'A-650/2-74-077

ptember  1974
                                          ^

Environmental  Protection Technology Series

M I M M M M Ml

                     I
                     55

                     V
                                                             ul
                                                             a
           • •_•_•_• • • •_•_• _•.*-•-

-------
                                  EPA-650/2-74-077
LIME/LIMESTONE  SCRUBBING
A PILOT  DUSTRAXTOR-KEY  WEST
                     by

              J. M. Craig, Ph.D.

      B. A. Bell, T. A. LiPuma, and J. K. Allison

            Engineering Science, Inc.
              7903 Westpark Drive
             McLean, Virginia  22101
             Contract No. CPA 70-61
              ROAP No. 21AQO-001
           Program Element No. 1AB013
        EPA Project Officer: Norman Kaplan

            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

                       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.
                                  ii

-------
                           TABLE OF CONTENTS

                                                                Page


Disclaimer Notice                                                 ii

List of Figures                                                   vi

List of Tables                                                  viii

Acknowledgements                                                 xii



INTRODUCTION                                                       1



PILOT PLANT EQUIPMENT                                              8

     Pilot Plant                                                   8

     Particulate Removal System                                    8

     Dustraxtor Absorber System                                   11


     Flue Gas Exhaust System                                      17

     Pilot Plant Schematic                                        17

     Reactant Handling and Waste Disposal Systems                 17

     Chemical Analysis Laboratory                                 21



TEST INSTRUMENTATION AND PROCEDURES                               22

     S02 Monitoring                                               22

     NO  Sampling and Analysis                                    23

     Particulate Sampling                                         24


     Slurry Sampling and Analysis                                 24

     Other Measurements                                           25

     Generalized Test Procedure                                   26
                                              i
     Test Procedure for Extended Operation                        28

     Test Procedure for Additive Operation                        28
                                   ili

-------
                      TABLE OF CONTENTS (CONTINUED)




                                                                Page




SUMMARY OF RESULTS                                                30



     System Operating Conditions and Characteristics              30



     S02 Absorption: Key West Program                             49



     SO. Absorption: Paducah Program                              59



     Particulate Collection in the Scrubber System                62



     NO  Absorption                                               62



     Sulfur Balance                                               62





DATA ANALYSIS                                                     68



     Introduction and Summary                                     68



     Key West Tests                                               70



     Paducah Test Program                                         79



     Liquid/Gas Ratio                                             81





CONCLUSIONS                                                       85



     Pilot Plant Equipment                                        85



     Sampling and Analysis                                        85



     SO- Absorption Efficiency                                    86



     NO  and Particulate Removal                                  87
       A


     Scale Formation                                              87



     Full Scale System                                            87
                                 iv

-------
                  TABLE OF CONTENTS (CONTINUED)



                                                          Page

APPENDIX A:  DETAILS OF ANALYSIS TECHNIQUES                 89

      SO, Apparatus and Procedure                           89

      Dust Loading Apparatus and Procedure                  91

      Nitrogen Oxide Apparatus and Procedure                91

      Calcium - EDTA Titrlmetric Method                     96

      Magnesium - EDTA Titrlmetric Method                   97

      Chloride - Argentometric Method                      100

      Nitrite                                              101

      Nitrate - Brucine Method                             103

      Sulfite                                              106

      Sulfate - Gravimetric Method                         108

APPENDIX B:  ANNUBAR FLOW ELEMENT, CALCULATION             111
             EQUATIONS, AND ENGINEERING SPECIFICATIONS

APPENDIX C:  SUMMARY OF SCRUBBER OPERATING CONDITIONS      116

APPENDIX D:  SUMMARY OF TEST DATA                          131

APPENDIX E:  UNITS OF MEASURE                              144

-------
         LIST OF FIGURES
Figure No.
2-1
2-2
2-3
2-4
2-5
2-6
2-7
4-1
4-2
4-3
4-4
4-5
4-6
4-7
4-8
4-9
Title
Mobile Pilot Plant Equipment Arrangement
Pilot Plant Dry Dust Collector Outside Detail
Typical Cross Section of Dust rax tor Scrubber
Pilot Plant Scrubber (Dustraxtor) Details
Dustraxtor Outside Return Apparatus and
90° V-Notch Weir
No. 10 Clarage Blower Induced Draft Fan Curve
Schematic Flow Diagram of Flue Gas and Slurry
Solutions with Locations of Sampling Points
Salt Water Entrained in the 12" Tube at
Selected Pressure Drops and Gas Flow Rates
River Water Entrained in the 8" Tube at
Selected Pressure Drops and Gas Flow Rates
Liquid/Gas Ratios for 8" & 12" Dia. Dustraxtor
Tubes at Selected Pressure Drops and Gas
Flow Rates
Variation of Inlet SO. Concentration, Key
West - 1971
Variation of Inlet SO- Concentration,
Paducah - 1971
Absorption Efficiency as a Function of Gas
Flow and Pressure Drop: Salt Water
Absorption Efficiency as a Function of Feed
Flow Rate and Pressure Drop: Salt Water
Summary of Coral Salt Water Experimental
Design Points
Summary of Fredonia Valley Limestone Salt
Page
9
10
12
14
16
18
20
31
32
34
42
43
50
52
53
55
Water Experimental Design Points
                   vi

-------
                     LIST OF FIGURES (CONTINUED)

Figure No.                    Title                             Page

  4-10        SO. Absorption Efficiency as a Function of          57
                Gas Flow Rate and Liquid/Gas Ratio

  4-11        Summary of Secondary Reactants Salt Water           58
                Experimental Design Points

  4-12        Summary of Fredonia Valley Limestone River          60
                Experimental Design Points

  5-1         Predicted Response vs. Actual Response for          76
                Independent Data from the Multiple Linear
                Regression Formula

  5-2         Summary of Significant Interactions in Coral        77
                Experiments

  5-3         Absorption Efficiency as a Function of              83
                Liquid/Gas Ratio

  A-l         S02 Sampling Apparatus                              90

  A-2         Particulate Sampling Train                          92
                                 vii

-------
                         LIST OF TABLES

Table                         Title                              Page

 1-1         Pilot Plant Test Series                                5

 2-1         Dustraxtor Mixer Specifications                       15

 2-2         Tank Mixer Specifications                             19

 2-3         Slurry Pump Specifications                            19

 4-1         Summary of Salt Water Tests Chemical Analyses          36

 4-2         Summary of Coral Marl Chemical Analyses                36

 4-3         Summary of Fredonia Limestone Chemical Analyses        37

 4-4         Summary of Hydrated Lime Chemical Analyses             37

 4-5         Summary of Dolomite Chemical Analyses                  38

 4-6         Summary of Precipitated Calcium Carbonate -            38
               Chemical Analyses

 4-7         Summary of Shawnee No.  9 Fredonia Limestone            39
               Chemical Analyses

 4-8         Summary of Shawnee No.  9 Aragonite Chemical            39
               Analyses

 4-9         Summary of Shawnee No.  9 Saltwater/Fredonia            40
               Limestone Chemical Analyses

 4-10        Summary of Shawnee No.  10 Aragonite Injection          40
               Chemical Analyses

 4-11        Summary of Shawnee No.  10 Fredonia Limestone           41
               Injection Chemical Analyses

 4-12        Summary of Inlet Flue Gas Conditions                  44
               Key West Unit No.  3

 4-13        Summary of Inlet Flue Gas Conditions                  44
               Shawnee Units No.  9 and 10

 4-14        X-Ray Diffraction Analyses of Scale Formation          45
               on the Tube

 4-15        Degree of Deposition at Various System Locations       46
               Key West Unit No.  3 and Unit No.9
                                  viii

-------
                     LIST OF TABLES (CONTINUED)


Table                            Title                           Page

 4-16        Degree of Deposition at Various System Locations      48
               TVA Sh«vn«« Unit No.10

 4-17        Settling Rate of Limestone Slurry Effluent            48

 4-18        SO. Removal Efficiency for Secondary Reactant Tests   56
               Key West No.3

 4-19        Pilot Plant Particulate Removal Efficiency            63
               Paducah Unit No. 9

 4-20        Average of NO  Measurements                           63
               Key West Unft No. 3

 4-21        Test Results of NO  During the Key West Test          64
               Series          x

 4-22        Average NO  Measurements                              65
               Paducah ¥est Series

 4-23        Test Results of NO  During the Paducah Test           65
               Series          *

 4-24        Sulfur Balances for Selected  Tests at Key West        66
               and Paducah

 5-1         Summary of Factorial Experiments                      68

 5-2         Key West Test Program Experimental Design             69

 5-3         Coral Reactant	Analysis of  the  Variance             72
               Key West Test Program

 5-4         FVL Reactant	Analysis of the Variance               73
               Key West Test Program

 5-5         Secondary Reactant Test Program at Key West           78
               Test Site

 5-6         Secondary Reactants	Analysis of the Variance        78
               Key West Test Program

 5-7         Paducah Test Program Experimental Design              80

 5-8         FVL Reactant	Analysis of the Variance               82
               Paducah Test Program
                                ix

-------
                    LIST OF TABLES (CONTINUED)


'Table

  C-l        Scrubber Operating Conditions—
 1            Series S-XX—Key West Initial Salt
             Water Tests

  C-2        Scrubber Operating Conditions—               120
             Series C-XX—Key West Coral

  C-3        Scrubber Operating Conditions—               122
             Series F-XX—Key West Fredonia Valley
             Limestone

  C-4        Scrubber Operating Conditions—               125
             Series HL-XX—Key West Lime, Dolomite,
             Precipitated CaCO_, And Recycled Limestone

  C-5        Scrubber Operating Conditions—               126
             Series P-XX—Paducah-Shawnee No. 9 Limestone

  C-6        Scrubber Operating Conditions—               129
             Series—PA-XX, and PS-XX—Paducah-Shawnee No.
             9 Aragonite and Simulated Key West Limestone

  C-7        Scrubber Operating Conditions—               130
             Series IPA-XX and IP-XX—Paducah-Shawnee No.
             10 Aragonite and Limestone Injection

  D-l        Data Consolidation—Series S-XX— Key West    132
             Initial Salt Water Tests

  D-2        Data Consolidation—Series C-XX— Key West    134
             Coral

  D-3        Data Consolidation—Series F-XX— Key West    136
             Fredonia Valley Limestone

  D-4        Data Consolidation--Series HL-XX, D-XX,       139
             PC-XX, and X-XX--Kwy West Lime, Dolomite,
             Precipitated CaCO~, and Recycled Limestone

  D-5        Data Consolidation—Series P-XX—Paducah-     140
             Shawnee No. 9 Limestone

  D-6        Data Consolidation—Series PA-XX, and         142
             P8-XX—Paduch-Shawnee No. 9 Araeonite and
             Simulated Key West Limestone

-------
                   LIST OF TABLES (CONTINUED)

Table                         Title

 D-7       Data Consolidation—Series IPA-XX and IP-XX   143
           —Paducah-Shawnee No. 10 Aragonite and Lime-
           stone Injection
                              xi

-------
                               ACKNOWLEDGEMENTS

The authors wish to acknowledge the helpful guidance of the Environmental
Protection Agency participants with whom this project was discussed and
especially F. Princiotta and E. L. Plyler of the Control Systems Laboratory.

Experiments during the first phase of this project were carried out at
the main power plant of the City Electric System (CES), Key West, Florida.
The cooperation of the CES and all Its personnel played a key part In
executing this phase of the study.  Particular thanks go to R. Garcia
of the CES for his assistance during the Key West operating phase.

The second phase of the study was carried out at the Shawnee Power
Station of the Tennessee Valley Authority (TVA), Paducah, Kentucky.  The
cooperation of the TVA personnel and on-site EPA personnel played a key
part in executing this phase of the study.

Some of the reactants utilized in this study were supplied gratis by
interested parties.  Special thanks go to Dixie Lime & Stone, Ocala,
Florida (aragonite); Fredonia Valley Quarries, Fredonia, Kentucky
(limestone); and Frank Toppino & Sons, Key West, Florida (coral marl).

The pilot plant was supplied by Zurn Industries, Inc., Erie, Pennsy-
lvania.  The efforts of Zurn's Air System Division, Birmingham, Alabama
contributed significantly towards keeping the pilot plant running
through certain critical periods.
                                xil

-------
                               SECTION 1
                             INTRODUCTION

Abatement of the air pollutants commonly associated with electric
power generation—particulates, sulfur dioxide  (SO.), and oxides of
nitrogen  (NO )--is a primary goal of the public utility industry and
government.  Particulates can be controlled through use of commer-
cially available and proven technology; however, processes for control
of the oxides of sulfur and nitrogen formed in the combustion process .
have not reached the same state of commercial availability.  SO- is
formed when sulfur-bearing fuels, either solid or liquid, are used in
the combustion process; hence, two SO. control strategies become
immediately apparent:  use low-sulfur or sulfur-free fuel or remove
the SO. from the flue gas.  The use of low-sulfur or sulfur-free fuel
depends upon the availability of these fuels.  Currently, the demand
greatly exceeds the supply.  Therefore, removal of SO. from flue gas
is receiving considerable attention from both the public utility
industry and government.

The major problem in removing SO. from flue gases is not the techni-
cal problem of SO. absorption or adsorption, but how to accomplish
this economically without adversely affecting the reliability of
electrical generation.  Handling the required volumes of flue gas by
standard chemical engineering unit operations (such as absorption
towers) is a complex engineering problem which must be thoroughly
investigated in a broad-based research and development program prior
to commercial operation of any process.  Thus, a primary task of the
Control Systems Laboratory, Office of Research and Development., Environ-
mental Protection Agency, has been to investigate a variety of research
and development processes potentially amenable to this problem.

-------
Wet scrubbing of flue gases by various aqueous absorbents capable of
reacting with SCL has been one of the most vigorously investigated
approaches for control of S0?.  The obvious technical advantage of
wet scrubbing is the well-established unit operation of particulate
collection and gas absorption by aqueous scrubbing processes in other
industries.  The Environmental Protection Agency, therefore, undertook
a program to develop the potential of the calcium base or limestone
process for SO- control from power plants.  The process appeared to
be technically simple and the least expensive to install among the
potential first-generation wet scrubbing processes.

There are many variations of the lime/limestone wet scrubbing process
including:
     •  Boiler injection with subsequent wet scrubbing
        Open-loop, tail-end lime/limestone wet scrubbing
        Closed-loop, tail-end lime/limestone wet scrubbing
The dry limestone injection into a power plant boiler followed by wet-
lime scrubbing has been studied on a large prototype test facility by
the Tennessee Valley Authority (TVA).  Other studies are currently
underway on both open and closed-loop tail-end systems since potential
problems related to limestone injection into a boiler are eliminated
with these processes.

Initial pilot plant testing of the Zurn Air Systems Division (ZASD)
limestone wet scrubbing system on boiler flue gas took place in 1969
with favorable results.  Based upon the favorable results from this
early effort, Zurn Industries, Inc. proposed a cost-shared pilot plant
study with the National Air Pollution Control Administration (now EPA)
to investigate certain variables important to the success of the tail-
end limestone wet scrubbing process.  On April 30, 1970, Engineering-
Science, Inc. was awarded a contract for an in-depth study of the
tail-end limestone wet scrubbing process utilizing the ZASD scrubber.

-------
'the  contract  included  the  design,  fabrication, and operation of a
limestone wet scrubbing pilot plant.

The  object  of this  contract was  to evaluate S09 removal capabilities
of this  limestone wet  scrubbing  system  on  flue gases from both oil-
fired  and pulverized coal-fired  boilers while utilizing a series of
different calcium based reactants.

Several  additional  goals of the  program were the:
         Investigation  of potential operating difficulties
      •   Generation  of  data for a scaled-up design
         Evaluation  of  the  process  economics

The  test program was divided into  two major segments:
      •   Key West, Unit No. 3, City Electric System, Key West, Fla.
      •   TVA Shawnee Units  No. 9  and No. 10, Paducah, Ky.

The  Key  West  phase  of  the  study  took place while the pilot plant was
installed on  the No. 3 unit of the Utility Board of the City of Key
West,  Florida.   This unit  was a  20 MW base-loaded steam generator
firing 1-2% sulfur  No. 6 fuel oil.  The sulfur content of the fuel was fairly
constant due  to the method of mixing of fuel shipments; however, on
the  days that fresh fuel was received,  the sulfur concentration in the
flue gas varied widely.

The  Paducah phase of the study took place  while the pilot plant
was  located between the No. 9 and  No. 10 units of the TVA Shawnee
generating  station. Both  of these units were 150 MW steam generators
firing 2-4% sulfur  pulverized coal. The No. 9 unit was conventional while
.the  No.  10  unit had been modified  to allow dry limestone injection for
testing  under another  EPA  contract.

A series of statistical design tests was developed to evaluate
the  effects of design  characteristics,  reactant properties, catalysts,
and  operating conditions on primary dependent variables such as:

-------
        SCL absorption
        Scale formation
      •  Corrosion
        Operating problems
      •  Process costs
The test program resulting from the experimental designs included over
200 tests related to:
        Reactant type
        Stoichiometry
      •  Particle size
        Slurry concentration
        Catalysts
        Liquid/gas ratio
      •  Pressure drop

Twelve series of tests were conducted at Key West and Paducah from
January 1971 through August 1971 to study the aforementioned variables.
Table 1-1 is a list of the test series conducted during this project,
the test site location, and test dates.

The initial salt water tests (S-XX series) were performed to determine
the operational limitations of the scrubber and monitoring equipment.
During this period mechanical and electrical malfunctions were identi-
fied and eliminated to produce a reliable system.  Once accomplished,
the testing program proceeded to the primary test series.

The first series of reactant tests (C-XX series) used a crushed coral
slurry to evaluate its potential use as a reactant for the full scale
demonstration scrubber.  A half factorial statistical design schedule
was followed to gain maximum scrubber operation information from a
minimum number of test runs.  The data generated were later compared
with tests which used the more commonly used slurries such as lime-
stone and lime.  In addition, specific additives which purportedly
enhance SO- removal were investigated during this testing period.
                                   4

-------
                 Table 1-1.  PILOT PLANT TEST SERIES
 ..Test
 identi-
fication
          Test series
     Dates
S-XX


C-XX

F-XX


HL-XX

D-XX

PC-XX


X-XX


P-XX


PS-XX


PA-XX


IPA-XX


IP-XX
Key West^ Initial Salt Water Shake-
  down Test Series

Key West - Coral Test Series

Key West - Fredonia Valley Lime-
  stone Test Series

Key West - Lime Test Series

Key West - Dolomite Test Series

Key West - Precipitated Calcium
  Carbonate Test Series

Key West - Recycled Limestone
  Test Series

Paducah - Shawnee No. 9 Limestone
  Test Series

Paducah - Shawnee No. 9 Simulated
  Key West Limestone Test Series

Paducah - Shawnee No. 9 Aragonite
  Test Series

Paducah - Shawnee No. 10 Aragonite
  Injection Test Series

Paducah - Shawnee No. 10 Limestone
  Injection Test Series
1/7/71 - 1/18/71
1/19/71 - 1/28/71

1/28/71 - 5/25/71

2/6/71 - 5/20/71


4/16/71 - A/20/71

A/20/71 - it/28/11

A/27/71 - A/30/71


A/29/71


6/1A/71 - 8/6/71


7/10/71


7/9/71 - 8/3/71


6/23/71 - 6/25/71


7/1A/71 - 7/28/71

-------
Fredonia Valley Limestone (FVL), the second principal reactant to be
evaluated in this study, was used as a primary standard in gauging
the scrubbing efficiency of the coral slurry.  As with test series
C-XX, a half factorial test design was followed during F-XX series to
gain maximum information concerning scrubber operation.

Test series HL-XX was conducted with a lime slurry which normally
produced SO- removal efficiencies in the 90% and greater range.  The
purpose of this test series was to establish the maximum S0~ removal
efficiency obtainable with the ZASD scrubber design.  However, lime
slurry was considered a secondary reactant for this program due to
the high purchase costs.

A fourth reactant, dolomite, was investigated during test series D-XX
as a potential scrubbing agent possessing lower reactivity and high
magnesium oxide content.

Precipitated calcium carbonate, considered a secondary reactant for
these test purposes, was used as the S0» scrubbing agent during test
series PC-XX.  This material is a highly reactive calcium carbonate
material and was therefore included in this program.

A special test run (X-XX) was performed with a slurry derived from
"spent" coral-limestone material that had made one pass through the
scrubber.  The purpose of this experiment was to investigate the
availability of unused calcium carbonate in a "spent" slurry.

The pilot plant was relocated to a coal-fired power plant in Paducah,
Kentucky, after the Key West series had been completed.  Scrubber
operating conditions during these test series duplicated or simulated
those during the Key West tests.  In addition, several new conditions
were investigated.

The first Paducah test series (P-XX) studied the effect of using a
Fredonia limestone and untreated river water slurry in removing S02
from flue gases generated by a coal-fired power plant.  During the

-------
majority of the test runs, a fly ash collector preceded the scrubbing
system; however, several tests were conducted with the dry collector
bypassed to investigate its effect on the total system.  A half fac-
torial statistical design was followed during this test series.

A secondary test series (PS-XX) was conducted in Paducah to investi-
gate the effects of slurry ionic strength on S02 removal efficiency.
An attempt was made to simulate the Key West salt water slurry by
adding rock salt to river water.

A new reactant was introduced into the program in Paducah during the
PA-XX test series.  The use of aragonite during the Key West tests had
been considered; however, it was unavailable in sufficient quantities
or desired grind size.  Aragonite was reported as a highly reactive
calcium carbonate material  possessing a crystal lattice structure
different from the limestone or coral  and was therefore included in
the Paducah test series.

A brief two day experimental test series (IPA-XX) was designed to
investigate the effect of injecting dry aragonite into the boiler
while attempting to remove S0_ gases in the scrubber with river water.
Similar tests were conducted with dry limestone during test series
IP-XX.

An addendum to this report, available through EPA's Control Systems
Laboratory, contains the raw data collected during the course of this
study.

-------
                               SECTION 2
                          PILOT PLANT EQUIPMENT

 The flexible design of the pilot plant allowed for modifications
 to suit the varied physical configuration expected at each of
 the plant.sites.  Discussed in this section are the physical arrange-
 ments and specifications of system components as installed in Key
 West and Paducah, general descriptions of the operation of each com-
 ponent, and descriptions of the sampling and laboratory facilities.

 PILOT PLANT
 A mobile 8 x 40 foot flat bed trailer supported the entire pilot plant,
 including a scrubber, a primary particulate collection device, pumps,
 mixing tanks and a test house.  After utilities and duct connections
 to the flue gas system were provided, the pilot plant became self-suffi-
 cient.  Figure 2-1 illustrates the mobility of the system and includes
 pertinent design information.

 PARTICULATE REMOVAL SYSTEM
 Approximately 1% of the flue gas leaving the boiler passed through a
 primary particulate removal system (Figures 2-1 and 2-2).  In theory,
 the MTSA-3-9 CYTOCD Whirlex collector (Figure 2-2) removed parti-
 cles by centrifugal action and change in air flow direction.  The unit
 was designed for primary removal of larger size particles generated
 during the coal-fire phase, and was not intended to meet the high
 removal efficiencies required by stringent emission regulations.
 Since the Key West tests (dealing with oil-fired flue gas) and the
 Shawnee No. 10 phase (investigating limestone injection) did not require
 particulate removal, the collector was bypassed by means of a built-in
 damper.  However, the particulate removal system was utilized during
 a greater part of the Shawnee No. 9 phase of testing.
_
 Although EPA policy is to use metric units in all of its documentations,
 certain non-metric units are used in this report for clarity.  Readers
 more familiar with metric units may use the conversion factors in
 Appendix E.
                                    8

-------
        «jcc wn
                                                                                          \-BCTEMUI. M* WCM
                                                                     ELEVATION \-HOPKW SAMHX POUT
HO  2-1 MOBILE PILOT PLANT EQUIPMENT AHBAM6FMFIIT

-------
                                               BYPASS DAMPER
                                         >I8"SQ.
                       INLET  ELEVATION
SIDE   ELEVATION
FIG. 2-2  PILOT  PLANT DRY  DUST  COLLECTOR  OUTSIDE  DETAILS

-------
The material of construction was carbon steel which appeared to be
satisfactory for this application.  However, the bypass damper for the
scrubber system, located on the inlet side of the collector, was made
of 304 stainless steel.  This damper served an on-off function only
and was not used to control the flue gas flow rate to the particulate
collector.  Sample ports were provided in the connecting ductwork for
particulate, SO,, and NO  sampling across the collector as well as
               ^        X
across the entire system.

Fly ash removal was accomplished by removing the bolted sealed blind
flange while the system was down.  Because the collector and ductwork were
not insulated, condensate built up in the fly ash hopper.  It was
necessary to drain this section daily, prior to start-up.

DUSTRAXTOR ABSORBER SYSTEM
The objective of scrubbing for SO. was to economically transfer maxi-
mum quantities of the contaminant from the flue gas to the scrubbing
liquid.  Physical and chemical equilibrium and rate relationships
all limit the mass transfer of SO. in any system.  However, the amount
of mass transfer depends not only upon the equilibrium relationships
but also upon the contacting scheme.  The Dustraxtor employs many
contacting schemes.

Gas leaving the particulate collector passes into a modified single-
tube Dustraxtor.  As shown in Figure 2-3, the design is essentially
a type of turbulent contact absorber (TCA).  The unit consists of
a flooded collecting tube through which the flue gas passes.  The
collecting tube is Installed vertically in the Inlet plenum chamber
directly above the recycle hopper so that the bottom of the tube is
a short distance above the liquid level in the hopper.

Flue gas enters the inlet plenum chamber, sweeps over the surface of
the scrubbing liquid, and is directed up the collecting tube.
The gas passing between the collecting tube inlet and the collecting
bonnet carries scrubbing slurry from the bonnet surface upward Into

                                  11

-------
         'GULL-WING" DEFLECTOR
                                                          GAS OUTLET TO FAN
          DUSTRAXTOR WALL
             GAS/SLURRY MIXTURE

               SLURRY DOWNCOMER
   90° V-NOTCH WEIR-
COLLECTING
TUBE
                                                                    GAS INLET
                                                                    BONNET
                                                                    HOPPER
   WEIR BYPASS VALVE
FIG. 2-3 TYPICAL CROSS SECTION OF DUSTRAXTOR SCU8BER
                                     12

-------
the collecting tube.  The shearing action of the gas atomizes the
scrubbing slurry into a dense spray as the gas slurry mixture con-
tinues up the collecting tube.  The result of this action is a highly
turbulent mixing zone which provides the intimate contact necessary
for the chemical reaction to occur.  As the gas is discharged from
the collecting tube, it is directed into a curved gull-wing deflector
which acts as a mechanical separator and forces the slurry downward
onto the tube sheet.  This shower effect provides an additional mixing
zone  in which absorption can occur.  The cleaned gas is then discharged
via the flue gas exhaust system as  the  scrubbing solution is returned
by gravity to the  recycle hopper.   To summarize, the stages of gas-
liquid contact in  this  type  of  scrubber are:
     1.  The initial shearing action as the flue gas passes through
         the slot between the collecting bonnet and tube.
     2.  The highly turbulent mixing within the collecting tube
         where the gas is in intimate contact with slurry droplets.
     3.  The impingement of the gases and slurry upon the surface of
         the deflector.
     A.  The passage of gas through a highly turbulent curtain of
         slurry being discharged from the deflector.
The contacting schemes found in the Dustraxtor could be categorized
as countercurrent, co-current, and crosscurrent flow.  Since there
were three contacting schemes occurring at the same time in this scrub-
ber, a theoretical analysis of the total mass transfer mechanism
would have been extremely difficult and only of academic interest.
Therefore, this type of an analysis was not conducted.

The Dustraxtor dimensions are shown in Figure 2-4.   This unit differed
from the typical Dustraxtor design in the following ways:
     1.  An exterior weir system was added between the tube sheet
         downcomer and the hopper.
     2.  The capability was added to allow alteration of the tube
         diameter by substitution of different diameter prefabricated
         tubes.
                                 13

-------
                        4-V
-------
The exterior weir provided a means of measuring the quantity of slurry
taken up into the tube, which was later used to calculate the liquid/
gas ratios.  The details of the external return apparatus and the 90°
"V-notch" weir are shown in Figure 2-5.

The liquid level in the scrubber was maintained by a level controller.
The slurry composition was maintained indirectly by adjustment of the spent
slurry flow rate.  The level controller in turn maintained the desired
liquid level by adjusting the fresh slurry feed rate.

In order to avoid problems with solids settling, a mechanical mixer
was installed in the Dustraxtor to help keep solids in suspension.  The
mixer specifications are listed in Table 2-1.  The sump had a 3 inch.
gravity-feed drain which was used to empty the scrubber after each test.

                Table 2-1.  DUSTRAXTOR MIXER SPECIFICATIONS
Trade Name;               Lightnin Mixer
Manufacturer;             Mixing Equipment Company, Inc.
                          138 Mt. Read Boulevard
                          Rochester, New York 14603
Model;                    N33-33, Fixed Mounted Propeller Type
Serial No.;               7012653
Design Specifications;    Motor - 1/3 HP/115V/60 Hz/1 phase/1750 rpm,
                                  totally enclosed
                          Shaft - 304 SS, 3/4 inch diameter x 49
                                  inches, 1750 rpm
                          Propeller - 316 SS, 3.8 inch diameter

The material of construction for the Dustraxtor and connecting ductwork
to the particulate  collector was 304 SS.  All piping to and from the
scrubber was galvanized  steel.  All sample ports on the scrubber were
either stainless  steel or polyvinyl chloride.  A stainless steel butter-
fly valve, located  in the bottom of the exterior weir, was used to bypass
the measurement system.
                                    15

-------
  Downcomer
  Exit
                              Weir Bypass Valve
                                                          Hopper
FIG. 2-5  DUSTRAXTOR OUTSIDE RETURN APPARATUS AND 90° V-NOTCH WEIR
                                     16

-------
A slide gate valve, installed on the 10-inch diameter discharge duct
was used to control the flue gas flow rate through the entire pilot
plant.  The pressure drop across the Dustraxtor was controlled by
adjusting the balance vent line control valve.  Sample ports were pro-
vided in the connecting ductwork for particulate, S0_, and NO  sampling
                                                    £        X
across the Dustraxtor, as well as across the entire system.  Liquid
sample ports were provided on the Dustraxtor body for tube sheet and
hopper liquor sampling.  A set of spray nozzles for gas saturating pur-
poses was installed in the ductwork between the particulate collector
and the Dustraxtor.

FLUE GAS EXHAUST SYSTEM
The scrubbed flue gas leaving the Zurn scrubber passed through an
induced-draft fan with an inlet damper control, to a vertical stack
containing sampling ports.  The, induced-draft fan was a No. 10 Clarage
Blower; the fan curve is shown in Figure 2-6.  The I-D fan had a paddle-
blade impeller which, because of particulate carry-over and subsequent
"plating-out" on the blades, became unbalanced and failed twice during
the study.  A field review of the I-D fan failures indicated that the
cast stainless steel spider, which secures the blade, fractured due
to fatigue caused by the fan imbalance.

PILOT PLANT SCHEMATIC

Figure 2-7 shows the overall flow diagram of the pilot plant.   The
location of thermometers and flow measuring devices are also shown
In this Figure.  Additional Information concerning the monitoring
system is included in Section 3: Test Instrumentation and Procedures.

REACTANT HANDLING AND WASTE DISPOSAL SYSTEMS
The calcium based scrubbing slurry was prepared in mix tanks.   The mix
tanks were equipped with steam coils and gear driven portable mixers.

                                   17.

-------
              24 i-
                                                    STATIC  PRESSURE
                                          BRAKE  HORSEPOWER
                        1000
2000     3000    4000     5OOO
     FLOW RATE OF AIR - scfm
6000
                                                                                  3450 RPM
                                                                                  Std. Air Wt. = 0.075 Ib/ft3
FIG. 2-6 NO. 10 CLARAGE BLOWER INDUCED DRAFT FAN CURVE

-------
 The steam coils were not used during the study;  however,  they were
 available for use if desired.  The mixers had a  304 stainless steel
 shaft, a 316 stainless steel 10 inch diameter propeller,  and were driven
 by a 1/3 HP, 115 volt, 1750 rpm, totally enclosed electric motor.
 Detailed tank mixer specifications are shown in  Table 2-2.
                  Table 2-2.  TANK MIXER SPECIFICATIONS
Trade Name;
Manufacturer:
Model;
Serial No.;
Design Specifications;
     Lightnin Mixer
     Mixing Equipment Company, Inc.
     138 Mt. Read Boulevard
     Rochester, New York 14603
     ND-1, Portable Mixer
     None, Mixco Order No. M315836
     Motor - 1/3 HP/115V/60HZ/1 phase/1750 rpm,
             totally enclosed
     Shaft - 304 SS, 5/8 inch diameter x
             48 inches, 1750 rpm
     Propeller - 316 SS, 10 inch diameter
 The scrubbing slurry from the mix tanks was fed by gravity to a high
 silicon iron Mark II Durcopump manufactured by the Duriron Company, Inc.
 The pump specifications are noted in Table 2-3.  The slurry feed rate
 from the pump to the scrubber was measured by a flowmeter located near
 the scrubber.  Excess slurry was returned to the mix tanks through a
 recirculation line shown in Figure 2-7.
                 Table 2-3.  SLURRY PUMP SPECIFICATIONS
 Manufacturer:
 Pump Design:
 Performance:
Duriron Company, Inc.
N. Findlay and Thomas Streets
Dayton, Ohio 45401
Series - 1-1/2 x 1 H - 6/60; Size - Mark II GPI
Packing - Standard; Alloy - Superchlor
Impeller Diameter - 6 inches; Shaft Wet End - Superchlor
20 gallons per minute
36 feet of water total differential head
NPSH - 1.5 feet of water net positive suction head
0.9 maximum brake horsepower
                                     19

-------
N>
O
                                                •GAS OUTLET
                     NC£  MIX TANK

                  DRAIN^
                                           SLURRY  PUMP
                                                                                                                    GAS INLET


                                                                                                   FLOW MEASURING ELEMENT
                     REACTANT8
                     FEED WATER
LEGEND
X  Slurry  Samples
o  Temperature, Dry Bulb
&  Temperature, Wet Bulb
O  S02, NOX, And Partlculate Sample Point
NC-Normally Closed
NOrNormally Open
FM-Flow Measurements
M  Mixer
         FIG. 2-7 SCHEMATIC FLOW DIAGRAM OF FLUE GAS AND SLURRY SOLUTIONS  WITH LOCATIONS OF  SAMPLING POINTS

-------
The required quantity of the calcium-based reactant was weighed out
and dumped into each mix tank containing fresh or salt water depending
upon the location of the pilot plant.  The amount of reactant was set
by the required slurry concentration.  Adequate time was allowed for
proper mixing of the calcium-based reactant and the  water before
pumping into the scrubber system.  Fresh slurry was fed to the scrubber
through the flowmeter while the spent slurry, from the scrubber system,
passed out the level controller.

During the Key West testing, the spent effluent was deposited in a
holding pond where the liquid was allowed to evaporate or permeate the
coral soil.  During the Shawnee testing, the effluent was deposited
in a storm sewer, mixed with other plant waters, and pumped to the
Mississippi River.  Each disposal technique was satisfactory for the
pilot scale system.

Several operating problems concerning the reactant handling system were
noted during the study.  Due to infrequent maintenance checks, the feed
pump packing developed a leak (a marked reduction of pump pressure was
experienced) and became progressively worse as the project continued.
A more serious problem was encountered at the Shawnee site.  Settleable
solids in the waste material often clogged  the flexible hoses.   This
necessitated frequent flushing with a fresh water stream to prevent
back-up into the scrubber.

CHEMICAL ANALYSIS LABORATORY
During the Key West testing, the City Electric System power plant lab-
oratory was utilized for all chemical analyses.  The laboratory was
equipped with instrumentation necessary for conducting the analyses
noted in Appendix A.
                                   21

-------
                               SECTION 3
                  TEST INSTRUMENTATION AND PROCEDURES

In order to facilitate movement of the pilot plant, basic but highly
reliable instrumentation was utilized for this study.  The SO. con-
centrations were determined with an electrochemical type sensor which
was calibrated daily with guaranteed-analysis calibration gas.  All
temperatures were determined with properly calibrated thermometers,
and gas flow rates wete determined with a calibrated sharp-edge orifice.
Standard sampling techniques were employed for particulates, SO., and
NO .  Details of the analytical techniques are in Appendix A.
  X
Test procedures were established based upon the.experimental design
required for each phase of the test program.  Day-to-day operation of
the pilot plant was held as constant as possible, with the only varia-
tion occurring in the level of the variable set for each test.  The
test procedures employed at the two sites, Key West and Shawnee, were
varied only where necessary in order to comply with necessary conditions,

S02 MONITORING
One of the most important analyses of the test program was the measure-
ment of the SO. concentration in the flue gas.  SO. sampling probes,
fabricated from 6 inch lengths of 1/4 inch diameter SS tubing, were
located in the entrance and discharge ducts of the Dustraxtor.  Analyses
were performed for the most part by a Dynasciences Model SS-330 monitor.
Selection of this monitor was based in part on the results of limited
test work conducted earlier.

The Dynasciences monitor operated on the principle of a fuel cell.
SO. was absorbed on a sensing electrode to form activated species capa-
ble of undergoing electro-oxidation.  The resulting current was directly
proportional to the partial pressure of SO. in the gas mixture.  The
current was amplified and the output recorded on a meter and a 10 mV
recorder.

                                  22

-------
The instrument specifications indicated linear response over the entire
range of SO. concentration.  Overall accuracy was specified at + 1% full
scale with the use of an external potentiometric recorder.  Response
time was specified as 90% of full scale in 1 minute.  The instrument
was to exhibit no response to N_, 0_, NO , CO, CO^, water vapor or
hydrocarbons.  In addition, the instrument was very portable and well
suited to the field environment experienced in pilot plant operations.

In operation, the Dynasciences Model SS-330 did not meet all the speci-
fications.  Response time was not as rapid as claimed; the instrument
was sensitive to changes in sampling gas flow rate (slight changes in
operating pressure), and changes in ambient temperature.  Also, the
instrument stability was considerably less than specified.  Electro-
chemical cell life was better than claimed by the manufacturer and once
the initial operating problems were corrected, SO,, monitoring operations
were very reliable.  In general, the instrument proved satisfactory for
the conditions experienced during the pilot plant operation.

NO  SAMPLING AND ANALYSIS
  X
NO  samples were taken randomly during certain test periods.  NO  probes
  X                                                             X
were located at the entrance and discharge ducts of the Dustraxtor
virtually in line with the SO- sampling probes.  The phenoldisulfonic
acid method was used for analysis.

A typical NO  sampling and analysis consisted of adding 25 ml of an
absorbing solution (hydrogen peroxide and dilute sulfuric acid) to a
250 ml (nominal) evacuation flask.  The flask was evacuated and attached
to a purged sample line; the stop cock was opened; and the gas sample
was drawn into the flask.

NO  was converted to nitric acid by the absorbent solution and reacted
  A
with phenoldisulfonic acid to produce a yellow compound which was
measured colorimetrically.  Color was measured with a photometer and
compared with calibration curves from solutions containing a known
quantity of nitrate.

                                 23

-------
A more complete and descriptive explanation of the phenoldisulfonic
acid method used in these tests is given in the American Society for
Testing and Materials (ASTM), Standard Method of Test for Oxides of
Nitrogen in Gaseous Combustion Products (Phenoldisulfonic Acid Proce-
dure) , ASTM Designation: D1608-60 (1967).  The analytical technique
associated with this test procedure was very tedious and time consuming.
However,.since other procedures had not been adequately demonstrated
on power plant flue gases, it was necessary to use this method to
obtain reproducible data.

PARTICULATE SAMPLING
Particulate sampling was conducted at three points in the pilot plant:
the entrance to the dry dust collector, the entrance duct to the Dust-
raxtor, and the Dustraxtor exit duct.  The sampling train consisted of
a stainless steel probe and nozzle, a glass fiber filter and a series
of three Greenberg-Smith impingers.  During the first attempts at
particulate sampling, the glass fiber filter clogged as a result of the
high moisture content present in the flue gas (the filter was not
heated).  The filter medium was therefore eliminated from the train
during the remaining tests.  The remainder of the particulate sampling
system included a gas meter and vacuum pump.

Isokinetic sampling was performed by regulating the sample flow tate
to correspond with the calculated velocity at the sampling point.
This method was justified over the null balance procedure for this
study to conform with suggested EPA test method.  Uniform flow rates
were demonstrated by observing the flue gas orifice over an extended
period of time.

SLURRY SAMPLING AND ANALYSIS
During each test, representative slurry samples were taken at the
following locations:
         Slurry feed
         Scrubber discharge
                                 24

-------
         External weir
     •   Hopper sump
         Tube sheet
Figure 2-6 illustrates the various sampling points.  During every test
at least one 100 ml slurry sample was taken for analysis -of the following
parameters:
     •   PH
     •   Calcium
         Magnesium
         Chloride
         Nitrate
         Nitrite
         Sulfite
         Sulfate

Sampling was performed in a manner to avoid oxidation; the samples
were placed in an ice water bath for transportation to the laboratory.
pH was measured in the field as well as in the laboratory.  Standard
gravimetric, titrimetric, and colorimetric methods were used by the
chemist following procedures outlined in American Society for Testing
Materials (ASTM) 1970 Annual Book of ASTM Standards. Part 23, Water;
Atmospheric Analysis, and the American Public Health Association,
Standard Methods for the Examination of Water and Wastewater, 12th
Edition.  The slurry samples which contained solids and liquid were
not separated prior to the analysis.
                                                                    •
OTHER MEASUREMENTS
Field measurements for pH of the slurry samples were made with a
Sargent-Welch Model PBX pH meter and laboratory ph measurements were
made with a Sargent-Welch Model LSX pH meter.

Flue gas flow rate was measured by an Ellison Instrument Division,
12 inch "Annubar," type 740.  The Annubar is a primary flow element
utilising a form of the classical Bernoulli energy balance equation
to d«t«nd.n« flow rate.  The equation used was:

                                25

-------
                             2  \/^r
                 Qn = 7.9  SND   ' '
A full description of the symbols of this equation, and other equations
available for use with this flow element, were supplied by Ellison
Instrument Division, Boulder, Colorado, and are reproduced in Appendix B.
The Annubar included an interpolating tube with equal annuli segments, an
equalizing element, and a downstream element for measuring the down-
stream pressure (static pressure less the Impact pressure of the flow).
The system had a non-clog design which was desirable for operation in
the pilot plant.  The unit was calibrated against standard pitot tube
traverses prior to initial testing in Key West.  Detailed drawings and
specifications of the Annubar were supplied by Ellison Instrument
Division and are also reproduced in Appendix B.

GENERALIZED TEST PROCEDURE
A typical testing day in which three tests were usually completed is
described below.  The experimental design established the operating
levels for the variables under study.  These variables were set during
the test period.

In preparation for a run with a cold start (that is, the scrubber had
been idle long enough to allow the ducts to accumulate condensation),
the unit was first filled with salt or fresh water and allowed to run
for a minimum of 30 minutes to attain operating temperatures and to
flush the accumulated condensate out of the ductwork.  Normal conditions
for this warm-up and flushing were a gas flow rate of 1000-1450 scfm,
liquid feed rate of 10-15 gpm and a AP of 9-12 in. H20.

While waiting for the system to reach operating conditions, the two
mix tanks were flushed and cleaned; the proper quantity of reactant for
                                 26

-------
the test was calculated, weighed out, and mixed in one tank; and the
liquid for the warm-up operation was pumped from the remaining tank.
After warm-up, the system was shut-down (fan off, pump off, reactant
feed valve off, inlet damper closed) and drained immediately.  While
the Dustraxtor was draining, the proper quantity of reactant was
weighed out and mixed in the second feed tank forming the scrubbing
slurry-

When the scrubber was drained, the drain valve and weir butterfly shut-
off valves were closed.  Slurry was then pumped into the scrubber at
maximum flow rate until the Dustraxtor was full.  The sequence of events
concerning the controls were as follows:
     1.  Hopper mixer - on
     2.  Reactant flow - 2-5 gpm
     3.  Level control - 50% open
     A.  Inlet damper - open
     5.  Fan - on

Once the system was in operation, the levels were set for the various
operating parameters.  During the first hour of a test, the gas flow
rate, slurry flow rate, and pressure drop were monitored to assure that
the unit would approach equilibrium at the predetermined conditions.
Based upon preliminary test results, steady state was assumed after
the system had been in operation for 2 hours and the SO- concentration
at the pilot plant exit remained constant.

Prior to final data acquisition, the S0» monitor was calibrated and
the inlet SO- concentration was determined.  During this time, the
inlet gas conditions were recorded (temperature, flow rate, pressures
and SO. concentration).  The required stoichiometric ratio based
on the inlet SO,,, was calculated and adjusted as necessary.  The
SO- monitor was then used to analyze the Dustraxtor outlet gas.  When
the effluent SO- concentration stabilized, outlet conditions were
recorded (temperature, flow rate, pressure, and S02 concentration).
                                 27

-------
During a test, several slurry samples were collected and placed in an
ice water bath for storage prior to laboratory analyses.  The samples
collected and stored in 8-ounce plastic bottles included the following:
slurry feed, scrubber discharge, weir overflow, tube sheet, and Dust-
raxtor hopper samples.

After a test was completed, operating conditions were changed for the
next test.  The unit was shut down, drained, and refilled with reactant
for another test.  There was usually a working period of about 1/2 hour
before the system was ready for another test.  Following this procedure,
about three 2 1/2 hour tests were completed every day.

TEST PROCEDURE FOR EXTENDED OPERATION
The start-up for extended operation did not differ from that of a normal
test day.  Test conditions were specified to be held constant for approx-
imately an 8 hour period.  No variables were changed during this test
period.

TEST PROCEDURE FOR ADDITIVE OPERATION
The start-up for additive experiments (catalyst—Fed- or inhibitor--
hydroquinone) did not differ from that of a normal test day.
All experimental procedures outlined above were followed with the
following additions and modifications:
     1.  Two identical reactant tanks were mixed: one contained the
         the test additive, the other did not.
     2.  The scrubber was filled and started up on reactant without
         the test additive.
     3.  Inlet and outlet conditions were recorded as outlined earlier.
         During some experiments, scrubber effluent samples were not
         taken at this time.
     4.  After stabilization of the outlet conditions, the correct
         amount of additive was added to the scrubber weir simulta-
         neously with changing the reactant feed to the tank with
         additive.  The feed rates remained constant.
                                     28

-------
5.  The outlet conditions were allowed to re-stabilize after
    additive addition; they were recorded again and scrubber
    effluent samples were collected.
                              29

-------
                           SECTION 4
                       SUMMARY OF RESULTS

Results of both the short term and extended tests were consolidated
into one set of results for each test site.  Because of the great
volume of data gathered during the test program, only a summary of the
results is presented in this report.  An addendum, available through
EPA's Control Systems Laboratory, contains all test data taken during
the 10 months of field work.  Appendices C and D contain scrubber
operating conditions and data summary tables, respectively.

SYSTEM OPERATING CONDITIONS AND CHARACTERISTICS
A summary of the Important system operating conditions and character-
istics including data important to future design and optimization of
the Dustraxtor absorber is given below.

Liquid Entrainment Relationships
The relationships of the parameters governing liquid entrainment in the
Dustraxtor unit are presented in Figures 4-1 and 4-2.  Approximately
170 data points are represented showing the association between the
liquid entrainment and gas flow rate at selected pressure drops (AP)
from 6.5-14.0 in. H_0.

Figure 4-1 applies to a single 12-inch diameter Dustraxtor tube.  Each
curve, representing a constant pressure drop condition, exhibits a
point of maximum liquid entrainment which becomes more apparent as
pressure drop increases.  Thus, at a AP of 6.5 inches, a + 350 scfm
variation in gas flow rate about the maximum entrainment point results
in a 15 gpm reduction in liquid entrainment.  Compared to this, at a
AP of 12.0 inches, a decrease in entrainment by 15 gpm is caused by a
gas flow variation of only + 170 scfto.  Another characteristic of this
maximum point is its tendency to occur at progressively lower gas flow
                                  30

-------
       2100
       1900
       1700
    to
    3
       1300
        1100
        900
        TOO
        500
                  40      80      120      (60    200     240     280     320     360     400     440     480

                                                 LIQUID  ENTRAINED-gpm
FIG. 4-1. SALT WATER ENTRAINED IN THE  12" TUBE  AT  SELECTED PRESSURE  DROPS AND GAS  FLOW RATES

-------
co

NJ
                      1000  -
                      900  -
                      800 -

                   I
                   8
                   O
                      700 -
                      600 -
                       500 -
                      400 -
                      300
20
40
                                                    60       80       100       120



                                                         LIQUID  ENTRAINED -gprn
160
180
         FIG. 4-2. RIVER WATER  ENTRAINED IN  THE 8" TUBE AT SELECTED  PRESSURE DROPS  AND GAS FLOW RATES

-------
rates as pressure drop increases.  As an illustration, a maximum
entrainment of 112 gpm is realized at about 1300 scfm for a 6.5 inch
AP.  At a 12 inch AP the maximum entrainment (396 gpm) has dropped to
900 scfm and at a 14 inch pressure drop the maximum has fallen to
700 scfm, nearly half of the flow rate and velocity of the gas stream
for the 6.5 inch AP maximum.

As gas flow rate and corresponding gas velocity increase above the
maximum entrainment point, all of the curves converge upward toward
a greatly reduced range of entrainment volumes.  This trend is so
pronounced that if assumed to continue at the same rate, extrapolation
of the data would indicate a minimal gain of 50 gpm in liquid entrain-
ment at a 2700 scfm flow rate when the AP changed from 6.5 to 14.5
in.i.H-O.

Figure 4-2 is a corresponding plot of the scrubber characteristics
utilizing a single 8 inch diameter Dustraxtor tube.  The same general
shape and trends appear for the 8 inch tube as were exhibited for the
12 inch tube.  However, the maximum entrainment points occured at
lower gas flow rates and higher gas velocities  and were 1/2 to 1/4
the order of magnitude found for the 12 inch Dustraxtor tube.

No attempt has been made to analyze the actual physical mechanism of
entraining and lifting the slurry from the surface within the hopper
or its transportation through the scrubber; hence, a complete under-
standing of the dynamic principles illustrated by the data of Figures
4-1 and 4-2 is not possible.  The development of a model to establish
the analytical relationship of liquid entrainment as a function of
AP and gas flow was not attempted.

Liquid/Gas Ratios
Figure 4-3 is a plot of liquid/gas ratios versus gas flow rates for both
the 8 inch and 12 inch Dustraxtor tubes.  Two corresponding pressure
drops are depicted for each size tube (9 and 12 inches for the 8 inch
                                     33

-------
co
                   3000
                   2600
   2200
I
r
3 MOO
                   1000
                    600
                    200
                                100       200      300       400      500
                                                 LIQUID/GAS  RATlO-gal/Mscf
                                                                600
700
800
900
        RG. 4-3. UQUID/GAS RATIOS FOR  8  8 12  DIA. DUSTRAXTOR TUBES AT SELECTED PRESSURE DROPS AND GAS FLOW  RATES

-------
diameter tube; 8.5 and 12 inches for the 12 inch diameter tube).  These
curves were wholly derived from the data points shown in Figures 4-1
and 4-2.  For the sake of clarity, the actual data points have been
eliminated.  All of the curves show an expected maximum liquid/gas
ratio at some relatively low gas flow rate.  The magnitude of this max-
imum increases with increasing values of AP; however, the point of max-
imum liquid/gas ratio occurs at a slightly lower gas flow rate than
the maximum entrainment volume (Figure 4-1 and 4-2).  As an example:
maximum entrainment volume for a 12 inch AP and a 12 inch Dustraxtor
tube occurs at about 850 scfm while the maximum liquid/gas ratio,
680 gallons per thousand cubic feet (gal./Mcf), occurs at a significantly
lower gas flow rate of about 420 scfm.  Also at high flow rates the
liquid/seas ratio:.seems to fall off very rapidly, approaching zero
for the 12 inch tube at 2700 scfm and at 1400 scfm for the 8 inch tube.
Therefore, from Figure 4-3, it would appear that liquid/gas ratio is
independent of pressure drop at higher gas flow rates.

Finally, Figure 4-3 exhibits the same trend as Figures 4-1 and 4-2
regarding tube diameter; that is, increasing tube diameter results in
an increased liquid/gas ratio.

Characteristics and Analysis of Feed Reactants and Discharge
While lack of sufficient instrumentation and funds prevented a detailed
study of the chemical reaction kinetics and mechanisms occurring during
the scrubbing process, analyses were made of the chemical constituents
important to the understanding of the chemical processes occurring in
                                                             • I     I I
the Dustraxtor absorber.  These analyses were made for pH, Ca  , Mg  ,
Cl , NO, , N02 , SO.  , and SO,  .  A summary of the chemical analyses
results for the scrubber feed and discharge streams of each reactant
tested during the test program are shown in Tables 4-1 through 4-11.

Characteristics and SOp Analysis of Influent Flue Gases
A summary of the influent flue gas characteristics for both the oil-
fired Key West unit No. 3 and coal-fired Shawnee units No. 9 and 10 is
                                  35

-------
                Table 4-1.  SUMMARY OF SALT WATER TESTS
                            CHEMICAL ANALYSES
                            (Key West)

Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum
PH
Feed
7.6
7.8
7.4
Discharge
2.6
2.9
2.3
Ca
(mg/1)
475
1040
410
432
530
420
Mg
(mg/1)
1385
1510
1215
1375
1440
1360
Cl
(ms/1)
23,207
25,000
19,800
21,667
23,500
20,000
N03
(tng/1)
0.40
0.60
0.02
0.20
0.80
<0.01
NO,
(mg/D
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
SOo
(mg/1)
1
1
1
32
127
3
S04
(mg/1)
2640
2740
2125
3171
3840
2800
                 Table 4-2.  SUMMARY OF CORAL MARL
                             CHEMICAL ANALYSES
                             (Key West)
             PH
  Ca     Mg
(mg/1)  (mg/11
  Cl     N03
(mg/11  (mg/1)
 N02    S03    S04
(mg/1)  (mg/1)  (mg/1)
Scrubber Feed
Average     7.7    541
Maximum     8.0   2180
Minimum     7.5    400
Scrubber Discharge
Average     5.5   1614
Maximum     6.4   2800
Minimum     4.7    880
        1431   22,400    0.9
        1920   30,000   16.0
        1340   19,000    0

        1471   23,180    3.0
        1875   30,000   15.2
         584   12,500    0
           2    2872
           7    3840
           1    2220
                 0.14
                 0.40
                 0
                 2.27   6419
                10.40  40,000  6100
                 0      1110   3500
                                36

-------
Table 4-3.  SUMMARY OF FREDONIA LIMESTONE




            CHEMICAL ANALYSES




            (Key West)

Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum
PH
Feed
7.6
7.9
7.1
Discharge
5.6
6.1
4.5
Ca
(mR/1)
492
600
400
1551-
1960
1120.
Mg
(mR/1)
1401
1460
1340
1498
1940
875
Cl
(OR/D
22,520
26,500
20,000
22,900
31,500
15,000
Table 4-4. SUMMARY OF
N03
(ma/l)
0.50
2.80
0
1.60
5.80
0
HYDRATED
N02
(mg/1)
0.20
0.60
0
2.20
7.92
0
LIME
SOo
(mg/1)
1
2
1
5601
23300
900

S04
(mg/1)
2810
3040
2750
5183
5400
3700

CHEMICAL ANALYSES


Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum

PH
Feed
11.4
11.6
11.1
Discharge
8.3
8.7
7.7

Ca
(mg/1)
2800
2960
2680
80
100
56
(Key
Mg
(mg/1)
158
267
97
2697
3340
2330
West)
Cl
(mg/D
20,900
21,500
20,500
21,200
22,000
20,500

N03
(mg/1) 1
0.70
0.80
0.60
0.70
1.20
0.20

NOo
taxi)
0.08
0.15
0.01
0.22
0.35
0.13

S03
(mg/1)
2
3
1
25,400
37,300
11,900

S04
(mg/D
2815
2840
2760
4203
4850
3660
                    37

-------
                       Table 4-5.  SUMMARY OF DOLOMITE
                                   CHEMICAL ANALYSES
                                   (Key West)
Ca
pH (mR/1)
Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum
Feed
7.6
7.7
7.5
Discharge
5.2
5.4
4.9
485
560
440
599
840
116
Mg
(mg/1)
1386
1410
1340
1563
1850
1410
Cl
(rng/1)
21,500
21,500
21,500
20,100
21,500
18,000
N03
(mg/1)
1.20
1.60
0.80
2.70
3.20
1.80
N02
(me/1)
0.50
0.90
0.10
0.03
0.09
0.01
SO 3
(mR/1)
2
3
1
515
600
450
SO*
(mR/1)
2963
3000
2920
4740
6740
4050
         Table 4-6.  SUMMARY OF PRECIPITATED CALCIUM CARBONATE
                     CHEMICAL ANALYSES
                     (Key West)
             PH
  Ca     Mg      Cl
(mg/1)  (mg/1)   (mg/1)
 N03    N02
(mg/1)  (mR/1)
 S03    SO*
(mg/1)  (mg/1)
Scrubber Feed
Average     7.6    467
Maximum     7.6    480
Minimum     7.5    440
Scrubber Discharge
Average     5.9   1133
Maximum     6.2   1640
Minimum     6.0    880
        1355   21,300
        1385   22,500
        1340   20,500

        1360   16,200
         804   20,500
         655   13,000
 1.10   0.10
 1.20   0.14
 1.00   0.07

 0.90   0.18
 1.40   0.27
 0.40   0.09
   2    2800
   2    2840
   2    2760

 1919   3677
 3500  41200
 1000   3250
                                   38

-------
Table 4-7.    SUMMARY OF SHAWNEE NO. 9 FREDONIA
              LIMESTONE CHEMICAL ANALYSES
              (Paducah)

Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum

Feed
8
8
8
pH
.2
.5
.0
Ca
(mg/1)
24
32
12
Mg
(mg/1)
4.9
7.4
3.5
Cl
(mg/1)
14
15
13
NOo
(mg/1)
3.30
5.50
1.50
NO 2
{mg/1)
0
0
0
.27
.35
.16
SOo
(mg/D
4
5
2
so4
(mg/1)
43
131
0
Discharge
5
5
5
.3
.6
.0
1063
1440
600
Table 4-8.
61.9
97.0
48.5
373
800
100
7.50
15.50
3.50
SUMMARY OF SHAWNEE NO.
0
0
0
9
.10
.16
.09
1585
2600
950
1654
2479
910
ARAGONITE
CHEMICAL ANALYSES
(Faducah)

Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum

Feed
8
8
8
pH
.1
.1
.1
Ca
(mg/1)
36
38
34
Mg
(mg/1)
11.3
12.0
9.5
Cl
(mg/1)
60
90
30
N03
(mg/U
3.9
5.0
2.5
NO 2
(mg/1)
0
0
0
.20
.22
.18
S03
(mg/1)
2
2
1
so4
(mg/1)
108
156
41
Discharge
6
6
5
.1
.3
.8
570
1000
360
48.6
48.6
48.5
211
350
150
3.8
5.5
3.0
0
0
0
.34
.55
.16
478
780
300
1061
2060
580
                         39

-------
Table 4-9.   SUMMARY OF SHAWNEE NO. 9 SALTWATER/FREDONIA
             LIMESTONE CHEMICAL ANALYSES
             (Paducah)
PH
Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum

Feed
8.
8.
8.
2
2
1
Ca
(mg/1)
60
60
60
Mg
(mg/1)
8.
8.
8.
5
5
5
Cl
(mg/1)
12,000
12,000
12,000
N03
(mg/1)
2
2
2
.50
.50
.50
NO?
(mg/1)
0.40
0.40
0.40
SOo
(mg/1)
1
1
1
S04
(mg/1)
58
58
58
Discharge
5.
5.
5.
Table
6
7
4
4-10
1040
1280
800
48.
48.
48.
. SUMMARY
CHEMICAL
5
5
5
OF
9,500
10,000
9,000
SHAWNEE
3
3
2
NO.
.30
.50
.00
0.17
0.17
0.16
10 ARAGONITE
865
1150
580
2120
2880
1360
INJECTION
ANALYSES
(Paducah)
PH
Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum
Feed
7.
8.
7.
9
1
8
Ca
(mg/1)
58
92
24
Mg
(mg/1)
2.
3.
2.
7
5
4
Cl
(mg/1)
10.8
13
10
N03
(mg/1)
35
52
1
.20
.00
.60
N02
(mg/1)
0.17
0.18
0.16
803
(mg/1)
4
8
2
804
(mg/1)
125
205
24
Discharge
7.
7.
7.
8
9
4
520
600
440
60.
72.
48.
7
9
5
200
250
150
3
8
1
.90
.00
.60
4.20
6.00
3.00
37
80
15
1021
1230
740
                            40

-------
      Table 4-11.   SUMMARY OF SHAWNEE NO. 10 FREDONIA LIMESTONE
                    INJECTION CHEMICAL ANALYSES
                    (Paducah)
Ca
PH (mg/1)
Scrubber
Average
Maximum
Minimum
Scrubber
Average
Maximum
Minimum
Feed
7.9
8.1
7.4
Discharge
5.8
6.9
4.9

37
64
28

643
1320
320
Mg
(mg/1)

5.7
6.0
5.0

33.6
48.6
24.3
Cl
(mg/1)

15
15
15

135
200
100
N03
(mg/l)

2.00
3.10
0.20

5.20
7.50
2.40
NO?
(mg/1)

0.39
1.50
0.20

0.61
3.60
0.18
S03
(mg/1)

2
2
1

557
1700
100
SO,
(mg/1)

61
80
24

1200
2080
310
presented in Tables 4-12 and 4-13, respectively.  In addition, plots of
the variation in entering SO- concentration with time at both test
sites are shown in Figures 4-4 and 4-5.  The major variations in inlet
SO  concentration at the Key West site were attributed to a change in
sulfur content of the fuel oil.  This was either due to stratification
of different sulfur content fuels in the fuel storage tank or a change
in the total sulfur content in the tank, the latter resulting from
the mixing of fuel oil shipments in the tank.  Similarly, S0? variations
at the Shawnee test site were attributed to the wide variation of sulfur
present in the coal supplied to the plant.  In addition, during the
month of July, TVA was conducting precipitator efficiency tests on
Shawnee No. 10.  These tests frequently increased dilution air leakages
into the flue gas upstream of the pilot plant entrance.

Scale Deposition
Deposition of scale in piping, fans, pumps, tanks, sumps, and on the
absorber internals varied from nondetectable to moderately heavy during
the Key West and Shawnee No. 9 test programs.  The degree and rate of
                                  41

-------
                      I
                     CC
                     Ul
                     O
                     CD
                     3
                     CC
                     O
                     en

                     o
                     UJ
                     <
                     oc
                     Ul
                     o
                     CO
1000
                           500
                                    JAN.
                       FEB.
    MAR.


TIME
APR.
MAY
FIG. 4-4. VARIATION  OF INLET S02 CONCENTRATION, KEY WEST - 1971

-------
                        2500
                    1.  2000
                    ex

                    r


                    UJ
                    U

                    CO
                         1500
O


£
m






§







8


U   1000




in
                         500
                                 JUN.
                                             JUL.


                                           TIME
                                    AUG.
FIG. 4-5. VARIATION  OF INLET  S02 CONCENTRATION, PADUCAH - 1971




                                           43

-------
      Table 4-12.  SUMMARY OF INLET FLUE GAS CONDITIONS
                   (Key West Unit No. 3)


Average Dry Bulb Temperature             231°F
Average Wet Bulb Temperature             130°F
Average SO- Concentration                703 ppm
       Table 4-13.  SUMMARY OF INLET FLUE GAS CONDITIONS

                    (Shawnee Units No. 9 and 10)


Average Dry Bulb Temperature             191°F
Average Wet Bulb Temperature             119°F
Average S0? Concentration               1699 ppm
                             44

-------
deposition was not quantified except in terms of how rapidly the
scrubbing system was affected.  While scale deposits did occur, the
only interference with the system operation was a fan failure caused
by build-up on the blades.

The deposits resulted primarily from precipitation of calcium salts and
deposition of the reactant material.  The composition of the deposit on
the tube was identified by x-ray diffraction and is presented in Table
4-14.  Deposits were not found in the mix tank, pumps, or related
piping.  Deposits were found on the weir hopper and inside scrubber
wall at the water line, the tube at the liquid/gas interface, the mist
elimination section, and in the fan housing.  It should be noted that
all deposits except those found on the tube and in the fan were very.
minor and did not interfere with the system performance over a 6 month
period.  The deposits on the tube occurred moderately fast and the rate
of formation appeared to increase with slurry concentration.  The tube
deposits took the form of stalactites which grew directly into the gas
path,  rf allowed to grew for-an extended time period, these deposits
would interfere with system performance.  Table 4-15 summarizes the
degree of deposition at various locations in the pilot plant during
both the Key West and TVA Shawnee Unit No. 9 tests.

             /Table 4-14.  X-RAY DIFFRACTION ANALYSES OF SCALE
                          FORMATION ON THE TUBE
                          (Key West)
Compound
CaSO, •
CaC03
MgCO
• 2H20
Quantity (wt%)
30.5
68.0
1.5
                                  45

-------
      Table 4-15.  DEGREE OF DEPOSITION AT VARIOUS SYSTEM LOCATIONS
                   (Key West Unit No. 3 and Shawnee Unit No. 9)
                                                           Rate of
	Location	Degree	Formation
Mix Tanks                         negligible
Pumps & Related Piping            negligible
Weir Hopper (water line)             slight                 slow
Absorber Body (water line)           slight                 slow
Tube  Before                   moderately heavy           moderate
      After                       negligible
Mist Eliminator                      slight                 slow
Fan                                  slight              very slow
 Mote the change when wash water was added to control scale formation
 at this location.

-------
The deposits at the tube lip were controlled later in the test program
by continually washing the tube with a small spray of fresh water in
the vicinity of the tube lip.  Fresh water was used because it was
readily available at the pressure required.  It is assumed that salt
water would be entirely suitable for this purpose.  The design require-
ments to control scale at the tube lip necessitated a ring attached to
the tube approximately 6 inches above the bottom of the lip.  The ring
had a series of 1/16 inch diameter holes drilled in such a manner
that the tube was continually wetted down to the lip.  Once this
modification was installed and adjusted, stalactite-type growth into
the gas stream was eliminated.  However, during the limestone injec-
tion testing on Unit No. 10 the scale deposition was so heavy that
the spray ring proved ineffective.  Therefore, operation in this
manner for extended periods of time (16-24 hours) was not possible.

The deposits during testing on Unit No. 10 resulted primarily from
impaction of calcined limestone on the exterior surface of  the tube.
This resulted from the heavy particulate loading to the absorber and
an inadequately designed entrance for this heavy loading.   This deposi-
tion interfered with the operation because calcined limestone impacted
on the tube at the entrance, built up, and fell in large cakes into the
hopper.  Table 4-16 summarizes the scale and sludge deposition locations
in the pilot plant while scrubbing flue gas from TVA Shawnee Unit No. 10,

Settling Characteristics
Table 4-17 summarizes the results of one test to determine  the settling
characteristics of the slurry effluent.  This was done by measuring the
time required for the turbid portion of a slurry sample (interface
between slurry and clarified liquor) to settle in an undisturbed
graduated cylinder.  The exact location of the Interface was defined
by the graduations on the side of the cylinder.  The majority of
the settling appeared to occur during the first 10 minutes.  These
data were collected to aid in the design of a clarifler or  other
solid-waste handling equipment which might be required on a larger
sized unit.
                                  47

-------
      Table 4-16.  DEGREE OF DEPOSITION AT VARIOUS SYSTEM LOCATIONS

                        (TVA Shawnee Unit No. 10)
Location
Mix Tanks
Pumps &. Related Piping
Weir Hopper (water line)
Absorber Body (water line)
Tube3
Mist Eliminator
Fan
Degree
negligible
negligible
slight
slight
severe
slight
slight
Rate of
Formation
-
-
slow
slow
rapid
slow
very slow
a
 Scale at tube lip controlled by wash water.  Major depositing occurred
 by impaction of calcined limestone carried in gas stream from Unit No.
 10 during limestone injection test program.
         Table 4-17.  SETTLING RATE OF LIMESTONE SLURRY EFFLUENT
Interface Location
(ml in 500 ml graduated cylinder)
500
400
350
300
250
200
150
100
48
Time
(Minutes)
0
3.5
4.6
5.8
7.3
8.5
9.7
10.7
144.0
Note: sample taken from scrubber discharge during test P-32 which
      utilized 325 mesh Fredonia Valley limestone in a 3% slurry
      concentration.
                               48

-------
Pry Collector
No attempt was made at quantifying the operating parameters or the
performance of the dry collector regarding S0_ or NO  removal.  While
                                             £      X
an attempt was made to quantify the particulate removal efficiency of
the dry collector during the Shawnee No. 9 tests, procedural errors
and equipment malfunctions caused the tests to be voided.  Since suffi-
cient data had been obtained concerning particulate removal across the
entire pilot plant (dry collector and Dustraxtor scrubber) no further
attempt was made to measure the removal in the dry collector separately.

The dry collector was bypassed, when not in use, by means of built-in
dampers and gas ducts at both the Key West and Shawnee No. 10 test sites.
The fly ash collected from Shawnee No. 9 flue gas had the character-
istics expected of coal fly ash; however, the ash was not chemically
analyzed.

S02 ABSORPTION:  KEY WEST PROGRAM
In this section some of the more general results of the Key West pro-
gram are summarized.  An addendum to this report, available through
EPA1s Control Systems Laboratory, contains all the data collected.
Section 5 is a more complete analysis of the data collected during
the factorial design experiments.

Salt Water Tests
A total of forty three preliminary tests were conducted with no reac-
tant in the salt water feed using a 12 inch diameter tube size.  The
purpose of this series was to determine the operating characteristics
of the equipment, to establish base line data and to calibrate the
Instrument system.  Thirty of the tests provided useful data from
which S02 removal efficiencies could be calculated.  Two levels of
gas flow and pressure drop were used, these being the same as would
later be utilized in the experimental design tests.  The arithmetic
average SO. removal efficiency was 39.6% with sea water alone.  Figure
4-6 summarizes the effects of the two gas flow and pressure drop levels.
                                   49

-------
            50
          b 40
            30
            "tt
I2X)
4-6.  ABSORPTION EFFICIENCY AS A  FUNCTION  OF  GAS FLOW  AND PRESSURE
  SALT WATER
                                 50

-------
Three levels of salt water feed flow rate were tested during  the pre-
liminary shakedown period; these were  2, 5,  and  10 gpm.  Little change
in absorption efficiency was measured  between the two lower rates.
Figure  4-r7  summarizes  the effects  of the 2 and 10 gpm levels.

Coral Marl  Tests

A total of thirty six tests were conducted with coral as the reactant of
which thirty four provided useful data.  All tests were conducted with a
12 inch diameter tube.  Sixteen of these were the factorial experiment,
ten were Independent data for validating the derived prediction equation,
and eight were centerpolnt and additive tests.  Except for the last eight
tests, each of the five independent variable factors was tested at two
levels.  The average SO. removal efficiency using coral was 74.3%;

Figure  4-8  illustrates the absorption  (expressed as SO- absorption
efficiency) as a function of the Independent variables studied during the
experiments with the coral reactant.   The independent variables were: gas
flow rate,  stoichiometry, reactant particle  size, slurry concentration,
and scrubber pressure drop.  An increase in  stoichiometric ratio or a
decrease in gas flow rate at a fixed pressure drop, produced  a signifi-
cant increase in SO, absorption efficiency in the Dustraxtor.  It should
be noted 'that the pilot plant system was operated in a once-through con-
figuration.  These results should  not  be extrapolated to a closed-loop,
recycle system with different ionic strength scrubbing slurries.

Four centerpoint experiments were  conducted  in which four of  the
variables were at the midpoint of  the  high and low level.  The fifth
variable, particle size, was not centered because of the unavailability
of the  proper size reactant.  The  average SO. removal efficiency for
the centerpoint experiments was 71.1%. This compares closely with the
74.3% average SO- removal for all  Key  West coral tests and indicates
there is a  linear response to all  variable factors within the range
tested.  Extrapolation beyond this range, however, is not recommended.
                                    51

-------
                  90
                 40
                  30
                                                                  I2X>
FIG. 4-7. ABSORPTION EFFICIENCY  AS A FUNCTION OF  FEED FLOW RATE AND
PRESSURE  DROP: SALT WATER
                                        52

-------
     100
      90
      60
   e  70
   I
   o
   bj
   O  60
   b.
   U.
   U
   D.
   CC
   O
   m
   CD
      40
   o"
   CO
      30
      20
       10
                     6.5" A P
                  I
             LEGEND
100  Or  325   Refer*   To Particle Size (Mesh)
1000 Or  20OO  Refers   To  Gas  Flow  Rote (ACFM)
   ?) Refers  To  Slurry Concentration (Wt.% I
12" Diameter Tube
                                            I
 I
I	I
I
I
                  LO       2.0       3.0             1.0       2.0

                    CaC03   TO   S02    STOICHIOMETRIC   RATIO
                                             3.0
FIG. 4-a SUMMARY  OF  CORAL  SALT WATER EXPERIMENTAL  I'ESIGN   POINTS

                                       53

-------
Three tests were conducted In which additions were made to the slurry
to test the effect of Fed., as a catalyst, and hydroquinone, as an
inhibitor.  These compounds were recommended by EPA because they have
been reported to affect SO- absorption by influencing the oxidation of
sulfite to sulfate in the scrubber medium.  Results were inconclusive
concerning this claim since there did not appear to be a significant
change in the S02 absorption capability with the additives tested.
When the. FeCl3 was added directly to the mixture in the weir, an im-
mediate decrease in SO- outlet concentration was observed on the SO.
monitor.  This response, however, was only temporary and the outlet
concentration soon returned to its original level.

No noticeable results were detected in the case of the addition of
hydroquinone nor in the case of the addition of both catalyst and in-
hibitor.

Fredonia Valley Limestone Tests
A total of forty three tests were conducted with limestone as the re-
actant of which  forty one  provided useful data.  Sixteen of those were
the factorial experiment, nine were independent data for validating the
derived prediction equation, four were centerpoint tests, and twelve were
comparative tests of the 8 inch and 12 inch scrubber tubes.  The average
SO. removal efficiency using Fredonia Valley limestone was 73.72.

Figure 4-9 illustrates the absorption efficiency as a function of the
five independent variables studied during the experiments with Fredonia
Valley limestone.  An increase in stoichiometry, a decrease in gas flow
rate, or an increase in pressure drop each resulted in an increase in
SO. absorption during the limestone tests.  These same variables caused
similar effects on SO. absorption during the coral reactant tests.

Four centerpoint experiments were conducted in which four of the vari-
ables were at the midpoint of the high and low level.  As before, par-
ticle size was not centered.  Three of the tests resulted in high S02
                                 54

-------
     100
      90
      80
H
 •
u
y
      70
      60
   ^  50
   a:
   o
   CO
   m

    M
   O
   CO
      30
      20
       IO
                    65" A P
                  1.0
                                                 12.0" A P
                                      LEGEND

                         IOO  Or  325  Refers  To  Particle Size (Mesh)

                         IOOO Or  2OOO Refers  To   Cos  Flow  Rate (ACFM)

                       l)or(T) Refers To Slurry  Concentration (Wt %)

                         12"  Diameter  Tube
                         2.0
3.0
1.0
2.0
3.0
                    CaC03   TO   S02    STOICHIOMETRIC   RATIO
FIG. 4-9  SUMMARY OF FREDONIA VALLEY  LIMESTONE
SALT WATER EXPERIMENTAL  DESIGN  POINTS
                                       55

-------
removal efficiencies, ranging between 84 to 86%.  These values are con-
siderably higher than expected results.  It is suspected that a malfunction
of the S02 monitor occurred.  The fourth value was 70.6% which is much
closer to the average value from all of the tests in this series.

Twelve tests with limestone demonstrated the increased absorption
efficiency of the 12 inch scrubber tube.  This series consisted of seven
tests with the 12 inch tubes at various gas flow rates, all other variables
held constant, and five similar tests with the 8 inch tube.  The results
of the experiment are shown in Figure 4-10.

Secondary Reactant Tests
Four tests with each of three secondary reactants were conducted.  These
were designed for factorial experiments', but one test for one reactant
was voided due to carry-over into the sampling line rendering that test
series inadequate for analysis.  One final test was conducted with spent
reactant.  All tests were conducted with a 12 inch tube diameter.  The
average SO. removal efficiencies for the four reactants are shown in
Table 4-18:

          Table 4-18.   S02 REMOVAL EFFICIENCY FOR SECONDARY
                        REACTANT TESTS
                        (Key West No. 3)

	Reactant	•  Per Cent Removal (%)	
         Hydrated Lime                             93.0
         Dolomite                                  46.8
         Precipitated Calcium
            Carbonate                              78.3
         Spent Reactant                            69.9

Figure 4-11 summarizes the results of the hydrated lime and dolomite  for
which adequate data were available.
                                   56

-------
Ul
                  100
                3? 90
                o
                UJ
                o
Ul

o

a.
§70
w
CQ
                UJ
                a
                x
                2 60
                Q

                CC
                u.
                   50
                   40
                                                    SCFM (I2"DIA)
                                 L/6 (12 DIA)
                                                                            SCFM  (8  DIA)
                             TEST CONDmONSKEY WEST,
                               M3 STOtCHIOMETRY
                               12
                               1% SLURRY  CONCENTRATION
                               325 MESH FREDONIA VALLEY  LIMESTONE
L/6 (8 DIA)
                                           500                    1000                   1500
                                     GAS  FLOW RATE-scfm AND LIQUID TO GAS RATIO-  gal/Mscf
                                                                                               2000
        FIG. 4-IQSO2  ABSORPTION EFFICIENCY AS A  FUNCTION  OF GAS FLOW RATE  AND LIQUID/GAS RATIO

-------
       120
       100
        90
S
O
UJ
1
t
§
(O
Q
        70
       60
       50
       40
 LEGEND
                                Hydroted Lime  (Co(OH)2)
                                 Dolomite (Co^MgCOj)
                     1000
                                12.0
                                                        1000
                      ln. H20)
 1000 Or  2000 Refers To Gas Flow Rate (ACFM)
 Particle Size - No Control, Agricultural Grade
 Slurry Concentration 1% By Wt.
                                              STOICHIOMETflIC RATIO
                                            (Ca(OH)2 Or CaC03-MgC03)
                                            (To S02                    )
FIG. 4-11 SUMMARY OF SECONDARY REACTANTS SALT WATER EXPERIMENTAL DESIGN POINTS
                                       58

-------
S02.ABSORPTION:  PADUCAH PROGRAM
In this section some of the more general results of the Paducah program
are summarized.  An addendum to this report, available through EPA's
Control Systems Laboratory, contains all the data collected.  Section 5
is a more complete analysis of the data collected during the factorial
design experiments.
Fredonia Valley Limestone Tests
A total of thirty five tests were conducted with limestone reactant in a
fresh water medium at the TVA Shawnee No. 9 unit.  Thirty three provided
useful data.  Of these, sixteen were the factorial experiment, eight
were centerpolnt tests, five were additive tests, and four were tests with
the dry collector bypassed.  No tests were run for independent validating
purposes.
The average SO. removal efficiency for the sixteen factorial experiment
tests was 57.1%.  This compares badly with the 73.8% average removal     i.
efficiency obtained in the sixteen factorial experimental tests conducted
at Key West using Fredonia Valley limestone with a salt water medium.
The variation in results is not entirely unexpected since there were many
differences in the two test series, nanely: tube size, stoichlometric
ratio, slurry concentration, slurry medium and gas flow rate.
Figure 4-12 illustrates the absorption  (expressed as SO- absorption
efficiency) as a function of some of the more important parameters studied
during the experiments on Unit No. 9.  An increase in stoichlometric ratio
or decrease in gas flow rate for each pressure drop produced a statistically
significant increase in SO. absorption efficiency in the Dustraxtor.  Again,
it should be noted that the pilot plant was operated in a one-through con-
figuration.  The results should not be extrapolated to a closed-loop, re-
cycle system with different ionic strength scrubbing slurries.

Five additive tests were conducted.  In three tests 200 ppm of Fed. were
added at the weir to study the catalytic effects (see Coral Marl Tests
paragraph above).  There were none:  the average SO- removal efficiency
was 67.9% before the addition of the catalyst and 65.2% after.  Similarly.
                                     59

-------
a*
 •
>-

z
bJ
u
u.
b.
bJ
g


Q.


g
OT
ID
Q
x
o
o
         oc
         CO
               100
               90
               80
               70
               60
               *>
               40
               30
               20
               to
                               6.0"
                                                  12.0" A P
                                  LEGEND

                        100 Or 325  Refers  To  Porticle  Size (Mesh)

                        500 Or 800  Refers  To  Gas

                         (Flow Rate (scfm)

                        8" Diameter  Tube

                        Slurry Concentrdtion 3%  By Wt.


                                      I

                                      I
                             1.0
                               ZO
1.0
2.0
                       CaC03  TO    S02    STOICHIOMETRIC    RATIO
FIG. 4-12 SUMMARY  OF  FREDONIA  VALLEY  LIMESTONE

RIVER  WATER  EXPERIMENTAL  DESIGN  POINTS
                                         60

-------
hydroquinone was tested in two cases as an inhibitor.  Before addition
    *
the mean removal efficiency was 69.4%.  After addition it was 66.4%.
Based on knowledge of the SO- monitor's accuracy, drift characteristics,
and response time, it was decided that effects on removal efficiencies
amounting to less than 5% could not be separated from Instrument and ex-
perimental error.

A total of twelve centerpoint tests were conducted.  Four tests were per-
formed at centerpoint conditions with the dry collector bypassed.  This
was done to evaluate the effects of coal fly ash on SO. absorption ef-
ficiencies.  The average SO. removal efficiency for the four tests was
65.7%.  The remaining eight centerpoint tests were conducted with the
dry collector in line and resulted in an average SO. removal of 63.4%.
The difference in average S0« removal for the two centerpoint test series
is within instrument and experimental error.  Therefore, no conclusion
can be made concerning the effect of fly ash on S02 removal efficiencies.

Aragonite Tests
Eight tests were conducted with aragonite using an 8 inch diameter tube.
Four of these were the factorial experiment, two provided independent data
at centerpoint, and two were dry collector bypass tests.  All were at a
3% slurry concentration.  The average S02 removal efficiency was 76.2%.
The two tests with the dry collector bypassed, at centerpoint, showed an
average of 67.7%.

Simulated Key West Salt Water Tests
Three tests were conducted with Fredonla Valley Limestone in simulated
salt water using an 8 inch tube diameter.  /
ciency of 76.5% was obtained at centerpoint
salt water using an 8 inch tube diameter.  An average SO. removal effi-
Injection Tests
In this  test series  conducted on TVA  Shawnee Unit No.  10, the mobile
pilot plant received flue gas laden with calcined limestone or aragonite,
                                  61

-------
as supplied from the boiler:, injection system.  A total of seventeen in-
jection tests were conducted; four with aragonite injection, thirteen with
limestone injection.  During all injection tests only river water was
used as the scrubbing medium.  The recorded stoichiometry is the injection
stoichiometry of dry reactant to SO- as determined by TVA equipment op-
erators.  The tests conducted with aragonite injection resulted in an
average S02 efficiency of 85.9%.  The limestone tests gave an average SO
efficiency of 77.1%.

PARTICULATE COLLECTION IN THE SCRUBBER SYSTEM
Since the city of Key West was using fuel oil exclusively, no particulate
testing was conducted at this site, and no quantitative results were
available.  From deposits observed above the scrubber mist eliminator
section and in the fan and fan ductwork, it was concluded that particu-
lates were generated during the scrubbing processes.
A number of tests utilizing the apparatus described in Section 3
Particulate Sampling paragraph, were conducted to quantify the particulate
removal at the Paducah test site (Shawnee No.9).  The results of the
successful tests are presented in Table 4-19.  The average result was dry
collector and Dustraxtor.  No attempt was made to determine the size
range or composition of the particulates.

NO  ABSORPTION
Tables 4-20 through 4-23 present NO  absorption data.  Tables 4-21 and
                                   A
4-23 present individual test results.  Tables 4-20 and 4-22 show the
average inlet and outlet measurements.  Although these averages show a
reduction in NO  across the scrubber, the conclusion that significant
absorption occurred is unwarranted because of the large standard deviations
of the measured concentrations and of the experimental error associated
with the analytical techniques.

SULFUR BALANCE
Table 4-24 shows sulfur balances from selected factorial design tests.
These were selected as representative of both the Key West (coral and
limestone) and the Paducah (limestone) tests.  The consistently low exit

                                 62

-------
        Table 4-19.  PILOT PLANT PARTICULATE REMOVAL EFFICIENCY

                     (PADUCAITUNIT NO. 9)
Test
Number
P-2
P-5
P-7
P-12
P-17
PA-5
PA-6
AP
(in. H.O)
L
6.0
12.0
6.0
12.0
9.0
9.0
9.0
Gas
Flow
(gcfnO
422
620
617
417
520
520
520
Particulate Loading v Particulate
(gr/scf) .Removal
Influent Effluent Efficiency^
5.83
4.77
5.21
5.16
5.62
4.92
5.76
Average
0.13
0.08
0.11
0.05
0.07
0.08
0.09

97.77
98.32
97.89
99.03
98.75
98.37
98.44
	 98.37%

            Table 4-20.  AVERAGE OF NO  MEASUREMENTS
                                      A

                         (KEY WEST UNIT NO. 3)
                                Average NO
                                    (ppm)
                 Standard Deviation
                	(ppm)	
Key West

   Pilot Plant Entrance

   Pilot Plant Exit
440.9

390.6
97.0
85.4
                                    63

-------
Table 4-21.  TEST RESULTS OF NO  DURING THE KEY WEST TEST SERIES
                               A
Test No.
S-38
C-19
C-26
C-27
C-28
C-29
C-30
P-3
F-4
F-5
F-16
F-17
F-20
F-21
F-22
F-23
F-24
F-25
F-26
F-27
F-28
HL-1
HL-2
D-3
D-4
PC-1
PC- 2

Influent Effluent
Concentration Concentration
(ppm) fppm}
357
425
387
195
392
645
475
392
475
417
547
388
195
310
416 -
407
536
512
501
187
462
462
425
510
175
416
489
Average NOx.
310
401
362
404
356
596
452
307
504
398
415
300
228
247
374
408
402
400
472
325
551
475
376
472
365
392
471
Absorption (ppm) .
NO
j£
Absorption
(ppm)
47
24
25
-
36
49
23
85
-
19
132
88
33
63
42
-
134
112
29
-
-
-
49
38
-
24
18
. . 54
Removal
(T)
13.2
5.6
6.5
-
9.2
7.6
4.8
23.2
-
4.6
24.1
22.7
-
20.3
10.1
-
25.0
21.9
5.8
-
-
-
11.5
7.5
-
5.8
3.7

                                 64

-------
Table 4-22.  AVERAGE NO  MEASUREMENTS




             (PADUCAH TEST SERIES)

Unit No. 9
Pilot Plant Entrance
Pilot Plant Exit
Unit No. 10
Pilot Plant Entrance
Pilot Plant Exit
Average -NQX '
(ppm)

710.8
672.0

76A.O
723.5
Standard Deviation
(ppm)

176.6
173.4 .

101.8
108.2
Table 4-23.  TEST RESULTS OF N0x DURING THE PADUCAH TEST SERIES
Test No.
P-5
P-7
P-8
P-16
PA-5
PA-6
IP-6
IP- 7
IP- 8
••
Influent Effluent
Concentration Concentration
( ppm) (ppm)
437
178
782
645
892
798
836
700
692
Average NO
402
811
741
608
846
763
800
845
647
Absorption (ppm) . . .
NO
Absorption
(ppm)
35

41
37
46
35
36
-
45
. . 39
Removal
(%)
4.2
-
5.2
5.7
5.2
4.4
4.3
.
6.5

                           65

-------
       Table  4-24.   SULFUR BALANCES  FOR SELECTED TESTS AT

                    KEY WEST AND PADUCAH
Test
Number
C-20
C-21
C-22
C-23
C-34A
F-8
F-18
F-20
F-26
F-42
P-l
P-2
P-3
P-5
P-8
P-13
P-20
P-21
P-22
P-23
. . Liquid Stream
Sulfur Flow
Feed
0.0432
0.0140
0.0288
0.0173
0.0200
0.0080
0.0320
0.0064
010033
0.0189
0.0002
0.0004
0.0000
0.0001
0.0005
0.0002
0.0001
0.0001
0.0003
0.0003
(Ibs/min)
Disch
0.0600
0.0200
0.0445
0.0245
0.0242
0.0143
0.0489
0.0104
0.0053
0.0222
0.0055
0.0085
0.0065
0.0067
0.0084
0.0066
0.0074
0.0104
0.0198
0.0082
Gas St
Sulfur Flew
Influent
0.0486
0.0486
0.0989
0.0988
0.0932
0.0267
0.1027
0.1004
0.0513
0.0912
0.0902
0.0826
0.0822
0.1030
0.0896
0.1146
0.0084
0.0861
0.0877
0.0812
:ream
(Ibs/min)
Effluent
0.0071
0.0131
0.0431
0.0249
0.0247
0.0072
0.0532
0.0508
0.0128
0.0268
0.0432
0.0379
0.0245
0.0395
0.0558
0.0729
0.0345
0.0323
0.0303
0.0303
Total
1 Liquid & Gas Stream
Sulfur Flow ""• (Ibs/min)
Influent
0.0918
0.0626
0.1277
0.1161
0.1132
0.0347
0.1347
0.1068
0.0546
0.1101
0.0904
0.0830
0.0822
0.1031
0.0901
0.1148
0.0885
0.0861
0.0878
0.0813
Effluent
0.0671
0.0331
0.0876
0.0494
0.0489
0.0215
0.1021
0.0612
0.0181
0.4900
0.0487
0.4640
0.0310
0.0462
0.0642
0.0789
0.0419
0.0427
0.0413
0.0342
Sulfur
Accounted
For (X)
73.1
52.9
68.6
42.5
43.2
62.0
75.8
57.3
33.2
44.5
53.9
55.9
37.7
44.8
71.3
68.7
47.3
49.6
47.0
42.1
Key West:
     Average Sulfur Accounted for 	  55.3%
     Maximum Sulfur Accounted for 	  75.8%
Paducah:

     Average Sulfur Accounted for 	  51.8%
     Maximum Sulfur Accounted for 	  71.3%

Total Average Sulfur Accounted for 	  53.57%
                            66

-------
 sulfur values might be attributed to the analytical technique used on the
 liquid discharge stream.  Slurry samples were sometimes allowed to stand
 up to 24 hours before the total sulfur analysis (as SO,  )  was performed.

This time period allowed sulfur to crystallize out of solution as the
dihydrate, CaSO,.2H.O.  Although slurry samples were thoroughly mixed
prior to aliquot removal, the rapid settling that occurred made rep-
resentative sampling difficult.  In addition, it is suspected that the
CaSO,.2H.O may not have been completely digested in the time allotted.
Either of these two situations could account for the low sulfur results
shown in Table 4-24.

-------
                             SECTION 5
                           DATA ANALYSIS

INTRODUCTION AND SUMMARY
     Five factorial experiments were conducted with the reactants to
determine the effects of the several independent variables.  The five
factors investigated were: gas flow rate (GF) ; pressure drop (AP);
stoichiometric ratio (SR) ; particle size (PS) ; and slurry concentration
(SC).  These experiments are summarized in Table 5-1.
               TABLE 5-1.  SUMMARY OF FACTORIAL EXPERIMENTS
                     Key West
Key West
Paducah
Type of
Design
Reactants
Studied
Factors
Studied and
Their Levels
Factors found
Statistically
Significant
at 95% level
Half Replicate of
a 2^ • 16 Runs
*-
Coral (C) and
Fredonia Valley Lime-
Stone (FVL)
GF(scfm) 1450 775
AP(in.H-O) 12.5 6.5
SR(mole7mole) 1:3 1:1
PS (mesh) 325 100
SC(%) 5 1
FG.tf.SR )
GFSC, PSC)
GF,/!P,SR for FVL
Half Replicate of
a 2 -4 runs
Dolomite (D) and
Hydrated Lime (HL)
GF(acfm) 2000 1000
AP 12.0 6.5
SR 1:3 1:1
PS Uncontrolled
SC 11
GF.AP.SR for D
£P,SR for HL
Full Replicate
of a 2 "16 runs
Fredonia Valley
Limestone (FVL)
GF(scfm)600 400
Ap 12.0 6.0
SR 1:2 1:1
PS 325 200
SC 33
GF,AP,SR for FVL
     A factorial experiment is designed to test the significance of a number
of different operating factors simultaneiously.  The proper application of
the method permits a rapid and Inexpensive method for determining the effects
of. each of the factors over the specified range.  The interpretation,of the
results is difficult, however, and sound engineering judgement must be used
in assigning statistical significance to the conclusions.  This judgement
                                  68

-------
        Table 5-2.  KEY WEST TEST PROGRAM EXPERIMENTAL DESIGN
REACTANTS:
             Coral (C)
             Fredonia Valley Limestone (FVL)
Run No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Factors
GF AP SR PS SC
+
+
+
+ + - + -
+
+ - + +
+ + +
+ + +
+
+ . - + +
+ - + +
+ + - - +
+ + +
+ - + - +
+ + - +
+ + + + +
Responses
C
64.6
56.5
73.1
70.6
85.3
67.6
90.1
88.8
77.5
58.2
80.5
55.6
83.3
74.7
93.3
75.0
(% SO- Removal)
2FVL
71.1
48.3
76.6
76.4
78.9
74.4
90.4
82.6
73.7
50.0
75.0
49.4
88.5
68.7
90.5
86.0
                          Total Response.  .
                          Average Response,
1195.2
  74.7
1180.5
  73.8
                                   69

-------
may be based on knowledge of and experience with the physical and chemical
processes involved and the methods in which the factors act; comparison
of results with other similar experiments; or conducting additional tests
of factors whose significance is questionable.

In treatment of the data, classical analysis of the variance techniques
are used, details of which may be found in any elementary statistics text.
Important assumptions Involved in the use of these techniques include:
      o  Independence of the tests
      o  Normal distribution of the experimental errors
      o  Linearity of response (percent SO. removal) to each of the factors
         in the range tested.

In general the analysis showed that in these experiments the three factors
GF, AF and SR were statistically significant at the 95% confidence level.
This means that one can be 95% confident that SO. removal efficiency is
different for the two levels tested for each factor.  No conclusion can
be reached concerning the effect of levels outside the range tested nor
can it be concluded that one factor is more effective than another.

KEY WEST TESTS

The main test series at Key West consisted of half-replicate 2  factorial
experiments for each of two reactants, coral and Fredonia Valley limestone
(FVL).  The factors and levels are listed below:
               Factor                            Design Levels
                                              High (+)         Low (-)
       GF - Gas flow (scfm)                     1450             775
       AP - Pressure drop (in.H.O)              12.0             6.5
       SR - Stoichiometric ratio                 1:3             1:1
       PS - Particle size (mesh)                 325             100
       SC - Slurry concentration (%)               5               1
Table 5-2 shows the experimental design and the response measured as per-
cent SO. removal for the main Key West test program.
                                 70

-------
A half replicate 2  factorial experiment permits determination of all
main effects and two-factor interactions.  Each main effect is aliased
with a four-factor interaction and each two-factor interaction is aliased
with a three-factor interaction.  If all third order and above interactions
are considered small  (to be shown in the analysis of the Paducah tests,
following), the design permits independent estimates of the main effects
and two-factor interactions.  However, if all two factor interactions are
considered Important, there remain no degrees of freedom to estimate the
error.  Thus, the investigator must estimate the error mean square by the
engineering judgement mentioned above or by assuming, a priori, that some
specific two factor interactions are negligible.

The means of the response  from the two reactants, 74.7% for coral and
73.8% for Fredonia Valley  limestone, were tested for significant difference.
There was found to be no significant difference at the 99% confidence level.
Furthermore, the correlation coefficient between the sixteen pairs of data
points was calculated to be 0.89.  These facts might indicate that the
physical and chemical processes involved were similar for each of the two
reactants and that any main effect or interaction that is assigned to ex-
perimental error in the one case should be so assigned in the other case.
In view of the lack of substantiating information concerning the estimation
of the error, an a priori  judgement was made that differences in average
effect of less than 2% are within the range of experimental error.

Tables 5-3 and 5-4 show the analysis of the variance for the Key West coral
and limestone tests respectively.  If the calculated effect is below this
value in either test, the  values are used for the estimate of the error.
With this scheme only three main effects and two Interactions would remain.
An exception to this scheme is made in the main effect of particle size.
It has been shown that coral undergoes certain physical changes in the
presence of sea water in which the particles are retained in a colloidal
suspension, regardless of  the original particle size distribution.  For this
reason, particle size main effect is assigned to the error sum in the case
of coral but not in the case of limestone.
                                     71

-------
    Table 5-3.  CORAL REACTANT	ANALYSIS OF THE VARIANCE
                   KEY WEST TEST PROGRAM

Total
GF
AP
SR
PS
SC
GFAP
GFSR
GFPS
GFSC
APSR
APPS
APSC"
SRPS
SRSC
PSSC

Total Effects
1195.2
-101.2
58.8
122.0
-14.4
2.0
7.2
8.4
6.0
-42.0
12.8
25.2
-38.4
-36.8
-12.0
7.2

Average Effects
74.7
-12.7
7.4
15.3
-1.8
0.3
0.9
1.1
0.8
-5.3
1.6
3.2
-4.8
•^4.6
-1.5
0.9
TOTAL 	
Sum of
Squares

640.1
216.1
930.2
13.0
0.3
3.2
4.4
2.3
110.3
10.2
39.7
92.2
84.6
9.0
3.2
2158.8
Mean
Square

640.1
216.1
930.2
Error
Error
Error
Error
Error
110.3
Error
Error
92.2
Error
Error
Error

F Ratio

37.7
12.7
54.7





6.5


5.5




Error sum of squares - 169.9
Error mean square    -  17.0
FQ>95 (1,10) - 4.96
FQ 99 (1,10) = 10.04
                              72

-------
     Table 5-4.   FVL  REACTANT	ANALYSIS OF THE VARIANCE
                   KEY WEST TEST PROGRAM

Total
GF
AP
SR
PS
SC
GFAP
GFSR
GFPS
GFSC
APSR
APPS
APSC
SRPS
SRSC
PSSC

Total Effects
1180.5
108.9
73.3
139 . 5
43.1
-16.9
32.7
35.7
32:5
-38.3
4.7
14.3
-33.3
-5.9
31.7
-8.7

Average Effects
73.8
-13.6
9.2
17.4
5.4
-2.1
4.1
4.5
4.1
-4.8
0.6
1.8
-4.2
-0.7
4.0
-1.1
TOTAL ....

Sum of
Squares

741.2
335.8
1216.3
116.1
17.8
66.8
79.7
66.0
91.7
1.4
12.8
69.3
2.2
62.8
4.7
2884.6

Mean
Square

741.2
335.8
1216.3
116.1
Error
Error
Error
Error
91.7
Error
Error
69.3
Error
Error
Error

F Ratio

21.2
9.6
34.8
3.3




2.6


2.0




Error sum of squares - 314.2
Error mean square    -  34.9
 0.95
 0.99
              10'60
                               73

-------
Some further justification for the 2% cut-off value is provided by tests
conducted with limestone in Paducah.  In this test series, where aliasing
does not occur, all interaction effects are small.  A comparison of the
average effects for the two Key West and the one Paducah test series shows
considerable stability from one test to another.

As can be noted in the case of coral, three main effects and two inter-
actions are statistically significant at a confidence level of 95% pr
greater.  In the case of limestone, only the three main effects are
significant.  This is determined by comparing the F ratio, defined as
                         p    m  Factor Mean Square
                                 Error Mean Square
with the significance values in the tables.  For instance in Table 5-3
the F ratio for GF is
                              .  640.1    m
                         F       17.0         37'7
which is greater than Fn QQ • 10.04.  Therefore, gas flow rate has a
                       U . yy
statistically significant effect on percent SO. removal by coral.

Prediction equations have been developed for relating response to the
significant factors:

  Y- - 74.7 + I-12.7GF + 7.4AP 4- 15.3SR - 5.3(GFoSC) - 4.8( PoSC)]  (5-1)
   c
  YFVL - 73.8 + [-13.6GF + 9.2AP + 17.4SR ]                         (5-2)

where Y_ and Y    are percent SO. removal for coral reactant and limestone
       L      r VL>               f.
reactant, respectively.

     „,,    gas flow (scfm) - 1112
     GF "           675
     AP    pressure drop (H^O) - 9.25
     or -           5.5
     CD    stoichlometric ratio - 2
     SR .            -
     cr    slurry concentration - 3
     J>L -            ^

These equations are given in this way to show their deviation.  As can be
                                   74

-------
Seen, the basic equations Include the average total response, Table 5-2,
and the average effects of each of the significant factors and inter-
actions, Tables 5-3 and 5-4,   The constants in the definitions of the
variables are the average of the test design levels and the range of test
design levels.

 There were  nineteen additional tests  in the coral and limestone  series
 at  Key  West which were not  part of  the  factorial  experiment.   These
 tests provided  independent  data to  examine the  accuracy of  the prediction
 equations.   Figure  5-1 is a plot of predicted response versus  actual  re-
 sponse  for  these' nineteen data points.

Figure 5-2  illustrates  the  response expected from the  two significant
interactions in the coral reactant series.  It  appears  that high slurry
concentration increases the  change in response  due to  changes  in gas flow
rate and decreases the  change  in response  due to  pressure drop changes.

Two additional reactants, dolomite and hydrated lime were tested in
half-replicate 2  experimental design programs  at Key  West.  Table 5-5
shows the design  criteria and  the results  of the  tests.  In  this experi-
mental  design each main factorial effect is allased with a two-factor
interaction.  Further,  the  second order Interaction cannot be  estimated
nor can the error mean  square.  A partial  analysis of  the variance is
given in Table 5-6.
                                      75

-------
      90
       80
       70
      60
       50
      40
              O Coral

              & Limestone
        30        40       50       60       70        80

                                ACTUAL  RESPONSE-(%)
90
FIG. 5-1. PREDICTED RESPONSE  VS. ACTUAL RESPONSE  FOR INDEPENDENT DATA FROM THE
 MULTIPLE  LINEAR REGRESSION FORMULA
                                     76

-------
    u
      70
      60
                                            80
                                          I
70
             L                H
          SLURRY  CONCENTRATION
60
                                                                    i_
       L               H
     SLURRY CONCENTRATION
RG. 5-2. SUMMARY OF  SIGNIFICANT INTERACTIONS IN CORAL EXPERIMENTS
                                     77

-------
            Table 5-5.  SECONDARY REACTANT TEST PROGRAM
                        AT KEY WEST TEST SITE
SECONDARY REACTANTS:
                Dolomite (D)
                Hydrated Lime (HL)
TEST PROGRAM
Response (% SO,, Removal)
Test No.
1
2
3
4
AP
6.5"
12.0"
6.5"
12.0"
Stoichiometry
1:1
1:1
1:3
1:3
Gas Flow Rate
2000 acfm
1000 acfm
1000 acfm
2000 acfm
D
32.6
51.8
51.7
51.2
HL '
86.4
94.9
93.4
97.5
     Slurry concentration - 1% by wt.
     Particle size - no control due to use of commercially
                     available materials.
       Table 5-6.  SECONDARY REACTANTS— ANALYSIS OF THE VARIANCE
                   KEY WEST TEST PROGRAM
Effect
Total
Gas Flow
Pressure

(GF)
Drop (AP)
Stoichiometric Ratio
(SR)
Total

Dolomite
Total
Effect
187.3
-19.7
18.7
18.5

Mean
Effect
46.8
-9.8
9.4
9.2

Mean
Square

97.0
87.4
85.6
270.0
Hydrated Lime
Total
Effect
372.2
-4.4
12.6
9>6

Mean
Effect
93.0
-2.2
6.3
4.8

Mean
Square

4.8
39.7
23.0
67.5
                                  78

-------
Prediction equations for the secondary reactants are:
      Y   -  46.8 +  [-9.8GF + 9.4AP + 9.2SR]                      (5-3)
  and  D
          -  93.0 +  [-2.2GF + 6.3AP + 4.8SR]                      (5-4)
where Y_ and Y_ are percent SCL removal for dolomite reactant and hydrated
       u      tlL               /
lime reactant respectively.
      „_     gas flow  (acfm) - 1500
      GF  "          1000
      AT>     pressure  drop  (in.H,.0) - 9.25
      AP  .            -_-*-,    y-*
      co     stoichiometric ratio - 2
      SR  -            -

An analysis of the variance in response to the two different reactants
showed a significant difference of the means at  the 99.51  confidence
level.
PADUCAH TEST PROGRAM
                  4
A full replicate 2  factorial test program was  conducted on the No. 9
unit at Paducah, using FVL as a  reactant.  The  factors  and levels are
listed below.  Slurry concentration was  constant at  3 in this  test series.

               Factor                        Design  Level
                                       High  (+)          Low  (-)
         GF - Gas Flow (scfm)             600              400
         AP - Pressure drop  (in.H.O)     12.0              6.0
         SR - Stoichiometric ratio        1:2              1:1
         PS - Particle size  (mesh)        325              200
Table 5-7 shows the experimental design  and  the response measured as per-
cent SO- removal.  The full  replicate design permits determination of  all
main effects and Interactions.   With the criteria established  earlier,  i.e.
that effect values less  than 2%  in any test  are considered error, all  two-
factor Interactions would be used to estimate the error mean square.
Further, if all two-factor interactions  are  so  considered, certainly higher
                                   79

-------
         Table 5-7.  PADUCAH TEST PROGRAM EXPERIMENTAL DESIGN
REACTANT:  Fredonia Valley Limestone
SLURRY CONCENTRATION:  3% wt.
Run No.
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
Factors
GF AP SR PS
+
+ +
+ _ +
+ + +
+ + - +
+ + + +
+ - - +
+ - + +
+
_
+ +
+
+
+ - +
+ +
+ + +
Response (% SO. Removal)
Z
52.0
54.0
70.2
83.7
61.7
67.3
34.7
37.8
53.2
46.3
73.9
70.0
36.3
42.1
62.6
67.7
                                      Total Response   913.5
                                      Average Response  57.1
                                  RO

-------
order interactions should be as well.  Table 5-8 shows the analysis of the
variance for the Paducah limestone test program.  It is seen that the three
main effects, gas flow, pressure drop, and stoichlometric ratio, are signi-
ficant at the 95% confidence level.  A prediction equation for this test
series is given below:

      Yp  - 57.1 + [-11.6GF + 25.1AP + 5.7SR]
where Y  is percent SO. removal.
      __    gas flow rate (scfm) - 500
      GF  "           200
      AP  - pressure drop (in.H?0) - 9
                        6      *"^
      SR  - stoichtometric ratio - 1.5
No independent data are available for testing the accuracy of the prediction
equation.

LIQUID/GAS RATIO
Some evidence has been seen that only three  factors are significant in
determining the S0« removal efficiency.  Two of these factors, gas flow
and pressure drop (for a given size scrubber tube), uniquely determine the
liquid/gas ratio.  Figure 5-3 is a plot of absorption efficiency versus
liquid/gas ratio for five of the reactants.  The efficiencies for high and
low levels of stoichiometrlc ratios are shown as envelopes around the
mean response for each reactant.  As can be noted, the efficiency Increases
rapidly with liquid/gas ratio up to about 140 gal/Mcf.  After that point,
little additional Increase is achieved.  Furthermore, it would appear that
increased efficiency is gained by increased  reactivity of the slurry above
that from Increased stoichlometric ratio.

The three Paducah salt water simulation test points lie on the Key West
Limestone curve indicating that the difference in this curve and the
Paducah curve results from the different media  (salt water and river water)
rather than the different size tubes.
                                   81

-------
    Table 5-8.   FVL REACTANT - ANALYSIS OF THE VARIANCE
                   PADUCAH TEST PROGRAM
Total Effects
Total
GF
AP
SR
PS
GFAP
GFSR
GFPS
APSR
APPS
SRPS
GFAPSR
GFAPPS
GFSRPS
APSRPS
GFAPSRPS
913.5
-93.1
200.7
45.9
9.3
16.1
-6.7
-23.7
10.3
9.1
2.5
-6.7
1.1
-6.9
17.7
-11.3
Sum of
Average Effects Square
57.1
-11.6
25.1
5.7
1.2
2.0
-0.8
-3.0
1.3
1.0
0.3
-0.8
0.1
0.9
2.2
-1.4
TOTAL 	


541.7
2517.5
131.7
5.4
16.2
2.8
35.1
6.6
4.1
0.4
2.8
0.1
3.0
19.6
8.0
3295.0

Mean
Square

541.7
2517.5
131.7
Error
Error
Error
Error
Error
Error
Error
Error
Error
Error
Error
Error
F Ratio

62.3
289.4
15.1












Error sum of squares - 104.1
Error mean square    -   8.7
FQ>95 (1,12) = 4.75
F     (1,12) = 9.33
                            82

-------
oo
u>
                                                        100       120       140        160

                                                          LIQUID/GAS  RATIO-gol/Mcf
200
       FIG.5-3. ABSORPTION EFFICIENCY AS  A FUNCTION OF UQUID/GAS RATIO

-------
Further, at a given liquid/gas ratio, residence time in the 8 inch tube
is longer than in the 12 inch tube (Figures 4-1 and 4-2).
                                 84

-------
                             SECTION 6
                            CONCLUSIONS

The following conclusions are based upon field observations and analysis
of the data.

PILOT PLANT EQUIPMENT
(1)  Locating the scrubber fan at the outlet of the system caused solids
     build up on the blades which ultimately resulted in fan failure due
     to imbalance.
(2)  Corrosion and pitting were experienced on the interior walls of the
     316-SS scrubber body and the non-PVC piping and valving.  This
     corrosion and pitting appeared to be more severe, both in degree
     and rate, when using silt water.  However, this corrosion and
     pitting did not result in failure of the scrubber body during the
     ten months of pilot plant operation.
(3)  Condensation was experienced in the non-insulated dry collector
     and ductwork.
(4)  The long hoses used to transport scrubber waste discharge frequently
     clogged due to low flow rates.
(5)  The pilot plant spray nozzles, intended for precontacting the
     inlet flue gas, were rendered useless due to clogging and corrosion
     before an evaluation could be made.

SAMPLING AND ANALYSIS
 (1)  An "ice-trap"  condenser  immediately following  the  S0_  sampling
     probe  operated in a satisfactory manner with the Dynasciences
     SS-330 monitor.
(2)  The Dynasciences  SO- monitor performed satisfactorily.
(3)  The use of one SO. monitor  to measure both  inlet and  outlet flue  gas
     proved cumbersome and reduced the accuracy  of  the  calculated
     scrubber efficiency.
                                85

-------
 (4)  The phenoldisulfonic acid analysis for determining concentrations
      of NO  proved limited.
           x
 (5)  The Annubar velocity measuring device performed very well, pro-
      ducing accurate readings in a minimal length of time.
 (6)  The use of a non-heated probe and filter in the particulate
      sampling train caused condensation and ultimate clogging of
      the filter.

 S02 ABSORPTION EFFICIENCY
 (1)  Absorption efficiency increased significantly with increased pressure
      drop across the scrubber.
 (2)  Absorption efficiency increased significantly with decreased gas
      flow rate through the tube.
 (3)  Absorption efficiency increased significantly with increased stoi-
      chiometric ratio.
 (4)  There was no significant change in absorption efficiency resulting
      from a change in reactant particle size.
 (5)  There was no significant change in absorption efficiency resulting
      from a change in slurry concentration.
 (6)  The analysis of the factorial experiments showed no significant
      interaction between or among the five factors listed above.
 (7)  Absorption efficiency is an increasing function of liquid to
      gas ratio.
 (8)  For a given tube size and gas flow rate, liquid to gas ratio
      increased with  increasing pressure drop.
 (9)  For a given tube size and pressure drop, liquid to gas ratio was
      a function of gas flow rate with maximum value near 600 scfm.
(10)  For a given pressure drop and gas flow rate, liquid to gas ratio
      increased with tube size.
                                 86

-------
(11)  Absorption efficiency was different with different reactants with
      decreasing efficiency as follows:
          Hydrated Lime
          Precipitated Calcium Carbonate
          Limestone
          Coral
          Spent Coral
          Dolomite
(12)  In the tests conducted, residence time within the tube had no
      apparent effect on absorption efficiency.
(13)  Addition of the catalyst (FeCl3> and/or the inhibitor (hydroquinone)  to
      the slurry produced no significant effect on absorption efficiency.

NO  AND PARTICULATE REMOVAL
  X
 (1)  NO  removal in the Dustraxtor scrubber utilizing limestone type
      reactants was negligible.
 (2)  Particulate removal by the total pilot plant system (dry collector
      and Dustraxtor) was excellent though evidence of re-entrainment
      of slurry solids was noted.

SCALE FORMATION
 (1)  Except during the injection tests, all scale formation that threatened
      the operation of the pilot plant was eliminated by the Installation of
      an annular fresh water spray ring around the lower few inches of the
      Dustraxtor tube.

FULL SCALE SYSTEM
As Indicated by results of the pilot test program on a single-tube
open-loop Dustraxtor, scale-up to a full size unit (capable of handling
50,000 to 60,000 scfm) should not present any problems other than those
already encountered and discussed on the pilot scale level.  However, to
handle larger quantities of boiler stack gases, the designer should
consider the potential unequal gas distribution in the scrubber as the
                                  87

-------
number of tubes is increased.  One solution might be to investigate the
possibility of using fewer tubes of larger diameter (e.g., 16 or 18 inch)
to increase scrubber capacity.
                                  88

-------
                              APPENDIX A
                     DETAILS OF ANALYSIS TECHNIQUES

S02 APPARATUS AND PROCEDURE
Sampling of flue gas for SO. determination was accomplished by passing
the sample gas through a Dynasciences Model SS-330 S0~ monitor, the
output of which was continuously recorded on a strip chart recorder.
Once the instrumentation was "debugged", operation of the equipment
was simple and required little operator time.  Based upon wet chemical
analyses, it was found to be very accurate.

The complete sampling train is shown in Figure A-l.  The sampling
probe used at all three sample locations consisted of stainless steel
tubing 6 inches long x 1/4 inch O.D.  A 500 ml flask was placed after
the sampling probe to trap any moisture which may have condensed in
the sampling tube.  The dry trap was completely immersed in an ice
bath to cool the sample gas to approximately 32°F.  A single acting
diaphragm pump drew the gas sample through approximately 15 feet of
Tygon tubing.  A tee located in the sample line was fitted with a 1/4
inch needle valve.  Excess gas, pumped by the diaphragm pump, which
did not flow through the monitor was vented.

The Dynasciences SO. monitor is an electrochemical gas analyzer.   The
sample gas is passed over a selective permeable membrane, where the
SO. Is absorbed on a sensing electrode to form activated species
capable of undergoing electro-oxidation.  The resulting current: is
directly proportional to the partial pressure of S02 in the gas mixture.
The current is amplified and the output of the amplifier recorded.

The sampling procedure is as follows:
     1.  Connect the sampling train to zero gas (nitrogen) and zero
         the SO. monitor.
     2.  Connect the sampling train to the calibration gas (guaranteed
         analysis gas - sulfur dioxide and nitrogen) and calibrate the
         SO. monitor.
                                  89

-------
                     1/4" SS Probe
              Ice Bath
Tubing
                                                 Needle  Valve

                                                 Vent
                  Oynasciences
                  SS-330
                  Monitor
FIG. A-1. S02  SAMPLING APPARATUS
                                     90

-------
     3.  Connect the sampling train to the pilot plant making sure
         that all joints are gas-tight.  Make sure the appropriate
         sampling line is open and connected with the sampling train.
     4.  Turn on the pump and adjust the gas flow rate through the
         SO, monitor (approximately 1 scfm).

DUST LOADING APPARATUS AND PROCEDURE
The patticulate sampling train is shown in Figure A-2.  The probe,
consisting of 1/4 inch stainless steel tubing, was connected to the
first Greenberg-Smith impinger by a 4 foot length of Tygon tubing.
Impingersf:No. 1 and No. 2 were each filled with 200 ml of distilled
water.  Impinger No. 3 was dry and used as a water trap.

In operation, the gas velocity through the duct was calculated before
the start of the particulate sampling by means of the Annubar flow
measuring element.  The apparatus was assembled and the probe inserted
into the center of the duct, open end upstream.  Initial dry gas meter
readings were recorded and the vacuum pump started.  The flow rate
through the dry gas meter was determined with a stop watch by measuring
the time required for 0.1 ft3 of gas to be pulled through the meter.
The flow was then regulated by means of a gas control valve until
the flow rate through the probe equalled the flow rate through the
duct and an isokinetic condition existed.  Between 10 and 20 standard
cubic feet of gas  was drawn through the train.  The temperature
and pressure inside the dry gas meter were recorded from the attached
thermometer and vacuum gauge; the vacuum pump was shut off and the
final dry gas meter reading was recorded.

NITROGEN OXIDE APPARATUS AND PROCEDURE
Principle
Stack gas samples are collected in evacuated flasks containing an
absorbent consisting of hydrogen perioxide in dilute sulfuric acid.
The nitric acid formed by the oxidation and absorption of nitrogen
oxides Is used to nitrate phenoldisulfonic acid which, when
                                  91

-------
    Pnbi  la Duel
               Impinger No.l
                 (H20)
                                                                              Tomporoton
Impinger No.2
  (H20)
Impinger No. 3
    Dry
Drr  Gis Motor
Vacuum Pump
FIG. A -2. PARTICULATE  SAMPLING  TRAIN

-------
reacted with ammonium hydroxide, forms a yellow compound (5-nitro,
6 hydroxy, 1, 3-benzenedisulfonic acid, triammonium salt).  The
intensity of the color produced is proportional to the concentration
of nitrogen oxides in the sample and is measured spectrophotometri-
cally at 420 my.

Interference
Inorganic nitrates, nitrites, or organic bearing compounds easily
oxidized to nitrates and interfere with this method. Reducing agents, such
as S0_, when present in high concentrations, may Interfere by reacting
with the hydrogen peroxide in the absorbing reagent to leave an in-
sufficient amount for reaction with the nitrogen oxides.  Halides tend
to Interfere, if present, and give lower results.

Reagents
     Hydrogen peroxide solution - Dilute 10 ml of 30 percent
        H-O- to 100 ml in a 100 ml volumetric flask with water.
     Sulfuric acid (0.1 N) - Dilute 2.8 ml of concentrated H.SO, to
                                                            £•  •§•
        1 liter with water.
     Absorbing reagent - Dilute 6 ml of 3 percent H_0. to 1 liter with
        0.1 N H.SO,.  This solution is stable and may be used for at
        least 30 days.  Analyses in this laboratory have shown that
        the percent H.O. in the absorbing reagent remained constant
        over a 49 day period.
     Sodium hydroxide (1 N) - Dissolve 40 gm of NaOH pellets in water
        and dilute to 1 liter.
     Ammonium hydroxide (concentrated)
     Sulfuric acid (fuming)
                                   93

-------
     Phenoldisulfonic acid solution - Dissolve 25 grams of pure
        white phenol in 150 ml of concentrated H.SO. on a steam bath.
        Cool and add 75 ml fuming sulfuric acid.  Heat to 100°C for 2
        hours.  Store in a dark stoppered reagent bottle.
     Potassium nitrate solution (standard) - Dissolve 0.5495 grams
        of KNO- in 1 liter of water in a volumetric flask.  Dilute
        100 ml of this solution to 1 liter in a volumetric flask.
        One ml of the final solution is equivalent to 0.025 mg NO..

Collection of Samples
Emission sources containing oxides of nitrogen are sampled by a grab
sampling technique using an evacuated 250 ml flask.
The following procedure is used for the collection of samples: Add
25 ml of absorbing solution to the sample flask.  Evacuate the flask
to the vapor pressure of the absorbing solution (approximately 20 mm
Hg).  Disconnect the vacuum pump line and accurately measure the
vacuum in the flask.  Connect the flask to the sample line and allow
the flask to fill with a sample of stack gas until there is very little
or no vacuum left.  Measure precisely the final vacuum in the flask
and record the flask temperature.  Shake the flask for 15 minutes and
allow to stand overnight to ensure complete reaction and absorption
of the nitrogen oxides.
Analysis
Transfer the contents of the collection flask to a 250 ml beaker.
Wash the flask three times with 10 ml of water and add to the beaker.
For a blank, add 25 ml of absorbing solution and 30 ml of water to a
250 ml beaker.  Proceed as follows for both the sample and blank:
Add 1 N NaOH dropwlse to the beaker until the solution is alkaline to
litmus paper.  Evaporate to dryness on a steam bath and allow to cool.
Add 2 ml of phenoldisulfonlc acid solution to the residue and triturate
thoroughly with a glass stirring rod.  Make sure all the residue comes in
                                  94

-------
contact with the solution.  Add 1 ml HO and 4 drops of concentrated
H-SO, .  Heat the solution on the steam bath for 3 minutes with occasional
stirring.
Allow the solution to cool, add 20 ml H.O, mix well, and add 10 ml of
concentrated NH.OH, dropwlse, with constant stirring.  Transfer the
solution to a 50 ml volumetric flask.  Wash the beaker three times with
5 ml portions of water.  Dilute to 50 ml and mix thoroughly.  Transfer a
portion of the solution to a centrifuge tube and centrifuge for several
minutes.  If no centrifuge is available, filter the solution.
Determine the absorbency of each sample at 420 my.  If the absorbency
is outside the range of the calibration curve (e.g., absorbency >0.6),
make a suitable dilution of the sample and blank and determine the
absorbency.  Obtain the number of milligrams of N0» present in the
sample from a previously prepared calibration curve, where absorbency
was plotted versus concentration.
Calculations
Calculate the concentration of oxides of nitrogen as N0_ in parts per
million by volume as follows:
                       ppm
                                         s
where    C   =  concentration of N0~, mg
         V   =  gas  sample volume at 70°F and 29.92 in. Hg, liters
         o
Calculate the  volume  of gas sampled at standard conditions of 70 °F,
29.92  in. Hg.
                                       Vf(Pf-P±) x 530°R
               Volume  of gas sampled =  Tf x 29.92 in. Hg

where    Vf  = flask volume, liters
         Pf  = final flask pressure, in. Hg
         P.  = initial  flask pressure, in. Hg
         T-  = flask temperature, °R

                                  95

-------
CALCIUM—EDTA TITRIMETRIC METHOD

Principle

When EDTA (ethylenediamine tetraacetic acid or its salts) is added to
water containing both calcium and magnesium, it combines first with the
calcium that is present.  Calcium can be determined directly using EDTA
when the pH is made sufficiently high so that the magnesium is largely
precipitated as the hydroxide and an indicator is used which combines
with the calcium only.  Several indicators are available that will give
a color change at the point where all of the calcium has been complexed
by the EDTA at a pH of 12-13.

Interference

Under conditions of this test, the following concentrations of ions
cause no interference with the calcium hardness determination: copper,
2 tog/1; ferrous iron, 20 mg/1; ferric iron, 20 mg/1; manganese, 10 mg/1;
zinc, 5 mg/1; lead, 5 mg/1; aluminum, 5 mg/1; tin, 5 mg/1.  Ortho-
phosphate will precipitate calcium at the pH of the test.  Strontium
and barium Interfere with the calcium determination and alkalinity
in excess of 30 mg/1 may cause an indistinct endpoint with hard waters.

Reagents

     Sodium hydroxide, IN- Dissolve 40 g NaOH and dilute to 1 liter
        with distilled water.
     Murexide (ammonium purpurate) was prepared by mixing it with NaCl.

a.  Analytical reagent grade disodium ethylenediamine tetraacetate
dihydrate, also called (ethylenedinitrilo) tetraacetic acid disodium
salt (EDTA), Na0H0C. -H.. 00QN • 2H.O, is commercially available.  Weigh
               / L 1U J.Z o i    /
3.723 g of the dry reagent, dissolve in distilled water, and dilute
to 1,000 ml.  Check the titer by standardizing against standard calcium
solution.
                                  96

-------
b.  The technical grade of the disodium salt of EDTA dlhydrate may
also be used if the titrant is allowed to stand for several days and
is then filtered.  Dissolve 4.0 g of such material in 800 ml distilled
water.  Standardize against standard calcium solution.  Adjust the
titrant so that 1.00 ml - 1.00 mg CaCO..

Because the titrant extracts haedness-producing cations from soft glass
containers, store preferably in polyethylene and secondarily in Pyrex
bottles.  Compensate for gradual deterioration by periodic restandardi-
zation and a suitable correction factor.

Procedure

Because of the high pH used in this procedure, the titration should be
performed immediately after the addition of the alkali.

Use 1 ml of sample diluted to 50 ml with water.

Add 2.0 ml NaOH solution, or a volume sufficient to produce a pH of
12-13.  Stir.  Add 0.1-0.2 g of the indicator mixture selected (or
1-2 drops if a solution is used).  Add EDTA titrant slowly with continuous
stirring to the proper endpolnt.  When using murexide, the endpoint may
be checked by adding 1 or 2 drops of titrant in excess to make certain
that no further color change occurs.

        Calculation:  1 ml EDTA = 0.4008 mg Ca"*"1"
                        _l_l_
                      Ca   mg/1 or ppm = Vol. EDTA x 0.4008 x 1000

MAGNESIUM—EDTA TITRIMETMC METHOD

Principle

EDTA and its soldum salts form a chelated soluble complex when added to
a solution of certain metal cations.  If a small amount of a dye such as
Eriochrome Black T is added to an aqueous solution containing calcium and
magnesium ions at a pH of 10.0 + 0.1, the solution will become wine red.
                                  97

-------
 If  EDTA is  then added as  a titrant,  the  calcium and magnesium will be
 complexed.   After sufficient  EDTA has been added to complex  all  the
 magnesium and calcium,  the solution  will turn  from wine  red  to blue.
 This  is the endpoint  of the titration.   Magnesium ion must be present
 to  yield a  satisfactory endpoint  in  the  titration.  A small  amount of
 complexometrically neutral magnesium salt of EDTA is therefore added
 to  the  buffer,  a step which automatically introduces sufficient
 magnesium and at the  same time  obviates  a blank correction.

 The sharpness of the  endpoint increases  with increasing  pH.  The
 pH, however, cannot be  increased  indefinitely  because of the danger
 of  precipitating CaCO~  or Mg(OH)2» and because the dye changes color
 at  high pH  values.  The pH value  of  10,0 + 0.1 recommended in this
 procedure is a  satisfactory compromise.   A limit of 5 minutes is set
 for the duration of the titration in order to  minimize the tendency
 toward  CaCO- precipitation.

 Interference

 Some metal  ions  interfere with this procedure by causing fading or in-
 distinct  endpolnts.  This interference is  reduced by the addition of
 certain  inhibitors to the water sample prior to  titration with EDTA.

 Suspended or  colloidal  organic matter in  the sample may also interfere
with the endpoint but may be overcome by evaporating the aliquot to
 dryness on a  steam bath, followed by heating in a muffle furnace at
 600°C until  the  organic matter is completely oxidized.   Dissolve the
 residue in 20 ml 1 N HC1,  neutralize to pH 7 with 1 N NaOH, and make
 up to 50 ml with distilled water; cool to room temperature and continue
 according to  the general procedure.

 Titration Precautions
 Titrations are best conducted at or near normal room temperatures.   The
 color change  becomes impractically slow as the sample approaches freezing
 temperature.  Indicator decomposition presents a problem in hot water.
                                  98

-------
The. pH specified in the recommended procedure may result in an environ-
ment conducive to CaCO_ precipitation.  Although the titrant can
slowly redissolve such precipitates, a drifting endpoint will often
yield low results.  A time limit of 5 minutes for the overall procedure
minimizes the tendency for CaCO- to precipitate.  The following three
methods also combat precipitation loss:
a.  The samples can be diluted with distilled water to reduce the
CaCO- concentration.  The simple expedient of diluting a 25 ml aliquot
to 50 ml has been incorporated in the recommended procedure.  If pre-
cipitation occurs at this dilution, modification b or c can be fol-
lowed.  Reliance upon too small an aliquot contributes a systematic
error originating from the buret-reading error.

b.  If the approximate hardness of a sample is known or is ascertained
by a preliminary titration,  90 per cent or more of the titrant can be
added to the sample before the pH is adjusted with the buffer.

c.  The sample can be acidified and stirred for 2 minutes to expel
C0? before pH adjustment with the buffer.  A prior alkalinity deter-
mination can indicate the amount of acid to be added to the sample
for this purpose.

 Reagents
     Buffer solution - Dissolve 1.179 g disodium salt of EDTA di-
        hydrate and 0.644 g MgCl2 • 6H20 in 50 ml distilled water.
        Add this solution to 16.9 g NH^Cl and 143 ml cone. NH.OH
        with mixing and dilute to 250 ml with distilled water.
     Eriochrome Black T is mixed with NaCl and used as an
        Indicator.
     Standard EDTA titrant, 0.01M.   Analytical reagent grade
        disodium ethylenediamlne tetraacetate dihydrate, also
        called (ethylenedinitrilo)  tetraacetic acid disodium
        salt (EDTA), Na2H2C10H12°8H2 * 2H2°' is commerclally
        available.  Weigh 3.723 g of the dry reagent, dissolve
        in distilled water, and dilute to 1,000 ml.   Check the
        titer by standardizing against standard calcium solution.
                                  99

-------
Procedure

The aliquot of sample taken for the titration should require less than
15 ml of EDTA tltrant.  The duration of titration should not exceed
5 minutes measured from the time of the buffer addition.

Dilute 1 ml of the sample to 50 ml with distilled water.  Add 1-2 ml
of buffer solution.  Add an appropriate amount of dry-powder indicator.
Add EDTA to the endpoint.

Calculation
                        11
1 ml EDTA « 0.2431 mg Mg   (The calcium concentration, determination
by the EDTA method, is required for this calculation since the volume
                                                          11        [ |
of EDTA used in the above titration is consumed by both Ca   and Mg
ions.)
     Vol. EDTA Mg""" - Total EDTA Vol. - Vol. EDTA Ca""" (from prior Ca"""
                                                             titration)
     mg Mg"""/! - Vol. EDTA Mg""" x 0.2431 x 1000

CHLORIDE—ARGENTOMETRIC METHOD

Principle

In a neutral or slighly alkaline solution, potassium chromate can be used
to indicate the endpoint of the silver nitrate titration of chloride.
Silver chloride is quantitatively precipitated before red silver chromate
is formed.

Interference

Substances in amounts normally found in potable waters will not interfere.
Bromide, iodide, and cyanide register as equivalent chloride concentrations.
Sulfide, thiosulfate, and sulfite ions interfere.  However, sulfite can
be removed by treatment with hydrogen perdxide in a neutral solution,
while sulfite and thiosulfate can be removed by treatment with hydrogen
peroxide in alkaline solution.  Orthophosphate in excess of 25 mg/1
                                 100

-------
interferes by precipitation as silver phosphate.  Iron in excess of
10 mg/1 will interfere by masking the endpoint.

Reagents

     Chloride-free water - If necessary, remove any chloride
        impurity from distilled water by redistillation from
        an all-pyrex apparatus or passage through a mixed bed
        of ion-exchange resins.
     Potassium chromate indicator solution - Dissolve 50 g
        K2CrO^ in a little distilled water.  Add silver nitrate
        solution until a definite red precipitate is formed.
        Allow to stand 12 hrs , filter, and dilute filtrate
        to 1 liter with distilled water.

        Standard silver nitrate titrant, 0.0141 N - Dissolve 2.396 g
           AgN03 in distilled water and dilute to 1,000 ml.  Stan-
           dardize against 0.0141 N NaCl.  Store in a brown bottle.
           Standard silver nitrate solution, exactly 0.0141 N,  is
           equivalent to 0.500 mg Cl per 1.00 ml.

 NITRITE

 Principle

 Diazotized sulfanilic acid,  formed by the reaction between sulfanilic
 acid and NO., forms a reddish-purple azo dye by coupling with  riapthy-
 lamine hydrochloride at pH 2 to 2.5.  The nitrite concentration is
 determined by spectrophotometrically measuring this dye at 520 m .

 Interference

 This method is not interfered with by relatively large amounts, up to
 1,000 times, of the alkaline earths, zinc, nickel, arsenate, benzoate,
 borate, bromide, chloride, fluoride, iodate, molybdate, nitrate, phosphate,
                                 101

-------
sulfate, and thlocyanate.  Numerous heavy metals such as gold, lead,
bismuth, iron, or mercury interfere by precipitation and others because
of colored salts.  Aliphatic amines react with nitrites to liberate
gaseous nitrogen.   Ammonia does not interfere in the small concentrations
usually encountered.  Strong reducing or oxidizing agents should be
absent.

Reagents

    Sulfanilic acid solution - Dissolve 0.60 g sulfanilic acid
       in 70 ml hot distilled water, cool, add 20 ml cone. HC1,
       dilute to 100 ml with distilled'water, and mix thoroughly.
    Naphthylamine hydrochloride solution - Dissolve 0.60 g 1-
       naphthylamine hydrochloride and 1 ml cone. HC1 in
       distilled water and dilute to 100 ml.
    Sodium acetate solution, 2M - Dissolve 16.4 g NaC2H,02 or
       27.2 g NaC2H302 • 3H20 in distilled water and dilute
       to 100 ml.  Filter if the solution is not clear.
    Stock sodium nitrite solution - Dissolve 0.492 g NaN02 in
       1,000 ml nitrite-free distilled water.

    Standard sodium nitrite solution - Dilute 100.0 ml stock
       sodium nitrite solution to 1,000 ml; then dilute 50.0
       ml of this solution to 1,000 ml with sterilized nitrite-
       free distilled water, add 1 ml chloroform, and preserve
       in a sterilized bottle; 1.0 ml = 0.5 ug  [N] or 1.6 yg
       N0~.
    Manganese sulfate solution -   Dissolve 480 g MnSO,  •  4 H90 or
        400 g MnS04  •  2  H20 or 364  g MnSO^ • H20 in distilled water,
        filter,  and  dilute to 1 liter.
     Potassium permanganate solution -  Dissolve 0.4 g KMnO,  in 1
        liter distilled water.
                                102

-------
      Ammonium oxalate solution - Dissolve 0.9 g
         in 1 liter distilled water.
      Nitrite-free water - Add 1 ml cone.  H_SO,  and 0.2  ml man-
         ganous sulfate solution to 1 liter distilled water and
         make pink with 1 to 3 ml potassium permanganate solution.
         After 15 minutes decolorize with  ammonium oxalate solution.

Procedure

     Using appropriate dilutions of standard sodium nitrite solution,
prepare a curve for various N0~  concentrations ranging from 0.05 to
1.4 mg N0~/I and plot against absorption at 320 m  on a spectrophotometer
using a light path of 2 cm.

Place a 10 ml sample  In  the  tube.  Measure 1.0 ml sulfanilic acid
solution into the  diluted sample, mix, and allow to stand  at least 3
minutes and not more than 10 minutes for diazotization.   The pH of
this solution should be  about  1.4
Add 1.0 ml napthylamine hydrochloride solution and 1 ml sodium acetate
solution.  This should buffer  the system  to a pH of 2.5.  Dilute to
50 ml and mix well.  After 10  minutes, but before 20 minutes, measure
the intensity of  the  reddish-purple  color in  a spectrophotometer, a
filter photometer, or by comparison  in Nessler  tubes.

Calculation

Read N0= mg/1 directly from  the  calibration curve.

NITRATE—BRUCINE METHOD

Principle

The reaction between nitrate and bruclne  yields a sulfur yellow color
employed for colorlmetric estimation.  The color system does not obey
Beer's law,  although in plotting transmlttance against nitrate concentration
                                 103

-------
 a smooth  curve  Is produced.   It  is necessary  to  develop  color simulta-
 neously in  a  series  of  standards and  samples.  The  intensity of  the
 color  is  measured at 410 m .

The intensity of the maximum color produced varies more or less inverse-
ly with the temperature, while the rate of color development varies
more or less directly with the temperature.  The temperature generated
upon mixing sulfuric acid with water can be controlled by adjusting
the acid concentration.   Both the acid concentration and the reaction
time have been selected to yield optimum results and to compensate for
any normal variations in room temperature.

 Interference

 All strong  oxidizing or reducing agents  interfere.   The  presence of
 oxidizing agents may be determined by the addition  of  orthotolidine
 reagent as  in the measurement of residual chlorine.  The interference
 by residual chlorine may be eliminated by the addition of sodium arsenite,
 provided  that the residual chlorine does not  exceed 5  mg/1.  A slight
 excess of sodium arsenite  will affect the determination.   Ferrous and
 ferric iron and quadrivalent manganese give slight  positive interferences,
 but in concentrations less than  1 mg/1 these  interferences are minimized
 by the use  of sulfanilic acid.   Chlorides do  not interfere.

 Reagents

     Stock nitrate solution - Dissolve O.Z218 g  anhydrous potas-
        sium nitrate, KNO-, and  dilute to  1,000  ml with distilled
        water.  This solution contains 100 mg/1  N.
     Standard nitrate solution - Dilute  100.0 ml stock nitrate
        solution to  1,000 ml wi.th distilled water; 1.00 ml =
        10.0 pg N.
                                 104

-------
    Sodium arsenite solution - Dissolve 5.0 g NaAsCK and dilute
       to 1 liter with distilled water.   (CAUTION:  Toxic; take
       care to avoid ingestion.)
    Brucine-sulfanilic acid - Dissolve 1  g brucine sulfate and
       0.1 g sulfanilic acid in approximately 70 ml hot distilled
       water.  Add 3 ml cone. HC1, cool and make up to 100 ml.
       This solution is stable for several months.  The pink
       color that develops slowly does not affect its usefulness.
       (CAUTION:  Brucine is toxicj  take  care to avoid ingestion.)
    Sulfuric acid solution - Carefully add 500 ml of cone. H SO,
       to 74 ml distilled water.  Cool to room temperature before
       use.  Keep tightly stoppered  to prevent absorption of
       atmospheric moisture.

Procedure

Prepare a calibration curve by plotting concentrations from 0.1 to
2.6 mg/1 N0~ against absorption at 410 m  on a spectrophotometer using
a light path of 2 cm.

Color Development

Carefully pipet 2.00 ml of sample containing not more than 10 mg/1
nitrogen Into a 50 ml beaker.  Add 1.0 ml brucine-sulfanilic acid reagent,
using a safety plpet.  Into a second 50 ml meaker measure 10 ml H.SO,.
(An automatic buret Is convenient for this purpose.   The intensity of
color Is affected slightly by the heat capacity of the containers.  The
concentration of H-SO, has been chosen so that normal variations In
heat capacities of beakers will not affect the result.  It is important,
however,  that only 50 ml beakers be used.)  Mix the contents of the two
beakers by carefully adding the sample with brucine-sulfanilic acid
reagent to the beaker containing acid.  Pour from one beaker to the other
four to six times to ensure mixing.  Allow the treated sample to remain
In the dark for 10+1 minutes.  (The beaker may conveniently be covered
                                105

-------
with a cardboard carton during this period.)  While the sample is
standing for color development, measure 10 ml distilled water into the
empty beaker.  After 10 minutes, add the water to the sample and mix as
before.  Allow to cool in the dark for 20-30 minutes.  Set the blank at
100 per cent transmittance at a wave-length of 410 m**.  It is advisable
to run a series of standards with each set of samples.  With a proper
arrangement of work, as many as twelve samples may be determined in a
batch along with eight standards.

Calculation

Read NO.mg/1 directly from the calibration curve.

SULFITE

Principle

An acidified water sample containing sulfite is titrated with a
standardized potassium iodide-iodate tltrant.  Free iodine is released
when the sulfite has been completely oxidized, resulting in the
formation of a blue color in the presence of starch Indicator.

Interference

The presence of other oxldlzable substances in the water such as organic
matter and sulfide will result in higher titratlon values for sulfite
than are actually present.  Nitrite, on the other hand, will combine with
sulfite in the acid medium to destroy both, leading to low results.  No
interference occurs with the Dual-Purpose Dry Starch Indicator Powder
because the sulfonic acid in this proprietary compound destroys the
nitrite.  Copper ion rapidly accelerates the oxidation of sulfite solution.
Certain heavy metals may also react in a manner similar to coppper.  Proper
sampling and Immediate fixing by acid addition should minimize those
difficulties.
                                 106

-------
Reagents

    .Sulfuric acid, 1+1.
    Starch solution - To 5 g starch (potato, arrowroot, or
      soluble) In a mortar, add a little cold distilled water and
      grind to a paste.  Pour into 1 liter of boiling distilled
      water, stir, and allow to settle overnight.  Use the clear
      supernant.  Preserve by adding either 1.3 g salicylic acid,
      4 g zinc chloride, or a combination of A g sodium proplonate
      and 2 g sodium azide to 1 liter of starch solution.

    Standard potassium iodide-iodate titrant, 0.0125 N - Dissolve
       0.4458 g anhydrous potassium iodate, KIO, (primary stan-
       dard grade dried for several hours at 120°C) , 4.35 g
       potassium iodide, KI, and 0.31 g sodium bicarbonate,
       NaHCO., in distilled water, and dilute to 1,000 ml.
            J                                   «.
       This titrant is equivalent to 0.500 rag SO- per 1.00 ml.


 Procedure
 Collect a fresh water sample with as little contact with air as possible.
 Cool hot samples to
 filter the samples.
Cool hot samples to 50 C or below in the cooling apparatus.   Do not
 Add 1 ml H.SO, (or 1 g dual-purpose starch indicator) to a 250 ml
 Erlenmeyer flask or other titrating vessel, then measure 50 ml water
 sample in a graduated cylinder, and transfer to the flask.  Add 1 ml
 starch Indicator solution or 0.1 g starch powder.  Titrate with potassium
 iodide-iodate titrant until a faint permanent blue color develops in the
 sampel, signaling the end of the titration.  View the color changes against
 a white background.
                                  107

-------
 Calculation

     mg/1  S0° equals A x N x 40.000
                       ml sample
     where:  A = ml of titrant used for sample
             N - normality of KI-KIO..

SULFATE -- GRAVIMETRIC METHOD

 Principle

 Sulfate is precipitated in a hydrochloric acid medium as barium sulfate
 by  the addition of barium chloride.  The precipitation is carried out
 near the boiling temperature and, after' a period of digestion, the
 precipitate is filtered, washed with water until free of chlorides,
 ignited or dried, and weighed as BaSO,.

 Interference

 The gravimetric determination of sulfate is subject to many errors, both
 positive and negative.  In potable waters where the mineral concentration
 is  low, these may:be of minor importance.  The analyst should be familiar
 with the more common interferences, however, so that he may apply
 corrective measures when necessary.

 Reagents

    Methyl red indicator solution - Dissolve 0.1 g methyl red
       sodium salt in distilled water and dilute to 100 ml.
    Hydrochloric acid- 1+1.
    Barium chloride solution - Dissolve 100 g BaCl2 • 2H20 in
       1 liter distilled water.  Filter through a membrane filter
       or  hard-finish filter paper before use; 1 ml of this
       reagent is capable of precipitating approximately 40 mg
       S0=
                                 108

-------
    Asbestos cream - Add 15 g acid-washed medium-fiber asbestos,
       which is prepared especially for Gooch crucible determina-
       tions, to 1 liter distilled water.  Remove the fine
       material from the asbestos before use by repeated decan-
       tations.
    Silver nitrate-nitric acid •>• Dissolve 8.5 g AgNO» and 0.5 ml
       cone. HNO_ in 500 ml distilled water.

Procedure

Adjust the  clarified sample — treated if necessary to remove inter-
fering agents — to contain approximately 50 mg of sulfate ion in a
250 ml volume.  Adjust the acidity with HC1 to pH 4.5-5.0 using a pH
meter or the orange color of methyl red Indicator.  Then add on additional
1 to 2 ml HC1.  Lower concentrations of sulfate ion may be tolerated if
It Is Impractical  to concentrate the sample to the optimum level, but
In such cases  It Is better to fix the total volume at 150 ml.  Heat the
solution to boiling and, while stirring gently, add warm barium chloride
solution slowly until precipitation appears to be complete; then add
about 2 ml  In  excess.  If the amount of precipitate is small, add a total
of 5 ml barium chloride  solution.  Digest the precipitate at 80°-90°C,
preferably  overnight but for not less than  2 hours.

Prepare an  asbestos filter mat in a Gooch crucible by using suitable
suction apparatus.  Wash with several portions of hot distilled water,
dry, and ignite at 800°C for 1 hour.  Cool the crucible in a desiccator
and weigh.

Mix a small amount of ashless filter paper pulp with the barium sul-
fate, and filter at room temperature.  The pulp aids filtration and
reduces the tendency of the precipitate to creep.  Wash the precipi-
tate with small portions of warm distilled water until the washings
are free of chlorides,  as indicated by testing with silver nitrate-
                                109

-------
nitric acid reagent.  Dry the filter and precipitate and Ignite at
800°C for 1 hour.  Do not allow the filter paper to flame.  Cool In a
desiccator and weight.

Calculation

     »g/l S0; equals  mgBag°4 * 411'5
                         ml sample
                                  110

-------
                APPENDIX B

ANNUBAR FLOW ELEMENT, CALCULATION EQUATIONS,

      AND ENGINEERING SPECIFICATIONS

               Supplied by
        Ellison Instrument Division
             Boulder, Colorado
                   111

-------
                                                                                               SECTION  C
             ELLISON
            ANNUBAR  FLOW
      CALCULATION  REPORT
                                                                            • TUX +m-mo» • CABLB i
Annubar Primary Elements, like other differential mea-
•uring flow elements,  utilize  a form of the classical
Bernoulli energy  balance equation to determine flow
rate.  But, unlike  most other flow elements, Annubar's
differential  pressure signal is  consistent and uniform
for • given pipe size and flow condition.  Since it is
uniform, only the  operating range of the Instrument or
control system needs to be sized.  This feature reduces
calculations and also allows  future changes to a sys-
tem's flow rate  without the necessity of resizing and
changing primary elements.
Sizing the operating range of  the instrument to be used
with an Annubar element is made easy with the charts
and formulas shown below. If you  prefer to have Elli-
son compute your instrument's operating range there is
a small charge for this engineering service.
                                   HELPFUL HINTS FOR SIZING
1. Select one of the calculation methods shown below
   according to your accuracy requirements.
2. When flow Is  liquid, check your result against the
   •Quick-Size* chart in the Annubar catalogue, E-100.
   This is a fast way to double check calculations. ,
3. If your system's flow rates are extremely high or ex-
   tremely low, it may be desirable to change the pipe
   •tee for the  metered section. Changing diameters of
   the metered section will not affect accuracy providing
   sufficient upstream and downstream lengths of pipe
   are provided	see Table II.
4. Contact your local Ellison Engineering Representative
   or our factory if you need any further information
   or help In sizing your  instrument's operating range
   .... we are here to serve you.
                                    EASY INSTRUMENT SIZING
Select flic  calculation method that meets your needs:
A. QUICK-SIZE CHART - fastest  sizing  for water
   flows — see chart  on the fold-out section of the
   Annubar catalogue,  E-100.
8. STANDARD 'PLANT* EQUATIONS - for general
   industrial metering needs .... see "B" below.
 C. PRECISE  "THEORETICAL" EQUATIONS  - for
    highest accuracy requirements and laboratory work
    .... see *C~ on next page.
 D. SLIDE RULE CALCULATOR - convenient for fast
    sizing of gases, liquids, and steam.  Request form
    E-A7 from your Ellison Representative.
                              -B" STANDARD  'PLANT"  EQUATIONS

                  lb.  Liquid volume flow rate.
                                                 .or    b, =
                                                                Q.Gi
                                                              SND*
                 2b. Liquid weight flow rate,
                                                .or
                 3b. -Gas volume flow rate at standard condition,*
                            = 7.9SND'-
                 4b.  Cat or steam weight flow rate,
                                                  .or
                                                        h.=
                                                112

-------
                              -pai T!W 4 t«e
2-to2'i-Plpe -Kf.O.tT*
3- to 4- Pipe _ Kf - 0.877
5- lo 6" Pip. _ Kf • O.S7*
8' Pipe - Kf « 0.8*7
10- Pipe - Kf « O.tM
12- Pipe - Kf - 0.90*
14- Pipe - Kf - O.tIT
16- Pipe - Kf - O.SM
18- Pipe - Kf - O.MS
20- Pipe - K* •» O.S4*
24* Pipe - K« . O.M*
     F. = Velocity distribution factor:
          F, = 0.82 IDT tnmlUoci «nd nubuleu Bov.

 V. = Gas adiabatic compression factor. Velocities below
     12,000 ft/min. use 1.00.
     k =  lUUo of ifecllk kwb. or tamtroolc txpoateL CUM wkkk
          follow dM perhcl *M law (Diatomic faeet vaA u Oiyfta,
          Hy^rofta, NBrofem. Air am) etc.) ban k = 1.40 naw
               	                        (M4)
                                                              P. = Upnnui itttlc piwrare to

                                                              P, = T«erf »r Imptcl pfeMun
          • Mr** vWi dlf Hal wulpmtM h raraeoMwM.

W. = Weight flow rate in convenient unit*. See Table L
                                            (p. 148)
                                                      .  113

-------
r, =Specific weight at flowing conditions in pounds per
    cubic foot Including compressibility.
T, = 14.7+ PSIG of line           520	
           14.7          460 +line temp («F) X  r'
Abo, 1, = specific gravity of gas at flowing conditions
times the weight of air (*/ft.3) at conditions equal to
                     flowing condition*. See Figure 1 and 3.  (p. 333,372)
                     7i = Specific weight of gas at base conditions in pounds
                          per cubic foot. 7, = Specific Gravity of gas at base
                          conditions times the weight of air  ('/ft.')  at base
                          conditions. Air = .0765*/ft> at standard  (60'F./
                          14.73 psia) base conditions.               (p. 78)
                       TABLE I - Factor "N" Values for Various "n" Units
Qi, Volume
GPM
GPH
CFM
CFH
LPM
"hn- Units of Differential Pressure, Dry Calibration
IN. of H,0»
5.667
340.0
0.7576
45.46
21.45
IN. of Hg. •
20.88
1252.0
2.791
167.5
79.02
Kg/cm
112.5
6750.
15.04
902.5
425.8
P.S.L
29.84
1790.
3.990
239.4
113.0
     W., Weight
PPM
PPH
47.25
2835.
174.1
10440.
938.0
56280.
246.8
14930.
                                   •Column readings corrected to 68* F.
FIGURE 1. VALUES OF
                                                               FOR AIR
                              100           200           300

                                            TEMPERATURE. °F.
                                      400
500
                                              114

-------
LOCATED 3/16"
UPSTREAM FROM
WELD COUPLING ON
OPPOSITE SIDE
       "C~-SENSING PORTS ARE
       LOCATED IN APPROPRIATE
       CENTERS OF CONCENTRIC
       ANNUL!.
                                            ASMS PAR UW- IS DRILL OR BURN
                                            TWO 1" DIAMETER HOLES AND
                                            PROVIDE 1/4" MINIMUM WELD
                                            BEAD • ALIGN TO PIPE AXIS. (TYPICAL)
                                                 2-ASTM SPEC. AtOB GRADE NO. 2
                                                 2/3000 LBS. 1/2" MPT FORGED
                                                 STEEL WELD COUPLINGS
                                                 ARE SUPPLIED.
                                                                                       REFER TO
                                                                                       SECTION DETAIL
   TOP FLAT ON HEX
   TO BE SET PARALLEL
   TO PIPE AXIS
                                                                                                                        PERMANENT METAL
                                                                                                                        TAO WITH 3" CHAIN.
                                            2MOP.8.I
                                            RATED METAL
                                            COMPRESSION
                                            FERRULE
0)3 TO 4 PIPE DIAMETERS IS RECOMMENDED FOR DOWNSTREAM SIDE.
  6 OR MORE PIPE DIAMETERS IS F.JCOMMENDED FOR UPSTREAM SIDE
  AFTER VALVES. ELBOWS * ETC.-SEE FORM E-TfL
  PERMANENT TAG SHOWING MIN., NORM. * MAX. DESIGNED FLOWS, METER
  READINGS FOR DESIGNED FLOWS, TAG NO., LINE SIZE.SER. NO.ft METERED FLUID
                                                                                               OI*T>HieM •TAMOAHO OOHVOKATION
                                                                                                 LklBOM IMSTHUMBMT OIVIBION
                                                                                         741  TO 744 ANNUBAR FLOW ELEMENTS
* SPECIFY PIPE  SCHEDULE OR  I.D. ft O.D.
1961-OIETERICH STANDARD CORP., BOULDER, COLO.

-------
               APPENDIX C




SUMMARY OF SCRUBBER OPERATING CONDITIONS
                   116

-------
TABLE C-l.  SCRUBBER OPERATING CONDITIONS—SERIES S-XX—KEY WEST INITIAL SALT WATER TESTS
Test
Number
SI
S2
S3
S4
S5
S6
S7
S8
S9
S10
Sll
S12
S13
S14
S15
S16
S17
S18
S19
PURPOSE
Preliminary
Preliminary
Preliminary
Preliminary
-
-
'-
-
-
-
Break in
Break in
Break in
Break in
Break in
Break in
Break in
-
Break in
TUBE
SIZE
(inches)
.12
12
12
12
-
-
-
-
-
-
12
12
12
12
12
12
12
-
12
GAS FLOW
RATE
(scfm)
2000
1013
1013
935
-
-
-
-
-
-
1572
788
783
789
1538
1538
1652
-
784
PRESSURE
DROP
(inches H20)
8.5
6.5
6.5
6.5
V
V
V
V
V
V
12.0
6.5
6.5
6.5
12.0
12.0
12.0
V
6.5
STOICHIOMETRIC RATIO
/ Ib-mole reactant \
\lb-mole entering SO./
. . -
-
- ,•
1:1 '
) I D
) I D
) I D
) I D
) I D
) I D
-
-
-
-
-
-
-
) I D
-
PARTICLE
SIZE
(mesh)
-
-
-
325
-
-
-
-
-
-
• -
-
-
-
-
-
-
-
- •
SLURRY
CONCENTRATION
(weight %)
-
-
-
3
-
-
-
•
-
-
-
-
-
-
-
-
-
-
-

-------
           TABLE C-l (contd).   SCRUBBER OPERATING CONDITIONS—SERIES S-XX—KEY WEST INITIAL SALT
                                                    WATER TESTS
Test
Number
S20
S21
S22
S23
S24
S25
S26
S27
S28
S29
S30
S31
S32
S33
S34
S35
S36
S37
PURPOSE
Break in
-
-
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
Break in
TUBE
SIZE
(inches)
12
-
-
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
GAS FLOW
RATE
(scfm)
778
-
-
795
795
1496
1478
1483
1495
1488
1480
1492
1507
1486
810
776
776
1500
PRESSURE
DROP
(inches H20)
6.5
- V 0
- V 0
12.0
12.0
6.5
6.5
6.5
6.5
6.5
6.5
12.0
12.0
12.0
12.0
12.0
12.0
12.0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO./
-
ID
ID - ;
-
-
-
-
-
-
-
-
-
-
-
-
-
-
—
PARTICLE
SIZE
(mesh)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
—
SLURRY
CONCENTRATION
(weight %)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
~
00

-------
TABLE C-l (contd).  SCRUBBER OPERATING CONDITIONS—SERIES S-XX—KEY WEST INITIAL SALT
                                         WATER TESTS
Test
Number
S38
S39
SAO
S41
S42
S43
PURPOSE
Break iu
Break in
Break in
Break in
Break in
Break in
TUBE
SIZE
(inches)
12
12
12
12
12
12
GAS FLOW
BATE
(scfin)
1468
784
782
781
780
778
PRESSURE
DROP
(inches HjO)
6.5
12.0
12.0
12.0
12.0
6.5
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO./
-
-
-
- •
-

PARTICLE
SIZE
(mesh)
-
-
-
-
-

SLURRY
CONCENTRATION
(weijrht X)
-
-
-
-
-


-------
                         TABLE C-2.   SCRUBBER OPERATING CONDITIONS—SERIES C-XX—KEY WEST CORAL
Test
Number
Cl
C2
C3
C4
C5
C6
C7
C8
C9
CIO
Cll
C12
C13
C14
CIS
C16
C17
CIS
C19
PURPOSE
-
Shakedown
Shakedown
Shakedown
Shakedown
Statistical
Duplicate C6
Statistical
Extra
Extra
-'
Duplicate C6
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
TUBE
SIZE
(inches)
-
12
12
12
12
12
12
12
12
12
-
12
12
12
12
12
12
12
12
GAS FLOW
RATE
(scfm)
-
781
779
1468
1467
1463
1478
775
779
1477
-
1370
721
1470
771
1452
772
1498
1460
PRESSURE
DROP
(inches H20)
- V 0
6.5
6.5
6.5
• 6.5
12.0
12.0
6.5
6.5
6.5
- V 0
12.0
12.0
12,0^
6.5
6.5
12.0
6.5
12.0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
lb-mole entering SO-j
1 D
1:1.9
1:1,8
l:l'
1:1
1:0.9
1:1
1:1
1:3
1:1
ID
1:1
1:3
1:3
1:3
1:1
1:1
1:3
1:3
PARTICLE
SIZE
(mesh)
-
325
325
325
325
325
325
325
325
325
-
325
325
325
325
325
325
325
100
SLURRY
CONCENTRATION
(weight %)
-
1
1
1
1
1
1
1
1
1
-
1
1
5
5
5
5
1
1
ro
o

-------
TABLE C-2 (Cont.)  SCRUBBER OPERATING CONDITIONS-cSERIES C-XX-—KEY WEST  CORAL
Test
Number
C20
C21
C22
C23
C24
C25
C26
C27
C28
C29
C30
C31
C32
C33A
C33B
C34A
C34B

PURPOSE
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Centerpoint
Centerpoint
Statistical
-
Duplicate C17
Centerpoint
Catalyst
Centerpoint
Inhibitor
Inhibitor
Inhibitor +
Catalyst
TUBE
SIZE
(inches)
12
12
12
12
12
12
12
12
12
-
12
12
12
12
12
12
12

GAS FLOW
RATE
(scfm)
794
794
1457
1483
788
1482
1146
1152
788
-
782
1120
1120
1120
1120
1120
1120

PRESSURE
DROP
(inches H20)
6.5
12.0
6.5
6.5
' 12.0
12.0
9.0
9.0
6.5
- V 0
12.0
9.0
9.0
9.0
9.0
9.0
9.0

STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO.j
1:3
1:1
1:1,
1:3
1:3
1:1
1:2
1:2
1:1
ID
1:1
1:2
1:2
1:2
1:2
1:2
1:2

PARTICLE
SIZE
(mesh)
100
100
100
100
100
100
325
325
100
-
325
325
325
325
325
325
325

SLURRY
CONCENTRATION
(weight %)
1
1
1
1
5
5
3
5
5
-
5
3
3
3
3
3
3


-------
            TABLE C-3.  SCRUBBER OPERATING  CONDITIONS—SERIES  F-XX--KEY WEST FREDONIA VALLEY LIMESTONE
Test
Number
Fl
F2
F3
F4
F5
F6
F7
F8
F9
F10
Fll
F12
F13
F14
F15
F16
F17
F18
PURPOSE
_
T
Statistical
Extra
Extra
Extra
Statistical
Duplicate F3
Statistical
Statistical
Statistical
Statistical
Statistical
Extra
Statistical
Statistical
Statistical
Statistical
TUBE
SIZE
(inches!
_
-
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
GAS FLOW
RATE
(scfm)
_
-
776
766
1462
1476
1475
777
1468
769
1497
1469
694
781
1480
798
796
1468
PRESSURE
DROP
(inches H.O)
- V 0 I
- V 0 I
6.5
6.5
6.5
6.5
12.0
6.5
12.0
6.5
6.5
6.5
12.0
12.0
12.0
6.5
12.0
6.5
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering S02j
D _
D
1:1'
1:3
1:1
1:1
1:1
1:1
1:3
1:3
1:1
1:3
1:3.3
1:1
1:3
1:3
1:1
1:1
PARTICLE
SIZE
(mesh)

-
325
325
325
325
325
325
325
325
325
325
325
325
100
100
100
100
SLURRY
CONCENTRATION
(weight Z)

-
1
1
1
1
1
1
5
5
5
1
1
5
1
1
1
1
Ni
ts>

-------
      .TABLE C-3 (Cont.). SCRUBBER OPERATING CONDITIONS-SERIES F-XX—KEY WEST FREDONIA VALLEY LIMESTONE
Test
Number
F19
F20
F21
F22
F23
F24
F25
F26
F27
F28
F29
F30
F31
F32
F33
F34
F35
F35A
PUBPOSE
Statistical
Statistical
Statistical
Statistical
Centerppint
Centerpoint
Centerpoint
Statistical
Duplicate F7
Duplicate F9
Duplicate F22
Duplicate F19
Special
Special
Special
Special
Special
Special
TUBE
SIZE
(inches)
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
12
GAS FLOW
RATE
(scfm)
.800
1453
1460
765
1107
1140
1152
790
1440
1445
778
778
1317
1110
678
784
455
1452
PRESSURE
DROP
(Inches H20)
6.5
12.0
6.5
12.0
9.0
9.0
9.0
12.0
12.0
12.0
12.0
.6.5
12,0
12.0
12.0
12.0
12.0
12.0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO,/
1:1
1:1
1:3,
1:3
1:2
1:2
1:2
1:1
1:1
1:3
1:3
1:1
1:3
1:3
1:3
1:3
1:3
1:3
PARTICLE
SIZE
(mesh)
100
100
100
100
100
325
325
325
325
325
100
100
325
325
325
325
325
325
SLURRY
CONCENTRATION
(weight %)
5
5
5
5
3
3
3
5
1
5
5
5
1
1
1
1
1
1
N>
OJ

-------
         TABLE C-3 (Cont.).  SCRUBBER OPERATING CONDITIONS—SERIES F-XX—KEY WEST FREDONIA VALLEY LIMESTONE
10
Test
Number
F36
F37
F38
F39
F40
F41
F42
PURPOSE
Special
Special
Special
Special
Special
Special
Centerpoint
TUBE
SIZE
(inches)
12
8
8
8
8
8
12
GAS FLOW
RATE
(scfm)
1093
1052
755
557
486
1320
1129
PRESSURE
DROP
(inches H-O)
12.0
12.0
12.0
12.0
12.0
12.0
9.0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering S0? j
1:3
1:3
1:3;
1:3
1:3
1:3
1:2
PARTICLE
SIZE
(mesh)
325
325
325
325
325
325
100
SLURRY
CONCENTRATION
(weight X)
1
1
1
1
1
1
3

-------
          TABLE C-4.  SCRUBBER OPERATING  CONDITIONS—SERIES HL-XX--KEY WEST LIME,  DOLOMITE,  PRECIPITATED

                                            CaCO,, AND RECYCLED LIMESTONE
Test
Number
HL1
HL2
HL3
HL4
Dl
D2
D3
D4
PCI
PC2
PCS
PC4
XI
PURPOSE
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
-
Recycle
TUBE
SIZE
(inches)
12
12
12
12
12
12
12
12
12
12
12
-
12
GAS FLOW
RATE
(scfo)
1445
790
782
1455
1430
1430
772
772
1431
778
773
-
1110
PRESSURE
DROP
(inches HjO)
6.5
12.0
6.5
12.0
6.5
12.0
6.5
12.0
6.5
12.0
6.5
- V 0

STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO*/
1:1
1:1
1:3
1:3
1:1
1:3
1:3
1:1
1:1
1:1
1:3
ID

PARTICLE
SIZE
(mesh)
-
-
-
-
-
-
-
-
-
-
-
—

SLURRY
CONCENTRATION
(weight %)
1
1
1
1
I
1
1
1
1
1
1
—
3
ts>
in

-------
              TABLE C-5.  SCRUBBER OPERATING CONDITIONS—SERIES P-XX--PADUCAH-SHAWNEE NO. 9 LIMESTONE
Test
Number
Pi
P2
P3
P4
P5
P6
P7
P8
P9
P10
Pll
P12
P13
P14
P15
P16
P17
P18
PURPOSE
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
Statistical
-
—
TUBE
SIZE
(inches)
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
-
—
GAS FLOW
RATE
(scfm)
442
422
420
426
620
620
617
615
415
415
417
417
617
618
619
619
-
—
PRESSURE
DROP
(inches H20)
6.0
6.0
12.0
3
' 12.0
12.0
6.0
6.0
6.0
6.0
12.0
12.0
6.0
6.0
12.0
12.0
- V 0
- V 0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO-/
1:1
1:2
1:1.
1:2
1:1
1:2
1:1
1:2
1:2
1:1
1:2
1:1
1:1
1:2
1:1
1:2
ID
E D
PARTICLE
SIZE
(mesh)
325
325
325
325
325
325
325
325
325
200
200
200
200
200
200
200
-
^
SLURRY
CONCENTRATION
(weight %)
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
-
'
to

-------
         TABLE C-5. (Cont.)  SCRUBBER OPERATING CONDITIONS—SERIES P-XX- —PADUCAH-SHAWNEE NO.9 LIMESTONE
Ki
Test
Number
P19
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
P31
P32
P33
P34
PURPOSE
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Catalyst
Catalyst
Catalyst
Inhibitor
Inhibitor
Dry collector
Bypassed
Dry Collector
Bypassed
Dry Collector
Bypassed
TUBE
SIZE
(inches)
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
GAS FLOW
RATE
(scfm)
518
518
517
520
520
520
520
520
522
522
522
517
517
523
620
619
PRESSURE
DROP
(inches H20)
9.0
9.0
9.0
9.0
' 9.0-
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
6.0
12.0
STOICHIOMETRIC RATIO
(Ib-mole reactant \
Ib-mole entering SO./
1:1.5
1:1.5
1:1*5
I:lv5
1:1.5
1:1.5
1:1,5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1.5
1:1
1:1
PARTICLE
SIZE
(mesh)
325
325
325
325
200
200
200
200
325
325
200
325
325
325
325
325
SLURRY
CONCENTRATION
(weight X)
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3

-------
        TABLE C-5.  (Cont.)  SCRUBBER OPERATING CONDITIONS—SERIES P-XX- —PADUCAH-SHAWNEE NO.9 LIMESTONE
to
oo
Test
Number
P35

PURPOSE
Dry Collector
Bypassed
TUBE
SIZE
(inches)
8

GAS FLOW
RATE
^scfm)
524

PRESSURE
DROP
(inches H20)
9.0

STOICHIOMETRIC RATIO
/ Ib-mole reactant \
\lb-tcole entering SO./
1:1.5

PARTICLE
SIZE
(mesh)
325

SLURRY
CONCENTRATION
(weight 7.)
1


-------
          TABLE C-6.  SCRUBBER OPERATING CONDITIONS—SERIES PA-XX, and PS-XX—PADUCAH-SHAWNEE,  NO.  9
                                 ARAGONITE AND SIMULATED KEY WEST LIMESTONE
Test
Number
PA1
PA2
PA3
PA4
PAS
PA6

PA7
PAS
PS1

PS2

PS3

PURPOSE
Statistical
Statistical
Statistical
Statistical
Dry Collector
Bypassed
Dry Collector
Bypassed
Centerpoint
Centerpoint
Simulated
Salt Water
Simulated
Salt Water
Simulated
Salt Water
TUBE
SIZE
(Inches)
8
8
8
8
8
8

8
8
8

8

8

GAS FLOW
RATE
(scfm)
620
620
386
386
520
520

522
522
525

525

525

PRESSURE
DROP
(inches BLO)
6.0
6.0
12.0
12.0
9.0
9.0

9.0
9.0
9.0

9.0

9.0

STOICHIOMETRIC RATIO
/ Ib-mole reactant \
\lb-mole entering SO./
1:1
1:2
1:1;'
1:2
1:1.5
1:1.5

1:1.5
1:1.5
1:1.5

1:1.5

i.-2

PARTICLE
SIZE
(mesh)
325
325
325
325
325
325

325
325
325

325

325

SLURRY
CONCENTRATION
(weight %)
3
3
3
3
3
3

3
3
3

3

3

VO

-------
TABLE C-7  SCRUBBER OPERATING CONDITIONS—SERIES  IPA-XX AND  IP-XX—PADUCHA-SHAWNEE NO.  10
                           ARAGONITE  AND LIMESTONE  INJECTION
Test
Number
IPA1
IPA2
IPA3
IPA4
IP1
IP2
IP3
IP4
IPS
IP6
IP7
IPS
IP9
IP10
IP11
IP12
IP13
PURPOSE
Statistical
Statistical
Statistical
Statistical
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Centerpoint
Special
Special
Special
Special
Special
TUBE
SIZE
(inches)
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
8
CAS FLOW
RATE
(scfm)
615
622
419
421
516
525
523
513
522
526
512
513
618
618
614
617
368
PRESSURE
DROP
(inches H20)
6.0
12.0
6.0
12.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
9.0
12.0
12.0
12.0
6.0
6.0
STOICHIOMETRIC RATIO
/ Ib-mole reactant \
\lb-mole entering SO./
1:104
1:2
1:2 ,-
1:1
1:1.85
1:3.0
1:0.68
1:1. li
1:2.46
1:1.31
1:1.47
1:1.47
1:0.95
1:0.95
1:1.45
1:0.99
1:0.99
PARTICLE
SIZE
(mesh)
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-

SLURRY
CONCENTRATION
(weight X)
-
-
-
—
-
-
-
-
-
-
-
-
-
-
-
-


-------
     APPENDIX D




SUMMARY OF TEST DATA
        131

-------
                 Table D-l.   DATA CONSOLIDATION — SERIES S-XX—KEY WEST  INITIAL SALT WATER TESTS
Teat
Hunber
S5
S6
S7
S8
S9
S10
Sll
S12
S13
S14
S15
S16;
S17
S18
S19
S20
S21
S22
S23
GAS HIWIDITY
(Ib/lb dry air)
IN
-
-
-
-
-
-
0.130
0.078
0.085
0.085
0.086
0.069
0.089
-
0.106
0.084
-
-
0.093
OUT
-
-
-
-
. -
-
0.072
0.071
0.063
0.050
0.092
0.084
0.076
-
0.078
0.066
-
-
0.059
DEW POINT
CP)
IN
-
-
-
-
-
-
138.0
118.8
121.5
121.3
121.8'
115.0
123.0
-
128.6
120.8
-
-
124.2
OUT
- -
-
-
-
-
- •
116.0
115.7
112.2
104.5
124.0
121.0
118.0
-
118.5
113 5
-
'
110.0
DRY GAS
FLOW HATE
(Ib/nln)
-
-
-
-
-
. -
107.12
56.29
55.57
55.99
109.05
110.78
116.81
-
54.58
55.26
-
•
56.01
LI00ID FLOW
RATE
Ub/Bln)
VOID
VOID
VOID
VOID
VOID
VOID
2644.9
656.9
656.9
656.9
2644.9
2644.S
2644.9
VOID
656.9
656.9
VOID
VOID
3764.8
GAS VELOCITY
(ft/.ec)

-
-
-
-
-
46.26
21.58
21.67
21.71
44.56
44.56
46.90
-
21.81
21.88
-
-
22.88
LIQUID/GAS RATIO
(gal/lOOOft3)
-
-
-
-
-
-
142.20 -
75.71
75.42
75.27
147.62
14 7'. 62
140.27
-
74.90
74.68
-
-
340.45
Ub/lb)

• -
; -
-
-
- -
21.85
10.83
10.90
10.81
22.33
22.33
20.79
- '
10.88
10.96
-
-
61.50
ABSORPTION PARAMETERS
;ntering SO;
;oncentratla
(DPS)

-
-
-
-
-
505
470
510
495
520
510
460
-
655
805
-
.
672
Leaving S02
3onc2ntratior
fpoml

-
-
-
-
-
245
350
360
305
320
290
190
-
485
540
-
-
375
SO, Removal
Efficiency
(Z)

-
-
-
-
—
51.5
25.6
29.4
38.4
38.4
43.2
58.7
i
26.0
32.9
-
-
44.2
ho
               NOTE:  Tests Si through S4 were preliminary tests.  No calculations were performed.

-------
            Table D-l (cont)    DATA CONSOLIDATION~SERIES  S-XX — KEY WEST  INITIAL SALT WATER TESTS
Test
Number
S24
S25
S26
S27
S28
S29
S30
S31
S32
S33
S34
S35
S36
S37
S38
S39
SAO
S41
S42
S43
CAS HPMIDITT
(Ib/lb dry air)
IN
0.094
0.089
0.086
0.093
0.091
0.092
0.088
0.086
0.092
0.078
0.074
0.085
0.085
0.085
0.089
0.081
0.086
0.082
0.085
0.084
OUT
0.078
0.100
0.090
0.077
0.090
0.088
0.079
0.087
0.077
0.068
0.046
0.060
0.068
0.087
0.083
0.069
0.061
0.045
0.076
0.055
DEW POINT '
CF)
IN
124.7
122.8
121.8
124.3
123.7
124.0
122.5
121.8
124.0
118.7
117.0
121.4
121.3
121.3
122.8
119.8
121.7
120.1
121.5
121.0
OUT
118.5
126.5
123.3
118.1
123.1
122.3
119.2
122.0
118.0
114.0
102.0
110.5
114.0
122.0
120.5
114.8
110.7
101.9
117.8
107.7
DRY GAS
FLOW RATE
(Ib/nln)
55.96
105.78
104.79
104.47
105.51
104.92
104.76
105.79
106.26
106.14
58.07
55.07
55.07
106.45
103.80
55.84
55.45
55.58
55.35
55.26
LIOUID FLOW
BATE
(lb/Bln)
3764.8
750.8
750.8
750.8
750.8
750.8
750.8
2883.8
2883.8
2883.8
3131.2
2883.8
2883.8
2644.9
802.0
3131.2
3131.2
3131.2
3131.2
802.0
GAS VELOCITY
(ft/sec)
22.88
43.40
43.71
43.61
43.33
43.71
42.95
42.65
43.12
42.87
22.07
21.88
21.94
43.18
43.29
21.75
21.77
21.79
21.71
21.84


LIQUID/GAS RATIO
(gal /1000ft3)
340.45
43.03
42.72
42.82
43.10
42.72
39.29
167.33
166.34
167.33
352.88
327.84
326.89
152.33
46.08
358.05
357.70
. 357.35
358.75
91.35
(lb/lb)
61. 5(
6.5:
6.6(
6.5!
6.5:
. 6.5'.
6.5!
25. 1<
24.8!
25. 2(
50. 2(
48. 2(
48. 2(
22.91
7.0!
51.8;
52.01
52. o;
52.1:
13.3!
ABSORPTION PARAME
entering S02
;oncentratla
(ppro)
690
565
622
620
630
630
625
518
505
500
, 482
465
510
465
450
443
445
465
480
410
Leaving S02
Concentratioi
(oonO
405
408
438
430
462
450
445
325
290
260
225
270
254
257
322
236
242
210
242
220
ERS
SO. Removal
EEficiency
«>
41.3
27.8
29.6
30.6
26.7
28.6
28.8
37.3
42.6
48.0
53.4
41.9
50.2
44.7
28.5
46.7
45.6
54.8
49.6
46.3
U)

-------
Table D-2.  DATA CONSOLIDATION—SERIES  C-XX—KEY WEST CORAL
Test
Number
Cl
C2
C3
C4
C5
C6
C7
C8
C9
CIO
Cll
C12
C13
C14
CIS
C16
C17
C18
C19
GAS HfMIDITY
(Ib/lb dry air)
IN
_
0.076
0.087
0.029
0.087
0.089
0.091
0.094
0.085
0.087
.'
0.095
0.089
0.094
0.079
0.092
0.091
0.075
0.078
OUT
— -
0.073
0.083
0.094
0.097
0.103
0.103
0.091
0.077
0.093
•" J
0.103
0.077
0.103
0.093
0.106
0.097
0.068
0.072
DEW POINT
(°F>
IN
—
117.9
122.0
122.7
122.3
122.6
123.5
124.8
121.2
122.2
-
124.9
123.0
124.5
119.0
124.0
123.5
117.5
118.7
OUT
' —
116.5
120.5
124.6
125.5
127.8
128.0
123.6
118.0
124.4
.-
127.8
118.2
127.8
124.2
128.8
125.6
114.1
116.0
DRY GAS
FLOW RATE
(Ib/min)
-
54.87
55.18
104.47
104.69
104.40
104.31
54.75
54.93
104.82
-
95.99
51.03
103.46
55.02
102.57
54.39
107.30
104.29
LinUID FLOW
RATE
(Ib/mln)
V 0 ID
704.4
704.4
807.5 .
807.5
2439.6
2439.6
704.4
704.4
807.5
VOID
2903.5
2663.0
2983.4
723.6
829.4
2982.4
807.5
2903.5
GAS VELOCITY
(ft/sec)

21.75
21.81
43.29
43.39
43.39
43.93
22.05
21.84
43.18
-
46.26
23.58
44.14
22.11
42.44
21.35
41.97
40.95
LIOUID/GAS RATIO
(gal/1000ft3)
-
80.00
79.77
46.08
45.97
138.88
137.20
78.92
79.69
46.19
-
155.05
279.03
162.50
78.69
47.00
335.98
47.52
175.13
(lb/lb)
- » -
11.71
11.74
7.14
7.15
21.66
21.44
11.80
11.74
7.10
—
27.52
47.97
7.08
12.19
7.42
50.17
7.00
25.83
ABSORPTION PARAMETERS
intering SOj
;oncentrattor
(ppm)

438
465
454
436
492
436
435
414
415
^
695
690
960
960
980
980
920
715
Leaving SOj
Concentration
(Dora)

97
100
173
174
154
128
147
75
180
-
240
67.5
240
155
410
191
298
80
SO. Removal
Efficiency
(H

77.9
78.5
61.9
60.1
68.7
70.7
64.6
81.9
56.6
-
65.5
90.2
75.0
83.8
58.2
80.5
67.6
88.8

-------
                   Table D-2 (cont) .  DATA CONSOLIDATION — SERIES  C-XX —KEY WEST CORAL
Test
Number
C20
C21
C22
C23
C24
C25
C26
C27
C28
C29
C30
C31
C32
C33A
C33B
C34A
C34B
CAS HUMIDITY
(Ib/lb dry air)
IN
0.074
0.075
0.081
0.081
0.081
0.082
0.088
0.084
0.081
. - •
0.083
0.083
0.083
0.085
0.085
0.085
0.085
OUT
0.061
0.069
0.084
0.088
0.089
0.103
0.101
0.101
0.088
-
0.095
0.094
0.094
0.101
0.101
0.101
0.101
DEW POINT
CF)
IN
117.0
117.5
119.6
119.6
119.6
120.2
122.4
121.0
119.6
. -
120.5
120.5
120.5
121.3
121.3
121.3
121.3
OUT
110.6
114.8
121.0
122.4
122.9
128.0
126.9
127.0
122.4
- .
124.8
124.5
124.5
126.9
126.9
126.9
126.9
DRY GAS
FLOW RATE
(Ib/min)
56.92
56.87
103.78
104.95
56.13
105.47
81.10
81.83
56.13
-
55.60
79.63
79.63
79.48
79.48
79.48
79.48
LIOUID FLOW
RATE
(Ib/mln)
704.4
2903.5
807.5
829.4
2735.4
2982.4
1689.2
1689.2
723.6
.V 0- I D
2982,4
1436.7
1436.7
1436.7
1436.7
1436.7
1436.7
GAS VELOCITY
(ft/aec)
21.50
21.43
41.59
42.55
22.28
42.29
32.25
32.47
21.75

21.75
32.36
32.36
32.57
32.57
32.57
32.57
LIOUID/CAS RATIO
(Ral/lOOOft3)
80.95
334.65
47.96
46.88
295.24
169.59
127.63
126.80
80.00
-
329.76
108.20
108.20
107.49
107.49
107.49
107.49
(lb/lb)
11.52
47.49
7.20
7.26
45.08
26.14
19.14
19.04
11.92
. . -'
49.53
16.66
16.66
16.66
16.66
16.66
16.66
ABSORPTION PARAME
entering SO;
:oncentratla
(ppm)
735
735
815
800
750
750
780
825
770
-
810
1000
1000
1000
1000
1000
1000
Leaving SO,
Concentratlot
(npra)
108
198
355
202
48
128
225
243
166
-
45
262
230
312
265
265
255
ERS
SO. Removal
Efficiency
(Z)
85.3
73.1
56.4
74.7
93.6
82.9
71.2
70.5
78.4
-
94.4
73.8
77.0
68.8
73.5
73.5
74.5
U>
(j:

-------
               Table D-3.  DATA CONSOLIDATION — SERIES F-XX — KEY WEST FREDONIA VALLEY LIMESTONE
Test
Number
Fl
F2
F3
F4
F5
F6
F7
F8
F9
F10
Fll
F12
F13
F14
F15
F16
F17
F18
F19
GAS HI'MIDITY
(Ib/lb dry air)
IN
-
-
0.092
0.077
0.091
0.090
0.085
0.079
0.092
0.087
0.093
0.094
0.092
0.079
0.080
0.075
0.075
0.082
0.077
OUT
- .
-
0.099
0.088
0.094
0.093
0.098
0.083
0.103
0.093
0.110
0.099
0.082
0.075
0.072
0.061
0.070
0.091
0.088
DEW POINT
CF)
IN
' -
-
123.9
118.2
123.5
123.2
121.4
119.0
123.9
122.1
124.3
124.7
124.0
119.0
119.5
117.5
117.5
120.5
118.2
OUT
' -
-
126.3
122.5
124.6
124.2
126.0
120.6
127.5
124.2
130.0
126.3
120.5
117.5
116.0
111.0
115.5
123.5
122.7
DRY GAS
FLOW RATE
(Ib/min)
-
-
54.72
54.77
103.18
104.27
104.68
55.45
103.51
54.22
105.46
103.39
48.94
55.79
105.52
57.16
57.02
104.47
57.20
LIOUID FLOW
RATE
(Ib/raln)
VOID
VOID
704.4
704.4
704.4
807.5
2903.5
704.4
2982.4
723.6
829.4
807.5
2903.5
2735.4
2903.5
704.4
2903.5
807.5
723.6
CAS VELOCITY
(ft/aec)

-
22.49
22.22
43.46
43.08
43.25
21.94
43.12
22.07
45.16
44.77
20.39
20.99
41.06
21.43
21.43
43.08
22.13
LIOUID/GAS RATIO
(gal/lOOOft3)
-
-
77.36
78.32
40.04
46.31
165.85
79.30
166.30
78.85
44.17
44.55
351.72
313.45
174.68
81.19
334.70
46.31
78.62
Ub/lb)
-
-
11.79
11.94
6.26
7.10
25.56
11.77
26.38
12.22
7.20
7.14
54.33
45.49
25.48
11.46
47.37
7.14
11.75
ABSORPTION PARAMETERS
Entering S02
;oncentratia
(ppm)

-
491
494
465
415
415
413
414
414
414
414
510
920
775
780
780
840
775
Leaving S02
Concentration
(DDIH)

—
142
83
210
172
97.5
111
52
48
207
116
49
460
135
165
183
435
143
SO- Reir.oval
Efficiency
-
—
71.1
83.2
54.8
58.6
76.5
73.1
87.4
88.4
50.0
71.9
90.4
50.0
82.6
78.8
76.5
48.2
81.5
OJ

-------
Table D-3 (cont).   DATA CONSOLIDATION — SERIES F-XX—KEY WEST FREDONIA VALLEY LIMESTONE
Test
Number
F20
F21
F22
F23
F24
F25
F26
F27
F28
F29
F30
F31
F32
F33
F34
F35
F35A
F36
F37
CAS HliMIDITT
(Ib/lb dry air)
IN
0.086
0.088
0.078
0.085
0.085
0.084
0.083
0.088
0.081
0.090
0.089
0.077
0.086
0.092
0.092
0.089
0.089
0.082
0.088
OUT
0.110
0.10
0.073
0.088
0.088
0.088
0.089
0.095
0.103
0.094
0.103
0.080
0.070
0.069
0.069
0.069
0.069
0.076
0.062
DEW POINT
CD
IN
121.6
122.5
118.7
121.5
121.3
121.0
120.5
122.5
120.0
123.4
123.0
118.5
121.5
123.8
123.8
123.0
123.0
120.0
122,5
OUT
130.0
126.5
116.5
122.5
122.5
122.6
122.8
125.0
127.7
124.6
127.6
119.5
115.5
115.0
115.0
114.7
114.7
118.0
111.5
DRY GAS
FLOW RATE
(Ib/nln)
103.02
103.33
54.64
78.56
80.90
81.83
56.17
101.91
102.93
54.96
55.01
94.16
78.77
47.81
55.28
32.17
102.66
77.78
74.45
LIOBID FLOW
RATE
Ub/min)
2982.4
829.4
2735.6
1689.2
1689.2
1689.2
2982.4
2903.5
2982.4
2982.4
723.6
2663.0
2783.2
2439.6
3152.6
1846.9
2293.6
2663.0
755.9
GAS VELOCITY
(ft/.ec)
42.06
42.76
21.16
31.41
32.04
32.57
21.60
42.23
42.23
21.96
22.17
38.51
31.62
18.67
21.75
12.43
42.95
31.83
68.05
LIOUID/GAS RATIO
(gal/lOOOft3)
170.53
46.65
310.93
131.08
128.48
126.38
332.02
169.85
169.85
326.57
78.47
170.80
217.45
322.73
358.05
366.89
131.92
2.06.67
61.75
(lb/lb)
26.66
7.38
46.44
19.82
19.24
19.04
49.03
26.19
26.80
49.78
12.08
26.26
32.56
46.73
52.22
52.72
20.51
31.6^
9.33
ABSO PTTON PARAMETERS
entering SO;
:oncentratler
(ppm)
830
830
720
780
780
820
780
810
800
855
855
922
965
1015
1015
1015
1015
1000
965
Leaving SOj
^oncencratloi
Coom)
420
260
30
120
110
122
195
185
112
82.5
225
204
155
57
70
38
180
144
286
SO- Removal
Efficiency
(I)
49.4
68.7
95.8
84.6
85.9
85.1
75.0
77.2
86.0
90.4
73.7
79.4
. 83.9
94.4
93.1
96.3
82.3
85.6
70.4

-------
             Table D-3 (cont) .  DATA CONSOLIDATION—SERIES F-XX—KEY WEST FREDONIA VALLEY LIMESTONE
Test
Number
F38
F39
F40
F41
F42
KAS HlftlDITY
(Ib/lb dry air)
IN
0.088
0.086
-
0.086
0.086
OUT
0.058
0.054
-
0.070
0.101
DEW POINT
CF)
IN
122.3
121.7
-
121.8
121.5
OUT
109.0
107.0
-
115.3
126.8
DRY CAS
FLOW RATE
(Ib/mln)
53.43
39.49
-
93.33
80.05
tlOUID FLOW
RATE
(Ib/rain)
953.5
919.2
859.0
438.1
1436.7
CAS VELOCITY
(ftysec)
47.56
34.34
30.66
85.96
33.74
LIOUID/CAS RATIO
(gal /1000ft3)
111.45
149.23
155.76
28.33
103.77
(lb/lb)
16.40
21.43
22.95
4.34
16.53
ABSORPTION PARAMETERS
Entering SO;
:oncentratici
(DPHI)
965
975
975
965
970
Leaving SOj
Concentrator
(ppm)
183
116
94.5
415
285
SO- Reaoval
Efficiency
(2)
81.0
88.1
90.3
57.0
70.6
00

-------
              Table D-4.   DATA CONSOLIDATION— SERIES HL-XX, D-XX, PC-XX, AND X-XX—KEY WEST LIME,




                              DOLOMITE, PRECIPITATED CaCOg, AND RECYCLED LIMESTONE
Test
Number
HL1
HL2
HL3
HL4
Dl
D2
D3
D4
PCI
PC2
PC3
PC4
XI
CAS HIWIDITY
(lb/lb dry air)
IN
0.080
0.088
0.085
0.087
0.089
0.089
0.085
0.085
0.085
0.100
0.083
-
0.088
OUT
0.096
0.077
0.077
0.089
0.094
0.086
0.081
0.094
0.094
0.092
0.087
- .
0.097
DEM POINT
CF)
IN
119.5
122.5
121.6
122.1
123.0
122.7
121.2
121.2
121.5
126.5
120.4
• -" '
122.5
OUT
125.5
118.1
118.1
123.0
124.5
121.7
120.0
124.6
124.5
124.0
122.3
-
125.5
DRY GAS
FLOW RATE
(Ib/nln)
103.02
55.91
55.50
102.36
101.11
101.11
54.79
54.79
101.55
55.42
54.96
-
78.56
LinUID FLOW
RATE
(Ib/nln)
807.5
2903.5
704.4
2903.5
807.5
2903.5
704.4
2903.5
807.48
2903.5
704.4
VOID
1689.2
GAS VELOCITY
(ft/aec)
41.59
21.81
21.75
41.91
42.44
42.44
22.03
'22.03
42.44
21.90
21.92
-
31.62
LIOUID/GAS RATIO
(Ral/lOOOft3)
47.96
328.79
80.00
171.14
47.00
169.00
79.00
325.63
47.00
327.52
79.38
-
130.20
(lb/lb)
7.26
47.73
11.70
25.92
7.33
26.37
11.85
48.84
7.33
48.47
11.83
-
19.76
ABSORPTION PARAME
Entering SO-
:oncentratlor
(own)
750
216
775
815
860
860
880
830
860
825
830
-
860
Leaving S02
Concencratloi
(oon)
102
40
51
20
580
420
425
400
280
120
150
-
258
•ERS
SO. Removal
Efficiency
a>
86.4
94.9
93.4
97.5
32.6
51.2
51.7
51.8
67.4
85.5
81.9
-
70.0
VO

-------
Table D-5.  DATA CONSOLIDATION — SERIES P-XX — PADUCAH-SHAWNEE NO. 9 LIMESTONE
Test
Number
PI
P2
P3
P4
P5
P6
P7
P8
P9
P10
Pll
P12
P13
P14
P15
P16
P17
P18
P19
CAS HliMIDITY
(Ib/lb dry air)
IN
0.071
0.074
0.072
0.048
0.059
0.051
0.061
0.077
0.059
0.061
0.054
0.054
0.060
0.051
0.051
0.051
—
- _ •
0.059
OUT
0.054
0.049
0.051
0.040
0.065
0.063
0.060
0.066
0.033
0.053
0.059
0.059
0.066
0.058
0.068
0.068
—
_
0.068
DEW POINT
CF)
IN
115.9
117.0
116.0
103.0
110.0
105.1
111.0
118.3
110.0
110.8
107.0
107.0
110.5
105.5
105.5
105.5
-
_
110.0
OUT
107.0
104.5
105.0
97.5
113.0
112.0
110.5
113.1
92.0
106.3
110.0
110.0
113.5
109.5
114.0
114.0
-
-
114.0
DRY GAS
FLOW RATE
(Ib/mln)
31.78
30.256
30.17
31.30
45.08
45.42
44.78
43.97
30.17
30.12
30.46
30.46
44.82
45.15
45.22
45.22
-
-
37.66
LIOUID FLOW
RATE
Ub/min)
281.0
281.0
800.4
698.2
698.2
698.2
289.5
289.5
289.5
289.5
698.2
698.2
289.5
289.5
698.2
698.2
VOID
VOID
604.6
GAS VELOCITY
(ft/sec)
26.02
24.83
24.59
24.11
36.76
37.62
37.25
37.58
24.64
24.64
24.02
24.02
37.58
37.24
37.24,
37.24
-
-. -
31.13
LIOUID/GAS RATIO
(gal /1000ft3)
60.55
63.46
182.52
162.38
106.49
104.06
43.59
43.20
65.89
65.89
163.02
163.02
43.20
43.59
105.13
105.13
-
<-
108.90
(lb/lb)
8.26
8.65
24.75
21.29
14.63
14.63
6.09
6.11
9.06
9.06
21.75
21.75
6.09
6. Of
14.65
14. 6f
-
- -
15. 1<
ABSORPTION PARAMETERS
Entering SO;
^oncer.tratior
(prrc)
2450
2350
2350
1900
1995
1500
1500
1750
940
940
2200
2200
2230
1450
1470
1470
-
-
1900
Leaving S02
Concer.tratioi
(ran)
1175
1080
700
310
765
490
980
1090
440
505
575
660
1420
840
550
475
-
-
730
SO- Removal
EEficiencv
(X) 	
52.0
54.0
70.2
83.7
61.7
67.3
34.7
37.7
53.2
46.3
73.9
70.0
36.3
42.1
62.6
67.7
-
-
61.6

-------
Table D-5 (cont).  DATA CONSOLIDATION-.-SERIES P-XX — PADUCAH-SHAWNEE NO. 9 LIMESTONE
Test
Number
P20
P21
P22
P23
P24
P25
P26
P27
P28
P29
P30
P31
P32
P33
P34
P35
GAS HIiMIDITT
(Ib/lb dry air)
IN
0.061
0.060
0.053
0.056
0.060
0.'055
0.054
0.059
0.059
0.054
0.059
0.062
0.069
0.077
0.076
0.069
OUT
0.068
0.068
0.058
0.056
0.056
0.061
0.061
0.065
0.065
0.063
0.068
0.065
0.072
0.074
0.072
0.072
DEW POINT
CF)
IN
111.0
110.5
106.6
108.1
108.0
107.0
107.0
109.7
109.7
107.0
110.0
111.5
114.5
118.1
117.7
114.6
OUT
114.0
114.0
109.3
107.9
107.9
110.7
110.7
113.0
113.0
111.7
114.4
113.0
116.0
117.0
116.0
116.3
DRY GAS
FLOW RATE
(Ib/oln)
37.59
37.56
38.02
37.92
37.77
37.84
37.99
37.95
37.95
37.66
37.59
37.48
37.60
44.33
43.30
37.74
LIOUID FLOW
RATE
(Ib/roln)
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
604.6
289.5
698.2
596.3
CAS VELOCITY
(ft/sec)
31.27
31.42
30.94
31.42
31.61
31.61
31.61
31.27
31.27
30.84
30.80
30.80
31.23
37.91
38.10
31.27


LIOUID/GAS RATIO
(gal/1000ft3)
108.40
107.90
109.57
107.90
107.25
107.25
107.25
108.40
108.40
109.91
110.08
110.08
108.56
42.82
102.76
108.40
(lb/lb)
15.16
15.19
15.10
15.10
15.10
15.10
15.10
15.04
15.04
15.04
15.19
15.19
15.01
6.06
14.65
14.78
ABSORPTION PARAMETERS
Entering SO*
;oncentratiot
(ppm)
2050
2000
2025
1875
2025
1650
1850
1650
1700
1850
1730
1700
2100
2300
2420
2620
Leaving S02
;oncentratlor
fpotn)
800
750
700
700
775
525
715
500
550
775
560
590
760
1125
975
1040
SO. Removal
Efficiency
(Z)
61.0
62.5
65.4
. 62.7
64.2
63.2
61.4
69.7
67.6
58.1
67.6
65.3
63.8
51.1
59.7
60.3

-------
Table D-6.   DATA CONSOLIDATION — SERIES PA-XX, AND PS-XX ~ PADUCAH-SHAWNEE





             NO. 9 ARAGONITE AND SIMULATED KEY WEST LIMESTONE
Test
Number
PAl
PA2
PA3
PA4
PAS
PA6
PA7
PAS
PS1
PS2
PS 3
CAS Hl'MTDITY
(Ib/lb dry air)
IN
0.060
0.060
0.056
0.056
0.087
0.086
0.088
0.086
0.058
0.058
0.059
OUT
0.081
0.078
0.051
0.049
0.064
0.069
0.064
0.064
0.063
0.068
0.057
DEW POINT
<*F)
IN
110.5
110.5
108.0
108.0
122.2
122.0
122.3
121.7
109.0
109.0
110.0
OUT
120.0
118.7
105.0
103.7
112.5
115.0
112.0
112.3
112.0
114.2
108.3
DRY GAS
FLOW RATE
(Ib/mln)
45.04
45.04
28.15
28.15
36.84
36.87
36.94
37.01
38.21
38.21
38.17
LIOUID FLOW
RATE
(Ib/mln)
289.5
289.5
698.2
698.2
698.2
698.2
698.2
698.2
604.6
604.6
604.6
CAS VELOCITY
(ft/sec)
37.05
37.05
21.25
21.25
31.75
31.89
30.56
30.84
30.37
30.37
30.37
LIOUID/CAS RATIO
(Ba I/ 1000ft3)
43.81
43.81
184.27
184.27
123.31
122.75
128.13
126.93
111.64
111.64
111.64
Ub/lb)
6.06
6.06
23.49
23.49
17.44
17.44
17.37
17.37
14.96
14.96
14.96
ABSORPTION PARAMETERS
Sneering SO;
:oneentraflot
(pen)
2050
2050
1600
1200
2500
2700
2350
2100
2000
2020
2020
Leaving S02
Concentratloi
(DP:H)
780
700
160
84
820
860
600
590
450
525
525
SO. Removal
Efficiency
«>
62.0
65.9
90.0
93.0
67.2
68.1
74.5
71.9
77.5
74.0
74.0

-------
                   Table D-7.   DATA CONSOLIDATION —SERIES IPA-XX AND IP-XX— PADUCAH-SHAWNEE




                                         NO.  10 ARAGONITE AND LIMESTONE INJECTION
Test
Number
IPAl
IPA2
IPA3
IPA4
IP1
IP 2
IP3
IP 4
IPS
IP6
IP 7
IPS
IP 9
IP10
IP 11
IP12
IP13
CAS HIiMIDITT
(Ib/lb dry air)
IN
0.052
0.055
0.053
0.049
0.051
0.063
0.061
0.051
0.053
0.052
0.052
0.075
0.063
0.063
0.085
0.087
0.085
OUT
0.061
0.067
0.064
0.052
0.054
0.051
0.51
0.059
0.057
0.055
0.051
0.089
0.068
0.068
0.070
0.071
0.071
DEW POINT
CF>
IN
106.0
108.0
106.5
104.0
105.0
111.7
111.0
105.3
106.2
106.0
105.7
117.5
111.9
111.9
121.3
122.2
121.3
OUT
111.0
113.8
112.3
105.5
106.8
105.3
105.0
109.5
108.7
107.8
105.5
122.7
114.3
114.3
115.4
115.5.
115.5
DRY GAS
FLOW RATE
(Ib/oin)
45.01
45.40
30.64
30.90
37.80
38.03
37.96
37.58
38.10
38.50
37.48
36.75
44.77
44.77
43.57
43 = 71
26.12
LinuiD FLOW
RATE
(Ib/oin)
289.5
698.2
698.2
698.2
698.2
698.2
698.2
698.2
698.2
698.2
698.2
698.2
800.4
800.4
800.4
289.5
272.5
GAS VELOCITY
(ft/sec)
39.25
37.62
25.73
24.92
31.18
30.37
30.56
31.42
31.18
30.51
31.23
33.42
38.05
38.05
38.58
38.48
22.30


LIODID/CAS RATIO
(gal/1000ft3)
41.36
104.06
152.13
157.09
125.57
128.93
128.13
124.62
125.57
128.33
125.38
117.14
117.94
117.94
116.34
42.18
68.52
(lb/lb)
6.11
14.58
21.64
21.54
17.57
17.27
17.34
17.68
17.37
17.24
17.71
17.68
16.82
16.82
16.93
6.09,
9.62
ABSORPTION PARAMETERS
Altering SO;
;oncentratlot
(Don)
750
415
415
510
420
700
975
1080
420
495
250
2000
2625
2625
2450
1520
1250
Leaving S02
;oncentratioi
(oom)
180
31.5
73
30
37.5
46.5
262
285
40
75
30
600
1050
1050
460
540
350
SO- Removal
EEflciency
(X)
76.0
92.4"
81.2
94.1
91.1
93.4
73.1
73.6
90.5
84.8
88.0
70.0
60.0
60.0
81.2
64.5
72.0
u>

-------
                            APPENDIX E
                         UNITS OF MEASURE
    It ±e EPA policy to express all measurements in metric units.
If undue costs or difficulty in clarity result from implementing
this practice, British units may be employed and a conversion table
provided.  Such is the case with this document, the first draft
of which was submitted before promulgation of the EPA policy.
                      British
Metric
Length
Power
Volume
Pressure
Density
Mass
Temperature
1 inch (in.)
1 foot (ft)
1 horsepower (HP)
1 cubic foot (cf)(ft3)
1 gallon (gal)
1 inch of water (in.H-O)
1 inch of mercury (in.Hg)
1 pound per square inch
(psi)
1 pound per cubic foot
(Ib ft'3)
1 ounce (oz)
1 pound (Ib)
1 ton (T)
1 grain (gr)
1° Fahrenheit (°F)
1° Rankine (°R)
2.54 centimeters
0.3048 meters
746 watts
28.3161 liters
3.7853 liters
2419 dynes per square
centimeter
338639 "
689476 " "
0.0160 grams per
cubic centimeter
28.3495 grams
453.5923 grains
907.1846 kilograms
0.0648 grams
(5/9)(°F-32) °Centigrade
(5/9)(°R) °Kelvin
                                144

-------
TECHNICAL REPORT DATA
(Please read Iiutraetioni on the reverse before completing)
'OUT NO. 2.
'A-650/2-74-077
LE AND SUBTITLE
ne/Limestone Scrubbing in a Pilot Dus traxtor- -
Cey West
THORis»Burke A Bell; Terrence A. LiPuma;
J. M. Craig, Ph.D.; and J. K. Allison
1FORMING OR«ANIZATION NAME AND ADDRESS
];ineering Science, Inc.
)3 Westpark Drive
Lean, Virginia 22101
ONSOMINO AGENCY NAME AND ADDRESS
A, Office of Research and Development
RC-RTP, Control Systems Laboratory
search Triangle Park, NC 27711
3. RECIPIENT'S ACCESSION NO.
i. REPORT DATE
September 1974
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO.
10. PROGRAM ELEMENT NO.
LAB013; ROAP 21AQO-001
11. CONTRACT/GRANT NO.
CPA 70-61
13. TYPE OF REPORT AND PERIOD COVERED
Final; January-July 1971
14. SPONSORING AGENCY CODE
IPPLEMENTARY NOTES
          report gives results of a 7-month series of nearly 200 tests of the Dust-
tor limestone wet scrubbing system in 1971, both in Key West, Florida, and at
Vs Shawnee Plant in Kentucky. At Key West, No. 6 fuel oil containing 1-2.2% sulfur
 burned; at Shawnee, 2-4% sulfur pulverized coal was burned.  The tests included
tematic variation of stoichiometry, reactant particle size, slurry concentration,
ssure drop, and gas flow rate. Reactants tested included coral marl,  Fredonia
ley limestone,  dolomite, lime, aragonite, and precipitated calcium carbonate.
its also included evaluation of spent reactant material, boiler injection of dry arag-
:e,  addition of an inhibitor and catalyst, and effects on particulate and NOx removal.
?r installing an annular fresh  water spray ring to reduce scale formation, the
traxtor worked satisfactorily. SO2 removal efficiencies varied up to 90-plus %,
ending on the reactant used. Absorption efficiency increased significantly with
-eased pressure drop, decreased gas flow rate, increased stoichiometric ratio,
 increased liquid-to-gas ratio. Other variables and an inhibitor and catalyst had no
lificant effect.  NOx removal  in the scrubber was negligible.  Particulate  removal
;he total pilot plant system was excellent.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Pollution Limestone Desulfurizatior
ubbers Marls Nitrogen
thing Dolomite Oxides
1 Oil Aragonite
1 Calcium
cium Oxides Carbonates
3TRIBUTION STATEMENT
imited
b.lDENTIFIERS/OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Dus traxtor
Particulates
19. SECURITY CLASS (This Report)
Unclassified
2O. SECURITY CLASS (This page)
Unclassified
c. COSATI Field/Group
13B , 07D
07A
13H
21D, 11H
07B
21. NO. OF PAGES
157
22. PRICE
arm J21O-1 (»-7J)
                                      145

-------