS for commercial products that
substances, Vienna, 2017. https://www.unodc.org/documents/
consumers can access easily such as dietary supplements, tea leaves, scientific/Global_Synthetic_Drugs_Assessment_2017.pdf. Accessed
etc. In order to preemptively prevent the illegal distribution of NPS, November 25, 2020.
these established analytical methods, which can be applied to 3. World Drug Report 2018 (United Nations publication, Sales No. E.18.
XI.9). https://www.unodc.org/wdr2018/prelaunch/WDR18_Booklet_
commercial products, are necessary. These illegal NPS adulterants can
1_EXSUM.pdf. Accessed November 25, 2020.
pose a potential threat to public health, especially because of their 4. European Monitoring Centre for Drugs and Drug Addiction 2017,
potential undesirable side effects and drug interaction risks. High-risk drug use and new psychoactive substances, EMCDDA
Therefore, we believe that these results will not only assist in the Rapid Communication, Publications Office of the European Union,
Luxembourg. https://www.unodc.org/documents/scientific/Global_
protection of public health safety from illegal NPS, but also contribute
Drugs_Assessment_2017.pdf. Accessed November 25, 2020.
to the development of other similar structural compound analysis. 5. Sutherland R, Bruno R, Peacock A, et al. New psychoactive
substances: Purchasing and supply patterns in Australia. Hum
ACKNOWLEDGEMEN T Psychopharmacol. 2017;32(3):e2577.
6. European Monitoring Centre for Drugs and Drug Addiction, Europol,
This study was supported by National Institute of Food and Drug
EU drug markets report: a strategic analysis. http://www.emcdda.
Safety Evaluation, the Ministry of Food and Drug Safety in 2020 europa.eu/publications/joint-publications/drug-markets. Accessed
(Grant No. 18181MFDS421). November 25, 2020.
KIM ET AL. 11 of 11

7. Harris CR, Brown A. Synthetic cannabinoid intoxication: A case series 23. Lee JH, Park HN, Leem T, et al. Identification of new synthetic
and review. J Emerg Med. 2012;44(2):360-366. cannabinoid analogue APINAC (adamantan-1-yl 1-pentyl-1H-
8. Hermanns-Clausen M, Kneisel S, Szabo B, Auwarter V. Acute toxicity indazole-3-carboxylate) with other synthetic cannabinoid MDMB(N)-
due to the confirmed consumption of synthetic cannabinoids: Clinical Bz-F in illegal products. Forensic Toxicol. 2017;35(1):45-55.
and laboratory findings. Addiction. 2012;108(3):534-544. 24. Banister SD, Longworth M, Kevin R, et al. Pharmacology of valinate
9. Theunissen EL, Hutten NRPW, Mason NL, Toennes SW, and tert-leucinate synthetic cannabinoids 5F-AMBICA, 5F-AMB, 5F-
Kuypers KPC, Ramaekers JG. Neurocognition and subjective ADB, AMB-FUBINACA, MDMB-FUBINACA, MDMB-CHMICA, and
experience following acute doses of the synthetic cannabinoid JWH- their analogues. ACS Chem Nerosci. 2016;7(9):1241-1254. https://doi.
018: Responders versus nonresponders. Cannabis Cannabinoid Res. org/10.1021/acschemneuro.6b00137
2019;4(1):51-61. 25. (ICH) International Conference on Harmonization, Harmonized
10. Kehr J, Ichinose F, Yoshitake S, et al. Mephedrone, compared with Tripartite Guideline, Validation of Analytical Procedures: Text and
MDMA (ecstasy) and amphetamine, rapidly increases both dopamine Methodology Q2(R1). European Medicines Agency (EUMA), London.
and 5-HT levels in nucleus accumbens of awake rats. Br J Pharmacol. 2005.
2011;164(8):1949-1958. 26. Jettanajit A, Nhujak T. Determination of mycotoxins in brown rice
using QuEChERS sample preparation and UHPLC–MS-MS.
11. Bretteville-Jensen AL, Tuv SS, Bilgrei OR, Fjeld B, Bachs L. Synthetic
J Chromatogr Sci. 2016;54(5):720-729.
cannabinoids and cathinones: Prevalence and markets. Forensic Sci
27. Kim NS, Kim J, Lim NJ, Lee JH, Park S, Kang H. Simultaneous
Rev. 2013;25(1–2):7-26.
determination of illegal drug substances in dietary supplements
12. Valente MJ, Guedes de Pinho P, de Lourdes Bastos M, Carvalho F,
for gout and osteoporosis using ultra-performance liquid
Carvalho M. Khat and synthetic cathinones: A review. Arch Toxicol.
chromatography and liquid chromatography-quadrupole-time-of-
2014;88(1):15-45.
flight mass spectrometry. J Pharm Biomed Anal. 2020;179:113003.
13. Shulgin AT, Carter MF. Centrally active phenethylamines.
28. Kim NS, Kim KY, Yoo GJ, et al. Determination of 26 anti-diabetic
Psychopharmacol Commun. 1975;1(1):93-98.
compounds in dietary supplements using a validated UPLC method.
14. Hansen M, Phonekeo K, Paine JS, et al. Synthesis and structure- Food Addit Contam, Part a. 2018;35(3):387-394.
activity relationship of N-benzyl phenethylamines as 5-HT2A/2C 29. Bonano JS, Glennon RA, De Felice LJ, Banks ML, Negus SS. Abuse-
agonists. ACS Chem Nerosci. 2014;5(3):243-249. related and abuse-limiting effects of methcathinone and the synthetic
15. Machado Y, Coelho Neto J, Lordeiro RA, Alves RB, Piccin E. “bath salts” cathinone analogs methylenedioxypyrovalerone (MDPV),
Identification of new NBOH drugs in seized blotter papers: 25B- methylone and mephedrone on intracranial self-stimulation in rats.
NBOH, 25C-NBOH, and 25E-NBOH. Forensic Toxicol. 38(1):203-215. Psychopharmacology (Berl). 2013;231(1):199-207.
https://doi.org/10.1007/s11419-019-00509-7 30. Batisse A, Fortias M, Bourgogne E, Grégoire M, Sec I, Djezzar S. Case
16. Kristofic JJ, Chmiel JD, Jackson GF, et al. Detection of 25C-NBOMe series of 21 synthetic cathinones abuse. J Clin Psychopharmacol.
in three related cases. J Anal Toxicol. 2016;40(6):466-472. https://doi. 2014;34(3):411-413.
org/10.1093/jat/bkw035 31. Oliver CF, Palamar JJ, Salomone A, et al. Synthetic cathinone
17. Coelho Neto J, Andrade AFB, Lordeiro RA, et al. Preventing adulteration of illegal drugs. Psychopharmacology (Berl). 2019;236(3):
misidentification of 25I-NBOH as 2C-I on routine GC–MS analyses. 869-879.
Forensic Toxicol. 2017;35(2):415-420. 32. Anthony WTB, Kenneth SW. Intercomparision study on accurate
18. Odoardi S, Fisichella M, Romolo FS, Strano-Rossi S. High-throughput mass measurement of small molecules in mass spectrometry. J Am
screening for new psychoactive substances (NPS) in whole blood by Soc Mass Spectrom. 2003;14:1086-1098.
DLLME extraction and UHPLC-MS/MS analysis. J Chromatogr B 33. Juan S, Rico P. Joint Fao/WHO Food Standards Programme Codex
Analyt Technol Biomed Life Sci. 2015;1000:57-68. https://doi.org/10. Committee on Residues of Veterinary Drugs in Foods (CX/RVDF
1016/j.jchromb.2015.07.007 12/20/10) December 2011.
19. Ares AM, Fernández P, Regenjo M, Fernández AM, Carro AM,
Lorenzo RA. A fast bioanalytical method based on microextraction by
packed sorbent and UPLC-MS/MS for determining new psychoactive SUPPORTING INF ORMATION
substances in oral fluid. Talanta. 2017;174:454-461. https://doi.org/ Additional supporting information may be found online in the
10.1016/j.talanta.2017.06.022 Supporting Information section at the end of this article.
20. Medsafe: Medsafe is issuing a warning that Elevacity Elevate Smart
Coffee (Version 3) should not be consumed – statement under
section 98 of the Medicines Act 1981. https://www.medsafe.govt. How to cite this article: Kim NS, Lim NY, Choi HS, Lee JH,
nz/safety/Alerts/ElevacityElevateSmartCoffee.asp. Accessed
Kim H, Baek SY. Application of a simultaneous screening
November 30, 2020.
21. Swissmedic: 13 new psychoactive substances added to Narcotics List method for the detection of new psychoactive substances in
2020. https://www.swissmedic.ch/swissmedic/en/home/news/ various matrix samples using liquid chromatography/
mitteilungen/narcotics-list-extended-2019.html. Accessed November electrospray ionization tandem mass spectrometry and liquid
30, 2020.
chromatography/quadrupole time-of-flight mass
22. Lee JH, Park HN, Kim NS, Park S, Lee Y, Kang H. Development of a
specific fragmentation pattern-based quadrupole-Orbitrap™ mass
spectrometry. Rapid Commun Mass Spectrom. 2021;35:e9067.
spectrometry method to screen drugs in illicit products. Sci Justice. https://doi.org/10.1002/rcm.9067
2020;60(1):86-94.