EPA-650/2-73-006
June 1973
Environmental Protection Technology Series
llillliB
II
HI
mm*
::?;-,,.-,.-,
:S:SS;:S
msm
-------
EPA-650/2-73-OQ6
DEVELOPMENT OF X-RAY
FLUORESCENCE SPECTROSCOPY
IN ELEMENTAL ANALYSIS
OF PARTICULATE MATTER
PHASE II: EVALUATION OF COMMERCIAL MULTIPLE
CRYSTAL SPECTROMETER INSTRUMENTS
by
L. S. BirksandJ. V. Gilfrich
Naval Research Laboratory
Washington. D. C. 20375
Interagency Agreement No. EPA-IAG-085(D)
Program Element No. 1A1010
EPA Project Officer: Jack Wagman
Chemistry and Physics Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D .C. 20460
-------
This report has been reviewed by the Environmental Protection Agency and
approved for publication. Approval does not signify that the contents
necessarily reflect the views and policies of the Agency, nor does
mention of trade names or commercial products constitute endorsement
or recommendation for use.
11
-------
CONTENTS
Abstract iv
Problem Status iv
Authorization iv
INTRODUCTION 1
EXPERIMENTAL RESULTS .. 4
DISCUSSION 6
Number of Elements 6
Tube Targets and Power .. 6
Crystals 6
Detectors 6
Data Handling 6
Sample Handling 7
REFERENCES 7
111
-------
ABSI’RACT
Four commercial multiple crystal spectrometer x-ray
analyzers were evaluated for use in the elemental analysis
of air pollution particulate samples. Fourteen to twenty-
four elements can be measured simultaneously in these
instruments. 100 second detection limits of 1 to 10 ngf cm 2
were achieved for about one-half of the elements examined.
Any. one of the commercial instruments is capable of per-
forming quantitative analysis of the particulate matter filtered
out of the atmosphere or source emissions. Some actual
pollution samples were analyzed in all four instruments to
demonstrate suitability.
PROBLEM STATUS
This report is the final report by the X-Ray Optics Branch
on one phase of the problem; work is continuing on other phases
of the problem.
AUTHOR IZATION
NRL Problem P04- 06
EPA-NRL Interagency Agreement No. 690114
This report has been reviewed by the Environmental Protection
Agency and approved for publication. Approval does not signify
that the contents necessarily reflect the views and policies of the
Agency, nor does mention of trade names or commercial products
constitute endorsement or recommendation for use.
Manuscript submitted May 29, 1973.
iv
-------
Development of X-Ray Fluorescence Spectros copy
for Elemental Analysis of Particulate Matter
in the Atmosphere and in Source Emissions
INTRODUCTION
In a previous report 1 it was concluded that multiple-crystal-
spectrometers offer the most efficient method of performing large
scale x-ray fluorescence analysis of air pollution particulate samples.
Based on that conclusion, an investigation was inaugurated to evaluate
the commercially available equipment of that type. Most large-scale
industrial x-ray analysis is presently performed using these multiple
crystal spectrometer instruments and there are four manufacturers of
such equipment. Fourteen to twenty-four elements can be measured
simultaneously. In addition to the combination of speed and resolution
of these instruments a further advantage is that each spectrometer
channel can be optimized for the individual element it is measuring
(best divergence, best crystal, best detector). This is important for
air pollution particulate samples where elements from sodium to lead
must be measured. Table I lists the characteristics of the four com-
mercial instruments as described in the manufacturer’s literature.
TABLE I cOMPARISoN OF MULTICHANNEL INSTRUMENTS
ARL Rigaku Philips Siemens
7 1000 “ SI IVIULTIX ” PW 1270 MRS-3
Max. No. of Spectrometers 24 24 14 17
No. of Positions Occupied 3 3 Not Available 2
by Scanner
No. oF Positions Occupied All Vacuum 2 All Vacuum All Vacuum
by Vac. Spectr.
X-Ray Tubes Available MachlettOEG-75 MachlettOEG.75 Philips FAA 100/3.5 Siemens AG-61
W, Pt, Rh-3 kW W, Ag. Rh-3 kW cr, Cu , Rh. Ag , Pt, Au- Mo, W, Au-3kW
cr-a. 5kW cr-2.skw 3kW, W, Mo-3.SkW cr-a.6kw
Rh-l. 5kW
crystal Optics curved and Ground Flat Flat Logarithmic curve
Read Out Integrator Integrator Scaler Scaler
X-Ray Incidence Angle 900 90° 52° 450
X-Ray rake-Off Angle 30°, 350, 40° 30° 23°, 550 30°, 3B°, 52°, 60°
1
-------
BLrks and Gilfrich
Figure 1. Geometry of the Laboratory Spectrometer
With the cooperation of the x-ray equipment manufacturers, sets
of standards prepared at the Naval Research Laboratory (NRL) along
with air pollution samples collected by the Environmental Protection
Agency (EPA) were taken to each of the manufacturers’ applications
laboratory and measured by NRL personnel using the multiple spec-
trometer instruments.
As noted in the previous report, the most significant modification
made to the laboratory equipment during the Phase I investigation was
to design the sample holder so that primary radiation illuminated only
the sample on its filter substrate and after being transmitted through
the substrate, was trapped in an area which could not be viewed by the
measuring system. This is illustrated for the laboratory wavelength
dispersion spectrometer in Figure 1. One specific goal of the Phase II
effort was to determine how the specimen holders of the conirnercial
instruments could be similarly modified within the constraints im-
posed by their mechanical construction. All four of the instruments
evaluated were amenable to the use of a sample cup as shown for a
composite instrument in Figure 2. For the measurements reported
here, three of the commercial instruments used plastic sample cups,
while the fourth instrument used a gold-plated metal cup. Because
these multi-channel instruments have the spectrometers arranged on
an arc or circle around the sample, it was not possible to avoid having
some of the spectrometers view a portion of the sample cup illumin-
ated by the primary beam as shown in the figure. Therefore the back-
ground was not as low as might have been possible if more extensive
modifications could have been made as shown in Figure 3.
LUCITE
SAMPLE
HOLDER
CRYSTAL SPECTROMETER
2
-------
NRL Report 7617
SAMPLE
BETWEEN LAYERS
OF MYLAR
PUMPING
ORTS
SPECIMEN HOLDER
WITH SAMPLE CUP
Figure 2. Geometry of a composite multiple spectrometer instrument,
shown for an end-window x-ray tube and both curved and flat crystal
x-ray optics. Side window tubes present an analogous situation.
Figure 3. Modifications to the sample holder to decrease the back-
ground due to fluorescence and scattering of the primary x-ray beam.
a.) Aperture
b.) Large Sample Cup
APERTURE
Dir
a.
b.
3
-------
Birks and Gilfrich
A second goal of Phase II was a comparison of the sensitivity and
detection limits for the four commercial instruments. To accomplish
this, actual pollution samples, as well as the calibration standards of
individual elements prepared at NRL, were measured on each of the
instruments. In some instances spectrometers were not immediately-
available for all of the elements of interest but presumably would be
available for a purchased instrument for a particular application. For
instance, in the air pollution problem, it is desirable to measure As
K$ and Pb L in order to avoid the interference between As Ka and
Pb La. In addition to measuring the samples the ease of instrument
disassembly, servicing and recalibration was demonstrated by each of
the manufacturers.
EXPERIMENTAL RESULTS
In an attempt to be perfectly objective in reporting the results of
this investigation, the only data which are being shown are the detec-
tion limits measured on the filter paper standards and some actual
results on pollution samples measured with each of the four instru-
ments. Table II lists detection limits for the elements of interest with
the number of check marks n each range indicating the number of
instruments giving detection limits within that particular range. The
purpose of the somewhat cryptic presentation of data is to avoid giving
the impression of endorsement to any individual manufacturer.
TABLE II 3cy DETECTION LIMITS
ng/cm 2 on Filter Paper (100 sec. measurement)
1 - 5
5 - 10
10 - 50
50-200
200-500
Na
xx
x
Mg
xxx
x
Al
xx
xx
Si
xx
xx
S
xx
xx
Cl
xx
x
K
xx
x
Ca
xx
x
V
x
x
Fe
x
xx
Co
xx
Ni
xxxx
Cu
x
xxx
Zn
xx
x
As
(Ku)
x
Cd
(La)
x
x
Pb
(La)
xx
4
-------
NRL Report 7617
TABLE III POLLUTION SAMPLES
Concentration ( gIcm 2 )
Na Mg Al Si S Cl K Ca V Fe Co Ni Cu Zn Cd Pb
Incinerator C-7
NRL
Instrument A
0.63 0.35 20 100 21 1.4 0.05 2.5 0.30 ND 2.5 58 0.33 59
13 0._14__0.40_0._14 10 84 6.6 0.93__0.02__1.6 0.05__0.23__2._1 78 0.37 55
Incinerator C-9
NRL
InstrumentA
B
C
D
ND ND ND 3.6 0.04 0.24 MD ND ND ND ND 0. 12 ND ND
0.04 0.01 ND 0.02 0.10 2.2 ND ND ND ND ND ND ND ND 0.002 ND
ND ND ND ND 0.13 2.3 0.04 0.14 ND 0.11 0.01 ND ND ND 0.63
ND ND ND 0.03 ND 1.8 ND ND ND 0.08 ND ND
0.05 0.02 ND 0.003 ND 0.15 0.03 ND 0.07
Incinerator N-6
NRL
Instrument A
B
C
D
ND 0.30 63 70 15 2.0 ND 2.0 ND 0.30 5.0 120 ND 120
38 0.73 0.31 0.41 68 65 16 1.8 0.09 2. 1 0.09 0.60 6.6 200 0.79 110
13 0.89 0.39 0.38 40 >56 51 2.0 1.2 2.0 0.08 2.6 90 3.1 86
30 1.7 0.34 0.95 88 73 48 1.9 ND 0.23 4.4 250
2.6 0.31 0.37 76 2.0 5.7 0.39 5. 1 130
incinerator B-I
NRL
Instrument A
B
C
D
ND 0.08 3.2 17 3.5 0.30 ND 0.30 0.04 0.40 0.80 13 ND 4.4
6.0 0.04 0.15 0.08 2.1 16 1.0 0.19 ND 0.22 ND 0.20 ND 18 0.08 4.6
0.81 ND 0.17 0.09 2.6 IS 3.6 0.31 0.11 0.23 0.01 MD 6.5 0.31 4.8
3.3 0.14 0.17 0.27 3.0 15 3.5 0.17 ND ND ND 16
0.35 0.22 0.12 3.3 0.25 0.40 ND 0.29 4.2
Incinerator B-3
NRL.
InstrunientA
B
C
D
ND 1.1 14 33 10 4.9 ND 2.5 ND 0.40 1.7 37 ND 34
11 0.2! 1.1 0.48 10 38 4.2 1.6 0.12 2.0 0.05 0.20 1.3 61 0.53 35
3.2 0.61 1.2 0.53 13 42 13 1.7 0.26 l.8 0.03 ND 2! 1.2 34
7.6 0.75 1.6 1.4 17 40 15 1.6 ND 0.11 0.89 65
1.3 1.4 0.53 17 1.8 4.8 0.14 1.0 37
Incinerator T-16
NRL
Instrument A
ND 0.30 8. 1 1.8 2.5 0.40 ND 0.30 ND 0.30 0.90 13 ND 11
5.9 0.03 0.02 0.23 5.0 7.3 3.2 0.37 ND 0.28 ND 0.05 0.46 19 0.65 12
Incinerator T-17
NRL
Instrument A
1. 1 0.60 20 5.3 6.7 1.5 ND 0.80 ND 0.50 2.0 32 ND 29
12 0.06 0.10 0.51 12 3.6 9.1 1.2 ND 0.81 ND 0.10 1.4 40 1.5 30
Power 12
NRk
Instrument A
B
C
D
6.2 8.6 2! ND 2.8 5.6 0.07 14 0. 10 0.70 I. I 0.50 ND 1.0
0.73 0.45 8.7 7.6 16 ND 3.8 4.2 0.23 17 0.04 0.20 0. 10 0.45 0.01 ND
0.52 0.83 11 9.0 17 ND 2.7 5.0 0.20 16 0. 19 ND 0.22 0.41 ND
0.61 0.77 13 24 20 ND 2.6 4. 5 0.04 0.21 ND 0.53
1.0 12 9.4 21 5.0 37 0.30 0.53 0. 14
Cement IA-B(g)
NRL
Instrument A
ND 0.03 0.30 0.30 0.06 0.20 ND 0.20 ND 0.40 0.80 0.30 ND ND
0.06 ND ND 0.004 0.21 ND 0. 14 0.04 MD ND ND ND ND ND
Cement 4D-B(g)
I”I L
I’istrument A
ND 0. 10 0.30 0.70 0.20 0.50 ND 0.3’) ND 0.30 1.0 0.30 ND ND
0. 12 0.04 0.03 0. 16 0.27 0.38 0.28 0.66 ND 0. i4 ND ND ND ND 0.002 ND
Cement IA-B(o)
NRL
Instrument B
C
D
ND 0.20 0.30 0.60 0.20 0.90 ND 0 10 ND 0.20 1.0 0.20 ND 0.60
ND ND ND 0.20 0.20 0.67 0. 13 0.32 ND 0.07 ND ND ND ND ND
ND ND 0.03 0.47 0.27 0.10 0.17 0.31 ND ND ND ND
ND 0.06 0.20 0.40 0.45 0.24 0.03 ND 0.09
ND Not Detectable
Table III shows quantitative analyses for some of the EPA samples
with the results coded so that the instruments are identified only as
HA, B, C and D. Previous x-ray analyses at NRL are also shown.
The overall estimates of concentration are similar from instrument to
instrument but occasionally one element will appear low or high by
more than a factor of two. Inasmuch as the calibration curves all
showed deviation only within statistical iimits(2) the most logical spec-
ulation concerning these differences in results seems to be that the
particles are not uniformly deposited and that different areas were ex-
amined by different instruments. There is no system to the discrep-
ancies in terms of instrument, element, concentration or sequence of
measurement.
5
-------
Birks and Gtlfrich
DISCUSSION
The experimental results show that all four of the commercial
instruments can measure the desired elements in the concentration
range of interest for air pollution. Simple modification of the speci-
men holder to use a sample cup improved the detection limits to the
range of 1 to 10 ng/cm 2 for almost half of the elements studied. More
extensive modification, as illustrated in Figure 3, should improve this
even further.
All of the instruments are convenient to operate and service; a
relatively short period of time was required (about an hour or two) to
change the x-ray tube, realign or replace a spectrometer or compon-
ents thereof and to put the instrument back into operation.
Table I listed a comparison of the physical characteristics of the
four commercial instruments, a discussion of which follows:
Number of Elements . The smallest number of elements which can
be measured simultaneously is 14. Depending on the ultimate decision
of EPA concerning the number of critical elements to be measured,
this may or may not be adequate. A scanning spectrometer in addition
to the fixed spectrometers would make it possible to analyze for addi-
tional elements. Under certain circumstances incorporation of an
energy dispersion system would allow semiquantitative analysis of
several additional elements.
Tube Targets and Power . All of the instruments can be equipped
with an adequate selection of x-ray tubes operable at 2.5kW or higher.
Crystals . Two of the instruments use curved crystals while the
other two instruments use flat crystals. However, on the basis of the
work reported here, there does not seem to be any compelling justifi-
cation to consider one type of crystal more desirable than the other.
Detectors . All of the instruments use gas proportional counters
or their equivalent and have similar efficiencies. However, only two
instruments employ pulse amplitude discriminators on all channels.
The ability to discriminate against higher order diffraction of shorter
wavelengths is desirable in the pollution measurements because of the
wide range of concentrations encountered and because of the require-
ment to minimize the background. Therefore it seems necessary that
pulse height analyzers be used on all channels.
Data Handling . Two of the instruments use scaler circuits to read
,out the number of x-ray photons collected by each channel whereas the
other two use the detector output to charge a capacitor and then read
the integrated charge as intensity. The scaler read-out is somewhat
more readily interpreted statistically but this is relatively unimportant.
6
-------
NRL Report 7617
The capacitors do have one undesirable feature, however, namely a
limited amount of charge can be accumulated. Thus it must be decided
in advance what instrument multiplier to use so that the charge is large
enough to be easily readable but not so large that the integrator satur-
ates. Although this may be a minor difficulty in many cases, the fact
that it can be troublesome does, to some extent, indicate the des ir-
ability of using scalers.
Sample Handling . All of the instruments were modified by using a
sample cup to support the filter and reduce the scattered background,
as was illustrated in Figure 2. This is necessary in order to achieve
the detection limits listed in Table II. If EPA should implement tenta-
tive plans to use rolls of filter-paper tape or frame-mounted filters
for sampling, automatic sample handling could be expected to improve
the efficiency of the analyses. The manufacturers of all four instru-
ments have indicated a willingness to consider such an automated
sample-handling system. In fact, some of the instruments have de-
signs for sample-handling devices which can process up to 160 samples
without operator attention or, in some cases, provide for the analysis
of samples as fast as they can be prepared by automated ancillary
equipment from a process stream.
REFERENCES
lttDevelopment of X-Ray Fluorescence Spectroscopy for Elemental
Analysis of Particulate Matter in the Atmosphere and in Source
Emissions,” L. S. Birks, J. V. Gilfrich and P. G. Burkhalter,
Environmental Protection Agency Report R2-72-063, Oct. 1972.
2 L. S. Birks, X-Ray Spectrochemical Analysis , 2nd ed., Wiley-
Interscience, New York, 1969, pp. 80-82.
7
-------
UNCLASSIFIED
Secunt’ , Classification
DOC UME NT CONTROL DATA R & D
(Srcurity classilicaS,on of t,Uc. bodj o abstract tind ,ndexui.t annotation must be entered s,hcn the overall report is classified)
I ORIGINATING ACTIVITY (Corporate author) — So. REPORT SECURITY CLASSIFICATION
Naval Research Laboratory Unclassified
Washington, D. C. 20375 Sb GROUP
3 REPORT TITLE
Development of X-Ray Fluorescence Spectroscopy for Elemental
Analysis of Particulate Matter in the Atmosphere and in Source Emissions
Phase II : Evaluation of Commercial Multiple Crystal Spectrometer Instruments
4 DESCRIPTIVE NOTES (Type of report end inclusive dates)
A final report on one phase of the problem.
5 AUTHORISI (Fir .et name, middle initial, last name)
L. S. Birks and J. V. Gilfrich
S REPORT DATE
June 15, 1973
7e TOTAL NO OF PAGES 7b NO OFREFS
14 2
Be CONTRACT OR GRANT NO
NRL PrI)’ .-r1 P04-06
b PROJECT NO
EPA Interagency Agreement No.690114
C
d
Ba ORIGINATORS REPORT NUMBERISI
NRL Report 7617
Sb OTHER REPORT NOISI (Any other numbere thai may be assIgned
this report)
10 DISTRIBUTION STATEMENT
Approved for public release; distribution unlimited.
II SUPPLEMENTARY NOTES
12 SPONSORING MILITARY ACT! VITY
Office of Research and Monitoring
U. S. Environmental Protection Agency
Washington, D C. 20460
-
iS . sTnACT
Four commercial multiple crystal spectrometer x-ray analyzers were evaluated
for use in the elemental analysis of air pollution particulate samples. Fourteen to
twenty-four elements can be measured simultaneously in these instruments. 100
second detection limits of 1 to 10 ng/cm 2 were achieved for about one-half of the
elements examined. Any one of the commercial instruments is capable of per-
forming quantitative analysis of the particulate matter filtered out of the atmosphere
or source emissions. Some actual pollution samples were analyzed in all four
instruments to demonstrate suitability.
F0RM 1473
1 NOV 65
S/N 0101.807.6801
(PAGE 1)
9
UNCLASSIFIED
Security Classafication
-------
UNCLASSIFIED
Air Pollution
Particulate Samples
X-Ray Fluorescence Analysis
Multiple Crystal X-Ray Analyzers
Security ClassUtcetIOn
4
KEY WORDI
LINK A
LINK B
LINK C
ROLE
WY
ROLE
WY
ROLE
WY
D D FORM I
NOVSIIII473 (BACK) 10
(PAGE 2)
UNCLASSIFIED
Security Classification
------- |