EPA/600/A^9S/201
 Yaacov Mamane1, Robert D. Willis2,  Robert  K.  Stevens3, and
 John L.  Miller4


 SCANNING ELECTRON MICROSCOPY/X-RAY FLDORE6CENCE
 CHARACTERIZATION  OF  POST-ABATEMENT DD8T
REFERENCE:  Mamane,  Y., Willis, R.  D.,  Stevens,  R.  K.,  and
Miller, J.  L.,  "SCANNING  ELECTRON MICROSCOPY/X-RAY
FLUORESCENCE  CHARACTERIZATION OF POST-ABATEMENT  DUST,"  Lead
in Paint. Soil  and Dust;  Health Risks.  Exposure  Studies.
Control Measures. Measurement Methods,  and  Quality
Assurance.  ASTM STP  1226. Michael E.  Beard  and S. p. Allen
Iske, Eds., American Society for Testing  and Materials,
Philadelphia, 1994.
ABSTRACT:  Scanning  electron microscopy  (SEM)  and  laboratory
X-ray fluorescence  (XRF) were used to characterize post-
abatement dust collected with a HEPA filter.   Three size
fractions of resuspended dust (0-150 ^m, 2.5-15 ^m,  and <2.5
jiirv) were collected on teflon filters and analyzed  by energy-
dispersive XRF.  Automated SEM was used to determine the
size, morphology, and chemistry of individual  particles from
0.2 M^ to greater than 250 fim.  Minerals associated with
construction materials, paint fillers, and soil were the
dominant species in  all size fractions.  Lead-rich particles
were found in all sizes and could be grouped into  three
categories: lead-only (including lead oxide and lead
carbonate), mixed lead/minerals, and automotive lead.
Isolated lead oxide  or lead carbonate particles derived from
paint pigments were  the dominant form of the lead-bearing
particles in the size fraction <15 ^m.


KEYWORDS:  lead, dust, scanning electron microscopy,  X-ray
fluorescence, post-abatement, paint
'Associate  Professor,  Environmental  Engineering Department,
Technion, Haifa 32000, Israel; Research  Scientist,  ManTech
Environmental, Research Triangle Park, NC 27709;  3Chief,
Source Apportionment Research Branch, U.S. Environmental
Protection Agency, Research Triangle Park, NC 27711; 4Senior
Consultant, US EPA, Research Triangle Park, NC 27711.

-------
                  GENERAL DISCLAIMER
 This document may have problems that one or more of the following disclaimer
                        statements refer to:
This document has been reproduced from the best copy furnished by the
sponsoring agency. It is being released in the interest of making
available as much information as possible.

This document may contain data which exceeds the sheet parameters. It
was furnished in this condition by the sponsoring agency and is the best
copy available.

This document may contain tone-on-tone or color graphs, charts and/or
pictures which have been reproduced in black and white.

The document is paginated as submitted by the original source/

Portions of this document are not fully legible due to the historical nature
of some of the material. However, it is the best reproduction available
from the original submission.

-------
 INTRODUCTION

      Knowledge of the  sources, size-distribution, and  lead
 content  of  lead-bearing particles is critical to assessing
 risk  and controlling environmental lead exposure.  Much
 information can be  learned from field samples by combining
 bulk  analytical techniques with microanalysis of individual
 particles.  This paper discusses results of a preliminary
 effort to characterize a lead-rich post-abatement dust
 sample using a combination of laboratory X-ray Fluorescence
 (XRF) for bulk analysis and manual and computer-controlled
 Scanning Electron Microscopy  (CCSEM) coupled with energy-
 dispersive  X-ray microanalysis (EDX) for individual particle
 analysis.   XRF provides rapid, quantitative, multielement
 analysis thus providing an "elemental context" for lead
 measurements and enabling interelement relationships to be
 investigated.  Scanning Electron Microscopy is an excellent
 complement  to the XRF technique.  Recent papers [1,2,3,4]
 demonstrate the power of CCSEM in apportioning sources of
 environmental lead  based on size, morphology, and
 composition of individual particles.

     Major  objectives in the present study were to determine
 if lead  concentrations varied as a function of particle size
 and to identify sources for the lead-bearing particles.
EXPERIMENTAL METHOD

Sample Description

     The sample used in this study, designated as dust D-5,
was prepared for the Atmospheric Research and Exposure
Assessment Laboratory of the U.S. Environmental Protection
Agency.  The sample was developed for the purpose of
establishing protocols and evaluating methods for the
analysis of lead-contaminated dust.  Details of the dust
collection, sample preparation, and bulk analysis of the
dust are given by Williams et al.[5].  The sample consisted
of lead-rich dust collected with a HEPA filtered vacuum
system during the abatement clean-up process in several
homes which had lead-based paint removed or encapsulated.
The majority of the dust particles were therefore expected
to be products of abatement activities.  The bulk lead
concentration in D-5 was previously determined to be 4550
Mg/g of dust based on a round-robin analysis of the sample
by atomic absorption (AA)  and inductively-coupled plasma
(ICP)  techniques.

XRF Analysis

     Energy-dispersive XRF analyses were carried out using
the XRF facility of the Source Apportionment Research Branch

-------
 (SARB)  of the EPA [6].   The  XRF  spectrometer measures  ng/cm2
 concentrations for elements  heavier than magnesium  for
 aerosol samples collected  on filter substrates.  Quantitation
 is based on thin-film standards  containing  known concentra-
 tions of selected elements.   Quality control standards
 including NIST thin-film SRMs containing certified  lead
 concentrations were measured before and after the sample
 analyses to monitor changes  in the operating condition of
 the spectrometer and  as  additional calibration checks.

      Samples were prepared for both XRF and SEM  analysis by
 aerosolizing sieved fractions of the dust and collecting the
 dust on teflon or polycarbonate  membrane filters.   Similar
 procedures for preparing XRF and SEM samples were used by
 Batterman et al.[7] in characterizing soils and  street dust.
 The parent D-5 material  had  been sieved at  60 mesh  to  remove
 particles >250 jim [5].   Prior to resuspension the dust was
 manually sieved a second time by the authors into two  size
 fractions:  150-250 /im and  below  150 ^m.  Figure  1 shows the
 particle resuspension chamber.   Dust sample was  introduced
 from the top of the chamber.   The chamber was designed such
 that particles smaller than  about 30 ^m aerodynamic diameter
 were suspended in the chamber by air forced through the
 glass frit at the base.  It  is possible that the original
 size distribution of  the D-5  dust was altered due to mass
 fractionation in the  resuspension process,  but such effects
 were beyond  the scope of this study; some preferential loss
 of  smaller particles  for example might be expected  due to
 electrostatic forces  between  particles and  the system  walls.

      A  Versatile Air  Pollution Sampler (VAPS) [8,9] was
 connected  to  the glass resuspension chamber via  a PM-15
 inlet operating at 32 1pm.  The  purpose of  the VAPS was to
 generate XRF  and SEM  samples  in  different size fractions.
 Particles  entering the inlet  impinged on a  virtual  impactor
 with a  cut-point of 2.5  /im, were partitioned into a coarse
 fraction (2.5  to 15 /im,  aerodynamic diameter) and two  fine
 fractions  (<2.5  pm),  and were  collected on  pre-weighed 47mm
 teflon  filters.   To avoid making large corrections  for X-ray
 attenuation,  dust  loadings for XRF samples  were  limited to
 less  than  200  M<3  cm'2.   Thus,  the quantity of dust collected
 on each  filter  ranged between  0.1 mg and 2  ing.

     Nine  aliguots of the fine sieved material were
 resuspended and  analyzed by XRF.  The nine  samples  are
 summarized in  Table 1 below and  included four fine  frac-
 tions, three coarse fractions, and two "total"  fractions.
 The  latter were  collected by removing the PM-15  inlet and
 replacing the VAPS with  a single filter holder operating
with a flow of  25  1pm; "total" fraction samples  in principle
 included particles ranging in size from zero up  to  the
maximum particle size (nominally 150 /im)  of the  fine sieved
dust.  "Total"  filters were not expected to proportionately

-------
    Dean
                               Excess Air
                                v_7
                                X
                          PV-15 Intel
                            32 L/mm
t    t
                   of
               Nyciepore •
                Finer
t    t
                                /\

Fmt



Con-Be
t


Fine
                                               Glass
                      Nuctepore Finer
                      Teflon Fiiief
m I 2.0 L/min
1
tSUmin



Pumps
     FIGURE 1—Schematic diagram of  the dust resuspension
chamber.   Dust particles injected  from the top  of the
chamber were size-selected using a Versatile Air  Pollution
Sampler  (VAPS) and collected on filters for analysis by XRF
and SEM/EDX.

-------
 represent  the  <150  jjm  size distribution of the D-5 dust both
 because  the  resuspension  chamber was designed only to
 aerosolize particles less than  30 fim and because the larger
 particles  have a higher probability of settling directly
 into  the sampling inlet.  Therefore, data obtained from
 "total"  samples was interpreted only in terms of the
 differences  between "total" and fine or course samples.
 Filter tare  weights were  measured after allowing the filters
 to equilibrate for  at  least 12  hours in a temperature and
 humidity-controlled room.  After sampling, the loaded filter
 was again  equilibrated in the balance room for at least 12
 hours before weighing.
TABLE 1—Post-abatement dust samples analyzed by XRF and SEM

     Sample ID      Particle Size, jim        Analysis
Fl
F2
F3
F4
C2
C3
C4
Tl
T2
F5
C5
FS1
CS1
0 - 2. 5"
0 - 2.5'>
0 - 2.5'>
0 - 2.5"
2.5 - 15 '>
2.5 - 15 »
2.5 - 15 »
0 - 150 ])
0 - 150 '>
0.2 - 4*
1.5 - 15 2)
5 - 50*
40 - 35025
XRF
XRF
XRF
XRF
XRF
XRF
XRF
XRF
XRF
CCSEM
CCSEM
CCSEM
CCSEM
Notes:  1) Aerodynamic diameter.
        2) Geometric diameter.

     Four additional samples were prepared for automated SEM
analysis.  Samples F5 and C5  (Table 1) were made with
resuspended dust as described above but collected on
polycarbonate nuclepore filters for improved SEM imaging.
Samples FS1 and CS1 were prepared by manually sprinkling
fine sieved dust (<150 ^m) and coarse sieved dust (150-250
^m) respectively directly onto carbon planchets for SEM
analysis.  No attempt was made to preserve the original size
distribution of the dust.  The particle sizes given in Table
1 for the four SEM samples are the effective diameter
criteria used by the SEM to search for particles in the
automated mode.

     SEM samples were analyzed by computer-controlled
SEM/EDX at R.J. Lee Group, Inc. (Monroeville, PA).   The
resulting particle data and image files were processed and
interpreted at the EPA using the Zeppelin Microimaging
System (R.J.  Lee Group, Inc.)  developed for interpreting

-------
 CCSEM data off-line.   The  software  assigns  a  chemical
 classification to each particle  based  on the  four  dominant
 elements in the particle's X-ray spectrum.  Data for each
 sample are then summarized in the following tables:

      1.  Number of particles in each chemical  class,
 percentages,  and average particle geometric diameter.
      2.  Size,  area and mass distribution of particles by
 average  diameter and  chemical class.  (A density is assigned
 to each  particle based on  its elemental composition).
      3.  Average elemental  composition  of the  chemical
 classifications provided in the  first  table.
      4.  Mass  and number distribution by aerodynamic
 diameter.
RESULTS

XRF Analysis

     Elemental concentrations in ng/cm2 were measured by XRF
for the nine resuspended dust samples listed in Table l.
These values were converted to Mg/g concentrations using the
measured area and deposited mass for each filter.  During
preparation of the samples it was found that a substantial
fraction of the deposited dust could be shaken off during
handling of the filter.  The problem was most severe for the
heavily loaded coarse and total filters.  Mass losses
occurring after the loaded filter was weighed and before XRF
analysis would cause concentrations to be underestimated as
well as bias the size distribution.  In order to put bounds
on the measured concentrations, all teflon filters were
weighed again immediately after XRF analysis to determine
sample losses due to handling.

     High concentrations of aluminum, silicon,  sulfur,
potassium, calcium, titanium, iron, zinc, strontium, barium
and lead were measured in all three size fractions of D-5.
Figure 2 shows the estimated concentrations of these
elements in the three size fractions analyzed by XRF.  The
abundances shown in Fig.2 are based on the assumption that
two-thirds of any dust mass losses due to handling occurred
before the XRF analysis.  If larger fractions of sample were
lost from the coarse and total filters,  their elemental
concentrations would be underestimated in Fig.2. Thus, the
apparent increase in elemental concentrations with
decreasing particle size may be artificial.   Uncertainties
in the concentrations reported in Fig.2  are approximately +
20% for fine fraction data and ±40% for the coarse and
total fraction results.

     Not surprisingly,  calcium had the highest  elemental
concentration in the post-abatement dust:  calcium carbonate

-------
 is the major constituent of cement,  while gypsum (CaSO4)  is
 a major constituent of plaster and wallboard,  and is
 frequently used as a filler in paints.   Some of the
 aluminum,  silicon,  and potassium is  probably associated with
 soil-derived dust or street dust.  SEM  analysis of
 individual particles suggests  that aluminosilicates are also
 associated with paint as paint fillers.   Kaolinite and talc
 for example are common fillers in paints.   Sulfur,  titanium,
 iron,  zinc,  strontium,  barium,  and lead are all commonly
 found  in paint  pigments.
      03
      O
      C
      (0
      "D
      C
      D
      ID
3.b-
-
.5-
•"

-
,5-
n.






,
S



-
1

|





S
!





P
1





1
1




"V
1
•i m
I










(
1
I
I
                                                 0-2.5 om
                                                 ED
                                                 2;5 -15 um
                                                 0 - 30 urn
             Si (x.1)  Ca(x,1)    Fe    Sr(xlO)  Pb(x10)
                  S  •'    Ti    Zn(x10)  Ba(x10)

     FIGURE 2—Estimated elemental concentrations measured
by XRF in three size fractions of post-abatement  dust.   The
increase in concentrations with decreasing  particle  size may
be an artifact associated with uncertainties  in the  sample
masses.
     The estimated lead concentration in the  fine  fraction
samples is 2980 ± 600 pg/g.  This is considerably  less  than
the bulk concentration of 4550 Mi/9 determined  by  AAS and
ICP and adopted as the consensus value.  This may  represent
a true difference between the lead concentration in  the bulk
dust and in the fine fraction, or there may be  additional
errors in the XRF analysis which have not been  accounted
for.  The above result is however similar to  the bulk
concentration of 2485 M9/g determined by XRF  in the  round-
robin analysis of D-5 [5].

-------
 SEM Analysis

     Table  2  summarizes CCSEM results obtained on samples
 F5,  C5,  and FS1 with the SEM operated in the secondary
 electron node (SE).   (Sample CS1 was analyzed only  in the
 backscattering mode).  Particles were automatically sized
 and analyzed  by EDX.  Images of particles containing high-Z
 elements were automatically collected and stored on optical
 disk for off-line  review.  Each particle was assigned to one
 of  several  chemical classes based on the particle's X-ray
 spectrum.

     Table  2  shows that the composition of the three size
 fractions is  similar.  As expected, minerals associated with
 construction  materials and paint fillers dominate the three
 size fractions.  The titanium-rich particles are probably
 fragments of  paint chips.  Examination of individual
 particles showed that the titanium was generally accompanied
 by  aluminum silicate and calcium or gypsum.  Iron-rich
 particles in  general were also rich in calcium, silicon, and
 sulfur.
      TABLE 2—CCSEM analysis of D-5 samples (SE mode)

Sample              F5             C5             FS1
/Particles
Size, ^m
385
0.2-4°
500
1.5-15"
232
5-50»
Chem. class                  Percent by number
Carbonates
Gypsum
Alum-Silicate
Quartz
Ti-rich
Fe-rich
Pb-rich
39
26
19
7
5
3
12>
36
22
24
10
4
3
0.52)
42
9
19
19
7
4
0.4
Notes:  1) Average geometric diameter.
        2) Based on CCSEM analysis in Backscatter mode.


     Fine and coarse fraction samples were analyzed by CCSEM
operating in the backscatter mode.  The intensity of the
backscattered electron (BSE) signal increases with atomic
number.  By setting a threshold on the BSE signal one can
exclude light elements from automated searches and greatly
enhance the efficiency of automated searches for particles
containing heavier elements such as iron,  copper, zinc,
barium, and lead.  With the BSE threshold set to exclude
elements lighter than iron, 100 fine and 161.5 coarse fields

-------
 of view were  scanned,  equivalent, to analysis  of  a population
 of 9600 and 18800  particles  respectively.  A  total  of  216
 fine  and 300  coarse  metal-rich particles were found in the
 BSE search.   Examination  of  their x-ray spectra  revealed
 that  the majority  of these particles were  iron-rich, lead-
 bearing,  or barium or  zinc-rich  particles  associated with
 paint materials.   (Barium was almost always accompanied by
 sulfur indicative  of barium  sulfate pigment).

        The  abundance of lead-rich particles in the  fine and
 coarse fractions was respectively around 1 and 0.5  particles
 per field of  view  at a magnification of 1000X.   Their
 average sizes were 0.8 pm and 2.5 ^m respectively.   Table 3
 shows an estimate  of the  lead concentration in the  fine and
 coarse fractions based on the the BSE and  SE  data.   The
 results agree well with the  XRF  results for the  same size
 fractions.
     TABLE  3—CCSEM-based estimate of Pb concentrations
                in Fine and Coarse fractions.

Fraction
Fine Pb
Fine Total
Coarse Pb
Coarse Total
A
/Particles
per field "
0.99
96.25
0.48
116.28
B
Avg. dia.
(^m)
0.8
1.6
2.5
3.6
C
Density
(g/cm3)
5
2.7
5
2.7
Relative
Loading
(AxB3xC)
2.53
1064
37.7
14650
Notes: 1) Magnification •= 1000X.
Estimated Pb in Fine fraction = 2.53/1064 = 2380
Estimated Pb in Coarse fraction = 37.7/14650 = 2570
     Fine sieved dust sprinkled by hand onto a carbon
planchet was analyzed in the size range of 5 to 50 Jim and
the results are as follows: 232 particles were found in 34
fields at magnification of 300X.  In the SEM backscatter
mode only one 7.6 pm lead-rich particle was found in 35
fields of view representing an estimated concentration of
2400 ppm as calculated above.  Time limitations precluded an
extended search of more fields in order to obtain better
statistics for this size range.

     Lead-bearing particles—The speciation of lead
particles was one of the objectives of this study.  Off-line
examination of the morphologies and X-ray spectra of the
lead-rich particles identified three major groups of
particulate lead:

-------
      1.   Particles  Containing only Lead—This group
 consisted of  particles composed entirely of Pb,  excluding O
 and C.   Many  of  these particles appeared as cubes or
 hexagonals.   This is  the morphology of basic carbonate of
 white lead, the  pigment commonly used in leaded  paints.
 This was the  most abundant  class of lead-rich particles
 found in the  fine and coarse  fractions of D-5.   Manual SEM
 showed the presence of these  particles down to 0.25 jam.  The
 hexagonal morphology  easily distinguishes these  particles
 from combustion-generated lead-only particles which
 typically appear as chain aggregates.   Figure 3  is a
 photomicrograph  of  a  typical  lead carbonate particle.
   000
5.000
JO. 000
15. 000
20. 000
                           ENERGY
                        keV
     FIGURE 3—Scanning electron micrograph and X-ray
spectrum of a lead carbonate particle in D-5.  The hexagonal
symmetry is characteristic of basic white lead carbonate
coitunonly used in paint pigments.

-------
      2.   Mixed Lead/Minerals—Minerals  enriched in lead,
 apparently as  lead  oxide or lead  carbonate particles
 attached to gypsum,  calcite,  or aluminosilicate surfaces
 were  found in  all size  ranges in  the  samples analyzed.
 These are probably  fragments  of gypsum  wallboard or other
 construction materials  that had been  painted with leaded
 paint.   Figures 4 and 5 are photomicrographs of paint chips
 rich  in  lead.
  000
5.000
10.000
15. ODD
20. ODD
                          ENERGY
                        KgV
     FIGURE 4—Backscattered electron image of a leaded
paint chip approximately 200x300 ^m.  Areas rich in heavy
elements appear bright in the BSE mode.  X-ray analysis of
the bright areas showed high lead concentrations.  The X-ray
spectrum above was collected from an area adjacent to the
lead-rich region and indicates calcium and sulfur (probably
as gypsum), titanium, and aluminum silicate.

-------
           5.000
10. 000
15. 000
20. 000
                        ENERGY
           keV
     FIGURE 5—Secondary Electron  (top)  and  Backscattered
Electron (bottom) images of a leaded paint chip.  The bright
ridge in the BSE image was rich in  lead,  zinc,  barium,  and
calcium as shown in the X-ray spectrum.   Analysis of areas
adjacent to the lead-rich region showed  gypsum,  titanium,
and aluminum silicate.

-------
      3.   Automotive Lead—Because the  D-5  dust  sample was
 comprised of post-abatement dust,  the  contribution from
 street dust vas expected to be  snail.   Nevertheless,  several
 particles containing both lead  and bromine and  possibly
 associated vith automotive emissions vere  found in D-5.
 These particles vere all smaller  than  three microns.
 DISCUSSION AUD CONCLUSIONS

      The primary objective  of  this  limited  study vas to
 characterize post-abatement dust using a combination of
 laboratory XRF for bulk  analysis of size-selected  fractions,
 and SEM/EDX for analysis of individual particles,  especially
 lead-rich particles.   The results confirmed the expected
 dominance of construction-related and paint-derived  (pigment
 and mineral filler) particulates in the post-abatement dust.
 Approximately 1% of the  9600 fine-fraction  particles
 (0.2-4  pm)  and 0.4% of the  18800 coarse particles
 (1.5-15 pm)  analyzed by  CCSEM  vere  lead-rich.  In  both size
 fractions the majority of these particles vere isolated lead
 oxide or lead carbonate  particles derived from paint
 pigments.   In the coarse fraction a larger  fraction of the
 lead appeared as lead  carbonate particles attached to other
 minerals or as mixed lead/mineral particles.

      One of the goals  in this  study was to  determine how
 lead concentrations vary with  particle size.  DeVoe at NIST
 has recently measured  lead  concentrations in sieved
 fractions of lead-rich dusts and found the  highest lead
 concentrations in particles less than 50 ^m [10].  This
 finding is  consistent  with  the high concentration  of lead
 oxide or lead carbonate  particles observed  by CCSEM in the
 size fraction below 15 M»*  Quantitative differences, if
 any,  in the  lead content of D-5 in  the fine, coarse, and
 total size  fractions analyzed  by XRF were unfortunately
 obscured by  the large  uncertainties in the  mass deposited on
 the XRF samples.   Also,  because of  probable mass
 fractionation effects  associated with the particle
 redeposition process,  the size distribution of the analyzed
 samples cannot be  assumed to be representative of  the
 original dust.

      Problems  which limited the present study should be
minimized in future studies vith the recent acquisition of a
new XRF spectrometer by  the Source Apportionment Research
Branch.  The  new system  will enable quantitative,
multielement analyses  on bulk dust  samples without
restriction  to particle  size,  thus eliminating the need for
resuspension.  The ability to analyze larger samples will
minimize the potential for non-representative sampling which
may have contributed to uncertainty in the XRF data.
Additional computer-controlled SEM/EDX analysis of samples
collected from  fluidized bed resuspensions should  further

-------
enhance our ability to characterize and determine the
sources of lead particles in the environment.
REFERENCES

[1]  Casuccio, G. S., Demyanek, M. L., Dunmyre, G. R.,
     Henderson, B. c., and Stewart, I. M., "Characterization
     and Identification of Lead-Rich Particles: A First Step
     in Source Apportionment," Proceedings of the 204th
     Symposium of the American Chemical Society, Washington
     DC, August 23-2S, 1992, in press.

[2]  Vander Wood, T. B. and Brown, R. S., "The Application
     of Automated Scanning Electron Microscopy/Energy
     Dispersive X-ray Spectrometry to the Identification of
     Sources of Lead-Rich Particles in Soil and Dust,*1
     EnvironmentalChoices Technical Supplement. July/August
     1992, pp. 26-32.

[3]  Hunt, A., Johnson, D. L., Watt, J. M., and Thornton,
     I., "Characterizing the Sources of Particulate Lead in
     House Dust by Automated Scanning Electron Microscopy,"
     EnvironmentalScience and Technology. Vol. 26, No. 8,
     1992, pp. 1513-1523.

[4]  Johnson, D. L. and Hunt, A., "Speciation of Lead in
     Urban Soils by Computer Assisted SEM/EDX - Method
     Development and Early Results," Proceedings of the 1993
     Boulder Conference on Lead in Paint, Soil, and Dust,
     Boulder CO, July 25-29, 1993, this publication.

[5]  Williams, E. E., Binstock, D. A.,  Estes, E. D., Neefus,
     J. D.,  Meyers, L.  E., and Gutknecht, W.  F.,
     "Preparation and Evaluation of Lead-Containing Paint
     and Dust Method Evaluation Materials," Proceedings of
     the American Chemical Society Symposium on Lead
     Poisoning in Children:  Exposure,  Abatement and Program
     Issues, Washington DC,  August 24-25, 1992, in press.

[6]  Dzubay, T.  G., Stevens, R. K., Lewis,  C. W.,  Hern, D.
     H., Courtney, W. J.f Tesch,  J. W., and Mason, M. A.,
     "Visibility and Aerosol Composition in Houston, Texas,"
     EnvironmentalScience and Technology.  Vol. 16, 1982, p.
     514.

[7)  Batterman,  S. A.,  Dzubay,  T. G.,  and Baumgardner,  R.
     E., "Development of Crustal  Profiles For Receptor
     Modeling,"  Atmospheric  Environmentf  Vol. 22,  No. 9,
     1988, pp. 1821-1828.

[8]  Gofer,  W. R.  Ill,  Stevens, R. K.,  Winstead,  E. L.,
     Pinto,  J. P., Sebacher, D. I., Abdulraheem, M. ¥.,
     Al-Sahafi,  M.,  Mazurek, M. A., Rasmussen,  R.  A.,

-------
     Cahoon, D. R., and Levine, J. S., "Kuwaiti Oil Fires:
     Compositions  and Source Smoke," Journal Of Geophysical
     Research. Vol. 97, 1992, pp. 14521-14525.

 [9]  Stevens, R. K., Pinto, J., Conner, T. L., Willis, R.,
     Rasmussen, R. A., Mamane, Y., Casuccio, G., Benes, I.,
     Lenicek, J.,  Subrt, P., Novak, J., and Santroch, J.,
     "Czech Air Toxics Study (CATS): Project Summary,"
     Proceedings of the 86th Annual Meeting of the Air and
     Waste Management Association, Denver CO, June 13-18,
     1993, in press.

 [10] Jim DeVoe, private communication.
ACKNOWLEDGEMENTS

     The assistance of Mr. Bradley Henderson of R.J. Lee
Group, Inc. in the analysis of samples by CCSEM, and Mr.
Robert Kellogg of ManTech Environmental Technology Inc., in
the analysis of filter samples by XRF is gratefully
acknowledged.  The authors are grateful to Ms. Karen Blume
for her assistance and many helpful suggestions during the
course of the project.
DISCLAIMER

The information in this document has been funded in part by
the U.S. Environmental Protection Agency under contract
(#68-09-0131) to Acurex and contract (#68-DO-0106)  to
ManTech Environmental Technology, Inc.   It has been subject
to Agency review and approved for publication.  Mention of
trade names or commercial products does not constitute
endorsement or recommendation for use.

-------
TECHNICAL REPORT DATA
1. REPORT NO.
EPA/600/A-93/201
2.
4. TPTLE AND SUBTITLE
SCANNING ELECTRON MICROSCOPY/X-RAY FLUORESCENCE
CHARACTERIZATION OF POST-ABATEMENT DUST
7. AUTHOR(S)
Yaacov Mamane', Robert D.
John L. Miller"
Willis2, Robert K.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
'Technion, Haifa, Isreal 32000
:Mantech Environmental Technology Inc., Res<
Triangle Park, NC 27709
3Atmospheric Research and Exposure Assessmei
Laboratory, US Environmental Protection Age
Research Triangle Park, NC 27711
"Senior Consultant, US Environmental Proted
Agency, Research Triangle Park, NC 27711
Stevens3,
=arch
it
ncy,
:ion
12. SPONSORING AGENCY NAME AND ADDRESS
Atmospheric Research and Exposure Assessment
Laboratory, Office of Research and Development,
US Environmental Protection Agency
Research Triangle Park, NC 27711-0047

5. REPORT DATE
6.PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT
NO.
10.PROGRAM ELEMENT NO.
1 1 . CONTRACT/GRANT NO.
In-house
13. TYPE OF REPORT AND PERIOD COVERED
Journal Article
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
Scanning electron microscopy (SEM) and laboratory X-ray fluorescence (XRF) were
used to characterize post-abatement dust collected with a HEPA filtered vacuum.
Three size fractions of resuspended dust (0-30 pm, 2.5-15 jjm, and <2 . 5 /jm) were
collected on teflon filters and analyzed by energy-dispersive XRF. Automated SEM
was used to determine the size, morphology, and chemistry of individual particles
from 0.2 ^m to greater than 250 ^m. Minerals associated with construction
materials, paint fillers, and soil were the dominant species in all size fractions.
Lead-rich .particles were found in all sizes and could be grouped into three
categories: lead-only (including lead oxide and lead carbonate), mixed
lead/minerals, and automotive lead. Isolated lead oxide or lead carbonate
particles derived from paint pigments were the dominant form of the lead-bearing
particles in the size fraction <15 pm.
17.
KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS

18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC


b. IDENTIFIERS/ OPEN ENDED c.COSATl
TERMS

19. SECURITY CLASS (This Repori) 21. NO. OF PAGES
UNCLASSIFIED 16
20. SECURITY CLASS (TJiis Page) 22. PRICE
UNCLASSIFIED

-------