Instrumentation for Ambient Ultrafine
Particle Measurement
historic perspectives
and
recent developments
---Susanne Hering
Aerosol Dynamics Inc.
Fresno First Street
December 2000 - Feburary 2001
R = 0.97
slope = 0.71 ± 0.003
intercept = 1120 ± 90
3
80x10
Ultrafine Particle Number Concentration (10 - 96 nm)
Ultrafine Particle Number Concentration (10 - 96 nm)
Ultrafine Particles Dominate
Particle Number Concentrations
60
40
20
0
3
14x10
12
Angiola (San Joaquin Valley farming area )
December 2000 - Feburary 2001
R= 0.95
slope = 0.82 ± 0.007
intercept = -470 ± 30
10
8
6
4
2
0
0
3
20
40
60
80x10
Total Particle Number Concentration (10 - 360 nm)
0
2
4
6
8
10
12 14x10
Total Particle Number Concentration (10 - 360 nm)
Rural and Urban Sites : Similar Relationship between
Ultrafine Particle Number Concentrations
and Total Particle Number Concentrations
3
Simplest Indicator
for Ultrafine
Particles:
Bricard et al, 1976
Particle Number
Concentration
Hering et al, 2005
1889: First Measurement of Atmospheric
Particle Number Concentrations
John Aitken,
“On the Number of Dust
Particles in the
Atmosphere”,
Transactions of the
Royal Society of
Edinburgh, 1889
How Aitken’s
Instrument
Worked
Receiving flask
into which
sample is
expanded, with
counting stage
lens
• Humidified air sample
• Expanded adiabatically
• Droplets formed around
particles, which then
settled
filter
• Counted manually
• Repeated with dilution
Sample collection flask
What Aitken Observed:
“The reason of the greater number of particles in the
room than that found outside was due to the particles
produced by the two gas flames burning in the room at
the time.”
• 1912: Wilson Cloud Chamber
– 1929: Nobel Prize for particle physics
– Determined precise expansion ratios for avoiding
homogeneous and ion-induced nucleation of
particles
• 1930s: Scholz: Automated expansion for
particle counting
• 1950s: Vonnegut: Automated counting by
recognizing particles grow to uniform size.
1950s: Vonnegut, Automated Counting
Automated, but not continuous
( condensing vapor: water )
1970s: Thermally Diffusive
Condensation Particle Counter (CPC)
Bricard et al, 1976
Automated & Continuous ( condensing vapor: butanol or other alcohol )
Widely used, suitable as detector for size distribution instruments
Many models, sold by several companies
Can you have a Continuous, Automated Particle
Counter without Butanol?
• Challenge: too small for direct optical detection
• Approach: Create region of supersaturation to
activate particle growth => form droplets
• Why Supersaturation? Equilibrium vapor pressure
over a droplet is greater than over a flat surface due
to free energy associated with surface (surface
tension)
• Kelvin Relation:
Pdroplet = Pflat surface exp( 2 σ v / kTρ
ρR)
Surface tension
Particle radius
Thermally Diffusive CPCs: Operational Principle
• Saturate flow with vapor
• Flow into cold-walled tube
• Vapor condenses on particles
• Requires slowly diffusing vapor (e.g. butanol)
Saturator, 35°C
Thermal Diffusivity,
air = 0.215 cm2/s
Condenser, 10°C
Optics Head
Mass Diffusivity,
butanol = 0.081 cm2/s
Note: diffusivity of water = 0.265 cm2/s > air
Does not work well with water
2003: Water Condensation Particle Counter (WCPC)
• Cold flow enters warm wet-walled tube
• Water vapor diffuses more quickly than flow warms
• Supersaturation, particle activation and growth occurs
inside of a warm, wet walled tube.
2003 WCPC
(~5 nm) :
Optics Head
Saturator, 20C
Condenser, 60C
Saturator, 12C
2004:
Nano-WCPC (~2.5 nm)
Condenser, 75C
Thermal Diffusivity,
air = 0.215 cm2/s
Mass Diffusivity,
water = 0.265 cm2/s
1.9
WCPC Approach
Comparison of
Centerline
Water
Saturation
Ratios
1.8
Centerline Saturation Ratio
1.7
1.6
1.5
Traditional
Approach
1.4
1.3
1.2
1.1
1.0
0
1
2
3
4
Non-Dimensional Axial Position (x/UR2)
WCPC: Inlet flow at 20°C & 100%RH,
Wetted walls at 60°C
---------------------------------------------------Traditional: Inlet flow 60°C & 100%RH
Wetted walls at 20C
Water Condensation Particle Counter
Hering et al, 2005
First laminar-flow WCPC Response to
Ambient Aerosols & Vehicle Emissions
1.0
Detection Efficiency
0.8
0.6
0.4
TSI-3785
Ambient Aerosol (Berkeley, CA)
Vehicle Tunnel Aerosol (Caldecott)
0.2
0.0
2
3
4
5
6
7
8
9
2
3
4
5
6
10
Particle Diameter (nm)
Tunnel Measurements with Antonio Miguel, Arantza Eiguren-Fernandez, UCLA
Calculated Supersaturation Profiles within the
Ultrafine Water – CPC
1.0
2.37 nm, S=2.38
2.17 nm, S=2.59
0.8
Radial Position, r/R
2.10 nm, S=2.69
2.05 nm, S=2.76
0.6
2.00 nm, S=2.83
1.96 nm, S=2.89
0.4
1.92 nm, S=2.96
0.2
0.0
0.0
0.1
0.2
µ v = πα v z/Q
0.3
0.4
αv = vapor mass diffusivity
z = axial distance
Q = volumetric flow rate
0.5
Calibration of the Ultrafine WCPC
Water Residue Particles
100
WCPC-410/WCPC-410x (%)
80
Efficiency for TSI-3786
(water residue particles)
60
Positvely Charged Particles
D50 = 2.4 nm
Negatively Charged Particles
D50 = 2.1 nm
40
TSI-3786
20
0
2
3
4
1
5
6
7
8
9
10
Particle Diameter (nm)
Comparison of Water CPC and TSI Ultrafine,
Queen’s College, NY (Univ. at Albany, EPA Supersite)
Ambient air inlet
Size-Selection
by
1.5 lpm
0.6 lpm
TSI 3025
Electrical Mobility
1 lpm
Filtered
room air
WCPC
Compare:
0.2-sec average
11-sec running average
(TSI 3785)
counts/cm3 measured below nanoSMPS
(not ambient size distribution – no charging correction)
(ultrafine)
Comparison to Butanol Ultrafine CPC with Nano-DMA
Particle counts per 0.2 sec window (average over two scans)
TSI-3025: 1.1±1.0 particles (0.5cc/s)
WCPC:
35 ± 6 particles (16.7 cc/s) --- much better counting statistics
30
10
Run 74
2/3/2004 18:15
WCPC-2sec
WCPC-20sec
TSI3025-2sec
TSI3025-20sec
Particle Number during SMPS Scan (cm-3)
Particle Number during SMPS Scan (cm-3)
12
8
6
4
2
0
4
5
6 7 8 9
2
3
10
4
5
6 7 8 9
100
Run 74
2/3/2004 18:15
WCPC-0.2sec
WCPC-10sec
TSI3025-0.2sec
TSI3025-10sec
25
20
15
10
5
0
3
4
Diameter (nm)
In collaboration with ASRC, University at Albany
5 6 7 8 9
2
3
10
Diameter (nm)
4
5 6 7 89
100
Total Particle Number Concentrations for Traffic Emissions
Ultrafine WCPC (TSI-3786) compared to Butanol UCPC (TSI-3025)
3
Freeway Tunnel
Vehicle Tunnel Particle Concentration (#/cm3)
200x10
Ultrafine Water CPC: TSI-3786
Butanol Ultrafine CPC: TSI-3025
150
100
50
0
1:00 PM
3/1/2005
2:00 PM
3:00 PM
4:00 PM
5:00 PM
6:00 PM
Ambient Sampling In Riverside, California
Ultrafine WCPC – 3786 & Butanol UCPC - 3025
50000
60x103
July 20 - August 14, 2005
SOAR - ICAT Study, Riverside, CA
40000
Water CPC TSI-3786
Butanol CPC T3025
One-Minute Data
One-to-One Line
slope=1.065, R=0.993
50
Ultrafine Water CPC (#/cm3)
Indicated Particle Number Concentration (#/cm3)
Riverside SOAR -ICAT
30000
20000
40
30
20
10000
10
0
9:00 PM
7/25/2005
12:00 AM
7/26/2005
3:00 AM
6:00 AM
9:00 AM
0
0
Pacific Standard Time
3
10
20
30
40
50x10
Butanol Ultrafine CPC (TSI-3025, #/cm3)
2005: Micro-Environmental WCPC
Size:
7” x 7” x 5”
Weight:
5 lb
Power
12 V, 30 Watts
Features:
internal data logging
one week unattended
up to 106 particles/cm3
in single count mode
Response to Near-Monodisperse Aerosols
Response Relative to Butanol UCPC
1.0
0.8
0.6
MICRO-ENVIRONMENT AL W ater-CPC
DMA-Selected Aerosol, 10/15/05
10% mobility window with T SI "long" DMA
Reference: TSI-3025 Butanol Ultraf ine CPC/ Eff
0.4
Freeway-influenced Ambient (Berkeley)
Isopropanol Residue
0.2
TSI-3781 prototype
0.0
4
5
6
7
8
9
2
3
4
10
5
6
7
8
9
100
Particle Diameter (nm)
2
Particle Number Concentratin (#/cm3)
ME-WCPC Measurements in a Residential Kitchen
6
10
Residential Kitchen
Butanol CPCs
T3025
T3022
Water CPCs
T3786
QME2
QME3
T3785_Calc
5
10
4
10
Oven on at 6:05 pm
3
10
6:00 PM
1/7/2006
7:00 PM
8:00 PM
Summary
• 1890’s
Aitken, first explorations of airborne particles number
concentration
Identified combustion as source of particles
• 1920s-1950s
Advances on Aitken’s approach: automated instruments
• 1970’s
First continuous flow condensation particle counters
widely used, especially as detectors for mobility size distributions
• 2003
Introduction of continuous water-based condensation counters
Acknowledgements
• California Air Resources Board, who provided funding
for many of the field comparisons.
• ASRC, University at Albany and the EPA Supersites
Program who made possible the measurements in
New York City.
• TSI Inc., who loaned equipment for our field and
laboratory testing.
• Quant Technologies LLC, who provided the
engineering and detailed instrument design.