United States Industrial Environmental Research EPA-600/7-79-151
Environmental Protection Laboratory July 1979
Agency Research Triangle Park NC 27711
Chemical Analysis of Wet
Scrubbers Utilizing
Ion Chromatography
Interagency
Energy/Environment
R&D Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, US. Environmental
Protection Agency, have been grouped into nine series These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields
The nine series are
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161
-------
EPA-600/7-79-151
July 1979
Chemical Analysis of Wet Scrubbers
Utilizing Ion Chromatography
by
Tobias R. Acciani and Ray F. Maddalone
TRW Defense and Space Systems Group
One Space Park
Redondo Beach, California 90278
Contract No. 68-02-2165
Task No. 214
Program Element No. INE624
EPA Project Officer: Frank E. Briden
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
US. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
TABLE OF CONTENTS
PAGE
FIGURES iv
TABLES v
1. INTRODUCTION 1
2. SAMPLE COLLECTION PLAN 3
2.1 Type of Scrubber 3
2.1.1 Lime/Limestone Wet Scrubber 3
2.1.1.1 Chemistry of Lime/Limestone
Scrubbing 3
2.1.1.2 Scrubber Design 4
2.1.1.3 Sampling Points and Species of
Interest 4
2.1.2 Dual Alkali Scrubber 7
2.1.2.1 Chemistry of Dual Alkali Scrubbing ... 7
2.1.2.2 Scrubber Design 8
2.1.2.3 Sampling Points and Species of
Interest 10
2.2 Sampling Collection (Lime/Limestone and Dual Alkali). . . 13
2.2.1 Sample Acquisition 13
2.2.2 On Site Solution and Slurry Sample Treatment ... 13
2.2.3 Solution and Slurry Sample Size Requirement ... 14
2.2.4 Solid Samples 17
2.3 Sample Presc-rvation 17
2.3.1 Procedures for Sample Preservation 19
3. ANALYTICAL PROCEDURES 2l
3.1 Introduction 21
3.1.1 Column Packings 22
3.1.2 Practical Ion Chromatography 24
3.2 Ion Chromatography - Definitions 27
3.2.1 Retention Time 27
3.2.2 Resolution 28
3.2.3 Column Capacity - Overloading 28
3.2.4 Temperature 31
3.2.5 Pre-columns 31
3.2.6 Sample Filtration 31
11
-------
TABLE OF CONTENTS (Continued)
PAGE
3.3 Solid Sample Preparation 32
3.4 Cation Analysis 32
3.4.1 Sodium and Potassium Analysis 34
3.4.2 Magnesium and Calcium Analysis 34
3.5 Anion Analysis 35
3.5.1 Chloride, TOS and Sulfate Analysis 36
3.5.2 Sulfur and Nitrate Analysis 35
3.5.3 Carbonate Analysis 36
3.5.4 Hydroxide Analysis 36
4. QUALITATIVE ANALYSIS 38
4.1 Retention Time Tables 38
5. QUANTITATIVE ANALYSIS 40
5.1 Determination of Sample Concentration 40
5.2 Verification of Precision and Accuracy of
Analytical Scheme 48
6. REFERENCES 55
iii
-------
FIGURES
Number Page
1. Schematic Flow Diagram of the Venturi/Spray Tower
Scrubber at the EPA/TVA Shawnee Limestone Test
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
Breakdown of Species of Interest and Process Stream
Location for the Venturi/Spray Tower Scrubber
Schematic Flow Diagram of a Dual Alkali Pilot Plant . . .
Breakdown of Species of Interest and Process Stream
Location for a Dual Alkali Scrubber
Pressure Filtration Apparatus .
On-Line Liquid Slurry Filtration System
Plot of Species Concentration vs Time (Immediate
Sample Preservation) -.
Plot of Species Concentrations vs Time (no Immediate
Sample Preservation Required)
Structure of Varian Aerograph Anion and Cation
Exchange Resin
Cross Section of a Macroreticular and Pellicular
Resins
Use of Peroxide to Resolve the Peak Overlap of
Nitrate/Sulfite
Ion Chromatogram of Sulfite, Sulfite and Peroxide, and
Nitrate
Example of Peak Height Measurements
Calibration Curve for Chloride Ion Analysis
Example of Quantification by Peak Height and Peak Area . .
Ion Chromatogram of the Dionex Anion Standard Solution . .
An Illustration of the Nonlinear Ion Chromatographic
Detector Response
Calibration Curve for Sulfate Ion Analysis
Ion Chromatogram of the Magnesium/Calcium Analysis ....
6
9
11
. 15
16
18
18
23
25
29
,30
41
42
43
45
46
49
52
iv
-------
FIGURES (Continued)
Number Page
20. Ion Chromatogram of the Sodium Analysis 53
21. Ion Chromatogram of the Chloride/Sulfate Analysis 54
TABLES
1. Recommendations for Solution Preservation 20
2. Extraction Approaches for Solid Samples 33
3. Retention Time of Various Ions Found in Wet Scrubbers ... 39
4. Precision of Peak Height Measurements 47
5. Direct Comparison Between TVA and TRW Sample Analysis
for the Venturi Spray Tower Scrubber 50
-------
1.0 INTRODUCTION
This manual has been prepared for the Process Measurement Branch of
the Industrial Environmental Research Laboratory of the Environmental Pro-
tection Agency, Research Triangle Park, North Carolina, as Task 48, Con-
tract No. 68-02-2165. Task 48 is devoted to chemical analysis of wet
scrubbers utilizing ion chromatography and is under the direction of EPA
Project Officer, Frank Briden.
The intended use of this manual is to provide the user with enough
background to develop a sampling and analysis program for a wet scrubber
utilizing ion chromatography as the main analytical technique. The first
part of the manual describes a sampling program for two different types of
wet scrubbers. The second part evaluates the use of ion chromatography to
analyze wet scrubber samples.
The two wet scrubbers described in the sampling program are the
venturi/spray tower (limestone) scrubber which is part of the EPA/TVA
Shawnee Test Facility, Paducah, Kentucky and the Arthur D. Little dual
alkali pilot scrubber. These two wet scrubbers were selected for inclusion
here because the Shawnee scrubber is representative of the current tech-
nology and the dual alkali scrubber is representative of future technology.
The sampling section is broken down into three subdivisions: scrubber
type, sample collection and sample preservation.
The analysis section of this manual describes the theory and practice
of ion chromatography as applied to wet scrubber samples. The analysis
section consists of three subdivisions: analytical procedures, qualitative
analysis, and quantitative analysis. The analytical procedures subdivi-
sions consists of the anion and cation and analysis scheme for wet scrubber
samples, the theory of ion chromatography, and current problem areas
involving the use of ion chromatography for wet scrubber related samples.
The qualitative analysis subdivision describes the use of retention time
to identify the various ions found in wet scrubber samples. The quantita-
tive analysis subdivision discusses the method of peak height and calibra-
tion curves for correlation of the ion chromatographic detector response
-------
to sample concentration. Also, included in this subdivision is the veri-
fication of the precision and accuracy of the analytical scheme presented
in the analytical procedures section.
-------
2. SAMPLE COLLECTION PLAN
In order to develop a successful sample collection plan for a lime/
limestone or a dual alkali wet scrubber, the basic concepts of flue gas
scrubbing must be understood. This section discusses the sampling criteria
necessary for collecting a representative sample from a wet scrubber. The
sample collection plan is broken down into three subdivisions: the scrub-
ber type, sample collection, and sample preservation. The subsection on
the scrubber type will discuss lime/limestone and dual alkali wet scrubbers.
The subsections on sample collection and preservation will discuss the
techniques employed which are independent of the type of scrubber.
2.1 TYPE OF SCRUBBER
This section of the manual is written so that the reader will have a
clear understanding of the concepts of scrubbing flue gas by the lime/
limestone or dual alkali systems. Each scrubber type is discussed indi-
vidually with regard to: chemistry, scrubber design, sampling points, and
species of interest for control of the scrubber.
2.1.1 Lime/Limestone Wet Scrubber
2.1.1.1 Chemistry of Lime/Limestone Scrubbing --
The Shawnee venturi/spray tower wet scrubber is a representative
example of most lime/limestone wet scrubbers. The term "lime/limestone"
scrubber means that the flue gas could be scrubbed with either lime or
limestone. The overall reaction of lime with S02 is:
CaO + S02 -ğ CaS03 (l)
The absorbent slurry liquor pH for lime is maintained between 7 and
9 at the scrubber inlet and runs 4.5 and 5.5 at the outlet. The stoichi-
ometric ratio (moles of lime added per moles of S02 absorbed) ranges from
1.0 to 1.2.
The overall reaction of limestone with S02 is:
CaCO., + S00 + CaSO, + C00
32 32
-------
In limestone based scrubbers the stoichiometric ratio (moles of
limestone added per moles of S02 absorbed) is 1.1 to 1.6. The absorbent
slurry liquor pH for limestone is maintained at a much lower pH value
than lime, between 5 and 6. The reaction products for both lime and
limestone scrubbers are actually a mixture of calcium sulfite hemihydrate
(CaS03 1/2 H20) and calcium sulfate dihydrate (CaS04 ' 2H20, an oxida-
tion product). The pH value of the limestone slurry at the outlet of the
scrubber runs the same as lime, between 4.5 and 5.5.
2.1.1.2 Scrubber Design --
Figure 1 shows the schematic flow diagram of a venturi/spray tower
lime/limestone wet scrubber. Flue gas entering the scrubber passes
through the venturi (particulate removal) and upward through the spray
tower counter-current to the scrubbing liquor that is recirculated from
the effluent hold tank (D-101). The pH of the effluent liquor is adjusted
with a slaked lime slurry in the effluent hold tank. A bleed stream
from the effluent hold tank is routed through the oxidation tank and
supersaturation tank to a clarifier. The oxidation tank is the vessel
where any SCL~ ions in the scrubber effluent are oxidized to SO^" ions.
Sulfate ions and their compounds have better handling properties (less
disposal volume) than sulfite ions and their compounds. The clarifier
overflow is returned to the effluent hold tank via the fill tray
receiving tank (D-407) and process water hold tank (D-103). The filter
catch is discharged to a disposal pond.
2.1.1.3 Sampling Points and Species of Interest
The main objective for sampling a lime/limestone wet scrubber is
to monitor the chemical species in the scrubber liquid. By sampling at
key points and for certain species, the scrubber can be controlled so
that it is operated efficiently and effectively. This section describes
the sampling and analysis requirements for each process stream.
Each number on Figure 1 refers to process streams where samples
are taken. Each process stream has been classified into one of three
groups; solution, slurry, and solubles in cake. Figure 2 shows an
overall view of the types of analyses for each process stream line for
the Shawnee lime/limestone scrubber. The important species are calcium,
-------
SCRUBBER GAS
(TO REHEATER)
D-101 SPRAY TOWER EFFLUENT HOLD TANK
D-102 CLARIFIER
D-103 PROCESS WATER HOLD TANK
D-108 OXIDATION TANK
D-109 SUPERSATURATION TANK
D-407 FILL TRAY RECEIVING TANK
Figure 1. Schematic Flow Diagram of the Venturi/Spray Tower Scrubber
at the EPA/TVA Shawnee Limestone Test Facility
-------
ILIME/LIMESTONE SCRUBBER]
TYPE OF PROCESS
SOLUTION
STREAM
LOCATION
7,8,9
STREAM
METHOD OF ANALYSIS
ION CHROMATOGRAPHY
SPECIES OF INTEREST
1 . SULFATE
2. SULFITE
3. CHLORIDE
4. CARBONATE
5. CALCIUM
6. MAGNESIUM
7. SODIUM
8. POTASSIUM
1
"YPE
OF PROCESS STREAM
SLURRY
STREAM
LOCATION
1,2,3,4,5,6
METHOD OF ANALYSIS
ION CHROMATOGRAPHY
SPECIES OF INTEREST
1 . SULFATE
2. SULFITE
3. CHLORIDE
4. CARBONATE
5. NITRATE
6. CALCIUM
7. MAGNESIUM
8. SODIUM
9. POTASSIUM
TYPE OF PROCESS STREAM
SOLUBLES
IN CAKE
STREAM
LOCATION
10
METHOD OF ANALY!
ION CHROMATOGRAI
>IS
>HY
SPECIES OF INTEREST
1 . SULFATE
2. SULFITE
3. CHLORIDE
4. CARBONATE
5. NITRATE
6. CALCIUM
7. MAGNESIUM
8. SODIUM
9. POTASSIUM
Figure 2. Breakdown of Species of Interest and Process Stream Location for the Venturi/Spray Tower Scrubber
-------
sulfate, sulfite, magnesium, and chloride. The concentration of these
species is determined in all the process streams throughout the scrubber
unit. The calcium and sulfate concentration is determined for lime/
limestone utilization data. In addition, the sulfite concentration is
determined before and after the scrubber so that the scrubbing efficiency
can be determined. Magnesium is an additive for increased scrubber
efficiency so its concentration must be monitored and controlled for
maximum efficiency. Chloride reacts with magnesium and will deplete
the magnesium additive so the chloride concentration must be monitored
and kept extremely low. Also, chloride will react with the stainless
steel equipment and pipes. Chloride is the only species which is not
added to the system for scrubbing, but enters as a component of the flue
gas and river water.
2.1.2 Dual Alkali Scrubber
2.1.2.1 Chemistry of Dual Alkali Scrubbing --
A number of processes are similar and can be considered technically
dual alkali. In these processes, a soluble sodium based alkali (NaOH,
NaHC03, Na2S03, Na2CCL) absorbs S02 from the flue gas in the scrubber,
then a calcium based alkali (Ca(OH)2, CaO, CaC03) reacts with the S02 rich
scrubber effluent liquid to precipitate the insoluble CaSO., 1/2 H20,
CaS04 2H20, and regenerate the sodium based soluble alkali for recycle
to the scrubber system. This manual only considers the sodium sulfite/
lime or limestone based dual alkali process.
The overall scrubber reaction of sodium sulfite with SCL is:
Na2S03 + S02 + H20 -ğ 2NaHS03 (3)
The overall regeneration reactions are:
Ca(OH)2 + 2NaHS03 -ğ Na2S03 + CaS03 1/2 H20 + 3/2 HgO (4)
CaC03 + 2NaHS03 + 1/2 H20 + Na2S03 + CaS03 1/2 H20 + C02
-------
2.1.2.2 Scrubber Design
Interest in dual alkali scrubbing has grown because there are two
main difficulties with lime/limestone scrubbers: scaling (CaSO^ plating
on interior walls which causes plugging of scrubber), and disposal of
CaS03 1/2 H20 and CaS04 2H20 which generally necessitates the use of
holding ponds.
Scaling is greatly reduced in dual alkali systems because sodium
sulfite is the active scrubbing species not lime or limestone. Instead of
calcium ions and sulfate ions forming CaSO^ 2^0, sodium sulfite in the
dual alkali system reacts with S02 forming sodium bisulfite which is
soluble. Since the lime or limestone employed for the regeneration reac-
tion is maintained in a separate system component, the operation of the
scrubber section of the dual alkali system should be free from plugging
of lines caused by scaling. This is extremely important for continuous
operation because scaling can be controlled and limited to a smaller part
of the scrubber system. In the dual alkali system, scaling problems will
occur in lime or limestone feed areas, but will not directly affect the
scrubber unit. Figure 3 shows a schematic flow diagram of a dual alkali
pilot plant. The calcium slurry is monitored at sampling points 4, 5, and
6 of Figure 3 and any soluble calcium (sample point 10) is minimized by
the addition of Na2C03 or Na2S04.
The waste products from dual alkali scrubbers have better handling
properties than lime/limestone scrubbers. Dual alkali scrubbers produce
a high gypsum sludge (CaS04 2H2
-------
TRAY SCRUBBER
WITH DEMISTER
AND 2-TRAYS
FLU E GAS
\
SCRUBBED GAS t T
-9* MIX TANK
Na0CO,. OR Na.SOA
f- J L 4
VENTUR1
SCRUBBER
H20
Ca(OH)9 OR CaCO
SETTLER
REACTOR
SYSTEM
Figure 3. Schematic Flow Diagram of a Dual Alkali Pilot Plant
-------
feed stream which is a slurry with solids (CaO or CaC03) content of 8%
where dual alkali scrubbers employ an absorber feed stream which is a
1 to 2 molar sulfite solution with little or no solids present.
The flue gas enters the venturi and moves upward through the tray
scrubber; the recycle liquor (mostly sodium bisulfites) drops into the
recycle tank and is transferred to the reactor system. At the reactor
system, lime or limestone is added to the liquor to convert the HSO" to
SOI. This solution contains calcium which must be removed before entering
the scrubber. The process of adding Na9COv which reduces the dissolved
Cm *5
calcium ion concentration in the regeneration liquor, is known as soften-
ing. The purpose of softening the scrubbing liquor before recycling to
the scrubber is to assure that soluble calcium is minimized which reduces
the gypsum scaling potential in the scrubber.
2.1.2.3 Sampling Points and Species of Interest
Control of process stream concentrations is extremely important for
dual alkali scrubbing. Each number in Figure 3 refers to process streams
where samples are taken. Each process stream has been classified into one
of four groups: solution, slurry, solid, and solubles in cake. Figure 4
shows an overall view of the types of analyses to be performed on the
samples from a dual alkali wet scrubber. The major species of interest
are: calcium, total oxidizable sulfur (TOS), carbonate, magnesium,
chloride, and sulfate.
The calcium concentration must be known throughout the scrubber sys-
tem because of the potential scaling problems. The most important sampling
point for calcium is the feed line into the tray scrubber (point 8, Fig-
ure 3). Any buildup of a gypsum scaling will prohibit the flue gas from
contacting the sodium sulfite solution which would cause poor scrubbing
efficiency. A second important area for calcium monitoring is the regen-
eration of the soluble alkali (points 12, 4, 5, 6; Figure 3). The con-
centration of calcium must be known for lime or limestone utilization
data.
Total oxidizable sulfur (TOS) value refers to sulfur in the +4 oxida-
tion state.
TOS (mole/liter) = [SOJj] + [HSO~])
10
-------
TYPE OF PROCESS STREAM
SOLUTION
DUAL -ALKALI
SCRUBBER
TYPE OF PROCESS STRtAM
T
STREAH * STREAM *
LOCATION LOCATION
1,3.5,8,9,10 4,6
METHOD OF ANALYSIS
ION CHROMATOGRAPHY
METHOD OF ANALYSIS
ION CHROMATOGRAPHY
YPE OF PROCESS STRE
SOLID
STREAM .
LOCATION
2,7,11,12
AM
METHOD OF ANALYSIS
ION CHRCHATOGRAPHY
SPECIES OF INTEREST
1. TOTAL OXIOIZABLE
SULFUR (TOS)
2 . SULf ATE
3. TOTAL SULFUR
4. HYDROXIDE
5. CARBONATE
6. CALCIUM
7. SODIUM
8. CHLORIDE
9. MAGNESIUM
10. NITRATE
SPECIES OF INTEREST
1. CALCIUM
2. HYDROXIDE
3. CARBONATE
* NUMBERS REFER TO FIGURE 1
TYPE OF PRO
SOLUBLES
CESS STREAM
IN CAKE
STREAM *
LOCATION
7
METHOD OF ANALYSIS
ION CHROMATOGRAPHY
SPECIES OF INTEREST
1. TOS
2. SULFATE
3. TOTAL SULFUR
4. HYDROXIDE
5. CARBONATE
6. CALCIUM
7. TOTAL SODIUM
8. MAGNESIUM
9, ALKALINITY IN LIME
10, CALCIUM IN LIME
11. CARBONATE IN LIME
12. CHLORIDE IN LIME
13. CARBONATE IN SODA ASH
14. CHLORIDE IN SODA ASH
15. MAGNESIUM IN SODA ASH
16. ALKALINITY IN LIMESTONE
17. CALCIUM IN LIMESTONE
18. CARBONATE IN LIMESTONE
19. MAGNESIUM IN LIMESTONE
20. CHLORIDE IN LIMESTONE
SPECIES OF INTEREST
1. TOTAL SOLUBLE SOLIDS
2. SOLUBLE TOS
3. SOLUBLE SULFATE
4. SOLUBLE TOTAL SULFUR
5. SOLUBLE HYDROXIDE
6. SOLUBLE CARBONATE
7. SOLUBLE CALCIUM
8. SOLUBLE SODIUM
9. SOLUBLE MAGNESIUM
Figure 4. Breakdown of the Species of Interest and
Process Location for a Dual Alkali Scrubber.
-------
The TOS value is important because of the oxidation reactions:
S03 + 1/2 02 + SO^ (5)
HSO~ + 1/2 02 + SO^ + H+
These reactions have the effect of deactivating the sodium solutions.
TOS species are sampled in the process stream lines which feed into and
out of the venturi and tray scrubbers, (points 8, 1, 2, 3; Figure 3). The
filter cake is also analyzed for TOS because any loss of these species
will drastically cut the efficiency of the scrubber.
The carbonate concentration like calcium must be known throughout the
scrubber. The carbonate ions serve two functions, removal of calcium
(softening):
Ca++ + Na2C03 -* 2Na+ + CaC03 (6)
and removal of S02 from the flue gas:
Na2C03 + 2 S02 + H20 + 2NaHS03 + C02 (7)
The sampling points for carbonate ions are the feed lines to the sodium
makeup tank and tray feed tank (points 11, 9; Figure 3). The feed line
into the tray scrubber (point 8, Figure 3) contains the carbonate ions for
scrubbing the flue gas.
Magnesium, unlike carbonate, calcium, or TOS, is not a basic chemical
to the scrubber, rather it is an additive used to increase scrubbing
efficiency. The magnesium concentration is determined in all samples from
the scrubber for utilization data.
The sulfate concentration is another important quantity which must be
known through the scrubber unit. Sulfates are the reaction products of the
oxidation of TOS species. Increases in the sulfate concentration indicate
that the effective scrubbing potential of the dual alkali scrubber is being
reduced through the loss of sulfite. The efficiency of the dual alkali
12
-------
Scrubber is measured by maintaining high concentration of TOS while mini-
mizing the sulfate level. The outlet process stream lines from the tray
scrubber (points 1, 2, 3) are the important points where sulfates would be
found and should be monitored.
2.2 SAMPLE COLLECTION (LIME/LIMESTONE AND DUAL ALKALI)
There are three possible types of process streams found in lime/lime-
stone and dual alkali wet scrubbers: solid, slurries and solutions.
Figures 2 and 4 show a breakdown of each stream for both scrubbers accord-
ing to their type and the species of interest. For lime/limestone wet
scrubbers, most process streams are slurries with the solids content for
each stream depending on the location. Feed lines into the lime/limestone
spray tower and recycle stream lines will be below 10% solids while other
lines will be above 10%. For dual alkali wet scrubbers, all three types
of process streams will be found. Feed lines into the dual alkali tray
scrubber are solutions; feed lines into the mixing tanks are solids and
all other process streams are slurries with varying solids content. The
following section describes how to sample the general types of process
streams regardless of the type of scrubbers.
2.2.1 Sample Acquisition
Sample acquisition is extremely important because the sample must be
representative of the source. The factors which determine representative-
ness of the sample are stream homogenity and flow rates. In general, most
process streams for wet scrubbers have areas where high enough flow rates
exist for good homogenity. Grab samples can be taken from these places.
For the scrubbers shown in Figures 1 and 2 built-in sampling ports already
exist, so grab samples will be easy to obtain. The areas where stratifica-
tion of slurries may occur are in the clarifier and hold tanks for these
scrubbers. This problem can be solved by taking grab samples from the
feed line into the clarifier and hold tank or at their exit lines.
2.2.2 On Site Solution and Slurry Sample Treatment
Most samples taken from either scrubber system can be taken using
grab techniques. If the sample contains solids, they are removed back in
the laboratory using N~ pressure filtration system to separate the liquids
13
-------
and solids. Filtration accomplished using a Gelman Acropore 1.2 um filter
in the apparatus shown in Figure 5. This approach is preferred over
gravity or Buchner filtration, since $2 contact is minimized preventing
SOI oxidation.
Due to the chemical nature of several streams, the distribution of
inorganic species in the liquid phase will be disturbed unless the liquid
and solid phases are immediately separated. At most wet scrubber sites,
demister wash return, recycle slurry, and absorber sludge bleed-off
streams will require on-line filtration.
The on-line filtration apparatus is shown in Figure 6. This system
uses a large (142 mm) filter holder with a Gelman Acropore 1.2 ym filter
to separate the solids from the slurry. The valving is arranged, so that
all the lines can be purged prior to sampling. While a pump is shown, most
streams are under sufficient pressure (especially downstream of a pump) to
force the slurry through the filter. The samples are filtered directly
into plastic containers which have been first rinsed with high purity HNO~
and then rinsed thoroughly with deionized water. When the solutions are
placed into the plastic container, exposure to air should be minimized to
prevent oxidation. The solution should be filled to the top of the con-
tainer to also minimize oxidation. Analysis for sulfite should be con-
ducted on site and immediately after the sample is taken. Because
carbonate is one of the species of interest, bubbling and shaking of the
solution should also be minimized.
2.2.3 Solution and Slurry Sample Size Requirement
An important factor which is always considered in sampling process
streams is the size of sample required. Normally, there are two factors
that determine how much sample must be collected. First, the amount of
sample collected must be sufficient for the testing and analysis proce-
dures to furnish accurate and precise results and provide enough sample
to reduce the stastical sampling error. A 1 L grab sample will be suffi-
cient to provide approximately five hundred ml of filtered sample.
Because of the restricted filtration capability of the on-line filter, as
little as 100 ml of filtrate is sufficient.
14
-------
01
RING
STAND
[FROM N,
TANK '
CLAMP
GELMAN PRESSURE
FILTRATION SYSTEM
TEFLON TUBE
250 MIL
ERLENMEYER
Figure 5. Pressure Filtration Apparatus
15
-------
cr>
SLU RRY STREAM
TEFLON
TUBING
SAMPLE LINE
FOR GRAB
SAMPLES
,TEFLON OR PLASTIC
DIAPHRAGM PUMP
VENT
LINE
TEFLON
TUBE
VENT
RUBBER
STOPPER
PLASTIC
SAMPLE
CONTAINER
Figure 6. On-Line Liquid Slurry Filtration System
-------
For ion chromatographic analysis, 100 yL of the sample is analyzed,
but one millimeter of sample is used to fill the sampling loop and con-
necting lines. Since the species of interest for the wet scrubber analy-
ses are in concentrations greater than 100 ppm, the 100 yL analyzed sample
is sufficient to meet the sensitivity requirements based on the instrument
detection limits.
2.2.4 Solid Samples
Solid samples will be obtained by filtration of the solids from the
slurry streams or taken from the filter cake produced by either the lime/
limestone or double alkali FGD systems. Prior to analysis, it is recom-
mended that these samples be rinsed with a 50% acetone soluton to remove
most of the free water and to expedite the drying process. The rinse is
performed by placing the solids in a precleaned Buchner funnel, pouring the
acetone/water solution over the solids, and filtering. Approximately
20-30 grams of material can then be dried to constant weight in an explo-
sion proof oven set at 60°C.
2.3 SAMPLE PRESERVATION
Chemical preservation should not be used when analyzing process
streams where the species of interest are in equilibrium. Reagents like
strong mineral acids, when added to a sample for preservation, will cause
a shift in equilibrium forcing a species into or out of solution. Sample
preservation should only be employed when there is a long time period
between sample acquisition and analysis. The time period between sample
acquisition and analysis depends on the species of interest. If sulfite
is the species of interest, then time is important because sulfite will
gradually oxidize to sulfate once exposed to air. The analyst will have
to run experiments to determine the allowable time period between sample
acquisition and analysis.
These experiments would consist of taking ten similar samples, then
allowing each sample to stand for a specific time period before analysis.
All samples will be analyzed for concentration by ion chromatography for
a specific ion or species vs the time the sample was allows to stand. The
plot would look something similar to Figures 7 or 8 and would give the
analyst an indicator when sample preservation would be necessary.
17
-------
o
o
o
01 23456789
TIME (Mr)
Figure 7. Plot of Species Concentration vs Time
(Immediate Sample Preservation Required)
i.
Q.
O
O
8 10 12 14 16 18 20
TIME (Hr)
Figure 8. Plot of Species Concentration vs Time
(No Immediate Sample Preservation Required)
18
-------
Figure 7 indicates the analysis would have to occur immediately after sample
acquisition where Figure 8 indicates the analyst would have a few hours
before preservation is necessary. The following section describes the
procedure which should be employed for sample preservation.
2.3.1 Procedures for Sample Preservation
Plots of species concentration vs the time period between sample
acquisition and analysis will indicate whether a sample should be preserved.
Samples from process streams whether dual alkali or lime/limestone will be
separated into either solids or solutions after filtration. Solution
samples and filtrate from slurry samples will be further divided into five
groups depending on the species to be analyzed and preserved, according to
the procedures in Table 1. Solid samples, filter cake samples, and solids
from slurry samples after the 50% acetone rinse will not require the addi-
tion of any preservation agent, but are placed into sealed plastic con-
tainers. Exposure to air should be minimized in all sampling activities.
19
-------
TABLE 1. RECOMMENDATIONS FOR SOLUTION PRESERVATION
Measurement
Container
Preservative
Holding Time
Total oxidizable sulfur
Sulfate
Hydroxide
Acidity
Carbonate
Chloride
Plastic
Cool, 4°C
24 hrs
ro
O
Calcium
Sodium
Magnesium
Plastic
HN03 to pH
6 mos
PH
Plastic
Cool, 4°C
Detection on site
6 hrs
Total Sulfur
Plastic
Cool, 4°C
Zn Acetate
24 hrs
Nitrate
Plastic
Cool, 4°C
H S04 to pH 2
24 hrs
-------
3. ANALYTICAL PROCEDURES
This section describes the theory and practice of ion chromatography.
It presents the anion and cation analysis scheme for wet scrubber samples.
Also, included in this section will be definitions and discussions of
problem areas involving the use of ion chromatography for wet scrubber
related samples.
3.1 INTRODUCTION
Ion-exchange chromatography is a process whereby ions of the same
charge in solution compete for positions with other ions of the same
charge, which are bound to the surface of a solid.
Resin
Resin
A" + Y"
Functional Group
and Backbone
Y" + A"
Functional Group
and Backbone
(8)
where
A" is the sample ion
Y" is the mobile phase
(anion exchange reaction)
As indicated in Equation (8), sample ion A" is exchanging with the
mobile phase ion Y" which is bound to a functional group or the active
site of a resin. The functional group is always opposite in charge to the
exchanging ions. The rate at which the ions exchange positions is deter-
mined by the ion's affinity for the functional group. The physical char-
acteristics of size and charge determines the ions affinity or interaction
with the functional group of the resin. Ions which have a charge that can
be polarized toward the functional group will have a stronger interaction
with the resin thus a slower rate of exchange. The attraction between an
ion and the resin can be thought of as an ion-pair. In practice, the rate
of exchange (ion-pair interaction) determines the elution order on the
chromatogram. The slower the rate of exchange, the stronger the ion-pair
interaction resulting in a longer elution time.
21
-------
A second factor which also can control an ion-exchange reaction is
the mobile phase. The mobile phase is usually an acid or base. Consider
a weak acid HX, the dissociation of the acid is:
HA + H20 t H30+ + A" (9)
The degree of the dissociation can be controlled by the pH of the
systems. An increase of H ions causes the equilibrium in Equation 9 to
shift to the left. The resulting shift causes less A" ions to be avail-
able for interaction with the anion exchange resin or less A" ions compet-
ing with sample ions for the resin which may result in longer elution times.
When ions elute from the resin, they are in a background of the mobile
phase ions. For many years, research was spent in attempting to develop a
universal detector which would be compatible with the mobile phase ions.
2
In 1975, Bauman, Small and Stevens presented a paper which established
the principles of modern ion chromatography. These workers minimized the
chemical effects of the mobile phase so that a conductivity meter could be
used as the detector system. Bauman, Small and Stevens placed a second
ion-exchange column behind the first which suppressed or neutralized the
mobile phase but did not interfere or alter the sample ions. The second
ion-exchange column is always opposite in chemical nature (acid or base)
to the mobile phase. If the mobile phase ions are protons, the second
ion-exchange column is a hydroxide ion exchanger for neutralization.
Modern ion chromatography employs an anion or cation ion-exchange column
followed by a second acid or base ion-exchange column to produce a deion-
ized water background with the anion or cation present as an acid or base
respectively.
3.1.1 Column Packing
The solid material or resin where the ion-exchange occurs is referred
to as the column packing. The column packing is usually spherical par-
ticles of divinyl benzene cross-linked polystyrene which forms the backbone
of the resin. The chemical functional groups fixed to the backbone gives
the resin its chemical exchange characteristics. Figure 9 shows the two
22
-------
CH-CH2-CH2-CH-fCH-CH,
H-
CH-CH2-CH2-CH-CH2
CATION RESIN
CH,
OH
k
9
H-CH,-CH0-CH-/CH-CH
V
H-
CH-CH2-CH2-CH-CH2-(cH-CH2yn-
OH"
CH
ANION RESIN
Figure 9. Structure of Varian Aerograph Anlon and Cation
Exchange Resins
23
-------
different Varian Aerograph ion-exchange resins with -S03" for cation
analysis or
,^
N R
R
for anion analysis.
There are only two types of column packings employed for ion chroma-
tography, high capacity or macroreticular and low capacity or pellicular.
Capacity refers to the amount of active sites or functional groups avail-
able for ion-exchange. High capacity packings are porous because they
contain both micropores and rigid pores. Low capacity packings are only
slightly porous because they have a solid inert core with a thin coherent
film of ion-exchange material on the surface. Figure 10 illustrates the
difference between high and low capacity ion-exchange packings.
The first ion-exchange column is called a separator and is usually a
low capacity column. The second column, called a suppressor, neutralizes
the mobile phase and is usually a high capacity column. Also, the mobile
phase is usually referred to as the eluent in ion chromatography.
3.1.2 Practical Ion Chromatography
The following example illustrates the use of practical ion chroma-
2 + + +
tography. A sample containing Li , Na , and K enters an ion chroma-
tography which has been set up for a cation separation. The first column
is a cation-exchanger (low capacity-sulfonated). The eluent is dilute
nitric acid which has been constantly pumped through the column before
the sample enters, therefore, the function group is occupied by H* ions
before the sample is injected into the column. When the sample enters
the first column, Li+, Na+, and K , now try to displace H+ from the active
site or function group of resin. Lithium ion being the smallest ion of
the three will have the least affinity to the resin because its charge 1s
hardest to polarize. Potassium ion being the largest will be most
attracted because its charge will polarize and form a strong ion-pair.
The order of elution from the first column will be Li+, Na+ and K+ in a
H , N03" background. The N03" ions from the eluent and all other anions
24
-------
MICRO PORES
MACRO PORES
MACRORETICULAR ION EXCHANGER RESIN
(HIGH CAPACITY)
SOLID
INERT
CORE
ION EXCHANGE FILM
PELLICULAR ION EXCHANGER RESIN
(LOW CAPACITY)
Figure 10. Cross Sections of a Macroreticular and Pellicular Resins
25
-------
will not be retained by the first column. A key point to realize is that
some of the protons used in the original eluent are now bound to the
resin. Each time a Li + , Na , or K ion becomes bound to the resin, it
replaced a proton on the resin. Later, the Li , Na , and K ions would
be displaced by a proton from the eluent before exiting the column. The
rate of exchange between the coupling eluent ions and the cations of
interest is determined by their concentrations and their individual affin-
ities for the function group as previously described. Because of the con-
stant exchanging between the eluent and the sample ions, no regeneration
of the first column is necessary in ion chromatography. The eluent con-
stantly provides the necessary protons for regeneration. As the separated
ions elute the first colunn, they directly enter the second column.
The second column, or suppressor, contains a basic hydroxide exchange
resin which neutralizes the acid eluent. Two reactions take place simul-
taneously in the suppressor column:
Li+, Na+, or K+ + N03"+ Resin OH" + (IQ)
Li+, Na+ or K+ + OH" + Resin NO3
H+ + N03" + Resin OH" -Ğ HgO + Resin NOg" (n)
The first reaction (10) is an ion-exchange reaction between NO-" and
OH" which essentially liberates OH" from the resin and converts the alkali
metal chloride to their hydroxide. The second reaction (11) is a stripping
reaction where H will pull a OH" ion from the resin. The sample ions are
not affected by the second column and remain in the same elution order.
As the sample ions elute the second column, they are now in a deionized
water background and pass into the conductivity cell for detection.
Because the suppressor column is depleted of OH" ions, it must be
regenerated. The suppressor column is regenerated by flushing with NaOH:
NaOH + Resin NO^ -ğ Resin OH~+ Na+ + N03" (12)
26
-------
The Dionex instrument has a semi-automatic procedure for regeneration and
this is usually done at the end of each day.
3.2 ION CHROMATOGRAPHY - DEFINITIONS
This section will define specific key concepts involving ion chroma-
tography. These concepts are retention time, overloading or column capac-
ity, resolution, temperature, columns, and filtration. Other chromato-
graphic concepts like temperature programming, solvent efficiency,
programmed eluent gradient solution, etc. will not be discussed because
they are currently not employed in ion chromatography. Ion chromatography
is a relatively new technique and most research has been involved in the
application areas using standard columns. As ion chromatography develops,
research will find new packing material for more types of separation, or
develop techniques like programmed gradient elution. The concepts defined
in this section will help the analyst avoid problem areas which could
occur during an I.C. analysis.
3.2.1 Retention Time
Retention time refers to the time between the injection point and the
peak maximum.
RETENTION
TIME
A
RETENTION
TIME
.. ltl. RETENTION
T A L i
Retention times are used for qualitative analysis. The most important
factor for all types of chromatographic systems is that the retention time
must be reproducible. For ion chromatography, the retention time of a
species may vary by ħ20 sec. Any large deviation in retention time usually
indicates a change in column conditions. Standard mixtures are analyzed
to monitor column performance and correct for changes in column
conditions.
27
-------
3.2.2 Resolution
Resolution refers to the separation of two consecutive peaks. Usually
increasing column length or reducing the acidity or basicity of the eluent
will improve resolution of two consecutive peaks. Ion chromatography
offers distinct advantages over other chromatographic techniques through
3
the use of wet chemical techniques and deviatives . Figure 11 shows an
ion chromatogram where S03~ and N03" peaks are not completely resolved.
The use of wet chemical techniques can completely resolve the S03~/NO-~
problem without any changes to the columns or eluent. The addition of
perixode to a second sample will oxidize SO-~ peak. The second sample is
run on the ion chromatography and the difference between the first ion
chromatogram and the second will determine the S03~ concentration. Fig-
ure 11 shows the second ion chromatogram with the addition of peroxide.
The initial peak at 2.5 min mark of the ion chromatogram indicates the
peroxide and peroxide by-products. Figure 12 shows a chromatogram uf a
100 ppm S03~ standard (9 min RT). The large peak at 13.8 min is SO/.
Figure 12 shows the same S03~ standard with the addition of H202, which
converts all the SO-~ to S0,~. Figure 12 shows the chromatogram of a
100 ppm standard of N03" (9 min RT). Notice that the NO.," response is
much stronger than the S03~ response for the same concentration. The
detector response is a function of concentration and ion conductance.
3.2.3 Column Capacity - Overloading
The term column capacity refers to the amount of active sites avail-
able for ion-exchange. Separator columns are low capacity resins which
can be easily overloaded because the active sites are scattered on the
surface. The term overloaded means that all the active sites on the resin
are being occupied and there is an excess of ions still competing for the
active sites. Overloading a column's capacity is indicated by a reduction
in the retention time of the operation and changes in peak shape. Both
reduction in retention time and a change in peak shape occur simultane-
ously. The peak shape for most species on an ion chromatogram are usually
sharp spikes or gaussian. When overloading occurs, the shape of the peak
moves from gaussian to skewed with increased tailing.
28
-------
COLUMNS
ELUENT
WONEX ANION SEPAIATOt 3X500i
DION6X ANION SUfPKSSOt 4X150.
0. 008 M N.HCOj/0, 0025 M
30% FLOW
AT
NO} AND 100 |
2-1/1
SO3'2 AND
A - 10 X NO3"/SO3"
30 X SO,"8
INJ
100 ffn> NOj" AND 100 ppm SOj"
Figure 11. Use of Peroxide to Resolve the Peak
Overlap of NHrate/Sulflte
29
-------
COLUMNS
DIONEX
DIONEX
AN) ON
AN) ON
SEPARATOR
SUPPRESSOR
3 X 500 mm
6X250 mm
ELUENT
0.003 M NoHCOj/0.0025 M
30% FLOW B
A - 30 X NO,
INJ
B = 30 X SO4 ~2
Cğ10XH2O2
-2.5
100 ppm NO3
100 ppm SO3"2 AND H
A-10XSO,
B - 30 X SO
-2
INJ
TOO ppm
Figure 12. Ion Chromatograms of Sulfite, Sulfite and
Peroxide, and Nitrate
30
-------
B
RETENTION TIME I \ RETENTION TIME
PEAKA-9MIN. / \PEAKB-7MIN.
9 79
The change in peak shape and the decrease in retention time is caused
by similar ion agglomeration at the overloaded active site. Ions at the
overloaded site are not absorbed but just associated with the others.
These ions tend to be more loosely held and migrate faster than those ions
directly associated with the functional group causing the skewed peak
shape. The tail of the overloaded or skewed peak will always coincide with
the tail of the gaussian peak. Comparing the tailing of overload peaks to
standard peaks sometimes can be used to distinguish or identify certain
unresolved ions.
3.2.4 Temperature
In general, temperature does not affect an ion chromatography analysis
of scrubber samples. Temperature changes of 3-5°C/24 hr period can be
tolerated without significant degradation of retention time or calibration
curves. Temperature changes above l°C/hour are not acceptable.
3.2.5 Pre-columns
Pre-columns are short (5 to 20 mm) columns packed as the separator,
and are placed before the separator column. Since separator columns are
expensive, pre-columns are employed to extend the separator column life.
By holding up or filtering out trace impurities in the sample, the pre-
columns prevent contaminants from contacting the separator column.
3.2.6 Sample Filtration
Samples should be filtered before injection into the ion chromatograph.
In the sample collection section, it was recommended to filter the wet
scrubber sample with a Gelman Acropore 1.2 ym filter before injection. If
samples contain particulate matter and are not filtered before injection
31
-------
into the I.C., they will eventually restrict the flow of the element and
problems with pressure drop will develop.
3.3 SOLID SAMPLE PREPARATION
All liquid samples can be run directly after dilution as long as the
results are on scale. Sample preparation is required for all solid samples
prior to I.C. analysis. The solids from both the lime/limestone and dual
alkali systems will have to be treated in a similar fashion so that
cations and species like S03~, S0.~, Cl" and CO-" can be analyzed.
Table 2 summarizes the extraction procedures that may be used to solubil-
ize components of the filter cake. A distinction between water soluble
and acid soluble species is made for the dual alkali systems, since
NaSO^ or NaSO. might be contained in the cake. The two acid extraction
O H
procedures are designed to stabilize the scrubber related solids (CaSO,,
CaS03, CaC03, CaO) and leave the fly ash. The difference in sulfate
values between the oxidizing acid extraction and the acid purging pro-
cedures is the sulfite concentration. In systems where the spent slurry
streams are oxidized prior to filtration, a direct measure of the sulfite
is preferred. In this case the N,, purge stream is collected in a 1%
H202 trap and analyzed for S04~, which can be related back the original
concentration of sulfite in the solids.
3.4 CATION ANALYSIS
The cations of interest for both dual alkali and lime/limestone wet
+ J. XJ. -II.
scrubbers are: Na , K , Mg and Ca . The analytical procedure for
cation analysis utilizes the Dionex Ion Chromatograph. Since the cations
of interest for the wet scrubber analysis are in concentrations greater
than 100 ppm, the Minimum Detectable Limit (MDL) for all cations in this
procedure is one ppm. The Dionex ion chromatograph using this analytical
procedure will have much lower MDLs for all cations. However, the objec-
tive of this manual is to apply ion chromatography to wet scrubber
samples, not to evaluate the capabilities of the instrument. The cation
analysis procedure contained in this manual is based on the procedures
called for in the Dionex instruction manual, as modified by TRW for these
types of samples.
32
-------
Table 2. Extraction Approaches for Solid Samples
System
Extraction System
Procedure
Species Extracted
Dual Alkali
Lime/
Limestone
- Dual Alkali
CO
CO
Lime/
Limestone
- Dual Alkali
1. 02 free
2. 0.1 m HCl/3% H202
3. 0.1 m HCl with
N purging
Extract 1 g of solids with
10 mL 02 free air, filter
and dilute to 50 mL.
Extract 1 g of solids with
10 nL of 3% H202 followed
by 10 mL of 0.1 N HCl.
Dilute to 100 mL.
Place 1 g of solids into a
small impinger. Add 10 mL
of 02 free 0.1 M HCl to
impinger and begin sparging
with 02 free N2- Continue
until no more solids
dissolve. Dilute to 100 mL.
(NOTE: For non-water
soluble C03= a dilute NaOH
trap should be added to
collect the C02 for
analysis).
- Cl"
- Water soluble S0.~, S03", C03"
- Cations
- Total sulfur (as S0.~)
- Cations
-so,
- SO-
(Total Sulfur - S04 )
- TOS (S03~)
- C0,~ (if NaOH trap used)
-------
3.4.1 Sodium and Potassium Analysis
Sodium and potassium are run on the Dionex cation separator (6 x 250
mm) and suppressor (9 x 250 mm) columns. The pump is set for 40% flow
rate with 0.005 N HN03 as eluent. The retention time for 10 ppm Na+
is seven minutes and twelve minutes for 10 ppm K .
Trace metals sometimes tend to build up in the Dionex cation separator
column and this is reflected in changes in retention time and peak height.
By flushing the cation separator column with ultra pure 1 N nitric acid
for 15 to 30 minutes, the trace metal buildup will usually be removed. In
general, it is a good practice to always flush the ion chromatograph with
deionized water for 30 minutes, after a set of analyses or column changes.
This includes flushing the detector system and not just the columns.
3.4.2 Magnesium and Calcium Analysis
Magnesium and calcium are separated with a Dionex cation separator
(6 x 250 mm) and suppressor (9 x 250 mm) columns. The pump is set for 40%
flow rate with 0.001 M p-phenylenediamine dihydrochloride as eluent. The
ti
+2
+2
retention time for 10 ppm Mg is eight minutes and 12 minutes for
10 ppm Ca
The same type of columns are used in the Mg /Ca+2 separation as in
the Na /K separation but the two groups of separations must be run on a
different set of columns. Since the p-phenylenediamine dihydrochloride
eluent tends to deteriorate the columns and the Na+/K+ separation is poor
on columns that are exposed to this eluent. Because of the cost of the
columns, it is more practical to run the two groups of separations on two
sets of columns rather than the one set. Nitric acid does not work as an
eluent for the Mg+2/Ca+2 separation.
The deteriorate.! of the cation separator is very easy to identify
because the column turns from white to dark gray to black. The nitric acid
eluent does not deteriorate or change the color of the cation separator
column and, therefore, it is very easy to distinguish between two sets of
columns. Whenever p-phenylenediamine dihydrochloride is employed as an
eluent, it should be made up fresh every day. This is a modification of
the Dionex procedure which states that p-phenylenediamine dichydrochloride
should be made up fresh every week. TRW found that this eluent lost its
34
-------
+2 +2
ability to force the Mg , Ca separation, and the deterioration of the
column was more rapid when the eluent was not fresh. After using
p-phenylenediamine dihydrochloride, the ion chromatographic should be
flushed with deionized water overnight, including the conductivity cell.
This can be done by filling the water reservoir (4 liters) and employing
a flow rate of 20%. Sometimes, flushing the columns with 1 N nitric acid
for one hour will remove some of the dark gray or black in the separator
column followed by the overnight deionized water rinse. It is extremely
+2 +2
important to rinse the conductivity cell after each Mg /Ca run because
p-phenylenediamine dihydrochloride tends to cause a buildup of particles
in the cell.
3.5 ANION ANALYSIS
The anions of interest for both dual alkali and lime/limestone wet
scrubbers are: sulfite, sulfate, carbonate, chloride and nitrite. The
anion analysis procedure was not developed by TRW but was taken directly
from the Dionex instruction manual. The MDL for sulfite, sulfate and
nitrite is one ppm. Chloride could vary between 0.1 and 10 ppm depending
on conditions, and carbonate's MDL is 5 ppm. In the chloride analysis,
the MDL improves if eluent is added to the sample before analysis. The
addition of eluent to the sample suppresses the water dip on the ion
chromatogram which occurs immediately before the chloride peak. The
carbonate usually has a high MDL because of the carbonate found in the
blanks.
The analytical scheme is based on the retention time for sulfate
which is the last species of interest to elute from the columns. The
conditions stated below will cause sulfate to elute in 15 to 17 minutes,
but the sulfite and nitrate peak will only be partially resolved.
Conditions can be changed so that sulfite and nitrate will be completely
resolved, but the retention time of sulfate will increase to about
45 minutes. The main problem with increased retention time is that peak
heights or areas become very difficult to determine and thus introduces
errors into the analysis. With shorter retention times, the peaks are
gaussian and, therefore, much easier to quantitate. Since wet scrubbers
samples normally have very little nitrate, this scheme uses the shorter
analysis time procedure.
35
-------
3.5.1 Chloride, TOS and Sulfate Analysis
The first anion sample run determines the amounts of chloride, total
oxidizable sulfur (TOS) and sulfate. TOS species are sulfite and bisulfite.
Both sulfite and bisulfite elute as a single peak because bisulfite cannot
exist in the eluent which has a pH of 10. The columns employed are the
Dionex anion separator (3 x 500 mm) and suppressor (6 x 250 mm). The pump
is set for 30% flow rate with 0.003 M NaHC03/0.0024 M Na2C03 as eluent.
The retention time for 10 ppm Cl is 4.5 min, ten min for TOS and 16 min
2
for SO, . If sulfite/nitrate separation becomes a problem, then the
eluent strength should be changed to 0.003 M NaHC03/0.0012 M Na2CO.,.
3.5.2 Sulfur and Nitrate Analysis
If one is interested in total sulfur or determining nitrate but not
increasing analysis time, a second anion sample should be run. The ion
chromatographic columns and conditions are the same except that the anion
sample is pretreated with H^Op. The peroxide oxidizes the sulfur in
solution to S0.~, so that the total sulfur can be determined, and it
removes any S03 /N0~3 peak overlap.
3.5.3 Carbonate Analysis
Carbonate is run on a BioRad A650W suppressor column (200-400 mesh,
6 x 500 mm). The pump is set for a 30% flow rate with deionized water as
the eluent. The BioRad resin separates out C03 from all other anions by
ion exclusion. The retention time for 10 ppm C03 is 6 minutes where all
other anions will elute in 2 minutes. The carbonate determination employs
only a single column, the suppressor is not used. Bicarbonate is con-
verted to carbonate by the eluent and total carbonate is analyzed in a
fashion analogous to S03=/HS03".
3.5.4 Hydroxide Analysis
Hydroxide cannot be determined by ion chromatography. The alternate
method for determination of hydroxide is titration. Because the wet
scrubber samples also contains C03~ ions, the titration will require two
indicators (phenolphthalein and methyl orange). The sample is titrated
with standard acid, first to the phenolphthalein endpoint (pH 9-10), and
then further to the methyl orange endpoint (pH 4-5). The following
example illustrates the dual endpoint titration technique.
36
-------
Example: A 25.00 ml aliquot of a solution containing NaOH and/or
Na2C03 and/or NaHC03 requires 24.23 ml of 0.100 N HC1 to the phenolphthalein
endpoint and an additional 22.37 ml from the phenolphthalein to the methyl
orange endpoint. Assuming that endpoint errors are negligible, calculate
the composition of the sample.
Since AJ24.23 + 22.37, or 46.60 ml) is less than twice A (24.23 ml),
the sample must contain NaOH as well as Na2C03.
To the phenolphthalein endpoint:
OH" + H30 * 2H20
£ + H.O -* HCO:
O O w
mmoles NaOH + \
mmoles Na0CO-, /
mmoles HC1 used to \
phenolphthalein endpoint /
0.1000 x 24.23 = 2.423
(13)
To the methyl orange endpoint:
OH" + H30 + 2H20
C03 + 2H30+ -> H2C
mmoles NaOH +
2 x mmoles Na2C03
2H20
/ mmoles HC1 used to \
methyl orange endpoint/
= 0.1000 x 46.60 = 4.660
(14)
From the difference between Eqs. (13) and (14), the number of mmoles
of Na2C03 in the 25 ml aliquot is 4.660 - 2.423, or 2.237. The Na2C03
concentration is therefore 2.237/25.00, or 0.0895 M.
By difference, the number of mmoles NaOH in the 25 ml aliquot is
2.423 - 2.237, or 0.186. The NaOH concentration is therefore 0.186/25.00
or 0.00744 M.
37
-------
4. QUALITATIVE ANALYSIS
This section is devoted to the identification of the various species
found in the process streams of wet scrubbers. In general, species of
interest for wet scrubbers are found in concentrations of greater than
100 ppm. Compound and species identification by ion chromatography (I.C.)
is based upon the retention time of standard solutions. Since most resins
employed for I.C. are low capacity, the retention time can vary with
sample concentration. As stated in the previous section, the retention
time of various species analyzed by I.C. can vary approximately ħ20 sec.
where any larger time deviations usually indicate changed column conditions.
In this manual, all retention times refer to 10 ppm standards of the stated
species.
4.1 RETENTION TIME TABLES
The retention times for the various wet scrubber species are listed in
Table 3. The retention time was determined for a 10 ppm standard solution
of each species. The standard solution consisted of dissolving the
appropriate amount of soluble salt in deionized water. Because I.C. is a
separation technique, it should not suffer from matrix interferences like
atomic absorption. I.C. interferences usually center around the incomplete
resolution of two species. Table 3 is a reference table which will give
an analyst some approximate indication where and in what order species will
separate. An analyst should develop his own retention time table.
38
-------
TABLE 3. RETENTION TIME OF VARIOUS IONS FOUND IN WET SCRUBBERS
CJ
to
Sppcips
Cl"
503
N03
S04
C03
Ca+2
Mg+2
Na+
K+
Cone.
10 ppm
10 ppm
10 ppm
10 ppm
10 ppm
10 ppm
10 opm
10 ppm
10 ppro
1
Retention
Time
4.5 min
10
8
16
6
12
8
7
12
Dionex
Separator
Col umri
Anion
3 x 500
Anion
3 x 500
Anion
3 x 500
Anion
3 x 500
BioRad
6 x 500
Cation
6 x 250
Cation
6 x 250
Cation
6 x 250
Cation
6 x 250
Dionex
Suppressor
Col un'.n
Anion
6 x 250
Anion
6 x ?50
Anion
6 /. 250
Anion
6 y ?Ği
None
Cation
9 x 250
Cation
9 x 250
Cation
9 x 250
Cation
9 x 250
Flow Rate
30£
1
(
30^
I
30J
i
1 - - -
30*
30V
40%
H'
.
*
40i
t
i
i
40%
'
40%
Fluent
0.003M NaHC03/
0.00?5M Na2C03
0.003M NaHCO,/
O.OOPBM Na2CD3
0.003M NaHCO-/
0.0025M Na?C03
I
0.003M NaHCO.,/
O.Cin25M Ma2CD3
Water
'---- - -j
for both
cations
0.001 M
p-phenylene-
diamine
dihydrochloride
.005N KN03
:005N HN03
i
Temp.
24"C
24 "C
24'JC
24 rC
24' C
-
24°C
24 C
24 C
24 C
-------
5. QUANTITATIVE ANALYSIS
This section is devoted to the quantification of the species of
interest found in the process streams of wet scrubbers. This section will
describe the ways for relating the ion chromatographic response to sample
concentration and the statistical treatment of data.
5.1 DETERMINATION OF SAMPLE CONCENTRATION
The ion chromatographic response is quantified using peak height and
calibration curves. Peak height is measured as the distance from the base-
line to peak maxima as shown in Figure 13. The peak height is plotted vs
sample concentration for a given conductivity scale and the resulting curve
is known as an ion chromatographic calibration curve. The calibration
curve is linear within a conductivity scale, but may plateau at higher ion
concentration levels. A calibration curve is constructed by first preparing
standard solutions of known ion concentration. The standard solutions are
run on the ion chromatograph and their peak height is determined. Finally,
the plot of sample concentration vs peak height is constructed. For
routine analysis, calibration curves are very easy to use and concentra-
tions can be read directly from the curve. Other techniques such as single
point external standard (or internal standard) can also be employed.
When constructing the calibration curve, many data points should be
taken between the limits of the conductivity scale. This will give the
working range for the linearity of the calibration curve. Figure 14 shows
that the linear range for Cl" ions is from less than 1 ppm to 10 ppm and
covers a peak height range from 1 to 25 cm, the full range on the strip
chart recorder. The calculation of the slope of the curve is determined
by a "least squares" fit of the data points.
Chromatograms can also be quantitated by measuring the peak area. A
series of experiments have shown that peak height is better than peak area
for quantification. Figure 15 shows the results between quantification by
peak area and peak height. Peak areas were calculated by full width at
half maximum (FWHM), baseline triangulation, and planemetry. The results
indicate that peak height is a more consistent way for quantification of
the ion chromatographic response.
40
-------
Figure 13. Example of Peak Height Measurements
41
-------
26
24
22
20
18
16
14
12
10
8
6
4
2
0
C1~, SLOPE - 2.59
y-INTERCEPT - -0.15 CM
CONDUCTIVITY SCALE 10X
6
8
9
10
Figure 14. Calibration Curve for Chloride Ion Analysis
42
-------
Na+, 5 PPM. 10X
Na+.5PPM. 10X
PEAK HEIGHT
FWHM
TRIANGULATION
PLANEMETRY
6.97 CM
6.60 MM2
7.13 MM2
1.37
PEAK HEIGHT
FWHM
TRIANGULATION
PLANEMETRY
RELATIVE
ERROR
7.00 CM 0.4%
7.35 MM2 10.2%
7.32 MM? 2.6%
1.42 3.5%
Figure 15. Example of Quantification by Peak Height and
Peak Area
43
-------
Determination of peak area by electronic integration is difficult.
The peak shape from a chromatographic technique usually depends on the
detector and type of column packing, and under most conditions all peaks
will have the same shape. Ion chromatography does not follow the general
characteristics stated above. The peak shape for each species is different
for the same chromatographic conditions. Figure 16 is an ion chromatogram
of a solution which contains F", Cl", N02, P03" , Br", N03~, and SO.'2.
Each ion has its own basic shape, F" and Cl" are sharp spikes whereas SO."2
is a broad gaussian. Electronic integration works by calculation of the
change in slope of the peaks. Since each peak on the ion chromatogram has
its own shape, the electronic integrator is normally set for one peak
shape and cannot adjust to a different peak shape. If an analyst sets up
- -2
the electronic integrator for Cl , then SO^ cannot be determined. Also,
species in high concentrations have a great deal of peak tailing, and an
electronic integrator has difficulty determining where the peaks, ends.
Before an analyst uses a standard gas chromatograph integrator, he should
check its operation for the above flaws. There are electronic integrators
available (like the Spectra Physics SP4000) which have the capability to
record peak height, and peak shape and these would be more adaptable to
I.C. analysis.
A second reason why ion chromatography should not be quantified by
electronic integration is the conductivity meter response. The meter
response is only linear within a single scale or attenuation but nonlinear
from one scale to another or changes in attenuation. The change in
linearity between scales is significant enough to force the analyst to
develop separate calibration curves for each scale, as in Figure 17. The
Dionex instrument has nine scales but the analyst can employ dilution
techniques to reduce the number of calibration curves. Usually,
establishing calibration curves for the 1,3,10,30 yMHO scales gives an
analyst a good workable concentration range.
The quantification of ion chromatography by peak height also has some
problems. During the evaluation of the ion chromatograph, it was observed
that the peak height for the same sample could change from day to day.
Table 4 illustrates the changes in peak heights of various ions over a
period of two months. The data in Table 4 shows that at the early part of
44
-------
F-
COLUMNS
DIONEX ANION
DIONEX ANION
\
SEPARATOR
SUPPRESSOR
3x500 MM
6 x 250 MM
ELUENT
0.003 M NaHCO3/0.0025 M Na2CO3 AT 30% FLOW
SO
-2
INJ
Figure 16. Ion Chromatogram of the Dionex
Anion Standard Solution
45
-------
IK
1000 ppm
-2
COLUMNS
ELUENT
DIONEX ANION SEPARATOR 3 X 500 mm
DIONEX ANION SUPPRESSOR 6 X 250 mm
0.003 M NaHCXy 0.0025
AT 30% FLOW
100X
100 ppm SO4
-2
10X
10 ppm SO
-2
IX
1 ppm
-2
INJ
10
PEAK HEIGHT =4.5 cm
PEAK AREA = 0.33
INJ
13 '
PEAK HEIGHT = 5.5 cm
PEAK AREA = 0.29
INJ
PEAK HEIGHT = 4.6 cm
PEAK AREA = 0.27
INJ
PEAK HEIGHT =5.Ocm
PEAK AREA = 0.28
Figure 17. An Illustration of the Nonlinear IonNChromatographic Detector Response
-------
Table 4. PRECISION OF PEAK HEIGHT MEASUREMENTS
Determinations
Ions
F"
Cl"
N0~
Br"
NO-
so4"2
Scale
30X
10X
10X
10X
10X
30X
1
20.7
14.2
4.8
6
11.7
8.2
2
20.0
13.7
4.7
6
11.2
8.3
3
20.1
13.9
4.2
5.7
11.5
8.0
4
20.8
14.35
5
5.8
11.4
8.4
5
23.3
10.5
1.7
3.5
8.4
8.1
6
23.5
10.1
1.7
1.2
2.8
8.3
Average
21.4
12.8
3.6
4.7
9.5
8.2
Standard
Deviation
1.6
1.9
1.6
2
3.5
0.15
-------
the evaluation, determinations 1 through 4, the peak heights were con-
sistent. As the column aged, the peak height became less consistent for
_ _ o
certain species. This is shown with Br and NO, while SO, vlaues
indicate no loss. A column will also lose its ability to separate ions
because of buildup of trace metals and impurities on the column packing.
Upon further examination of the above problem, it was observed that
the slope of the calibration curve did not change, only the y-intercept.
This observation is shown in Figure 18. The results of Figure 18 means
that once a calibration curve is established, the analyst could continue
to use it even though the column may be aging. The analyst must check the
slope of the calibration curve by running two standards at the beginning
and end of the analysis. Uhen there is a change in the peak height of the
standard, an analyst can make the connection by using the equation for a
straight line:
y = mx + b
where y is the peak height, m is the slope of the curve, x is the concen-
tration of the species, and b is the y-intercept. The analyst finds the
y-intercept from the standard run on the day. The slope has already been
known for the calibration curve and the peak height is directly measured
from the ion chromatogram. By placing the corrected y-intercept into the
equation for a straight line, the concentration of the ion is determined.
5.2 VERIFICATION OF PRECISION AND ACCURACY OF ANALYTICAL SCHEME
It should be emphasized that changes in the column or flow conditions,
will cause changes in the qualitative and quantitative results obtained by
this technique. The analyst should monitor and compensate for changes
using the appropriate standards. The accuracy and precision of the
analytical scheme was tested using real samples from the lime/limestone
wet scrubber at the EPA/TVA Shawnee Test Facility in Paducah, Kentucky.
Table 5 shows a direct comparison between TVA and TRW sample analyses for
the lime/limestone wet scrubber. The relative error for each set of values
is listed. Whenever large (> 20%) errors were found, an alternate
analysis was employed as a check.
48
-------
24
22-
20
18
16
14
12
10
8
6
4
2
0
X
9
V
X
X
*
X
X
V
100
SO/2. CONDUCTIVITY SCALE - 100X
TUESDAY AND WEDNESDAY
SLOP - 0.06
y-INTERCEPT - -0.2 CM
THURSDAY
SLOPE * 0.06
y-INTERCEPT - -2.6 CM
200
300
400
Figure 18. Calibration Curve for Sulfate Ion Analysis
49
-------
TABLE 5. DIRECT COMPARISON BETWEEN TVA AND TRW SAMPLE ANALYSIS FOR THE VENTURI SPRAY TOWER SCRUBBER
in
O
Species
Calcium
Magnesium
Sodium
Sulfate
Chloride
Method of Analysis
Atomic Absorption (TVA)
Ion Chromatography (TRW)
Atomic Absorption (TVA)
Ion Chromatography (TRW)
Atomic Absorption (TRW)
Atomic Absorption (TVA)
Ion Chromatography (TRW)
Atomic Absorption (TRW)
Ion Exchange (TVA)
Ion Chromatography (TRW)
Potentiometric (TVA)
Ion Chromatography (TRW)
Concentration (ppm), Relative Error,
and TVA Sample Identification
#5253
604 ,
551 9/°
8219 yM
10506 gr
11025 OA
186 Ğ*
172 8%
32994
34000 J*
2676
2726 *
15254
624 ,
618 n
5269 ,y
5030 D7°
192 2%
197 1%
195 '*
17584 fiy
16209 a*
1418 ,~
1454 6%
#5255
618 ,y
614 l/0
4629 ,fiy
5536 16/0
oq 46%
yy 1%
102 J*
17700 £Ğy
16660 °*
1418 ,-
1485 5%
#5256
667 iqy
541 I9%
81 79 ooy
10014 ^£°
10490 D7°
92 48%
31442 qy
34404 y7°
2614 9,
2675 ^*
-------
The magnesium values reported by TRW differed greatly from the TVA
values and a second determination was carried out by atomic absorption at
TRW. The results of the atomic absorption analyses by TRW were in agree-
ment with the ion chromatography values. The magnesium results indicated
that there must be a shift in the solution equilibrium because of the time
difference between sampling at Paducah and analysis at TRW. A check of
the pH of the sample solutions showed that all the solutions decreased in
acidity to almost neutral or slightly basic. Figure 19 shows an ion
chromatogram of the magnesium/calcium analysis.
The chloride ion values were very consistent for both analytical
techniques employed. Chloride is completely ionized in solution and
should be less susceptible to changes in pH. The relative differences
between the reported values ranged from 1.8% to 4.5%.
Sodium ions are also completely ionized in solution and their values
should not vary with changes in pH. Only two of the four TVA determinations
for sodium showed good agreement with the TRW results. The two values
which disagree are approximately one-half less than the TRW values. Again,
atomic absorption was used as a check and the TRW values agreed with the
ion chromatography analyses. Figure 20 shows an ion chromatogram of the
sodium analysis.
Process streams which contain both sulfate and sulfite ions when
sampled will constantly increase in sulfate due to oxidation. The TRW
values for sulfate show both high and low values compared to the TVA
results, but both sets still showed good agreement. The variations between
the TVA and TRW analyses seem to be due to sampling and time and not
instrumental problems. Figure 21 shows an ion chromatogram of the
chloride/sulfate analysis.
Real samples could not be acquired from a dual alkali wet scrubber
because none were in operation. The results of the Shawnee tests would
indicate that ion chromatography is capable of analyzing samples from a
dual alkali process. Remembering that one-half of the dual alkali process
involves lime and/or limestone, these process streams are very similar to
the Shawnee scrubber streams and should be amiable to I.C. analysis.
51
-------
COLUMNS
ELUENT
SAMPLE
DIONEX CATION SEPARATOR 6 X 250 mm
DIONEX CATION SUPPRESSOR 9 X 250 mm
0.001M P-PHENYLENEDIAMINE
DIHYDROCHLORIDE AT 40% FLOW
1851 DILUTION FACTOR 1/100
100 XMg 8 mm 10,014ppm
3X Co 14 min 541 ppm
INJ
Figure 19. Ion Chromatogram of the Magnesium/Calcium Analysis
.52
-------
COLUMNS
ELUENT
SAMPLE
DIONEX CATION SEPARATOR 6 X 250mm
DIONEX CATION SUPPRESSOR 9 X 250 mm
0.005 N HNO3 AT 30% FLOW
1851 DILUTION FACTOR 1/100
No 10X 193 ppm 5.5 min
INJ
mm
Figure 20. Ion Chromatogram of the Sodium Analysis
.53
-------
COLUMNS
ELUENT
DIONEX ANION SEPARATOR 3 X 500 mm
DIONEX ANION SUPPRESSOR 6 X 250 mm
0.003 M NaHCOj/0.0025 M Na2CO3 AT 30% FLOW
SAMPLE
1851 DILUTION 1/1000
Cl" SOX 2.45ppm 3.5 min 2675 ppm
SO4~ 10X 15.45 ppm 13.5 min 34,404 ppm
INJ
Figure 21. ion Chromatogram of the Chior1de/Sulfate Analysis
54
-------
6. REFERENCES
Snyder, L. R. and Kirsland, J. J., Introduction to Modern Liquid
Chromatography, Wiley-Interscience, New York, 1974, P. 283.
Small, H., T. S. Stevens and W. C. Bauman, "Novel Ion Exchange
Chromatographic Method Using Conductimetric Detection." Anal. Chem.
47:1801 (1975).
Mulik, J. D., Todd, G., Estes, E., Puckett, R., Sawicki, E., and
Williams, D. "Ion Chromatographic Determination of Atmospheric
Sulfur Dioxide," Ion Chromatographic Analysis of Environmental
Pollutants. Sawicki, E., Mulik, J. B., and Wittgenstein E., Eds.
Ann Arbor Science, Ann Arbor, Mich., 1978, p. 23-40.
Blaedel, W. J. and V. W. Meloche, Elementary Quantitative Analysis.
Harper and Row, New York, 1963, p. 796-7.
55
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-79-151
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Chemical Analysis of Wet Scrubbers Utilizing Ion
Chromatography
5. REPORT DATE
July 1979
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Tobias R. Acciani and Ray F. Maddalone
8. PERFORMING ORGANIZATION REPORT NO
9. PERFORMING ORGANIZATION NAME AND ADDRESS
TRW Defense and Space Systems Group
One Space Park
Redondo Beach, California 90278
10. PROGRAM ELEMENT NO.
INE624
11. CONTRACT/GRANT NO.
68-02-2165, Task 214
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Task Final; 9/77 - 9/78
14. SPONSORING AGENCY CODE
EPA/600/13
15 SUPPLEMENTARY NOTES jERL-RTP project officer is Frank E. Briden, MD-62 919/541-
2557.
16. ABSTRACT
The report describes the key elements required to develop a sampling
and analysis program for a wet scrubber utilizing ion Chromatography as the main
analytical technique. The first part of the report describes a sampling program for
two different types of wet scrubbers: the venturi/spray-tower (limestone) scrubber
which is part of the EPA/TVA Shawnee Test Facility, Paducah, Kentucky; and the
Arthur D. Little dual alkali pilot scrubber. The sampling section of the report des-
cribes the scrubber type, sample collection, and sample preservation. The analysis
section of the report, describing the theory and practice of ion Chromatography as
applied to wet scrubber samples, describes analytical procedures , qualitative analy-
sis, and quantitative analysis. The analytical procedures portion covers the anion
and cation analysis scheme for wet scrubber samples, the theory of ion Chromato-
graphy, and current problems involving the use of ion Chromatography for wet scrub-
ber related samples.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COSATI Field/Group
Pollution
Chromatography
[onization
Scrubbers
Analyzing
Calcium Carbonates
Alkalies
Pollution Control
Stationary Sources
Ion Chromatography
Dual Alkali System
13 B
07D
07B,07C
07A,13I
14B
IBU'
FEME!
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (Thispage)
Unclassified
21. NO. OF PAGES"
61
22. PRICE
EPA Form 2220-1 (ğ-73)
56
------- |