Upload
phungdan
View
213
Download
1
Embed Size (px)
Citation preview
10/31/2009
1
Aerosol Chemical Speciation MonitorACSM
Instrument description and sample data
N.L. Ng, T. Onasch, A. Trimborn, S. Herndon, M. Canagaratna, D. Sueper, D. Worsnop,
J. Jayne
Aerodyne Research, Inc.
AAAR Meeting Oct 27, 2009
Size: 19”D x 21”W x 33”H
Aerosol Chemical Speciation MonitorACSM
Weight: 140 lbs (64 kg)
Power: 300W
Data acquisition via Ethernet connection – basic laptop is sufficient.
ACSM-002 SN 140-100
10/31/2009
2
ACSM, aka…
• Ehm-Eye-Ehn-Eye MINI• Mini-AMS• Baby AMS• AMS Light• Cheap AMSp• Ponche• Sammy
Aerosol Chemical Speciation MonitorACSM
• Continuous monitoring of non-refractory aerosol composition by thermal particle vaporization aerosol mass spectrometry.
Sulfate, Nitrate, Chloride, Ammonium, Organics
• Builds on ARI Q and ToF AMS Hardware and analysis concepts.
lower cost, lower sensitivity.
• Designed for monitoring, long term unattended operation.
Performance demonstrations in progress
10/31/2009
3
Research grade
Particle Beam Generation
Aerodynamic Sizing
Particle Composition
Aerosol Mass Spectrometer
Fast DataResearch gradeTOF mass spectrometer
Thermal Vaporization
Beam Chopper
Particle TOF Region
Fast DataSystemTimersDACs
Vaporization &
Electron Impact
Ionization
Aerodynamic Lens
Particle TOF Region
Pumps (x5)
Particle Beam Generation
Particle Composition
Aerosol Chemical Speciation Monitor
No Sizing
Commercial grade
Thermal Vaporization
&
LaptopComputer
RG
A
gMass Spectrometer
Pfeiffer Prisma
Particle Inlet (1 atm)
Electron Impact
Ionization
Aerodynamic Lens40-1000 nm
Pumps (x3)Slow scan rates, 1 sec/amu. Can’t time resolve SI or SP
10/31/2009
4
ACSM Designed Around Pfeiffer Prisma RGA
• Prisma electronics supports:• Prisma electronics supports:– 6mm diameter rods, 200 amu range.– 1 mA/mbar sensitivity to Ar (200 amu head)– Ethernet connectivity with OPC1 interface.– A Windows CE computer/OS.
B ilt i di it l d l I/O– Built-in digital and analog I/O.
1OPC is a standard software interface which enables data communication between applications of different manufacturers. OPC stands for Openness, Productivity, Collaboration (formerly OLE for Process Control).
60
40
amps
( x1
0-9 ) Pump 4 installed
Pump 4 removed
Removal of Pump4
20
0
Sig
nal,
a
504540353025201510
AMU
10-9
10-8 Pump 4 installed Pump 4 removed
• Lower Air signal• Slightly larger water signal
10-12
10-11
10-10
Sign
al
14012010080604020
AMU
Slightly larger water signal• Organic background ok
10/31/2009
5
RG
A
Internal StandardNaphthalene Effusive Source
• m/z marker
I i i• Ion transmission
• Ionization efficiency reference
h h l100
ty EI spectra from NIST for Napthalene
1 µm pin hole
naphthalene molecules
80
60
40
20
0
Rel
ativ
e In
tens
i
14012010080604020
AMU
EI spectra from NIST for Napthalene
1011 s-1
Ion Transmission1.4
1.2
1.0ensi
ty
16 mm (QAMS) 6 (P i )
N2
0.8
0.6
0.4
0.2
0.0
Nor
mai
lzed
Ion
Inte 6 mm (Prisma)
Ne
ArKr
Xe
14012010080604020
m/z
Compares 16mm (QAMS) to 6mm (ACSM)
10/31/2009
6
1.2
1.0
ion
Noble gas mix ratio of 6mm and 16mm Quad
Naphthalene
Ion Transmission Correction
0.8
0.6
0.4
0 2Rel
ativ
e Io
n Tr
ansm
issi Current estimate
0.2
0.0
R
200150100500
m/z
Naphthalene provides an in-situ measure of IT
30x10-12
25mps
)
34
32 Naphthalene Sig ChamberT
Naphthalene Signal Varies with Chamber Temperature
25
20
15
10
5
Nap
htha
lene
Sig
nal (
am
3/9/2009 3/13/2009 3/17/2009
30
28
26
24
22
Cham
ber Temp., C
-10.8
-10.6
g)
Date/Time
-11.4
-11.2
-11.0
Log(
Nap
. Si
3.40x10-33.383.363.343.323.30
1/ChamberT (K)
This is a vapor pressure measurement…
10/31/2009
7
30x10-12
25amps
)
34
32C
Naphthalene Sig Naphthalene T-corrected ChamberT
Temperature Corrected Naphthalene Signal
20
15
10Nap
htha
lene
Sig
nal (
a
30
28
26
24
Cham
ber Temp., C
5
3/9/2009 3/13/2009 3/17/2009
Date/Time
24
22
A measure of SEM gain
Anthracene for AMS systems?Anthracene vapor pressure 1.70E-05naphthalene vapor press 0.078Naph/Anthra vp 4.59E+03
10/31/2009
8
Automated Valve System for Instrument Zero
3 way Valve
GA
Sample mode
Filter
Filter mode
Filter
3-way Valve
RG
Aerosol mass is determined from difference of ‘Sample – Filter”
Automated ValvesFixed pinhole assemblyFilter
To Cyclone
Achim’s (beer) Shelf
10/31/2009
9
ACSM Inlet Schematic
Sample Flow~3 LPM from cyclone
100 um pin-hole
~3 LPM from cyclone
iso-kinetic splitNominal flo 10 000cc/minAMS aerodynamic lens
1/2 OD x 0.41” ID tube
Pump
P
pNominal flow 10,000cc/minReynolds No. ~1300
Aerosol mass is determined from difference of ‘Sample – Filter” mass spectra
1008060
e (
x10-1
2 )
Difference Sample Filt
10-11
10-9
10-7
gnal
(am
ps)
4020
0Diff
eren
ce Filter
Naphthalene
Chemical species are determined following Allan et al, 2003Sulfate, Nitrate, Chloride, Ammonium, Organics
10-13
10
Sig
14012010080604020
m/z
10/31/2009
10
Instrument Stability - Allan Variance
0.2m-3
6:00 PM10/5/2009
12:00 AM10/6/2009
6:00 AMDate/Time
10-4
10-3
10-2
rianc
e (σ
2 )
0.20.0
-0.2µg m
Org
5 min
Va
4 5 6 7 8
1032 3 4 5 6 7 8
1042 3
Integration Time (s)
Measured under (ideal) laboratory conditions
Estimated 3σ Detection Limits
• 1 hr DLs are 100
2
46
1000
-3)
NH4 Org SO4 NO3Chl < 0.2 µg m-3
• 24 hr DLs are < 0.05 µg m-3
12
46
102
46
100
3σ D
L (n
g m
- Chl
Detection limits depend on background signal levels in the spectrum
20151050
Signal Integration Time (hours)
10/31/2009
11
Ratio of xAMS / ACSM Detection LimitsValues are 30 min 3σ
HTOF-w HTOF-v Ctof QAMSHTOF w HTOF v Ctof QAMSOrg 3.9 63.7 73.8 3.0SO4 2.1 45.3 107.1 1.5NO3 3.3 35.9 86.7 3.3NH4 18.0 70.9 168.4 7.7Chl 1.9 8.2 24.6 3.1
DeCarlo et al, 2006 Anal. Chem, 78, 8281-8289
10-9
10-8
s)
MS Difference in Beam Block mode
13
10-12
10-11
10-10
Sign
al (a
mps
Beam Block Mode (filter on)
10-13
222120191817161514131211
AMU
A consequence of a shorter chamber is a more intense molecular beam
10/31/2009
12
10-2
3.6
3.4
3.2 sign
al d
ata
12:00 PM10/6/2009
6:00 PM 12:00 AM10/7/2009
6:00 AM
datBeam Block mode, 5 min 1 scan open, 1 scan closed
10-4
10-3
Alla
n V
aria
nce
(σ2 )
4 5 6 7 8 9
1032 3 4 5 6 7 8 9
1042 3
Integration Time (s)
ORG
40x10-3 a
12:00 PM10/6/2009
6:00 PM 12:00 AM10/7/2009
6:00 AM
dat
10-6
10-5
10-4A
llan
Var
ianc
e (σ
2 )
4 5 6 7 8 9
1032 3 4 5 6 7 8 9
1042 3
Integration Time (s)
40x10200
-20-40 si
gnal
dat
a
SO4
Org does not signal average as well in BB mode. Looks like Filter switching is better. But, other species are not as affected?
Atomizer
Setup for Mass Based IE Determination
By-passDrier
Must consider lens transmission and multiply charged diameters from DMA
ACSMDMA
CPC
AerosolDiluter
filter
Mixing Tube
Input Mass = ρ x Volume(size) x Number Reported Mass
Plot Measured Mass vs Input Mass
10/31/2009
13
700
600
00s x1
0-12 ) 5.6
5.4
s x1
0-9 )
NH4NO3 Ionization Efficiency Calibration
NO3 NH4500
400
300
200
100
0NO
3 io
n si
gnal
s (a
mps
76543210
Slope=8.62e-11
5.2
5.0
4.8
4.6
NH
4 io
n si
gnal
s (a
mps
2.52.01.51.00.50.0
Slope=4.05e-10
4
NO3 mass (ug/m3) NH4 mass (ug/m3)
Calibration based on calculated mass delivered from atomizer/DMA/CPC system
NO3 = NO+ + NO2+ NH4 = NH+ + NH2
+ + NH3+
2.5
2.0
0-10
Correlation of AN-IE with naphthalene Signal
1.5
1.0
0.5
0 0
IE (a
mp/
mas
s)x1
0.0140120100806040200
Naphthalene Sig (Temp corrceted)x10-12
Want to demonstrate that naphthalene signal is an effective measure of IE
10/31/2009
14
• Spain, March 2009 – 3 weeks• Houston May 2009 – 6 weeks
Recent Field Deployments of the ACSM
Houston, May 2009 6 weeks• Queens, NY June 2009 – 8 weeks
NYSDEC
http://www.dec.ny.gov/chemical/54360.html
ACSM Data Sample Queens, NY 2009
10
8
6
4gs (u
g/m
3)
NH4 Org SO4 NO3 Chl
4
2
0
Load
ing
7/16
/200
9
7/21
/200
9
7/26
/200
9
7/31
/200
9
8/5/
2009
8/10
/200
9
8/15
/200
9
8/20
/200
9
8/25
/200
9
8/30
/200
9
9/4/
2009
9/9/
2009
UTC Time• 8 weeks unattended 29.0% 6.7% 8 weeks unattended continuous measurements • Data upload to a FTP site 0.2%
12.4%
51.7% cp5
15.9 µg m-3
average
10/31/2009
15
ACSM vs HTOF-AMS Time TrendsQueens, NY 2009
Org4
25201510
50
HTOF AMS
NO3
NH4
Chl
12
µg m
-3
43210
43210 0.20
0.150.100.050.00
Qi Zhang, Yele Sun SUNY Albany
SO4
12840
7/15/2009 7/19/2009 7/23/2009 7/27/2009 7/31/2009
Local Time
0
cep8
0.20
0.15
Results from PMF AnalysisOOA and HOA Factors
OOA – Oxidized Organic
0.12
0.08Rel
ativ
e Io
n In
tens
ity 0.10
0.05
0.00
Aerosol“Photochemically Aged Aerosol”
HOA – Hydrocarbon-like Organic Aerosol
0.04
0.0014012010080604020
AMU
like Organic Aerosol“Primary Particles”
Queens Data, 2009
10/31/2009
16
OOA and HOA Time seriesACSM and AMS Comparison
12 ACSM OOA ACSM HOA AMSOOA 8
4
010
8
6
µg m
-3
AMS
HOA
OOA
7/16/2009 7/20/2009 7/24/2009 7/28/2009 8/1/2009
Date/Time (EST)
4
2
0
Qi Zhang, Yele Sun SUNY Albany
0.0 0.0
ACSM AMS
PMF Mass fractions
0.6
0.4
0.2HOA = 27%
OOA = 71%
0.6
0.4
0.2HOA = 26%
OOA = 73%
Fract_Mass1.0
0.8 Residual 2%
Fract_Mass1.0
0.8 Residual 1%
10/31/2009
17
Future Work for ACSM
• Vaporization cage to capture and sample “bounced” particles.
• PM2.5 lens.
• Particle sizing module.• Integrated aerosol drier for RH treatment.
• Continued sw development.p
• Continue field demonstrations.
• Magnets
1200
1000Small Covered Ionizer
with magnets no magnets
Magnets on the Ionizer Average Single Particle Signal
800
600
400
200
NO
2+
03002001000-100-200
time (x10-6 sec)
Increases effective IE 2-3x
10/31/2009
18
Flow Control ModuleBlower with active flow
regulation.3-20 LPM 25mBar
Cyc
lone
Iso-kinetic split ACSM
3 20 LPM, 25mBar50W, 24VDC14” x 4” x 6”~5 lbs
MFMFilterBlower
Flow Control Elec.
Portable Met Monitor Portamon• RH 10-90%
• Temp• Barometric Pressure, 30-100 kPa,
resolution of 1.5Pa, ~12 cm air.• Wind Speed, spinning cup ~0.2
m/s/pulse• Wind Direction, 10 bit=360degree• Solar • RTC• GPS• Data logging to USB thumb drive or
RS232• Data rates: 1, 2, 5, 10, 15, 30, 60 secs, , , , , ,