EPA-650/2-74-024
March 1974
                           Environmental Protection Technology Series




-------
                               EPA-650/2-74-024
DEVELOPMENT OF  SAMPLING
AND ANALYTICAL METHODS
      OF LIME/LIMESTONE
    WET SCRUBBING  TESTS
                  by

        K. Schwitzgebel, F. B. Meserole,
  C. M. Thompson, J. L. Skloss, and M. A. McAnally

             Radian Corporation
          8500 Shoal Creek Boulevard
            Austin, Texas 78766
           Contract No. CPA 70-143
            ROAP No. 21ACY-25
          Program Element No. 1AB013
      EPA Project Officer:  Robert M. Statnick

          Control Systems Laboratory
      National Environmental Research Center
    Research Triangle Park, North Carolina 27711
               Prepared for

     OFFICE OF RESEARCH AND DEVELOPMENT
    U.S. ENVIRONMENTAL PROTECTION AGENCY
          WASHINGTON, D.C.  20460

               March 1974

-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication.  Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
                                  11

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
                         ACKNOWLEDGEMENTS

           The authors wish to acknowledge the assistance of
 OAF personnel under whose  guidance this program was  carried out.
 Mr. Julian Jones was ORD's Project Officer from the  beginning
 of the contract until December, 1971.  Dr. Robert  Statnick (ORD)
 directed the program starting in January, 1972.  We  appreciate
 the cooperative spirit  of  both project officers.

-------
                  8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 .454.4797
                           ABSTRACT

          This study was carried out to develop appropriate
sampling and analytical methods to be used at OAF's  test
facility at Shawnee.   The three problem areas encountered  in
analyzing thermodynamically unstable slurry streams  as  encoun-
tered in lime/limestone based S02 wet scrubbing processes are
sampling, sample handling and chemical analysis.  A  positive
pressure filtration was found to minimize the mass transfer
phenomena during the filtration step to an acceptable  level.
Quenching of the filtered liquid was chosen to avoid change  in
sample composition.  Two sets of analytical methods  were
selected for application at Shawnee.  The back-up methods are
based on atomic absorption and wet chemical procedures.  The
rapid field methods are based on X-ray fluorescence, atomic
absorption, and wet chemical analyses.

          The X-ray fluorescence spectrometer was automated  by
interfacing it with a NOVA 1200 minicomputer.  Additional
peripheral devices have the function of processing all raw  data
The raw data are input to the system with a card  reader, a
teletype, or a CRT.  The final results are stored on a magnetic
tape.  A hard copy is provided by a printer.

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
                        TABLE OF CONTENTS
                            VOLUME  I
                                                       PAGE
 1.0       INTRODUCTION 	     1

 2.0       PROBLEM DEFINITION  	     3

 3.0       SAMPLING 	     8

 4.0       SAMPLE HANDLING	    11
                                                    • •
 5.0       LIQUID PHASE CHARACTERIZATION  	    12
 5.1       Wet Chemical Procedures  	    13
 5.1.1     Chloride Determination  	    13
 5.1.2     COS Determination 	    17
 5.1.3     Sulfur Dioxide Determination  	    19
 5.1.4     Total Sulfur Determination  	    19
 5.1.5     Total Nitrogen 	    21
 5.1.6     Determination of Nitrite and Nitrate  ....    22
 5.2       Atomic Absorption Procedures  	    23
 5.2.1     Determination of Ca, Mg, K, and  Na  	    26
 5.2.2     Determination of Catalytically Effective
           Trace Elements 	    27
 5.3       X-Ray Fluorescence  	    28
 5.3.1     Physical Phenomena  	    28
 5.3.2     Description of Available Equipment  	    29
 5.3.3     Preliminary Tests 	    29
 5.3.4     Description of Sequential
           Instrumentation  	    30
 5.3.5     Matrix Interference Corrections  	    32

-------
                  S500 SHOAL CREEK BLVD. • P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -451-1797
                                                      PAGE
5.3.6     Summary of Selected  Values  for Matrix
          Interference Coefficients and
          Associated Uncertainties  	   36
5.3.7     Calibration Procedures  for  X-Ray
          Fluorescence Spectrometry  	   36
5.3.8     Mathematical Background for Computer
          Calculation of  Calibration  Parameters ...   39
5.3.9     Fluorescence Counting Rate  Measurements  .   39

6.0       SOLID PHASE CHARACTERIZATION 	   41
6.1       Chemical Composition 	   41
6.2       Phase Identification 	   42

7.0       FIELD STUDIES  	   44

8.0       USE OF THE RAW  DATA  	   53

9.0       DATA HANDLING SYSTEM 	   57
9.1       Laboratory Data Analysis Hardware 	   59
9.2       Laboratory Data Analysis Software .,	   60
9.2.1     Executive System  	   60
9.2.2     Application Routines 	   61
9.2.3     Diagnostic Routines  	   61

10.0      SUMMARY 	   63

11.0      BIBLIOGRAPHY 	   67

-------
                  ssooSHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
                                                      PAGE
                            VOLUME II

1.0       INTRODUCTION  	    1

2.0       PROCESS DESCRIPTION AND PROBLEM
          DEFINITION  	    3
2.1       Shawnee Test  Facility  	    4
2.2       Process Chemistry 	    7
2.3       Required  Procedures 	   13

3.0       LIQUID PHASE  ANALYSIS  	   15
3.1       Literature  Review* 	   16
3.2       Experimental  Evaluation of Atomic
          Absorption  Spectrophotometry* 	  179
3.3       Experimental  Studies of X-Ray Fluorescence
          Spectrometry* 	  251
3.4       Experimental  Evaluation of Methods for
          Individual  Species* 	  314
3.5       Selected  Referee Methods for Liquid
          Phase Analyses* 	  421
3.6       Selected  Field  Methods and Data Analysis
          System  	  485

4.0       SOLIDS CHARACTERIZATION 	  491
4.1       Phase Identification Using X-Ray
          Diffraction 	  497
4.1.1     Theory of Powder Diffraction 	  497
4.1.2     Instrumentation 	  500
4.1.3     Compilation of  X-Ray Powder Diffraction
          Data  	  504
*  Detailed  contents  of these sections are given at the front
of these  sections.
                               vi

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. 80X9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512 - 454-4797
                                                       PAGE
 4.2        Methods to Measure Particle Size and
           Surface Area 	   511
 4.2.1     Phase Separation 	   512
 4.2.2     Methods for Particle Size Determination  .   513
 4.2.3     Surface Area and Pore Size
           Determination 	   526
 4.3        Solids Dissolution and Analysis Methods  .   533
 4.3.1     Analysis for Calcium, Magnesium and
           Total Sulfur 	   533
 4.3.2     Analysis Method for the Determination  of
           Total Carbonate in Solids 	   540
 4.3.3     Analysis for Sulfite 	   546

 5.0        SAMPLING TECHNIQUES 	   548
 5.1        Theory of Sampling Fluid Phase Streams  ..   549
 5.2        Recommended Procedure for Sampling and
           Rapid Separation of Unstable Slurries  ...   556
 5.2.1     Pump 	   556
 5.2.2     Filter Holder and Membrane  	   558
 5.2.3     Sample Train 	   558
 5.2.4     Procedure 	   559
 5.3        Recommendations for Collecting Liquid
           Samples and Fixing Unstable Solutions  ...   560
 5.3.1     Fixing Solution for Carbon Dioxide
           Analysis 	   561
 5.3.2     Fixing Solutions for Sulfite Analysis  ...   564
 5.3.3     Fixing Solutions for Calcium Analysis  and
           for Sulfate (Total Sulfur) Analysis  	   564
 5.4        Field Sampling 	   566
                                Vll

-------
Radian Corporation
8500 SHQAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 - 454-4797
                                                      PAGE
6.0        FIELD TESTS OF SELECTED METHODS 	  567
6.1        GAP In-house Test Data Conditions  	  569
6.2        Tidd Plant  Test Data 	  573
6.2.1      Process Description for the Tidd Plant
           Scrubbing Unit 	  573
6.2.2      Sampling Procedures 	  577
6.2.3      Analytical  Methods 	  579
6.2.4      Results 	  580
6.3        Key West Test Data 	  592
6.3.1      Pilot Unit  and Test Conditions  	  592
6.3.2      Results of  Chemical Analyses of the
           Liquid Phase 	  596
6.3.3      X-Ray Diffraction Results  	  599
6.3.4      Results of  the Chemical Analyses of Key
           West Solids 	  609
6.4        Colbert Test Data 	  613
6.4.1      Pilot Unit  and Test Conditions  	  613
6.4.2      Results of  Chemical Analyses of the
           Liquid Phase 	  617
6.4.3      X-Ray Diffraction Results  	  621
6.4.4      Results of  the Chemical Analyses of
           Colbert Solids 	  630
6.5        Shawnee Test Data 	  635
6.5.1      Pilot Unit  and Test Conditions  	  635
6.5.2      Results of  Chemical Analyses of the
           Liquid Phase 	  637
6.5.3      X-Ray Diffraction Results  	  637
6c5.4      Results of  the Chemical Analyses of
           Shawnee Solids 	  647
                                viii

-------
                       S500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797










                                                                    PAGE







7.0         SUMMARY  	    650







8.0         BIBLIOGRAPHY  	    652

-------
                   aSOO SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -454-4797
                                                       PAGE
                           VOLUME  III
1.0        INTRODUCTION 	

2.0        EXECUTIVE SYSTEM
3.0       APPLICATION SOFTWARE  	    10
3.1       X-Ray Operation  	    10
3.2       System Commands  	    27

4.0       DIAGNOSTIC TEST ROUTINES  	    69

5.0       DATA STORAGE AND INPUT  	    74

          APPENDIX A - PROGRAM  WRITE-UPS 	   105

          APPENDIX B - PROGRAM  LISTINGS 	   182

          APPENDIX C - CONSIDERATIONS FOR SOLVING
             X-RAY FLUORESCENCE  MATRIX CORRECTIONS  .   354

-------
                  ssoo SHOAL CREEK BLVD. •  P.O. BOX 9943 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
1.0       INTRODUCTION

          The Office of Research and Development has  sponsored
several approaches for sulfur dioxide removal from  flue  gases
emitted by coal and oil-fired power stations in the past years.
One of the most advanced control strategies is S03  removal  based
on lime/limestone wet scrubbing techniques.  Since  the summer
of 1972, three different scrubbing units were tested  at  Shawnee
a venturi scrubber, a turbulent contact absorber, and a  marble
bed.  The goal of these tests was to demonstrate the  long term
reliability of these units and the extraction of engineering
design information such as:

             vapor-liquid mass transfer characteristics
             in the scrubbers,

             solid-liquid mass transfer rates through-
             out the system, and

             scaling potential.

The mathematical description of these problem areas depends on
the knowledge of equilibrium partial pressures and  important
activity products.  These quantities can be calculated  from the
chemical composition of the scrubber solutions obtained  by
chemical analysis.

          Radian was granted a contract in  1970  to  select ap-
propriate referee and field chemical analysis methods to be
used at Shawnee.  This final report consists of  three volumes.
Volume I summarizes the major findings of  the  literature and
experimental results in broad terms.  It starts  with  a  short
process description and problem definition.  The problem areas

-------
                  8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 . TELEPHONE 512 -454-4797
encountered in sampling and sample handling  are  then  discussed.
Analytical methods selected for  the  liquid and solids analyses
are described in the following sections.  The data  collected
at TVA's Colbert Steam Plant  are presented as an example of
methods testing in the field.  The analytical results obtained
from the scrubber effluent are further  processed to demonstrate
the extraction of equilibrium partial pressures  and important
activity products necessary for  process evaluation.

          X-ray fluorescence  proved  to  be a  rapid and accurate
procedure for the determination  of sulfur in aqueous  scrubber
samples.  Calcium, chlorine,  and potassium are additional ele-
ments  that are detectable by  X-ray fluorescence.  The X-ray
spectrometer was interfaced with a minicomputer  and several
peripheral devices such as  teletype, CRT, card reader, disk,
magnetic tape, etc.,  to automate the fluorescence analysis and
to facilitate the data handling  problem associated with data
reduction.

          Volume II  is a  detailed  description of the  literature
findings and experimental effort leading to  the  methods of
choice.  Details of  the automated  X-ray fluorescence  unit and
the data handling systems are presented in  Volume III.
                               -2-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • A USTIN. TEXAS 78766 • TELEPHONE 512 • 454.4797
 2.0        PROBLEM DEFINITION

           The basic equipment arrangement for limestone injec-
 tion  wet  scrubbing (LIWS)  processes is shown in Figure 2-1.
 The  three streams entering the system are flue gas, particulates
 and make-up water.  Three  streams leaving the unit are cleaned
 stack gas,  solid  waste products, and scrubbing liquor.  The
 composition of the incoming streams provides a means of predict-
 ing the  liquor composition on a qualitative basis.  The important
 species  in  the LIWS process are:

   Group  I              Group II                 Group III
   Calcium              Sodium                 Trace elements
   Sulfite              Potassium              Iron
   Sulfate              Magnesium              Cobalt
                        Chloride               Nickel
                        Nitrate                Copper
                        Nitrite                Manganese
                        Carbonate

          The components  listed in Group I are the most important.
 They  dominate the process  by participating in the gas-liquid
 and liquid-solid  mass  transfer steps.   The species listed under
 Group  II  contribute to the process performance in three ways.
 First, they  influence  solubilities which are dependent on the
 ionic  strength of the  solution.   Second, they form ion pairs
 with  Group  I  compounds.  Finally,  they influence  the driving
 force  for the mass transfer rates.  The components in this group
 form  very soluble compounds with the exception of magnesium
 hydroxide and calcium  carbonate.   In a closed loop operation
 there  is  a  buildup of  the  soluble  compounds, since the only
                               -3-

-------
GAS SPECIES
   FG, SG
 I. S02
 2. C02
 3. NOX
 A. HZ0
 5. 02
 6. CO
 7. N2
FLUE GAS
  FG
      LIMESTONE
       FLY  ASH
       SOLIDS
         LA
       t. CoO
       2. MgO
       3. CoS04
       4. MgS04
       5. CoS03
       6. MgSOj
       7. CoC03
       8. MgC03
       9. FLY ASH
      10. SOLUBLE No
      II. SOLUBLE Cl
STACK GAS
    SG
     I
              VTiiTEfT
              MAKEUP
                WM
 SCRUBBER
     S
SCRUBBER FEED
                                         SF
                                   'Li
 PROCESS
  WATER
HOLD TANK
    P
      SCRUBBER
      BOTTOMS
        SB
                                              SLURRY RECYCLE  SR
 SCRUBBER
 EFFLUENT
HOLD TANK
    E
                                                      CLARIFIER
                                                        LIQUID
                                                         CL
    CLARIFIER
      FEED  _
       CF
CLARIFIER
    C
                                   CLARIFIER
                                   BOTTOMS
                                  * CB
                                FILTER
                                  F
                             FILTER
                             LIQUID
                              FL
                                   FILTER
                                   (BOTTOMS
                                     FB
                                        PROCESS  SOLID  SPECIES
                                          (CF, SR, CB, FB.SF]
                                   I. CoO
                                   2. Co(OH)2
                                   3. Co CO 3
                                   4. CcS03 • xH20
                                   5. CoSO-; • xH20
                            6. MgO
                            7. Mg(OH)2
                            8. MgC03-xH20
                            9. MgS03 • xH20
                            10. FLY ASH
   PROCESS
   LIQUID
   SPECIES
SB.CF.SR, CB,
FB.CL.FL.SF

  I.H*
 2. OH~
 3. HS03
 4. SOf
 5. SO?
 6. HC03
 7. COf
 0. HSOJT
 9. H2S03
 10. H2C03
 II. Co++
 12. CoOHf
 13. CoS03
 14. CoCOj
 15. CoHCOj
 16. CoSO/,
 17. CoKOj
 10. NOj
 19. Mg+ +
20. MgOH*
21. MgSO<»
22. MgHCOj
23. MgS03
24. MgCOj
25. No +
26. NoOH
27. NoC03
28. NollCOj
29. NoS04
30. NoNOj
31.cr
                      FIGURE 2-1   -  WET SCRUBBING  SCHEME
                                         -4-

-------
                  8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 . TELEPHONE 512 -454-4797
stream in which they can leave the scrubbing unit  is  the  liquor
adherent to the solids.  This fact must be kept  in mind when
selecting analytical methods.  The procedures must give accurate
results in those cases where the soluble species build up  to
a high level.  The implications for the selection  of  methods
for sulfate and sulfite will be discussed later.

          The third group is comprised of species  leached  from
the fly ash and impurities in the limestone.  The  concentration
of these elements is never very high, since it  is  limited  by
the solubility of the hydroxides in the alkaline parts of  the
scrubbing unit.  Their importance is based on the  fact that
they are catalysts for sulfite oxidation, even  if  present  in
the parts per billion range.

          The process simulations, performed under CPA Contract
No. 70-45, "Study of the Limestone Injection Wet Scrubbing
Process," gave a valuable basis for estimating  anticipated con-
centration ranges.  Estimation was necessary since no data on
a closed loop system operated over an extended  period of  time
were available at the time of analytical method  development.

          As a general rule, the higher the accuracy  demand of
an analysis, the higher are its costs.  This fact  raises  the
question as to the ultimate use of the analytical  results. The
accuracy requirements for routine, day-to-day operation are less
stringent than the requirements for process analysis.  One key
objective of the tests at Shawnee is the collection of engineer-
ing design information.  From an engineering point of view the
following areas are of ultimate interest:

             gas-liquid mass transfer rates in
             the scrubber
                              -5-

-------
                  SSOO SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
             dissolution and precipitation rates
             as function of liquor composition

             scaling potential

          The driving force term in the mass  transfer  equations
describing these rates is a function of the difference of  the
actual process conditions and the equilibrium conditions of the
system.  In other words, the rates are a  function  of  the dif-
ference of two activity expressions.  The closer the  system
approaches equilibrium the more severely  analytical errors will
influence rate correlations.  For LIWS processes the  analyses
of the species listed in Group 1 are, therefore, the  most
important.  Error propagation calculations showed  that the error
in these analyses should not be greater than  about 270.  The
concentration of the species influencing  the  ionic strength
(Group II) must be  known within about 470. The  accuracy require-
ments for the trace elements effective as catalysts are still
less stringent.  Twenty to fifty percent  is considered to  be
sufficient.

          An example of how to calculate  partial pressures and
important solubility (activity) products  from chemical analysis
data is shown in Chapter VIII of this volume.  These  data  are
necessary inputs to evaluate scrubber performance  from a chemical
engineering point of view.

          The analytical results are  influenced by three steps:

             solid-liquid  separation

             sample handling
                               -6-

-------
                     S500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 •  TELEPHONE 512-454.1797
               actual  analysis





These problem  areas  will  be discussed  next
                                     -7-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN . TEXAS 78766 • TELEPHONE 51 2 - 454-4797
 3.0       SAMPLING

           The scrubbing system can be divided into an acidic
 and a basic part.   The environment is acidic in the scrubber
 itself and in the  pipe between the scrubber and the effluent
 hold tank.  The solutions circulated in the rest of the system
 are alkaline.  For sampling purposes it should be noted that
 the scrubbing slurry, especially in the acidic part of the
 system, is not in  thermodynamic equilibrium.  The sorbent tends
 to dissolve and sulfite and sulfate tend to precipitate.  The
 technique often used to sample this stream is collection of a
 slurry sample in a beaker and filtration through a Buchner
 funnel.  This technique results in only semi-quantitative results
 for the follow-on  chemical analysis for three reasons:

           1.  Loss of acidic gases (SOS, C0a)
               especially if a vacuum is used.

           2.  Solid-liquid mass transfer during
               the  sampling procedure.

           3.  Sulfite oxidation by air oxygen.

 Because of these sources of error much of the pilot plant data
 collected using this method must be considered to be qualitative
 in nature and not  suitable for the extraction of engineering
 design information.  In-line, positive pressure filtration was
 the sampling method selected after field tests at several pilot
 units (see Figure  3-1).  The sampling apparatus consists of a
 positive pressure  pump, a membrane filter holder and lines and
 valves to control  sampling and purge rates.  Flow rates used in
 the tests were about 1300 ml/min.  The residence time of the
 slurry is about 2.3 seconds in the filter and approximately
 seven seconds in the entire sampling equipment.

                               -8-

-------
          Reagent Reservoirs
Process
Stream
                                                -{Xh
                             Positive
                           Pressure Pump
                Filter Holder
\
                                        Purge to
                                         Drain
          Sample
        Container
       FIGURE 3-1  -  TYPICAL  SLURRY SAMPLING TRAIN

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • PO. BOX 9948 • AUSTIN. TEXAS 78766 •  TELEPHONE 512-454-4797
           The degree of mass transfer  in  the  filter cake, which
 is  by nature a good contacting device, was  checked by taking
 consecutive samples and plotting the chemical analysis results
 as  a function of the filtered volume.  Extrapolation to zero
 volume of filtrate represents the true aqueous phase composition,
 With the exception of carbonate, the amount of solids dissolved
 or  precipitated in the filter cake was within the experimental
 error of the chemical analyses.

           Loss of acidic gases is avoided by  the  positive pres-
 sure filtration, and air oxidation of  sulfite is  prevented by
 fixing the sample immediately.
                               -10-

-------
                  esoo SHOAL CREEK BLVD  • P.O BOX ssas • AUSTIN. TEXAS 78766 • TELEPHONE 512 -454-1797
4.0       SAMPLE HANDLING

          After filtration care mast  be  taken that the liquid
samples do not undergo  further change.   This  is especially true
for the sulfite analysis.  Sulfite  losses  can occur by:

             Evaporation from acidic  samples

             Oxidation  by air oxygen

             Interaction with nitrites

All three sulfite  losses can be avoided  by quenching the sample
in a solution of pH = 6 with known  iodine  content (see Chapter
5.3.2 of Volume II).  Nitrites can  be formed  by absorption of
NO and N02 from the flue gas.  The  degree  of  NO*  absorption
was unknown at the time of method  selection.

          Carbonate losses from acidic  liquid can be avoided
by quenching the sample in a solution of pH - 10.  EDTA must
be added to the buffer  in order to  avoid calcium carbonate
precipitation at this pH (see Chapter 5.3.1 of Volume II).

          Sulfate  in the presence  of  sulfite  is determined as
the difference between  the total sulfur  and the sulfite sulfur.
In order to avoid  sulfite losses and  sulfate  precipitation, the
sample for total sulfur analysis is quenched  in a H3 Og-water
solution.  Hydrogen peroxide oxidizes the  sulfite.  Dilution
with distilled water prevents sulfate precipitation in the
sample bottle (see Chapter 5.3.3 of Volume II).
                               -11-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 51 2 - 454-4797
 5.0       .LIQUID PHASE CHARACTERIZATION

           The literature was surveyed through 1970 for analytical
 methods which might be applicable to the solutions of interest.
 The sources consulted were:

           1.  Kolthoff and Elving, "Treatise on
               Analytical Chemistry"

           2.  Biannual Reviews on Analytical
               Chemistry

           3.  1969 Book of ASTM Standards

           4.  FWPCA Methods  for Chemical Analysis
               of Water and Wastes

           5.  Chemical Abstracts

           6.  Pertinent Original Articles

           The present chapter discusses the wet chemical
 procedures for the determination of chloride, C03 in aqueous
 solution,  sulfite in aqueous solution, total sulfur as sulfate,
 total nitrogen and for the analysis of nitrite and nitrate.
 The methods chosen are applicable for highly concentrated
 scrubber solution as encountered in closed loop operation.

           Atomic absorption  proved most suitable for the analysis
 of calcium, magnesium, sodium, and potassium.  The catalytically
 active transition elements,  iron, manganese, cobalt, nickel,
 and copper, are determined after chelation with diethyl-dithio-
 carbatnate  and extraction into methyl isobuthyl-ketone.
                               -12-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -454-4797
           X-ray fluorescence proved to be an  accurate  and  rapid
 procedure for the determination of total sulfur, which is  a
 very critical analysis in limestone based wet  scrubbing solutions
 In addition, calcium, chlorine, and potassium  in the aqueous
 phase can be analyzed by this approach.

           The methods selected are described  in this chapter
 in a succinct form.  The results of the literature  survey  and
 a detailed description of the interference studies  performed
 experimentally are presented in Volume II of  this final report.
 5.1       Wet Chemical Procedures

 5.1.1     Chloride Determination

           A summary of the relevant methods for determination
 of  chloride ion in aqueous solution, based on  the  literature
 survey,  is given in Table 3.1-6 of Volume II.  The methods
 covered  in this table fall into one of the following  broad
 categories.

           1.   Gravimetric as silver chloride

           2.   Volumetric (visual end point)

           3.   Spectrophotometric

           4.   Nephelometry

           5.   Flame Photometry

           6.   Potentiometric measurements
                               -13-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 • 454-4797
           7.   Amperometrie (current) measurements

           8.   Coulometric measurements

           Most chloride determinations are based on the reaction
 of  chloride ion with either silver (I) or mercury (II).  The
 methods  proposed by Volhard,  the mercurimetric .titration for
 chloride and  the potentiometric method described by Shiner and
 Smith (SH-014, FI-019) were tested in the laboratory.   The
 Volhard  method was  found to yield good results down to a chloride
 concentration of 0.02 M (Volume II,  page 315).  The mercurimetric
 procedure suffered  interferences from iron ion found in some
 pilot plant samples (Volume II, page 327).

           The potentionmetric determination of chloride proved
 to  be the most satisfactory wet chemical procedure (Volume II,
 page  319).  A Fisher Automatic Titralyzer (Model 740)  was used
 to  check this procedure.   This is an automated potentiometric
 titrator designed to do volumetric analyses fully automatically.
 It  incorporates  into a single, integrated system an electrometer,
 an  automatic  buret,  a digital data recording system, and an
 automatic sample  changing device.

           The potential between a metallic silver indicator
 electrode and a  silver-silver chloride reference electrode is
 measured  as a function of the amount of standard silver nitrate
 solution  added.   The reaction of silver nitrate with chloride
 ion may  be  represented in the following way.

           Cl"  + Ag+ ->  AgCl  (solid)                         (5-1)

 The potential  of  the  metallic silver electrode depends  on the
 amount of  silver  ion  in solution.
                              -14-

-------
RSQ 131*1 COrpOrdtlOn  8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -454-4797
           In practice the potential at the equivalence  point
 is  determined by a manual titration or by addition of a  known
 amount of standard silver nitrate solution to a known amount
 of  a standard chloride solution.  This known equivalence point
 potential is then set as the end point potential on  the  Titra-
 lyzer and all other titrations are stopped when this potential
,is  reached.

           Interferences

           Bromide and iodide will be determined as equivalent
 chloride concentrations.  Ferricyanide causes high results  and
 must be removed.  Chromate and dichromate interfere  unless
 reduced to the chromic state.  Concentrations of ferric  iron,
 if  substantially higher than the amount of chloride, will
 interfere.  Ferrous ion and phosphate do not interfere.

           Results of Chloride Titrations

           Accuracy and precision are excellent.  The relative
 error using artificial scrubber samples never exceeded  .5%  in
 a series of 15 titrations.

           This method for determining chloride was also applied
 to  field samples.  For the Key West series of samples,  triplicate
 titrations were run for each sample.  Recovery tended to run  a
 little low ranging from 98.2% to 99.8% for 22 samples.

           For the Colbert Test Series one unspiked aliquot  and
 one spiked aliquot were run for each sample.  The results of
 the recovery studies are given in Table 5-1.  Recoveries ranged
 from 99.4% to 101.8% for 15 samples.
                               -15-

-------
Radian Corporation
6500 SHOAL CREEK BLVD. • P.O. BOX 9948  • A USTIN , TEXAS 78766 • TELEPHONE 51 2 - 454-1797
                                     TABLE 5-1
RECOVERY STUDIES


Sample Name
Scrubber Effluent
Scrubber Spray
F-12 Overflow
F-13 Recycle
Clarifier Overflow
Limestone Feed
Make Up Water
Scrubber Effluent
Scrubber Spray
F-12 Overflow
F-13 Recycle
Scrubber Effluent
Scrubber Spray
F-12 Overflow
F-13 Recycle
ON COLBERT CHLORIDE DETERMINATIONS
(1)
Titration
of
Unspiked
Aliquot
ml AgN03
2.22
.32
1.91
1.88
2.15
1.81
.05
2.38
.53
2.30
2.18
2.30
.49
4.86
4.56
(2)
Titration
of
Spiked
Aliquot
ml AgN03
7.26
5.32
6.92
6.95
7.12
6.82
5.06
7.44
5.50
7.32
7.19
7.36
5.49
9.90
9.65


Difference
(2)-(l)
5.04
5.00
5.01
5.07
4.97
5.01
5.01
5.06
4.97
5.02
5.01
5.06
5.00
5.04
5.09


Percent
Recovery
100.8
100.0
100.2
101.4
99.4
100.2
100.2
101.2
99.4
100.4
100.2
101.2
100.0
100.6
101.8
                                         -16-

-------
                  SSOO SHOAL CREEK BLVD. •  P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -451-4797
          The recoveries are all well within  the  accuracies
required for chloride determinations and  indicate that no
serious interferences for  this method of  chloride determination
were found in these sets of field  samples.
5.1.2     C0a Determination

          The literature  findings  on  methods  used for measuring
COS content  in  liquid and  solid  samples  are  summarized in
Chapter 3.1.8 of Volume II.   The procedures  described can be
grouped into one of  the following  categories:

             Volumetry

             Colorimetry

             Gravimetry

             Evolution  techniques

An instrument based  on  C02 evolution  in  an acid pool with sub-
sequent CO.,  detection by  a nond isper sive infrared analyzer was
found to be  suitable for  C0a  determinations  in aqueous solutions
(Total Carbon System, Oceanography International).  Interferences
caused by nitrates,  nitrites,  and  sulfite could be overcome by
using a 10%  KR, P04 solution  in the reactor.

          Accuracies of better than two  percent were obtained
in analyzing synthetic  scrubber  solutions.  Table 5-2 gives
supporting data.   The analysis time per  sample ranges from five
to six minutes.
                               -17-

-------
                                                TABLE 5-2
 Run
 No.
  1
  2
  3
  4
 i
»-•
T5
  6
  7
  8
  9
 10
   me
Injected
  46.4
  96.1
 146
 146
 146
 146
 146
 146
 146
 146
INFLUENCE
OF NITRITE, NITRATE AND SULFITE ON THE TOTAL
USING THE "TOTAL CARBON
NO;
mtration
ss/Liter
0.5
0.5
0.5
0.5
	
«__
0.5
	
___
0.5
NO;
Concentration
Moles/Liter
0.5
0.5
0.5
	
0.5
	
	
0.5
	
.0.5
so;
Concentration
Moles/Liter
0.05
0.05
0.05
	
	
0.05
	
	
0.05
0.05
SYSTEM" CQce
Actual C0a
Concentration
mg/Liter
100
100
100
100
100
100
	
	
	
	
C02 ANALYSIS

:anography International)
Experimental C03
Concentration Percent
- mg/Liter Error
102.0
100.6
100.2
102.0
100.5
101.0
0.0
0.0
0.0
0.4
- 2.0
- 0.6
- 0.2
- 2.0
- 0.5
- 1.0
0.0
0.0
0.0
- 0.4

-------
                  KOO SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
5.1.3     Sulfur Dioxide Determination

          The literature finding  concerning methods for
determination are summarized  in Volume  II,  page 113.   The
procedures are based on:

             Polarography

             Atomic absorption spectroscopy

             Fluorescence

             Visible and ultraviolet  spectro-
             photometry

             Amperometry

The amperometric procedure  was found  to be  the most straight-
forward approach.  The sample is  added  to an excess iodine
solution buffered to pH = 6.0 - 6.2  to  inhibit sulf ite-nitrite
and nitrite-iodine interaction  (DE-029, SE-015).  Iodine remain-
ing after stoichiometric S0a  oxidation  is titrated with standard
sodium arsenite solution.   An amperometric  dead stop method for
end point detection is used.   The error is  2-4% in the presence
of 20 mrnoles nitrite.


5.1.4     Total Sulfur Determination

          The methods described  in the  literature for the
determination of sulfate in aqueous  solutions are summarized
in Volume II, Chapter 3.1.10.
                              -19-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIM. TEXAS 78766 • TELEPHONE 512 - 454.4797
           The published procedures can be divided into five
 groups :

           1.   Gravimetric procedures

           2.   Direct titrimetric procedures
               •
           3.   Indirect titrimetric methods

           4.   Colorimetric techniques

           5.   Acidimetric methods

 The  most elegant method is the titrimetric approach using thorin
 as  end point  indicator (AM-002, MA-039, FR-003, FR-009).  Un-
 fortunately,  this method is subject of severe anion interference
 caused by the presence of phosphate, fluoride, nitrate, and
 chloride.

           The barium chloranilate spectrophotometric procedure
 was  extensively  checked (BE-024, PR-007, FE-004).   It is based
 on  the precipitation of barium sulfate upon interaction of
 barium chloranilate with sulfate ion.  The intensity of the
 colored  chloranilate is measured at 530 nm.  This  procedure is
 amenable to automation (Volume II, page 366).  Nitrate and
 chloride showed  interferences if present in high concentrations.
 The  method showed,  in addition, pH-sensitivity and problems in
 the  removal of the  very fine precipitate.

           Barium sulfate precipitation and backtitration of
 excess barium using EDTA and Eriochrome Black T as indicator
 gave satisfactory results (Volume II, page 358).  Interfering
            3+    3+
 cations  (Fe  , Al  , Ca and Mg) are removed using a cation
                               -20-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9918 « AUSTIN. TEXAS 78766 • TELEPHONE 512 - 454-4797
 exchange resin in the sodium form.  A disadvantage of the
 procedure is the long digestion time (KO-015, TH-007, SI-005,
 SH-006).

           The titrimetric procedure proposed by Dollman (DO-006)
 was  finally accepted (Volume II, page 383).  A sample aliquot
 is  passed through a strong acid type ion exchange resin in  the
 hydrogen form.  Sulfate and other anions are converted to the
 corresponding acids.  The column effluent is quantitatively
 retained.  All acids except H^ S04 and Hg P04 are volatilized
 at  75°C.  Titration with a standard base completes the deter-
 mination.  Phosphate, if present, interferes.  It can be deter-
 mined by the same technique by volatilizing the HgS04.  The
 sulfate value is then corrected.  The accuracy of the method,
 if  carefully applied, is better than 17» (see Tables 3.4-13,
 3.4-14, 3.4-15 of Volume II).
 5.1.5     Total Nitrogen

           The nitrogen containing compounds found in lime/
 limestone based scrubbing solution can be quantitatively reduced
 to ammonia in the presence of a metallic reducing agent (Volume
 II,  page 138).   Devarda's alloy (50% Cu, 4570 Al, and 5% Zn)  in
 a  strong NaOH solution was found to be satisfactory (EN-020,
 KO-050,  MU-020).   The reduction cannot be performed in an acidic
 medium due to hydrogen sulfide evolution.  Ammonia is distilled
 into hydrochloric acid.  Excess HC1 is backtitrated.

           The error determined in the presence of sulfite was
 found to be smaller than 170 at nitrate levels of 100 to 160  mg
 in the aliquot  (Volume II, page 399).
                               -21-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9918 •  AUSTIN. TEXAS 78766 • TELEPHONE 51 2 - 45*1-4797
 5.1.6     Determination  of  Nitrite and Nitrate

           The literature  findings  for the quantitative analysis
 of nitrites are presented in  Volume II, Chapter 3.1.12.  Table
 3.1.11 shows examples of  the  principal types of methods which
 can be used to determine  nitrite  in aqueous media.  They can
 be classified in the following  categories:

           *  Ultraviolet

              Manometric

              Amperometric

              Coulometric

              Chemical

           Procedures for  nitrate  determination are similarly
 numerous.  They encompass:

              Nitration

              Reduction

              Ultraviolet

              Electrochemical

              Manometric
                               -22-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 7B766 • TELEPHONE 51 2 - 454-4797
              Gravimetric

              Other methods

 Key examples are presented in Volume II, Table 3.1-12.

           The simultaneous determination of nitrate and nitrite
 proposed by Wetters and Uglum (WE-008) was found suitable for
 analysis of scrubber solutions.   The method is based on the
 fact that both nitrate and nitrite absorb in the ultraviolet.
 Nitrite has an absorbance maximum at 355 nm while nitrate shows
 a maximum at 302 nm.  Calcium, magnesium, sodium, potassium,
 carbonate, sulfite, sulfate and chloride showed negligible
 interference.  Samples with nitrite concentrations of 100-1000
 mg/£ can be determined with better than 3% accuracy using 1 cm
 cells.   Samples with nitrate concentrations of 20-500 mg/£ and
 nitrite concentrations of 20-100 mg/1 can be determined with
 better  than 570 accuracy using 10 cm cells.  Tables 5-3 and 5-4
 give supporting data.  The experimental results are described
 in more detail in Volume II, page 400.

           Nitrate concentrations at lower levels can be deter-
 mined using the procedure by West and Ramachandran (WE-012).
 This method is based on a reaction of nitrate with chromotropic
 acid.  The absorbance is measured at 410 nm.  Beer's Law is
 fulfilled for a nitrate concentration between 0 and 60 ppm.
 Experimental details of the procedure are presented in Volume
 II, page 416.


 5.2       Atomic Absorption Procedures

           The application of atomic absorption procedures for
 the analysis of scrubber liquors is discussed in detail in

                               -23-

-------
                                                   TABLE 5-3
N3
4>
 i
"355
.505
.505
.504
.202
.101
.052
"30 a
.763
.219
.225
.539
.614
.055
Simultaneous Determination of Nitrate and Nitrit.e
in pH 6 Simulated Filter Bottoms
Corrected A3oa
.561
.117
.024
.458
.574
.024
NO "-Added
Cmg/jO
1000
1000
1000
400
200
100
NO ~ Found
*mg/A)
1004
1004
1002
400
198
100
(1 cm cells)
% Error
-0.4
-0.4
-0.2
0.0
+1.0
0.0
NO ~ Added
(mg/j&)
5000
-1000
200
4000
5000
200
NO " Found
(~g/A)
4855
1000
200
3924
4965
200
7, Error
+2.9
0.0
0.0
+1.9
+0.7
0.0

-------
                                             TABLE 5-4





i
NJ
Ln
^3 38
.505
.501
.499
.201
.100
.048
.025
^303
.762
.317
.223
.536
.609
.079
.388
Simultaneous
Determination of Nitrate and Nitrite in
pH 11.3 Simulated Filter Bottoms
Corrected A303
.560
.117
.0234
.456
.569
.060
.288
N0a" Added
(mg/jl)
1000
1000
1000
400
200
100
500
NO "Found
1010
1002
998
404
204
100
500
(1 cm Cells)
7. Error
-1.0
-0.2
+0.2
-1.0
-2.0
0.0
0.0
NO 3~ Added
(mg/jl)
5000
1000
200
4000
5000
500
2500
N03~ Found
(mg/l)
4870
1000
199
3974
4950
508
2483
7o Error
+2.6
0.0
+0.5
+0.7
+1.0
-1.7
+0.7
*A11 runs were made  using delonized water as a reference and NaNOa and  NaN03  as  standards.

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.0.80X9948 • AUSTIN. TEXAS 78766 • TELEPHONE 51 2 • 454-4797
 Volume II,  Section 3.2.  This method was found suitable for
 the  analysis of calcium, magnesium, sodium, potassium, and
 trace elements.
 5.2.1      Determination of Ca,  Mg. K, and Na

           The  experimental studies to define optimum concentra-
 tion ranges  and  interferences encountered in lime/limestone
 based  wet  scrubber solutions are described in detail on pages
 181-233  of Volume II.   A Perkin-Elmer Model 403 Atomic Absorption
 instrument was used.
                                    o
           The  calcium line at 4227 A was measured.  Interferences,
mainly  from sulfate,  could be suppressed by using a 1% LaCl3,
57o  HC1  solution  in  the final dilution step.  The absorbance  is
a linear  function  of  calcium concentration in the range 0-7 mg/jl
Ca.   The  accuracy  in  synthetic scrubber solutions was found  to
be  about  2%.
                                              o
           Magnesium was measured at the 2852 A line.  Interfer-
ences could also be suppressed by using a 170 LaCl3 , 570 HC1
solution  in the  final solution step.   Absorbance is a linear
function  of concentration in the 0-0.7 mg Mg/£ range.  The error
in  analyzing wet scrubbing solutions  is approximately 270.
                                            o
           The  absorbance of sodium at 5890 A is linear up to
a sodium  concentration of about 1.5 mg/£.  Interferences are
suppressed  by  170 LaCl3 , 57o HC1, with the exception of potassium
interference.  The  standards should contain approximately the
same  amount of potassium than the test solution for accurate
determinations.  Accuracies using synthetic scrubber solutions
were  found  to  be approximately 370.
                              -26-

-------
                  SSOO SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
          Sulfate, calcium, magnesium, and sodium were  found
                                                      o
to interfere with the potassium determination at 7665 A.   The
addition of 170 LaCl3 , 5% HC1 suppresses  the  interference  partly.
The Mg, Ca, and Na content in the standard must be matched to
that in the sample.  The optimum potassium concentration  ranges
up to 6 mg/Ji.  Errors in analyzing artificial scrubber  solutions
using matched standards were approximately 4%.
5.2.2     Determination of Catalytically Effective  Trace  Elements

          Cobalt, copper, iron, manganese,  and  nickel  are reported
to catalyze sulfite oxidation by oxygen even  if  present  in the
ppb range.  These elements can be determined  by  atomic absorption
once they are extracted from the original phase.  The  elements
of interest are concentrated in the extraction  step.   In
addition, the instrument sensitivity  is increased two  to  five-
fold (PE-037) when an organic, instead of an  aqueous phase,  is
aspirated into the atomic absorption  spectrophotometer.

          The system dithizone, 8-quinolinol  and  acetyl  acetone
(chelating agent)-ethyl propionate  (solvent)  extracted Co, Cu,
Fe, and Ni but not Mn.  It was abandoned in favor of methyl
isobutyl ketone (solvent) and diethyldithiocarbamate (chelating
agent) (JO-012).

          Using this method cobalt, copper, iron, and  manganese
can be determined at approximately  the 10 ppb level to better
than ± 25%.  Nickel can only be determined  to 100 ppb  with this
accuracy due to flame emission effects at the low wavelength
at which nickel is measured.  Chapter 3.2.3 in  Volume  II  gives
more details.
                               -27-

-------
                  SBOO SHOAL CREEK BLVD .  P.O. BOX 9948 . AUSTIN. TEXAS 7B766 • TELEPHONE 512-454-4797
5.3       X-Ray Fluorescence

          The evaluation of the  literature on methods  to  deter-
mine total sulfur (in the form of sulfate) revealed  that  no
rapid field method existed giving accuracies of better  than  3%.
X-ray fluorescence spectrometry was  investigated  as  a  possible
means to determine sulfur rapidly and accurately.  No  literature
could be found describing the application of X-ray fluorescence
to the analysis of liquid samples of composition  comparable  to
samples taken from SOS removal processes.  In addition  to sulfur,
calcium, potassium, and chlorine may be determined rapidly using
X-ray fluorescence spectrometry.

          This section of the report presents a brief  discussion
of the physical basis for X-ray  fluorescence spectrometry and
a description of the types of instrumentation available.   This
section also contains the report of  the experimental measure-
ment of relevant matrix interference coefficients and  the mathe-
matical description of their application  to calibration and
measurement procedures.  Chapters 3.3 and 4.1 of  Volume II dis-
cusses the application of X-ray  fluorescence in more detail.
5.3.1     Physical Phenomena

          If a material  is bombarded with  X-rays  of  sufficiently
high energy (short wavelength),  the atoms  of  the  material  will
give off characteristic  X-rays.  The wavelength of  the  emitted
(fluorescent) X-rays will be characteristic of the  kinds  of
atoms (elements) in the  material,  and  the  intensity  of  a  given
wavelength will indicate the concentration of that  particular
element.  Then by measuring the  intensity  of  the  fluorescent
X-radiation at different wavelengths it  is possible  to  deter-
mine the elements present in a given material and their
                               -28-

-------
                  ssoo SHOAL CREEK BLVD. • P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512-451-4797
concentrations.  Absorption of  the fluorescent  radiation by
the sample or secondary fluorescent  effects  are potential sources
of error and are discussed later.
5.3.2     Description of Available Equipment

          In conventional X-ray fluorescence  instruments,  the
fluorescent radiation is dispersed so  that  the  angle  of
dispersion is dependent upon  the wavelength.  X-ray detectors
are used to determine the intensity  of the  radiation  at each
angle.  X-ray dispersion is usually  accomplished  by directing
the X-rays onto a crystal which acts in  the same  way  as a
diffraction grating.

          Energy dispersive X-ray fluorescence  utilizes a  semi-
conductor detector which responds linearly  to the energy of
the incoming X-ray and a multichannel  analyzer  which  sorts the
resulting voltage pulses according to  height  and  counts the
number of pulses in each energy band to  produce a complete
energy spectrum of the X-rays.  Such systems  do not have as
much resolution as wavelength dispersive instruments,  and  if
radioactive elements are used as sources of existing  radiation,
they are not as sensitive to  low concentrations of elements in
the sample.  This type of instrument v?as not  tested for Lhis
application.
5.3.3     Preliminary Tests

          Wavelength dispersive  instruments  are  available
which measure one element after  another  (sequential)  or which
measure several elements simultaneously  (multichannel).
                               -29-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN, TEXAS 78766 • TELEPHONE 512 - 454-4797
 Instruments of both types were tested in the manufacturers'
 laboratories using simulated scrubbing liquors.  The multi-
 channel instrument yielded more precise analyses for chlorine.
 The sequential instrument yielded more precise analyses for
 sulfur and calcium.  The precision of potassium analyses was
 comparable on the two instruments.  Sulfur and calcium were
 the elements for which the more accurate analyses were required;
 therefore, the sequential instrument was chosen as being more
 suitable for this particular application.
 5.3.4      Description of Sequential Instrumentation

           The spectrometer arrangement for the sequential X-ray
 spectrometer  chosen is shown in Figure 5-1.  The sample to be
 analyzed is  irradiated from below; the excited characteristic
 X-radiation  is collimated by a Soller slit and reflected at the
 analyzer crystal at various angles, depending on the wavelength
 (Bragg reflection condition).   The intensity of a spectral line
 is  measured  by the proportional counters and the associated
 electronics.   The wavelength to be counted is set for each ele-
 ment  in a given sample in sequence.  A given analyzing crystal
 can cover only a finite wavelength range so for large wavelength
 changes one  crystal must be replaced by another.  The angle
 setting mechanism is continuously variable and by careful choice
 of  crystals  a large- number of  elements may be determined.

           The system chosen was operated manually for deter-
 mination of matrix interference coefficients (described below).
 Before installation at Shawnee a minicomputer and data acquisition
 system were  interfaced with the X-ray fluorescence spectrometer
 system so that operation of the spectrometer system was controlled
 by  the computer and data reduction was automatic.
                               -30-

-------
 Primary X-ray Beam
 (Continuum -f Anode Characteristic)
SampFc—_r
         I ,
X-ray Tube
Wavelength Measured, X
Secondary Emission Lines
(Sample Characteristic)
                                                                    Analysing Crystal-
                                                                    Changer
                                                                    Flov/
                                                                    Proportional
                                                                    Counter
                                                                    -Scintillation
                                                                     Counter
           FIGURE 5-1   -  BEAM PATH IN  THE SEQUENTIAL X-RAY SPECTROMETER
                                           -31-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 7B766 « TELEPHONE 512 - 454-4797
 5.3.5     Matrix Interference Corrections

           Quantitative X-ray fluorescence analysis, as with
 most other analytical procedures, is subject to interferences.
 The intensity of the fluorescent radiation of element i can be
 reduced or increased by another element j present  in the  sample.
 Reduction of intensity is caused by absorption effects and
 increased intensity is observed if secondary excitation occurs.
 In the analysis of S03 scrubbing liquors, absorption effects
 are the dominant interfering phenomena and secondary excitation
 effects can be neglected if they are present at all.

           Figure 5-2 is a plot of counting rate versus concen-
 tration of chlorine in the presence of varying amounts of sulfur
 The error bars represent counting errors for individual measure-
 ments.  The equations for the lines were obtained  using the non-
 linear least squares program outlined later in this section.

           The general equation relating concentrations of inter-
 fering elements to the observed intensities may be written:
                             -Ik., c.
                              •i  LJ  J
           Ni = ai + biC± exp J                            (5-2)

 where the i subscripts refer to the element being  measured and
 the j subscripts to the interfering elements.  The definitions
                                              •
 of the terms in this equation are listed below.

           N. = the counting rate for fluorescent
                radiation f$r the element i as cal-
                culated from number of counts and
                the time required to obtain those
                counts (counts/sec).  The counting
                rate is the instrumental measure
                of the intensity of the fluorescent
                radiat ion.
                               -32-

-------
 I

OJ

OJ
            £000 -
            /e>oo -
             (600 -
          O
          O
          u
          CO
             MOO -
             IZOO -


                                                                   11  ~!"~T~"  . •  | •  •
                                                                  •:;• •  ; h ••  :  h;1; r;  I;
                                                                     .......... i    ......
                                                                                 I  '
                          .01      .02.      .03      .04.      .OS     .06      .07      .OS>



                       FIGURE 5-2   -   CONCENTRATION OF  CHLORINE (MOLES  PER LITER)



-------
Radian Corporation
8500 SHOAL CHEEK BLVD. •  P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 51 2 • 454.4797
           a.  = the intercept of the calibration
                line.   It is the counting rate at
                the setting for measuring element
                i when no element i is present in
                the sample (counts/sec).

           b.  = the slope of the calibration line
                for element i when no interfering
                elements are present
           c.  -  the  concentration of element i in
                the  sample measured (moles/liter).

          k. .  -  matrix interference coefficient for
                j  as the interfering element when
                element i is being measured (liters/
                mole).

           c.  :=  The  concentration of interfering
                element j (moles/liter).

           Reliable  literature values for the matrix interference
coefficients  (k..)  were not available.   Matrix interference
coefficients  of importance in analyzing lime/limestone scrubber
liquors  were  determined as part of the  in-house testing and
preparation of  the  sequential X-ray fluorescence system.

           Matrix  interference coefficients were determined using
solutions  containing  kno^n concentrations of interfering ele-
ments  as well as  known concentrations of the element being
measured.  In the simplest case, three  solutions were used which
had  the  following concentration specifications.   One solution
would  contain no  element i.  Measurement of the fluorescent
                               -34-

-------
                  S500 SHOAL CREEK BLVD. •  P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512.454-4797
intensity of this solution would determine a..  The other  two
solutions would contain equal amounts of element i but different
amounts of element j, the interfering element.  Practical  con-
siderations (prevention of the formation of a precipitate  or a
volatile species) dictated that in some cases more than one
interfering element had to be present.  In practice.there  were
never more than three interfering elements present.  We now  let
the subscript j stand for the element, the effect of which we
are trying to measure, and the subscripts i and p indicate oLher
interfering elements which may be in solution.  If appropriate
substitutions are made into Equation (5-2) , the resulting
equations solved simultaneously and rearranged we obtain:
                             - c
                               C
-k.   (c 0 - c i )
   IP \  p2    pi/
        ^N.9   a.)   "ije Vu£2   "£l/    ip  \"P2    pi
k.  - 	^ ~  L	      (5-3)
  J                      r-   — r
Equation (5-3) is the basic equation used  to calculate  all  the
values of k-. which were obtained by measurements  on  three
solutions.

          Matrix interference coefficients were determined  for
the four elements to be measured by X-ray  fluorescence,  sulfur,
chlorine, potassium, and calcium as element i.  Interfering
elements were taken to be nitrogen, sodium, magnesium,  sulfur,
chlorine, potassium, calcium.  The self-interference  of an
element (k..) was found to be zero in  the  concentration range
of interest.  Typically, three determinations  of each matrix
interference coefficient were made by  the  method outlined above

          In order to obtain more accurate data for certain key
interactions and for those systems which for practical  reasons
did not contain large excesses of the  interfering  element,
                               -35-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 - 454-4797
 several sets  of data were taken using 10-16 known solutions.
 The  data for  these solutions were correlated using a computer
 to perform a  non-linear least squares program which calculated
 the  coefficients a., b., and k..  (Equation 5-2) iteratively so
 as to minimize the error between observed counting rates and
 counting rates calculated using the coefficients and known con-
 centrations.   Mathematical details and computer printouts are
 given in Section 3.3.4.4, Volume II of this report.
 5.3.6     Summary of Selected Values for Matrix Interference
           Coefficients and Associated Uncertainties

           The selected values of k^. are listed in Table 5-5.
 In cases where least squares values are available, they are
 used.  In some cases the average of two least squares calculations
 was chosen.   If sets of least squares data were not available,
 then averages of the values obtained on sets of three solutions
 were selected.  These are the values which are used in the
 Shawnee data handling computer program to calculate matrix
 corrections.

           Table 5-5 lists the errors to be expected for a
 solution of  a composition which might reasonably be found at
 Shawnee.  The sixth column lists the errors if matrix interference
 corrections  are neglected at all, and the last column the un-
 certainties  of the results caused by errors of the k. . ' s .
 5.3.7     Calibration Procedures for X-Ray Fluorescence
           Spectrometry

           Referring back to Equation (5-2), we note that counting
 rates that are measured, N., are related not only to  the

                               -36-

-------
                                         TABLE  5-5
SUMMARY OF ERRORS TO BE EXPECTED IN CONCENTRATION MEASUREMENTS  IF MATRIX  INTERFERENCES ARE IGNORED
      AND OF UNCERTAINTIES IN CONCENTRATION MEASUREMENTS DUE  TO UNCERTAINTY IN VALUES OF It.
Concentration
of Measured
Element Interfering Element 1
Measured 1 Element i iir.oles/liter)
Limestone
S
S
s
s
s
s
Totals
Cl
Cl
Cl
Cl
Cl
Cl
Totals
K
K
K
K
K
K
Totals
Ca
Ca
Ca
Ca
Ca
Ca
Totals
Sodium
System
N
Na
Mg
Cl
K
Ca
for Sulfur
N
Na
MS
S
K
Ca
for Chlorine
N
Na
Mg
S
Cl
Ca
for Potassiun
N
Na
Mg
S
Cl
K
for Calcium
Carbonate Svstem

.025
.025
.025
.025
.025
.025

.025
.025
.025
.025
.025
.025

.001
.001
.001
.001
.001
.001

.020
.020
.020
.020
.020
.020


X Relative
Error In
Concentration
Concentration .. , Measurement
of Interfering * " If Matrix
Element J or Klj Correction
(moles/liter) Used is Ignored

.002
.001
.017
.025
.001
.020

.002
.001
.017
.025
.001
.020

.002
.001
.017
.025
.025
.020

.002
.001
.017
.025
.025
.001



.03*
.05
.10*
.009*
.01
.1**

.02
.08
.11
.26*
.03
.2*

.03
.08
.11
.28
.26
.08

.03*
.08
.09*
.3**
.29**
.41



.008
.005
.17
.025
.001
.2
.4
.004
.008
.2
.65
.003
.4
1.3
.006
.008
.2
.8
.65
, .18
1.8
.006
.008
.15
.9
.72
.04
1.8

Uncertainty
In Value
of ku

.01
.01
.02
.005
.01
.1

.02
.01
.04
.01
.01
.2

.01
.02
.04
.01
.01
.04

.01
.01
.03
.2
.03
.06


1 Relative
Uncertainty In
Concentration
Measurement of
Element i Due
to Uncertainty
in kj.

.003
.001
.04
.01
.001
.2
.3
.004
.001
.008
.02
.001
.4
.4
.002
.002
.11
.03
.025
.08
.3
.002
.001
.05
.6"
.05
.006
.7

                         .015
.035
                                                     .05
.18
.01
                                                                                            .03
*    Obtained by non-linear  least  squares calculation using at least 10 data points.(Tables 3.3-8 - 3.3-20)
**   Average of 2 values  obtained  from non-linear least squares calculations on  2  sets  of  data  points.
(Estimated  concentrations for limestone runs were obtained  from data  for Colbert Steam Plant Pilot
Unit -  October,  1971.  The concentrations listed  are  approximately %  the maximum concentration
found for a given element.  Thus, a dilution of approximately  1:1 la assumed  In line with the
standard procedure for obtaining type X samples.)
                                                 -37-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 • 454-4797
 concentration of the element for which the analysis is run, c-,
 the concentration of interfering species, c., and the matrix
 interference coefficient, k..,  but also to a. and b. which are
 respectively the intercept and  slope of the calibration line.
 These last two parameters are dependent not only on the settings
 and characteristics of the entire X-ray excitation and measuring
 system but also on the characteristics of the individual plastic
 membrane used as a window on the bottom of the sample cup.

           Measurements on sulfuric acid solutions using a set of
 nine hostaphane membranes revealed significant variation from
 one membrane to the next.  The  necessity of calibrating each
 plastic membrane was established.

           Calibration will also be required if any change is
 made in the settings of the X-ray excitation and measuring
 system.  In normal field operation this is likely to occur when
 the P-10 gas used in the flow proportional counter is changed
 or in the instances of instrumental failure and subsequent
 repairs.

           The following calibration procedure is recommended.
 For a given membrane all elements will be calibrated at once.
 Two to four calibrating solutions each containing different
 amounts of sulfur, chlorine, potassium, and calcium will be used
 The counts obtained in analyzing deionized water determine a..
 The minimum total number of calibration solutions is three so
 that the calibration line may be calculated by the least squares
 program outlined below.  Five solutions including deionized
 water are a practical number to establish the calibration
 constants.
                              -38-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  PO. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 • 454.4797
 5.3.8     Mathematical Background for Computer Calculation
           of Calibration Parameters for XRF System

           The computer program is designed to calculate the
 intercept and slope of the best straight line as fitted by a
 least squares calculation through three to five data points.
 It  is anticipated that each datum point will represent the
 counting rate for a different concentration of the element under
 consideration.

           The program optimizes the intercept and slope in an
 iterative manner by minimizing the difference between the
 measured counting rates and those calculated from trial values
 of  a. and b. and the known concentrations.  Matrix interference
 corrections are made for the interfering elements in the calibrat'
 ing solutions.

           The mathematical basis for the program is given in
 more detail in Section 3.3.5.1 of Volume II.  Details of the
 computer program are given in Volume III.

           Detailed instructions for the preparation of suitable
 calibrating solutions are given in Volume II, Section 3.6,
 "Selected Field Methods and Data Analysis System."
 5.3.9     Fluorescence Counting Rate Measurements

           Referring again to Equation (5-2), we have listed  the
 values of k..  (in Table 5-5) and have outlined the method for
 obtaining a.  and b. from the calibration procedure.  N. is
 obtained directly from the output of the X-ray measuring system,
 and  if we can obtain values of c.  we will be in a position to
 calculate c-.

                               -39-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 * AUSTIN. TEXAS 78766 • TELEPHONE 512 - 454-4797
           The concentrations of interfering elements  (the c.'s)
 come from two different sources.  The concentrations  of  nitro-
 gen, sodium, and magnesium are determined by methods  other  than
 X-ray fluorescence.  These concentrations, when available,  may
 be put into the computer and the corrections calculated  directly,
 The concentrations of sulfur, chlorine, potassium, and calcium
 will usually be determined by X-ray fluorescence and  their  con-
 centrations must be calculated simultaneously by an interative
 process.   For computational purposes the interfering  elements
 are divided into two groups.  The group j = 1 through J  are
 elements  determined by XRF.   The elements i = j+1 through L are
 determined externally.  The logarithm of Equation (5-2)  is  taken
 and the above definitions used.
 0n (—^~	j = ^ Ci - Z ^ij °j
                   ci
 The s subscripts for N, a, and b refer to a particular  sample
 membrane .

           The set of non-linear equations obtained when known
 values are substituted into Equation (5-4) is solved  iteratively
 by assuming that the corrections are small to obtain  initial
 values, then defining an error function and minimizing  it.
 Mathematical details are given in Section 3.3.5.2 of  Volume II.
                               -40-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 51 2 •45«-4797
 6.0        SOLID PHASE CHARACTERIZATION

           The characterization of solids comprises two steps,
 namely (1)  determination of the chemical composition and (2)
 crystalline phase identification.  The first task is achieved
 by chemical analysis  after dissolution of the solids, the second
 by use of X-ray diffraction.   The problem area is discussed in
 detail in Section 4.0 of Volume II on pages 491-547.
 6.1        Chemical Composition

           The  components of interest in the solids include
 calcium,  magnesium,  total sulfur,  carbon dioxide, and sulfur
 dioxide.   Total sulfate is calculated as the difference of
 total  sulfur and sulfite sulfur.

           Approximately 1 g of solid sample (dried at 75°C) is
 dissolved in a hydrogen peroxide,  hydrochloric acid solution.
 The  dissolution step is carried out in a stoppered flask to
 avoid  loss of  SOS.  Sulfate,  sulfite, and carbonate dissolve
 completely under these conditions.  Sulfite sulfur is oxidized
 to sulfate sulfur.  The solution is brought to volume.  Total
 sulfate  is determined by the  ion exchange alkalimetric procedure
 or by  X-ray fluorescence.  Calcium is determined by X-ray fluo-
 rescence  or atomic absorption.  Atomic absorption is the method
 of choice for  magnesium.

           Carbon dioxide in solids is determined by an evolution
 technique (see Sections 3.4.4 and  4.3.2 of Volume II).  C03 from
 solids is liberated  by acidifying the sample with sulfuric acid
 in a closed system,  which includes a carbon dioxide absorber,
 a  gas  scrubber, an expansion  baldder and a circulating pump.
                              -41-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9918 • A USTIN. TEXAS 78766 • TELEPHONE 51 2 - 454-4797
 The carbon dioxide combines with barium hydroxide  solution  of
 known normality content to form BaC03 precipitate.   The  excess
 hydroxide is titrated with standard hydrochloric acid  to the
 phenolphthalein end point.

           About 0.1 to 0.4 g of original sample is dissolved  in
 a  buffered iodine solution for sulfite determination.  Excess
 iodine is back titrated with arsenite solution using the dead
 stop technique for end point determination.  The procedure  is
 described in more detail in Sections 3.5.4 and 4.3.3 of  Volume
 II.
 6.2        Phase Identification

           The crystalline phases contained in the solids  are
 identified by X-ray diffraction.  The principle of X-ray  dif-
 fraction techniques are summarized in Section 4.1 of Volume II.

           A finely ground powder is irradiated with monochromatic
 X-radiation.   The lattice planes of the fine crystals diffract
 the  X-rays in a discrete direction if the Bragg equation  is
 fulfilled.

           nX  = 2di •  sin ei                                (6-1)

            X  = wavelength of the incident X-ray
                      o
                beam (A)

            n  = order  of the reflection

           d.  = distance between reflecting crystal
                planes
                              -42-

-------
Radian Corporation
                  8500 SHOAL CREEK BLVD.
                                P.O. BOX 9948 • AUSTIN. TEXAS 78766 •  TELEPHONE 512 • 454-4797
           0. = reflection angle from  crystal  planes
                of distance d.
 The intensity of the diffracted  radiation  is  a  function of the
 atomic structure of the crystalline  phase  which is reflected
 in the atom form and structure factors.  The  set of lattice
 spacings d. and the intensity of  the diffracted lines is unique
 for each distinct crystalline phase  and  is used for phase
 identification of unknown material.

           The instrumentation used  to record  the X-ray pattern
 is either a powder camera (Debye-Scherrer  method)  or a goinometer,

           Section 4.1.3 of Volume  II shows the  X-ray patterns
 of the following phases potentially  present in  solids from lime/
 limestone wet scrubbing processes.
1.  CaO                           9.
2.  Ca(OH)s                      10.
3.  CaCQs  (aragonite)            11.
4.  CaC03  (calcite)              12.
5.  CaS03-%HsO                   13.
6.  yCaS04  (soluble  anhydride)   14.
7.  pCaS04  (insoluble  anhydride)15.
8.  CaSO,-2HpO                   16.
                                       CaS04 -%HaO
                                       MgO
                                       Mg(OH)s
                                       MgC03-3HsO
                                       MgSOg -3^0
                                       MgS03 • 6H^ 0
                                       CaMg(C03)8  (dolomite)
                                       SiOa  (from fly ash)
                                -43-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512 • 454-4797
 7.0       FIELD STUDIES

           The sampling, sample handling, and analytical methods
 described in previous sections were developed and  tested by
 analyzing data from several pilot units.  Samples  were collected
 during

              OAF in-house studies (Volume II, page 569),

              pilot plant runs at the Tidd Plant in
              Brilliant, Ohio (Volume II, page 573),

              pilot plant runs at Key West (Volume  II,
              page 592),

              pilot plant studies at TVA's Colbert
              Steam Plant (Volume II, page 613),

              pilot plant studies at Shawnee (Volume II,
              page 635).

 Representative results will be presented here for  the pilot
 studies  at TVA's Colbert steam plant (see Volume II for the
 other  systems).   The system arrangement at Colbert is shown in
 Figure 7-1.   Samples were taken and analyzed at the scrubber
 effluent (Sample Point 2),  scrubber spray (Sample  Point 1),
 effluent hold tank F-12 overflow (Sample Point 3), and the
 process  liquor tank F-13 (Sample Point 4).  The results of the
 liquid and solid phase analyses are shown in Tables 7-1 through
 7-3  for  the  liquids and Tables 7-4 through 7-6 for the solids
 analyses.   The buildup of inerts is very small in this arrange-
 ment since most  of the fly  ash was removed by the raw water
 spray.   The  accuracy of the methods is reflected in the total
 ionic  imbalance.

                               -44-

-------
         Venturi Rod
           Scrubber
Raw Water Spra
Flue Gas	
From Boiler
                   Water, Flyash
                     to Sewer
                                           Flue Gas
                                           to Stack
                                  affluent Hold Tank  Process Liquor Tank
                                  (F 12,  500 gal)      (F  13,  500  gal)
                                                                                          Limestone Solids
Limestone Slurry
   Feed Tank
   (120 gal)
                                                                                         ©
                                                                        Clarifier  (3000  gal)

                        FIGURE 7-1 - TVA LIMESTONE WET  SCRUBBER PILOT UNIT

-------
               TABLE 7-1
Results of Liquid Sample
Colbert Plant. Run 1
Sample
Sample Point
Designation (See Fig. 2-1) K Na Ca
Scrubber Effluent #1
Scrubber Effluent #2
Scrubber Effluent #3
Scrubber Spray
F-12 Overflow #1
' F-12 Overflow #2
ON F-12 Overflow #3
F-13 Recycle #1
F-13 Recycle #2
F-13 Recycle #3
Clarifier Overflow #1
Clarifier Overflow #2
Clarifier Overflow #3
Lir.estone Feed
Raw Water
2
2
2
1
3
3
3
4
4
4
6
6
6
5
8
0.31 33.8
0.56 36.0
0.56 35.2
0.11 0.23 6.40
0.28 0.55 34.3
0.53 36.1
36.2
0.28 0.46 31.4
0.51 32.4
32.2
0.30 0.62 25.7
26.5
26.2
0.22 0.55 25.1
0.03 0.19 0.55
Mg
10.1
10.5
10.7
0.26
10.6
10.4
10.6
9.9
10.0
9.7
10.6
11.1
10.9
10.0
0.14
Sulfite
4-
Sulfate
47.7
48.8
49.4
9.25
47.0
46.9
48.3
43.2
44.2
44.0
32.4
33.9
33.4
31.1
0.0
Sul-
fite
15.4
15.9
16.2
7.0
13.3
13.3
12.7
14.0
14.1
14.0
3.5
3.3
3.3
1.2

Anal ysis
5-27-71
Sul-
fate
32.3
32.9
33.2
2.25
33.7
33.6
35.6
29.0
30.1
30.0
28.9
30.6
30.1
29.9
0.0
Imbalance
^ipos' Zi *
Total m^ Zi Temperature
C°* CL N N°a (^les) P" CF>
3.18
3.75 9.0 2.73 .614 -3.95 5.31 101
3.70
2.48 1.3 0.67 .036 +1.42 5.42 109
3.64
4.14 7.7 1.17 -2.72 5.55 101
4.86
3.75
4.80 7.7 1.43 .551 -2.66 5.57 98
4.25
3.61
5.95 8.5 1.90 .614 -2.12 6.13 89.5
4.55
2.75 7.3 2.10 .555 -2.89 6.94 95
0.66 0.20 0.30 .025
(Concentrations ara given In m moles per liter)

-------
                                                              TABLE 7-2
Results of Liquid
Sample Analysis
Colbert Plant . Run 2

Sample
Sample Point
Designation (See Fie. 2-
Scrubber Effluent #1
Scrubber Effluent 02
Scrubber Effluent 03
Scrubber Spray
F-12 Overflow 01
F-12 Overflow 02
F-12 Overflow #3
F-13 Recycle #1
F-13 Recycle 02
F-13 Recycle 03
2
2
2
1
3
3
3
4
4
4


., K Na Ca
0.29 0.48 35
35
0.48 36
0.16 0.27 7
0.28 0.47 31
0.48 32
32
0.30 0.48 30
30
30
.5
.3
.2
.10
.6
.9
.0
.3
.5
.2

Mg
12.3
12.3
12.5
0.34
11.8
12.1
11.9
11.9
11.9
11.9
Sulfite
Sulfate
47.4
48.0
48.5
8.92
42.3
43.6
43.2
40.0
39.6
39.7

Sul-
fite
12.2
12.2
11.8
6.2
10.1
10.1
10.1
6.70
6.40
6.50
5-28-7


Sul-
fate
35.
35.
36.
2.
32.
33.
33.
33.
33.
33.
2
8
7
72
2
5
1
3
2
2
1



C0a CL
3
3
3
1
3
5
4
3
3
4
.25
.98 9.6
.39
.42 2.1
.75
.25 9.2
.93
.41
.75 8.7
.02

Total
N

1.73

0.40

1.33


1.10

                                                                                                          NO,
                                                                                                         .690
                                                                                                         .044
                                                                                                         .678
                                                                                                         .614
                                                                                                                 Imbalance
                                                                                                                 (m moles)
                                                                                                                                  Tcirperature
-2.21     5.52       99
+2.36     3.38     104
                                                                                                                   -3.24     5.88     100
                                                                                                                   -0.66     5.68     100
Clarifler  Overflow
                                                                                                  0.95   .560
                                              (Concentrations  are given in m moles per  liter)

-------
                                                                      TABLE 7-3
00
 I
Sample
Designation
Scrubber Effluent #1
Scrubber Effluent #2
Scrubber Effluent #3
Scrubber Spray
F-12 Overflow #1
F-12 Overflow #2
F-12 Overflow #3
F-13 Recycle #1
F-13 Recycle #2
F-13 Recycle #3
Sample
Point
(See Fie. 2-1)
2
2
2
1
3
3
3
4
4
4
K
0.28 0
0
0
0.12 0
0.30 0

0
0.29 0
0
0
Na
.49
.51
.52
.25
.49

.50
.49
.48
.48
Ca
31.6
34.1
32.5
6.9
31.2
30.3
30.8
29.7
29.9
30.1
Results
Colbert
Mg
12.2
12.5
12.4
0.27
12.7
12.6
12.9
12.4
12.7
12.6
of Llauid
Plant
Sample
. Run 3
Sulfite
•f
Sulfate
42
43
44
9
41
41
41
39
39
39
.4
.8
.1
.23
.1
.4
.7
.2
.0
.0
Sul-
fite
10.3
10.5
9.1
6.5
8.5
9.0
8.7
5.3
5.6
5.4
Analysis
5-28-71
Sul-
fate
32.1
33.3
35.0
2.73
32.6
32.4
33-0
33.9
33.4
33.6
CO,
3.07
4.55
5.16
1.45
4.75
4.95
5.75
3.23
3.93
3.93
CL

9.3

2.0

9.8


9.2

Total


  N
                                                                                                          3.50   .662
                                                                                                          0.17
                                                                                                           1.27    .723
1.23   .649
              Imbalance

             ^W Zi "
             mi    , Zi


             (n moles)
                 -2.69
                 +1.78
                 -1.87
                                                                                                                           -1.12
                                                                                                                                      pH
                          5.67
                          3.19
                          5.77
                          5.96
                                                                                                                                          Temperature
                                   101
102
101
                                                                                                                                              101
         Clarifler Overflow
                                                                                                          0.94    .588
                                                       (Concentrations are  given  in m moles per liter)

-------
                                               TABLE 7-4



KelRht
Sanple of
Point Results of X-Ray Analyzed
Saffiple Designation 	 (see FlR. 2.1) 	 Analysis 	 Sample ( YR)

Scrubber Effluent 2 I. CaCO, (C.Iclte)
2. CaSO. -2H.O (Cypsun) 982
3. CnSO,-VI,0


Scrubber Spray 1 1. CnCO, (Calclte)
3. Ca(Al,SI,0.) -4H.O 1003
(poss Ible)
6. S10. (possible)

r-12 Overflow 3 1. CaCO, (Cslclte)
2. CaSO. -211,0 (Gypsum) 993
4. CnSO,-\,H.O
(A l.o unidentified
compound) major peak

F-13 RecycU 4 1. CnCO, (Calclte)
2. CaSO.-2H,0 (Gypsum) 990
4. c«so,-yi.o



Clarlfler Overflow 6 1. CaCO, (Cnlclte)
2. CaSO.-JH.O (Gypaun) 342
4. CaSO.-l,H.O



Clarlfler Bottoei 7 1. CaCO, (Cslclte)
2. CaSO.-2H,0 (Cypmn) 1011
4. CaSO, • VI ,0



Lloe>tona F«ed 5 1. CaCO. (Calclte)
2. CaSO.-2H,0 (Cypaun) 984
4. CaSO.-^H.O
(Unidentified ccoponmta
present)

Results of Solid Sample Analyala
Colbert Flant Hun 1 5-27-71

Slurry Sulflta Sulfate +
Undltaolved Concentration and Sulflte +
Sollda (OR) (R/i) C« MR Sulfate Sulfate Sulflte Carbonate Ca + MR Carbonate Solids Compofl 1 t Ion
CnSO.-ljH.O 36.61
CaSO.-2H,0 18.41
29 143 7.97 0.50 3.91 1.07 2.84 4.09 8.47 8.00 MRCO, 4.21
CaCO, 40.61
Insoluble 3.01
Total 102.81
701 12.: 2.79 0.23 .10 0.01 0.09 2.89 3.02 2.99 MpCOJ " K91
CaCO. 26.91
Insoluble 69.11
Total 99.31
CaSO,-VH.O 23.01
20 99.6 8.»3 0.53 2.33 0.57 1.78 6.24 8.98 8.59 CaSO. -2H.O 9.81
MgCO , t» . J*
CaCO, 61.01
Insoluble 2.07.
Total 100.31
CaSO,-\H.O 30.21
33 59.4 | 07 0 52 3 36 1.02 2.34 5.20 8.59 8.56 CnSO.-2H.O 17.51
MpCO. 4.41
CnCO, 47.11
Insoluble 3.31
Total 102.51
CaSO,-HH.O 46.51
4 1-5 j in 0 19 5 36 1.76 J.60 1.40 7.59 7.76 CaSO..2H.O 30 31
MgCO, 1.61
CaCO. 20.41
Inaoluble 1.27.
Total 100.01
CaSO,-\H.O 15.11
71 — 8.18 0.35 1.96 0.79 1.17 6.55 8.53 8.50 CaSO..2H.O 13.61
MnCO, 3.01
CaCO, 62.21
Insoluble 7.01
Total 100.91
CaSO,-VH.O 14 si
26 1*1 j.g4 c.63 1.67 0.52 1.15 7.73 9.47 9.40 CaSO.-2H.O 9.01
HgCO. 5.31
C.CO. 71.71
Iniolubl* 3.01
Total 103.61
vO
I

-------
                                                  TABLE 7-5



Saople
Point
Saaole Designation (see riR. 2.1)

Scrubber Effluent 2 1.
2.
4.


Scrubber Spray 1 1
J.
6.


f-12 Overflew 3 1
2.
4.



f-13 R.cycla 4 1.
2.
4.






Weight Weight
of of
Reaults of X-Ray Analyzed Undlssolved
	 Analrsla 	 Sample (maj 	 Solids 	

CaCO. (Calclte)
C«SO.. 211.0 991 66
CaSO.-^H.O


CaCO. (Calclte)
Ca(Al.S1.0.)-4H.O 993 779
(poiilblr)
S10, (poislble)


CaCO. (Calclte)
CaSO.-2H.O 1007 41
• jn.o


CaCO. (Calclte)
CaS04-2H.O 987 47
CaSO.-VH.O



Results of Solid Sample Analrnli
Colbert Plant Run 2 5-28-71
AnalTaes of Acid Soluble Material Concentration In m moles Per Cram of Solid



Slurry Sulflte lulf't. t
COTC"5"tl°° r. -. 5..?^t. Sulfate Sulflt. Carbonate Ca t «R Carbonate Sollda Cor:po.ltlon
C.SO.-SH.O
CaSO.-2H.O
88.0 8.17 0.49 3.0* 1.02 2.04 J.43 8.66 8.49 MjCO,
Inaofuble
Total
CaSO.-ljH.O
1J.7 1.46 0.10 0.11 0.03 .08 1.44 1.56 1.55 CnSO, -2H.O
CaCO!
Insoluble
Total
CaSO.'Vl.O
B6.0 g.j9 o.4» J.79 1.02 1.77 5.47 8.77 8.26 CoSO.-ZH.O
CaCO.
Insoluble
Total
CaSO.-VH.O
»-° 8.09 0.49 3.30 1.22 2.08 J.JO 8.J8 ».6C M!|S;'2Hl°
CaCO
Inaofubla
Total
26.51
17.51
4.11
51.11
6.71
105.91
1.01
o'.si
13.51

94.21
22.81
17.51
4.01
55.01
4.11
10J.41
26.91
21.01
4.11
l»l 91
4.81
105.71
o
I

-------
TABLE 7-6



Weight Weight
Sonple of o!
Seerele PtlU"«Uon lit* Tin. 2.1) Analysis Sample (olO Solids
Scrubber Effluent 2 1. CaCO, (Calclte)
2. C«SO -2H 0 (Cypeua) 100] 33
4. CaSO,-SH.O
Scrubber Spr«» 1 1. CaCO (Calclte)
i. C.(A!.SI 0.)-4«,0 9»6 784
(po.slble)
6. S10. (poaslble)
F-12 Overflow 3 1. C.CO, (Calclte)
2. CaSO. -2HO 1006 42
4. -CaSO.-SH.O
r-13 R««7CU » 1. C«CO. (Celclte)
2. C«SO. -2H.O 1014 44
4. CeSO.-^H.O
Unidentified Ccnponent
Preeent In Significant
Result! of Solid Sample Analj£eie_
(o^tert Pl>m Ryn ] ;-2g-?l
Analveei of A-rld Soluble Katertal Concentration In ra nolea Per Crftm of Solid
Slurry Sulflt* Sulfate »
CB/I) Ce rU Sulfate Sulfite Sulflte Carbonate Ca + Ki Carbonate Solid. Co
CaSO.-iH.
CaSO. -2H,
85. 2 7.98 0.53 3.11 1.16 1.95 5.41 8.51 8.52 HgCO,
C.CO,
I naolubla
Totel
CeSO, -iH.
17.2 1.51 O.m .076 .006 0.07 1.57 1.61 1.63 C.SO. -2H,
MgCO,
CaCO
Insoluble
Total
CaSO.-fcH.
CaSO. -2H.
100 >.10 0.59 2.78 1.05 1.73 5.38 8.69 8.16 rujCO,
CaCO,
Insoluble
Totel
C.SO -^H,
CeSO. -2H.
87.9 7.63 0.51 3.4* 1.40 2.04 5.00 8.36 8.44 MgCO.
C.CO.
Inioluble
Total




n?os It Ion
0 25. 21
0 20.01
1.51
48.71
101.91
0 1.01
0 0.11
o.ai
14.41
78.61
94^91
0 22.31
0 18.11
5.01
53.21
102 !BI
0 26.31
0 24.11
4.31
44.11
10) '.It

-------
Radian Corporation
                  8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 •  TELEPHONE 51 2 • 451-4797
    Imbalance = ^ K'^)     - J'
                        pos    L    i  i neg

          m. = molar concentration of charged species

          z. = charge number

          The pH measurements  and  analytical results shown in
Tables 7-1 through  7-3 were used  to calculate this imbalance.
The imbalance should ideally be  zero for zero errors in the
analytical determinations.  Another source of ionic imbalance
is the presence of  species for which no analysis was made.

          The results of  the solid phase analyses are presented
in Tables 7-4 through 7-6.  The  concentrations of the solid
species add up to nearly  10070  with exception of the solids of
the scrubber spray  which  contain most of the fly ash.  Compounds
leached from the fly ash  for which no analysis was made may be
responsible for the low values found.
                                -52-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -454-4797
 8.0        USE OF THE RAW DATA

           It was mentioned earlier that the results of the
 chemical analyses have no value per se.  They gain their value
 in the chemical engineering framework within which they are used.
 Dominant points of interest are:

              mass transfer characteristics in the
              scrubber

              solid-liquid mass transfer rates

              scaling potential

           Therefore, analytical results like those presented
 in the previous chapter must be processed further.  As an example,
 the  scaling tendency of the scrubber effluent of Run 3 given
 in Table 7-3 will be determined.  This task is solved by con-
 sidering the ionic equilibria in the aqueous phase.  The results
 of the chemical analysis listed in Table 7-3, the pH value, and
 the  temperature were used as inputs for computer calculations.
 This  equilibrium program distributes the eight key species into
 30 important complexes (see Table 8-1) and calculates the
 equilibrium partial pressures of sulfur dioxide and carbon
 dioxide.   Individual activity coefficients are calculated using
 an extended Davies equation.  The resulting activities of the
 individual ionic species for the scrubber effluent are listed
•in Table 8-1.
           The  activities of Ca  ,  S03,  and S04 are given as
 7.25  x  10"3,  1.22  x 10"4,  and 5.84 x 10"3, respectively.  The
 ratios  of  activity product to solubility product constant at
 38°C  for CaS03 -^HgO and CaS04-2HsO are  10.6 and 1.79, respectively
                               -53-

-------
                    BSOO SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 •  TELEPHONE 512.454.4797
                              TABLE  8-1
             DISTRIBUTION ACTIVITIES OF KEY SPECIES
10 June  1971    14:41:52.539
Temperature 38.340 Deg.  C
                             INPUT MOLES
S02 = 1.00-02
C0a = 3.07-03
S03 -•= 3.34-02
!a,0 = 3.85-04

Component
H,,0
H+
OH"
HS03"
S03'~
S04~~
HC03~
C03~~
N03"
HS04"
H^S03
HaC03
Ca"^
CaOH+


N305 = 1.
HCL = 9.
AQUEOUS
Molality
2.6-06
1.6-08
6.8-03
3. 5-04
1.9-02
7.0-04
4.3-08
3.4-03
2.3-06
1.1-06
2.2-03
2.0-02
3.0-09


75-03 CaO = 3.27-02
30-03 HS0 = 5.55+01
SOLUTION EQUILIBRIA
Activity
2.1-05
1.2-08
5.2-03
1.2-04
5.8-03
5.4-04
1.5-08
2.4-03
1.8-06
1.1-06
2.3-03
7.2-03
2.3-09


MgO = 1.24-02


Activity
Coefficient
1.0-00
8.3-01
7.7-01
7.7-01
3.4-01
3.0-01
7.7-01
3.4-01
7.0-01
7.7-01
1.0+00
1.0+00
3.6-01
7.7-01
                                  -54-

-------
                   S500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454.4797
                                                      Activity
Component
CaS03
CaC03
CaHCOg +
CaS04
CaN03 +
M ++
Mg
MgOH+
MgS03
MgHC03 +
MgS04
MgC03
Na+
NaOH
NaC03 "
NaHC03
NaS04 "
NaN03
CL"
Molality
2.6-03
2.0-07
1.0-04
10.0-03
7.0-05
8.0-03
2.1-08
3.3-04
1.9-05
4.1-03
1.2-07
7.4-04
1.9-12
2.3-10
1.7-07
2.5-05
5.5-07
9.3-03
Activity
2.6-03
2.0-07
7.8-05
1.0-02
5.4-05
2.8-03
1.6-08
3.3-04
1.5-05
4.1-03
1.2-07
5.8-04
1.9-12
1.8-10
1.8-07
1.9-05
5.6-07
7.1-03
Coefficient
1.0+00
1.0+00
7.7-01
1.0+00
7.7-01
3.5-01
7.7-01
1.0+00
7.7-01
1.0+00
1.0+00
7.8-01
1.0+00
7.7-01
1 . 0+00
7.7-01
1.0+00
7.6-01
PS03 =  1.46-06 ATM
PC03 =  9.37-02 ATM
           Molecular Water  = 9.99-01  KGS
Specified pH ••=  5.670
Ionic  Strength  =  1.08-01
Res.  E.  N. =  -2.692-03
                                 -55-

-------
                  asoo SHOAL CREEK BLVD. •  P.O. BOX 99^8 • AUSTIN. TEXAS 78766 • TELEPHONE 512 .asa.4797
This shows that  the  solution is highly supersaturated  with
respect to CaS03-%HaO and moderately supersaturated  with
respect to CaS04 -2H20.   These numbers will be  of  value in con-
junction with  scaling studies to define scaling  tendency.

          Another  number of importance for engineering cal-
culations is the partial pressure of S0a .  From  Table  8-1, it
is seen that Pcri  in the scrubber effluent is  1.46  x 10   atm
               O Ug
or about 1.5 ppm.   This is a necessary input  for  SQS  vapor-
liquid mass transfer calculations.  In similar fashion other
activities can be  calculated if required for  the  description
of solid-liquid  mass transfer rates.
                               -56-

-------
                  asoo SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 .154-4797
9.0       DATA HANDLING SYSTEM

          The Shawnee laboratory data analysis  system  is  designed
to perform data storage, laboratory computations,  and  report
generation tasks associated with the laboratory  operations.   The
system is basically a card oriented system using marked-sense
card input to ease the problem of converting data  to a machine
readable format.  In addition, the system is designed  to  provide
automatic operation of an X-ray fluorescence spectrometer with
automatic calibration and matrix corrections performed upon
the results.  The X-ray analysis results are entered automatically
without operator intervention.  The system is  represented diagram-
matically in Figure 9-1.

          For each set of analyses defined by  a  time,  sampling
point, sample type (i.e., line out, steady-state,  or exception),
and run number, a data packet  is created on disk to  store all
raw data and computed results  associated with  that set of
analyses.  After all data for  that particular  set  of analyses
has been entered into the data processing system and all  cal-
culations performed, the completed data packet  is  transferred by
the operator from the disk to  magnetic tape and  by means  of  the
line printer a hard copy is prepared.  The data  analysis  system
may be commanded to prepare sample taking schedules  and sample
analysis schedules.

          Volume III of this report contains a detailed
description of the system, including complete  operating instruc-
tions.  The paragraphs below give a very brief description of
the hardware and software components of  the system.
                              -57-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
 FIGURE  9-1 -  LABORATORY  DATA ANALYSIS  SYSTEM USED  AT SHAWNEE
                                   -58-

-------
                  8500 SHOAL CREEK BLVD. • P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 512 .454.4797
9.1       Laboratory Data Analysis Hardware

          In addition to the X-ray fluorescence  spectrometer,
the hardware system consists of the following  major  components:

          Nova 1200 (20 K)/Jumbo Chassis
          Fixed Head Disk (256 K)
          Magnetic Tape Transport (9  track/800
            BPI/10V reels/24 IPS)
          Automata Marked Sense, Punched  Card
            Reader
          ASR 33 Teletype
          KSR 35 Teletype
          Beehive CRT Terminal
          Data Products Line Printer  (80  column,
            1100 LPM)

          The fixed disk is used for  storage of  all  data in
data handling programs.  The Beehive  CRT  terminal is used as the
basic input/output device for the user  of the  data analysis sys-
tem.  The ASR 33 Teletype is used as  a  backup  to  the CRT and to
supply the capability of the paper tape reader/punch.   The KSR
35 (without paper tape feature) is included for  the  basic control
of the X-ray equipment.  As mentioned previously, the  system
input is basically card oriented using  marked  sense  cards.  The
magnetic tape transport provides the  capability  of storing data
on magnetic tapes for transfer to other sites  as  well  as storage
of the basic data handling system, the  disk operating  system
with FORTRAN compiler, and diagnostic routines for the major
hardware components.  The line printer  provides  a permanent hard
copy of all data.
                               -59-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUST IN. TEXAS 78766 • TELEPHONE 512 - 454-4797
 9.2       Laboratory Data Analysis Software

           The software system can be divided logically  into
 three parts:  (1) executive system, (2) application routines,
 and (3) diagnostic routines.
 9.2.1     Executive System

           The executive system for the laboratory data handling
 system is an extension of the vendor supplied disk operating
 system.  The system provides comprehensive file handling capabil-
 ities and protection.  The executive system allows the execution
 of the laboratory data system as well as program generation and
 development software including a FORTRAN compiler, editor, de-
 bugger, etc.  The executive system retrieves appropriate files
 from disk storage as commanded from the user operating input
 device.

           The executive system provides the user a variety of
 commands to perform the laboratory data handling.  These include
 data input, report generation, generation of magnetic tape files,
 and automatic operation of the X-ray fluorescence spectrometer.
 Upon receiving a data input command, the system reads the data,
 retrieves the appropriate application programs, performs neces-
 sary calibrations and computations, and stores the raw data as
 well as the resultant computed values.  The executive system
 also performs bookkeeping functions such as scheduling samples
 to be  taken and scheduling the analysis of these samples.
                               -60-

-------
                  BGOO SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454.4797
9.2.2     Application Routines

          The laboratory application routines  actually  perform
the function as instructed by the executive  system.   The  ap-
plication routines are initiated under  user  control;  however,
the user is not required to insure  that  appropriate  disk  files
are input into the computer memory.  Most  application routines
are written in FORTRAN which allows easy modification if  lab-
oratory computations are changed or if  new procedures are
implemented in the laboratory.
9.2.3     Diagnostic Routines

          This software package operates  independently  of the
previously mentioned packages.  The primary  function  is for
trouble-shooting hardware failures to  localize  the  equipment
that is malfunctioning, and for performing preventative main-
tenance.  These routines guide the user  through series  of tests
on each peripheral device to check all phases of operation.
The test routines may be input from any  of three input  devices:
(1) the teletype paper tape reader, (2)  the  card reader,  and
(3) the magnetic tape unit.  This allows  for diagnostic testing
even if one of the input units becomes inoperative.   The print-
out for a typical set of analyses is given in Figure  9-2.
                               -61-

-------
                                      RESULTS  OF  SAMPLE  ANALYSES

I
SAMPLE It)
SAMPLE POINT
TEMPERATURE(C)
CONDUCTIVITY
PH
3616
FIELD
0,0
,0B00E 0
.000I3E 0
J RUN NUMBER 
-------
K3QI3D \sOrpOr3tlOn  asoo SHOAL CREEK BLVD. •  P.O. BOX sate • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
 10.0      SUMMARY

           The chemical analysis of the  key  species  in  lime/
 limestone based sulfur dioxide removal  processes  is  the  basis
 for  the engineering evaluation of the system  performance.   The
 problem area encompasses five individual  steps.

           1)  Sampling

           2)  Sample Handling

           3)  Sample Analysis

           4)  Data Processing

           5)  Data Evaluation

 Figure 10-1 shows the methods selected  to accomplish the
 individual tasks.

           The pH and the temperature of the slurry  are measured
 in situ.   A positive pressure filtration  system separates  the
 liquid from the solids.  The solids content of the  slurry  is
 calculated from the solid to liquid ratio.

           The liquid is thermodynamically unstable  and must be
 quenched  by appropriate techniques.  An aliquot is  diluted in
 an aqueous hydrogen peroxide solution.  Sulfite is  oxidized
 and  sulfate precipitation is avoided by the dilution involved.
 Total sulfur, calcium, chloride, potassium, sodium,  magnesium,
 and  catalytically active trace constituents such  as  manganese,
 cobalt, copper, iron, and nickel are determined in  this  sample
 using the procedures as indicated by Figure 10-1.
                               -63-

-------
SAMPLING
                        QUENCHING
                        In Aqueous
                        Hydrogen Peroxide
                         X-RAY DIFFRACTION
                         DISSOLUTION
                            tn  HCL,

                         H.,00 - Solution
                         DISSOLUTION

                         In Buffered Iodine
                         CO,   -  Evolution
                         in  Closed  System
                          SAMPLE HANDLING
TOTAL SULFUR h* '"" Exchnnr.e Titrlmctrlc
Proc. or X-Ray
CALCIUM
CHLORIDE
by Atomic Absorption or
X-Ray Fluorescence
by Polentlonctric Iteration or


POTASSIUM
by Atonic Absorption or
y
                                                  SODIUM by Atomic Absorption
                                                  MAGNESIUM by  Atomic  Absorption

                                                  TRACE-ELEMENTS  by Atomic  Absorption
                                                   SULFITE  by  BacktItratlon  of  Excess
                                                               Iodine
                                                   CARBONATE  by  CO,   - Evolution and
                                                                      Detection by NDIR
                                                  TOTAL NITROGEN by modified Kjeldahl
                                                                             Procedure
                                                  NITRATE  by Colorlmetry or UV
                                                              Spectrophotometry
                                                  NITRITE  by UV Spcctrophotomccry
                                                 |  SOLIDS  CONTENT  of  Slurry
                                                 J  pH an:
                                                        id TEMPERATURE
                                                  CRYSTALLINE PHASES of Major Consclcutcnts
                                                  TOTAL SULFUR by Ion Exchange Titrlir.etric
                                                                  Procedure or X-Ray
                                                                  Fluorescence
CALCIUM  by  Atonic Absorption  or  X-Ray
             Fluoresccnce
                                                   MAGNESIUM by Atomic Absorption
                                                   INERTS In weight 7.
SULFITE sulfur by Backtlcracion of excess
                  Iodine
CARBONATE by CO, absorption  In  Ba(OH).
                                                                 SAM-'LE ANALYSIS                     DATA PROCESSING


                                                FIGURE 10-1:     SCRUBBER SLURRY ANALYSIS SCHJEMZ
                                                                                                                                       DATA EVALUATION

-------
                  8500 SHOAL CREEK BLVD. •  P.O BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512-454-4797
          A second aliquot is quenched directly  in a buffered
iodine solution thus avoiding any sulfite losses.  Excess  iodine
is back-titrated using standard arsenite and a dead stop pro-
cedure for end point detection.

          A sample to determine carbonate is quenched  in an
alkaline EDTA solution.  The EDTA prevents calcium carbonate
precipitation.  The alkaline environment lowers  the COS partial
pressure of the sample.  The C0a determination involves evolution
from an acidified sample and determination by a  nondispersive
infrared analyzer.

          Total nitrogen, nitrate, and nitrite are determined
by a modified Kjeldahl procedure and by ultraviolet spectroscopy.
A colorimetric procedure based on the  interaction of nitrate
with chromotropic acid is specific for nitrate.

          The X-ray fluorescence unit was interfaced with  a
minicomputer for rapid field measurements and data reduction.
Data not measured by X-ray fluorescence are  entered into the
system through a card reader, a CRT or a teletype.  The system
stores all the data on a magnetic tape.  In  this fashion they
can be read directly in a larger computer for chemical equilibrium
calculation.  A hard copy of the data  is provided by the printer.

          The solid sample obtained in the filtration  step is
processed in a similar fashion.  Part  of the crystals  are  finely
powdered.  Subsequent X-ray diffraction determines  the crystalline
phases of the major constituents.

          The solid sample is dissolved prior to chemical  analysis
Dilute hydrochloric acid containing hydrogen peroxide  is used  to
dissolve approximately 1 g solids.  Total sulfur, calcium,
                               -65-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9918 • AUSTIN. TEXAS 78766 • TELEPHONE 5! 2-454-4797
 magnesium and weight percent inerts are  determined  in this
 sample.  The analytical methods are the  same  as  those chosen for
 liquid sample analysis.

           Sulfite is determined on a  solid  sample dissolved in
 a buffered iodine solution.  Excess iodine  is back-titrated
 with arsenite solution and a dead stop technique for end point
 detection.

           The carbonate analysis  is performed by dissolving a
 solid sample in sulfuric acid  in  a closed system.  The evolved
 COS is absorbed in barium hydroxide solution.  Excess barium
 hydroxide is back-titrated to  the phenolphthalein end point.
 The data processing in the field  is done by the  minicomputer
 and the peripheral devices.

           The analysis scheme  was checked and developed by
 sampling and analysis of several  pilot plants.   They included
 the GAP in-house test facility and pilot plants  at  the Tidd
 Plant, Key West, Colbert Steam Plant, and at  Shawnee.
                                -66-

-------
                  8500 SHOAL CREEK BLVD. •  P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 -154.4797
11.0      BIBLIOGRAPHY
AM-002    American Society for Testing  and Materials,  1969
            Book of. ASTM Standards,  Part  23 :   Water,  Atmospheric
            Analysis, Philadelphia,  1969.

BE-024    Bertolacini, R. J. and J.  E.  Barney,  Anal.  Chem. 29.,
            281-83 (1957).

DE-029    DeBerry, David W., "Procedure for  the Determination
            of Total S03 in Aqueous  Solutions," Radian Technical
            Note 200-004-04, Radian  Corporation,  Austin,  Texas,
            May, 1970.

DO-006    Dollman, G. W. , Env. Sci.  Tech. 2.,  1027-29  (1968).

EN-020    Environmental Protection Agency, Water  Quality  Office,
            Analytical Quality Control  Lab.,  Methods  for  Chemical
            Analysis of Water and Wastes, Washington,  D.C., GPO,
            1971.

FE-004    Federal Water Pollution Control Administration,
            FWPCA Methods for Chemical  Analysis of Water  and
            Wastes , U. S. Department of Interior, FWPCA,
            Division of Water Quality Res.,  Anal. Qua 1. Contr.
            Laboratory, Cincinnati,  Ohio, November,  1969.

FI-019    Fisher Scientific Instruments Div.,  Bull.  No. 68-2A,
            Cat. No. 9-319-100, Pittsburgh,  Pennsylvania.

FR-003    Fritz, J. S., S. S. Yamamura, Anal.  Chem.  27,  1461-
            64, 1955.
                               -67-

-------
Radian Corporation
8500 SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 51 2 • 454-4797
 FR-009    Fritz, J. S., M. Q. Freeland, Anal.  Chem.  26.,  1593-5,
             (1954).

 JO-012    Joyner, et al., Env. Sci. and Tech.  I,  417  (1967).

 KO-015    Kolthoff, I. M. and P. J. Elving, Treatise  on  Analytical
             Chemistry, Part II, Vol. 7, Interscience  Publishers,
             N.Y., 1961.

 KO-050    Kolthoff, I. M. and E. B. Sandell, Textbook of
             Quantitative Inorganic Analysis, 3rd  Ed.  MacMillan,
             N.Y., 1952.

 MA-039    Macchi, G., B. Cescon and D.'Mameli-D1Errico,  Archo
             Oceanogr. Llmnol. 16, 163-71  (1969).

 MU-020    Muller, Gerhard-Otfried, Praktikum der  quantitativen
             chemischen Analyse. Leipzig,  S. Hirzel  Verlag,  1957.

 PE-037    Anal. Method for A. A. Spec., March  1971  supp.

 PR-007    Prochazkova, L., Zeitschr. Anal. Chem.  182,  103-7
             (1961).

 SE-015    Seel, F. and E. Degener, Z.  fur Anorg.  und  Allgem.
             Chemie 284, 101-17 (1956).

 SH-006    Shaw, W. M. , Anal. Chem. 3_0, 1682-9  (1958).

 SH-014    Shiner, V. J. and Morris L.  Smith, Anal.  Chem.  28.,  1043-
             45 (1956).
                               -68-

-------
                   SSOO SHOAL CREEK BLVD. • P.O. BOX 9948 • AUSTIN. TEXAS 78766 • TELEPHONE 512 - 451-4797
SI-005     Sijderius, R., Anal.  Chim. Acta 11, 28  (1954).

TH-007     Thompson, C. M., Radian Technical Note  200-004-13,
             Sept. 4, 1970.

WE-008     Wetters, J. H. and  K.  L. Uglum, Anal. Chem.  42. (3),
             335-40 (1970).

WE-012     West, Philip W. and T. P. Ramachandran,  Anal. Chim.
             Acta 35., 317-24  (1966).
                                -69-

-------
TECHNICAL REPORT DATA
(Please read Inunicliuns on llie reverse before completing)
1 REPORT NO.
EPA-650/2-74-024
2.
4. TITLE AND SUBTITLE
Development of Sampling and Analytical Methods
of Lime/Limestone Wet Scrubbing Tests
7.AUTHORis)K . schwitzgebel , F. B. Meserole, C. M.
Thompson, J. L.Skloss , and M. A. Me Anally
9. PERFORMING ORGANIZATION NAME AC
Radian Corporation
8500 Shoal Creek Blvd.
Austin, Texas 78766
JD ADDRESS
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION-NO.
Mir criT Ml
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
RAD-073-013
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ACY-25
11. CONTRACT/GRANT NO.
CPA 70-143
13. TYPE OF REPORT AND PERIOD COVERED
Final
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
i6.ABSTRACTThe report' gives results of B. study to develop appropriate sampling and
analytical methods to be used at EPA's test facility at Shawnee. Three problem
areas developed in analyzing the thermodynamically unstable slurry streams
encountered in lime/limestone-based SO2 wet scrubbing processes: sampling,
sample handling, and chemical analysis. Positive-pressure filtration was found to
lower the mass transfer phenomena during the filtration step to an acceptable level.
Quenching of the filtered liquid was chosen to avoid changing sample composition.
Two sets of analytical methods were selected for application at Shawnee: the
back-up methods are based on atomic absorption and wet chemical procedures; and
the rapid field methods are based on X-ray fluorescence, atomic absorption, and
wet chemical analysis. The X-ray fluorescence spectrometer was automated by
interfacing it with a NOVA 1200 minicomputer. Additional peripheral devices have the
function of processing all raw data. The raw data are input to the system with a card
reader, a teletype, or a CRT. The final results are stored on a magnetic tape. A
hard copy is provided by a printer.
17.
a. DESCRIPTORS
KEY WORDS AND DOCUMENT ANALYSIS
b. IDENTIFIERS/OPEN ENDED TERMS
Air Pollution Sulfur Oxides Air Pollution Control
Sampling Desulfurization Stationary Sources
Analyzing Scrubbers Atomic Absorption
Slurries X-Ray Fluorescence
Calcium Oxides Automation
Limestone Spectrometers
13. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
20. SECURITY CLASS (This page)
Unclassified

c. COSATI Field/Group
13B
14B
7B
20F
7A
8G
21. NO. OF PAGES
22. PRICE
EPA Form 2220-1 (9-73)

-------