Evaluation of characteristic deuterium distributions of ephedrines and methamphetamines by NMR spectroscopy for drug profiling

Article abstract of DOI:10.1021/ac701639j

We have established a method for quantitative analysis of the deuterium contents (D/H) at the phenyl, methine, benzyl, N-methyl and methyl groups of l-ephedrine/HCl, d-pseudoephedrine/HCl and methamphetamine/HCl by ²H NMR spectroscopy. Comparison of the 5 position-specific D/H values of l-ephedrine/HCl and d-pseudoephedrine/HCl prepared by three methods (chemical synthesis, semichemical synthesis, and biosynthesis) showed that chemically synthesized ephedrines and semi-synthetic ephedrines have highly specific distributions of deuterium at the methine position and at the benzyl position, compared with the other positions. The classification of several methamphetamine samples seized in Japan in terms of the D/H values at these two positions clearly showed that the methamphetamine samples had been synthesized from ephedrines extracted from Ephedra plants or semisynthetic ephedrines but not from synthetic ephedrine. This isotope ratio analysis method should be useful to trace the origins of seized methamphetamine in Southeast Asia.

Full text of DOI:10.1021/ac701639j

Anal. Chem. 2008, 80, 1176-1181
Evaluation of Characteristic Deuterium
Distributions of Ephedrines and
Methamphetamines by NMR Spectroscopy for Drug
Profiling
†
‡
‡
†
†
Teruki Matsumoto, Yasuteru Urano, Yukiko Makino, Ruri Kikura-Hanajiri, Nobuo Kawahara,
†
,‡
Yukihiro Goda, and Tetsuo Nagano*
National Institute of Health Sciences, 1-18-1 Kamiyoga, Setagaya-ku Tokyo 158-8501, Japan, and Graduate School of
Pharmaceutical Sciences, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan
We have established a method for quantitative analysis
of the deuterium contents (D/H) at the phenyl, methine,
benzyl, N-methyl and methyl groups of l-ephedrine/HCl,
d-pseudoephedrine/HCl and methamphetamine/HCl by
2
H NMR spectroscopy. Comparison of the 5 position-
specific D/H values of l-ephedrine/HCl and d-pseu-
doephedrine/HCl prepared by three methods (chemical
synthesis, semichemical synthesis, and biosynthesis)
showed that chemically synthesized ephedrines and semi-
synthetic ephedrines have highly specific distributions of
Figure 1. Three routes for methamphetamine manufacture from
l-ephedrine or d-pseudoephedrine.
deuterium at the methine position and at the benzyl
position, compared with the other positions. The clas-
sification of several methamphetamine samples seized in
Japan in terms of the D/H values at these two positions
clearly showed that the methamphetamine samples had
been synthesized from ephedrines extracted from Ephe-
dra plants or semisynthetic ephedrines but not from
synthetic ephedrine. This isotope ratio analysis method
should be useful to trace the origins of seized metham-
phetamine in Southeast Asia.
profiling has been used to enable the identification of the synthetic
2-5
route for methamphetamine manufacture from ephedrines. For
example, route-specific impurities can be detected by HPLC and
GC/MS. Chiral analysis of seized methamphetamine may also be
useful; for example, if methamphetamine is racemic, P-2-P may
have been used as a precursor. By the analysis with HPLC and
GC/MS, it is impossible to get the differences among ephedrines
produced by different methods. Stable isotope ratio mass spec-
trometry (IR-MS) has recently attached much interest for drug
Methamphetamine, a stimulant drug that activates adrenergic
nerve in the brain, is controlled by the Stimulant Control Law in
Japan. The precursor of methamphetamine is mainly l-ephedrine
or d-pseudoephedrine from Ephedra plants, which are used as a
profiling. Natural isotope analyses of MDMA,6
-11
heroin,
12-16
(2) Qi, Y.; Evans, I. D.; McCluskey, A. Forensic Sci. Int. 2006, 164 (2-3), 201-
210.
(
(
3) Dayrit, F. M.; Dumlao, M. C. Forensic Sci. Int. 2004, 144 (1), 29-36.
4) Makino, Y.; Urano, Y.; Nagano, T. J. Chromatogr., A. 2002, 947 (1), 151-
154.
1
therapeutic drug in Southeast Asia, including Japan. There are
two major synthetic approaches for the preparation of metham-
phetamine: one is synthesis from 1-phenyl-2-propanone (P-2-P),
which is a precursor for the Leuckart method or reductive
amination, and the other is synthesis from l-ephedrine or d-
pseudoephedrine via Birch reduction, the Nagai method, or Emde
method (Figure 1). To prevent the production of illicit metham-
phetamine, it is important to control, monitor, and evaluate the
precursors of seized methamphetamine. Impurity analysis for drug
(
(
5) Lambrechts, M.; Rasmussen, K. E. Bull Narc. 1984, 36 (1), 47-57.
6) Carter, J. F.; Sleeman, R.; Hill, J. C.; Idoine, F.; Titterton, E. L. Sci. Justice
2
005, 45 (3), 141-144.
(7) Palhol, F.; Lamoureux, C.; Chabrillat, M.; Naulet, N. Anal. Chim. Acta 2004,
10, 1-8.
8) Palhol, F.; Lamoureux, C.; Naulet, N. Anal. Bioanal. Chem. 2003, 376 (4),
86-490.
(9) Phillips, S. A.; Doyle, S.; Philp, L.; Coleman, M. Sci. Justice 2003, 43 (3),
53-160.
10) Carter, J. F.; Titterton, E. L.; Murray, M.; Sleeman, R. Analyst 2002, 127
6), 830-833.
5
(
4
1
(
(
*
To whom correspondence should be addressed. Phone +81-3-5841-4850.
(11) Billault I.; Courant F.; Pasquereau L.; Derrien S.; Robins R.J.; Naulet N.
Fax +81-3-5841-4855. E-mail: tlong@mol.f.u-tokyo.ac.jp.
Anal. Chim. Acta 2007, 593, 20-29.
(12) Zhang, D.; Sun, W.; Yuan, Z.; Ju, H.; Shi, X.; Wang, C. Eur. J. Mass Spectrom.
2005, 11 (3), 277-286.
†
National Institute of Health Sciences.
‡
Graduate School of Pharmaceutical Sciences, The University of Tokyo.
(1) Ecstasy and Amphetamines: Global Survey 2003; United Nations: New York,
(13) Ehleringer, J. R.; Cooper, D. A.; Lott, M. J.; Cook, C. S. Forensic Sci. Int.
2
003.
1999, 106 (1), 27-35.
1176 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008
10.1021/ac701639j CCC: $40.75 © 2008 American Chemical Society
Published on Web 01/15/2008

Figure 2. 2H NMR spectra of l-ephedrine & d-pseudoephedrine using TFA as a shift reagent.
cocaine,13,17 cannabis,18,19 and methamphetamine20 by IR-MS
have been reported. The analysis of site-specific deuterium
content by H NMR is also potentially useful for drug profiling.
Armellin et al. have reported that the position-specific deuterium
content of methamphetamine synthesized from P-2-P depends
on the synthetic method.23 However, the most popular metham-
phetamine synthetic routes employ l-ephedrine or d-pseudoephe-
chemicals were as follows: semichemical synthetic d-pseudoephe-
drine from Emmellen (India, sample a) and BASF (Germany,
sample b); semichemical synthetic l-ephedrine from Emmellen
(India, sample c) and BASF (Germany, sample d); synthetic
l-ephedrine from Daiichi Fine Chemical Co. Ltd. (Japan, sample
e); biosynthetic l-ephedrine from Dainippon Sumitomo Pharma
Co. Ltd. (Japan, sample f), galenical market (People’s Republic of
China, sample g) and galenical market (Xinjian Uygur, sample
h). d-Methamphetamine/HCl samples A-H were synthesized
from d-pseudoephedrines and l-ephedrines by the Nagai method
(Figure 1). Sample e was also used to examine the deuterium
abundance in products obtained by the various synthetic methods,
Nagai method (sample E), Emde method (sample I), and Birch
reduction (sample J), as shown in Figure 1, in our laboratory.
Twelve samples of d-methamphetamine crystals had been seized
by law enforcement agencies in Japan, and sample P was
d-methamphetamine/HCl from Dainippon Sumitomo Pharma Co.
Ltd. All chemicals were reagent grade.
2
21,22
drine as a precursor instead of P-2-P. Schmidt et al. employed
2
H
NMR spectroscopy to discriminate the biosynthetic
2
4
pathways of l-ephedrine. However, the position-specific D/H
values of l-ephedrines are likely to be a more powerful tool for
discriminating the sources of ephedrines used as precursors for
preparing illegal methamphetamine/HCl. In this paper, we
focused on methamphetamine prepared from l-ephedrine and
d-pseudoephedrine. We show that the methamphetamine
precursor can be discriminated on the basis of the (D/H) values
2
by using H NMR, and we describe the relationship of the (D/H)
values of the precursor ephedrines and the product methamphet-
amine.
Synthesis. Samples A-H by Nagai method: A mixture of
l-ephedrine/HCl or d-pseudoephedrine/HCl (10.0 g), aqueous
hydroidodic acid (55∼58 wt %, 20 mL), and red phosphorus (1.5
g) was heated at 140 °C for 4.5 h. After cooling, the mixture was
basified (pH 12) with 2 N NaOH(aq) and d-methamphetamine was
EXPERIMENTAL SECTION
Materials and Chemicals. Six l-ephedrine/HCl and two
d-pseudoephedrine/HCl samples were used. The origins of these
extracted with CH
evaporated under vacuum, and HCl-saturated Et
After filtration, the crystals were dried under vacuum to give
d-methamphetamine/HCl in 65% yield.
Sample I by Emde method: Step 1, to a solution of l-ephedrine/
HCl (9.3 g) in benzene (50 mL) was added dropwise a solution of
2
Cl
2
. The organic phase was dried (Na
2 4
SO ) and
(
(
(
(
(
(
(
(
(
(
(
14) Besacier, F.; Guilluy, R.; Brazier, J. L.; Chaudron-Thozet, H.; Girard, J.;
Lamotte, A. J. Forensic Sci. 1997, 42 (3), 429-433.
15) Dautraix, S.; Guilluy, R.; Chaudron-Thozet, H.; Brazier, J. L.; Lamotte, A. J.
Chromatogr., A 1996, 756 (1-2), 203-210.
16) Desage, M.; Guilluy, R.; Brazier, J. L.; Chaudron, H.; Girard, J.; Cherpin,
H.; Jumeau, J. Anal. Chim. Acta 1991, 247, 249-254.
17) Ehleringer, J. R.; Casale, J. F.; Lott, M. J.; Ford, V. L. Nature 2000, 408,
2
O was added.
3
11-312.
18) Denton, T. M.; Schmidt, S.; Critchley, C.; Stewart, G. R. Aust. J. Plant Physiol.
001, 28, 1005-1012.
2
SOCl (20 mL), and the mixture was kept at 60 °C for 3 h. After
2
cooling, the precipitate was collected by filtration to afford
chloropseudoephedrine/HCl in 94% yield. Step 2, to a solution of
chloropseudoephedrine/HCl in pure water (20 mL) was added
19) Liu, J. H.; Lin, W.-F.; Fitzgerald, M. P.; Saxena, S. C.; Shieh, Y. N. J. Forensic
Sci. 1979, 24 (4), 814-816.
20) Kurashima, N.; Makino, Y.; Sekita, S.; Urano, Y.; Nagano, T. Anal. Chem.
2004, 76 (14), 4233-4236.
Pd-BaSO
temperature under H
4
(6 g). The mixture was stirred for 4 h at ambient
. After filtration, the solution was worked
21) Armellin, S.; Brenna, E.; Fronza, G.; Fuganti, C.; Pinciroli, M.; Serra, S.
Analyst 2004, 129 (2), 130-133.
22) Hays, P. A.; Remaud, G. S.; Jamin, E.; Martin, Y. L. J. Forensic Sci. 2000,
2
up as described under the Nagai method to give d-methamphet-
amine/HCl in 75% yield.
Sample J by Birch reduction: to liquid ammonia on a dry ice-
acetone bath (-78 °C) was added dropwise an anhydrous THF
solution of l-ephedrine free base (1.8 g), and the mixture was
4
5 (3), 552-562.
23) Armellin, S.; Brenna, E.; Frigoli, S.; Fronza, G.; Fuganti, C.; Mussida, D.
Anal. Chem. 2006, 78 (9), 3113-3117.
24) Schmidt, H.-L.; Roland A.; Werner R. A.; Eisenreich W. Phytochem. Rev.
2003, 2, 61-85.
Analytical Chemistry, Vol. 80, No. 4, February 15, 2008 1177

Table 1. D/H Values of the d-Pseudoephedrine and l-Ephedrine (ppm)a
sample
phenyl
methine
benzyl
N-methyl
methyl
a (semichemical synthesis)
b (semichemical synthesis)
c (semichemical synthesis)
d (semichemical synthesis)
e (chemical synthesis)
f (biosynthesis)
158.4 ( 3.2
141.6 ( 1.6
150.5 ( 8.6
144.3 ( 7.0
158.6 ( 2.8
138.0 ( 1.7
130.6 ( 2.9
128.7 ( 6.2
26.7 ( 5.4
33.3 ( 4.0
26.9 ( 10.0
13.7 ( 3.8
192.1 ( 21.9
53.0 ( 8.4
52.9 ( 11.2
45.6 ( 6.6
641.1 ( 29.3
818.5 ( 10.7
665.0 ( 44.9
641.5 ( 23.3
54.9 ( 27.2
118.2 ( 13.0
124.1 ( 6.9
120.0 ( 15.9
121.1 ( 2.4
140.7 ( 1.4
140.6 ( 2.4
132.5 ( 2.8
139.0 ( 2.0
99.8 ( 1.5
99.1 ( 2.1
98.8 ( 1.4
118.6 ( 2.2
110.8 ( 0.9
109.3 ( 2.1
105.5 ( 1.9
137.5 ( 1.2
129.5 ( 2.4
128.9 ( 1.7
125.9 ( 2.6
g (biosynthesis)
h (biosynthesis)
a
(
D/H)i ratios and standard deviations measured from eight spectra.
Table 2. D/H Values of Methamphetamines Synthesized by the Nagai Method (ppm)a
sample
phenyl
methine
benzyl
N-methyl
methyl
A
B
C
D
E
F
142.8 ( 2.3
142.3 ( 1.5
159.3 ( 3.9
141.1 ( 1.5
148.3 ( 1.4
136.8 ( 1.1
132.0 ( 1.7
130.8 ( 2.0
44.7 ( 6.1
64.2 ( 3.9
27.1 ( 7.7
62.6 ( 0.6
250.7 ( 10.3
70.9 ( 2.4
61.2 ( 3.4
63.3 ( 3.0
458.6 ( 4.4
617.8 ( 6.0
460.1 ( 13.2
492.5 ( 7.7
20.6 ( 4.5
72.3 ( 4.3
66.8 ( 3.6
62.4 ( 2.6
137.2 ( 2.1
164.1 ( 1.5
151.3 ( 4.5
154.1 ( 2.7
106.6 ( 1.3
86.5 ( 0.5
79.9 ( 1.1
81.2 ( 0.6
121.9 ( 0.9
107.3 ( 1.6
120.5 ( 1.7
107.2 ( 1.2
136.2 ( 1.2
134.9 ( 1.2
122.5 ( 0.8
126.2 ( 1.2
G
H
a
The samples A-H were prepared from the ephedrines a-h, respectively, in Table 1. (D/H)i ratios and standard deviations measured from
eight spectra.
Table 3. D/H Values of Methamphetamines Synthesized from Sample e by Various Methods (ppm)a
sample
phenyl
methine
benzyl
N-methyl
methyl
e
158.6 ( 2.8
148.3 ( 1.4
144.4 ( 1.4
144.2 ( 2.4
192.1 ( 21.9
250.7 ( 10.3
233.6 ( 3.0
199.5 ( 8.3
54.9 ( 27.2
20.6 ( 4.5
72.6 ( 0.8
41.6 ( 5.7
139.0 ( 2.0
106.6 ( 1.3
112.1 ( 0.8
105.8 ( 3.2
137.5 ( 1.2
136.2 ( 1.2
136.3 ( 1.2
136.1 ( 2.9
E (Nagai method)
I (Emde method)
J (Birch reduction)
a
(
D/H)i ratios and standard deviations measured from eight spectra.
2
2
stirred. Then, small pieces of lithium metal (300 mg) were added
to the mixture during 5 min with stirring, retaining the blue
color of the solution. To the reaction mixture was added ethanol
H NMR Spectroscopy. H NMR spectra were recorded on
an ECA-600 spectrometer (JEOL Ltd, Japan) at 92.13 MHz, using
a 10 mm broad-band tunable probe, under CPD (WALTZ se-
quence) proton decoupling conditions. All samples were run
unlocked at 298 K. At least eight spectra were run for each sample,
collecting 7000 scans and using the following parameters: acquisi-
(
0.6 mL), followed by excess ammonium chloride as a
quenching agent. The mixture was warmed to room temperature,
ammonia was removed by evaporation, and the residue was
dissolved in pure water (20 mL). d-Methamphetamine was
1 1
tion time of 5 times T max, repetition time of 6 times T max,
extracted with CH
described under the Nagai method to give d-methamphetamine/
HCl in 50% yield.
Sample Preparation. An amount of 400 mg of l-ephedrine/
HCl or 1 g of d-pseudoephedrine/HCl and 0.45 g of maleic acid
2
Cl
2
. The organic phase was worked up as
1843.11-2764.26 Hz spectral width, and 4K or 8K memory size,
for adjustment of acquisition time and repetition time. Backward
linear prediction was used to reconstruct the first few data points
of the FID, replacing points which were corrupted by ringdown
and Fourier-transformed with a line broadening of 2 Hz. The peak
areas involving partially overlapped signals were determined
through the deconvolution routine of the JEOL NMR software
Delta using a Lorentzian line shape. The signals of d-pseudoephe-
drine/HCl and l-ephedrine/HCl, obtained from an analysis of the
proton spectrum taken under the same conditions as the deute-
rium spectrum, were as follows: d-pseudoephedrine/HCl, 7.19-
7.11 (5H, m, phenyl), 4.46 (1H, d, J ) 9.7 Hz, benzyl), 3.23 (1H,
dq, J ) 6.9 Hz, 9.7 Hz, methine), 2.59 (3H, s, N-methyl), 0.84 (3H,
d, J ) 6.9 Hz, methyl); l-ephedrine/HCl, 7.20-7.10 (5H, m,
phenyl), 5.00 (1H, d, J ) 3.1 Hz, benzyl), 3.25 (1H, dq, J ) 3.1 Hz,
6.9 Hz, methine), 2.63 (3H, s, N-methyl), 0.91 (3H, d, J ) 6.9 Hz,
methyl). The signal assignment of the spectrum of d-metham-
(Wako Pure Chem. Ind. Ltd, Japan) as an internal standard were
weighed into a 10 mm NMR sample tube (Shigemi Inc., Japan)
and dissolved in 2.0 mL of deuterium-depleted water (Isotec Inc.).
Then 1.0 or 0.4 mL of trifluoroacetic acid was added as a shift
reagent. The signal of the benzyl position of l-ephedrine/HCl or
1
d-pseudoephedrine/HCl was observed near that of H
2
O by H
2
NMR. Though the H NMR signal overlaps with the broad signal
of DHO, it was possible to obtain quantitative values by using
TFA as a shift reagent (Figure 2). In addition, 0.4-0.9 g of
methamphetamine/HCl and 0.45 g maleic acid mixed in a
10 mm NMR tube were dissolved in 3.0 mL of deuterium-depleted
water.
1178 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008

2
2
.48 (3H, s, N-methyl), 1.03 (3H, d, J ) 6.6 Hz, methyl). In the H
1
NMR experiment, the benzyl signal detected at 2.65 ppm by H
NMR in d-methamphetamine/HCl overlapped the signal based
on the N-methyl group. Quantitative analysis of the spectra was
done by curve fitting using the Lorentz function. The standard
deviations of the quantitative values of phenyl, N-methyl, and
methyl signals in the examined samples were less than 9 ppm
{
signal-to-noise ratio (S/N) > 30} as shown in Tables 1 -3. The
standard deviations of the methine and benzyl signals of some
samples were not so good due to the low deuterium abundance,
but there was little quantitative overlap. The unlocked state during
data acquisition in NMR experiments was found to have little effect
on the quantitative analysis under the condition used.
Figure 3. Distribution of D/H value of the d-pseudoephedrine and
l-ephedrine.
Determination of the Site-Specific Isotope Ratio. The (D/
i
H) value for each spectrum was calculated from eq 1 using maleic
acid as an internal reference. The isotopic ratio of maleic acid,
having exchangeable protons within the molecule, (D/H)MA, was
precisely calibrated on the VSMOW scale (155.8 ppm) via the
dimethylsulfoxide (146.3 ppm).
phetamine/HCl was similarly assigned as follows (δ, ppm): 7.17-
7
.06 (5H, m, phenyl), 3.29 (1H, m, methine), 2.83 (1H, dd, J ) 6.2
Hz, 13.7 Hz, benzyl), 2.64 (1H, dd, J ) 7.9 Hz, 13.7 Hz, benzyl),
Figure 4. (a) Scatter plot of D/H values of the methine signal (X axis) and the benzyl signal (Y axis) of precursor ephedrines, (b) carbon and
nitrogen isotope ratio of various ephedrines by IR-MS quoted from ref 26.
Analytical Chemistry, Vol. 80, No. 4, February 15, 2008 1179

(
D/H) ) P m M S (D/H) /P m M S
(1)
i
ref ref
s
i
ref
i
s
ref ref
where P
position i and in the reference. S
the signal. M , m , Mref and mref are, respectively, the molecular
i
and Pref are the stoichiometric numbers of hydrogens in
i
and Sref are the surface areas of
s
s
weight and mass of the sample and the reference used.
RESULTS AND DISCUSSION
2H Distribution of d-Pseudoephedrine/HCl and l-Ephe-
drine/HCl. The results of the quantitative analyses of d-pseu-
2
doephedrines/HCl and l-ephedrines/HCl by H NMR spectros-
copy are shown in Table 1 and Figure 3. All samples showed
comparable D/H values at the phenyl, N-methyl, and methyl
positions, while characteristic distributions of deuterium were
observed at the methine and benzyl positions. Sample e showed
the highest D/H value at the methine position and the smallest
at the benzyl position, and samples a-d showed higher values at
the benzyl position. Butzenlechner et al. reported that benzalde-
hyde from natural sources and benzaldehyde prepared by catalytic
Figure 5. Scatter plot of D/H values of the methine signal (X axis)
and the benzyl signal (Y axis) of (a) synthesized methamphetamines
and (b) seized and commercial methamphetamines.
2
oxidation of toluene showed mean δ( H) values of -125 ( 14 ‰
2
5
and +777 ( 20 ‰, respectively. It is presumed that the higher
values of samples a-d at the benzyl position may be due to
synthesis via synthetic benzaldehyde from toluene as the precur-
sor. Deuterium abundance of ephedrines differed greatly at the
methine and benzyl positions from the other positions. The
preparation method of ephedrines appears to affect the deuterium
contents at both positions. In a scatter plot of absolute values
2H Distribution of d-Methamphetamine Prepared from
l-Ephedrine/HCI by Various Synthetic Methods. Table 3
shows the D/H values at each position of the precursor l-
ephedrine e and the d-methamphetamines/HCl synthesized from
l-ephedrine/HCl by three synthetic methods: Nagai method (E),
Emde method (I), and Birch reduction (J). The cluster of sample
E, I, and J in Figure 5a corresponds to cluster I in Figure 4a but
not clusters II and III in Figure 4a. Thus, the variability of
deuterium distribution by synthetic method of methamphetamine
did not show large characteristic differences as compared with
the variability of D/H values of precursors shown in Figure 4a.
(
Figure 4a), three kinds of ephedrines could be distinctly
separated into the individual groups. We previously reported that
two groups of semichemical synthetic ephedrines could be
1
5
13
discriminated on the basis of the values of δ N and δ C using
2
6
IR-MS. Five samples (a, b, e-g) in Figure 4a are the same as
those in Figure 4b, which was reported in ref 26. In Figure 4b, it
is not clear if semichemical synthetic pseudoephedrine (sample
b) and biosynthetic ephedrine (sample f) belong to the same group
or not. However, sample b (semichemical synthetic) and sample
f (biosynthetic) were clearly discriminated by the D/H values at
the benzyl and methine positions, as shown in Figure 4a. Thus,
2
H Distribution in Seized and Commercial Methamphet-
amines. Figure 5b shows a comprehensive scatter plot of
deuterium abundance at the methine and benzyl positions for the
12 seized methamphetamine samples together with a commercial
methamphetamine (P). From this plot, the studied methamphet-
amines could be grouped into two clusters. Cluster II includes
methamphetamines synthesized from biosynthetic ephedrines.
Cluster III includes methamphetamines which originated from
precursors prepared by a semichemical synthetic process. The
classification of several methamphetamine samples seized in Japan
in terms of the D/H values at these two positions clearly showed
that the methamphetamine samples had been synthesized from
ephedrines extracted from Ephedra plants or semisynthetic
ephedrines but not from synthetic ephedrine. Thus, our findings
are expected to be helpful to monitor and identify the sources of
ephedrines used for the manufacture of methamphetamine and
thereby to assist enforcement authorities in the strategy of
precursor control.
2
quantitative analysis by H NMR appears to be more effective for
discrimination of preparation methods of ephedrines.
2
H Distribution of d-Methamphetamine/HCl Prepared
From l-Ephedrine/HCl or d-Pseudoephedrine/HCl. The
results of the quantitative analyses of methamphetamines prepared
2
from three kinds of ephedrine by H NMR spectroscopy are
reported in Table 2. All methamphetamines were synthesized by
the Nagai method.
Figure 5a shows a scatter plot of deuterium abundance at the
methine and benzyl positions in the synthesized methamphet-
amines. The distribution is clearly similar to that of the corre-
sponding precursor shown in Figure 4a. Thus, the characteristics
of deuterium abundance of the precursor were well-correlated with
those of the product methamphetamine. It is therefore possible
to obtain chemical information on the synthetic method of
ephedrines used as a precursor, as we had expected.
CONCLUSIONS
We have established a method for quantitative analysis of the
position-specific deuterium contents (D/H) of ephedrines/HCl and
2
2
methamphetamine/HCl by H NMR spectroscopy. H NMR
spectroscopy is suitable for position-specific isotope detection, in
contrast to IR-MS which gives only overall abundance. We
compared five position-specific D/H values of l-ephedrine/HCl and
(
25) Butzenlechner, M.; Rossmann, A.; Schmidt, H.-L. J. Agric Food Chem. 1989,
7, 410-412.
26) Makino, Y.; Urano, Y.; Nagano, T. Bull. Narc. 2005, 57, 63-78.
3
(
1180 Analytical Chemistry, Vol. 80, No. 4, February 15, 2008

d-pseudoephedrine/HCl prepared by three methods (chemical
synthesis, semichemical synthesis, and biosynthesis). Semichemi-
cal synthetic ephedrines characteristically showed a specific high
distribution of deuterium at the benzyl position. As there is a large
difference at the benzyl and methine positions compared with the
other three positions, the D/H values at the benzyl and methine
positions would be useful as indicators to classify the origin of
ephedrines. The difference between biosynthetic and semichemi-
cal synthetic ephedrines was easier to discern from the combina-
tion of the D/H values at the benzyl and methine positions than
from the combination of δ N and δ C obtained by IR-MS. The
five position-specific D/H values of l-ephedrine used as a precursor
were well-correlated with those for methamphetamine synthesized
from it. The several seized and commercial methamphetamines
could be classified into two groups by plotting the D/H values of
methine versus benzyl position. The D/H values at these two
positions of several methamphetamine samples seized in Japan
suggested that the samples had been synthesized from ephedrines
extracted from Ephedra plants or semisynthetic ephedrines but
not from synthetic ephedrines. This isotope ratio analysis method
should be useful to trace the origins of seized methamphetamine
prepared from ephedrines.
ACKNOWLEDGMENT
The authors are grateful to JEOL engineers for technical
assistance. The present work was supported by a research grant
from the Ministry of Health, Labor and Welfare, Japan.
1
5
13
Received for review August 2, 2007. Accepted November
27, 2007.
AC701639J
Analytical Chemistry, Vol. 80, No. 4, February 15, 2008 1181