Full text of DOI:10.1080/02786820117809

This article was downloaded by: [University Of Maryland]
On: 16 October 2014, At: 17:37
Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer
House, 37-41 Mortimer Street, London W1T 3JH, UK
Aerosol Science and Technology
Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/uast20
Design and Testing of a Personal Porous-Metal
Denuder
Chuen-Jinn Tsai , Cheng-Hsiung Huang , Si-Ho Wang & Tung-Sheng Shih
Published online: 30 Nov 2010.
To cite this article: Chuen-Jinn Tsai , Cheng-Hsiung Huang , Si-Ho Wang & Tung-Sheng Shih (2001) Design and Testing
of a Personal Porous-Metal Denuder, Aerosol Science and Technology, 35:1, 611-616
To link to this article: http://dx.doi.org/10.1080/02786820117809
PLEASE SCROLL DOWN FOR ARTICLE
Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”)
contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors
make no representations or warranties whatsoever as to the accuracy, completeness, or suitability
for any purpose of the Content. Any opinions and views expressed in this publication are the opinions
and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of
the Content should not be relied upon and should be independently verified with primary sources of
information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands,
costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or
indirectly in connection with, in relation to or arising out of the use of the Content.
This article may be used for research, teaching, and private study purposes. Any substantial or
systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution
in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at
http://www.tandfonline.com/page/terms-and-conditions

Aerosol Science and Technology 35: 611–616 (2001)
c 2001 American Association for Aerosol Research
Published by Taylor and Francis
278-6826=01=$12.00 C .00
°
0
Design and Testing of a Personal Porous-Metal Denuder
1
1
1
2
Chuen-Jinn Tsai, Cheng-Hsiung Huang, Si-Ho Wang, and Tung-Sheng Shih
1
Institute of Environment Engineering, National Chiao Tung University, Hsin Chu, Taiwan
Institute of Occupational Safety and Health, Council of Labor Affairs, Taipei, Taiwan
2
coiled denuder with performance comparable to that of an an-
A personal denuder made of porous-metal disc has been de- nular denuder. Koutrakis et al. (1993) and Sioutas et al. (1996)
signed and tested. The entire casingandsubstrate support are made
of Te on, and the sampling  ow rate is 2 L/min. The sampler con-
sists of a two-stage cascade impactor (having cut-off aerodynamic
diameters of 9.5 and 2.0 m, respectively) to collect liquid par-
ticles and two porous-metal discs (diameter: 2.54 cm; pore size:
developed a glass honeycomb denuder/Ž lter pack system to col-
lect atmospheric gases and particles. The system is consider-
ably smaller than the annular denuder system and can be eas-
ily used for large Ž eld studies. Wai et al. (1994) developed a
1
00 m; thickness: 0.317 cm) to collect basic and acidic gases, re- high efŽ ciency compact diffusion denuder using porous-metal
spectively. The denuder was tested for gas collection efŽ ciency and
capacity at a gas concentration of two times the permissible expo-
sure limit (PEL, promulgated by Taiwan Institute of Occupational
discs. The small size of the denuder makes it possible to de-
sign a compact atmospheric and/or indoor denuder sampling
system.
Safety and Health (IOSH)), with relative humidity (RH) of 80
§
±
Wai et al. (1994) reported that when using 1% (w/v, g/mL)
5
% and temperature of 30
§
3 C. The test data indicate that the
gas collection efŽ ciency is high, and the capacity is sufŽ cient for sodium carbonate/1% (w/v)glycerol coating inthe porous-metal
the acidic/basic gas sampling in the workplace. Using 5% (w/v,
g/mL) sodium carbonate/1% (w/v) glycerol coating on the porous-
metal disc, the collection efŽ ciency is 91.2
standard deviation), 95.08
0.06% and 100
pacity is 4.47, 7.2, and 2.5 mg for HNO , HCl and HF, respectively.
The collection efŽ ciency for NH
disc, the collection efŽ ciencies of SO2 and HNO3 were higher
than 99% and 93% at 10 L/min, respectively. If a 2% sodium
carbonate/1% glycerol coating was used, the gas collection ca-
pacity of the porous-metal disc was as high as 8.4 mg for SO2.
§
§
0.26% (average
0.04%, and the ca-
§
§
3
for the porous-metal disc with The collection efŽ ciency and capacity have not been determined
§
3
4
3
% (w/v) citric acid coating is 96.39
3.6 mg.
0.13%, and the capacity is
–
for other gases. The particle loss in the size range 0.1 2.5 ¹m
for the porous-metal disc was below 3%.
In this study, a personal porous-metal denuder intended for
workplace acidic aerosol sampling was designed and tested. The
INTRODUCTION

ow rate is 2 L/min. This denuder has a two-stage cascade im-
Diffusion denuder is a sampler for removing gases from an pactor with the cut-off aerodynamic diameter of 9.5 and 2.0 ¹m.
aerosol stream to measure their concentrations separately. Gas Two porous-metal discs (diameter: 2.54 cm; pore size: 100 ¹m;
or vapor molecules diffuse rapidly to the wall of a diffusion thickness: 0.317 cm; P/N 1000, Mott Inc., Farmington, CT) were
sampler and are adsorbed onto the wall coated with a suitable placed downstream of the cascade impactor to remove basic and
material. Particles are collected at the downstream Ž lter. The acidic gases, followed by a 37 mm Ž lter holder to collect parti-
gas concentration can be determined by extracting the coated cles smaller than 2.0 ¹m. The cascade impactor was tested for
substrates and analyzing the samples.
Original denuders were made of straight cylindrical tubes discs were tested for gas collection efŽ ciency and capacity of
Ferm 1979). Later, Possanzini et al. (1983) designed an annu-
particle collection efŽ ciency and wall loss. The porous-metal
(
HNO3, HCl, HF, and NH3 at two times the permissible exposure
lar denuder system with much better gas collection efŽ ciency limit (PEL, promulgated by Taiwan IOSH), with a 80 § 5% rel-
±
and absorptive capacity. Pui et al. (1990) designed a compact ative humidity (RH) and a 30 § 3 C temperature. Currently, the
PEL of Taiwan IOSH for HNO3, HCl, HF, and NH3 is 2, 5, 3,
±
and 50 ppm, respectively. The condition at 80% RH and 30 C is
Received 31 January 2000; accepted 25 May 2000.
Address correspondence to Chuen-Jinn Tsai, Institute of Environ-
a stringent condition and is speciŽ ed in the standard procedure
for certifying a reference sampling and analysis method of the
mental Engineering, National Chiao Tung University, No. 75 Poai
Taiwan IOSH
(1996).
Street, Hsin Chu, Taiwan. E-mail: cjtsai@cc.nctu.edu.tw
611

6
12
C.-J. TSAI ET AL.
Figure 1. Schematic diagram of the present personal denuder.
PRESENT DESIGN
34.6 mm, and the total length is 90 mm. The sampling  ow rate is
The schematic diagram of the personal denuder sampler is 2 L/min.
shown in Figure 1 and its detailed drawing is shown in Figure 2.
The cascade impactor is mainly used to remove and clas-
The sampler is made of Te on to minimize interaction between sify large liquid particles that may be collected by the down-
internal surfaces and reactive gases. The outside diameter is stream porous-metal discs and interfere with gas concentration
Figure 2. Detailed drawing of the present personal denuder. (a) Cascade impactor, (b) casing, (c) porous-metal disc, and (d) Ž lter
holder.

DESIGN AND TESTING OF A PERSONAL POROUS-METAL DENUDER
613
measurement. Liquid particles collected by the cascade im- can be calculated as follows:
pactors can be analyzed further by the ion chromatograph to
M2
determine the concentration of particles. Each stage of the cas-
cade impactor has a single round nozzle. The diameter of the
nozzle is 7.2 and 1.9 mm for the Ž rst and second stage, respec-
tively. Since most of acidic aerosol particles are liquid in the
workplace, a removable porous-metal disc (OD: 1.2 cm; other
dimensions are the same as PN/1000, Mott Inc.) is used as the
impaction substrate to make use of its capillary action to prevent
particle overloading problems.
´
(%) D
£ 100%;
2
[1]
[2]
1
M C M
M3
Loss(%) D
£ 100%;
M1 C M2 C M3
where M1; M2, and M3 are the mass concentration of the after-
Ž
lter, impactor plate, and the other portions of the impactor,
respectively. An aerodynamic particle sizer (APS; TSI Model
310A) was used to check the monodisdersity and steadiness of
In the loading test, collection characteristics of liquid parti-
cles on a  at impaction plate and a porous-metal substrate were
compared. Monodisperse oleic acid particles of 10.6 ¹m inaero-
3
the testing aerosols before and after each test.
3
dynamic diameter, 5–10 #/cm in concentration were sampled
for 2 h. The total loaded particle mass was about 0.7 mg. The
results showed that liquid particle permeated into the porous-
metal substrate and no liquid droplets remained on the surface.
Therefore the collection efŽ ciency of subsequent particles is not
expected to be affected. In comparison, particles pile up into
a large liquid droplet on the  at impaction surface, which was
subsequently reentrained.
The Ž rst and second porous-metal discs were coated with
citric acid and sodium carbonate/glycerin to remove ammonia
and acidic gases, respectively. Different coating concentrations
were tested in order to Ž nd a suitable concentration which has
a high gas collection efŽ ciency and capacity for 8 h continu-
ous sampling in the workplace. Fine acidic aerosol particles not
collected by the impactor and denuder discs are collected by
the 37 mm after Ž lter. If volatilization loss of particles from
the after Ž lter is a problem, several other Ž lter materials can
be used after this Ž lter to absorb volatilized gases. However,
volatilization loss is not the objective of this study and was not
investigated.
Particle Loss in the Porous-Metal Disc
Particle loss in the porous-metal disc was determined using
monodisperse polystyrene latex particles (PSL) generated by
an atomizer (Retec X-70/N, Cavetron Corp., Portland, OR) and
dried by a silica gel diffusion drier. The PSL particles were
further neutralized by a Kr-85 neutralizer before being intro-
duced into a mixing chamber. From the chamber, particles were
sampled through the porous-metal disc by a laser aerosol spec-
trometer (PMS Model LAS-X, Particle Measuring Systems Inc.,
Boulder, CO) to determine the inlet and outlet particle con-
centrations (N and N ). Particle loss in the porous-metal disc,
1
2
Lossp(%), was calculated as
N2
N1
Loss (%) D 1 ¡
£ 100%:
[3]
p
Gas Collection EfŽciency and Capacity
of the Porous-Metal Disc
Figure 3 shows the test set up to measure the gas collection
efŽ ciency of the porous-metal disc for HNO3, HCl, or HF gases.
Desirable test gas concentration was generated by aerating clean
EXPERIMENTAL METHOD
Particle Collection EfŽciency and Wall Loss
of the Cascade Impactor
The particle collection efŽ ciency and wall loss were deter-
mined using monodisperse oleic acid test particles as described
in Tsai and Cheng (1995). The particles containing  uorescein
dye tracer were generated by a vibrating oriŽ ce monodisperse
aerosol generator (VOMAG; TSI Model 3450, TSI Inc., St.
Paul, MN). The aerosols were dried and the charge neutral-
ized before being introduced into the testing stage of the cas-
cade impactor. After sampling for 5–30 min, the impactor and
the after-Ž lter were washed with distilled water buffered with
0
.001 N NaOH in known volumes. The washed solution was
measured using a  uorometer (10-AU Fluorometer, Turner
Designs, CA) to determine the dye concentration of different
portions of the impactor and after-Ž lter. Assuming that the dye
concentration is proportional to the mass concentration of col- Figure 3. Test set up for measuring gas collection and capa-
lected particles, the collection efŽ ciency ´(%) and wall loss (%) city.

6
14
C.-J. TSAI ET AL.
air through a bubbler containing a known concentration of acidic
solution. A hot plate was used to heat up the bubbler to facilitate
gas generation. To generate NH3 gas, a standard gas bottle was
used instead of a bubbler. The test gas was then mixed in a
mixing bottle with humid air coming from a humidiŽ er (also
a bubbler) containing deionized water. Heating tape was used
throughout the sampling line to prevent gas condensation on
the wall. By adjusting the concentration of bubbling liquid, hot
plate temperature, and  ow rate of the mixing humid air, it was
possible to obtain the desirable test gas concentration at two
±
times the PEL, 30 § 3 C and 80 § 5% RH.
A quartz Ž lter was used to collect small particles that might
be generated in the test system before introducing the test gas to
the porous-metal disc. Two impingers with proper absorbing so-
lutions were used in series to collect gas that penetrated through
the porous-metal disc. After sampling, the porous-metal disc
was extracted with distilled deionized water in a low-pressure
chamber at 0.2 atm (Wai et al. 1994).
The concentration of the test gas collected on the disc and
impingers (M4 and M5) was determined by an ion chromato-
graph (Model 4500i, Dionex Corp., CA). Two impingers were
used to collect test gas, and M5 is the sum of gas concentra-
tions of the two impingers. The second impinger was used to
check whether the Ž rst one broke through. The gas collection
efŽ ciency of porous denuder, ´g(%), was calculated as
Figure 4. Particle collection efŽ ciency and wall loss of the
cascade impactor.
RESULTS AND DISCUSSION
Results from Particle Experiments
Particle collection efŽ ciency and wall loss of the cascade im-
pactor is shown in Figure 4. The cut-off aerodynamic diameter
for the Ž rst and second stage is 9.5 and 2.0 ¹m, respectively. The
cut-off diameter is smaller than that predicted using Marple’s
theory (1970), which should be 18.4 and 2.43 ¹m for the Ž rst
and second stage, respectively. This is due to partial entrainment
M4
´
g(%) D
£ 100:
[4]
M4 C M5
The gas collection efŽ ciency was measured at the sampling
time of 1, 2, 3, and 4 h. At each sampling time, the test was
repeated three times to improve the precision of the experiment.
Breakthrough time was deŽ ned as the time at which the col-
lection efŽ ciency dropped below 95%. Gas collection capacity,
expressed in mg, was calculated by the sampling air volume at
the breakthrough time multiplied by the test gas concentration.
Two different coating concentrations were used for the test.
For acidic gases, 10 ml, 3 or 5% (w/v ing/mL) sodiumcarbonate,
1
% (w/v)glycerol in1:1 (v/v)methanol/water solution was used.
For ammonia gas, 10 ml, 2 or 4% (w/v) citric acid in ethanol
was used. The coating solution concentration is higher than that
of the denuder used for atmospheric sampling (Wai et al. 1993;
Sioutas et al. 1996) since the latter was found to be insufŽ cient
for the high challenging gas concentration. After coating, the
porous-metal discs were dried by passing nitrogen gas through
them.
Careful QA/QC procedure in this study showed that the me-
thod detection limit was 1.5, 4.0, 1.5, and 15.1 ppb, in terms
of gas phase concentration, for HNO3, HCl, HF and NH3, re-
spectively, based on 8 h sampling at 2 L/min. The corresponding
recovery efŽ ciency from the porous-metal disc for the above
4
3
gases was 94.7 § 0.4, 101.5 § 1.0, 103.6 § 2.2, and 98.1 §
.8%, respectively.
Figure 5. Particle loss in the porous-metal disc.

DESIGN AND TESTING OF A PERSONAL POROUS-METAL DENUDER
615
of the air streamlines into the porous substrate, resulting in ad- Results from Gas Experiments
ditional particle collection. The collection efŽ ciency does not
Figures 6a and b show the gas collection efŽ ciency of the
go to zero when the aerodynamic particle diameter approaches porous-metal disc for acidic and basic gases, respectively. Using
zero. In Figure 4, wall loss for the Ž rst and second stage of the a 3% sodium carbonate coating, the gas collection efŽ ciencies
impactor is shown to < 5.7% and 1.2%, respectively.
3
are all close to100%, as shown in Figure 6a, for HNO , HCl, and
Figure 5 indicates that the maximum particle loss in the HF when the sampling time is <3.0 h. However, the efŽ ciency
porous-metal disc is <9% for particle aerodynamic diameters drops to 85.89 § 0.36% (average § standard deviation) and 90.1
smaller than 2.0 ¹m. Loss can be much higher for larger particle § 0.45% for HNO and HCl, respectively, at the sampling time
3
sizes. Since particles larger than 2.0 ¹m are removed by the cas- of 4.0 h. Increasing the sodium carbonate concentration to 5%,
cade impactor, particles collected in the porous-metal discs are the efŽ ciency at 4.0 h is increased to 91.2 § 0.26% and 95.08
not expected to interfere with gas concentration measurements. § 0.06% for the above two gases, respectively. For HF, the gas
collection efŽ ciency remains high at 97.57 § 0.05% and 100 §
0
5
.4% for the sodium carbonate coating concentration of 3 and
%, respectively.
For 3% sodium carbonate coating concentration, the break-
through time (time at which the collection efŽ ciency drops to
5%) is 3.35, 3.5, and 4.0 h for HNO3, HCl, and HF, respec-
9
tively. The maximum breakthrough time is assumed to be 4.0 h
as in the case of HF. The breakthrough time is increased to 3.58
and 4.0 h for HNO3 and HCl, respectively, when the coating
concentration is increased to 5%. The gas collection capacity of
the porous-metal disc is calculated to be 4.18, 6.3, and 2.5 mg
for HNO , HCl, and HF, respectively, at 3% coating concen-
3
tration. It is increased to 4.47 and 7.2 mg for HNO3 and HCl,
respectively, when the coating concentration is increased to 5%.
Since the maximum breakthrough time is 4.0 h, the capacity for
HF remains the same at 2.5 mg for 5% coating concentration.
Similarly, Figure 6b shows that different citric acid coating
concentrationsalsoresult indifferent NH collectionefŽ ciencies
3
and capacities by the porous-metal disc. Increasing the concen-
tration of citric acid coating solution from 2 to 4% increases the
efŽ ciency at 4.0 h from 92.8 § 0.14 % to 96.39 § 0.13% and
the breakthrough time is increased from 2.9 to 4.0 h. The corre-
sponding collection capacity for 2 and 4% coating concentration
is 24.36 mg and 33.6 mg, respectively.
CONCLUSION
This study extended the work of Wai et al. (1994) and showed
that it is possible use the porous-metal disc as a personal denuder
for sampling high concentrations of acidic and basic gases in
the laboratory. The gas collection efŽ ciency and capacity of the
denuder with suitable coating material and concentration will be
sufŽ ciently high for the 8 h sampling in the workplace, providing
that the gas concentration is below the PEL.
Because loss of particles in the porous-metal disc may lead
to interference with the gas concentration measurement, incor-
porating a particle classiŽ er to remove large particles before the
porous-metal discs is necessary. This study uses a 2 stage cas-
cade impactor for this purpose. To prevent a particle overloading
problem, porous-metal discs were used as impaction substrates
to collect high concentration liquid aerosol particles that may
Figure 6. Gas collection efŽ ciency versus sampling time. (a) exist in the workplace. The experimental data showed that the
acidic gas and (b) ammonia.
cut-off diameters are smaller than that predicted by Marple’s

6
16
C.-J. TSAI ET AL.
theory, presumably due to excess particle collection because of Marple, V. A. (1970). A Fundamental Study of Inertial Impactors, Ph.D. thesis,
Mechanical Engineering Department, University of Minnesota, Minneapolis,
MN.
Possanzini, M., Febo, A., and Liberti, A. (1983). New Design of a High-
Performance Denuder for the Sampling of Atmospheric Pollutants, Atmos.
partial air penetration into the porous-metal substrates. Further
study of the particle collection characteristics and the loading
effect of the impactor is necessary.
Environ. 17:2605–2610.
Pui, D. Y. H., Lewis, C. W., Tsai, C. J., and Liu, B. Y. H. (1990). A Compact
Coiled Denuder for Atmospheric Sampling, Environ. Sci. Technol. 24:307–
ACKNOWLEDGMENT
The authors would like to thank the Taiwan Institute of
Occupational Safety and Health (IOSH), Council of Labor
Affairs, for the Ž nancial support of this project under contract
number IOSH87-A105.
312.
Sioutas, C., Wang, P. Y., Ferguson, S. T., and Koutrakis, P. (1996). Laboratory
and Field Evaluation of an Improved Glass Honeycomb Denuder/Filter Pack
Sampler, Atmos. Environ. 30:885–895.
Taiwan IOSH (Institute of Occupational Safety and Health). (1996). CertiŽca-
tion Procedure of Sampling and Analysis Reference Methods for Hazardous
Pollutants in Work Environments, IOSH technical report no. IOSH85-T-011
(in Chinese).
Tsai, C. J., and Cheng, Y. H. (1995). Solid Particle Collection Characteristics
on Impaction Surfaces of Different Designs, Aerosol Sci. Technol. 23:96–
REFERENCES
Ferm, M. (1979). Method for Determination of Atmospheric Ammonia, Atmos.
Environ. 13:1385–1393.
Koutrakis, P., Sioutas, C., Ferguson, S. T., and Wolfson, J. M. (1993). Devel-
opment and Evaluation of a Glass Honeycomb Denuder/Filter Pack System
to Collect Atmospheric Gases and Particles, Environ. Sci. Technol. 27:2497–
106.
Wai, S. P., Pui, D.Y. H., Lee, C. T., and Liu, B. Y. H. (1994). A Compact Porous-
Metal Denuder for Atmospheric Sampling of Inorganic Aerosols, J. Aerosol
Sci. 25:923–934.
2501.