United States
Environmental Protection
Agency
Environmental Monitoring Systems
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S4-83-010 June 1983
Project  Summary
Validation  of Samplers for
Inhaled  Paniculate  Matter

Walter John, Stephen M. Wall, and Jerome J. Wesolowski
    Methods for testing inhalable par-
ticle samplers have been developed
and applied to the dichotomous sam-
pler and the size-selective hi-vol sam-
pler. The sampling effectiveness of the
inlet to the dichotomous sampler was
measured and found to be excessively
dependent on wind speed. A modifica-
tion to improve the performance was
designed as a retrofit to the existing
inlet   Measured wall losses  in the
dichotomous sampler were small. The
fine fraction is found to be correlated
to error in nozzle concentricity, which
is not within specifications  for the
typical commercial sampler.
  Penetration of the size-selective hi-
vol by solid particles largerthan the 15-
/xm cutoff was measured by three
methods.  Near the cutoff, monodis-
perse spray-dried particles that were
sized  aerodynamically by a new laser
sedimentation velocimeter were used.
In the 35 to 60 jam range, glass beads
and A/C test dust particles were pro-
duced by a new sonic fluidized micro-
bed generator.  Finally, ambient par-
ticles that were collected on a Nucle-
pore after filter were analyzed by scan-
ning electron microscopy.  All three
tests show that excess penetration of
the size-selective hi-vol by solid par-
ticles was acceptably small. Additional
testing was conducted in ambient air
using an array of side-by-side samplers
including two dichotomous samplers,
two monocut samplers, and a cyclone
sampler.

  This Project Summary was developed
by the Environmental Monitoring Sys-
tems  Laboratory, Research Triangle
Park North Carolina, to announce key
findings of the research project that is
fully documented in a separate report
of the same title (see Project Report
ordering information at back).

Introduction
  The Environmental Protection Agency
(EPA) is considering revisions to the
National  Ambient Air Quality Standards
(NAAQS) for paniculate matter.  Once
promulgated, the revised standards would
regulate only those particles that can be
deposited in the human respiratory sys-
tem.   The  rationale  for a  standard for
inhalable particles, those particles smaller
than  15 jam  aerodynamic diameter,*
which are  deposited in the respiratory
system below the trachea, has been pre-
sented by Miller, et al.
  A network of inhalable particle sam-
plers has been deployed across the United
States by the Environmental Monitoring
Systems Laboratory at Research Triangle
Park, North Carolina (EMSL/RTP) to field
test the  sampling methodology and to
acquire a data base in support of the new
standard. Early results revealed that the
samplers have some technical problems.
These experiences emphasize the need
for rigorous testing of the new samplers
because they will generate data that will be
the basis for control decisions.
  During the present project, the dicho-
tomous and size-selective hi-vol samplers
used in the EPA network were tested in
the laboratory.  Additional testing was
carried out on side-by-side samplers in
ambient air. The new testing methods that
were developed include measurement of
nozzle alignment in the dichotomous sam-
pler,  precision wall  loss measurements,
and tests  for  particle  bounce.   These
'After the present work was completed, the EPA staff
recommended a cutpoint of 10 /u.m for the revised
paniculate matter standard. Modified versions of the
1 5 Mm samplers are now available. Some, but not all,
of the test results presented here are outdated, but the
test methodology is still current.

-------
methods should be useful for the formu-
lation of acceptance criteria for new can-
didate samplers and for quality assurance
testing of commercial samplers.

Results
  Laboratory testing of the dichotomous
samplers  included measurement of the
critical dimensions of the  inlets and the
virtual impaction stage, determination of
the effect of wind speed on the effective-
ness  of inlet sampling,  and  precision
measurement of the  wall losses in the
samplers. Testing of inlets to commercial
dichotomous samplers  reveals  that the
critical dimensions adhere closely to man-
ufacturing tolerances.   However,  wind
tunnel and static chamber tests reveal that
the inlet sampling effectiveness  is exces-
sively dependent on wind speed.  Flow
visualization studies show strong chan-
neling of the flow inside the  inlet   A
theoretical model  including  inertial and
gravitational forces  accounts for the test
data and  provides  a  basis for  an  inlet
modification  consisting of an  enlarged
entrance  slit and an inertial impactor.
Tests of two variations of this modification
show that this is a promising approach
that permits a low-cost retrofit to existing
inlets. The design also facilitates a change
in particle size cutpoint
  The cutoff (coarse fraction vs. particle
diameter) for the virtual impaction stage of
the dichotomous sampler  was measured
with  DOP-uranine aerosol.  The results
agreed with independent  measurements
by Loo and  by McFarland.  A casting
method was developed to measure the
critical  dimensions of the dichotomous
sampler stage. The nozzle concentricity
was out of tolerance in three out of five
samplers.  Design  faults  were  revealed
and a simple method was suggested for
improvement of the concentricity.
   Laboratory tests with  monodisperse
aerosol show that the 50%  particle size
cutpoint of the impactor stage of  the
dichotomous samplers averages 2.4 ju.m
and the wall loss at the cutpoint averages
2.3%, both acceptable figures. Wall losses
peaked at the cutpoint and increased again
above 10 jum. Most of the losses were on
the collection nozzle  (Figures 1 and  2).
While the wall losses were small for 3-jum
particles,  the fine fraction  correlated with
error in nozzle concentricity (Figure 3);
therefore, close adherence to the toler-
ance on the nozzle concentricity is recom-
mended.  The magnitude of the wall loss
agreed with the measurements by Loo but
did not agree with the results of McFarland
The losses at 2.5 ju.m are about twice as
large as those  reported by  Loo for the
prototype.  This increase is probably caused
by larger  tolerances in  the  commercial
samplers for nozzle concentricity and col-
lection nozzle profile.  However, the wall
j.7 a
                             7
                           Beckman
        Sierra
Figure  1.    Schematic drawing of dichot-
            omous sampler impactor parts
            that were washed separately to
            recover wall losses.
   1.0

   0.1
  0.01
       1                 10

           Aerodynamic Diameter, urn
Figure 2.    Wall losses measured for indi-
            vidual parts of the dichotomous
            samplers (parts are labelled as
            in Figure 1).
            WT = Total wall loss
                + We + We + Wo>.
losses are acceptable in the commercial
units. The high precision of the measure-
ments obtained with standard laboratory
equipment indicates that the present meth-
odology provides a basis for a standard
approach to sampler wall loss measure-
ments.
  Laboratory testing of the size-selective
hi-vol  sampler inlet included measure-
ment of sampling effectiveness for liquid
particles and determination of excess pen-
etration by large solid particles. Measure-
ments made with DOP-uranine to test the
sampling effectiveness of the size-selective
hi-vol inlet under static conditions agreed
with  previous  measurements made  by
Wedding and by McFarland and Ortiz in a
wind tunnel at 2 km/h. These data and the
previously described results for the dicho-
tomous sampler show that independent
laboratories  can  use  liquid aerosols to
obtain consistent results for cutoff and
effectiveness measurements.
  Three methods have been developed for
measurement of the penetration of a size-
selective sampler by large solid particles.
(1)  Near the cutoff, monodisperse spray-
dried particles were used from a vibrating
orifice aerosol generator and sized aero-
dynamically  with a laser  sedimentation
velocimeter.    (2) Particles much larger
than the cutoff were produced by a new
sonic fluidized microbed generator using
presized samples of glass beads and A/C
test (Arizona  Road) dust  (3)  Penetration
by large ambient particles was measured
by microscopy after they were collected on
a Nuclepore after filter.
  The three methods were applied to the
size-selective hi-vol sampler.  Method 1
indicated a  25% excess penetration by
solid vs. liquid particles  at  the  15-fim
cutpoint (Figure 4).  However, this pene-
tration would not ordinarily lead to a serious
error in sampled mass because the cut-
point for solid particles (1 7 jum) is only
slightly larger than 15 /im and the test
particles are bouncier than  most ambient
particles.  Method 2  showed 3 to 4%
penetration (by mass) of particles that had
aerodynamic diameters in the range of 35
to  60 jum (Figure 5).  Seven  percent
penetration was observed for 62 /im A/C
test dust particles, but this may be caused
by  breakup  of the  particles,  an  inherent
problem with A/C test dust Glass beads
were excellent test particles. Examination
of the impaction plate showed  that the
particles did not adhere to the plate in the
impaction zone but were collected near
the bases of  the collection tubes. There-
fore, the particles bounced, but were not
reentrained  in the exit flow.   Method 3

-------
            20
                                     3.0 fan DOP—Uranine
       .§'
       .8
             10
             0

            to




           0.75




            0.5




           0.25
                                    I
                                                          A
                                                    R= 0.97
                               A
            R = 0.94
                ^  s
                      Tolerance
                                                        I
                                  50                  100                 150

                                Error in Nozzle Concentricity, ftm
Figure 3.    A:  Fine fraction of the dichotomous sampler vs. error in nozzle concentricity.
            B:  Wall loss vs. error in nozzle concentricity.
gave a penetration efficiency of 5% for
ambient particles that had aerodynamic
diameters greater than 30 /xm;  these
results agree with those of method 2.
  Ambient air  testing on  side-by-side
samplers shows that the standard devia-
tion of the coarse fraction of the  dicho-
tomous sampler is more than twice that of
the fine fraction and nearly twice that of
the monocut samplers. The fine fractions
correlated closely with the mass concen-
tration from the Air and Industrial Hygiene
Laboratory (Al H L) cyclone, and the su m of
the coarse and fine fractions correlated
closely with the mass concentration from
two monocut inhalable particle samplers.
One dichotomous sampler had a consis-
tent bias of about -10% relative to the
other samplers; the cause of this bias was
undetermined.
Conclusions and
Recommendations
  The experience from the validation pro-
gram for inhalable particle samplers em-
phasizes the importance of rigorous test-
ing.  Precise tests with laboratory aerosol
can lead to correction of problems before
costly tests are conducted in ambient air.
Sampler inlets need to be wind tunnel
tested to verify that wind speed sensitivity
is  within acceptable limits.   Low  wind
speeds (not more than 2 km/h) should be
included to ensure that particle sedimen-
tation within the inlet is  not significant.
Samplers should be tested  with mono-
disperse liquid particles for efficiency and
wall  loss measurements  and with  solid
particles to  determine possible excess
penetration  resulting from bounce and
reentrainment.
  This study resulted in the development
of several testing techniques which are
recommended for future work. The method
for wall  loss measurements is capable of
accuracies approaching 0.1 %  and requires
only standard apparatus.  Three  methods
for measurement of solid particle penetra-
tion have been developed.
  Additional work  is needed for adapta-
tion of the laboratory solid particle tests
used in  wind tunnel testing.  Static sam-
pling is sufficient for testing of an impac-
tion stage in a sampler. Use of A/C test
dust is not recommended because of its
friable nature.
  The following recommendations are
based on the current test  results:
  (1) The inlet to the dichotomous sam-
     pler should be replaced with one
     that is insensitive to wind speed.
  (2) Accurate  nozzle alignment in the
     commercial dichotomous samplers
     should be ensured  by the simple
     design change  identified in  the
     research report.
  (3) The size-selective hi-vol, which per-
     formed well under all testing, should
     continue to receive consideration as
     a candidate sampler.
References
Miller, F. J., D. E. Gardner, J. A. Graham,
R. E.  Lee, Jr., W. E. Wilson, and J. D.
Bachman.  Size  Considerations  for
Establishing  a  Standard for Inhalable
Particles.  J.  Air  Poll.  Control  Assoc.
29:610-615, 1979.

Suggs, J. C., C. E. Rodes, E. G. Evans, and
R. Baumgardner.  Inhalable Particulate
Network, Annual  Report: Operation  and
Data Summary, April  1979-June 1980.

-------
g
I
     700
      80
      60
      40
      20
                     \       \     I   I  I  I   I  I  I
               •   John & Wall
                    Okm/h
                    OOP

               D   John & Wall
                    Okm/h
                    Potassium biphthalate
         -    X
                   Wedding
                     2km/h
                     Oleic acid

                   McFarland & Ortiz
                     2km/h
                     Oleic acid
                                  I
                                                               D
                                            I  I  I   I
     4      6    8   10

Aerodynamic Panicle Diameter,,
                                                              20
                                                                           40
Figure 4.
           Sampling effectiveness of the size-selective hi-vol measured at zero wind speed
           for liquid (OOP) and solid (potassium biphthalate) particles. Data from wind tunnel
           tests at 2 km/h wind speed are shown for comparison.
                                                       EPA-600/4-81-037, U.S. Environmental
                                                       Protection  Agency,  Research Triangle
                                                       Park, NC 27711.

                                                       Loo, B. W., Unpublished progress report
                                                       to C. E.  Rodes.

                                                       McFarland,  A.   Unpublished  progress
                                                       report to C. E. Rodes.

                                                       McFarland, A. R. and C. A. Ortiz. Progress
                                                       Report  -- Aerosol Characterization of
                                                       Ambient Particulate  Samplers Used in
                                                       Experimental  Monitoring Studies. Texas
                                                       A & M University Research Foundation,
                                                       College Station, Texas, October 1979.

                                                       Wedding, J.  B.  Unpublished progress
                                                       report,  Colorado  State  University,  Ft.
                                                       Collins, Colorado, 1981.
      10
                0  Glass Beads

                X  A/C Test Dust

                                         i
Figure 5.
                         20              40               60               80

                               Aerodynamic Diameter, urn

            Penetration of the size-selective hi-vol by large glass beads and A/C test dust.

                                     4

-------
W. John, S. M.  Wall, and J. J. Wesolowski are with Air and Industrial Hygiene
  Laboratory, California Department of Health Services, Berkeley, CA 94704
Ralph Baumgardner is the EPA Project Officer (see below).
The complete report, entitled "Validation of Samplers for  Inhaled Paniculate
  Matter," (Order No. PB 83-191 395; Cost: $11.50. subject to change) will be
  available only from:
        National Technical Information Service
        5285 Port Royal Road
        Springfield, VA 22161
        Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
        Environmental Monitoring Systems Laboratory
        U.S. Environmental Protection Agency
        Research Triangle Park, NC 27711
                                                                                . S. GOVERNMENT PRINTING OfFKt 1983/659-095/1953

-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
         PS   0000329
         U S  ENVIR  PROTECTION  AGENCY
         REGIUN  5 LIBRARY
         230  S OEAR8URN  STREET
         CHICAGO IL 60604

-------