1

Elemental Analysis (EA) of Organic Species Elemental Analysis (EA) of Organic Species with Electron Ionization Highwith Electron Ionization High--Resolution Mass Resolution Mass

Spectrometry (EISpectrometry (EI--HRMS)HRMS)Allison C. Aiken,1,2 Peter F. DeCarlo,2,3 Ingrid Ulbrich,1,2 Ken Docherty,2

Jose L. Jimenez1,2

University of Colorado at Boulder, Boulder, CO, 80309, U.S.1Department of Chemistry and Biochemistry, 2Cooperative Institute for Research in the Environmental Sciences(CIRES), 3 Department of Atmospheric and Oceanic Sciences (ATOC)

Jesse H. Kroll, Doug R. Worsnop Aerodyne Research Inc., 45 Manning Rd., Billerica, MA, 01821, U.S.

Qi Zhang, Yele SunDepartment Of Environmental Health Sciences, University at Albany, SUNY, Albany, NY, U.S.

Paul ZiemannDepartment Of Environmental Sciences, University of California, Riverside, CA, U.S.

Rami Alfarra, Andre Prevot, Josef Dommen, Jonathan Duplissy, Axel Metzger, Urs Baltensperger Laboratory of Atmospheric Chemistry, Paul Scherrer Institut, Switzerland

Monday, October 1, 20078th AMS User’s Mtg. DRI, Reno, NVFunding: NASA, EPA, NSF

Bulk Organic Aerosol Analysis• Historically poorly characterized

– Thousands of species– No technique or combination can identify all species present in Organic

Aerosol (OA) mass– Approaches that characterize bulk organic types

• Elemental Analysis (ex: Atomic O/C)– Traditional analysis with offline filters (~ 1 mg sample)– NAMS: Nano-aerosols using laser ablation MS (Wang, Zordan, Johnston

2005 Anal. Chem.)– Thermal method for Organic O (Pang, Turpin, Gundel 2006 AS&T)– FTIR Functional group analysis (S. Gilardoni, L. Russell et al. 2007 JGR) – STXM C and O Edges (Tivanski/M. Gilles 2007 J. Phys. Chem. A)– First application with unit mass resolution (UMR) AMS data (Zhang 2005

ACP)

• This work using with high-resolution AMS (HR-ToF-AMS)– High time and size resolution– No filter sampling artifacts– Applicable to many AMS Field Data sets

2

Elemental Analysis (EA) by EI-HRMS • EI MS Ion intensities α original mass concentration• Identify all ions (HRMS)• Determine Elemental Mass

– Atomic and Mass Ratios– Calibrate method

• Possible frag. biases

600400200

0

Raw

Sig

nal (

A.U

.)

57.1057.0557.00m/z

C3H5O+ C4H9

+Nonanal spectra (AMS)

1086420

Per

cent

of T

otal

Sig

nal (

%)

100908070605040302010

m/z

CxHy+

CxHyOz+

OHx+

1086420

Per

cent

of T

otal

Sig

nal (

%)

100908070605040302010

m/z

1086420

CxHy+

CxHyOz+

OHx+

CHO

Fitting - P. DeCarlo et al., Anal Chem, 2006.

OH

C

Atomic Ratios:O/C = 0.06 (0.11)H/C = 1.8 (2.0)

Potential Positive Ion EI-MS Biases• Fragmentation with respect to electronegativity

– Possibility for more electronegative ions to be preferentially lost as neutrals

– Do the HRMS fragments represent the original composition?– Largest issues for Oxygen

• Rearrangements (ex: McLafferty):

• Is quantitative elemental analysis still possible despite these biases?

Original Molecule:

e-

+

+ Detected Ion

e-Lost Neutral

3

NIST EI Database Analysis• National Institute of Standards and Technology (NIST) Database of EI

MS– Not HRMS, Unit mass resolution (UMR)– Limited range of compounds (16 < MM < 90)

• 20 compounds– Complete MS used– ~ 3 replicates each (from different labs)

• Atomic H/C and O/C determined, some bias but calibration is possible

2.0

1.5

1.0

0.5

0.0

EA

O/C

2.01.51.00.50.0Atomic O/C

NIST fit f(x) = 0.86(x) ± 0.07 (95% CI)R2 = 0.95

5

4

3

2

1

0

EA

H/C

543210Atomic H/C

NIST fit f(x) = 0.75(x) ± 0.05 (95% CI)R2 = 0.92

R2 = 0.95R2 = 0.95

HR-ToF-AMS Laboratory Experiments• 35 Organic Compounds• 114 < MM < 426• Organic Aerosols (CHNO)

– POA’s (S&P, 1998)– Markers: Levoglucosan, Cholesterol

• EA biases specific to this data– Compounds – Sampling Conditions

• Can calculate Organic-Mass-to-Organic-Carbon ratios (OM/OC)

0.8

0.6

0.4

0.2

0.0

EA O

/C

0.80.60.40.20.0Atomic O/C

AMS fit f(x) = 0.60(x) ± 0.05 (95% CI)R2 = 0.88

2.5

2.0

1.5

1.0

0.5

0.0

EA H

/C

2.52.01.51.00.50.0Atomic H/C

AMS fit f(x) = 0.93(x) ± 0.02 (95% CI)R2 = 0.96

0.30

0.20

0.10

0.00

EA N

/C

0.300.200.100.00Atomic N/C

AMS fit f(x) = 0.77(x) ± 0.13 (95% CI)R2

= 0.78

30% avg uncertainty 17% avg uncertainty

R2 = 0.95R2 = 0.95

R2 = 0.78

4

Applicability of the Calibrations• Laboratory standard calibration• Application to Ambient/Data sampled “in Air”

– Calibrations in progress– Preparing a panel

• Changes to the fragmentation table• Email release

• Will email the group when paper is published• A whole new “ANIMAL”

– ANalysis of Ions Measuring Atomic Levels– Maybe even APES!

– Analytical Procedure for Elemental Separation

Mexico City Ground EA (T0, MILAGRO)

0.7

0.6

0.5

0.40.3A

tom

ic O

/C

3/11/06 3/12/06 3/13/06 3/14/06 3/15/06

2.01.51.00.50.0

OM

/OC

12840C

O2+ /O

rg(%

)

0.00

Atom

ic N/C

0.00

Atom

ic H

/C

0.7

0.6

0.5

0.40.3A

tom

ic O

/C

3/11/06 3/12/06 3/13/06 3/14/06 3/15/06

2.01.51.00.50.0

OM

/OC

12840C

O2+ /O

rg(%

)

0.00

Atom

ic N/C

0.00

Atom

ic H

/C

0.7

0.6

0.5

0.40.3A

tom

ic O

/C

3/11/06 3/12/06 3/13/06 3/14/06 3/15/06

2.01.51.00.50.0

OM

/OC

12840C

O2+ /O

rg(%

)

0.00

Atom

ic N/C

1.81.71.61.51.41.3At

omic

H/C

0.7

0.6

0.5

0.40.3A

tom

ic O

/C

3/11/06 3/12/06 3/13/06 3/14/06 3/15/06

2.01.51.00.50.0

OM

/OC

12840C

O2+ /O

rg(%

)

0.15

0.10

0.050.00

Atom

ic N/C

1.81.71.61.51.41.3At

omic

H/C

0.7

0.6

0.5

0.40.3A

tom

ic O

/C

3/11/06 3/12/06 3/13/06 3/14/06 3/15/06

2.12.01.91.81.71.6

OM

/OC

12840C

O2+ /O

rg(%

)

0.15

0.10

0.050.00

Atom

ic N/C

1.81.71.61.51.41.3At

omic

H/C

AVG’s: O/C = 0.5, H/C = 1.5, N/C = 0.05, OM/OC = 1.84

5

Fast Measurements from Aircraft over Mexico City

P. DeCarlo et al., C-130, MILAGRO (2006)

40

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

Organic Mass

40

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

Organic Mass NO340

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

Organic Mass NO3

Over Mexico City Mexico City OutflowRegional Background

40

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

1.00.80.60.4

O/C

2.42.22.01.8O

M/O

C

1.61.41.21.00.80.6

H/C

O/C Atomic Ratio Organic Mass NO3

Regional Background Mexico City OutflowOver Mexico City

40

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

1.00.80.60.4

O/C

2.42.22.01.8O

M/O

C

1.61.41.21.00.80.6

H/C

O/C Atomic Ratio H/C Atomic Ratio Organic Mass NO3

Regional Background Mexico City OutflowOver Mexico City

40

30

20

10

0

Org

anic

Mas

s (µ

g/m

3 )

7:00 PM3/8/2006

8:00 PM 9:00 PM 10:00 PM 11:00 PM 12:00 AM3/9/2006

Time (UTC)

1.00.80.60.4

O/C

2.42.22.01.8O

M/O

C

1.61.41.21.00.80.6

H/C

OM/OC O/C Atomic Ratio H/C Atomic Ratio Organic Mass NO3

Regional Background Mexico City OutflowOver Mexico City

Atomic (O/C) Ratio

0

0.2

0.4

0.6

0.8

1

Mex G

rnd AM

Mex G

rnd PM

Mex P

lane C

ity

Mex P

lane O

utflow

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

PSI Isop

rene

UC-Rive

rside

Toluen

e

Primary, Secondary, and Ambient OATotalHOABBOAOOA-IOOA-IISOA-C

Mass (OM/OC) Ratio

11.21.41.61.8

22.2

Mex G

rnd A

M

Mex G

rnd P

M

Mex Plan

e City

Mex Plan

e Outf

low

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

Mex G

rnd

Mex P

lane

PSI Isop

rene

UC-Rive

rside

Toluen

e

TotalHOABBOAOOA-IOOA-IISOA-C

6

Conclusions• EA successful with HR-ToF-AMS

– High time resolution with very small sample sizes– Biases in fragmentation can be calibrated– EA can be used to estimate OM/OC

• O/C for aerosol types – ~ 30% avg. error (lab standards, potentially better for Ambient)– “Aged” SOA/OOA > “Fresh” SOA/OOA ~ Chamber SOA > POA– BBOA ~ Fresh OOA

• N/C is low for OA in Mexico City– ON ~ 20% of TN– Contrast with fog (Q. Zhang, AAAR 2007)

• Need for intercomparisons of emerging methods• Calibrations for Ambient measurements to come

– Email release of panel and paper announcement