EPA 600 2 76-265
October 1976
Environmental Protection Technology Series
VALENCE STATES OF SULFUR IN
POLLUTION SAMPLES BY X-RAY ANALYSIS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1 Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-76-265
October 1976
VALENCE STATES OF SULFUR IN
POLLUTION SAMPLES BY X-RAY ANALYSIS
by
J. V. Gilfrich
M. C. Peckerar
L. S. Birks
Naval Research Laboratory
Washington, D. C. 20375
Interagency Agreement EPA-IA6-D6-F344
Project Officer
Jack Wagman
Emissions Measurement and Characterization Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U. S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, N. C. 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U. S. Environmental Protection Agency, and approved for publi-
cation. Approval does not signify that the contents necessarily reflect
the views and policies of the U. S. Environmental Protection Agency, nor
does mention of trade names or commercial products constitute endorsement
or recommendation for use.
n
-------
ABSTRACT
A flat single crystal spectrometer was configured to
measure the valence band x-ray spectra of various forms of
sulfur. While most different valence states showed differ-
ences in the structure of the K$ band, particular emphasis
was put on distinguishing sulfide from sulfate in samples
simulating pollution particulate collections. The relative
fraction of sulfide and sulfate in samples containing as
low as 25 yg/cm2 total sulfur can be measured with an
accuracy of about 10%.
111
-------
INTRODUCTION
The atomic number range from 11 (Na) to 17 (Cl) in-
cludes elements where the KB x-ray emission results from
the transition of valence electrons (from the M shell) to
fill a vacancy in the K shell. As such it displays struc-
ture associated with the chemical combination of the
element. The chemical form of sulfur present in pollution
samples is important in the data interpretation of the
elemental analysis of such samples. The desirability of a
routine method for identifying the valence state of the
sulfur is clearly evident/ particularly if it could be
carried out concurrently with the elemental x-ray analysis.
This fine structure in the valence-band x-ray spectra
has been studied for many years because it is a sensitive
measure of chemical combination. Such measurements usually
require the high resolution of a double crystal spectrom-
eter,-'" By the choice of an appropriate crystal a single-
crystal spectrometer can be configured so as to achieve
resolution adequate to perform measurements of this type.
In this report we illustrate a particular configura-
tion of a single-crystal, conventional x-ray fluorescence
analyzer with the best practical resolution to measure the
structure of the K3 emission from various forms of sulfur.
Further we demonstrate the ability to observe this struc-
ture in specimens containing sulfur at levels which are
consistent with those which occur in many types of particu-
late pollution samples. But perhaps of the most importance,
data are presented which show that, at total sulfur con-
2
centrations on the filter as low as ^ 25 yg/cm , sulfate
-------
and sulfide forms can be distinguished from one another,
and their relative amounts determined with reasonable
accuracy.
-------
SUMMARY
A conventional wavelength-dispersion x-ray fluorescence
spectrometer was used to identify and measure the sulfide
and sulfate forms of sulfur at concentrations approximating
pollution samples. It was necessary to substitute a NaCl
crystal for the usual crystal in order to obtain enough
resolution to distinguish the sulfide and sulfate features
in the fine structure of the S K$ line. Accuracy of the
measurement of sulfide/sulfate ratio was demonstrated to be
about 10%.
-------
CONCLUSIONS
The results of the measurements reported here indicate
that sulfur in air particulate samples can be identified as
the sulfate or sulfide form; if both forms are present, their
proportions can be determined quantitatively. Measurements
made using a flat, single crystal, high resolution x-ray
spectrometer are capable of about 10% accuracy in delineating
the distribution between the two forms, at concentrations
which might be expected in air particulate samples.
Extension of this technique to other elements is, of
course, possible. The measurement of the valence state of P
and Cl, immediate neighbors of S in the periodic table, should
be analogous to what is reported here. Phosphorous has been
2
studied in a variety of compounds using a photographic
Johann-type spectrograph; an application to chlorine has also
been reported. While the experimental details in these
references differ from the technique reported here, the
principles are the same.
On the other hand, the valence band transitions in some
elements may require more specialized instrumentation. The
L-lines of atomic numbers 24 to 30 (Cr to Cu), for instance,
are low energy, less than 1 keV, and are difficult to measure
4
at best. However, Henke and Taniguchi have shown that it is
possible to make measurements even as low in energy as 100 eV,
using a more or less conventional single crystal spectrometer
with a specially-designed low energy excitation source. The
intermediate energy region (100 eV to 1-2 keV) can be directly
attacked by such an approach.
-------
RECOMMENDATIONS
This program was limited to the study of sulfide and
sulfate forms of sulfur in pollution samples. The results
indicate the potential of this technique. It is recommended
that the technique be implemented for the examination of
particulate pollution samples.
In order to carry out the implementation, the Naval
Research Laboratory (NRL) has agreed to design and construct
for the Environmental Protection Agency (EPA) a single crystal
spectrometer channel for the multichannel x-ray analyzer
presently in use at the EPA laboratories in Research Triangle
Park, North Carolina. This component will be suited specifi-
cally to make the measurements appropriate to the determination
of the valence state of sulfur during the period when the
analyzer is performing the determination of the elemental
composition of the sample.
In addition, NRL is presently constructing for EPA a
compact x-ray sulfur analyzer intended for on-site use.
Because of the success of the program to measure valence,
this sulfur analyzer has been redesigned to enable it to
perform the valence measurements as well as the analysis for
total sulfur.
-------
EXPERIMENTAL
EQUIPMENT
The single-crystal spectrometer used for this work is a
standard x-ray fluorescence analyzer (Philips PW 1410). No
modifications were necessary except to insure that the primary
beam from the x-ray tube can be scattered into the spectrometer
only by the sample itself; this is our standard configuration
so that the background is minimized when analyzing "thin"
samples. The instrument was operated manually although
automated instruments of this type would be more efficient.
Operating conditions are listed in Table 1.
Table 1. OPERATING CONDITIONS
X-ray tube:
Crystal:
Detector:
Collimator:
Vacuum:
Cr target; 45 kV, 45 mA.
Freshly cleaved (200) Nad;
2d = 5.641 A.
Gas flow proportional counter;
90% Ar, 10% CH4 at atmospheric
pressure; aluminized 6 ym
Mylar window.
A9 = 0.07°
* 100 ym Hg
SAMPLE PREPARATION
Initially, bulk samples were prepared by loading plastic
x-ray sample cups with reagent chemicals and closing the con-
tainer with 6 ym mylar. Although the primary interest was in
differentiating between sulfide and sulfate, other forms such
-------
as elementary sulfur, sulfite and organic sulfur (thiourea)
were prepared. The K£ spectra of these bulk samples indicated
significant differences in fine structure (as will be shown
later) and suggested that micro-samples should be prepared.
Since the sulfide and sulfate compounds which were pre-
pared in bulk were water soluble, the deposition of solutions
on filter paper seemed like a reasonable method of preparing
micro-samples. Solutions of Na2S and Na2SC>4 were deposited
on Whatman filter paper at concentrations such that the mass
loading on the filter was approximately 50 yg of sulfur per
square centimeter. X-ray measurements of these filters re-
produced the spectra from the bulk samples of those compounds.
Initially the mixtures of the sulfide and sulfate solutions
were prepared at different ratios to produce intermediate
o
samples having a total sulfur content of about 50 yg/cm .
X-ray results were disappointing in that these mixtures
seemed unstable; the spectra were not the mixture of the two
forms as expected. Similarly, precipitating one of the forms
of sulfur and depositing the other form as a solution onto the
filter containing the precipitate did not produce adequate
mixtures. In this case the solution deposited non-uniformly.
The method of preparation which produced satisfactory
mixed samples in evaluating the proposed technique involved
separate precipitation of each form of sulfur and sequential
filtering onto Millipore. The sulfide was precipitated with
ii -f"4-
a Cd solution and the sulfate with Ba . The use of rela-
tively high atomic number cations has two beneficial effects:
1.) the samples effectively simulated actual collections where
significant quantities of heavy elements may be present; and
2.) the Cd and Ba elemental concentration can be measured by
x-ray fluorescence to confirm the sulfide-to-sulfate ratio
present in the sample. One other type of sample was prepared
at the 50-50 ratio: Filters containing pure end members of CdS
2
and BaSO., at about 40 ug S/cm each, were cut in half and re-
assembled between 6 ym Mylar sheets to form two mixed samples.
7
-------
SPECTRA
Figure 1 shows the K3 spectra from bulk samples. The
different forms of sulfur can be distinguished from one an-
other by the position of the main Kg peak, the degree of
asymmetry of this peak, and the presence of absence of peaks
on the high or low angle side of the main peak. The most
striking sifference occurs between the sulfide and sulfate
where the sulfate is characterized by a high-angle secondary
peak (commonly referred to as a satellite but more accurately
described as a molecular orbital feature). The fine spectral
features are in general agreement with those reported in a
recent paper by Hurley and White,^ who were concerned with
measuring the chemical form of the sulfur in bulk samples of
coal.
Figure 2 shows the Kg spectrum from a mixed sample of
o
precipitated material at a total sulfur content of 40 yg/cm
where approximately 50% of the sulfur is present as the
sulfide and the balance as sulfate. The features used in the
data interpretation are labelled "A" and "B".
-------
Q
LU
N
o:
o
L±J
I-
ELEMENTAL
S
20 ANGLE
Figure 1. K3 spectra of sulfur in various compounds. (The
2 eV bar is an estimate of the instrumental resolution with
which the measurements were made).
-------
CO
LJ
H
Z
"
BSULFIDE
"A"(SULFATE
26 ANGLE
Figure 2. 8KB spectrum of 50% S as CdS, 50% S as BaS04,
10
-------
CALIBRATION
CALCULATED CALIBRATION CURVE
A calibration curve for mixed samples can be calculated
from measurements of separate sulfate and sulfide end-members.
This calibration curve plots the ratio of Peak A to Peak B
(from Figure 2) against the weight fraction of sulfur which
is present as sulfide (or sulfate). For the pure sulfate A/B
is 0-437; for pure sulfide it is 0.048. For mixed samples
the ratio is simply
(CSulfate) (APure Sulfate*
(CSulfide) (APure Sulfide*
(BPure Sulfate)
(BPure Sulfide*
(1)
where C = weight fraction
A = intensity measured at Peak Position A
B = intensity measured at Peak Position B.
A typical calculated calibration curve is shown in Figure 3,
resulting from the calculation listed in Table 2, the end-
number intensities being averaged from three sets of data.
The curve is not linear because there is some intensity above
background at each peak position from each of the forms of
sulfur.
11
-------
0.5
0.4
0.3
A/B
0.2
O.I
X - END MEMBERS
20 40 60 80
100
SULFATE S
TOTAL S *
Figure 3. Calculated calibration curve.
12
-------
Table 2. CALCULATION OF CALIBRATION CURVE
H
LO
Intensities (c/100 s)
Sulfate KB* Sulfide K3
Distribution
of Sulfur Sulfide Sulfate S Sulfide Sulfate £
100% as CdS 1651 0 1651 34286 0 34286
(Measured)
100% as BaS04 0 6412 6412 0 14681 14681
(Measured)
Ratio
(Sulfate K$V
Sulfide K3)
0.048
0.437
75% as CdS
25% as BaSO,
} 1238 1603 2841 25714 3670 29384
0.097
50% as CdS
50% as BaSO,
} 826 3206 4032 17143 7340 24483
0.165
25% as CdS
75% as BaSO,
} 412 4809 5222 8572 11011 19583
0.267
-------
0.5
0.4
0.3
A/B
0.2
0.1
X- END MEMBER
A- HALVED END-MEMBER
O - SEQUENTIAL PRECIPITATES
20
40 60
SULFATE S
TOTAL S
80
100
Figure 4. Intermediate composition measurements.
14
-------
RESULTS AND DISCUSSION
The results of the measurements on the intermediate
compositions are shown on Figure 4, the two different types
of samples being distinguished by symbol. Table 3 lists the
numerical results of the measurements obtained by using the
calibration curve to convert the intensity ratios to per-
cent sulfate. One would expect that the use of intensity
ratios would make these measurements independent of the
total sulfur content. A comparison of samples number 2 and
3 confirms this premise within the limits of this small
range. Sample number 3, in fact, shows a more accurate
7 2
result at 26 yg/cnr total sulfur than number 2 at 43 yg/cm
in spite of the calibration curve being prepared from end-
o
numbers at the 40 pg/cin level.
In the preparation of the sequentially filtered pre-
cipitates there was obvious non-uniformity on some filters;
these samples were not used. Some of the uncertainty in
the measurements undoubtedly arises from the non-uniformity
in those samples where it was not obvious. The "halved end-
member" samples (Nos. 5 and 6), which should not be plagued
by this non-uniformity problem, give the best results of
all. If the worst result (No. 2) is considered an outlier
and dropped, the relative standard deviation (if meaningful
for these few samples) improves from 11.5 to 7.5%.
15
-------
Table 3. RESULTS FOR INTERMEDIATE COMPOSITION
cn
Sample
No.
Total
S
(ug/cm^) as
Sequentially Filtered
1
2
3
4
Halved
5
6
42
43
26
46
End Members
36
43
% of Total
Sulfide as
Precipitates
74
57
47
31
53
48
sa
Sulfate .
-
26
43
53
69
47
52
% of Total Sb
as Sulfide as Sulfate
72.5 27.5
48 52
41 59
27 73
55 45
46 54
% Difference
in measuring
Sulfate
5.3
21
11
5.8
4.2
3.8
R.S.D.= 11
.5
on XRF measurement for Cd and Ba.
3Based on A/B measurement and use of calibration curve in Figure 2.
-------
REFERENCES
1. Goshi, Y. and K. Yanagase. J. Fuel Ass. (Japan).
52 (No. 560):1973.
2. Fichter, M. Spectrochim. Acta, Part B, 30:417, 1975.
3. Whitehead, H. C. and G. Anderman. 167th National
Meeting, ACS, Los Angeles, Calif., Mar. 31 to Apr. 5,
1974.
4. Henke, B. L. and K. Taniguchi. Valence Band Spectros-
copy in the Ultrasoft X-Ray Region (50 to 100 A). In:
Advances in X-Ray Analysis, Vol. 19, Gould, R. W. et al.
(ed.) Dubuque, Kendall/Hunt Publ. Co., 1976, p. 627-
641.
5. Birks, L. S., J. V. Gilfrich, and P. G. Burkhalter.
"Development of X-Ray Fluorescence Spectroscopy for
Elemental Analysis of Particulate Matter in the At-
mosphere and in Source Emissions," Environmental
Protection Agency, Washington, B.C., Report No. EPA-
R2-72-063, October 1972, p. A-2-1 to A-2-5.
6. Hurley, R. G. and E. W. White. Anal. Chem. 4_6:.2234,
1974.
17
-------
TECHNICAL REPORT DATA
(Phase read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/2-76-265
2.
3. RECIPIENT'S ACCESSION>NO.
4. TITLE ANDSUBTITLE
VALLENCE STATES OF SULFUR IN
POLLUTION SAMPLES BY X-RAY ANALYSIS
5. REPORT DATE
Octohpr 1Q7fi
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)'
J. V. Gilfrich, M. C. Peckerar, and L. S. Birks
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Naval Research Laboratory
Washington, D. C. 20375
10. PROGRAM ELEMENT NO.
1AD605
11. CONTRACT/GRANT NO.
EPA-IAG-D6-F344
12. SPONSORING AGENCY NAME AND ADDRESS
13. TYPE OF REPORT AND PERIOD COVERED
Environmental Sciences Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
Research Triangle Park, N. C. 27711
Interim 6/75-6/76
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES
16. ABSTRACT
A flat single crystal spectrometer was configured to measure the
valence band x-ray spectra of various forms of sulfur. While most
different valence states showed differences in the structure of the
Kg band, particular emphasis was placed on distinguishing sulfide from
sulfate in samples simulating pollution particulate collections. The
relative fraction of sulfide and sulfate in samples containing as low
o
as 25 yg/cm total sulfur can be measured with an accuracy of about 10%.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b. IDENTIFIERS/OPEN ENDED TERMS
COS AT I Field/Group
*Air pollution
Sulfur inorganic compounds
Sulfur organic compounds
*Sulfur
*Valence
*X-ray fluorescence
3B
37B
)7C
)7D
>OF
3. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
19. SECURITY CLASS (ThisReport)
IINPI ASSTFTFD
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
UNCLASSIFIED
22. PRICE
EPA Form 2220-1 (9-73)
18
------- |