novel miniature ftms for analysis of corrosives and...

SIEMENS 4th Workshop on Harsh-Environment MS October 8, 2003

Transportable Miniature FTMS for Analysis of Corrosives and Chemical

Warfare Agents

Wayne V. Rimkus, Dean V. Davis, Kenneth Gallaher

Siemens Applied Automation, 500 West Highway 60, Bartlesville, OK 74003


Introducing the Quantra• Rugged• Transportable• Accurate• Fast• Cheap• Small (by FTMS standards!)• 1 torr to 2 Atmosphere sampling• Self contained


The relative size, absence of external pumping requirements and full battery backup of critical systems lead to the portability of the system. This translates to “field-ability” .

24”

42”

Miniature FTMS Quantra


Transportable

The author transporting the FTMS system – a small truck with a ramp was used. The Quantra is on wheels and is in the crate.

System was setup in a hotel room. Elapsed time - truck to 20,000 resolution – 15 minutes.


Unique Design Features

• Permanent Magnet (1 tesla)• Ion Getter Pump (10-10 torr)• Pulsed Valve – picoliter sampling• Filament current in Nanoamps• Battery Backup – All Critical Systems• LOD 10 ppm, with trapping < 10 ppb (ppt?)• Resolution at m/z 131 is 20,000• Ion Manipulation

– Ion Ejection– MS/MS– Chemical Ionization– Negative Ion Detection


Measurement ChamberSensor

Valve manifold

Ion pump

power supplyVacuum chamber

crimp tube


Quantra Valves


Theory of Fourier TransformIon Cyclotron Resonance

Mass Spectrometry

FTICR or FTMS


The FT-ICR Measurement Cell

Trapping Plates

Transmitter Plates

Detector Plates


FT

Time DomainThe FT produces a Frequency Domain

Subsequent non linear inversion yields Mass/Charge


Mixed Hydrogen, Helium, Deuterium sample

Resolution > 540,000

m/z 4 Spectrum


Resolving PowerFTMS vs Quadrupoles

NitrogenCarbon Monoxide

What the FTMS “sees”What the Quadrupole “sees”

m/z 28


Resolution Zoom

Resolution > 8,000

Quantra exhibits very high resolving power


Quantra Instrument Sensor and Electronics

Example Applications

• Trace organics in bulk gases

• Identification of materials during catalysis

• Headspace and Solids analysis

• Residual Solvents

• Ambient air monitoring

• Characterization of semi-conductor reaction mixtures

• Incinerator monitoring

• Chemical Threat


MS/MS in Direct effluent analysis

FTICR Spectral Trace of SF6 W ithout N 2 M atrix Ejection

14 14.5

28

29

3032

40 44 1270

5000

10000

15000

Ion

Abu

ndan

ce (a

rb. u

nits

)

50 100 150

M ass (m /z)

Sulfur Hexafluoride not readily seen


Study of process effluent with Nitrogen ejection

F T IC R S p e c tra l T ra c e o f S F 6 W ith N 2 M a tr ix E je c t io n

4 1

7 0

8 9

1 0 8

1 2 7

0

5 0

1 0 0

Ion

Abu

ndan

ce (a

rb. u

nits

)

5 0 1 0 0 1 5 0

M a s s (m /z )

N2 completely ejected


Isolated SF5+ Ion for MS/MS

0

2000

4000

6000

8000

10000

12000

20 40 60 80 100 120 140

SF5+ 126.964 amu


MS/MS Spectrum of SF5+

0

500

1000

1500

2000

2500

20 40 60 80 100 120 140

SF4+ 107.964

amu

SF3+ 88.966

amu

SF2+ 69.968

amuSF+ 50.970amu

SF5+ 126.962

amu


Fermenters


Fermenter 1Headspace Spectrum

0

2000

4000

6000

8000

10000

10 20 30 40 50 60

10 sets of “Mass Doublets” – Isobars

1 Triplet (Nominal mass 16 – O, CH4, NH2)


Mass 16

0

500

1000

1500

2000

2500

3000

15.99 16 16.01 16.02 16.03 16.04 16.05

0

500

1000

1500

2000

2500

3000

15.99 16 16.01 16.02 16.03 16.04 16.05

O

NH2 CH4


Mass 17

0

50

100

150

200

250

16.99 17 17.01 17.02 17.03 17.04 17.05

0

50

100

150

200

250

16.99 17 17.01 17.02 17.03 17.04 17.05

OHNH3


Mass 43

0

500

1000

1500

2000

42.98 43 43.02 43.04 43.06 43.08 43.1 43.12 43.14 43.16

0

500

1000

1500

2000

42.98 43 43.02 43.04 43.06 43.08 43.1 43.12 43.14 43.16

C2H3O

C3H7


Summary Of Components DetectedPeak Number Observed mass Fragment Assignment Theory Corrected obs Corrected Delta

1 12.0029 C C 12.0000 11.9989 -0.00112 14.0069 N N 14.0037 14.0023 -0.00143 14.0226 CH2 CH2 14.0156 14.0180 0.00244 14.7103 noise? noise 14.7055 5 15.0281 CH3 e/butane/propane 15.0234 15.0232 -0.00026 15.99996 O H2O 15.9949 15.9948 -0.00017 16.0239 NH2 NH3 16.0187 16.0187 0.00008 16.0362 CH4 CH4 trace 16.0312 16.0310 -0.00029 17.0084 OH H2O 17.0027 17.0029 0.000210 17.0322 NH3 NH3 17.0265 17.0267 0.000211 18.0167 H2O H2O 18.0106 18.0109 0.000312 19.9884 Ar+2 Ar+2 19.9812 19.9820 0.000814 25.0149 C2H butane/propane 25.0078 25.0070 -0.000815 26.0235 C2H2 butane/propane 26.0157 26.0153 -0.000416 26.0121 CN HCN 26.0031 26.0039 0.000817 27.0195 HCN HCN 27.0109 27.0110 0.000118 27.0313 C2H3 butane 27.0235 27.0228 -0.000719 28.0058 CO CO 27.9949 27.9970 0.002120 29.0117 CHO acid? 29.0027 29.0026 -0.000121 29.0224 HN2 HN2 29.0140 29.0133 -0.000722 30.0067 NO NO 29.9980 29.9973 -0.000723 32.0002 O2 O2 31.9898 31.9902 0.000427 39.0352 C3H3 propane 39.0235 39.0231 -0.000429 39.9747 Ar Ar 39.9624 39.9623 -0.000132 41.051 C3H5 propane 41.0391 41.0383 -0.000833 42.0224 C2H2O acetone 42.0106 42.0094 -0.001234 42.0598 C3H6 butane 42.0469 42.0468 -0.000135 43.0327 C2H3O acetone 43.0184 43.0194 0.001036 43.0677 C3H7 butane &? 43.0548 43.0544 -0.000437 44.0006 CO2 CO2 43.9898 43.9870 -0.002838 44.0145 N 2O N2O 44.0010 44.0009 -0.000139 45.0117 COOH acid? 44.9977 44.9978 0.000140 50.0291 C4H2 butane 50.0157 50.0137 -0.002042 58.0619 C3H6O acetone 58.0419 58.0441 0.002243 58.0949 C4H10 butane 58.0782 58.0771 -0.0011

Average delta

0.00015 AMU


Fermenter 2 N2, Ar, and CO2 Ejection

0

500

1000

1500

2000

2500

10 20 30 40 50 60 70

0

500

1000

1500

2000

2500

10 20 30 40 50 60 70

Arbitrary Y / Mass (m/z) Paged X-Zoom CURSOR

File # 1 = REG-FERM1B-T7_P#3 @258 Seconds 2/28/03 11:17 AM Res=None

Profile data

SO2

NO2

CH3S

Ar

ArH

36ArH2S

NO or CH2O

N2

HCN& C2H3

CN &C2H2

Ar++

H20

O

N

C

OH & NH3

Sulfur Eating Bacteria

Hot Spring


m/z 17

-100

-50

0

50

100

150

200

250

16.85 16.9 16.95 17 17.05 17.1 17.15 17.2 17.25

-100

-50

0

50

100

150

200

250

16.85 16.9 16.95 17 17.05 17.1 17.15 17.2 17.25



Profile data

OHObserved 17.0046Theory 17.0027

NH3Observed 17.0283Theory 17.0265

DifferenceObserved 0.0237Theory 0.0238

Fermenter 2


m/z 27

0

100

200

300

400

26.99 27 27.01 27.02 27.03 27.04 27.05 27.06

0

100

200

300

400

26.99 27 27.01 27.02 27.03 27.04 27.05 27.06



Profile data

HCNObserved 27.0126Theory 27.0109

C2H3Observed 27.0249Theory 27.0235

DifferenceObserved 0.0123Theory 0.0126

Fermenter 2


Fermenter 3 m/z 28

0

500

1000

1500

27.94 27.96 27.98 28 28.02 28.04 28.06 28.08

0

500

1000

1500

27.94 27.96 27.98 28 28.02 28.04 28.06 28.08



Profile data

COObserved 27.9978Theoretical 27.9949

N2Observed 28.0086Theoretical 28.0061

DifferenceObserved 0.0107Theoretical 0.0115

Sulfur Eating Bacteria

Deep Sea Ocean Vent


Fermenter 3

0

200

400

600

28.94 28.96 28.98 29 29.02 29.04 29.06 29.08

0

200

400

600

28.94 28.96 28.98 29 29.02 29.04 29.06 29.08



Profile data

N2HObserved 29.0160Theoretical 29.0140

Protonated 28 ions

COHObserved 29.0051Theoretical 29.0027

Difference Observed 0.0109Theoretical 0.0113

29 AMU


Fermenter 332 AMU

0

200

400

600

31.92 31.94 31.96 31.98 32 32.02 32.04 32.06

0

200

400

600

31.92 31.94 31.96 31.98 32 32.02 32.04 32.06



Profile data

O2Observed 31.9908Theoretical 31.9898

SObserved 31.9744Theoretical 31.9721

DifferenceObserved 0.0164Theoretical 0.0177


Fermenter 3 m/z 64

-20

0

20

40

60

80

100

63.7 63.8 63.9 64 64.1 64.2 64.3 64.4

-20

0

20

40

60

80

100

63.7 63.8 63.9 64 64.1 64.2 64.3 64.4



Profile data

Observed 63.9466S2 Theoretical 63.9441error of -0.002not SO2 Theoretical 63.9619error of 0.0152(error for CH3S at 47 is -.0014)


observed theory Adjusted Diff. fragment assignment

14.0171 14.0031 0.0014 N N215.0414 15.0235 -0.0024 CH316.0137 15.9949 -0.0018 O O216.0489 16.0313 -0.0006 CH4 CH417.0228 17.0027 -0.0015 OH H2O18.0319 18.0106 -0.0010 H20 H2O19.0429 ?20.0061 19.9812 -0.0011 Ar++ Ar26.0536 26.0157 -0.0021 C2H2 Ethane27.0517 27.0189 0.0052 HCN HCN27.0634 27.0234 -0.0020 C2H3 Ethane28.0436 27.9949 0.0027 CO CO28.0504 28.0061 -0.0153 N2 N229.0492 29.0027 -0.0040 COH COH29.0594 29.014 -0.0028 N2H N2H30.0456 29.998 -0.0027 NO NO32.0428 31.9898 -0.0032 O2 O239.095 39.0234 -0.0027 C3H3

40.0369 39.9624 -0.0028 Ar41.1163 41.0391 -0.0022 C3H542.1271 42.047 -0.0020 C3H643.1023 43.01839 -0.0028 CH3O44.0795 43.9898 -0.0055 CO245.0866 44.9977 -0.0014 CHO253.1565 53.0391 -0.0017 C4H555.1796 55.0548 -0.0014 C4H756.1913 56.0626 -0.0014 C4H858.179 58.0418 -0.0020 C3H6O acetone

65.2052 65.0391 -0.0013 C5H5 benzene67.2285 67.0547 0.0001 C5H768.2394 68.0626 0.0017 C5H870.2652 70.0783 0.0011 C5H10 cyclopentane?78.2719 78.047 0.0027 C6H6 Benzene91.3506 91.0548 0.0045 C7H7 Toluene92.3709 92.0626 -0.0019 C7H8 Toluene

0.00037 4/10 of a millimass unit

Average difference 0.4 Millimass Unit

Data from a Cigarette “smoked”

via an aspirator35 ml volume


Corrosives


Spectrum #1 - NF3

NF2

NF3

5,550 countsNormal Electron Ionization

systems will fail within 1hour NF3 consumes the filament


Spectrum #2000 - NF3

17 hours later NF2

NF3

5,450 counts The Quantra has a double advantage the filament is

impervious to these corrosives and the detection system does not rely on any impact of ions to count the

signal – No Multiplier Fadeout


Calibration curve

5

4.54

3

2

10.5

y = 1290.5x - 230.38R2 = 0.9996

0

1000

2000

3000

4000

5000

6000

7000

0 1 2 3 4 5 6

% NF3

Inte

nsity

calibration curve NF3 Linear (calibration curve NF3)

m/z 52 peak

NF3 in Nitrogen calibration


m/z 52 Peak from NF3

quant plot

0

1000

2000

3000

4000

5000

6000

7000

8000

150 250 350 450 550 650 750 850 950 1050

scan number

inte

nsity

mass 52

7 point Calibration Stability

Serial Dilution

Random Introduction


Random validation points

actual calc 95% confidence % NF3 %NF3 Conc RSD

1.25 1.21 1.28 1.14 3.513.33 3.19 3.26 3.11 1.332.56 2.45 2.49 2.40 0.914.38 4.24 4.35 4.13 1.353.15 3.02 3.07 2.97 0.851.01 0.99 1.05 0.93 3.763.70 3.61 3.73 3.49 1.735.00 5.01 5.13 4.89 1.28

1 – 4% RSD


Semiconductor Gas Sampling

Semiconductor applications (Etch process and abatement) involving highly corrosive fluorocarbon species were readily analyzed with the Quantra Mini FTMS. Response times are similar to FTIR (data presented in part at 50th ASMS - 2002).

Once again the design features of this miniature FTMS allow aggressive and corrosive species to be directly sampled with no ill effects to the mass spectrometer. Placement of the filament, very low filament current, very high vacuum, extremely small sample volume and fast pumping are equally important in this design scheme and yield an extremely rugged system.


CF4/CHF 3 Etch Process Emissions (FTICR)

0.00

0.50

1.00

1.50

2.00

2.50

0 5 10 15 20

Scan Number

Ion

Abu

ndan

ce (a

rb. u

nits

)

CF3+CF2+

0

200

400

600

800

1000

50 100 150 200

CF3+

C2F4+

CHF2+

C2F5+

C3F5+C2F3

+ C3F7+ C4F7

+CF2

+

CN+, C2H2+

COF2+

O+

C+

N+

0

200

400

600

800

1000

50 100 150 200 0

200

400

600

800

1000

50 100 150 200 0

200

400

600

800

1000

50 100 150 200

CF3+

C2F4+

CHF2+

C2F5+

C3F5+C2F3

+ C3F7+ C4F7

+CF2

+

CN+, C2H2+

COF2+

O+

C+

N+

CF3+

C2F4+

CHF2+

C2F5+

C3F5+C2F3

+ C3F7+ C4F7

+CF2

+

CN+, C2H2+

COF2+

O+

C+

N+

FTMS Spectrum During CF4/CHF3/Ar Etch Process

16

51 66

69

85

93

100119

131

138 169181

0

200

400

600

800

1000

50 100 150 200 250 300

File # 1 = DATA2_P#3 @ 15 Seconds 9/18/01 4:09 PM

C 4F8 Etch of X LK (Spin-O n Low K )

2629

M ass-to-Charge (m /z)

Abu

ndan

ce (a

rb. u

nits

)

16

2629

51

69

85

119

0

100

200

300

400

500

50 100 150 200 250 300

File # 1 = DATA2_P#4 @ 21 Seconds 9/18/01 4:15 PM

C 4F6 Etch of X LK

M ass-to-Charge (m /z)

Abu

ndan

ce (a

rb. u

nits

)

Courtesy of Motorola APRDL – Austin Texas


Process Monitoring During High-Density Plasma Etch

C4F8 Etch of F54P

0

50

100

150

200

250

0 20 40 60 80 100

Time (sec)

Ion

Inte

nsity

(arb

. uni

ts)

Mass 16Mass 26Mass 29Mass 41Mass 42Mass 51Mass 66Mass 69Mass 81Mass 82Mass 85Mass 93Mass 100Mass 113Mass 119Mass 131Mass 138Mass 169Mass 181

CF3+

C3F5+

C2F5+

C2F4+C4F7

+

Portability and effective sample introduction make the Quantra a particularly useful instrument for process monitoring. Generally it is impervious to aggressive species that will damage filaments or electron multipliers.



0

50

100

150

200

250

300

350

400

450

500

0 50 100 150 200 250Time (sec)

Rela

tive

Peak

Inte

nsity

(arb

. uni

ts)

O+H3O+HF+C2H+C2H2+C2H3+N2+NO+CF+CF2+CHF2+CH2F2+COF2+CF3+C2F3+SiF3+C3F3+C2F4+C2F5+C3F5+

CF3+

C2F4+

C3F5+

Emission Trend Plot During an Etch Process

Very corrosive Fluorocarbon species – analyzed with no adverse affects to mass spectrometer performance

Mass-to-charge (m/z)

Rel

ativ

e In

tens

ity (a

rb. u

nits

)

0

100

200

300

400

20 40 60 80 100 120 140 160 180 200

CF3+

O+

SiF3+

C2F4+

C2F5+

C3F5+

C2F3+

Fluorocarbon fragments are dominant species

Mass-to-charge (m/z)

Rel

ativ

e In

tens

ity (a

rb. u

nits

)

0

100

200

300

400

20 40 60 80 100 120 140 160 180 200

0

100

200

300

400

20 40 60 80 100 120 140 160 180 200

CF3+

O+

SiF3+

C2F4+

C2F5+

C3F5+

C2F3+

Fluorocarbon fragments are dominant species Predominance of

Fluorocarbon species here indicated a plasma instability – this was corroborated by FTIR



Phosgene Measurement


PhosgeneLibrary

Phosgene spectrum

QuantraPhosgene spectrum

Library match of the low resolution Quantra spectrum was quite good.


%Phosgene

Phosgene reactor

startup modeFaulty Flow Controllers


Phosgene reactor

Shutdown mode

Faulty Flow Controllers


Chemical Warfare SurrogatesDMMP, DBBP

Chemical Warfare surrogates were obtained from Aldrich. Samples of DMMP (Dimethylmethylphosphonate C3H9PO3) and DBBP (Dibutylbutylphosphonate C12H27PO3) were analyzed using a direct liquid inlet. The samples were dissolved in methanol and a few microliters were injected into the mass spectrometer. Additional methanol laced DMMP and DBBP samples were added to a gasoline matrix. The aim of this experiment was two fold: 1. Determine what accuracy of mass measurement could be obtained on these compounds, and 2. Determine whether that accuracy would be sufficiently high enough to “target” these surrogates using their exact mass in a complex matrix.


DMMP Dimethylmethylphosphonate C3H9PO3 spectrum

(Actual mass)/Measured mass

(46.9687)/46.9706

(63.0000)/63.0030

(78.9949)/78.9952(94.0184)/94.0183

(109.0050)/109.0080

(124.0289)/124.0341

Actual Measured Delta46.9687 46.9706 -0.001963.0000 63.0030 -0.003078.9949 78.9952 -0.000394.0184 94.0183 0.0001

109.0050 109.0080 -0.0030124.0289 124.0341 -0.0052

Std Dev 0.00196


DMMP in a Gasoline Matrix

(Actual mass)/measured mass

(94.0183)/94.0179

(78.9949)/78.9942

Actual Measured Delta94.018 94.0179 0.000478.995 78.9942 0.0007

avg 0.0005


DBBP Dibutylbutylphosphonate C12H27PO3 Spectrum

Mol.wgt. 250.1698


DBBP Dibutylbutylphosphonate C12H27PO3 Spectrum

Hydrocarbon Matrix

(Actual mass)/Measured mass

(57.0704)/57.0723

Lock mass 91.0548

(96.9843)/96.9824

(138.9949)/138.9896

Actual Measured Delta57.0704 57.0723 -0.001996.9843 96.9824 0.0019

138.9949 138.9896 0.0053

avg 0.0018


DMMP

Unlikely chemical species

As can be seen from this typical exact mass fragment calculator, it is possible to “target” a compound such as DMMP, using a few millimass unit tolerance, thus drastically reducing the possible “hits” at any given mass. The mass accuracy provided by this mini FTMS is more than adequate for this type of an analysis.


Vacuum inlet

• Adsorption/Desorption• Membrane Probe• Direct Exposure Probe• Solids Probe• Liquid Injection• High Temperature Nozzle Inlet • Cryo Trapping


Standard assembly for the Vacuum inlet System

AccumulationVolume

InjectionValve

ToQuantra

MeasurementCell

VacuumPump

BlockValve

PumpDischarge

+–

Direct ExposureProbe (DEP)

End changes for each type

of inlet device

Pressure range in Vacuum inlet 1 – 10 torr

Accumulation volume with heater and controller

Metering or block Valve

Small Vacuum pump and Vacuum gauge system


High Temperature Expansion Nozzle

(Molecular Leak)

AccumulationVolume

InjectionValve

ToQuantra

MeasurementCell

VacuumPump

BlockValve

PumpDischarge

Expansion Nozzle(High Temp)


2,6 di-tert-butylnapthaleneUsing expansion nozzle

0

2000

4000

6000

8000

50 100 150 200 250

0

2000

4000

6000

8000

50 100 150 200 250

Arbitrary Y / Mass (m/z) Paged Y-Zoom CURSOR

File # 1 = NOZZLE1_P#10 @411 Seconds 7/1/03 10:54 AM Res=None

Profile data

Melting point 146oC Boiling point 336oC


Vacuum inlet – MIMS Probe

AccumulationVolume

InjectionValve

ToQuantra

MeasurementCell

VacuumPump

BlockValve

PumpDischarge

+–



MIMS/HR-FTMSTypical Isobaric Interferences – aliphatic ketone vs alkane

m/z 43

m/z 57 m/z 72

Methyl Ethyl Ketone & N-Pentane


MIMS/HR-FTMSm/z 43 and 57 doublets

57.0340

C3H5Ofrom MEK 57.0704

C4H9from Pentane

43.0184COCH3from MEK

43.0547C3H7 fromN-Pentane


Vacuum inlet Adsorption/Desorption

AccumulationVolume

InjectionValve

ToQuantra

MeasurementCell

VacuumPump

BlockValve

PumpDischarge

+–


Vacuum Lock

Trapping Media


1 ppm Benzene and Toluene 8 ml/min for 5 minutes

(carboxen 563)

The interference is the S34 isotope of CS2

m/z 77.940 vs. benzene at m/z 78.047

~1200 resolution


100 ppb Benzene and Toluene 8 ml/min for 1 minute

(mol sieve 5A)


10 ppb Benzene and Toluene 20 ml/min for 15 minutes

(mol sieve 5A)


Blank


Quantra as a field unit

At ambient temperature the Quantra will draw 300 watts, with heaters on approximately 700 watts, runs from a

generator or a voltage inverter, and typically uses a laptop


Conclusion:

Adverse effects from corrosive and aggressive species are minimized due to design features of a new miniature FTMS.

The system is capable of process analysis as well as transportable field analysis of species ranging from phosgene gas and various fluorocarbons to NF3 and chemical weapons surrogates with little or no detriment to the mass spectrometer.

While pure elemental analysis, requiring 5 ppm or less mass accuracy is not readily seen above m/z 100 with this instrument, targeting species by their exact mass with subsequent identification – even in a complex background is definitely possible.

The system works very well in the field!


Acknowledgements:

Victor Vartanian, Brian Goolsby,

Motorola – Austin, Texas

Steve Beu

Beu Consulting – Austin, Texas

Dean Davis, Lonnie O’bannion

Siemens Applied Automation – Bartlesville, OK


Questions?

[email protected]

918 519 3071 Cell

http://www.sea.siemens.com/ia/product/aiquantra.html

Applications

Budget Quote

Specifications

novel miniature ftms for analysis of corrosives and...

Documents