Embed Size (px)
Citation preview
RT: 22.57 - 31.20
23 24 25 26 27 28 29 30 31Time (min)
20
40
60
80
100
20
40
60
80
100
20
40
60
80
100
Rel
ativ
e A
bund
ance
20
40
60
80
10023.07
30.6928.7128.42
28.1227.52
23.07
30.6928.7128.42
28.1127.52
28.6627.49
28.6627.49
NL: 3.20E4
NL: 4.28E4
NL: 2.51E5
NL: 3.21E5
Heinz Mehlmann, Dirk Krumwiede, Thermo Scientific Bremen, Germany
Maximized productivity for PBDE, Dioxin and PCB analysis using DualData Mode with Magnetic Sector GC-HRMS
ABSTRACTPurpose: Added to the intrinsic high sensitivity and robustness - due to the large ionsource - for Dioxin and POPs analysis of a Magnetic Sector High Resolution MassSpectrometer the attachment of 2 GCs to one single MS strongly increases its flexibility(Figure 1a). Furthermore on the basis of such a Dual GC Magnetic Sector MassSpectrometer the concept of DualData XL acquisition was developed (Figure 1b). Thisconcept and its technical implementation allow to strongly increase the samplethroughput. Effectively the productivity in number of samples injected on one single MSis almost doubled.
Methods: For all gas chromatographic analyses a certain amount of ‘dead’ time is anintrinsic part of the measurement. The dead time is the time before the first relevantpeak is detected and after the last relevant peak elutes. Dioxin analyses are typicallyconducted using 60 m columns that result in run times of 50-60 minutes. The dead timefor such analyses can be up to 20-30 minutes per sample.
The dead time can be almost eliminated by performing alternate staggered injectionsusing two GCs coupled to one MS. First of all both GCs are running in parallel.Secondly the timing is such that the chromatogram section with peaks from one GCfalls into the chromatogram section with no peaks from the other GC and vice versa.To realize a time-controlled staggered injection sample sequence specific software isneeded and some hardware modifications inside of each GC need to be implemented.These modifications ensure that at any point of time the flow of one GC only – timesection with peaks - is guided into the mass spectrometer for measurement while theflow of the other GC – time section without peaks - is directed into waste. A timecontrolled dynamic flow switching system was developed using a proprietary microfluidic channel device (MCD) to switch flow between vacuum purge and MS (seeFigure 4).
Results: Chromatograms with and without the MCD device are practically undistinguishable from one another in terms of peak shape or sensitivity. The concept was proven by numerous experiments and is by now validated in full productiondioxin analysis laboratories. The Thermo Scientific DualData XL Module for the Thermo Scientific™ DFS™ Magnetic Sector GC-HRMS can be used for a range of POPs analyses applications such as Dioxins, PCBs or PBDEs. Also a combination of different applications is possible. With an increase of more than 90% the sample throughput for e.g. Dioxins is almost doubled.
FIGURE 1a. Thermo Scientific DFS Magnetic Sector GC High Resolution Mass Spectrometer coupled to two Thermo Scientific TRACE 1310 GC.
FIGURE 2. Timescale of a typical dioxin analysis.
INTRODUCTIONThe sequence timeline on a standard dual GC configuration is interrupted by segments of dead time (see Figure 3a). In the sequence timeline of a DualData XL Module equipped dual GC configuration a staggered injection is possible with no dead time between each data acquisition (see Figure 3b).
FIGURE 3a. Normal dual GC configuration: The gray area represents dead time, where no data is acquired.
FIGURE 3b. DualData XL acquisition sequence. Only the relevant retention time windows of both GCs are monitored.
Each GC was equipped with a dynamic flow switching system which allows to divide the GC eluate to waste or to the MS source (see Figure 4). The ability to cut out parts of the chromatogram in order to prevent any cross contamination caused by the second GC run was checked by a measurement of a EPA 1613 CS5 standard. Furthermore the ability to cut out the solvent was proved by monitoring the MS Ion source pressure while injection solvent up to a volume of 10ul into the system. The solvent don’t reach the source, when the valves are in ON position. In the meantime the flow into the ion source will be compensated by He delivered by a computer controllable carrier gas module in order to keep the pressure in the ion source constant .
FIGURE 4. Principle of the dynamic flow switching system of the DFS GC-HRMS DualData XL Module (left) and inner view of a micro fluidic channel device (MCD) wafer (right).
In a typical experiment, the first GC run was started, and during the wait time of 20minutes, while the solvent peak as well as other compounds of no interest eluted all GCeluate was diverted to waste. After 20 minutes, the GC eluate was directed to the ionsource of the MS and MS data acquisition started. At approximately the same time, asecond sample was injected to the second GC, running the same process as the firstone. (i.e. during the first 20 minutes no GC eluate was directed towards the MS. Oncethe first GC finished cooling down the oven to start condition, another injection occurred,and the same scheme as denoted above repeats). This resulted in two GCs runningsimultaneously with staggered sample injections. Only the retention time windows ofinterest from each GC were directed to the MS for data acquisition. Figure 5 illustratesthe measurement sequence from both of the GCs providing consecutive data files fortarget compound quantification.
FIGURE 5. Flow of the analytical column of GC A is guided into the MS and isacquired while the flow of GC B is directed into waste (upper schematic) and viceversa (lower schematic).
MATERIALS AND METHODSThe configuration used in this study consists of two Thermo Scientific™ TRACE 1310™GC equipped with the DualData XL Module using two columns coupled to the ThermoScientific™ DFS™ Magnetic Sector GC-HRMS. The mass spectrometer was set up in amultiple ion detection mode (MID) at a resolution of 10,000 (10% valley definition). FC43and PFK was used as a reference compound to provide the inherent lock and calimasses. The Thermo Scientific™ TriPlus RSH Autosampler with extended x-rail servedboth GCs from one common sample tray. Typically one μL of sample was injected. Amethod 1613 CS1 – CS5 calibration standard (1:10 diluted from Cambridge IsotopeLaboratories) and CS3 / CPM 8290/1613 was used as well as method 1668 and 1614standards to demonstrate the chromatographic performance of the system. A low levelpooled blood sample in the range of 20fg/ml of 2,3,7,8 TCDD in dirty matrix was used todemonstrate the performance in terms of sensitivity.
RESULTSPerformance: The analytical performance with DualData XL Module and conventionalGC/MS configuration was compared using the same set of polychlorinated dioxins andfurans, PCBs and PBDE samples as model compounds. Sensitivity was compared byusing low concentrated PCDD/PCDF standards as well as a low level pooled bloodsample. The GC separation integrity, ruggedness and long-term stability of the columnswitching system have been proven in unattended sample.
RT: 38.44 - 43.53
38.5 39.0 39.5 40.0 40.5 41.0 41.5 42.0 42.5 43.0 43.5Time (min)
0
50
1000
50
1000
50
1000
50
1000
50
100
Rel
ativ
e A
bund
ance
0
50
1000
50
1000
50
10039.39 41.8841.1641.01
42.91
41.8841.1639.39 41.01 42.89
41.8641.1440.99 42.89
41.8641.1440.99 42.88
40.16 42.5142.19
40.16 42.19 42.5142.08
42.18 42.4942.06
42.18 42.4942.06
NL: 5.27E5
NL: 4.28E5
NL: 5.76E5
NL: 1.07E6
NL: 3.87E5
NL: 3.06E5
NL: 7.43E5
NL: 5.95E5
native HxCDF (373.8202 m/z)
native HxCDF (375.8172 m/z)
13C12 Standard HxCDF (383.8633 m/z)
13C12 Standard HxCDF (385.8604 m/z)
native HxCDD (389.8151 m/z)
native HxCDD (391.8121 m/z)
13C12 Standard HxCDD (401.8554 m/z)
13C12 Standard HxCDD (403.8524 m/z)
FIGURE 6a. Example of peak integrity of Dioxin trace analysis (Hexa CDD/F) usingthe Dual Data Acquisition.
native TCDD (319.8960 m/z)
native TCDD (321.8930 m/z)
13C12 Standard TCDD (331.9363 m/z)
13C12 Standard TCDD (333.9333 m/z)
It was found that the effects due to some increased dead volumes and a disturbed flowpath caused by the MCD were negligible. No additional peak tailing was observed andthe requirements of EPA 1613 to separate the 2,3,7,8 TCDD to the next eluting TCDDwith a valley of better than 25% was easy achievable .
2,3,7,8-TCDD
native TCDD (319.8960 m/z)
native TCDD (321.8930 m/z)
13C12 Standard TCDD (331.9363 m/z)
13C12 Standard TCDD (333.9333 m/z)
< 10%
FIGURE 7a. Separation of the 2,3,7,8-TCDD to the next eluting TCDD. The valley isbelow 25% as requiered by EPA1613.
Column: Thermo TR-Dioxin 60m x 0.25mm ID x 0.1µm film; Oven: 150 (2)-25-230(25)-6-320(6)
FIGURE 6b. 1μl Blood sample in dirty matrix in the range of 20 fg/μl TCDD.
FIGURE 1b. DualData XL Module on a Thermo Scientific TRACE 1310 GC.
Column: Thermo TR-Dioxin 60m x 0.25mm ID x 0.1µm film; Oven: 150 (2)-25-230(25)-6-320(6)
Column: Thermo TR-Dioxin 60m x 0.25µm x 0.1µm film; Oven: 120 (2)-10-220(10)-3-235(7)-4.6-310(1)
ConclusionIt has been demonstrated that the Dual Data option allows a higher sample throughput by no loss in performance such as peak shape or sensitivity.
Increase of productivity up to double sample throughput.
Excellent peak shape using MCD wafer technology.
No loss in sensitivity compared to a standard dual GC System.
Applicable to different POPs such as Dioxins, PCBs and PBDEs.
References1. U.S. Environmental Protection Agency, Method EPA 1613 Rev. B, Washington,
October 1994.
2. European Committee for Standardization, EN 1948, Brussels, December 1996.
3. Application Note AN30098, Thermo Fisher Scientific, Bremen 2006.
Cambridge Isotope Laboratories is a trademark of Cambridge Isotope Laboratories, Inc. All other trademarks are the property of Thermo Fisher Scientific and its subsidiaries.
This information is not intended to encourage use of these products in any manners that might infringe the intellectual property rights of others.
Presented at the 8th International Symposium on Flame Retardants (BFR 2017), York, 7-10 May 2017.
RT:5.67 - 26.50
6 8 10 12 14 16 18 20 22 24 26Time (min)
0
50
1000
50
1000
50
1000
50
100
Rel
ativ
e A
bund
ance 0
50
1000
50
1000
50
100 8.16 8.36
7.887.25
10.8311.2510.31
9.66
13.6112.40
14.37 15.64
15.7717.24
15.6315.9316.4917.22
20.1117.9619.73 20.23 21.79
18.08
21.7818.89 21.9919.00 22.4417.81 20.11
25.50
NL:7.60E5m/z= 327.8815-327.9015 MS D90311A20NL:4.34E5m/z= 405.7981-405.8181 MS D90311A20NL:3.09E5m/z= 485.7010-485.7210 MS D90311A20NL:3.47E5m/z= 563.6120-563.6320 MS D90311A20NL:3.93E5m/z= 643.5200-643.5400 MS D90311A20NL:1.97E5m/z= 721.4310-721.4510 MS D90311A20NL:1.97E4m/z= 799.3230-799.3430 MS D90311A20
RT: 16.39 - 23.75
17 18 19 20 21 22 23Time (min)
0
20
40
60
80
100
Rel
ativ
e A
bund
ance
0
20
40
60
80
100
Rel
ativ
e A
bund
ance
0
20
40
60
80
100
Rel
ativ
e A
bund
ance
0
20
40
60
80
100
Rel
ativ
e A
bund
ance
RT: 20.27AA: 439636
RT: 17.14AA: 181218
RT: 21.67AA: 202188
RT: 20.08AA: 206028
RT: 20.85AA: 199819 RT: 23.54
AA: 152336
RT: 20.27AA: 276265
RT: 20.08AA: 131558
RT: 17.14AA: 115798
RT: 21.67AA: 127874
RT: 20.85AA: 121786 RT: 23.54
AA: 102118
RT: 18.28AA: 197852
RT: 20.25AA: 202457RT: 17.12
AA: 164137 RT: 21.64AA: 182319 RT: 23.52
AA: 161317
RT: 18.28AA: 124122
RT: 20.25AA: 125141RT: 17.12
AA: 103963 RT: 21.66AA: 113758 RT: 23.52
AA: 100446
NL:1.37E5m/z= 325.6574-326.0329 MS ICIS X4-210313-1S3
NL:8.59E4m/z= 327.7226-328.0981 MS ICIS X4-210313-1S3
NL:6.17E4m/z= 337.7672-338.1427 MS ICIS X4-210313-1S3
NL:3.98E4m/z= 339.6447-340.2079 MS ICIS X4-210313-1S3
Flexibility: Also other POPs such as PCBs and PBDEs can be run with the Dual Dataoption as well as combinations of different applications per GC on one Dual DataSystem. (e.g. Dioxins on GC1 60m column and PBDEs on GC2 using a 15m column)
FIGURE 8. Example of PCB trace analysis (5Cl PCB) (left). Example of PBDEs trace analysis (right).
Sample: 1µl CS4-7474-62 CVS; Column: Restek Rtx@-1614 15m x 0.25mm ID x 0.1µm film; Oven: 120 (3)-10-300(6)
Productivity: The amount of samples running a staggered Sequence of dioxin andfuran analysis using DualData XL Module was compared to a standard dual GCconfiguration during a time frame of 12 hours. The analysis was done on Thermo TR-Dioxin 60m x 0.25mm ID x 0.25µm film in each GC with a total runtime of 43 min. Theacquisition was started after 20min. In standard dual GC mode 16 samples wereanalyzed compared to 32 samples with the DualData XL Module.
FIGURE 9. Sample throughput in a timeframe of 12 hours. In standard dual GC configuration 16 samples (left) were measured compared to 32 samples with the DualData XL Module (right).
