Evidence for different gas/particle partitioning of water-soluble organic compounds

in two urban atmospheres with contrasting emissions

X. Zhang, J. Liu, E. T. Parker and R. J. Weber

Georgia Institute of Technology

CalNex Data Analysis Workshop

May 16-19, 2011

Sacramento, CA

Gas and Particle Water-Soluble Organic Carbonmeasurements during CalNex

WSOCg:

Mist Chamber – TOC(H > 103M/atm)

TOC

WSOCp:

PILS – TOC (PM2.5)

One 5-min sample every 10 mins

Determine WSOC partitioning

Fp = WSOCp

WSOCp + WSOCg

(Hennigan et al., GRL, 2008)

Fp - the fraction of total WSOC in the particle phase

Investigate Fp as a function of:

BVOC & AVOC

SVOC Water (haze, cloud)

stableOH, O3 NO3

KH

1) Relative humidity (LWC) 2) Organic carbon mass

CalNex (May-Jun, 2010)

Atlanta (Aug-Sep, 2010)

SVOC Organic Aerosol

BVOC & AVOC

OH, O3 NO3

Two data sets:

CalNex: Fp and Relative Humidity

Nitrate aerosols partition to liquid water, but WSOC don’t..

Unlike NO3-, Fp peak in the

afternoon (lowest RH, highest T during a day)

NO3- WSOC

0.16

0.14

0.12

0.10

0.08

Fp

12:00 AM 6:00 AM 12:00 PM 6:00 PM

5

4

3

2

NO

3- ,

µg

/m3

95

90

85

80

75

70

RH

RH Fp

NO3-

0.20

0.15

0.10

0.05F

p

100806040Relative Humidity, %

J. Liu et al., poster #36

CalNex: Fp and Organic carbon mass

Partitioning Theory (Pankow et al, 1994)

KP CP /MO

Cg

760 R T fOM106 MWOM pi

oOH O3 NO3

SVOC

Organic Aerosol

VOC

p0

Partitioning coefficient

Pankow-type partitioning of WSOC dominates in LA

0.20

0.15

0.10

0.05

0.00

Fp

6543210

Organic Cabon, µgC m-3

Estimated VOC Emissions in LA and Atlanta

4x1012

3

2

1

0

Iso

pre

ne

Em

issi

on

(mo

lecu

leC

cm

-2 s-1

)

LA bio ATL bio

30x1012

20

10

0

AV

OC

Em

ission

(mo

lecu

leC

cm-2

s-1)

LA anthro ATL anthro

4.8 times 1.7 times

Isoprene Anthrop. VOCs

(Müller et al., ACP, 2008)

LA: Anthrop. VOCs dominant

ATL: Biogenic VOCs dominant

SOA formation in Atlanta very different from LA

In Atlanta (and other sites in eastern US) no clear diurnal trend to SOA

WSOCp

5

4

3

2

1

Fin

e p

art

icle

WS

OC

, µ

gC

/m3

8/6/2010 8/11/2010 8/16/2010 8/21/2010 8/26/2010 8/31/2010 9/5/2010 9/10/2010

Eastern Standard Time

AtlantaSummer 2010: WSOC(SOA) with no diurnal trendWSOCp

Rain

Atlanta: Fp dependence on RH & OC0.23

0.22

0.21

0.20

0.19

0.18

0.17F

p

12:00 AM 6:00 AM 12:00 PM 6:00 PM

0.40

0.35

0.30

0.25

0.20

0.15

NO

3- ,

µg/m

3

80

70

60

50R

H

RHFp

NO3-

0.30

0.25

0.20

0.15

0.10

0.05

Fp

76543210

Organic Cabon, µgC m-3

OCRH0.35

0.30

0.25

0.20

0.15

0.10

Fp

10080604020Relative Humidity, %

70%

0.30

0.25

0.20

0.15

0.10

0.05

Fp

76543210

Organic Cabon, µgC m-3

Atlanta

RH < 70 RH > 70

Atlanta: When RH is below 70% (mostly later after noon)

12

8

4

0

10080604020Relative Humidity, %

10

8

6

4

2

0

ATL 8AM-9AM ATL 2PM-3PM

RH frequency distribution

8AM -9AM

2PM –3PM

Segregate the data by RH </ > 70%

0.23

0.22

0.21

0.20

0.19

0.18

0.17

Fp

12:00 AM 6:00 AM 12:00 PM 6:00 PM

0.40

0.35

0.30

0.25

0.20

0.15

NO

3- ,

µg/

m3

80

70

60

50

RH

RHFp

NO3-

70%

During dry period (RH below 70%), OC partitioning likely to dominate;

0.23

0.22

0.21

0.20

0.19

0.18

0.17

Fp

12:00 AM 6:00 AM 12:00 PM 6:00 PM

0.40

0.35

0.30

0.25

0.20

0.15

NO

3- ,

µg

/m3

80

70

60

50

RH

RHFp

NO3-

Atlanta: Examine Fp dependence during daily transition from wet to dry periods

12

8

4

0

10080604020Relative Humidity, %

10

8

6

4

2

0

ATL 8AM-9AM ATL 2PM-3PM

8AM -9AM

2PM –3PM

“wetter period” [5AM, 11AM]

“drier period” [12PM, 6PM]

0.30

0.25

0.20

0.15

0.10

0.05F

p

76543210

Organic Cabon, µgC m-3

0.30

0.25

0.20

0.15

0.10

0.05

Fp

76543210

Organic Cabon, µgC m-3

0.35

0.30

0.25

0.20

0.15

0.10

Fp

10080604020Relative Humidity, %

0.35

0.30

0.25

0.20

0.15

0.10

Fp

10080604020Relative Humidity, %

RH OC

RH OC

Evidence from other species and studiesCalNex vs. Atlanta formic acid partitioning (J. Liu et al., poster#36)

0.35

0.30

0.25

0.20

0.15

0.10

Fp

12108642Organic carbon

9am to 1pm 12pm to 6pm

0.40

0.35

0.30

0.25

0.20

0.15

Fp

10080604020

Relative Humidity

9am to 1pm 12pm to 6pm

5-month (May – Sep, 2007) in Atlanta by Hennigan et al.

RH OC

Summary In LA, WSOC partitioning consistent with absorption to organic mass

(Pankow’s theory) with no clear dependence on particle water (RH); August In Atlanta, WSOC partitioning has a transition from evidence

for partitioning to particle water at night into late morning to some evidence for partitioning to organic mass in afternoon when particle is dry; Also, Formic acid partitioning in Atlanta related to LWC, not so much in LA (J. Liu et al., poster#36) Overall, during a longer time period (May-Sept), WSOC partitioning

most obvious to LWC (Hennigan et al., 2009) Why the difference?

Different VOC mixture in the two urban atmospheres; Different dispersion/dilution patterns.

Implication: is it appropriate to apply one mechanism to predict SOA in all types of environments?

Characteristics of WSOCp in the SoutheastUrban/Rural Diurnal Tends-AMIGAS

Average over more than 1 month

Time of day

~ 70 km apart

Atlanta

Yorkville (NW of Atl)

Atlanta signal riding on large regional signal some evidence for a daytime increase (urban photochemical SOA?), 20%

WSOCp

Small, but measurable, primary WSOCp at urban site

Summer 2008

Zhang et al. 2011

urban

rural

13