EPA-600/7-78-045b
Office of Research and Development Laboratory _ _ . ^ nfo
Research Triangle Park. North Carolina 27711 MaFCn iy/<
CONTROLLING SO2 EMISSIONS
FROM COAL-FIRED
STEAM-ELECTRIC GENERATORS:
WATER POLLUTION IMPACT
(Volume II. Technical Discussion)
Interagency
Energy-Environment
Research and Development
Program Report
z
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
EPA REVIEW NOTICE
This report has been reviewed by the participating Federal Agencies, and approved
for publication. Approval does not signify that the contents necessarily reflect
the views and policies of the Government, nor does mention of trade names or
commercial products constitute endorsement or recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-78-045b
March 1978
CONTROLLING SO2 EMISSIONS FROM
COAL-FIRED STEAM-ELECTRIC
GENERATORS: WATER POLLUTION IMPACT
(Volume II. Technical Discussion)
by
R.L Sugarek and T.G. Sipes
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78766
Contract No. 68-02-2608
W. A. 10
Program Element No. EHE624A
EPA Project Officer: Julian W. Jones
Industrial Environmental Research Laboratory
Office of Energy, Minerals and Industry
Research Triangle Park, N.C. 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, D.C. 20460
-------
TABLE OF CONTENTS
Figures
Tables
List of Abbreviations . . .
Metric Conversion Factors
1.0 INTRODUCTION 1
1.1 Objective and Background 1
1.2 Presentation of Results 2
2.0 RESULTS AND CONCLUSIONS 5
2.1 Results of Water Consumption Calculations . 5
2.1.1 Uncontrolled Power Plant Water
Consumption 5
2.1.2 SOX Control System Water Consumption 7
2.1.3 Conclusions 9
2.2 Results of Wastewater Characterizations ... 9
2.2.1 Uncontrolled Power Plant Wastewaters 9
2.2.2 SO,. Control Process Wastewaters .... 13
X
2.3 Effect of SOX Control System Wastewaters
on Receiving Streams 17
2.4 Examination of Water Treatment Technology
Applicable to SOX Control System Waste-
water Streams 18
2.5 Comparison of Model Plant Systems 19
3.0 PROCESS DESCRIPTIONS 22
3.1 Coal-Fired Power Plant Water System:
Uncontrolled for SOV Emissions . 22
j\
3.1.1 Process Description . . 23
3 . 2 Lime Wet Scrubbing Process 30
3.2.1 Process Description 30
3.3 Limestone Wet Scrubbing Process 35
3.3.1 Process Description 36
-------
TABLE OF CONTENTS (Continued)
Page
3.4 Wellman-Lord Sulfite Scrubbing Process ..... 40
3.4.1 Process Description ................. ^
3.5 Magnesia Slurry Absorption Process ......... ^'
3.5.1 Process Description ................. ^°
3.6 Double Alkali Wet Scrubbing ................ 5Zl-
3.6.1 Process Description ................. 55
3.7 Physical Coal Cleaning ..... . ............... 61
3.7.1 Process Description ... .............. 63
3.8 SOa Conversion Processes ................... 69
3.8.1 Sulfuric Acid Production ............ 69
3.8.2 Sulfur Production ................... 72
4.0 WATER CONSUMPTION ............................... 77
4.1 Coal-Fired Power Plant Water Consumption;
Uncontrolled for SO Emissions ............. 77
X
4.1.1 Cooling Water System ................ 81
4.1.2 Ash Handling System ................. 85
4.1.3 General Services Water System ....... 91
4.1.4 Boiler Makeup Water Requirement ..... 93
4.2 Lime Wet Scrubbing Water Consumption ....... 94
4.2.1 Evaporation in the Absorber ......... 94
4.2.2 Occlusion in the Solid Waste ........ 95
4.3 Limestone Wet Scrubbing Water Consumption .. 96
4.3.1 Evaporation in the Absorber ......... 96
4.3.2 Occlusion in the Solid Waste ........ 97
4.4 Wellman-Lord Sulfite Scrubbing Process Water
Consumption ................................ 97
4.4.1 Evaporation in the Scrubber ......... 98
4.4.2 Particulate Sluicing Requirement .... 98
4.4.3 Water Loss Association with Purge
Solids Drying ....................... 99
4.4.4 Water in S02 Product Stream ......... 99
111
-------
TABLE OF CONTENTS (Continued)
Page
4.4.5 Condenser Cooling Water Slowdown ... 100
4.5 Magnesia Slurry Absorption Process Water
Consumption 100
4.5.1 Evaporation in the Scrubber 101
4.5.2 Particulate Sluicing Requirement ... 101
4.5.3 Water Losses Associated With Drying. 102
4.6 Double Alkali Wet Scrubbing Water Con-
sumption 102
4.6.1 Evaporation in the Scrubber 103
4.6.2 Particulate Sluicing Requirement ... 103
4.6.3 Occlusion in the Solid Waste 104
4.7 Physical Coal Cleaning Water Consumption .. 104
4.8 SOz Conversion Processes Water Consumption. 106
4.8.1 Sulfuric Acid Production 106
4.8.2 Elemental Sulfur Production 107
4.9 Model Systems Makeup Water Requirement .... 107
4.9.1 Base Uncontrolled Power Plant Water
Requirements for the Model Plant
Systems 108
4.9.2 SOX Control Strategy Water Require-
ments for the Model Plant Systems .. 112
4.9.3 Matrix Presentation of Model Plant
Water Requirements 112
5.0 CHARACTERIZATION OF PROCESS WASTEWATERS 118
5.1 Characterization of Wastewaters From a
Power Plant Not Equipped With an FGD System 119
5.1.1 Power Plant Wastewater Sources 119
5.2 Characterization of Wastewaters from the
Lime -Wet Scrubbing Process 158
5.2.1 Base Case Water Balance 158
5.2.2 Description of the Water System .... 160
5.2.3 Purge Characteristics . 163
IV
-------
TABLE OF CONTENTS (Continued)
5.3 Characterization of Effluents from the
Limestone Wet Scrubbing Process
5.3.1 Base Case Water Balance .
5.3.2 Description of the Water System 164
5.3.3 Purge Characteristics 168
5.4 Characterization of Wastewaters from the
Wellman-Lord Sulfite Scrubbing Process 168
5.4.1 Base Case Water Balance 169
5.4.2 Description of Water System 172
5.4.3 Slowdown Characteristics 173
5.5 Characterization of Wastewaters from the
Magnesia Slurry Absorption Process 173
5.5.1 Base Case Water Balance 173
5.5.2 Description of the Water System 177
5.5.3 Slowdown Characteristics 178
5.5.4 Purge Characteristics 178
5.6 Characterization of Wastewaters from the
Double Alkali Wet Scrubbing Process 178
5.6.1 Base Case Water Balance 179
5.6.2 Description of the Water Systems .... 183
5.6.3 Slowdown Characteristics 184
5.6.4 Purge Characteristics 184
5.7 Characterization of Wastewaters from the
Physical Coal Cleaning Process 185
5.8 Characterization of Wastewaters from S02
Conversion Processes 186
6.0 EXAMINATION OF PURGE CHARACTERISTICS AND
APPLICABLE TREATMENT TECHNOLOGY 187
6.1 Magnesia Slurry Absorption Process Purge ... 187
6.1.1 Purge Characteristics 187
6.1.2 Effect on Receiving Streams 188
6.1.3 Treatment Technology 188
v
-------
TABLE OF CONTENTS (Continued)
Page
6.2 Double Alkali Sulfate Purge ............... 190
6.2.1 Purge Characteristics .............. 191
6.2.2 Effect on Receiving Stream ......... 192
6.2.3 Treatment Technology 192
6 .3 Prescrubber Slowdown 195
6.3.1 Slowdown Characteristics 196
6.3.2 Effect on Receiving Stream 197
6.3.3 Treatment Technology 198
6.4 Cooling Water System Slowdown . 198
6.4.1 Purge Characteristics 199
6.4.2 Effect on Receiving Streams 200
6.4.3 Treatment Technology 200
6.5 Possible Lime/Limestone Purge 200
6.5.1 Purge Characteristics 201
6.5.2 Effect on Receiving Streams 202
6.5.3 Treatment Technology ............... 202
6.6 Lime/Limestone/Double Alkali Solid Waste .. 204
6.7 Existing Water Treatment Technologies
Applicable to Wastewater from FGD Systems . 204
6..7.1 Lime-Soda Softening 204
6.7.2 Reverse Osmosis 205
6.7.3 Ion Exchange 207
6.7.4 Vapor Compression Distillation 208
6.7.5 Multistage Flash Evaporation 209
REFERENCES 211
APPENDIX A A-l
VI
-------
FIGURES
Number
3.1-1 Power plant water system: recirculatory water
management 25
3.1-2 Power plant water system: once-through water
management • 26
3.2-1 Process flow diagram for the lime wet scrubbing
process 31
3.3-1 Process flow diagram limestone wet scrubbing
process 37
3.4-1 Process flow diagram for the Wellman-Lord sulfite
scrubbing process 42
3.5-1 Process flow diagram for the magnesia slurry
absorption process 49
3.6-1 Process flow diagram for the double-alkali
scrubbing process 56
3. 7-1 Generalized coal cleaning process 64
3.8-1 Typical flow diagram of a single absorption
contact sulfuric acid plant 71
3.8-2 Typical process flow diagram of the allied
chemical SO2 reduction process 74
4.1-1 Power plant water system: system #1 once-through
water management 79
4.1-2 Power plant water system: system #2, partial
recirculatory water management 80
4.1-3 Power plant system: system #3, recirculatory
water management 82
4.1-4 Power plant water system: system #4, zero
discharge water management 83
5.1-1 Sources of wastewater in a fossil-fueled steam-
electric plant 120
5.2-1 Process flow diagram lime wet scrubbing process.. 159
VII
-------
FIGURES (Continued)
Number Page
5.3-1 Process flow diagram limestone wet scrubbing
process . 165
5.4-1 Process flow diagram for the Wellraan-Lord sulfite
scrubbing process . 170
5.5-1 Process flow diagram for the magnesia slurry
absorption process 175
5.6-1 Process flow diagram for double-alkali wet
scrubbing 180
Vlll
-------
TABLES
Number
2.1-1 BASE CASE 500 MW SYSTEM: MODEL PLANT WATER
CONSUMPTION ................ . .......... • • • • ..... • 6
2.1-2 BASE CASE 500 MW SYSTEM: FGD SYSTEM WATER
CONSUMPTION ................. ......... ....... • • • • 8
2.2-1 CHARACTERIZATION OF POWER PLANT WASTEWATER
STREAMS ...................... . . ......... . ....... 10
2.2-2 CHARACTERIZATION OF SOX CONTROL SYSTEM PROCESS
WASTEWATER STREAMS .......... ...... .............. 14
2.5-1 SOV CONTROL SYSTEM MAKEUP WATER REQUIREMENTS .... 19
X
4.1-1 CHARACTERISTIC COOLING SYSTEM OPERATION ..... .... 86
4.1-2 SLUICE WATER REQUIREMENT ........................ 87
4.1-3 SYSTEM #2: ASH SLUICE MAKEUP REQUIREMENT ....... 89
4.1-4 SYSTEM #3; ASH SLUICE MAKEUP REQUIREMENT ....... 90
4.1-5 SYSTEM #4: ASH SLUICE MAKEUP REQUIREMENT ....... 92
4.1-6 GENERAL SERVICES MAKEUP WATER REQUIREMENT ....... 93
4.9-1 EPA/OAQPS ALTERNATIVE CONTROL SYSTEMS FOR MODEL
PLANTS .......................................... 109
4.9-2 BASE CASE: MODEL POWER PLANT WATER CONSUMPTION.. Ill
4.9-3 BASE CASE: FGD SYSTEM WATER CONSUMPTION ........ 113
4.9-4 MODEL PLANT SYSTEM WATER REQUIREMENTS ........... 114
5.1-1 CHEMICAL TREATMENT SUMMARY FOR RECIRCULATING
COOLING SYSTEMS ................................. 128
5.1-2 CHARACTERISTICS OF ONCE -THROUGH ASH POND
DISCHARGES ................ . ..................... 13 2
5.1-3 COAGULATING AND FLOCCULATING AGENT
CHARACTERISTICS ....... ..... ..................... 135
5.1-4 ION EXCHANGE MATERIAL TYPES AND REGENERANT
REQUIREMENT .............. . ...................... 139
IX
-------
TABLES (Continued)
Number Page
5.1-5 RECOMMENDED LIMITS OF TOTAL SOLIDS AND SUS-
PENDED SOLIDS IN BOILER WATER FOR DRUM BOILERS.. 142
5.1-6 CHEMICAL ADDITIVES COMMONLY ASSOCIATED WITH
INTERNAL BOILER TREATMENT 143
5.1-7 PLANT DATA RELATING TO WATER QUALITY PARAMETERS
FOR COAL PILE RUNOFF 146
5.1-8 OPERATIONAL CLEANING OF A HIGH PRESSURE, ONCE-
THROUGH BOILER 149
5.1-9 OPERATIONAL CLEANING OF A LOW PRESSURE, DRUM
BOILER 150
5.1-10 OPERATIONAL CLEANING, MAIN CONDENSER WATER-SIDE. 151
5.1-11 DATA FOR BOILER FIRE SIDE WASHING OPERATIONS;
INCREASE IN POLLUTANT QUANTITY PER WASHING CYCLE 153
5.1-12 DATA FOR AIR PREHEATER WASHING OPERATIONS;
INCREASE IN POLLUTANT QUANTITY PER WASHING
CYCLES 155
5.2-1 WATER BALANCE: LIME WET SCRUBBING PROCESS . 161
5.3-1 WATER BALANCE: LIMESTONE WET SCRUBBING PROCESS. 166
5.4-1 WATER BALANCE: WELLMAN-LORD SULFITE SCRUBBING
PROCESS 171
5.5-1 WATER BALANCE: MAGNESIA SLURRY ABSORPTION
PROCESS 176
5.6-1 WATER BALANCE: DOUBLE ALKALI WET SCRUBBING,
LIME REGENERANT 181
5.6-2 WATER BALANCE: DOUBLE ALKALI WET SCRUBBING,
LIMESTONE REGENERANT 182
6.3-1 USGS STREAM CLASSIFICATIONS 197
6.5-1 RANGE OF CONCENTRATION OF CONSTITUENTS IN
SCRUBBER LIQUORS STUDIED 203
x
-------
LIST OF ABBREVIATIONS
acf
BOD
Btu
°C
cf
cm3
COD
EPA
°F
FGD
g
gpm
gal
J
Kcal
kg
1
L/G
Ib
m3
mg
MJ
MM Btu
MPa
MW
Nm3
NSPS
OAQPS
Pa
ppm
psi
actual cubic foot
Biochemical Oxygen Demand
British thermal unit
degrees Centigrade
cubic foot
cubic centimeter
Chemical Oxygen Demand
Environmental Protection Agency
degrees Fahrenheit
Flue Gas Desulfurization
gram
gallons per minute
gallon
Joule
kilocalories
kilogram
liter
liquid to gas ratio
pound
cubic meters
milligrams
megajoule
million Btu
megapascals
megawatt
normal cubic meter
New Source Performance Standard
Office of Air Quality Planning
and Standards
Pascal
parts per million
pounds per square inch
-------
s second
scf standard cubic foot
ss suspended solids
TDS Total Dissolved Solids
Ug microgram
xii
-------
To Convert From
Btu/lb
gpm
gal
gal/1000 scf
Ib/min
lb/105 Btu
psi
METRIC CONVERSION FACTORS
To
J/kg
m3/s
m3
m3 /Nm3
kg/s
Ug/J
MPa
Multiply By
2.324 x 103
6.309 x 10~5
3.785 x 10"3
.000135
7.567 x 10"3
0.4303
6.9 x 10"3
XLll
-------
1.0 INTRODUCTION
!•1 Objective and Background
The objective of this study is to define and assess
the effects of alternative SOV control systems on the water
?x "
consumption and wastewaters from coal-fired power plants. The
approach used was to calculate the water requirements for model
power plant water systems and model SOX control systems provided
by EPA. Effluent streams from uncontrolled coal-fired power
plants and each of the processes involved in the S0x control
strategies were characterized. A comparison of the calculated
water requirements and characteristic wastewaters of the power
plant and the SOX control systems was made to assess the impact
on receiving waters.
The processes comprising the alternative S0x control
systems examined in this study include:
Lime Wet Scrubbing
Limestone Wet Scrubbing
Wellman-Lord Sulfite Scrubbing
Magnesium Slurry Absorption
Double Alkali Wet Scrubbing
Physical Coal Cleaning
SOz Conversion
This study is one task in a -comprehensive program by
the Office of Air Quality Planning and Standards to review the
New Source Performance Standards for SOa emissions from coal-fired
steam generating plants. The comprehensive impacts of two alter-
native revised standards, and the existing NSPS are being examined,
The existing NSPS allows an emission rate of 0.52 ug S02/J
-------
(1.2 Ib S02/MM Btu) of heat input. One alternative standard
requires 0.22 yg S02/J (0.5 Ib S02/MM Btu) of heat input.
This standard has the same form as the existing NSPS and thus
allows a credit for physical coal cleaning or use of low sulfur
coal. The second alternative standard requires 90% removal of
S02 from stack gases, regardless of original sulfur content
in the coal. 108 model plant system permutations were defined
for evaluation. These model systems are discussed in detail in
Section 4. Generally, these model systems allow an analysis of
the impacts of the three alternative NSPS for S02, accommodating
the variables of; type of flue gas desulfurization (FGD) system
(5 types), sulfur content of the coal, size of the steam genera-
tor (25,100,500 and 1000 Mw), and degree of coal cleaning.
An examination of the model plant water consumption
calculations indicates that the application of the SC> control
X
systems to coal-fired power plants will generally require 10-
15 percent additional water consumption.
An examination of the wastewater characterizations
indicates that there are no significant wastewaters associated
with the SOX control systems. There are significant wastewater
streams from uncontrolled power plants. The insignificant
wastewater streams from S0x control processes can be treated
with developed technology to allow reuse or discharge. The
effect of these streams on receiving stream water quality is
expected to be negligible.
1.2 Presentation of Results
The presentation of the results of this assessment
is organized in the following manner:
-------
* Process Descriptions
Process descriptions for a power plant uncontrolled
for SOX emissions and for each process involved in the various
control strategies are presented in Section 3. Process descrip-
tions are given to allow a basic understanding of the process
function, components, operating variables, and process flows.
A simplified flow diagram for each process is included.
* Assessment of the Impact of S0y Control Strategies
on Coal-Fired Power Plant Water Consumption
The make-up water requirements for each of the systems
and operations in the uncontrolled power plant and each of the
processes involved in the various S0x control strategies are
discussed in Section 4, For purposes of discussion a base
case calculation for a 500 MW power plant, burning 3.5% sulfur
coal with an average heating value of 28 MJ/kg (12,000
Btu/lb) is used. The results of calculations for the remaining
model plant systems are presented in tables at the end of
Section 4. A detailed presentation of the calculation methods
and assumptions is presented in Appendix B.
* Characterization of Process Wastewaters
Section 5.0 characterizes the composition and quantity
of wastewaters from an uncontrolled power plant, and each of the
processes involved in the various S0x control strategies.
Wastewater quality varies widely due to differences in influent
water quality and the use of different water management techniques
It is therefore necessary to characterize the wastewater streams
in a general manner. Characteristics of example compositions
-3-
-------
and flow rates of effluent streams are given to illustrate
ranges of operation wherever possible. An assessment of rela iv
impacts is made,
. Examination of SO Control System Wastewater Char-
acteristics and Applicable Treatment Technology
A detailed examination of each of the wastewater
streams for SOX control technologies and applicable water treat-
ment technology is presented in Section 6. This examination
includes a characterization of effluent composition, an assess-
ment of effect on receiving stream water quality, and possible
water treatment technologies.
-4-
-------
2.0 RESULTS AND CONCLUSIONS
This section presents the major findings and conclu-
sions of the study. It summarizes the results of water con-
sumption calculations, the characterization of process waste-
waters, the assessment of effluent effects on receiving stream
water quality, possible water treatment technology, and a com-
parison of the model plant systems. A base case 500 MW power
plant burning 3.5% sulfur coal with an average heating value of
28 MJ/kg (12,000 Btu/lb) is used. This base case is more
completely defined in Appendix A.
2.1 Results of Water Consumption Calculations
Water consumption calculations were performed to
determine the additional requirement for power plants controlled
by alternative SO control strategies. Calculations were per-
X
formed for four uncontrolled power plant water systems. This
was done to characterize the wide range of current practice.
Calculations were also performed for each of the five FGD
systems under study, the physical coal cleaning process, and
the S02 conversion processes. In each calculation, specific
requirements were determined, and their sum was taken as the
process requirement. A detailed discussion of the water con-
sumption calculations is given in Section 4.
2.1.1 Uncontrolled Power Plant Water Consumption
The results of the calculations for uncontrolled
model power plant water systems are summarized in Table 2.1-1.
Once-through cooling requires enormous quantities of water.
The use of recirculatory cooling systems reduces this require-
ment significantly, even at low cycles of concentration. Ash
handling systems also require large quantities of water, but
-5-
-------
TABLE 2.1-1. BASE CASE 500 MW SYSTEM: MODEL POWER PLANT WATER CONSUMPTION
System
Number
1
2
3
4
System
Description
Once-through
Partial Recirculatory
Recirculatory
Zero Discharge
Cooling Water
System
m'/s (g
13 (210
0.44 (7
0.32 (5
0.25 (4
Ash Handling General Services Boiler
System Wiiter Makeuji
pm)
,000)
,000)
,000)
,000)
in3/s (8P">)
0.23 (3600)
0.07 (1100)
0.02 (300)
0 (0)
mVs
0.05
0.05
0.05
0,01
(gl>"0 ro'/s
(750) 0.0006 W
(750) 0.0006 (»)
(750) 0.0006 (9)
(190) 0.0006 (9)
Total l-'resti Water
Makeup Requiremenl
11.5 (214,
0.57 (9,
0.18 (60,
0.2r> (4,
000)
000)
000)
000)
These power plant water systems are discussed In detaJL In Section 4.1
The base case is a 5QO Mw power plant operating at an efficiency of 37%; 3,5% S coal; average heating value of 2B HJ/ltg (12,000 Btu/lh) .
All power plant water systems are once-through; refer to Figure 4.1-1.
Recirculatory cooling at 2.5 cycles of concentration, once-through ash handling, and once-through general services water; refer to Figure 4.1
Reci.rculatory cooling at 5.0 cycles of concentration, 50% recirculatory ash handling, and recycle of general service water blowdown to the
ash handling system; refer to Figure A.1-3. *
All power plant water systems are recirculatory; refer to Figure 4.1-4.
-------
this requirement can be reduced through the use of cooling system
blowdown as makeup, and/or recirculatory practices. The general
service water requirement was estimated from plant data as
being minor. A reduction credit is possible in these cal-
culations through collection and reuse of these waters in
the cooling water or ash handling systems. The boiler makeup
requirement is not significant.
2.1.2 SOX Control System Water Consumption
The results of water consumption calculations for the
five FGD systems examined in this study are summarized in Table
2.1-2. The evaporative loss caused by the adiabatic saturation
of the hot flue gases in the prescrubber or the absorber re-
quires the significant fraction of the makeup to the FGD water
systems in all cases. The Wellman-Lord Sulfite Scrubbing pro-
cess also has a substantial requirement for makeup to the
condenser cooling water system. Other large FGD water require-
ments are for a prescrubbing system blowdown, and occlusion of
water in the solid wastes. These two requirements apply only
to specific FGD systems, as indicated in the table.
Also included in Table 2.1-2 are the results of
calculations of the water requirements for two S02 conversion
processes. The Wellman-Lord Sulfite Scrubbing Process and the
Magnesia Slurry Absorption Processes produce concentrated S02
product streams. These streams can be converted to either sul-
furic acid or elemental sulfur. The calculations indicate a
significant requirement for makeup to the product acid cooling
water system, and an insignificant requirement for elemental
sulfur production.
-7-
-------
TABLE 2.1-2. BASE CASE 500 MW SYSTEM: FGD SYSTEM WATER CONSUMPTION11
i
CO
I
Water8 Lime Wet
Requirement Scrubbing
m3/s
Evaporative loss, 0.030
Loss in solid waste, 0.005
Prescrubber blowdown,
Cooling water blowdown.
Loss with solids drying,
Loss in product SO2 stream,
S02 Conversion Requirement:
Sulfuric acid
Elemental Sulfur
Total 0.035'
Liquid to gas ratio required .005-. 015
Eor scrubber - m'/Nm' (gal/1000 scf)
(gpm)
<480)
(80)
(560)
(35-
110)
Limestone Wet
Scrubbing
m'/s (gpm)
0.030 (480)
0.005 (80)
0.035 (560>
.005-. 015 (35-
110)
Wellman-Lord
Sulfite Scrubbing
nVs
0.030
0.003
0.014
0.002
0.003
0.008
0.002
0.058
0.050
0.004
.002
-------
Not indicated in the table is the water requirement
for the physical coal cleaning process. Make-up is required to
replace losses due to occlusion in the solid waste, and drying
losses. The makeup requirement was estimated to be 0.015 m3/s
(240 gptn) for a 500 MW power plant.
2.1.3 Conclusions
A review of Table 2.1-2 indicates that S0x control
through the use of an FGD system has water requirements ranging
from 0.035 m3/s (550 gpm) to 0.058 m3/s (920 gpm) , dependent
upon the FGD system used. In comparison to the water require-
ment for once-through cooling shown in Table 2.1-1, this is
insignificant. In comparison to the requirements of the recir-
culatory systems, FGD systems increase the water requirement
6 to 22 percent with the majority of the cases falling between
10 and 15 percent additional requirement.
2.2 Results of Wastewater Characterizations
Wastewaters were characterized for uncontrolled power
plants, and for each of the processes involved in the alterna-
tive SOX control strategies.
2.2.1 Uncontrolled Power Plant Wastewaters
Wastewater quality is highly dependent on the power
plant water management system, the influent water quality,
plant layout, and treatment practices. Because of these rea-
sons, it was necessary to characterize the Wastewaters from
each source in a general manner. Typical flow rates and compo-
sitions are given in Table 2.2-1 to the extent possible. A
review of Table 2.2-1 indicates that power plants are a signifi-
cant source of wastewaters. The sources of largest concern are
-9-
-------
TABLE 2.2-1. CHARACTERIZATION OF POWER PLANT WASTEWATER STREAMS
Power Plant System
Effluent Stream
Characteristics
Cooling Water
once-through
discharge
recirculatory
system
blowdown
Ash Handling
bottom ash
sluice water
fly ash
sluice water
The effluent water quality is
essentially equivalent to that
of the influent water. There
may be slight differences due
to the presence of corrosion
products or traces of chemicals
used in treatment practices.
Soluble species enter the system
in the makeup water, and are con-
centrated to levels ranging typi-
cally from 1500 to 10,000 mg/£
(ppm). Intimate contact of air
and water in a cooling device al-
lows the entry of particulate
matter and soluble gases into the
system. Traces of chemicals used
for treatments to prevent scale
and corrosion may also be present.
These chemicals may include in-
organic polyphosphates, chelating
agents, polyelectrolyte anti-
precipitants, and organic/polymer
dispersants for scale control,
chromate, zine, phosphate, sili-
cates, or certain proprietary
organics for corrosion inhibition.
Bottom ash forms as a fused ma-
terial and therefore has little
impact on sluice water quality.
It has excellent settling charac-
teristics, and is chemically inert
with water.
Fly ash has poorer settling
characteristics than bottom ash,
however, low turbidities are ob-
served with adequate residence
times for sedimentation. Fly ash
contains a broad spectrum of sol-
uble inorganic salts which can be
leached into the sluice water
resulting in sodium, potassium,
calcium, magnesium, chloride,
sulfates, etc., in solution.
(Continued)
-10-
-------
TABLE 2.2-1 (Continued)
Power Plant System
Effluent Stream
Characteristics
Water Conditioning
sedimentation
clarifier
underflow
filtration
backwash water
Lime/lime-soda
softening
clarifier
underflow
ion exchange
regeneration
waste streams
evaporation
blowdown
This stream has high suspended
solid concentrations, and traces
of coagulants and flocculants
such as alum, aluminate, ferric
chloride, or copperas.
This stream has a high suspended
solids content.
This stream contains traces of
coagulants and flocculants such
as alum, aluminate, ferric
chloride, or copperas. Its
hardness is typically 50 mg/H
as CaCOs, and its pH is approxi-
mately 10.
The backwashing stream from this
process has a high suspended
solids content. The spent re-
generate has extreme pH, and
contains high concentrations of
ions eluted from the exchange
material. The eluted ions rep-
resent the chemical species that
were removed from water during
the service cycle of the process.
Impurities in the feedwater are
concentrated. Thus, the water
quality of the blowdown is de-
pendent on this influent water
quality, and the degree of con-
centration in the evaporator.
The total dissolved solids level
is typically 1000 to 2000 mg/i
(ppm) and the pH is 9 to 11.
-11-
-------
TABLE 2.2-1 (CONTINUED)
Power Plant System
Effluent Stream
Characteristics
Steam Generation
boiler blowdown
General Services
equipment clean-
ing and washing
waste streams
coal pile
runoff
Boiler feedwater impurities are
concentrated, and a blowdown is
taken to maintain a desired
level of suspended and dissolved
solids. This stream contains
a high level of dissolved solids,
traces of corrosion products, and
traces of chemicals used for
treatment to prevent scale. These
chemicals may include inorganic
phosphates or chelating agents
such as EDTA or NTA. The pH
ranges from 8.0 to 9.5.
These streams may have extreme
pH, high suspended and dissolved
solids contents, and high
oxygen demands (BOD and/or COD) .
These streams may contain
detergent constitutents.
This stream may have a pH as
low as 2 to 3, and has high
suspended and dissolved solid
content.
-12-
-------
the recirculatory cooling system blowdown, ash sluicing water,
and wastes from water conditioning operations. These streams
have high flow rates, and contain high concentrations of various
suspended and dissolved solids. Also, these streams contain
trace amounts of coagulants, flocculants and metals. Some of
these streams have extreme pH.
2.2.2 SOX Control Process Wastewaters
A review of Table 2.2-2 indicates that there are no
significant wastewater streams from any of the processes involved
in the alternative SOX control systems.
The Wellman-Lord and Magnesia Slurry FGD systems
require a prescrubbing system blowdown. (With high chloride
coals, a prescrubber may also be required for Double Alkali.)
This stream has a high chlorides concentration, but otherwise
has the same composition as fly ash sluice water. It has
approximately 1 percent of the flow rate of the fly ash sluice
water. The high chlorides concentration is usually diluted in
the ash pond, before discharge to a receiving stream.
The Wellman-Lord Sulfite Scrubbing Process and the
sulfuric acid plant require a cooling water system blowdown.
This blowdown is equivalent in composition to the power plant
cooling system blowdown, and is 3 to 5 percent of the flow rate.
The Double Alkali Wet Scrubbing Process and the
Magnesia Slurry Absorption Process may require small purges.
These purges cannot be discharged directly to a receiving stream
because of their very poor quality, but water treatment technol-
ogy is available. Reuse or discharge is possible after treatment.
-13-
-------
TABLE 2.2-2. CHARACTERIZATION OF SOX CONTROL SYSTEM
PROCESS WASTEWATER STREAMS
SOX Control Process
Effluent Stream
Characteristics
Lime/Limestone
Wet Scrubbing
Possible Scrubbing
Liquor Purge
Solid waste
Wellman-Lord
Sulfite Scrubbing
Prescrubbing
System
Slowdown
Condenser Cooling
Water System
Slowdown
There are no wastewater streams
associated with these processes
in normal closed loop pond
operation.
Aerospace Corporation has
reported that in catastrophic
situations or in operating con-
ditions below 50 percent of
design loading, a purge may be
necessary. This purge would
have the same composition as
the scrubbing liquor (BO-203).
This study does not consider the
potential impact of the solid
waste on surrounding water
quality. Aerospace Corporation
is studying this specific problem
in another task in this program
for the OAQPS.
The prescrubber blowdown may
have a chloride concentration
ranging from 10-20,000 mg/2, (ppm)
and a suspended solids content of
approximately 5 percent. Trace
constituents of the liquor
will be similar to those of
fly ash sluice waters. The
blowdown rate is approximately
1 percent of the fly ash- sluice
water requirement.
The quality of this stream is
equivalent to the quality of
the power plant cooling water
system blowdown. The blowdown
rate is approximately 5 percent
of the power plant blowdown rate.
(Continued)
-14-
-------
TABLE 2.2-2. (Continued)
SOX Control Process Effluent Stream
Characteristics
Magnesia Slurry
Absorption
Prescrubbing
System
Slowdown
Intermittent
Purge
Double Alkali
Wet Scrubbing
Possible
Prescrubbing
System
Slowdown
(Not Required
in Normal
Operation)
Sulfate Purge
The prescrubber blowdown may have
a. chloride concentration ranging
from 10-20,000 mg/& (ppm) and a
suspended solids content of approx-
imately 5 percent. Trace constit-
uents of the liquor will be similar
to those of fly ash sluice waters.
The blowdown rate is approximately
1 percent of the fly ash sluice
water requirement.
In the developmental stages of
this process an intermittent purge
was taken to remove impurities
that enter the system with the
makeup water and makeup MgO-
McGlammery et al (MC-076) have
estimated that a purge rate of
approximately 63 cm /s (1 gpm) will
be necessary for a 500 MW power
plant. The purg-e will contain
1.2 percent MgSOs, 15 percent
, and various trace impurities.
The prescrubber blowdown may have
a chloride concentration ranging
from 10-20,000 mg/£ (ppm) and a
suspended solids content of approx-
imately 5 percent. Trace constit-
uents of the liquor will be similar
to those of fly ash sluice waters.
The blowdown rate is approximately
1 percent of the fly ash sluice
water requirement.
A purge may be taken off of the
clarifier supernate to remove
sodium sulfate and nonsulfur/
calcium solubles from the system.
The purge stream has high sodium,
sulfate, sulfite, and nonsulfur/
calcium soluble species concen-
trations. The nonsulfur /calcium
(Continued)
-15-
-------
TABLE 2.2-2. (Continued)
SOX Control Process
Effluent Stream
Characteristics
Solid Waste
Physical Coal
Cleaning
SO2 Conversion
Processes:
Elemental Sulfur
Production
Sulfuric Acid
Production
Product Acid
Cooling Water
System Slowdown
species are impurities which
enter the system in makeup water
and makeup lime or limestone.
This study does not consider the
potential impact of the solid
waste on surrounding water
quality, as discussed under
Lime/Limestone Wet Scrubbing.
Modern physical coal cleaning
plant water systems operate in
closed loops. There are no
wastewater streams from these
plants.
There are no wastewater streams
from this process.
The quality of this stream is
equivalent to the quality of
the power plant cooling water
system blowdown. The blowdown
rate is approximately 3 percent
of the power plant blowdown
rate.
-16-
-------
• The Lime/limestone Wet Scrubbing Process should have
no wastewater streams in normal closed loop operation. All pond
water is recycled to the process. Net water gains by the pond
due to rainfall can be adjusted for by cutbacks in raw water
makeup to the process. However, a catastrophic condition, load-
ing at less than 50 percent of design capacity, or operator
error may require a purge. Water treatment technology is
available to handle this stream to allow reuse or discharge.
2. 3 Effect of SOx Control System Wastewaters on Receiving
Streams
The scrubbing liquor purges for the Magnesia Slurry
Absorption and Double Alkali Wet Scrubbing Processes should not
be discharged directly to a receiving stream. Some treatment of
these streams is necessary to allow reuse or discharge.
The prescrubber blowdown' is a wastewater source of
high chloride concentration. The other stream components are
equivalent to fly ash sluice waters. The normal treatment pro-
cedure is to route this stream to the ash pond. The chlorides
are diluted to 70 mg/& (ppm) in an ash pond overflow of 0.13 m3/s
(2000 gpm) for the base case. The additional impact on the re-
ceiving stream is expected to be minimal, but in site specific
instances treatment of the concentrated stream may be required.
• The Wellman-Lord condenser cooling water system
and the sulfuric acid plant product cooling water system require
blowdowns. The blowdowns are of equivalent quality, and 3 to 5
percent of the quantity of corresponding power plant cooling
water system blowdowns. Therefore, the additional impact of
this wastewater on receiving stream water quality is expected
to be negligible.
-17-
-------
Aerospace Corporation has reported that a purge of
lime/limestone scrubbing slurry may be required in catastrophic
situations, or in situations where these systems are operated
below some critical design loading level (approximately 50
percent) (BO-203). This purge cannot be discharged directly to
a receiving stream.
2 .4 Examination of Water Treatment Technology Applicable
to SOx Control System Wastewater Streams
Magnesia Slurry Absorption Process Purge - Several
purge treatment techniques have been suggested by McGlammery,
et al (MC-076) for the disposal and/or partial recovery of the
sulfate purge. Conventional treatment techniques, such as re-
verse osmosis, vapor compression distillation, flash evaporation,
evaporation in a deadend pond, or softening-ion exchange, could
be used to treat the wastestream from the recovery processes.
Double Alkali Sulfate Purge - Several methods for
sulfate removal have been suggested by Kaplan (KA-227). If a
purge is necessary to maintain a desired level of soluble
nonsulfur/calcium species, the constituents remaining after
sulfate removal can be removed with developed water treatment
technology (i.e., reverse osmosis, vapor compression distillation,
flash evaporation, or softening-ion exchange). If oxidation has
been limited, the purge is discharged with the solids as occluded
water.
• Prescrubber Blowdown - As recirculatory systems
become predominant as a national zero discharge goal approaches,
treatment of this stream will become necessary. Developed water
treatment technologies that can be applied to this stream include
reverse osmosis, vapor compression distillation, flash evaporation,
and softening-ion exchange.
-18-
-------
Cooling System Slowdown - Recirculatory practices
and required treatment applicable in power plant cooling systems
can be applied to this stream as required.
Possible Lime/Limestone Purge - Aerospace Corpora-
tion has reported that conventional lime-soda treatment will
allow reuse of this stream within power plant water systems.
If followed by a reverse osmosis treatment, the stream would
be suitable for discharge and use in public water supply (BO-203)
Other water treatment technologies are applicable but less fav-
orable economically.
2.5 Comparison of Model Plant Systems
The major makeup water requirements for the five al-
ternative SOx control systems are given in Table 2.5-1 for a
500 MW power plant. The makeup water replaces losses due to
evaporation in the absorber or prescrubber, occlusion in the
solid waste, blowdown from the prescrubbing system, and blowdown
of cooling water systems. The dominant makeup requirement
replaces evaporative losses in the prescrubber or absorber.
This loss alone demands 60 to 90 percent of the total makeup
for the various FGD systems. Large amounts of makeup water are
required for the physical coal cleaning process, and for makeup
to cooling water systems in specific SOx control systems.
-19-
-------
TABLE 2.5-1. SOX CONTROL SYSTEM MAKEUP WATER REQUIREMENTS*
Evaporative loss
Cooling water system:
Wellman-Lord Sulfite Scrubbing
Sulfuric Acid Production
Physical Coal Cleaning
Prescrubber Slowdown
Occlusion in solid wastes
ra3
0.
0.
0.
0.
0.
0.
/s
030
014
008
015
003
005
gpm
480
220
130
240
54
75
aThe example requirements given in table are for a base case
500 MW power plant.
An examination of the model plant system calculations
(presented in Section 4.9) allows the following conclusions:
In general, for a given size power plant and percent
sulfur removal, the water requirements for the five FGD systems
increase in the following order:
Double Alkali < Lime < Limestone < Magnesia Slurry < Wellman-Lord.
If a prescrubber is used with Double Alkali, its water require-
ments will be greater than those of the Limestone process.
The water requirement increases in direct propor-
tionality with the power plant size.
In nonregenerable FGD systems, increased sulfur
removal increases the amount of water occluded in the solid
waste by direct proportion. Doubling the amount of .sulfur
-20-
-------
removed, however, increases the total makeup requirement on the
order of 10 to 15 percent.
In control systems using physical coal cleaning,
and reduced S02 removal in the FGD system, water consumption is
increased by the requirement for the coal cleaning process. The
major FGD loss is due to evaporation in the scrubber, and this
loss is constant regardless of the initial coal sulfur content.
Increasing the amount of sulfur removal in the FGD
system does not significantly alter the makeup requirement due
to the dominant evaporative loss.
In control systems using low sulfur coals, a large
increase in water consumption is seen if an FGD system is required,
FGD system removal of 50 and 90 percent of the sulfur content
require essentially the same amounts of water due to the dominant
evaporative losses. Burning low sulfur coal without an FGD
requires no makeup water. FGD applications to 500 MW power
plants burning low sulfur coal require 0.035 m3/s (550 gpm)
makeup.
-21-
-------
X
3.0 PROCESS DESCRIPTIONS
This section presents process descriptions for the
water system of a coal-fired power plant uncontrolled for S0>
emissions, and for each process involved in the alternative
SO control systems. These descriptions provide a basic under-
standing of the process, system components, operating variables,
and process flows. A simplified flow diagram for each process
is included. Descriptions are included for the following
processes:
Coal-fired Power Plant Water System: Uncon-
trolled for SOX Emissions,
Lime Wet Scrubbing ,
Limestone Wet Scrubbing,
Wellman-Lord Sulfite Scrubbing ,
Magnesia Slurry Absorption,
Double Alkali Wet,
Physical Coal Cleaning , and
SOa Conversion.
3.1 Coal-Fired Power Plant Water System: Uncontrolled
for .SQX Emissions
A fossil-fuel fired power generating station uses
large quantities of water. The primary -use is for the conden-
sation of exhaust steam from turbines in the condenser cooling
system. Other power plant water uses include steam generation,
ash handling, water conditioning, cleaning operations, and
miscellaneous operations.
-22-
-------
The major consumers of water in a power generating
station unequipped with an FGD system are the condenser cooling
system and the ash handling system. There are several process
variations of each of these two systems. However, all can be
classified as either once-through systems or recirculatory sys-
tems. The quantities of water required/discharged in the recir-
culatory systems are much less than those for once-through systems
The circulating water flow rate in the recirculatory system is
comparable to the discharge flow rate in the once-through system.
Although the water requirement is less in recirculatory systems,
impurities are concentrated, and may cause scale formation. A
blowdown stream is taken from the system to avoid excessive
buildup of dissolved species.
Other power plant processes which use and consume
water are steam generation, water conditioning, cleaning oper-
ations, and miscellaneous operations. The water consumption
in these processes, however, is insignificant in comparison to
the consumption in the condenser cooling system.
3.1.1 Process Description
Water management in a fossil-fuel power generating
station must consider the following water consumptive pro-
cesses and operations:
1) Condenser Cooling System
2) Ash Handling System
3) Steam Generating System
4) Water Conditioning Operations
5) Cleaning Operations
6) Miscellaneous Operations.
-23-
-------
Figures 3.1-1 and 3.1-2 show simplified flow diagrams
for power plant water systems which use recirculatiory and once-
through processes, respectively, in both the condenser cooling
and ash handling systems.
Condenser Cooling System
Approximately 45 percent of a fossil-fuel-fired gen-
erating station's energy is removed and ultimately discharged
to the environment by the condenser cooling system. Basically,
two condenser cooling systems are employed by the electric utility
industry: 1) once-through system and 2) recirculating system.
In once-through cooling systems, the total cooling
water flow for heat removal is discharged as wastewater effluent.
After passing through the condenser, the cooling water is dis-
charged to a heat sink, e-g.} a river, lake, or pond, where the
heat is dissipated. Thermal pollution of the heat sink is the
major problem associated with once-through cooling systems.
Recirculating cooling systems employ cooling devices
such as cooling towers, spray ponds, canals, etc., which allows
the reuse of recirculated cooling water. These devices promote
cooling primarily by evaporating a portion of the recirculating
water flow. Thus, impurities that come into the system through
makeup water or other sources are concentrated. A blowdown stream
is withdrawn from the system to control the concentration of im-
purities. The quantity of blowdown is set by the maximum concen-
tration of a limiting impurity, e.g., dissolved solids, that can
be tolerated in the system, or by the solubility limit of scaling
salts such as calcium carbonate or sulfate.
-24-
-------
DRIFT
, EVAPORATION
WATER
SOURCE
I
-FOR BOILED
MAKE-UP
ASH
PONO
BOTTOM
ASH
3LUICH
FLY
ASH
SLUICE
1
.
I
'
Figure 3.1-1.
Power plant water system:
water management,
recirculatory
-25-
-------
WATER SOURCE
RETURN TO
WATER SOURCE
RETURN TO
WATEH SOURCE
Figure 3.1-2
Power plant water system: once-through water
management.
-26-
-------
Ash Handling
Ash, a solid by-product of coal combustion, is formed
in a power plant boiler as bottom or fly ash. Bottom ash settles
in the boiler firebox and must be removed from the boiler in
order to maintain system operability. Fly ash is entrained with
the flue gas, leaves the boiler as particulate matter, and is
normally collected in flue gas cleaning equipment. The convey-
ance of both bottom ash and fly ash to points of disposal con-
stitutes ash handling. Ash handling systems employ either pneu-
matic or hydraulic mechanisms for ash transportation. Hydraulic
(wet sluicing) systems produce wastewater streams.
Coal-fired generating stations require formal ash
handling facilities due to the quantity of ash produced during
coal combustion. The ash content of U.S. coals range from 6 to
20 weight percent. The average value is approximately 11 weight
percent (EN-127). The distribution between bottom ash and fly
ash is greatly influenced by boiler furnace design and operating
mode. Ash distribution can affect the water balance for a hy-
draulic ash handling system. The chemical differences between
fly and bottom ashes can also affect sluicing water quality.
Bottom ash generally forms as a fused, clinker-type
material and is removed by wet sluicing to the ash pond. Hy-
draulic design considerations dictate a sluice water of 5 to 10
percent solids. In practice, slurries of less than 1 percent
solids are used depending on such factors as plant design, lo-
cation, and operating circumstances (AY-007).
Fly ash can be collected in the dry form by cyclones,
fabric filters, dry electrostatic precipitators, etc., and in
a water slurry by wet scrubbers, wet electrostatic precipitators,
etc. Fly ash, collected in either the wet or dry form, is
-27-
-------
commonly sluiced to ash ponds for sedimentation of the suspended
fly ash solids. The minimum sluice water requirement is set by
hydraulic design considerations. For fly ash, sluice water com-
positions range from 5 to 10 percent solids. However, as wxth
bottom ash sluicing, sluice water compositions of less than 1
percent solids are used in certain cases (AY-007).
In a once-through ash sluicing system, both bottom and
fly ashes are sluiced to a disposal pond where the ash settles.
The overflow from the disposal pond is then discharged. In a
recirculating system, a portion or all of the pond overflow is
recycled to the system. Scale problems may result if soluble
salts are leached from the fly ash and concentrated in the re-
circulating system. This may be avoided by treating a slip-
stream with lime or soda-ash softening.
Steam Generation
Power plant boilers are either of the once-through or
the drum-type design. Once-through designs are employed exclu-
sively in high pressure, super-critical boilers. No wastewater
t
streams are produced by operation of once-through systems asso-
ciated with their operation. Drum-type boilers operate at sub-
critical conditions. Steam generated in the drum-type units is
in equilibrium with liquid boiler water. Boiler water impurities
are, therefore, concentrated in the liquid phase as steam is
generated in these units. These impurities are ultimately re-
moved in a liquid blowdown stream.
Water Conditioning Operations
Water requirements for conditioning operations include
water required for backwashing filters in lime softening processes
and regenerating demineralizer resins. Some degree of water
-28-
-------
treatment is practiced in all power plants. Water is treated
primarily to remove suspended solids and/or dissolved salts.
Sedimentation and filtration are used for removing suspended
solids. Lime/lime-soda softening, ion exchange, and evaporation
are used to remove dissolved solids. Water conditioning schemes
employ these basic processes singly or in multiple combinations.
Cleaning Operations
Heat transfer surfaces in the boiler and steam con-
denser are chemically cleaned prior to plant start-up. In addi-
tion, operational cleaning occurs during the plant's service
life. Operational cleaning removes scale and corrosion products
that accumulate on the boiler's steam-side and on the water-side
of the steam condenser. The frequency of chemical cleaning
varies from power plant to power plant. The active reagents in
cleaning solutions are acidic or alkaline in nature depending
on the deposits to be removed. Ninety percent of cleaning oper-
ations employ acidic solutions. Acid solutions attack all forms
of alkaline scale (i.e.s CaC03, Mg(OH)2, etc.), silica scale,
and corrosion deposits containing iron. The majority of these
compounds contain hydrochloric acid in solution strengths ranging
from 5.0 to 7.5 percent (AY-007). Alkaline solutions are employed
to remove deposits passive to acid attack and to neutralize re-
siduals resulting from acid cleanings.
Miscellaneous Operations
A number of miscellaneous operations also constitute
sources of plant wastewater. For the most part, the impact of
wastewater from these miscellaneous operations is small compared
to those discussed previously; their nature is highly varied de-
pending upon specific plant characteristics. Three typical mis-
cellaneous operations are auxiliary cooling systems, water intake
screen washings, and laboratory and sampling operations.
-29-
-------
3.2 Lime Wet Scrubbing Process
In the lime flue gas desulfurization process S02 is
removed from the flue gas by wet scrubbing with a slurry of
calcium oxide (lime). The principal reaction for absorption
of S02 by the lime slurry is:
S02 f s + GaOx N + %H20 •> CaS03"%H20, s. (3.2-1)
(g) (s) u;
Oxygen absorbed from the flue gas or surrounding atmosphere causes
the oxidation of absorbed S02. The calcium sulfite formed in the
principal reaction and the calcium sulfate formed through oxi-
dation are precipitated as crystals in a hold tank. The crystals
are recovered in a solid/liquid separator. Waste solids disposal
is accomplished by ponding or landfill. The clear liquor is re-
cycled.
3.2.1 Process Description
The design of a lime wet scrubbing system includes the
following process areas:
1) S02 Absorption,
2) Solids Separation, and
3) Solids Disposal.
Figure 3.2-1 shows a generalized flow diagram for the
lime wet scrubbing process.
S02 Absorption
SO2 is removed from flue gas in a wet scrubber by ab-
sorption into a circulating slurry of lime. Calcium sulfite is
-30-
-------
u>
I-1
I
LIME
REHEATER
FAN
TO STACK
S02 ABSORBER
FLUE GAS
'STEAM
•*- MAKE-UP WATER
SLURRY
EFFLUENT HOLD TANK
SECOND STAGE
SOLID-LIQUID
SEPARATOR
OR
SETTLING POND
SOLID-LIQUID
SEPARATOR
SOLID WASTE
Figure 3.2-1. Process flow diagram for the lime wet scrubbing process.
-------
formed in the principal absorption reaction, and calciuin sulfate
is formed as the result of oxidation in a secondary reaction.
The calcium sulfite and sulfate formed in the scrubber are then
precipitated in a hold tank. A 10-15 percent solids slurry is
recycled to the absorber from the hold tank. A bleed stream is
sent to solids dewatering for subsequent disposal.
The flue gas can be pretreated for particulate removal
with an electrostatic precipitator or particulate scrubber.
Particulates can also be removed in the S02 absorber, although
this increases the solids load in the S02 scrubbing system. ' In
addition, it is believed that some components of fly ash catalyze
the oxidation of sulfite to sulfate, thus increasing the poten-
tial for sulfate scaling. The selection of particulate removal
method is based on assessments of operational reliability and
the economics of installing particulate control devices.
The feed material for a lime scrubbing process is
usually produced by calcining limestone. Carbide sludge, an
impure, slaked lime by-product of acetylene production, has also
been used successfully at two installations.
The absorption of S02 from flue gas by a lime slurry
constitutes a multiphase system involving gas, liquid and solid
phases. The reaction of gaseous S02 with the lime slurry yield-
ing calcium sulfite hemi-hydrate is shown in Equation 3.2-1.
S02(g) + CaO(g) 4- %H20 * CaS03-%H20(s). (3.2-1)
The solid sulfite is only slightly soluble in the scrubbing liquor
and will precipitate to form an inert solid for disposal. Some
C02 may be absorbed from the flue gas and will react in a similar
manner to form solid calcium carbonate.
-32-
-------
In most cases some oxygen will also be absorbed from
the flue gas or surrounding atmosphere. This leads to oxidation
of absorbed S02 and precipitation of solid CaSCU-2H20. The re-
action for this step is:
S°2(g) + ^°2(g) + Ca°(s) + 2Ez° * GaSO,-2H20(g).
(3.2-2)
The extent of oxidation can vary considerably, from almost zero
to 40 percent. In some systems treating dilute S02 flue gas
streams, sulfite oxidation as high as 90 percent has been ob-
served. The mechanism for sulfite oxidation is not completely
understood. The rate is known to be a strong function of oxygen
concentration in the flue gas and liquor pH. It may also be in-
creased by trace quantities of catalysts in fly ash entering the
system.
Several types of gas-liquid contactors can be used as
the S02 absorber. These differ in S02 removal efficiency as
well as operating reliability. Four types of contactors are
generally used for S02 removal:
venturi scrubbers,
spray towers (horizontal and vertical),
grid towers, and
mobile bed absorbers (such as TCA marble bed
and turbulent contact absorber).
The liquid to gas ratio (L/G) typically ranges between 0.005-
0.015 m3/Nm3 (35-110 gal/1000 scf) depending upon the type of
contactor. Simple impingement devices are placed downstream
from the absorber to remove mist entrained in the flue gas.
-33-
-------
The absorber effluent is sent to a hold tank for
precipitation of calcium sulfite and sulfate. The tank is
equipped with an agitator to prevent settling of solids and to
maintain uniform effluent composition. Additional streams
entering the tank include settling pond water, clarifier over-
flow, and makeup lime slurry. The hold tank is sized to allow
sufficient residence time for dissipation of supersaturation
and precipitation of calcium sulfite and sulfate. Too little
residence time in the hold tank can cause scaling as a result
of nucleation of calcium sulfite and sulfate solids in the
scrubber.
Solids Separation
A bleed stream is taken off the effluent hold tank to
be dewatered. This step, necessary to minimize the land area
needed for sludge disposal, varies depending on the application
and type of disposal.
For systems with on-site pond disposal, solids may be
pumped directly from the effluent hold tank to the pond area.
Clean overflow liquor from the pond would then be returned to
the system. If necessary, a thickening device such as a clarifier
or centrifuge can be used to increase the solids content.to a
maximum of about 40 weight percent. Additional dewatering to
60-70 percent solids can sometimes be achieved by vacuum fil-
tration.
Solids Disposal
The lime flue gas desulfurization process is a non-
regenerative or "throwaway" process. Sludge disposal is one of
the main disadvantages of "throwaway" FGD systems as compared
-34-
-------
to "recovery" processes. The quantity of sludge produced is
large in weight and volume; it requires a large waste pond or
landfill area for disposal.
On-site disposal is usually accomplished by sending
the waste solids to a large pond, where settling of the solids
occurs. Pond water is recycled to the process hold tank for
reuse.
Stabilization methods are currently under development
to convert the sludge to structurally-stable, leach-resistant
landfill material. When on-site disposal is not possible, the
stabilized material could be trucked to an off-site landfill.
3.3 Limestone Wet Scrubbing Process
In the limestone flue gas desulfurization process, S02
is removed from the flue gas by wet scrubbing with a slurry of
calcium carbonate. The principal reaction for absorption of
S02 is:
GaC°3 + %Ez° * CaS°3'%H2° + C°
(s) z 3'2(s) 2(g)- --
Oxygen absorbed from the flue gas or surrounding atmosphere may
cause the oxidation of absorbed S02 . The calcium sulf ite formed
in the principal reaction and the calcium sulfate formed via
oxidation are precipitated as crystals in a hold tank. The crys
tals are then sent to a solid/ liquid separator where the solids
are removed. Waste solids disposal is accomplished by ponding
or landfill. The clear liquor is recycled.
-35-
-------
3.3.1 Process Description
The design of a limestone wet scrubbing system can be
divided into the following process areas:
1) S02 Absorption,
2) Solids Separation, and
3) Solids Disposal.
Figure 3.3-1 shows a generalized flow diagram for the
limestone wet scrubbing process.
302 Absorption
SO2 is removed from flue gas in a wet scrubber by ab-
sorption into a circulating slurry of calcium carbonate. Calcium
sulfite is formed in the principal absorption reaction; calcium
sulfate is formed as the result of oxidation in a secondard re-
action. The calcium sulfite and sulfate formed in the scrubber
are then precipitated in a hold tank. A 10-15 percent solids
slurry is recycled to the absorber from the hold tank. A bleed
stream is sent to solids dewatering for subsequent disposal.
The flue gas can be pretreated for particulate removal
with an electrostatic precipitator or particulate scrubber. Par-
ticulates can also be removed in the S02 absorber, although this
increases the solids loading in the S02 scrubbing system. In
addition, it is believed that some components of fly ash catalyze
the oxidation of sulfite to sulfate, thus increasing the poten-
tial for sulfate scaling. The selection of particulate removal
method is based on assessments of operational reliability and the
economics of installing particulate control devices.
-36-
-------
I
u>
REHEATER
S02 ABSORBER
FLUE GAS
'STEAM
•*- MAKE-UP WATER
LIME-_
STONE"
CRUSHING
AND
GRINDING
SLURRY
EFFLUENT HOLD TANK
FAN
TO STACK
SECOND STAGE
SOLID-LIQUID
SEPARATOR
OR
SETTLING POND
SOLID-LIQUID
SEPARATOR
SOLID WASTE
Figure 3.3-1. Process flow diagram limestone wet scrubbing process.
-------
The absorption of S02 from flue gas by a limestone
slurry constitutes a multiphase system involving gas, liquid,
and solid phases. The reaction of gaseous S02 with the slurry
yielding calcium sulfite hemi-hydrate is shown in Equation 3.3-1:
SOZf. + CaC03,x + %H20 + CaS03'%H2Of . + C02( .. (3.3-1)
(.gj {s} \*> j \e>/
The solid sulfite is only slightly soluble in the scrubbing liquor
and will precipitate to form an inert solid for disposal.
In most cases some oxygen will also be absorbed from
the flue gas or surrounding atmosphere. This leads to oxidation
of absorbed S02 and precipitation of solid CaSOn.' 2H20. The re-
action for this step is:
(3.3-2)
s°2(g) + %02(g) + CaC03(-s) + 2H20 + CaSO* • 2H20,0^ + C02
Although the extent of oxidation can vary considerably, it nor-
mally ranges from almost zero to 40 percent. In some systems
treating dilute SO flue gas streams, sulfite oxidations as high
as 90 percent have been observed. The mechanism for sulfite
oxidation is not completely understood. The rate is known to be
a strong function of oxygen concentration in the flue gas and
liquor pH. It may also be increased by trace quantities of cata
lysts in fly ash entering the system.
Several types of gas-liquid contactors can be used as
the S02 absorber. These differ in S02 removal efficiency as
well as operating reliability. Four types of contactors are
generally used for S02 removal:
venturi scrubbers,
spray towers (horizontal and vertical),
-38-
-------
grid towers, and
• mobile bed absorbers (such as TCA marble bed and
turbulent contact absorber).
The liquid to gas ratio (L/G) typically ranges between 0.005 -
0.015 m3/Nm3 (35-110 gal/1000 scf) depending upon the type of
contactor. Simple impingement devices are placed downstream from
the absorber to remove mist entrained in the flue gas.
The absorber effluent is sent to the hold tank for
precipitation of calcium sulfite and sulfate. The tank is
equipped with an agitator to prevent settling of solids and to
maintain uniform effluent composition. Additional streams enter-
ing the tank include settling pond water, clarifier overflow, and
makeup lime slurry. The hold tank is sized to allow sufficient
residence time for dissipation of supersaturation and precipita-
tion of calcium sulfite and sulfate. Too little residence time
in the hold tank can cause scaling as a result of nucleation of
calcium sulfite and sulfate solids in the scrubber, resulting in
scaling.
Solids Separation
A bleed stream is taken off the effluent hold tank to
be dewatered. This step, necessary to minimize the land area
needed for sludge disposal, varies depending on the application
and type of disposal.
For systems with on-site pond disposal, solids may be
pumped directly from the effluent hold tank to the pond area.
Clean overflow liquor from the pond would then be returned to
the system. If necessary, a thickening device such as a clarifier
or centrifuge can be used to increase the solids content to a
-39-
-------
maximum of about 40 weight percent. Additional dewatering to
60-70 percent solids can sometimes be achieved by vacuum fil-
•it
tration.
Solids Disposal
The limestone flue gas desulfurization process is a
non-regenerative or "throwaway" process. Sludge disposal is one
of the main disadvantages of "throwaway" FGD systems as compared
to "recovery" processes. The quantity of sludge produced is
large in weight and volume, and requires a large waste pond or
landfill area for disposal.
On-site disposal is usually accomplished by sending
the waste solids to a large pond where settling of the solids
occurs. Pond water is recycled to the process hold tank for
reuse.
Stabilization methods are currently under development
to convert the sludge to structurally-stable, leach-resistant,
landfill material. When on-site disposal is not possible, the
stabilized material could be trucked to an off-site landfill.
3.4 Wellman-Lord Sulfite Scrubbing Process
The Wellman-Lord Sulfite Scrubbing Process is a regen-
erable flue gas desulfurization process marketed by Davy Powergas.
It is based on the ability of a sodium sulfite solution to absorb
S02 and form a solution of sodium bisulfite. The sodium bisulfite
solution can be thermally regenerated to produce a concentrated
stream of S02 and the original sodium sulfite solution. The con-
centrated S02 stream can be processed to produce elemental sulfur,
sulfuric acid, or liquid S02. The regenerated sodium sulfite
-40-
-------
solution is recycled to the absorber. In the absorption phase
of the process, sulfates formed by oxidation of sulfites are
removed from the system in a purge of sodium sulfate and sulfite
solids.
3.4.1 Process Description
The Wellman-Lord Process consists of five processing
areas:
1) Gas pretreatment,
2) Absorption,
3) Regeneration,
4) Purge treatment, and
5) SOa conversion.
A simplified process flow sheet appears in Figure 3.4-1.
The gas pretreatment and absorption sections are essentially the
same as those found in most aqueous scrubbing systems. No unique
equipment is used in any of the processing areas with the possible
exception of the S02 conversion step, which is licensed technology.
Gas Pretreatment
Flue gas is pretreated in a venturi or tray-type pre-
scrubber to cool and humidify the gas, and to reduce fly ash and
chlorides. The humidification and cooling step prevents the
evaporation of excessive amounts of water in the absorber. The
potential for scaling and plugging problems is reduced by the
removal of fly ash; in a well designed prescrubber, 99 percent
of the chlorides can be removed, thus reducing the potential for
stress corrosion.
-41-
-------
ABSORBER
VENTURI
PRESCRUBBER
TO STACK
-P-
ho
I
Figure 3.4-1. Process flow diagram for the Wellman-Lord sulfite
scrubbing process„
-------
Flue gas, exiting the electrostatic precipitator at a
temperature of 150°C (300°F) , is passed through a venturi or tray-
type prescrubber. There the gas is cooled to around 55°C (130°F)
and humidified. The venturi is preferred because 70 to 80 percent
of the remaining fly ash and 95 to 99 percent of the chlorides are
removed. Although a tray-type prescrubber is satisfactory for
cooling and humidifying the gas with low pressure drop, it provides
lower reductions in fly ash and chlorides. The fly ash and other
solids collected by the prescrubber are pumped to the ash disposal
pond as a slurry of approximately 5 percent solids. Since ab-
sorption of chlorides and some SOa and SOa can cause the slurry
to become fairly acidic, neutralization is accomplished with lime
when necessary.
Absorption
Cooled and humidified gas from the prescrubber passes
upward through an absorption tower, where SOz is removed by ab-
sorption into the sodium sulfite scrubbing solution. After the
cleaned gas is reheated to about 80°C (175°F) (so that it has the
proper dew point and buoyancy), it is then exhausted to the at-
mosphere. The scrubbing solution is sent to regeneration and
purge treatment.
Davy Powergas offers two types of absorption units: a
packed tower for small volume applications and a valve tray tower
for large volume applications. The valve tray unit, built in a
square configuration, includes three to five trays depending on
the inlet S02 concentration and the degree of desulfurization
required. Three to four absorbers would be used in a 500 Mw
installation. Because sodium sulfite has a large capacity to
absorb S02, the feed liquor flow rate is low. Recirculation is
practiced on each stage to maintain good hydraulic-characteristics.
-43-
-------
With recirculation, the L/G ratio is kept at approximately
.0004 m3/Nm3 (3 gal/1000 scf) per tray.
The absorption of S02 proceeds according to Equation
3.4-1;
Na2S03 + S02 + H20 * 2NaHS03 (3.4-1)
Makeup sodium carbonate also reacts with S02 in the
absorber to form sodium sulfite by Equation 3.4-2:
Na2C03 + S02 ->• Na2S03 + C02 (3.4-2)
A very important side reaction is the oxidation of sulfite to
sulfate by oxygen in the flue gas as in Equation 3.4-3:
Na2S03 + %02 + Na2SCU (3.4-3)
Some sodium sulfate is also formed by absorption of S02 from
the flue gas as in Equation 3.4-4.
2Na2S03 + S03 4- H20 -> 2NaHS03 + Na2SCK (3.4-4)
Cooled and humidified gas from the prescrubber is
passed up through the absorption tower, where the S02 level is
reduced by at least 90 percent. The cleaned gas is reheated by
heat exchange with high-pressure steam and exhausted to the at-
mosphere. Although alternatives to this exist, the use of steam
allows coal to be used indirectly rather than premium fuels such
as oil or natural gas .
-44-
-------
Regeneration
Regeneration of sodium sulfite is accomplished by re-
versing the absorption reaction through the addition of heat.
The absorber product and purge centrate (discussed in the follow-
ing section) enter a double effect evaporator where S02 and water
vapor are driven off. The S02 and water vapor are subjected to
partial condensation to remove most of the water and produce a
concentrated S02 stream. The evaporator bottoms high-solids
sodium sulfite solution, condensate, and makeup Na2C03 are mixed
in a dissolving tank and recycled to the absorber.
The regeneration of sodium sulfite proceeds according
to Equation 3.4-5:
2NaHS03— j— Na2S03 + H20 + S02 (3.4-5)
Because of the higher temperatures, there is an in-
creased formation of thiosulfate by the disproportionation re-
actions 3.4-6 and 3.4-7:
6NaHS03 -»• ZNazSOi, + Na2S203 + 2S02 + 3H20 (3.4-6)
2NaHS03 + 2Na2S03 -f 2Na2SCK + Na2S203 + H20 (3.4-7)
The combined stream of absorber product liquor and purge centrate
is split between the two evaporator effects, each of which oper-
ates under a vacuum. Fifty-five percent goes to the first effect,
with 45 percent going to the second. The first effect operates
at 95°C (200°F) and is heated with low pressure steam by an ex-
ternal shell and tube exchanger. The S02 and water vapor driven
off overhead from the first effect are used to heat the second
effect which operates at about 75°C (170°F) . The approximately.
-45-
-------
45 percent undissolved solids content in each effect is primarily
sodium sulfite. The regeneration reaction is limited by the
equilibrium concentration of sulfite ion in solution. Fortunately,
since sodium sulfite is less soluble than sodium bisulfite, it is
continuously removed from solution by crystallization, thus driv-
ing the reaction forward. The evaporator product is sent to the
dissolving tank.
The SO2 and water vapor overhead from the evaporators
is subjected to partial condensation to remove most of the water
and concentrate the S02. The condensate, containing several
hundred ppm of dissolved S02, is steam-stripped to lower this
value. Along with a small amount of makeup water and sodium
carbonate, the stripped condensate is sent to the dissolving tank.
There it is agitated with the sodium sulfite slurry from the evap-
orators to provide absorber feed. The SO2 stream exiting the
condenser contains only 5-10 percent water. It is compressed
and sent to an SOz conversion process.
Purge Treatment
A sidestream of the absorber product liquor is drawn
off to the purge treatment area for the removal of sodium sulfate
in a chiller/crystallizer.
About 15 percent of the absorber product liquor is sent
to the purge treatment area for removal of sodium sulfate. This
stream is precooled by heat exchange with the cold, sulfate-free
product. When the purge is cooled to 0°C (32°F) in a chiller/
chrystallizer, a mixture of sodium sulfate and sulfite is crys-
tallized out. The sodium sulfate and sulfite slurry is centri-
fuged to produce a 40 percent solids cake. Just prior to steam
drying, a secondary purge stream is drawn off the evaporators and
-46-
-------
added to the cake. This secondary purge removes thiosulfate
formed in the regeneration phase of the process. The resulting
product is a crystalline mixture of anhydrous sodium sulfate
(70 percent), and sodium sulfite (30 percent), with small amounts
of thiosulfates, pyrosulfites, and chlorides. The sulfate-free
supernatant liquor from the centrifuge is heated by passing it
through the feed cooler. It is then returned to the product
liquor stream entering the evaporator loop. The vent gases from
the dryer are cleaned to remove dust and returned to the main
flue gas stream before it enters the prescrubber.
SO2 Conversion
Several processes can be used to convert the concentrated
SOa stream produced by the Wellman-Lord Process into a more useful
form. The SOa can be converted to sulfur by several processes
that have been demonstrated or are under development. S02 can
also be converted to sulfuric acid in a contact sulfuric acid
plant. The acid production is less consumptive of fuel and re-
ducing media, but the acid produced is more difficult to store
and ship. The SOa Conversion Processes are discussed in detail
in Section 3.8.
3.5 • Magnesia Slurry Absorption Process
The Magnesia Slurry Absorption Process is a regenerable
flue gas desulfurization process. S02 is removed from the flue
gases by wet scrubbing with a slurry of magnesium oxide. Mag-
nesium sulfite is the predominant species formed in the absorption
reaction shown in Equation 3.5-1:
Mg(OH)2 + S02 + MgS03 + H20. (3.5-1)
-47-
-------
The absorber effluent is centrifuged. The liquor is sent to
the slurry tank for combination with makeup water, makeup MgO,
and regenerated MgO to form the slurry feed for the scrubber.
The magnesium sulfite cake is dried to remove free and bound
water. Magnesium oxide is then regenerated in a calciner by
thermal decomposition of the magnesium sulfite according to
Equation 3.5-2:
MgS03 -* MgO + S02. (3.5-2)
The concentrated S02 gas stream can be used to promote sulfuric
acid or elemental sulfur.
3.5.1 Process Description
The design of the magnesia scrubbing system can be
divided into five process areas:
1) Gas Pretreatment,
2) S02 Absorption,
3) MgSOa/MgSO^ Separation and Drying,
4) MgO Regeneration and S02 Recovery, and
5) Sulfur Production.
Figure 3.5-1 is a simplified flow diagram for the process.
Gas Pretreatment
Flue gas is pretreated in a venturi or tray-type pre-
scrubber to cool and humidify the gas, and to reduce fly ash and
chlorides. The humidification and cooling step prevents the
evaporation of excessive amounts of water in the absorber. The
-48-
-------
REHEATER
VO
I
SOj ABSORBER
FLUE GAS
EFFLUENT HOLO TANK
•s^
TO STACK
MAKE-UP M)O
MAKE-UP WATER
COKE
CALCINEF)
Figure 3.5-1. Process flow diagram for the magnesia slurry absorption process.
-------
potential for scaling and plugging problems is reduced by the
removal of fly ash which, containing vanadium and iron compounds,
can catalyze the oxidation of MgSCh to MgSCU . In a well designed
prescrubber 99 percent of the chlorides can be removed, thus re-
ducing the potential for stress corrosion
Flue gas, exiting the electrostatic precipitator at a
temperature of 150°C (300°F) , is passed through a venturi or tray-
type prescrubber. There the gas is cooled to around 55° C (130°F)
and humidified. The venturi is preferred because 70 to 80 percent
of the remaining fly ash and 95 to 99 percent of the chlorides
are removed. Although a tray-type prescrubber is satisfactory
for cooling and humidifying the gas with low pressure drop, it
provides lower reductions in fly ash and chlorides . The fly ash
and other solids collected by the prescrubber are pumped to the
ash disposal pond as a slurry of approximately 5 percent solids.
Since absorption of chlorides and some SO 2 and SO 3 can cause the
slurry to become fairly acidic, neutralization is accomplished
with lime when necessary.
SOa Absorption
Development of the magnesia scrubbing process has
followed at least three major technical routes since the early
1930 's. Process variations include the use of magnesium sulfite/
magnesium oxide slurries having a basic pH, the use of magnesium
sulfite in acidic solution (clear liquor process).
The presence of manganese promotes desulfurization,
oxidation of magnesium sulfite to sulfate, and decomposition of
magnesium sulfate by roasting. The use of magnesium sulfite
in an acidic solution (pH less than 6.0) produces a clear liquor
which can be used on coal-fired systems where one scrubber is
-50-
-------
used to remove both the particulates and S02. A clear solution
is used so that the ash can be filtered out. Because the vapor
pressure of the scrubbing solution is higher, the process is re-
stricted by a lower S02 removal efficiency than the basic slurry
process. Since the basic slurry process is the most advanced
system it will be evaluated in this study.
An aqueous slurry of magnesium hydroxide and magnesium
sulfite (pH range 6.5 to 8.5) is used to absorb the S02 according
to Equations 3.5-1 and 3.5-3:
Mg (OH) 2 + SO 2 -> MgS03 + H20, (3.5-1)
MgS03 + H20 4- S02 + Mg(HS03)2. (3.5-3)
Sulfite oxidation gives rise to sulfates in the system by the
following reaction:
MgS03 + %02 -»• MgSO^. (3.5-4)
The test facility at Boston Edison reported sulfate concentration
to be in the 15-20 weight percent range for solids shipped to
the regeneration facility. As illustrated in the following equa-
tions, the sulfite and sulfate solids precipitate as hydrated
crystals:
MgS03 + 6H20 -*• MgS03'6H20, (3.5-5)
MgS03 + 3H20 •> MgS03-3H20, (3.5-6)
+ 7H20 ->• MgS0^7H20. (3.5-7)
-51-
-------
The bisulfite in the spent scrubbing liquor is reacted with
magnesium hydroxide which is formed by slaking fresh and recycled
magnesium oxide:
Mg(HS03)2 +Mg(OH)2 + 4H20 * 2 (MgS03•3H20), (3.5-8)
MgO + H20 -»• Mg(OH)2. (3.5-9)
Cooled and humidified gas from the presrubber is passed
through the absorption tower where the S02 level is reduced by at
least 90 percent. The cleaned gas, reheated So that it has the
proper dew point and buoyancy, is then exhausted to the atmosphere.
The gas is reheated by heat exchange with high-pressure steam.
Although alternatives to this reheating method exist, the use of
steam allows coal to be used indirectly rather than premium fuels
such as oil or natural gas.
When MgO slurry enters the absorption tower and absorbs
S02, MgSOs crystals are formed. A bleed stream is sent to a
centrifuge in the first step of MgO recovery. After makeup water,
recovered MgO, and makeup MgO are added to the recycle slurry,
the resulting slurry is returned to the absorber as scrubbing
solution feed.
In a 500 MW plant, four 125 MW scrubber trains are used
so that the scrubbers will be of reasonable size. This also en-
ables the plant to run at a 375 MW capacity while maintenance is
performed on one scrubber train.
MgS03./MgSOi» Separation and Drying
A bleed stream is taken off the absorber effluent stream
as a 10-15 percent solids slurry and sent to a thickener to produce
-52-
-------
a 40 percent solids slurry. This slurry is then centrifuged to
recover a wet cake of" MgS03 /MgS04 hydrate crystals. The wet
crystals are discharged from the centrifuge through a vertical
chute into a screw feeder. The feeder acts as a seal and provides
a continuous flow of wet solids into a rotary or fluid-bed dryer.
The rotary kiln type dryer is presently used in the.three U.S.
magnesia scrubbing demonstration units. Combustion gases from
an oil-fired burner are used to dry the crystals. A portion of
the gases is recycled to the dryer chamber for temperature con-
trol; the remainder is exhausted to the stack. The dried MgS03/
MgSOit is discharged from the dryer and conveyed to the calciner
for MgO regeneration and S02 recovery.
MgO Regeneration and SO2 Recovery
To generate MgO and S02, the dried MgSQ3/MgSOt» crystals
are calcined in an oil-fired rotary kiln or fluidized-bed reactor.
The thermal decomposition reaction of MgS03 is shown in Equation
3.5-2:
MgS03 + MgO + S02. (3.5-2)
MgSOit is reduced in the presence of carbon as shown in Equation
3.5-10:
MgSO^ + %C + MgO + S02 + %C02. (3.5-10)
Two installations have used rotary kilns to regenerate
the magnesium oxide. However, rotary kilns have high dust losses
and require a hot cyclone and venturi scrubber for magnesium
solids recovery. In a fluidized-bed calciner, most of the MgO
formed would go overhead with the S02 and combustion gases.
Thus, this method would also require separation equipment.
-53-
-------
The optimum calcining temperature in the reactor is set
to achieve decomposition of all of the MgSOa/MgSO^ solids without
"dead burning" the MgO. "Dead burned" MgO is chemically unreactive
and not effective for further S02 removal. Operating temperatures
in the 800°C (1500°F) range have been used in the rotary calciner.
SO2 Conversion
After dust removal, the sulfur dioxide-rich gas from
the calciner is sent to either a sulfur or sulfuric acid production
unit. The gas stream from the magnesium oxide calciner is well
suited for sulfuric acid production. The calciner off-gas is
saturated with water at 40°C (100°F) and contains about 8-10 per-
cent S02. Acid production is less consumptive of fuel and re-
ducing media, but the acid produced is more difficult to store
and ship. The S02 can also be converted to elemental sulfur by
several processes that have been demonstrated or are under de-
velopment. The S02 Conversion Processes are discussed in detail
in Section 3.8.
3. 6 Double Alkali Wet Scrubbing
The double alkali flue gas desulfurization process is
a "throwaway" process that removes S02 from the flue gas by wet
scrubbing with a sodium sulfite liquor. In a second step, a
waste sludge of calcium sulfite and sulfate is formed, as is a
regenerated sodium sulfite scrubbing liquor. Separating the
absorption and waste production steps has the advantage of
scrubbing the flue gas with a more concentrated soluble alkali.
This permits the use of lower liquid-to-gas ratios in the double
alkali process as compared with lime/limestone processes. Further-
more, the amount of soluble and slurried calcium in the scrubber
is minimized, thus offering the opportunity for better scale con-
trol.
-54-
-------
3.6.1 Process Description
A number of processes can be considered double alkali
processes. In the United States, most of the developmental work
has emphasized sodium-based double alkali systems using lime for
regeneration. Double alkali systems using an ammonia/calcium
base have been tested, but they suffer the disadvantage of poten-
tially producing a visible ammonium salt plume from the scrubbing
system. The following process description will be limited to
sodium/calcium-based processes.
The design of a double alkali system can be divided
into four process areas:
1) S02 Absorption,
2) Waste Production and Sorbent Regeneration,
3) Solids Separation, and
4) Solids Disposal.
Figure 3.6-1 shows a generalized flow diagram for a double
alkali system.
SO2 Absorption
Normally, gas from the electrostatic precipitator
passes through an absorption tower, where S02 is removed by
absorption into a sodium hydroxide or sodium sulfite scrubbing
solution. The gas may be saturated in a presaturation section
of the absorber before it enters the absorber itself. The
cleaned gas, reheated to 80°C (175°F) so that it has the proper
dew point and buoyancy, is then exhausted to the atmosphere.
The scrubber effluent liquor is regenerated with lime or
-55-
-------
FLUE GAS
REHEATER
FAN
SCRUBBER
t
Cn
LIME OR
LIMESTONE
SOLID / LIQUID
SEPARATOR
*- STACK QA3
MAKE-UP
-------
limestone in a reaction tank. The calcium sulfite and sulfate
solids formed in the reaction tank are removed from the system
in a solid/liquid separator. The separator liquor is recycled
to the absorber.
A prescrubber (for gas saturation and chloride removal)
on a separate liquor loop is not normally used in a double
alkali system as it is in a Wellman-Lord or Magnesia Slurry
system because of water balance problems. If a prescrubber is
used, the major water loss in the system (evaporation by the
flue gas) occurs in the prescrubber. The only water loss in the
remainder of the system is the water lost with the solid waste.
This small loss may not allow addition of enough water for lime
slaking, Na2C03 solution, and cake washing. The limitation of
cake washing would result in a high sodium make-up requirement
and a high dissolved solids content in the waste. Normally,
enough chloride is removed in the liquor discharged with the
solid waste to prevent excessive chloride build-up. In applica-
tions with coals having a very high chloride content, however,
this mechanism for chloride removal may not be adequate, and a
prescrubber may be required for chloride removal in spite of the
previously mentioned difficulties.
The principal reactions for the absorption of S02
from the flue gas are shown in equations 3.6-1 and 3.6-2:
2NaOH + S02 + Na2S03 + H20, (3.6-1)
Na2S03 + S02 + H20 + 2NaHS03. (3.6-2)
-57-
-------
In double alkali systems where lime is used as the regenerant,
the pH range varies over the hydroxide/sulfite/bisulfite range
so that reactions indicated by Equations 3.6-1 and 3.6-2 both
occur. In double alkali systems using limestone as the regen-
erant, the pH range is limited to the sulfite/bisulfite range
and only the second reaction (Equation 3.6-2) occurs.
A very important side reaction is the oxidation of
sulfite to sulfate caused by the absorption of oxygen in the
flue gas:
Na2S03 + %02 + Na2SCU. (3.6-3)
The absorber effluent is sent to the reaction tank.
The double alkali processes can be operated in either
the "dilute" or "concentrated" mode. Here these terms refer to
the concentration of active alkali (sulfite). In general,
dilute systems are more suited to applications in which oxida-
tion is expected to be relatively high, whereas concentrated
systems are favored in applications where oxidation is expected
to be low.
In order to minimize the potential for gypsum
(CaSO,,-2H20) scaling in the scrubber, a softening step is used
to reduce the dissolved calcium concentration in the scrubber
feed liquor. In dilute double alkali systems, carbonate soften-
ing is generally employed. Such systems utilize soda ash
(Na2C03) and carbon dioxide (C02) to precipitate the dissolved
calcium as carbonate:
Na2C03 + Ca(OH)2 + 2NaOH + CaC03i (3.6-4)
-58-
-------
CO2 4- Ca(OH) 2 + CaC03 + H20. (3.6-5)
The main function of the soda ash is to replenish the sodium
losses in the system (solid disposal). The carbon dioxide
supplies the additional softening required. In concentrated
systems, sulfite softening is used:
(3.6-6)
Na2S03 + Ca(OH)2 + %H20 + 2NaOH + CaS03.%H20.
From economic and chemical utilization standpoints, the system
should be operated with minimum softening to avoid scrubber
scaling.
Waste Production and Sorbent Regeneration
The scrubber effluent liquor is sent to a reaction
tank where the sorbent is regenerated with lime or limestone.
Solid calcium sulfite and sulfate are formed as shown for the
lime system by equations 3.6-7 through 3.6-9:
Ca(OH)2 + 2NaHS03 -»• Na2S03 + CaS03.%H20 + (3.6-7)
3/2H20,
Ca(OH)2 + Na2S03 + %H20 * 2NaOH + CaS03.%H20, (3.6-8)
Ca(OH)2 + Na2SO, + 2H20 •»• 2NaOH + CaS04.2H20. (3.6-9)
The corresponding reactions for the limestone system are given
by equations 2.6-10 and 3.6-11:
CaC03 + 2NaHS03 + Na2S03 + CaS03-%H20 + C02+ (3.6-10)
%H20,
-59-
-------
CaC03 + 2NaHSCK + H20 -> NaaSO^ + CaS0lt-2H20 + (3.6-11)
CO 2
Solids Separation
The calcium sulfite and calcium sulfate solids formed
in the reaction tank are withdrawn from the system in a solid/
liquid separator. After make-up alkali and water are added,
the separator effluent liquor is recycled to the scrubbing
loop. A liquid purge stream is required to remove soluble
sodium sulfate. Failure to allow for sulfate removal from
double alkali systems will ultimately result in 1) precipitation
of sodium sulfate somewhere in the system if active sodium is
made up to the system or 2) in the absence of make-up, eventual
deterioration of the S02 removal capability due to the loss of
active sodium from the system.
Solids Disposal
The double alkali flue gas desulfurization process is
a non-regenerative or "throwaway" process. Sludge disposal is
one of the main disadvantages of "throwaway" FGD systems when
compared to "recovery" processes. The quantity of sludge pro-
duced is large in weight and volume, and requires a large waste
pond or landfill area for disposal.
On-site disposal is usually performed by sending the
waste solids to a large pond. Settling of the solids occurs
and pond water is recycled to the process hold tank for reuse.
Stabilization methods are currently under development
to convert the sludge to structurally stable, leach-resistant,
landfill material. These methods could be used when ori-site
disposal is not possible. The stablized material can then be
trucked to an off-site area for landfill.
-60-
-------
3-7 Physical Coal Cleaning
First practiced in the United States in 1880, coal
cleaning is an established technology in the mining industry.
Since 1945 the annual percentage of coal cleaned has tripled.
At present, nearly 50 percent of the annual U.S. coal produc-
tion is physically cleaned. To date, coal cleaning operations
have not been utilized to control sulfur oxide emissions. The
primary functions of coal preparation plants have been to remove
rock and ash from coal and to produce coking grade coals for
use in metallurgical processes. In these operations, sulfur
removal has not been optimized. However, pilot plant studies
to optimize sulfur removal have been reported; currently General
Public Utilities Corporation and New York Electric and Gas Cor-
poration are constructing a full-scale facility to achieve
compliance with sulfur oxide emission regulations (DA-189).
Physical coal cleaning removes impurities from coal
via a mechanical separation process. In most cleaning opera-
tions, this separation of impurities is based on a gravity
difference between coal (which is relatively light) and
contaminants such as pyrite (FeS2), ash, and rock (which are
heavier) (PA-003).
Sulfur occurs in a coal seam in three forms: pyritic,
organic, and sulfate. In any given coal the amount of sulfate
sulfur is negligible. The total sulfur content may vary from
less than 1 percent to over 8 percent, with most coals in the
2 to 5 percent range. The total sulfur content distribution
between the organic and pyritic forms ranges from 5 to 60 per-
cent and 40 to 95 percent, respectively.
-------
Since organic sulfur in coal is chemically bound and
requires a chemical extraction process for removal, physical
coal cleaning is restricted to removal of the pyritic sulfur
from coal.
The pyritic sulfur content of coal is present in
many forms and particle sizes. These range from coarse sized
particles, which are relatively easy to remove, to finely dis-
seminated particles which could almost be classified as inherent,
Not only is the distribution and nature of pyrite particles
different in various coal seams, but the amount and variety of
sulfur forms will vary in the same coal bed from one area to
another. Therefore, it is difficult to draw conclusions as to
which physical cleaning methods will best suit a particular
coal seam.
The potential for sulfur reduction in coal by appli-
cation of conventional physical coal cleaning technology is
limited. The U.S. Bureau of Mines investigated the sulfur
release potential of 322 coal mines representing a large per-
centage of utility coal sources. The results of this study
indicated that the physical removal of sulfur is both coal and
process specific (DE-064). Individual coals respond uniquely
to the various unit operations utilized in cleaning facilities.
The generalized results obtained from the 322 mines
studied in the Bureau of Mines investigation indicate that an
average cleaning process operating at 90 percent yield has the
potential to reduce the total sulfur content of the coal by 30
percent. Of the 322 mines sampled, less than 30 percent of the
coal could be cleaned to a sulfur content of 1 percent (DE-064).
-62-
-------
3.7.1 Process Description
Although it is not possible to describe a universally
applicable coal cleaning process, certain processing areas which
are common to most cleaning operations can be identified.
Figure 3.7-1 is a flow diagram of a coal cleaning facility
depicting common process areas without detailing specific unit
operations (LO-071, CO-380).
The following process areas are found in most coal
cleaning facilities. Listed under each area are various opera-
tions which may be utilized in an individual cleaning process.
Initial Coal Preparation
1) Storage
Bins, Silos, and Hoppers
2) Rough Gleaning/Primary Breaking
Tramp Iron Removal
Scalping Screens
Crushing Equipment
3) Raw Coal Sizing
Shaking Screens
• Vibrating Screens
Fine Coal Processing
1) Dry Cleaning
Airflow Cleaner
Air Tables
• Centrifugal/Electrostatic Separator
-63-
-------
TRAMP
IRON
REMOVAL
RECinCULATINO PLANT WATER
COARSE COAL
CLEANING
SLIMES
REFUSE
(REJECT STREAM)
REJECT STREAMS
COAL FLOWS
. WATER FLOWS
FINE COAL'CLEANING
DRY ' WET
PROCESSlPnOCESS
REFUSE
Figure 3.7-1. Generalized coal cleaning process.
-------
2) Wet Cleaning
Jigs
Heavy-Medium Cyclones
Water Tables
Hydrocyclones
Spiral Classifiers
• Froth Flotation
3) Desliming
Coarse Coal Processing
Jigs
Heavy-Media Separators
Hydrocyclones
Launders
Water Management/Refuse Disposal
1) Dewatering
Sieve Bends
Screens
Thickners
Cyclones
2) Drying
Centrifuges
Filters
Thermal Dryers
3) Water Recovery
-65-
-------
4) Refuse Disposal
Fire Prevention
Prevention of Ground Water Pollution
1) Initial Coal Preparation
Prior to the actual cleaning process, run-of-mine
(R.O.M.) coal must undergo initial preparation. This involves
preliminary crushing of the coal to remove large rock fractions
and to liberate entrained impurities such as clay, rock, and
other inorganic materials, including pyrite. The first crushing
step is followed by a screening operation and secondary crush-
ing. A second screening step produces two product streams from
this process area: one containing a fine fraction (usually less
than 6.5 mm) and the other containing coarse particles (nomi-
nally 76 x 6.5 mm). These two coal flows are then fed to their
respective process areas where the actual cleaning operation
takes place (CO-380, LO-071).
2) Fine Coal Processing
The fine coal processing area of a preparation plant
can employ both wet and dry cleaning operations. In plants
utilizing a dry coal cleaning process, fine coal from the ini-
tial preparation step flows to a feed hopper and then to an
air cleaning operation. This cleaning operation can employ one
of several devices which rely on an upward current of air
traveling through a fluidized bed of crushed coal. Separation
is effected by particle size and density. Product streams
from a dry cleaning process are sent directly to the final coal
preparation step, while reject streams are usually processed
further in wet cleaning operations (FI-102).
-66-
-------
In operations utilizing wet methods to effect fine
coal cleaning, the process feed stream of less than 6.5 mm coal
is slurried with water as it enters the fine coal processing
area of the plant. This slurry is then subjected to a desliming
operation which removes a suspension containing approximately
50 percent of minus 200 mesh material (FI-102) . The cutoff
size for this separation is usually in the range of 28 to 48
mesh (PE-030) . This desliming operation is necessary because
the presence of slimes adversely affects the capacity and effi-
ciency of fine cleaning units.
Subsequent to desliming, the oversize coal fraction
(greater than 28 mesh) is pumped to the fine coal cleaning
process. Here, fine coal particles undergo gravity separation
in one of several wet cleaning devices. This removes a percen-
tage of ash and pyritic sulfur to produce a clean coal product.
The product stream from this operation is fed to the drying area
of the plant; refuse material is further processed in the water
treatment section.
The slimes removed from the fine coal stream are fed
to a froth flotation process. Other material, such as reject
from dry cleaning operations, may also be treated in the flota-
tion process. This process consists of "rougher" and "cleaner"
sections which are comprised of cells of flotation machines.
Upon entering the flotation process area, the slime suspension
is treated with a frothing agent. This agent selectively floats
coal particles in the flotation machines while allowing pyrite
and ash impurities to settle. Processing slime in the "rougher"
cells produces a reject stream and a low grade product. The low
grade product is further processed in the "cleaner" cells to
produce a clean coal product. This final float product is then
sent to the dewatering area for further handling, while reject
-67-
-------
material from both rougher and cleaner sections is processed in
the water treatment and recovery area.
3) Coars e Coal Pro cess ing
Feed to the coarse coal processing area of the plant
consists of oversize material (76 x 6.5 mm particles) from the
initial preparation area. This feed stream is slurried with
water prior to cleaning, since coarse coal cleaning operations
employ wet processing equipment to remove impurities from the
coal. The coarse coal slurry is fed to one of the many types
of process equipment currently employed in coarse coal cleaning.
Here, impurities are separated from the coal due to differences
in product and reject density. It is common practice to remove
a middling fraction from the separation operation and process it
further by means of recycle or by feed to another cleaning
proces. These cleaning operations result in removal of two
streams from the coarse coal processing area: a product and
reject stream. Subsequent to-the coarse cleaning operation,
the product stream is pumped to the dewatering and drying area
of the plant, while the reject stream is processed in the water
treatment and recovery area.
4) Water Management/Refuse Disposal
Dewatering and drying equipment handle the product
flows from both the fine and coarse coal preparation areas.
Typically, cleaning plants employ mechanical dewatering opera-
tions to separate coal slurries into a low-moisture solid and
a clarified supernatant. The solid coal sludge produced in the
dewatering step can be mechanically or thermally dried to
further reduce the moisture. The supernatant from the dewatering
process is returned to the plant's water circulation system.
-68-
-------
The water treatment and recovery section of a cleaning
plant processes refuse slurries containing both coarse material
and reject slimes. Here, the refuse slurry is dewatered, typi-
cally in thickeners and settling ponds. The supernatant from
this operation is returned for reuse in the plant, while the
refuse is buried and revegetated to prevent burning.
The coal product from the dewatering and drying area
of the plant can be further processed. This may involve crush-
ing and screening operations to separate the product into various
product sizes. The cleaned and sized product is then loaded
into rail cars for shipping.
3.8 SQ2 Conversion Processes
The Wellman-Lord and the Magnesium Slurry Wet Scrubbing
Processes produce an S02 rich product gas stream which can be
converted to either sulfuric acid or elemental sulfur. Several
available processes and several processes in the developmental
stages are suitable for FGD by-product production. In a recent
study for EPRI (OT-Q51), Radian chose the Allied Chemical Sulfur
Plant, and a single absorption contact sulfuric acid for eco-
nomic comparisons on the basis of data availability. These two
processes will be described, and the subsequent water effluent
and water consumption quantities evaluated.
3.8.1 Sulfuric Acid Production
On a capital cost basis, a single absorption contact
process is the suggested method for conversion of FGD S02
streams into sulfuric acid. The tail gas from the acid unit
would be sent back to the S02 scrubber to prevent S02 emissions.
An alternative to treating the tail gas would be the use of a
-69-
-------
Double Contact/Double Absorption Acid Unit (DC/DA). This
facility would result in added costs to the acid production unit
whereas routing the tail gas back through the FGD process would
result in increased FGD costs for processing the extra gas. An
economic comparison on a site specific basis is needed to deter-
mine the best choice. In this study, it is assumed that the FGD
system can handle the relatively small increase (about 5 percent)
in load from routing the gases back through the system.
Process Description
Basically, the single absorption contact process oxi-
dizes sulfur dioxide to sulfur trioxide over a vanadium pent-
oxide catalyst. The sulfur trioxide then combines with water
in an absorber where the product acid is formed. Single absorp-
tion contact acid plants operate at sulfur dioxide conversion
efficiencies of about 97 percent. A typical flow diagram of
the process is illustrated in Figure 3.8-1.
The feed gas to the process should consist of 8.4 to
9.0 volume percent SOz and 8.6 to 9.2 volume percent Oz on a
dry basis. This feed gas first passes through a drying tower.
The dry gas then passes through a series of heat exchangers to
bring the gas temperature to 435°C (815°F). The hot gas passes
through the first three catalyst beds with intermediate heat
exchange to remove the heat generated by the exothermic reaction
shown below:
catalyst
S02 + %02 * S03. (3.8-1)
After the third catalyst bed, the gas is fed to an
absorber where the S03 combines with the water in the circulat-
ing acid. Before being mixed with the flue gas going to the
-70-
-------
S02 STREAM
TO
ATMOSPHERE
CONVERTER
Q
TANK
CAR
FGD SCRUBBED
COOLINQ
WATER
96% ACID
ABSORPTION
TOWER
ON j
PRODUCT
STORAGE
TANK
PRODUCT
STORAGE
TANK
v
A ~"
' i
' I
ACID
COOLERS
93% AGIO
COOLERS
PUMP
TANK
98* AGIO
COOLERS
Figure 3.8-1. Typical flow diagram of a single absorption
contact sulfuric acid plant.
-------
SOa absorber, the vapor from this absorber passes through a
demister to remove any entrained acid mist.
Sulfuric acid from the absorption towers flows to the
acid pump tank where it is diluted to the required strength for
absorption. The acid is pumped through a series of coolers be-
fore returning to the absorption towers. The product acid is
removed from the absorption tower system through product coolers
and sent to storage.
3.8.2 Sulfur Production
Elemental sulfur is currently produced by reacting an
S02 stream with a reducing agent at elevated temperatures.
Three processes have been applied to an S02 source from an FGD
process: the Allied Chemical Process using natural gas, the
RESOX Process using anthracite coal, and the BAMAG Process using
a medium Btu town gas. In addition, preliminary work is being
done to investigate the use of coal gasification reducing gas
in the Allied Chemical Process. Because the RESOX and BAMAG
processes are in early developmental stages, the Allied Chemical
Process will be selected as the process for elemental sulfur
production for water effluent and water consumption evaluations
in our study.
Allied Chemical Corporation has developed and commer-
cialized a process for direct, catalytic reduction of S02 to
elemental sulfur using natural gas as a reductant. The first
plant to use the process operated successfully for two years
using a 12 percent SOa stream from a sulfide-ore roasting facil-
ity.
The process may be joined to a regenerable FGD process
that produces a S02-rich gas stream with a low oxygen content.
-72-
-------
This application is being demonstrated at the D. H. Mitchell
Station of Northern Indiana Public Service Co. (NIPSCO) at Gary,
Indiana. There, the process has been combined with the Wellman-
Lord SO 2 Recovery Process to provide an FGD system for a 115 MW
coal-fired boiler. Acceptance testing is scheduled for July 1977
Process Description
A process flow diagram for an Allied Chemical Process
with an S02 feed stream from an FGD system is shown in Figure
3.8-2. The plant consists of three main sections: 1) gas puri-
fication, 2) S02 reduction, and 3) sulfur recovery. The gas
purification system, which is designed to remove excess water
vapor and gaseous and solid impurities, is not required for all
of the FGD processes.
The principle function of the catalytic reduction
section is to increase the H2S/S02 ratio in the gas stream to
approximately the stiochiometric ratio of 2:1 required for the
Glaus reaction, while achieving maximum formation of elemental
sulfur. The primary reaction system may be summarized in the
following equations :
C!U + 2S02 -> C02 + 2H20 + S2 (3.8^2)
+ 6S02 -»• 4C02 + 4H20 + 4H2S + S2 (3.8-3)
In the reduction section, the S02 stream, which has been combined
with preheated natural gas, first passes through a four -way flow
reversing valve and a final preheat reactor. The heated stream
then enters the primary reactor where over 40 percent of the
total recovered sulfur is formed. The reactor, which uses a
-73-
-------
i
--j
-P-
SO2 GAS
GAS PURIFICATION AND COOLING
HOT GAS HEAT EXCHANGER
t
MAIN BLOWER
NATURAL GAS
REDUCTION
REACTOR
SULFUR CONDENSER
SULFUR CONDENSER
FEED GAS HEATER
TO UTILITY BOILER
OR INCINERATOR
SULFUR STORAGE
Figure 3.8-2. Typical process flow diagram of the allied
chemical S02 reduction process.
-------
catalyst developed by Allied Chemical that is stable up to
1100°C (2000°F), achieves efficient methane utilization and
provides minimum formation of undesirable side products. Care-
ful control of the reaction conditions is necessary to achieve
chemical equilibrium in the single reactor.
From the reduction reactor, the gas passes through
a second heat regenerator where it gives up its heat to the
packing. Direction of flow through the two heat regenerators
is periodically reversed to interchange their functions of heat-
ing and cooling.
After condensing sulfur in a steam generator, the gas
stream enters a two-stage Glaus reactor system where H2S and S02
react to produce elemental sulfur and water. At this point,
product sulfur is again removed from the gas by condensation.
Residual H2S in the Glaus plant effluent gas is oxidized to S02
by recycling the gas stream back to the boiler. The residual
S02 is then recovered in the absorber of the original FGD
recovery process.
According to the developer, this process can be applied
directly to S02 streams containing as low as four percent S02,
where the oxygen content is not over five percent. Processing
streams with low S02 concentrations will, however, be costly
as compared to more concentrated streams. When higher oxygen
concentrations are encountered, provisions must be made to
dissipate the excess heat produced as a result of methane oxida-
tion. The Canadian plant has demonstrated that this process is
capable of converting better than 90 percent of the S02 from the
entering gas stream. Operation at one-third of design capacity
with constant operation efficiency has been established. The
major disadvantage of this method of sulfur production is the
need for methane as a reductant.
-75-
-------
Since the availability of methane in the future is
predicted to decrease substantially, a process using a CO/H2
coal gasification reducing gas appears to be an attractive alter-
nate route to elemental sulfur production. Allied Chemical has
made a preliminary evaluation of using a CO/H2 reducing gas.
The primary reactions in such a process would be as follows:
2CO + S02 -»• 2C02 + %S2 (3.8-4)
2H2 + S02 -> 2H20 + %S2 (3.8-5)
The configuration of such a process would probably be very simi-
lar to the present Allied Chemical process using methane reduc-
tant.
-76-
-------
4.0 WATER CONSUMPTION
This section assesses the impact of S0x control tech-
nologies on coal-fired power plant water consumption. Many of
the systems in a coal-fired power plant, the FGD systems under
study, the physical coal cleaning process, and the S02 conver-
sion processes, have large circulating water requirements.
Fresh water makeup is required due to evaporation, solids occlu-
sion, and blowdown losses. This section presents the results
of calculations to determine the effect of the SO control
technologies on raw water makeup requirements. First, the
water consumption of a power plant uncontrolled for SO emis-
X
sions is calculated. Then, the water consumption for each of
the various S0x control strategies is calculated. A matrix
presentation of the results by model plant system is made. A
more detailed presentation of the calculations is made in
Appendix A. A base case 500 MW power plant is used for discus-
sion.
4.1 Coal-Fired Power Plant Water Consumption;
Uncontrolled for SOX Emissions
The systems and operations in a coal-fired power
plant that require fresh water makeup are the
cooling water system,
ash handling system,
boiler makeup,
water conditioning operations,
equipment cleaning operations, and
miscellaneous.
-77-
-------
Due to the general unavailability of data on inter-
mittent and miscellaneous water consumptive practices, water
conditioning, equipment cleaning, and miscellaneous operations
will be addressed as "general service water". Plant data from
water recycle/reuse studies conducted by Radian for the EPA
(NO-106, NO-137, HA-636, GA-203, CH-387) will be used to charac-
terize these requirements.
Current water management in the power generation
industry uses two basic processes: 1) once-through, and
2) recirculatory. Plant layout and water management practices
vary widely in power plant water systems. Two extremes in water
management are those in which: 1) once-through techniques are
used in all systems, and 2) recirculatory practices are maxi-
mized in all systems. The total recirculatory system ultimately
results in zero-discharge. The once-through system requires/
discharges enormous quantities of water. Because of costs
associated with water conditioning operations and wastewater
treating, some recirculatory use of water is common practice.
And because of the potential for scaling and fouling of lines,
total recirculatory use of water is limited when fresh water is
readily available. As the nation approaches a goal of zero
discharge, recirculatory systems are becoming more prevalent.
To characterize the range of water consumption patterns typical
of current practices, Radian has chosen to characterize four
model power plant water management systems. These represent
the full range of power plant water management systems used
today. The four systems are:
System #1: All power plant water systems are once-
through; Figure 4.1-1
System #2: Recirculatory cooling at 2.5 cycles of
concentration, once-through ash handling,
and once-through general services water;
Figure 4.1-2.
-78-
-------
WATER SOURCE
RETURN TO
WATER SOURCE
RETURN TO
WATER SOURCE
Figure 4.1-1. Power plant water system: system #1 once-through
water management.
-79-
-------
WATER
SOURCE
SURGE POND
GENERAL
SERVICE
WATER
WATER
TREATMENT FOR
BOILER MAKE-UP
DRIFT
ASH PONO
RETURN TO
WATER SOURCE
Figure 4.1-2.
Power plant water system: system #2, partial
recirculatory water management.
-80-
-------
System #3: Recirculatory cooling at 5.0 cycles of
concentration, 50 percent recirculatory
ash handling, and recycle of general
service water blowdown to the ash hand-
ling system; Figure 4.1-3.
System #4: All power plant water systems are
recirculatory; Figure 4.1-4.
4.1.1 Cooling Water System
In coal-fired steam/electric power plants, the heat
of combustion produces steam to power turbine generators. The
steam is subsequently condensed and returned to the boiler for
further service. Approximately 45 percent of a fossil-fuel
fired generating station's energy is removed and ultimately
discharged to the environment by the condenser cooling system
(DI-139). To calculate the total cooling water requirement,
a power plant efficiency of 37 percent (MC-147) was used. For
a 500 MW power plant, 610 MW (35 MM Btu/min) heat removal capacity
is required. If a 10°C (20°F) rise in cooling water temperature
is assumed in the condenser, a circulating flow of 13 m3/s
(210,000 gpm) is required.
In once-through cooling systems, the makeup water
requirement is equal to the circulating rate, i.e. 13 m3/s
(210,000 gpm).
In recirculatory systems, the required blowdown to
achieve the desired cycles of concentration is a function of
drift (entrained water carried out by the exhaust air) and
evaporation:
r - B+D+E
L B+D
-81-
-------
WATER
SOURCS
_L
SURGE
PONO
GENERAL
SERVICE
WATER
RETURN TO
WATER SOURCE
ASH PONO
RETURN TO
WATER SOURCE
Figure 4.1-3. Power plant system: system #3,
recirculatory water management.
-82-
-------
QHtFT
, EVAPORATION
WATER
SOURCE
WATER
|TR£ATM6NT)
-FOR SOILED
MAKE-UP
ASH
POND
SLUICE
FLY
ASH
SLUICE
^
r
Figure 4.1-4. Power plant water system: system #4, zero
discharge water management.
-83-
-------
where: C = cycles of concentration,
B = blowdown,
D = drift, and
E = evaporation.
A power plants ability to operate its cooling system at high
cycles of concentration is limited by the maximum concentration
of a limiting impurity (hardness, dissolved solids, suspended
solids) or by the solubility limit of scaling salts (calcium
sulfate, calcium carbonate, etc.)- Recirculating water compo-
sition is determined by concentration of the constituents in
the makeup water, and possible treatment practices. Therefore,
the ability of a power plant to operate at high cycles of con-
centration is a site-specific problem.
For model plant water consumption calculations, makeup
water rates are calculated for three possible ranges of power
plant cooling tower operation. Makeup water requirements are
calculated for 2.5, 5.0, and 13.5 cycles of concentration.
The makeup requirement (M) is determined by blowdown (B),
drift (D) , and evaporation (E) .-
M = B + D + E.
The makeup requirement was extrapolated from plant
data from three sites (NO-106, NO-130, CH-387). A general
correlation was drawn for makeup requirement vs. cycles of
concentration for a 1000 MW power plant (RA-352). This extra-
polation assumes that the heat load is proportional to the power
plant capacity, and that the evaporation and drift rates are
similar. The two data sources showed excellent agreement. It
is also assumed that the plant data for evaporation and blow-
down rates are'characteristic values. The four power plant
-84-
-------
cooling systems are characterized in Table 4.1-1.
4.1.2 Ash Handling System
In calculating the water requirement for ash handling,
Radian has assumed that 1) fly ash is collected in an ESP and
wet sluiced to the ash pond as a 5 weight percent slurry, and
2) bottom ash is sluiced to the same pond as a one weight percent
slurry. These are characteristic industry values from water recy-
cle/reuse studies that Radian performed for the EPA (NO-106,
NO-137, HA-636, GA-203, CH-387). It is also assumed that 75
percent of the ash forms as fly ash and exits in the flue gas,
while 25 percent of the ash forms as bottom ash (MC-147). Six
representative coals were chosen for the model plant calcula-
tions. The compositions are shown in Table A.1-2 (Appendix A).
Coal usage rate and ash content determine both the amount of
ash to be sluiced and the sluice water requirement. The sluice
water requirement for each coal is shown in Table 4.1-2.
Sources for the ash sluicing water vary with the
power plant water management systems.
SYSTEM #1: It is assumed that raw water is the only
source for the ash sluicing water and that ash sluicing is
once-through. Therefore, the water makeup requirement equals
the sluicing requirement.
SYSTEM #2: It is assumed that cooling tower blowdown
is the source for the ash sluicing water and that ash sluicing
is once-through, as shown in Figure 4.1-2. The cooling tower
blowdown rate from the previous section is 0.16 m3/s (2500 gpm).
-85-
-------
I
CO
CTv
I
TABLE 4.1-1. CHARACTERISTIC COOLING SYSTEM OPERATION
System II
1
2
3
4
Cycles of
Description Concentration
Once-through
Partial
recirculatory
Recirculatory
Zero Discharge
1
2.5
5.0
13.5
Makeup
m'/s (gpm)
13 (210,000)
0.44 (7,000)
0.32 (5,000)
0.25 (4,000)
B lowdown
mVs
13
0.16
0.06
0.02
(gpm)
(210,000)
(2,500)
(900)
(300)
Evaporation
m /B (gpm)
0.27 (4200)
0.25 (3900)
0.23 (3670)
Drift
ma/s (gpm)
0.019 (300)
0.013 (200)
0.002 (30)
-------
TABLE 4.1-2. SLUICE WATER REQUIREMENT
I
CO
•-J
1
Coal
1
2
3
4
5
6
Fly
mVs
0.076
0.049
0.085
0.088
0.045
0.046
Ash
(gpm)
(1200)
( 780)
(1350)
(1400)
( 720)
( 730)
Sluice Water
Bottom
mVs
0.125
0.082
0.14
0.15
0.076
0.076
Requirement
Ash
(gpm)
(2000)
(1300)
(2250)
(2400)
(1200)
(1200)
Total
mVs
0.20
0.13
0.23
0.24
0.125
0.13
(gpm)
(3100)
(2050)
(3600)
(3800)
(2000)
(2050)
-------
Coals #2, #5, and #6 require less ash sluicing water
than available through use of cooling tower blowdown. The
excess blowdown could be used for other plant requirements or
ponded before discharge. No raw water is required to sluice
ash from these coals.
Coals #1, #3, and #4 require more ash sluicing water
than available through cooling tower blowdown. It is assumed
that this extra requirement is made up with raw water. Cycles
of concentration in the cooling tower may also be reduced, or
weight percent solids in the slurry could be increased. The
makeup water requirement for each coal is shown in Table 4.1-3.
SYSTEM #3: It is assumed that the cooling tower blow-
down is one source for ash sluicing water, as shown in Figure
4.1-3. Another assumption requires that the ash sluicing system
recycle 50 percent of the sluice water. A second source for
sluice water is the collected general services water blowdown
(discussed in Section 4.1.3). Table 4.1-4 summarizes the
requirements and sources of ash sluice water for each coal.
With a slight solids increase, coals #2, #5, and
#6 can essentially operate without any raw water
makeup requirement.
Coals #1, #3, and #4 require makeup water in the
quantities shown in Table 4.1-4. It is probable
that a power plant with a water management program
of this nature will probably collect "general
services" blowdown (discussed in the following
section) for use as makeup. Coal #1 can operate
without any raw water makeup by a slight solids
change. Coals #3 and #4 require raw water makeup
in the quantities shown in Table 4.1-4.
-88-
-------
TABLE 4.1-3, SYSTEM #2: ASH SLUICE MAKEUP REQUIREMENT
CO
Coal
1
2
3
4
5
6
Total
Sluice Requirement
m3/s (gpm)
0.19
0.13
0.23
0.24
0.12
0.12
(3100)
(2090)
(3600)
(38€0)
(1900)
(1950)
Cooling Tower
Slowdown
m3/s (gpm)
0.16
0.13
0.16
0.16
0.12
0.12
(2500)
(2050)
(2500)
(2500)
(1900)
(1950)
Raw Water
Makeup
m3/s (gpm)
0.038
0
0.063
0.082
0
0
(600)
0
(1000)
(1300)
0
0
-------
TABLE 4.1-4. SYSTEM #3: ASH SLUICE MAKEUP REQUIREMENT
Coal
1
2
3
4 -
5
6
Total
Sluice Water
Requirement
m'/s (gpm)
0.
0.
0.
0.
0.
0.
20
13
23
24
12
12
(3100)
(2050)
(3600)
(3800)
(1900)
(1950)
Makeup
Sluice Water
Requirement
m3/s (gpm)
0
0
0
0
0
0
.098
.063
.11
.12
.060
.061
(1550)
(1000)
(1800)
(1900)
( 950)
( 970)
Available
Cooling Tower
Bloudoun
m9/s (gpm)
0.057
0.057
0.057
0.057
0.057
0.057
(900)
(900)
(900)
(900)
(900)
(900)
Available
Makeup Water Gen. Serv.
Requirement Blowdown
m3/s (gpm) rna/s (gpm)
0.041
0.0082
0.057
0.063
0.0036
0.0046
(650) 0.035 (560)
(130)
(900) 0.035 (560)
(1000) 0.035 (560)
( 57)
( 73)
Raw Water
Requirement
ms/s (gpm)
-
0.021 (330)
0.027 (430)
o
I
-------
SYSTEM #4: It is assumed that cooling tower blowdown
is the source for ash sluicing makeup, as shown in Figure 4.1-4.
In this power plant water system, the cooling tower operates at
13.5 cycles of concentration and the ash sluicing system is total
recycle. Because the ash settles as a 40-50 percent solids
sludge, and assuming a 5 weight percent ash sluice slurry, total
recycle translates to 95 percent recycle of the water stream.
Table 4.1-5 shows the total sluice water requirement, makeup sluice
requirement, and available cooling tower blowdown for each coal.
It can be seen that no additional raw water makeup
requirement is necessary for any coal. The excess cooling
tower blowdown is available for use as general services water.
4.1.3 General Services Water System
The general services water system is defined to
include water conditioning, boiler and condenser cleaning,
boiler fireside and air preheater washing, the auxiliary cool-
ing system, and general power plant water use. Data is gener-
ally unavailable to define the water consumption for each of
these processes and operations. Therefore, to characterize
the water requirement for the general services water system,
plant data from a water recycle/reuse study (NO-106, CH-387,
GA-203) will be used. The data indicates that an assumption
of 95 cm3/s (1.5 gpin) general services water requirement
per megawatt will give a reasonably accurate number. Thus for
a 500 MW power plant, the general services water requirement
is 0.047 m3/s (750 gpm) . Data from the Georgia Power Company,
Plant Bowen (NO-106) indicate that approximately 75 percent
of the general services water could be available for use as
cooling tower makeup or ash sluicing makeup. This rate is
0.035 m3/s (560 gpm). Table 4.1-6 shows the general services
makeup water requirement for each of the systems discussed below:
-91-
-------
TABLE 4.1-5. SYSTEM #4; ASH SLUICE MAKEUP REQUIREMENT
Total Sluice Water
loal Requirement
m3/s (gpm)
1
2
3
4
5
6
0.20
0.13
0.23
0.24
0.12
0.12
(3100)
(2050)
(3600)
(3800)
(1900)
(1950)
Makeup Sluice Water
Requirement
m3/s (gpm)
0.0095
0.0063
0.011
0.012
0.0061
0.0061
(150)
(100)
(180)
(190)
( 96)
( 97)
Available Cooling
Tower Slowdown
m3/s (gpm)
0.019
0.019
0.019
0.019
0.019
0.019
(300)
(300)
(300)
(300)
(300)
(300)
VD
fO
-------
TABLE 4.1-6. GENERAL SERVICES MAKEUP WATER REQUIREMENT
System #1
System #2
System #3
System #4
Makeup Water
mVs
0.047
0.047
0.047
0.012
Requirement
(gpm)
750
750
750
190
Description
once-through
once-through
recirculated
recirculated
to ash sluicing
to cooling tower
SYSTEM #1: All flows are once-through with no
attempt being made to reuse any waters
SYSTEM #2: It is assumed that no attempt is made
to reuse this water.
SYSTEM #3:
SYSTEM #4:
The 0.035 m3/s (560 gpm) recoverable
general services water is required for
ash sluicing makeup. The general
services requirement is 0.047 m3/s
(750 gpm) and the recycle advantage
lowers the raw water requirement for
ash sluicing.
The recycle advantage is achieved by
combining the recovered general services
water with the cooling tower makeup.
4.1.4
Boiler Makeup Water Requirement
Slowdown is required to avoid excessive concentration
of impurities in the liquid phase in the boiler. The typical
blowdown rate for a drum-type steam boiler is 0.1% (AY-007) of
the steam generating rate. For a 500 Mw power plant operating
-93-
-------
at 37 percent efficiency, approximately 570 cm3/s (9 gpm)
makeup water is required.
4-- 2 Lime Wet Scrubbing Water Consumption
The lime wet scrubbing process requires fresh makeup
water for losses due to: 1) evaporation of water in the ab-
sorber, and 2) occlusion in the solid waste. Each model plant
system requires a separate calculation due to different flue
gas composition and mass rate. For purposes of discussion we
will assume one of four equivalent scrubbing trains for a 500
MW power plant burning a 3.5% S coal, with 28 MJ/kg (12,000
Btu/lb) average heating value. Assumptions for each other case
are discussed in detail in Appendix B.
4.2.1 Evaporation in the Absorber
The evaporative water loss in the absorber is due to
the quenching of hot flue gas by adiabatic saturation. Water
evaporated from the scrubbing liquor requires makeup. If there
is no sensible heat transfer to the scrubbing liquor, the follow-
ing heat balance equation applies:
MCpAT = MwX
where M = mass flow rate of the flue gas,
C = heat capacity of the flue gas,
AT = the change in temperature of the flue gas,
li^ = mass rate of water evaporated, and
A = heat of vaporization of water.
To perform the calculation, a final flue gas outlet
temperature is assumed, and M^ is calculated. A new gas
-94-
-------
composition is calculated, and compared to the saturated gas
composition at the assumed temperature in psychrometric charts.
A new outlet temperature is assumed and the calculation is
repeated until the gas compositions are equivalent. In the
discussion case, the outlet temperature is 52° C (125°F) and
0.0076 m3/s (120 gpm) of water are evaporated.
4.2.2 Occlusion in the Solid Waste
The lime wet scrubbing system disposes of the scrub-
bed sulfur as solid waste. Water is lost both as water of hy-
dration and water occluded in the 60 percent solids sludge
waste. In the discussion case, 90 percent removal of the sulfur
in the flue gas was assumed, as were 25 percent oxidation and
1.05 lime stoichiometry. (Oxidation can vary considerably, but
normally ranges from about 0 to 40% (DT-R-Q51). Twenty-five
percent was chosen as an average value. The assumption of lime
stoichiometry of 1.05 is based on Radian experience at the
Paddy's Run Station of Lewisville Gas & Electric.) The amount
of sulfur to be removed was calculated from known flue gas rate
and composition. Assuming that 25 percent of the CaSOs was
oxidized to CaSCU , the water of hydration was calculated. With
a lime stoichiometry of 1.05 the excess regenerant was calculated,
The final solids concentration in the sludge was assumed to be
60%. This solids level can be achieved through clarification
followed by ponding and settling, or by vacuum filtration and/or
centrifugation. In the discussion case, 1.5 kg/s (200 Ib/min)
of waste solids with 1.1 kg/s (150 Ib/min) of water of hydra-
tion and occluded water were produced. The water of hydration
produced was approximately 0.13 kg/s (17 Ib/min).
-95-
-------
4.3 Limestone Wet Scrubbing Water Consumption
The limestone wet scrubbing process requires fresh
makeup water for losses due to: 1) evaporation of water in the
absorber, and 2) occlusion in the solid waste. Each model plant
system requires a separate calculation due to different flue
gas composition and mass rate. For purposes of discussion we
will assume one of four equivalent scrubbing trains for a 500
MW power plant burning a 3.5% S coal, with 28 MJ/kg (12,000
Gtu/lb) average heating value. Assumptions for other cases are
discussed in detail in Appendix A.
4.3.1 Evaporation in the Absorber
The evaporative water loss in the absorber is due to
the quenching of hot flue gas by adiabatic saturation. Water
evaporated from the scrubbing liquor requires makeup. If there
is no sensible heat transfer to the scrubbing liquor, the
heat balance equation presented in Section 4.2.1 applies.
To perform the calculation, a final flue gas outlet
temperature is assumed, and the mass flow rate of water is
calculated. A new gas composition is calculated, and compared
to the saturated gas composition at the assumed temperature in
psychrometric charts. When a new outlet temperature is assumed,
the calculation is repeated until the gas compositions are
equivalent. In the discussion case, the outlet temperature is
53°C (128°F); 0.0076 m3/s (120 gpm) of water are evaporated per
train.
-96-
-------
4.3.2 Occlusion in the Solid Waste
The limestone wet scrubbing system disposes of the
scrubbed sulfur as solid waste. Water is lost both as water of
hydration and water occluded in the 60 percent solids sludge
waste. In this case, 90 percent removal of the sulfur in the
flue gas was assumed, as were 25 percent oxidation and 1.20
limestone stoichiometry. The amount of sulfur to be removed
was calculated from known gas rate and composition. Assuming
that 25 percent of the CaS03 was oxidized to CaSCK , the water
of hydration was calculated. The excess regenerant was calcu-
lated with a limestone stoichiometry of 1,20. A final solids
concentration of 60 percent was assumed in the sludge. This
concentration can be achieved through clarification followed
by ponding and settling, or by vacuum filtration and/or centri-
fugation. In the discussion case 1.7 kg/s (220 Ib/min) of
waste solids with 0.13 kg/s (17 Ib/min) water of hydration, and
1.1 kg/s (150 Ib/min) occluded water were produced. The total
water loss was 1.25 kg/s (165 Ib/min) per train.
4.4 Wellman-Lord Sulfite Scrubbing Process Water
Consumption
The Wellman-Lord process requires fresh makeup water
for losses associated with: 1) evaporation in the prescrubber,
2) sluicing of particulates, 3) drying of the purge solids,
4) water content in the SOz product stream, and 5) condenser
cooling water blowdown. The largest water loss is due to
evaporation in the prescrubber. Particulate slurry and conden-
ser cooling makeup requirements are an order of magnitude less.
Drying losses and water in the product stream are again another
order of magnitude less. Because calculation of each of these
losses is dependent on the model plant case, a separate calcu-
lation is required for each. For the purposes of discus-sion
-97-
-------
we will assume a base case of one of four equivalent scrubbing
trains for a 500 MW power plant burning a 3.57= S coal, with
28 MJ/kg (12,000 Btu/lb) average heating value. Assumptions
for each other case are discussed in detail in Appendix A.
4.4.1 Evaporation in the Scrubber
The Wellman-Lord process requires that all particu-
lates and chlorides are removed from the flue gases before
entering the absorber. This is commonly accomplished by pre-
scrubbing the gases in a venturi scrubber. The flue gases are
cooled from approximately 150°C (300°F) to 50° C (125°F) by
adiabatic saturation. The evaporative scrubber loss occurs in
the prescrubber rather than the absorber. It is assumed that
only water is evaporated by the hot flue gas. There is no
sensible heat transfer to the scrubbing liquor. Therefore,
the heat balance equation presented in Section 4.2.1 applies.
-To perform the calculation, a final flue gas outlet
temperature is assumed, and the mass flow rate of water is
calculated. A new gas composition is calculated, and compared
to the saturated gas composition at the assumed temperature in
psychrometric charts. When a new outlet temperature is assumed,
the calculation is repeated until the gas compositions are
equivalent. In the discussion case the outlet temperature is
approximately 53°C (128°F); 0.0076 m3/s (120 gpm) of water are
evaporated per train.
4.4.2 Particulate Sluicing Requirement
A blowdown stream is necessary in the prescrubbing
loop to maintain desired circulating suspended and dissolved
solids concentrations. A chlorine balance was performed to
-98-
-------
determine the required blowdown to maintain a dissolved solids
concentration of 20,000 mg/1 (ppm). The required blowdown to
maintain particulates at five weight percent was also calculated.
It was found that the desired level of suspended solids (particu-
lates) concentration was the blowdown limiting factor. For the
discussion case, 0.83 kg/s (110 Ib/min) of water were required
for makeup to replace blowdown water in each train.
4.4.3 Water Loss Association with Purge Solids Drying
From water balance calculations, 0.006 kg/s (11.4
Ib/min) of purge solids were delivered to the dryer. The centri-
fuge cake was assumed to contain 0.084 kg/s (11.1 Ib/min) of
water based on calculations performed by Radian for a compara-
tive FGD systems evaluation (OT-051). When the cake is dried,
100 percent of the water evaporates. When the dryer combustion
gases and the water vapor are routed to the prescrubber, the
water vapor exits the system with the flue gases. The hot
dryer gases evaporate additional water in the prescrubber. The
amount of this water loss is calculated on the basis of mass
rate of required combustion gases (MC-147) and adiabatic satura-
tion. The total water loss for the discussion case is 0.53 kg/s
(70 Ib/min) per train.
4.4.4 Water in SQa Product Stream
It is assumed that the product S02 stream contains
10 weight percent water (OT-051). The amount of S02 in the
product stream is calculated based on 90 percent removal of the
S02 in the absorber and 92 percent evolution of S02 from the
sulfur in the coal (MC-147). The water was calculated as 10
weight percent of the gas stream or 0.072 kg/s (9.5 Ib/min)
per train.
-99-
-------
4.4.5 Condenser Cooling Water Slowdown
In the regeneration loop, the absorber effluent is
sent to double effect evaporators. The overhead from the first
evaporator passes through a heat exchanger to provide heat for
the second effect. If it assumed that a condenser cooling water
is required to condense 50 percent of the overhead from both
effects, 0.155 m3/s (2450 gpm) of cooling water are required in
the discussion case. This assumes a 10°C (20°F) rise in cooling
water temperature in the condenser. A cooling tower operating
at five cycles of concentration would require 0.0035 m3/s (56
gpm) of makeup water, and 0.0007 m3/s (10.5 gpm) of wastewater
would be blowndown per train.
4.5 Magnesia Slurry Absorption Process Water Consumption
The magnesia slurry absorption process requires fresh
water makeup for losses due to: 1) evaporation of water in the
prescrubber, 2) sluicing of particulates, and 3) water losses
associated with drying the magnesium sulfite in the regeneration
loop. The largest water loss is due to evaporation in the
prescrubber. Both the sluicing requirements and the loss in
the dryers are approximately an order of magnitude smaller.
Because calculation of each of these losses is dependent on the
model plant case, a separate calculation is required for each.
For the purpose of discussion we will assume a base case of one
of four equivalent scrubbing trains for a 500 MW power plant
burning a 3.5 percent S coal, with 28 MJ/kg (12,000 Btu/lb)
average heating value. Assumptions for each other case are
discussed in detail in Appendix A.
-100-
-------
4.5.1 Evaporation in the Scrubber
The magnesia slurry absorption process requires that
all particulates and chlorides are removed from the flue gases
before entering the absorber. This is commonly accomplished
by prescrubbing the gases in a venturi scrubber. The flue
gases are cooled from approximately 150°C (300°F) to 50°C
(125°F) by adiabatic saturation. The evaporative scrubber loss
occurs in the prescrubber rather than the absorber. It is
assumed that only water is evaporated by the hot flue gas.
There is no sensible heat transfer to the scrubbing liquor.
Therefore, the heat balance equation presented in Section
4.2.1 applies.
To perform the calculation, a final flue gas outlet
temperature is assumed, and the mass flow rate of water is
calculated. A new gas composition is calculated, and compared
to the saturated gas composition at the- assumed temperature in
psychrometric charts. A new outlet temperature is assumed, and
the calculation is repeated until the gas compositions are
equivalent. In the discussion case the outlet temperature is
approximately 53°C (128°F); 0.0075 m3/s (120 gpm) of water are
evaporated per train.
4.5.2 Particulate Sluicing Requirement
A blowdown stream is necessary in the prescrubbing
loop to maintain desired circulating suspended and dissolved •
solids concentrations. A chlorine balance was performed to
determine the required blowdown to maintain a dissolved solids
concentration of 20,000 mg/1 (ppm). The blowdown required to
maintain particulates at five weight percent was also calculated.
It was found that the desired level of suspended solids
-101-
-------
(particulates) concentration was the blowdown limiting factor.
For the discussion case 0.83 kg/s (110 Ib/min) of water were
required for makeup to replace blowdown water per train.
4.5.3 Water Losses Associated With Drying
In the magnesia slurry absorption process, a bleed-
stream containing approximately 10 percent solids is passed
through screens for thickening to 40 percent solids. The
MgS03-6H20 is thermally converted to MgS03-3H20 and centrifuged
to 95 percent solids. For the discussion case 2.0 kg/sec
(260 Ib/min) of MgS03-3H20 and 0.11 kg/s (15 Ib/min) of
water are fed to the dryer. In the dryer, the water of hydra-
tion and free water evaporate. The water exits with the dryer
combustion gases and is sent to the stack. The water loss is
0.71 kg/sec (94 Ib/min) per train.
4.6 Double Alkali Wet Scrubbing Water Consumption
The double alkali wet scrubbing process requires
fresh makeup water for losses due to: 1) evaporation of water
in the absorber, and 2) occlusion in the solid waste. If a
prescrubber is required to remove chlorides, an additional
water loss would be incurred from the blowdown stream necessary
to maintain desired circulating concentrations of suspended
and dissolved solids in the prescrubber loop. The water loss
due to evaporation in the absorber is the largest water loss.
The water loss due to occlusion in the solid waste is approxi-
mately an order of magnitude less, as is the prescrubber
blowdown if required. Since calculation of each of these
losses is dependent on the model plant case, a separate calcu-
lation is required for each. For the purpose of discussion we
will assume a base case of one of four equivalent scrubbing
-102-
-------
trains for a 500 MW power plant burning a 3.5% S coal, with 28
MJ/kg (12,000 Btu/lb) average heating value. Assumptions for
each other case are discussed in detail in Appendix A.
4.6.1 Evaporation in the Absorber
For application to high chloride coals, the double
alkali wet scrubbing process may require that chlorides be
removed from the flue gases by a prescrubber before entering
the absorber. Under normal conditions, however, a prescrubber
will not be used and the evaporative loss occurs in the
absorber or in a presaturation chamber in the absorber. It is
assumed that only water is evaporated by the hot flue gas.
There is no sensible heat transfer to the scrubbing liquor.
Therefore, the heat balance equation presented in Section 4.2.1
applies.
To perform the calculation, final flue gas outlet
temperature is assumed, and the mass flow rate of water is
calculated. A new gas composition is calculated and compared
to the saturated gas composition at the assumed temperature in
psychrometric charts. A new outlet temperature is assumed, and
the calculation is repeated until the gas compositions are
equivalent. In the discussion case, the outlet temperature is
approximately 53°C (128°F) and 0.0076 m3/s (120 gpm) of water
are evaporated per train.
4.6.2 Occlusion in the Solid Waste
The double alkali wet scrubbing system disposes the
scrubbed sulfur as solid waste. Water is lost both as water of
-103-
-------
hydration and water occluded in the 60 percent solids sludge
waste. In the discussion case, 90 percent removal of the sulfur
in the flue gas was assumed as were 25 percent oxidation and
1.05 lime stoichiometry. The amount of sulfur to be removed
was calculated from known flue gas rate and composition.
Assuming that 25 percent of the CaS03 was oxidized to CaSCK ,
the water of hydration was calculated.
Excess lime and limestone were calculated for each
regenerant. A final solids concentration of 60 percent was
assumed in the sludge. In the discussion case, 1.5 kg/s
(200 Ib/min) of waste solids were produced with lime regenerant,
and 1.7 kg/s (220 Ib/min) of waste solids were produced with
limestone regenerant. The water losses were 1.1 kg/s (150
Ib/min) with lime regenerant and 1.25 kg/s (165 Ib/min) with
limestone regenerant per train.
4.7 Physical Coal Cleaning Water Consumption
It is inherently difficult to describe a generalized
coal cleaning process because a large number of processes and
operations can be used in common processing areas. It is even
more difficult to describe those processes optimized to remove
pyritic sulfur. Sulfur in this state is present in many
particle sizes and forms, which may be inherent or easily
removed. The processes to be used and the extent to which the
coal is crushed are dependent upon the amount of sulfur to be
removed and its form within the particular coal. Because of the
-104-
-------
wide variability in physical coal cleaning plant layouts and
ranges of operations, it is possible to characterize water con-
sumption only in a general sense. It is possible to attribute
water losses to certain specific causes such as drying opera-
tions and occlusion in solid wastes. However, these directly
attributable losses generally- indicate minimum losses. There-
fore, plant data presented in Coal Preparation (LE-218) were
used to characterize water makeup requirements for coal cleaning
plants. These data indicate that in 1962, water consumption in
coal preparation plants in the United States averaged 3.3 percent
of the circulating water flow rate. New plants are designed to
operate closed loop (LE-218). It is impossible to ascertain
the extent to which improvements in water consumption may have
been made. The water consumption varies from approximately 1.5
percent to 27 percent in the values for various states, with
most values being between 1.5 percent and 5.0 percent. Water
balance calculations have shown that minimum losses due to
occlusion in waste coal and thermal drying or increased coal
moisture content are in the range of 1-2 percent of the circu-
lating flow rate (Appendix A). Thus, if it is assumed that
modern coal preparation plants operate with water consumption
averaging 3.3 percent of the circulating flow, this may be
reasonable in regard to realistic average water management prac-
tices. With this assumption, the water makeup requirement for
the base case 500 MW power plant burning coal that was initially
3.5 percent sulfur coal with an average heating value of 28
MJ/kg (12,000 Btu/lb) is 15 kg/sec (2000 Ib/min).
-105-
-------
4.8 SO a. Conversion Processes Water Consumption
Two of the five FGD systems in our study produce a
concentrated S02 product stream, thus requiring either lique-
faction of the S02, or conversion into elemental sulfur or
sulfuric acid. Both product S02 streams can be converted to
either elemental sulfur or sulfuric acid. However, for tabu-
lation of water requirements it will be assumed that the magne-
sium slurry product S02 stream will be converted to sulfuric
acid, while the Wellman-Lord product stream will be converted
to elemental sulfur (as described in Detailed Cost Estimates of
Advanced Effluent Desulfurization Processes (MC-147)).
4.8.1 Sulfuric Acid Production
Makeup water is required in sulfuric acid production
for: 1) stoichiometric requirement, and 2) cooling water blow-
down. The S02 stream is oxidized to S02 over a vanadium
pentoxide catalyst. The S02 combines with water in an absorber
to form the product acid. The acid is cooled and sent to
storage. The stoichiometric requirement is calculated on the
basis of one mole of water per mole of S02 sent to conversion.
Thus 0.75 kg/s (99 Ib/min) of water are required for the
total 500 MW discussion case conversion S02 stream. McGlammery,
et al (MC-147) state that the sulfuric acid cooling water
requirement is 0.30 m3/s (4750 gal/min) for the discussion case
500 MW power plant. With a cooling tower operating at five
cycles of concentration, 0.0069 m3/s (110 gal/min) makeup water
are required and 0.00013 m3/s (20.5 gal/min) of wastewater are
blown down.
-106-
-------
4.8.2 Elemental Sulfur Production
Although several water systems exist in an elemental
sulfur production unit, the water requirement is insignificant.
Steam is produced in the sulfur condensers at a rate of approxi-
mately 1.7 kg/s (225 Ib/min) (MC-147) for the compressor
seals. The blowdown from these closed loop operations is insig-
nificant in comparison to the other FGD water requirements.
4.9 Model Systems Makeup Water Requirement
The NSPS for SOa emissions from coal-fired steam
generating plants is currently under review by OAQPS. This
review is considering the comprehensive impacts of the existing
NSPS and two alternative revised standards. The existing NSPS
allows an emission rate of 0.52 yg S02/J (1.2 Ib S02/MM Btu) of
heat input. One alternative standard requires 0.215 yg S02/J
(0.5 Ib S02/MM Btu) of heat input. This standard has the same
form as the existing NSPS and thus allows a credit for physical
coal cleaning or use of low sulfur coal. The second alternative
standard requires 90 percent removal of SOa from stack gases,
regardless of original sulfur content in the coal.
In order to assess all the various impacts so that a
f
comprehensive conclusion may be drawn, 108 model plant systems
have been chosen by the OAQPS as a common base for evaluations.
These systems are listed in Table 4.9-1. Generally, these cases
will allow analysis of the impacts of the three standards. The
type of FGD system, sulfur content of coal, size of steam
generator, and degree of coal cleaning will be the variables
examined.
-107-
-------
"The first alternative standard deals with the
existing NSPS and serves as a baseline for comparing
impacts for plants ranging in size from 25 to 1000 mega-
watts when burning coals with average sulfur contents
of 3.5 and 7 percent. It also provides a basis for com-
paring the impact of a revised standard on eastern and
western plants burning typical low sulfur western coal
and on a plant using flue gas desulfurization (FGD) in
conjunction with coal washing. The second alternative
standard represents a 90 percent removal of S02 by FGD
on all plants regardless of the sulfur content of the
coal burned. It does not give any credit for coal
cleaning. The third alternative standard has the same
format as the existing NSPS; that is, it is based on a
fixed mass emission rate. Therefore, plants can use
combinations of FGD and coal cleaning, and total removal
efficiency will vary depending on the sulfur content
of the coal. The third alternative standard is similar
in stringency to the second alternative standard in that
emissions of 0.4 to 0.5 Ib of S02/mm Btu represent about
90 percent S02 removal efficiency on a typical 3.5 per-
cent sulfur coal.
To limit the scope of work but still provide compara-
tive information, the five viable FGD systems (lime,
limestone, magnesium oxide, double alkali, and Wellman-
Lord) are considered only in Cases l(a) and 2(a). This
is recommended so that the variations in impacts of the
alternative FGD systems will be shown and to provide basic
information needed to answer questions which will cer-
tainly arise during revision of the NSPS. For example,
use of a limestone FGD system results in a sludge dis-
posal problem while use of the magnesium oxide FGD does
not result in a sludge. The remainder of the analysis
for the recommended alternative is limited to the lime/
limestone systems which are the predominant systems used
by the domestic industry and which, due to cost, will
probably continue to be the first choice of most of the
domestic industry for the near future." (CU-077).
4-9.1 Base Uncontrolled Power Plant Water Requirements
for the Model Plant Systems
To characterize the range of current water consumption
patterns, Radian has chosen to characterize four model power
plant water management systems. The four systems are:
-108-
-------
TABLE 4.9-1. EPA/OAQPS ALTERNATIVE CONTROL SYSTEMS FOR MODEL PLANTS
Plant sizes to
be considered, MW
FGD Systems
to be considered
Alternative Standards and Model Plant Systems
25; 100; 500; 1000
25; 100; 500; 1000
25; 500
25; 500
25; 100; 500; 1000
25; 100; 500
25; 500
25; 500
Lime/limestone
Lime/limestone
Lime/limestone
Lime/limestone
1. The existing NSPS of 0.52 pg S02/J (1.2 Ib S02/MM
Btu) heat input.
a. ^80 percent S02 removal on a plant burning a
typical coal of 3.5 percent sulfur.
b. A plant burning a typical 7 percent sulfur coal
with about 90 percent S02 removal by FGD.
c. High and low heating value western and eastern
low sulfur coals without FGD for a typical
eastern plant.
d. High and low heating value western low sulfur
coals without FGD for a typical western plant.
2. a. 90 percent SO2 removal by FGD on a typical coal
of 3.5 percent sulfur and a typical coal of
7 percent sulfur.
b. 90 percent S02 removal by FGD on a plant burning
typical high and low heating value western coals
of 0.8 percent S (western plant).
3. 0.215 ]jg S02/J (0.5 Ib S02 emission/MM Btu) heat input.
a. 70 to 75 percent S02 removal on a plant burning
typical high and low heating value western coals
of 0.8 percent S (western plant).
b.l 40 percent sulfur removal by coal washing of
a 3.5 percent sulfur coal and 85 percent removal
by FGD.
b.2 40 percent sulfur removal by coal washing of a
7 percent sulfur coal and 95 percent removal by FGD.
aThe five systems to be considered are lime, limestone, magnesium oxide, double alkali, and Wellman-Lord,
-------
System #1: All power plant water systems are
once-through.
System #2: Recirculatory cooling at 2.5 cycles
of concentration of once-through ash
handling, and once-through general
services water.
System #3: Recirculatory cooling at 5.0 cycles of
concentration, 50 percent recirculatory
ash handling, and recycle of general
service water blowdown to the ash
handling system.
System #4: All power plant water systems are
recirculatory.
Each of the process water requirements and methods of
their calculation have been previously discussed in this sec-
tion. Table 4.9-2 summarizes the results of the calculations
for Systems #1 through #4 for the discussion case 500 MW power
plant. In the once-through system, since the cooling water
requirement dominates, FGD water requirements are insignifi-
cant by comparison. In recirculatory systems operating at 2.5,
5.0, and 13.5 cycles of concentration, ash handling, general
services, and FGD water requirements become significant. Sys-
tem #3 has been chosen as representative of "typical" power
plant water requirements for comparison to FGD requirements.
This system may have lower water requirements than power plant
systems in current operation, but it will become more predomi-
nant as the national zero discharge goal nears. This system
characterizes water requirements in the midrange of the three
recirculatory systems. Model systems base power plant water
requirements calculated for System #3 are shown in Table 4.9-4.
-110-
-------
TABLE 4.9-2. BASE CASE: MODEL POWER PLANT WATER CONSUMPTION
System System3
Number Description
1 Once-through
2 Partial Recirculatoryc
3 Recirculatory
4 Zero Discharge
Cooling Water Ash Handling
System System
md/s (gpm) m3/s (gpm)
13 (210,000) 0.23 (3600)
0.44 (7,000) 0.069 (1100)
0.32 (5,000) 0.021 (330)
0.25 (4,000) 0 (0)
General Services
Wuter
m3/s (gpm)
0.047 (750)
0.047 (750)
0.047 (750)
0.012 (190)
Boiler
Makeup
IBS/S (gpm)
0.0006 (9)
0.0006 (9)
0.0006 (9)
0.0006 (9)
Total
Makeup
,«3/s
13.5
0.56
0.38
0.26
Fresh Water
Requirement
(gpin)
(215,000)
(8,850)
(6,100)
(4,200)
The base case is a 500 MH power plant operating at an efficiency of 37%; 3.5% S coal; average heating value of 28 MJ/kg (12,000 Btu/lb).
All power plant water systems are once-through; refer to Figure 4.1-1.
Q_
Recirculatory cooling at 2.5 cycles of concentration, once-through ash handling, and once-through general services water; refer to Figure 4.1-2.
Recirculatory cooling at 5.0 cycles of concentration, 50% recirculatory ash handling, and recycle of general service water bloudown to the
ash handling system; refer to Figure 4.1-3.
All pouer plant water systems are recirculatory; refer to Figure 4.1-4.
-------
4.9.2 SO Control Strategy Water Requirements for the
X
Model Plant Systems
Each of the SOX control strategy water requirements
and methods of calculation have been discussed in previous
sections. Table 4.9-3 summarizes the results of each of these
calculations for the base case 500 MW power plant. This table
illustrates that the method used to calculate the total FGD
process water requirement was to define and calculate specific
requirements with their sum being the total requirement. The
most significant requirement is evaporative loss in the scrub-
ber. Cooling water requirements for the Wellman-Lord Sulfite
Scrubbing Process and SC-2 conversion to sulfuric acid are also
significant. Other significant requirements are approximately
an order of magnitude less.
4-9.3 Matrix Presentation of Model Plant Water Requirements
Table 4.9-4 presents the results of water require-
ment calculations for each model plant. A detailed discussion
of the methods and assumptions used in calculating these re-
quirements are given in Appendix A.
-112-
-------
TABLE 4.9-3. BASE CASE: FGD SYSTEM WATER CONSUMPTION1
Water3
Requirement
Evaporative loss,
Loss in solid waste,
Prescrubber blowdown.
Cooling water blowdown.
Loss with solids drying,
Loss in product S02 stream,
S02 Conversion Requirement:
Sulfuric acid
Elemental Sulfur
Total
Liquid to gas ratio required
for scrubber - m3/Nm* (gal/1000
Lime Wet
Scrubbing
ma/s (gpm)
0.030 <480)
0.005 (80)
0.035 (560)
.005-. 015 (35-
scf) 110)
Limestone Wet
Scrubbing
mj/s (gpm)
0.030 (480)
0.005 (80)
0.035 (560>
.005-. 015 (35-
110)
Wellmi
Sulflte
m'/s
0.030
0.003
0.014
0.002
0.003
0.008
0.002
0.058
0.050
0.004
.002
in-Lord
Scrubbing
0.036 (580)e
003-.004 (20- .
30)
.002 (15)B
Double Alkali
Lime
Kegenerant
m'/s (gpm)
0.030 (480)
0.005 (72)
d 0.035 (552)°
0007-.OQ2 (5-15)
Wet Scrubbing
Limestone
Regenerant
m'/s (gP««)
0.030 (480)
0.005 (79)
0.035 (559)C
0.007- (5-15)
.002
3 The base case is for a 500 MH power plant operating at an efficiency of 37%, burning 3.5JE burning 3.5% sulfur coal with an average heating
value of 28 MJ/kg (12,000 Btu/lb).
This value is assumed for model plant calculations.
c The average of the two double alkali systems is used for model plant calculations.
This is the total water requirement if sulfuric acid is produced.
e This is the total water requirement if sulfur is produced.
^ Separate scrubbing loops are provided for each of 3 trays.
*= Prescrubber.
Make-up water requirement.
-------
TABLE 4.9-4. MODEL PLANT SYSTEM WATER REQUIREMENTS
-p-
i
Case
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
Power
Plaut
Capacity
MU
1000
1000
1000
1000
1000
500
500
500
500
500
100
100
100
100
100
25
25
25
25
25
1000
1000
500
500
100
100
25
25
so*
Control
Strategy
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Lime
Limestone
Lime
Limestone
Lime
Limestone
Z Sulfur
Removal
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
76
88
sa
68
88
88
88
88
88
Coal a
Type
03
03
03
113
#3
ill
#3
03
03
03
#3
03
#3
113
03
#3
03
#3
03
03
04
04
04
04
04
04
04
04
System 03
Power Plant
Makeup Water
Requirement
m'Vs
0.76
0.76
0.76
0.76
0.76
0.38
0.38
0.38
0.38
0.38
0.076
0.076
0.076
0.076
0.076
0.019
0.019
0.019
0.019
0.019
0.76
0.76
0.39
0.39
0.76
0.76
0.020
0.020
(gp»)
(12,000)
(12.000)
(12,000)
(12,000)
(12,000)
(6,100)
(6,100)
(6,100)
(6,100)
(6,100)
(1,200)
(1,200)
(1,200)
(1,200)
(1,200)
(300)
(300)
(300)
(300)
(300)
(12,000)
(12,000)
(6,200)
(6,200)
(1.200)
(1,200)
(310)
(310)
SO Control
Strategy
Makeup Water
Requirement
m'/s
0.069
0.069
0.095
0.082
0.069
0.034
0.034
0.048
0.042
0.034
0.0069
0.0069
0.0095
0.0088
0.0069
0.0017
0.0017
0.0024
0.0023
0.0017
0.079
0.082
0.039
0.040
0.0082
0.0082
0.0020
0.0020
(gpm)
(1.100)
(1,100)
(1,500)
(1,300)
(1.100)
(540)
(540)
(760)
(670)
(540)
(110)
(110)
(150)
(140)
(110)
(27)
(27)
(38)
(37)
(27)
(1,250)
(1,300)
(620)
(640)
(130)
(130)
(31)
(32)
Total
Model System
Makeup Water
Requirement
m3/a
0.82
0.82
0.88
0.82
0.82
0.42
0.42
0.43
0.43
0.42
0.082
0.082
0.088
0.088
0.082
0.021
0.021
0.021
0.021
0.021
0.82
0.82
0.43
0.43
0.082
0.082
0.021
0.021
(gpm)
(13,000)
(13.000)
(14,000)
(13.000)
(13,000)
(6,600)
(6,600)
(6,800)
(6,800)
(6,600)
(1,300)
(1,300)
(1,400)
(1,400)
(1,300)
(330)
(330)
(340)
(340)
(330)
(13,000)
(13,000)
(6,800)
(6,800)
(1,300)
(1,300)
(340)
(340)
(Continued)
-------
TABLE 4.9-4. MODEL PLANT SYSTEM WATER REQUIREMENTS (Continued)
Case
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
Power
Plant
Capacity
m
500
500
500
25
25
25
500
500
25
25
500
500
1000
1000
1000
1000
1000
500
500
500
500
500
100
100
100
100
100
25
SOX
Control % Sulfur
Strategy Removal
_
-
-
-
-
-
-
-
-
-
Coal Cleaning/Lime
Coal Cleaning/Limestone
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Line
Limestone
Wellnmn-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellman-Lord
Magnesium Oxide
Double Alkali
Lime
N/A
N/A
N/A
H/A
N/A
H/A
N/A
N/A
N/A
N/A
40/39
40/39
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
Coal a
Type
#1
n
#3
tl
112
03
01
112
»i
#2
*5
US
n
03
03
03
03
*3
03
*3
#3
03
03
03
03
03
03
03
System 03
Power Plant
Makeup Water
Requirement
mVs
0.37
0.37
0.38
0.018
0.018
0.020
0.37
0.37
0.018
0.018
0.37
0.37
0.76
0.76
0.76
0.76
0.76
0.38
0.38
0.38
0.38
0.38
0.076
0.076
0.076
0.076
0.076
0.020
(gpm)
(5.800)
(5,800)
(6,100)
(290)
(290)
(310)
(5,800)
(5,800)
(290)
(290)
(5,800)
(5,800)
(12,000)
(12,000)
(12,000)
(12,000)
(12,000)
(6,100)
(6,100)
(6,100)
(6,100)
(6,100)
(1,200)
(1,200)
(1,200)
(1,200)
(1,200)
(310)
SOX Control
Strategy
Makeup Water
Requirement
mVs
0.049
0.049
0.070
0.070
0.095
0.088
0.070
0.035
0.035
0.049
0.044
0.035
0.0069
0.0069
0.0095
0.0088
0.0069
0.0017
(gpm)
N/A
N/A
N/A
N/A
N/A
N/A
N/A
N/A
H/A
N/A
(770)
(770)
(1,100)
(1,100)
(1,500)
(1.400)
(1,100)
(550)
(550)
(770)
(700)
(550)
(110)
(110)
(150)
(140)
(110)
(27)
Total
Model System
Makeup Water
Requirement
m'/s
0.37
0.37
0.38
0.018
0.018
0.020
0.37
0.37
0.018
0.018
0.41
' 0.41
0.82
0.82
0.88
0.82
0.82
0.42
0.42
0.44
0.43
0.42
0.082
0.082
0.088
0.082
0.082
0.021
(gpm)
(5,800)
(5,800)
(6,100)
(290)
(290)
(310)
(5,800)
(5,800)
(290)
(290)
(6,500)
(6,500)
(13,000)
(13,000)
( 14 , 000)
(13,000)
(13,000)
(6,600)
(6,600)
(6,900)
(6,800)
(6,600)
(1,300)
(1,300)
(1,400)
(1,300)
(1,300)
(340)
(Continued)
-------
TABLE 4.9-4. MODEL PLANT SYSTEM WATER REQUIREMENTS (Continued)
Case
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
Power
Plant
Capacity
MM
25
25
25
25
1000
1000
1000
1000
1000
500
500
500
500
500
100
100
100
100
100
25
25
25
25
25
500
500
500
500
30x
Control
Strategy
Limestone
Mellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellnjan-Lord
Magnesium Oxicje
Double Alkali
Lime
Limestone
Hellman-Lord
Magnesium Oxide
Double Alkali
Lime
Limestone
Wellraan-Lord
Magnesium Oxide
{louble Alkali
Line
Limestone
Uelluian-Lord
Magnesium Oxide
Double Alkali
Lime
Lime
Limestone
Limestone
X. Sulfur
Removal
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
90
Coal a
Type
#3
43
03
03
IH:
04
04
04
D4
04
04
04
04
#4
44
04
04
04
04
04
04
04
114
04
III
02
ii
02
System S3
Power Plant
Makeup Mater
Requirement
m'/s
0.020
0.020
0.020
0.020
0.76
0.76
0.76
0.76
0.76
0.39
0.39
0.39
0.39
0.39
0.076
0.076
0.076
0.076
0.076
0.020
0.020
0.020
0.020
0.020
0.37
0.37
0.37
0.37
(BIM)
(310)
<310)
(310)
(310)
(12,000)
(12,000)
(12,000)
(12,000)
(12,000)
(6.20U)
(6,200)
(6,200)
(6,200)
(6,200)
(1,200)
(1,200)
(1,200)
(1,200)
(1,200)
(310)
(310)
(310)
(310)
(310)
(5,800)
(5,800)
(5,800)
(5, BOO)
SOX Control
Strategy
Makeup Hater
Requirement
ma/s
0.001B
0.0024
0.0022
0.0017
0.082
0.082
0.11
0.11
0.82
0.040
0.040
0.056
0.054
0.040
0.0082
0.0082
0.011
0.011
0.0082
0.0022
0.0023
0.0031
0.0029
0.0022
0.035
0.034
0.035
0.034
(SP»i)
(2B)
(38)
(35)
(27)
(1.300)
(1,300)
(1.800)
(1,700)
(1,300)
(630)
(640)
(890)
(850)
(630)
(130)
(130)
(180)
(170)
(130)
(35)
(36)
(49)
(46)
(35)
(550)
(540)
(550)
(540)
Total
Model Syatem
Makeup Uater
Requirement
m'/a
0.021
0.022
0.021
0.021
0.82
0.82
0.88
0.88
0.82
0.43
0.43
0.45
0.44
0.43
0.082
0.082
0.088
0.088
0.082
0.022
0.022
0.023
0.023
0.022
0.40
0.40
0.40
0.40
(gpm)
(340)
(350)
(340)
(340)
(13,000)
(13,000)
(14,000)
(14,000)
(13.000)
(6,800)
(6,800)
(7,100)
(7,000)
(6,800)
(1,300)
( 1 , 300)
(1,400)
(1,400)
(1,300)
(350)
(350)
(360)
(360)
(350)
(6,300)
(6,300)
(6,300)
(6,300)
(ConciuueJ)
-------
TABLE 4.9-4. MODEL PLANT SYSTEM WATER REQUIREMENTS (Continued)
Power
Plant
Case Capacity
HU
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
"coal H
coal 02
coal #3
coal /ft
coal #5
coal lit
100
100
100
100
25
25
25
25
500
500
500
500
25
25
25
25
500
500
25
25
500
500
25
25
0.8% S;
0.8% S;
3.5% S;
7.0% S;
2.0% S;
4.0% S;
S°K
Control %
Strategy
Liine
Lime
Limestone
Limestone
Lime
Lime
Limestone
Limestone
Lime
Lime
Limestone
Limestone
Lime
Lime
Limestone
Limestone
Coal Cleaning/Lime
Coal Cleaning/Limestone
Coal Cleaning/Lime
Coal Cleaning/Limestone
Coal Cleaning/Lime
Coal Cleaning/Limestone
Coal Cleaning/Lime
Coal Cleaning/Limestone
Sulfur
Removal
90
90
90
90
90
90
90
90
70
70
70
70
70
70
70
70
40/85
40/85
40/85
40/85
40/91
40/91
40/91
40/91
19 MJ/kg (8,000 Btu/lb); 6% ash; 30%
26 MJ/kg (11,000 Btu/lb); 6%
28 MJ/fcg (12,000 Btu/lb); 12%
28 MJ/kg (12,000 Btu/lb); 12%
26 MJ/kg (11,000 Utu/lb); 6%
27 HJ/kg (11,500 Btu/lb); 6%
ash; 15%
ash; 2.
ash; 5.
ash; 15%
ash; 15%
Coal a
Typo
01
#2
01
02
#1
112
til
• 112
til
112
ill
It 2
III
U2
III
02
*5
05
05
05
06
06
06
lib
H20
H20
6% H20
7% H20
li.,0
H20
Syst
Powe
Make
Keou
«?/s '-
0.073
0.073
0.073
0.073
0.018
0.018
0.0)8
o.oia
0.37
0.37
0.37
0.37
0.018
0.018
0.018
0 . 018
0.37
0.37
0.018
0.018
0.37
0.37
0.018
0.018
em tf 3
r I'lmit
i|> Water
i cement
(ispm)
(1,150)
(1,150)
(1,150)
(1,150)
(290)
(290)
(290)
(290)
(5,800)
(5,800)
(5,800)
(5,800)
(290)
(290)
(290)
(290)
(5.800)
(5,800)
(290)
(290)
(5,800)
(5,800)
(290)
<290)
SOX Control
Strategy
Makeup Water
Requirement
II? /S
0.0069
0.0069
0.0069
0 . 0069
0.0017
0.0017
0.0018
0.0017
0.034
0.031
0.034
0.032
0.0017
0.0015
0.0017
0.0016
0.050
0.050
0.0025
0.0025
0.052
0.052
0.0026
0.0026
(8P«0
(110)
(110)
(110)
(110)
(27)
(27)
(28)
(27)
(540)
(490)
(54Q)
(500)
(27)
(24)
(27)
(25)
(790)
(790)
(39)
(40)
(820)
(830)
(41)
(41)
Total
Model System
Makeup Water
Requirement
m*t*
0.082
0.082
0.082
0.082
0.020
0.020
0.020
0.020
0.40
0.40
0.40
0.40
0.020
0.020
0.020
0.020
0.42
0.42
0.021
0.021
0.42
0.42
0.021
0.021
(iS PII>)
( 1 , 300)
(1,300)
(1,300)
(1,300)
(320)
(320)
(320)
(320)
(6,300)
(6,300)
(6,300)
(6,300)
(320)
(310)
(320)
(320)
(6,600)
(6,600)
(330)
(330)
(6,600)
(6,600)
(330)
(330)
-------
5.0 CHARACTERIZATION OF PROCESS WASTEWATERS
This section assesses the impact of S0x control
technologies on power plant wastewater streams. Wastewaters of
power plants uncontrolled for S0x emissions are characterized.
In addition, wastewaters from the five subject FGD systems, the
physical coal cleaning process, and two S02 conversion processes
are characterized. Because water management at steam/electric
utilities is highly site-specific, it is necessary to character-
ize the wastewater streams in a general manner. A quantitative
characterization of typical power plant wastewater streams is not
possible. Wherever possible, characteristic compositions and
quantities of effluent streams are given to indicate typical
ranges of operation. Impact is assessed on the basis of a gen-
eral comparison of effluents. Fortunately, this comparison is
easily made, due to the significant impact of uncontrolled power
plant wastewaters on the surrounding water quality. The minor
nature of the wastewater streams associated with SO control
X
technologies is also a contributing factor.
Wastewater streams are characterized for the following
processes:
• Coal-fired Power Plant, Uncontrolled for SO
Emissions x
Lime Wet Scrubbing
Limestone Wet Scrubbing
Wellman-Lord Process
Magnesia Slurry Absorption Process
Double Alkali Wet Scrubbing
Physical Coal Cleaning
-118-
-------
• Allied Chemical SOx Reduction Process
• Single Absorption Contact Sulfuric Acid Plant
5.1 Characterization of Wastewaters from a Power Plant
Not Equipped with an FGD System
Wastewater effluents are discharged from many sources
in steam/electric power plants. Because of the wide variability
in individual plant layouts, water intake quality, selection and
operation of process units, and water management strategies,
however, not all plants have the same effluents. Discharges
from similar sources at different plants may also have highly
dissimilar characteristics in terms of stream flow and composi-
tion. This is in part attributable to wide variations in raw
feed water quality, ash composition, coal composition, quantity
of ash, and selected slate of boiler water and cooling tower
additives. There are also several options for handling wastes
such as recycle, combining with other waste streams, slip
stream or full stream treatment, and discharge of once-through
streams. The last option is becoming less practicable, however,
due to promulgation of zero discharge standards.
The main point that must be considered in addressing
water treatment technologies is that water management at steam/
electric utilities is highly site-specific. Attempts to general-
ize about aqueous discharges and their treatment should be care-
fully qualified since a single description of any given effluent
is not generally applicable to all power plants. In this section
a description of all major utility effluents will be presented.
5.1.1 Power Plant Wastewater Sources
Figure 5.1-1 graphically summarizes the sources of
wastewater in a fossil fueled generating station and depicts
the interrelation of the various processes producing wastewater.
-119-
-------
10 Af MO
to
o
I
.
ciUu.cM.-s Fteiooic.c.uAUiiJ$
Reference: BU-087
Figure 5.1-1- Sources of wastewater in a fossil-fueled steam-electric plant
-------
The basic sources of wastewater are:
cooling water system,
ash handling,
water conditioning,
boiler blowdown,
coal pile runoff,
equipment cleaning,
general plant drainage,
• process spills and leaks, and
miscellaneous sources.
Flue gas desulfurization wastes will be discussed in the follow-
ing sections.
The frequency of waste discharge can be any of the
following:
Continuous - Wastewater is discharged at a fairly
constant rate without interruption as long as the
plant is operating.
Intermittent - Wastewater is discharged on a regu-
lar or scheduled basis (e.g., every shift, daily)
when the plant is operational.
Periodic - Wastewater is discharged at infrequent
intervals (monthly or yearly) which may or may not
be of regular frequency.
Significant water parameters which can be used to
characterize wastewater streams from fossil-fueled power plants
include:
-121-
-------
Suspended Solids - undissolved solid matter
which is present in water,
Dissolved Solids - dissolved matter which is
present aqueous solution,
gH - a measure of the acidity of alkalinity of
a waste stream,
Hardness - the soluble calcium and magnesium
content of a waste stream,
Toxic Chemical Species - aqueous constituents
such as heavy metals, certain organic compounds,
etc., that are generally present in trace
concentrations but which are highly detrimental
to receiving bodies of water,
Oily Wastes - the insoluble organics in water
such as oils, greases, etc.
Biochemical Oxygen Demand - the oxygen required
to stabilize soluble and/or insoluble impurities
in water by biochemical reaction, and
Chemical Oxygen Demand - the oxygen required
to convert water impurities to their
oxidized forms through the action of oxidizing
agents.
-122-
-------
Identification of trace toxic chemicals in utility
effluents is currently underway. Preliminary results of a
literature review, utility survey, and preliminary screening
sampling study indicate that approximately 31 of the
126 "unambiguous" toxic substances may be found in utility
streams. An in-depth sampling program is now in progress
under EPA sponsorship to further define these streams.
The possible sources of these pollutants may be any
of the following:
Cooling System Treatment Condensate Neutralizer
Corrosion and Scale Inhibitor Cleaning Product
Corrosion Inhibitor Ash Constituent
Biocide - Cooling System Construction Material
Algacide Cooling Tower Material
Insulation Lab Reagent
Maintenance Material Instrument Use
Transformer Fluid
In the following discussion, each of the wastewater
sources potentially present in utilities is described. To the
greatest extent possible the composition and flow rate of the
streams will be characterized.
Cooling Water System
Approximately forty-five percent of a fossil-fuel
fired generating station's energy is removed and ultimately
discharged to the environment by the condenser cooling system.
Basically, two condenser cooling systems are employed by the
electric utility industry: 1) once-through system and
2) recirculating system.
-123-
-------
Once-Through Cooling Water
Once-through cooling systems are unique since the
total cooling water flow for heat removal is discharged as a
wastewater effluent. This cooling water flow rate is approxi-
mately 100 cm3/Kcal of heat removal for every 10°C of cooling
water temperature rise (12 gal/1000 Btu of heat removal for
every 10°F of cooling water temperature rise). After passing
through the condenser, the cooling water is discharged to a
heat sink (i.e., river, lake, pond) where the heat is dis-
sipated.
Due to the nature of the once-through system, the
chemical composition of the effluent water is essentially equiv-
alent to that of the influent water. Water quality parameters
such as total dissolved and suspended solids, pH, etc., are
largely governed by the characteristics of the cooling water
source, and they are not significantly altered by the operation
of the cooling system. Slight changes in the chemical composi-
tion between influent and effluent for these systems may occur,
however, due to 1) formation of corrosion products and/or 2)
addition of treatment chemicals such as chlorine.
Water-side corrosion of the main condenser will
result in corrosion products (i.e., metal oxides) appearing
in the cooling water effluent. Condenser metallurgy is care-
fully selected, however, to minimize water-side corrosion rates
to the extent that negligible quantities of corrosion products
appear in the effluent cooling water (EN-127).
Extensive cooling water treatment is normally pre-
cluded in once-through systems due to the large quantities
of raw water used. However, chemical treatment with biocides
-124-
-------
is often necessary to control biological growths of algae and
slime that accumulate on condenser surfaces, retard heat trans-
fer, and obstruct cooling water flow. Chlorine is by far the
most common biocide used; however, on rare occasions sodium
hypochlorite is used instead.
Cooling water is chlorinated by "shock" or "slug"
treatment methods in which a large dose of chlorine is added
intermittently. The duration of "shock" treatment ranges from
five minutes to two hours, but typically lasts 30 minutes. The
frequency of chlorine treatment ranges from one to ten times per
day (typically once per shift or three times per day) (MA-230) .
During chlorination, free residual chlorine is kept between 0.1
and 0.2 mg/JL (ppm) in the condenser effluent. When using seawater
as a coolant, chlorine residuals as high as 12 mg/£ (ppm) may be
used to deter the presence of eels and jelly fish as well as to
inhibit the mussel and crustacean growths on the condenser (MA-230)
Recirculating System: Cooling Tower Slowdown
Recirculating cooling systems employ cooling devices
such as cooling towers, spray ponds/canals, etc., which allow
the reuse of cooling water. These devices promote cooling pri-
marily by evaporating a portion of the recirculating water
flow. Impurities and contaminants that come into the system
with makeup water and other sources become concentrated. A
blowdown stream is withdrawn from the system to control the
concentration of impurities and contaminants. This stream
represents the recirculating cooling system wastewater. Blow-
down quantity is set by the maximum concentration of a limiting
impurity (i.e., hardness, dissolved solids, suspended solids)
that can be tolerated in the system or by the solubility limit
of scaling salts such as calcium sulfate, calcium carbonate, etc.
-125-
-------
(Example limits are 5 weight percent suspended solids and 20,000
mg/£ (ppm) dissolved solids.) The blowdown rate typically ranges
between 0.5 and 3.0% of the recirculating water flow (DO-051).
The recirculating flow is approximately 100 cm3/Kcal of heat
removal for every 10°C of cooling water temperature rise (12 gal/
1000 Btu for 10°F).
The blowdown from recirculating cooling systems has
the same chemical composition as does the recirculating cooling
water. The major factors that influence cooling water composi-
tion include:
Makeup water characteristics
Chemical treatment of the recirculating
cooling water
Intimate contacting of air-water in the
cooling device.
Makeup water to recirculating cooling systems replen-
ishes water loss due to evaporation, entrainment (or drift) and
blowdown. Makeup water brings soluble chemical species such as
"f" 4- ' I i [ |
sodium (Na ), potassium (K ), calcium (Ca ), magnesium (Mg ),
chloride (Cl~), nitrate (NO?), sulfate (SOT) , and carbon dioxide
(HCOa and GOT) into the system. The degree of concentration of .
these species is governed by the operating characteristics of
the cooling system, such as blowdown, drift and evaporation rates
Soluble constituents in makeup water become concentrated to
levels typically ranging from 1,500 to 10,000 mg/A (ppm) before
being removed in the blowdown stream (MA-230). The chemical
species contributing to the salinity of the blowdown is pri-
marily determined by makeup water composition.
-126-
-------
Chemical treatment is commonly practiced in recircu-
lating cooling systems to control corrosion, scale, biological
fouling, and solids deposition. Table 5.1-1 summarizes some of
the treatment methods employed and also presents their impact
on the quality of the blowdown stream.
The intimate contact between air and water in the
cooling device enables particulate matter and soluble gases to
be scrubbed from the air contacted. Airborne solids captured
by the cooling water significantly contribute to the solids
that accumulate in the cooling system. It is estimated that
up to 8070 of the suspended solids in recirculating systems
originally come into the system as airborne particulates
(GL-028). Upon dissolution, water soluble particulates will
increase the concentration of dissolved species of the cir-
culating water as well. Soluble gases give rise to anionic
species in the cooling water. For example, carbon dioxide
(C02), nitrogen oxides (NO ), sulfur oxides (SO ) yield
X X ^^
carbonates (C07 and HCO"^) , nitrates (NO3) , and sulfates (SO^)
in the cooling water, respectively, when these gases are
scrubbed from the air.
Leaching of preservatives from treated wood cooling
towers constitutes an additional source of potentially hazar-
dous components in cooling water blowdown. Preservatives
commonly used include acid copper chromate (ACC), chromated
copper arsenate (CCA), creosote and pentachlorophenol. The
extent of leaching is .currently under investigation.
Additional potential contaminants which may apply in
some cases include insecticides and herbicides from agricultural
runoff, or phenolic compounds from vegetation decay, most of
which are considered toxic. Chlorine addition to control
-127-
-------
TABLE 5.1-1. CHEMICAL TREATMENT SUMMARY FOR RECIRCULATING COOLING SYSTEMS
Treatment Objective
CorLOSion Inhibit Ion
Scale Control
00 '
Biological Fouling
(aIgac, slimes,
fungi) Control
Chemical Additive
Chroma te
Zinc
Phosphate
Si I I caeca
1'ropri etary Organic tt
Ac i <1
1 no r gan I c l*o I y pho a -
phates
Ctie lat ing Agents
Poiyolectrolyte
Antiprec1 pi tanta
Organlc/l'olymer
Chlorine
lly|>ocltlorf te
(Jh I o rophtHia t c- s
Thlocyanatca
Organic Sulfur
Typical AddiHve
Concert I rations in fll owdown
10-50 B.,-;/* as CrG..
8-35 nig/ft as Zn
15-60 mg/fl. as l'Ok
3-10 ing/I aa organic
< 05 ing/t residual C)»
'v- 30 mg/t rofltdudl con-
ccntrntions
Comment s
Cuoliitg water |>H la main-
tained between 6.5 and 8.0.
2-5 nig/ I
1-2
20-50
Reference
Chroma it; | reatntenl ha a |>ecn (he 1 i'adi. -
L ifm/i 1 cttrfoa i«
h i ghl y I oxl c l.i) atjual It: 1 i f e . wai e-i
treat mont vtiiuUn s are now of f er i ii)',
« 1 1 *> rnL n pec t to
sea 1 ing Halts wit houi. preci pi tat t on
do not i nhitji t sea 1 e pn-i: i pi tai ion ,
bur prevent prec i pi I uti?d sa 1 1 H from
set: 1 1 1 n|', and ail hurl rig to heal, t i an;; f tr
sui facet;
UI oc I dos usod to control It Jo lofj cat
foul ing are oi I her the o« I di /. ing or
mm-(;xi dl v.i HR, l.y{u?s , i>xtiH z Ing 1> Jo-
cldes (cliioiJut: ami hypoch lot i t e) ktave
hcc-n di Bcussttd for once - through cool ing
sysl ems i n ihik "Once -Through Cool i ng
Wat er" sec I ion . These hi oc i d.-s at e
useil In roe i roul at i ug cool i ng ay si IMUS
In a fashion si mi tar to i hut dc'.scr ibed
for once-through sy.siuius Nun oxi dl y.i ug
hloc ides (ch 1 orophon.U e:i , thtocyanar t>s ,
01 c,ani c sul fm compo\tnds , etc ) are
eiti|i) oy ed when 01 |u-i cluiini cal addl t i vts
fiuc.h r»3 organic coi ros i on inhlbi tors .
scale con l rot agcnt.s . or .so 1 i du con tin k
agt;nts are tic si royed hy the con von i i onai
ox id I ^lug hioci des .
AY -00 7
AY-007
MA- 230
EW-127
AY-007
MA- 2 30
Suspended So I i d:i
DlspcirsLori
I'annins
l.lgnlns
Pfoprietary Organics/
I'alyiuerti
Polyclccrrolyiea/Non-
ion i c INklyuu^rs
m|-/C
Olieiiucii t dl ;;(>(• r.s;jn(« maim *iin au.-;pfiidt*d AY-00?
solids from nettling and adhc-iriug lo DO-048
ht*al: r > ann f(*r «ui faces .
-------
biological fouling can result in chlorination of these or other
hydrocarbons entering with the makeup and result in highly un-
desirable reaction products.
Ash Handling
Ash is a solid by-product of coal combustion and
appears in a power plant boiler in two distinct forms: bottom
ash and fly ash. Bottom ash must be removed from the boiler in
order to maintain system operability. Fly ash is normally
collected in flue gas cleaning equipment. The conveyance of both
bottom ash and fly ash to their ultimate points of disposal con-
stitutes ash handling.
Ash handling systems employ either pneumatic or
hydraulic mechanisms for ash transportation. This section
addresses only the hydraulic (or wet sluicing) systems. Fly
ash collected dry has a market in some locations, and there-
fore, would not only eliminate one major wastewater source,
but would be a salable by-product. The usual method of opera-
tion, however, is to wet sluice the ash to a pond for settling,
or truck the ash dry to some other location for off-site dis-
posal. Wet sluicing of ash is a major source of wastewaters.
Coal-fired generating stations require formal ash
handling facilities due to the quantity of ash produced during
coal combustion. The ash content of U.S. coals ranges from
6 to 20 wt %. The average value is approximately 11 wt %
(EN-127). The distribution between bottom ash and fly ash is
greatly influenced by boiler furnace design and operating mode.
The ash distribution can affect the water balance for a hy-
draulic ash handling system. The chemical differences between
fly ash and bottom ash can also affect sluicing water quality.
-129-
-------
Bottom ash generally forms as a fused, clinker-type
material and is removed by wet sluicing. Hydraulic design con-
siderations dictate the minimum sluice water requirements as
10-20 kg (tons) per kg (tons) of bottom ash transported. In
actual practice, as high as 165 kg (tons) of water per kg (ton)
of bottom ash are used depending on such factors as plant design,
location, and operating circumstances (AY-007). Bottom ash has
excellent settling characteristics; therefore, sluice water will
be relatively free of suspended solids if adequate residence
time is supplied for sedimentation. The chemical composition of
sluice will not significantly change from sluice influent due
to the chemically inert nature of bottom ash with water (SC-267,
AS-054).
Fly ash is collected in the dry form by cyclones,
fabric filters, dry electrostatic precipitators, etc., and in
a water slurry by wet scrubbers, wet electrostatic precipitators,
etc. Fly ash collected in either the wet or dry form is commonly
sluiced to ash ponds for sedimentation of the suspended fly ash
solids. Sluice water in the pond may be 1) discharged as a
waste effluent, 2) recycled for additional ash sluicing, or
3) evaporated where meteorological conditions are favorable.
The minimum sluice water quantities are set by hydraulic design
considerations. For fly ash, the minimum sluice water require-
ment ranges from 10 to 20 kg (tons) per kg (ton) of fly ash
transported. However, as with bottom ash sluicing, the sluice
water requirement may be as high as 165 kg (tons) per kg (ton)
of ash (AY-007). Flows range from 4.5 to 150 m3 per day/MW
(1200 to 40,000 gpd/MW), while a typical rate for coal is 40 m3
per day/MW (10,000 gpd/MW).
Although fly ash has somewhat poorer settling charac-
teristics than does bottom ash, low turbidities are observed in
-130-
-------
sluice effluents if adequate retention time in the ash pond is
provided. The residence time necessary varies for different
fly ash. It is determined by the settling velocity of the par-
ticles and the rate of flow through the ash pond. The particle
settling velocity depends on particle size. Most fly ash
settles compactly in a pond volume of about 0.6 m3/metric ton
dry fly ash solids (20 ft3/ton solids) (IF-001). Fly ash con-
tains a broad spectrum of soluble inorganic salts which give
rise to sodium, potassium, calcium, magnesium, chlorides, sul-
fates, etc., in solution. The level of these dissolved solids
in solution may range from a few hundred to many thousand mg/A
(ppm). In addition, varying concentrations of approximately 30
different trace elements have been detected in both bottom ash
and fly ash sluice water (AS-054). Table 5.1-2 presents ash-
pond effluent analyses for a large coal-fired plant where sepa-
rate bottom ash and fly ash ponds are employed.
Sluice water pH is also affected by soluble chemical
species in fly ash. Fly ash from pulverized coal burning units
contain alkaline species such as oxides of sodium, potassium,
calcium, and magnesium (Na20, K20, CaO, and MgO) . Dissolution
of these salts can increase sluice water pH to levels on the
order of 10.0. On the other hand, fly ash from cyclone fur-
naces can yield sluice water pH as low as 5.5 due to adsorption
of acidic species such as S02, S03, HC1, etc., on fly ash
surfaces (AS-054). The type of coal also influences sluice pH.
Western coals generally produce ashes with high lime content,
while eastern coal ashes contain lower levels of alkaline
species and, in some cases, higher chloride levels, resulting
in acidic ash pond effluents.
-131-
-------
TABLE 5.1-2. CHARACTERISTICS OF ONCE-THRO UGH ASH POND DISCHARGES
Parameter
Flow mVs
(gpa)
Tocal alkalinity
(as CaCO,)
Phen. alkalinity
(as CaCOj)
Conductivity
(uaihos/cm)
local hardness
(as CaCOj)
PH
Dissolved solids
Suspended solids
Aluminum
Ammonia (as M)
Arsenic
la rims
Beryllium
C^Hin'tni
Calcium
Chloride
Chromium
Copper
Cyanide
Iron
Lead
Magnesium
Manganese
Mercury
Nickel
Total Phosphate (aa ?>
Selenium
Silica
Silver
Sulface
Zinc
Min
0.20
(3,100)
-
0
615
185
3.5
141
2
3.6
0,02
<0.005
0.2
<0.01
0.023
94
5
0.012
0.16
<0.01
0.33
<0.01
9.4
0.29
<0.0002
0.06
<0.01
<0.001
10
<0.01
240
1.1
Flyash Pond
Ave
0.39
(6,200)
-
0
810
260.5
4.4
SOS
62.5
7.19
0.43
0.010
0.25
0.011
0.037
136
7.12
0.067
0.31
<0.01
1.44
0.053
13.99
0.48
0.0003
1.1
0.021
0.0019
12.57
<0.01
357.5
1.51
Bottom Ash Pond
MAX
0.56
(8,300)
-
0
1125
520
6.3
320
256
3.3
1.4
0.023
0.4
0.02
0.052
180
14
0.17
0.45
<0.01
6.6
0.2
20
0.63
0.0006
0.13
0.06
0.004
15
<0.01
440
2.7
Min
0.28
(4,500)
30
0
210
76
4.1
69
5
0.5
0.04
0.002
<0.10
<0.01
<0.001
23
5
<0.005
<0.01
<0.01
1.7
<0.01
0.3
0.07
<0.0002
0.05
0.01
<0.001
6.1
<0.01
41
0.02
Ave
1.00
.(16,000)
35
0
*
322
141.5
7.2
167
60
3.49
0.12
0.006
0.15
<0.01
0.0011
40.12
3.38
0.009
0.065
<0.01
5.29
0.016
5.35
0.16
0.0007
<0.059
0.081
0.002
7.4
<0.01
48.75
0.09
Max
1.45
(23,000)
160
0 .
910
394
7.9
404
657
3.0
0.34
0.015
0.30
<0.01
0.002
67
15
0.023
0.14
<0.01
11
0.031
9.3
0.26
0.0026
0.12
0.23
0.004
S.6
<0.01
80
0.16
Units are mg/Z (ppm) unless otherwise indicated
SOURCE: RI-160
-132-
-------
Water Conditioning
Some degree of water treating is practiced in all
power plants to remove suspended solids and/or dissolved mineral
salts. The basic water treating processes used by utilities
include:
sedimentation )
/ suspended solids removal
filtration J
lime/lime-soda
softening
dissolved solids removal
ion exchange
evaporation "
Water conditioning schemes employ these basic processes singly
or in multiple combinations. The subsections below characterize
the wastewater expected from each basic process.
S edimentat ion
Sedimentation is a solid-liquid separation process in
which suspended solids in dilute concentrations are separated
from water and concentrated by gravity settling. Turbid water
is charged to a clarifier and the suspended solids settle and
concentrate at the vessel's bottom. The clarifier underflow
constitutes the waste stream for the process. It ranges from
1.0 to 4.0% of the total water charge. The solids content of
the underflow rarely exceeds 5.0 wt% (AY-007) and typically is
on the order of 2.5 wt% (EN-127).
The liquid phase composition of the clarifier under-
flow is influenced by coagulating the flocculating agents that
-133-
-------
may be added to the process in low concentrations. Table 5.1-3
characterizes the types of coagulants and flocculants that are
commonly used as well as their normal concentrations in clarifi-
cation equipment. Other chemical characteristics of the clari-
fier underflow's liquid phase are essentially equivalent to that
of the influent water to the process.
Filtration
Filtration, another solid-liquid separation process,
is used extensively to remove suspended solids from water used
in the electric utility industry. During the service cycle,
turbid water is passed through the granular filter media of the
pressure filter which collects solid impurities in the water.
These solids accumulate in the media until they are removed by
periodic backwashing of the filter.
The wastewater generated by the filtration process
comes as a result of filter backwashing. This process involves
passing a large surge of water in an upward direction through
the filter media. Backwashing expands the filter bed and provides
sufficient turbulence for solids to be freed from the media and
swept from the filter. The backwash quantity is highly variable;
it can be as high as 6.07=> of the water treated. Backwash
effluent quality has a high suspended solids content which
varies with the solids-holding filter capacity, the backwash
efficiency, and the backwash water quantity. Filter backwash
from plants producing an alum sludge typically contains 40 to
100 mg/£ (ppm) suspended solids. Other water quality parameters
such as TDS, pH, hardness, etc. are roughly equivalent to the
backwash influent, and thus, are largely determined by the
choice of backwash water.
-134-
-------
TABLE 5.1-3.
]OAGULATING AND FLOCCULATING AGENT CHARACTERISTICS
Coagulant/Flocculant
Purpose
Normal Dosage (rog/t)
Alum
A12(SO,,) 3. 14 H20
Main coagulant
To assist coagulation with aluminate
5-50
2-20
Aluminate
Na2Al20,,
Main coagulant
To assist coagulation with alum
5-15
^2
(0.1 to 0.05 of alum
dosage)
Ferric Chloride
FeCl3-6H20
Main coagulant
5-50
to
Ln
I
Copperas
FeSOH-7H20
Main coagulant
5-50
Weighting Agents Coagulant Aid
(bentonite, kaolin, montmorillonite)
Adsorbents Coagulant Aid
(powdered carbon, activated alumina)
Polyelectrolytes
(inorganic activated silica and or-
ganic polymers)
Coagulant Aid
<2
SOURCES: AY-007, TE-111
-------
Lime/Lime-Soda Softening
Lime/lime-soda softening is employed to treat the
hardness and alkalinity content of water. Calcium and mag-
nesium are precipitated as calcium carbonate (CaC03) and mag-
nesium hydroxide [Mg(OH)2], respectively. Hydrated lime
[Ca(OH)2] and/or soda ash (Na2C03) are added to the process
to drive chemical precipitation. Precipitates of CaC03 and
Mg(OH)2 are separated from the treated water by sedimentation.
The wastewater problem for the softening process arises
from removal of precipitated salts. Many similarities between
softener wastes and sedimentation wastes (described in the
Sedimentation Section) exist. The solids concentration of the
softener waste is typically 2.5 wt 70 but may be as high as 5.0
wt %. The waste quantity varies with influent water hardness
and process softening efficiency.
The liquid composition of the waste is similar to the
softened water effluent. It contains coagulants and flocculants
to enhance sedimentation efficiency. Total hardness is typi-
cally on the order of 50 mg/£ (ppm) as CaC03 (ST-135) . Waste-
water pH is approximately 10.0 due to the alkaline reagents
(soda ash and/or lime) added to the process.
Ion Exchange
Ion exchange is used to provide softening and deminer-
alization capability in power plant water treatment systems.
Water softening by ion exchange can be used for condenser cool-
ing systems, low pressure boilers, and other evaporative pro-
cesses which have a water concentrating effect. Demineraliza-
tion is used exclusively as a water treatment for high pressure
boilers.
-136-
-------
Ion exchange resins are the heart of the process and
they are classified as either the cationic or anionic type
depending upon their ion affinity. As the exchange resins
reach their ion load capacity, they are removed from service
and are regenerated. Regeneration is accomplished in three
distinct steps:
Backwashing for resin bed expansion and
solids removal
Elution of ions from the resin with a
regenerant solution
Rinsing excess regenerant and eluted ions
from the resin bed.
All three regeneration steps contribute to wastewater generated
in the ion exchange process.
Backwashing loosens and expands the resin bed and
removes suspended solids that accumulate on the resin during
service flow. This process is accomplished by an upward flow
of water through the resin bed. The backwash flow rate is ad-
justed to expand the bed volume to 80-100% greater than its
settled depth; this flow is maintained for a duration of 10 to
15 minutes. Flow rates of 3.4 - 4.1 m3/sec/m2 (5-6 gallons per
minute per ft2 of bed cross-sectional area) are typical (EN-127)
Raw influent to the process is generally employed as backwash
water. The chemical quality of the backwash water will not
be significantly altered as it passes through the ion exchange
bed.
After backwashing, the ion exchange resins are
treated with regenerant solutions to rejuvenate ion exchange
capability. The regenerant solutions elute ions from the ion
-137-
-------
exchange resins and replace them with species from the regen-
erant solution. Hydrogen cycle cation exchange resins are
regenerated with acid solutions while anion exchange resins
are regenerated with alkalis. The spent regenerant solutions
represent the aqueous waste associated with this step of the
regeneration sequence.
Spent regenerant solutions contain ions that are
eluted from the ion exchange material plus the excess regenerant
that is not consumed during regeneration. The eluted ions
represent the chemical species that were removed from water
during the service cycle of the process. For example, regenera-
tion of a hydrogen cycle cation exchange resin will elute such
species as sodium, potassium, calcium, magnesium, etc. Similarly,
ions eluted from anion exchange units will include chloride,
nitrate, bicarbonate, sulfate, carbonate, etc. The distribution
of these species in spent regenerant solutions vary with influent
water to the ion exchange process .
The excess regenerant required for ion elution varies
with the ion exchange resins employed. Table 5.1-4 presents a
summary of ion exchange material types and the regenerant re-
quirements of each. With the exception of sodium cycle ion
exchange, excess regenerant creates an effluent of extreme pH.
Spent regenerant from cation exchange units is acidic (low pH).
Alkaline (high pH) regenerants are characteristic of anion
exchange units.
The final step of regeneration is rinsing spent re-
generant solution from the ion exchange bed. The rinse water
quantity varies with each resin type, but it is typically
8.0 m3 of water per m3 of resin for cationic resins and 10 m3
of water per m3 of resin for anionic resins (EN-127). Rinse
-138-
-------
TABLE 5.1-4. ION EXCHANGE MATERIAL TYPES AND REGENERANT REQUIREMENT
Ion Exchange Material
Description of Operation
Hegenerant Solution
Theoretical Amount
Cation Exchange
Sodiura Cycle
Sodium cycle ion exchange is used as a water
softening process. Calcium, magnesium, and
other divalent cations an; exchanged for
more soluble sodium cations. I.e.,
211 -»
-------
water quality characteristics range from those of the spent
regenerant solution to those of treated water.
Evaporation
used to
Evaporation is a demineralization process sometimes
_^, _w treat boiler feedwater whereby raw feedwater is distilled
to produce a pure condensate. Feedwater impurities concentrated
in the evaporator are removed as a waste blowdown stream.
Evaporator blowdown has a high dissolved solids con-
tent. The concentration of the dissolved solids varies with
"the level of dissolved salts in the influent water and with the
degree of concentration in the evaporator. The degree of con-
centration is limited so that the evaporator operates within the
solubility limit of calcium, magnesium, and other scaling salts.
Therefore, as the scaling_potential of influent water increases,
the allowable degree of concentration in the evaporator
decreases.
The blowdown quantity necessary to maintain accept-
able scaling potentials ranges from 10-40% of the water charged
to the process (AY-007). This corresponds to maximum TDS of
approximately 3000 mg/£. However, concentrations on the order
of 1000-2000 mg/£ are more commonly observed (EN-127, AY-007).
The distribution of soluble species in the blowdown is similar
to the distribution for influent water. However, blowdown pH
is typically 9 to 11 due to the thermal decomposition of car-
bonates in the evaporator (EN-127).
-140-
-------
Boiler Slowdown
Power plant boilers are either of the once-through or
the drum-type design. Once-through designs are employed ex-
clusively in high pressure, super-critical boilers and have
no wastewater streams directly associated with their operation.
Thus, they will not be considered further in this section.
Drum-type boilers, on the other hand, operate at sub-critical
conditions where steam generated in the drum-type units is in
equilibrium "with liquid boiler water. Boiler water impurities
are, therefore, concentrated in the liquid phase as steam is
generated in these units. These impurities are ultimately re-
moved in a liquid blowdown stream, the wastewater from this
system.
The blowdown from drum-type boilers generally con-
tains soluble inorganic species tbat occur in natural waters
(i.e., Na*, K+, Cl", S07, etc.); precipitated solids containing
the calcium/magnesium cation; soluble and insoluble corrosion
products of iron, copper, and other metals; plus a variety of
chemical compounds added to the system. Dissolved solids are
present in excess of all other boiler water impurities. The
concentration of impurities in drum-type boiler blowdown is
largely governed by boiler operating conditions, such as pressure.
Table 5.1-5 presents recommended limits of total and suspended
solids in drum-type boilers as a function of drum pressure.
A number of chemical additives may be present in the
boiler blowdown as a result of internal boiler water treatment.
Internal treatment is designed to control scale formation,
corrosion, pH, and solids deposition in the boiler system. A
summary of these internal treatment control practices is
presented in Table 5.1-6.
-141-
-------
TABLE 5.1-5. RECOMMENDED LIMITS OF TOTAL SOLIDS AND SUSPENDED
SOLIDS IN BOILER WATER FOR DRUM BOILERS
Limits Recommended for Total (Dissolved and Suspended) Solids
Drum Pressure
0
2.1
3.1
4.1
5.2
6.2
6.9
10.3
MPa
- 2.1
- 3.1
- 4.1
- 5.2
- 6.2
- 6.9
- 10.3
- 13.8
psi
0
300
450
600
750
900
1000
1500
- 300
- 450
- 600
- 750
- 900
- 1000
- 1500
- 2000
>13.8 >2000
Total Solids
mg/£ (ppm)
3500
3000
2500
2000
1500
1250
1000
750
15
Limits Recommended for Suspended Solids
Below Over
4.1 MPa 4.1-6.9 MPa 6.9-13.8 MPa 13.8 MPa
(600 psi) (600-1000 psi) (1000-2000 psi) (2000 psi)
Drum Pressure
Total solids, mg/£
Total hardness as
mg/Z CaC03
Iron, mg/&
Copper, mg/£
Oxygen, mg/£
PH
Organic
0
0.1
0.05
0.007
8.0-9.5
0
0
0
0
0
8
0
.05
.03
.007
.0-9.5
0.15
0
0.01
0.005
0.007
8.5-9.5
0
0.05
0
0.01
0.002
0.007
8.5-9.5
0
*No value reported.
Source: BA-185
-142-
-------
TABLE 5.1-6. CHEMICAL ADDITIVES COMMONLY ASSOCIATED
WITH INTERNAL BOILER TREATMENT
Control
Objective
Candidate Chemical Additives
Residual Concentration
in Boiler Water
Reference
Scale
di- and tri-rsodium phosphates
Ethylene diaminetetracetic
acid (EDTA)
Nitrilotriacetic acid (NTA)
Alginates
Polyacrylates
Polymethacrylates
3-60 mg/Jl as PO,,
20-100 mg/Jl
10-60 mg/«.
up to 50-100 ing/i
up to 50-100 mg/Jl
up to 50-100 ing/A
BA-185
EN-127, AY-007, BL-036
EN-127. AY-007, BL-036
AY-007, BL-036
AY-007. BL-036
AY-007, BL-036
-P-
u>
Corrosion Sodium sulfite and catalyzed
sodium sulfite
Hydrazine
Morpholine
Sodium hydroxide
Sodiurn carbonate
Ammonia
Morpholine
Hydrazine
less than 200 mg/£
5-45 mg/ft
5-45 mg/Jl
Added to adjust
boiler water pH
to the desired •
level, typically
8.0 - 11.0.
MA-230, BL-036
AY-007
AY-007
EN-127, AY-007, BA-185, BL-036
Solids
Deposition
Starch
Alginates
Polyacrylamides
Polyacrylates
Polymethacrylates
Tannins
Lignin derivatives
20-50 mg/Jl
20-50 mg/Jl
20-50 mg/Ji.
20-50 mg/i
20-50 mg/Jl
<^200 mg/Jl
<200 mg/Jl
AY-007
AY-007
AY-007
AY-007
AY-007
AY-007
AY-007
-------
The blowdown quantity of modern, high-pressure
boiler ranges from effectively zero to an upper limit of 2.07=
of the steam generation rate. The blowdown rate is typically
0.1% of the steam generation rate (AY-007). Much higher blow-
down rates, typically 10%, are associated with lower pressure
steam generating systems where makeup is not demineralized.
Boiler blowdown may be performed in either an intermittent or
continuous fashion.
Coal Pile Runoff
Coal-fired power stations maintain reserve fuel on
the plant premises in active and/or inactive coal storage
piles. Active coal storage is open and is exposed to all am-
bient conditions. Inactive coal piles are commonly sealed with
a tar spray or some other impervious covering which provides
protection from the weather. Runoff from active coal storage
piles is of primary concern in this section.
Precipitation runoff from active coal storage piles
presents a potential problem of stream and ground water pol-
lution. This runoff commonly exhibits extreme pH and contains
soluble chemical species and suspended solids. The primary
cause of runoff contamination is a reaction mechanism similar
to the one that produces acid mine drainage. Inorganic sulfur
in the coal reacts with moisture and oxygen in air to produce
sulfuric acid.
When rainwater seeps into the coal pile, sulfuric
acid is leached from the coal. The pH of the runoff effluent
can be as low as 2-3 units. The acidic nature of this water
drives the dissolution of inorganic salts that are present in
the coal. In addition to a high sulfate anion concentration,
-144-
-------
the runoff contains high concentrations of cations such as
iron, aluminum and manganese. Traces of cadmium, beryllium,
nickel, chromium, vanadium, zinc, and copper have also been
reported. Coal fines and other insoluble material appear in
the runoff as suspended solids.
Table 5.1-7 presents plant data for coal pile runoffs.
Coal type has a great influence on runoff characteristics. For
example, some coals such as are burned at Plant 5305 have suf-
ficient alkalinity to neutralize all of the sulfuric acid formed.
The resulting effluent pH in such cases ranges from 6.5 to 7.5.
The higher pH range decreases the solubility of many inorganic
salts, thus affecting runoff effluent quality. The runoff at
Plants 1729, 3626, 0107, on the other hand, is very acidic be-
cause of the high-sulfur fuel burned. Other factors causing
variations in the effluent quality besides coal type are coal'
pile history and runoff flow rate.
The quantity of runoff effluent is a strong function
of coal pile area and local meteorological conditions. Coal
pile area is primarily determined by generating station size.
Power plants store from 600 to 1,800 cubic meters (0.5 to 1.5
acre-feet) of coal for each MW of generating capacity. The
storage piles are typically 8 to 12 meters (25 to 40 feet)
in height (AY-007). This corresponds to a coal storage area
of 50 to 250 square meters (0.013 to 0.060 acres) for each
MW of capacity, depending on pile height. An annual pre-
cipitation rate of one meter (40 inches), for example, will
result in an annual runoff of between 50 and 225 cubic meters
(13,000 to 60,000 gallons) per MW of generating capacity. The
typical runoff rate is 76,000 - 95,000 cubic meters (20 to 25
million gallons) per year at most coal-fired generating stations
(EN-127, AY-007).
-145-
-------
TABLE 5.1-7
PLANT DATA RELATING TO WATER QUALITY
PARAMETERS FOR COAL PILE RUNOFF
Plane Cade
* Ai!uJUai:y (ag/i)
200 (rag/i)
C00 (ag/O
TS
TDS
tss
Aczonia
Hicrata
?iigs?norau3
Turbidity
Acidic?
local Hacdnass
Suliace
Chloride
Altanisia
Cironiua
Copper (ag/l)
Iron (a?/i)
Magaesiuai (ag/Z)
Zinc (smj/t)
Sodium Oog/1)
Pa
34Q2
5
0
1080
1330
720
610
0
0.3
-
. 505
-
130
525
3.5
-
0
1.6
0.153
-
1.6
1260
2.3
3401
0
0
1080
1330
720
610
0
0.3
-
505
-
130
525
3.6
-
0
1.6
0.1S3
1.5
1260
2.3
3936
0
10
306
9999
7743
22
1.77
1.9
1.2
-
-
1109
5731 .
431
-
0.37
-
-
34
2.43
ISO
3
1325
-
-
35
60QO
5300
200
1.35
1.3
-
-
-
1850
361
-
-
0.05
-
0.06
174
0.0006
-
4.4
i;:6
32
3
1099
3549
247
3302
0.35
2.25
0.23
-
~
-
133
23
-
,
-
-
-
0.08
-
7.3
1729 3626
-
,
-
-
- 23970
100
.
-
.
-
- 21700
-
6337 19000
-r
1200
15.7
1.3
0.353 4700
-
12.5
-
4.7 2.1
-0107 5305
0 21.36
-
-
45000
44050
950
-
-
-
3.37
27310 3. s8
-
21920
-
325
0.3
3.4
93000 1.0
-
23
-
2.3 5.7
5303 5J05
14.32 36.41
-
-
-
-
-
-
-
-
2.77 5.13
10.25 3.34
-
-
-
-
.
-
1.05 3.9
-
-
-
5.5 6.6
* All concentrations (except pH) are expressed in mg/2..
Source: EN-127
-146-
-------
Chemical Cleaning
Operational cleaning of heat transfer surfaces is
designed to remove scale and corrosion products that accumulate
on the boiler's steam-side and the water-side of the steam
condenser. The frequency at xvhich chemical cleaning is needed
varies from plant to plant. For example, the mean time between
boiler chemical cleanings is approximately 36 months. However,
plant data indicate extreme variations in frequency ranging from
once in 7 months to once in 100 months (EN-127).
The active reagents in cleaning solutions are acidic
or alkaline in nature depending primarily upon the deposits
they are to attack. Ninety percent of all cleaning operations
employ acidic formulations that attack all forms of alkaline
scale (i.e., CaCOs, Mg(OH)2, etc.), silica scale, and corrosion
deposits containing iron. The majority of these formulations
contain hydrochloric acid in solution strengths ranging from
5.0 to 7.57=, (AY-007) . Other acid solutions contain the fol-
lowing constituents which are present alone or in various
combinations (EN-127, AY-007, BA-185, BE-162):
Inorganic Acids Organic Acids
Hydrochloric (HC1)
Sulfuric (H2SOO Citric [HOC(CH2C02H)2C02H]
Sulfamic (NH2S03H) Formic (HC02H)
Phosphoric (HaPOO Hydroxyacetic (HOCH2C02H)
Nitric (HN03)
Hydrofluoric (HF)
Representative data for three separate cleaning
operations are presented in Tables 5.1-8 through 5.1-10. The data
are for cleaning the steam-side of a high pressure, once-through
-147-
-------
boiler and a low pressure, drum boiler plus the water-side of
a main condenser, respectively. Alkaline and acidic solutions
are shown for both cleanings while only an acidic solution is
shown for condenser cleaning.
Spent chemical cleaning solutions usually have ex-
treme pH, high dissolved solids concentrations, and significant
oxygen demands (BOD and/or COD). The pH of spent solutions
ranges from 2.5 to 11.0 depending on whether acidic or alkaline
cleaning reagents are employed. The dissolved solids include
sodium, hardness, heavy metals, chloride, bromide, and
fluoride. Tables A-8 through A-10 report only iron and copper
concentrations for heavy metals. However, additional metal
constituents may include nickel, zinc, and aluminum. Heavy
metals for combined boiler cleaning wastes follow the general
concentration trend of (EN-127):
iron > copper > nickel > zinc > aluminum
The quantity of cleaning wastes varies directly with
liquid holding volume of equipment to be cleaned (BA-185),.
For example, Tables A-8 and A-9 show spent alkaline and acid
cleaning solution quantities as equal to the tube-side boiler
volume. Rinse solution quantities are generally one or two
times this volume.
•148-
-------
TABLE 5.1-8. OPERATIONAL CLEANING OF A HIGH PRESSURE, ONCE-THROUGH BOILER*a
1
-p-
1
pE
Acidity, ppm total hot,
as CaC03
Alkalinity, ppm as CaC03
NH 3 , (%)
T- ++
Fe , ppm
Cu
BOD
Suspended Solids, ppm
Volume, gal
m3
Temperature,
as drained, °F
°c
Stage 1
11.0
_
90,000
1
-
720+
-
500
30,000
110
100
40
Stage 1
Rinse
9.0
_
9,000
0.1
-
75
-
5
60,000
220
100
40
Stage 2
2,5-3.0
27,000
(2)
60 , 000
-
high
100
30,000
110
200
90
Stage 2
Rinse
6.0-7.0
low
(2)
600
-
high
5
60,000
220
200
90
use of ammonium salts of organic acids raises this number,
/ 2\
v 'Capacity-20,000 gallons; solvent system Stagel, ammonium persulfate solvent;
Stage 2, inhibited 370 organic acid; deposit inventory, 200 Ibs copper as Cu
and 1500 Ibs iron as
Source: BE-162
-------
TABLE 5.1-9. OPERATIONAL CLEANING OF A LOW PRESSURE, DRUM BOILER
0)
pH
Acidity, ppm total hot
as CaC03
Alkalinity, ppm as CaCOa
NH3(7o)
Chloride, ppm
T, ++
Fe , ppm
M Cu, ppm
Ui
*p Bromide , ppm
Suspended Solids , ppm
Volume, gal
m3
Temperature, as
drained, °F
°C
Stage 1
11.0
_
90,000
1
-
' -
720+
1,500
500
30,000
110
150
65
Stage 1
Rinses
9.0
_
9,000
0.1
-
-
75
150
5
60,000
220
150
65
Stage 2
<1.0
70,000
-
0.01
68,000
6,000
75
15
500
30,000
110
150
65
Stage 2
Rinses
7.0-9.0
-
1,000
-
<6,800
600
—
100
60,000
220
125
52
^''Capacity, 30,000 gallons; solvent system Stage 1, bromate solvent; solvent
system Stage 2, inhibited 570 hydrochloric acid and copper complexing
agent; deposit inventory, 200 Ibs copper as Cu and 1,500 Ibs iron as
Fe203.
Source: BE-162
-------
TABLE 5.1-10. OPERATIONAL CLEANING,
MAIN CONDENSER WATER-SIDE(l
PH
Acidity, hot, ppm CaC03
Alkalinity, ppm CaCOa
Ca, Mg, ppm
Fe, ppm total
Cu, dissolved, ppm
Suspended solids , ppm
Temperature, as
drained, °F
°c
Solvent
2.0
30,000
_
5,000
11,000
1,000
2,000+
140
60
Rinses
7.0-9.0
-
1,000
50
100
50
200
120
49
^'Capacity - 1000 gallons; solvent'system, 10%
sulfamic acid, inhibited, 17. NaCl; Rinse -
Na2C03 + sodium phosphates; depository inven-
tory, 100 Ibs Ca and Mg salts, 100 Ibs Fe203,
10 Ibs copper.
Source: BE-162
-151-
-------
Washing Operations
Floor and equipment washing operations produce ef-
fluents containing suspended solids, detergent constituents,
oily wastes, and a broad spectrum of soluble inorganic species.
Water quality parameters for washing operations are highly de-
pendent on the specific area or equipment item washed . Two
major washing operations at fossil-fueled power plants concern
removal of deposits consisting of fuel ash, soot, combustion
additives, etc., from the boiler fireside and air preheater.
Boiler firesides are commonly washed by spraying
high pressure water against boiler tubes while they are still
hot. In some cases alkaline wash water is used. Waste ef-
fluents from this washing operation contain an assortment of
dissolved and suspended solids. Wash frequencies, waste vol-
umes, and effluent water quality data for two generating sta-
tions are presented in Table 5.1-11. It should be noted that
acid wastes are common for boilers fired with high sulfur
fuels. Sulfur oxides adsorb onto fireside deposits and im-
part low pH and a high sulfate content to the waste effluent
(BA-185). These deposits can also be a source of iron, nickel,
chromium, vanadium, and zinc depending on the fuel type and
fuel additives.
Air preheaters employed in power stations are either
the tubular or regenerative types. Both are periodically
washed to remove deposits that accumulate. The frequency of
washing is typically once per month; however, frequency varia-
tions ranging from 4 to 180 washings per year are reported
(EN-127). Many air preheater designs are sectionalized so
that heat transfer areas may be isolated and washed without
•152-
-------
TABLE 5.1-11. DATA FOR BOILER FIRE SIDE WASHING OPERATIONS-
INCREASE IN POLLUTANT QUANTITY PER WASHING CYCLE
?Uac Coda
cyciaa/yr
3&CC3 7o i£3&, 3,"
1000 jal
.Ukaliaisy, T-b
'**
COD, Ib
kg
Tocal Solids, Ib
'*«
Tocal Qissoivad Solids, Lb
*3
Toe 3.1, Suaoaaead Salids, ib
'*3
Sulzaea, Lb
kg
Gsiorida, Lb
kg
Asmoaia, Lb
kg
Hieraea, Lb
'hosohorous, Lb
kg
Ear daas a , Lb
Chranu.ua, Lb
'*«
Capper, ib
Irpa, Lb
«a aesium ib
"kg
Mickal, Lb
kg
Sodiua, Lb
kg
Ziae, Lb
30D, Lb
'*3
Turbidicy, JTu
3410
2
2525
720
-240
-109
U34
315
40861
13551
35127
LiJ9^3
j3?T
1736
L1949
5423
0
0
1.49
0.63
14.73
6.7
11.1
5.04
35409
16076
0.0299
O.OU6
..
900
403.9
U.36,9
3425
30.02
13.63
0
0
23.72
13.042
o-
0
476
•411
3
90.3
24
-5.99
-2.72
19
8.63
4002
1317
3002
1363
113.09
54.07
299. 4
135-9
13.01
3. IS
0.039
0.013
0.7
0.313
0.257
0.117
791.41
359.3
0.998
0.433
0.249
0.113
30
13.53
190.35
36.42
-
"
9 ,
4. Q9
2
0.908
0
0
98
SOURCE: EN-127
-153-
-------
shutdown of the entire unit (BA-185). Higher wash frequencies
are expected for air preheaters employing this design feature.
Fossil fuels with significant sulfur content will
produce sulfur oxides which adsorb on air preheater deposits.
Water washing for these deposits produces an acidic effluent.
Alkaline reagents are often added to wash water to neutralize
acidity, prevent corrosion of metallic surfaces, and maintain
an alkaline pH. Alkaline reagents might include soda ash
(Na2C03), caustic soda (NaOH), phosphates and/or detergent.
Preheater wash water contains high solids content
(both suspended and dissolved). The solids primarily include
sulfates, hardness, and heavy metals, including large quanti-
ties of copper, iron, nickel, and chromium. The levels of
metals in preheater deposits are usually much higher than those
in fireside deposits. Air preheater wastes may also be a source
of oily matter and pol-ynuclear hydrocarbons . Data for air
preheater wash water discharges are presented in Table 5.1-12.
General Plant Drainage
General plant drainage refers to liquid that accumu-
lates in a floor and yard drains in the process area of a
power plant as a result of precipitation runoff.
Plant drainage generally contains a high level of
suspended solids consisting of such materials as soil, dust,
coal fines, fly ash, etc. that are entrained in the runoff
flow. Any significant degree of dissolution of these solids
will also add to the dissolved salts present in the water.
The specific characteristics of runoff vary radically from
plant to plant and from time to time for a given plant. The
-154-.
-------
TABLE 5.1-12. DATA FOR AIR PREHEATER WASHING OPERATIONS-;
INCREASE IN POLLUTANT QUANTITY PER WASHING CYCLES
?l3uc Cads
Clsaaing Trafluency
i'/ciis/^r
3acca Voluaa, a!
1QOO jai
COD, Uo
'*S
Toeal Solids, lb
'
-------
general trend is that highest impurity levels are observed as
initial runoff occurs. Subsequent runoff is observed to have
lower impurity concentrations„
Process Spills and Leaks
Liquid spills and leaks are commonly associated'with
overfilling of storage vessels; tank or pipe ruptures; failure
of valves, pump seals, etc. Waste effluent characteristics
resulting from spills and leaks depend upon the type of fluid
that escapes containment. Potential fluids include:
acid, 'alkalis, and brine solutions for ion
exchange regenerants as well as other water
treating chemicals
fuel oil, transformer oil, and circuit
breaker oil
water used in plant operation such as
cooling water.
Miscellaneous Operations
Several miscellaneous sources of wastewater at power
plants are listed below:
laboratory and sampling operations
auxiliary cooling system(s)
water intake screen washings
others
-156-
-------
Although the impact of wastewater from these miscellaneous
operations is less significant than those discussed in previous
sections, these sources nevertheless contribute to the total
wastewater problem.
Quantitative data which characterize wastewater from
these miscellaneous sources is limited. The only data availa-
ble are for auxiliary cooling systems, which remove heat from
mechanical equipment items such as those listed below:
bearing and/or gland cooling for pumps,
fans, and other rotary equipment
air compressor water jackets
generator cooling
Auxiliary cooling systems can be either the once-through or
closed-cycle types.
Once-through auxiliary cooling systems do not usually
involve chemical treatment, with the exception of chlorination.
Thus, water quality of the waste is determined by the influent
water to the system. During chlorination of these systems,
residual chlorine levels in the effluent are approximately the
same as presented for condenser cooling systems, about 0.1 to
0.2 mg/A (ppm) . However, the frequency of chlorination is
considerably reduced. (See sections entitled Once-Through
Cooling Water and Cooling Tower Slowdown.)
The flow through the once-through auxiliary cooling
circuit ranges from 0.000032 to 0.0022 m3/s (0.5 to 35 gpm) per
MW of rated generating capacity. The typical flow is approxi-
mately 0.00066 m3/s (10 to 11 gpm) (AY-007). This total flow
represents the wastewater stream for once-through auxiliary
cooling.
-157-
-------
Wastewater from a closed-cycle auxiliary cooling
system is a blowdown stream. Water in the closed cooling cir-
cuit is treated to control corrosion with inhibitors such as
chromates (at levels up to 250 mg/£) and borates or nitrates
(at levels ranging from 500 to 2000 mg/A). Water pH is main-
tained between 9.5 and 10 by addition of caustic soda or soda
ash (AY-007). Alternatively," some plants use stream condensate
ammonia and hydrazone in this cooling circuit (EN-127). The
water recirculation rate is typically 0.0015 to 0.0016 m3/s
(23-25 gpm) per MW of rated generating capacity. Blowdown
rates vary from zero to 0.0019 m3/day (0-5 gpd) (EN-127,
AY-007).
5.2 Characterization of Wastewaters from the
Lime Wet Scrubbing Process
A description of the Lime Wet Scrubbing Process is
given in Section 3.2. The following section presents a closer
examination of the water system. A water balance for one of
four equivalent scrubbing trains for 500 MW power plant is given.
An examination of the water system flows allows the conclusion
that there are no wastewater streams in the lime wet scrubbing
process in normal operation.
5.2.1 Base Case Water Balance
In calculating an example water balance to illustrate
the water flows in a lime wet scrubbing system, Radian chose a
base case for ease of comparison. The base case is calculated
for one of four equivalent scrubbing systems for a 500 MW power
plant burning 3.5 percent sulfur coal with an average heating
value of 28 MJ/kg (12,000 Btu/lb). The detailed calculations
are presented in Appendix B. Figure 5.2-1 illustrates the water
-158-
-------
VD
I
LIME
SOj ABSORBER
SECOND STAGE
SOLID-LIQUID
SEPARATOR
OR
SETTLING POND
SOLID-LIQUID
SEPARATOR
Figure 5.2-1. Process flow diagram lime wet scrubbing process.
-------
system. The results of the water balance calculations are pre-
sented in Table 5.2.1. Several of the circulating flow rates
are characteristic flows rather than design flows, as indicated.
Flue gas first enters an ESP where 99 percent of the
fly ash particulates are removed and sluiced to the ash pond.
The flue gases, which contain., some percentage of water, then
enter the absorber where SQz is removed. The gases are cooled
and saturated by evaporation of water from the circulating lime
slurry. The saturated flue gas exits the absorber, is reheated
for proper bouyancy, and discharged to the stack. The absorber
effluent is sent to a hold tank where calcium sulfite and sulfate
crystals are precipitated. A 10-15 weight percent slurry is
recirculated to the absorber. A bleed stream is sent to solids
dewatering. The supernatant from the solid liquid separator is
recycled to the hold tank. The underflow is typically 30-40
percent solids by weight. Either second stage solid-liquid
separation or a settling pond is used to recover more of the
liquor. This liquor is then recycled to the hold tank. The final
sludge is typically 50-60 percent solids in a settling tank, or
60-70 percent solids if a vacuum filter and/or centrifuge is used.
Makeup water is therefore required to compensate for loss due to
evaporation in the absorber and to occlusion in the solid waste.
5-2.2 Description of the Water System
The water systems in the lime wet scrubbing process
can be designed to operate in closed loops, as exemplified by
the lime scrubbing systems at Louisville Gas and Electric's
Paddy's Run and Pennsylvania Power Company's Bruce Mansfield
stations (PE-259). In a closed system, the lime slurry is
recirculated between the effluent hold tank and the absorber.
A bleed stream is taken for solids dewatering and disposal. The
solids are generally removed in a clarifier underflow as a 30-40
weight percent sludge. The clarifier supernatant is recycled
-160-
-------
TABLE 5.2-1.
WATER BALANCE: LIME WET SCRUBBING PROCESS
(one of four equivalent scrubbing trains)
Stream No.
Figure 5.2-1
Description
Stream Rate,
kg/s (Ib/min)
Water,
kg/s (Ib/min)
1
Gas
to
Absorber
160
(21,000)
7.64
(1000)
2
Gas
to
Air
Heater
179
(22,000)
15.1
(2000)
3
Make-up
Water
to
Absorber
8.3
(1100)
8.3
(1100)
It*
Recycle
Slurry
to
Absorber
940
(125,000)
830
(110,000)
5*
Slurry
to
Hold
Tank
940
(125,000)
830
(110,000)
6
Lime
Feed
0.63
(84)
7
First Stage
S/L Separator
to
Slurry Feed
Tank
4.9
(650)
4.9
(650)
8**
Liine
Slurry
to
Hold
Tank
5.6
(740)
4.9
(650)
9
Slurry
Bleed
Stream
15
(2000)
14
(1800)
10
Solid/Liquid
Separator
Overflow to
Hold Tank
6.3
(830)
6.3
(830)
11
S/L
Separator
Underflow
to Fond or
Filtration
3.9
(520)
2.4
(320)
12
Second Stage
S/L Overflow
to Hold
Tank
1.3
(170)
1.3
(170)
13
Solid
Waste
2.6
(350)
1.1
(150)
*These values are approximate characteristic values based on MC-147.
**This slurry is 12.Tt solids aa in McClammery, et al. (MC-147).
Assumptions;
a) 105% stolctiiomatric lime
b) 3.5% sulfur in coal (dry basis)
c) 12% Ash (as fired basis)
d) 28 MJ/kg (as fired basis)(12,000 Btu/lb)
e) 92% of sulfur in coal evolves as SQi
f) 99% removal of particulates in ESP prior to absorption
g) 90% SOi. removal
h) pond evaporation equals rainfall
i) 752 of ash evolves as fly ash
j) first stage S/L separator produces 40% solids sludge
k) ponding or second stage S/L produces 60% solids sludge
1) 10% solids reclrculating slurry
-------
to the effluent hold tank. If the waste solids are to be disposed
of off-site, it is usually more economical to dewater them further
by vacuum filtration and/or centrifugation to produce a final
solids content of 60-70 weight percent. The filter cake is then
transported to the disposal site. The filtrate and/or centrate
are recycled to the effluent hold tank. When disposed of on-site,
the clarifier underflow is pumped to large settling ponds where
additional solids settling occurs. Clear supernatant is withdrawn
from the pond and recycled to the scrubbing loop.
Water makeup is required to replace water losses due
to evaporation in the absorber, and to occlusion in the solid
waste. In certain cases, the water occluded in the solid waste
can have an impact on surrounding water quality. This aspect of
FGD-associated water impacts, however, is addressed in the report
on solid waste impact (RO-359). Large influxes of rain water
into the settling pond are handled by increasing the flow of pond
supernatant back to the hold tank and decreasing the water makeup
rate.
A purge stream may, however, be required from a lime
FGD system because of several possible process problems. Fresh
water must be added to the system for pump seals and demister
washing. When the boiler is turned down so that flue gas flow
is decreased, fewer solids are produced and, thus, the liquor
bleed stream is reduced. This decreases the amount of water lost
from the system. The amount of fresh water required for pump
seals and demister washing does not decrease, however, so that
a water imbalance necessitating a purge may occur. Changes in
operation from design conditions, such as operating at lower
sulfur coal or oxidizing sulfite sludge to sulfate (which can be
dewatered more easily) could also decrease the amount of water
lost and necessitate a purge. In addition, catastrophic occur-
rences could necessitate a quick blowdown to prevent scaling.
Operator error could also result in a purge. In many cases, the
-162-
-------
need for a purge can be avoided by proper design and operation
of the process. Treatment technologies are available to handle
these purge streams.
5.2.3 Purge Characteristics
The possible purge from the lime system would have the
same quality as the recirculated clarifier supernatant. This
liquor differs from system to system due to coal type, fly ash
collection facilities, and scrubber operation. The liquor is
saturated with dissolved calcium sulfite and/or calcium sulfate
salts. Sodium and chloride ions are also present in high con-
centrations. Most trace elements and toxic species are controlled
below 1 mg/fc (BO-203).
5.3 Characterization of Effluents from the Limestone
Wet Scrubbing Process
A description of the Limestone Wet Scrubbing Process
is given in Section 3.3. The following section presents a closer
examination of the water system. A water balance for one of four
equivalent scrubbing trains for a base case 500 MW power plant is
given. An examination of the water system flows permits the
conclusion that there are no wastewater streams in the limestone
wet scrubbing process in normal operation.
5.3.1 Base Case Water Balance
In calculating an example water balance to illustrate
the water flows in a limestone wet scrubbing system, Radian
chose a base case for ease of comparison. The base case is
calculated for one of four equivalent scrubbing systems for a
500 MW power plant burning 3.5 percent sulfur coal with an
average heating value of 28 MJ/kg (12,000 Btu/lb). The detailed
-163-
-------
calculations are presented in Appendix A. Figure 5.3-1
illustrates the water system. The results of the water balance
calculations are presented in Table 5.3.1. Several of the cir-
culating flow rates are characteristic flows rather than design
flows, as indicated.
Flue gas first enters an ESP where 99 percent of the
fly ash particulates are removed and sluiced to the ash pond.
The flue gases, which contain some percentage water, then enter
the absorber where S02 is removed. The gases are cooled and
saturated by evaporation of water from the circulating limestone
slurry. The saturated flue gas exits the absorber, is reheated
for proper bouyancy and discharged to the stack. The absorber
effluent is sent to a hold tank where calcium sulfite and sulfate
crystals are precipitated. A 10-15 weight percent slurry is
recirculated to the absorber. A bleed stream is sent to solids
dewatering. The supernatant from the solid liquid separator is
recycled to the hold tank. The underflow is typically 30-40
percent solids by weight. Either second stage solid-liquid
separation or a settling pond is used to recover more of the
liquor. This liquor is then recycled to the hold tank. The
final sludge is typically 50-60 percent solids in a settling
tank or 60-70 percent solids if a vacuum filter and/or centrifuge
is used. Makeup water is therefore required to compensate for
loss due to evaporation in the absorber and to occlusion in the
solid waste.
5.3.2 Description of the Water System
The water systems in the limestone wet scrubbing pro-
cess can be designed to operate in closed loops as exemplified by
the limestone scrubbing system at Central Illinois Light Company's
Duck Creek No. 1A and Kansas City Power and Light's La Cygne
Station (PE-259). In a closed system, there are no aqueous
-164-
-------
FAN
TO STACK
Ui
LIMESTONE
SECOND STAGF.
SOLID-LIQUID
SBPAHATOR
OR
ETTI.ING
•OLID-LIQUID
SOLID WASTE
Figure 5.3-1. Process flow diagram limestone wet scrubbing process.
-------
TABLE 5.3-1. WATER BALANCE: LIMESTONE WET SCRUBBING PROCESS
(one of four equivalent scrubbing trains)
Stream No.
Description
Stream
Rate, kg/a
(Ib/min)
Water, kg/s
(Ib/min)
1
Gas
to
Absorber
160
(21,000)
7.6
(1000)
2
Gas
to
Air
Beater
170
(22,000)
15
(200)
3
Make-up
Water
to
Heater
9.1
(1200)
9.1
(1200)
4*
Recycle
Slurry
to
Absorber
1500
(200,000)
1400
(ISO, 000)
5*
Slurry
to
Hold
Tank
1500
(200,000)
1400
(180,000)
6 1**
Limestone First Stage
Feed S/L Overflow
to
Slurry Feed
Tank
1.3 0.76
(170) (100)
0.76
(100)
8
Limestone
Slurry to
Hold
Tank
2.0
(270)
0.76
(100)
9
Slurry
Bleed
Stream
17
(2200)
15
(2000)
10
Solid-
Liquid
Separator
Overflow
to
Hold Tank
11
(1500)
12
(1600)
11
Solid-
Liquid
Separator
Underflow
to Pond or
Filtration
4.3
(570)
2.6
(350)
12
Second Stage
S/L Separator
Overflow to
Hold Tank
1.4
(180)
0.64
(B5)
13 -
Solid
Waste
2.9
(390)
1.3
(170)
*These values are approximate characteristic values.
**This stream is 6.3X solids as in McGlammerry, et al. (MC-147).
Assumptions!
cn
Cft
i
a) 120% stoichiometric limestone
b) 3.5% sulfur in coal (dry basis)
c) 12% ash (us fired basis)
d) 28 MJ/kg (as fired basis)(12,000 Btu/lh)
e) 12% of sulfur in coal evolves as S02
f) 99% removal of partlculatea in ESP prior to absorption
g) 90% SO2 removal
b) pond evaporation equals rainfall
i) 752 of ash evolves as fly ash
j) first stage S/L separator produces 40% solids sludge
k) ponding or second stage S/L produces 60% solids sludge
1) 10% solids recirculating slurry
-------
discharges. The limestone slurry is recirculated between the
effluent hold tank and the absorber. A bleed stream is taken
for solids dewatering and disposal. The solids are generally
removed in a clarifier underflow as a 30-40 weight percent sludge.
The clarifier supernatant is recycled to the effluent hold tank.
If the waste solids are to be disposed of on-site, the sludge is
usually ponded and allowed to settle. A final sludge of 50-60
weight percent solids can be achieved with the clear supernatant
liquor from the sludge pond recycled to the effluent hold tank.
If the waste solids are to be disposed of off-site, it is usually
more economical to dewater them further by vacuum filtration
and/or centrifugation to produce a final solids content of 60-70
weight percent. The filter cake is then transported to the
disposal site. The filtrate and/or centrate are recycled to the
effluent hold tank.
A net water makeup is required to replace water losses
in the process due to evaporation in the absorber, and to occlu-
sion in the solid waste. In areas where rainfall exceeds solar
evaporation rates, more supernatant liquid is returned from the
settling pond than sent to the pond. This influx of water into
the closed loop limestone scrubbing system can be offset by
reducing the raw water makeup to the process. In certain cases,
the water occluded in the settled solid wastes can have an impact
on surrounding water quality. This aspect of the FGD associated
water impacts, however, is addressed in the report on solid
waste impact (RO-359).
A purge stream may be required from a limestone FGD
system because of several possible process problems. Fresh
water must be added to the system for pump seals and demister
washing. When the boiler is turned down so that flue gas flow
is decreased, fewer solids are produced and, thus, the liquor-
bleed stream is reduced. This decreases the amount of water
lost from the system. The amount of fresh water required for
-167-
-------
pump seals and demister washing does not decrease, however, so
that a water imbalance necessitating a purge may occur. Changes
in operation from design conditions, such as operating at lower
sulfur coal or oxidizing sulfite sludge to sulfate (which can
be dewatered more easily) could also decrease the amount of
water lost and necessitate a purge. In addition, catastrophic
occurrences could necessitate a quick blowdown to prevent
scaling (BO-203). Operator error could also result in a purge.
In many cases, the need for a purge can be avoided by proper
design and operation of the process. Treatment technologies are
available to handle these purge streams.
5.3.3 Purge Characteristics
The possible purge from the limestone system would
have the same quality as the recirculated clarifier supernatant.
This liquor differs from system to system due to coal type, fly
ash collection facilities, and scrubber operation. The liquor
is saturated with dissolved calcium sulfite and/or calcium
sulfate salts. Sodium and chloride ions are also present in
high concentrations. Most trace elements and toxic species are
controlled below 1 mg/2. (BO-203) .
5 .4 Characterization of Wastewaters From the Wellman-Lord
Sulfite Scrubbing Process
A description of the Wellman-Lord process is given in
Section 3.4. The following section presents a closer examination
'of the water system. A water balance is given for a base case
condition for one of four equivalent scrubbing trains applied to
a 500 MW power plant. An examination of the water system shows
that one wastewater stream is associated with the prescrubbing of
flue gases. This stream contains particulates and chlorides, but
is insignificant in comparison to the wastewater streams from
power plants.
-168-
-------
5.4.1 Base Case Water Balance
In calculating an example water balance to illustrate
the water flows in a Wellman-Lord sulfite scrubbing system,
Radian has chosen a base case for ease of comparison. The case
is calculated for one of four equivalent scrubbing systems for
a 500 MW power plant burning 3.5 percent sulfur coal with an
average heating value of 28 MJ/kg (12,000 Btu/lb). The detailed
calculations are presented in Appendix A. Figure 5.4-1 illus-
trates the water system. The results of the water balance cal-
culations are presented in Table 5.4-1. Several of the circulat-
ing flow rates are characteristic flows rather than design flows,
as indicated.
Flue gas first enters an ESP where 99 percent of the
particulates are removed and sluiced to the ash pond. The hot
flue gases then enter a prescrubber,. where the remaining
particulates and 99 percent of the chlorides are scrubbed from
the gases. The gases are cooled and saturated by the evaporation
of water in the prescrubber. A bleedstream is taken from the
scrubber circulating slurry to maintain desired suspended and
dissolved solids levels. This stream is then routed to the ash
pond. The scrubbed flue gases enter the absorber where S02
is removed and absorbed into a circulating sodium sulfite solution.
The cleaned gases exit the absorber, are reheated, and exhausted
to the stack. No additional evaporation of water takes place
in the absorber. The absorber effluent is sent to double effect
evaporators where S02 and water are driven off. The overhead is
passed through condensers and most of the water is condensed.
The water is then sent to the dissolving tank for recombination
with the evaporator underflow and make-up sodium carbonate. The
resulting solution is used for absorber feed. The product S02
stream contains 5-10 percent water.
-169-
-------
Figure 5.4-1. Process flow diagram for the Wellman-Lord
sulfite scrubbing process.
-170-
-------
TABLE 5.4-1.
WATER BALANCE: WELLMAN-LORD SULFITE SCRUBBING PROCESS
(one of four equivalent scrubbing trains)
Stream No.
Figure 5.4-1
Description
Stream
Kate, ke/s
(lb/min)
Water, kg/s
(lb/min)
1 2
Flue Dryer
Gas Gas
to to
Scrubber Scrubber
160 5.9
(21,000) (780)
7.6 0.083
(1000) (11)
3
Total
Gas
to
Scrubber
170
(22,000)
7.6
(1000)
4* 5. 6
Scrubber Scrubber Slurry
Liquor Make-up to
Product Hater Ash Pond
300 9.1 0.91
(40,000) (1200) (120)
290 9.1 0.83
(38,000) (1200) (110)
7*
Circulating
Slurry
to
Scrubber
300
(40,000)
290
(38,000)
8
Gas
to
Absorber
170
(23,000)
16
(2100)
9 10* 11
Gas Absorber Absorber
to tlquor Liquor
Reheat Product to
Evaporator
170 11 9.8
(23,000) (1500) (1300)
16
(2100)
12
Absorber
Liquor
to
Evaporator
1.7
(230)
13
Centra te
to
Evaporator
1.5
(200)
Stream No.
Figure 5.4-1
Description
Stream
Rate, kg/s
(lb/min)
Water, kg/s
(lU/min)
14 15
Purge Purge
Solids .Solids
to
Dryer
0.17 O.OS3
(23) (11)
i 0.083
(11)
16
First
Effect
Evaporator
Feed
6.7
(890)
17 18 19
Second First Second
Effect Effect Effect
Evaporator Evaporator Evaporator
Feed Overhead Overhead
4.5 4.0 2.6
(590) (530) (350)
20
SO 2
Product
Stream
0.71
(95)
0.071
(9.5)
21
Condensate
6.2
(820)
6.2
(820)
22 23 24
Make-up First Second
Na2CO3 Effect Effect
Slurry Slurry
0.22 2.6 1.8
(30) (350) (240)
0.16
(21)
25
Slurry
to
Dissolving
Tank
4.5
(590)
26
Feed
Solution
CO
Absorber
11
(1500)
^Characteristic values.
Assumptions
a) 3.5% sulfur in coal (dry basi
b) 12% ash (as fired basis)
c) 28 MJ/kg (as fired basis) (12,
d) 92% of sulfur in coal evolves
s)
000 Btu/lb)
as SO 2
e) 99Z removal of particulat.es in ESP prior
f) 902 S02 removal
to scrubbing
g) 75% of ash evolves as fly ash
h) 15% of circulating slurry is
i) double effect evaporator
J) 602 overhead in each effect
k) 10% H20 in product S02 stream
sent to purg
e treatment
-------
Approximately 15 percent of the absorber effluent is
sent to purge treatment to remove sulfates and other impurities
from the system. The purge stream is cooled and centrifuged to
remove 70 percent sodium sulfate and 30 percent sodium sulfite
crystals. The centrate is routed to the evaporators and the crys-
tals are dried. The dryer gases are routed to the prescrubber
to evaporate additional water therein. The dryer gases and evap-
orated water exit through the absorber and stack with the flue
gases.
Water losses in the Wellman-Lord process are due to
evaporation in the prescrubber, prescrubber blowdown, loss in
the product S02 stream, and drying of purge solids. In addition,
the blowdown of the condenser cooling water system has evapora-
tive and drift losses.
5.4.2 Description of Water System
The water systems in the Wellman-Lord sulfite scrubbing
process operate in closed loops. The prescrubbing system recir-
culates a slurry of approximately 5 weight percent solids. This
system requires a small blowdown to maintain the suspended and
dissolved solids at desired levels. The absorber also operates
in a closed loop, with 85 percent of the scrubbing solution being
circulated through double effect evaporators. Regeneration of
the scrubbing solution and production of a concentrated SO 2
stream are thus accomplished. The remaining 15 percent is sent
to purge treatment. The liquor is then either recycled to the
evaporator loop or is evaporated into the dryer combustion gases
which exit through the scrubber and absorber. Thus, the only
effluents from the Wellman-Lord process are the prescrubber
blowdown and the solid by-product. The solid is currently sold
to the paper industry. The prescrubber blowdown is the only
wastewater.
-172-
-------
5.4.3 Slowdown Characteristics
The size of the prescrubber blowdown for the base case
calculation was limited by the particulate loading (Appendix A).
Because 99 percent of the particulates are removed in an ESP
prior to the gases' entering the scrubber, the size of this
stream is relatively small compared to the current power plant
ash sluicing requirement. The quality of this stream will be
similar to that of ash sluicing waters (see Table 5.1-1) with
the addition of high chloride concentrations (about 10,000 mg/A).
The effect on receiving streams and treatment technologies will
be discussed in Section 6.
5.5 Characterization of Wastewaters From the Magnesia
Slurry Absorption Process
A description of the magnesia slurry absorption pro-
cess is given in Section 3.5. The following section presents a
closer examination of the water system. A water balance is given
for a base case condition for one of four equivalent scrubbing
trains applied to a 500 MM power plant. An examination of the
water system indicates that the wastewater streams associated
with the magnesia slurry process are insignificant in comparison
to current power plant wastewaters. The wastewater streams
include: 1) a prescrubber blowdown for suspended and dissolved
solids control, and 2) an intermittent purge from the absorber
loop to remove impurities which enter the system in makeup
streams and as products of oxidation.
5.5.1 Base Case Water Balance
In calculating an example water balance to illustrate
the water flows in the magnesia slurry absorption process, Radian
has chosen a base case for ease of comparison. The base case
-173-
-------
is calculated for one of four equivalent scrubbing systems for a
500 MW power plant burning 3.5 percent sulfur coal with an
average heating value of 23 HJ/kg (12,000 Btu/lb). The detailed
calculations are presented in Appendix A. Figure 5.5-1 illus-
trates the water system. The results of the water balance calcu-
lations are presented in Table 5.5-1. Several of the circulating
flow rates are characteristic flows rather than design flows, as
indicated.
In the magnesia slurry absorption process, flue gas
first enters an ESP where 99 percent of the particulates are re-
moved and sluiced to the ash pond. The hot flue gases then enter
a prescrubber where the remaining particulates and 99 percent of
the chlorides are scrubbed from the gases with water. The
gases are also cooled and saturated by the evaporation of water
from the scrubbing slurry. A blowdown is taken from the cir-
culating scrubbing solution to maintain the desired suspended
and dissolved solids levels. This stream is then routed to
the ash pond. The scrubbed flue gases enter the absorber,
where SOz is removed and absorbed into a circulating magnesia
slurry. The cleaned gases exit the absorber, are reheated, and
exhausted to the stack. No additional evaporation of water
takes place in the absorber. The absorber effluent is recycled.
A bleed stream is sent to regeneration; makeup slurry is added
to the recirculated slurry prior to entering the absorber. The
bleed stream is passed through thickeners to produce a 40 weight
percent solids slurry. It is then thermally treated to release
some water of hydration, and centrifuged. The liquor from each
of these operations is recycled to the slurry tank. The centri-
fuge cake is then dried with the dryer gases being exhausted to
the stack. The magnesia sulfate crystals are calcined to release
S02 and regenerate MgO. The regenerated MgO is recycled to the
slurry tank, while the S02 product gas is sent to S02 conversion.
-174-
-------
Ol
I
TO A3II POND
Figure 5.5-1. Process flow diagram for the magnesia slurry
absorption process.
-------
TABLE 5.5-1.
WATER BALANCE: MAGNESIA SLURRY ABSORPTION PROCESS
(one of four equivalent scrubbing trains)
Stream No.
Figure 5.5-1
Description
Scream
Rate, kg/s
(Ib/mln)
Water, kg/s
(Ib/min)
Stream No.
Figure 5.5-1
Description
Stream
Rate, kg/3
(Ib/mln)
Water, kg/s
(Ib/iain)
1
Gas
to
Scrubber
160
(21,000)
7.6
(1000)
•
14
S/L Underflow
to
Dryer
2.1
(280)
0.12
(15)
2*
Scrubber
Slurry to
Surge
Tank
300
(40,000)
290
(38,000)
15*
Dryer
Gas
to
Stack
17
(2300)
0.71
(94)
3*
Scrubber
Recycle
Slurry
300
(40,000)
290
(38,000)
16
Feed
to
Calclner
1.4
(180)
—
4
Make-up
Water to
Scrubber
8.3
(1100)
8.3
(1100)
17
Recycle
MgO
0.53
(70)
5
Scrubber
Effluent
to
Ash Pond
0.91
(120)
0.83
(110)
18
Make-up
Water to
Slurry
Tank
0.71
(94)
0.71
(94)
6 7* 8* 9
Gas Absorber Recycle Slurry
to Effluent Slurry to S/L
SO 2 to Separator
Absorber Hold Tank
170 440 410 34
(22,000) (58,000) (54,000) (4500)
15 390 360 31
(2000) (52,000) (48,000) (4100)
19
Make-up
MgO
0.037
(5)
10* 11* 12** 13
Make-up Recycle Gas S/L
MgO Slurry to Overflow
Slurry to Air to
S02 Absorber Heater Slurry Tank
32 430 170 32
(4300) (57,000) (22,000) (4300)
29 390 15 32
(3900) (52,000) (2000) (4300)
!
*These values are approximate characteristic stream values.
**Thls value is calculated based on gas flow rates in MC-147.
Assumptions:
a) 105% stoichiometrlc magnesia
b) 3.5% sulfur in coal (dry basis)
c) 12% ash (as fired basis)
d) 28 MJ/kg (as fired basis)(12,000 Btu/lb)
e) 75% of ash evolves as fly ash
f) 92% of sulfur in coal evolves as SOj
g) 99X removal of particulates in ESP prior to scrubbing
h) 90% S02 removal
i) pond evaporation equals rainfall
j) first stage S/L separator produces a 403! solids slurry
k) centrifuge produces a 95% solids cake
1) 102 solids recirculating slurry
-------
Water losses in the magnesium slurry process are due
to evaporation in the prescrubber, prescrubber blowdown, and
drying of solids in the regeneration loop.
5.5.2 Description of the Water System
The water systems in the magnesia slurry absorption
process operate in closed loops (MC-076). The prescrubbing
system recirculates a slurry of approximately 5 weight percent
solids. This system requires a small blowdown to maintain the
suspended and dissolved solids at desired levels. Most of the
absorber effluent is recirculated; the regenerative bleedstream
operates in closed loops. The liquor from the thickeners and
centrifuge are recycled to the slurry tank. Water lost in
drying the crystals exits as a vapor through the stack.
Thus, the only continuous wastewater associated with
the magnesia slurry process is the prescrubber blowdown.
A second possible wastewater is an intermittent purge
from the absorber loop. Impurities enter the system in makeup
water and makeup MgO, and are concentrated. In addition,
sulfite ion is oxidized to sulfate. .Because of developmental
status of this process, little information is available to
predict the effect of long-term build-up of MsSCu. McGlammey,
et al (MC-076) have estimated that a purge of approximately
63 cm3/s (1 gal/tain) for the base case might be necessary.
These authors have suggested several possible treatment tech-
nologies which will be discussed in detail in Section 6.
-177-
-------
5.5.3 Slowdown Characteristics
The size of the prescrubber blowdown for the base case
calculation was limited by the particulate loading (Appendix A).
Because 99 percent of the particulates are removed in an ESP
prior to the gases entering the scrubber, the size of this
stream is relatively small compared to the current power plant
ash sluicing requirement. The quality of this stream will be
similar to that of ash sluicing waters (see Table 5.1-1) with
the addition of high chloride concentrations (about 10,000 mg/£).
The effect on receiving streams and treatment technologies will
be discussed in Section 6.
5.5.4 Purge Characteristics
The impurities in the purge will come from makeup water
and makeup MgO. The MgO can be expected to contain silica,
ferric oxide, alumina, chloride, sulfate, and calcium oxide
impurities. Makeup water quality may vary considerably, but may
contain calcium oxide, sulfate, chloride and trace metal impuri-
ties. McGlammery, et al (MC-076) assume that the purge will be
a clarified solution of approximately 1.2 percent MgSOs and 15
percent MgSCs .
5 .6 Characterization of Wastewaters from the Double
Alkali Wet Scrubbing Process
A description of the double alkali wet scrubbing pro-
cess is given in Section 3.6. The following section presents a
closer examination of the water system. A water balance is given
for a base case condition for one of four equivalent scrubbing
trains applied to a 500 MW power plant. An examination of the
-178-
-------
water system shows that there are no continuous wastewater
streams associated with the process. If a prescrubber is used
for chloride removal, there will be a wastewater stream consist-
ing of prescrubber blowdown. This stream would contain par-
ticulates and chlorides, but is insignificant in comparison to
current wastewater streams from power plants.
5.6.1 Base Case Water Balance
In calculating an example water balance to illustrate
the water flows in a double alkali wet scrubbing system, Radian
has chosen a base case for ease of comparison. The base case
is calculated for one of five equivalent scrubbing systems for a
500 MW power plant burning 3.5 percent sulfur coal with an
average heating value of 28 MJ/kg (12,000 Btu/lb). The detailed
calculations are presented in Appendix A. Figure 5.6-1 illus-
trates the water system. The results of the water balance
calculations are presented in Tables 5.6-1 and 5.6-2 for lime
and limestone regenerant, respectively. Several of the circu-
lating flow rates are characteristic flows and not design flows.
In the double alkali wet scrubbing process, flue gases
containing water first enter an ESP where 99 percent of the
particulates are removed and sluiced to the ash pond. The hot
flue gases then enter the absorber, where S02 is removed and
absorbed into a circulatory sodium sulfite solution. The gases
are cooled and saturated by evaporation of water from the
scrubbing slurry. This saturation may occur in a presaturation
section of the absorber. The cleaned gases exit the absorber,
are reheated and exhausted to the stack. A bleed stream is
taken off the absorber effluent and sent to a reaction tank.
There either lime or limestone regenerant is added to form the
-179-
-------
oo
o
STACK QAS
MAKE-UP i MAKE-UP
ALKALI WATER
Note: Streams
> not shown.
Figure 5.6-1. Process flow diagram for double-alkali wet scrubbing,
-------
TABLE 5.6-1.
WATER BALANCE: DOUBLE ALKALI WET SCRUBBING,
LIME REGENERANT
(One of four equivalent scrubbing trains)
Stream No.
Figure 5.6-1
Description
6
Gas
to
Absorber
7
Abaorber
Liquor
Product
8
Recycle
Liquor
9
Bleed
Stream
to
Reaction
Tank
10
Make-up
Liquor
11
Feed
Liquor
to
Absorber
12
Gas
to
Reheat
13
Lime
Stream 160 25Q 220 3Q 36 25Q
Rate, kg/s (21,000) (33,000) (29,000) (3900) (4800) (33,000) (22,000) (280)
(Ib/min)
Water, kg/s 7.6
(Ib/min) (1000)
1.5
(200)
Stream No.
Figure 5.6-1
Description
Stream
Rate, kg/s
(Ib/min)
Water, kg/s
(Ib/min)
14
Slurry
to
S/L
Separator
32
(4200)
30
(4000)
15
S/L
Overflow
to
Make-up
System
30
(3700)-
16
S/L
Underflow
to
Vacuum
Filtration
3.9
(520)
2.4
(320)
17
Filter
Cake
Wash
Water
2.3
(300)
2.3
(300)
18
Solid
Waste
2.6
(350)
1.1
(150)
19
Centrate
to
Make-up
System
3.6
(470)
20 21*
Sulfate Make-up
Purge Water
and
Alkali
4.9
(650)
4.9
(650)
* 0.027 kg/s (3.6 Ib/min) alkali
Assumptions:
a) 105% lime stoichioraetry
b) 3.5% sulfur in coal (dry basis)
c) 12% ash (as fired basis)
d) 28 MJ/kg (as fired basis)(12,000 Btu/lb)
e) 92% of sulfur in coal evolves as S02
f) 99% removal of particulate in ESP prior to scrubbing
g) 90% SOZ removal
h) pond evaporation equals rainfall
i) 75% of ash evolves as fly ash
j) first stage S/L separator produces 40% solids underflow
k) second stage S/L or ponding produces 60% solids sludge
1) solid waste washed with 2 displacement washes
-181-
-------
TABLE 5.6-2. WATER BALANCE: DOUBLE ALKALI WET SCRUBBING,
LIMESTONE REGENERANT
(One of four equivalent scrubbing trains)
Scream No.
Description
Stream
Rate, kg/s
(Ib/min)
Mater, kg/s
(Ib/miti)
6 7
Gas Absorber
to Liquor
Absorber Product
160 250
(21,000) (33,000)
7.6
(1000)
8 9 10 11 12
Recycle Bleed Make-up Feed Gas
Liquor Stream Liquor Liquor to
to to Reheat
Reaction Absorber
Tank
210 32 39 250 170
(28,000) (4300) (5200) (33,000) (22,000)
15
(2000)
13
Limestone
4.3
(570)
3.0
(400)
Stream Ho.
Description
Stream
Rate, kg/s
{ Ib/min)
Water, kg
(Ib/min)
* 0.027 kg/s
14 15
Slurry S/L
to Overflow
S/L to
Separator Make-up
System
36 33
(4800) (4300)
35
(4600)
(3.6 bl/min) alkali
Assumptions:
16 17 18 19 20
S/L Filter Solid Centrate Sulfate
Underflow Cake Waste to Purge
to Mash Make-up
Vacuum Water System
Filtration
4.2 2.5 2.9 3.8
(550) (330) (390) (500)
2.5 2.5 1.3
(330) (330) (170)
a) 120% limestone stoichiometry
b) 3.5% sulfur in coal (dry basis)
c) 12% ash (as fired basis)
d) 28 MJ/kg (as fired basis) (12,000 Btu/lb)
e) 92% of sulfur in coal evolves as S02
f) 99% removal of particulate in ESP prior to
g) 90% SOj removal
h) pond evaporation equals rainfall
i) 75% of ash evolves as fly ash
21*
Make-up
Alkali
3.3
(430)
3.3
(430)
scrubbing
j) first stage S/L separator produces 40% solids underflow
If.) second stage S/L or ponding produces 60% solids sludge
1) solid waste washed with 2 displacement washes
-182-
-------
solid waste and to regenerate the scrubbing solution. The
reaction tank slurry is sent to a solid/liquid separator,
where the supernatant is recycled to the makeup system. The
40 percent solids underflow is sent to a vacuum filter. The
filter cake is washed to recover soluble sodium. The filtrate
and wash water are recycled to the makeup system. The final
filter cake is 60 percent solids. The main portion of the
absorber effluent is combined with regenerated liquor and
recirculated as absorber feed.
Thus, water losses in the double alkali wet scrubbing
process are due to evaporation in the absorber and occlusion in
the solid waste. If a prescrubber is used, prescrubber blowdown
would be an additional water loss.
5.6.2 Description of the Water Systems
The water systems in the double alkali wet scrubbing
process operate in closed loops. If a prescrubber is required,
this system would require a small blowdown to maintain the
suspended and dissolved solids at desired levels. Approximately
90 percent of the absorber effluent is recirculated. A bleed
stream is regenerated with lime or limestone in a reaction tank,
but all supernatant liquors from the solid liquid separations
are recycled to the makeup system. The water occluded in the
solid waste can have an impact on surrounding water quality in
some cases, but this aspect of the FGD-associated water impacts
is being addressed in the report on solid waste impact (RO-359).
-183-
-------
Thus, there are no continuous wastewaters associated
with the double alkali system in normal operation. Scrubber
liquor would have to be treated for sulfate removal, as discussed
in Section 6.2. In some cases, a purge may be necessary to
control the concentration of nonsulfur/calcium species in the
system. Such a purge should not be required in normal operation,
however. As discussed earlier, a prescrubber may be required
under certain conditions, and would necessitate a blowdown
stream.
5.6.3 Purge Characteristics
The sulfate purge characteristics are the same as
those of the recirculated clarifier overflow. The purge con-
tains sodium sulfate, sodium sulfite, and soluble nonsulfur/
calcium species in varying amounts. The nonsulfur/calcium "
species enter the system in the lime and/or limestone, and in
the makeup water. In systems having common prescrubber and
absorber circulating loops, these species enter with the fly
ash and flue gas. The soluble nonsulfur/calcium contaminant in
highest concentration is probably sodium chloride. Contamina-
tion results from the absorption of HC1 from the flue gas.
5.6.4 Blowdown Characteristics
If a prescrubber is required for chloride removal,
the size and quality of the blowdown stream would be comparable
to that of blowdowns from the Wellman-Lord and Magnesia Slurry
systems.
-184-
-------
5.7 Characterization of Wastewaters from the Physical
Coal Cleaning Process
The physical coal cleaning plants vary widely in types
of processes used and in plant layout. Generally, water systems
in new plants operate in closed loops (LE-218). Solid refuse is
allowed to settle in ponds and the pond water is recycled for
reuse in the processing areas. Some water is occluded in the
solid refuse, while some is lost due to thermal drying operations
or increased coal moisture content.
Coal cleaning by the wet process method uses about
.0063 to .0083 m3 of water per kg of coal (1,500 to 2,000 gallons
of water per ton of coal processed) . Contaminants consist of
suspended solids, which are chiefly fine clay and coal, and
dissolved solids, which may contain iron, aluminum, calcium,
magnesium, sodium, and potassium. Water effluents may also
contain surface-active organic compounds such as alcohols or
kerosene, which are added in some coal cleaning plants to
enhance frothability in the process. In a modern plant, all
liquid waste streams are routed to holding ponds to allow
settling of the suspended solids. The clear supernatant liquid
is then recycled to the process. Thus, no liquid effluents
result directly from the cleaning process (LE-218). Water
runoff from refuse piles, however, may contribute to water
pollution. Water contaminants in refuse pile runoff include
sulfuric acid, sulfates, manganese, and iron in varying concen-
trations. For example, in effluents from four different refuse
sites, the concentration of sulfates varied from 690-9500 tng/l;
of manganese, 3.5 to 120 rngA; and of iron, 6.2 to 3400 mgM
(MA-411).
-185-
-------
5. 8 Characterization of Wastewaters from SO2 Conversion
Processes
The Allied Chemical Process for elemental sulfur
production has insignificant water systems which operate in
closed loops. However, the production of sulfuric acid requires
significant quantities of cooling water.
Elemental Sulfur Production
The water system in the Allied Chemical Process in-
cludes steam generation in the sulfur condenser, and cooling
water for compressor seals. Both of these systems have relatively
small requirements (as discussed in Section 4). The blowdown
from these systems is expected to be negligible in quantity.
The quality will be similar to that of cooling tower blowdown
streams or steam system blowdowns. The Allied Process, as
presented by Allied (HU-051), appears to have no liquid effluents
when operated on an SOa stream from an FGD process.
Sulfuric Acid Production
Product acid coolers require significant quantities
of cooling water. The effluent stream would be a blowdown
similar to those described in regard to power plant cooling
tower blowdowns. However, the quantities required are one to
two orders of magnitude less than those required by the power
plant. The recirculating water flow for sulfuric acid cooling
is about 0.32 m3/s (500 gpm), compared to about 13 m3 /s
(210,000 gpm) for the condenser cooling system of the base case
500 MW power plant. The blowdown rate typically ranges between
0.5 and 3.0% of the recirculating water flow (DO-051).
-186-
-------
6-° EXAMINATION OF PURGE CHARACTERISTICS AND APPLICABLE
TREATMENT TECHNOLOGY
This section examines the FGD-associated wastewater
streams in detail. This examination includes a characteriza-
tion of the effluent composition, an assessment of effect on
receiving stream water quality, and possible treatment technolo-
gies. The specific wastewater streams addressed in this sec-
tion include:
Magnesia Slurry Absorption Process Purge,
Double Alkali Sulfate Purge,
Prescrubber Slowdown,
Cooling Water System Slowdown,
Possible Lime/Limestone Purge, and
Lime/Limestone/Double Alkali Solid Waste.
6.1 Magnesia Slurry Absorption Process Purge
The magnesia slurry scrubbing system operates in a
closed loop, and is subject to buildup of impurities that could
lead to corrosion and scaling problems. Impurities enter the
system in makeup water and makeup MgO. In addition, some MgS03
is oxidized to MgSO,,. Because of the developmental stage of
the process there is little information available to predict
the effect of buildup of MgSO^ in continuously operated, closed
loop cycles. Some sulfate removal may prove necessary.
6.1.1 Purge Characteristics
The impurities in the purge will come from makeup
water and makeup MgO. The MgO can be expected to contain
-187-
-------
silica, ferric oxide, alumina, chloride, sulfate, and calcium
oxide impurities. Makeup water quality may vary considerably,
but may contain calcium oxide, sulfate, chloride and trace
metal impurities. McGlammery, et al (MC-076) assume that the
purge will be a clarified solution of approximately 1.2 percent
MgS03 and 15 percent MgSCK .
6.1.2 Effect on Receiving Streams
This purge would not be discharged directly to a
receiving stream. Economics make recovery of MgSCU desirable.
Furthermore, the Federal Water Pollution Control Act Amendment
of 1972 and the discharge limitations based on best practical
control technology available in 1977 and best available technology
economically achievable in 1983 would prohibit direct discharge.
6.1.3 Treatment Technology
McGlammery et al (MC-076) suggested three purge
treatment techniques:
1) sending a sidestream to a deadend pond;
2) concentration of the mother liquor until
MgS04 precipitates before sending to a
deadend pond, or treating;
3) "dissolving MgS03*6H20 slurry with a minimum
amount of sulfur dioxide, filtering the in-
soluble impurities, then reprecipitating sul-
fite with makeup MgO. The resultant crystals
would be filtered and returned to the system;
the mother liquor would be evaporated to
-188-
-------
recover MgSO*. and the supernatant of soluble
impurities discarded."
The first technique would be the simplest and probably
least expensive purge treatment. Evaporation ponds have been
used in water management throughout the chemical and electric
utility industries. Solar energy is used to evaporate water
from the pond. Dissolved solids in the waste stream fed to the
pond are, thus, concentrated and precipitated. The precipitated
dissolved solids are usually allowed to accumulate at the bottom
of the pond, but in some instances are periodically dredged from
the pond. The pond may have to be lined to prevent seepage of
dissolved chemicals into underground water supplies. If so
lined, there will be no effluents from the pond. Applicability
of evaporation ponds depends on the net evaporation rate (the
gross evaporation rate minus rainfall). Geographical areas with
less than 50 cm (20 in.) net evaporation rate are not suitable
for ponds. Water in a purge stream treated in an evaporation
pond is lost by evaporation; and is thus not available for reuse
in the system. Magnesium contained in the purge is also lost
from the system (EN-392, EN-394, MC-076).
Magnesium losses could be reduced by the second method,
concentrating the mother liquor until MgSCU precipitates. Evap-
oration should not be allowed to proceed beyond the point where
undesirable impurities such as NaCl precipitate with the MgSCK
(MC-076). This procedure alone would not remove insoluble im-
purities such as silica, ferric oxide, aluminum oxide, and fly
ash. After evaporation, however, the mother liquor can be sent
to a deadend pond (as described above) or treated by conventional
processes such as vapor compression distillation, reverse osmosis,
or softening-ion exchange. These processes will be discussed in
Section 6.7.
-189-
-------
A possible complete treatment of the purge could be
accomplished by method 3. The supernatant of soluble impurities
could be treated by conventional water treatment techniques such
as reverse osmosis, vapor compression distillation, flash evap-
oration, or softening-ion exchange. These processes will be dis-
cussed in Section 6.7. Pretreatment would be necessary before
reverse osmosis or ion exchange processes. For reverse osmosis
pretreatment is needed to reduce the concentration of calcium
and magnesium to prevent scaling on the membrane. The concen-
tration of impurities should be reduced before treatment by ion
exchange to avoid frequent regeneration of the resin.
6.2 Double Alkali Sulfate Purge
In double alkali systems, some of the sulfer removed
from the flue gas takes the form of soluble sodium sulfate because
of oxidation in the system. Some of the active sodium (sodium
associated with anions involved in SO2 absorption reactions, in-
cluding sulfite, bisulfite, hydroxide, and carbonate/bicarbonate)
is thus converted to an inactive state which does not take part
in S02 absorption. Converting NaaSCK back to active sodium is
relatively difficult. To do so, sulfate ion must be removed from
the system while sodium is left in solution. Alternatively, the
sodium sulfate may be removed from the system at the rate it is
being formed. This alternative is not desirable since it wastes
sodium. Furthermore, removal would generally be achieved by
purging the sodium sulfate from the system in the liquor which
is occluded in the wet solid waste product. The sulfate could
then be leached from the waste. Water runoff can lead to con-
tamination of surface water, and leaching and percolation of the
leachate into the soil can result in contamination of the ground-
water in the vicinity of the disposal site. Failure to allow
for sulfate removal will ultimately result in: 1) precipitation
-190-
-------
of sodium sulfate in the system if active sodium (Na2C03 or NaOH)
is made up, or 2) eventual deterioration of the S02 removal capa-
bility due to loss of active sodium, if no makeup is added.
When low sulfur coal is burned, the ratio of oxygen
to SO2 is greater than it is when high sulfur coal is burned.
Thus, a greater percentage of sulfite is oxidized to sulfate
with low sulfur coal and a larger purge stream must be treated
for sulfate removal.
Soluble nonsulfur/calcium species can also concentrate
in the system and could be controlled by a purge. If the solid
waste product is washed with excessive amounts of fresh makeup
water to recover the potentially leachable sodium salts, the
solubles can concentrate. Limited waste product washing provides
these nonsulfur/calcium solubles with an exit through occlusion
in the solid waste. As more sodium is recovered, more solubles
are recovered. A purge may be necessary to maintain a desired
level of concentration (KA-227).
6.2.1 Purge Characteristics
The sulfate purge characteristics are the same as
those of the recirculated clarifier overflow. The purge con-
tains sodium sulfate, sodium sulfite, and soluble nonsulfur/
calcium species in varying amounts. The nonsulfur/calcium
species enter the system in the lime and/or limestone, and in
the makeup water. In systems having common prescrubber and ab-
sorber circulating loops, these species enter with the fly ash
and flue gas. The soluble nonsulfur/calcium contaminant in
highest concentration is probably sodium chloride. Contamina-
tion results from the absorption of HC1 from the flue gas.
-191-
-------
6.2.2 Effect on Receiving Stream
This stream could not be discharged directly to
receiving streams due to the high levels of sodium, sulfate,
sulfite, and other soluble species.
6.2.3 Treatment Technology
Several methods for sulfate removal in double alkali
systems are discussed by Kaplan (KA-227) . These methods are:
1) precipitation of sulfate as
with the addition of lime (This method
applies only to dilute double alkali
systems) ;
2) co-precipitation of sulfate with calcium
sulfite in a mixed crystal or solid
solution;
3) addition of sulfuric acid;
4) formation of t^SO^ in an electrolytic
cell;
5) limitation of oxidation.
With the first two methods, sulfates are removed in the regener-
ation operation and a sulfate purge is not required. With. methods
(3) and (4), the purge treated for sulfate removal can be re-
turned to the system. With method (5), the purge is discharged
with the solid waste.
-192-
-------
If the active sodium concentration in the system is
sufficiently low (less than about 0.15 M) , sulfate can be re-
moved from the system by precipitation as gypsum with the addi-
tion of lime, according to Equation 6.2-1:
Na2SCK + Ca(OH)2 + 2H20 -> 2NaOH + CaS(K-2H20 (6.2-1)
Because the solubility product of calcium sulfite is much lower
than that of calcium sulfate, this reaction does not occur to a
significant extent in the presence of high concentrations of
sulfite ions. Thus, this method is not applicable for concen-
trated double alkali systems. This method for sulfate removal
has been.demonstrated in the dilute double alkali process devel-
oped by General Motors Corporation. This process has been ap-
plied to cleaning flue gas from coal-fired industrial boilers
at the 30-40 MW (equivalent) level. The treated effluent is a
clear solution of NaOH/Na2S03 which is recycled to the absorber.
The gypsum solid wastes are disposed of.
In concentrated systems, with a concentration of active
alkali greater than about 0.15 M, sulfates cannot be removed to
a significant degree by Equation 6.2-1. Under certain conditions,
however, sulfate is co-precipitated with calcium sulfite in a
mixed crystal or solid solution. This phenomenon has been de-
scribed by Borgwardt as it applies to lime/limestone systems.
Similar observations have been made by A. D. Little in their
double alkali investigations. Enough sulfate can reportedly be
removed by co-precipitation to allow successful operation without
a sulfate purge with system oxidation rates as high as 20% of
the SO2 absorbed. The treated effluent is a clear solution of
NaOH/Na2S03 which is recycled back to the absorber. The solid
wastes are disposed of.
-193-
-------
In the third method (Equation 6.2-2), sulfuric acid
is added to dissolve calcium sulfite, increasing the concentra-
tion of calcium ions in solution enough to exceed the solubility
product of calcium sulfate.
(6.2-2)
4- 2CaS03-%H20 + H2SCU + 3H20 -»• 2NaHS03 + 2CaSCU-2H20
This method requires more sulfuric acid than the stoichiometric
amount indicated in Equation 6.2-2 since any material that func-
tions as a base (such as unreacted lime or limestone) can consume
sulfuric acid. Because of the high sulfuric acid consumption,
this method may be uneconomical in applications with high oxida-
tion rates. The sulfuric acid addition method is used for sul-
fate removal in full scale double alkali applications in Japan.
It has also been pilot tested by A. D. Little in the U.S. The
liquid effluent produced is a clear solution of sodium bisulfite
which is recycled back to the main regeneration reactor. The
gypsum solids are sold in Japan, but would probably be landfilled
in the U.S.
The fourth method uses electrolytic cells developed by
Ionics to remove sulfates as sulfuric acid and sodium hydroxide,
according to Equation 6.2-3:
Electrolytic
Col 1
NazSOo + 3H20 - z£±± - ^ 2NaOH + EzSO^ + Ez + %02 (6.2-3)
The technology for the cells has been relatively well developed
by Ionics. In Japan, Kur eh a /Kawasaki have pilot tested the
Yuasa/Ionics electrolytic process for sulfate removal with their
double alkali process, and feel that it may be less expensive
than the sulfuric acid addition method. Energy consumption for
this method is relatively high, however. The sodium hydroxide
solution produced can be recycled to the absorber. The sulfuric
-194-
-------
acid produced will be dilute. If no use for it can be found, it
can be neutralized with lime and the resulting gypsum disposed
of by landfill (KA-227, OT-R-051).
Another approach to sulfate control may be to limit
oxidation by process and equipment design. If a sufficiently
low level of oxidation can be maintained, it may be possible to
remove sulfates by a small purge of sodium sulfate with the solid
waste product. This alternative, however, would increase the
potential for leaching from the solid waste (KA-227).
If a purge is necessary to maintain a desired level of
soluble nonsulfur/calcium species, the remaining constituents of
the double alkali purge can be removed with developed water
treatment technology. Aerospace Corporation has reported that
lime-soda softening reduces the concentrations of all constituents
except soluble sodium and chloride salts in lime and limestone
recirculating liquors (BO-203). Reverse osmosis can be used to
further treat the water to remove the sodium chloride salts and
other undesirable constituents. This results in a high quality
product water stream with a low concentration of dissolved solids.
Aerospace also reported that other existing treatment processes
such as multistage flash evaporation, vapor compression distil-
lation, and softening-ion exchange could be used instead of soft-
ening and reverse osmosis, but would generally be more expensive
(BO-203). These treatment processes will be discussed in Section .
6.7.
6.3 Prescrubber Slowdown
The Wellman-Lord Sulfite Scrubbing Process and the
Magnesia Slurry Absorption Process require that chlorides and
particulates are removed from the flue gases prior to entering
the absorber. (With high chloride coals, a prescrubber may also
-195-
-------
be required for Double Alkali.) The prescrubber operates in a
closed loop and requires a small blowdown to maintain desired
levels of suspended and dissolved solids. An ESP is assumed to
have removed 99 percent of the fly ash prior to entering the
prescrubber. The remaining 1 percent of the fly ash and 100
percent of the chlorides are removed in the prescrubber. In the
base case calculation (Section 4) the blowdown from the pre-
scrubber was sized by the suspended solids limitation of a 5
percent solids slurry. The chlorides concentration was approxi-
mately 10,000 mg/£ (ppm).
Composition of the coal would have some effect on the
quantity and composition of the blowdown stream. When coal with
a high chloride content is burned, either the concentration of
chlorides in the blowdown would be increased, or the amount of
blowdown would be increased to keep the chloride concentration
at a lower level. Coals with a high ash content may require an
increased blowdown to maintain a low level of suspended solids,
but since only 1 percent of the fly ash is assumed to be removed
in the prescrubber, the change in quantity of the blowdown stream
would be very slight.
6.3.1 Blowdown Characteristics
The prescrubber blowdown contains approximately 50,000
mg/£ (ppm) suspended solids, and 10,000 mg/Jl (ppm) chlorides in
the base case. Because many soluble inorganic salts can be
leached from the fly ash, there are a wide variety of trace ele-
ments. These trace elements vary widely with coal type, and no
typical analysis is practical. It can be assumed, however, that
the constituents of the prescrubber blowdown stream will be equiv-
alent to the constituents of the fly ash sluicing stream, with
the significant addition of chlorides. The prescrubber blowdown
is approximately 1 percent of the fly ash sluicing requirement.
-196-
-------
6.3.2 Effect on Receiving; Stream
The prescrubber blowdown stream is a wastewater source
of high dissolved chloride concentration. The other components
in the stream are of an equivalent nature, much reduced in
quantity as compared to the fly ash sluice water. Thus, the
major impact of this stream is the high chlorides concentration.
The normal treatment procedure in current use is to route this
stream to the ash pond to allow the suspended solids residence
time to settle. The chlorides are diluted from 10,000 mg/£
to approximately 70 mg/2- (if an ash pond overflow of 2,000 gpm
is assumed) , and discharged to a receiving stream. The effect
on receiving stream water quality is site specific and depends
on the nature of the receiving stream water quality and flow
parameters . Adverse impact is measured in relation to current
regulations. These are set to maintain current water quality.
The USGS Stream Classifications, given in Table 6.3-1, give an
indication of the effect on the biosystems in the receiving
streams.
TABLE 6.3-1. USGS STREAM CLASSIFICATIONS
Classification Total Dissolved Solids
(ppm)
Fresh 0-1,000
Slightly Saline 1,000-3,000
Moderately Saline 3,000-10,000
Very Saline 10,000-35,000
Briny > 35, 000
The impact of the ash pond overflow containing the di-
luted prescrubber blowdown is seen to be minimal, but certain
-197-
-------
receiving stream regulations may require treatment before dis-
charge. In some systems, ash pond overflow is reused for ash
sluicing instead of being discharged.
6.3.3 Treatment Technology
As the nation approaches a 1983 goal of zero discharge,
recirculating systems will become predominant. In this case, the
prescrubber blowdown stream will probably be treated separately,
rather than sent to the ash pond. Currently available treatment
technologies potentially applicable to this stream include reverse
osmosis, vapor compression distillation, flash evaporation, and
softening-ion exchange. These technologies will be discussed in
Section 6.7. Before these technologies could be applied, however,
the suspended solids content of the stream would have to be re-
duced, perhaps by sedimentation or filtration.
6.4 Cooling Water System Blowdown
The Wellman-Lord Sulfite Scrubbing Process, the
Allied Chemical SOa Reduction Process, and the sulfuric acid
production plant require cooling water. The most significant
requirement is for condensation of evaporator overhead in the
Wellman-Lord Process. The sulfuric acid plant requires cooling
water for product acid cooling. The SO a gas stream in the
Allied Chemical Process must be compressed, and cooling water is
required for compressor seal cooling. This requirement is in-
significant.
The circulating cooling water requirements, as defined
in Section 4 for the application of FGD systems to the base case
500 MW power plant, are:
-198-
-------
1) Wellman-Lord condenser cooling water 1.3 m3/s
(20,000 gpm)
2) Sulfuric acid product cooling, o 32 m3/s
and (5,000 gpm)
3) Compressor seal cooling, elemental 0 019 m3/s
sulfur process (300 gpm)
For comparative purposes, the circulating cooling
water requirement for the condenser cooling system of the
base case 500 MW power plant is 13 m3/s (210,000 gpm).
6.4.1 Purge Characteristics
Cooling water systems can operate with once-through
or recirculatory flows. Due to the nature of the once-through
system, the chemical composition of the effluent water is
essentially equivalent to that of the influent water. Recir-
culating cooling systems employ cooling devices such as cooling
towers, spray ponds/canals, etc., which allow the reuse of
cooling water. These devices promote cooling primarily by
evaporating a portion of the recirculating water flow. Impuri-
ties and contaminants that come into the system with makeup
water and other sources become concentrated in the system. A
blowdown stream is withdrawn from the system to control the con-
centration of impurities and contaminants.
Soluble chemical species brought into the system by the
makeup water are typically concentrated to levels ranging from
1,500 to 10,000 mg/£ (ppm). The chemical species contributing to
the salinity of the blowdown is primarily determined by the makeup
water. Chemical treatments to control corrosion, scale, biolog-
ical fouling and solids deposition also impact the blowdown water
quality. The intimate contact between air and water in the cool-
ing devices enables particulate matter and soluble gases to be
-199-
-------
scrubbed from the air. Up to 80 percent of the suspended solids
in recirculating systems are estimated to have originally come
into the system as airborne particulates (GL-028).
6.4.2 Effect on Receiving Streams
The quality of the FGD-associated cooling water stream
is similar to that of the power plant cooling water system. The
FGD-associated cooling water requirements are very small compared
to power plant cooling requirements (less than 10%). The blowdown
from the FGD cooling systems could thus be treated and discharged
along with the power plant cooling water blowdown, and would not
have a significant impact on receiving stream water quality.
6.4.3 Treatment Technology
Cooling water system blowdown is normally sent to a hold
pond to allow residence time for suspended solids settling. In
current practice, the supernatant is then discharged to a receiving
stream without further treatment for removal of dissolved solids.
As the nation approaches the 1983 goal of zero discharge, recircu-
latory systems will predominate. Such treatment technologies as
reverse osmosis, vapor compression distillation, flash evaporation,
and softening-ion exchange may be used to treat small blowdowns for
dissolved solids removed. These treatment technologies are dis-
cussed in Section 6.7.
6.5 Possible Lime/Limestone Purge
Under normal operating conditions no lime/limestone
system purge should be necessary. Normal operating conditions
are defined by design criteria, and include characteristics of
the boiler, coal, hardware, absorbent, and disposal facilities.
Under normal conditions, water is lost by evaporation in the
scrubber, and occlusion in the solid waste. Normally, sufficient
-200-
-------
dissolved solids are occluded in the solid waste to maintain
recirculated dissolved solids at desired concentrations.
In a recent study, Aerospace defined four situations
which might require a purge of scrubbing liquor to allow
.sufficient makeup of fresh water to avoid scaling (BO-203).
These include:
1) operation below some critical flue gas mass
load level (created by low boiler loading);
2) use of lower sulfur coal than that for which
the system was designed;
3) oxidation of sulfite sludge to a sulfate
sludge, with more liquor recovered due to
greater dewatering efficiency; and
4) need for a quick blowdown due to catastrophic
circumstances.
Aerospace estimated that a purge might become neces-
sary in the range of 50 percent of the maximum design criteria.
During start-up or shutdown, a temporary excess liquor problem
can usually be handled by the excess capacity of the disposal
pond. Operator error can also result in a purge stream.
6.5.1 Purge Characteristics
The possible purge from the lime/limestone system
would have the same quality as the recirculated clarifier super-
natant. This liquor differs from system to system due to coal
type, fly ash collection facilities, and scrubber operation.
-201-
-------
The liquor is saturated with dissolved calcium sulfite and/or
calcium sulfate salts. Sodium and chloride ions are also present
in high concentrations. Most trace elements and toxic species
are controlled below 1 mg/£ (ppm) (BO-203). Table 6.5-1 shows
the range of constituents in four scrubber liquors sampled by
Aerospace Corporation.
6.5.2 Effect on Receiving Streams
Aerospace Corporation has concluded that purged scrubber
liquor must be treated if it is to be discharged to a receiving
stream (BO-203). This statement was based on data from "Disposal
of By-Products from Nonregenerable Flue Gas Desulfurization Sys-
tems", which is a study conducted under EPA Contract 68-02-1010.
A secondary reason for not allowing direct discharge of these
liquors is the Federal Water Pollution Control Act Amendments of
1972. Requirements for 1977 and 1983 discharge limitations, based
on best practical control technology available and the best avail-
able technology economically available, also prohibit direct dis-
charge.
6.5.3 Treatment Technology
Aerospace Corporation has studied the feasibility of
using available and developmental water treatment technologies
applicable to scrubber liquors. They have concluded that lime-
soda softening would render the scrubbing liquors acceptable for
power plant general services use, but the liquor would still con-
tain a high concentration of dissolved solids. Lime-soda soft-
ening followed by reverse osmosis would make the liquor suitable
for public water supply. The softening step must precede treat-
ment by reverse osmosis because the high concentration of calcium
in the purged liquor would cause scaling of the membrane. Other
-202-
-------
TABLE 6. 5-1, RANGE OF CONCENTRATION OF CONSTITUENTS
IN SCRUBBER LIQUORS STUDIED
Constituents
Cadmium
Calcium
Chromium (total)
Cobalt
Copper
Iron
Lead
Magnesium
Manganese
Mercury
Molybdenium
Nickel
Potassium
Selenium
Silicon
Silver
Sodium
Tin
Vanadium
Zinc
Carbonate
Chloride
Fluoride
Sulfite
Sulfate
Phosphate
Nitrogen (total)
Chemical oxygen demand
Total dissolved solids
Total alkalinity (as CaCOj)
Conductance mho/cm
Turbidity, Jackson Units
pH
Range of constituent
concentrations
at potential
discharge points3
Aluminum
Antimony
Arsenic
Beryllium
Boron
0.03
0.09
<0.004
<0,002
8.0
0.3
2.3
0.3
0.14
46.
0.004
520.
0.01
0.10
<0.002
0.02
0.01
3.0
0.09
0.0004
0.91
0.05
5.9
<0.001
• 0.2
0.005
14.0
3.1
<0.001
0.01
420.
0.07
0.8
720.
0.03
<0.001
60.
3200.
41.
0.003
3.04
0.11
3000,
0.5
0.7
0.2
8.1
0.4
2750.
2.5
0.07
6.3
1.5
32.
2.2
3.3
0.6
2400.
3.5
0.67
0.35
4800.
10.
3500.
10,000.
0.41
0.002
390.
15,000^
150.
0.015
10.7
Samples obtained from: EPA/TVA, Shawnee, Steam Plant - venturi and
spray tower; EPA/TVA Shawnee Steam Plant - turbulent contact absorber;
Arizona Public Service Cholla Station - flooded disk scrubber and
absorption tower; and Duquesne Light Phillips Station - single- and
dual-stage venturi.
Includes all soluble species.
-203-
-------
existing treatment processes such as flash evaporation, vapor
compression distillation, and softening-ion exchange could be
used instead of lime-soda softening and reverse osmosis, but
were not economically competitive. These treatment processes
are discussed in Section 6.7.
6.6 Lime/Limestone/Double Alkali Solid Waste
The solid waste produced by the nonregenerable FGD
processes does not directly impact receiving stream water quality.
However, the solid waste can impact surrounding water quality
through leaching and percolation of soluble components of the
solid waste. This subject is discussed further in the report
on solid waste impact (RO-359).
6.7 Existing Water Treatment Technologies Applicable to
Wastewater from FGD Systems
Several existing technologies for water treatment have
been identified as potentially being applicable for treatment of
water from FGD processes. These technologies will be discussed
in this section.
6.7.1 Lime-Soda Softening
This process can be used to decrease the concentration
of calcium and magnesium ions in purge liquors from the lime/
limestone, magnesia slurry, and double alkali processes. It can
also be used as a pretreatment step before reverse osmosis or
ion exchange processes. Lime (CaO) and soda ash (Na2C03) added
to the liquor react with the major species in the liquor to pre-
cipitate the calcium and magnesium together with the heavy metals
Carbon dioxide is added to adjust the pH, and the solution is
-204-
-------
filtered and centrifuged. The precipitates are sent to the
disposal site (BO-203). Concentrations of hardness (magnesium
and calcium ions) in the product water can be reduced to 50 mg/£
(expressed as CaC03) (WE-324). An 80 to 90 percent reduction in
the concentrations of As, Cd, Mn, Pb, and Se and a 30 to 90 per-
cent reduction in hexavalent chromium can also reportedly be
achieved with this process. Boron will also be removed to some
extent (BO-203). Other dissolved solids such as sodium and
chlorides will not be removed, however. Thus, the product water
may have to be further treated by reverse osmosis or ion exchange
before it can be recycled to the process. Lime-soda softening
processes are in commercial operation.
6.7.2 Reverse Osmosis
Reverse osmosis could be used to decrease the concen-
tration of dissolved solids in the prescrubber blowdown from the
Magnesia Slurry or Wellman-Lord; cooling tower blowdown from the
processes requiring it; or purge streams from the Magnesia
Slurry, Double Alkali, or Lime/Limestone Processes. Because the
waste stream produced from reverse osmosis would be large in
volume (around 25 percent of the feedwater), another process,
such as vapor compression distillation, would have to be used in
conjunction with reverse osmosis to treat this waste stream.
The purge streams would contain relatively high concentrations
of calcium and/or magnesium ions and would have to be treated
(perhaps by lime-soda softening) to decrease these concentra-
tions before the streams could be treated by reverse osmosis.
Otherwise, the calcium or magnesium salts could cause membrane
fouling.
In reverse osmosis, water is forced through a semi-
permeable membrane which allows the passage of water but prevents
the passage of dissolved solids and other impurities. The
-205-
-------
resulting two products are deionized water and brine concentrate.
The net driving force for flow of water through the membrane is
the total applied pressure less the osmotic pressure of the sys-
tem. There are many inorganic and synthetic organic materials
that possess the property of semipermeability, but cellulose
acetate is the most common membrane material employed (GE-009).
Reverse osmosis membrances are subject to fouling from
many different fouling agents. The most significant of these
agents are:
biological growths
suspended solids or particulate matter
scale
manganese and iron
organics
To control membrane fouling, a feedwater pretreatment step is
often dictated and certain limitations are imposed on the process,
Process feed pretreatment can take several forms depending upon
each particular feedwater.
In addition to being a pretreatment consideration,
scale control also imposes restrictions or limitations on the
reverse osmosis process. Scale formation is driven by super-
saturation of a solution with respect to a chemical salt. Cal-
cium carbonate [CaC03], calcium sulfate [CaSCM, and magnesium
hydroxide [Mg(OH)2] are the primary scaling salts due to their
low solubilities. Therefore, these salts must be kept below
their saturation concentrations in order to avoid precipitation
and scale formation. The conversion of saline water to desalted
water by reverse osmosis is thus limited in the degree of feed-
water concentration that can be achieved. The degree of concen-
tration varies with the quality of feedwater to the process.
-206-
-------
A high degree of concentration may not be possible for wastes
from FGD purge streams because of the high concentrations of
calcium and/or magnesium.
Reverse osmosis units have typically demonstrated 90-
98% salt rejection for feedwater with about 10,000 ppm dissolved
solids (comparable to the concentration in prescrubber blowdown).
Product water TDS levels of less than 200 ppm can be achieved
with one stage, with 75% recovery of the feedwater. Multiple
stage RO systems can be used to achieve lower product TDS levels
(<50 ppm) . Reverse osmosis will also remove more than 95% of
organics and all colloidal particles down to 0.05 microns (DI-149)
The concentrated waste stream from a reverse osmosis unit will be
relatively large in volume. The actual volume will depend on the
amount of feedwater recovered. This waste stream can be further
concentrated by vapor compression distillation or, in some appli-
cations, sent to an evaporation pond.
Reverse osmosis units are commercially-marketed by a
number of companies. They have been used to treat cooling tower
blowdown water to recover deionized water for reuse, and to pro-
duce drinking water from sea water and inland brackish water.
6.7.3 Ion Exchange
Ion exchange is a commercially available water treat-
ment process that has been suggested as a possible means for
treating wastewater for FGD processes (BO-203). Because of the
high concentration of dissolved solids in these streams, however,
treatment by ion exchange may be prohibitively expensive. It
has been estimated that processes such as reverse osmosis would
be more economical than ion exchange for treating water with an
inlet TDS concentration of 1000 mg/£ (ppm) or more (WE-324) .
-207-
-------
The wastewater from FGD would generally have more than 10,000
mg/£ (ppm) of IDS.
Ion exchange is based on the reversible interchange of
ions between a solid and liquid phase in which there is no per-
manent change in the structure of the solid (ST-135). As water
containing soluble impurities flows across a bed of ion exchange
resin, ions in the resin are replaced with different ions from
the water. When the resin has reached its load capacity, it is
regenerated by rinsing with a regenerant solution that contains
a high concentration of the type of ions contained in the orig-
inal resin. When water with a high concentration of IDS is
treated by ion exchange, a large quantity of impurities is re-
moved and frequent flushing and regeneration of the beds is re-
quired. Wastewater produced from ion exchange includes backwash
water, spent regenerant solution, and rinse water.
6.7.4 Vapor Compression Distillation
Vapor compression distillation can be used to further
concentrate the waste stream from a reverse osmosis unit. It
can also treat the prescrubber blowdown from the Wellman-Lord
or Magnesia Slurry Processes; cooling tower blowdown from the
processes requiring it; or purge streams from the Magnesia
Slurry, Double Alkali, or Lime/Limestone Processes. ^
Vapor compression distillation (also known as "brine
concentration") uses a falling film evaporator along with a vapor
compression thermodynamic cycle to concentrate the TDS in waste
streams while producing low TDS product water. The pH of the
waste stream to be treated is adjusted to between 5.5 and 6.5
for decarbonation. The stream is then heated by heat exchange
with the hot product condensate, deaerated, and combined with
-208-
-------
the slurry concentrate in the evaporator sump. The brine slurry
is constantly circulated from the sump to the top of the evapor-
ator tubes. As the slurry falls as a film on the inside of the
tubes, part of the water is vaporized by steam condensing on the
outside of the tubes. This vapor is compressed and introduced to
the shell side of the evaporator. The steam condenses on the out-
side of the tubes, and the condensate is pumped through the feed
preheater. This product condensate contains less than 10 mg/&
(ppm) TDS, and is suitable for reuse in the power plant or FGD
process. Approximately 90% of the inlet water can be recovered
when treating water with an inlet TDS of 10,000 mg/£ (ppm) (com-
parable to the water quality of the prescrubber blowdown) (RE-259)
The brine concentrate, which will be about 10% of the original
stream volume and contain most of the dissolved solids, can be
sent to an evaporation pond or mechanical drying system for final
disposal. Vapor compression distillation processes are commer-
cially available. They have been installed in electric power
generating stations in the western and southwestern states to
recover deionized water from cooling tower blowdown (LE-299).
Vapor compression distillation is an energy intensive process
requiring approximately 90 kw-hr/1000 gal of water processed
(RE-259). Most of this energy goes into driving the vapor com-
pressor.
6.7.5 Multistage Flash Evaporation
Multistage flash evaporation could be used to treat
the prescrubber blowdown from the Wellman-Lord or Magnesia Slurry
Processes; cooling tower blowdown from any of the processes;
or purge streams from the Magnesia Slurry, Double Alkali, or
Lime/Limestone Processes. In this method, the wastewater stream
is heated and flash evaporated in a series of stages under
progressively lower pressures. Each stage of the evaporator
has a heat exchanger for preheating the incoming liquid and a
-209-
-------
vacuum chamber for flash evaporation. The vapor is used for
preheating the incoming liquid, and is condensed. The condensate
from each stage is collected. This product water will have a low
concentration of TDS (less than 50 ppm), and is suitable for re-
use in the system (even as boiler feedwater) or for discharge.
A portion of the residual concentrated wastewater is mixed with
the influent stream for recycle through the evaporator. The
remaining concentrated waste can be sent to an evaporation pond
or mechanical drying system for final disposal. The amount of
waste produced will depend on the evaporator design and equipment
operating characteristics. For example, if water with an inlet
TDS of 10,000 ppm (comparable to prescrubber blowdown or lime/
limestone purge) is treated to a product concentration of less
than 50 ppm TDS and a waste concentration of about 100,000 ppm
TDS, the waste stream will be about 10 percent of the feedwater.
Multistage evaporators have been used in the chemical process
industry for many years, and have also been used for desalting
sea water to produce drinking water (BO-203).
-210-
-------
REFERENCES
AY-007 Aynsley, Eric and Meryl R. Jackson, Industrial Waste
Studies: Steam Generating Plants, Draft Final
Report. Rosemont, 111. , Freeman 'Labs . , Inc., 1971.
BA-185 Babcock & Wilcox, Steam/Its Generation and Use. 38th
ed. New York, 197IT '
BE-162 Bell, William E. & E. Dennis Escher, "Disposal of
Chemical Cleaning Waste Solvents". Mat. Protect.
Perfor. 9, 15 (1970). ~
BL-036 Blake, R. T. , "Proper Feedwater Treatment Helps
Minimize Pollution", Plant Eng. 1970 (Oct), 34.
BO-203 Bornstein, L. J., et al., Reuse of Power Plant Desul-
furization Waste Water , Final'"Report. EPA-6007 2-76-
024, PB250731TEl Segundo, California, Aerospace
Corp., Feb. 1976.
CH-R-387 Christman, Peter G., Water Recycle/Resuse Alternatives
— ^he (Pennsylvania Power and Light) Montour Station,
DrafT"~Report. DCN 77-200-118^02, EPA,"Contract No.
68-03-2339, TN 200-118-11. Austin, TX, Radian Corp.,
April 1977.
CL-028 Cline, M. A. A., B. A. Thrush, and R. P. Wayne,
"Kinetics of the Chemiluminescent Reaction Between
NO and 03", Trans. Faraday Soc. 60(494), 359-70 (1964)
CO-380 "Coal Preparation and Unit-Train Loading", Coal Age
1970 (July), 188.
CU-077 Cuffe, Stanley T., Alternative Control Systems to be
Considered for Cost and Environmental Impact Analysis-
Program to Review the NSPS for Emissions from Steam
Generators, Research Triang!e~Park, N.C. EPA, April
1977.
DA-189 -Davis, John C., "Coal Cleaning Readies for Wider
Sulfur-Removal Role", Chem. Eng. 83 (5), 70 (1976).
DE-064 Deurbrouck, A. W. , Sulfur Reduction Potential of_ the
Coals of the United States. Report of Investigations
7633. Pittsburgh, Pa., Pittsburgh Energy Research
Center, 1972.
-211-
-------
REFERENCES CONTINUED
DI-R-139 Dickerman, J. C., et al. , An Investigation of the Use
of Coal in the HL&P Generating System, Interim Report,
2~~vols. DCN 77^ZUQ-162-QI.Austin, TX, Radian Corp.,
Feb. 1977.
DO-048 Donohue, John M., "Making Cooling Water Safe for Steel
and Fish, Too", Chem. Eng. 7J3(22) , 98 (1971).
DO-051 Donahue, John M., "Chemical Treatment", Ind. Water
Eng. 7(5), 35 (1970).
EN-127 Environmental Protection Agency, (Office of Air and
Water Programs, Effluent Guidelines Div.), Development
Document for Proposed Effluent Limitations Guidelines
and New~Source Performance Standards for the Steam
Electric Power Generating Point Source Category. 1974.
FI-102 "Fine-Coal Treatment and Water Handling", Coal Age 6_6
(12), 67 (1961).
GA-R-203 Gathman, Wayne and J. G. Noblett, Jr., Water Recycle/
Reuse Alternatives at the Public Service of Colorado
Comanche Plant. Tech. Note 200-118-09, EPA Contract
No. 68-03-2339. Austin, TX, Radian Corp., Aug. 1976.
GE-009 Gentry, Robert E., Jr., "Reverse Osmosis: A Pleasant
Inversion", Env. Sci. Tech. 1, 124-31 (1967).
GL-028 Glover, G. E., "Cooling Tower Slowdown Treatment Costs",
Industrial Process Design for Water Pollution Control,
Vol. 2. New York, AICHE, pp. 74 ff.
HA-R-636 Hargrove, 0. W. and J. G. Noblett, Jr., Water^Recycle/
Reuse Alternatives at the Arizona Public Service Four
Corners Stations. Technical Note 200-118-07, EPA
Contract No. 68-03-2339, Reviewed by D. M. Ottmers, Jr.,
Austin, TX, Radian Corp., July 1976.
HU-051 Hunter, William D. , Jr., "Application of S02 Reduction
in Stack Gas Desulfurization", Presented at the Flue
Gas Desulfurization Symposium, New Orleans. May 1973.
IF-001 Ifeadi, C. N. and H. S. Rosenberg, "Lime/Limestone Sludge
Disposal Trends in the Utility Industry", Presented at
the Flue Gas Desulfurization Symposium, Atlanta, Ga.,
Nov. 1974.
-212-
-------
REFERENCES CONTINUED
KA-227 Kaplan, Norman, "Introduction to Double Alkali Flue
Gas Desulfurization Technology", Presented at the
EPA Flue Gas Desulfurization Symposium, New Orleans,
Louisiana. March 1976.
LE-218 Leonard, Joseph W. and David R. Mitchell, eds., Coal
Preparation, 3rd edition. New York, The American
Institute of Mining, Metallurgical and Petroleum
Engineers, Inc., 1968.
LO-071 Lowry, H. H. ed., Chemistry of Coal Utilization, 2
vols. & supplementary volume. N.Y., Wiley,1945,
1965 (supplementary volume).
MA-411 Martin, John F., "Coal Refuse Disposal in the Eastern
United States". News Env. Res. Cincinnati. Dec. 1974.
MC-076 McGlamery, G. G., et al-, Conceptual Design and Cost
S tudy. Sulfur Oxide Removal from Power"Plant Stack
Gas . Magnesia Scrubbing-Regeneration: Production of
Concentrated Sulfuric Acid. EPA-R2-73-2^T. Muscle
Shoals, Ala., TVA, 1973"!
MC-147 McGlamery, G. G., et al., Detailed Cost Estimates for
Advanced Effluent Desu1furizati.on Processes, Final
Report. EPA-600/2-75-006.Muscle Shoals, Ala., TVA,
Jan. 1975
MA-230 Marshall, Win. L. , "Cooling Water Treatment in Power
Plants", Ind. Water Eng. 9(2), 38 (1972).
NO-R-106 Noblett, James G., Jr., Water Recycle/Reuse Alterna-
tives at the Georgia Power Company Plant Bowen.
Technical Note 200-118-08, Radian Project No.
200-118, EPA Contract No. 68-03-2339. Austin, TX.,
Radian Corp., Aug. 1976.
NO-R-137 Noblett, James G., Jr., Water Recycle/Reuse Alterna-
tives at Montana Power C omp any's Co1s trip P1ant.
DCN 77r2'00-188-04, Technical Note 200-118-12, EPA
Contract No. 68-03-2339, Reviewed by D. M. Ottmers.
Austin, TX., Radian Corp., May 1977.
OT-R-051 Ottmers, D. M., Jr., et al., Evaluation of Regenerable
Flue Gas Desulfurization Processes, revised report, 2
voT?.71?PRI RP 535-1, Austin, TX. ,' Radian Corp. ,
July 1976.
-213-
-------
REFERENCES' CONTINUED
PE-030 Perry, John H., Chemical Engineers Handbook, 4th ed.
New York, McGraw-Hill, 19FT7
PE-259 PEDCo Environmental, Inc., Flue Gas Desulfurization
Systems, Jan. Feb., March, 1977, summary report, EPA
Contract No. 68-02-1321, Task No. 28, Cincinnati,
Ohio, 1977.
RA-R-352 Radian Corporation, An Inves tig at ion p_f the Potential
for Utilization of 'SaTine Ground' Water in Energy-
Related Processes, Draft Final Report. DCN 77-200-
152-01 (06), FE-200-1, ERDA Contract No. E(49-18)-
2000. Austin, TX., Jan. 1977.
RO-359 Rossoff, J. and P. P. Leo, The Solid Waste Impact of
Controlling SO2 Emissions from Goal-Fired Steam
Generators. El Sungundo, CA., Aerospace Corp., Oct.
1977. ^~
RI-160 Rice, James K. and Sheldon D. Strauss, "Water-Pollution
Control in Steam Plants", Power 12£(4), Sl-20 (1977).
ST-135 Strauss, Sheldon D., "Water Treatment", Power 117(6),
SI-S24.
WE-003 (Paul) Weir Company, An Economic Feasibility Study of
Coal Desulfurization,~Z vols . FB 176 845, PB 176 855",
ChTcago, 111., Oct. 1965.
WE-324 Weber, Walter J., Jr., Physiochemical Processes for
Water Quality Control, New York, Wiley, 1972.
-214-
-------
APPENDIX A
-------
A.O EXPLANATION OF CALCULATIONAL METHOD
This section presents example calculations, and dis-
cussions of the assumptions used in estimating the water con-
sumption of uncontrolled power plants, and SO control strategies,
X
A-2
-------
A.I Selection of Representative Coal Ultimate Analyses
The model plant systems require the selection or
approximation of representative ultimate analyses for the
following four coals:
coal #1) 0.8% S (dry basis), 8,000 Btu/lb (AR basis), 6% ash, 30% H20
coal #2) 0.8% S (dry basis), 11,000 Btu/lb (AS. basis), 6% ash, 15% H20
coal #3) 3.5% S (dry basis), 12,000 Btu/lb (AR basis), 12% ash, 2.6% H20
coal #4) 7.0% S (dry basis), 12,000 Btu/lb (AR basis), 12% ash, 5.7% H20
Also, the determination of the ultimate analyses fol-
lowing coal cleaning operations, where 4070 of the sulfur is to
be removed, is required for the following two original coals:
coal #5) 3.5% S (dry basis), 12,000 Btu/lb (AR basis), 12% ash, 2.6% H20
coal #6) 7.0% S (dry basis), 12,000 Btu/lb (AR basis), 12% ash, 5.7% H20
Table B.l-1 shows the ultimate analyses of several
example coals.
Coal #1 - The ultimate analyses of coal #1 was approximated by
taking midrange values of the three low heating val-
ues western coals listed in Table B.l-1, i.e., coals
15, 16, and 17.
Coal #2 - Coal 14 is the only high heating value western coal
listed in Table B.l-1. From a comparison of proximate
analyses, this coal seems to have a representative
ultimate analysis. Therefore, the ultimate analysis
of coal #2 is taken to be the same as coal 14.
A-3
-------
TABLE A. 1-1. EXAMPLE COAL ULTIMATE ANALYSES
-p-
No.*
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
* Coals
Type
PA
PA
VA
WV
PA
PA
PA
PA
KY
OH
IL
UT
IL
KT
WY
WY
m
Western
Eastern
1 -»• 17 are
C
79.84
79.45
70.00
84.21
77.52
76.74
75.42
72.66
79.94
67.39
64.24
69.83
59.88
63.48
53.89
47.10
42.46
70.7
53.13
72.7
69.9
from Steam
H2
1.78
2.21
3.24
4.47
4.16
4.15
4.48
4.62
5.14
4.75
4.39
4,90
4.31
4.00
3.62
3.56
2,86
4.7
3.70
5.3
4.9
N2
0.25
0.77
0.77
1.21
1.30
1.38
1.21
1.45
1.50
1.17
1.28
1.49
1.13
1.02
1.14
0.57
0.53
1.1
1.00
1.1
1.3
S
0.71
0.60
0.62
0.74
1.68
1.68
2.20
1.82
0.70
4.00
2.70
0.90
3.20
0.43
0.30
0.55
0.40
3.4
0.39
1.0
1.1
02
1.96
1.95
2.55
2.51
2.08
2.68
2.84
4.96
6.26
6.16
7.26
10.33
7:18
9.57
12.07
11.83
12.15
10.3
14.17
9.0
7.1
Ash
9.7
11.9
20.2
5.1
10.3
10.2
10.2
11.2
3.3
9.1
11.7
6.4
9.0
7.0
3.7
4.8
4.2
7.1
4.62
8.9
13.7
H20
4.5
2.5
2.0
1.0
1.3
1.5
1.5
1.5
2.5
3.6
5.8
5.2
12.2
14.1
25.0
31.0
37.0
2.7
23
2.0
2.0
Btu/lb
12,745
12,925
11,925
14,715
13,800
13,720
13,800
13,325
14,480
12,850
11,910
12,600
11,340
11,140
9,345
8,320
7,255
12,400
13,135
12,640
Coals 18 & 19 are from Tech. Note 200-118-09
Coals 20-1-21 are from FA-084
-------
Coal #3 - The ultimate analysis used for coal #3 is the same as
that used by McGlammery, et al (MC-147) . A comparison
of this analysis with the analyses of coals 7, 10, 11,
and 13 in Table A.1-1 shows similar values.
Coal #4 - Coal 10 is the only high sulfur coal listed in Table
B-l-1, and is 4% sulfur rather than 7% sulfur. Values
of 12,000 Btu/lb, 5.7% H20, and 12% ash were assumed.
The remaining properties of coal #4 were chosen to be
similar to those of coal 10.
Coals #5 and #6 are the product coals from washing
Coals #3 and #4. It is assumed that 1) 407o of the original
sulfur content has been removed, 2) 50% of the original ash
has been removed (RA-215), 3) the yield is 85% (ZI-014, MA-495,
RA-215), and 4) that the coal product is dewatered to 1570
(FI-102, LE-218, CO-380).
The ultimate analyses of the six representative coals,
chosen for use in our study calculations, are shown in Table
A.1-2.
A-5
-------
'TABLE A.1-2. SIX REPRESENTATIVE COALS - ULTIMATE ANALYSIS
Ultimate
Analysis
H20
C
H2 .
S
N2
02
Ash
Coal No.
1
30.0
47.3
3.5
0.6
0.6
12.0
6.0
2
15.0
63.8
4.0
0.7
1.0
9.5
6.0
3
2.6
70.5
4.5
3.4
1.0
6.0
12.0
4
5.7
63.6
5.0
6.6
1.1
6.0
12.0
5
15.0
66.6
4.3
1.9
0.9
5.7
5.6
6
15.0
63.1
5.0
4.0
1.1
5.9
5.9
Average
Heating Value
Btu/lb
8,000
11,000
12,000
12,000
11,100
11,500
-------
A-2 Base Case Assumptions
A set of design criteria constituting base case was
chosen to facilitate discussions and comparisons in the body of
the report. Also, in Appendix B, this base case is used to
illustrate the detailed calculational procedures. This section
defines the base case.
A.2.1
The following assumptions are the same as those used
by McGlammery, et al to calculate detailed mass balances for
FGD systems (MC-147) :
500 MW power plant
377> conversion efficiency
3.57o sulfur coal (dry basis)'
127o ash (as fired basis)
12,000 Btu/lb (as fired basis)
9270 of sulfur in coal evolves as S02
757° of ash evolves as fly ash
2070 excess air
137o leakage
9070 S02 removal
In addition, the base case that we have used assumes
that:
0.13 Ib HaO/lb dry air - humidity of excess air
457o of total heat is lost to condenser cooling system
A-7
-------
are:
20°F cooling water temperature used in the condenser
99% of fly ash removed in an ESP prior to entering
the scrubbing system
5% sluicing of fly ash
1% sluicing of bottom ash
The assumptions applicable to the. FGD systems studied
Lime: Adiabatic saturation of the flue gases in the
scrubber
105% stoichiometric lime
pond evaporation equals rainfall
first stage S/L separator produces a 40%
solids sludge
ponding or second stage S/L separator produces
a 60% solids sludge
10% solids recirculating slurry
25% oxidation
Limestone:
adiabatic saturation of the flue gases in the
scrubber
1207» stoichiometric limestone
pond evaporation equals rainfall
25% oxidation
10% solids recirculating slurry
first stage S/L separator produces a 40%
solids sludge
ponding or second stage S/L separator produces
a 60% solids sludge
A-8
-------
Wellman-Lord:
15% of circulating slurry is sent to purge
treatment
double effect evaporators
607o overhead in each effect
107o H20 in product S02 stream
Magnesia Slurry:
10570 stoichiometric MgO
1070 solids recirculating slurry
first stage S/L separator produces a 4070 solids
sludge
thermal decomposition of MgS03-6H20
9570 solids cake from centrifuge
Double Alkali:
10570 stoichiometric lime
1207» stoichiometric limestone
pond evaporation equals rainfall
first stage S/L separator produces a 4070
solids sludge
second stage S/L separator or ponding produces
a 607o solids sludge
25 7o oxidation
A-9
-------
A.3
Calculation of Flue Gas Rates and Compositions
To calculate flue gas rates and compositions, a coal
ultimate analysis was converted to a molal composition, and the
amount of excess air calculated. Then the volumes of combustion
products were calculated based on 100 Ib of fuel fired. A coal
usage rate was then calculated, and from this the total flue gas
volume was calculated. The following calculations for the base
case illustrate the calculational method used for all the model
systems.
A.3.1
Base Case Calculations
Basis: 100 Ib coal
Ultimate Analysis Mol. Wt. Moles
H20
C
H2
S
N2
02
Ash
2.6
70.5
4.5
3.4
1.0
6.0
12.0
18
12
2
32
28
32
0.14
5.88
2.25
0.11
0.04
0.19
For 1007o combustion:
C
H2
S
02 in coal
Moles
Moles 02
6.93
Moles Air
5.88
2.25
0.11
5.88
1.13
0.11
7.12
0.19
33.00
A-10
-------
With 20% excess air and 13% leakage:
Air
02
H20 in excess air =
Excess air =
Excess 02 -
33.00 (1.33) = 43.89 moles
9.22 moles
(43.89) (0.013) (29/18) = 0.92
moles
10.89 moles
2.29 moles
Combustion products:
Moles
Mol. Wt, Ib
CO 2
H20
S02
Na
02
5,88
3.31
0.10
35.71
2,29
46.29
Avg.
44
18
64
28
32
Mol. Wt.
285.72
59.58
6.40
971.88
73.28
1,369.86
= 29,6
Flue gas volume:
V =
NRT = (46.29 moles)
p ~(14.7 psia)
= 27,369 acf
psia ft3
mole °R
= 16,624 scf/100 Ib coal fired
Coal usage rate:
500 MW @ 37% eff = 4.492 x 10 9 Btu/hr
Coal AHV = 12,000 Btu/lb
Usage rate = 374,342 Ib/hr
A-ll
-------
Flue gas rate:
(374.342 Ib/hr) (166.24 scf/ib) (L^J
o u mm
= 1,037,177 scfm
For one of four equivalent scrubbing trains the flue
gas rate would be: 259,000 scfm.
A. 3 . 2 Model Systems Results
Each type of coal will require a characteristic coal
usage rate, and therefore the flue gases resulting from coal
combustion will have differing compositions and rates. In this
study, it is assumed that the conversion efficiency does not
vary with power plant size. Therefore, to calculate flue gas
rates for the model systems it is necessary to make separate
calculations for each coal type, but power plant size and flue
gas rate are directly proportional and thus results can be
scaled. Table A:3-l shows the flue gas rates and compositions
for each of the example coals for a 500 MW power plant.
TABLE A.3-1. FLUE GAS RATES AND COMPOSITIONS FOR EXAMPLE COALS
(500 Mw plant; one of four trains)
Flue Gas
Composition
Gas Rate**
C02
H20
S02
N2
02
Coal No.*
1
307.2
12.41
12.63
0.06
70.27
4.63
2
265.9
12.91
8.81
0.05
73.41
4.83
3
259.3
12.70
7.15
0.22
74.98
4.95
4
254.4
12.32
5.67
0.46
76.50
5.04
5
282.5
12.56
8.69
0.14
73.76
4.86
6
271.9
11.94
9.51
0.27
73.45
4.83
*A11 composition values are in volume percent
**Gas rates are given for 500 MW plants; given in 1000 scfm for one of four
trains
A-12
-------
A. 4 Water Requirements For A 500 Hfrj Power Plant
The processes and operations in a coal-fired power
plant that require water consumption are:
cooling water
ash handling
boiler blowdown
water conditioning
equipment cleaning, and
miscellaneous.
For the purpose of calculating a power plant water
balance for an estimation of the impact that an FGD installa-
tion would have on power plant water consumption, a 500 MW
power plant was chosen for comparative basis. Due to general
unavailability of data on intermit tant and miscellaneous water
consumptive practices, water conditioning, equipment cleaning,
and miscellaneous operations will be addressed as "general ser-
vice water", and plant data from preliminary water recycle/re-
use studies currently being conducted by Radian under EPA con-
tract will be used to characterize these flows.
Current water management in the power generation in-
dustry uses two basic techniques: 1) once-through, and 2) re-
circulatory. There is a wide variability in plant layouts, and
many combinations of once- through, and recirculatory practices
are used in various power plant water systems. The two extremes
in water management would be where: 1) once- through techniques
are used in all systems, and 2) recirculatory practices are max-
imized in all systems. The total recirculatory system ultimately
results in zero discharge. The once-through system requires
enormous quantities of water. Current practice falls somewhere.
A-13
-------
between these two extremes, but because of the large number of
available process options and possible combinations, there is
no "typical" water system. Water balances will be calculated
for power plant water systems that operate 1) once-through,
2) with recirculatory cooling at 2.5 cycles of concentration,
and once-through ash handling, 3) with recirculatory cooling
at 5 cycles with 50% recycle ash handling and 4) with zero
discharge of concentration, and once-through ash handling.
These four cases should characterize representative water con-
sumption in the current coal-fired power generation industry.
Hereafter, they will be referred to as cases #1 through #4.
A. 4.1
Cooling Water System
Assume: 1) "45% of a fossil-fuel fired generating
station's energy is removed and ultimately
discharged to the environment by the
condenser cooling system" (DI-139)
2) power plant conversion efficiency is 37%
(MC-147)
.'. For a 500 MW power plant 1,351 MW of ener-
gy are produced from the coal burned
(1.351MW) (0.45)
608 MW
34.6 x 106 Btu/min
of energy transfer
to the condenser
cooling system
Assume: 20°F temperature rise of the cooling water
A-14
-------
Q = MC AT
" C~AT = 7>469 gpm
P
This is the circulating or once- through water require-
ment.
In recirculatory systems:
r = B+EH-E make-up
B+D ~~~%+D
where: C = cycles of concentration
B = blowdown
D - drift
E = evaporation
We will calculate the make-up requirement for water
systems operating at 2.5, 5.0, and 13.5 cycles of concentration,
From Figure 6.1-3 (RA-352), the following table
of make-up requirement for a 1,000 MOW power plant versus cycles
of concentration can be constructed:
TABLE B.4-1. COOLING SYSTEM MAKE-UP FOR A 1,OOOMW PLANT
Cycles of Concentration Make-up Requirement (103gpm)
2 15.6
3 12.2
5 9.8
6 9.2
13 8.2
14 8.2
A-15
-------
If it is assumed that at proportional air flow rates
the drift and evaporation rates are proportional, then the make-
up versus cycles of concentration can be directly scaled from
1,000 MW to a 500 MW plant. Therefore the same general curve
would apply. The adjusted valued are shown in Table A.4-2.
TABLE A. 4-2. COOLING SYSTEM MAKE-UP FOR A 500 MW PLANT
Cycles of Concentration Make-up Requirement (103gpm)
2
3
5
6
13
14
7.8
6.1
4.9
4.6
4.1
4.1
Plant data from water recycle/reuse studies at the
Colstrip (NO-137), Comanche (GA-203), and Montour (CH-387)
stations, adjusted by proportionally to equivalent 500 MW
plants indicate:
TABLE A,4-3. ADJUSTED PLANT COOLING SYSTEM DATA
Cycles of Concentration Make-up Requirement (103gpm)
Montour 2.6 8.0
Comanche 5.0 5.5
Colstrip 13.5 4.3
Thus, the two cooling water system data sources show
excellent agreement. To fully characterize the plant cooling
water systems, it was assumed that the plant evaporation and
A-16
-------
drift data are characteristic. Table A.4-4 summarizes the four
cooling water system characteristics.
TABLE -A.4-4.. CHARACTERISTIC COOLING SYSTEM OPERATION
Cycles of
System # Description Concentration Make-up Slowdown Evaporation Drift
(gpia) (gpm) (gpm) (gpm)
1
2
3
4
once-through
partial
recirculatory
recirculatory
zero discharge
1
2.5
5.0
13.5
210,000
7,000
5,000
4,000
210,000
2,500 4,200
900 3,900
300 3,670
300
200
30
A.4-2 Ash Handling System
Assumptions: 1) 757o of the original ash content of the coal
forms as fly ashs and 25% forms as bottom ash.
2) 997o of the fly ash is collected in an ESP and
wet sluiced to an ash pond as a S70 slurry.
3) Bottom ash is sluiced as one percent slurry.
Table A.4-5 shows the fly ash and bottom ash rates for
each of the six example coals.
Assuming 1% bottom ash sluicing and 57o fly ash sluicing
the sluice water requirement is shown in Table A.4-6,
The source for ash sluicing waters varies with power
plant water management systems.
. For system #1 it is assumed that raw water is the
only source for the ash sluicing water, and that
ash sluicing is once-through.
A-17
-------
TABLE A.4-5. BOTTOM ASH AND FLY ASH RATES
Coal Coal Usage Rate
Ib/hr
1 * 646,659
2 431,061
3 374,342
4 395,167
5 427,207
6 412,348
Bottom Ash
Ib/hr
9,700
6,466
11,230
11,855
5,981
6,082
Fly Ash
Ib/hr
29,100
19,398
33,691
35,565
17,943
18,246
TABLE A. 4-6. SLUICE WATER REQUIREMENT
Coal
1
2
3
4
5
6
Bottom Ash Sluice Fly
(gpm)
1,939
1,293
2,245
2,370
1,196
1,216
Ash Sluice
(gpm)
1,163
776
1,347
1,422
717
729
Total Sluice
(gpffl.)
3,102
2,069
3,592
3,792
1,913
1,945
A-18
-------
For system #2 it is assumed that cooling tower blow-
down is the source for the ash sluicing water, and
that ash sluicing is once-through. The cooling
tower blowdown rate is 2,500 gpm.
Coals #2, #4, and #5 require less ash sluicing water
than available through use of cooling tower blowdown. The ex-
cess blowdown could be used for other plant requirements or
ponded for discharge. No raw water is required for ash sluicing
these coals.
Coals #1, 3, and 4 require more sluicing water than
available at 2.5 cycles of concentration. In this situation in
many operating procedures, cooling tower blowdown would be in-
creased. This reduces the cycles of concentration as shown in
the following calculation.
Assuming the same evaporation and drift values, the
cycles of concentration to achieve desired blowdown in these
cases would be:
TABLE A.4-7
Coal
1
3
4
Desired Blowdown
(gpm)
3,102
3,592
3,792
Cycles of Concentration
2.30
2.13
2.08
The raw water requirement is increased to equal the
demand for ash sluicing. Passing the water through the cooling
tower rather than adding the raw water as make-up directly to
A-19
-------
the ash sluicing system allows the cooling tower a greater margin
of safety in regards to scaling.
A second alternative would be to increase the % solids
in the ash slurry. Bottom ash slurries are often in the 1-2%
range with many systems operating with 570 solids. Fly Ash
Slurries operate in the 5-10% solids range. Often fly ash slur-
ries have higher solids.
If it is assumed that bottom ash is slurried at 2%
solids and bottom ash is slurried at 10% solids, the total
sluice requirement would be reduced by half. This reduces the
required sluice water to:
TABLE A.4-8
Coal
1
2
3
4
5
Total Sluice Water Requirement
(gpm)
1,551
1,035
1,796
1,896
957
With the reduced requirement for sluice water, at the
desired 2.5 cycles of concentration, the excess blowdown would
be routed to a pond before ultimate disposal. Because higher
solids concentration slurries lead to handling problems, the
power plant could utilize the excess blowdown and operate at
as low slurry concentrations as possible. Therefore the ash
sluicing requirement is equal to the cooling tower blowdown and
therefore there is no additional water make-up requirement for
ash sluicing.
A-20
-------
A third alternative is that the excess sluice water
requirement is met with raw water ash sluice make-up. 1) Because
we want to characterize a cooling system operating at 2.5 cycles
of concentration (and therefore do not want to reduce cycles of
concentration), 2) Because we have defined a 1% slurry of bottom
ash and 5% slurry for fly ash as typical operating conditions,
and 3) because at least one of the five power plants currently
under study by Radian for water recycle/reuse alternatives in
an EPA program uses raw water for make-up ash sluice in addition
to the cooling tower blowdown, we will choose this third
alternative for calculation of water consumption in our study.
This choice will not lead to optimal water use (lowering cycles
of concentration is a better alternative) , but it is a current
practice at some power plants and the water requirement will
characterize a medium consumption pattern. Table A.4-9 shows
the total ash sluice requirement, the raw water ash sluice make-
up, and the blowdown for coals #1 through #6 under the third
alternative.
TABLE A. 4-9-. SYSTEM #2 SLUICE MAKE-UP WATER REQUIREMENT
Coal
1
2
3
4
5
6
Total Ash
Sluice Requirement
(gpm)
3,102
2,069
3,592
3,792
1,913
1,945
Raw Water
Make-Up
(gpm)
602
0
1,092
1,292
0
0
Cooling Tower
Blowdown
(gpm)
2,500
2,069
2,500
2,500
1,913
1,945
For system #3 it is assumed that cooling tower
blowdown is the source for ash sluicing water.
With this power plant operating at 50 cycles of
A-21
-------
For system #4 it is assumed that cooling tower blow-
down is the source for ash sluicing make-up. In this
power plant water system the cooling tower operates
at 13.5 cycles of concentration, and that the ash
sluicing system is total recirculatory. Because the
ash settles as a 40-50% solids sludge, and assuming
a 5 wt % ash sluice slurry, total recycle translates
to 95% recycle of the water stream. Table A.4-11
shows the total sluice water requirement, make-up
sluice requirement, and available cooling tower
blowdown for each coal.
TABLE A.4-10.
Coal
1
2
3
4
5
6
Total
Sluice Water
Requirement
(gpm)
3,102
2,069
3,592
3,792
1,913
1,945
Make-Up
Sluice Water
Requirement
(gpm)
1,551
1,035
1,796
1,396
957
973
Available
Cooling tower
Slowdown
(gpm)
900
900
900
900
900
900
Make-Up
Water
Requirement
(gpm)
651
135
896
996
57
73
Available
Gen Serv
Blowdown
(gpm)
563
563
563
Raw HZ0
Reg
(gpm)
333
433
Coals #2, #5, and #6 can essentially operate without
any raw water make-up requirement by a slight % solids increase.
Coals #1, #3, and #4 require make-up water in the quantities
shown in Table A.4-10. It is probable that a power plant with
a water management program of this nature will probably collect
"general service" blowdown (discussed in the following section)
and this water will be used as make-up. If raw water is used
as make-up, the general services blowdown will probably be fed
into the cooling tower circulating stream thus reducing cooling
tower raw water demand. We will consider the general services
blowdown as ash sluicing make-up. Therefore, coal #1 does not
have any additional raw water requirement if a slight increase
in solids slurry is made. Coals #3 and #4 require the raw make-
up water shown in Table A.4-10.
A-22
-------
• For system #4 it is assumed that cooling tower blow-
down is the source for ash sluicing make-up. This
power plant water system assumes that the cooling
tower operates at 13.5 cycles of concentration, and
that the ash sluicing system is total recycle. Be-
cause the ash settles as a 40-50% solids sludge,
• and assuming a 5 wt % ash sluice slurry, total
recycle translates to 95% recycle of the water
stream. Table A.4-11 shows the total sluice
water requirement, make-up sluice requirement, and
available cooling tower blowdown for each coal.
TABLE A.4-11.
Total Make-up Available
Sluice Water Sluice Water Cooling Tower
Coal Requirement Requirement Slowdown
(gpm) (gpm) (gpm).
1
2
3
4
5
6
3,102
2,069
3,592
3,792
1,913
1,945
155
103
180
190
96
97
300
300
300
800
300
300
It can be seen that no additional make-up requirement
is necessary for any of the coals.
A.4.3 General Services Water System
Data from water recycle/reuse studies being conducted
by Radian for the EPA were used to characterize this water
A-23
-------
requirement. Data from the Comanche (GA-203), Montour (CH-387),
and Bowen (NO-106) power plants are shown in Table A.4-12.
TABLE A.4-12. PLANT GENERAL SERVICES WATER REQUIREMENT DATA
Capacity General Services Water gpm/MW
Comanche
Montour
Bowen
MW
700
1,500
3,180
(gpm)
' 960
1,900
11,000
1.37
1.27
3.46
The Bowen power plant collects the general services
water blowdown and feeds it into the cooling tower system. It
operates essentially once-through. Therefore, the value for the
Bowen general services water will be taken as a high range
value. We will assume a value of 1.5 gpm general service water
requirement per megawatt rated power plant capacity as a con-
servative number. For a 500 MW plant this is a raw water re-
quirement of 750 gallons per minute. From Bowen data 1,400 gpm
out of 1,900 gpm are collected and routed to a pond (NO-106).
This indicates approximately 75% of the general services water
requirement might be available for use as cooling tower make-up
or ash sluicing make-up. This is 563 gpm in our case . Table
A.4-13 shows the general services requirement by system.
TABLE A.4-13.
System
#1
n
#3
#4
Water
Requirement (gpm)
750
750
750
187
Description
once- through
once -through
recirculated to ash sluicing
recirculated to cooling tower
A-24
-------
For system #1 - all flows are once-through with no attempt
being made to reuse any waters.
For system #2 - we will assume that no attempt is made to reuse
this water.
For system #3 - the 563 gpm recoverable general services water
is required for ash sluicing make-up. This re-
cycle advantage was considered in reducing the
requirement for make-up ash sluicing waters.
For system #4 - The recycle advantage is achieved by combining
the recovered general services water with the
cooling tower make-up.
A.4.4 Boiler Make-up Water Requirement.
Assume that the typical blowdown rate for a drum
type steam boiler is 0.1% of the steam generation rate (AY-007).
For a 500 MW power plant operating at 37% conversion efficiency,
approximately 75 MM Btu/min are used converting water to steam.
Assuming a latent heat of vaporization of 1,000 Btu/lb, and a
blowdown rate of 0. !%> of the steam generation rate, 9.2 gpm
of make-up water are required.
A-25
-------
A. 5 Water Requirements For SO Control Strategies
The SO control strategies require significant amounts
of fresh water for make-up to large circulating streams. The
largest make-up requirement is for replacement of water evapora-
ted in the scrubber, • Fresh water make-up is also required for
loss due to occlusion in solid wastes, prescrubber blowdown,
cooling water blowdown, loss with solids drying, loss in product
S02 streams, and S02 conversion process requirements.
A.5,1 Evaporative Scrubber Water Losses
The hot flue gases enter either a prescrubber, or the
S02 scrubber, and contact a recirculating liquor. The gases are
adiabatically cooled and saturated evaporating a portion of the
liquor. Fresh water make-up is required to replace this lost
water.
Base Case Calculations -
CO 2
H20
N2
02
S02
NOX
Mole %
12.55
7.76
74.55
4.86
0.22
0.06
Mol.
44
18
28
32
64
--
Wt, lb/100 moles
552.2
139.7
2,087.4
155.5
14.1
2,948.9
Avg Mol. Wt. = 29.5
flue gas rate = 260,000 scfm per train
A-26
-------
n - IS - (14' 7 Psia> (260 onn
= 723.98 moles/min
Mfg - 21,375,4 Ib/min
Inlet temperature - 310°F
Assume an outlet temperature of 125 °F:
Hfg ' MfgyT
= (21,357.4 Ib/min) (0.26 Btu/°R-lb) (310°R-125°R)
= 1,027,291 Btu/min
Hw = MW^W - Hfg
= (1,027,291 Btu/min)/(ls020 Btu/lb)
= 1,007.1 lb/tnin of HaO evaporated
= 55.95 moles H20/min
New flue gas composition:
water in original gas = (723.98 moles/min) (0. 0776 mol fract H20)
= 56.18 moles H20
A-27
-------
% H20 = ' x 100% - 14.4 mole % H20
From psychrometric charts @ 125°F, saturated gases
contain 12.79 mole % H20.
Assume an outlet temperature of 128°F:
Hfg = MfgAT = VAw = Hw
MC AT
= P
= 55.045 moles of H20/min
55.045 + 56.18
new gas composition = -
723.98 + 55.045
= 14.277 mole % H20
From psychrometric charts @ 128°F the saturated gas
contains 14.105 mole % H20.
The agreement is very close. Therefore the outlet
temperature is taken as 128°F, and 991 Ib/min of water are
evaporated per train.
Model Plant Systems
To calculate the evaporative water losses for each of
the model plant systems it is assumed that there is no variation
with FGD system. The amount of water evaporated is directly
proportional to the amount of flue gas, and therefore, can be
scaled directly with plant size. A separate calculation is
A-28
-------
necessary for each coal type due to different coal usage rates,
and flue gas compositions. The evaporative losses for one of
four equivalent scrubbing trains for a 500 MW power plant for
each of the six coal types is shown in Table A. 5-1.
TABLE A.5-1. EVAPORATIVE SCRUBBER WATER LOSSES
T outlet, °F
Water Loss, gpm
1
137
130.9
2
131
119.1
CoaJ
3
128
118.8
. No.
4
124
119.9
5
131
126.5
6
132
120.5
A.5.2 Prescrubbing Water Make-up Requirement
In the Wellman-Lord and Magnesia Slurry FGD systems,
the flue gases must be scrubbed to remove particulates and
chlorides before the gases enter the absorption tower. The
evaporative scrubber water loss occurs in the prescrubber, and
there is no further evaporation in the absorber. Additional
water make-up is required to replace a prescrubber blowdown.
The blowdown is taken to maintain desired suspended and dissolved
solids concentrations.
For our calculations, it was assumed that 99% of the
particulates were removed in an ESP prior to the prescrubber.
100% of the remaining particulates, and 100% of the chlorides
are assumed to be removed from the flue gases in the prescrubber.
The particulates are sluiced in a 5% solids slurry to the ash
pond, and the limiting concentration of dissolved solids is
20,000 ppm.
Assumptions: make-up water 40 ppm chlorine
flue gases 30 ppm chlorine
A-29
-------
Dissolved Solids Slowdown Requirement
Chlorine Balance
Water Balance
chlorine input = chlorine output
F-fcl+M-Mcl - B-bcl
(1)
Water in = Water out
F*f
H20
M = B + E
(2)
M = make-up, Ib/min
B = blowdown, Ib/min
E = evaporation, Ib/min
F = flue gas rate, Ib/min
•cl
LH,0
'cl
M
cl
weight fraction of chlorine
in the flue gas
weight fraction of water in
the flue gas
weight fraction of chlorine
in the blowdown
weight fraction of chlorine
in the make-up water
Combining (1) and (2) and solving for B:
E-M , + F-f,
B =
cl
•cl
'cl
(3)
Table A.5-2 shows the results of the calculations
for each of the coals.
A-30
-------
TABLE A.5-2. DISSOLVED SOLIDS SLOWDOWN REQUIREMENT
Coal
1
2
3
4
5
6
F
Ib /min
24S747
21,768
21,357
21,088
23,119
22,146
E
Ib/min
4,364
3,972
3,964
4,000
4,220
4,020
B
Ib/min
45.85
40.60
39.97
39.63
43.12
41.26
fcl * 3° Ppm = 3xl0"5
M , = 40 ppm = 4x10"5 Ibd/lbM
B . = 20,000 ppm =0.02 Ibd/lbB
Suspended Solids Slowdown Requirement
Table A.5-3 shows the water requirement for sluicing
the particulates removed in the prescrubber. The sluice is 57o
solids.
TABLE A.5-3. SLUICE WATER REQUIREMENT
Coal
1
2
3
4
5
6
Water Requirement, Ib/min
97
65
112
119
60
61
A-31
-------
Thus, the prescrubber blowdown is limited by the de-
sired suspended solids concentration. For the base case, for
all four scrubbing trains, the make-up requirement is 54 gpm.
A.5.3 Water Loss Due to Occlusion in the Solid Waste
The lime, limestone, and double alkali FGD systems
produce a by-product solid waste. This solid waste is assumed
in this study to be settled or dewatered to a 60% solids sludge,
Water loss due to occlusion in the sludge requires fresh water
make-up.
Base Case Calculation
gas rate: 260,000 scfm x a = 723 Ib mole/min
S02 content: 0.0022 volume fraction
Assume 90% S02 removal:
(0.90) (723 lbm^les)(.OQ22) = 1.43 m°^nS°2
Assume 25% oxidation and coprecipitation of sulfite and sulfate
0.36 moles CaSCL 1.07
1.07 moles CaS03
15 mole CaSO, =
85 mole CaS03
.•. 1.26 moles Ca(.85S03•.15SOX) • %H20
0.17 moles CaS04-2H20
A-32
-------
Assume a lime stoichiometry of 1.05:
excess lime regenerant, recarbonated to CaS03
GaC03 = (0.5)(1.43) - 0.7 lb moles CaC03
Sludge solids:
Ca(.85 SO,-.15 SO,) - %H20 (1.26)(130) = 163.8 Ib/min
CaS0^ (0.17) (172) = 29.2 Ib/tnin
(0.07)(100) = 7.Q Ib/min
200.0 Ib/min
these solids contain 17.4 Ib/min of water of hydration
Assume 60% solids sludge:
occluded water = 133 Ib/min +17.4 Ib/min
= 150.4 Ib/min
Model Plant Systems Calculations
Each model plant system, excluding variations in power
plant size, requires a separate calculation. Because the amount
of water occuluded in the solid waste is directly proportional
to the power plant size, these systems' water losses can be
scaled from the values calculated for 500 MW plants. The
results of the calculations are shown in Table A.5-4.
Cooling Water Slowdown
The Wellman-Lord condenser cooling system, and the
product acid cooling system in the contact sulfuric acid plant,
A-33
-------
TABLE A.5-4. WATER LOSS: OCCLUDED WITH SOLID WASTE
Case %S Removal
1 a 76.4
76.4
76.4
76.4
b 88.2
88.2
c not applicable
d not applicable
e 62.5
62.5
2 a 90
90
90
90
90
90
90
90
b 90
90
90
90
3 a 70
70
70
70
FGD System
L
LS
DL
DS
L
LS
L
LS
L
LS
. DL
DS
L
LS
DL
DS
L
LS
L
LS
L
LS
L
LS
Coal
#3
#3
#3
#3
#4
#4
#5
#5
#3
#3
#3
#3
#4
#4
#4
#4
#1
#1
#2
#2
n
n
#1
#1
Water Loss
(Ib/min)
128
141
128
141
303
332
46
51
150
165
150
165
309
339
309
339
49
53
35
38
27
30
38
42
(Continued)
A-34
-------
TABLE A.5-4. WATER LOSS: OCCLUDED WITH SOLID WASTE (Continued)
Case
bi
b2
%S Removal
85
85
90.8
90.8
FGD System
L
LS
L
LS
Coal
#5
#5
#6
#6
Water Loss
(Ib/min)
98
110
196
214
L - LIME
LS - LIMESTONE
DL - DOUBLE-ALKALI: lime regenerant
DS - DOUBLE-ALKALI: limestone regenerant
A-35
-------
used for conversion of product SOa streams, require fresh water
make-up due to losses of evaporation, drift, and blowdown.
These values were scaled from the requirements presented by
McGlammery, et al (MC-147).
A.5.5 Loss with Solids Drying
The Wellman-Lord and Magnesia Slurry FGD systems con-
tain solids drying operations. The water loss associated with
these operations is relatively small, but requires make-up to
the system. This make-up requirement was calculated from
water balances (Section B.6) and 100% loss was assumed. The
results were scaled for each model plant system.
A.5,6 Loss in the Product S02 Stream
The product stream from the Wellman-Lord process
was assumed to contain 10% water. The quantity of water lost
was calculated by sulfur removal requirement, and scaled for
each of the model plant systems.
A-36
-------
A.6.0 WATER BALANCES FOR FGD SYSTEMS
Water balances were calculated for the base case con-
ditions for the five FGD systems studied in this project. Tables
showing the flows, and water compositions of the various streams
are given in Section 5 of the report, when discussing each FGD
water system. The method of calculation is straightforward, and
the major assumptions are listed below the table for each system.
A-37
-------
TECHNICAL REPORT DATA
(Please read fnitfuctions on tne reverse before completing!
3E?C3T NO.
EPA-600/7-78-045b
2.
3. RECIPIENT'S ACCESSION-NO.
s AND SUBTITLE Controlling SO2 Emissions from Coal-
Fired Steam-Electric Generators: Water Pollution
Impact (Volume n. Technical Discussion)
5. REPORT DATE
March 1978
5. PERFORMING ORGANIZATION CODE
7. AUTHQRlSi
8. PERFORMING ORGANIZATION REPORT NO.
R, L. Sugarek and T. G. Sipes
9. PERFORMING ORGANIZATION NAME ANO ADDRESS
Radian Corporation
8500 Shoal Creek Boulevard
Austin, Texas 78766
10. PROGRAM ELEMENT NO.
EHE624A
11. CONTRACT/GRANT NO.
68-02-2608, W.A. 10
12. SPONSORING AGENCY NAMS ANO ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OP REPORT AND PERIOD COVERED
Task Final: 4-12/77
14. SPONSORING AGENCY CODE
EPA/600/13
is.SUPPLEMENTARY NOTES T£RL-RTP project officer is Julian W. Jones, Man Drop 61,
919/541-2489.
is. ABSTRACT
rep0ri- gives results of one task in a comprehensive program to review
the New Source Performance Standards (NSPS) for SO2 emissions from coal-fired
steam-electric generating plants. The results compare two alternative standards
to the existing NSPS (1. 2 Ib SO2/million Btu of heat input): (1) 0. 5 lb SO2 /million Btu
of heat input, allowing credit (as does the existing NSPS) for physical coal cleaning
or use of low sulfur coal; and (2) 90% removal of SO2 from stack gases, regardless
of original coal sulfur content. The comparisons are in terms of their effect on the
quality and quantity of power plant waste water effluents and on the amount of plant
water consumption. Potential effects of SO2 control system effluents on the environ-
ment are evaluated, and alternative treatment processes are discussed. A total of
108 plant systems were discussed, including combinations of three NSPS, five flue
gas desulfurization (FGD) systems , five coal types , four plant sizes , and sulfur
removal by coal cleaning. Volumes and quality of wastewater streams varied very
little from one alternative NSPS to another: all streams can be treated adequately
using commercially available technologies. However, the alternative standards
increase total water consumption 8-11%, depending on the .FGD process used. Physi-
.coal cleaning plus lime/limestone scrubbing increases total water consumed 8-12%.
17.
KHY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIPIgRS/OPSN ENDED TERMS
c. COSATI Field/Croup
Pollution
Sulfur Dioxide
Flue Gases
Desulfurization
Water Pollution
Coal
Waste Treatment
Combustion
Steam-Electric
Power Generation
Calcium Oxides
Limestone
Wastewater
Pollution Control
Stationary Sources
13B
07B
21B
07A,07D
08G,21D
10A
13. OI5THI3UT1Q.N STATSMSN1
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OP PAGES
265
20. SECURITY CLASS I This page)
Unclassified
22. PRICE
SPA Form 2220-1 (9-73)
------- |