Journal of Analytical Toxicology, Vol.

12, March/April 1988

Confirmation of Marijuana, Cocaine, Morphine,

Codeine, Amphetamine, Methamphetamine,
Phencyclidine by GC/MS in Urine Following
ImmunoassayScreening
S . J . Mul(~* a n d G . A . C a s e l l a

New York State DSAS Testing and Research Laboratory, 80 Hanson Place, Brooklyn, New York 11217 and
Department of Psychiatry, SUNY Health Science Center at Brooklyn, New York 12203

Abstract
Rapid, reliable, sensitive, qualitative, and quantitative
methods using small urine volumes (0.2-0.5 mL) were
developed primarily for confirmation of marijuana, cocaine,
benzoylecgonine, ecgonine methyl ester, morphine, codeine,
amphetamine, methamphetamine, and phencyclidine. Using
capillary gas chromatography/mass spectrometry (GC/MS)
and selected ion monitoring (SIM), mass spectra were
obtained for each analyte. Samples were prepared by
hydrolysis where applicable, organic solvent extraction, and
derivatization where necessary. Confirmation was achieved
by comparing abundance of major ions and retention time of
the total ion current (TIC) of an analyte with those of the
appropriate analytical standard. Quantitation was achieved
and calibration curves derived by obtaining the molecular ion
ratios of that analyte/internal standard (IS) over a
concentration range of 10-300 ng/mL (0.16-4.0 ng total
injected into GC/MS). The overall extraction efficiency for
these analytes ranged from 53~ to 96~ Statistically
significant cut-off values Oo< 0.01) were obtained for each
analyte. The slope, y-intercept, and coefficient of
determination (r 2) were calculated for each analyte. All of the
GC/MS methods were extensively tested against urine
samples determined positive or negative by immunoassay
(IA) and are now used in our laboratory.

Introduction

In the past few years urine drug abuse testing has entered the
workplace thus placing the burden of both scientific validity and
legal defensibility upon the analytical toxicologist. This responsibility has led to the conclusion that gas chromatography coupled
with mass spectrometry ( G C / M S ) offers the best means for unequivocal identification of drugs of abuse or their metabolites
in biological materials (1). Although G C / M S and GC methods
are available for marijuana (2-5), morphine and codeine (6-8),
amphetamine and methamphetamine (2,9), cocaine, benzoylecgonine and ecgonine methylester (2,10-14), and phencyclidine
(15-17), rapid, reliable, and sensitive assays requiring small

* Correspondence should be addressed to S.J. Mul~ at the OSAS address

102

volumes of urine were needed in our laboratory in order to meet

the daily requirements of high volume drug abuse testing. Consequently we d e v e l o p e d methods using capillary gas
chromatography with electron impact mass fragmentography
( E I / M F ) operating in the selected ion monitoring mode (SIM).
In this report we present these methods in detail and demonstrate
quantitative use through the addition of an internal standard
(IS) to the assays.

Materials and Methods

Equipment
A Model 5890A Hewlett-Packard G C with 5970B mass selective detector was used for the analyses. The data system in use
was the HP 310 computer with the H P 9133H Winchester disc
drive. The manufacturer provided tile operating application and
three dimensional display software for use with the HI:' 59970C
M S / M S D Chemstation. The MSD was operated in the electron
impact (El) mode at 70 eV with an ion source temperature of
280~ and an m/z range that varied with the drug or metabolite
analyzed (see below). The instrument was autott, ned daily with
perfluorotributylamine (PFTBA). The G C / M S intcrface temperature was 280~ A fused-silica capillary column (HP-190916-312), 12.5 m 0.2 m m i.d., consisting of cross-linked
dimethyl silicone was used. The helium flow rate was 0.65
m L / m i n at a linear velocity of 34.7 cm/s. The operating temperature of the column varied with the analytes (see below) and
the injection was splitless.
Urine drug screening
The EMIT d . a . u J ' (Syva Corp.) was utilized for most drugs,
and samples indicating drugs at or above cut-off levels were
analyzed by the G C / M S procedures described below. The
radioimmunoassay used was either the Abuscreen"; (Roche
Diagnostics) or the U r i n e - T H C (lmmunalysis Corp.).
Drug analysis, extraction, derivatization, and GC/MS

Marijuana." ll-nor-gxg-tetrahydrocannabinol-9-carboxylic
acM
(THC-COOH). To a 15-mL silanized centrifuge tube, add 0.5
mL urine and 50 ng (in 0.05 mL ethanol) o f 9-carboxy-llnor-2~%tetrahydrocannabinol-5-"H3 (trideuterated THC-COOH)
(Research Triangle Institute). Alkalinize with 0.5 mL of 5~

Reproduction (photocopying) of editorial content of this journal is prohibited without pubhsher's permission.

Journal of Analytical Toxicology, Vol. 12, March/April 1988

NaOH by mixing on a vortex. Hydrolyze for 15 min in a boiling

water bath; place in freezer to cool. Add 3-4 drops of conc.
HCI to lower pH (<2) and add 3 mL of hexane/ethyl acetate
(7:1, v/v). Shake sample for 10 min on Eberbach shaker at low
speed. Centrifuge at 2500 rpm for 5 min to separate the aqueous
and organic phases and transfer the upper organic phase to a
clean, 15-mL centrifuge tube. Evaporate extract to dryness in
water bath (50-60~ under nitrogen. Derivatize sample by
adding 200 #L of tetramethylammonium hydroxide (TMAH)/
dimethylsulfoxide (DMSO) (1:1, v/v) plus 100 #L of iodopropane. (0.2M TMAH in methanol, Pierce Chem. Co.; DMSO
and iodopropane, Sigma Chem. Co.). Cap the tube tightly and
heat for 15 min at 60~ (heating block or hot water bath). Allow
to cool at room temperature. Add 3 mL of 0.IN KOH plus 3
mL of hexane and shake for 5 min. Centrifuge 5 rain, then
transfer upper organic phase to a clean, 15-mL silanized tube
and evaporate to dryness under N~. Dissolve residue in 25 #L
of ethyl acetate and inject 1/~L into the GC/MS. An alternative
derivatization method is to add to each tube (after hydrolysis,
extraction, and evaporation) 50 #L of n-methyl-n-trimethylsilyltrifluoroacetamide (MSTFA, Regis Chemical Co.), cap the tube
tightly and heat for 15 rain at 60~ allow tube to cool at room
temperature, and inject 1 p.L directly into the GC/MS. Standard control urine (analyzed simultaneously) consisted of 50 ng
(50 #L) of THC-COOH (Research Triangle Institute) plus 50
ng of deuterated THC-COOH (50 #L) as the internal standard
in 0.5 mL of blank urine. The injector temperature is 250~
column temperature 165~ (30 s) to 250~ at 50~
Selected
ion monitoring (SIM) is utilized to detect major ion peaks of
m/z 341,385, and 428 for the TMAH-catalyzed derivative of
THC-COOH (dipropyl ester) and m/z 344, 388, and 431 for
the TMAH derivative of the deuterated THC-COOH (IS). Major
ion peaks of m/z 371, 473, and 488 were obtained for the
MSTFA derivative of THC-COOH (TMS) and m/z 374, 476,
and 491 for the MSTFA derivative of the deuterated THCCOOH (IS).
Cocaine (benzoylecgonine). To a 15-mL silanized tube, add
0.2 mL of urine and 100 ng (0.1 mL) of the internal standard
scopolamine (Sigma Chem. Co.) and bring to final volume of
1 mL with distilled water. Add 0.3 g of solid NaHCO~/K2COj
(2:1) to the sample (pH 9) and mix on a vortex. Extract sample
by adding 3 mL of chloroform/isopropanol (90:10, v/v) and
shake for 10 rain at low speed on an Eberbach shaker. Centrifuge 5 min at 2500 rpm and then siphon off upper aqueous
phase. Add 1 mL of water, shake, and centrifuge as described.
Transfer the lower organic layer with a Pasteur pipette to a clean,
15-mL tube, being very careful not to carry over any of the
aqueous phase. Evaporate to dryness under N~ in a heated water
bath (50-60~
Derivatize sample by adding 50 #L of pentafluoropropionic anhydride (PFPA, Pierce Chem. Co.) and 25
#L of pentafluoropropanol (PFP, PCR/SCM Chem.). Heat for
15 min in a heating block at 60~ and evaporate to dryness under
N2. Dissolve residue in 25 #L ethyl acetate and inject 1 #L into
GC/MS. Standard control urine (analyzed concurrently) consisted of 50 ng (50 #L) of benzoylecgonine (Alltech-Applied
Science) and 100 ng (100 #L) of scopolamine (IS) in 0.2 mL
of blank urine adjusted to l mL with water. The injector
temperature is 250~ column temperature 165~ (l.0 min) to
225~ at 30~
SIM is used to detect major ion peaks of
m/z 272, 300, 316, and 421 for the PFP derivative of benzoylecgonine and m/z 108, ll9, 138, and 449 for the PFP
derivative of scopolamine.
Morphine and codeine. To a 15-mL silanized tube add 0.2
mL of urine, 250 ng (25 ttL) of nalorphine as the internal standard, and 2-3 drops of cone. HCI. Cap and hydrolyze 30 min
at 100~ in a heating block or boiling water bath. Cool and
bring to final volume of l mL with distilled water, add about
0.3 g of solid NaHCO3/K2CO3 (2:1), and mix by vortexing. Extract sample with 3 mL of chloroform/isopropanol (90:10, v/v)

by shaking for 10 min at low speed on an Eberbach shaker and

centrifuge 5 min at 2500 rpm. Remove upper aqueous phase,
add 1 mL of distilled water to the tube, shake and centrifuge
as described, then carefully transfer the lower organic phase
(being careful not to carry over the aqueous phase) to a clean,
15-mL tube and evaporate to dryness under N2 in a heated water
bath (50-60~
Derivatize the sample by adding 25 #L of pentafluoropropanol (PFP), 50 #L of pentafluoropropionic anhydride
(PFPA, Pierce Chem. Co.), heat for 15 min in a heating block
at 90~ evaporate to dryness under N2, dissolve residue in 25
#L of ethyl acetate, and inject 1 #L into GC/MS. A standard
control urine (analyzed simultaneously) contains 50 ng (50 ~L)
of morphine, 50 ng (50 #L) of codeine, and 250 ng (25 p.L) of
nalorphine in 0.2 mL of blank urine. The injector temperature
is 250~ the column temperature 165~ (1 min) to 240~ at
30~
Using SIM, the major ion peaks for the PFP derivatives of morphine are m/z 357,414, and 577; for codeine m/z
282, 388, and 445; and for nalorphine (IS) m/z 357,440, and 603.
Amphetamine and methamphetamine. To a 15-mL silanized
tube add 0.2 mL of urine, 50 #L of 50~ NaOH, and 25 ng (25
#L) of phenylcyclohexylamine(Abbott Labs) as internal standard. Bring to final volume of 1 mL with distilled water. Extract
the sample by adding 3 mL of chloroform/isopropanol (90:10,
v/v) and shaking for 10 min at low speed on an Eberbach
shaker. Centrifuge for 5 min at 2500 rpm. Transfer lower organic
phase to a clean tube and add 25/~L of glacial acetic acid and
50 #L of trifluoroacetic anhydride (TFAA, Pierce Chem. Co.).
Evaporate to dryness under N~ in a heated water bath (40-50~
Reconstitute residue in 0.5 mL of CHCL3/C3H,OH and add
50 #L TFAA by mixing on vortex, then evaporate to dryness
as described. Dissolve residue in 25 p.L of ethyl acetate and inject
1 p.L into the GC/MS. Standard control urine (analyzed at the
same time) containing 25 ng (25/~L) of amphetamine, 25 ng
(25 ~L) of methamphetamine, and 25 ng (25 #L) of phenylcyclohexylamine (IS) in 0.2 mL of blank urine is made alkaline and
brought to a final volume of 1.0 mL with distilled water. The
injector temperature is 250~ the column temperature 110~
(0.5 min) to 140~ at 25~
SIM is used to obtain major
ion peaks for the TFA derivatives of amphetamine at m/z 115,
118, and 140; for methamphetamine at m/z 110, 118, and 154;
and for phenylcyclohexylamine (IS) at m/z 189,217, and 271.
Phencyclidine (PCP). To a 15-mL silanized tube add 0.2 mL
of urine, 50 #L of 50~ NaOH, and 50 ng (50 #L) of ketamine
(Alltech-Applied Sci.) as internal standard and bring to final
volume of 1 mL with distilled water. Extract with 3 mL of
chloroform/isopropanol (90:10, v/v) as described previously
and transfer the lower organic phase to a clean, 15-mL tube
and evaporate under N2 in a heated water bath (50-60~
Dissolve residue in 25 #L of ethyl acetate and inject 1 #L into
the GC/MS. A standard control urine (run concomittantly) containing 25 ng (25 #L) of PCP and 50 ng (50 #L) of ketamine
in 0.2 mL blank urine is made alkaline and brought to a final
volume of 1.0 mL with distilled water. The injector temperature
is 250~ the column temperature 100~ (1 min) to 190~ at
20~
The major ion peaks using SIM are m/z 186, 200, 242,
and 243 for PCP and m/z 180, 209, and 237 for ketamine (IS).
Cocaine (C) and ecgonine methyl ester (EME). Utilizing the
method described for PCP, both C and EME may be identified
in an extract from 0.2 mL of urine. A standard control urine
(analyzed simultaneously) should contain 100 ng (100 #L) of
cocaine and 100 ng (100 #L) of ecgonine methyl ester in 0.2 mL
of blank urine. Alternatively, if quantitation is required, 50 ng
(50/~L) of ketamine may be added to the urine and extracted

103

Journal of Analytical Toxicology, Vol, 12. March/April 1968

as described. Injector temperature is 250~ and column temperature 100~ (1 min) to 190~ at 20~
hold 8.5 min, then
to 225~ at 50~
The abundant ion peaks monitored (SIM)
for cocaine are m / z 82, 182, and 303 and for ecgonine methyl
ester are m / z 82, 96, and 199.

Results and Discussion

Although these assays were primarily developed to confirm
positive results of screening by immunoassay, they may also
be quantitated by the addition of an internal standard (IS) to
each analyte assay. Such calibration curves were obtained by
adding known quantities of each analyte to control urines along
with the appropriate internal standard. The ratio of the molecular ion of the analyte to the molecular ion of the internal standard (except for amphetamines) was then determined and plotted
against the concentration (ng/mL or ng total) of the analyte.
The results obtained with the TMAH-catalyzed derivative of
THC-COOH (15-200 ng/mL, 0.3-4 ng total) appear in Figure
IA. Excellent linearity was obtained with the molecular ion
ratios m / z 428/431, The sensitivity of this assay (cut-off) was
equal to or greater than 15 ng/mL (0.3 ng total), The results
obtained with the MSTFA derivative of THC-COOH (10-100
ng/mL, 0.1-1 ng total) appear in Figure 1B with linearity achieved
using the molecular ion ratio m / z 488/491. The sensitivity (cutoff) obtained was equal to or greater than 10 ng/mL (0.1 ng
total) of the MSTFA derivative TMS-THC-COOH. The slopes
and the y-intercepts as well as the coefficient of determination
(rz) were calculated, The latter value for the TMAH derivative
was 0.99067 and for the MSTFA derivative was 0.98094,
In Figure 2 appear the calibration curves and the slope and
y-intercepts for the PFP derivatives of morphine and codeine.
Linearity was achieved with both of these drugs over the concentration range of 25-300 ng/mL (0.2-2.4 ng total). The sensitivity (cut-off) of ~he GC/MS assays for both analytes was

~ ~ 1.0]

TMAH-THC-COOH
SLOOIE:O0045

.,.c,o,: 001.

at least 50 ng/mL (0.4 ng total) with molecular ion ratios of

m / z 577/603 for morphine and m / z 445/603 for codeine with
nalorphine as the internal standard for both drugs. The coefficient of determination for morphine was 0.99724 and for
codeine was 0.99491.
In Figure 3 appear the quantitative calibration curves for the
TFA derivatives of amphetamine and methamphetamine, including their slopes and y-intercepts. Linearity with these drugs
was obtained over the range of 25-250 ng/mL (0.2-2 ng total).
The sensitivity level (cut-off) was equal to 25 ng/mL for amphetamine and methamphetamine at select ion ratios of m / z
140/271 and m / z 154/271, respectively, with phenylcyclohe
amine (PCHA) as internal standard. The coefficient of determination for amphetamine was 0.99325 and for methamphetamine 0.99439.
The calibration curve for phencyclidine (PCP) over the concentration range of 10-1(90 ng/mL (0.08-0.8 ng total) appears
in Figure 4A. Linearity was achieved over this range with a maximum cut-off sensitivity of I 0 ng/mL using molecular ion ratios
of m / z 243/237 for phencyclidine and the internal standard
ketamine, in Figure 4B appears the calibration curve for the
PFP derivative of benzoylecgonine, the primary metabolite of

PFI~PHINE
~S'E-OPE=O.O015

~.,,,o

0.4 7

INTERCEPT=-0.O001

J,,,,,e ~ O D E I N E
/.,/~
SLOPE:O.O012
,NTERCEPT=O.O037

v.,^
"/

"~

9 ~j

ol
o_o

T I I

02550

250 300 4~--ng/ml

100 150

(0.2)(0.4) (0,8) (12)

(Z0) (2.4) C)------r--TOTALng

Figure 2. Calibration curve for PFP derivative of morphine and codeine

with molecular ion ratio of m/z 577/603 for morphine/IS (nalorphine)
vs. ng/mL of morphine added to urine (total ng analyzed). Molecular
ion ratio of m/z 445/603 for codeine/IS (nalorphine) vs. ng/mL of codeine
added to urine (total ng analyzed). Each data point is the mean value
of at least 3 determinations.

A i~o,4

~03)(06)<10)
N

Bi

(zo)

(3~0)

MSTFA-THC-COOH
9 SLOPE:O.0207

6,O5~0-

~
~

~ ~ 4.0~ . ~-

,o

~ ~ ~" -

o.s-

z~O.O 0

10 20 30 40 50 60
(0.1) (0,2) (03) (04) (oS) ((16)

80
(06]

7
Q

~ TFA-AMPHETAMINE

E6PE,o.o,o2o

i INTERCEPT0 0624

1QO~ n g / ~
(10( ~TOTALnQ

50

Figure 1. (A) Calibration curve for TMAH-catalyzed derivative of 11-norAg-THC-9-carboxylicacid OHC-COOH). Molecular ion ratio of m/z 428/431
for analyte/IS (deuterated THC-COOH) vs. ng/mL of analyte added to
urine (total ng analyzed). (B) Calibration curve for MSTFA derivative of
THC-COOH. Molecular ion ratio of m/z 488/491 for analyte/IS (deuterated
THC-COOH) vs. ng/mL of analyte added to urine (total ng analyzed).
Each data point on each curve is the mean value of at least 3
determinations.

104

T'FA-METM/ I f ~'k~HETAMINE
SLOPE~O.0234
INTIERCEPT=O.1730

(4o~

75 1OO 125 150

2.$
(O2) (04) (0,6) (08) (10) (t2)

200
(1.6)

250 =
(ZO) 9

n0/r~

TOTALng

Figure 3. Calibration curve for the TFA derivative of amphetamine with

select ion ratio of m/z 140/271 for amphetamine/IS (phenylcyclohexylamine) vs. ng/mL of amphetamine added to urine (total ng analyzed).
Select ion ratio of m/z 154/271 for methamphetamine/IS (phenylcyclohexylamine) vs. ng/mL of methampbetamine added to urine (total ng
analyzed). Each data point is the mean value ol at leasl 3 determinations.

Journal of Analytical Toxicology, Vol. 12, March/April 1988

cocaine. Linearity of response was obtained over the concentration range 25-125 n g / m L (0.2-1.0 ng total). The assay was
sensitive (cut-off) at 50 n g / m L (0.4 ng total) with molecular
ion ratios of m / z 421/449 for benzoylecgonine and the internal
standard scopolamine. The slopes and y-intercepts of these
analytes also appear in Figure 4. The coefficient of determination calculated for PCP was 0.99749, and for benzoylecgonine
was 0.98588.
In Figure 5A appear the mass fragmentation spectra of the
MSTFA derivative of T H C - C O O H (500 ng), of which 10 ng
(1 ~L) was injected into the G C / M S and scanned from m / z 70
to 500. A retention time of 7.345 min was obtained for the total
ion current (TIC). The selected ion current profiles are illustrated
in Figure 5B showing the abundant ions for TMS-THC-COOH
of m / z 371,473,488 (m + ) after extraction of 50 ng from 0.5
mL urine and injection of 1 ng in 1 #L into the GC/MS. In
Figure 5C appear the mass fragmentation spectra of the TMAHcatalyzed derivative of T H C - C O O H (500 ng), of which 20 ng
(1 #L) was injected into the G C / M S and scanned from m / z 60
to 450. For the TIC, a retention time of 10.395 was obtained.
The selected ions monitored for the TMAH-catalyzed derivatives
of THC-COOH were m / z 341,385, and 428 (m + ). They appear
as ion currents in Figure 5D. Either derivative of THC-COOH
may be used effectively to identify and confirm marijuana use.
Although the MSTFA derivative of THC-COOH was easy to
prepare, it unfortunately readily contaminated the ion source
of the instrument. In order to avoid this problem, the dipropyl
derivative of THC-COOH was prepared with TMAH and
iodopropane, even though additional steps were required.
The pentafluoropropionic anhydride derivative of morphine
(40 ng in 1 #L) was injected into the GC/MS and scanned over
the range m / z 70-625 for the mass spectra (Figure 6A). A retention time of 5.324 min was observed for the TIC of morphine.
Morphine (50 ng) extracted from 0.2 mL of urine, of which 2
ng in 1 #L was scanned in the SIM, revealed three major ion
peaks at m / z 357,414 (base peak), and 577 (m + ) (Figure 6B).
The mass fragmentation spectra for the PFP derivative of
codeine (40 ng in 1 #L) over the range scanned m / z 55-450
appear in Figure 6C. A retention time of 5.593 min for the TIC

of codeine was observed. The molecular ion m / z 445 and 388,

as well as the base peak ion rn/z 282, were monitored after extraction of 50 ng codeine from 0.2 mL of urine with 2 ng in
1 #L injected into the GC/MS. The codeine ion current profile
appears in Figure 6D.
The mass fragmentation spectra for the trifluoroacetic anhydride derivative of amphetamine and methamphetamine
appear in Figure 7A and 7C, respectively. These two drugs at
a concentration of 20 ng in 1 #L were each scanned over the
range m / z 40-250. A TIC retention time of 2.527 min was
obtained for amphetamine and 3.687 min for methamphetamine. From 0.2 mL of urine, 12.5 ng of amphetamine plus
12.5 ng of methamphetamine was extracted into chloroform/
isopropanol (90:10, v/v) with a subsequent derivatization with
TFAA as described and 1 #L injected into the G C / M S (0.5 ng
of each analyte) and scanned for selected ions. In Figure 7B
appear the ion current profiles obtained at rn/z 115, 118, and
140 (base peak) for amphetamine. The molecular ion (m + ) of
231 is barely detectable and therefore of little use in quantitation
or identification. For methamphetamine (Figure 7D), major
peaks and abundances were obtained at m l z 110, 118, and 154
(base peak). The methamphetamine molecular ion 245 was
barely detectable and therefore of little quantitative value.
In Figure 8A appear the mass fragmentation spectra for the
PFP derivative of benzoylecgonine, of which 20 ng (1 #L) was

Scam 03 (7,345 mln) or V3:TNC.D

//

s.mE+i P+

+,o+++/

t++

~+t

./

,,,|
100

+. . . . .

?i

290

,,
300

400

M&~s/Charge
Ion 32t,gg amu. Crem V3:T U/HC.D

2-gE4 t
1.5E4 t

10000]

%
7.1

~a

;~

"

PHENCYCMO~E(PCP)

SLO~=O.OmS

INTERCEPT=O.068 9

,.s

C ~2.eE4~

62

O.~,,! ,,,
100

~ ++i
6000]
5000]

z~m-m 2 . 0 ~

147

193

SLOPIE-'O.OO74
INTERCEPT:-O.O811

\
tO.4
Time

(0.4)

10.61

(0.81

(1.01 II

[i~!l

413

299

S~ "

.11 ,~ ,,

300

H~g~/Charge
]on 34).O8 amu. from V3:TTHC3U.D
Ion 385.~ imu. from V]:TTHC3U,D
Ion 42Bl~[~,mu from V3:TTHC3UD

2000~-'
1000

7.G

of V3:TTHC.D

207

7,5

7+4

200

L0,2
(0,(~1 (0.2)

m~n)

.1 H ,.,,,I ,. ,.,

~'t

~\

7.3

Time (mln, }

Sczm 63 (IO,395

~~
4.~[4]

0.5

~= o.o

7.2

...........
Lg.~
(mln I

, ,J
400

+......
10.~

tl.O

TO'[~I_i~

Figure 4. (A) Calibrationcurve for phencyclidine (PCP). Molecular ion

ratio of mlz 2431237 for PCP/IS (ketamine) vs. ng/mL of PCP added
to urine (total ng analyzed). (B) PFP derivative of benzoylecgonine.
Molecularion ratio of mlz 4211449 for benzoylecgoninellS(scopolamine)
vs. ng/mL of analyte added to urine (total ng analyzed). Each data point
on each curve is the mean value of at least 3 determinations.

Figure 5. (A) Mass spectra of TMS-THC-COOH. (B) The selected ion

current profile for TMS-THC-COOHof m/z 371,473, and 488 with percentage abundancein reference to the major ion m/z 371 (100), 26.7, and
18.4, respectively.(C) Mass spectra of the TMAH-catalyzedTHC-COOH.
(D) The selected ion current profile for TMAH-THC-COOH of m/z 341,
385, and 428 with percentage abundance ions in reference to major
ion m/z 341 (100), 27.5 and 17.1, respectively.

105

Journal

injected into the G C / M S and scanned from m / z 70 to 450. A

retention time of 4.543 min (TIC) was observed. From 0.2 mL
of urine (Figure 8B), 50 ng of benzoylecgonine was extracted
and derivatized with PFPA. One microliter (2 ng) was injected
into the GC/MS and scanned to obtain selected ion current profiles of m / z 272, 300 (base peak), 316, and 421 ( m + ) .
In Figure 8C appear the mass fragmentation spectra for phencyclidine (PCP) over the range m / z 70-300 for 10 ng in 1 /~L.
A retention time of 8.150 min (TIC) was observed for PCP.
PCP was analyzed in the SIM mode by extracting 0.2 mL of
urine containing 20 ng of PCP. Final concentration of PCP
injected into the G C / M S was 0.8 ng in 1-#L. The selected ion
current peaks monitored (Figure 8D) for PCP were m / z 186,
200 (base peak), 242, and 243 (m + ).
The methods employed to extract and identify PCP may also
be utilized to extract cocaine and its metabolite ecgonine methyl
ester (EME). Selected ion monitoring for cocaine provided
values o f m / z 82, 182 (base peak), and 303 (m + ) at a retention
time of 10.353 min for the TIC. For ecgonine methyl ester,
selected major ion peaks were m / z 82 (base peak), 96, and 199
(m + ). A retention time of 4.341 min for EME total ion current
was also obtained. Quantitation may be achieved for these
analytes by adding ketamine (50 ng) as an internal standard and
determining the ratios of the molecular ion of cocaine or EME
metabolite to ketamine.
Two primary extracting solvents were used: hexane/ethyl

Scan

L3 ( 5 . 3 2 4

mln)

of Analytical

Scan 39 ( 2 5 2 7
4, gE5

114

"

302

20~

352

Ion 4 t 4 ~

amu.

30~
Has~/Charge

577

/ 5~ (/

4gg

500

60

#rom DAT~:H4 9 8 D

225

Ion
Ion

1/65

268
'

341

68081
4800 ]"
28Of, J

5.$8

22D

',

S '"'-.

5.6g

. . . . . . . . . . . . . .

r~me I m l n l

42

LIg

rain) Of

2,0

].~

D,qTA:R HETN,[,

,4FI
8

,oo ,,, oo .2XL"g4.o,o

4 80~
0 I

"~\

/
....

2,0
Time ( m l n . )

490

/
/ "

27,

388

5,28

Figure 6. (A) Mass spectra of the PFP-morphine. (B) The ions in the
selected ion current profile for PFP-morphine were m/z 414, 577, and
357, with percentage abundance ions in reference to the base peak of
m/z 414 (100), 20.6 and 9.5, respectively. (C) Mass spectra of the PFPcodeine. (D) The ions in the SICP for PFP-codeine were m/z 282,445,
and 388 with percentage abundance ions in reference to the base peak
of m/z 282 (100), 55.6 and 6.0, respectively.

106

/I

!.3808 ]

3 80981

1.~2

188 120 14~1 168 tO0 tOO

Has~/Charge
14~.~O ainu. ~rom D~TA:ALJI9 4 ~)
1;8 8~s~,,j
#tom [JI4Tr
4 i
/

,004

"-1

445

2~8
]g(]
klass ,Charge
Ion 282 t~O ainu. from DRTA:H4 9,e ~
Ion 4 4 5 . ( ~ amu. frou, [JRTA:H4 9 / 8 . [ I
'

88

SCan 128 ( 3 . 6 8 ?

!"
1282
t/19

14g

]
1.0[5

,,51
"-=

2.4

2. gg5

~RTRI r4 R~!PN L)

2.oEsI
ecs] . /,~

Scan 3~ 15597 ~Inl .o,f DATR:H2.D

1988

B
.

o#

~
'118

9l

gg~Jg

m:~

~. oc5

i(~08~i

V o l . 12, M a r c h / A p r i l

acetate (7:1, v/v) for THC-COOH and chloroform/isopropanol

(90:10, v/v) for all other analytes. The extraction efficiencies
or recoveries obtained using the method described by B.D. Paul
et al. (5) with the aforementioned methods were 84~ for TMSTHC-COOH, 73% for the TMAH derivative of THC-COOH,
96% for PFP-morphine, 91 ~ for PFP-codeine, 61~ for TFAamphetamine, 53% for TFA-methamphetamine, 65~ for
PFP-benzoylecgonine, and 87~ for PCP.
The sensitivity of these assays may be increased in some cases
by increasing the volume of urine extracted 2- to 3-fold, increasing the volume injected into the G C/ MS (2-3 p.L), or decreasing
the volume of ethyl acetate used to dissolve the residue following
extraction and derivatization of the analyte. These changes,
however, may also affect the signal to background ratio.
Using the methods described in this report, the statistically
significant cut-off value for each analyte Lo<0.01) was the
lowest concentration analyzed (Figures 1-4) except for morphine, codeine, and benzoylecgonine where the value was 50
ng/mL at the p<0.01 level of significance.
A cursory examination revealed that the stability of the PFP
and TFA derivatives as well as the THC-COOH derivatives normally did not appear to exceed 72 h at - 2 0 ~
The internal standards used in these assays were chosen for
functional ease of application, the primary purpose being quantitation with confirmation of the positive sample. Obviously,
deuterated standards of the parent drug or metabolite make the

Of DATA:H2 D

G,OE58OES]

Toxicology,

'~ 1

Ion ll(~.Oe
Ion 1 1 8 . 8 8

a~:~. from DATk:A-URMPH. D

~LI~, from DRTIq~R-URHPH.D

/\

1808

~\

3.5

4.g
Tlme [mln. )

4.5

Figure 7. (A) Mass spectra for TFA-amphetamine. (B) The selected ion
current profile for TFA-amphetamine of m/z 140, 118, and 115 with
percentage abundance ions in reference to the base peak of m/z 140
(100), 65.9 and 9.1, respectively. (C) Mass spectra for TFA-methamphetamine. (D) Major ions in the SICP for TFA-methamphetamine
were m/z 154, 110, and 118 with percentage abundance ions in reference
to the base peak of m/z 154 (100), 33.3 and 31.8, respectively.

Journal of Analytical Toxicology, Vol. 12, March/April 1988

best possible G C / M S internal standards, if available and if in

pure form. Unfortunately, the deuterated standards are very
expensive and are controlled, thus adding more controls to an
already heavily regulated process.
Our criteria for confirmation of a positive urine sample by
comparison with an extracted standard analyte are (1) a retention time for the total ion current (TIC) with an acceptable
variance of less than _ ! 070, (2) selected ions monitored with
an acceptable variance o f +_20~ in the ion ratios, and (3) computer match quality of usually 98~ or greater.
These methods are sensitive, rapid, reproducible, and currently in routine use in our laboratory. One trained technician
working through a normal shift may prepare and submit about
25 urine samples to G C / M S analysis. Obviously, the number
of technicians, the availability of G C / M S instruments, and the
use of automatic sampler devices will enormously enhance the
daily confirmation capability.

Acknowledgments
The authors are extremely grateful to Ms. Ann DePace for
technical assistance, to Mr. Arthur Kramer for statistical analysis o f the data, to Mr. Jed Shaw for the illustrations, and to
Mrs. Elizabeth McLeod for typing the manuscript.

so,~ , ,

.....

5 BE4J 62

I-,.,c*~

.k,

.....

200
H~ss/Ch~rge

IOB

Ion 3 0 0 . 0 0
Ion 4 2 1 f I ~

00~101

,2,

][ . . . . .
300

4e0

i.~)u, f r o m DFII'FhAI.D
imu. from DRTA~AI.D

I
B
4.s

4.5
Scan

4,7

5,0E41
4 . BE4]

C p o ,1
~2.0s

4.8

4.9

142 (e.15g mlnl of D~Tg:PCP.D

~oa

243

z,z !

41

Gu, .,.u.
50

5.0C*1
,4, BE41

100

150
200
Hi~/Ch~rge
Ion 200.~ ~mu. fPOm DRTR:P PCPU.B
Ion 242.00 ~ . from DRTR:P:PCPU.D
IOn 243.~ ~mu\. r
DRTR:P PCPU.D
[On I ~ D R T R r P 2 P C P U . D

o :i:iii
0--"

e.t

e.2

Time

(~ln,)

9.3

e.4

Figure 8. (A) Mass spectra for PFP-benzoylecg0nine.(B) The selected

ion current profile for PFP-benzoylecgoninewas m/z 300, 421. 316,
and 272 with percentageabundanceions in referenceto the base peak
of m/z 300 (100), 30.6, 16.5, and 13.4, respectively. (C) Mass spectra
for phencyclidine(PCP). (D) The major ions in the SICP for PCP were
m/z 200,242, 243, and 186 with percentageabundanceions in reference
to the basepeak of m/z 200 (100), 38.5, 34.0, and 21.8, respectively.

References
1. Urine Testing for Drugs of Abuse, R.L. Hawks and C.N Chiang,
Eds. NIDA Research Monograph No. 73, DHHS publication
number (ADM) 87-1481, U.S. Government Printing Office, 1986,
pp. 30-42.
2. GC/MS Assays For Abused Drugs in Body Fluids, R.L. Foltz,
A.F. Fentiman, and R.B. Foltz, Eds. NIDA Research Monograph
No. 32, DHHS Publication number (ADM) 80-1014, U.S. Government Printing Office, 1980, pp. 5-198.
3. T.S. Baker, J.V. Harry, J.W. Russell, and R.L. Meyers. Rapid method
for the GC/MS confirmation of 11-nor-9-carboxy-Ag-tetrahydrocannabinol in urine. J. Anal. Toxicol. 8 : 2 5 5 - 5 9 (1984).
4. H.H. McCurdy, L.J. Lewellen, L.S. Callhan, and P.S. Childs.
Evaluation of the ion trap detector for the detection of 11-norL@-THC-9-carboxyfic acid in urine after extraction by bonded-phase
adsorption. J. Anal. Toxicol. 10:175-77 (1986).
5. B.D. Paul, L.D. Mell, J.M. Mitchell, and R.M. McKinley. Detection
and Quantitation of urinary 11-nor-delta-9-tetrahydrocannabinol9-carboxylic acid. A metabolite of tetrahydrocannabinol by capillary gas chromatography and electron impact mass fragmentography. J. Anal. Toxicol. 11 : 1-5 (1987).
6. N.B. Wu Chen, M.I. Schaffer, R.I. Lin, and FLJ. Stein. Sirnuttaneous quantitation of morphine and codeine in biological
samples by electron impact mass fragmentography. J. Anal. Toxicol. 6:231-34 (1982).
7. J. Fehn and G. Meggs. Detection of O~-monoacetylmorphine in
urine samples by GC/MS as evidence for heroin use. J. Anal,
Toxicol. 9:134-38 (1985).
8. B.D. Paul, L.D. Mell, J.M. Mitchell, J. Irving, and A.J. Novak.
Simultaneous identification and quantitation of codeine and morphine in urine by capillary gas chromatography and mass spectroscopy. J. Anal. Toxicol. 9 : 2 2 2 - 2 6 (1985).
9. N. Narasinhachari and R.D. Friedel. Quantitation of biologically
important primary amines as their iosthiocynate derivatives by
gas chromatography using nitrogen detector and validation by
selected ion monitoring. Clin. Chim. Acta 110:235-43 (1981).
10. M.J. Kogan, K.G. Verebey, A.C. DePace, R.B. Resnick, and S.J.
Mul$. Quantitative determination of benzoylecgonine and cocaine
in human biofluids by gas-liquid chromatography. Anal. Chem.
49:1965-69 (1977).
11. D.M. Chinn, D.J. Crouch, M.A. Peat, B.S. Finkle, and T.A. Jennison.
Gas chromatography-chemical ionization mass spectrometry of
cocaine and its metabolites in biological fluids. J. Anal 7oxicol.
4:37-42 (1980).
12_ J.J. Ambre and 1". Ruo. Ecgonine methyl ester. A major metaboiite
of cocaine. J. Anal, Toxicol. 6 : 2 6 - 2 9 (1982).
13. P.M. Folk and B.C. Harrison. Use of DB-1 capillary columns in
the GC/FID analysis of benzoylecgonine. J. Anal Toxicol, 9:
273-74 (1985).
14. J. Ambre. The urinary excretion of cocaine and metabolites in
humans: A kinetic analysis of published data. J. Anal. Toxicol.
9" 241-45 (1985).
15. E.J. Cone, W. Buchwald, and D. Yousefnejad. Simultaneous determination of phencyclidine and monohydroxylated metabolites in
urine of man by mass fragmentography with methane chemical
ionization. J. Chromatogr. Biomed. Appl. 223:331-39 (1981).
16. J.R. Woodworth, M. Mayersohn, and S.M. Owens. Quantitative
analysis of phencyclidine and metabolites by capillary column gas
chromatography. J. Anal Toxicol. 8 : 2 - 6 (1984).
17. E.J. Halsztynska and E.E Domino. Quantitation of phencyclidine,
its metabolites and derivatives by gas chromatography with
nitrogen-phosphorus detection: Application for in vivo and in vitro
biotransformation studies. J. Anal Toxicol. 10:107-15 (1986).

Manuscript received June 2, 1987;

revision received October 19, 1987.

107