United States
Environmental Protection
Agency
Air and Energy Engineering
Research Laboratory
Research Triangle Park NC 27711
Research and Development
EPA/600/S2-85/025 May 1985
c/EPA Project Summary
Speciation of Hazardous
Inorganic Compounds by
Fourier Transform Infrared
(FTIR) Spectroscopy
Tobias R. Acciani and Eugene A. Burns
.V
Recent advances in instrumentation
for Fourier transform infrared (FTIR)
speetroscopy have provided the oppor-
tunity to extend the application of
infrared methodology to characterize
hazardous inorganic compounds in solid
inorganic emissions. The improved in-
strumentation, consisting of multiple-
scan computer enhancement methods,
has increased the signal-to-noise ratio
so that low concentrations of infrared
active bands are now measurable. Also,
the ability to use spectral subtraction
methods, together with improved infra-
red detectors, permits trace level infra-
red analyses, where a few years ago this
was not possible. Increasing the Infrared
spectral scanning range to the far infra-
red (i.e., down to 100 cm'1) permits
characterization of metal-oxygen bend-
ing modes. The result of this enhanced
technology allows the speciation of the
hazardous Inorganic compounds.
In this task, gaps in the available,
relevant inorganic compound spectral
data base were identified and a series of
reference infrared spectra ware ob-
tained. These reference spectra served
as a basis for characterizing hazardous
solid inorganic compound emissions
collected from a metal smelting opera-
tion, a steel mill, and a fluidized-bed
desulfurization control process. Special
studies were conducted to determine
the effect of water on far infrared
spectra (600 to 100 cm'1), and typical
environmental samples were examined
using techniques developed in this
effort. In addition, studies were under-
taken (1) to determine the effect of
matrix interferences on the pollutant
spectra, and (2) to examine potential
matrix iaolation techniques consisting
of spectral subtraction methods and
chemical extraction approaches. Fur-
ther, the use of attenuated total reflec-
tance (ATR) was studied as a way to
measure potentially hazardous inorgan-
ic compounds which accumulate on the
surface of fly ash and as a technique to
facilitate sample preparation.
This study demonstrated that FTIR
was a useful analytical chemistry tech-
nique for inorganic compound specia-
tion. The mid infrared region provided
information regarding the anion groups,
and the far infrared region provided
information regarding the metal and Its
oxidation state, principally from the
metal oxygen bending modes. A chart
showing the absorption frequency as a
function of 19 inorganic compounds
over the mid and far infrared ranges was
prepared and found useful in making
structural assignments.
Using FTIR instrumentation, it ap-
pears that infrared speetroscopy will be
a valuable tool for inorganic compound
characterization, just as conventional
infrared speetroscopy has been for
characterization of organic compounds.
This Pro/act Summary was developed
by EPA 's Air and Energy Engineering
Research Laboratory, Research Triangle
Park. NC. to announce key findings of
the research project that is fully docu-
mented in a separate report of the same
title (see Project Report ordering Infor-
mation at back}.
-------
Introduction
Recent studies have shown that the
toxiclty of fly ash, other coal-fired utility
waste, and high temperature metal proc-
essing emissions may be related to the
trace elemental composition and specific
compounds on the surface of the waste
and within the waste. The compound
configuration is important for two rea-
sons:
1. The Inherent toxicity is a function of
the specific compound, and
2. The ability of a material to be
mobilized (e.g., by a leaching proc-
ess) and become available and
introduced into the environment is
dependent on the compound config-
uration.
An example of the variability in toxicity for
five mercury compounds shows that,
while the mercury content varies by only
0.2, the calculated LD8o varies by a factor
of over 16. In addition to the toxicity
variation among these compounds, there
are considerable differences in solubility
and, hence, the potential for the metal to
become mobilized, which is also com-
pound-assemblage(molecular configura-
tion) related.
In a combustion process, it has been
shown that trace metals concentrate on
the surface of particles. Consequently,
the concentration of the trace metal
available is much higher during the initial
contact and not directly related to the
average concentration of the trace metal
in the bulk amorphous fly ash. Conse-
quently, to obtain and understand the
environmental impact of fossil fuel com-
bustion products, waste incineration, or
metal smelting processes, it is essential
to establish methodology capable of char-
acterizing the specific compound form or
molecular assemblage of the trace metals
present on the surface of the particulate
matter.
Researchers have examined X-ray
methodology and time-resolved solvent
leaching as approaches to meet this
objective; however, with limited success
because of experimental problems inher-
ent with these methods; e.g., the amor-
phous chemical structure of waste solid
particles and extensive characterization
times removing just one compound at a
time, respectively. However, infrared
spectroscopy permits the opportunity to
identify and characterize non-crystalline
inorganic compounds and the use of
Fourier transform infrared (FTIR) spectro-
scopy extends the application of infrared
to qualitate the presence of specially
active compounds at low concentrations
through (1) enhanced signal-to-noise
' capability, and (2) the facile ability to use
spectral subtraction methods. The devel-
opment of improved FTIR instrumentation
permits measuring infrared vibration and
bending modes of metal-oxygen bonds in
an active far infrared range extending to
lOOcm'1.
The objective of this effort was to
evaluate the applicability of FTIR spectro-
scopy for characterization of toxic com-
pounds at low to trace concentration
levels in hazardous inorganic wastes. The
infrared spectral range for this study was
4000 to 100 cm'1. Potassium bromide
pellet sample preparation techniques
were used to measure in the spectral
range of 4000 to 400 cm'1; polyethylene
sample preparation techniques were used
for the spectral range of 500 to 100 cm"1.
A systematic study was undertaken to
identify gaps in available, relevant in-
organic compound spectra and to gener-
ate needed reference infrared spectra.
Further studies were undertaken to as-
sess the effect of water on far infrared
spectra andto characterize environmental
samples for trace level compounds for
solid waste collected from ducted gaseous
emissions from a copper smelter, a steel
mill, and a fluidized-bed combustion
control process.
Results
Results are discussed in terms of data
gaps, experimental efforts, effects of
water, and characterization of environ-
mental samples.
Data Gaps in Available,
Relevant Inorganic Compound
Spectra
An extensive literature review was
undertaken to determine how infrared
spectroscopy had been applied to the
characterization of inorganic matter. This
review centered on the lack of direct
application of infrared spectroscopy, both
inthemidand(especially)thefar infrared
regions. During the review, a limited
amount of inorganic compound spectra
were available, but the far infrared spec-
tra of inorganic compounds were almost
nonexistent. Based on the review, a list of
96 environmentally significant com-
pounds whose spectra were not available
was compiled. Fourteen of these com-
pounds were not commercially available.
Of the 82 available compounds which did
not have valid FTIR spectra in the 4000 to
1000 cm'1 spectral range, 19 were sel-
ected for generation of infrared spectra.
Nine of the compounds selected wera
sodium salts of nitrate, nitrite, chromate.l
dichromate, metavanadate, orthovana-
date, selenate, sulfate, and sulfite. Other
compounds selected were ammonium
nitrate and sulfate, calcium and lead
sulfates, and differing oxidation states for
arsenic (III) and (V) oxides, mercury (I) and
(II) chlorides, and mercury (I) and (III)
sulfates. The discrete characterization of
various anions was facilitated because
sodium was the central metal group
cation for most of these compounds.
Consequently, its influence over the
metal/non-metal-oxygen bending and
stretching frequencies observed would
be held constant in direct comparison
with the spectra; its influence was ex-
pected to be second order of importance.
In general, the mid infrared region was
used to characterize the anion assem-
blage groups because they consisted of
two or more different atoms. For most
inorganic compounds, ionic bonds exist
between the metal and the anion assem-
blage; these bonds are weak mid infrared
absorbers. Ionic bonds usually absorb in
the far infrared region and are observed
as a lattice-type bond vibration. For
central metal identificaiton, on the other
hand, the far infrared region is generally
used. The effect of metal oxidation states
and the related absorption frequencies!
arising from metal-oxygen bending modes
are observed in the far infrared region.
Experimental
In this project the Digilab Model FTS-
15C FourierTransform Infrared Spectrom-
eter was used. This device is an automatic
ratio recording infrared spectrophoto-
meter with resolution of 0.25 cm'1. A
Model 396 Michelson Interferometer with
a KBr beamsplitter for mid infrared range
and Mylar beamsplitters for far infrared
was used. In the mid infrared range(4000
- 400 cm'1), a dual beam optics and
triglycine sulfate detector was used, and
in the far infrared range (500 -10 cm"1), a
filtered triglycine sulfate detector was
used.
The reference spectra were generated
using standard KBr pellet pressing meth-
odology with a concentration of reference
compound of 5 mg/g KBr pellets. In the
far infrared region, a polyethylene pellet
was prepared using a concentration of 20
mg/g polyethylene. To assist in far infra-
red sample preparation, polyethylene was
purified by extraction in hot xylene fol-
lowed by reprecipitation with ethanol.
The filtered product was washed with
ethanol, dried, and then sized to
-------
20-mesh sieve. Two mg of pure sample
nd 100 mg of the purified polyethylene
were ground into flakes (using a Wig-L-
Bug) and then transferred to a special
aluminum die for hot pressing at 100°C.
A 0.04-cm spacer was used to ensure a
uniform optical path thickness.
Figure 1 summarizes the mid and far
infrared absorbence frequency character-
istics for the 19 organic compounds
studied and for water over the range
1500 to 100 cm'1. This chart permits
qualitative characterization among these
compounds by examining their spectral
absorbences. Detailed discussions of the
differences among these spectra are
provided in the full report.
Effect of Water
The far infrared spectra of inorganic
compounds are greatly affected by the
presence of water. Water alters the
spectrum because the crystal structure of
the compound changes from one form to
another, depending on the extent of
hydration. The spectra of water in calcium
sulfate as a function of the three config-
urations— CaSO< • 2 HgO (dihydrate),
CaSO« • Vi H20 (hemihydrate), and CaS04
(anhydrous)—showed significant differ-
ences in the 400 to 100 cm'1 region. In
addition to water in the sample, water
vapor in the optical beam of the spec-
trometer is also a significant concern. To
this end, purging the spectrometer with
inert gas was found to effectively elimi-
nate this problem. Use of a spectrometer
capable of evacuation of the optical
compartment would be highly desirable
to minimize the time required to purge the
last vestiges of water from the spectrom-
eter. A simple distinguishing fact about
atmospheric water bands in the far
infrared region is that they are very sharp
and narrow, where normal infrared ab-
sorbence bands (in the sample) tend to be
much broader.
Characterization of
Environmental Samples for
Trace Level Compounds
Environmental samples were obtained
from EPA, and a preliminary interpreta-
tion of the spectra (recorded in terms of
inorganic compound species) was made.
The samples were from ducted gaseous
emissions from (1) a copper smelter, (2) a
steel mill, and(3)afiuidized-bedcombus-
tion (FBC) control process. In general, the
inorganic species were identified by
comparison with the active frequencies
"or the compounds shown in Figure 1.
Compound
HgCI
HgCIt
HgtfO*
HgSO*
AstO»
ASA,
Na£r&,
Na£rOt
(NHJiSO*
NHtNOt
/V»aS04
/VaaSOs
PbSO*
Na£eO*
Na3VOt
NaVO,
NaNOt
NaNOi
CaSO<-2H&
H0
15t
Frequency, cm"1
1500
/ >139°
C
c
a
70
3
1
P
J
1
C
P
I
131.
D
P
70
Mid Infrared
1100 900 700
£
C
C
d
c
i
3
I
at
77
3
]
mm
mm
p
]
I
3
I
00
q
<
n
ail
D
0
1
a
9C
it
ii
i
c
1
a
10
?
b
ID
mmm
mm*
C
'
o
c
7C
Ql
C
C
1
II
a
0
i
>0
1
500 450
C
ID
1
3
II
BC
I
K
C
C
q
d
d
c
a
c
c
c
c
d
c
10
b
3
1
3
p
p
P
b
p
p
i
3
3
3
3
4t
}
a
0
400
a
D
a
41
C
c
a
>o
Far Infrared
3SO 300 250
=1
C
D
a
31
3
c
D
C
>o
i
tm^
mmt
D
31
3
D
C
70
|__
C
C
c
c
mmm
26
3
MH
•••
C
200
D
C
mmm
m
C2_J
m
c
c
mmm
3
mmm
C
3
0
LJ
•••
)
3
mmm
•M
mmm
mmm
3
={
LJ
mmm
700 1
M ° ^
3
3
I
J
]
f
d
I
J
C
3
a
3
200
U
••
1
\
C
I
D
•••
a
3
3
mmm
ISO 101
Figure 1. Mid end fer infrared absorbence frequency characteristics for inorganic compounds
study.
The mid infrared spectrum of the
particulate matter collected at a copper
smelter indicated the presence of sulfate
at 1100 cm"1 and (possibly) a halogen at
600 cm'1. The far infrared spectra clearly
identified the presence of arsenic (III)
oxide. Examination of the mid infrared
spectrum of the sample also validated the
presence of arsenic (III) oxide and sulfate.
The mid infrared spectrum of a steel
mill sample confirmed the presence of
nitrate or ammonium at 1400 cm'1 and
sulfate or silicon at 1100 cm'1. The far
infrared spectrum of the steel mill sample
did not provide any useful information.
The mid infrared spectrum of the FBC
sample indicated the presence of sulfate
at 1100 cm'1. A comparison between the
mid infrared spectrum of the FBC sample
and a reference calcium sulfate dihydrate
spectrum showed similarities at different
specific frequencies. The far infrared
spectrum did not provide any supplemen-
tal information.
Additional efforts were undertaken to
isolate the matrices of these samples
(either by FTIR spectral subtraction meth-
ods or chemical treatment), and also to
examine the usefulness of attenuated
total reflectance (ATR) methodology for
assisting in characterizing complex envi-
ronmental samples. For the samples
examined, spectral subtraction was use-
ful only in characterizing smelter emis-
sions. Once again, a clear-cut arsenic (III)
oxide content was identified. The spectra
-------
of the other two samples were too general
for use of spectral subtraction because of
a complex matrix system.
The alternative approach for assisting
and isolating the spectrum of pollutants
in environmental samples was by chem-
ical treatment of the samples to solubilize
either the pollutant or the matrix. The key
step was to achieve a selective difference
in the stabilization of either component.
It was found that water leaching did not
cause the desired differential soluble ratio
for the three environmental samples. Acid
extraction did not assist in further char-
acterization of the smelter sample; how-
ever, the presence of calcium sulfate in
the FBC sample was confirmed, and
further information regarding the steel
mill sample was provided from the far
infrared spectra, indicating the presece of
ferric nitrate.
Attenuated total reflectance (ATR) was
used to discern the spectra of inorganic
compounds on the surface of the partic-
ulate matter. The ATR is made possible by
differences in the refractive indices of the
sample and a KRS crystal (thallium bromo
iodide) used to obtain the attenuated total
reflectance spectra. Conventional trans-
mission infrared spectroscopy measures
the bulk composition of the sample. As a
consequence, differences between ATR
and transmission infrared spectra indi-
cate differences between the surface
composition and the bulk composition of
the sample.
A major advantage of ATR is that
sample preparation is minimized and does
not introduce impurities (e.g., water) that
may be present in infrared sample matri-.
ces. Sample preparation for the ATR
experiments in this project consisted of
placing a layer of powdered sample on
each side of the KRS crystal. In examining
the three environmental samples, the
ATR and transmission spectra were not
found to be significant, indicating compar-
able surface and bulk composition of the
collected sorbed water.
Conclusions
FTIR spectroscopy was shown to be a
useful analytical chemistry technique for
inorganic compound speciation. The mid
infrared region provided informaion re-
garding ionic radicals and assemblages,
and the far infrared region provided
information about the metal and its
oxidation state, principally from the metal-
oxygen bending modes. This project clear-
ly demonstrated the feasibility of FTIR for
qualitative characterization of inorganic
compounds. It appears that FTIR will be a
valuable tool for inorganic compound
characterization, just as conventional
infrared spectroscopy has served for
many years to characterize organic com-
pounds: inorganic compounds have char-
acteristic fingerprint infrared spectra
which are similar to the fingerprint region
for organic compounds.
Conclusions of the project include:
1. Little information is available in the
literature regarding FTIR and its
application to inorganic compound
speciation.
2. Of the 19 compounds evaluated,
each had its own characteristic
infrared spectrum.
3. A useful compound identification
chart (Figure 1) was prepared show-
ing a set of absorption frequencies
for each inorganic compound stud-
ied. This chart was used in making
structural assignments based on a
set of absorbence frequencies.
4. Incorporation of water (in the form
of water of hydration) modified the
far infrared spectrum of inorganic
compounds because of its change
of molecular structure.
5. Sample matrix interfering absorp-
tions made quantitation of the
pollutant in the sample difficult
unless its concentration was at
levels of 1 percent or higher.
6. The feasibility of FTIR to identify
toxic inorganic compounds in envi-
ronmental samples from smelter
flue gases, fluidized-bed combus-
tion effluent, and steel mill effluent
particulate matter was demonstrat-
ed.
7. Attenuated total reflectance (ATR)
permits the measurement of dif-
ferences between surface composi-
tion particles and their bulk com-
position.
Recommendations
Additional information that will extend
the utility of infrared characterization of
environmental samples includes: (1) in-
creasing the library of available FTIR
reference spectra, (2) further verifying
the applicability of FTIR in characterizing
environmental samples, and (3) develop-
ing improved methodology for character-
izing environmental sample matrix mater-
ials and concentrating toxic pollutants in
environmental samples. More reference
spectra will permit the development of
expanded correlation tables to cover the
total range of probable inorganic com-
pounds that may be emitted from conrv«
mercial manufacturing, power conven
sion, and hazardous waste treatment
methods (e.g., incineration).
Improved sample preparation tech-
niques will increase the differential sepa-
ration/isolation of the pollutant to be
measured from its matrix, by concentrat-
ing either the matrix or the pollutant.
Once the background matrix is identified,
usefulness of FTIR will increase signif-
icantly as an analytical tool. If a back-
ground matrix cannot be identified, only
compounds present in the environmental
sample at levels of 1 percent and above
can be identified readily.
The utility of FTIR for inorganic com-
pound characterization in environmental
samples can be enhanced by incorporat-
ing a pollutant concentration step. Qual-
itative analysis depends on the type of
material present in the sample; some
compounds are weak infrared absorbers
and can be masked out by the matrix,
while other compounds have strong
absorptions that can be readily identified.
For characterization analysis to be suc-
cessful, the chemical nature of the matrix
has to be identified. Special sample
preparation techniques must be employed
both to (1) facilitate the identification of
the matrix material, and (2) concentrate^
the pollutant to be identified. All sample!
preparation techniques must emphasize
concentration of the pollutant to increase
sensitivity.
The findings of this effort clearly show
the feasibility of FTIR for characterizing
inorganic structural compound assem-
blages. The partial success in identifying
pollutants in environmental samples rec-
ommends (1) further study using addi-
tional environmental samples, (2) new
approaches for selective differentiation of
the pollutant and its matrix, and (3)
examination of a wider range of potential
compounds.
-------
-------
-------
-------
T. AccianiandE. Bums are with Systems, Science andSoftware(S-Cubed), La Jo/la,
CA 92038.
Frank £. Briden is the EPA Project Officer (see below).
The complete report, entitled "Speciation of Hazardous Inorganic Compounds by
Fourier Transform Infrared (FTIR) Spectroscopy," (Order No. PB85-188 969/A S;
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:
Air and Energy Engineering Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
* U.S. GOVERNMENT PRINTING OFFICE: 1986-559-018/27092
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
OC00329 PS
U S ENVIR PROTECTION AGENCY
REGION 5 LIBRARY
230 S DEAPBCRN STREET
CHICAGO IL £Q<$Q<
------- |