Current - Leco

March 2009
Nonderivatized Drug-Screen Analysis in Urine
with Automated Solid Phase Microextraction and
Comprehensive Two-Dimensional GC–TOF-MS
This research investigates the potential use and practical application of comprehensive multidimensional gas
chromatography–time-of-flight-mass spectrometry (GC × GC–TOF-MS) in forensic toxicology analysis for a
diverse range of illicit drugs in urine without sample derivatization. Conventional methods for the analysis
of drugs in urine use time-consuming sample derivatization for GC–MS analysis to enhance chromatographic
performance. This study utilizes GC × GC to increase peak capacity and resolution in combination with TOF-
MS detection followed by data processing with deconvolution software algorithms for the identification and
quantitation of nonderivatized drugs in urine.
John Heim and Scott Pugh
Current trends indicate a global rise in the abuse of
both prescription and illicit drugs. In the “National
Drug Threat Assessment Summary for 2009,” published
by the National Drug Intelligence Center, the U.S.
Department of Justice reported that more than 35 million
individuals used illicit drugs or abused prescription drugs
in 2007. In September of 2008, nearly 100,000 inmates in
federal prisons were convicted for drug offenses, representing
approximately 52% of all federal prisoners (1). Forensic
crime laboratories are continually challenged to find faster
and more efficient analytical methods and instrumentation
to cope with ever-increasing sample backlogs. This experimental
research explores the capabilities and utilization of
automated sample handling using solid-phase microextraction
(SPME) for nonderivatized urine samples along with
comprehensive two-dimensional gas chromatography (GC ×
GC) coupled with time-of-flight-mass spectrometry (TOF-
MS) as a practical instrumental resource to increase forensic
drug screening laboratory throughput and productivity.
Recent forensic drug-screening applications have moved
toward utilization of liquid chromatography with tandem
MS detection (LC–MS-MS) to eliminate labor-intensive and
time-consuming sample derivatization before analysis (2).
Conventional MS methods for drug screening in urine use
Figure 1: Total ion chromatogram comprising a two-dimensional contour
plot of the drug mix standard spiked into an 8-mL aliquot of urine at 250
ng/mL analyzed by automated SPME-GC × GC–TOF-MS. The contour plot
is labeled with the 11 nonderivatized drugs identified with NIST library
matches greater than 75%. Note that methamphetamine and ecstasy
chromatograph with significant peak tailing; however, they are both labeled
with one arrow in the elution region for clarity of this illustration. Over 9000
peaks were found with an S/N of 50 or more for this analysis.

Figure 2: This is the extended range (1) calibration
curve for methamphetamine from 10 to 250 pg/
μL with a correlation coefficient value of 0.9991.
sample derivatization techniques before
GC–MS analysis. This exploratory
research utilizes automated SPME-GC
× GC–TOF-MS for various drugs of
abuse analysis in urine without timeconsuming
sample derivatization.
Methamphetamine, cocaine, diacetylmorphine,
codeine, oxycodone,
ecstasy, acetylcodeine, monoacetylmorphine,
hydrocodone, benzodiazepine,
and LSD spiked as a standard
mixture into urine were identified by
this research. The variety of drugs utilized
for this study represent distinct,
as well as similar, chemical classes and
functionalities.
The objectives of this research
demonstrate the detectability of nonderivatized
drugs in complex biological
samples such as urine. This article
describes the calibration linearity capabilities
over the range of 10–1000
ng/mL for drug screening in urine
using SPME of nonderivatized samples
coupled with GC × GC–TOF-MS. The
feasibility of this analysis is described
by illustrating the advantages of GC ×
GC and TOF-MS. Data from this research
will show trace level part-perbillion
(ppb) identification of targeted
drugs in complex sample matrices.
The use of automated SPME applied
to nonderivatized urine samples, coupled
with comprehensive two-dimensional
chromatography and TOF-MS
detection, demonstrates that this is a
favorable technique for qualitative and
quantitative analysis.
Figure 3: The calibration curve for
diacetylmorphine from 10 to 1000 pg/μL with
a correlation coefficient value of 0.9797.
Experimental
A multiple-drug standard mixture prepared
from Sigma Aldrich standards
(St. Louis, Missouri) was spiked at concentrations
from 10 to 1000 ppb into 8mL
aliquots of urine. Automated sample
preparation was conducted on the
nonderivatized spiked urine standards
using an MPS2 autosampler (Gerstel,
Linthicum, Maryland) equipped with
a SPME prepstation. The extracted
sample was then subjected to thermal
desorption in the GC injection port.
Aliquots of urine spiked with the drug
standard mixture were prepared at the
concentrations of 10, 50, 250, 500, and
1000 ng/mL. Hexachlorobenzene was
added to each sample as an internal
standard at 500 ng/mL before extraction.
The samples required no further
sample preparation prior to SPME
extraction and GC × GC–TOF-MS
analysis. Each sample was placed in
a 10-mL glass SPME autosampler vial
and sealed. Extraction was conducted
on 8-mL aliquots spiked with the drug
stock standard mixture at the specified
concentration ranges from 10 to 1000
ng/mL using the MPS2 Prepstation
(Gerstel). The results were generated
with a Pegasus 4D GG × GC–TOF-MS
instrument (LECO, St. Joseph, Missouri)
equipped with a Gerstel MPS2
autosampler and SPME prepstation.
The LECO Pegasus 4D was equipped
with an 7890 gas chromatograph (Agilent,
Wilmington, Delaware) featuring
a LECO quad jet dual-stage thermal
m
modulator and secondary oven. LECO
ChromaTOF software with True Signal
Deconvolution was used for all acquisition
control and data processing. The
autosampler was a single-rail Combi
Pal (CTC, Zwingen, Switzerland)
equipped with an SPME sample agitator/prepstation
and a SPME fiber-conditioning
station. Automated sample
extraction was performed using a
50/30 μm DVB–Carboxen–PDMS Stable
Flex SPME fiber (Supelco-Sigma-
Aldrich, Bellefonte, Pennsylvania) (3).
The SPME method was developed in
the ChromaTOF software autosampler
methods section using the CTC
Combi PAL option. The sample agitator
was set to “ON” at a speed of 200
rpm and extraction temperature of 37
°C. The SPME time was set for 30 min,
and sample desorption time in the GC
injection port was 2 min. The SPME
fiber was then conditioned in the fiber
bakeout station at 270 °C for 40 min
before concurrent sample extraction
for the next analysis.
A 30 m × 0.25 mm × 0.25 μm film
thickness, Rxi -5ms,(Restek Corp.,
Bellefonte, Pennsylvania) GC capillary
column was used as the primary
column for the analysis. In the GC ×
GC configuration, a second column
1.5 m × 0.18 mm id. × 0.18 μm film
thickness Rtx-200 (Restek Corp.) was
placed inside the LECO secondary GC
oven after the thermal modulator. The
helium carrier gas flow rate was set to
1.5 mL/min at a corrected constant
flow via pressure ramps. The primary
column was programmed with an
initial temperature of 40 °C for 2.00
min and ramped at 6 °C/min to 290
°C for 10 min. The secondary column
temperature program was set to an initial
temperature of 50 °C for 2.00 min
and then ramped at 6 °C/min to 300
°C with a 10-min hold time. The column
temperature offset between the
primary and secondary column oven
temperature program was +10 °C. The
liquid nitrogen–cooled thermal modulator
was set to +25 °C relative to the
secondary oven and a modulation time
of 5 s was used. The MS mass range
was 45–550 m/z with an acquisition
rate of 200 spectra/s. The ion source
chamber was set to 230 °C, and the

detector voltage was 1650V with an
electron energy of −70eV.
Results and Discussion
A total of 11 drugs of abuse and metabolites
were detected in this analysis
of nonderivatized drugs spiked in
urine. methamphetamine, cocaine,
diacetylmorphine, codeine, oxycodone,
ecstasy, acetylcodeine, monoacetylmorphine,
hydrocodone, benzodiazepine,
and LSD were identified
by comparison match to the NIST 05
mass spectral library. This analysis
illustrates the capability of comprehensive
two-dimensional chromatography
coupled with TOF-MS to
provide maximum chromatographic
separation and resolution along with
deconvolution algorithms to accurately
identify trace parts-per-billion levels of
drugs of abuse in urine without timeconsuming
sample derivatization. On
average, 9000 peaks were found per
sample with an S/N of 50, confirming
the complexity of the sample matrix
and difficulty of this type of analysis.
The chromatogram in Figure 1 illustrates
the increased peak capacity and
resolution power of the extended chromatographic
plane made available by
comprehensive two-dimensional chromatography.
The TOF mass spectrometer
that was used acquires continuous
full-range nonskewed mass spectra at
acquisition rates of up to 500 spectra/
s, thereby achieving the fast detection
needed for accurate identification of
peaks in a complex biological matrix
such as urine.
Detectability of Nonderivatized
Drugs in Urine by Automated
SPME-GC × GC–TOF-MS
The minimum detectable concentrations
calculated were based upon the
data collected from this research of
urine samples spiked with the drug
standard mixture at the concentration
range of 10–1000 ppb. Relative limits
of detection were calculated for seven
drugs by extrapolation from the ratio
of concentration versus signal to noise.
The relative limits of detection indicate
that drug screening in urine without
sample derivatization is sensitive
to very low parts-per-billion or even
Figure 4: This illustration shows the benefits of TOF-MS to allow fast acquisition rates, which
provide the data density and nonskewed mass spectra required to facilitate the deconvolution
algorithms that successfully identify trace components even in heavy sample matrices.
Hydrocodone (peak 7279) is identified with a similarity of 89.3% library match as well as two
other components in approximately 40 ms.
(a)
(b)
Figure 5: (a) Peak true deconvoluted mass spectrum for peak 7279 and (b) the library match for
the drug hydrocodone with a match similarity of 893.
parts-per-trillion levels by automated
SPME-GC × GC–TOF-MS analysis.
These results highlight the increased
peak detectability realized by thermally
modulated two-dimensional
chromatography, which provides very
narrow peak widths and increased
peak height and thereby maximizes
chromatographic efficiency.
Calibration Linearity for Eight
Nonderivatized Drugs of Abuse
Five-point calibration curves were
developed from 10 to 1000 ng/mL for
eight nonderivatized drugs of abuse in
this research. Good calibration linearity
was observed for all eight components,
even for two components that
exhibited poor peak shape. Methamphetamine
and ecstasy are known to
give poor chromatographic performance
and exhibit significant peak
tailing. The “peak combine” feature in
the software was used to overcome the
chromatographic peak tailing and designate
accurate peak areas for methamphetamine
and ecstasy. The extended
range capability in the calibration

Table I: The relative minimum detection limits of seven nonderivatized drugs spiked in urine at parts-per-billion levels.
*The relative limits of detection were calculated as an extrapolation from a ratio of concentration versus signal to noise.
Name *Concentration R.T. (s) Similarity Unique mass Quant. masses Quant. S/N Area
Methamphetamine 0.212 ng/mL 1160, 1.445 820 72 91 11786 408058871
Ecstasy 0.145 ng/mL 1555, 1.490 917 135 135 17286 13151439
Cocaine 0.038 ng/mL 2285, 1.820 909 82 182 65374 260235677
Codeine 0.065 ng/mL 2445, 1.770 948 162 162 38418 10242005
Oxycodone 0.277 ng/mL 2565, 2.090 926 315 315 9009 2699361
Heroin 0.082 ng/mL 2640, 2.015 920 327 327 30427 7137490
LSD 25 9.830 ng/mL 3095, 3.700 796 221 221 254 71671
Table II: Calibration table showing a partial table used to develop calibration curves in the software. The “Range, Concentration”
column allows the user to define either range 1 or 2 with extended range capabilities. The “Masses” column defines the ion fragment
mass used as the quantitation mass for each analyte and calibration range. Six out of the eight components listed achieved
correlation coefficient values of 0.9500 or greater. All components had correlation coefficient values greater than 0.9000.
Name
Absolute R.T.
(sec , sec)
Range,
conc.
Min valid
conc.
Max valid
vonc.
Masses
Correlation
coefficients
Curve
order
Type
Methamphetamine 1070 , 1.740 2 250 1500 134 0.99483 1 Analyte
Ecstasy 1555 , 1.490 1 5 1500 135 0.91266 1 Analyte
Hexachlorobenzene
(INTERNAL STANDARD)
1765 , 1.520 1 250 750 284 NA 1 ISTD
Cocaine 2285 , 1.805 2 250 1500 303 0.90428 1 Analyte
Codeine 2445 , 1.769 1 5 1500 162 0.95206 1 Analyte
Hydrocodone 2500 , 1.990 1 5 1500 242 0.96399 1 Analyte
6-Monoacetylmorphine
2555 , 1.855 1 5 1500 268 0.94953 1 Analyte
Oxycodone 2560 , 2.140 1 5 1500 315 0.98705 1 Analyte
Diacetylmorphine 2640 , 2.015 1 5 1500 327 0.97971 1 Analyte
feature of the software was utilized to
define different quantitation masses of
two overlapping calibration ranges for
the same analyte. The extended range
calibration capability was applied in
the calibrations for methamphetamine
and cocaine to compensate for ion
fragment mass signal saturation from
particular fragment ions in the mass
spectrum. The software used here allows
the user to define a second quantitation
mass spectral ion fragment
that is not signal saturated, thereby allowing
greater dynamic concentration
range for a specific analyte in a single
calibration. The nonderivatized drugs
in urine analysis achieved linearity of
95% or greater for six out of the eight
drugs selected in this research. All of
the components achieved at least 90%
linearity for the concentration range
from 10–1000 ng/mL.
TOF-MS Mass Spectral
Deconvolution
TOF-MS produces continuous nonskewed
spectra across the full mass
range, which is sampled simultane-
ously from the ion source. Mass spectra
obtained by TOF-MS maintain
uniformity across the entire peak for
a specific analyte. This allows deconvolution
algorithms of coeluted chromatographic
peaks to be accomplished
more effectively and accurately in comparison
with deconvolution of skewed
mass spectra obtained from scanning
mass spectrometers. The illustration
in Figure 4 shows how successful mass
spectral deconvolution is achieved in
a compressed time window of 40 ms
analyzed by TOF-MS. The peak markers
labeled at peak 7277, 7278, and 7279
in the contour plot on the left in Figure
4 are then illustrated by the onedimensional
chromatogram of the
unique ion mass for each component
on the right-hand side of the figure,
designated by arrows for each of the
three selected peaks. The overlapping
peaks observed in the one-dimensional
chromatogram for the unique masses
of the three selected analytes illustrate
the importance of acquiring uniform
nonskewed mass spectra across the
entire chromatographic peak. This
allows deconvolution algorithms to
provide accurate spectral information
for peak identification. The fast
acquisition rates available with TOF-
MS provided the data density needed
to deconvolute the drug hydrocodone
successfully from two other components
in the time window of 40 ms.
Effective deconvolution of nearly
coeluted peaks is observed in Figure
5a by the deconvoluted mass spectrum
for hydrocodone, Peak 7279, previously
illustrated in Figure 4 in the contour
plot and first dimension chromatograms.
The mass spectrum in Figure
5b is the NIST 05 library mass spectra
for hydrocodone. The match of 893 out
of 1000 shows a high degree of mass
spectral similarity. This example illustrates
the value of nonskewed full mass
range high density data available with
TOF-MS and the impact of such data
on performing accurate identifications
even in complex sample matrices.
Conclusions
This investigation employed GC ×
GC–TOF-MS as well as SPME with-

out labor- and time-intensive sample
derivatization to optimize qualitative
and quantitative capabilities for drugscreening
analysis in urine. Automated
SPME-GC × GC–TOF-MS analysis results
show that parts-per-billion levels
of drugs from various chemical classes
can be detected successfully. This experimentation
achieved calibration linearity
of 90% or greater for eight drugs
with a concentration range from 10 to
1000 ng/mL in heavy sample matrices.
The results of this study indicate that
trace-level screening of drugs in urine
can be performed without time-consuming
and complex sample derivatiza-
tion, providing accurate and positive
identifications by automated SPME
coupled with GC × GC–TOF-MS. GC
× GC–TOF-MS combined with automated
SPME achieves the chromatographic
resolution, analyte sensitivity,
and mass spectral integrity that can
identify trace parts-per-billion levels of
drugs accurately in complex sample matrices
without sample derivatization.
References
(1) National Drug Threat Asssessment
2009, National Drug Intelligence Cen-
ter, U.S. Department of Justice, pp.
1–2, www.usdoj.gov/ndic/pubs31/
31379/31379p.pdf
(2) S. Robinson and S. McDonnel, Clin-
cal/Toxicology Application Notebook,
May 2008, p. 6
(3) Supelco Application Bulletin 901A,
Solid Phase Microextraction/Capillary
Analysis of Drugs, Alcohols, Organic
Solvents in Biological Fluids, Supelco-
Sigma-Aldrich Co.
John Heim and Scott Pugh are
with LECO Corporation, Life Science and
Chemical Analysis Center, St. Joseph,
Michigan. ◾
© Reprinted from Current Trends in Mass Spectrometry Supplement, March 2009 Printed in U.S.A.