Novel Measurements of Volatility- and Polarity-Separated ... · Masters: Xiaochen Zuo, Mayura...
-
Upload
dangnguyet -
Category
Documents
-
view
218 -
download
0
Transcript of Novel Measurements of Volatility- and Polarity-Separated ... · Masters: Xiaochen Zuo, Mayura...
Novel Measurements of Volatility- and Polarity-Separated Organic Aerosol Composition and Associated Hygroscopicity to
Investigate the Influence of Mixed Anthropogenic-Biogenic Emissions on Atmospheric Aging Processes
Brent J. WilliamsWashington University in St. Louis
Department of Energy, Environmental and Chemical Engineering
U. Colorado – Boulder and CIRES (Jimenez)U. California – Berkeley (Goldstein)Aerosol Dynamics Inc. (Berkeley, CA)Aerodyne Research Inc. (Billerica, MA)
U. Minnesota (Millet)Georgia Tech (Ng, Weber)Penn State University (Brune)WashU (Turner)
R835402
Acknowledgements: Participating Lab Members
Research Scientist: Munkhbayar BaasandorjMasters: Xiaochen Zuo, Mayura PatankarUndergrads: Hagan, Dhawan, Geyer, Streff, Sussman, Wilcosky
PhD
• Atmospheric primary and secondary organic gases and particles originate from biogenic and anthropogenic emissions, and bio/anthro interactions have been observed to alter production of secondary organic aerosol (SOA).
• Impacts health and climate
• *Uncertainty remains in determining sources and evolution of organic aerosol (OA)
• *There is a need for improved chemical characterization of complex OA composition
• *We want manageable parameterizations of this chemistry to enable modeling of complex atmospheric OA (mixture can contain tens of thousands of compounds). Approach through novel data analysis methods and new measurement techniques.
Motivation
Organic aerosol is an intermediate in the oxidationof most organics to CO2.
Secondary Organic Aerosol (SOA)
CO2
CO
C40
Organic Aerosol
gas
particle
Adapted from Kroll et al., Nature Chemistry, 2011
Oxidative Evolution of Organic Material in Atmosphere
CH4
Oxidative Evolution of Organic Material in Atmosphere
• A small molecule like isoprene (C5) only crosses edge of the condensed phase• More likely for larger molecules (e.g., C10, C15) to cross through condensed phase• Aerosol Yield from smaller molecules may be very sensitive to altered reaction paths
OrganicAerosol
Organic Gases
Adapted from Kroll et al., Nature Chemistry, 2011
Spracklen et al., ACPD, 2011
Our goal is to contribute to a better understanding of the anthropogenic
enhancement of biogenic SOA production
Primary Organic Aerosol Secondary Organic Aerosol
OUTLINE
• Instrument Development
• Novel Data Analysis Methods
• Field observations: SOAS – Centreville, AL
• Field observations: SLAQRS – East St. Louis, IL
• Laboratory-based oxidation studies
• Conclusions
Our methods for In-Situ Organic Aerosol Chemical Characterization
Thermal desorption Aerosol Gas chromatograph (TAG)
Volatility and Polarity Separator (VAPS)
Aerosol Mass Spectrometer (AMS)
Collection and Thermal Desorption Cell
GC Oven
QuadMS
30m Non-Polar Column
Helium
Pola
rity
Volatility
R+
ToF-MS
Particle Time of Flight Chamber
Displaying Organic Ions Only
High Resolution MS Vol. & Pol.- Separated MS
Canagaratna et al, MS Reviews
2007
Williams et al, AS&T 2006
Goldstein et al, J.ChromA 2008Martinez et al, AS&T 2016
15.00 20.00 25.00 30.00 35.00 40.00 45.000
500000
1000000
1500000
2000000
2500000
3000000
3500000
4000000
4500000
5000000
Time-->
AbundanceTIC: 33702-12.DTIC: 33702-13.D (*)
Chromatography-Separated MS
Collection and Thermal Desorption Cell
Transfer Line
Polar Column
Modulator
ToFMS
Helium
- Faster- More Mass- Parameterized Chemistryfor models
Organic Aerosol Analysis in Perspective
Tim
e R
eso
luti
on
High
Low
100
80
60
40
20
0
Co
mp
lete
nes
s(%
of
Mas
s A
nal
yzed
)
Chemical Resolution
Useless
Instrument
AMS
Useless
Instrument
Perfect
Instrument
TAG
AMS
Increase
completeness
Increase
classification
Adapted from Hallquist et al., ACP, 2009
VAPS – Martinez et al., AS&T, 2016
VAPS1 second
1 hour
1-5 minute
20-30 minute
Aerodyne Research Inc. (US-DOE – SBIR, Phases I, II): Volatility and polarity separated total organic aerosol using thermal desorption modulated chromatography
Vehicle
Secondary Vehicle
Biogenic
Secondary Biogenic
Biomass Burn
SecondaryBiomass Burn
Industrial
Secondary Industrial
Group covarying chemical signals:Positive Matrix Factorization (PMF)
Match to known source profiles Factors can also represent atmospheric transformation processes
Sample and separate using complementary techniques:AMS (Mass Spec. of organic and inorganic components)VAPS (organic components with additional chemical separation)TAG (individual organic “marker” compounds)
In-Situ Measurements
x Time Series
OUTLINE
• Instrument Development
• Novel Data Analysis Methods
• Field observations: SOAS – Centreville, AL
• Field observations: SLAQRS – East St. Louis, IL
• Laboratory-based oxidation studies
• Conclusions
New binning method developed to directly insert a collection of TAG chromatograms or VAPS “vapograms” into PMF.
• Saves from having to integrate individual compounds• Incorporates the unresolved complex mixture (UCM)
Zhang et al, AS&T 2014
Chromatogram “Vapogram”
600x103
500
400
300
200
100
0
TIC
50403020
Retention Time (min)
Total signal Column bleed
Resolved peaks
UCM
Bin size determines degree of chemical resolution
Pola
rity
Volatility
Binned PMF data matrix can be arranged to separate for:
Zhang et al, AS&T 2014 + In-Prep.
1) Major chromatogram components(e.g., compounds with similar mass spectral features or functionality)
Chromatogram Deconvolution PMF
2) Major study-time components(e.g., Sources or Transformations)
Source Apportionment PMF
10
8
6
4
2
0
Mass C
once
ntr
atio
n (
mg m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
CDT
AMS Organics VAPS_Isoprene SOA VAPS_Terpene SOA3 VAPS_Terpene SOA2 VAPS_BBOA VAPS_Alkanes VAPS_Terpene SOA1
Alkanes
Acids
OUTLINE
• Instrument Development
• Novel Data Analysis Methods
• Field observations: SOAS – Centreville, AL
• Field observations: SLAQRS – East St. Louis, IL
• Laboratory-based oxidation studies
• Conclusions
Observations from two recent field studies
St. Louis Air Quality Regional Study (SLAQRS) Aug.3 – Oct.15, 2013
Southern Oxidant and Aerosol Study (SOAS) Jun.1 – July.15, 2013
Modeled Isoprene Emissions
Millet et al. (2008)
SLAQRS
SOAS
Now known that isoprene can produce SOA….
Formation of Isoprene SOA(anthropogenic influences)
isoprene+OH,O2
+H2O
+SO4-2
+NO3-
+H2O
+SO4-2
+NO3-
RO2
+HO2
IEPOX
+NO
+OH, NO2
MPAN
+OH
MAE
Low
-NO
xPa
thH
igh
-NO
xPa
th Gas phase
Pa
rtic
le p
ha
se
Emission
Organosulfates?Organonitrates?Oligomers?
Figure from Surratt et al.
Goldstein, Fung, et al, 2009
(satellite observations)
+NO3,O2
…some to
Particle
12
8
4
0
6/1/13 6/21/13 7/11/13 7/31/13 8/20/13 9/9/13 9/29/13
20
15
10
5
0
6/1/13 6/21/13 7/11/13 7/31/13 8/20/13 9/9/13 9/29/13
120
80
40
0
6/1/13 6/21/13 7/11/13 7/31/13 8/20/13 9/9/13 9/29/13
Alabama ground site (SOAS) East St. Louis (SLAQRS)
SOAS Isoprene (deGouw & Goldstein et al.)
SOAS NOx from SEARCH CTR database (ARA)
pp
bp
pb
SOAS SO2 from SEARCH CTR database (ARA)
pp
b
Isoprene
NOx
SO2Precursor to Anthropogenic Sulfate Aerosol
Xu et al., PNAS, 2015AMS Factors from SOAS
AMS only finds 1 type of isoprene SOA, although there are various formation pathways. AMS may not easily differentiate.
Some lab evidence showing LO-OOA may be terpene SOA.
Can VAPS offer more information?
VAPS - SOASRural Alabama(lots of Isoprene and Terpene SOA)
VAPS – SLAQRSUrban St. Louis (more POA)
2D-VBS Oxidation pathways(Donahue et al., ACPD 2012)
B)
Tota
l Io
n S
ign
al
Po
lari
ty
Tota
l Io
n S
ign
al
Martinez et al, AS&T 2016
First Field deployments for VAPS
A)
POA
10.5
0-0.5
-1-1.5
-2
OS c C5H8 Isoprene
C10H16 Terpenes
C)
Column Bleed
TerpeneSOA2 O:C=0.36
TerpeneSOA3 O:C=0.5
Alkanes
Terpenes O:C=0.12
TerpeneSOA1 O:C=0.19
IEPOX-derived
Organic Sulfates
Acetic Acid
Thermal Decomp. – LVOCs (m/z44) NO3
-, SO42-
Decomposition
Acetic Acid
OrganicSulfates
IEPOX-derived
TerpeneSOA1
Terpenes
Alkanes
TerpeneSOA3
TerpeneSOA2
GCColumn
Chromatogram Deconvolution PMF on Binned VAPS Data (SOAS)
Source Apportionment PMF on Binned VAPS Data (SOAS)
10
8
6
4
2
0
Mass C
once
ntr
atio
n (
mg m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
CDT
AMS Organics VAPS_Isoprene SOA VAPS_Terpene SOA3 VAPS_Terpene SOA2 VAPS_BBOA VAPS_Alkanes VAPS_Terpene SOA1
10
8
6
4
2
0
Mass C
once
ntr
atio
n (
mg
m-3
)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
CDT
AMS Organics VAPS_Isoprene SOA VAPS_Terpene SOA3 VAPS_Terpene SOA2 VAPS_BBOA VAPS_Alkanes VAPS_Terpene SOA1
10
8
6
4
2
0
Mass C
once
ntr
atio
n (
mg m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
AMS Organics GTech_BBOA GTech_Isoprene-OA GTech_LO-OOA GTech_MO-OOA
VAPS PMF
AMS PMF(Ng, GTech)
Correlations (Pearson r)
SV-TAG (UC-Berkeley) AMS
C5
alk
en
e t
rio
l1
C5
alk
en
e t
rio
l2
2-m
eth
ylte
tro
l 1
2-m
eth
ylte
tro
l 2
me
thyl
glyc
eri
cac
id
pin
icac
id
pin
on
icac
id
hyd
roxy
glu
tari
cac
id
levo
glu
cosa
n
Iso
pre
ne
OA
LO-O
OA
MO
-OO
A
BB
OA
Iso
pre
ne
OA
+ M
O-O
OA
VA
PS
Isoprene SOA 0.67 0.67 0.53 0.55 0.45 -0.04 0.29 0.55 0.22 0.43 0.04 0.65 0.28 0.67
Terpene SOA1 -0.18 -0.25 -0.10 -0.28 -0.46 0.56 0.27 0.17 0.19 0.24 0.56 0.14 0.28 0.06
Terpene SOA2 0.43 0.39 0.59 0.50 0.08 0.70 0.40 0.65 0.60 0.23 0.54 0.60 0.30 0.55
Terpene SOA3 0.44 0.32 0.51 0.31 0.17 0.73 0.79 0.55 0.35 0.66 0.86 0.55 0.70 0.65
BBOA 0.25 0.15 0.18 0.08 0.20 0.29 0.63 0.27 0.15 0.37 0.62 0.10 0.67 0.26
IsopreneSOA + TerpeneSOA3 0.78 0.69 0.73 0.60 0.40 0.47 0.75 0.76 0.40 0.80 0.67 0.87 0.73 0.92
TerpeneSOA1+2 + BBOA 0.31 0.20 0.43 0.23 -0.11 0.87 0.70 0.64 0.55 0.45 0.93 0.51 0.67 0.52
AM
S
IsopreneOA 0.78 0.67 0.69 0.61 0.68 0.40 0.62 0.52 0.25
LO-OOA 0.52 0.38 0.53 0.38 0.23 0.83 0.85 0.68 0.51
MO-OOA 0.77 0.74 0.79 0.73 0.43 0.54 0.54 0.85 0.46
BBOA 0.62 0.54 0.50 0.40 0.28 0.51 0.71 0.61 0.50
Isoprene+MO-OOA 0.79 0.70 0.81 0.68 0.40 0.55 0.69 0.80 0.51
Combinations of ComponentsProvides highest correlations
7
6
5
4
3
2
1
0
Ma
ss C
on
cen
tra
tion
(m
g m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
VAPS_Terpene SOA3 VAPS_Isoprene SOA AMS_IsopreneOA AMS_IsopreneOA + MO-OOA
6
5
4
3
2
1
0
Mass C
once
ntr
atio
n (
mg m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
VAPS_Terpene SOA2 VAPS_Terpene SOA1 VAPS_BBOA AMS_LO-OOA
BBOA16%
LO-OOA1.9 mg m-3
Isoprene SOA61%
Terpene SOA339%
Isoprene OA1.1 mg m-3
MO-OOA2.0 mg m-3
AMS VAPS
r = 0.92
r = 0.93
Terpene SOA232%
Terpene SOA152%
VAPS Indicates:
40% Isoprene SOA60% Terpene SOA3
52% Terpene SOA132% Terpene SOA216% BBOA
GTech AMS results (Xu et al., PNAS, 2015)
IEPOXIsoprene SOA
18%
BBOA15%
TerpeneSOA117%
TerpeneSOA210%
TerpeneSOA323%
OtherIsoprene SOA
16%
50%
34%
*VAPS analysis based only on 1 week time period, AMS analysis was over entire month
8
6
4
2
0
Mass C
once
ntr
atio
n (
mg m
-3)
12:00 AM6/21/13
12:00 AM6/22/13
12:00 AM6/23/13
12:00 AM6/24/13
12:00 AM6/25/13
12:00 AM6/26/13
0.4
0.3
0.2
0.1
0.0
AM
S N
O3
+ Mass C
once
ntra
tion (m
g m
-3)
VAPS_Terpene SOA3 VAPS_Terpene SOA2 VAPS_Terpene SOA1 AMS_Particle NO3
+
AMS_LO-OOA
• Xu et al. (2015) suggests LO-OOA is largely from terpene BVOC+NO3 and locally produced.• VAPS Terpene SOA3 is in excess of LO-OOA and has highest O:C ratio. It may represent a
more aged – regional component.• Or altered chemistry……need additional lab studies to test
• Currently developing Isoprene/Terpene + NO3 lab studies + analysis of standards
• Many of the compounds observed by VAPS do not exist in current “underivatized” mass spectral libraries!
OUTLINE
• Instrument Development
• Novel Data Analysis Methods
• Field observations: SOAS – Centreville, AL
• Field observations: SLAQRS – East St. Louis, IL
• Laboratory-based oxidation studies
• Conclusions
image: Google Earth
Ozarks Plateau
St. LouisMO
St. Louis Air Quality Regional Study
(SLAQRS)
Aug.3 – Oct.15, 2013
East St. Louis, IL
Millet et al. (U.Minnesota)
Turner et al. (WashU)
Weber et al. (GTech)
0.300.250.200.150.100.050.00
Rela
tive Inte
nsity
1009080706050403020
m/z
0.120.080.040.00
0.100.080.060.040.020.00
0.120.080.040.00
0.120.080.040.00
Isoprene SOA37%
LO-OOA27%
MO-OOA10%
HOA111%
HOA216%HOA
OOA79%
1) PMF results from SLAQRS, East St. Louis: (AMS Data)
TAG chromatogram
Temperature of CTD and GC oven
Williams et al., AMTD 2015
ThermalDecompositionWindow
CompoundWindow
Made an interesting discovery during SLAQRS(TAG and VAPS decomposition signal)
The thermal decomposition signal seems to contain information on the fraction of the aerosol that traditional operation of TAG and VAPS are not capable of detecting
r = 0.56
r = 0.81
r = 0.67
r = 0.65
A large fraction of the IEPOX signal is in the Decomposition Window
m/z 53m/z 82
Tracking major ions in the decomposition window reveals some good correlations to major AMS components:
Nit
rate
Sulf
ate
Org
anic
sIs
op
ren
e IE
PO
X
Williams et al., AMTD 2015
2) PMF results from SLAQRS, East St. Louis: (TAG decomposition signal)
0
30
60
90
120
6.12 7.58 9.04 10.50 11.96 13.43 14.89
Rel
ativ
e si
gnal
(TI
C)
x 1
00
00
Retention time (min)
Nitrate+OOA_carboxylic acidsNitrate+benzene+tolueneIsoprene OOAIsoprene OOA_sulfatesulfate-1Sulfate-2Sulfate-3Carbon-rich sulfateChloroform
Indicators that this factor represents fresh isoprene SOA, not chemically processed
Isoprene SOA enhancement from presence of acidic seed aerosol
Isoprene SOAIsoprene
SOA+sulfate
Zhang et al., In-Prep
Ozarks
4) PMF results from SLAQRS, East St. Louis: (TAG individual compounds)
30x10-3
20100
Fraction
of sig
nal
octa
no
ic a
cid
be
nzo
ic a
cid
h
exylb
enzen
e
nona
no
ic a
cid
tr
ide
cen
e
na
ph
thale
ne, 1
-meth
yl-
trid
ecan
e
na
ph
thale
ne, 2
-meth
yl-
ph
thalic a
cid
decan
oic
acid
hep
tylb
en
zen
e
1-t
etr
adecen
e
3-m
eth
ylp
hth
alic a
cid
te
tra
de
ca
ne
4-m
eth
ylp
hth
alic a
cid
acen
aph
thyle
ne
acena
ph
the
ne
1-p
enta
decen
e
pen
tade
can
e
do
deca
no
ic a
cid
n
onylb
en
zen
e
1-h
exa
de
cen
e
flu
oren
e
hexa
de
can
e
decylb
enzen
e
1-h
epta
decen
e
hep
tade
can
e
pri
sta
ne
un
decylb
en
zen
e
octa
decen
e
phen
an
thren
e
octa
decan
e
anth
ra
ce
ne
ph
yta
ne
2,3
-na
ph
tho
barre
len
e
do
decylb
en
zen
e
nona
de
cen
e
nona
de
can
e
ph
en
an
thre
ne
, 1
-meth
yl
ph
en
an
thre
ne
, 2
-meth
yl
an
thra
cen
e, 1
-m
eth
yl
an
thra
cen
e, 2
-m
eth
yl
he
xa
de
ca
noic
acid
na
ph
thale
ne, 2
-ph
enyl-
eic
ose
ne
trid
ecylb
en
zen
e
eic
osa
ne
flu
oran
then
e
hen
eic
osen
e
tetr
ad
ecylb
enzen
e
ace
phe
na
nth
ryle
ne
hen
eic
osan
e
pyren
e
ph
en
an
threne
, 2
,3,5
-tr
imeth
yl-
do
co
sen
e
flu
ora
nth
en
e,
1-m
eth
yl-
pe
nta
decylb
en
zen
e
do
co
san
e
flu
ora
nth
en
e,
2-m
eth
yl-
flu
ora
nth
en
e,
4-m
eth
yl-
rete
ne
ph
ena
nth
rene
, 1-e
thyl-2
-meth
yl
pyren
e,
1-m
eth
yl-
pyren
e,
2-m
eth
yl-
pyren
e,
4-m
eth
yl-
tric
osen
e
tric
osan
e
he
xad
ecylb
en
ze
ne
tetr
aco
se
ne
tetr
aco
sa
ne
cyclo
pen
ta(cd)p
yre
ne
he
pta
decylb
en
zen
e
ch
rysen
e
pen
taco
sen
e
pen
taco
san
e
octa
de
cylb
en
ze
ne
hexaco
sen
e
hexaco
san
e
no
nad
ecylb
en
ze
ne
hep
taco
sen
e
hep
taco
san
e
eic
osylb
en
zen
e
octa
co
sen
e
octa
co
san
e
he
ne
ico
sylb
en
ze
ne
nona
co
sen
e
nona
co
san
e
tria
co
nte
ne
tria
co
nta
ne
hen
tria
co
nte
ne
hen
tria
co
nta
ne
dotr
iaco
nte
ne
dotr
iaco
nta
ne
tritria
co
nte
ne
ind
eno
(1,2
,3-cd
)pyre
ne
tritria
co
nta
ne
tetr
atr
iaco
nte
ne
te
tra
tria
co
nta
ne
pe
nta
tria
co
nte
ne
pe
nta
tria
co
nta
ne
m
eth
ylfura
n
ben
zen
e
tolu
en
e
benzald
eh
yd
e
ph
eno
l hyd
roqu
ino
ne
Un
kno
wn
2
Un
kno
wn
4
levo
glu
co
se
non
e
iso
pren
e
trim
eth
ylfuran
on
e
no
na
nal
Eth
ylid
ene
he
pte
nolid
e
naph
thale
ne
M
eth
yle
thylp
he
nol
Be
nzo
thia
zole
C
um
inic
-ald
eh
yd
epin
on
ald
eh
yd
e
Dim
eth
ylp
rop
ylm
eth
ylb
enze
na
min
e
Hep
tam
eth
yln
on
en
e
Bip
he
nyl
Pe
lletie
rin
e
Dio
xa
sp
iron
on
an
edio
ne
Co
um
arin
bis
dim
eth
yle
thylc
yclo
he
xa
die
ne
dio
ne
silo
xa
ne
1
phth
alim
ide
nic
oty
rin
e
ditertb
uty
lphe
no
l dib
en
zofu
ra
n
eth
oxyna
ph
thale
ne
alp
ha
cala
co
ren
e
die
thyltolu
am
ide
d
ieth
ylp
hth
ala
te
fluo
reno
l silo
xa
ne
2
ben
zop
he
no
ne
dih
yd
roo
cty
lfura
non
e
tetr
ahyd
ron
on
ylp
yran
on
e
pen
tade
ca
no
ne
eth
ylh
exylb
en
zoa
te
he
pta
me
thyln
on
ene
_2
flu
oren
on
e
dib
enzo
thio
phe
ne
form
yln
orn
icotin
e
silo
xa
ne
3
hydra
zon
eflu
ore
no
ne
trim
eth
ylp
en
tad
eca
non
e
terp
he
nyl
he
xad
eca
ne
itrile
he
xad
eca
no
ica
cid
-m
eth
yle
ste
r
an
thra
quin
on
e
hexa
de
can
oic
acid
_e
thyle
ste
r no
rab
ieta
tetr
ae
ne
me
thylid
en
e_
mix
ture
2
sim
on
ellite
dip
he
nylc
yclo
he
xan
e
he
pta
de
can
enitrile
be
nzyln
aph
tha
len
e
p_
terp
he
nyl
be
nzo
(a)p
he
na
zin
e
hexa
de
can
oic
acid
_b
uty
leste
r b
en
zylb
iphe
nyl
hexa
hyd
rod
ime
thylp
he
nan
thre
ne
ca
rboxylic
me
thyla
bie
tate
tra
eno
ate
to
dte
nh
au
ptd
op
pe
lto
pfte
er_co
mp
oun
d
de
hyd
roab
ietica
cid
o_
terp
he
nyl
m_
terph
enyl
trip
he
nylh
exan
e2
mon
op
alm
itin
b
is(2
-eth
ylh
exyl)
ph
thala
te
mo
no
ste
ari
n
qua
terp
henyl
ole
an-1
2-e
n-3
.beta
-ol is
om
er1
2
8-nor-1
7.b
eta
.(H
)-h
opa
ne
ole
an-1
2-e
n-3
.beta
-ol is
om
er2
(17.a
lpha
.H,2
1.b
eta
.H)-
hop
an
e
50403020100
x1
0-3
403020100
x10
-3
403020100
x10
-3
30
2010
0
x10
-3
403020100
x10
-3
2520151050
x10
-3
403020100
x1
0-3
30
2010
0
x10
-3
50403020100
x10
-3
403020100
x10
-3 Diesel Trains (hopanes, hydrocarbons, SE point)
Biomass Burning (levoglucosenone)
Isoprene SOA (methylfuran, trimethylfuranone)
BBQ Wood-Charcoal (cuminic aldehyde)
BBQ Meat (heptadecanenitrile, monopalmitin)
Anthropogenic SOA (benzoic acid, phthalic acid)
Terpene SOA (pinonaldehyde, cuminic aldehyde, a-calacorene) + Phthalates
Diesel Vehicle (hopanes)
Volatile Fuel Emissions (C23-26 alkanes, alkenes, monopalmitin)
Traffic (PAH,b-PAH, acephenanthrylene)
Mixed Semivolatile (anthro: benzothiazole, toluene, benzene; bio: pinonaldehyde)
15
10
5
0
Mas
s
2015105
Hour of Day
108
6420
16
12
8
4
0
20
15
10
5
0
20
15
10
5
0
16
12
8
4
0
20
15
10
5
0
15
10
5
0
12
8
4
0
1086420
302520151050
Hour of Day
mid
nig
ht
mid
nig
ht
No
on
Same as Isoprene OOA from Decomp..
Zuo et al., In-Prep
Currently working to combine SLAQRS PMF results from:
1) AMS2) TAG (decomposition signal)3) TAG (binned compounds + UCM)4) TAG (integrated compounds)5) VAPS
Created a bit of a data monster, but learning a lot as we now make comparisons!
Good correlations amongst the various Isoprene SOA factors and major Anthropogenic HOA/POA factors.
Study Average (black dots)
Winds from Ozarks:Active NO3 chemistry to remove Isoprene
Winds from Ozarks:No active NO3 chemistry= results in elevated morning O3
Millet et al., ES&T (In-Review)
Observations of Nighttime Chemistry during SLAQRS
Diurnal Average NO3+Isoprene reaction rateJune 2013 climatology, surface to 20.89 km integral
McKeen, Brown (NOAA)
OUTLINE
• Instrument Development
• Novel Data Analysis Methods
• Field observations: SOAS – Centreville, AL
• Field observations: SLAQRS – East St. Louis, IL
• Laboratory-based oxidation studies
• Conclusions
Design for Laboratory-based Oxidation Studies
UV Lamps
PAM ReactorVacuum &Exhaust
O2
UV lamp(O3 generator)
Measurements(Online and Offline)
N2
HumidifierFilter
VOCs, SO2, NOx
PAM Bypass (Optional)
Seed Particles
Denuder
Emissions & Combustion Chamber
Controlled Vent Purified Dry Air Input
Particle/Gas Selection: Cyclone (PM1,PM2.5), DMA (monodispersed), or Teflon Filter (gases only)
Ambient Air,Dilution Air,orNebulized Aerosol
2-way Ball Valve
3-way Ball Valve
POA, SVOCs, IVOCs, VOCs (various sources)
VOC I/SVOC LVOC Real Emissions
Potential Aerosol Mass (PAM) Reactor
O3, NO, NO2, SO2, CO, CO2
Volatility+Hygroscopicity-TDMA, SMPSProton Transfer Reaction Mass Spectrometer (PTR-MS)
Volatility & Polarity Separator – Aerosol Mass Spectrometer (VAPS-AMS)
Thermal desorption Aerosol Gas Chromatograph (TAG)
Building and Adding to Chemical Databases
Diesel
Oak Wood
Hamburger
…..
Thermal desorption Aerosol Gas chromatograph (TAG)
Volatility and Polarity Separator (VAPS)
Aerosol Mass Spectrometer (AMS)
…..
Oak Leaves
Oak Wood
Pine Needles
Toluene SOA
8 6 4 2 0
Volatility
350
300
250
200
150
100
50
0
Polarity
IsopreneSOA
8 6 4 2 0
Volatility
300
250
200
150
100
50
0
Polarity
TolueneSOA
…..
Diesel
For purpose of this project we are characterizing oxidation of Isoprene + several Terpenes(over a range of: oxidation states, humidity, oxidants, acidity, NOx)
Connecting chemistry to hygroscopicity
Currently attempting to map hygroscopicity onto 2D VAPS maps of volatility- and polarity-separated OA functional groups.
Multi-Channel TDMA hygroscopicity, volatility, etc.(developed through EPA Early Career award)
8 6 4 2 0
Volatility
300
250
200
150
100
50
0
Polarity
VolatilityPo
lari
ty
Conclusions• Our novel research tools are offering new insights on the chemical composition of
atmospheric OA
• A combination of new binning analysis techniques, incorporation of thermal decomposition data, and controlled lab studies are improving our understanding of complex ambient chemical signatures observed by TAG, VAPS, and AMS
• VAPS data from SOAS (Alabama) suggests major contributions to OA loadings from Terpene SOA and additional contributions from Isoprene SOA not previously reported.
• OA observed during SLAQRS (St. Louis Region) was composed of a mix of anthropogenic sources with biogenic events delivered from the Ozarks. Nighttime chemistry plays an important role during these periods of biogenic influence. Larger-scale studies should be performed in this region of interest.
R835402