Journal of Analytical Toxicology, Vol. 24, October 2000
GC-MS Analysisof Methamphetamine Impurities:
Reactivity of (+)-or (-)-Chloroephedrine and cis- or
trans- 1,2- Dimethyl-3-phenylaziridine
Veeravan Lekskulchai 1, Karen Carter 2, Alphonse Poklis1, and William Soine2,*
I Department of Pathology, School of Medicine, Medical College of Virginia and 2Department of Medicinal Chemistry,
School of Pharmacy, Medical College of Virginia, Virginia Commonwealth University, Richmond, Virginia 23298
Abstract
S-(+)-Methamphetamine is frequently found as the only isomer in
urine specimensfrom methamphetamineabuseres.Enantiomerically
pure S-(+)-methamphetaminecan be synthesizedfrom ephedrine or
pseudoephedrinevia chloroephedrineintermediates.These
intermediates are unstableand capable of cyclizing to form cis- and
trans-l,2-dimethyl-3-phenyl aziridine. Studieswere done to
determine if these intermediatescould be detected when usinga
common gas chromatographic-massspectrometric analytical
method (derivatization with heptafluorobutyricanhydride, HFBA)
for toxicologicalscreeningof methamphetamine.Analysisof (+)- or
(-)-chloroephedrine after extraction into hexane and derivatization
with HFBA indicated thai both pseudoephedrineand ephedrine
were the major compounds detected. Direct derivatization of a
hexane solution of cis-l,2-dimethyl-3-phenyl aziridine yielded only
the derivativesof ephedrine and pseudoephedrine,indicating that
the aziridine intermediate is responsiblefor the formation of the
ephedrine or pseudoephedrine.Thesestudiesindicate thai the
aziridine intermediateswould not be detected in methamphetamine
samples following HFBA derivatization.
Introduction
The occurrence of MPTP, a potent neurotoxin that was a synthetic impurity present in an illicitly synthesized meperidine analogue, was a dramatic example of the dangers associated with
drug impurities present in clandestinely synthesized drugs.
Metharnphetamine ("speed") has developed into a major problem
in the United States. Prior to the early 1970s, pharmaceuticalgrade rnethamphetarnine was readily available. Currently, clandestinely synthesized rnethamphetamine is used almost
exclusively and ranks among the top 20 drugs in emergency room
* Author to whom corresp<mdence should be addressed: William H. Soine, Ph.D., Department of
Medicinal Chemistry, Sch(x)l of Chemistry, P.O Box 980540, Ri( hmond, VA 2 ~298-0540. E-mail
[email protected].
602
and medical examiner mentions. The acute lethal dose of
amphetamines is generally thought to be in the range of 20-25
mg/kg. However, as tolerance develops, dosages as high as 800 mg
have been given without significant effect (1). Chronic abusers
may use as much as 15,000 mg per day (2). The concentration of
rnetharnphetamine in blood and tissues in fatalities associated
with methamphetamine is highly variable (3). Metharnphetarnine
concentrations in fatal cases have ranged from less than 1 rng/L
to over 14 rng/L (4-7); however, there does not appear to be any
correlation between the levels of methamphetamine and
amphetamine in blood to metharnphetarnine-caused deaths (8,9).
Often, low levels of amphetamines are thought to be incidental
findings, but very low concentrations of metharnphetarnine (0.7
rng/L) have been observed in patients dying of what is now
described as "classic stimulant toxicity" with agitation, hypertension, tachycardia, and hypertherrnia (10).
Surprisingly, there is very little information on the occurrence
of synthetic impurities in biological samples associated with
methamphetarnine. The only report to date was the occurrence of
c~-benzyl-N-methylphenethylamine that was present in 2 of 80
rnetharnphetamine-positive urine samples (3,11). This impurity
occurs during the synthesis of phenyl-2-propanone used in the
reductive amination of metharnphetarnine. According to the
DEA, this synthetic route occurred in only 16% of the clandestine
laboratories seized in 1993. In the same study, the ephedrine
reduction method was used in 81% of the laboratories (12).
Although multiple recipes have been reported for the conversion
of (-)-ephedrine or (+)-pseudoephedrine, the synthetic routes
depicted in Figure 1 for the conversion of (-)-ephedrine to S-(+)rnetharnphetamine are often used. When using either PCI5,
SOC12, HI (red phosphorus), or I2/NaI, the common product or
intermediate from the halogenated intermediates is cis- or trans1,2-dimethyl-3-phenylaziridines (13-15). The presence of these
intermediates and impurities (including the aziridines) have been
detected at varying levels in clandestinely prepared methamphetamine (13-i6), and there is an anecdotal report that the
halogenated and aziridine impurities "ruin the finer aspects of the
rneth [sic] high" (17).
Reproduction(photocopying)of editorialcontentof this journalis prohibitedwithoutpublisher'spermission.
Journal of Analytical Toxicology, Vol. 24, October 2000
Analytical methods for qualitative identification of (+)- or (-)chloroephedrine or ca- or trans-l,2-dimethyl-3-phenylaziridine
have been described for the analysis of drug formulations
(14,16). Methods for detecting these impurities in biological
samples have not been reported. A common approach used for
the identification of methamphetamine in plama or urine
requires extraction and derivatization with an anhydride (e.g.,
heptafluorobutyric anhydride, HFBA) followed by analysis by gas
chromatography-mass spectrometry (GC-MS) (18,19). This
derivatization procedure was evaluated to determine if it could
be used to detect (+)- and (-)-chloroephedrine or cL~-and trans1,2-dimethyl-3-phenylaziridine in aqueous samples during the
analysis of methamphetamine. Unexpectedly, it was observed
during these studies that all of these synthetic intermediates
were primarily identified as HFBA derivatives of ephedrine and
pseudoephedrine.
chloride is added dropwise with stirring. Once addition was complete, the ice bath was removed, and the reaction mixture was
allowed to stir at room temperature for 1 h. The mixture was again
cooled in an ice bath and layered with 100 mL of cold diethyl ether.
White crystals began to form immediately. The crystals were filtered and washed with 10 mL of acetone and dried to give 10.55 g
(47.9 retool) of a mixture of (•
(75:25)
based on proton NMR using the peak heights for N-CH3 HCI salts
at 2.87 and 2.79 for the (+)- and (-)-isomers, respectively.
Recystallization in absolute ethanol (75 mL) gave 7.34 g (33.3
retool, 67% yield) (+)-chloroephedrine.HCl (> 99% single
isomer): melting point, 200-201~ (Lit 200-201~ (21). 1H NMR
(CDCI3) 8:2.87 (s, 3H, N-CH3). 1H NMR (D20) & 1.16 (d, 3H, CH3,
J = 7 Hz), 2.79 (s, 3H, N-CH3), 3.89-3.99 (m, 1H, N-CH, J = 15, 7
Hz), 5.2 (d, 1H, CI-CH, J = 9 Hz), 7.45-7.52 (m, 5H, ArH). No
decomposition was observed in D20 after 28 days (pH approximately 5.5-6.0). Using pseudoephedrine.HC1 (5 g, 24.8 mmol)
under identical conditions gave 5.6 g (25.4 retool) (•
chloroephedrine.HCl (60:40) based on 1H NMR.
Experimental
(-)-Chloroephedrine.HCl. (-)-Chloroephedrine.HCl was synthesized by the method of Erode (20). Phosphorus pentachloride
Reagentsand chemicals
(16 g, 76.8 mmol) was dissolved in 20 mL chloroform. The solu(-)-Ephedrine and (+)-pseudoephedrine as the free base were
tion was cooled in an ice bath, and 7 g (34.7 retool) of (+)-pseupurchased from Aldrich Chemical Co. (Milwaukee, WI). The (-)doephedrine.HC1 was added slowly while stirring. After 2 h, 15
ephedrine.HCl and (+)-pseudoephedrine.HCl were purchased
mL cold ethanol was added dropwise to destroy an unreacted
from Spectrum Health Care Products (New Brunswick, NJ).
PC15.Crystals were allowed to form for no more than 2 h while the
HFBA was purchased from Regis Chemical Co. (Morton Grove,
reaction mixture was stored in the freezer. The white precipitate
IL). The other chemicals and reagents were high-performance
was filtered to give 2.1 g (10.4 retool, 30% yield) of (-)liquid chromatography or reagent grade. Melting points were
chloroephedrine.HCl (> 98% single isomer): melting point
determined on a Thomas Hoover melting point apparatus. Proton
195-197~ (Lit 197-198~ (21). It must be noted that the reacnuclear magnetic resonance spectra (1H NMR) were obtained on
tion mixture or additional work-up of the mother liquor will often
a Varian Gemini 300 MHz spectrometer.
precipitate out a higher percentage of (+)-chloroephedrine (up to
(+)-Chlorophedrine.HCl. (+)-Chlorophedrine was synthesized
25%) with the (-)-chloroephedrine. The mixture can be further
according to the method of Erode (20). (-)-Ephedrine.HCl (10 g,
purified by recrystallization of this mixture using a chloro49.6 retool) is dissolved in 20 mL chloroform. The solution is
form/ethanol azeotrope and allowing it to stand in the freezer for
cooled to ice bath temperature and 20 mL (274 mmol) thionyl
seven days. Starting with 2.5 g (12.4 retool) of (•
ephedrine.HCl (25:75), recrystallization gave 350
Table I. Retention Times and Mass Spectral Data for Amphetamine,
mg (1.7 retool) (-)-chloroephedrine.HC1 (> 99%
Methamphetamine, and Common Synthetic Impurities as their HFBA
single isomer): melting point 203-205~ (Lit
Deriviatives*
197-198~ (20). 1H NMR (CDCI3)& 2.79 (s, 3H,
N-CH3).
1H NMR (D20) & 1.26 (dd, 3H, CH3, J -- 7
tR* Base M§
M+H§
Hz),
2.78
(s, 3H, N-CH3), 3.71-3.81 (m, 1H, NCompound
(min) m/z m/z(%) m/z(%)
Other ions,m/z (%)
CH), 5.55 (d, 1H, CI-CH, J = 4 Hz), 7.40-7.57 (m,
5H, ArH). No decomposition was observed in D20
(+)-Amphetamine
7.36
91
331
ND
240 168 192 117 118
(2)
(99) (15) (4) (31) (31) after 28 days (pH approximately 5.5--6.0).
(+)-Methamphetamine 9.55 254
345
ND
210 91 118 117 169
r
was synthe(2)
(50) (33) (11) (7) (6)
sized from (+)-chloroephedrine.HCl (> 99%
(+)-Chloroephedrine
7.28 254
ND
ND
210 117 344 153 38o
single isomer) using the method described by
(38) (13) (3) (3) (1)
Tanakaetal. (22). A mixture of cis- and trans-1,2(-)-Chloroephedrine
6.61 254
ND
ND
210 344 117 153 381
dimethyl-3-phenylaziridinewas synthesized using
(37) (31)(14)
(3) (2) the procedure described by Windahl et al. (23).
(-)-Ephedrine
(+)-Pseudoephedrine
9.98
254
ND
ND
11.07 254
ND
ND
210 344 105 117 115
(31) (7) (7) (6) (6)
210 344 105 117 115
(29) (9) (6) (6) (6)
* All spectra reported were with relative total ion abundance < 1,200,000. Chromatographic conditions are
described in the section on GC-MS analysis.
t tR' retention time. ND; not detected
From (•
(44:56), a mixture
ofcis- and trans-l,2-dimethyl-3-phenylaziridine
(30:70,based on N-CH3 absorbance) was obtained.
Sample preparation for HFBA derivatization
(+)- and (-)-chloroephedrine.HCl and
cis-l,2-
dimethyl-3-phenylaziridine stock standard solu603
Journal of Analytical Toxicology,Vol. 24, October 2000
tions (2 mg/mL) were prepared in deionized water. Standard solutions were prepared fresh just prior to analysis. 3~vomilliliters of
standard solution was alkalinized with 300 ~L of 12N sodium
hydroxide, and the analytes were extracted into 2 mL of hexane.
The hexane layer was transferred into a 13 x 100-ram screwcapped test tube; 100 IJL of HFBA was added, and the tube was
capped. The mixture was incubated at 75~ for 20 min, then evaporated under a stream of air at 40~ The resultant residue was
dissolved in 50 ~L of hexane, and 2 IJL was injected into the
GC-MS.
Resultsand Discussion
The GC-MS method was evaluated for the analysis of (+)- and
(-)-chloroephedrine and cis- and trans-l,2-dimethyl-3-phenylaziridine after derivatization with HFBA. After derivatization of
(+)-chloroephedrine, the HFBA derivative of (+)-chloroephedrine eluted at 7.28 min and accounted for 60% of the material
injected (based on relative peak heights) along with ephedrine
(5.51 rain, 9%) and pseudoephedrine (6.03 min, 32%). After
derivatization of (-)-chloroephedrine, the HFBA derivatives of
(-)-chloroephedrine was not observed. Instead, the HFBA derivative of (+)-chlorophedrine (1%), ephedrine (74%), and pseuGC-MS analysis
doephedrine (23%) were observed. The HFBA derivative of
GC-MS analysis was performed on the Varian Star 3400Cx
(-)-chloroephedrine could be formed by derivatization of a
(Varian, Sugar Land, TX) ion-trap GC-MS equipped with a 30-m
CH2Cl2solution of (-)-chloroephedrine.HCl, and it eluted at 6.61
x 0.25-mm i.d. capillary column with a 0.25-]Jm film thickness of
rain. No peak at 6.61 min was detected after a basic extraction of
5% phenyl 95% methylpolysiloxane cross-linked (DB-5, J&W
(-)-chloroephedrine with hexane. After extraction of cis-l,2Scientific, Folsom, CA). The GC was operated in the splitless
dimethyl-3-phenylaziridine from a basic solution and derivatizamode with a helium carrier gas linear velocity of 20 mL/min. The
tion with HFBA, the products detected were ephedrine (12%),
transfer line was set at 280~ The GC parameters for the direct
pseudoephedrine (80%), and an unidentified material eluting at
analysis of the amines used temperature programming, 60~ to
5.46 rain (Rs = 0.76, 8%).
240~ at 10~
(injection port, 230~ The GC parameters
The ions associated with the HFBA derivatives are listed in
for the HFBA derivatives used temperature programming, 70~
Table I. No chloro substituted ions were observed for the HFBA
to 240~ at 10~
(injection port, 190~ The ion-trap MS
derivativesof (+)- or (-)-chloroephedrine gave mass spectra comwas operated in the electron impact mode at 70 eV with ion conparable to HFBA derivatives of ephedrine or pseudoephedrine.
trol on Auto. Mass ranged from m/z 40 to 600 with a scan time of
Therefore, the differentiation of these compounds is based upon
0.7s.
their different retention times. It would be
expected that the free base form of the chloroephedrine isomers cyclized only to their respecHI, red P
tive aziridine products during the analysis. In
contrast,
during the HFBAderivatization of (+)~~~NHCH3
3
or (-)-chloroephedrine, both HFBA derivatives of
S-Methamphetamine ephedrine and pseudoephedrine were detected.
Ephedrineor
This indicates that during the derivatization proPseudoephedrine
cedure a carboniurn ion is formed at the benzylic
position once the aziridines are formed allowing
LIAIH4
epimerization to occur. Therefore, the ratio of
or PCI5 ~4~
or PdlC
ephedrine and pseudoephedrine detected after
analysis cannot be used to determine the ratio of
(+)- or (-)-chloroephedrine to cis- or trans-l,2HCH3
~NHCH3
dimethyl-3-phenylaziridine before analysis.
It appears that if the chloroephedrines or
4"
~
CH3
aziridines were carried through an extraction and
(+)-Chloroephedrine
(-)-Chloroephedrine acyl derivatization sequence comparable to that
used for methamphetamine, the chloroephedfines or aziridines did not give derivatives or mass
spectra consistent with their structures. The
observations in this study may explain why these
halogenated intermediates and aziridine impurities have not been previously detected or reported
by forensic toxicologists. Also, if methamphetamine is detected in a urine or blood specimen and ephedrine or pseudoephedrine is also
trans-1,2-Dimethylcis-1,2-Dimethylidentified, it cannot be assumed that ephedrine or
3-phenylaziridine
3-phenylaziridine
pseudoephedrine were co-administered with
methamphetamine. They may be present as artiFigure I. Common clandestine methods for conversion of (-)-ephedrine or (+)-pseudofacts
of illicit methamphetamine synthesis.
ephedrine in (+)-methamphetamine via a halogenated intermediate.
No pharmacological or toxicologicalstudies on
••NHCH
soc,,\
604
J
Journalof Analytical Toxicology,Vol. 24, October2000
the intermediates described in this study have been done.
Whether these impurities would present an acute or chronic
problem in humans when co-administered with methamphetamine is unknown.
12.
13.
14.
References
15.
1. E.G. Zalis, G.D. Lundberg, and R.A. Knutson. The pathophysiology
of acute amphetamine poisoning with pathologic correlation.
J. Pharmacol. Exp. Ther. 158:115-127 (1967).
2. E.G. Zalis and L.F. Parmley. Fatal amphetamine poisoning. Arch.
Intern. Med. 112:822-826 (1963).
3. K.A. Moore, J.S. Daniel, M. Fierro, A. Mozayani, and A. Poklis. The
detection of a metabolite of 0r
synthesis in a mixed drug fatality involving methamphetamine.
J. Forensic Sci. 41:524-526 (1996).
4. R.H. Cravey and D. Reed. Intravenous amphetamine poisoning:
report of three cases.J. Forensic Sci. 10:109-112 (1970).
5. T. Kojima, I. Une, and J. Yashiki. CI-massfragmentographic analysis
of methampbetamine and amphetamine in human autopsy tissues
after acute methamphetamine poisoning. Forensic Sci. Int. 21:
253-258 (1983).
6. R. Matoba, S. Onishi, and I. Shikata. Cardiac lesions in cases of
sudden death in methamphetamine abusers. Heart Vessels l(suppl
1): 298-300 (1985).
7. R.C. Baselt and R.H. Cravey. Disposition of Toxic Drugs and
Chemicals in Man. Year Book Medical Publishers, Littleton, MA,
1989.
8. S.B.Karch, B.G. Stephens, and C.-H. Ho. Methamphetamine-related
deaths in San Francisco: demographic, pathologic, and toxicologic
profiles. J. Forensic Sci. 44:359-368 (1999).
9. S.B. Karch. The Pathology of Drug Abuse. CRC Press, Boca Raton,
FL, 1993.
10. T. Fukunaga, Y. Mizoi, J. Adachi, and Y. Tatsuno. Methamphetamine
concentrations in blood, urine and organs in fatal cases after abuse.
Jpn. J. Leg. Med. 41:328-334 (1987).
11. K.A. Moore, A. Ismaiel, and A. Poklis. (~-BenzyI-N-methylphenethylamine (BNMPA), an impurity of illicit methamphetamine synthesis:
16.
17.
18.
19.
20.
21.
22.
23.
III. Detection of BNMPA and metabolites in urine of methamphetamine users. J. Anal. Toxicol. 20:89-92 (1996).
Diversion Quarterly. U.S. Department of Justice, Drug Enforcement
Administration, Office of Diversion Control. 1:4 (1994).
H.F. Skinner. Methamphetamine synthesis via hydriodic acid/red
phosphorus reduction of ephedrine. Forensic Sci. Int. 48:123-134
(I 990).
T.S. Allen and W.O. Kiser. Methamphetamine from ephedrine: 1.
Chloroephedrines and aziridines. J. Forensic Sci. 32" 953-962
(1987).
T.S.Cantrell, B. John, L. Johnson, and A.C. Allen. A study of impurities found in methamphetamine synthesized from ephedrine.
Forensic Sci. Int. 39:39-53 (1988).
F.T. Noggle, J. DeRuiter, and C.R. Clark. Liquid chromatographic
determination of the enantiomeric composition of methamphetamine prepared from ephedrine and pseudoephedrine. Anal.
Chem. 58:1643-1648 (1986).
Uncle Fester. Secrets of Methamphetamine Manufacture Including
Recipes for MDA, Ecstasy, and Other Psychedelic Amphetamines,
5th ed. Loompanics Unlimited, Port Townsend, WA, 1999, p 114.
E.M. Thurman, M.J. Pederson, R.L. Stout, and T. Martin.
Distinguishing sympathetic amines from amphetamine and methamphetamine in urine by gas chromatography/mass spectrometry.
J. Anal. ToxicoL 16:19-27 (1992).
T. Yamamoto, R. Takono, T. Egashira, and Y. Yamanaka.
Simultaneous determination of methamphetamine and its metabolites in monkey urine by gas chromatography/mass spectrometry
with selected ion monitoring. J. Anal. Toxicol. 13:117-119 (1989).
H. Erode. Uber diastereomerie Ill. Chlor- und brom-ephedrine. Helv.
Chim. Acta 12:384-399 (1929).
T. Taguchi and M. Kojima. The formation and chemical properties of
diastereomeric 1,2-dimethyl-3-phenylaziridine. Chem. Pharm. Bull.
7:103-107 (1959).
K. Tanaka, T. Ohmori, and T. Inoue. Analysis of impurities in illicit
methamphetamine. J. Forensic Sci. 56:157-165 (1992).
L. Windahl, M.J. McTigue, J.R. Pearson,S.J.Pratt,J.E.Rowe, and E.M.
Sear. Investigations of the impurity found in methamphetamine synthesized from pseudoephedrine by reduction with hydriodic acid
and red phosphorus. J. Forensic Sci. 76:97-114 (1995).
Manuscript received March 30, 2000;
revision received July 10, 2000.
605