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.       -I—I.
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

-------