Simultaneous Measurements of Carbon, Hydrogen, Nitrogen, Sulfur, and Oxygen with

Thermal/Optical Analysis

John G. Watson1, Gustavo M. Riggio1, Xiaoliang Wang1, L.-W. Antony Chen1,2, Xufei Yang3, Jana Diab4,

Ralf Zimmermann4, and Judith C. Chow1 1Desert Research Institute, Reno, NV

2University of Nevada, Las Vegas, NV 3Montana Tech, Butte, MT

4University of Rostock, Rostock, German

Presented at:

11th International Conference on Carbonaceous Particles in the Atmosphere Berkeley, CA

August 11, 2015

Motivation • More than 100,000 thermal/optical analyses (TOA) are

performed worldwide on quartz-fiber filters each year, including long-term trends networks in the U.S., Canada, and China

• It is desirable to obtain more information from these analyses beyond simple carbon fractions (e.g., organic and elemental carbon [OC and EC]) at no added cost

• New developments in detector technology show potential for expanding the components quantified by thermal methods

Objectives

• Identify approaches for expanding thermal analysis from carbon (C) to hydrogen (H), nitrogen (N), sulfur (S), and oxygen (O) and their associated compounds

• Demonstrate that the long-term OC/EC trends record can be maintained by detector modifications

Approach 1: Emulate the AMS for filters (TOA-QMS*)

Riggio, G.M. (2015). Development and application of thermal/optical- quadrupole TOA-QMS mass spectrometry for quantitative analysis of major particulate matter constituents., M.S. Thesis, University of Nevada Reno, Reno, NV.

*TOA-QMS: Thermal/Optical Analyzer with Quadrupole Mass Spectrometry

The IMPROVE temperature program is simplified for testing Temperature Steps (in helium atmosphere):

• 80 oC – Desorption of H2O • 580 oC – Combustion of most species (OC4 of IMPROVE_A) • 840 oC – Combustion of remaining species (EC3 of IMPROVE_A)

Chow et al. (2007). The IMPROVE_A Temperature Protocol for Thermal/Optical Carbon Analysis: Maintaining Consistency with a Long-Term Database. J. Air & Waste Manage. Assoc. 57:1014-1023.

Air infiltration during sample insertion is reduced with higher pressure and an argon sheath

Argon sheath chamber

Increase sample oven pressure from 5 to 10 psi

Reduce O2 from 15.5 to 8.5 ppm

CO

O2

Ar

TOA-QMS spectra are similar , but not identical, to AMS spectra

• Potential causes of differences -Heating rate

-Particle collection medium

-Thermal desorption

-Ionization

SO+

SO2+

NO+

NO2+

AMS Spectra from: Allan et al. (2004). A generalised method for the extraction of chemically resolved mass spectra from aerodyne aerosol mass spectrometer data. J. Aerosol Sci., 35(7):909-922. Jimenez et al. (2003). Ambient aerosol sampling using the Aerodyne aerosol mass spectrometer. J. Geophys. Res., 108(D7):doi:10.1029/2001JD001213.

Signal/response is determined by analysis of quartz filter samples of nebulized NH4NO3, (NH4)2SO4, and oxalic acid solutions.

NO3-

SO4=

=

NH4+ OC

Application to ambient samples shows good correlation, but systematic biases

LOD 1.59 µgcm-2

LOD 0.07 µgcm-2

(58 Fresno, CA, samples; Dec 2000 – Feb 2001) NH4

+

LOD 0.96 µgcm-2

SO4=

NO3-

LOD 4.22 µgcm-2

OC

Concentration variations are similar to those obtained from speciation analyses

(Fresno supersite, 58 samples; Dec 2000 – Feb 2001)

Largest concentration of species between 16:00 and 24:00 Local Standard Time

Haze Period

Approach 2: Oxidize thermally-evolved products to simpler compounds (TOA-O-QMS/NDIR)

Flow Control Network

CHNS Reactor(MnO2)

C→CO2, H→H2O, N→NOx/N2, S→SO2

NDIR CO2 Detector

Carrier/ReactionGases

98%

He,

2% O

2

He

He,

CH

4

He,

O2,

NO

, SO

2

CalibrationGases

Oven

Filter LoadingPush Rod

UV-VIS-NIRLight Source(λ=405-980 nm)

ReflectanceDetector

TransmittanceDetector

OpticalFibers

Filter

FilterHolder

ThermocoupleHeated Fused

Silica Capillary

MassSpectrometer

Vent

Four-Way Solenoid Valve

FlowSplitter

Outputs:Reflectance/

Transmittance Spectra

O

Mass Spectra

C, H, N, S

O Reactor(Ni/C)

O→CO

Oxidation Oven

(MnO2)CO→CO2

Oxidation ReactorC→CO2

FIDMethanatorCO2→CH4

From Oven

DRI Model 2001 Carbon Analyzer

C

DRI Model 2001Carbon Analyzer

Outputs:

ThermalFragmentation

Oven650°C

Time (min)20 40 60 80

Ion

Sign

al (a

.u.)

0.0

2.0e+4

4.0e+4

6.0e+4

8.0e+4

1.0e+5

1.2e+58.0e+51.0e+61.2e+6

Ove

n Te

mpe

ratu

re (°

C)

0

200

400

600

800

1000

m/z=44 (CO2+)

m/z=18 (H2O+) m/z=30 (NO+)

m/z=64 (SO2+)

CalibrationCH4 Injection

Temperature

100% He 98% He / 2% O2

140°C280°C

480°C

580°C

740°C

840°C(a)

m/z=28 (CO+, N2+)

time vs Temp Time vs m/z18 Time vs m/z28 Time vs m/z30 Time vs m/z64

Time (min)20 40 60 80

ND

IR S

igna

l (m

V)

40

60

80

100400500

Ove

n Te

mpe

ratu

re (°

C)

0

200

400

600

800

1000

140°C

280°C480°C

580°C

CalibrationO2 Injection

NDIR

Temperature

100% He(b)

Existing thermal/optical protocols can be adapted to quantify C, H, N, S, and O

y = 1.02x - 0.58R² = 0.98

0

20

40

60

80

0 20 40 60 80OC

by D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

OC by DRI Model 2001 C Analyzer (µg/cm2)

OC (DRI CHNSO vs. Model 2001 C Analyzer)

y = 1.07x - 0.12R² = 0.97

0

5

10

15

20

0 5 10 15 20EC b

y D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

EC by DRI Model 2001 C Analyzer (µg/cm2)

EC (DRI CHNSO vs. Model 2001 C Analyzer)

Fresno and Baltimore ambient samples (N=87)

Instrument signals are linear with C, H, N, S, and O quantities for model compounds

y = 0.046xR² = 0.998

0.0

1.0

2.0

3.0

4.0

0 20 40 60 80

MS

m/z

=44

(CO

2+ ) S

igna

l

Mass of C (µg)

C:

SulfanilammideL-CystineCO2CH4

y = 0.132xR² = 0.994

0.0

0.5

1.0

1.5

2.0

2.5

0 5 10 15 20M

S m

/z=1

8 (H

2O+ )

Sig

nal

Mass of H (µg)

H:

SulfanilammideL-Cystine(NH4)2SO4NH4NO3CH4

y = 0.017xR² = 0.955

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

0 20 40 60 80

MS

m/z

=30

(NO

+ ) S

igna

l

Mass of N (µg)

N:

SulfaniammideL-Cystine(NH4)2SO4NH4NO3

y = 0.012xR² = 0.993

0.0

0.2

0.4

0.6

0.8

0 20 40 60 80

MS

m/z

=64

(SO

2+ ) S

igna

l

Mass of S (µg)

S:

SulfanilammideL-Cystine(NH4)2SO4

Calibration compounds: • Ammonium nitrate: NH4NO3; • Ammonium sulfate: (NH4)2SO4; • Benzoic acid: C7H6O2; • Carbon dioxide: CO2; • L-Cystine: C6H12N2O4S2; • Methane: CH4; • Nonadecanol: C19H40O; • Pentadecanoic acid: C15H30O2; • Sulfanilamide: C6H8N2O2S

y = 0.0168xR² = 0.986

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 20 40 60 80

ND

IR S

igna

l (m

V)

Mass of O (µg)

Benzoic acidNonadecanolPentadecanoic acid

O:

NDIR signal is linear with O quantities in calibrated chemicals

Benzoic acid: C7H6O2; Nonadecanol: C19H40O Pentadecanoic acid: C15H30O2

Challenges: • H2O bound to filters and particles • Intrusion of ambient O2 into the analyzer

CHNSO concentrations are comparable with

other methods

y = 0.95x - 1.77R² = 0.93

0

20

40

60

80

0 20 40 60 80TC b

y D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

TC by by Thermo CHNSO Analyzer (µg/cm2

TC (DRI vs. Thermo CHNSO Analyzer)

y = 0.54x + 3.62R² = 0.69

0

5

10

15

20

0 5 10 15 20

H b

y D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

H by Thermo CHNS0 Analyzer (µg/cm2)

H (DRI vs. Thermo CHNSO Analyzers)

y = 0.99x - 3.19R² = 0.95

0

20

40

60

0 20 40 60N b

y D

RI C

HN

SO A

naly

zer

(µg/

cm2 )

N by Thermo CHNSO Analyzer (µg/cm2)

N (DRI vs. Thermo CHNSO Analyzers)

y = 1.00x + 0.13R² = 0.99

0

10

20

30

40

50

60

0 20 40 60

N b

y D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

N (NH4+ + NO3

-) by ion chromatography (µg/cm2)

N (DRI CHNSO Analyzer vs. IC)

y = 0.95x - 0.03R² = 0.94

0

5

10

15

20

0 5 10 15 20S by

by

DRI

CH

NSO

Ana

lyze

r (µg

/cm

2 )S (SO4

=) by Ion chromatograph (µg/cm2)

S (DRI CHNSO Analyzer vs. IC)

y = 1.10x - 5.50R² = 0.92

0

20

40

60

80

100

120

0 20 40 60 80 100 120O b

y D

RI C

HN

SO A

naly

zer (

µg/c

m2 )

O by Thermo CHNSO Analyzer (µg/cm2)

O (DRI vs. Thermo CHNSO Analyzer)

• Thermo Flash EA1112 CHNS/O Analyzer

• Dionex Model ICS-3000 Ion Chromatographs (IC)

Fresno and Baltimore ambient samples (N=87)

Composition varies between summer and winter (Fresno, California)

Abundant (NH4)2SO4 in summer (Decompose at 200‒400 °C; OC2 at 280 °C in 100% Helium) (EC/TC=0.23)

Abundant NH4NO3 in winter (Dissociation starts at room temperature; OC1 at 140 °C in 100% Helium) (EC/TC=0.28)

0.0

3.0

6.0

9.0

12.0

15.0

OC1 OC2 OC3 OC4 EC1 EC2 EC3

Con

cent

ratio

n (µ

g/cm

2fil

ter)

Fresno Summer

OSNHC

Seasonal variability in the CHNS-O composition of Fresno ambient samples (N = 67)

Source profiles vary between smoldering and flaming wood smoke for thermal carbon fractions

Smoldering wood smoke shows lower EC:TC and higher O:C ratios than flaming smoke.

0

100

200

300

400

500

600

OC1 OC2 OC3 OC4 EC1 EC2 EC3

Con

cent

ratio

n (µ

g/cm

2fil

ter)

Flaming Wood Smoke (SDKQ073)

OSNHC

EC/TC=0.02

EC/TC=0.25

Fraction Molar Ratios H:C O:C

OC 1.33 0.27 EC 0.14 NA TC 1.04 0.20

Fraction Molar Ratios H:C O:C

OC 1.55 0.41 EC 0.93 NA TC 1.54 0.40

0

30

60

90

120

150

OC1 OC2 OC3 OC4 EC1 EC2 EC3

Con

cent

ratio

n (µ

g/cm

2fil

ter)

Smoldering Wood Smoke (BIOQCF101)

OSNHC

Elemental analysis of organic standards shows consistent O/C and

H/C relationships 1:1

Lower N/C ratio may be due to inadequate number of samples and unaccounted N species

Approach 3: Detect thermal output with photon ionization time-of-flight mass spectrometry (TOA-PI-TOFMS)

Diab et al. (2015). Hyphenation of a EC/OC thermal-optical carbon analyzer to photo ionization time-of-flight mass spectrometry: A new off-line aerosol mass spectrometric approach for characterization of primary and secondary particulate matter. Atmos. Meas. Tech. Discuss., (8):269-308. Grabowsky et al. (2011). Hyphenation of a carbon analyzer to photo-ionization mass spectrometry to unravel the organic composition of particulate matter on a molecular level. Anal. Bioanal. Chem., 401(10):3153-3164.

Soft ionization doesn’t fragment components, mass spectra are more complex, but individual compounds are quantified

Phenanthrene

Pyrene

Benz(a)anthracene or chrysene Retene

Benzo(a)pyrene or benzo(a)fluoranthene

Phenols

*OC1-OC3 are OC fractions evolved at 140, 280, and 480°C in helium atmosphere following IMPROVE_A protocol

*

*

*

Propene

C2H4O2+

Pentene or butene Furfurylalcohol

Guaiacol Octonoic acid or naphthol

Ethylguaiacol or vanillin

Methylguaiacol

Guaiacol

Vinylguaiacol

Distinct temperature/ion profiles are discernable, even without identifying individual compounds

Diesel Exhaust Gasoline Exhaust Grabowsky et al. (2011). Hyphenation of a carbon analyzer to photo-ionization mass spectrometry to unravel the organic composition of particulate matter on a molecular level. Anal. Bioanal. Chem., 401(10):3153-3164.

These approaches seem to be feasible, but not yet practical

Approaches 1 and 2

Approach 3

Mini mass spectrometers are demonstrating sufficient sensitivity for ambient concentrations

Torion Technologies Inc. American Fork, UT, http://torion.com/home.html.

Microsaic Systems. Abingdon, UK, http://www.microsaic.com/home/

Aston Labs, Purdue University, Lafayette, IN, http://aston.chem.purdue.edu/research/instrumentation/miniature-mass-spectrometers.

Challenges for Enhanced Chemical Characterization of Filter Samples

• Perfecting, evaluating, and making more efficient procedures for additional characterization

• Modifying instrumentation and procedures to incorporate more specific analyses methods into long-term chemical speciation networks to obtain more information from existing samples

• Maintaining continuity and consistency with the long-term trends data sets

• Developing more detailed source profiles with these methods for speciated inventories and source apportionment

Acknowledgements

• U.S. National Science Foundation (CHE 1214163)

• National Park Service IMPROVE Carbon Analysis Project (C2350000894)