EPA
ill Protection
Officf of Industrial Environmental Research
Research and Laboratory
Development Research Triangle Park, North Carolina 27711
EPA-b'OU
MAGNESIA SCRUBBING
APPLIED TO A COAL-
FIRED POWER PLANT
Interagency
Energy-Environment
Research and Development
Program Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S.
Environmental Protection Agency, have been grouped into seven series.
These seven broad categories were established to facilitate further
development and application of environmental technology. Elimination
of traditional grouping was consciously planned to foster technology
transfer and a maximum interface in related fields. The seven series
are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from
the effort funded under the 17-agency Federal Energy/Environment
Research and Development Program. These studies relate to EPA's
mission to protect the public health and welfare from adverse effects
of pollutants associated with energy systems. The goal of the Program
is to assure the rapid development of domestic energy supplies in an
environmentally—compatible manner by providing the necessary
environmental data and control technology. Investigations include
analyses of the transport of energy-related pollutants and their health
and ecological effects; assessments of, and development of, control
technologies for energy systems; and integrated assessments of a wide
range of energy-related environmental issues.
REVIEW NOTICE
This report has been reviewed by the participating Federal
Agencies, and approved for publication. Approval does riot
signify that the contents necessarily reflect the views and
policies of the Government, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use.
This document is available to the public through the National Technical
Information Service, Springfield, Virginia 22161.
-------
EPA-600/7-77-018
March 1977
MAGNESIA SCRUBBING
APPLIED TO
A COAL-FIRED POWER PLANT
by
George Koehler
Chemico Air Pollution Control Company
One Perm Plaza
New York, New York 10001
Contract No. 68-02-1870
Program Element No. EHB528
EPA Project Officer: C.J. Chatlynne
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
THIS PAGE INTENTIONALLY LEFT BLANK
-------
ABSTRACT
A full-size demonstration of the magnesia wet-scrubbing
system for Flue Gas Desulfurization (FGD) was conducted on
a coal fired utility boiler. The system was designed to de-
sulfurize one-half the flue gas from a 190 MW rated capacity
generating unit firing 3.5% sulfur coal. The FGD installa-
tion was equipped with a first-stage wet scrubber for parti-
cle emissions control followed by the magnesia unit.
Twenty-eight hundred hours of operation were logged at
the generating station, and the FGD system's ability to re-
move 90% of the inlet S02 was shown. The particle control
capability of the unit was also demonstrated by reducing
particle emissions to less than 0.01 gr/SCFwhen the unit
was operated in series with an electrostatic precipitator.
In a test program using only the wet scrubbing unit for par-
ticle emissions control a collection efficiency of 99.6%
was achieved.
Magnesia was regenerated and recycled successfully.
The SO- produced during regeneration was used to manu-
facture commercial grade sulfuric acid which was marketed.
Correlations were developed to determine SO, removal for
varying boiler loads and fuel sulfur content, and to control
regeneration of acceptable alkali. Several other studies of
the process technology and process chemistry were undertaken
as part of the work.
111
-------
TABLE OF CONTENTS
PAGE
1.0 Summary 1
1.1 Operational Effect of Design 5
Variation From the Oil Fired Case
1.2 Effect Of Process Chemistry 7
Variations on System Performance
2.0 Introduction & Background 9
3.0 Process Description 16
3.0.1 Facility Description 19
3.0.2 Pollution Abatement System 21
Description
3.1 Detailed Process Description - FGD 25
System
3.1.1 Particle Removal 25
3.1.2 SO- Removal 31
3.2 Magnesia Regeneration System 34
3.3 Acid Plant 40
3.3.1 Acid Plant Modifications 41'
3.3.2 Acid Production 44
4.0 Pollution Abatement System Performance 48
4.1 Development Test Program 50
4.1.1 October 1974 Details 58
4.1.2 November 1974 Details 58
4.1.3 December 1974 Details 60
4.1.4 January 1975 Details 61
4.2 Utilities Consumption 62
4.3 MgO Consumption 62
4.3.1 MgO Losses During Preliminary 65
Operations
4.4 MgO Regeneration - Inventory 66
4.5 Post Operation FGD System Inspection 69
4.5.1 1st Stage Piping 70
4.5.2 2nd Stage Piping 74
4.5.3 Corrosion of Stainless Steel 74
5.0 Performance Test Results 76
5.1 Particulate Emission 76
5.2 SO Emissions 78
5.3 Particle Size Measurements 82
5.4 Coal Analysis 82
IV
-------
TABLE OF CONTENTS
PAGE
6.0 Process Chemistry 85
6.1 1st Stage Operation 85
6.2 Process Chemistry 88
6.2.1 Absorber Reactions 89
6.2.2 Formation of Oxysulfates 90
6.3 Magnesium Sulfite Chemistry 91
6.3.1 Conditions Governing the 91
Formation of Magnesium
Sulfite Hydrates
6.3.2 Dehydration of Magnesium Sulfite 93
Hydrates
6.3.3 Mass Spectroscopy of the Two 95
Hydrates of Magnesium Sulfite
vs. Temperature
6.4 Pulverization 96
7.0 Correlation of Process Data 99
7.1 SO- Removal 101
7.1.1 First Stage S02 Absorption 101
7.1.2 Effect of Pulverization of 102
Regenerated MgO
7.1.3 Reduction in Activity of 106
Magnesia
7.2 Centrifuge Operation 110
7.3 Dryer Operation 113
7.4 Calciner Feed Rate 113
8.0 Data 120
8.1 Monthly Average Operating Conditions 120
8.2 Data Listings 120
9.0 List of Publications 134
10.0 Conversion From English To Metric Units 137
Appendix I - Post Operation Inspection Report 138
Appendix 2 - Thermal Dehydration of Magnesium 153
Sulfite Hydrates - Using Thermo-
Analytical Techniques
Appendix 3 - Mass Spectrometric Study of the 201
Thermal Degradation of MgSO-j.3H~0
& MgS03.6H20
-------
TABLE OF CONTENTS
FIGURES
PAGE
1 Two Stage Scrubber/Absorber Vessel 17
2 FGD Gas Flow Schematic - Dickerson No. 3 - 18
PEPCO
3 The Dickerson Regenerative Wet Scrubbing 22
System - Schematic Process Flow Diagram
4 PEPCO Regenerative Wet Scrubbing System - 26
Process Flow Diagram
5 Magnesia Regeneration System 35
6 Sulfuric Acid Plant - Process Flow Diagram 43
7 Tail Gas Scrubbing System - Process Flow 46
Diagram
8 Effect of Pulverization on S02 Efficiency 97
9 Process Data Flow Diagram 100
10 SO, Removal Efficiency vs. Time 104
11 Effect Of MgO Particle Size on S02 Removal 107
Efficiency
12 Observed vs. Predicted S02 Removal Efficiency 108
13 Effect Of Centrifuge Feed Rate on Separa- 111
tional Efficiency
14 Centrifuge Operation - Ib/min Solids 112
Separated vs. Feed Rate
15 Comparison of PEPCO and Boston Edison Dryer 114
Operation
16 Dryer Operation - Outlet Gas temperature vs. 115
% Dry Solids
17 Comparison of MgSO3 Oxidation In Dryer 116
18 Calciner Operation - Effect Of Feed Rate On 117
MgS03 In Product
VI
-------
TABLE OF CONTENTS
PAGE
TABLES
1 Pertinent Data On Boiler Design, Operation 20
and Atmospheric Emissions
2 PEPCO Regenerative Wet Scrubbing System - 27
Stream Properties and Composition
3 Wet Scrubbing System - Equipment List 28
4 Magnesia Regeneration System - Stream 36
Properties and Composition
5 Calciner Product - Dry Screen Analysis 39
6 Sulfuric Acid Unit Feed Gas Composition 42
7 Tail Gas Scrubbing System - Stream 47
Properties and Composition
8 Performance Test Designations 53
9 Summary Of Operations FGD System - 55
Performance Test Phase
10 Summary Of Operations Calciner System - 57
Performance Test Phase
11 Utilities Consumption - FGD System 63
12 MgO Makeup - FGD System 64
13 Inventory Summary By Week - FGD System 67
14 Inventory Summary By Week - Regeneration 68
System
15 Operating History - FGD System 71
16 Particle Emissions Test Results - FGD System 77
17 Estimate Of Electrostatic Precipitator 79
Efficiency
18 SO Emissions Test Results - FGD System 80
19 Removal Efficiency For Particle Size Ranges 83
20 Coal Analyses - Composite To Bunker #3 84
21 Chemical Composition At Various Streams In 86
The First Stage - Fly Ash Removal System
22 Properties Of Different Fraction Of PR-289 98
23 S0~ Removal In First Stage 103
24 Wet Screen Analysis of MgO Belt Samples 105
25 Manufacturer's Analysis of Magnesia 109
Shipped To Dickerson Station
26 Size Distribution Of Dryer Products 119
27 Comparison Of Boston and PEPCO size Range 119
Averages
28 Operating Conditions For The FGD System - 121
Monthly Data Averages
29 Stream Compositions For The FGD System - 122
Monthly Data Averages
30 Operating Conditions For The Regeneration 123
Unit - Monthly Data Averages
31 Stream Compositions For The Regeneration 124
Unit - Monthly Data Averages
32 Run Dates and Data Listings 125
vn
-------
TABLE OF CONTENTS
PHOTOGRAPHS
3-2 Centrifuge Case
3-5 Centrifuge Plow - Failed
3-11 Centrifuge Plow - Normal
4-1 Dryer
4-3 Bearing Support - Dryer Feed Screw
4-6 Dryer Burner Block Refractory
10-7 2nd Stage Venturi Throat
10-10 2nd Stage Venturi Throat
PAGE
151
151
151
151
151
151
152
152
Vlll
-------
1.0 EXECUTIVE SUMMARY
The*Chemico-Basic magnesia slurry SO- recovery process
for control of sulfur oxides emissions from fossil fuel
fired utility boilers has been demonstrated on an oil fired
boiler (The Magnesia Scrubbing Process As Applied To An Oil
Fired Boiler, EPA 600/2-75-057). This process, with some
modifications developed during early operations in New England,
was next installed on a coal fired boiler at Potomac Electric
Power Company's Dickerson Generating Station as a system to
control both particle and sulfur dioxide emissions from the
plant. The regeneration and SO, recovery steps of the process
continued to be carried out at the Environmental Protection
Agency's calcination facility at Essex Chemicals' Rumford Acid
Plant in Rhode Island, which had originally been constructed
for the New England S02 Control Project.
Major distinctions between these two FGD units are divided
into process design variation and process chemistry differences.
In the former category the principal variations between the
design of the previously reported oil fired application and
this coal fired boiler system were:
A) The addition of a first stage venturi scrubber
for particle control only (no alkali is added
to this stage). Flue gas exiting this first
stage is lowered in temperature to near the
adiabatic saturation point through the normal
transport mechanisms.
B) Magnesia alkali added to the second stage for
SO- control is slurried with centrate returned
from the centrifuge instead of process water.
Since gas entering the second stage is saturated
with water vapor little or no evaporation takes
*Wherever the term '?Chemico" is referred to in this
report, the term "Chemico Air Pollution Control
Company" shall be substituted therefore.
-------
place; thus it is possible to eliminate
fresh water inputs to this stage.
C) The use of a co-current, rather than a
counter current dryer to dry the centri-
fuge cake. This design eliminates at least
one problem encountered with the counter-
current design used in oil fired application.
Dust exiting with the dryer off-gas can be
separated and fed directly to the dryer pro-
duct discharge without adding the pneumatic
conveying equipment necessary with counter-
current units.
D) The use of "wet" fan after the two-stage
scrubber unit to provide the induced draft
for the venturies instead of a hot, forced
draft type. An important element of the pro-
gram was the taking of flue gas containing coal
fly ash directly from the boiler, use of the
"wet" fan eliminated the erosion problem associated
with dry fans handling dirty flue gas and took
advantage of the lowered temperature and volume
handled after the scrubbers.
The major difference in process chemistry between the
two plants was the production of the magnesium sulfite hexa-
hydrate in this coal fired boiler application rather than
the magnesium sulfite trihydrate produced during operations
at the oil fired generating plant. Processing of the hexa-
hydrate proved similar in many cases, to the experiences
encountered when processing the trihydrate.
Another process chemistry difference was the evidence
that magnesium oxysulfates were formed, sometimes in large
quantities, during the slaking of MgO with centrate. This
resulted in a subsequent reduction in SC>2 removal efficiency.
- 2 -
-------
Because of an overlap with the operations of the oil
fired station the regeneration plant was not available to
the Dickerson facility when the FGD system was ready for
its inaugural operation. The lack of a regeneration faci-
lity plus problems discovered during the start up period
limited the initial PEPCO operations to less than 1000 hours
in the nine months period between September 1973 and June
1974.
On July 1, 1974 when the Rumford facility became avail-
able, an operating and development program was initiated
at PEPCO's Dickerson Station. In the next six months another
1790 hours of operation were achieved despite a number of
new problems. All major objectives were completed during
the period:
1) Operations were initiated using regenerated
MgO.
2) Particle removal efficiency of 99%+ was
verified.
3) S02 removal efficiency was measured, improved
and verified at 90%.
4) Operating conditions for continuous running
were established, with emphasis on MgO slurry
mixing, control and makeup.
5) MgO loss points were discovered and corrected.
6) An operations evaluation program was established.
7) Material balance measurements were made.
8) The SO2 removal correlation developed in the
previous study was improved to include the
effect of regenerated MgO properties.
9) Preliminary correlations were developed to
define operating variable influence on dryer
and centrifuge system performance.
- 3 -
-------
10) Laboratory investigations of fundamental
system properties were continued.
Some important objectives were only partially attained
because of a number of shut-downs in the latter period which
limited the time available for operation. Thus, a long dura-
tion run was not realized. The longest uninterrupted runs
were a twelve day continuous period in October and another
of eleven days in December. Despite these limited operations
two recycles of the magnesium oxide inventory were achieved
during the program.
While a number of causes contributed to the interrup-
tions the major "shut-down" problems was deterioration of
the piping in the first and second stage recycle loops of
the FGD plant. In the second stage extensive corrosion had
occurred in the unprotected steel pipe at some time during
the preliminary operations. In the first stage recycle pipe,
which was rubber lined, a number of failures occurred which
were attributed to the supply of off-specification pipe
and fittings during construction.
After a piping replacement program and other repairs
to the FGD plant had been completed in preparation for a
demonstration run, the project was terminated when a rou-
tine overhaul of the power plant's generator system in
January 1975 revealed a defect in the turbine case. The
resultant five month outage required for repair of the tur-
bine extended past the close-out date of the E.P.A. con-
tract for access to the acid plant and operation of the re-
generation facility, thus forcing the end of the project.
Despite the limited operations and lack of extended
operating data a number of conclusions concerning the oper-
ability at the plant can be gathered from the information
that was obtained.
- 4 -
-------
1.1 OPERATIONAL EFFECT OF DESIGN VARIATION FROM THE
OIL FIRED CASE
A) Particle removal efficiency for both the case
in which gas was first cleaned in an electro-
static precipitator and the case where dirty
flue gas was taken directly from the boiler
exceeded process guarantees. With the elec-
trostatic precipitator in the circuit an aver-
age outlet loading of 0.002 gr/ACF was attained
with the unit handling its full rated capacity.
When treating dirty flue gas an average removal
efficiency of 99.6% was obtained at full load
operation. The outlet loading from the scrub-
ber system for this case was 0.011 gr/ACF.
B) Slurrying magnesia feed with centrate posed some
problems. It appears that part of the MgO feed
is complexed as a magnesium oxysulfate. This
resulted in the formation of lumps and gels which
plugged feed lines in the early operations. This
problem was solved by:
1) Providing an intensive mixer between the
MgO feeder and slurry tank.
2) Slurrying the MgO feed with the entire
recycle centrate stream to reduce concen-
tration.
3) Adding a steam sparger to the slurry tank
to heat the slurry to 180°F.
These fixes diminished the problems to a level
that allowed more routine system operation.
A reduction in S02 removal efficiency was also
noted over a period of time during which virgin
MgO was the major part of the feed. Continued
- 5 -
-------
operations showed a recovery in efficiency
when using regenerated MgO.
Performance tests for S02 removal run on
recycled MgO gave an average 90% and 93.9%
S02 removal efficiency with the scrubber
operating at full rated capacity with S02
inlet concentrates of 771 PPM and 1418 PPM,
respectively.
An explanation of the reduced efficiency when
using virgin MgO is the more rapid formation of
the complex, magnesium oxysulfate, with this
more reactive material.
C) The co-current dryer system used in the PEPCO
operation had a higher efficiency than the unit
used in the oil-fired case. Extensive modifica-
tions of the counter-current dryer/ to prevent
material sticking to the walls,included elimina-
tion of most of the internal lifters. This would
account for the higher co-current efficiency.
A number of instances were encountered during the
PEPCO operations in which centrifuge cake feed
was reported as sticking to the dryer walls.
This was corrected several times "on the run" by
reducing load. There were some cases where large
deposits in the dryer resulted in a temporary shut-
down.
It should be noted that the dryer had been designed
to accept a set of chains, but these had not been
installed pending the outcome of the development
test work. Because of the unanticipated shut-down
of the entire operation this potential was not
demonstrated.
- 6 -
-------
Dryer product from the PEPCO operation also
proved more difficult to handle in this re-
generation step, and resulted in a forced cut-
back in calciner feed rate to one half that
obtained on the oil fired application. The
reduced feed rate was correlated with a higher
percentage of fines in the PEPCO dryer product,
with only 22% as + 25 mesh in the PEPCO dryer
product versus 58% in the same mesh size for
the Boston Edison dryer product.
D) The wet fan performance and service was
excellent. There were no signs of deterioration,
wear or corrosion after all operations comple-
tion in either E.S.P. treated or untreated flue
gas modes.
1.2 EFFECT OF PROCESS CHEMISTRY VARIATIONS ON SYSTEM
PERFORMANCE
A) Production of magnesium sulfite hexahydrate
rather than magnesium sulfite trihydrate resulted
in little or no relief from the problems of "wet"
centrifuge cake encountered in the oil fired appli-
cation.
In normal operations, after attaining equilibrium,
the centrifuge cake contained a mix of 82% as
hexahydrate, 18% as trihydrate for its magnesium
sulfite component. Materials handling problems
were solved, as in the oil fired case, by design-
ing the system to take any consistency of the cen-
trifuge cake.
Improvements were made in redesigning the centri-
fuge discharge chute with a vertical wall on the
solids discharge side of the machine and by lining
-------
the chute with plastic to overcome the sticking
tendency of the cake.
B) Processing problems resulting from the formation
of the complex S04Mg MgO H20 have been described
in A-2 above. Laboratory experiments indicate
that this material is formed more readily when
slaking a highly reactive magnesia (as virgin
MgO grade used in this operation). Performance
improves when the feed is composed of a high per-
centage of recycled MgO. It appears that the
complex formation occurs during the step
M9°(s) + H20 (1) >Mg (OH) 2 (s)
as subsequent laboratory tests showed no oxy-
sulfate formation when the starting material was
magnesium hydroxide.
C) Other Findings
Explanations were developed for the contradictions
in the findings of various researchers on the
thermographic analysis of the magnsium sulfite tri-
and hexa-hydrates. It was shown that the dehydra-
tion can be characterized as one-step in an open,
non-equilibrium process whereas in a closed, equi-
librium process it proceeds at the expected two
step dehydration. No evidence was found for the
existance of any lower hydrate.
This work led to a simple and accurate method of
analysis for the ratio of the two hydrates using
thermogravemetric analysis (T.G.A.) techniques.
- 8 -
-------
2.0 INTRODUCTION AND BACKGROUND
The Chemico-Basic magnesia slurry process for S02
control had been developed in laboratory and pilot plant
studies over a number of years. The first prototype plant
employing this system was intalled on an oil fired boiler
at Boston Edison Co.'s Mystic Station on a program (New
England SO- Control Project) that spanned four years from
the signing of the contract to the completion of the test
work in July 1974.
Concurrently with the development of the "New England
Project" Chemico pursued other potential applications for
this efficient flue gas desulfurization method.
Applications were investigated that would further
demonstrate the broad range of its use, particularly the
advantages in those areas where a regenerable, non sludge-
producing sulfur oxides control method was mandated.
Simultaneously, forward thinking utilities, such as
Potomac Electric Power Company, were also investigating a
number of process schemes that would provide a means of con-
trol of their sulfur oxides emissions in order to satisfy the
new limitations being imposed by municipal, state and Federal
governments. A principal criterion in selection of the pro-
cess for control of the recognized pollutants was the econo-
mics associated with the alternate of a low sulfur content
fuel supply.
In late 1970 Chemico submitted a proposal to PEPCO
for a prototype demonstration plant using the Chemico-
Basic process. The prototype location was set for PEPCO's
Dickerson generating station site approximately 35 miles
northwest of Washington, D.C. in Montgomery County, Maryland.
- 9 -
-------
This site consists of 1000 acres in a rural area (but with
suburban expansion from Washington developing within a few
miles of the plant) bordered on the west by the Potomac
River and the Chesapeake and Ohio Canal National Historic
Park. The site had also been planned by PEPCO as the loca-
tion of a 1700 MW expansion of generating capacity which
was to be coal fired in accordance with company considera-
tions for a balance of fuel types. PEPCO had successfully
developed methods for adequate disposal of its fly ash from
the Dickerson Station; therefore, a major consideration for
future planning was the selection of a process for sulfur
oxides control which would be acceptable at this environ-
mentally sensitive site. Regenerable processes were a major
candidate because of the complexity that would be added to
operation of the power plant by the requirement to dispose
of the larger volume of residue that would be produced in
the throw-away processes. The Chemico-Basic magnesia F.G.D.
system was eventually selected by PEPCO and engineering for
the plant to be built at Dickerson Station commenced in
July 1971. Construction of the plant started in July 1972
and was completed in August 1973.
A limited test program had been planned for the PEPCO
operation as it was hoped that the major operating problems
would have been developed and corrected in operation at the
first prototype in Boston by the time the Dickerson plant
was in operation. However, a series of events including de-
layed start up of both the Boston Edison and the PEPCO pro-
ject, unanticipated problems found on starting up the processes,
and the lack of a regeneration facility for PEPCO's early
runs resulted in a restriction on the operation of the plant.
Those operations that were conducted evidenced areas requir-
ing additional, intensive investigation and development.
- 10 -
-------
Some of these were:
1) Identification of the cause and remedy of
agglomerates in the MgO slurry system.
2) Determination of losses and loss points as
discrepancies in production and consumption
calculations pointed to leaks of large mag-
nitude in the magnesia system.
3) Determination of the extent and rate of cor-
rosion as samples obtained from rail cars of
MgSO3 shipped to Rumford showed much higher
levels of iron contamination than any experienced
in the Boston Edison Operation.
4) Establishing actual SO- and particle removal
efficiencies.
In order to provide information and solutions to these
problems, a more comprehensive development and test program
was prepared by Chemico and accepted by PEPCO in June 1974.
A major factor in planning of the new program was the
extension of the data gathering and interpretation methods
that had been used for the successful operations at Mystic
station to operations of Dickerson. This immediately made
the large data and correlation file which had been developed
during the previous program accessible for extension and com-
parison with data generated from the coal fired application.
The program was divided into a series of tasks and phases
in order to assure that the development proceeded in a fashion
that would establish priorities first, attaining the funda-
mental requirements, and second leading to demonstration of
the system reliability, if time permitted.
The first phase of the program was to determine control
parameters that would allow sustained plant operations under
conditions that established system equilibrium. Complica-
- 11 -
-------
tions were encountered during this period because of
need to continue shake-down operations with new and modi-
fied equipment.
The tasks slated for completion during this first
phase were:
1) Evaluation of changes to the MgO slurry
system.
2) Determination of the complete material
balance at the plant.
3) Measurement of system pH throughout the
plant.
4) Preliminary measurements of SO, removal.
5) Preliminary operation with regenerated
MgO from the calcining facility.
6) Evaluation of equilibrium conditions in
the fly ash system with the precipitator
in operation.
7) Coordination of operations during the pre-
liminary system testing (conducted by an
outside testing service) in order to pro-
vide initial S02» SO., and particle control
data.
The second phase in the program had as its goal the
optimization of operation of the system at the Dickerson
site and determination of the process conditions at the
regeneration facility necessary to provide an alkali suit-
able for reuse after several cycles. Some differences be-
tween the oil and coal fired applications could be expected
resulting from coal ash and/or first stage particle removal.
These differences would affect the:
1) Efficiency of SO- removal.
- 12 -
-------
2) Degree and rate of oxidation of the
salts formed.
3) Efficiency of regeneration of MgO.
4) Activity of the regenerated MgO.
Evaluation of these areas required an investigation
of the degree of contamination of MgO slurry with unre-
moved particulate matter and assessment of the S02 removal
efficiency of each stage for various cases. This was to
be done concurrently with the determination of the effect
of fly ash contamination on the regenerated MgO and acid
plant in long term operations.
In order to achieve these results in the shortest term
possible, extensive use was made of the data file complied
from the operating and analytical information logged during
the Boston Edison operations. Using data processing methods,
conditions at the Dickerson plant could be quickly analyzed
and compared with information in the file. In this way trends
indicating potential operating or process difficulties could
be rapidly identified and the necessary corrective action
taken so that maximum utilization of the intergrated plants
would be achieved. The plan involved investigation of several
parameters, such as system pH, pressure drop, recycle and gas
flow and the resultant ash concentrations and solids levels
in the various streams, dryer temperature, and liquid flow
rates.
Most variable excursions would occur as a natural effect
of process deviations. With data recording, sampling and
chemical analysis on a routine schedule, they would be mea-
sured and their effects evaluated. Some of the variables
such as gas flow and recycle rates were to be artifically varied
in order to reduce the danger of gross contamination of the
absorbent before the influence of the build-up of inerts on
- 13 -
-------
the process could be assessed.
In order to determine the effect of changes in these
parameters a factorial experimental program was to be used
where possible.
It can be anticipated that the widest excursions in
operating conditions would occur during the earliest part
of the program as the operators gain skill in running the
plant and as deliberate changes in conditions are made to
overcome process and mechanical problems. As operations
continue, these variations diminish or disappear as greater
control is gained over the system so that steady conditions
will be experienced during the latter part of this period.
Phase three of the program was to be a time extension
so that continuous operations under the optimized condi-
tions developed during Part II could be achieved. The dura-
tion of this phase would be contingent upon the time required
to achieve three additional cycles of the inventory. Over-
lap with the previous period would result in a total of five
regenerations of the magnesia and replacement of 1/2 of the
inventory. This phase was never implemented.
Program Schedule
Phase I of the program, the conclusion of the shake-
down operations and. the preliminary operational testing,
was estimated to require a one to two month period for com-
pletion. System feasibility as well as the suitability of
the installed equipment was to be demonstrated in this
Phase I. With this information plant design and equipment
selection could be undertaken for any similar installation.
The second phase of the program, investigation of
varying operating conditions on performance for system
optimization and the determination of the quality of regener-
ated magnesia to obtain the required levels of SO- control
- 14 -
-------
was estimated to require an additional four months. At
the conclusion of the testing program the detailed infor-
mation on operating values were to be used to fix control
parameters for the final phase.
Phase three, a continuous operating period of three
to four months' duration was to conclude the program.
A measure of the success of the program that was
achieved stems from the reorientation of the program from
the limited process guarantee tests to the more expansive
one described in the preceding sections. In order to ac-
complish those goals, however, it was necessary to expand
the socpe of work that was being undertaken at the Dickerson
site. Provision had to be made for a staff of chemists
and technicians to provide the analytical and testing sup-
port, technicians and operating engineers were necessary
to provide supervision and advisory services for the plant
operations and an overall organization had to be created
to direct the project, assimilate the data, implement the
changes necessary to solve the system problems and to co-
ordinate the operations being conducted at both Dickerson
and Rumford.
This was accomplished through a joint agreement be-
tween PEPCO, Chemico and the Environmental Protection
Agency in which PEPCO provided the funding for the bulk of
the operations at both plants, Chemico provided the project
management function and EPA funded the analytical program
and its support. EPA also provided funding for other
emissions testing programs conducted to determine the effect
of transient conditions.
- 15 -
-------
3.0 PROCESS DESCRIPTION
The pollution abatement system installed at PEPCO's
coal burning Dickerson Station on the Number 3 boiler is
composed of a particle control scrubber and an S02 control
absorber, both contained in a single, two stage unit
(Fig 1). Particle control is effected by use of a Chemico
designed variable throat venturi as a first stage followed
by a venturi absorber using the Chemico-Basic Magnesium
Oxide System for Flue Gas Desulfurization (FGD) as the
second stage. The particle control stage, including the
recycle and >slurry streams and fly ash thickening units
are independent and separate from the SO- control stage,
thus preventing fly ash contamination of the magnesia used
as the absorbent in that stage. Both the particle control
streams and the F.G.D. streams are operated in a closed
loop mode.
The original system concept was to provide a prototype
demonstration unit designed to process one-half of the flue
gas from the Unit 3 boiler, as shown in Fig. 2. In one
configuration the flue gas to be treated is diverted from
the stack after the electrostatic precipitator by using the
venturi scrubber induced draft fan. A second configuration
allows the ESP to be bypassed through new ductwork so that
tests could be conducted with the full particle loading
from the coal fired boiler treated in the wet scrubber unit.
In either case, the flue gas cleaned in the first stage
was processed for S0? removal in the second stage.
In the Chemico-Basic process for sulfur dioxide removal
from flue gas the S02 is absorbed in a slurry containing
- 16 -
-------
FIGURE 1
- 17 -
-------
PR6.CIPITATOR
I.D.FAKJ STACK
SCRUBBER
SCROBBBg
FAU
FGD 6AS PLOW SCHEMATIC
DICKERSOKJ MC 3 " PEPCQ
- 18 -
-------
magnesia, precipitating the solid magnesium sulfite. The
magnesium sulfite, after appropriate steps to reduce its
moisture content, is then processed at a separate installa-
tion to regenerate the magnesia and recover the SO, for
use in chemical manufacture. In this case the regenera-
tion was accomplished at the E.P.A. facility at Essex
Chemicals' Rumford Acid Plant in Rhode Island which had been
built for the New England S02 Control Project, with the
recovered S02 used in the manufacture of sulfuric acid.
3.0.1 Facility Description
Dickerson Station is located on the Potomac River
outside the town of Dickerson, Maryland. The plant is
situated in a rural, non industrialized area about 35
miles northwest of Washington, D.C. Coal is delivered
to the plant by rail.
The station has three electric generators each rated
at 190 MW. A fourth generator, rated at 800 MW is scheduled
for installation nearby on the plant site by 1982. The in-
stalled 95 MW FGD system is sized to handle approximately
one-half the exhaust gas flow from Unit No. 3. Pertinent
data on this Unit are given in Table 1. Unit No. 3 has a
dry-bottom coal-fired boiler that was designed by Combustion
Engineering and installed in 1962.
The coal presently burned has an average gross heating
value of 11,700 BTU/lb. Average ash and sulfur contents
are 14 percent and 2 percent, respectively.
The boiler is fitted with an electrostatic precipitator
(ESP) designed and installed by Research-Cottrell in 1962.
Particle collection efficiency was estimated to be 94 per-
cent. The FGD system is installed so that it can receive
exhaust gas either from the outlet or breeching ahead of
the ESP.
- 19 -
-------
TABLE 1
PERTINENT DATA ON PLANT DESIGN, OPERATION
AND ATMOSPHERIC EMISSIONS
Boiler data - Dickerson No. 3 - PEPCO
Rated Generating Capacity MW
Average Capacity Factor (1974), %
Boiler Manufacturer
Year Placed In Service
Unit Heat Rate, BTU/kwh
Maximum Coal Consumption, Ton/Hr.
Maximum Heat Input, BTU/hr.
Stack Height Above Grade, Ft.
Flue Gas: Design Rate, ACFM
Maximum Rate, ACFM
Flue Gas Temperature, °F
Emission Controls:
Particulate Matter
so2
Particle Emission Rate:
Allowable, g/scf dry
SO- Emission Rates:
Allowable, Ib/Million BTU
190
81
Combustion Engineering
1962
9180
74.5
1744 x 106
400
450,000
650,000
245
ESP with a venturi
scrubber on half
of gas flow.
Venturi absorber on
half of the gas flow.
0.03
(1)
1% sulfur coal equivalent
- 20 -
-------
3.0.2 Pollution Abatement System Description
A schematic representation of the pollution abate-
ment process installed at the power plant is shown in
Fig. 3. The principal element of the system is the two
stage venturi scrubber/absorber which combines in a single
vessel both particle removal and S0_ control. Each of the
two functions are separated in the unit with the flue gas
being treated sequentially, first to remove fly ash then
to reduce its S02 content to acceptable levels. Besides
providing the means of cleaning the flue gas, this sequence
assures that a negligible amount of fly ash will be cap-
tured in the magnesia slurry used for S02 removal in the
second stage. This is necessary in order to prevent a
buildup of coal fly ash which would result in the contam-
ination of the regenerated MgO. A further advantage of
this design is the ability to use the particle removal
stage independently should it be desirable to continue the
unit in operation without SC^ control.
The Chemico/Basic1s Magnesium Oxide System for the
recovery of sulfur dioxide from plant flue gases utilizes
the sulfur dioxide absorption characteristics of an aqueous
slurry of magnesium sulfite, magnesium sulfate, and magnesium
oxide. The process is accomplished in several steps:
Absorption
Dewatering & Drying
Materials Handling
Calcination
with the regeneration/recovery step (calcination) performed
at a remote site.
Absorption
In the absorption step the process chemistry which
describes the removal of S02 from the flue gas can be most
- 21 -
-------
THE DICKERSON REGENERATIVE WET SCRUBBING SYSTEM
POTOMAC ELECTRIC POWER CO. PROTOTYPE
PRECIPITATOR/SCRUBBER - ABSORBER
MgO ADDITIVE SYSTEM FOR S02 RECOVERY
SCHEMATIC PROCESS FLOW SHEET
VENTURI
SCRUBBER/ABSORBER
1ST STAGE
ELECTRO-STATIC
PRECIPIIAIOI
TO DRY ASH HANDLING SYSTEM
OUST COUfCTOR
RECYCLED POND
WATER
TRANSFER
TANK
CRYSTAL
ORYEI
M0O FROM ACID PLANT
MgSOi 1O ACID
Figure 3.
-------
simply explained as the diffusion of S02 through the flue
gas to a liquid surface, then absorption of the S02 with the
hydrated form of MgO, i.e.
Mg° + S0
(aq) 2 - 3 (3-1)
The MgSO_ produced has a low solubility and can be separated
from the absorbing slurry as a solid.
In the process, the flue gas containing sulfur oxides
enters a venturi absorber of special design, and contacts
the absorbing media which is an aqueous slurry of magnesium
oxide, magnesium sulfite, magnesium sulfate, and a small
percentage of other components. The process of SO- removal
that occurs is explained by conventional mass transfer prin-
ciples.
The venturi absorber can be considered as similar to
a co-current, packed vessel. In the venturi the liquid
slurry is introduced and flows downward on a surface over
which passes an accelerating gas stream. The high velocity gas
stream flowing over the liquid causes wave motion on the fluid
surface, the waves increase in amplitude and finally disperse
as fine droplets. Thus, the whole mass of liquid can be dis-
tributed in the form of atomized droplets in the gas stream.
There are several advantages in using the venturi as
an absorber. The absorption surface is dispersed into and
flows with the gas stream during the time absorption is oc-
curing, thus eliminating the problems of plugging associated
with conventional packed towers (the surface area per unit
volume is approximately equivalent to dumped 3 inch rachig
rings) . Due to system dynamics, this surface area relation
is relatively invariant over wide turn-down ratios, and high
removal efficiencies can be maintained over the normal operat-
ing range of the power plant's boiler.
- 23 -
-------
Dewatering & Drying
A stream from the absorption system enters a centri-
fuge where the solids which were formed by the absorption
reaction in the slurry are separated. This bleed stream
is controlled in order to maintain a constant solids con-
tent in the recycle slurry by removal of product magnesium
sulfite and unreacted magnesium oxide and any precipitated
magnesium sulfate. The system is operated so that the ab-
sorbed S02 is removed as an equivalent amount of the mag-
nesium sulfur compounds.
The wet centrifuge cake containing hydrated magnesium
sulfite, magnesium oxide and magnesium sulfate plus other
solids removed in the venturi absorber-centrifuge system
is passed to a rotary dryer to remove both unbound water
and most of the water of crystallization. The dry product
is easy to store, and the removal of water reduces shipping
costs.
Materials Handling
The anhydrous magnesium sulfite and magnesium sulfate
produced in the dryer is conveyed to a storage silo for
transportation by rail or truck to the recovery acid plant.
The same transport is used to return regnerated magnesia to
the magnesium oxide silo at the power plant on the return
trip.
Calcination System
Calcination is the process used for regeneration. The
magnesium sulfite which has been separated and dried is
thermally decomposed as represented by the following reaction:
MgS03 >MgO + S02 (3-2)
The recovered S02 is used in the production of sulfuric acid
and the regenerated MgO returned to the process for reuse.
The dry product transported to the regeneration acid plant
from the power plant is received, weighed and pneumatically
- 24 -
-------
conveyed to a storage silo. It is fed to a direct-fired
rotary kiln at a metered rate, and calcined to both generate
sulfur dioxide gas and regenerate magnesium oxide. Coke
can be added to provide a reducing atmosphere, as necessary,
to reduce the residual magnesium sulfate to the oxide and
sulfur dioxide. The hot flue gas containing sulfur dioxide
and dust enters a cyclone where essentially all the dust is
removed and returned to the calciner. The flue gas then en-
ters a venturi scrubber for final dust cleaning. At the
same time, the gas is cooled and adiabatically saturated.
The saturated flue gas is further cooled in a direct
contact packed tower to meet the requirements of the acid
plant water balance. The cleaned flue gas then enters the
drying tower of an existing 50 T/D acid plant. At the Essex
Chemicals' installation, the resultant product made from the
recovery of the sulfur dioxide is 98% sulfuric acid.
The regenerated magnesia is cooled, conveyed to the
magnesia storage silo, and recycled back to the power plant.
3.1 DETAILED PROCESS DESCRIPTION
The process flow sheet for the facility installed at
Dickerson is shown in Fig. 4, material balances are given in
Table 2 with the description of each item given in the Equip-
ment List, Table 3.
3.1.1 Particle Removal
The first stage of the system uses a Chemico variable
throat venturi. In this design, a 60° tapered plug is sus-
pended by a height adjustment mechanism in a 60° converging
section of the scrubber. By stroking the mechanism through
its entire adjustment the throat opening can be varied allow-
ing up to a two foot maximum spacing. In normal operation,
with standard liquor circulation rates, the throat width has
been adjusted to maintain an eleven inch pressure difference
across the particle removal stage.
- 25 -
-------
VENTURI
5CRJBBIR/46SON6ER
\V STACE
UfCTRO-STATIC
PftfCIPlTATOR
1A/ET MIST
FAN ELIMINATOK
TO Z* SI Aft*
DUSTCOLLECTOR
ICINTR4TC
^ TANK
TPAHSFIR
TANK
CRYSTAL
DRYIK
FIOM 4CI6 PIAMT
F I SURE A
P6PCO REGENERATIVE
WET SCRUBBING SYSTEM
-------
TABLE
STREAM PROPERTIES ANO COMPOSITION
SO2 AND FLY ASH REMOVAL SYSTEM
M
STREAM
NUMBERS
Temp.
Flow
•F
GPM
Solids Content.%
H2O
MgS03
MRSO«
MgSO3>6H2O
MgSO4.7H2O
MgO
Inert*
Fly Ash
Total
Fuel
STREAM
NUMBERS
Temp.
Pressure
Flow
S02
R2O
Dry Gas
Partlculate
Lb/Mln
Lb/Mln
Lb/Min
Lb/Mln
Lb/Mln
Lb/Min
Lb/Min
Lb/Mln
Lb/Mln
Btu/Hr
7
117
5.165
2.0
42. 870
875
43. 745
1
8
117
980
2.0
8.166
166
8.332
2
9
140
13.2
1.102
166
1.268
1
10
LIQUID AND
10.621
13.3
82. 850
95. 520
4
11
12 13 14 15 16
SOLID STREAMS
170
13.2
1.322
178
9.9
9.9
2.4
0
1.522
5
38.1
18.7
10.4 300
86.3
4.9
178 2.0
9.9
9.9 9.9 45.3 45.3
2.4 2.4 2.4 2.4
0 0
210.6 105.5 47.7 347.7
16.6x10*
6
GAS STREAMS
•F
In. W. G.
ACFM
Lb/Mln
Lb/Min
Lb/Mln
Lb/Mln
259
-11
295. 000
62.5
581
15.820
166.6
117
-26
263. 000
62.5
1.209
15.820
0.85
121
-36
286. 000
6.25
1.420
16.457
0.89
121
0
260. 000
6.25
1.420
16.457
0.89
450
-5.5
21.600
0.22
70
•0.6
10.200
-------
TABLE 3
EQUIPMENT LIST
Motor
Horsepower
Item Description
MgO Make-Up Tank
Dust Collector-MgO Silo
Thickener
Transfer Tank
Distribution Box
Sump Tank
Mother Liquor Tank
Dust Collector-Dryer
MgO Storage Silo
MgS03 Silo
MgO Feed Pump
First Stage Recycle Pump
Second Stage Recycle Pump
Underflow Pump
Transfer Pump
Sump Pump
Mother Liquor Pump
Induced Draft Fan
MgO Agitator
Mother Liquor Agitator
MgO Weigh Feeder
Dryer Feed Conveyor
Dryer Discharge Conveyor
Dryer Discharge Elevator
MgS03 Conveyor
MgSO, Weigh Feeder
Two-Stage Venturi Scrubber
Mist Eliminator
Variable Speed Coupling
Code No.
G-101
G-102
G-201 A,B
G-202
G-203
G-204
G-301
G-401
1-101
1-401
J-101 A,B
J-201 A,B
J-202 A,B,C
J-203 A,B,C,D
J-204 A,B
J-205
J-301 A,B
K-201
M-101
M-301
0-101
0-401
0-402
0-403
0-404
O-405
R-201
R-202
R-203
Quantity
1
1
2
1
1
1
1
1
1
1
2
2
3
4
2
1
2
1
1
1
1
1
1
1
1
1
1
1
1
(HP)
10
350
250
125
30
20
3500
5
3
1
2
2
2
- 28 -
-------
TABLE 3 (Cont'd)
EQUIPMENT LIST
Motor
Horsepower
if HP ) '
Item Description Code No. Quantity v '
Centrifuge R-301 1 200
Rotary Dryer R-401 1 40
Dryer I.D. Fan K-402 1 100
Instrument Air Dryer V-101 1
- 29 -
-------
Liquor is circulated in this particle removal stage
by recycle pumps, J-201 A & B (one operating, one spare,
each with a design capacity of 5700 GPM and driven by a
350 HP motor) through rubber lined distribution piping to
the upper level of the scrubber R-201. The recycle flow
is split into two streams before reentering the scrubber.
The flow in each stream is adjusted and set by pinch valves
to provide uniform irrigation of the converging surfaces.
The outer converging section is supplied by ten 4" tangen-
tial nozzles while the adjustable cone is supplied by three
6" lines feeding a single bull nozzle.
Flue gas enters the vessel and passes downwards through
the irrigated converging section. As the gas is accelerated
through the throat the liquid film is atomized and distributed
through the gas stream. Particle collection occurs on the
dispersed droplets through the mechanisms of impaction, dif-
fusion and attraction. The flue gas and dispersed liquor
continue through the central down-comer to the basin area
of the first stage where the gas turns 180° upwards to pass
through the first stage mist eliminators. The liquid drop-
lets are first disengaged in the turning zone and fall into
the first stage liquor pool, droplets small enough to be
carried with the gas stream are captured by the slot type
mist eliminator. The cleaned gas continues out of the first
stage through the annular space between the outer shell and
the wall of the first stage reservoir.
A bleed stream, about 15% of the total flow from the
recycle pumps, is sent to a distribution box and thence to
one of two thickeners, G-201 A & B, each of these is 40' in
diameter with a rake mechanism feeding a central underflow
discharge. The thickened underflow, approximately 20 GPM
at 40 percent solids, is discharged to a distribution tank
where water from the lower pond is added and the mixture is
- 30 -
-------
pumped to the upper settling pond. The pond overflow cas-
cades through a total of four settling ponds in series and
the clear water from the lowest is again returned to the
dilution tank.
The thickener overflow drops into a transfer tank and
is returned to the first stage recycle loop by transfer
pumps J-204 A & B, (one operating, one spare, each designed
for 1190 GPM and driven by a 125 HP motor) so that closed
loop operation is maintained.
It should be noted that the mechanisms for collection
of particulate matter or sulfur dioxide both depend on the
formation of droplets dispersed in the flowing gas. In the
Chemico venturi the mechanism of atomizing the liquid by the
accelerated gas stream is the same for either stage. How-
ever, the mode of operation of the stages is different.
In the case of particle collection a specific droplet size
distribution will be found to provide the maximum collection
efficiency and the size distribution is directly related to
a gas velocity flowing over the liquid surfaces. In the
case of gas adsorption it is necessary to maintain a uniform
specific surface area (i.e. area of droplet surface per volume
of gas) which in the venturi, is self regulating, after initial
adjustment, over a range of gas flows. Thus, first stage
efficiency is maintained by operation at a fixed pressure
drop to keep the best droplet size distribution while second
stage efficiency is controlled by operation at fixed liquor
recycle rates to maintain a uniform surface area of the dis-
persed droplets.
3.1.2 Detailed Process Description - SO2 Removal
In this system magnesia, both regenerated and make-up
material, is transferred from pneumatic discharge hopper
trucks to the elevated MgO storage bin through a 4 in.
pneumatic unloading system. The MgO bin, 1-101, is 25 ft.
- 31 -
-------
in diameter and 46.5 ft high. Magnesia is fed from the
silo, which is equipped with a vibrating hopper bottom,
0-103, to a 3760 gallon capacity MgO make-up tank G-101 by
an adjustable weigh feeder, 0-101, with mother liquor to
make the desired MgO slurry composition. A small pre-mix
tank is interposed between the weighing system and the
steam heated make-up tank to insure dispersion of the MgO
powder and to act as a vapor seal. Heated magnesia slurry
is added to the recycle stream by the MgO make-up pump,
J-101 A, B. The magnesia slurry addition rate is controlled
by the operator to maintain the pH of the absorbing slurry
at the desired value.
The recycle stream for the absorber is circulated at
a rate of 10,767 GPM by pumps, J-202 A, B and C, (the capa-
city of each of these 250 HP pumps is half the design flow
of the recycle stream) to provide a slurry dispersion within
the vessel sufficient for the desired S0~ removal.
The recycle stream itself is split into two streams
for distribution before reentering the venturi absorber.
The flow in each stream is adjusted and set by plug valves
to provide a uniform irrigation to the converging surfaces.
Distribution of absorbing slurry is made by ten, six inch
tangential nozzles supplying the outer converging section
while the inner cone is supplied by a single central nozzle.
The slurry enters the upper part of the absorber stage
with the flue gas which is pulled through the system by the
I.D. fan K-201 driven by an 3500 HP motor, designed to handle
286,000 ACFM at a discharge head of -26 in. water. The gas
and slurry mixture passes through the throat area into a di-
verging section, then into a central downcomer. To exit the
vessel the flow of the cleaned flue gas turns one hundred
eighty degrees upwards. In this step most of the larger
liquid droplets are disengaged from the flowing gas stream
and fall to the slurry pool in the conical base of the absor-
- 32 -
-------
ber. The treated flue gas continues upwards through a slot-
type mist eliminator to further remove any entrained liquid
before the gas exits to wet fan K-201. The fan wheel of
K-201, driven through a variable speed drive, is continuously
sprayed with water to prevent solids build up on the blades.
Entrained liquid in the gas stream leaving the fan is removed
in Mist Eliminator vessel R-202 before passing the stack.
A stream of slurry, approximately 1.6% of the total
recycle, is taken from the discharge of the recycle pumps
to the centrifuge, R-301. This is a 36 in. x 72 in. horizontal,
solid bowl unit driven by a 200 HP motor, and in normal
operations it removes 50% of the solids in the liquid stream
going to it.
The solids separated in the centrifuge contain an
amount of SO- (as MgS03 or MgSO.) equivalent to the amount
removed in the absorber. The centrate is discharged directly
to an agitated tank, G-301 of 1375 gallon capacity which
serves as a pump tank for this system, and then is returned
to the MgO slurrying system by pumps J-301 A & B. The basin
of the absorber is the reservoir for the system and contains
approximately 20,000 gallons of slurry. The other sources
of water entering the system are small amounts added as pump
seal water and an additional quantity used as a mist elimina-
tor wash; the latter is an intermittent addition.
The centrifuge cake separated in the centrifuge is fed
to a dryer by screw conveyor O-401, a 35 ft. long unit with
a 12 in. diameter ribbon flight. The dryer is a rotary unit
8 ft. in diameter by 50 ft. long. Heat is supplied from an
integral 18 ft. long oil fired combustion chamber at a maxi-
mum heat release rate of 26.3x10 BTU/hr.
The dryer off-gas exits in to a cyclone, G-401, designed
for 97% removal of the solids entering with the gas. From
the cyclone dust collector, the gas discharges via a booster
- 33 -
-------
fan, K-402, to the venturi absorber where, mixed with the
entering flue gas, it is cleaned of its remaining particle
load. The separated fines collected by the cyclone returns
directly to the dry product pneumatic conveying system dis-
charging to the product silo.
The dryer product is discharged to the dryer discharge
conveyor, D-402, 21.75 feet long unit with a 9 inch diameter
solid flight screw. It then exits into a MgSO., weigh feeder
0-405 at the boot of the MgSO, discharge elevator, 0-403,
which is designed to handle 5 tons/hour of dried magnesium
sulfite. The product then travels through another conveyor,
0-404, located on top of the MgS03 storage silo. The stor-
age silo is 66.5 feet high and 25 feet in diameter, and is
equipped with a vibrating discharge hopper, 0-406.
The product MgS03 is finally gravity loaded into a
waiting hopper truck for shipment to the MgO regeneration
plant.
3.2 DETAILED PROCESS DESCRIPTION - MAGNESIA REGENERATION
SYSTEM
Magnesium sulfite from the absorption system is received
at the regeneration facility. The regeneration section flow
sheet is shown in Fig 5 . Material balances are given in Table
4. The salt is unloaded from hopper trucks to the sulfite
silo 1-401 through a 4 in. pneumatic conveying system similar
to that used for MgO at the power station. The sulfite silo
was an existing tile structure, 25 ft. in diameter x 45 ft.
high, which was water-proofed before this use. Sulfite is
unlaoded from this silo through a pair of inclined screw
conveyors, MgS03 conveyors 1 and 2, 0-509 and 0-510, equipped
with 12 in. diameter helicoid flights, which carry the
material from the below grade discharge spout to the boot
of the MgS03 elevator.
This elevator, O-511, a continuous discharge unit 54
ft. high, is designed to handle 9 TPH of MgSO,. It discharges
- 34 -
-------
ACIP
-------
TABLE 4-
STREAM PROPERTIES AND COMPOSITION
MgO REGENERATION SYSTEM
STREAM
NUMBERS
Temp °F
Pressure PSIG
Flow GPM
MgfHSO )2 Ib/Min
MgO Ib/Min
MgSO, Ib/Min
u
MgSO4 Ib/Min
Inerts Ib/Min
Fly Ash Ib/Min
Total Solids Ib/Min
H,O Ib/Min
Total Flow Ib/Min
Fuel Oil Ib/Min
STREAM
NUMBERS
Temp. ° F
Pressure In. W. C
Flow ACFM
Flow SCFM
Total Dry Gas Ib/Min
HJD Vapor Ib/Min
Total Wet Gas Ib/Min
SO Ib/Min
O, Ib/Min
Dry Gas MW
3 4 5 6 7 8 9 10 11 12 13
8.24
72.50
4.12
1.89
0.25
87.0
87.0
1
2
2
LIQUID & SOLID STREAMS
300
38.35
1.81
0.25
40.41
40.41
2
250
1.5
12
14
160
91
760
160 '
2
0.88
1.5
0.1
0.1
13.4
15.88
90
119
99.2
150
125.5
1045
150
80
65
Amb.
1.5
12
80
205
1708
GAS STREAMS
100
-45
2.980
2.460
206.2
9.0
215.2
44. 7
1.8
34.4
70
0
2.140
2.100
161
29
100
-45
640
529
39.0
0.8
39.8
29
U)
CTi
-------
to the MgS03 feed bin, G-506, a 6 ft. diameter by 11 ft.
high vessel equipped with a vibrating bottom, which also
serves as a surge bin for the weigh feeder 0-514. This
weigh feeder is continuously variable to a maximum capacity
of 3.6 TPH and discharges to the calciner conveyor 0-512,
a 33 ft. long horizontal unit equipped with a 9 in. dia-
meter helicoid screw. Coke is also discharged to this
screw which serves to mix the two components as calciner
feed prior to processing.
The coke is pneumatically conveyed to the coke feed
bin G-505 9 ft. diameter x 19 ft. high, which also serves
as the storage bin for this material. It is metered to the
calciner conveyor by a weigh feeder which is also continuously
variable up to a maximum capacity of 2.5 pounds per minute.
The two streams (3 and 4 of Fig. 5 ) enter at points 9 feet
apart and mix in the remaining 22 ft. section of the conveyor
before reaching the calciner elevator O-513, a 43 ft. high
centrifugal discharge unit. This elevator feeds directly to
the calciner R-501.
The calciner is a refractory lined, oil fired, rotary
kiln, 7 ft. 6 in. ID and 120 ft. long. Rotational speed is
variable between 1.5 and 2 RPM using a variable diameter
pulley drive, and the kiln has a slope of 3/8 in. per foot.
The calciner product empties into four tube coolers attached
to the shell and equipped with internal flights to contact
the existing hot regenerated magnesia with incoming air which
serves a secondary combustion air in the calciner, thus cool-
ing the calciner product before it empties to the MgO con-
veyor #1, 0-503.
This is a 9 in. diameter helicoid screw conveyor,
54% ft. long, which elevates the product from the calciner
tube cooler discharge seventeen feet to an enclosure housing
the MgO processing equipment. The regenerated MgO is dis-
- 37 -
-------
charged from 0-503 to a magnetic pulley for tramp iron
separation. The MgO then passes through a 1 in. mesh
vibrating screen and enters a high speed pulverizer which
reduces it to the finished grind shown in Table 5.
The cleaned, cooled, and pulverized regenerated mag-
nesia flows by gravity to the MgO elevator 0-504, a con-
tinuous discharge design, 89 ft. high, which brings the
material to the top of the MgO storage silo where it is
loaded by means of a horizontal screw conveyor (MgO Con-
veyor #2) 0-505 equipped with a 9 in. helicoid screw.
The MgO silo, 1-502, 25 ft. in diameter and 45 ft.
high is equipped with vibrating hopper bottom, 0-506, and
elevated on a structural steel support for direct gravity
loading of the returning trucks.
The gas from the calciner, containing SO- and products
of combustion, as well as a small percentage of excess air,
is first partially cleaned of particles in the cyclone
dust collectors P-075. This is a dual cyclone array, de-
signed for a 1 in. pressure drop. The collected solids are
returned to the calciner with the feed to the unit.
The partially cleaned calciner gas containing 8-10%
S02 is further cleaned in a venturi scrubber of Chemico's
special design, operated at a pressure drop of approximately
25 in. of H-O where it is also adiabatically saturated. Next
the gas enters the separator tower section, which is an in-
tegral part of the venturi equipment. The lower section of
this 4*s ft. diameter vessel serves a cyclonic liquid separa-
tor and the upper section, containing eight feet of 3*s in
pall rings is irrigated with cooled weak acid to further re-
duce the temperature of the gas to 100 F in order to main-
tain the acid plant's water balance. A slip stream of cool-
ing liquor is stripped of dissolved SO- in the weak acid
- 38 -
-------
TABLE 5
DRY SCREEN ANALYSIS
Screen Unpulverized Pulverized
Size (Tyler) Calciner Product Calciner Product
+50 30.3 % 5.3 %
+100 18.1 9.5
+200 9.4 8.8
+325 13.1 49.8
-325 29.1 26.6
- 39 -
-------
stripping tower P-502, a small (15 in. diameter x 14 ft.
high) packed contactor.
The stripped SO- joins the main gas stream and is
ducted to the acid plant in 18 in. diameter, FRP pipe.
3.3 ACID PLANT
The regeneration system was installed at the Rumford
acid plant of Essex Chemical Company located in Rumford,
RI. The Rumford plant is located 55 miles from the instal-
lation at Mystic Station and haulage of the dryer and cal-
ciner products between these two sites was done by truck.
Both plants were equipped to load their respective products
from elevated silos and receive their feed by pneumatic un-
loading.
The plant has been producing sulfuric acid since
1929 when it was built by Chemical Construction Corporation
and for the past several years all acid produced was sold
in the merchant market with no captive use. Markets for
the acid made at this plant are manufactured of detergents,
dyestuffs, Pharmaceuticals, aluminum sulfate, tanning
chemical, steel pickling, boiler water treatment, lead-acid
batteries, galvanization, etc. Sales in this market were
at the full published price, at the time of this program,
approximately $46 per ton based on 100% H-SO,. No change
in use or pricing was imposed when marketing the acid produced
from MgO regeneration.
The plant is a contact sulfuric acid plant which uses
sulfur as raw material. In the Chemico process the molten
sulfur is injected into the furnace using a spray burner,
with the process air first dried in a tower against 93% sul-
furic acid. The catalyst is vanadium pentoxide.
When first built in 1928 the Rumford plant was an in-
door plant typical of its time and had a capacity of about
- 40 -
-------
20 tons per day of 100% sulfuric acid. Substantial
modifications were made to the plant in 1948 by Chemico
to increase its capacity to 50 tons per day. The modifi-
cations included improved converters, a converter heat ex-
changer, waste heat boiler and economizer all installed out-
doors. In addition, the cast iron cooling section was en-
larged and moved outside.
This was the plant that was modified in 1971 to
accept the calciner off-gas essentially converting it to a
metallurgical (roaster) gas plant; however, in the modifi-
cation the capability to continue to burn sulfur and aug-
ment the SOj from the regeneration plant was retained.
Typical feed gas analyses to the acid unit are shown
in Table 6.
The regeneration of magnesia, described in the pre-
vious section, produces an off-gas from the calciner approxi-
mately 100 fold richer in S0_ than the power plant combus-
tion gases treated in the S02 Absorption System. The cal-
ciner gas is of sufficient strength to be used as a feed for
the manufacture of sulfuric acid.
The small, sulfur burning acid plant (Fig. 6) required
some modifications to enable it to accept the calciner gas.
Provisions were also made during these modifications allow-
ing the plant to burn sulfur as an alternative source of
S02 or to operate on a combined feed from both combustion
of sulfur and gas from the regeneration plant.
3.3.1 Acid Plant Modifications
The principal element replaced in the acid plant was
Main S02 Blower, K-901. The original blower handled only
air required for the conversion of SO- to S03 and was not
designed to be gas tight (as required when feeding the acid
plant with gas containing SO-), or capable of the required
- 41 -
-------
TABLE 6
SULFURIC ACID UNIT FEED GAS
COMPOSITION/ MOLE %
Source Of Gas N2 cq2 q2 H20 S02
Regeneration Section 73 657 9
Sulfur Burning 79 12 9
- 42 -
-------
kf
soa RJCM puoe
a AS eflie I*
.'I
*
(A)
I
mneiM
•»(•
*,
PUMP
TO WES
CD-
93X PUMP
L
X
TOW6.C
PRODUCT
ORX
TOWER
35t AGIO
•nraj
98^
a
eooueas
PRODUCT
TO
«-To
PUMP TANKS
• CM. R6TUBW
TO ATMO&PM&Q&
JS
IM1»1». '"»•« M
iiia 1 i fj alia I •
ioono J—L f^isaa^,! 1
J?™,
'. ACID TO STOBAQEl *
MO PLOW
98% PRODUCT
o sTon
/ v
-------
control of suction pressure for operation of the re-
generation plant. The replacement blower is an axial flow
compressor of 5,240 CFM capacity driven by a 200 HP motor.
It is capable of a suction pressure of up to 44 in. H-O for
the venturi pressure drop, duct and equipment losses, and
calciner draft at the regeneration plant, and a 75 in. H20
discharge pressure for the acid plant. The unit was de-
signed to handle either air or a typical feed gas as shown
below:
ACID PLANT FEED GAS
S02 6.8%
02 9.4%
N2 74.7%
C02 8.1%
CO 0.9%
The blower was equipped with an adjustable recycle
control in order to accommodate the variable feed rate of
the regeneration plant.
Another important element added to the acid plant was
the Cold Heat Exchanger E-901. This piece of equipment
supplies the heat to the incoming cold calciner gas equiva-
lent to that available in the gas when burning sulfur in the
sulfur furnace. This is necessary to ensure that the gas
entering the first mass of the primary converter is hot
enough to sustain the reaction. The exchanger, E-901, is
a vertical shell and tube unit 3 ft. diameter x 9 ft. high
containing 1700 sq. ft. of surface. Hot gas exiting the
secondary converter enters the tube side on E-901 and heats
the calciner gas passing through the shell side to 540 F.
3.3.2 Acid Production
In the acid plant, calciner gas enters (Stream 1 on
Fig. 6 ) and is diluted with sufficient air through the air
filter for conversion and passes to the acid plant's drying
- 44 -
-------
tower where it is contacted with 93% sulfuric acid to re-
move the water from the gas. Entrained liquid is removed
in F-901A, the Dry Tower. Any S02 absorbed by the 93% acid
is stripped in the new Tower (F-903), a 30 in. diameter x 15 ft. high
ceramic tower containing a 10 ft. bed of 1% in. saddles,
and returned to the main gas stream. The small amount of
additional air is used for trim of the oxygen concentration.
The gas is pressurized by the S02 blower and enters the Cold
Heat Exchanger, passes to the Converter Heat Exchanger, E-902
and finally enters the converters H-901 A & B. The heat
balance for the conversion step is maintained by using the
gas from the second mass of the Primary Converter, H-901A,
to heat the gas from the Cold Exchanger, and the gas leaving
the fourth mass of the secondary converter, to heat the
incoming feed.
Next, the stream containing 6.7% SO., is contacted with
98% H2S04 in the Absorption Tower F-902 A & B. The gas
streams from these towers are first demisted in R-902 A & B,
then the remaining SO- is removed in the Tail Gas Scrubbing
section shown in Fig. 7 and Table 7 to reduce the concentra-
tion of S02 leaving the plant to allowable levels.
The absorbant used in the Tail Gas Scrubbing section
is NaOH solution.
In order to allow a rapid change to sulfur burning,
when the plant is operated on 100% calciner gas, the sulfur
furnace was equipped for oil firing. The combustion products
from this operation are vented to the atmosphere through a
short stack which could be bypassed. In normal operation,
however, sulfur was burned concurrently because of a low
MgS03 feed rate. This procedure had the advantage of allow-
ing a rapid change should the calciner gas flow be interrupted.
This alternative was possible because of the several dampers
which had been incorporated in the ductwork.
- 45 -
-------
, WATtK
CE. TA,
FSZOKA
OM A BATCH
TO
NU!
PJMP
—BECVCUE
-RjfAP
— J-107.
PKS.URE 7
TAIL GAS. SCRUBBING SYSTEM
-------
TABLE 1
STREAM PROPERTIES AND COMPOSITION
TAIL GAS SCRUBBER SYSTEM
STREAM NUMBERS
Temperature ° F
Pressure IN WC
Flow ACFM
Dry Gas Ih/Min
H2O Ih/Min
Wet Gas Ib/Min
SO2 (Design) Ih/Min
^ SO2 PPM Dry
•"• By Volume
1
160
0
4,250
279
0
279
3.0
5000
2
165
14
4140
279
0
279
3.0
5000
3
77
0
5600
276.3
5.6
281.9
0.3
500
STREAM NUMBERS
Temperature ' F
Flow GPM
NaOll Ib/Min
Na -SO, Ih/Mm
i 3
H2O Liquid Ib/Min
Total Solution Ib/Min
4 5 678
77
152
299
1266
1565
Amb.
0.6
3.75
3.75
7.5
77
3.0
5.9
23.6
29.5
Amb.
3.0
24.6
24.6
77
76*
149. 5
133
782.5
-------
4.0 POLLUTION ABATEMENT SYSTEM PERFORMANCE
Construction and check out of the system at Dickerson was
completed by September of 1973 and gas was first passed to
the unit on September 14th. In the following three month
period 600 hours of operation were logged in the initial
operation and debugging of the system. Most operating periods
were of short duration, from 10 to 24 hours, before a system
problem would force a halt. The major problem area encountered
in the earliest runs was the MgO feed system, where plugging
occurred in both the mix tank and slurry lines. The plugging
was partially remedied by two modifications: the installation
of a premix tank before the slaking tank to assure complete
wetting of the MgO powder with recycle liquor, and addition of
a steam sparging system to the tank. Both these remedies
have also been found necessary to correct similar problems in
the oil-fired application.
Other shut-downs in this period were caused by leaks that
developed in the venturi scrubber internal piping between first
and second stages. The leaks occurred in expansion joints
where sections of the recycle lines left the first stage re-
servoir. Testing showed the joints supplied were off-specifi-
cation stainless steel that was not resistant to this acidic
stream.
Despite these problems two sequential six-day runs inter-
rupted by a 24 hour boiler outage, were attained in the pre-
liminary operations. While these runs did show the operability
of the entire system their usefulness in assessing chemical
stability was limited because of high entrainment losses from
the venturi which prevented equilibrium conditions from being
attained in the recycle streams. A low S0_ removal efficiency
was also noted in the preliminary testing.
- 48 -
-------
An inspection of the scrubber in January 1974 showed
that off-specification material, unresistant to the acidic
environment of the first stage, had also been used in bolts,
nuts, hanger rods, spray nozzles and some piping in the
vessel. A major shut-down extending over the next six months
was taken to replace corroded parts and repair damaged
areas. A policy was also adopted to test each replacement
item to ensure that it was of the specified type 316 stain-
less steel. A brief verification run was made at the be-
ginning of May and the new problems which it uncovered were
corrected in the following weeks, prior to starting regular
operations.
During this initial operating period MgO regeneration
was not available to the Dickerson plant. The EPA unit
located at Rumford, Rhode Island was being used exclusively
for regeneration of the MgO from the oil-fired system proto-
type. The regeneration plant was not scheduled to be avail-
able for Dickerson Station till July 1, 1974, when PEPCO
would then have exclusive use of the facility. In antici-
pation of this use, PEPCO sent three rail cars of MgS03 from
its operations to the Rumford facility for preliminary testing
of magnesium sulfite derived from coal fired operation.
Before operations with regenerated MgO were started at
the power plant some additional test work was conducted in
June 1974. The low S02 removal efficiency, first noted in
the debugging runs, was confirmed by these test runs and
attributed to a low (1" to 3"^^P) second stage pressure drop.
A restrictor, to reduce the second stage throat area by 40%,
was designed in order to increase the pressure drop to design
value.
The test runs exhausted the initial supply of MgO which
had been delivered for start up operations. The abatement
- 49 -
-------
system was idle for another six week period, awaiting a
new shipment of MgO, the return of regenerated magnesia
from Rhode Island and installation of the throat restric-
tor.
4.1 DEVELOPMENT TEST PROGRAM
The silos at the regeneration plant were completely
emptied of all Boston Edison MgSO3 in July 1974 and filled
with PEPCO material from the rail cars. (The
lack of any size reduction or lump breaking equipment in
the path between the Dickerson dryer and regeneration
plant calciner caused unloading problems then, and through-
out the entire program). The operation, after a brief one-
day adjustment period, produced a calciner product which
tests showed to be of acceptable quality for recycle.
When the PEPCO MgSO^ was processed in the calcining
facility, it was found necessary to add carbon to the feed
to effect the reduction of magnesium sulfate. The oil de-
rived calciner feed had contained sufficient carbon to satisfy
the reduction reaction needs as the absorber, in the oil-fired
application, treated the gas directly from the boiler. This
"scrubbed" out the soot and uncombusted fuel associated with
oil-fired applications which not only reduced the power
plants particle emissions but also provided the 0.5 to 1%
carbon required in the process for reduction of MgSO..
Because of the necessity to prevent fly ash contamina-
tion of the recycled magnesia in the coal-fired application,
the incoming flue gas at PEPCO was precleaned (by ESP and/or
scrubber) to remove the fly ash. This also removed any car-
bon sources from the flue gas before it entered the second
stage.
- 50 -
-------
A supply of petroleum coke was located and added in
the required proportions with the calciner feed. This
was continued throughout the integrated operations at the
acid plant with Dickerson Station.
These initial operations at the regeneration plant also
saw the return of the "slide" problem which had been en-
countered and partially solved for the oil-fired application.
(A "slide" was a description of a rapid emptying of a part
of the calciner charge before decomposition of all the MgS03
in it). The slides caused overloading of the materials
handling equipment on the kiln discharge, when they occurred,
which forced a reduction in the kiln feed rate. Analysis of
the product after a slide showed several percent of undecom-
posed MgS03. The frequency of occurrence of slides was
eventually related to the size distribution of the entering
feed. It was shown that a high percentage of fines in the
feed triggered them. By adjusting the methods of operation
of the dryer to reduce the amount of fines in the dried MgSO^
during the oil-fired application calciner feed rates up to
80 Ib/min were able to be achieved.
A different type dryer than the one used in the oil-fired
boiler application had been designed for the PEPCO operation.
(A counter-current design dryer had been installed at Boston
Edison Company's Mystic Station while a co-current design
dryer was installed at Dickerson Station). This difference
in dryer types was one of the major equipment differences be-
tween the oil-fired boiler and coal-fired boiler applications
in the MgO slurry dewatering process.
Analysis of the product from the PEPCO dryer showed that a
higher percentage of fines were produced from the co-current
dryer and that the fines could not be controlled by the tech-
- 51 -
-------
niques used at Mystic Station. The slide problem limited
calciner feed rates to 40 Ib/min rather than the 106 Ib/min
design rate. Because of the interrupted operations of the
plant a dryer investigation could not be undertaken during
the program, and calciner feed rates were held at 40 Ib/min.
However, this rate was sufficient to keep the FGD operation
supplied with recycle MgO.
At Dickerson Station a shipment of virgin MgO was received
July 29th and the first regenerated MgO was returned August
16th. With this inventory of material the performance test-
ing phase was started to provide preliminary reliability data
and optimization information.
Tests conducted over the following two weeks checked SO-
removal efficiency prior to installation of the throat re-
strictor and tested a new configuration for MgO pre-mixing
using an eductor mixer. (The mixing test was inconclusive
because the unit obtained was too small and plugged). The
MgO slurrying system was returned to the original configura-
tion, an agitated pre-mix tank. The throat restrictor was
also installed in preparation for the series of performance
tests which were to follow.
The first six series of performance tests were started during
the last week of August and the test program continued through
the months of September with 400 hours of operation logged.
Test series conditions included full to half boiler load
with full to half design flow to the scrubber/absorber, and
inlet gas taken after the precipitators or with no pre-clean-
ing. Series designations are shown in Tables . A descrip-
tion of these tests and the results are presented in another
section of this report. Also incorporated into the foregoing
investigation was an analysis program of the first stage re-
cycle liquor, thickener and pond and river discharge flows
- 52 -
-------
TABLE 8
PERFORMANCE TEST DESIGNATIONS
Test
Series
Boiler
Load
(% of Design)
Scrubber
Gas
(% of Design)
Scrubbed Gas
Pretreated
5A
5B
6
7
8
100
50
100
100
100
100
50
50
100
50
E.S.P
E.S.P
E.S.P
NONE
NONE
- 53 -
-------
for a number of chemical species of environmental signifi-
cance.
Following these tests the plant was readied for the
balance of the optimization program. Emphasis was placed
on the assessment of the effect of coal fly ash on the re-
generation and utilization of the recycled magnesia. A
regular sampling and analysis program was initiated for flue
gas monitoring and chemical determination of all major stream
compositions. Over the next four months, system operations
totaling 1350 hours provided much information on the effect
of regeneration operations on the absorption of sulfur dioxide/
operation of the centrifuge and dryer and the conditions of
scrubber/absorber operation.
Attainment of the major goals of the project, multiple
recycles of magnesia and conclusive demonstration of system
reliability were thwarted by the deteriorated condition of
the piping which evidenced itself almost immediately after
entering this phase of the demonstration program. As the
performance testing was completed the first of many failures
in the rubber lined, first stage recycle piping occurred
forcing a shut-down of the system. During the rest of the
operations first and second stage pipe failure was the most
common shut-down cause. A summary of operations during this
period is given in Table 9.
Some improvements were obtained when several sections
of the second stage recycle pipe were replaced but leaks
in the other sections and the unrectified first stage problem
limited operating periods to days rather than weeks. The
longest continuous run during this period was eleven days
from December 13th through December 23rd. Operations re-
sumed on December 27th and continued into January 1975 with
tests of first stage operation alone. These tests also
- 54 -
-------
TABLE 9
SUMMARY OF OPERATIONS FGD SYSTEM
PERFORMANCE TEST PHASE
Week Ending
11/01
1V08
11/15
11/22
11/29
12/06
12/13
12/20
12/27
01/03
01/10
01/17
01/24
01/31
Goal
System Repair
Repair
Operate at design load
Establish dependable &
continuous operation
Continuous Operation
Continuous Operation
Continuous Operation
Continuous Operation
Continuous Operation
Continuous Operation
Operate 1st Stage only
Continuous Operation
Continuous Operation
Continous Operation
Problem Areas
Centrate Punp
Bucket Elevator (1)
& Piping Leaks
Dryer Fan Drive
Piping Leaks
Piping Leaks &
valve failure
Piping Leaks
Piping leaks &
valve failure
Piping leaks
MgO inventory
depleted
Piping Leaks &
Conveyor Jam
Piping Leaks
Number Of
Interruptions
1
2
1
1
2
1
2
1
1
1
1
Avg.
Load
60-80%
60-70%
70-80%
75-80%
60-80%
60-85%
60-90%
80-90%
90-100%
60-70%
50-60%
50-60%
Hours Of
Operation
0
0
88
84
123
77%
114%
159%
83%
53
78
75
30
37%
1003.5
1) Marginal bucket elevator capacity limited operation
to average 70% of design
OVERALL AVAILABILITY 46.1%
-------
provided data on first stage SO, removal capacity. A
final series tested recycle liquor flow effect on S02 re-
moval efficiency.
Operation of the regeneration plant during this period
was sporadic because of the outages at Dickerson and the
resultant interruptions in feed to be calcined. The major
mechanical problem encountered at the regeneration plant
was a failure of the pinion gear in the main drive on the
calciner (due to misalignment and lack of lubrication). A
replacement was locally fabricated and the unit was only
out of service for two weeks during December. A summary
of operations at the regeneration facilities is given in
Table 10.
During the program approximately 2247 tons of MgSO~
were processed to MgO at the acid plant.
During January 1975 plans were made for major over-
haul of the PEPCO FGD system including replacement of all
"thin" second stage pipe and repair of several sections-of
first stage rubber lined pipe. This work was scheduled in
conjunction with PEPCO1s planned 12 week overhaul of its
No. 3 boiler and turbine. Repair of the FGD system was
for the most part, completed by April. PEPCO, however, dis-
covered a major defect in the No. 3 turbine during their
boiler overhaul which necessitated a number of extensions
of their boiler outage. Eventually the outage extended to
the end of July 1975. The outage extensions created
secondary problems in the financing of the project and the
program was terminated.
- 56 -
-------
TABLE 10
SUMMARY OF OPERATIONS CALdNER SYSTEM
PERFORMANCE TEST PHASE
Week Ending
11/01/74
11/08/74
11/15/74
11/22/74
11/29/74
12/06/74
12/13/74
12/20/74
12/27/74
01/10/75
01/17/75
01/24/75
01/31/75
Maintenance on Bucket Elevator
Routine Maintenance
Routine Maintenance - Accumulate MgS03 Inventory
Routine Maintenance - Accumulate MgSO3 Inventory
START-UP
MgO Production for PEPCO
MgO Production for PEPCO
MgO Production for PEPCO
Maintenance on Calciner Main Drive and MgO
Production for PEPCO
Complete Repairs on Calciner Main Drive and MgO
Production for PEOCO
Run Test on Boston Edison MgSO, without Coke
Addition
Test PEPCO MgSO3 without Coke Addition
MgO Production for PEPCO
MgO Production for PEPCO
Operated For
Hours
0
0
0
58
47
87.4
24
8.9
100.9
44.2
4.5
7.9
- 57 -
-------
4.1.1 October 1974 - Details
Regular operation of the FGD system at Dickerson was
resumed after the performance test work on September 30.
This first operation was stopped after three days when pro-
blems with the MgO premix tank forced an outage. The sys-
tem was restarted at 1 p.m. on the 3rd and ran continuously
for the next 12 days at approximately 50 to 80% of design
load. A failure of a 14" rubber lined pipe elbow in the
first stage recycle loop forced another outage at 4 p.m. on
the 14th.
The analytical program was fully implemented by October
16, 1974. The scrubber ran from a start up at 12:00 a.m.,
October 16 to 9:00 a.m., October 18. A planned shutdown was
taken to repair several leaks in the first stage recycle
lines and to attempt to increase the flows in MgO slurry feed
system. The scheduled repair was completed by noon, October
18, however, after operations resumed low flow of MgO slurry
resulted in loss of pH control in the second stage. Leaks
in the MgO magnetic flow meters also caused delays. Flue
gas treatment resumed at 3 a.m. October 23, 1974.
During the night of the 24th problems again occurred
in the MgO slurry system with line plugging reducing pH in
the second stage to 6.6; however, the situation was corrected
and the system ran till 9 a.m. on the 25th when the unit was
shut down for a boiler tube leak. The boiler was repaired
by October 28, but the scrubber system was waiting for main-
tenance and was not returned to service.
4.1.2 November 1974 - Details
During the 30 day period, November 1 to November 30,
1974, the scrubber operated 318.9 hours out of a possible
720 hours for a 44.3% availability. During the period of
operation the unit ran at 50% to 100% of design gas load
- 58 -
-------
resulting in a total average available capacity for the
month of November of 34%. Because of scheduled outage
in the earlier portion of the month for leak repair and
overall maintenance, most running time was logged in the
latter portion of the month.
Scrubber Operation and Comments
The scrubber was brought on-line at 2 a.m., November
10, at 80% of design flow. Attempts to increase the gas
flow to full load were hindered, as problems in the MgO-
slurry feed system prevented pH control at increased loads.
The system operated steadily, adjusting the scrubber gas
flow to maintain a pH of 5.7, till 2:20 a.m. on the llth,
when the scrubber was taken off-line due to a leak at the
inlet of the second stage bleed line.
Repairs were completed at 8:00 p.m. on the 13th.
The system was returned to operation and the scrubber load
was varied between 60 and 80% to compensate for station
load changes during this period. On November 15, at 10 a.m.,
a reduction to 1/2 load was made to repair a leak in the
centrifuge. When the centrifuge was diverted, the solids
in the second stage recycle slurry reached a high level
and the low load was maintained until the solids level was
centrifuged down. At this time it was noted that the flow
to the centrifuge had dropped to 55 gallons per minute,
which prevented operating the unit at design gas loading.
This was corrected by operating with both the main and
spare pump. The plant continued in operation with the
scrubber load gradually increased to 100%., holding pH in
the 7.0 to 7.3 range. (Normal operating practice was to
control pH by maintaining a steady MgO feed, adjusting the
gas flow to maintain the pH set point affected by variations
in the coals sulfur content).
- 59 -
-------
A leak developed in a first stage recycle control
valve and the unit was taken off the line to make repairs
on the 16th. Operations resumed on the 19th and continued
through the 21st when the belt drive failed on the dryer
I.D. fan.
The fan repairs were completed and gas was put back
into the scrubber at 3:15 p.m. on November 23, but by 11 p.m.
the No. 3 boiler developed a tube leak, forcing a shutdown.
This was repaired and operations resumed by 2:40 a.m. on the
24th, six hours later another interruption was needed to re-
pair a bearing on the dryer product screw conveyor. The
system was returned to operation 2:45 that afternoon and
operated continuously through the end of the month. Operat-
ing adjustment had to be made during this period to accommo-
date generating station coal feeder problems.
During this final fifteen day period the scrubber
operated a total of 241.6 hours out of a possible 360 hours
for a 67.1% availability (51.7% average available capacity
based on a 77% of design gas flow).
4.1.3 December 1974 - Details
During the 31 day period, December 1, 1974 to
December 31, 1974, the scrubber operated 430.8 hours of a
possible 744 hours, an availability of 57.9%.
During the month of December various new studies were
initiated and the wet chemical method for the specific deter-
mination of MgS03.6H20 and MgS03.3H20 in the mixture was
used to determine the hydrate ratio in the centrifuge cake
samples. Previously, the ratio of MgSO.,. 6H20/MgSO_ .3H_0
had been approximately determined on the basis of percent
combined water (water of crystallization) as determined on an
OHAUS moisture balance.
Chloride in the system was also measured as a function
of operating time.
- 60 -
-------
Scrubber/Absorber Operation and Comments
The run which had started on November 24 had to be
aborted on December 1 after a leak developed in the rubber-
lined bleed line to the thickener area. Operations were
again resumed on December 5 and continued till 4:30 a.m.
on the llth when leaks in the first stage bleed line again
forced an outage. During this period adjustments in operat-
ing conditions were used to control problems of MgO feed,
station coal handling and centrifuge cake buildup in the
discharge hopper.
Operations were resumed at 6 p.m. on the 13th and
this run continued till the 23rd when leaks in both the
first and second stage discharge headers forced a shutdown.
The system was restarted on the 27th but the run was
aborted on the 29th after 64 hours of operation with another
recycle pipe failure.
4.1.4 January 1975 - Details
Measurements of the pipe wall thickness and observa-
tions of the pipe involved in the numerous failures which
had been encountered showed that a major system overhaul and
pipe replacement program was necessary if any'sustained
operations were to be achieved. PEPCO has rescheduled an
overhaul of the No. 3 generating unit to 1975. This was a
five year inspection and overhaul of the generator originally
due in 1973. Plans were made to recondition the scrubber
system during the boiler outage scheduled to start in January 1975.
Some additional test work was conducted with the pollution
abatement system during January while waiting for the planned
boiler shutdown.
These tests included four days of operation with first
stage (particle control) only, and another series of tests
to determine the effect of variation of recycle flow rate on
SO- removal efficiency.
- 61 -
-------
4.2 UTILITIES CONSUMPTION
Electric power consumption (in MW/Day) of the particle
control and flue gas desulfurization system installed at
Dickerson Station is given in Table 11 for the major operat-
ing periods of the test program.
Also included in the table are data on average fuel
consumption for the dryer used in the FGD system along with
an approximate gas flow processed during the period. While
there is some effect caused by the variation in fuel sulfur
content on the amount of centrifuge cake produced it appears
that a significant factor in dryer fuel oil consumption is
the weather. Fuel consumption is seen to increase during
the winter months and decrease during the summer months
while the gas flow processed through the system remained the
same.
4.3 MgO CONSUMPTION
Magnesia consumption during the fifteen month period
of operation of the plant was high. It is evident that the
loss rate during the period of initial operations was higher
than the loss rate during the latter planned operations
phase.
MgO shipments to the plant were made corresponding to
these two periods. An initial inventory of 357 tons of
MgO was shipped to Dickerson between August and December
of 1973 for the start-up operation. An additional 281.3
tons of virgin magnesia wereadded as make-up between July
and November 1974 for the planned operations phase. The
addition of make-up MgO at Dickerson between July and
November is given in Table 12.
- 62 -
-------
TABLE 11
UTILITIES CONSUMPTION
FGD SYSTEM
Major
Operating
Periods
April 25-30, 1974
June 13-16, 1974
Aug. 13-17, 1974
Sept. 3-11, 1974
Oct. 16-18, 1974
Nov. 24 - Dec. 1, 1974
Dec. 6-10, 1974
Dec. 14-23, 1974
Dec. 27-29, 1974
Jan. 12-14, 1975
Aug. 12-14, 1975
Aug. 22-26, 1975
Approximate
Gas Flow
% of Design
66%
75%
70%
67%
80%
70%
70%
Average
Electric Power
Consumption
MWH/Day
38.6
39.8
34.8
27.7
41.8
53.3
49.0
45.5
57.5
50.0
50.0
33.8
Average
Dryer Fuel
Consumption
GPM
0.57
0.39
0.33
0.38
0.72
0.93
0.95
0.91
0.96
0.93
0.54
0.32
- 63 -
-------
TABLE 12
MgO MAKE UP
FGD SYSTEM
Date
7/29/74
9/1/74
10/24/74
11/4/74
Quantity Of Virgin
MgO Received
67.6 Tons
73.4 Tons
67.4 Tons
72.9 Tons
281.3 Tons
- 64 -
-------
4.3.1 Losses - Preliminary Operations
Estimation of dryer product generated at Dickerson
Station based on shipping records showed 485 tons of MgSO-
were produced from the initial charge of 357 tons of MgO.
This accounts for only 55% of the dryer product that should
have been produced based on the design material balance.
An inventory evaluation also showed that 144 tons of
MgO equivalent were lost during the same period giving a
loss rate of 40.4% of the material fed. Since no magnesia
regeneration was being performed during the early period
the losses were essentially confined to the FGD system
site. Part of the loss can be attributed to a number of
unmeasured spills anddischarges during this period, some
of these were:
1) Material discarded when cleaning out the MgO
slurry system after a plug.
2) Material lost when draining the scrubber for
repair work.
3) Spills of dryer product due to material handling
equipment overloads.
Another major loss point was entrainment and overflow
from the second stage of the scrubber. The few chemical
analyses of recirculating slurry taken during the early
period showed a maximum of 8% magnesium sulfate in this
stream. Since a concentration of 15% to 18% MgSO. is ex-
pected in the equilibrium system the low concentration in-
dicated a large bleed (or leak) from the magnesia recycle
loop.
Chemical analysis of the fan sump discharge during the
performance testing conducted in August and September 1974
showed high levels of magnesium salts and solids in this
stream. Analysis showed MgS04 concentrations as high as
- 65 -
-------
10% again indicative of excessive entrainment. The
entrainment was probably the major loss point sustained
during these preliminary operations.
4.3.2 Losses - Planned Operational Testing Period
Corrective action was taken to reduce the entrainment
from the scrubber system by lowering the operating liquid
level in the second stage reservoir from 80% to 40% to re-
duce gas velocity in that zone. Analysis of the fan sump
pit discharge in early October 1973 showed much lower MgS04
concentration. Losses from this source were reduced to less
than 0.5% (based on Mg content of the fan discharge) during
the final three months.
4.4 MgO REGENERATION
MgO inventory recording was started at Dickerson in
October 1974 and the weekly inventory records are summarized
for the period of operation in Tables 13 and 14 .
Test work at Dickerson in August had been carried out
using regenerated MgO; however, the losses, which had occurred
before the development program was initiated, necessitated re-
plenishment of the inventory with virgin MgO. During September
and October 141 tons of MgO were received and 109 tons of re-
generated alkali were returned from Rumford for operation of
the integrated system. During October the Dickerson system
operated 420 hours of a possible 647 hours, consuming 80% of
the alkali available and producing 482 tons of dryer product.
Thus, at the start of November, 141.4 tons of MgO and
100.6 tons of MgS03 were on hand at Dickerson. Of the MgO,
approximately 65% had been regenerated at least twice. Be-
fore operations were resumed at Dickerson 73 additional tqns
of virgin MgO was received and added to the silo. Inventory
on November 8th, for the resumption of Phase II of the de-
velopment program, was 213.3 tons of MgO and 100.6 tons of
- 66 -
-------
TABLE 13
INVENTORY SUMMARY BY WEEK-FGD SYSTEM
Week Ending
11/1/74
11/8/74
11/15/74
11/22/74
11/29/74
12/6/74
12/13/74
12/20/74
12/27/74
1/3/75
1/10/75
1/17/75
1/24/75
1/31/75
MgS03
Produced
0
0
32.8
98.4
142.5
54.0
116.8
157.9
83.6
65.0
0
85.9
8.0
(Tons)
Shipped
19.5
0
20.9
35.4
141.2
141.4
35.1
108.6
114.6
0
0
88.5
76.3
43.2
824.7
Closing^
MgO
110.4
182.3
150.5
105.2
110.1
131.0
68.2
26.9
73.9
48.9
48.9
54.8
118.3
105.1
(Tons)
MgSO
56.6
56.5
68.4
131.4
123.3
14.0
95.9
145.1
96.2
161.2
161.2
158.5
90.2
71.5
-------
TABLE 14
INVENTORY SUMMARY BY WEEK-REGENERATION SYSTEM
Week Ending
11/1/74
11/8/74
11/15/73
11/22/74
11/24/74
1 12/6/74
£ 12/13/74
1 12/20/74
12/27/74
1/3/75
1/10/75
1/17/75
1/24/75
1/31/75
MgO
Produced
0
0
0
30
26.4
50.3
13.3
4.9
0
48.2
56.5
30.6
2.5
4.4
(Tons)
Shipped
11.9
0
0
0
91.7
35.5
27.8
15.7
0
12.8
0
53.8
71.0
0
Closing
MgO
31
31
31
61
11.5
26.3
11.8
0
0
35.4
90.9
67.7
1.5
5.9
(Tons)
MgSO3
44
44
44
8
45.1
52.6
62.1
156
233.4
126.4
0
7.6
85.0
135.5
-------
MgSO, (total MgO equivalent 250.6 tons). The approximate
inventory on December 27th, after 730 hours of operation,
was 73.9 tons of MgO and 329.6 tons of MgSO, (total MgO
equivalent 196 tons). The inventory difference of 54.6 tons
is attributed to process losses, approximately 150 Ib/hr.
This may be contrasted with 223.7 Ib/hr. loss sustained in
the final operational period of the New England S02 Control
Project. As was the case in the oil-fired application, most
of these losses occurred from uncontrolled spills and bleed
streams at the regeneration plant; however, during this
period, November to January the regeneration plant only
operated a total of 226.3 hours accounting for the lower loss
rate during the PEPCO project.
By the end of December the entire inventory had been
processed through an additional cycle. Some limited runs
were conducted in January with 142 hours of two stage opera-
tion at Dickerson which initiated a new cycle of regeneration.
At Rumford, 194 tons of MgSO, remaining from the New
England S02 Control Project had been in storage since July
1974. The first rail car of this material (80 tons) was pro-
cessed with dryer product from Dickerson during January 1975
in a rate check of the calciner feed capacity.(D
An examination of the statistical summaries for the
Dickerson operations reveals a trend of a gradual increase in
unreacted MgO in the recycle liquor similar to that seen in
operations at Boston Edison's Mystic Station.
Period MgO in Recycle (Avg.) Standard Deviation
11/10-11/29 3.81% 2.90
12/1-12/29 3.95% 3.10
1/11-1/27 5.79% 1.40
4.5 POST OPERATION FGD SYSTEM INSPECTION
In January 1975, before the FGD system overhaul, the
Due to lack of operations at PEPCO and eventual termina-
tion of the entire program the regenerated MgO never found use
at Dickerson Station.
- 69 -
-------
major pieces of equipment in the plant were inspected to
determine their condition. The scrubber vessel was found
to be in good condition with only a few, small internal
areas where its plastic lining had failed. The fan was
found to be in excellent condition as was the secondary
mist eliminator vessel. Most other equipment, vessels and
pumps in the system were in good to satisfactory condition.
Major failures and unsatisfactory conditions were noted for
the first and second stage recycle piping. A report of
the inspection is given in Appendix 1.
An operating history of the plant showing the number of
runs started and their duration is given in Table 15.
Problems which had an adverse effect on operability of
the FGD system were divided into several catagories which
are described in the following sections.
4.5.1 1st Stage Piping
Problems with some portion of the 1st stage piping sys-
tem were the most frequent cause of scrubber shutdowns and
accounted for about one-third of the total. There are five
separate, but interrelated causes of failure that have been
recognized, as follows: (a) pulsation in the piping near
the inlet to the scrubber, (b) high liquid velocities at
some points in the system, (c) the presence of sharp ob-
jects in the system due to corrosion of stainless steel,
(d) rubber lining thinner than specified, (e) poor adhe-
sion of rubber lining to the steel pipe.
(a) Pulsation in the piping near the inlet to the
scrubber caused several failures of the pinch
valve liners. The valves were finally replaced
with rubber lined spool pieces. The pulsation
also contributed to the failure of the elbows
just before the scrubber inlet in all three of
the 6-inch lines to the center nozzle. The cause
- 70 -
-------
TABLE 15
OPERATING HISTORY
FGD SYSTEM
Date
1973:
Sept.
Oct.
Nov.
Dec.
1974:
Jan.
Feb.
Mar.
Apr.
May
June
July
Aug.
Sept.
Oct.
Nov.
Dec .
Number Of
Runs Started
2
2
4
3
1
0
0
5
1
0
0
6
6
4
5
6
1975:
Jan.
TOTALS
3_
48
Duration Of Runs,
Hours
46
90
142
249
192
0
0
174
100
0
0
248
326
344
318
420
131
2,780
- 71 -
-------
of the pulsation was eventually traced to the
hydraulic design of the piping system. The
pressure drop between the high point in the sys-
tem and the scrubber inlet was less than the
vertical distance/ resulting in an alternate fill-
ing and emptying of the line. Corrective action
has been taken by the installation of restrictive
orifices in each of the inlet pipes, immediately
before entrance to the scrubber. Also, new tan-
gential nozzles have been designed for a higher
pressure drop—sufficient to compensate for the
head between the restrictive orifice plate and
the nozzle outlet.
(b) Excessive liquid velocities caused failures in the
1st stage bleed control valve and in a 316 stain-
less steel pipe section immediately after the con-
trol valve. The corrective action taken (some-
time prior to August 1974) was the installation of
three eccentric orifice plates in horizontal piping
runs in order to increase the pressure at the down-
stream side of the control valve, thus permitting
the valve to operate in a more nearly open position
with reduced velocities through (and immediately
after) the valve. The eccentric orifice plates were
designed for velocities that would be adequately
low for normal conditions. However, with the actual
conditions, as will be discussed in (c), (d) and
(e), failures of the rubber lining occurred adjacent
to and downstream from the eccentric orifice plates.
The plates were removed, leading to a recurrence of
failures in and after the control valve. Correc-
tive action has now been taken by the installation
of a concentric orifice flange at the short, vertical,
discharge end of the line.
- 72 -
-------
(c) Sharp pieces of metal were present in the liquid
cycle on numerous occasions as a result of cor-
rosion failures of stainless steel as will be
described in a subsequent part of this section.
The adverse effects were aggravated by conditions
to be discussed in (d) and (e).
(d) The rubber lining as supplied was thinner than
the 1/4" specified. This was substantiated in
December 1974 by micrometer measurements of rub-
ber samples from five locations, selected at ran-
dom from pipe sections that failed.
Readings, in inches, were as follows:
0.160
0.154
0.150
0.133
Average
In every instance the original "waffle" pattern
was still on the rubber and there was no evidence
of any erosion. Replacement pipe linings have
been a full 1/4-inch thickness and there have been
no failures of any of the replacement linings.
(e) Adhesion of rubber lining to the steel has been
poor. This has not been measured quantitatively
but has been judged qualitatively by pulling por-
tions of the lining from pieces that have failed and
also by inserting a knife or fingernail between steel
and rubber flanges. The significance of poor adhe-
sion is that it allows a small failure to grow rapidly
into a large one that cannot be controlled with ex-
ternal patches.
- 73 -
-------
4.5.2 2nd Stage Piping
Failures of second stage piping became quite extensive
in October 1974, and several replacements had to be made.
Portions of the 16-inch and 20-inch piping that wore were
repaired by applying a polyester or epoxy coating, rein-
forced with glass fabric to the thin area.
Failures are ascribed to a combination of erosion and
corrosion. For long-term commercial use it is planned that
rubber-lined piping will be specified.
An extensive series of piping thickness measurements
were made in late December 1974, using an ultrasonic gauge,
with the primary objective of establishing which portions of
the piping system needed to be replaced (or covered) in
order to provide for three months of additional operation.
4.5.3 Corrosion of Stainless Steel
During early operation of the scrubber there were a
number of failures of stainless steel elements fabricated of
type 304 or other types that are less corrosion resistant
than type 316. Type 316 has performed with mixed results,
depending on the specific environment encountered.
The first stage tangential inlet nozzles corroded and
failed. It was concluded that the cause of failure in this
case was the condensation of strong sulfuric acid from the
flue gas onto the outer surface of the pipes—a mechanism
similar to that which has been encountered in air heaters
when flue gas is cooled to too low a temperature. Replace-
ment nozzles were fabricated in both glass reinforced poly-
ester plastic and stainless steel coated with glass re-
inforced polyester. Stress has been another cause of failure
of type 316 stainless steel exposed to first stage liquid
as evidenced by extensive attack where threads had been cut
or cold bends had been made. This was particularly notice-
able in the case of bolts in the south thickener and hanger
rods supporting the first stage demister piping.
- 74 -
-------
Replacements for the latter were heat treated after fab-
rication in order to prevent attack from this cause.
In the absence of either undue stress or conditions that
permitted condensation of strong acid from flue gas, type
316 provided satisfactory performance. Examples are the hold-
down straps for the mist eliminators, where there was general
light pitting but no measurable loss of thickness and short
sections of piping to the tangential nozzles, just before en-
trance to the scrubber, where thickness was very nearly equal
to specifications and there was no visual evidence of corro-
sion.
- 75 -
-------
5.0 PERFORMANCE TEST RESULTS
Performance tests of the particulate emissions and SO2
control system installed at Dickerson Station were conducted
by York Research Corporation. The testing period extended
from August 26, 1974 through September 21, 1974 following a
preliminary period in September for equipment set-up,
familiarization, establishing gas velocity profiles and gas
flow measurement.
The performance testing used methods detailed in the
Federal Register, Vol. 36, No. 247, 1974 for:
a) Gas Analyses by Method 3
b) Particle Sampling Outlet by Method 5
c) S02 Removal Efficiency by Method 6
d) SO, Determination by Method 8
Particle sampling at the inlet was conducted in accordance
with ASME PT6-27 (modified).
Sizing of the collected particle samples was done micro-
scopically. This method is semi-quantitave as the actual de-
minsions measured are the diameter or length of the particle.
This gives a general indication of the particles size and a
mathematical approximation of its mass.
The data have been abstracted from York Research Corpora-
tion 's Report Y-8513 dated January 31, 1975.
5.1 PARTICLE EMISSIONS
Results of the particle emissions tests are summarized.
in Table ig. Each result represents the average of four tests
per series. Additional data are also provided on gas flow
and composition.
- 76 -
-------
TABLE 16
PARTICLE EMISSIONS TEST RESULTS
FGD SYSTEM
Test(1> Boiler Load ESP(2) Xip(3) Flue Gas Flow
Series MW in H2O 103 ACFM
5A 183 Yes 10.2 299
i SB 96 Yes 11.1 161
-j
-j
6 185 Yes 10.6 153
7 183 No 10.7 282
8 176 No 10.0 127
Inlet
Outlet
Efficiency
Inlet
Outlet
Efficiency
Inlet
Outlet
Efficiency
Inlet
Outlet
Efficiency
Inlet
Outlet
Efficiency
Tgmp.
242
126
241
116
240
136
245
118
232
CO- 0- Gr/SCFD
% * 12% C02
12.2 6.5 0.190
0.002
99.0
11.9 6.7 0.101
0.005
95.0
12.3 6.4 0.120
0.007
94.2
11.9 6.5 3.654
0.014
99.6
11.6 7.1 3.276
0.017
99.5
Gr/ACF
0.131
0.002
98.5
0.070
0.004
94.3
0.085
0.007
91.8
2.474
0.011
99.6
2.212
0.014
99.4
Average of four (4) tests per test series
Indicates whether the flue gas treated by the scrubber stage passed
through the electrostatic precipitator (Yes) or not (No).
First stage (scrubber) pressure drop.
-------
As noted in Section 4.1 the conditions over which the
tests were conducted were varied to provide data on a range
of operational situations, taking gas either ahead of or
after the precipitator.
These results show the low particle emissions attain-
able using the Chemico wet scrubber. In the tests taken
after the precipitator outlet particle emissions were less
than O.Olgr/SCF for all cases despite the low inlet loading.
For the full boiler load, full gas flow case (5A) fly ash
emissions were less than 4 Ib./hr. When the system was
tested without the precipitator scrubber particle removal
efficiencies of 99.5% plus were obtained.
As the installation was designed to process only one-half
of the total gas flow from the boiler the results noted above
do not represent the actual stack emissions as this consists
of both the contribution from the scrubbed flue gas and the
particle load remaining in the unscrubbed portion.
An estimate of precipitator efficiency can be obtained
from the test results by using the inlet loading to the test
series before and after the precipitator as a measure of the
inlet and outlet condition of the precipitator. This esti-
mate is shown in Table 17.
The improvement in the efficiency of precipitator
operations, as shown in Table 17, as the amount of gas flow
to it decreases was confirmed in a later series of tests
which showed an increase to 99% efficiency in the ESP when
the gas flow was divided equally between the scrubber and
the precipitator.
5.2 S0v EMISSIONS
J^
Results of the sulfur oxides emissions tests are sum-
marized in Table 18 and represent the average of four de-
- 78 -
-------
TABLE 17
ESTIMATE OF
ELECTROSTATIC FBECIPITATOR EFFICIENCY
Efficiency (Based)
Boiler Particle Load (Avg.) On 3.6 GR/SCFD Source
Avg. Load Inlet Outlet Inlet load (Test Series)
MW Gr/SCFD Gr/SCFD
183 3.654 7
183 0.190 94.8% 5-A
96 0.101 97.2% 5-B
- 79 -
-------
CO
o
TABLE 18
SOX EMISSIONS TEST RESULTS
FGD SYSTEM
Test
Series
5A
5B
6
7
8
pd)
in H2O
15.1
6.6
5.1
14.7
5,2
Inlet
779
1373
800
1418
1419
SO, (PPM)
Outlet
78
157
137
88
156
SO, Removal
^ %
90
88.7
82.9
93.9
89.0
SO,
Inlet
34.6
47.5
2.9
1.8
(PPM)
Outlet
3.56
3.31
0.64
0.41
(1)
Second stage (absorber) pressure drop.
*Test results abstracted from York Research Corporation,
Final Report, Y-8513, Jan. 31, 1975.
-------
terminations per series. Except for tests series 5A
most results for SO, removal were reasonably close to the
average, in test series 5A removal efficiencies measured
in the individual tests ranged from 83.5% to 96.8%.
In general, the trends noted by these results (reduced
efficiency at lower pressure drop and lower inlet SO- con-
centration) are similar to those seen in the oil fired appli-
cation of this prototype FGD method. The correlation de-
veloped from the New England S02 Control Project has been
used to analyze the data from this project.
SO., inlet and outlet concentrations were also deter-
mined during the performance test series and show a signi-
ficant reduction across the two stage scrubber for both gas
taken before the electrostatic precipitator and the gas
taken after the electrostatic precipitator cases.
Maryland State Department of Health and Mental
Hygiene 10.03.39 "Regulations Governing the Control of
Air Pollution in Area IV", maximum allowable emission
of particulate for all solid fuel burning installations
greater than 200 million BTU/hr. furnace are 0.03 GR/SCFD.
- 81 -
-------
TABLE 19
REMOVAL EFFICIENCY FOR PARTICLE SIZE RANGES
Test
Series
5-1
5-2
5-3
5-4
Average
Removal Efficiency
Above 5^M
99.0
99.8
99.0
98.8
99.2
1 to
97
99
99
97
98
5 /KM
.5
.2
.9
.5
.4
Below l/t
-------
TABLE 20
OQAL ANALYSES - COMPOSITE TO BUNKER #3
DATE From
To
Total Moisture
Dry Volatile
Dry Ash
Dry S
Dry Btu
M&AF Btu
DATE From
TO
Total Moisture
Dry Volatile
Dry Ash
Dry S
Dry Btu
M&AF Btu
1974
11-1
11-5
6.03
22.58
21.85
2.36
11,684
14,951
12-18
12-24
12.40
22.99
17.53
2.42
12241
14843
11-6
11-12
5.93
22.31
20.43
2.36
11,826
14,862
12-25
12-31
7.90
24.66
15.51
2.41
12711
15044
11-13 11-21 11-27 12-1
11-19 11-26 11-30 12-10
8.67
21.58
19.11
2.15
12,096
14.954
1975
1-7
8.06
24.26
17.05
2.42
12467
15030
7.50
21.14
17.77
2.15
12,399
15,078
1-8
1-14
8.37
24.74
15.34
2.20
12809
15130
7.95
22.44
21.06
2.25
11,686
14,804
1-15
-121
3.18
22.21
13.09
2.26
13492
15524
9.66
23.36
19.23
2.13
11,934
14,775
1-22
1-31
8.19
25.12
17.80
2.06
12477
15142
12-11
12-17
10.69
25.18
19.50
2.17
11,911
14,796
2-1 2-5
2-4 2-11
8.70 9.78
23.67 24.62
16.68 16.09
2.28 1.88
12553 12514
15061 14914
Randan Sample
Arsenic 13.6 ppm
Tellurium 1.4 ppn
Selenium 1.1 ppn
- 84 -
-------
6.0 PROCESS CHEMISTRY
6.1 FIRST STAGE (PARTICLE CONTROL) OPERATION
While most attention was focused on the S02 removal
section of the system some work was done to determine the
composition of the various important streams in the first
stage liquor loop. Tests were conducted during the perfor-
mance test work carred out in August and September 1974.
These tests involved analysis for a number of chemical
species in the streams going to the first pond/ the first
stage bleed, thickener underflow and overflow, and the sump
pump discharge. Components analyzed were calcium,
potassium, aluminum oxide, magnesium, chromium, cadmium,
titanium oxide, silica, sulfates, sulfites, and chlorine.
In addition, the pH and total dissolved solids were also
measured.
In order to complete the data and provide some standard
of comparision a series of baseline analyses was done for
both the inlet and outlet of the pond system and the various
system water supplies during a period of limited activity of
the FGD system, prior to the inauguration of the performance
test work.
The data for both the baseline period and the performance
test period have been abstracted for a number of streams and
arepresented in Table 21. In these tests the baseline data
aregiven where available, while the compositions during the
testing period (samples were taken intermittently for the
period August 29 through September 21, 1974) are presented
as an average for all tests days for the first stage bleed,
the pond (Basin #1) inlet, and pond return (Make Up Water to
Dilution Tank).
- 85 -
-------
TABLE 21
Chemical Composition at Various Streams in the
First Stage - Fly Ash Removal System
Sampling
Point
First
Bleed
Inlet
Basin
Make
Water
Stage
to
#1
Up
to
Potassium
ppm
Avg. All Tests 24. 6
Baseline n.a.
Avg. All Tests 12. 9
Baseline 16.4
Avg. All Tests 10.9
Calcium
ppm
7.
11.
8.
10.
8.
8
4
8
1
8
Magnesium
ppm
562.
78.
755.
83.
123.
0
1
7
7
2
Iron
PPm
87.4
38.0
41.3
69.2
.4
Cadmium Copper Chromium
ppm ppm ppm
.94 1.2
.01 .23 0.6
.04 0.44
.01 2.42 0.64
.04 .08 .13
Mercury
ppb
12.7
2.0
2.0
4.7
Dilution Tank
Service
Water
Supply
Baseline
n.a.
13.3
83.6
,07
.49
0.74
2.0
- 86 -
-------
Table 21 (Cont'd)
Chemical Composition at Various Streams in the
First Stage - Fly Ash Removal System
Sampling
Point
First Stage
Bleed
Inlet To
Basin #1
Make Up
Water To
Avg.
All Tests
Baseline
Avg. All Tests
Baseline
Avg. All Tests
Sodium
ppm
17.0
16.2
21.8
10.0
Flouride
ppm
1.4
1.0
1.5
1.0
1.9
SO.
ppm
4921.0
1.0
2024.4
408.1
597.8
Dilution Tank
Service
Water
Supply
Baseline
45.1
1.0
39.2
Phosphate
ppm
32.4
2.1
18.0
16.3
7.8
2.2
Cl
537.0
224.0
64.2
- 87 -
-------
6.2 PROCESS CHEMISTRY
In an aqueous envirpnment the following series of reactions
can be envisioned for S02 absorption by pulverized solid
MgO:
1. Slaking of MgO
nyu(s) + H2U(1)
Mr-r /OTM /rii-iln\ . .
S0_ Absorption
Of")
°°2 (g)
S02 (Soln) + H20 —
Formation of MgSO,
— ^MrrfnvH fQnlrO
+ 2 -
>Mn -4- 5OH
-*-gf^ f^r^l r\\
tail 4. IICO —
+ ^ -2
and Mg(HSO,),
\v-±)
(6-2)
(6-3)
I fi-4}
\\) 1 1
(6-5)
(6-6)
(6-7)
(6-8)
Mg(OH) (Soln) + HS0 + (x-2)
x = 3 or 6
Mg(OH)2 (Soln) + 2H-
MgSO, x H2O
Mg(HS03)2
4. Other reactions (side):
-j — >Mg(HS03),+2
+S0
,
( ^H20
2HSO3-
co2 w ^ "2
-CO- (Soln;
:O3 (Soln)
+ HC03-
+ C03-2
2 MgS03 x H20 + 0^—»2 MgS04 + x
Mg (OH) 2 (Soln) +H2C03+3H2->'MgC02.5 H20
Mg (OH) 2 (Soln) + SO^-»-MgSO4 + HjO
(6-8)
(6-9)
(6-10)
(6-11)
(6-12)
(6-13)
(6-14)
(6-15)
(6-16)
(6-17)
- 88 -
-------
6.2.1 Absorber Reactions
In the system the main reaction in the second or SO2
removal stage is the neutralization of H2SO3 by hydrated
MgO to form insoluble hydrated MgS03. MgS03 crystallizes
out with either three or six molecules of water. The
nature of the hydrate, i.e., hexa- or tri- is partially
dependent upon the temperature in the absorber. (Hexa-
hydrate is stable at room temperature while the trihydrate
is stable above 108°F). The transition temperature of the
two hydrates, according to the literature, is 42.5°C
(108.5°F). During operations at PEPCO the temperature of
the recycle slurry averaged about 110°F. Since this tem-
perature in the absorber is higher than the transition tem-
perature it was expected that the trihydrate of magnesium
sulfite would be the main product as was the case during
the work on the Boston Edison Project. However, in actual
operation, at Dickerson Station the centrifuge cake
analysis showed almost 95%
- 89 -
-------
6.2.2 Formation of Oxysulfates
At PEPCO feed MgO was digested in mother liquor that
has as its main consitutent magnesium sulfate in solution
(approximately 15% dissolved MgSO.).
It is known that MgO may react with MgSO. solution to
form a jell like substance. This jell consists mainly of
basic magnesium sulfate, represented as: xMgS04 yMg(OH)2
zH-O. The values of x, y, and z are dependent upon the
concentration of reactants and temperature.
During early operations at PEPCO, when virgin MgO was
used in the unheated mixing tank considerable jell and lump
formations occurred. X-ray and chemical analysis of these
products showed the presence of 3Mg(OH)2MgS0.8H20 (i.e.,
oxysulfate in varying amounts admixed with quantities of
unreacted Mg(OH)2•
Laboratory studies, which were initiated to investigate
this problem concluded that:
1) The problem appeared to be confined to operation
with virgin MgO particularly at high MgSO. con-
centrations (15% MgSO, or more) in the centrate.
The problem is further aggravated when the MgO-
MgS04 mixture is allowed to stand for a time with
insufficient agitation.
2) When calcined MgO (regnerated) of bulk density
above 20 Ib/ft is used in the reaction no major
problem was observed as long as proper agitation
and temperature of digestion of virgin MgO were
maintained.
Because of lack of operating time the laboratory obser-
vations with regnerated MgO could not be completely confirmed
in the plant.
- 90 -
-------
6.3 MAGNESIUM SULFITE CHEMISTRY
Fundamental questions in the Chemico Basic Magnesia
Slurry Process are the chemistry of the formation and de-
hydration of the hexa and tri-hydrate forms of magnesium
sulfite, and the mechanism of MgSO,-MgS04 thermal decom-
position in the presence of carbon and a reducing atmos-
phere.
Understanding of the first would allow control of the
production of the hydrate type formed in the absorption
system thereby affecting economies in the separation and
drying of these materials.
Knowledge of the second might lead to process improve-
ments for production of richer SO- off-gas, with more re-
active MgO, or direct conversion to elemental sulfur.
Because of the complexity of these questions research
work in the three areas was divided between the Chemico
laboratory facilities at the Rumford, R.I., plant and the
New Jersey Institute of Technology, (N.J.I.T.) Department
of Chemical Engineering and Chemistry. Most of the instru-
ment studies, i.e., Mass Spectroscopy, I.R. Spectroscopy,
Differential Thermal Analysis, Thermogravimetric Analysis,
etc. were performed at N.J.I.T. The results of these studies
are given in the Appendicies.
The following section is a summary of the principal
findings resulting from the work.
6.3.1 Conditions Governing the Formation of Magnesium
Sulfite Hydrates
A literature survey showed the existence of only two
hydrates of magnesium sulfite: MgS03.6H20 and NgSO^-
The hexahydrate is more stable at ambient conditions.
The hydration energy of the two are as follows:
- 91 -
-------
MgS03 + 6H20 ->MgS03.6H20 -13.38 K Cal.
MgS03 + 3H20 >-MgS03.3H20 -11 k Cal.
The N.J.I.T. study of the reaction of S02 (at 320°F)
with MgO slurry at 120°, 135°, 149° and 176°F showed that
at 120° and 135°F the major product was MgS03.6H20 while
the trihydrate, MgS03.3H20 was the dominant product at
149° and 176°F. Presence of MgS04 in solution (up to 18%)
and passage of excess air did not have any significant in-
fluence on the nature of the hydrate formed.
At Rumford, the reactions of S02 (at 300°F) with MgO
(virgin and regenerated) were studied at 135°, 140-145°
and 150°F. The pH was maintained at 6.8-7.20. Samples
were collected periodically and the filtrates were returned
to the reactor. The results of the study agreed with the
conclusions noted above.
At 135 F the major product was the hexahydrate and tri-
hydrate was the predominant one at 150°F. The study at
140-145 F was interesting—it showed the gradual change
from hexahydrate (original product) to trihydrate (final pro-
duct) . The complete transormation of the hexahydrate to tri-
hydrate took place in 4-5 hours of run. The study (140-145°F)
also showed that the transition of hexahydrate to trihydrate
is accompanied by a reduction of crystal size (from 200 to
50 micrometers).
Analysis of centrifuge cakes from PEPCO and Boston sup-
port the validity of the contention that (a) the ratio of
the two hydrates at any temperature is controlled by the
equilibrium value at that temperature, (b) hexahydrate per-
sists even above the transition temperature and (c) the
rate of transition is slow (cf. Kovachev Ts. et al. f Khimiya
i Industriya. Sofia 42 (5) (1970) 209-211. Transition
temperature of the hydrates, 40 C or 104 F).
- 92 -
-------
Centrifuge Cake From
Boston (135°F)
2-20-74 to 2-21-74
Centrifuge Cake From
PEPCO (120°F)
12-10-73 to 2-15-73
Hours
Ratio of 3H20
6H20
Hours
Ratio of 3H20
6H20
0 0/100 24 16/84
4 12.5/87.5 48 9/91
12 30/70 74 11/89
16 37/63 97 18/82
20 49/51 130 7.5/92.5
24 55/45
28 58/42
32 70/30
The variation in PEPCO samples was probably due to tem-
perature fluctuations in the system (the PEPCO system did not
stablize at the time. However, analysis of centrifuge cakes
at later dates always showed a preponderance of hexahydrate
in PEPCO cakes).
6.3.2 Dehydration of Magnesium Sulfite Hydrates
DTA, TGA and DSC studies with the two hydrates of mag-
nesium sulfite, MgSO~ 6H20 and MgS03 3H20 showed that the
path of dehydration and temperature of dehydration are depen-
dent on the manner of heating, i.e., whether dehydration took
place in an enclosed vessel or open or flow-through vessel.
(Appendix 2).
For example:
CONDITIONS
DTA Study
MgS0 6H
CLOSED
Two endotherms
observed
(at 107° and 205°C)
OPEN
One endotherm
only
(at 90°C)
- 93 -
-------
CONDITIONS
DTA Study CLOSED OPEN
MgS03 3H20 One endotherm only One endotherm
(at 205°C) (at 160°C)
TGA study also showed similar behavior of the two hydrates:
CONDITIONS OF STUDY
TGA Study CLOSED OPEN
MgSO, 6H-0 Two endotherms One endotherm
(at 125° and 220°C) only (at 70°C)
MgS02 ^H2° One en(*otherm One endotherm
(at 200°C) (at 100°C)
This study also showed that TGA (enclosed environment)
can be used to determine quantitatively the amounts of the
two hydrates in a mixture of the two. Presence of inerts
(any substance which does not change weight between 175°
and 400°C) does not interfere with the determination.
The only potential interferring element relevant to mag-
nesia process is MgSO. 7H2O (~7H2° at 200°c)• Tne ^n~
terference by MgSO, 7H~0 can be easily avoided by washing
the sample with absolute ethyl alcohol (only MgSO. &H20
dissolves). The amounts of the two hydrates, in a mixture,
are calculated as follows:
%MgSO. 6H90 = % wt. loss at 175°C x 2 x 100
50.9
%MgSO^ 3H,0 = 100 x (% wt. loss at 400°C - wt. % loss atl75°C)
J 34.10
Additional dehydration studies were carried out in the
laboratory.
- 94 -
-------
The two hydrates were heated in a vacuum oven (12
inches of mercury) at 80°C (176°F) , 120°C (248°F) and
150°C (302°F) for extended periods of time. Samples were
withdrawn at intervals and chemically analyzed.
At 176°F and 248°F little or no change was observed
with MgS03 3H20 even after 25 days. At 176°F the hexa-
hydrate changed to trihydrate after 2 days (confirmed by
chemical and x-ray analysis) . At 248°F the hexahydrate de-
hydrated to yield a hydrate of apparent composition
MgS032H20 after 21 days.
At 302°F the hexahydrate changes to trihydrate within
two hours. Holding the compound at this temperature results
in a continuous, gradual loss of water. After 200 hours at
302°F the starting hexahydrate had an apparent composition
of MgS03%H20. When starting with the trihydrate it appears
to be stable up to 48 hours at the 302°F temperature. Then
it looses water rapidly to give the apparent hydrate
The rates of dehydration of the two hydrates at 330°F
are also different the hexahydrate loosing water more
rapidly than the trihydrate. The rates of dehydration at
330°F were studies using Ohaus moisture balance. To form
the anhydrous salt, MgSO.,, containing 10% moisture at 330 F
following retention periods were necessary:
for MgSO,6H20 10 minutes
MgSO,3H20 25 minutes
The dehydration study indicates that the water molecules are
more strongly bound in the trihydrate than in hexahydrate.
6.3.3 Mass Spectroscopy of the Two Hydrates of Magnesium
Sulfite vs. Temperature
The two hydrates were heated from 35° to 400°C in a
- 95 -
-------
Finnigan Quadrupole 1015 mass spectrometer. No other
species except SO- and EJO were observed. The plot of
log (80) vs. temperature showed that the two hydrates
decomposed in a similar manner (under vacuum conditions) .
(Appendix 3) .
6.4 PULVERIZATION
The SO- removal capability of MgO in the Chemico
magnesia system is dependent on both intensive and exten-
sive properties of the regenerated material.
Of the extensive properties, the size range of the
MgO particles returned for slaking is very important. Ex-
perimental data indicate that grinding the calcined magnesia
increases its reactivity. However, there is a limit to which
that reactivity can be increased.
Laboratory data now indicate that a size range of -100
to +200 mesh (between 75 and 150 micrometers) is satisfactory
for recycled, regenerated magnesia. This is shown in the
attached graph (Figure 3) • The greatest increase in MgO
activity occurs when +50 mesh size particles are reduced to
-50 to +100 mesh, additional improvement in SO, removal is
obtained when size is reduced to -100 to +200 mesh. Reduc-
tion to finer size has only a slight affect on activity in-
crease.
The properties of regenerated MgO (shown in Table 22)
are determined by both calcining conditions and contaminent
level.
- 96 -
-------
FIGURE 8
EFFECT OF PULVERIZATION ON S02 EFFICIENCY
60
50
o
2
O
z
w
H
o
H
fe
fa
W
O
co
40
30
20
10
UPPER
X
50 100 200
SCREEN SIZE (TYLER MESH)
300 325
(1)
Measure of uptake of S02 by measured quantity of regenerated MgO by
the method described in "The Magnesia Scrubbing Process as Applied
to an Oil Fired Power Plant" EPA 600/2-75-057 pg. 253 ff.
- 97 -
-------
TABLE 22
PROPERTIES OF DIFFERENT FRACTION REGENERATED MgO
(SAMPLE NO. PR-289)
Analysis
+50 Mesh
Fraction
-50 to
+100 Mesh
Fraction
-100 to
+200 Mesh
Fraction
-200 to
-325 Mesh
Fraction
Ground +50*
Mesh Fraction
to (-100 to
+200} Mesh
%MgO
%MgS04
%MgS03
S02 Eff.
89.96
1.73
0.50
13
89.46
2.43
1.42
37
90. :2
2.60
1.33
47
91.73
2.35
0.98
48
48
- 98 -
-------
7.0 CORRELATION OF PROCESS DATA
With the initiation of the planned operations program
process data were collected and recorded regularly. In order
to assimilate, store, analyze and deseminate this information
efficiently, a computerized method of handling the data was
employed. As a result, it was possible to effect a real-time
feedback to the process or correlated and trend results at the
time they are most useful, along with accumulating the infor-
mation in a permanent data bank available for analysis and
retrieval. The statistical computer program package utilized,
permits the use of the same format data file as input to a
wide range of sub-programs, including regression calculations,
x-y plotting, file listings, trend plots, etc. This flexibility
reduced the number of files that needed to be maintained, and
provided speed advantages.
Figure 9 illustrates the flow of information between
scrubber and calciner operating systems and the computer data
bank. Operating conditions and analyses are entered on punch
cards on a daily basis for primary storage and transferred to
computer disk files for futher processing. At monthly inter-
vals the following statistics were generated for all variables:
average, maximum, minimum, standard deviation, and a percent
change relative to a base period. The final stage in the data
flow was the integration of the operating log and analysis
files, plus the inclusion of appropriate time lags for the
scrubber and regeneration facilities to form single disk files
suitable for input to correlation and plotter computer programs.
The process data bank becomes the source of retrieval of pro-
cess information for continued analysis and for comparison with
- 99 -
-------
o
o
PATA
T
T
RLE
Process DATA FLow PIA
-------
results from other installations.
The process correlations developed in the previous "New
England SO, Control Project", EPA Ref. CPA 70-114, were de-
veloped primarily using the data bank as input to a stepwise
multiple linear regression program. This approach minimizes
the number of computer runs required to determine equations
that are statistically significant, prior to their evaluation
for consistency with observed data. F ratio tests are em-
ployed to establish significance levels for testing of equa-
tion variables, and coefficients are calculated by least
squares techniques. With respect to the SO- removal efficiency
correlation, log transformations were required by the regres-
sion program. Continued investigation of mass transfer co-
efficient in this program may lead to the use of non-linear
programs.
Normal excursions in the process variables are sufficient
to satisfy the required scope of the test program variation
in levels of operating data to fulfill the requirements of
this method of operations data analysis.
7.1 S02 REMOVAL
A correlation relating S02 removal efficiency, pressure
drop and S02 inlet concentration had been developed during the
previous operation of the MgO system prototype on the oil fired
boiler. Included in the formulation of the correlation were
data points from the performance testing of Dickerson. The
data accumulated during the additional operation of this pro-
ject was used to improve the correlation based on the varia-
tions of S02 removal efficiency and other operating factors
which were observed.
7.1.1 First Stage S02 Absorption
A series of tests were performed with the scrubber to
assess the SO- removal capability of the first (particle re-
moval) stage of the system. Results of these tests are given
- 101 -
-------
in Table 23. These tests showed an average removal efficiency
for the first stage of 6.7%.
7.1.2 Effect of Pulverization of Regenerated MgO
During the period of scrubber operation at PEPCO from
November 1974 to January 1975, the observed S02 removal
efficiency varied more than could be expected from normal
changes in operating conditions, i.e., pressure drop, S0_ in-
let concentration, etc. Efficiency decreased approximately
10% in November before recovering in December and January.
The efficiencies were adjusted to constant operating condi-
tions and plotted to show that a significant variation in SO^ removal
remained after adjustment of pH and pressure drop which was
assumed to be related to other process variables. Figure 10
is plot of removal efficiency vs. time illustrating this
phenomenon. The scale of the plot is expanded to show the
effect more clearly.
Laboratory studies indicated that a critical particle
size range was required to maintain activity of the regenerated
magnesia. In order to determine whether any relationship
existed between MgO particle size distribution and S02 removal
activity in the FGD system, a series of MgO samples were taken
from the weigh feeder at Dickerson for screen sizing and
activity measurements. Results of this analysis are shown in
Table 24 along with corresponding scrubber operating condi-
tions for the time frame over which the samples were taken.
This information was analyzed by linear regression tech-
niques and showed a significant particle size correlation with
SOj removal efficiency in the -50 to +100 mesh range. The
correlation is given in equation 7-1.
R =-7.68 - 0.0338 (P1QO)2 (7-1)
R = % reduction in S02 removal efficiency
material
MgO Feed
Where PIOQ = % (-50 to +100 mesh) material in
- 102 -
-------
Date
1/8/74
1/9/74
TABLE 23
0 REMOVAL IN FIRST STAGE
Time
1700
1900
2100
2300
2100
0300
0500
0700
0900
1100
1300
1500
Inlet
ppm
1300
1300
1390
1430
1300
1390
1500
1300
1070
1080
980
900
so2(1>
Outlet
ppm
1220
1290
1200
1340
1300
1360
1380
1300
990
900
900
800
Efficiency
%
6.2%
0.8
13.7
6.3
0.0
2.2
8.0
0.0
7.5
16.7
8.2
11.1
(1)
Average Removal Efficiency 6.7%
Abstracted from the continuous emissions monitoring
data obtained using a Flourescent SO- analyzer.
- 103 -
-------
FIGURE 10
o
*>.
A1*
D
J.
R
E.
M.
E
F
F.
19
10/.
4<
100 •
94 .
88 •
82 •
76 •
74
23 11/10 11/15 11/20 11/25 ll/
D 66 92 118 144 17
•
•
•
•
•
1
•
•»»••-••»
•*
•
• ••
•
• •
•
•
•
• •
• •• •
••
• •
•
•
•
•
••
• •••
•
•
•
•
•
•
•••
»••••••»
• •
• •
•
••
28 ll/
CLJLaS
• •
•
• ••
•
•
• •
•
•
• •
• •
•
'30
•
••
• ••
*•••
•
r '••
% SO0 REMOVAL EETICIEHCY VS. TIME 1975
12/7 12/9 12/14 12/17 12/20 12/22 12/29 1/13 1/25
2?0 26Q 29Q 320 350 380- 410 440 470 500 HRS./2
•
•
•
•
•
•
•
• •
•
•
• ••
• • •
•
•
••
•
•
•
•
•
•
• •
• ••
••
••
••
•
•*
••«»•••••
••
•
•
•
•
•••
• ••••
• •••• •
•• •
•
•
•
•
•
••
•• • •
•••• •
•
*
»••••••••
•
••
• •
• • ••
••
•
•
• •
••
•*
• *
• • •
•
•
•••
•
•
•
• •
• ••
•
•
•
• ••
•• •
•
•
•
•
•••••••••»
*Observed SO2 ren. eff. ad-justed to conmon basis of 9"^> P, 7 pH
1900 PPM Inlet SO,.
-------
TABLE 24
WET SCREEN ANALYSIS OF MqO BELT SAMPLES
Date (1974)
•Time
Tyler Screen
Mesh Size - %
+50
-50,+100
-100, +200
-200, +325
-325
£ Operating
w Conditions
1 P, In. H2O
SO2 Removal
Eff.,%
SO, Inlet
Cone. , PPM
Predicted (2)
SOj Removal
Efr. - %
(Actual-Predicted)
Efficiency/ %
11/20
1.0
0.8
2.6
4.1
91.5
(1)
9.5
85.1
1210.
85.0
0.1
11/21
0.8
0.5
1.5
2.2
95.0
(1)
9.3
86.2
1240.
85.0
1.2
11/24
0.5
0.6
1.3
2.7
94.9
(1)
8.9
83.8
910.
84.9
-1.1
11/25
0.3
0.8
1.7
3.3
93.9
(1)
8.2
84.1
920.
84.6
-0.5
12/1
4.6
14.5
8.2
6.9
65.8
(1)
6.4
76.0
1080.
76.4
-0.4
12/9
0800
5.2
5.4
12.2
11.8
65.4
4.2
79.0
1170.
79.5
-0.5
12/18
2000
3.6
5.7
10.3
13.8
66.6
8.4
82.0
1200.
83.5
-1.5
12/20
1200
6.6
6.8
13.3
10.1
63.2
3.6
80.6
960.
77.0
3.6
12/23
0400
0.7
1.9
7.7
13.6
76.1
8.5
83.6
840.
84.6
-1.0
12/27
1200
1.
2.
3.
2.
89.
5.
83.
900.
82.
0.
8
5
6
5
6
9
0
9
1
(1)
Average Conditions For Day
(2)
Fr°m Equation 7-1
-------
After determining the pulverized MgO particle size the
resultant correction can be subtracted from the efficiency
obtained from the correlation developed from the previous
operations. This basic correlation is given in equation 7-2.
S02 Removal = 1-Exp ((2.666 (AP)A (S02I)B(10)C -3)) 7-2
Efficiency (Base)
Where A = -1.014
B = -3.75 + 0.271 In S02I
C = 6 - 0.031 pH
SO~T - SO- inlet concentration (ppm)
AP = pressure drop, inches H-O
Figure 11 is a plot of the.predicted effect of the con-
centration of -50, +100 mesh size in the MgO on S02 removal
using equation 7-1 at constant process conditions. Results
show that SO- removal decreases with increasing concentration
of this size range, which is consistent with laboratory obser-
vations. Figure 12 is a plot of observed vs. predicted
efficiency during the subject time period.
7.1.3 Reduction in Activity of Magnesia
Equation 7-1, showing particle size effect also predicts
a 7.68% offset from removal efficiency calculated from 7-2,
a correlation based on AP, pH, and inlet S02 concentration.
An analysis of inventory records during the period covered
by the testing program revealed that the magnesia feed consis-
ted of a mix of virgin and regenerated MgO. The beginning in-
ventory on October 18th at Dickerson was 46 tons of regenerated
MgO in the following week 140 tons of virgin MgO were added as
makeup to the silo. About 22 tons of regenerated MgO were re-
turned to Dickerson in the period from October 18th through
November 29th while about 120 tons of MgO were consumed in the
operations there. Thus during the period of reduced efficiency
virgin material was the principal feed to the system. A
comparison of the manufacturer's analysis, Table 25,
- 106 -
-------
FIGURE 11
w
H
o
H
w
1
a-
O
CO
86
84
EFFECT OF MgO PARTICLE SIZE ON SO., REMOVAL EFFICIENCY
BASIS; PEPCO Operation Nov.-Dec. 1974
Adjusted to 7" A P, 7 pH, 1000 PPM S02 Inlet
82
80
78
76
74
6
10
12
14
% (-50, +100) MESH MgO
- 107 -
-------
FIGURE 12
88
86
U
§ 84
M
U
W
f^4
g
o
W
Q
W
EH
U
82
80
78
76
74
OBSERVED VS. PREDICTED S00 REMOVAL EFFICIENCY
^ £
74
76
78
80
82
84
86
88
OBSERVED SO2 REMOVAL EFFICIENCY - %
- 108 -
-------
TABLE 25
MANUFACTURERS ANALYSIS OF MAGNESIA
SHIPPED TO DICKERSON STATION
CaO
SiO2
R2°3
Cl
so3
-325 M
LOI
S.A.
7/12/74
L&N200196
1.62
0.7
0.51
0.18
0.69
99.75
6.22
66.7
8/20/74
L&N200487
1.78
0.60
0.54
0.26
0.78
99.09
5.46
52.4
10/7/74
L&N200257
1.94
0.61
0.43
0.12
0.67
98.71
4.33
79.3
10/25/74
L&N200304
1.96
0.62
0.70
0.19
0.64
99.61
4.57
94.7
Where R2^3 = Iron an(* aluminum oxides
-325 M = Percent less than 325 mesh
LOI = Loss on ignition
S.A. = Surface area
- 109 -
-------
for the virgin MgO supplied during the period of the
program showed little deviation from that supplied at the
beginning of the Development Program.
It has been noted (Sec. 6.3.4) that analysis of
materials from the MgO slurrying system in the earliest phase
of the project showed that the highly reactive virgin MgO
exhibited a stronger tendency to form magnesium oxysulfates,
which would hinder the S02 absorption efficiency of the
magnesia slurry. It is thought that the deviation noted
here is a measure of the additional reduction in efficiency
resulting from the formation of the complex when a feed of
virgin MgO is admixed with centrate.
7.2 CENTRIFUGE OPERATION
Centrifuge operations at both PEPCO and Boston Edison
were compared and analyzed with regard to process optimiza-
tion. The effect of centrifuge feed rate on efficiency and
solids removal rate was determined by regression analysis of
the operating data. Since the two centrifuges are mechanically
identical, it would be expected that similar operating charac-
teristics would be observed, and this was confirmed by analy-
sis. The following results were obtained:
1. Centrifuge efficiency, i.e., pounds solids re-
moved in centrifuge cake per pound solids in
centrifuge feed, was shown to be a function of
centrifuge feed rate with efficiency declining
with increasing throughput.
2. Solids removed from the system was shown to pass
through a maximum value as total feed rate to the
centrifuge is increased.
Figures 13 and 14 show these results for centrifuge operation
with average recycle compositions, normal rotational speed
and standard weir height.
- 110 -
-------
FIGURE 13
EFFECT OF CENTRIFUGE FEED RATE ON SEPARATIONAL EFFICIENCY
80 --
dP
B
B
W
H
u
H
h
Cn
W
§
H
EH
H
(0
70 - -
BOSTON
PEPCO
60
50
40 --
30
40
60 80 100
CENTRIFUGE FEED RATE GPM
120
180
160
-------
m
O
03
s
Q
>H
as
80
70
g 60
S
50
W 40..
30
20
10
FIGURE 14
CENTRIFUGE OPERATION
LB./MIN. SOLIDS SEPARATED VS. FEED RATE
BOSTON EDISON
50
100 150
CENTRIFUGE FEED RATE (GPM)
200
-------
7.3 DRYER OPERATION
Data from the operations at PEPCO were analyzed to
determine the influence of dryer operation conditions on
dryer efficiency. A scatter plot of dryer solids vs. pro-
duct temperature is shown in Fig.15. In addition, the data
collected during the Boston Edison operation were analyzed
and compared. The results of preliminary review of the
effect of dryer product and dryer outlet gas temperature
on product moisture content is shown in Fig. 16 . Variations
in the performance of the dryers is attributed to the use
of a counter-current unit at Boston and a co-current unit
at PEPCO. A statistical analysis of this data shows a
minor interaction between dryer product and dryer outlet
gas temperature.
An analysis of the change in the ratio of MgSO. to
MgS03 in the dryer feed and product streams, an indication
of oxidation accurring in the dryer, showed a slight decrease
in the MgSO^ concentration for PEPCO operations as compared
to the oil fired boiler application for the average range of
system operations. This is shown in Figure 17 .
7.4 CALCINER FEED RATE
A limitation on reliable operation of the integrated
plants was reduced calciner feed rate. A low (40 Ib/min.)
feed rate was necessary to control slides and unprocessed
MgS03 in the calcined product, Fig.18. Initial investigations
during the work at Mystic Station indicated a relation be-
tween particle size of the dryer product and frequency of
slides. Further investigation showed that variation in size
distribution of the dryer product was associated with varia-
tion in moisture content of the product. During the final
phases of the program at Boston the dryer was operated to
produce a product with a moisture content of approximately
- 113 -
-------
FIGURE 15
100
in
Q
M
ij
O
en
Q
<#>
90
80
70
100
COMPARISON OF PEPCO AND BOSTON EDISON DRYER OPERATION
(FOR RANGES OF DRYER GAS OUTLET TEMPERATURE)
PEPCO DRYER
BOSTON EDISON DRYER
150
200 250 300
DRYER PRODUCT TEMPERATURE,°F
350
400
-------
1014.33 —
97.06
89.80 _
82.53 —
I
<*>
DRYER OPERATION
PEPCO DATA
3/31/75
75-27
68.66.
210.00
2U2.66
28l«. 96
I
326.00
368.00
U16.66
OUTLET GAS TEMPERATURE °F
- 115 -
-------
FIGURE 17
COMPARISON OF MqS03 OXIDATION IN DRYER
EH
U
D
Q
§
On
OH
U
S3
H
16
14
12
10
8
n
O
en
O
I6
BOSTON EDISON DATA
PEPCO DATA
t l_
2 4 6 8 10 12 14 16
MqSO.
MgSOj
IN DRYER FEED
- 116 -
-------
FIGURE 18
CALCINER OPERATIONS ON PEPCO MgS03
EFFECT OF FEED RATE ON MqS03 IN PRODUCT
40
U
I
O
CO
3
OP
30
20
10
45
50
55
60
65
CALCINER FEED RATE LB/MIN.
- 117 -
-------
16% which eliminated the slides in the calcination step.
The investigations at Dickerson showed there was a signi-
ficant difference in the size distribution of the product
produced from its dryer compared to material produced at
Boston as shown in Tables 26 and 27.
Table 26 shows the average values and standard devia-
tions for 25 Boston dryer samples and 22 PEPCO dryer sam-
ples covering the following mesh size ranges: +25, -25,
to +50 -50 to +100 and -100 to +200. Table 25 presents
a statistical analysis of these results.
- 118 -
-------
TABLE 26
SIZE DISTRIBUTION OF DRYER PRODUCTS
Mesh Size
Range
+25
-25, +50
-50, +100
-100, +200
BOSTON AND
Mesh Size
Range
+25
-25, +50
-50, +100
-100, +200
BOSTON
25 Samples
Average Std .
58.1
14.9
9.4
4.7
TABLE 27
COMPARISON OF
PEPCO SIZE RANGE
Significance
Level
99% +
99% +
99% +
99% +
PEPCO
Deviation Average
10.2 22.3
4.5 29.5
4.4 23.5
2.0 11.1
AVERAGES
Difference
(Boston-PEPCO)
35.8
14.6
14.1
6.4
22 Samples
Std. Deviation
11.4
8.5
8.4
4.0
99% Confidence
Limits
27.2-44.5
9.2-19.9
8.8-19.4
3.9-8.9
- 119 -
-------
8.0 DATA
8.1 MONTHLY AVERAGE OPERATING CONDITIONS
Operating data are summarized as montly averages for
the planned operational testing period. Included in this
information are:
Table 28 Operating Conditions for the FGD System
Table 29 Stream Compositions for the FGD System
Table 30 Operating Conditions for the Regeneration
Unit
Table 31 Stream Composition for the Regeneration
Unit
8.2 DATA LISTING
Data recorded from the PEPCO operations for the
period October 15, 1974 through the end of the program in
January 1975 are given in Table 32 . Data were recorded at
two hour intervals, which is the interval between successive
listing in the table for extended runs.
- 120 -
-------
TABLE 28
OPERATING CONDITIONS FOR THE FGD SYSTEM
MONTHLY
VARIABLE
S02 (PPM)
IN
OUT
% REMOV
PH
POWER PLANT
RATE (MW)
DIFF. PRESSURES
(IN. H20)
TOTAL
MIST
DRYER
TEMPERATURES (F)
-ABSORBER
INLET
MAGNESIA SLURRY
-DRYER
GASOUT
PROD.
FLOWS (GPM)
CIRC.
CENTFD
DATA AVERAGES
11/74
1057.86
193.46
81.54
7.04
174.03
7.49
0.23
0.76
109.26
157.02
274.65
179.36
10152.78
72.74
12/74
1061.95
207.90
80.57
7.06
153.98
6.77
0.19
0.69
110.88
139.20
311.03
200.62
10823.13
83.09
1/75
1272.22
214.89
81.95
7.11
142.82
6.49
0.21
0.80
108.54
130.33
254.11
165.30
11546.30
154.61
CENTRIFUGE
TORQ.
34.75
40.38
43.94
- 121 -
-------
TABLE 29
STREAM COMPOSITIONS FOR THE FGD SYSTEMS
MONTHLY
VARIABLE
DRYER
% SOL.
% MgSO3
% MgS04
% MgO
CENTRIFUGE
% SOL.
% MgS03
% MgS04
% MgO
RECYCLE
% SOL.
PH
-FILTRATE
% MgS04
-CAKE
% MgO
MOTHER LIQUOR
% SOL.
-CAKE
% MgO
DATA AVERAGES
11/74
93.63
68.06
8.28
4.78
88.86
41.78
4.44
3.06
6.98
7.17
17.05
3.81
2.46
6.14
12/74
96.94
68.12
9.08
6.82
88.33
42.81
4.86
3.03
10.90
7.30
18.95
3.95
4.80
5.68
1/75
91.30
66.45
7.70
7.78
88.45
42.19
4.02
5.25
8.72
7.22
13.13
5.79
2.20
9.73
- 122 -
-------
TABLE 30
OPERATING CONDITIONS FOR THE REGENERATION UNIT
MONTHLY DATA AVERAGES
VARIABLE 10/74 11/74 12/74 1/75
CALCINER
-TEMPERATURES (F)
MDKILN 1260.59 1361.25 1451.97 1494.86
GASOUT 571.75 534.04 544.39 582.89
PRODEX 224.72 160.00 155.22 169.74
-SOLID FLOWS (PPM)
MgS03 50.20 51.62 56.61 55.40
COKE 0.43 0.32 0.30 0.31
NEUT. pH LIQ. 5.25 5.83 6.01 6.40
ACID PLANT
FEED GAS (%)
02 5.76 7.34 7.57 7.40
- 123 -
-------
TABLE 31
STREAM COMPOSITIONS FOR THE REGENERATION UNIT
MONTHLY DATA AVERAGES
VARIABLE 10/74 11/74 12/74 1/75
CALCINER FEED
% H20
% MgS03
% MgS04
% MgO
% CARBON
11.20
64.08
7.89
5.80
2.66
11.38
67.47
7.25
4.94
1.35
6.00
63.78
9.25
7.69
1.76
6.45
68.42
8.63
6.28
1.32
CALCINER PRODUCT
% MgS03 0.45 0.15 0.92 0.46
% MgS04 0.90 1.29 1.08 3.74
% MgO 89.78 94.62 93.57 93.41
BULK DENSTIY 27.61 22.73 22.41 21.31
- 124 -
-------
TABLE 32
RUN DATES AND DATA LISTINGS
RUN DATES
STARTING
10/15/74
11/24/74
12/6/74
12/14/74
12/27/74
1/12/75
ENDING
10/18/74
12/1/74
12/10/74
12/23/74
12/29/74
1/14/75
CASE NO.'S
1-31
128-214
218-277
279-384
386-415
421-451
The following identifies the abbreviated words on
Pages 126 - 133:
S02 IN PPM
% S02 REM
D GAS T
D PROD T
CIRC GPM
FAN AMPS
Parts Per Million Of SO- Into Scrubber
Percent S02 Removed In Scrubber
Dry Gas Temperature
Dry Product Temperature
Recycle GPM
Current required To Drive Fan
- 125 -
-------
CASE NO. i S02INPPM 2 isoz REM 3 PH
4 DELTA P 50 OAS T 60 PROD T 7 CIRC 0PM 8 FAN AHPS
N>
at
1
2
3
4
5
6
7
8
9
10-
1L
12
13
14
15
16
17
18
19
20
21
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
«3
44
45
46
4T
48
49
50
51
52
53
54
55
56
57
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING-
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING--
MISSING
MISSING
MISSING
MISSING
MISSING
U ¥ CC T Mf
HISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
**ISS[NG
MISSING
MISSING
MISSING
"ISSING
MISSING
"ISSING
"ISSING
MISSIN6
MISSING
MISSING
"ISSING
"ISSING
"ISSING
11 10.0000
1760.0000
1?60.0COO
1360.0010
"I5SIMG
MIS5:*JG
"ISSING
[?60.0000
I?™. 0000
IZC0.0090
1360. 3000
"ISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
- MISSING-
MISSING
MISSING
MISSING
MISSING
MISSING
- MISS-ING- —
MISSING
MISSING
MISSING
MISSING
MISSING
tl T C C T U^
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING .
MISSING
MISSING
MISSING
MISSING
MISSING
BUSSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
94.5900
9'. 8 100
93.9100
•n.eioo
"ISSING
*!SSiNr.
-ISSIN'J
3s. 2400
3<>.OOOQ
fl'.OOOP
3*. 6700
MISSING
7.3000
7.2000
7.2000
7.0000
6.9000
7.0000
.8000
.9000
.9000
.9000
.9000
.8000
.9000
6.9000
6.9000
T 1 A A A
" * i U 0 n
7.1000
6.9000
7.0000
7.0000
7.1000
7.0000—
6.8000
6.7000
6.9000
6.7000
6.9000
6.8000-
7.1000
7.1000
7.0000
6.3000
6.3000
6.9000
6.9000
6.7000
6.9000
6.9000
6.9000
7.0000 -
6.8000
6.9000
7.3000
7.3000
7.3000
7.4000
7.3000
MISSING
"ISSTNG
T.?COO
6.6000
7.1000
6.9000
6.9000
6.7000
6.8000
6.3000
4.0000
9.0000
8.6000
9.0000
6.0000
5.6000
5.5000
5.0000
5.5000
5.5000
5.5000
5.5000
5.5000
5.5000
6.0000
T...AAAA
TWUVH
5.5000
6.0000
5.2000
5.5000
5.5000
ff C A A A
-- - 5.5000
5.5000
4.0000
5.5000-
3.5000
'4.0000
4.5000
3.5000
3.5000
3.3000
7.0000
4.0000
9.0000
8.0000
8.6000
8.6000
8.6000
8.6000
8.6000 --
7.0000
7.0000
7.0000
8.5000
10.5000
13.4000
13.2000
MISSING
MISSING
2.5000
1.4000
1.5000
3.0000
3.0000
4.8000
2.0000
6.5000
MISSING
MISSING
MISSING
MISSING
240.0000
300.0000
380.0000
330.0000
280.0000
MISSING
MISSING
MISSING
MISSING
150.0000
145.0000
160.0000
160.0000
170.0000
10000.0000-
9800.0000
10000.0000
9800.0000
9800.0000
10000.0000
10000.0000
10000.0000
10000.0000
-270-.0000 — -'175vOOOO — JOOOOvOOOO
240.0000
240.0000
255.0000
220.0000
280.0000
-32Or«000-
275.0000
280*0000
280.0000-
280.0000
360.0000
260.0000
220.0000
230.0000-
225.0000
225.0000
-225.0000-
235.0000
235.0000
220.0000
MISSING
MISSING
210.0000-
220.0000
250.0000
240.0000
235.0000
235.0000
220.0000-
210.0000
210.0000
210.0000
225.0000
225.0000
210.0000
260.0000
MISSING
MISSING
265.0000
220.0000
270.0000
300.0000
300.0000
2T5.0000
290.0000
"ISSING
175.0000
165.0000
- 165.0000
160.0000
170.0000
1 O^ Aft Art
1*93* Uv VU
180.0000
180.0000
- 175-iOOOO
170.0000
170.0000
165-.0000
185.0000
175.0000
-165-.0000
160.0000
160.0000
1-65-sOOOO-
165.0000
165.0000
155.0000
MISSING
MISSING
- 155.0000
165.0000
170.0000
175.0000
175.0000
175.0000
— HO. 0000- •
160.0000
160.0000
160.0000
160.0000
168.0000
177.00CO
180.0000
MISSING
MISSING
MISSING
165.0000
170.0000
185.0000
185.0000
190.0000
175.0000
MISSING
10050.0000
9900.0000
9900.0000-
9900.0000
9900.0000
9800.0000
9800.0000
—9900.0000 -
MISSING
MISSING
M T **^ TMi*
MJ 55 trio
MISSING
MISSING
MISSING-;-
MISSING
MISSING
--- MISSING- -
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSIN& -
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
"ISSING
MISSING
-ISSING
MISSING
MISSING
MISSING
•'ISSING
12400.0000
190.0000- -
250.0000"
230.0000
230.0000
200.0000
200.0000
200.0000
200.0000
200.0000
--200.0OOO
200.0000
190.0000
- 200*0000
200.0000
200.0000
VAfl .Aft 01)
200.0000
200.0000
200.0000
200.0000
200.0000
~*AA AAAA
— cOuTvtrvv
200.000O
190.0000
-200*0000
190.0000
MISSING
l-9OrOOOfl
180.0000
180.0000
MISSING —
210.0000
180.0000
225.0000
225.0000
220.0000
220.0000
220.0000
220.0000
220.00OO
220.0000
220.0000
220.0000
220.0000
240.0000
280.0000
280.0000
150.0000
150.0000
160.0000
150.0000
150.0000
165.0000
175.0000
190.0000
150.0000
220.0000
-------
10
vj
NO.
53
59-
60
61
62
63
64
65
66
67
66
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83-
84
85
86
87
88
89
90
91
92
93
94
95-
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
11*
115
116
117
118
'« Z
1 S02INPF-
MISSING
— MISSING- -
MISSING
MISSING
MISSING
MISSING
MISSING
11*0.0000
1200.0000
1?00.0000
870.0000
1080.0000
MISSING
MISSING--
MISSING
MISSING
900.0000-
900.0000
960.0000
960.0000 -
900.0000
960.0000
1020.0000
1020.0000
960.0000
960-.-0000- "
960.0000
960.0000
1140.0000
1POO.OOOO
1260.0000
900.0000-
960.0000
900.0000
940.0000
1080.0000
900.0000
Q40-.0000
900.0000
900.0000
1080.0000
1080.0000
1080.0000
1020.0000
960.0000
960.0000
MISSING
780.0000
1560.0000
1380.0000
1380.0000
1180.0000
1320.0000
1140.0000
1140.0000
1150.0000
ipeo.oooo
1140.0000
1140.0000
1140.0000
1140.0000
sso2 REM
• -
rflSSINF
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
77.8900
79.0000
79.5000
75.8600
84.4400
MISSING
MISSINS
MISSING
MISSING
83.3300
82.0000
80.6300
77.5000
86.6700
83.1300
85.2900
84.1200
80.0000
64.3800
81.2500
81.2500
81 .0500
78.5000
80.0000
76.6700
79.3800
79.3300
80.1200
80.5600
84.0000
86.4300
86.6700
86.6700
86.1100
86.1100
86.1100
79.4100
79.3800
7^.0000
MISSING
85.3800
MISSING
86.9600
86.9600
8*. 9600
8*>.3600
84.7400
3&.210D
84.3500
S3. 3300
8*. 2100
84.2100
MISSING
8*.2100
3 PH
4
6.6COO
7.3000
7.3000
6.5000
6.8000
6.8COO
7.1COO
6.7000
6.8000
6.6000
8.0000
7.2000
6.7000
7.6009
7.6000
7.5000
7.0000
7.1000
6.9000
O.VOOO
7.0000
7.2000
7.3000- '
7.1000
7.3000
7.5000
7.5000
7.5000
7.1000
7.2000
7.0000
7.0000 -
7.0000
6.9000
7.1000
7.0000
7.2000
7.3000
7.2000
7.1000
6.8000
6.8000
7.1000
6.8000
6.8000
7.2000
7.3000
7.1000
6.8000
7.1000
7.1000
7.1000
7.2000
6.9000
7.1000
7.1000
7.1000
7.0000
7.0000
7.0000
7.?000
DELTA P '
5.5000
6.200D
6.2000
1.000
3.5000
3.5000
4.0000
2.5000
4.2000
5.0000
6.0000
6.0000
4.0000
3.0000-
3.0000
3.0000
6.5000
5.0000
'• D GAS T—
MISSING
-- MISSIMG
MISSING
260.0000
283.0500
282.0000
252.0000
3] O.OGOO
230.0000
265.0000
MISSING
MISSING
265.0000
—260.0000 -
310.0000
310.0000
295.0000
290.0000
5.2000 300.0000
^ _ A4WIA ^t A AAA.A
3 1 ruuuO 270 vO OOw ~
7.0000 300.0000
7.0000
8.3000
a. oooo
MISSING
- STIOOOO-
8.0000
B.OOOO
5.8000
3.0000
3.0000
5.2000
5.2000
7.0000
7.0000
7.0000
9.4000
12.0000
13.0000
12.5000
11.5000
11.5000
9.5000
M.ISSIN6
5.0000
1.2000
7.0000
9.5000
7.0000
9.2000
9.2000
9.2000
10.000
10.000
R.OOOO
8.5000
10.000
10.000
10.000
10.000
10.000
270.0000
310.0000
285.0000
300.0000
-- 260.0000 -
260.0000
260.0000
340.0000
MISSING
315.0000
315.0000
290.0000
300.0000
310.0000
300.0000
280.0000
245.0000
225.0000
280.0000
270.0000
270.0000
250.0000
200.0000 •
240.0000
238.0000
MISSING
MISSING
250.0000
255.0000
MISSING
• MISSING
"ISSING
240.0000
250.0000
300.0000
200.0000
200.0000
?oo.onoo
200.0000
200.0000
6 D PfOD T
MISSING
HISSING
MSSING
iso.oooe
'.62.0000
162.0000
180.0000
i '5. 0000
175.0000
165.0000
•"ISSINS
HISSING
150.000C
7 CIPC GPM
12400.0000
looeo.oooo-
10000.0000
1)200.0000
11400.0000
114CC.OGOO
H7CO.OOCO
12000.0000
11700.0000
12000.0000
77CO.OOOO
7500. 0000
9500.0000
-155.00OO 9400. 0000 -
155.0000
155.0000
160.0000--
175.0000
9100.0000
9300.0000
10000.0000
9400.0000
165.0000 9400.0000
17-3vOOOO 9400-.0000 —
170.0000 MISSING
170.0000
18C.OOOO
182.0000
185.0000
- 185.0000---
155.0000
135.0000
190.0000
MISSING
195.0000
lao.oooe -
170.0000
175.0000
135.0000
190.0000
188.0000
MISSING
MISSING
MISSING
MISSING
- - MISSING -
MISSING
MISSING
• MISSING
MISSING
MISSING
- MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
182.0000 MISSING-
186.0000
200.0000
190.0000
190.0000
190.0000
- 160.0009- •
150.0000
155.0000
MISSING
MISSING
175.0000
175.0000
180.0000
140.0000
130.0000
130.0000
170.0000
195.9000
1=0.0000
190.0000
150.000C
150.0000
140.0000
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
8500.0000
8200.0000
8600.0000
8600.0000
8600.0000
8600.0000
9500.0000
11200.0000
11200.0000
11200.0000
10400.0000
10700.0000
11100.0000
11100.0000
11300.0000
8 FAN AMPS
200.0000
220.0000 —
200.0000
:fo.oeoo
190.0000
18C.CCOO
200.0000
175.0000
190.0000
190.0000
220.0000
220.0000
200,0000
- 170.0000 —
180.0000
210.0000
220.0000-
200.0000
210.0000
— 175-5-00OO —
220.0000
220.0000
240.0000—
230.0000
200.0000
--235>-0000
250.0000
230.0000
200.0000 -
170.0000
170.0000
200.0000—
200.0000
220.0000
220.0000
MISSING
250.0000
280.0000
300.0000
300.0000
280.0000
280.0000
260.0000
240.0000 • -
180.0000
160.0000
240.0000
260.0000
240.0000
260.0000
260.0000
260.0000
260.0000
260.0000
230.0000
230.0000
260.0000
2*0.0000
260.0000
260.0000
270.0000
-------
to
GO
NO.
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
H8-
139
140
141
142
143
144
145
146
147
149
1*9
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
1«>5
166
io7
158
169
170
171
172
173
i7»
1^5
176
177
173
179
1 S02IKPPM 2 SSO2 REM 3 PH 4
114Q.OOOO
1140.0000
MISSING
MISSING
900.0000
1?00.0000
1680.0000
1680.0000
960.0000
960.0000
900.0000
900.0000
900.0000
900.0000- -
MISSING
MISSING
900.0000
960.0000
960.0000
^OO.OOOfl— -
900.0000
900.0000
900.0000
900.0000
960.0000
960.0000
900.0000
900.0000
900.0000
960.0000
960.0000
1020.0000
1020.0000
1020.0000
1020.0000
1020.0000
1020.0000
060.0000
1080.0000
1020.0000
1020.0000
1020.0000
1080.0000
1140.0000
1140.0000
1140.0000
1^20.0000
1380.0000
inSO.OOOO
1020.0000
1020.0000
1023.0000
060.0000
•JhO.OOOO
1 ''80.9000
aeo.oooo
11*0.0000
IPOO.OOOO
i?:o.aooo
1?00.0000
1?00.0000
79.4700
79.4700
HISSING
MISSING
81.3300
92.0000
94.6400
94.6400
81.&800
81.8800
82.0000
85.3300
85.2200
84.4400
MISSING
MISSING
82.6700
82.5000
85.6300
84.0000
84.0000
84.6700
84.6700
84.0000
84.3800
84.3800
82.0000
82.0000
83.3300
83.1300
8?. 5000
82.3500
a?. 3500
82.9400
8?. "400
fl?.9»00
dl.1800
80.6300
32.2?00
81.1800
81.1800
dC.OOOO
30.5600
80.5300
80.5300
79.4700
70.2400
73.3300
7°. 3300
30.0000
6'.9403
00 9300.0000
MISSING
MISSING
215.0000 -
203.0000
230.0000
— 228. -00 00
230.0000
220.0000
250.0000
265.0000
272.0000
240. -0000—
210.0000
220.0000
220.0000
220.0000
240.0000
230.0000
230.0000
230.0000
230.0000
230.0000
270.0000
290.000&
300.0000
280.0000
2*0.0000
220.0000
220.0000
230.0000
230.0000
225.0000
220.0000
220.0000
220.0000
225.0000
220.0000
220.0000
220.0000
220.0000
?25.0000
230.0000
230.0000
240.0000
240.0000
240.0000
310.00CO
155.0000
160.0000
155.0000
155.0000
165.0000
< tfj fl fl A rt
roa.uuuu
172.0000
163.0000
172.0000
175.0000
173.0000
160.0000-
160.0000
158.0000
153.0000
153.0000
159.0000
1-70.0000
170.0000
170.0000
170.0000
170.0000
180.0000
180.0000
180.0000
185.0000
190.0000
170.0000
170.0000
180.0000-
180.0000
180.0000
130.0000
170.0000
170.0000
170.0000
170.0000
170.0000
170.0000
165.0000
175.0000
170.0000
170.0000
175.0000
175.3000
175.0000
190.0000
9400.0000
9500.0000
9500.0000
MISSING
9600.0000
OTn ft A n A n
" r U 0 • v vv V
9700.0000
10000.0000
10100.0000
10100.0000
10000.0000
10000.0000
10500.0000
10400.0000
10100.0000
10100.0000
10000.0000
10200.0000 -
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSIN&
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
-ISSING
"ISSING
-ISSING
MISSING
MISSING
MISSING
"ISSING
"ISSING
8 PAN AMPS
230.0000
220.0000
220.0000
230.0000
260.0000
280.0000
280.0000
230.0000
250.0000
220.0000
230.0000-
250.0000
250.0000
250iOOOO
170.0000
200.0000
220.0000
220.0000
240.0000
2*&cOO
-------
to
vo
SE NO.
130
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
* AA —
199
200
201
202
203
204
205
206
207
208
209
210
211
213
313
214
215
216
217
218
219
220
231
222
223
334
225
236
227
228
239
230
231
232
333
?34
2J5
236
237
238
239
240
1 S03INPPM
1300.0000
1080.-0000
1080.0000
loec.oooo
1080.0000
1140.0000
1080.0000
1080.0000
1080.0000
1140.0000
1140.0000
1140.0000
1140.0000
1080.0000-
1140.0000
1080.0000
1020.0000
1020.0000
1140.0000
1140.0000-
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1080.0000
1080.0000
1080.0000
1020.0000
1020.0000
960.0000
960.0000
960.0000
1020.0000
1030.0000
1030.0000
1020.0000
1200.0000
1020.0000
1110.0000
1080.0000
1080.0000
1080.0000
1140.0000
ineo.oooo
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
1140.0000
"40.0000
1080.0000
1140.0000
1140.0000
1080.0000
2 sso2 BEM
a*..ooor
78.8900
78.8900
78.8900
78.3300
83.1100
77.7800
77.7800
77.7800
76.3200
7^.6600
71.6800
71.6800
75.5600" •
77.8900
76.1100
74.7100
74.7100
77.8900
W %7 i% A-
7 T.370O
76.3200
76.3200
77.3700
76.3200
74.7400
76.3200
7f .3200
76.3200
7*. 3200
75.5600
7*. 6700
76.6700
7«-.4700
76.4700
72.5000
75.6300
75.6300
79.4100
76.4700
76.4700
76.4700
MISSING
30.5900
83.7800
77.7800
82.7800
74.3300
83.6800
hi .6700
82.6300
76.3200
7«.3300
7*. 3300
7*. 3200
7*.3?00
7«..320P
7-5.0000
81.6700
73.9500
7«.950P
77.7800
3 PH
7.0000
6.9000-
6.7000
6.9000
6.9000
7.0000
7.1000
6.9000
6.9000
6.7000
6.7000
7.0000
7.0000
-7.0000
6.9000
7.0000
7.0000
7.1000
7.0000
7.0000"-
7.1000
7.0000
7.0000
7.0000
7.1000
7.0000-
6.9000
7.0000
7.0000
7.0000
7.1000
7.1000
7.1000
7.1000
7.1000
7.3000
7.3000
7.2000
7.0000
7.0000
7.0000
7.5000
7.1000
7.1000
7.1000
7.1000
7.0000
7.0000
7.1000
7.1000
7.0000
7.1000
7.1000
7.2000
7.2000
7.3000
7.3000
7.1000
7.0000
7.1000
7.0000
4 DELTA P
4.0000
4.0DOO —
5.1000
3.9000
6.0000
8.0000
5.8000
7.0000
7.0000
6.0000
6.0000
5.8000
S.1000
5.9000
6.3000
6.0000
5.5000
5.5000
6.0000
6.5000
6.4000
7.0000
6.8000
6.6000
MISSING
~ 6.600 0
6.5000
6.5000
6.5000
6.5000
6.5000
6.5000
6.6000
6.2000
5.4000
3.0000
3.0000
5.5000
5.3000
5.2000
5.2000
6.6000
8.0000
9.5000
9.5000
9.0000
8.4000
8.4QOO
8.4000
8.5000
9.0000
9.0000
9.0000
9.onoo
9.0000
9.0000
7.0000
8.4000
8.5000
6.5000
8.5000
> D OAS T
315.0000
300.0000
400.0000
345.0000
335.0000
335.0000
400.0000
340.0000
340.0000
280.0000
380.0000
365. GOOD
410.0000
420".00«0-
420.0000
435.0000
425.0000
450.0000
450.0000
- 450.000t>"- "
430.0000
410.0000
390.0000
365.0000
370.0000
360.0000-
360.0000
350.0000
350.0000
350.0000
330.0000
350.0000
340.0000
350.0000
375.0000
MISSING
HISSING
200.0000
250.0000
250.0000
250.0000
310.0000
225.0000
225.0000
225.0000
275.0000
280.0000
275.0000
270.0000
285.0000
270.0000
260.0000
260.0000
250.0000
350.0000
?50.0000
280.0000
310.0000
290.COOO
275. "000
300.0000
6 0 PROD T
180.0000
165.0000 —
205.0000
155.0000
185.0000
180.0000
•.95.0000
160.0000
160.0000
165.0000
1*5.0000
240.0000
340.0000
25-3. OD 00
351.0000
263.0000
356.0000
345.0000
245.0000
260.0000
260.0000
250.0000
240.0000
335.0000
235.0000
230.0000" -
330.0000
230.0000
310.0000
310.0000
190.0000
190.0000
200.0000
300.0000
250.0000
MISSING
MISSING
MISSING
185.0000
185.0000
185.0000
195.0000
170.0000
180.0000
210.0000
310.0000
210.0000
205.0000
210.0000
210.0000
200.0000
190.0000
190.0000
190.0000
190.0000
190.0000
295.0000
210.0000
190.0000
?00.0000
210.0000
7 CIPC GPM
MISSING
*!lSSI*iG —
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
- -MISSING —
MISSING
HISSING
MISSING
MISSING
HISSING
MISSING
HISSING
HISSING
HISSING
MISSING
MISSING
KISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
HISSING
MISSING
HISSING
MISSING
MISSING
MISSING
MISSING
MISSING
HISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
-ISSING
8 FAN AMPS
190.0000
190.0000- ~
1EO.COOO
170.0000
210.0000
230.0000
300.0000
330.0000---
190.0000
300.0000
200.0000
200.0000
300.0000
--208TOOOO- —
200.0000
200.0000
200.0000
200.0000
210.0000
21 0.0000
210.0000
210.0000
210.0000-
310.0000
210.0000
-210-.0000-- -
210.0000
210.0000
210.0000
210.0000
210.0000
210.0000 —
210.0000
210.0000
210.0000
170.0000
200.0000
200.0000
200.0000
200.0000
200.0000
210.0000
230.0000
240.0000-
240.0000
240.0000
230.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
240.0000
-------
OJ
O
NO.
241
242
243
244
245
246
2<»7
248
249
250
251
252
253
254
255
256
25T
258
259
260
261
262
263
264
265
266
267
263
269
270
271
272
273
274
275
276
277
278
279
280
261
282
283
284
285
286
287
288
289
290
291
292
293
29»
295
296
297
298
299
300
301
1 S02IHPPM
1140.0000
1140.0000
1140.0000
MISSING
1380.0000
1140.0000
1140.0000
1230.0000
1230.0000
1080.0000
1140.0000
1?60.0000
1140.0000
1380.0000
1380.0000
1380.0600
1170.0000
' 1020.0000
1020.0000
1080. ODOO
1020.0000
1020.0000
1140.0000
1140.0000
1?00.0000
1200.0000-
1020.0000
1320.0000
1140.0000
1140.0000
1140.0000
1140.0000
1080.0000
1080.0000
1140.0000
1140.0000
1200.0000
1200.0000
1?00.0000
1200.0000
1200.0000
1080.0000
1020.0000
1080.0000
10BO.OOOO
960.0000
1020.0000
060.0000
1020.0000
1080.0000
inPO.OOOO
IOPO.OOOO
inPO.OOOO
IOHO.OOOO
1050.0000
10?0.0000
In50. 0000
IOPO.OOOO
1140.0000
inpo.oooo
1140.0000
2 5S02 REM
S?.6300
81.5800
7B.9500
MISSING
80.4300
74.7400
74.7400
79.7000
78.7000
72.2200
73.6800
76.1900
73.6800
78.2600-
78.2600
78.2600
74.3600
75.8800
7*. 8800
77.2200
74.1200
74.1200
76.3200
76.3200
7&.0000
77.0000
70.5900
77.2700
73.6800
71.6800
73.6800
73.6800
72.7800
72.2200
7->.6800
73.6800
75.0000
75.0000
80.0000
80.5000
80.5000
77.7800
76.4700
81.6700
77.7800
74.3800
81 .7600
81.250P
44.7100
8?. 8900
8?.Q90P
37.89QO
in. 3300
6?. 6100
3&.OOOP
8?. 6500
9?.ooor
80.5600
8* .?! 0*
8=>.650C
8-4.680P
3 PH
7.0000
6.8000
7.0000
7.0000
6.7000
6.9000
6.9000
7.0000
7.0000
7.0COO
7.0000
7.1000
7.1000
7.0000
7.0000
7.1000
7.1000
7.0000
6.9000
7.0000
7.1000
7.0000
6.9000
6.9000
7.1000
7.0000
6.9000
6.9000
7.1000
6.9000
7.2000
7.3000
7.3000
6.9000
7.0000
7.0000
7.0000
7.0000
7.2000
7.2000
7.2000
7.2000
7.2000
7.2000
7.2000
7.1000
6.9000
6.9COO
6.9000
7.0000
7.2000
7.3000
7.2000
7.2000
7.2000
7.0000
7.0000
6.9000
7.0000
6. "000
7.0000
1ELTA P
8.5000
8.6000-
8.5000
8.5000
7.5000
8.4000
7.4000
6.2000
4.0000
4.0000
4.0000
4.0000
4.0000
4-.4000
4.4000
4.4000
4.2000
5.0000
4.4000
5.0000 —
4.5000
4.5000
6.2000
6.2000
6.0000
6.4000
7.0000
7.0000
5.2000
4.7000
1.000
1.2000
2.8000
4.2000
4.0000
5.7000
7.2000
5.8000
5.0000
5.0000
5.0000
5.5000
4.8000
6.2000
6.2000
2.5000
6.3000
6.0000
8.0000
7.0000
7.0000
7.0000
7.0000
6.5000
6.7000
6.7000
6.7000
6.5000
6.2000
6.0000
7.0000
^ c Gis r
340.0000
305.0COO
29C. 00.00
33C.C090
32C.OOOO
275.0000
250.0000
300.0000
400.0000
410.0000
400.0000
390.0000
450.0000
390.000*
300.0000
380.0000
350.0000
325.0000
325.0000
-325.000fr
320.0000
320.0000
320.0000
330.0000
320.0000
320.0000-
320.0000
325.0000
325.0000
325.0000
325.0000
325.0000-
325.0000
335.0000
352.0000
352.0000
350.0000
350.0000
325.0000
325.0000
325.0000
325.0000
325.0000
320.0000
320.0000
318.0000
280.0000
277.0000
270.00CO
265.0000
265.0000
265.0000
265.0000
270.0000
265.0000
265.0000
280.0000
280.0000
230.P09S
260.0000
285.0000
6 D PROD T
2^0.0000
215.0000
2?5.0GOC
215.0000
210.0000
205.0000
215.0000
210.0000
185.0000
200.0000
210.0000
225.0000
225.0000
7 CIRC QPH
MISSING
- - MISSING
MI £51 VG
MJSETNG
MISSING
MISSING
MISSJNG
MISSING
MISSING
MISSING
KISSING
MISSING
MI5SING
225,-oooc — MISSING —
200.0000
175.0000
220.0000
200.0000
200.0000
MISSING
MISSING
MISSING
MISSING
MISSING
200.0000 — MISSING
205.0000
205.0000
205.0000
205.0000
200.0000
210.0000—
205.0000
218.0000
224.0000
215.0000
215.0000
215.0000
192.0000
195.0000
210.0000
215.0000
220.0000
220.0000
210.0000
210.0000
210.0000
209.0000
215.0000
220.0000
215.0000
225.0000
190.0000
197.0000
195.0000
210.0000
200.0000
200.0000
200.0000
215.0000
PDO.OOOO
195.0000
195.0000
195.0000
?15.0000
205.0000
205.0000
MISSING
MISSING
MISSING
MISSING
"ISSING
MISSING
MISSING
HISSING
MISSING
MISSING
MISSING
- MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
1020Q.OOOO
10200.0000
10200.0000
10200.0000
10200.0000
1G200.0000
10200.0000
10200.0000
10200.0000
10200.0000
10200.0000
8 FAN AMPS
240.0000
240.COOO
240.QQCO
24C.OCOO
2^0.0000
230.0000
210.0000
200.0000
leo.oooo
leo.oooo
1PC.OOOO
180.0000
leo.oooo
- leo^oooo —
leo.oooo
leo.oooo
leo.oooo
180.0000
leo.oooo
— leovoooo
190.0000
190.0000
200.0000--
200.0000
200.0000
200.0000
200.0000
205.0000
200.0000
190.0000
160.0000
160.0000—
170.0000
180.0000
180.0000
190.0000
200.0000
205.0000
190.0000
190.0000
190.0000
190.0000
190.0000
200.0000--
200.0000
150.0000
200.0000
200.0000
220.0000
210.0000
210.0000
210.0000
210.0000
210.0000
220.0000
220.0000
220.0000
220.0000
230.0000
230.0000
220.0000
-------
CASE NO. i SOZINPPM z »soz PEM 3 PH
"4 DEL
5 D GAS T- 6 0 PROD T 7 CIRC GPM 8 FAN AMPS
H
I
302
303
304
305
306
307
308
309
310
311
31Z
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
336
339
340
341
342
343
344
345
346
347
3*8
3*9
350
351
352
353
354
355
356
357
358
359
360
361
362
1140.0000
1140.0000
1140.0000
1140.0000
930.0000
960.0000
"40.0000
900.0000
900.0000
900.0000
840.0000
640.0000
840.0000
840.0000
810.0000
1080.0000
940.0000
810.0000
1340.0000
900*0000
960.0000
960.0000
1020.0000
960.0000
960.0000
1080.0000
1 100.0000
1140.0000
1)40.0000
1140.0000
1700.0000
1700.0000
I'OO.OOOO
1140.0000
1 140. 0000
1080.0000
loeo.oooo
-1080.0000
1080.0000
1200.0000
1?60.0000
1320.0000
1380.0000
1?00.0000
1140.0000
1140.0000
llin.0000
1140.0000
10*0.0000
960.0000
000.0000
060.0000
1"20.0000
i •-.P.O.: ooo
IIP 1.000 3
^•1. 1000
•390.0000
390.0000
M51. 1000
1 ISC. 0000
I3po.oooo
ai.saoo
83.6800
a?. 1100
82.6300
83.8700
82.8100
82.1400
83.3300
8e.3300
8=.3300
84.2900
8*. 0000
85.0000
83.5700
77.0400
86.1100
8*. 0000
87.7800
8C.7100
8*. 3300
86.2500
86.6300
84.1200
83.7500
83.7500
84.4400
83.0900
83.6800
31.5800
31.5800
8^.0000
?.^n
"I . 11 •"*
7.0000
7.1000
7.0000
7.2000
7.3000
7.3000
7.3000
7.1000
6.9000
7.0000
7.1000
7.1000
7.2000
7.2000
7.3000
7.1000
7.1000
7.0000
7.1000
• 7.1000
7.0000
7.1000
7.0000
7.1000
7.1000
7.1000
7.0000
7.0000
7.1000
7.1000
7.00QO
7.0000
7.0000
7.0000
7.0000
7.0000
7.0COO
7.0000
7.0000
7.0000
7.1000
7. 1000
7.0000
7.0000
7.1000
7.0000
7.1000
7.0000
7.0000
7.0*00
7.1000
7.1000
7.3COO
7. isr.n
7. lor.-:
7.1000
'.1SOO
7.1)000
'.1000
T.30CO
7.0000
6.7000
6.5000
6.0000
5.5000
6.3000
6.0000
4.7000
6.0000
8.2000
8.3000
7.5000
7.0000
7.0000
7.2000
2.3000
8.0000
6.0000
9.0000
10.6000
10.4000-
12.0007)
10.000
9.8000
10.000
9.8000
10.000
9.5000
9.5000
7.5000
9.5000
9.4000
9.6000
8.4000
8.4000
3.0000
9.2000
3.2000
8.4QOO
8.4000
2.8000
6.6000
6.6000
7.0000
3.3000
3.5000
3.7000
3.5ngo
3.3COO
3. "5300
3.6000
3.60QO
3.6000
3.a?00
3. '-OP
3. ••'30
3.0-00
-.2:00
5.8-00
= .5."00
6.S'>00
*.5"00
280.0000
280.0000
MISSING
285.0000
280.0000
2AO.OOOO
305.0000
310.0000
310.0000
310.0000
300.0000
300.0000
MISSING
300.0000
300.0000
310.0000
310.0000
300.0000
280.0000
250.0000-
260.0000
260.0000
260.0000
265.0000
267.0000
272.0000
300.0000
300.0000
350.0000
350.0000
300.0000
320.0000
330.0000
320.0000
320.0000
325.0000
325.0000
335.0000
335.0000
MISSING
350.0000
370.0000
370.0000
370.0000
370.0000
385.0000
3P5.0000
380.0000
3B0.0100
385.COOO
3PO.OOOO
190.0000
»00.0000
400.0010
400.0030
4CO.On<10
410.0010
410.0000
415.0109
40-3.0000
340. ngno
200.0000
MISSING
MISSING
195.0000
192.0000
MISSING
IBS. 0000
185.0000
190.0000
203.0000
205.0000
200.0000
MISSING
200.0000
190.0000
195.0000
205.0000
210.0000
200.0000
185.0000
180.0000
180.0000
183.0000
160.~0000
175.0000
175.0000 -
165.0000
170.0000
220.0000
215.0000
190.0000
190.0000
185.0000
190.0000
185.0000
190.0000
190.0000
195.0000
192.0000
MISSING
185.0000
205.0000
230.0000
235.0000
220.0000
220.0000
?1Q.OOOO
715.0000
?1Q.OOOO
160.0000
160.0000
220.0000
?30.0000
315.0000
310.0000
305.0000
730.0000
'30.0000
'35.0000
740.0000
715.3000
10200.0000
10200.0000
10200.0000
10200.0000
MISSING
10000.0000
10500.0000
10800.0000
10800.0000
10800.0000
10400.0000
10600.0000
10600.0000
10600.0000—
HISSING
11900.0000
HISSING
11600.0000
11600.0000
11600.0000
11600.0000
11600.0000
10600.0000
10800.0000
10800.0000
10800.0000
10900.0000
10900.0000
11200.0000
11200.0000
11200.0000
11200.0000
11200.0000
11200.0000
10800.0000
1 1000.0000
11000.0000
11100.0000
11100.0000
11200.0000
11200.0000
11200.0000
11200.0000
10900.0000
10700.0000
10700.0000
10700.0000
1 1000.0000
1 1000.0000
11000.0000
I 1000.0000
1 1000.0000
11000.0000
1 1000.0000
11000.0000
10900.0000
10900.0000
1*900.0000
13900.0000
'.0900.0000
10900.0000
210.0000
210.0000
210.0000
210.0000
210.0000
KISSING
150.0000
200.0000
220.0000
220.0000
220.0000
220.0000
220.0000
220«000fr
160.0000
180.0000
200.0000
250.0000
250.0000
2so*oooa —
250.0000
250.0000
250.0000-
250.0000
250.0000
250.0000—
230.0000
230.0000
210.0000
230.0000
230.0000
230.0000
210.0000
210.0000
220.0000
220.0000
220.0000
220.0000
220.0000
170.0000
200.0000
200.0000
200.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
160.0000
180.0000
200.0000
200.0000
200.0000
200.0000
-------
00
10
NO.
3b3
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
403
409
410
411
412
413
414
415
416
417
416
'1?
<>20
421
422
IC^3^"'™Dhl 3
O w t i • • ^ " ™ ™
11»C.OOOO
1140.0000
1140. OGOO
1140.0000
1140.0000
1140.0000
1140.0000
11 10.00PO
11 10.0000
ireo.oooo
1080.0000
8*0.0000
900.0000
900.0000
900.0000
900.0000
900.0000
900.0000
900.0000
f» 4-0. 0000"
840.0000
960.0000
1060.0000
loao.oooo
780.0000
760.0000
900.0000
900.0000
900.0000
000.0000
960.0000
900.0000
000.0000
900.0000
1020.0000
900.0000
son. oooo
900.0000 -
1000.0000
1140.0000
1020.0000
1120.0000
1020.0000
960.0000 —
960.0000
960.0000
1180.0000
1040.0000
1090.0000
1140.0000
000.0000
1160.0000
1*00.0000
1160.0000
ln2C.OOOO
1440.9000
144C.OOOO
l?r;.oooo
!!<•' .0000
1PCC.OOOO
!!<•<:. 0000
asos REM
£0.0000
61.5800
81.0500
8? .5800
81.5800
30.5300
31.3200
31.8900
32.160P
32.2200
8?.220n
8?. 8600
33.3300
81.3300
8*4.3300
83.3300
81.3300
81.3300
83.3300
83.5700
84.2900
80.0000
77.2200
77.2200
70.4600
7". 4600
81.3300
82.6700
88.0000
88.0000
87.5000
8*.. 6700
36.6700
86.6700
88.2400
86.6700
87.3300
87.3300
Se.6000
86.8400
87.6500
60.2400
3Q.2400
8=.0000
8=.0000
86.2500
8*. HOP
8«..7300
AT. 2200
67.3700
ofc. 6700
7?.7000
30.4000
91.3200
61.1*00
SC.P300
6*. BSD"!
87.5000
81.5100
63.510"
80.530P
3 PH
6.9COO
o.eeoe-
7.000C
7.1GOC
f.iooe
7.0000
7.2000
7.1006
7.2000
7.1000
7.0000
7.2000
7.1000
7.1000
7.1000
7.1000
7.1000
6.9000
6.9000
7.0000
7.0000
7.0000
7.0000
7.1000
7.2000
7.2000
7.2000
7.2000
7.2000
7.2000
7.1000
7.1000
7.1000
7.1000
7.1000
7.1000
7.0000
7.0000
7.0000
7.0000
7.0000
7.0000
7.1000
7.1000
7.1000
7.2000
7.1000
7.1000
7.00"0
7.0000
7.2000
7.0000
7.0000
7.0000
7.0000
7.0000
7. OGOO
7.1000
6.9000
7.1COO
7. 2000
4 DELTA P
6.5900
6.5000
6.6000
6.6COO
6.8000
6.3000
6.2000
6.5000
6.5000
6.4000
6.3000
6.3000
7.0000
7.0000
7.0000
7.0000
7.0000
8.0000
8.5000
8.500-0
8.8000
8.5000
4.0000
2.2000
5.9000
5.9000
5,9000
5.9000
5.9000
6.5000
8.8000
9.2000
9.2000
9.2000
9.2000
9.4000
10.3000
10.3000
9.0000
9.0000
10.000
10.000
10.000
10.000
10.000
10.000
10.2000
10.000
10.000
10.000
10.000
3.5000
3.0000
5.0000
5.0000
5.5000
6.0*100
7.0100
7.0000
7.4000
6.3000 '
5 C CAS T-
335.0000
335.0fr60-
330.0000
330.0000
330.0000
330.0000
330. OGOO
335.0000
335.0000
335.0000
335.0000
335.0000
365.0000
365.0COO
365.0000
365.0000
365.0000
360.0000
350.0000
350.0000
340.0000
330.0000
MISSING
MISSING
"ISSING
Hissr^
260.0000
260.0000
260.0000
260.0000
265.00TOO
265.0000
260.0000
260.0000
255.0000
280.0000
285.0000
285.0000
285.0000
210.0000
210.0000
210.0000
200.0000
200.0000
200.0000
200.0000
MISSING
275.0000
275.0000
275.0000
275.0000
250.0000
260.0000
MISSING
•MSSING
"ISSING
P60.0000
260.noOO
260.0000
260.0000
? nooo
6 0 PROD T
205.0000
- 2os.ooac -
250.0000
3:5.0000
2G5.000C
250.0000
21C.GOGO
215.0000
210.0000
210.0000
205.0000
180.0000
197.0000
197. MOO
197.0000
197.0000
197.0000
210.0000
220.0000
— 230.0000-
225.0000
210.0000
MISSING
MISSING
MISSING
MISSING-
185.0000
185.0000
185.0000
1 85. 0000
185.0000
185.0000
175.0000
175.0000
180.0000
170.0000
19?. 0000
192.0000
192.0000
170.0000
170.0000
170.0000
170.0000
170.0000
170.0000
170.0000
175.0000
175.0000
200.0000
200.0000
200.0000
180.0000
180.0000
MISSING
MISSING
MISSING
170.0000
180.0000
190.0000
169.0000
170.0000
7 CIRC GPM
10900.0000
1C900.0000
1 1200.0000
11200.0000
112CO.OCOO
11200. OOOD
1 1200. SOOO
11000.0000
11000.0000
11000.0000
11000.0000
11000.0000
11000.0000
H 000. 0000
11000.0000
11000.0000
11000.0000
10900.0000
10800.0000
10*00.0000
10800.0000
10800.0000
9900.0000
11000.0000
10900.0000
1*900.0000
10900.0000
10900.0000
10900.0000
10900.0000
11000.0000
11000.0000
11000.0000
11000.0000
11000.0000
10850.0000
10850.0000
10850.0000
10850.0000
10850.0000
10850.0000
10850.0000
11200.0000
11200. oooe
11200.0000
11200.0000
10700.0000
11050.0000
11050.0000
11050.0000
11100.0000
11400.0000
11400.0000
11400.0000
11400.0000
"ISSING
HISSING
"ISSING
11200.0000
11700.0000
11700.0000
8 FAN AMPS
200.0000
200.0000
200.0000
20C.OOOO
200.0000
2CO.OOOO
2CO.OOOO
200.0000
200.0000
200.0000
200.0000
200.0000
210.0000
2 10. -0008- ~
210.0000
210.0000
210.0000
220.0000
220.0000
220^0000
220.0000
220.0000
200.0000
160.0000
200.0000
200.0000—
200.0000
210.0000
220.0000
240.0000
240.0000
260.0000-
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
2fO.OOOO
260.000*--
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
260.0000
180.0000
IPO. 0000
220.0000
220.0000
220.0000
230.0000
220.0000
230.0000
230.0000
210.0000
-------
CAac NO.
424
425-
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443-
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
1 auZl.'O'Ka
U40.0000
-1440.0000
1560.0000
1560.0000
1560.0000
1560.0000
1560.0000
1560.0000
1500.0000
1440.0000
1440.0000
1320.0000
1500.0000
870VOOOO-
870.0000
1020.0000
1020.0000
1060.0000
1080.0000
1020.0000
1080.0000
1140.0000
1080.0000
1200.0000
1200.0000
1200.0000
1200.0000
MISSING
1560.0000
1560.0000
1050.0000
1050.0000
1500.0000
1?00.0000
1500.0000
1500.0000
1500.0000
1380.0000
1140.0000
MISSING
1140.0000
1080.0000
1060.0000
1320.0000
1320.0000
nzo.oooo
1?60.0000
"ISSING
1140.0000
MISSING
NUMBER 3f CASES =»EAO
-S3IJC PElH
79.4700
82.0800
82.3100
81.5400
MISSING
MISSING
MISSING
MISSING
HISSING
MISSING
MISSING
MISSING
MISSING
68.2800
68.2800
76.4700
77.0600
76.7900
76.6700
7*i7100-
75.0000
75.2600
73.3300
MISSING
MISSING
MISSING
MISSING
MISSING
MISSING
HISSING
MISSING
MISSING
82.0000
3*. 0000
94.0000
34.0000
84.4000
89.1300
38.4200
MISSING
89.4700
39.4400
39.060r
38.64QO
39.0900
89.550C
89.520"
-ISSfNG
-ISST-MG
«ISSIN5
j PH
7.
6.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
- 7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
7.
6.
6.
7.
7.
7.
7.
7.
7.
6.
7.
7.
7.
»
1000
9000 - -
0000
1000
2000
1000
2000
1000
1000
2000
2000
1000
2000
2000
1000
2000
0000
0000
0000
oooo
0000
0000
1000
1000
0000
0000
1500
2000
1000
1000
9000
9000
4000
2000
1000
1000
1000
2000
9000
1000
2000
1000
MISSING
7.
7.
7.
7.
7.
7.
7.
2000
1000
1000
1000
5000
4000
4000
i/c-k f 5
6.5000
— 6-r2a»O—
6.1000
6.0000
6.0000
6.0000
6.0000
6.0000
6.2000
6.2000
6.2000
6.4000
6.0000
6.0000
6.3000
6.3000
6.8000
6.8000
6.9000
6.9000
6.9000
6.9000
7.0000
6.9000
7.0000
7. -00 00
6.5000
6.5000
3.4000
• 3.4000
1.2000
MISSING
4.5000
4.0000
4.3000
MISSING
4.4000
7.8000
3.0000
7.5000
9.0000
9.2000
9.3000
9.4QOO
9.0000
9.nnoo
B.flnoo
1.2000
10.8000
8.0000
0 GAS T" - 6 0 PROD T
260.0000
265.0000
265.0000
260.0000
263.0000
260.0000
260.0000
260.0000
265.0000
260.0000
260.0000
260.0000
260.0000
260.0000
270.0000
270.0000
285.0000
300.0000
310.0000
305.0000
305.0000
300.0000
325.0000
300.0000
310.0000
310.0000
305.0000
305.0000
MISSING
MISSING
MISSING
MISSING
MISSING
200.0000
200.0000
200.0000
200.0000
200.0000
210.0000
200.0000
210.0000
200.0000
170.0000
200.0000
200.0000
200.0000
200.0000
"ISSING
"ISSING
MISSING
170.0000
- '--170.-0000 -
190.0000
195.0000
198.0000
190.0000
188.0000
187.0000
190.0000
19?. 0000
190.0000
190.0000
190.0000
7 CIRC GPM
11900.0000
-11600.0000- -
11600.0000
11500.0000
11300.0000
11400.0000
11400.0000
11400.0000
11400.0000
11300.0000
11300.0000
11300.0000
11600.0000
8 FAN AMPS
210.0000
- -aiovoooo
210.0000
210.0000
210.0000
210.0000
210.0000
210.0000
210.0000
210.0000
210.0000-
210.0000
210.0000
- 145.0000—1-1000. 0000 ZIOrWOO
180.0000
180.0000
165.0000
175.0000
170.0000
— tro.oooo—
170.0000
170.0000
165.0000
174.0000
175.0000
175.0000- -
175.0000
175.0000
MISSING
MISSING
MISSING
MISSING-
MISSING
125.0000
120.0000
120.0000
125.0000
1-30.0000
155.0000
MISSING
150.0000
150.0000
145.0000
135.0000-
14Q.OOOO
135.0000
140.0000
••ISSING
MISSING
115.0000
11400.0000
11400.0000
12000.0000
11800.0000
11800.0000
i too o.oo oo- -
11800.0000
11800.0000
11800.0000
11800.0000
11800.0000
H800.0000--
11500.0000
11500.0000
11700.0000
11700.0000
11600.0000
11600.0000
11600.0000
11600.0000
11600.0000
11600.0000
11200.0000
11200.0000
11300.0000
11300.0000
11700.0000
-ISSING
MISSING
11700.0000
11700.0000
11700.0000
11700.0000
11400.0000
11400.0000
11400.0000
VARIABLE
Un UAUET UFAhl
NO.
1
2
3
4
5
6
7
a
9
S02INPPM
«S02 <»EM
PH
DELTA P
D GAS T
0 PRO^ T
CIRC RPv.
FAN AMPS
C TOROUE
I-lt *»"•
109?.
81.
7.
6.
287.
I"*.
1072C.
?lc.
37.
210.0000
210.0000
210.0000
210.0000
210.0000
-21 0-. 0000
210.0000
210.0000
210.0000 -
210.0000
210.0000
210.0000
210.0000
210.0000
190.0000 -
170.0000
160.0000
170.0000- -
200.0000
200.0000
200.0000
200.0000
200.0000
250.0000-
250.0000
240.0000
260.0000
250.0000
MISSING
240.0000 —
260.0000
260.0000
260.0000
160.0000
240.0000
240.0000
STANDARD
nC"U T ATT OM
UtV 1 A 1 I oN
136 163.471
?24 4.444
050 0.166
873 2.290
?15 57.256
475 24.719
461 911.543
PS2 28.681
770 8.549
TOTAL
FRFOII
r rvcwu
403
382
469
466
429
432
278
468
437
-------
9.0 LIST OF REFERENCES
1) Sulfuric Acid from the Stack. Chemical Week, 197 (3)
(1970).
2) Shah, I.S., Wechselblatt, P.M., Radway, J.E. SO-
Recovery from Smelters with Magnesium Base
SOg Recovery Process. AIME Environmental
Quality Conference, Washington, D.C. (June 1971).
3) Shah, I.S. Removing SO., and Acid Mist with Venturi
Scrubbers"! CEP^ (Hay 1971) Vol. 67, No. 5.
4) Kleiman, G. and Willett, H. Relative Economics of
Stack Gas Scrubbing US Residual Oil Desulfurization.
API Session on Desulfurization and SO, Recovery
San Francisco, Calif., (May 12, 1971)7
5) Shah, I.S., Quigley, C.P. Magnesium Base SOp Recovery
Process, A Prototype Installation.70tR AIChE
National Meeting, (August 1971).
6) Wechselblatt, P.M., Quig, R.H. Magnesium Base S02
Recovery Scrubbing Systems.71st AIChE National
Meeting, (February 1972).
7) Maxwell, M.A., Koehler, G.R. The Magnesia Slurry SOp
Recovery Process with a Large Prototype System.
65th AIChE Annual Meeting, (November 1972).
8) Quigley, C.P. Progress Report - Magnesium Oxide System
at Boston Edison Company's Mystic Station.
Electrical Worlds Technical Conference, Chicago,
(October 1972) .
9) Houston, P. and Koehler, G. Application of Magnesia
SO- Control System to a 150 MW Power Plant.
International Conference on S02 Control,
Manchester, England (April 1973).
10) Koehler, G.R. (Part I) and Quigley, C.P. (Part II)
Operational Performance of the Chemico Basic
Magnesium Oxide System at the Boston Edison
Company Flue Gas Desulfurization Symposium.
New Orleans, (May 1973). (EPA-650/2-73-038,
December 1973) .
- 134 -
-------
11) Radway, J.E. and Rohrbach, R.R. Progress Report on
the Chemico-Basic Magnesium Oxide Flue Gas
Desulfurization System at the Boston Edison
Company. Presented at the 30th Annual Meeting,
East Central Section APCA, Cleveland, Ohio
(September 26, 1973).
12) Maxwell, M.A. Application of the Magnesia Slurry SO^
Recovery Process to Stack Gas Desulfurization.
Paper presented at 25th Southeastern Regional
Meeting, American Chemical Society, Charleston, S.C.
(November 8, 1973).
13) Koehler, G.R. New England S02 Recovery Project -
System Performance.66th AlChE Annual Meeting,
Philadelphia (November 1973).
14) Koehler, G.R. Alkaline Scrubbing Removes Sulfur
Dioxide. Chemical Engineering Progress, 70, No. 6
74.
15) Koehler, G.R. and Dober, E. Magnesia SOg Absorption
Process Development. Flue Gas Desulfurization
Symposium, Atlanta, Ga. (November 1974)
(EPA-650/2-74-126) .
16) Quigley, C.P. and Burns, J.A. Assessment of Prototype
Operation and Future Expansion Study - Magnesia
Scrubbing Mystic Generating Station Boston,
Massachusetts (Ibid).
17) Erdman, D.A. Mag-Ox Scrubbing Experience at the Coal-
Fired Dickerson Station, Potomac Electric Power
Company"Washington, D.C.(Ibid).
18) Zonis, I.S., Olmsted, F., Hoist, K.A. and Cunningham,
D.M. The Production and Marketing of Sulfuric
Acid From the Magnesium Oxide Flue Gas Desulfurization
Process"! (Ibid) . "
19)- Koehler, G. Report of Operation of a Magnesia FGD System
on an Oil Fired Boiler. AIChE Symposium Series,
No. 148, (July 1975).
- 135 -
-------
20) Koehler, G., Chatlynne, C. The Magnesia Scrubbing
Process As Applied to an Oil-Fired Power Plant.
EPA-600/2-75-057 (October 1975).
21) Hess, H., Englick, J., Koehler, G. Magnesium Oxide
Flue Gas Desulfurization Process. Paper presented
at Conference on Air Quality Management in the
Electric Power Industry, Austin, Texas, (January 1976) .
22) Taylor, R., Gambarani, P., Erdman, D. Summary of Operations
of The Chemico-Basic MqO FGD System at The Pepco
Dickerson Generating Station. Paper presented at 1976
EPA Symposium On Flue Gas Desulfurization, New Orleans,
Louisiana.
- 136 -
-------
10.0 CONVERSION FROM ENGLISH TO METRIC UNITS
TO CONVERT FROM
Atmosphere (normal)
Atmosphere (normal)
Barrel (42 US gallons)
British thermal unit (BTU)
BTU/Hour
BTU/pound mass
BTU/pound mass - °F
Foot
Foot2
Foot3
Foot /minute
Foot-pound force
Gallon (US)
Gallon (US)/minute
Grain
Horsepower
Inch
Inch H20 (60°F)
Pound Force
Pound mass av
Pound mass av
2
Pound force/inch
Pound mass/foot
°Rankine
Ton mass (.US short)
Ton mass (US long)
TO
Bar
Pascal
Meter3
Joule
Watt
Joules/gram
Joules/gram - K
Meter
Meter2
Meter3
Meter /minute
Joule
Meter3
Meter /hour
Milligram
Kilowatt
Centimeter
Kilopascal
Newton
Kilogram
Metric ton (tonne)
Kilopascal
Kilograms/meter
°Kelvin
Kilogram
Kilogram
MULTIPLY BY
1.01325
101,325
0.15899
1055.1
0.29307
2.32600
4.18680
0.30480
0.09290
0.02832
0.02832
1.35582
0.00379
0.22712
64.7989
0.74570
2.5400
0.24884
4.44822
0.45359
0.0004536
6.89476
16.0185
0.55556
907.185
1016.05
_ 137 -
-------
APPENDIX I
POST OPERATION INSPECTION REPORT
DATE OF INSPECTION - Jan. 17-24, 1975
(NOTE: Refer to Figure 1-1, where numbers refer to follow-
ing sections)
Plumb Bob Control Mechanism
Operation has been satisfactory.
Upper Section of 1st Stage (5)
(a) The polyester coating was gone from the stem and the
top 8" of sloping sides of the plumb bob. The metal
is type 316L stainless steel at this point and measure-
ments indicated no loss of thickness.
(b) The coating was also gone from most of the 316 stain-
less steel lip projecting from the sloping wall just
above the throat.
(c) In the throat area there were about 10 spots ranging
in size from 12 square inches to 30 square inches
where the entire coating had been removed from the
metal. The metal is 316L stainless steel at this point.
It is possible that these coating failures were caused,
by debris from the 4 inch stainless steel nozzles
which had failed.
(d) Above the projecting lip on the sloping outer wall,
there were a number of holes in the lining where the
carbon steel had been attacked so that only the lining
on the other side of the steel plate prevented a hole
in the vessel. The figure is typical of this type of
failure.
- 138 -
-------
OOTUMft
FIGURE I- I
- 139 -
-------
(e) As was discussed previously in the section on corro-
sion of stainless steel all of the 4-inch tangential
nozzles were subjected to extensive failure. Two
were still in place, but were in shreds.
(f) The three six-inch stainless steel lines that feed the
center manifold were subjected to the same type of
attack as the four-inch tangential nozzles and pipes.
There was less severity because of the protective
coating of brick cast which had been applied over these
pipes. Repairs have been carried out by removing the
brick cast and coating these six-inch pipes with glass
reinforced polyester plastic.
(g) The center manifold was in good condition both inter-
nally and externally.
(h) There were two holes through the external wall of the
vessel adjacent to tangential inlet nozzles. It is
not known the extent to which these were caused by nor-
mal erosion or by impaction by partially failed inlet
nozzles. Except for these two failures the coating at
this portion of the vessel was in good condition.
Lower Section of 1st Stage /g^
(a) The polyester lining was in fair condition, with re-
pairs needed at several points in the throat section,
at the bottom of the skirt and on the outer surface
of the inner skirt.
(b) There was one small (3/8 inch diameter) hole in the
bottom of the cone near the vortex breaker where a
repair had been made.
- 140 -
-------
(c) There was considerable loss of metal by corrosion of
type 316 stainless steel from the hanger rods, clips,
clamps and bolts. As indicated previously, corrosion
was most severe in areas which had been stressed and
the replacement pieces have been stress-relieved before
installation.
1st Stage Mist Eliminator Section (?)
(a) The mist eliminators were clean and in excellent con-
dition. However, all of the J hooks and clips used
as hold-downs had failed and most were completely mis-
sing. These are among the sources of metal that have
contributed to failures of rubber-lined pipe. The 2
inch x 0.2 inch straps used as auxiliary hold-downs
for the mist eliminators have performed their function
well. There is evidence of slight pitting, but no loss
of thickness.
(b) The new clevis type 1st stage cone support hangers
showed slight signs of pitting, but are structurally
in excellent condition. These were of 316 stainless
steel and were installed in July 1974. The Inconel
625 pins installed with them were in excellent condi-
tion and showed no sign of attack.
(c) The 1st stage mist eliminator wash system has operated
satisfactorily.
Upper Section of 2nd Stage (8)
This portion of the scrubber was found to be in generally
good condition. There was a small hole (possibly 1/16
inch in diameter) in the stainless steel pipe just below the
base.
- 141 -
-------
Inspection Report (Information of two-stage scrubber
absorber and the recycle pipe have been presented in
the previous report).
2nd Stage Throat Area (9)
Failure of the coating in the area of the 2nd stage throat
was very extensive as illustrated in photographs 10-7 and
10-10. Assurance of adequate structural integrity was pro-
vided by a series of thickness measurements.
The throat restrictor provided substantial protection for
the inner surface of the throat except where there apparently
were gaps between the restrictor and the inner surface.
Planned repairs and changes include division into an inner
and outer ring, a reduction in the total width (to decrease
pressure drop), and caulking between the restrictor plates
and walls to eliminate flow and erosion at these points.
Gas Systems
(1) Fan
The fan has given excellent service throughout the
entire history of the scrubber. Inspection indicated
that the rotor, which is of high alloy stainless steel
is in a condition apparently as good as new. The
rubber lining of the fan housing is also in a condition
as good as new. The only problem was a failure in one
of the pipes which feeds water to the fan rotor. This
pipe had failed presumably as a result of continued
vibration. Repairs were effected by welding.
(2) Secondary Mist Eliminator
The secondary mist eliminator was clean and in excel-
lent condition except for some evidences of corrosion
of the various 316 stainless steel hanger rods, etc.
This was similar to the corrosion observed in the 1st
stage of the scrubber although substantially less in
extent.
- 142 -
-------
(3) Ducts and Dampers
There was a fair amount of build-up of fly ash in the
ducts as would be expected, but the ducts were other-
wise in good condition. The major problem with the
dampers has been leakage during shut down, permitting
a flow of gas backwards through the ducts and scrubber
and a progressive increase in fly ash build-up and de-
crease in damper effectiveness. At the conclusion of
operations in late January, whenever the scrubber was
shut down there was a very high flow of gas backwards
through the scrubber.
A seal ring has now been installed on the outlet dam-
per, with the expectation of drastically reducing (and
possibly eliminating) this problem. It is also expected
that better continuity of operation will be obtained
and this in itself will permit better performance of
the dampers on the few occasions when shut downs are
expected to be necessary.
Liquid Systems
(1) 1st Stage Recycle Pumps
These have performed excellently at all times. In-
spection through the suction piping indicates an
appearance "like new."
(2) Thickeners
These have performed satisfactorily. The south
thickener had a number of failures in the coating
which were repaired. The north thickener, in contrast,
had no coating failures. The bolts that hold the arms
of the thickener rakes were corroded excessively in the
south thickener, but little or no corrosion was observed
in the similar bolts in the north thickener.
- 143 -
-------
(3) Thickener Underflow System
This system has performed satisfactorily and inspec-
tion did not indicate any defects.
(4) Thickener Overflow Return System
This system has performed satisfactorily from a
mechanical standpoint and inspection did not indicate
any defects.
(5) 2nd Stage Recycle Pumps
These pumps have given generally satisfactory perfor-
mance. The impeller on each of the three pumps has
been replaced once. In a commercial installation it
is expected that rubber covered impellers would be
specified and that impeller replacement would not be
necessary. At the conclusion of the operating period
in late January 1975, performance curves were obtained
on the three pumps as indicated in Figure 1-2.
(6) Mother Liquor Tank
This tank has given satisfactory performance. The
gears on the agitator were replaced on one occasion.
Presumably their failure was due to a build up of
solids in the bottom of the tank leading to excessive
side thrusts on the agitator.
(7) Mother Liquor Pumps
Operation has been satisfactory.
(8) MgO Premix Tank
Operation of this tank has been satisfactory since
the changes were made in the summer of 1974.
- 144 -
-------
FIGURE 1-2
9000 _.
PERFORMANCE CURVES
2nd STAGE RECYCLE PUMPS
8000 __
7000 --
6000 --
Scrubber Level 24%
January 1975
5000 __
S
8
4000 __
3000 --
2000
1000 __
-0- "A" Pump
-D- »B" Pump
-A- "c" Pump
PSI
To Jo 30
'40
50
70
- 145 -
-------
(9) MgO Slurry Tank
Operation has been satisfactory except that the steam
sparger has plugged repeatedly. During the current
boiler outage, the location of the steam inlet line
has been changed. The line now enters through the
top of the tank and is so arranged that the sparger
may be unplugged if plugging should continue. It will
also be possible to change to other sparger designs if
such becomes necessary in order to eliminate plugging.
(10) MgO Slurry Feed Pumps
Operation has been satisfactory since October 1974,
at which time a larger pump was installed as the east
MgO pump. One remaining problem with this system is
that it is necessary to operate both pumps in order to
obtain correct reading of the temperature control ele-
ment. A change in the location of this element is
planned so that the operation may be carried out with
only one pump, reserving the other as a spare.
Solids Systems
(1) MgO Unloading
This system has operated satisfactorily in all respects
except that a leak developed in the elbow at the top of
the loading line, as was to be expected. Repairs were
made by welding which will be adequate for the remain-
ing three months of expected operation of this system.
For a commercial installation materials more resistant
to erosion would be specified for the elbows in unload-
ing lines.
(2) MgO Storage Bin
Operation has been satisfactory.
(3) MgO Weigh Feeder
Operation of this system has been acceptable since
- 146 -
-------
changes were made in the summer of 1974. However,
the reliability of this system leaves something to
be desired and for future commerical installations,
other designs would be investigated. There have
been several belt failures during the operation of
this system. Near the end of operation in late
January, the weigh feeder became most unreliable in
its operation and during the outage, moderately ex-
tensive repairs and replacements have been carried
out.
(4) Centrifuge
Operation of the centrifuge has been generally quite
satisfactory and very substantially improved over
operation experienced at Boston Edison Company. A
major difference between the two installations, of
course, is that at Boston Edison the centrifuge feed
contained primarily magnesium sulfite trihydrate where-
as at PEPCO the feed has been primarily magnesium sul-
fite hexahydrate. There are four areas that require
comment:
(a) There has been no evidence of the type of binding
between the conveyor and the bowl (internally)
that was so troublesome at Boston Edison. Rota-
tion of the centrifuge was continued most of the
time, although there were substantial periods
when the machine was shut down and binding might
have taken place. The washing procedures deve-
loped at Boston Edison were applied at PEPCO from
the start of the operations. The extent of inter-
nal wear, which is probably directly related to
binding, could not be determined. The desirability
- 147 -
-------
of dismantling the centrifuge for inspection
was recognized, but when weighed against the
cost involved, the decision was made to resume
operations without inspection of the internal
portions of the centrifuge.
(b) There was excessive erosion-corrosion of the
case at the liquid end, as shown in photograph
3-2. A patch was applied around the entire 180°
circumference of the casing. Wear continued and
the casing, which was originally 1/4 inch thick,
was worn through completely, exposing the patch
for a longitudinal distance of about 1-1/2 inches.
Although replacement patching internally with
stainless steel would probably have provided a
permanent cure to this erosion problem, the deci-
sion was made, in the interest of economy, to
patch the casing with carbon steel. A coupon of
316 stainless steel has been installed and obser-
vation will be made at the conclusion of the operat-
ing period as to the ability of type 316 to with-
stand this service.
(c) One plow failed and all of the plows were replaced
in September 1974. One plow from the new set
failed in January 1975 and a complete third set of
plows has now been installed. An improved design
of plow may be expected to eliminate this problem
in future installations. The photographs 3-5 (the
failed plow) and 3-11 (normal plows) suggest that
some machines are designed with three bolts per
plow.
- 148 -
-------
(5) Screw Conveyors
These have given generally satisfactory performance.
There is, of course, considerable evidence of wear,
but it is not regarded as excessive. The bearing at
the inboard end of the dryer feed screw was dislodged
due to a spreading of the bearing support as illustra-
ted in photograph 4-3. The screw continued to operate
without any inboard bearing, using the solids in the
trough as support, for several weeks. This bearing has
now been replaced.
(6) Dryer
The dryer has given excellent performance. Buildup
has been limited in amount and it has been possible
to clean the dryer usually by an adjustment in operat-
ing conditions, although occasionally it has been neces-
sary to use a pipe or rod for this purpose. At the con-
clusion of operations in late January, the dryer was
remarkably clean, as illustrated in photograph 4-1.
The refractory of the burner block and the metal casing
around it failed as indicated in photograph 4-6 and
have been replaced.
(7) Dryer Off-Gas Cyclone
The double gate at the base of this cyclone has never
operated properly and a rotary valve has now been in-
stalled to replace it. In the past, failure to remove
solids from the base of the cyclone has led to an ex-
cessive carry-over of fines and recycle to the scrubber.
This situation should be substantially improved as a
result of the installation of the rotary dust valve.
(8) Dryer Off-Gas Fan
Operation has been satisfactory.
- 149 -
-------
(9) Bucket Elevator
It appears that the bucket elevator was sized for a
product of higher density and possibly sized in ex-
pectation of continuous operation at a uniform rate.
In any event, the capacity of the bucket elevator has
been inadequate to handle the full load of the scrub-
ber system and has made it necessary to operate at rates
approximating 75% of gas flow capacity and somewhat
less than 75% of magnesium sulfite capacity. Larger
buckets have now been installed on the bucket elevator
and it is expected that capacity operation can be
achieved at this point in the system as well as else-
where.
(10) Magnesium Sulfite Storage Silo
Operation has been satisfactory.
(11) Magnesium Sulfite Loading System
Mechanical operation of this system is considered
satisfactory. The repeated problems that have been
experienced with excessive spillage of magnesium sulfite
during loading are considered due to operating problems
rather than to defects in the mechanical design or
condition of the equipment.
- 150 -
-------
2 o ^RS
151
-------
10-7
152
-------
APPENDIX 2
SULFUR OXIDE REMOVAL FROM POWER PLANT STACK GAS
Magnesia Scrubbing-Regeneration: Production of Sulfuric Acid
The Thermal Dehydration of Magnesium Sulfite
Hexahydrate (MgSO.'6H20) and Magnesium Sulfite
Trihydrate (MgS03-3H20);
A mechanistic Study using Thermo-analytical
Techniques and the Development of an Analytical
Method for Quantitating Mixtures of the Hydrates
by
Prof. Leonard Dauerman
Department of Chemical Engineering and Chemistry
New Jersey Institute of Technology
Newark, New Jersey
Prepared for
Chemical Construction Co.
New York, New York
February, 1975
- 153 -
-------
ABSTRACT
The thermal dehydration of MgS(>3.3H20 and MgS03.6H20 have been
studied by differential thermal analysis, differential scanning
calorimetry and thermal gravimetric analysis.
Similar studies by other groups led to contradictory conclu-
sions. In this investigation, these results were reconciled and
it was concluded that MgS(>3.6H20, under equilibrium conditions,
dehydrates in two steps through the intermediate formation of
MgS03.3H20.
One consequence of this work is the development of a simple
analytical procedure, using TGA, for the quantitative determination
of both hydrates in mixtures and in the presence of thermally in-
active material.
- 154 -
-------
TABLE OF CONTENTS
ABSTRACT 154
TABLE OF CONTENTS 155
LIST OF TABLES 156
LIST OF FIGURES 157
INTRODUCTION 158
EXPERIMENTAL 161
RESULTS AND DISCUSSION 163
APPLICATION 168
REFERENCES 172
TABLES 1-6 inclusive 173-178
FIGURES 1-22 inclusive 179-200
- 155 -
-------
LIST OF TABLES
No. Page
1. Starting Temperature of Thermal Dehydration 17.3
of MgS03.3H2Q
2. Starting Temperature of Thermal Dehydration 174
of MgS03.6H20
3. Calculated and Observed Values of Water 175
Content in Synthetic Mixtures
*. M8SO,.3H 0 Content from TGA 176
3 2
5. MgS03.6H20 Content from TGA
6. TGA Results of a Synthetic Mixture
Containing Thermally Inactive .'Mre
- 156 -
-------
LIST OF FIGURES
No. Page
1 DTA of MgS03.3H20 179
2 DTA of MgS03.6H20 18-°
3 DSC of MgS03.3H20 181
H DSC of MgS03.6H20 18.2
5 TGA of MgSO .3H20 183
6 TGA of MgS03.6H20 184
7 TGA of MgS03.3H_0 under self-generated atmosphere 18^5
8 TGA of MgS036H20 under self-generated atmosphere 186
9 TGA of Std #1 187
10 TGA of Std #2 1S8
11 TGA of Std #3 1£9
12 TGA of Std #U 19-°
13 TGA of Std #5 191
W TGA of Std #6 192
15 TGA of Std #7 193
16 TGA of Std #8 194
17 TGA of Std #9 195
18 TGA of Std #10 196
19 TGA of Std #11 19,7
20 TGA of a mixture of MgS03.3H20, MgS03.6H20 and 198
glass beads
21 Water content versus % MgS03.3H20 199
22 Water content versus % MgS03.GH20
- 157 -
-------
INTRODUCTION
The thermal dehydration of the two known stable hydrates
of magnesium sulfite, namely, magnesium sulfite hexahydrate
(MgSG^.BHjO) and magnesium sulfite trihydrate CMgS03.3H20)
have been studied. On the one hand, this work is of practical
importance because in the Magnesia Scrubbing Regeneration
Process (CHEMICO)^, the thermal dehydration of the magnesium
sulfite hydrates is a significant step. On the other hand,
this study is of scientific interest because the mechanism of
the dehydration has been the subject of controversy.
In the Magnesia Scrubbing Regeneration Process (CHEMICO),
sulfur dioxide in the flue gas is removed by scrubbing with a
magnesia (MgO) base slurry. Hydrates of MgSOs are formed;
also MgSOij is formed by oxidation. A portion of the slurry is
withdrawn and centrifuged. The solid is separated from the
liquid, which is recycled. The solids are then dehydrated in
a dryer and, afterwards, calcined at 1400°F-1600°F with coke
added to the solids. Heating is sufficient to effect the
thermal decomposition of MgSOs; the coke reduces MgSOq. to
MgSOs. The useful products of calcination are MgO, which is
recycled back to the slurry, and S02, which is converted to
sulfuric acid.
Turning to the scientific relevance of this study,
thermal dehydration studies have been reported by three groups
and the results were apparently contradictory. Okabe and
- 158 -
-------
Horiloused three different techniques to study the dehydration:
Differential Thermal Analysis '(DTA), X-ray and Infrared (IR).
From the DTA and X-ray results, they concluded that MgS03.6H20
loses three water molecules between 60°C and 100°C to form
MgS03.3H20. The latter at 200°C completely dehydrates to
yield amorphus anhydrous MgS03. But in the infrared investi-
gation, they reported that the spectra does not change when
the trihydrate goes to the anhydrous state at 200°C. It does
not seem plausible that the transformation suggested could
have occurred without any change in the infrared spectra.
The band in the 3500 cm region is obviously an 0-H stretching
band, therefore, if the salt is dehydrated it would be expected
that this band would disappear.
Two other investigations using DTA were carried out by
groups at the Tennessee Valley Authority (TVA). These studies
were not published, but are presented, in part, in an EPA-
sponsored critical analysis of the Magnesia Process prepared
by TVA *•"•'. In the above cited report, Jordan's work based upon
DTA leads to the conclusion that the thermal dehydration of
MgS03.6H20 takes place in one step starting at 100°C and that
MgS03.3H20 dehydrates in one step, starting at 160°C.
In the other study, Hatfield and co-workers reported that
MgS03.GH20 loses nearly all its water when heated in a stream
of argon or air at 104°C for 16 hrs. This, too, supports the
inference that the thermal dehydration of the hexa-
hydrate occurs in one step. It is also reported that
- 159 -
-------
.Sj^O is partially dehydrated when heated in air for 16
hrs. at 160°C.
In the above-cited EPA report, it is suggested that the
apparently contradictory results may be due to differences in
experimental conditions. It was speculated that the samples
were heated in sealed tubes in the work of Okabe and Hori,
although such information was not provided in that paper.
Without presenting a critical analysis of the effects of hav-
ing the samples open or closed to the atmosphere, it is
concluded in the TVA report that the results of the TVA
groups, in which the samples are open to the atmosphere, are
valid and that MgS03.6H20 dehydrates in one step to MgS03 at
100°C.
The purpose of this study is to reconcile the contra-
dictory results. Thermoanalytical techniques arc used
exclusively: Differential Thermal Analysis (DTA), Differential
Scanning Calorimetry (DSC) and Thermal Gravimetric Analysis
(TGA). Other groups in this laboratory have studied the
thermal dehydration using Mass Spectroscopy and Infrared
Spectroscopy. These results will be reported separately.
- 160 -
-------
EXPERIMENTAL
MgS03.3H20 (97.9%) and MgS03.6H20 (99.0%) used in this
study were laboratory prepared samples. Each hydrate was
studied individually; also synthetic mixtures of both hydrates
were studied.
The DuPont thermal analyzer was employed in this
investigation. This included the DuPont 900 differential
thermal analyzer (DTA) equipped with both the standard DTA
cell and the differential calorimetric cell (DSC). The DuPont
950 TGA unit,which is an attachment to the DuPont 900,was
also used.
DTA (9J is a thermal technique in which the heat effects,
associated with chemical or physical changes, are recorded as
a function of temperature or time as the substance is heated at
a uniform rate. Enthalpic changes, either-endothermic or
exothermic, are recorded. The sample temperature is contin-
uously compared with a reference material temperature; the
difference in temperature is recorded as a function of furnace
temperature or time. DTA is reported to have been first used
by LeChatelier'4'in 1887 for studying clay. Since that date
many developments have been introduced and the literature has
grown exponentially. DTA has been used for the study of the
thermal dehydration of hydrates. For example, the reader is
referred to the work of Wendlant and Hoiberg' '.
In contrast to DTA, in which the temperature difference
- 161 -
-------
between the sample and the reference is measured, in DSC it is
the heat necessary to equalize the temperature between the
sample and the reference which is measured. This technique
thus can be used to measure enthalpic changes quantitatively.
TGA is a technique in which a sample is continuously
weighed as it is heated at a linear rate. The resulting
thermogram gives information concerning the thermal stability
of the substance under investigation. TGA was first described
by Honda ^2^in 1915. Griffith* ^nas applied TGA to the
study of mixtures of hydrates and anhydrous salts.
Thermal methods of analysis are uniquely applicable for
study of dehydration processes. Dehydration can be observed
as endothermic changes in DTA, heats of dehydration in DSC,and
as weight losses in TGA.
- 162 -
-------
RESULTS AND DISCUSSION
First, the DTA results will be considered. A typical
thermogram for the 'dehydration of MgS03.3H20 is shown in
Figure 1. Only one endothermic transition is observed, start-
ing at 190°C. On the other hand, in the thermogram of
MgS03.6H20 shown in. Figure 2, two endothermic transitions are
observed. One starts at 90°C and the other coinciding with
the endotherm observed for the trihydrate, starts at 190°C.
The inference to be drawn from this data is that the
hexahydrate does degrade in two steps and that the two steps
involve a transition from the hexaform to the triform.
Next, studies were carried out using DSC. In this case,
it is the heat input rather than the temperature which is
measured. The DSC thermogram of MgS03.3H20 is shown in
Figure 3. Only one endothermic transition was observed
starting at 100° C with a. peak maximum at 160° C. Both DTA
and DSC analyses indicate that the thermal dehydration of
MgS03.3H20 is a one step process. From DSC data, it appears
that the dehydration of MgS03.3HpO starts at a low temperature,
100°C, in contrast to the DTA data from which it is inferred
that 190°C is the starting temperature.
The DSC thermogram of MgS03.6H20 is shown in Figure 4.
It is significant to note that only one endothermic transition
was observed starting at 45°C with a peak maximum at 90°C.
DSC results suggest that the thermal dehydration of
- 163 -
-------
takes place in one step starting at 45° C, in contrast to the
two step mechanism starting at 90°C to be drawn from the DTA
results.
Since the phenomena measured by DTA and DSC is the same,
it was expected that the results would be consistent.
Surprisingly, this was not the case. From the DSC data, it
appears that the dehydration of both hydrates starts at a
lower temperature and the hexaform dehydrates in one step and
not in two steps as was inferred from the DTA results.
The apparent contradiction between the DTA and the DSC
results can be rationalized by considering the relationship
between the dehydration reaction and the sample environmental
conditions. In any dehydration reaction water is liberated;
if the latter is continuously removed from the reaction
atmosphere two consequences are observed. First, the
dehydration starts at a lower temperature, and, secondly,
equilibrium is not attained.
On the one hand in the DSC studies, the sample is placed
in an open dish and is heated under a sweeping stream of
nitrogen. Under these conditions the liberated water is
continuously removed from the reaction environment. On the
other hand, in the DTA studies, the sample is heated in a
self-generated atmosphere because it is placed in a capillary
tube. Under the open conditions encountered in DSC, thermal
dehydration starts at a lower temperature, t5°C versus 90°C
- 164 -
-------
and only one endo thermic transition is observed for the hexar
form, whereas in the HA studies two were observed.
To confirm the above rationalization of the differences
between the DTA and the DSC studies, the thermal dehydration
was studied by another independent technique, TGA, in which
it was possible to heat the samples' either in an open
condition or in a self-generated atmosphere. TGA thermograms,
under open conditions, were obtained by the conventional
procedure in which the sample is placed in an open platinum
dish under a sweeping blanket of nitrogen. TGA, under self-
generated atmosphere, was achieved by placing the sample in
a capillary tube with a thermocouple inside. Then, the whole
tube was placed in the platinum dish.
In Figure 5, the TGA of MgS03.3H20, under open conditions
is shown. Thermal dehydration starts at 100° C and takes place
in one step. The weight loss of the sample is 34.0%, which
corresponds to the loss of 3 moles of water. The effect of
operating under open conditions is shown in Figure 6 for
MgS03. 6H20. Thermal dehydration of the hexaform starts at
70° C and takes place in one step. The weight loss of the
sample is 51%, which corresponds to the loss of 6 moles of
water.
The thermograms for the trihydrate and the hexahydrate,
respectively, under the conditions of a self-generated
atmosphere arc shown in Figures 7 and 8. MgSC^.SHgO, as
- 165 -
-------
shown in Figure 7, loses 34.5% of its weight in one step.
This is in a very good agreement with TGA under open conditions,
but thermal dehydration starts at 220°C, i.e., at a higher
temperature, because of the self-generated atmosphere con-
ditions. The weight loss for the hexahydrate under a self-
generated atmosphere, as shown in Fipure 8, is 51%, which
corresponds to the loss of 6 moles of water. It is significant
to note that the thermal dehydration of the hexaform starts
at a higher temperature and takes place in two steps, with a
weight loss of 25.5% in each step. In other words, MgS03.6H20
loses 3 moles of water in' each dehydration step.
Thus, when the TGA study of the hexaform is carried out
under a non-equilibrium condition, because of heating the sam-
ple under a sweeping blanket of nitrogen, equilibrium is not
attained due to the continuous removal of the liberated
water. As a result, thermal dehydration starts at a lower
temperature and takes place in one step. On the other hand,
if equilibrium is attained, because of heating the sample in
a self-generated atmosphere, the TGA results showed that
thermal dehydration starts at a higher temperature and takes
place in two steps. First, dehydration leads to the formation
of MgS03.3H20, then, to the anhydrous MgS03. These TGA
observations are consistent with the explanation of the
differences observed in the DTA and the DSC studies. The
results are summarized in Table 1 for MgS05.3H_0 and in
- 166 -
-------
Table 2 for MgSO .6H 0. The onset of the endotherms as a
function of the method are compared.
- 167 -
-------
APPLICATION
The observations made in the TGA studies were used as
bases for developing a new analytical method for quantitating
mixtures of the tri- and hexahydrates. It had been observed
that although the second weight loss for the hexahydrate
overlapped that of the trihydrate, nevertheless, from a
knowledge of the first weight loss of the hexahydrate, the
contribution of the hexahydrate to the second weight loss
could be calculated.
The accuracy of the method was investigated by analyzing
the thermograms of synthetic mixtures of the hexahydrate and
the trihydrate ranging from 10% to 90%. The thermograms are
shown in Figures 9-19.
From these thermograms, the water content and the mag-
nesium sulfite content could be calculated. First, the water
content calculations will be considered and compared to the
theoretical values. The water content of each hydrate in a
mixture can be obtained from the TGA thermogram, by taking
into account the fact that the weight loss in the first
dehydration step at 175°C represents the first three moles of
water in MgS03.6H20, i.e., 50% of the water content of the
hexaform. Thus the water content can be calculated, from the
TGA thermogram, as follows:
% H20 in MgS03.6H20 in a mixture = (% weight loss
in the first step at 175°C) (2)
- 168 -
-------
% H20 in MgS03.3H20 in a mixture = & weight loss in
the second step at 100° C) - (% weight loss in the first step at
175° C)
The theoretical values of the water content are calculated
as follows:
% H20 in MgS03.6H20 in a mixture = % MgS03.6H20 in the mixture
X 6H?0
MgS03.6H20
X 100
= % MgS03.6H20 in the mixture
x
= % MgS03.6H20 in the mixture
X 50.9
% H2Q in MgS03.3H20 in a mixture = % MgS03.3H20 in the mixture
MgS03.3H20 in the mixture
- X MO
= % MgS03.3H20 in the mixture
X 34.1
Both theoretical and observed values of the water content are
shown in Table 3. The percent water (observed values) is
plotted versus magnesium sulfite content in Figure 21 for
MgS03.3H20 and in Figura 22 for MgS03-6H20.
- 169 -
-------
The magnesium sulfite content can be calculated from the
TGA thermograms by considering the same aspects mentioned in
the water content calculations, and, taking into account that
6 moles of water represent 50.9% by weight of MgS03.6H20,
and that 3 moles of water represent 31.1% by weight of
MgSO^.S^O. Thus jhe percent of each hydrate in a mixture is
calculated as follows:
% MgS03.6H20 in a mixture = % weight loss at 175°C X 2 X 100
50.9
% MgS03.3H20 in a mixture = (% weight loss between 175° and
«M)0°C _% weight loss at 175°C)
X 100
3TTI
The data obtained is compared versus the theoretical values in
Table 4 for the triform and in Table 5 for the hexaform. The
accuracy of the method, as shown in Tables 4 and 5, was found
to be - 3% if no calibration curve is used. A value of less
than 1% is to be expected if a calibration curve is used.
Furthermore, the TGA method was tested in the presence of
TGA inactive material. This was achieved by analyzing a
mixture of MgS03.3H20, MgSG^.eh^O and glass beads. The mixture
was subjected to the same experimental conditions of a self-
generated atmosphere. The TGA thermogram is shown in Figure
20. From the results summarized in Table 6, it is concluded
that this method does not suffer any interferences due to the
presence of thermally inactive materials.
- 170 -
-------
Relevant to the nature of'the materials obtained in the
Magnesia Process, the quantitative analysis of mixtures of
MgSOj.SHgQ and MgS03.6H?0 has to be considered in the light
of the following contexts: one, only these two hydrates are
present; two, in the presence of MgO, three, in the presence
of another hydrate, MgSO^.7HpO. The latter is a common
product in the Magnesia Process.
When only the two hydrates are present, methods available
include total sulfite by iodine titration , X-ray
analysis , and a wet chemical method developed by Dr. Ray
which will be referred to as Ray's Method' '.
In the presence of MgO and/or MgSO,.. 7^0 the measurement
of the sulfite content cannot be used to determine the ratio
of the hydrates. From the available literature which is limit-
ed to the one paper previously cited , it is not clear as
to whether or not quantitative analysis is possible using the
X-ray method in the presence of MgO and/or MgSO^^HjO. Ray's
method is applicable in the presence of these materials.
The TGA method is simpler than Ray's method but it is
not applicable in the presence of MgSO^ 7H.O. On the other
hand, Dr. Ray*-7-' has developed a simple procedure for
quantitatively stripping MgSOg.. 71*20 from mixtures of
MgS03.3H20, MgS03.6H20 and MgO. Therefore it appears that the
TGA technique in conjunction with Ray's stripping procedure
is the optimum method.
- 171 -
-------
REFERENCES
(1) Griffith, E.J., Anal. Chem.. 29, 198 (1957)
(2) Honda, K., Sci. Rept. Tohoku Uni., 4,97 (1915)
(3) Karchmer, J.H., The Analytical Chemistry of Sulfur
and its Compounds. New York, Wiley Interscience,
Vol. 29, Part I pp 225-226 (1970)
(4) LeChatelier, H., Bull. Soc. Franc. Mineral.
10, 203 (1887)
(5) Okabe, T. and Hori, S., Tohoku University Technology
Report. 23(2), pp 85-89 (1959)
(6) McGlamery, G.G., Torstrick, R.L., Simpson, J.P., and
Phillips, J.F., Conceptual Design and Cost Study.
Sulfur Oxide Removal From Power Plant Stack Gas.
EPA-R2-73-244, pp 33-34, (1973)
(7) Ray, A.B., Chemical Construction Corporation, Private
Communication (1974)
(8) Shah, I.S., Chemical Construction Corporation, Recovery
of Sulfur Dioxide From Waste Gases. U.S. 3, 577, 219
(1971)
(9) Wendlant, W.W., Thermal Methods of Analysis, Second
Edition. New York, Wiley Interscience, pp 134-201
(1974)
(10) Wendlant, W.W., Hoiberg, J.A., Anal. Chem. Acta.
28, 506 (1963)
- 172 -
-------
Table 1
Starting Temperature of Thermal
Dehydration of MgSO?.3HgO
Technique Experimental Conditions Onset of the Endotherm
DTA self-generated 190°C
DSC open conditions 100°C
TGA open conditions 100°C
TGA self-generated 200° C
- 173 -
-------
Table 2
Starting Temperature of
Thermal Dehydration of MgSOg.SHnO
Technique Experimental Condition First Endotherm Second Endotherm
DTA self-generated 90° C 190° C
DSC open conditions 45°C not observed
TGA open conditions 70° C not observed
TGA self-generated 125°C 220° C
- 174 -
-------
Table 3
Calculated and
of Water
Observed Values
Content in
Synthetic Mixtures
i
M
Ul
1
Stdtt
1
2
3
4
5
6
7
8
9
10
11
Composition
%MRSOP.3H,0 *MKSOo6HoO
0.0%
10.29%
18. 96%
29.37%
38.24%
50%
59.84%
70.19%
79.34%
90.18
100%
100%
84. 71%
81. 04%
70. 63%
61. 76%
50%
40. 16%
29. 81%
20.66%
9.82%
0%
Calculated 1120 content in Observed HpO content In
Triform
0%
3.5%
6.46%
10%
13.03%
17.04%
20.39%
23.91%
27.03%
30.72%
34.1%
Ilex a Form
50. 9%
45.56%
41. 19%
35.9%
31.39%
25. 42%
20. 41%
15.15%
10.5%
4.99%
0%
Total Triform
50.9%
49.06%
47.65%
45.9%
44. 42%
42.46%
40.8%
39.06%
37.53%
35.71%
34.1%
0%
3.4%
6.6%
9.0%
12%
16.4%
19.8%
23%
26.4%
30.2%
34.8%
Hexaform
52%
44.4%
39.4%
35.0%
32%
26%
20.4%
16%
10.6%
5%
0%
Total
52%
47.8%
46%
44%
44%
42.4%
40.2%
39%
37%
35.2%
34.8%
-------
Table 4
MgSOa.3H^O Content from TGA
Stdtt
1
2
3
4
5
6
7
8
9
10
11
Theoretical Value
0%
10.29%
18. 96%
29.37%
38.24%
50.0%
59.84%
70.19%
79.34%
90. 18%
100.00%
Observed Value
0%
10. 26%
19.35%
26.39%
35.24%
48.1%
58.06%
67.45%
77. 42%
88.56%
102.0%
Deviation
0%
-.03%
+.39%
-2.98%
-3.00%
-1.9%
-1. 78%
-2.74%
-1. 92%
-1.62%
+2.0%
- 176 -
-------
Table 5
Stdft
1
2
3
4
S
6
7
8
9
10
11
MeS(h.6H,0
From
Theoretical Value
100%
89.71%
81.04%
70.63%
61.7696
50%
40.16%
29.81%
20. 66%
9.82%
0%
Content
TGA
Observed Value
102.2%
87.3%
78. 45%
68.8%
62.9%
51.11%
40.1%
31.45%
20.84%
9.83%
0%
Deviation
4-2.2%
-2.41%
-2.54%
-1.83%
+1.14%
+1.11%
-.06%
+1.64%
+.18%
+.01%
0%
- 177 -
-------
Table 6
TGA results of a synthetic
mixture containing thermally
inactive compound
Theoretical Value Observed Value
MgS03.3H20 W.3% «.2#
MgS03.6H20 30.096 29.3%
glass beads 25.7% 27.5%*
*This value was obtained by difference
- 178 -
-------
•Urt
SAMPLE: MgS03.3H20
Fig. No. 1
ORIGIN:
O
x
UJ
vo
O
O
Z
UJ
SIZE C. jng
R E F Sla_ss_beajls_.
PROG. MODE__heat_
RATg 10 JE ,START_3fl_°C
ATM._se.i.F-ii^incaJted. MM
SCALE
SHIFT
AT
.-a
RUN NO L
DATE l
OPERATOR.
RM
BASE LINE SLOPE.
ui
>
•o
m
TO
c
T °r (rHRHMFI • AMIMFM
-------
SAMPLE: MgS03.6H20
Fig. No. 2
ORIGIN:
O
x
u
CO
O
O
O
Z
UJ
SIZE.
RE F._.glass_ .Leads
PROG. MODE beat_
RATE in A ,START_IO_°C
ATM ._s_e 1 Jl-ar-nptMteil MM
SCALE
SHIFT
_T
Si I
J"
JLJN.
AT
2 '«-
fc lu
RUN NO ^JL
DATE 11/B/7U
OPFRATOR RM
BASE LINE SLOPE.
U
n
T, °C (CHROMEL: ALUMED*
• »l mSrUJCTIOM MMUAL fOI ICALf COMECTIOII
-------
I
M
03
t->
I
o
X
IU
SAMPLE: N«S03.3H20
Fig. No. 3
ORIGIN:
SIZE.
REF..
12 rug
pmpty pan
PROG. MQHE heat
RATE 10 A ,START_25_°C
ATM. No ,_, llL_MM
T
1
SCALR
SHIFT
RUN NO L
OPERATOR QM_
BASE LINE SLOPE Q_
tfl
>
S
•o
m
ido°c
•a
c
2
O
o
z
UJ
160°C|
T. °C (CHROMEL: ALUMEL)*
• tit wxraucnoH MINUM. roi ic*if eo»«icno»
-------
SAMPLE: MgSO-.BH 0
Fig. No.
ORIGIN:
SIZE.
5 mg
REF—empty pan
PROG. MODE heat
RATE 10 A ,START_12_°C
ATM. "2 3JL_MM
SCALE
SHIFT
_.!_
50
JL_IN.
AT
_L_fr
RUN NO
OPERATOR_RM_
BASE LINE SLOPE Q.
•o
r-
m
T. °C (CHROMEL: ALUMEL)*
-------
SAMPLE: MuS03.3H2Q
Fig. No. 5
SIZE ...20 mg.
s
X-AXIS
TEMP. SCALE. ..50 ^C_
inch
SHIFT 0 ._ inch
TIME SCALE (ALT.)
Star
t of D
^
shydra
N
\
50 100 150 2
tiori '
\
X
^^
Y-AXIS
SCALE .'I _ '"9-_
inch
(SCALE SETTING X 2)
SUPPRESSION. _50_._
00 250 JU
TEMPERATURE*. °C
—
— mg.
RUN NO. 1. _. DATE11/12/71
OPERATOR . RM . .
HEATING RATE . 1»> °_C_
min.
ATM. N2, 30 ml /min.
TIME CONSTANT 1... sec. .
u Jbu m
• WPIT COmiCTIOli FBI NOM-UHIAI
U '
IIT» BF CHROHIL-AIUIHI
5ti 500
THIIHOCOUPIIS
00
CJ
100
80
'10
20
-------
oo
ion
80
SAMPLE: MgS03.6H20
Fig. No. 6
SIZE... 20 . mg.
— --,
"\
X-AXIS
TEMP. SCALE 5C
SHIFT 0 ...
TIME SCALE (AL1
y
\
\
\
\
">>
°C
inch
inch
r.)
^
** — .
Y-AXIS
SCALE '* T»9-
inch
(SCALE SETTING X 2}
SUPPRESSION .5 mg.
'•
"! .. . i
RUN NO. . 1. . .DATE11/12/7I
OPERATOR KM
HEATING RATE 10 .. ^C_.
min.
ATM. N2, 30 ml/min. .
TIME CONSTANT . 1 sec. .
*•
UJ
«*0
20
50
100
150
350
*rn.r CORRECTION foa
400
450
500
«muri i»r«unrmin rs
-------
I-1
CO
(J\
I
100
80
f
h-*
§6n
2
*
20
SAMPLE:
Fig. No.
SIZE . 5
7
mg.
X-AXIS
TEMP. SCALE 5n _'
i
SHIFT . ° .. . inch
TIME SCALE (ALT.). .
>
>£
nch
ss
\
\
Y-AXIJ
SCALE 2 mg
inc
(SCALE SETTING X 2)
SUPPRESSION
•jait o
^
\
:.
: Dehy
V,
^ —
Jratio
^--^
\
60 mg
— my.
-_
1
*— ^
/
RUN NO. 1 DATEll/^/7'l
OPERATOR RM . .
HEATING RATE in ?C .
min.
ATM. sel P-generatod
TIME CONSTANT. 1. sec. .
34%
i
weigh'
.
. loss
V
*
SO 100 150 200 250 0 300 350 MOO HaU auu
TEMPERATURE , C • «m» comucno* roi MM-UMuin or CMOHIL-M.UMIL Tnfimocoun.»
-------
SAMPLE: MgS03.6H 0
Fig. No. 8
SIZE 5 mo.
—
>
X-AXIS
TEMP. SCALE .5
SHIFT _.. 0
TIME SCALE (ALT
\
\
V
—
— -.
o °c
inch
inch
.)
^>v
\
\
V
Y-AXIS
SCALE I «"9-..
inch
(SCALE SETTING X 2)
SUPPRESSION 60 . mg.
•—
"
"" -
"-
RUN NO. 1 . DATE1L/16/7M
OPERATOR RM
HEATING RATE 10 °C
mm.
ATM. self -generated
TIME CONSTANT . sec. .
51%
-
weighi
HMBM|BBaM
; lois-
:*
so
100
150
200
250
o/-
300
350
"mo
450
500
-------
SAM
Fij
SIZE
°LE:
10
3. No.
12
Std #1
0% MgS03.6H20
9
. mg.
"" -v
X-AXIS
TEMP. SCALE ..5
SHIFT 0 .
TIME SCALE (AL1
\
\
V
••
~ —
o °c
inch
inch
-.)
^
\
\
V
Y-AXIS
SCALE 2. . mg.
inch
(SCALE SETTING X 2)
SUPPRESSION. _.50_
— -— .
•
•=9BV
— —
rnci,
•«s=:
RUN NO. ^ DATE 11/17/71
OPERATOR . RM
HEATING RATE 10. °C.
min.
ATM. sol F-gencrated
TIME CONSTANT . .1 sec. .
JT
50 100 150 200 250 300 350 400 USO 500
TEMPERATURE C • *»«.» co««cno« »• non-imf/utiT* or CHIOMIL-M.UHIL THiDHOCoumi
-------
SAMPLE: Std #2
10.295$ MgS03.3H20
89.71% MgS03.6H20
Fig. No. 10
SIZE in . mg.
.
A
\
100
" -
\
\
\
200
X-AXIS
TEMP. SCALE K
SHIFT 0
TIME SCALE (A LI
^
"*
— ».
1
10 °C
inch
inch
r.)
>x_
Y-AXIS
SCALE 2 mg...
inch
(SCALE SETTING X 2)
SUPPRESSION 50 mg.
-
RUN NO. L DATE 11/2.0/7
OPERATOR RM
HEATING RATE 10 fC
min.
ATM. self-generated
TIME CONSTANT 1 sec. .
•
300 UOO 500 . 600 700 800 900 1UUI
TEMPERATURE . C • »«.» CORHICTIOM lot NON-LiMuin or CHROMI-ALUMII. iH»nocotin.tt
-------
1
M
CO
1
lot
80
*
mm
J]
£
Ml)
20
SAMPLE: Std #3
81.0'J?5 MgSO,.6H 0
3 2
Fig. No. 11
SIZE ...iS mg.
^
X-AXIS
TEMP. SCALE 50 !
-------
SAMPLE: Std #4
2'J.37% MgS03.3H20
70.03% MgS03.6H20
Fig. No. 12
SIZE. .10 mg.
•
X-AXIS
TEMP. SCALE 50 __°C_
inch
SHIFT.—. 0 inch
TIME SCALE (ALT.) _
\
V
_
—- ^.
•^
\
V
V
Y-AXIS
SCALE 2 mg.
inch
(SCALE SETTING X 2)
SUPPRESSION 50 .
• — — —
—
.
- . mg.
•^^— MMM
RUN NO. 1 DATE H/22/71
OPERATOR RM
HEATING RATE 10 *C
min.
ATM. sn 1C- generated
TIME CONSTANT 1 sec. .
M^^^BMM^BM .
SO 100 150 200 250 300 350 MUO 450 5UC
TEMPERATURE , C • «mr CMKCTHM to* ran Lmiuinr or CHHOnft.-M.uMti. imiiMocoun.»
vo
o
ion
80
20
-------
1
I-1
vo
M
1
ion
SAMPLE. Stil #5
38.21*% MgS03.3H20
01.70% MgS03.6H20
Fig. No. 13
SIZE.. 12 mg.
-
•^^^™
• • • i
X-AXIS
TEMP. SCALE 50 ^.Q_
inch
SHIFT... ° inch
TIME SCALE (ALT.)
\
v_
~ '
— • —
\
V
Y-AXIS
SCALE ^ .. >"9-_.
inch
(SCALE SETTING X 2)
SUPPRESSION... 50 mg.
"--^
—
50 100 150 200 25(1 m 30
TEMPERATURE*. °C
—
™ ^«^^™«^™
RUN NO. 1 DATE11/25/74
OPERATOR R"
HEATING RATE 10 . ?C .
min.
ATM . se 1 f -generated
TIME CONSTANT 1 sec. .
[) 350 400 'IbO 500
• AW¥ CORNCCTlOM fON NOM-LINCAftin Of CMOMlL ALUMIL TMIMUOCOUPUl
80
UJ
HO
20
-------
vo
to
SAMPLE: Std #6
50?5 M«S03.3H20
50% MtjS03.6H20
Fig. No. 1U
SIZE . 10 mg.
•MW^^
^
X-AXIS
TEMP. SCALE 50 ^C_
inch
SHIFT. .0 . inch
TIME SCALE (ALT.) _..
\
v_
'" —
• 1 -^
^N
\
\
V
Y-AXIS
SCALE 2 P'9
inch
(SCALE SETTING X 2)
SUPPRESSION mg.
-^_
«-•»--._
•• — -
RUN NO. I DATEH/ZS/TU
OPERATOR . RM _
HEATING RATE 10 °C .
min.
ATM. sol F-.v;enerated
TIME CONSTANT 1 sec. .
._
100
t
•
|c,n
UJ
20
50
100
150
200 2^0 . . 300
TEMPERATURE*,°C
350
'100
500
*»nr comii ITIOII »o« KOM iiNKmrr Of CMROMCL-UUMIL IHICMOCOUPLI*
-------
1
U)
1
100
80
H
gr.ii
Ul
MO
20
SAMPLE: Std #7
50. 8'J% MgSO .311-0
ll-0.10/o MgSO-3.611 0
Fig. No. 15
SIZE _12 mg.
bU it
— ~^
X-AXIS
TEMP. SCALE 50 °C_
inch
SHIFT 0 _ inch
TIME SCALE (ALT.)
V.
U 1
MHMMM^h.
1 -»
~\
\
Y-AXIS
SCALE 2 mg.
inch
(SCALE SETTING X 2)
SUPPRESSION ..50 mg.
^
— ^— ^
~
~^M^HM»
=B*BB^
RUN NO. 1 DATEL1/27/7II
OPERATOR RM
HEATING RATE 10 . °C .
min.
ATM. sc'l f-gonerated
TIME CONSTANT L sec. .
0 200 2SO f o 300 350 MOO M5M 501
TEMPERATURE , C • »*«.» COH«ICTIOH ton mom imt«Rir» or CH»OI«I uuuti IHIHMOCUUPK*
-------
SAMPLE: Std #8
70.19% MgS03.3H20
29.81% MgS03.6H 0
Fi«. No. 1C
SIZE 12 mg.
'
^
X-AXIS
TEMP. SCALE 50 !C_.
inch
SHIFT 0 inch
TIME SCALE (ALT.) .
^
^MIM^MnM
— — «^
^
\
>*v
Y-AXIS
SCALE .2 mg._
inch
(SCALE SETTING X 2)
SUPPRESSION 50 . _mg.
*• — ->
—
—
— — ^.««B
RUN NO. . * . DATE H/2S/7'
OPERATOR RM
HEATING RATE in °C.
min.
ATM. self-generated
TIME CONSTANT 1 sec. .
VO
*»
1
ion
80
t
LLJ
MO
20
"50
"100
150
200 25n _ ft 3no
TEMPERATURE*.°C
350
'ino
M50
500
• APPLY COHRCC1ION FOR NON LINIAH1IY Ol CHROMEL ALUM1L IMrRMOCOllPlIS
-------
SAMPLH. Std #'.l
JO. 3M& MgS03.3H20
2n. r>r,% Mgso3.6ii2o
Fig. No. 17
SIZE _._.!'» mg.
j
— =s
. X-AXIS
TEMP. SCALE 50 _.^C_
inch
SHIFT. 0 . . inch
TIME SCALE (ALT.) L_
^
^^•^•^••M.
— ^
\
\
V
Y-AXIS
SCALE 2 mg.
inch
(SCALE SETTING X 2)
SUPPRESSION ._5.°_..mg.
*-^.
~— ^_
RUN NO. 1 DMElV^7/7'
OPERATOR RM
HEATING RATE l» °C
mm.
ATM. self-generated
TIME CONSTANT L sec. ;
™«,
VD
Ul
ion
80
§•(>
UJ
'in
20
50
100
150
350 «mn l«5n soo
APPlr COBHtCTION fOII HOH-ilMAIIIK Of CMROIKl MUMCl IXKUOCOUPlll
-------
I
M
\0
ion
80
*
H
LlJ
«
'Ml
20
SAMPLE: Std #10
90.18ft MyS03.3H20
').&<>% MgS03.GH20
Fi«. No. 18
SIZE 10 . mg.
V
X-AXIS
TEMP. SCALE 5f) . _°C_
inch
SHIFT ... 0 . inch
TIME SCALE (ALT.) .. . _
—
-'
.
^
\
\
V
Y-AXIS
SCALE 2 . rr.g
inch
(SCALE SETTING X 2)
SUPPRESSION 50 . mg.
.
— — —
—»—«-««-
•V^K^^BHBM
RUN NO. 1 DATE 11 '30/7'
OPERATOR RM
HEATING RATE 10 °C
min.
ATM. sclf-^onnrnted
TIME CONSTANT 1 sec. .
•••
•
50 100 150 200 250 300 350 iJOO U50 50(
TEMPERATURE , C • »'«.» convcnM ron MM LINCMITT or CHUOMH W.UML TnCMiocourifS
-------
SAMPLE: Std #11
100?£ MRS03.3H20
Fig. No. 1«J
SIZE I1* mg.
-L
X-AXIS
TEMP. SCALE 5
SHIFT 0 .
TIME SCALE (ALT
bO 1UU 15U 20
n °c
inch
inch
'.)
^
>
\
V
Y-AXIS
SCALE 2 mg. .
inch
(SCALE SETTING X 2)
SUPPRESSION 50 mg.
•*• — ^
•^ —
-~
^•^^MM..
RUN NO. 3 D/'TEL?/1/7H
OPERATOR RM
HEATING RATE in °C _
™i"-
ATM. self-Kcnerbted
TIME CONSTANT 1 sec. .
0 250 ^ 300 3bll MdO 4bO 50(J
TEMPERATURE « C • »»n.t COMICTIOM r
-------
M
VO
00
1
inn
80
gm
IU
xo
o^
MO
20
SAMPLE: A mixture of:
'14.3% MgS03.r>n20
30.0% MyS03.3H20
25. 1% glass beads
Fig. No. 2n
SIZE 15 m0.
•••••^•••i
X-AXIS
TEMP. SCALE 5n _.°Q_
inch
SHIFT . n inch
TIME SCALE (ALT.) _ _ _
X
s^
.
->s
N
Y-AXIS
SCALE 2 mg.
inch
(SCALE SETTING X. 2)
SUPPRESSION .50 . . mg.
\
*
_
•.^.^H^MHI
RUN NO. 1 DATE 12/1/74
OPERATOR RM
HEATING RATE ln °C .
mm.
ATM. so If- ^operated
TIME CONSTANT 1 . sec. .
i^M^B«^^m
• m
50
100
150
350
400
450
500
• »»«.» COHHCCTION KM NON-llNCMir» Of CHHOMIL ALUHIl. TMf KMOCOUPHS
-------
90%
100%
-------
£3
JQUW 13 1 II (0 IKt «UI hCI AStlU C5
70%
60%
OQ
O
ro
ro
«JO%
100%
-------
APPENDIX 3
SULFUR OXIDE REMOVAL FROM POWER PLANT STACK GAS
Magnesia Scrubbing-Regeneration: Production of Sulfuric Acid
A Mass Spectrometric Study of the Thermal Degradation
of MgS03 • 3H20 and MgSO • 6H20
by
Prof. Leonard Dauerman
Department of Chemical Engineering and Chemistry
New Jersey Institute of Technology
Newark, New Jersey
Prepared for
Chemical Construction Co.
New York, New York
February, 1975
- 201 -
-------
INTRODUCTION
In the Magnesia process for the removal of SO. from the flue gas and
power plants, two hydrates of magnesium sulfite - the trihydrate and the hexa-
hydrate - are formed. At a later stage these hydrates are calcined; MgO is
formed which is recycled and, also, SO., which is converted into sulfuric acid.
It was the aim of this study to answer several questions pertaining
to the calcining process: How do the hydrates decompose under vacuum condi-
tions? Are trace amounts of H_S formed in the thermal decomposition? How
valid is the thermodynamic calculations on the partial pressure of SO. as a
function and temperature. (K. Schwitzgebel and P. S. Lowell, "Thermodynamic
basis for existing experimental data in Mg-SO.-O. and Ca-SO.-O. systems," Env.
Sci. & Tech..7. No. 13, 1147 (Dec. 1973).
The technique used in this study was mass spectrometry.
EXPERIMENTAL:
Samples of the hydrates were heated in the ionization source of the
Finnigan Quadrupole 1015 mass spectrometer by utilizing the solid inlet probe.
The temperature of the solid could be heated up to 400 C and the ambient pres-
sure was approximately 10 mm. Samples were heated to a given temperature
and spectra were taken after the spectral pattern had stabilized.
RESULTS AND DISCUSSION
Two series of runs are presented as representative of the data obtain-
ed. In Run //370, the spectra presented are those for the decomposition of
MgSO '3H20. Spectra are recorded from 35 to 400 C. The three parallel re-
cordings on each spectrum represent data recorded at three different sensitivities
- 202 -
-------
1:3:9. A similar set of records for MgSO '6H_0 are presented in Run #371 series
of spectra. As a reference record, the spectrum of SO-.H.O is shown in Figure 1.
To determine whether or not there is a difference in the thermal de-
composition between the two hydrates, the ratios of the peak heights of -- as
a function of temperature were calculated.
In making these calculations data from an additional run for the tri-
hydrate, and two additional runs for the hexahydrate were used. In Tables 1-5,
the relevant data taken from the spectra are tabulated, and the value of the
•j;-£ ratio is calculated. In Table 6, the ratios for the three runs for MgSO-
HoO j
• 6H70 are averaged at each temperature, and the logs are calculated. In Table 7,
similar operations are carried out on the two MgSO '3H_0 runs.
SO?
The log CUQ) as a function of temperature for the hydrates *s plot-
ted in Figure 1. The trend is similar. And in the light of the precision of
the data, it can not be concluded that there is a difference in the way that the
hydrates decompose.
Another question raised is whether H.S, even in trace amounts, is ob-
served. The peak characteristic of H.S is observed at an m/e equal to 34.
This ra/e is observed in some spectra, for example, Run #370 at 400°C. The ratio
of ^32/^64 is 0.00547. In spectra Figure //I, for SO., where the intensity of
m/e 64 is of a similar value the ratio is 0.00550. The fact that the difference
in the ratio is the within the error in measuring the intensities means that the
34 is not an independent peak due to a separate species. It is, therefore, con-
cluded that H-S does not form even in trace amounts when the hydrates ther-
mally decompose.
Note that SO- is observed at very low temperatures. In Run #370, i-fc *-s
observed at 35 C. But the pressure is ~ 10 mm. From Figure 2 in the previously
- 203 -
-------
cited paper by Schwitzgebel and Lowell, in which the equilibrium constant is
plotted against temperature for the system MgSO,-*-SO- + MgO, it is seen that at
35°C, the predicted value of the SO. pressure is 1.2xlO~ mm. This prediction
is consistent with the observation made in this study, therefore, it is con-
cluded that the equilibrium calculations are verified for the low temperature
region.
CONCLUSIONS
1. Under vacuum conditions, MgSO '3H.O and MgSO3'6H_0 degrade similarly.
2. H»S is not a thermal degradation product of the magnesium sulfite hydrates.
3. The thermodynamic calculations for the equilibrium MgSO_-«-MgO + S02 were
confirmed in the lower temperature region, 50-100 C.
- 204 -
-------
MASS SPECTRA OF MgSO3.6H20 & MgS03.3H2O
RUN SERIES 370, 371, REFERENCE MATERIAL
The numerous mass spectra taken at the temperature
intervals given in the table are not presented here
since the pertinent data have been reduced in the
subsequent tables and graphs.
- 205 -
-------
«
Tables 1-5. CALCULATION OF ^£ INTENSITY RATIOS AS A
H20
FUNCTION OF TEMPERATURES FOR RUNS OF VARYING
DETECTION SENSITIVITY FOR MgS03'3H20 and
MgS03-6H20.
- 206 -
-------
Table 1. MgSO '3H.O; Run #370; Detection Sensitivity, 1000 microamps.
Temp
35°C
50
100
150
200
250
300
350
400
64
17
8
11
13
13
28
42
110
350
m/e
18
1500
40
47
33
26
34
17
25
47
m/e
17
480
11
13
9
7
9
5
7
13
S02 64
H.O 18
0.01133
0.20
0.234
0.394
0.50
0.823
2.47
4.40
7.445
m = atomic weight
e = charge
- 207 -
-------
Table 2. MgS03'6H20; Run #371; Detection Sensitivity, 1000 Microamps
(1)
Temp
40°C
50
100
150
200
250
300
350
400
m — f
m/e
64
36
250
250
46
86
200
440
940
3250
m/e
18
4610
15600
1830
150
180
260
260
200
350
m/e m/e
17 16
1540 90
4900 320
580 40
38
45
52
50
42
68
S02 64
H20 = 18
0.007810
0.01603
0.1366
0.307
0.478
0.769
1.692
4.70
9.286
m = atomic weight
e = charge
- 208 -
-------
Table 3. MgS(>3'6H20; Run //372; Detection Sensitivity, 500 Hicroamps.
Temp
35°C
70
150
200
250
300
350
400
64
5
24
50
33
50
98
207
423
m/e
18
450
1674
315
80
61
63
35
72
m/e
17
150
558
100
22
18
18
10
18
S02 m 64
H,0 " 18
0.0111
0.01434
0.1587
0.4125
0.8197
1.556
5.914
5.875
m = atomic weight
e = charge
- 209 -
-------
Table 4. MgSO '3H.O; Run //373; Detection Sensitivity, 600 Microamps,
Temp
35°C
50
100
150
200
250
300
350
400
m/e (1>
64
- -
13
10
2
2
12
20
36
113
m/e
18
16
500
350
33
32
150
140
26
42
m/e
17
5
130
80
9
8
30
28
7
11
S°2 64
H20 " 18
- -
0.026
0.0286
0.0606
0.0625
0.080
0.143
1.385
2.69
* m = atomic weight
e = charge
- 210 -
-------
Table 5. MgSO *6H20; Run #374; Detection Sensitivity, 750 Microamps.
Temp
50
100
150
200
250
300
350
400
tf(1>
13
74
160
55
88
200
520
850
m/e
18
135
4200
960
160
115
98
65
100
m/e m/e
17 J.6
45
1400 91
320
36
30
22
17
27
S02 64
H20 = 18
0.0963
0.1771
0.1667
0.3438
0.7652
2.564
8.00
8.50
' m = atomic weight
e = charge
- 211 r-
-------
SO,
Table 6. Ave. Ratio of H20 as a function of T (MgSO '6H 0)
Temp ( C) Run #371
Run #372
Run #374
Ave
Log Ave
35
40
50
70
100
150
200
250
300
350
400
— —
0.007810
0.01603
—
0.1366
0.307
0.478
0.769
1.692
4.700
9.286
0.0111
—
—
0.01434
—
0.1587
0.4125
0.8197
1.556
5.914
5.875
~
—
0.0963
—
0.1771
0.1667
0.3438
0.7652
2.564
8.00
8.50
0.0111
0.007810
0.0562
0.01434
0.1569
0.2108
0.4114
0.7846
1.9373
6.205
7.887
-1.95
-2.107
-1.250
-1.843
-0.804
-0.676
-0.386
-0.105
0.287
0.793
0.897
- 212 -
-------
so3
Table 7. Ave. Ration of H20 as a function of I (MgSO^SI^O)
Temp(°C)
35
50
100
150
200
250
300
350
400
Run #370
0.01133.
0.200
0.234
0.394
0.500
0.823
2.47
4.40
7.445
Run #373
__
0.026
0.0286
0.0606
0.0625
0.080
0.143
1.385
2.69
Ave
0.01133
0.113
0.1313
0.2273
0.2813
0.4515
1.3065
2.893
5.068
Log Ave
-1.946
-0.947
-0.8817
-0.6434
-0.5509
-0.3453
1.161
0.4612
0.7048
- 213 -
-------
SO 9
Figure 2. Log fc-r) as a function of Temperature
««"
2 ave.
for MgSO -3H20 and
- 214 -
-------
1.2
1.0
0.8
0.6
0.4
0.2
0
-0.2
OJ§, -0.4 •
O -0.6
O
-J -0.8
-1.0
-1.2
-1.4
-1.6
-1.8
-2.0
-2.2
V TfC)
EQUAL CONCENTRATIONS OF
HjO AND S02
QMgS03-6H20
®MgS03 3H20
50 100
ISO 200 250 300 350 400
- 215 -
-------
APPENDIX A. Simulation of Calciner in the Laboratory; The Testing
of the Hypothesis that tUS is Formed When the Burner
Gas is Operated in the Fuel-Rich Region.
INTRODUCTION:
Since the previous study showed that H-S is not a product of the thermal
decomposition of the magnesium sulfite hydrates, and since H_S had been reported
as a product during calcination, it was conjectured that H_S forms as a result
of reactions between SO. and hydrocarbons present when the flame used to effect
thermal degradation is operated in the fuel-rich region.
A laboratory set-up was designed which simulates many features of a
calciner and, also, permits the immediate continuous analysis of volatilized
products.
EXPERIMENTAL:
An inverted premix burner and a movable platform on which the sample
is situated are housed within a pyrex glass pipe. See Figure A. In the center
of the platform Just above the sample is a quartz probe which leads to the Fin-
nigan 1015 Quadrupole mass spectrometer. Thus, volatiles are continuously and
rapidly characterized. In addition, the temperature in the region is sensed by
a thermocouple. Also, the pressure in the "cross" can be varied and controlled.
In practice, the pressure was maintained at 100 mm and the burner was operated
in the fuel-rich region.
RESULTS:
This study was not carried out in a comprehensive manner. In the survey-
type studies carried out, H.S was not- observed.
- 216 -
-------
LAB-SCALE CALCINER SIMULATOR
- 217 -
APPENDIX A
-------
LAB-SCALE CALCINER SIMULATOR
EXPLANATION OF NUMERALS
1. 02 feed
2. Fuel feed
3. Premlx burner
4. Lead for Tejla Coil ignition
5. Cooling water
6. NaCl window
7. Quartz probe
8. Thermocouple
9. Perforated stainless steel cover
10. Asbestos sheet
11. Stainless steel sample holder
12. Rubber stopper
13. To mass spectrometer
14. Position of thermal decomposition stage is adjustable
15. To vacuum
- 218 -
-------
APPENDIX B. Effect of Wettness of Hydrates on Aggregate Formation
During Calcining
INTRODUCTION:
The prevalent feeling had been that less aggregation, finer particles,
resulted from the calcining of "dry" hydrates. On the other hand, it was thought
in this laboratory that the volatilization of water during calcination should
break up the particles.
EXPERIMENTAL:
Samples of MgSO *3H_0 were heated in air and vacuum to 500 C. Also,
an extra-moist sample was heated in vacuum.
RESULTS:
Samples heated in air showed less aggregation than samples heated in
vacuum. Extra-moistened samples heated in vacuum show less aggregation than the
drier samples.
CONCLUSION:
Water decreases aggregation by explosive volatizatlon, thus rapid re-
moval by applying vacuum increases aggregation. Ideally, a wet sample should be
rapidly heated to as high a temperature as possible.
- 219 -
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-600/7-77-018
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Magnesia Scrubbing Applied to a Coal-Fired
Power Plant
S. REPORT DATE
March 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
George Koehler
I. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Chemico Air Pollution Control Company
One Penn Plaza
New York, NY 10001
10. PROGRAM ELEMENT NO.
EHB528
11. CONTRACT/GRANT NO.
68-02-1870
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 8/73-8/75
14. SPONSORING AGENCY CODE
EPA-ORD
15. SUPPLEMENTARY NOTES jjjRL-RTP project officer for this report is C. J. Chatlynne, Mail
Drop 61, 919/549-8411 Ext 2915.
i6. ABSTRACT rphe rep0rj. gjves results of a full-size demonstration of the magnesia wet-
scrubbing system for flue gas desulfurization (FGD) on a coal-fired utility boiler. The
system was designed to desulfurize half the flue gas from a 190-MW rated capacity
generating unit firing 3. 5% sulfur coal. The FGD installation was equipped with a
first-stage wet scrubber for particle emissions control, followed by the magnesia
unit. The FGD system was able to remove 90% of the inlet SO2 over 2800 hours of
operation logged at the generating station. Its particle control capability was also
demonstrated by reducing particle emissions to less than 0.01 gr/scf with the unit
operated in series with an electrostatic precipitator. A test program, using only the
wet-scrubbing unit for particle emissions control, achieved a collection efficiency
of 99. 6%. Magnesia was regenerated and recycled successfully. The SO2 produced
during regeneration was used to manufacture commercial grade sulfuric acid which
was marketed. Correlations were developed to determine SO2 removal for varying
boiler loads and fuel sulfur content, and to control regeneration of acceptable
alkali. Several other studies of the process technology and process chemistry were
undertaken as part of the work.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
c. cos AT i Field/Croup
Air Pollution
Flue Gases
Scrubbers
Magnesium Oxides
Desulfurization
Sulfur Dioxide
Electric Power
Plants
Coal
Combustion
Particles
Sulfuric Acid
Air Pollution Control
Stationary Sources
Magnesia Scrubbing
Particulates
13B
21B
07A
07B
07D
10B
21D
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
228
20 SECURITY CLASS (Thupage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
- 220 -
------- |