Presented to the Graduate Council of the University of North Texas in Partial Fulfillment of the Requirements for the Degree Of

Home , Phentermine, Propylamphetamine

3-7?

CHEMICAL IONIZATION (CI) GC/MS ANALYSIS OF UNDERIVATIZED

AMPHETAMINES FOLLOWED BY CHIRAL DERIVATIZATION TO IDENTIFY

d AND 1-ISOMERS WITH ION TRAP MASS SPECTROMETRY

THESIS

Presented to the Graduate Council of the University of North

Texas in Partial Fulfillment of the Requirements

for the Degree of

MASTER OF SCIENCE

(Pharmacology)

John A. Tarver B.S.,B.A.

May, 1991 Tarver, John A. Chemical Ionization (CI) GC/MS Analysis of Underivatized Amphetamines Followed by Chiral Derivatiza- tion to Identify d and 1-Isomers with Ion Trap Mass Spectro- metry. Master of Science (Biomedical Sciences), May, 1991,

26 pp., 9 figures, bibliography, 20 titles.

An efficient two step procedure has been developed

using CI GC/MS for analyzing amphetamines and related

compounds. The first step allows the analysis of underiv-

atized amphetamines with the necessary sensitivity and

specificity to give spectral identification, including

differentiation between methamphetamine and phentermine.

The second step involves preparing a chiral derivative of

the extract to identify d and 1-isomeric composition. TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS...... iv

INTRODUCTION .1......

EXPERIMENTAL

MATERIALS ...------... -....-..... 15

METHODS ------.---...... 15

RESULTS AND DISCUSSION ...-..-...... 17

CONCLUSION ------. . . - . . 23

APPENDIX...... ---. -----....-.-. ----.. -...... 24

REFERENCES ------.. . . --. . ...---.. . . . 28

iii LIST OF ILLUSTRATIONS

Figure Page

1. Full scan chromatogram and mass spectra of

amphetamine obtained in the electron impact

mode......

2. Full scan chromatogram and mass spectra of

methamphetamine obtained in the electron

impact mode...... 8

3. Full scan chromatogram and mass spectra of

amphetamine obtained in the chemical ioniza-

tion mode...... 9

4. Full scan chromatogram and mass spectra of meth-

amphetamine obtained in the chemical ionization

mode...... 10

5. Possible stereoisomers produced by derivatiz-

ing methamphetamine with trifluoro-l-prolyl

chloride...... 14

6. CI analysis of a 10 ng/ml urine standard

extract showing the spectra and signal:noise

ratio for both amphetamine and methamphetamine..18

7. Analytical curves demonstrating procedure

linearity ...... 19

8. Total ion chromatograms of TPC derivatives...... 20

iv Figure Page 9. Total ion chromatogram demonstrating spectral

differentiation between phentermine and meth-

amphetamine...... 22

V INTRODUCTION

Amphetamine and methamphetamine are sympathomimetic phenethylamines. Sympathomimetic agents are so called because their effects resemble stimulation of adrenergic nerves. As a result, these agents have profound effects on almost all systems. These include peripheral excitation or inhibition depending on the type of receptor present, card- iac excitation, metabolic increases and central nervous system (CNS) excitation. The differences in these drugs are seen in the degree to which each effects the above actions.

In comparison to other sympathomimetic agents, ampheta-

mine and methamphetamine have markedly more pronounced CNS effects. This is particularly true for methamphetamine which partitions more easily into the CNS because of its lipid solubility. These compounds have been used therapeutically in the treatment of obesity, narcolepsy, parkinsonism and behavior disorders. However, they have been largely replaced by other sympathomimetic agents for their peripheral effects with lesser CNS effects (1). With the decrease in therapeu- tic use amphetamine and methamphetamine are most often seen associated with cases of self administration and abuse.

Amphetamine and methamphetamine both occur as stereoisomers. Two molecules are isomers if they have the same

1 2

molecular formula but differ either in their structural

arrangement (structural isomers) or in their spatial ar-

rangement (stereoisomers). This is possible in any molecule

that contains asymmetric carbons. An asymmetric carbon is

one that is bound to four different atoms or groups.

Molecules that contain asymmetric carbons are said to

be chiral (handed) and they can be arranged spatially so

that they are mirror images of each other and cannot be

superimposed. These mirror images are called enantiomers.

Diastereoisomers, molecules with more than one asymmet-

ric carbon that are not superimposable, differ in energy

content so their chemical and physical properties are dif-

ferent and they can be resolved analytically. Enantiomers

differ only in the direction of rotation of plane polarized

light and therefore do not lend themselves to analytical

resolution. The nomenclature for enantiomers consists of

designation as d (dextrorotatory) or 1 (levorotatory),

alternatively + or -, indicating the direction of rotation

of plane polarized light. The absolute sterochemical config- uration about asymmetric carbons is designated by the sym- bols R and S.

The synthesis of optically active drugs usually produces racemic mixtures. These are mixtures that contain both

forms of the enantiomers. Enantiomers may have significantly different pharmacokinetic and pharmacodynamic properties

(2). 3

Because of these differences drug stereochemistry is an

area of growing importance in both clinical and forensic

toxicology. Methamphetamine exemplifies stereochemical im-

portance. Pharmacologically the d-isomer has much greater

central activity than the 1-isomer and forensically the d-

isomer is an illicit drug while the 1-isomer is an active

ingredient in an over the counter nasal decongestant (3).

The central psychic effects of normal dosages include

alertness, elevation of mood, elation and euphoria. Metham-

phetamine is also easily synthesized in clandestine labora-

tories (4,5) and is therefore widely encountered in illegal

trade and abuse.

As the dosage of methamphetamine used increases and

toxicity develops, hallucinations and paranoid delusions are

common. This is particularly true in cases of chronic intox-

ication rather than acute toxicity. The wide range of doses

required to produce toxicity in different individuals reduc-

es the predictability of onset and increases occurrence.

Toxicity can occur with as little as a single 2 mg dose

while some tolerant individuals may use as much as 1700

mg/day without apparent ill effects (1).

The abuse potential, the type of toxic manifestations

involved and the ease of synthesis by clandestine laborato-

ries all contribute to making the use of amphetamines and methamphetamines a social problem. For this reason they are 4 included in the list of drugs routinely screened for during urine drug testing.

Increased public awareness of the deleterious effects of drug abuse and the fact that U.S. companies spend around

$33 billion dollars annually (6) on drug testing reflects acceptance of drug testing. From 1985 to 1986 employee drug testing by Fortune 500 companies rose from 18 to 40 percent

(7). The impact that urine drug testing has on people's lives and careers causes a proportional concern regarding the accuracy and reliability of results. This concern is manifested in the legal system and demands that urine drug testing be legally defensible. Legally defensible testing systems must include: (a) witnessed urine collection; (b) maintenance of external and internal chain of custody docu- ments; (c) records of procedures and worksheets regarding instrument operation, maintenance and quality control, and records of personnel training and experience; (d) participa- tion in accreditation and proficiency programs; (e) reporting procedures and recall; and (f) storage of all specimens, whether they gave negative or positive results (6). Experts indicate that single procedure methods of testing are not legally defensible and that only screening followed by confirmation methods, EMIT-GC/MS, TDX-GC/MS and RIA-GC/MS, are fully defensible (8).

One ongoing problem with drug testing programs has been the lack of consistency which is obtained through common 5 methodology and defined criteria for positive results. This was first addressed by the National Institute of Drug Abuse

(NIDA) in drafting guidelines for accreditation of drug testing laboratories (9) and drug testing programs (1.0).

There are now two entities that offer accreditation for drug testing laboratories, NIDA and College of American Patholo- gist (CAP).

Methodologies required for accreditation are now con-

sistent but there is considerable evidence that the "cutoff"

limits used to determine positive/negative results as out-

lined by NIDA are much too high and results in false negative reporting. The effects of this are seen in that while drug testing programs have decreased the incidence of drug related incidents a serious problem still remains. The high,

500ng/ml (11), "cutoff" limits enables too many abusers to beat the system.

The "gold standard" for confirmations is gas chromatography with mass spectrometry (GC/MS) (8); however, there is still need for improvement. Most laboratories use quadrupole mass spectrometers in the selected ion monitoring (SIM) mode. This method relies on GC retention time plus the molecular structure evidence provided by the ratios of selected ions (conventionally three ions). Increased certainty can be obtained by using full scan spectra with chemical ionization mass spectrometry. 6

The reason full scan spectrometry is not used more, directly relates to the operational differences between quadrupole and ion trap spectrometers. The conventional quadrupole mass spectrometer using SIM must have three dis- tinct ions to use ion ratioing. Amphetamine, for example, has a base peak of 44 and virtually no mass ion +1 (136) of significant intensity that allows detection or ion ratioing

(Fig.1 & 2); whereas ion trap mass spectrometry allows full spectra identification even in the electron impact (EI) mode. The increased residence time of ions in a "trap type" spectrometer increases the sensitivity enough to differenti- ate them from background noise on a full scan. In addition, source operating pressures in a quadrupole make CI operation cumbersome and is not routinely supported by manufactures.

The low source pressure in an ion trap makes CI a simple operational mode to employ. CI produces a well defined spectra for amphetamine and methamphetamine (Fig. 3 & 4).

The need for improved analytical procedures for the analysis of amphetamines is exemplified by a consensus statement issued by the Department of Health and Human

Services which states: "For the amphetamine(s) a study should be undertaken to critically evaluate data for the purpose of recommending lower cut-off levels for both screening and confirmation. Laboratories should be able to resolve the d- and 1-isomers of methamphetamine and amphetamine" (12). 7

Rw'* t?;2) -269) DU 31199L1/t :2 II :024,24E $can Range:281 - 601 141912 eve32134

TOT

A t j

.00350 400 450 500 550 600 b:01 5:51 6:41 7:31 8:21 9:11 10:01 CHRO>

Spectrum # 31 Filenw : 1-5 I Acqsif'ed: Nov-10- 90 12:90:24 + 5:12 C'q-en t: 90(2)-260 9DEMIN IUL:=U N1 OK LUM ? EXTRE14L lase Pk: 44 Int: 998 Range: 43- R C: 141 7 Iw. =:9960 tine: 185 j W/o 44AGC

AMPHETAMINE ION TRAP-EI MODE

1 51 659 7710 40 69 80 109 ' 120 140 SPEC>

Figure 1. Total ion chromatogram from EI GC/MS of amphetamine. Note the low intensity of the molecular mass +1 (136) in the mass spectra even from this 500 ng/ml sample. 8

) 602 u :00:34 San 2-4 4 1. 1Scan32134

TOT

3 030 4W0 450 5H0 550 600 :01 5:51 6:41 7:31 8:21 9:11 10:01

SpeCtmxn 9 343F e ape: 1-5 I- No 6-19 12:90:24 + 5:44 bast Wke 58 Int: 2 902 Range; 43-1 2 RC: 314 ME.64 = 2202 199t 58 AGC time:93

JHT- 14ETHAMPHETANINE ION TRAP-EI MODE

65 91 43 5 j 1 7? , !3f .3 6e 88o100 120140 SPEC>

Figure 2. Total ion chromatogram from EI GC/MS of methamphetamine. Note the absence of the molecular mass +1 (150) in the mass spectra in this 500 ng/ml sample. 9

bromatograr S500G1CJ AC uiret ov- Z- ?9 S:34:49 vmnnt: SEIWN AmPligroI EBy it sef)-al,10 19N JL INJ JA' Scan ianst: 3 1.P0U Sb a" in : 4S2738 10w

NNO.. . - - - -r 350 400 450 sea W5: & ftf.651 6:41 7:3L S10 9:2 C1490>) Spectrum 3 19 Ti ei : 5 C e: Nov-02-1999 01:34:49 + 5:19 SERUM WANH ?T#NN 3I" G"TIUL INJ JAT Base P)C' 44 Int:_Z439 RaS.: 43-136 C. 5? LQ =.0z : 28439 AGc tite: 239 44

INT 'I 149

136

Alit of - r - - - 11 . .& I I v t--4 - i, "I A I T T- - - 40 90 .9Ie0 sFrc Ito

Figure 3. Total ion chromatogram from Ci GC/MS of amphetamine. The mass spectra demonstrates the intensity of the molecular mass +1 (136) obtained using CI. 10

CwoeAto9r SSMAC A qui red: Nov- i9- 09::34: 49 Cop ent: SNUMAMP$FAJINE SY C? 90(2,-28 Z iL0DEC/MIN LL INJ JAY ican lanet: 301-6U Scan: 367 st : 6114 & 6:08 lw/,: 452'73 19 . . A Ep TOT?

,>, - t ---. LL K~I 1~F~-' 35 499 450 5: 51 6:41 7:31 55 CHRO) 8:21 9: 1

Spectrupt * 967 !jleae: S50CI qred: Nov-92.-19 9 09:34049 + 6:08 Content: E M RUN ANN *H NE NY Cl pet- E/MN~ IHL 43 JOT 1wt: 36 4 Range: 43-1 R C 61546 1e. 4 : 36149 5,84 tto:253

M~r

91 119 45 69 S 97 l --.- I L 134- o If 4 15 Be 1eto SPEC> I2 14 160

Figure 4. Total ion chromatogram from CI GC/MS of methamphetamine. The mass spectra demonstrates the intensity of the molecular mass (150) obtained using CI. 11

Amphetamine related compounds present unique challenges

for drug testing laboratories. The simplicity of their structure and the large number of closely related compounds contribute to a high degree of cross reactivity with immunoassay methods. In addition, because of their structural and physicochemical similarities they have similar chromatographic elution times and in some instances, i.e. methamphetamine and phentermine, cannot be differentiated even by

their mass spectra when using EI mass spectrometry. Even

amphetamine compounds that chromatographically resolve are

difficult to identify by EI mass spectrometry, especially at

lower concentrations, because of the few fragment ions they

exhibit greater than 5% of the total ion current and their base peak ions of 44 and 58 are generic to primary and

secondary methyl amines.

In order to solve this problem many procedures have

been reported that use derivatization to improve the chromatographic properties and increase the mass and mass frag- ments of amphetamines (13-15). These range from relatively

complex and time consuming procedures to on column derivati-

zation. These improve the chromatographic characteristics

and mass fragment patterns, but the by products often dete-

riorate the column and contaminate the mass spectrometer

source faster resulting in decreased efficiency." In addi-

tion, some derivatizing agents in the presence of high 12 concentrations of ephedrine have been shown to artificially produce racemic methamphetamine (16).

Full scan CI mass spectrometry using an ion trap mass

spectrometer allows the necessary sensitivity and specificity to accomplish the task of characteristic identification without derivatization . With full scan CI spectra we obtain

improved certainty over SIM, greater sensitivity and actual

spectral differentiation between previously undifferentiated

compounds (17,18).

CI spectrometry is a "softer" ionization than EI. This

is accomplished by using an intermediate ion, obtained by EI

ionization of the methane reagent gas , to analyze analyte molecules. This process allows greater production of the molecular ion for identification and increases sensitivity

(S/N ratio) by reducing background.

Chromatographic isomeric resolution of d- and 1-isomers

requires bonding with another chiral compound. This bonding

creates diastereomers out of enantiomers and allows resolution. There are two approaches that accomplish this. The

first is through the use of chromatographic columns that

contain chiral stationary phases. As the analyte compounds are carried through the column they partition into the

stationary phase and transient bonds are formed converting the enantiomers into diastereomers and the partioning time

is then different and the enantiomers are resolved as diastereomers. This approach requires instrumentation be dedi- 13 cated for enantiomer analysis and is not feasible for most laboratories.

The second approach to enantiomer analysis is to use a chiral derivatizing agent to produce diastereomers of the analyte enantiomers (Fig. 5). These diastereomers can then be resolved on conventional (achiral) columns. For this to be of practical use the derivatizing reagent must be of sufficient enantiomeric purity to give minimal amounts of one of the possible set of spatial configurations. For example, in figure 5 the structures are those that occur when d-1 methamphetamine is derivatized with d-1 Trifluoro- acetylprolyl chloride. The possible products are two sets of enantiomers which are diastereomers (lS2S,lR2R are diastereomers of lS2R,lR2S). On an achiral column these will give two peaks with the enantiomers coeluting. If the derivatizing reagent were not enantiomerically pure there would be three products even for an enantiomerically pure analyte. These would still chromatograph as two peaks. Using a pure derivatizing reagent yields a single product for a pure analyte and two peaks for racemates. This procedure makes the reasonable assumtion that racemization does not occur around the chiral carbons during analysis (19).

Several chiral derivatizing reagents are available of sufficient purity to be acceptable. Two that have been used to derivatize amphetamines are (-)-menthylchloroformate (20) and S-(-)-N-Triflouroacetylprolyl chloride (TPC) (19). 14

C. 0 u t44f-4C))C)' e-.

Li.L 4 0 4.Jy-& C ~~-4 o N 'T 0 4-J '4

'H"' 0M u0, (n-4 *0+ mga t ..... l * ,-4,o I : F= i ) c 0 + 44 -r4.9 t 0Isel 0 ,i: s-s C)I .C 'H,

.r 5 V tO

S0 0j .41 )4 - SC C)9 4 4JC ii I 00 0~ 0-H G 0

LL- %- o 0M *N '4O 't'osJ C NO 01:1 - 1 '4it.. 'rA Q

If, Q 0 mEs

1-4 9 0--IU 0 U44 EXPERIMENTAL

Materials 1-Amphetamine(a-0533),d-amphetamine sulfate (A-5880),d- methamphetamine HCl (M-8750) were purchased from Sigma

Chemical Co. 1-methamphetamine, 1mg/ml in MeOH (014703) from

Alltech-Applied Science Labs.

S-N-Trifluoroacetyl-L-prolyl chloride (24850-9) and

(-)-menthylchloroformate (24530-5) were supplied by Aldrich

Chemical Co. Chloroform was purchased from Baxter

(GC60125-4) B&J GC2.

Instrumentation consisted of a Hewlett Packard 5890 GC

with a J&W 30 meter DB5 column coupled to a Finnigan 80 0T" ITD mass spectrometer equipped for CI analysis. Reagent gas was Methane 99.99% from Curtin Matheson Scientific, Inc.

Methods

Standards were prepared by spiking blank urine at a

concentration of 2000 ng/ml of each drug. These were used to

produce standard curves from 10 ng/ml to 2000 ng/ml which were used to demonstrate linearity and for quantitation

purposes. The extraction procedure used 2 ml specimens

spiked with 50 microliters of 100 mcg/ml N-Propylamphetamine

as the internal standard. The samples were made basic with

0.5 ml 1 N NaOH and extracted into 5 ml of Chloroform. After

15 16 centrifuging the aqueous layer was aspirated and the Chloro- form layer back-extracted with 5 ml 0.1 N H2SO4. This mixture was vortexed, centrifuged and the aqueous layer trans-

ferred to a 10 ml disposable centrifuge tube. One ml of 1 N

NaOH and 50 microliters of Chloroform were added, the solution was vortexed and centrifuged. One microliter of the

Chloroform layer was injected into the GC/MS for analysis of the underivatized analytes.

The (-)-menthylchloroformate derivatives were made by aspirating all of the aqueous layer, adding 50 microliters of (-)-menthylchloroformate to the Chloroform, capping tightly and incubating at 700 C for 30 minutes. One microliter of the resulting solution was analyzed by GC/MS.

The TPC (Trifluoroacetyl-L-prolyl chloride) derivatives were made by adding 200 Al of Chloroform to the initial extractions, centrifuging, aspirating all of the aqueous

layer, adding 50 Al of 1 M TPC, capping tightly and incubat-

ing at 900 C for 5 minutes. After incubation 20 Al triethyl- amine was added and vortexed thoroughly. One yl was analyzed by GC/MS. The faster on column derivatization (20) was not used because of an increase in interfering compounds at the retention time of amphetamines. The GC program used was 200*

C isothermal for 2 minutes then ramped at 20* C/min. to a final temperature of 2600 C. All analytes eluted in less than 9 minutes. RESULTS AND DISCUSSION

The CI analysis of amphetamine and methamphetamine allows excellent sensitivity and gives a strong molecular ion even at a concentration of 10 ng/ml (Fig. 6). As shown, the signal to noise ratios at this concentration are well above the limit of quantitation of 10:1. This procedure has good linearity (Fig. 7) but has a relatively small dynamic range due to the depletion of the reagent gas. This might be extended some by increasing the reaction time but then there is a concomitant loss of molecular ion.

The (-)-menthylchloroformate derivative proved unsatis- factory because of the excessive amount of background produced and the length (30 min.) of the program necessary to achieve separation.

The TPC derivatives performed well giving baseline resolution of enantiomers (Fig. 8). The detection limit for the TPC derivatives of amphetamines and methamphetamines based on a signal to noise of 3:1, is at 100 ng/ml due to the increased number of peaks and background.

Standards spiked only with 1-amphetamine and 1-methamphetamine were derivatized and analyzed to determine the percent of d-TPC present as a contaminant in the derivatizing reagent and therefore the lowest ratio between enantiomers to be indicative a racemic composition. The average d-

17 18

1s14*23 Pj . it: 3734 t 4:42 aoYI9:e: j m I64sea 7.J5458

4t4. AM IM *"ETm*"ptrrf tNE 111 "L O A --A, I H

M 4 I b-is -.. I1

SOMaaW mo

1 & - TX -r :. I,5 st w 7031 CHRO)

Figure 6. Total ion chromatogram from CI GC/MS of amphetamines (Scan 306), methamphetamine (Scan 338), and N-propylamphetamine as internal standard (Scan 412). Full spectra for amphetamine and methamphetamine are in designated inserts, and the signal to noise (S/N) is 95:1 and 65:1 for 10 ng/ml (0.4 ng on column) for ,amphetamine and methamphetamine, respectively. 19

CA raton ?let (lot Stds) ilter0ee p AI? CormtI o f6 "A tNE Co#poun Y 3 $taoN od Uiio gik Nr$a of ample/Area of Standa4) vs Amount of SWSpl CIC ste UinL7n)

AMHET?.NE STANDWD CUE WIT? Nnit .. 2.0. AN bO??Z L NE$ AT I AND Z IAIWARD EIATIONS ..

A 0.5Ww

104 "a 9W 3900.9ff 40l999 3911.00

libration tnt (J4t Stk) Pilenwl: AMM Corr lotion CO..f 0.391 METAMMETAMLEL Co@poUnd: 3 of S itanderd Deviatio 9.634 (Area of aIpNe/Area of StandArd) i V (Awunt of SaMple In.tected) (LintLin)

1,So 5C1A1ALNwR FIT

i.1 w

190.900 290060 390.90 40.00 S.beS

Figure 7. Analytical curves for amphetamine (A) and methamphetamine (B) demonstrating linearity. The dotted lines represent 2 S.D. of each curve. The correlation coefficient for amphetamine is 0.998 0.034 and 0.998 0.44 for amphetamine and methamphetamine, respectively. 20

O OC Jan-10-1991CAc;ir:A: 11:5 :1 5 re wis 1Wes ean:Ili t :21283 : 1/.:235

.469 29166

119 23 1237

T -or AAn

7P, 2111, 70:11s

e' A sa 6 ?-1TT7:41 :8:4 I M

hrvates of ahetaine(Sc Jan-2-1991amhi:25 (Sn 38i) 1 ehanp I t 1c3 496:1 IMxeam2a-S

166 125 343 3

3 sio,41% '1 r*7+ 41 i a

Figure 8. Total ion chromatograms from CI GC/MS of TPC derivatives of 1-amphetamine (Scan 370), d-amphetamine (Scan 382), 1-methamphetamine (Scan 453), d-methamphetamine (Scan 462),and 1-n-propylamphetamine (Scan 490), and d-n-propylamphetamine (Scan 497). Spectra of TPC derivatives of 1-amphetamine and d-amphetamine in TIC (A) and spectra of TPC derivatives of 1-methamphetamine and d-methamphetamine in TIC (B) are shown in designated inserts. 21

amphetamine-TPC present was 17.4% and the average d-metham-

phetamine-TPC present was 9.1%.

After this preliminary work a set of 26 urine specimens

that had been previously identified by immunoassay and GC/MS

as being positive for amphetamines were analyzed. This set

included one specimen which was positive for phentermine and

a wide range of amphetamine, methamphetamine concentrations

from at or near "cut-off" of 200 ng/ml to greater than 2000

ng/ml (see appendix A).

It was possible to identify phentermine in both the underivatized and derivatized analysis not only by retention time but also spectrally due to the presence of a signifi-

cant 133 mass. This unique mass is not seen with amphetamine or methamphetamine (figures 6 & 9). As seen with the standards, the concentration of each isomer had to be 100 ng/ml or greater in order to achieve a full scan spectral

identification of the derivatized analyte at the requisite

3:1 signal to noise. In a specimen containing a racemic mixture this would translate into a 200 ng/ml specimen.

However, it was possible to identify these compounds at concentrations below this detection limit by looking only at characteristic mass in the data, but defeats the advantage of full scan spectra identification. 22

"tog eire f-.199j, c 2 1:33:40 ean~li r: sa: Int : 2561 0 4,s42 Iwoz 3$40zf 58 58 NeEm,1144 9 133L59[S, I -=w m-- - II TOT

I - 'r -r I 2A f CHIRO)

Figure 9. Total ion chromatogram from CI GC/MS of amphetamine (Scan 303), phentermine (Scan 323), and methamphetamine (Scan 335). Full spectra of phentermine and methamphetamine demonstrating characteristic 133 ion for phentermine not found in the 'methamphetamine spectra are shown in designated inserts.. CONCLUSIONS

CI GC/MS analysis of amphetamines increases sensitivity and spectral identification confidence by increasing the percent molecular ion present. One slight disadvantage is that there is a fairly low upper limit to the linear range for quantitation due to depletion of reagent gas. However, the added benefit of spectral differentiation between Phent- ermine and Methamphetamine, via characteristic full scan identification of each, is a significant improvement over conventional techniques. In addition, the analysis of amphetamine and methamphetamine without derivatization precludes the artificial generation of methamphetamine from samples containing ephedrine.

The L-TPC derivative is an efficient way to identify enantiomeric components of the amphetamines. If used quali- tatively with a criteria of a greater than 0.2:1 (d:l) ratio between enantiomers to establish racemic composition, it should be a valuable addition to prevent inappropriate identification.

23 APPENDIX

24 25

pectr-e og .lf2 en- 1 ; w-1PC A"c1'eN: JIn-15-:991 G9I51'45 +6:

211 NOTE 133 P'ASS: PHENTrRMINE

-1 251 A 343

133 3 ~ 94

5 50 I L~ M f-I'

P4ETERPIN INT Nf li1- 33 MS

7 r

Specimen obtained from CI analysis of a forensic urine specimen, TPC-derivatized (A) and underivatized (B), containing phentermine. 26 Chro*atogr aI* C:UNEITPC Acquired: Jan-OV-*991 (,rfent: 2@Q(2)-260 10DEG/MN IU:L INJ Sean Range: 281 - 690 Scan: 456 Int = 723487 t 7:3? 1 00. 997285 58 34 3 119 166 251 - I

L-METHAMPIETANINE NOTE NO D-NETHAMPHETAMINE TOT- PRESENT r

rm j__ 30 350 400 450 500 550 602 5:01 5:51 6:41 7:31 8:21 9:11 10:01 14R0>

C:UN]R2TPC Acquived: jan-08-1991 13: 2:55 Coent: 200(2)-260 10DEG/lN IUL INJ Swcan Range: 281 - 766 Scan: 455 Int 783939 7:36 10718 2 1 w/ 58 343 119 2;1 n L-METAAMPHETAMINE NOTE NO D-METHAMPHETANINE F TOT J PRESENT

r I -Y-;-' "f 402 509 71: N:01 6:41 8:21 11:41 CHRO> chromatograph and spectra from the analysis of TPC-derivatives of forensic specimens identified by another laboratory as methamphetamine. Note the specimens contained only L-methamphetamine. 27

S ecimen I.D. JUrderivatized (ng/ml) TPC-Derivatized (ng/mI) 11287 AMph. = 1001 .T-Amph.-= 334 D-Amph, = 251

Meth. = 937 L-Meth, = 329

D-Meth. = 873 11297 Amph. = 187 L-Amph1 = 88

D-Amph. = 97 Meth. =232 L-Meth. = N.D. D-Meth. = N.D.

D-Amph. = 289 Meth. 55 L-Meth. = N.D.

D-Meth. = N.D.

11325 Amph. = 290 L-Amph. = 75

D-Amph. = 103 Meth. = N.D. L-Meth. = N.D.

D-Meth. = N.D.

1465 Amph.120Q T-Amph - 2000 D-Amph. 2000 Meth. 5595 L-Meth. = N.D. D-Meth. = N.D. 11546 -Amph..=..1192 .- 124. D-Ameh. 689

Meth. 2000 L-Meth. 2000 D-Meth._=.1692

1191Amph . 1283 T-Amph .= Nn.

ip D-AMph.F = 988

11598 p = 175 L-Amph. = N.D. D-mh. = 117 Meth. = 1240 L-Meth. = 793

D-Meth. = 314

D-Amph = N.D. Meth. = 38 L-Meth. = N.D. D-Meth. = N.D.

Data from the quantitation of forensic urine specimens that had previously been identified by immunoassay and GC/MS to be positive for amphetamines. Other specimens data not presented because they all quantitated greater than 2000 ng/ml and did not contribute to the comparative data. REFERENCES

1. The Pharmacological Basis of Therapeutics, 5th ed., A.G.

Gilman, L.S. Goodman, A.G. Gilman, G.B. Koelle, Eds. Mc-

Millan Publishing Company, New York,1970, pp 302-305, 480-

483 and 496-500.

2. G.P. James. Methamphetamine stereochemistry. AACC TDM

Vol. 10; 12:3-4 (1989).

3. R.C. Basselt. Disposition of Toxic Drugs and Chemicals in Man. Biomedical Publications, Davis, CA, 1978, pp 189-201.

4. D.K. Eaton and G.C. Harbison. Isolation and identification of major products and reactants of clandestine amphetamine laboratories. Forensic Consultant Services, Fort

Worth, Texas. Personal Communication.

5. M. Courtney and T.R. Ekis. Laboratory analysis of clandestine lab chemicals, reaction mixtures, and raw products.

DEA, Dallas, Texas. Personal communication.

6.Hawks, R.L., Chiang, C.N. Urine Testing for Drugs of

Abuse, research monograph 73. National Institute on Drug

Abuse, 1986.

7.Drug Testing in Major U.S. Corporations: A Survey of

Fortune 500. Raleigh, N.C., Noel Dunivant and Associates,

1985.

28 .29

8.Hoyt, David W., Finnigan, Robert E., Nee, Thomas, Shults,

Theodore F., Butler, Thorne J. Drug testing in the work-

place-Are methods legally defensible? JAMA, 1987, 258,

504.

9.Draft Standards for Accrpditation of Laboratories Engaged

in Urine Drug Testing. National Institute on Drug Abuse,

1987.

10.Scientific and Technical Guidelines for Drug Testing

Programs. U.S. Dept. of Health and Human Services, 1987.

11. Federal Register April 11, 1988.

12. Department of Health and Human Services Publication No.

(ADM)90-1684, 27(1990).

13. E.G. de Jong, S. Keizers and R.A.A. Maes. Comparison of the GC/MS Ion Trapping Technique with GC-FTIR for the Iden- tification of Stimulants in Drug Testing. J.Anal.Tox. 14:-

127-131 (1990).

14. R.J. Czarny and C.L. Hornbeck. Quantitation of Metham- phetamine and Amphetamine in Urine by Capillary GC/MS Part

II. Derivatization with 4-Carbethoxyhexafluorobutyryl Chlo- ride. J.Anal.Tox. 13:257-262 (1989).

15. D.A. Toseland and P.H. Scott. Determination of amphetamine as its N-acetyl derivative by gas-liquid chromatography. Clin.Chim. Acta 25:75-78 (1969). 30

16. Roger Foltz, Chairman, Amphetamine Conference 43rd

Annual Meeting, American Academy of Forensic Sciences, Feb.

18-23, 1991.

17. M. Uhrich. Chemical ionization with the INCOS 50 : The key to accurate and quantitative analysis of drugs. Research monograph 213, Finnigan MAT (1988).

18. M. Lehrer. Application of Gas Chromatography/Mass Spec- trometry Instrument Techniques to Forensic Urine Drug Test- ing. Clinics in Laboratory Medicine,vol. 10, No. 2, June

(1990).

19. J.H. Liu and W.W. Ku. Determination of enantiomeric N- trifluoroacetyl-L-prolyl chloride amphetamine derivatives by gas chromatography / mass spectrometry with chiral and achiral stationary phases. Anal. Chem. 53:2180-2184 (1981).

20. R.O. Hughes, W.E. Bronner and M.L. Smith. A GC/MS assay for amphetamines and methamphetamines with stereoisomer separation and reduced interferences. 1990 S.O.F.T. Annual

Meeting.

21. R.L.Fitzgerald, J.M.Ramos,Jr.,S.C.Bogema, and A. Poklis.

Resolution of methamphetamine stereoisomers in urine drug testing: Urinary excretion of R(-)-methamphetamine follow- ing use of nasal inhalers. J.Anal.Tox. 12:255-259 (1988).