Sample Preparation and Sample Presentation for
Direct Analysis in Real Time (DART)
Robert B.
1JEOL
*1,
Cody
A. John
1
Dane ,
USA Inc., Peabody MA;
Drew
2Nanoliter
2
Sauter
2
III
and A. D. Sauter
LLC, Henderson, NV
A DART Glow Discharge
Stimulant Drug
Mixture
Averaged
spectra
“Less is More”
100 ppb oxazepam in 200 ppm creatinine
100
Introduction: The DART [1,2] ionization mechanism is dominated by gas-phase ionmolecule reactions [3]. Like atmospheric-pressure chemical ionization (APCI), matrix
effects in positive-ion mode are related to the relative proton affinities of compounds
present in a mixture. Condensed-phase phenomena that cause suppression in
electrospray ionization (e.g. competition for the surface of a drying droplet) do not play a
role in DART ionization. Nevertheless, suppression can occur with the DART ion source if a
trace analyte with a relatively low proton affinity is associated with a large excess of a
compound with a relatively high proton affinity. Stout and coworkers have shown the
suppression of oxazepam in the presence of a large excess of creatinine [4].
284.1769
160
Rel. Abund.
234.1497
120
150.1270
50
287.2223
1
3
6
.
1
1
2
279.1243
271.1710
6
273.2043
274.2141
100
195.0891
2 ml droplet
285.1969
140
297.1940
289.2150
281.1675
80
299.1296
292.2170
304.2365
20
100
150
200
2
300
m/z
3
150.1269
100
250
195.0860
m/ z 270
275
280
285
290
295
300
Results: We observed that signal suppression for low-m/z components in the stimulant
mixture was clearly reduced for 50 nl sample introduction compared to 2 ml volumes (Figure
1). This led us to wonder if this was a general phenomenon that could be reproduced for
the documented example of oxazepam in the presence of excess creatinine. The results
from 50 nl droplets were compared to the results for 2 ml deposited on melting point tubes.
Suppression of oxazepam by creatinine was significantly reduced by using nanoliter
droplets, resulting in better detection limits. Oxazepam could barely be detected in an
actual urine sample at the 1 ng/ml level (not shown) but the ability to detect oxazepam in
the presence of other chemical interferences from urine was limited by the mass
spectrometer resolving power.
Improving signal by reducing sample quantity is a well-known phenomenon in mass
spectrometry and had been shown for FAB, ESI and other ionization methods. A significant
improvement in MALDI sensitivity was observed for nanoliter sample deposition [5]
305
Oxazepam_100pg_nl_series_repeat
Scan: 147,168 TIC=589078 Base=.11%FS #points=3566 RT=1.32
50 nL pipette
5
MH+
450
304.1562
4
1
350
400
350
234.1494
136.1126
50
50 nl droplet
300
More even response
for 1 mg/ml mixture
of 5 drugs
250
200
150
100
50
m/ z 270
0
100
150
200
250
300
350
275
280
285
290
295
300
305
MH+
Oxazepam_100pg_nl_series_repeat
Scan: 88 TIC=293770 Base=.11%FS #points=3566 RT=.73
m/z
282.2652
50 nl droplet
450
400
Figure 1. Stronger signals are observed for low-m/z compounds in stimulants mixture when 50 nanoliter droplets are used
(bottom) compared to 2 ml droplets (top). The compounds in the mixture are:
1 = amphetamine, 2 = methamphetamine, 3= caffeine, 4= methylphenidate, 5= cocaine.
350
285.2411
300
287.0481
299.2672
280.2671
250
200
150
271.2423
297.2400
273.2469
274.2471
295.2297
279.2363
277.1805
293.1961
289.0503
300.2732
303.2810
291.2297
Aqueous solutions containing 200 mg/ml (= 200 ng/ml) of creatinine were spiked with
oxazepam at levels of 100 pg/ml, 1 ng/ml, 10 ng/ml, 100 ng/ml, and 100 ng/ml. DEA-exempt
solutions (Alltech) of a 1 mg/ml methanol solution of 5 stimulant drugs were analyzed
without further dilution. Approximately 50 nanoliter droplets of each solution were manually
deposited within 1-2 mm in front of the exit grid of the DART ion source by using Nanoliter
LLC's patent-pending nanoLiter Cool Wave pipette. For comparison, 2 ml aliquots of the
solutions were pipetted onto the outside of the sealed end of 2 mm Pyrex melting point
tubes which was then positioned within 2 mm of the DART exit grid for analysis by using the
DART-SVP linear rail. All measurements were made in triplicate to average out variations in
signal intensity due to variations in sample placement.
303.2239
40
100
Experimental: An atmospheric pressure ionization time-of-flight mass spectrometer (JEOL
AccuTOF™) equipped with a DART ion source (DART-SVP ™ from IonSense™) was used
for all measurements. The DART ion source was operated with helium gas at a flow rate of
approximately 1.2 liters per minute. The DART gas heater was set to 350°C and the DART
exit grid was set to +250 V. The mass spectrometer atmospheric pressure interface was
operated with the following potentials: orifice 1 = 20V, orifice 2 and ring lens = 5 V. The RF
ion guide voltage was set to 1000 V to transmit and detect ions of m/z 100 and greater.
301.1835
294.2672
60
0
Rel. Abund.
Liquid samples have commonly been analyzed by DART ionization by dipping the sealed
end of a glass rod, such as a 2 mm-diameter melting point tube, into a liquid sample.
Alternatively, several microliters of sample are deposited onto the outside of the glass rod
which is then positioned within a few millimeters of the DART ion source exit grid. Here we
show that by presenting nanoliter droplets directly in front of the DART ion source,
suppression effects are drastically reduced, resulting in better detection limits and a more
uniform response for mixture components. The ultimate goal is to achieve efficient
ionization and transmission into the API interface of the entire sample.
Not detected
Oxazepam_100pg_2ul_series
Scan: 481 TIC=788098 Base=.04%FS #points=3566 RT=4.02
m.p. tube
304.1571
50
DI H2O
Oxazepam_0pt001_nl_series |
m/z 270
50 nl droplets
Sample quantity = 25 pg
MH+water
_______________
0.0
Oxazepam_0pt001_2ul_series |
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
DI H2O
2 ml droplets
Sample quantity = 2 ng
MH+water
_______________
= 87
MH+200 ppm creatinine
200 ppm creatinine in H2O
287.0532
Min.
1
2
3
4
5
6
Figure 2. RICs for 1 ng/ml solution of oxazepam in DI water (left) and in 200 ng/ml aqueous creatinine (right). Suppression of the
oxazepam signal is significantly reduced when 50 nl droplets are analyzed (top) compared to 2 ml droplets (bottom). The
analyte signal in 200 ppm creatinine is approximately 4x greater for the 50 nl droplets, despite the fact that the sample quantity
is 80x less than for the 2 ml droplets. The analyte signal in water is only reduced by a factor of 2 in going from 2 ml to 50 nl
samples. Note that the time scales are not identical. Nanoliter droplets can be introduced with the nanoliter pipette and
analyzed more rapidly than microliter droplets could be deposited on a melting point tube and introduced with the linear rail.
MH+
Oxazepam_1ng_nl_urine
Scan: 21 TIC=7789028 Base=.4%FS #points=538 RT=.18
m/z 287.057
1400
1200
280
285
290
295
300
305
Figure 4. 100 pg/ml oxazepam. Top: 2 ml droplet (200 ng/ml creatinine) on melting point tube.
Middle: 50 nl droplet (200 ng/ml creatinine).
Bottom: 50 nl droplet (DI H2O). Little or no suppression was observed for this concentration.
MH+200 ppm creatinine
200 ppm creatinine in H2O
287.0532
Min.
= 8
275
1000
800
Conclusions:
• Sample introduction as nanoliter droplets results in a significant reduction in
sample suppression.
• Nanoliter droplets permit better detection limits than the commonly-used
melting-point tube method, despite the smaller sample volumes used.
Detection limits reported here are 10 to 100 times lower than previously
reported values for the same solution concentrations.
• The working hypothesis is that an excess of reagent ions in the DART gas
stream results in complete ionization of the all of the components in the
nanoliter droplets. In contrast, if larger samples volumes are used, the
number of analyte molecules can exceed the number of reagent ions and
charge competition can occur.
• Shorter desorption times for the nanoliter droplets result in temporal focusing
of the signal, perhaps reducing discrimination of volatile components and
improving the signal-to-background.
600
400
200
m/ z 284.5
285
285.5
286
286.5
287
287.5
288
288.5
289
289.5
Figure 3. 1 ng/ml oxazepam in urine measured as a 50 nl droplet. The detection limits are determined by the ability to
separate analyte from background, not by suppression.
References
1. DART is a trademark of JEOL USA, Inc..
2. US Patent Number 6,949,741. Other patents pending.
3. Cody, R. B.; Laramée, J. A.; Durst, H.D. Versatile New Ion Source for the Analysis of Materials in Open Air
under Ambient Conditions. Anal. Chem., 2005, 77(8), 2297 - 2302 DOI: 10.1021/ac050162j
4. Stout, P. R.; Bynum, N.; Minden, E.; Miller, J. “Evaluation of AccuTOF DART for Postmortem Toxicology
Screening” NIJ Technology Transition Workshop.
<http://projects.nfstc.org/tech_transition/dart08/presentations/Stout.ppt>
5. Tu, T., Sauter, A.D., Sauter, A.D., Gross, M.L., Improving the Signal Intensity and Sensitivity of MALDI Mass
Spectrometry by Using Nanoliter Spots Deposited by Induction-Based Fluidics, Journal of the American Society
for Mass Spectrometry (2008), doi: 10.1016/j.jasms.2008.03.017.