PB-249 508
SULFURIC AGIO PLANT EMISSION^ DURING START-UP. SHUTDOWN,
AND MALFUNCTION
E. L. Calvin, et al
Catalytic, Incorporated
Prepared for:
Industrial Environmental Research Laboratory
i
January 1976
DISTRIBUTED BY:
Nation; I Technical Information Service
U. S DEPARTMENT OF COMMERCE
-------
EPA-600/2-76-010
January 1976
Environmental Protection Technology Series
SULFURIC ACID
uiHii
. -,'r'J.i.<-*V-i .'_-»• i-^Sd*^S*.7'•
EMISSIONS DURING
AND MALFUNCTION
: rvw./: .,,;
~ .."-v^eS-'. "
«•."•' s»ftu Vsv
?rt^
b-^:. •v.^.;t
"li
;-. .>-^ .!•• ,'•*
?^.v;
~'^ft" •
•-"' "'• -4 "*
^^ PROl^
;-,v--.i^^
rt,1^?^:-.' /
;.V.; -T^:
--'^^••^S^r^'^4^^ °* Uesearch and Development
U.S. Environmental Protection Agency
Rest rch Triangle Park, North Carolina 27711
-------
TECHNiCAL REPORT DATA
(Plcatc read Itta/vciium on liu rc> MS* frcjurt cumpteltngl
i REPORT NO.
EPA-600/2-76-010
4. TITLE AND SUBTITLE
2.
Sulfuric Acid Plant Emissions During Start-up,
Shutdown, and Malfunction
7. AUTHOR(S)
E.L. Calvin and F.P. Kodras
9. PERFORMING ORSANIZA , ,O.M NAME AND ADDRESS
Catalytic, Inc.
P.O. Box 11402
Charlotte, North Carolina 28209
3 RECIPIENT'S ACCESSION NO.
5. REPORT DATE
January 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO,
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21BAV-014
'1. CONTRACT/GRANT NO.
68-02-1322, Task 6
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final: 4/74-3/75
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
report gives results of a study of dual- absorption contact sulfuric acid
plants, as well as single -absorption plants equipped with vent gas cleaning systems
for removal of SO2 , to determine the relationship between process parameters and
air emissions. Processes studied were dual- absorption acid plants and single-
absorption acid plants equipped with sodium scrubbers, ammonia scrubbers, and
molecular sieve adsorbers. Emissions considered were SO2 and acid mist emis-
sions and vent gas opacity. Relationships were developed for normal operations
and compared to off-normal operations such as shutdown, start-up, malfunction, and
misoperation. Process parameters and emission relationships are presented in
statistical, tabular, and graphic form. Converter bed operating temperature ranges
were established and causes of SO2 and acid mist emissions are illustrated from
plant operating data.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
Air Pollution
Industrial Plants
Sulfuric Acid
Mist
Sulfur Dioxide
Opacity
Scrubbers
Sodium
Ammonia
Absorbers
(Materials)
fo.lOENTIFIERS'OPEN ENDED TERMS
Air Pollution Control
Stationary Sources
Molecular Sieves
PRICES SUBJECT TO CHANGE
c. COSATl Ficld/CIroup
13B
07B
07A
11G
3. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS I Thu Krport;
Unclassified
?6~Sf CURI Tfv'CLASS'm
Unclassified
21. NO. OF PAGES
353
2?. PRICE
EPA Form 2720-1 (9-73)
-------
SPA-600/2-76-010
SULFURIC ACID PLANT
EMISSIONS DURING
START-UP, SHUTDOWN, AND MALFUNCTION
by
E.L. Calvin and F.D. Kodras
Catalytic, Lie.
P.O. Box 11402
Charlotte, NC 28209
Contract No. 68-02-1322, Task 6
ROAP No. 21BAV-014
Program Element No. 1AB013
EPA Task Officer: R.V. Hendriks
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triaugle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
January 1976
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development,
U.S. Environmental Protection Agency, have been grouped into
five series. These five broad categories were established to
facilitate further development and application of environmental
technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in
related fields. The five series are:
1. Environmental JealtK Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This scries describes research performed
to develop and demonstrate instrumentation, equipment and
methodology to repair or prevent environmental degradation from
point and non-point sources of pollution. This work provides the
new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental Protection
Agency, and approved for publication. Approval does not signify that
the contents necessarily reflect the views and policies of the Agency, nor
does mention of trade names or commercial products constitute endorse-
ment or recommendation for use.
This document is available to the public through the National
Technical Information Service, Springfield, Virginia 22161.
«•
u
-------
TABLE 0" CONTENTS
Paqe
LIST OF FIG'JRES VI
LIST OF TABLES VIII
ACKNOWLEDGEMENTS x
SECTIONS
I. SUMMARY Ax:n CONCLUSIONS 1
TT T VT1 •">("•'tT'^IT-"VT A
.L X • -Li*.i.;\\>l^Ov-*lXOkiJ. • » • • « • • • • • • • • • • . 4
III. STUDY OBJECTIVE AND APPROACH 8
IV. PROCESS DESCRIPTION 13
Contact Sulfune Acid Process ; 13
"ingle Absorption Process 14
Dual Absorption Process 16
Sulfur Recovery Process 20
Sodium Scrubber System 21
Ammonia Scrubber System 24
Molecular Sieve Adsorption Process ... 29
Ins-rui-icr.T-^tio-i an:. Control 3"".
V. ' FTjRVEY 0!- AC IT PLANT EMISSION'S AND C-TCTROLS . 17
S--v ovt I'-:! or : lants for Field Surveys .... 37
Ty:>r of rnission Contrcl Systems 1C
PrniSHiOn Dat=. Obtained 44
VI. ANALYSIS OF PROCESS VARIABLES A?:D EMISSIONS . 50
irov:;;t.l i'lirsi "r-~ rat ions ^0
Ccnvarler Tor.perac'.:ro Control 51
."irurle Absorption -lants 59
i
-------
Page
Single Absorption Plants with Gas
Cleaning Systems 64
Dual Absorption Plants 65
Shutdown Operations 72
Start-Up Operations 79
Single Absorption Plants ?9
Single Absorption Plants with Vent Gas
Cleaning Systems 84
Dual Absorption Plants 89
Malfunction Conditions 94
Single Absorption Plants 94
Single Absorption Plants with Tail Gas
Cleaning 98
Dual Absorption Plants 107
VII. DISCUSSION 113
Emissions During N'ormal Operation 113
Steady State Emissions and Equipment
Limits 115
Feedstock Changes 121
EPA Performance Parameters 122
Shutdown Emissions 125
Planned Shutdowns 125
Unplanned Shutdown and Equipment
Malfunctions 127
Emissions from Misoperation of Plant . . . 134
Start-Up Procedures 135
New Plant and New Catalyst Start-Up. . . . 136
Cold Plant Start-Up - Dual Absorption. . . 137
IV
-------
Page
Warm Plant Start-Up « Dual Absorption. . . 139
Vent Gas Cleaning Systems - Single
Absorption 139
VIII. REFERENCES 142
IX. GLOSSARY 144
APPENDICES
Appendix A Process Descriptions 158
Appendix B Plant Start-Up Data 197
Appendix C Statistical Analysis 216
-------
LIST. OF FIGURES
No. Page
1. Single Absorption Sulfur Burning Contact Sulfuric
Acid Plant Process Flow Diagram 15
2. Dual Absorption Sulfur Burning Contact Sulfuric
Acid Plant Process Flow Diagram 18
3. Sodium Scrubber Tail Gas Cleaning System Process
Flow Diagram 22
4. Ammonia Scrubber Tail Gas Cleaning System Process
Flow Diagram 25
5. Molecular Sieve Tail Gas Cleaning System Process
Flow Diagram. . . 30
6. Dual Absorption Sulfur Earning Contact Sulfuric
Acid Plant P. and I. Diagram 33
7. E.P.A. Sulfur Dioxide Emission Standard for New
Contact Acid Plants 47
8. E.P.A. Acid Mist Emission Standard for New Contact
Acid Plants 48
9. Normal Range of Operation of Single Absorption and
Dual Absorption A';id Plants as Found in this Study . 57
10. Monthly Variations in Acid Mist Emission Over the
Past Six Years (Plant U,, Bright Sulfur Burned in
a Single Absorption Contact Acid Plant) 62
11. Normal Operating Range for Dual Absorption and
Single Absorption Acid Plants 71
12. Dual Absorption Average Emissions After Start-Up . . 76
13. Example of SO-, Emissions During Start-Up of Single
Absorption Acid Plant vs. Single Absorption Acid
Plant with Cominco-Type Ammonia Scrubber Plant U,. . 83
14. An Example of a Single Absorption Plant "U2"
Entering and Leaving Malfunction Operating
Conditions 88
15. Cold Start-Up of Typical Dual Absorption Acid
Plants 90
VI
-------
No.
16. An Example of a Dual Absorption Plant ("Q") Entering
and Leaving Malfunction Operating Conditions -
During Start-Up 93
17. Acid Mist Emission Average Operating Levels for
Both Dual (Plant "Q") and Single (Plant U, & U-)
Contact Acid Plants . . 7 ..... 96
18. Normal Range of Operation for Acid Mist Emissions
of New Dual Absorption and Single Absorption Acid
Pints Burning Bright and Dark Sulfur. ....... 97
19. Computer Plot of Acid Mist vs. Time. ........ 99
20. Malfunction of Single Absorption Acid Plant "K"
with Ammonia Scrubber 100
21. Malfunction of Single Absorption Acid Plant "K"
with Ammonia Scrubber .101
22. Malfunction of Single Absorption Acid Plant "K"
with Ammonia Scrubber. . 102
23. Malfunction of Siirvjle Absorption Plant "Y" with
Sodium Scrubber 106
24. Transient Process Conditions During Dual Absorption
Plant Malfunctions 107
VII
-------
LIST OF TABLES
No. Page
1. A List of Candidate Companies for the Su1furic
Acid Plant Malfunctions Emissions Survey 39
2. Summary of Emissions from Various Types of ,
Contact Sulfuric Acid Plants Surveyed During
Normal Operations 41
3. E.P.A. New Source Performance Standards Test
Data Dual Absorption Contact Acid Plant (Plant "Q") . 46
4A. Normal Steady State Operating Temperatures and
Conversions in Each Converter Stage of Contact
Acid Plants 51
4B. Converter Temperature and SO- Outlet Concen-
tration Plant "J"—Upset Conditions 55
4C. Converter Temperature and S02 outlet Concen-
tration Plant "J"—Normal Operation 56
5. Plant "X" Test Data 59
6. Acid Mist Test 61
7. Acid Mist Observations 63
8. E.P.A. Performance Test Results for SO2 and Acid
Mist Emission Control Effectiveness from Sodium
Scrubber (Plant "Y") 66
9. Effect of Primary Absorber in the Dual Absorption
Process (Plant "Q") 67
10. Summary of Test Results from Dual Absorption Plant
"Q" 69
11. The Effect of Downtime Shutdown Durations Upon
Etcluent S02 Concentrations (pprn) During Start-Up
(For Plant *J")
74
12. Summary of Stait-Up Data on Dual (DAP) and Sinqle
(SAP) Absorption Acid Plants for Figures 25 through
41 77
13. Start-Up Temperatures for Plant "U," Single Absorp-
tion Acid Plant with Ammonia ScrubEer 81
VIII
-------
No^ Page
14. Start-Up Plant "U^"-Ammonia Scrubber Start-Up
Conditions and Emissions 82
15. Start-Up Temperature Contact Acid Plant "U-"
Single Absorption Plant with Ammonia Scrubber. ... 86
16. Ammonia Scrubber Start-Up Conditions and Emissions
on Plant "U2" (with Mist Eliminator) 87
IX
-------
ACKNOWLEDGEMENTS
The assistance of the EPA Task Officer, Mr. Robert V. Hendriks,
is acknowledge-! with sincere appreciation. Mr. Donald Carey, Divi-
sion of Station,iry Enforcement, EPA, Washington, D. C., also
provide.-.i , iKiable information and guidance.
The personnel of the Technical Library, Continuous Monitoring
Branch, and Enforcement Branch, EPA, Research Triangle Park,
North Carolina furnished useful data during the preparation
of this report. This support is appreciated.
The EPA Renion IV personnel, Atlanta, Georgia and the Florida
Department ot. Pollution Control personnel, Winter Haven,
Florida, supported the project through use of their experience
and knowledge in performance testing.
Special assistance in obtaining useful operating data and
operating experience was provided by members of the Manufac-
turing Chemists Association, the Florida Phosphate Council,
and other cooperating non-member companies.
-------
SECTION I
SUMMARY AND CONCLUSIONS
Data were collected from twelve sulfuric acid plants to estab-
lish the relationship between atmospheric emissions (S0_» acid
mist and opacity) and process parameters. The data were col-
lected from single absorption and dual absorption contact acid
plants and single absorption plants with tail gas cleaning
systems. The tail gas cleaning systems studied were the sodium
scrubbing system (Wellman-Lord sodium bisulfite), the ammonia
scrubbing system and the molecular sieve absorption system.
Operating parameters were related to emissions during shutdown,
start-up, malfunctions and misoperation conditions. Conclu-
sions that can be drawn from the data collected and the plants
studied are presented in the table of Summary and Conclusions.
Obviously, all malfunctions and all aspects of operations con-
tributing to emissions could not be covered in this study.
Those that were covered represent occurrences where data was
available at the plants visited.
-------
AND CQNCmsioKS
CAOSZ of
nUMSIENT
CONDITION
Converter Bed Inlet
tavpereture Out of
Range
3oaJ Absorption Plant
Planned Shutdown
Dual Absorption Pluit
SAP With scrubber
Start-Op
Dual Absorption Plant
•tart-Dp
Dual Absorption Plant
Conversion Efficien-
cy Reduisd
Sulfur r««d Stopped
and Unit Purged of
•era Plant After
Short Ten Shutdown
«4 be.)
Cold Plant After
Long Ten Shutdown
(>4 Irs.)
Start-Op
Dual Absorption Plant
•Isoperation of
•odium and Ajwnonla
Scrubber Systems
Hlsoperation of
Ammonia Scrubber
•yst
Cold Plant After
Long Tern Shutdown
(>4 Irs.)
Operation at Cas
Plow Above Design
Limit
Blgh Gas Flow
Mates and Solution
Concentration Too
OOnsCTZVB ACTION
Manual Adjustment of
Dampers, Adjustment
or Repair of Instru-
ments
Manual Adjustment of
Valves and Control-
lers
Preheat of Plant Not
Required
PI* ». Must •* Pre-
heated Dp to S Days
Of
OCCUMIBfCe
Dally
DURATION
OP
CONTROL
2 Irs.
4/Yr. to i lour to
1 time/2 Tr. I neeJia
U/tr.
1-2 Times/
Tr.
4 ire.
J-» Days
Plant Receives Mini-
mum Beat Mfore SO.
Addition
Operate Scrubber with
Cas Plow Within De-
sign
1-4 Times/
Tr.
Dally
STACK OR
EMISSIONS CONTROL
SO]. 4°° * 'roper Operation
1,000 PPM and Control and
Maintenance
None Proper Procedure
for Shutdown to
Facilitate start-
Up
< 100 PPM Beat Conservation
SO. and Proper Stait-
Op Procedures
<100 PPM Complete Preheat
SO, and Reduced SO,
Inlet Cone. Stlrt-
Dp Period «-)
Days Oelnf Proper
Start-Op Proce-
dore
100 to 3000 Minimum Preheat
PM SO. Is Applied and
* Inlet SO. Con-
centration ie
Started At Pull
•ate
Range of Permissible Var- T«bl« 4A
iation Prom Temperature
Specifications on Inlet
Temperatures Is 1 to X
of Specified Temperature
Planned Shutdown Should
Not Cause CalliIon* on
Shutdown. Proper Shut-
down Klnimiies Start-Up
Emissions.
SO, Concentration Inlet
to Converter Host Impor-
tant to SO,
C-vUol
•laeion
Reduce 6a* 'low *•»•
and Increase Solution
Coooentration
1-4 Ire.
1-4 Irs.
Op to
1,200
Op to
1.100
Reduce Cat Plow
Rate. Proper
Instruments end
Procedures.
Operate With O*s
Plow Within De-
sign and Solution
Concentration At
7%. Proper In-
stoments and Pro-
cedures.
SO, Concentration Can
Be Kept Below NSPS
With Sufficient Preheat
Tine and Low Initial
Inlet SO, Concentrations.
This is ideal And Seldom
Deed Because of Produc-
tion Loss During Long
•tart-Op. Auxiliary
Pzeneaters Required for
Paster Beat Op.
Start-up Method Host
Cotewnly Used to
Hlnlmlie Start-Up Time
and Product Lois.
Balance With Above
Method Host Desirable
Moat Common Cause of Blgh
SO, Emissions is High Cas
Plow. More Conservative
Design Should Be
Operation Hlth Solution
Concentration at 41 (Spec*
If led Set Point) Hill Re-
duce KM, Consumption But
Mill Produce High SO]
emissions.
Figure 21
Teble 11 I
12
Table 11
Figure 12
1 16
Figure 22
t 2)
Figure 20
* 22
-------
cuss or
TMMIENT
COKDtTIO*
Hi.aH. Miter On Ajno-
al*.Scrubber Outlet
Inoperative
lot* of Sulfur flow
to Furnace
•Igb Sulfur Flow to
runic*
to«« of Dilution
Water ta Primary
Afcjorber
Loss of DMutlon
Water to Secondary
Absorber
feed of Dark Sulfur
or Sludge Mid
reed of Dark Sulfur
or Sludge Acid
Blfh Opacity a:id
High SOj Readings
On Raich T«»t
Low SO, talet
Concent*atloa
Blgh SO, Inlet
Concentration
Increased Absorber
Acid Concentration
Increased Absorber
Acid Concentration
Increased Water la
CORMCT1VI ACTION
Rector* Effective Op-
eration of Brink
filter
Restore turner or
Pu*p Operation or
Onpluq Burners
Reduce Sulfur flow
»o Purnece or Incrnsse
Air for Proper SO,
Ooacentration
Knstore Dilution
Water Plow to Puap
or
OCCURRENCE
1/2 Tear
frequent
frequent
Infrequent
DURATION
Of
CONTROL
Continuone
Until
Repaired
1-4 Sours
Depend*
Opifl
Depend*
Upon
Operator
Attention ,
Continuous
Until
••pel red
STACX
EMISSIONS
SO, Read-
Ing Up to
1.000 PPM
Blgh Opac-
ity
None
Sigh B02
SO, In-
crla*ed
Small SO.
and Opac*
ity In-
pRiuuriON
OR
CONTROL
Repair Leak in
•rink rilter In-
ternals. Place
Booster Blower
and Brink in
SAT* ice.
Use Clean Sul-
fur, Proper
Sulfur System
Maintenance and
operator Atten-
Control SO.
Inlet Concfn-
tration Consist-
ent with Catalyst
•CoBdltlOB)
Assure Adequate
Supply and In-
strument Mainte-
nance. Operator
Inspection.
HO, G«nv*tloo in
Pvim«c«
Restore Dilution
vster Plov to Puap
Teak
CTuuu.e reed, rilter
Sulfur
Cbsnge reed, Filter
Sulfur
Infrequent
Continuous
Until
(•paired
Continuous
OBU1
SO, and
Opacity
Increased
Acid Hist
Increaaed
Continuous Ac'.d Hist
Until Increase
read Ottaee
Assure Adequate
Supply and In-
strument, Mainte-
naace. operator
Inspection.
laproved Denis-
tars and rilter*
such A* Blink
Blgh Efficiency
leproved Deais-
ters and rilter*
Such As Brink
Blgb Efficiency.
Oee fi Mr
Brink SoawtiJM* Bypassed Table 1
on Start-Up Because of * 14
Lack of Steaa for Booster
Blower. Corrosion Can
Cauae Leak Around Pilter
Sleeve Connections.
Loss of Sulfur Peed Hill
Not Cause Riqh SO, Unless
Extended Period Reduces
Converter Temp.
Dual Absorption Plant
Normally Uae* H^her SO,
Inlet Concentration ThaA
Single Absorption. Cata-
lyst Condition Lialts
Concentration.
Reduced SO. Absorption in
Priiury Abiorber tncreates
SO. Exit Converter by
EqOlllbrlua Shift in Suc-
ceeding Converter Stage*.
SO. and Opacity Nay Bo
Increased.
Secondary Abeorber Will
Increaae SO. and Opacity
But Hill Not Increase SO.
Emissions
Additional Mater in Systeai
Causes Generation of Acid
Hist. Hodem Dual Absorp-
tion Plants Mill Remove
Most With Efficient De-
Bister*.
Presence of NO in Gas
Streu Causes formation
of Acid Mlat By Gas Phase
Reaction Through Combina-
Plqure 33
Plfure 2)
figure 23
figure .21
Figure 17
4 1»
tion
SO,
-------
SECTION II
INTRODUCTION
EPA promulgated emissions standard of performance for a
number of stationary sources including sulfuric acid plants
on December 23, 1971, based upon steady-state operations.
Subsequent revisions to Section 60.7 (c) provide for a
quarterly report to be submitted to EPA detailing " — excess
emissions as defined in the applicable subparts" (subpart
H for sulfuric acid plants of the regulation). "The report
shall include the magnitude of excess emissions as measured
by the required monitoring equipment reduced to the units
of the applicable standard, the date, and time of commence-
ment and completion of each period of excess emissions.
Periods of excess emissions due to start-up, shutdown, and
malfunction shall be specifically identified. The nature and
cause of any malfunction (if known) , the corrective action
taken, or preventive measures adopted shall be reported".
The standards for new or modified sulfuric acid plants are
summarized below:
SULFUR DIOXIDE
No more than 2 kg of sulfur dioxide (SO,) per metric ton of
acid (100 percent H^SO.) produced (4.0 pounds per short ton)
-------
ACID MIST
No more than 0.75 kg of acid mist (measured as H2SO4) per
metric ton of acid (4 Ib. per short ton) (100 percent H2S04)
produced.
No visible air pollutants shall be released to the atmosphere.
These standards of performance are applied to new contact-
process sulfuric acid and oleum facilities that burn elemental
sulfur, alkylation acid, hydrogen sulfide, organic sulfides,
mercaptans, or acid sludge. They do not apply to metallurgi-
cal plants that use acid plants as SO- control sy,stems, or to
chamber process plants or acid concentrators.
These performance standards were based on tests of sulfuric
acid plants, and the judgement of the Environmental Protection
Agency is that the standards can be met with present available
control systems at a reasonable cost.
PERFORMANCE TESTING
Testing must be performed no later than 60 days after achiev-
ing the design production rate, but no longer than 180 days
after initial start-up. These tests will be conducted at or
above the design production rate.
Owner or operator responsibilities:
1. To perform the tests.
2. Give minimum of ten days notification of scheduled teats.
3. Provide all equipment necessary to conduct tests.
-------
4. Provide safe access to the sampling site.
5. Furnish a written report of the test results.
EPA personnel have the right to perform additional tests at
any reasonable time under conditions based on representative
performance.
The various testing methods are specified in the standards
for each pollutant. Each test consists of three repetitions
of the applicable test method. The average of the repetitions
will be used to determine compliance.
STACK AND PROCESS MONITORING
The instrumentation required for monitoring stack emissions
and production rate is listed below:
1. Oxides of sulfur detector for continuously monitoring the
stack. It must read in volume ppm of oxides of sulfur as
sulfur dioxide.
2. A suitable flow meter for determining the production rate
cf sulfuric acid in tons per hour as 100 percent sulfuric
acid. An appropriate tank inventory system is also ac-
ceptable.
RECORDKEEPING AND REPORTING
All records are to bo kept by the facility for two years fol-
lowing the date of measurement and summary. The plant will
make available to EPA any records necessary to determine
performance. Emission data shall be irMe available to the
public, although no trade secrets, commercial, financial, or
-------
other private information will be disclosed.
The items to be recorded, and frequency of data to be
recorded is specified in the Standards of Performance.
-------
SECTION III
STUDY OBJECTIVE AND APPROACH
STUDY CBJECTIVE
The objective of this study is to establish the relationship
between sulfuric acid plant emissions (SO2f particulates, and
visible emissions) and process parameters for normal produc-
tion operations and for other operations such as start-up,
shutdown, and mechanical malfunctions. The relationships
developed are based upon data obtained from actual plant opera-
tions representing realistic operating conditions. Theroretical
considerations were used only where required to complete the
data used to establish the relationships. A maxirurn use of
operating data and statistical analysis using measured emis-
sions has strengthened the reliability of the process operating
parameters. The study is intended to establish (to the great-
est extent possible) an understanding of the cause and effect
relationships between process plant operating parameters and
emissions.
The project included the study of contact sulfuric acid
plants using double absorption process techniques that are
capable of meeting the EPA New Sources Performance Standards
(NSPS) under normal operating conditions and those using the
tail gas clean-up processes applied to the single absorption
type of plants. The study is further limited primarily to
plants using elemental sulfur combustion as a source of sul-
fur dioxide although one plant burning sludge acid was studied.
8
-------
This report compiles the available information that has been
developed during the short operating history of dual adsorp-
tion plants that are located in the United States. The report
also points out additional areas of needed information that
should be obtained from future detailed experience data.
APPROACH
A data collection and analysis plan was employed to provide
an analysis and definition of operations that have an influence
on emissions under normal steady-state operations and under
upset conditions. Data was gathered on the existing dual
absorption plants and single absorption plants with SOX scrub-
ber systems that clean up vent gases from the exhaust stack
of the acid plant. The data included periods of upset opera-
tions resulting from production equipment failures., process
control instrumentation failures, electric power and water
utility losses, start-up and shutdown operations, changes
in sulfur quality and SO- feed conditions, and product con-
centration changes.
The four main sources of data and information were: (1) EPA
v
survey of continuous monitoring records and studies, (2) acid-
manufacturing plant operator's log books, (3) designer-engi-
neering consulting firrus data and (4) acid plant equipment
manufacturers performance data. Additional information was
obtained from a technical literature search and in-house
files. Contact was also made with independent researchers.
-------
Five companies operating 12 acid plants were selected for
visits. They are representative of the present sulfuric acid
industry with typical sulfur-burning configurations using
double absorption processes and single absorption plants with
some type of sulfur oxide control equipment.
Plant selection criteria included the best available technology
for emission reduction to meet the EPA new plant performance
standards.
(1) Capacity should be representative of medium, or
large commercial scale operation.
(2) Ability to achieve 99.7 per cent SO2 removal effi-
ciency without secondary pollution.
(3) Completeness of process control instrumentation at
each plant equipped with stack emission monitoring
equipment.
(4) RecoCffftftfed leadership in pollution control system
applications and a cooperative attitude in supply-
ing reliable and complete data from both current
and historical records.
(5) Application of state-of-the-art design to plants
operating at full capacity without experimental
design factors.
(6) Complete malfunction notations and failure diagnos-
tics recorded on data logs and charts.
The inspection team consisted of two men. Standard data forms
for efficient transcription of data into the data processing
10
-------
system were used where possible. Data were obtained directly
from plant operations and by examination of existing data
logs in their original form. Where possible, these log sheets
were photocopied.
Five acid plant manufacutrers and designers of sulfuric acid
production and control equipment were interviewed by Catalytic
engineers. Process design and operation test data were col-
lected where available to supplement data collected from all
plant operators.
Data from completed EPA studies and test programs and current
emission testing programs on sufluric acid plants were ana-
lyzed in addition to the data from plant operators and de-
signers. EPA also provided plant operating data obtained
from performance tests made to establish the compliance of
sulfuric acid plants.
Computer programs were used for data reduction and statistical
analysis to establish the effect of process parameters on
emissions from the sulfuric acid plant.
An analysis was made of the important variables in the plant
operation as they affected emissions from the process. The
range of data was obtained during most the important expected
off-normal, start-up, shutdown, and normal operations. The data
was sorted and prepared for input to a multiple linear regres-
sion analysis program and standard statistical analysis. The
result of the regression correlations and statistical analysis
11
-------
assisted in determining the significance of each of the vari-
ables in controlling the emissions. Process parameter vari-
ations were correlated with operating conditions and probable
causes of off-normal operations.
12
-------
SECTION IV
PROCESS DESCRIPTION
CONTACT SULFURIC ACID PROCESS
The basic contact sulfuric acid process was patented in England
in 1831. Since this time continued improvement has refined the
process to the present highly efficient operation. Although
variations occur in the basic process resulting from differences
in feed stock or application, all contact acid plants contain
the same five basic operations. These basic operations are;
(1) burning sulfur or sulfur bearing feed stocks to produce
sulfur dioxide, (2) cooling the resulting SO- containing com-
bustion gas, (3) catalytic oxidation of the SO2 to SOj, (4)
cooling the resulting oxidized gas containing SO,, (5) absorp-
tion of SO, in strong sulfuric acid.
The simplest contact plant results from use of elemental
sulfur as feed stock. Where other feed stocks such as spent
acid or acid sludge containing moisture and organ! ^ are used,
additional processing steps must be added to remove the re-
sulting water and particulate matter before processing the
combustion products in the catalytic converter. Feed stock
variation also affects the sulfur conversion ratio, the volume
of exhaust gases, and the character and amount of pollution
emitted. The processes discussed in this chapter all use ele-
mental sulfur as feed stock.
More detailed process descriptions, material balances and
equipment lists for all processes are included in Appendix A.
13
-------
This appendix includes complete operating data for a typical
single absorption acid plant taken from actual plant opera-
tion.
SINGLE ABSORPTION PROCESS
A simplified process flow diagram for a single absorption
contact sulfuric acid plant burning elemental sulfur is pre-
sented in Figure 1.
In this process sulfur is burned with air to form a gas mix-
ture containing approximately eight percent sulfur dioxide,
13 percent oxygen, and 79 percent nitrogen. Combustion air
is pre-dir'.ed by passing it through a packed tower circulating
93 to 98 percent sulfuric acid. Pre-drying the air minimizes
acid mist formation and resultant corrosion throughout the
system.
Combustion products from the sulfur furnace pass through a
boiler to cool the gas and generate process steam. The com-
bustion products leaving the waste heat boiler contain sulfur
dioxide and excess oxygen. These gases are passed through a
multi-bed converter containing vanadium pentoxide catalyst
that promotes the combination of sulfur dioxide and oxygen
to produce sulfur trioxide (SO,). The catalytic oxidation of
SOj to SO, increases the temperature of the gas mixture in
the catalytic converter. The heat generated in the first
stages OL catalytic oxidation must be removed to control the
temperature for succeeding stages of conversion. This heat
14
-------
.£..&
,L* .0^1 ."Htf.t E^iuy^
812
*^
f"
— jOr— ••
eviEii*
iTnn'
JS>
r s / ;-r
lizZXZ
< *««
C'WvtS'M
rmucooun
mTin
,„..__—
-H
**—..
4
......
1
\ X
X*X
-«| Vt«u )
is=LJ
f
:
OLEUM MMUCT
SICMCC TAliK
-£
£SD SKIS'
ABSOW'flON
KID P! «N'
pftxtss FLOW . LUJI
FIGURE '
KEV
CASCS ..-.-»
-------
is removed in additional waste heat boilers to generate pro-
cess steam. Temperature of catalyst beds is also controlled
by introduction of cold air into the converter. This is the
air quench system. The temperature of the gases leaving the
converter ts approximately 806F to 815F with approximately
98 percent of the SO, converted to SO.,. These gases are then
passed through an economizer to preheat the boiler feedwater
going to the waste heat boilers.
Sulfuric acid is produced by passing the gases leaving the
economizer through an absorption tower where the SO3 is ab-
sorbed in hot 98.5 to 99.0 percent sulfuric acid. In the
absorption tower, sulfuric acid of desired strength is pro-
duced by controlling the acid concentration and watar make-
up and temperature of the feed acid. If fuming sulfuric acid
or oleum is required, the gases containing SO, leaving the
economizer are first passed through an oleum tower where SO,
is dissolved in oleum to produce oleum. The gas stream leav-
ing the oleum tower is further stripped of SO, by passing
through a normal acid absorber containing 98.5 to 98.8
percent sulfuric acid. The majority of single absorption
plants use a single absorption tower and produce a product
acid with a concentration of 98 to 99 weight percent sul-
furic acid.
A DUAL ABSORPTION PROCESS
The average single absorption sulfuric acid plant will produce
SO- emissions in the range of 1600 ppm to 2000 volume ppm oper-
16
-------
ating at normal efficiency. Since this emission level is in
excess of the four pounds of SO~ per ton of production allowed
by the current standards, some modification of the single
absorption process is required. The need for more efficient
sulfuric acid plants initiated the development if the dual
absorption plant. The dual absorption process can convert
99.7 to 99.8 percent of the sulfur dioxide to sulfur trioxide
for producing sulfuric acid.
A typical modern dual absorption plant burning elementa^. sul-
fvr is shown in Figures 2. The primary difference between the
single absorption and the dual absorption process is the
addition oC a primary SO, absorber for gas leaving the third
catalyst bed. One process uses this absorber after
the second bed. Since the addition of an absorber between
catalyst beds requires cooling and reheating the process gas
a change in the heat recovery system is al-so required.
Comparison of Figures No. 1 and 2 will show that the sulfur
combustion portions of the single and dual absorption plants
are similar. Combustion air is compressed and dried in a
drying tower with 93 to 98 percent sulfuric acid before use
for combustion of sulfur in the sulfur furnace. The hot gases
'v
from a sulfur furnace then pass through the waste heat boiler
to generate process steam. The waste heat boiler is designed
to permit further cooling of the combustion products to approx-
imately 795 to 820F in the No. 1 heat exchanger by reheating
part of the gas from the primary absorber. The cool gases are
17
-------
DuAl MbSOWTICN
-------
then passed through the first bed of the catalytic converter
where the gas temperature is increased to approximately 1100
to 1130F. The high temperature gas exiting the first catalyst
stage is cooled in No. 2 heat exchanger to approximately 820F
for reaction in the second catalyst bed. The heat is used to
reheat the remainder of the gas from the primary absorber.
Heat generated in the second catalyst bed is removed by a
steam superheater in the process steam Line from the waste
heat boiler. Heat generated in the third catalyst bed is re-
moved by an economizer in the boiler feedwater system before
the gas is fed to the primary absorption tower. In the
primary absorption tower the concentration of SO., in the gas
is reduced to about 100 ppm by contact with 98.5 percent
sulfuric acid. Cold gas leaving the primary absorption
tower must be reheated before introduction to the fourth
catalyst bed. This is accomplished by passing in parallel
through No. 1 and No. 2 heat exchangers.
Approximately 97 percent of the SO2 remaining in the gas
stream is converted to SO_ in the fourth catalyst bed. This
is a much high overall conversion rate than is possible in
the single absorption plant. The increased conversion
efficiency results from the lower partial pressure of SO,
in the gas permitting the reaction to be driven more nearly
to completion. The gases leaving the fourth absorption
bed are fr'nally copied in a second economizer heating boiler
feedwater before contacting 98.5 percent sulfuric acid in
the secondary absorption tower. The gases leaving the
19
-------
secondary absorption tower will contain approximately 100
to 300 parts per million SO- under normal operations and will
meet the existing emission standards without further processing.
If oleum is required, an oleum tower is inserted before the
primary absorber in a manner similar to the single absorption
process.
With the exception of the location of the primary absorber and
heat exchangers previously discussed, all major designers of
dual absorption sulfuric plants use the same basic equipment
configuration. Important differences between these designs
are found in the design details of the converter, heat ex-
changers and absorbers. Air quench is also used in some dual
absorption plants.
SULFUR DIOXIDE RECOVERY PROCESSES
The typical single absorption contact sulfuric acid plant is
not capable of meeting the current S02 emission standards
using an economically feasible converter design. There is
a lower limit to the concentration of SO, in the vent gas
imposed by equilibrium conditions within the catalytic con-
verter and the amount of catalyst that can be installed eco-
nomically in the plant. Because of these limitations the
single absorption sulfuric plant must be equipped with sulfur
dioxide recovery processes to reduce the SO^ concentration
in the vent gas to the acceptable level.
Many processes have been proposed to perform this operation
but only three have been applied commercially. These are
20
-------
the sodium scrubber process, the ammonia scrubber process,
and the molecular sieve adsorption process. These three pro-
cesses will be described in the following paragraphs.
Sodium Scrubber System
The sodium scrubber process for removing sulfur dioxide that
was included in this study is a wet regenerative system based
on a sodium sulfite/bisulfite cycle. This process, currently
being marketed by Davy Powergas Inc. under the Wellman-Lord
trademark, is represented by Figure 3 that depicts the
absorber and chemical regeneration areas. The reactions
that take place in the process can be simplified to the
following:
Absorption - SO- + Na_SO3 + H20 * 2 NaHSO,
Regeneration - 2 NaHSO., * Na2SO, + SO2 + H2O
Sodium sulfate (Na2SO.), which is non-regenerable in the nor-
mal process, is formed in the absorber as a result of the
reaction between the sulfite ion and oxygen or sulfur trioxide
'as follows:
Na2SO_ + 1/20_ > Na_SO.
2 Na2SO3 + SO.. + H2O > Na2SO. + NaHSO.,
The sodium sulfate formed is controlled at a level of approx-
imately five to ten percent by weight in the absorber feed
stream by maintaining a continuous purge from the system.
Caustic addition is required to replace the sodium lost in
the sodium sulfate purge. This caustic make-up solution
21
-------
"!'
•
A
\ /
ja%?°
tinge
S
MK^cm.uac. *
1
y
&.
/
/
/
i^2J
^
USSOLViNC
SODIUM SCmjBBER
gSsa«S!-o
FIGURE 3
-------
reacts with the sodium bisulfite in the absorber to form
additional sodium sulfite solution:
NaOH f NaHSO3 + Ma2SO3 + H2O
The caustic can also react with the S02 to form bisulfite as
follows:
NaOH -v SO2 > NaHSO3
The SO2~rich gas is contacted counter-currently in the absorber
by the sodium sulfite solution and passes out the top stripped
of SO_. An absorber with inlet concentration of 1560 parts
per million SO- and 95 percent scrubber efficiency can achieve
an emission of 86 parts per million SO_. The absorber bleed
stream, rich in sodium bisulfite, (NaHSO.), is discharged to
a holding tank. From the holding tank the solution flows
to the chemical regeneration area for recovery of the SO2 gas.
The sodium bisulfite solution flows to the evaporator crys-
tallizer in the chemical regeneration area where it is boiled
by indirect heating with steam, resulting in the decomposition
of the sodium bisulfite solution into SO2/ water vapor, and
a precipitate of sodium sulfite crystals.
The overhead gas from the evaporator-crystallizer passes
through a condenser to remove most of the water vapor. Con-
densate is recycled to the dissolving tank to dissolve the
sodium sulfite crystals and the product SO2 gas is recycled
to the sulfuric acid plant or Claus elemental sulfur plant.
s
Precipitate from the evaporator-crystallizer is processed
23
-------
through a centrifuge where sodium sulfite solids are discharged
to a dissolving tank for recycle to the absorber. The bulk
of the mother liquor from the centrifuge is recycled to the
evaporator-crystallizer for solids density control. A small
portion of the mother liquor is purged from the process to
control the level of sodium sulfate as previously stated. The
sodium ions lost in the purge stream are replaced by addition
of sodium hydroxide (50 percent NaOH) solution directly to
the absorber.
Ammonia Scrubber Process
Ammonia has been used for removing sulfur dioxide from vent
gases in the sulfuric acid and smelter industries for many
years. Several commercial ammonia scrubber processes are
available for application to sulfuric acid plants. All of
i_l.ese processes are based upon the same chemical reactions.
A flowsheet of this process is shown in Figure 4.
Ammonia provides one of the most vigorous reactions with sul-
fur dioxide of all the alkaline materials and produces
soluble by-products. The primary reaction in the ammonia
scrubber is the conversion of ammonium sulfite to ammonium
bisulfite by the reaction with sulfur dioxide. The ammonium
bisulfite is regenerated by reaction with ammonia to again
form ammonium sulfite. The equations for the two reactions
are as follows:
NH, + NH.HSO, --- »> (NH.)- SO,
3 43 423
24
-------
10W£»
MS SCKUBKII
I
....A
/\
I..
1
WOCO TOWt*
OS SOUBNR
Y
AI/.'/ONIA SCRUBBER
ww
ruovv
CHI»K>N
SCUICt
i..
wie \
VrS'EH/
\AMMQNI* 1
D
-------
In applying the ammonia scrubbing system to a sulfuric acid
plant exhaust gas, the gas is first treated by spraying water
into the exhaust duct tc humidify and cool the gases leaving
the absorber. The cool cases are then passed through an ab-
sorber counter current to the flow of the ammonia brine and
the S02 removed by reaction with ammonium sulfite. The con-
centration of ammonia sulfite is controlled fay the addition
of ammonia on the basis of pH. Operation with a large concen-
tration of ammonium sulfite promotes the reaction with SO2
but increases the usage of ammonia since a large amount of
ammonium sulfite is withdrawn with the spent scrubber liquor.
Operation with a large concentration of ammonium bisulfite
reduces the absorption capability of the brine and increases
the concentration of SO, leaving the absorber. Careful con-
trol of the pH of scrubbing liquor is required for efficient
operation of the system.
A secondary reaction taking place in the absorber is the oxi-
dation of the aramonium sulfite to ammonium sulfate by reaction
with excess oxygen in the exhaust gas stream. The chemical
equation for this oxidation process is as follows:
)S0 + 1/2 0 ----
The oxidation of ammonium sulfite to sulfate is undesirable
since ammonium sulfate is unreactive in the scrubbing mech-
anism and removes ammonia from the system in the largest
ratio to the sulfur content.
26
-------
The top of the absorber is equipped with a mist entrainment
separator to prevent the carry over of ammonia brine in the
gas stream. This separator is usually of the wire mesh type.
The primary disadvantage of removing SO- by absorption with
ammonia brine is the generation of particulate matter by gas
phase reaction between SO2 and ammonia to form ammonium sul-
fite. The ammonium bisulfite particulate is very finely
<
divided and passes through the absorber and is -'isible in
the vent gas as a blue haze. The quantity of haze produced
depends upon the partial pressures of the various gases in
the vapor phase and cannot be completely eliminated using pH
control in a single stage absorption. Where the particulate
haze is not permissible in the exhaust gas from sulfuric acid
plants, it has been standard practice to equip the plant with
a venturi scrubber or a fiber pad demister operating at very
hiqh efficiencies to remove the submicron particulate. Al-
though these high efficiency filters are satsifactory for
removal of the particulates, the high pressure drop across
the filters requires a large investment in additional blower
capacity and energy consumption. Similar costs are encountered
using a hi.jh energy venturi scrubber for removal of the par-
ticulate.
Various processes have been developed for reclaiming the sul-
fur v?lues from the ammoniacal brine withdrawn from the absorber
27
-------
processes. These include acidulation with sulfuric acid,
thermal decomposition of the ammonium bisulfite to sulfite
and SO2 , and mixing of the ammonium sulf i te-bisulfite liquor
into fertilizer solutions. The thermal decomposition process
is still in pilot plant development by TVA and was not applied
commercially in any of the plants studied.
In the acidulation process, the ammoniacal brine from the
absorber is mixed with sulfuric acid before being admitted
to a packed column. In this column Ihe reaction of sulfuric
acid with ammonium sulf ite-bisulf ite takes place, liberating
gaseous SO- and generating ammonium su]fate. The primary
reactions in the acidulation process are shown in the follow-
ing equations .
)S0 + HO -C- S(>
2
2 NHHSO + HSO --- » (NH)S0 + 2HO + 2
The packed column disengages the gaseous SO- from the ammon-
ium sulfate solution so the SO2 can be recycled back to the
sulfuric acid plant while the ammonium sulfate solution is
collected for further processing. The ammonium sulfate solu-
tion is either concentrated and crystallized to produce solid
ammonium sulfate or is included in mixed fertilizers. When
the acid plant is a part of a fertilizer complex, the ammonium
sulfate or ammonium sulf ite-bisulfite brine can be used in
the diammonium phosphate production process as was the case
in Plant "U" included in the study. The brine can also be
used in other fertilizer concentrates using standard processes
28
-------
for manufacturing solid or fluid products.
Molecular Sieve Adsorption Process
One of the newest SC^ recovery processes to be applied com-
mercially to sulfuric acid plant vent gas is the PuraSiv S
process development by Union Carbide. This process uses a
dry bed molecular sieve material for removing S02 from the
vent gas. The primary features incorporated in this system
are the high removal efficiency for SOj especially at low
SO2 concentrations, freedom from liquid chemical handling
problems, absence of waste products, and simple operation.
Evaluation tests on production models of this system have
been run very recently and data could not be fully evaluated
in this study. However, general process descriptions are
included because of the prospects of future applications.
A procsss flow diagram for the molecular sieve sulfur dioxide
recovery system is shown in Figure 5. The tailgas from the
sulfuric acid plant at approximately 170F is cooled to 100F
by pai.sing through a water cooler. Any sulfuric acid en-
trained from the absorption tower is collected from this
cooler in an acid circulation pot. Acid from the circulation
pot is continuously recycled to the cooler to aid in conden-
sation and collectJLoji^of the sulfuric acid mist. Since it
is undesirable for liquid sulfuric acid to enter the molecu-
lar siev-» adsorption tower, the gas is passed through a filter
containing packed fiber elements to remove entrained sulfuric
acid. The sulfuric acid collected in this filter is also
29
-------
-.0".
©-i
•fcj-'
-itn
\
»_..
'/•„
»/
J
^szu
o-
A
VJVF'tLAP SlCV
-------
returned to the acid circulation pot. This feed preparation
system for the PuraSiv operation is similar to that included
in many sulfuric acid plants to reduce the amount of sulfuric
acid entrainment from the absorber system.
After removal of all excess sulfuric acid mist from the gas
stream the gases are passed through a tower packed with gran-
ular PuraSiv S. In this tower the SO2 is removed from the
vent gas to a level of 15 to 20 parts per million. The molec-
ular sieve adsorption system consists of two adsorber towers
to permit regeneration of one tower while the second is being
used to ad?orb S0_.
When the. irolecular sieve in one adsorption tower nears satu-
ration, as indicated by an increase in the sulfur dioxide con-
centration in the vent gas, the towers are switched and the
saturated tower is put on the regeneration cycle. Regenera-
tion of the PuraSiv S is accomplished by passing a hot, dry
air ^stream through the bed in reverse direction to strip out
the adsorbed SOj.
Regeneration air is brought into the plant through an air
filter and a water cooled indirect cooler to remove as much
of the water vapor as pos3ible. The air is then passed
through a desiccant bed dryer to remove the remaining moisture
before being preheated to 200F in indirect fired air heaters.
The air is then used for stripping the SO2 in the regeneration
cycle. Stripping air leaving the regenerating adsorber flows
31
-------
to the acid plant for mixing with dilution air in the sulfur
furnace. This air contains from four to 0.3 percent SO-
depending upon the point in -the regeneration cycle. In this
manner the recovered SO- is returned to the sulfuric acid
plant converter for recovery as sulfuric acid.
To provide an uninterrupted flow of regeneration air two des-
iccant towers are used for air drying. As in the SO2 adsorp-
tion towers, one tower is on adsorption cycle while the second
is being regenerated. Regeneration is accomplished by passing
the dried hot air from the fired air preheaters through the
desiccant bed in reverse direction to strip out the adsorbed
water which is then vented to atmosphere.
A control system is provided to determine the point when re-
generation of both the SO2 adsorber and the dryer is required
and to automatically cycle the units to place a newly regener-
ated tower into operation. In this manner there is no inter-
ruption of cither process or regenerating air flow. The
regeneration air flow required in the SO- absorber is approx-
imately 20 percent of the design process flow. Since about
the same percentage of the total air flow is required for re-
generation of the air dryers, the air drying units are small
compared to the SO2 adsorbers.
The SO2 content of the recycled stripping air varies from
four to 0.3 percent SO2 throughout the regeneration cycle.
The addition of this SO2 to the SO2 concentration leaving
32
-------
the furnace will vary the SO- concentration entering the
converter also. During normal operation this variation of
SO2 entering the converter do«=>s not adversely affect the
operation of the sulfuric acid process. Also, since the ad-
sorption efficiency of the PuraSiv S is not highly dependent
upon flow rate or concentration of S02» normal upsets in
acid plant operation do not affect the concentration of SO-
in the vent gas from the PuraSiv S system. It has also been
found that the presence of CO- and NO in the vent gas as a
£* Jt *,
result of burning acid sludges or dirty sulfur do not affect
the adsorption efficiency of the PuraSiv S system.
Instrumentation and Control
Each manufacturer of sulfuric acid plants includes variations
in the instrumentation control systems as well as variations
in equipment and equipment arrangements. Although different
instruments and controls are included, certain process para-
meters are either controlled automatically or recorded to permit
manual control in all acid plants.
The most important instruments and controls are shown in the
process and instrument diagram of Figure 6. This diagram
shows a typical dual absorption acid plant as previously de-
scribed. The most important process parameters are those
affecting the temperatures and sulfur dioxide concentrations
in the converter, and the sulfur trioxide concentrations and
temperature of acid in the absorbers. All of these process
parameters directly affect the emissions of SO2 and acid mist
33
-------
J| <•>•'»> "I-j Vt'-. (..^ f ;.•{•, l-.a-i*-.
--.•I ««.>.-.t •..•••-. T
CX.llO — - I
(•a I
1CID S
TMN
rrr.
-------
from the unit.
The sulfuric acid production rate is controlled by regulating
the flow of molten sulfur and air to the sulfur furnace. These
flows are properly ratioed and controlled by the temperature
of the gas leaving the furnace. The SO- concentration in
this gas is higher than required for the converter feed, so
the gas is diluted by addition of dry air controlled by a
temperature controller on the feed to the converter. This re-
duces the SO2 concentration in the feed gas to the first stage
of the catalytic converter. The inlet temperature to the
converter must be precisely controlled for optimum conversion
of S02 to SO,. Additional temperature control j.s obtained
in the feed stream by regulating the amount of cold ceases
from the primary absorber flowing to the No. 1 heat exchanger.
Since the waste heat boiler must be operated to provide suf-
ficient steam for driving the turbine on the primary blower,
the balance in heat transfer between the No. 1 and No. 2
exchanger must be adjusted manually with by-pass valves to
properly set the range of heat transfer in the No. 1 heat
exchanger apd heat recovery boiler.
These by-pass valves and others around heat exchanger No. 2,
also are used to adjust the temperature distribution in the
various stages of the catalytic converter. For proper con-
version of SO- to SO.., inlet and outlet temperatures of each
stage must be carefully controlled within a narrow range.
Control and adjustment of the converter temperatures is
provided by a multi-point temperature recorder recording the
34
-------
inlet and outlet temperatures or. each catalyst section. Ob-
servation of the temperature profile across the converter
will permit quick analysis of the status of the conversion
efficiency of the unit. This multi-point temperature recorder
is the heart of the control system for the sulfuric acid plant.
Critical parameters around the primary absorber are the temp-
erature, flow, and concentration of the acid feed to the ab-
sorber. If these parameters are outside the acceptable range,
acid mist is generated in the absorber and is discharged from
the plant vent. Temperature of the absorbing acid is controlled
by adjusting the wnter flow on the inlet acid cooler while the
concentration of acid is adjusted by a concentration analyzer
controlling the amount of water make-up to the strong acid
exiting the absorber. The same process parameters are as im-
portant on the secondary absorber as on the primary unit, and
must be controlled as well.
Since the dual absorption plant was designed for the purpose
of reducing the SO- and acid mist emissions from the plant,
as well as increasing the efficiency of acid manufacturing,
the important measurement of the success of operation of the
plant is the SO analyzer installed in the vent gas stack.
This automatic analytical instrument continuously monitors
the SO concentration in the vent gas, indicating when one
£
of the process variables exceeds optimum range of operation.
When an excessive concentration of SO2 in the stack exists,
the instrumentation including the multi-point recorder on the
35
-------
converter, is used to diagnose the problem and correct the
emissions.
Many other instrument and control systems are required for
proper operation of a sulfuric acid plant. These include
instruments on such auxiliaries as the waste heat boiler,
economizer, concentrated acid sumps, boiler feedwater for
economizers and boiler, the air drying tower, and storage
units. Although these instruments are important to operation
of the plant, their effect on the emissions of the plant are
not as direct as those previously discussed. Most of the
data collected in our plant survey and presented in the next
section include the important process parameters discussed in
the first portion of this section and are analyzed in the
succeeding section.
36
-------
SECTION V
SURVEY OF ACID PLANT EMISSIONS AND CONTROLS
SELECTION OF PLANTS FOR FIELD SURVEYS
The primary, purpose of this study was to obtain information
showing the relationship between acid plant process variables
and emissions. The study was concerned with those plants
that could meet the air pollution control standards under
normal operation. Reporting only those plants that will meet
air pollution control standards limits the study primarily to
dual absorption acid plants and those single absorption plants
equipped with vent gas cleaning systems. Three vent gas
cleaning systems were selected as being properly proven in
commercial installations. These were the sodium scrubbing
system, the ammonia scrubbing system, and the Union Carbide
f
PuraSiv system.
The application of the dual absorption process is relatively
new in this country and only a limited number of these plants
are in operation. Vent gas cleaning systems are also rela-
tively new in commercial application, limiting the number of
plants available from which information could be obtained.
Because of the limited number of dual absorption plants avail-
able for study, comparative information from several single
absorption plants was included in the study to support informa-
tion obtained from dual absorption plants. Many process
variables can be correlated between the two plants since pro-
cess variables have similar effect on emissions in both.
37
-------
The newest gas cleaning system applied commercially is the
Union Carbide PuraSiv system. Only one plant is operating
commercially with sufficient experience to provide operating
data. This plant was not available for a visit because of the
proprietary nature of the process but data was obtained from
the performance test performed by EPA.^3^
After a review of the industry a list of tentative plants was
prepared (See Table 1) including all the known dual absorption
plants and examples of the three major gas cleaning processes.
A list of acid plant design companies was also compiled as a
possible alternate source of acid plant operating information.
Visits to several of the major plant design firms produced very
little useful data because they considered their plant designs
as proprietary to themselves and performance test data from
plants that they installed as proprietary to the clients.
After this initial effort attention was turned to obtaining
data from operating plants.
All of the candidate sulfuric acid plants were investigated
and five companies were selected for visits. The selection
was based upon the appropriateness of plant type and willing-
ness to provide the required information on the operation of
the plants. The EPA task officer approved the list of selected
plants to be visited.
During the field visits to the selected locations, contractor
personnel examined a total of 12 contact acid plants.
38
-------
TABLE 1
A LIST OF CANDIDATE COMPANIES FOR THE SULFURIC ACID PLANT MALFUNCTIONS EMISSIONS SURVEY
Single W/NH-> Scrubbing Single W/Na Scrubbing
vo
No. of
Plants Company
No. of
Plants Company
Dual Absorption on Line
No., of
Plants Company
New Dual Absorption Start-Up
No. of
Plants Company
(2)
(1)
(2)
(1)
(1)
CD
(2)
C. F. Industries (1) Olin, Inc.
Conserve, Inc. (1) Olin, Inc.
Olin, Inc. (1) Allied Chemical
Collier Carbon &
Chemical Company
Allied Chemical Co.
W. R. Grace
Texas Gulf
(1)
(1)
(1)
(1)
(1)
(1)
*
American Cyanamid
Miss. Chemical
N. L. Industries
Stauffer Chemical
Standard Oil of Cal.
American Smelting
and Refining Co.
(1)
(1)
(1)
(1)
(1)
(2)
(1)
(2)
(3)
Freeport Chem.
Agricultural Prod.
Texasgulf
N. L. Industries
Allied Chemical
C. F. Industries
Miss. Chemical
Agrico
IMC
-------
The location and type of plants visited are listed below:
No. of Plants Type of Contact Acid
Surveyed Plant Location
2 Single Absorption without North Carolina
Tail Gas Cleanup
4 Single Absorption with Florida
Ammonia Scrubber
3 Single Absorption with Maryland
Sodium Scrubber
3 Dual Absorption (1) Florida
(1) North Carolina
(1) Texas
12 - Total number of plants surveyed at the five (5)
locations
TYPE OF EMISSION CONTROL SYSTEMS
The acid plants visited are listed in Table 2 using
letter codes to protect the confidentiality of the plant data.
This table indicates the types of emission control systems
found at each plant. The information on plants "A" and "X"
was furnished from EPA records as these two plants were not
available for inspection. The data furnished on these two
plants are helpful in this study even though the plants could
not be visited.
The two (2) single absorption plants (M.,M2) located in North
Carolina have no tail gas cleaning systems. The single ab-
sorption plants (K, T, U,, U_) visited in Florida had ammonia
scrubbing systems for tail gas cleanup. Plant "U..", also uses
a mist eliminator on the downstream side of the ammonia scrubber.
40
-------
TABLE 2
SUMMARY J3F EMISSIONS FROM VARIOUS TYPES OF CONTACT SULFURIC ACID PLANTS SURVEYED DURING NORMAL OPERATIONS
Code Letters
Type of Sulfur
Type of Plant
H SO^, (100%) Pro-
auction Rate, tons/day
Conversion Efficiency
Percent
Stack-Gas Rate, scfm
x 10
SO- Entering Converter
Vol. 7.
S02 Removal Eff. %
SOy Emitted, ppm
by Vol.
Ib. per Ton ,.
Ib./scf x 10
SO. Emitted, mgSO./scf
Vol. % SOj J
ppm (vol. J SO
Mist H,SO, Emitted
1b./ton Acid
H.SO, Mist. Removal
Efficiency
mg/scf
Average SO-, mg/scf
Average S0_, mg/sef
Opacity 7.
Local Standard
SO. (new)
Q
Bright Sulfur .
Dual
1600-1650
99.7-99.9
96-106.6
8.6-8.9
99.7
101-134
1.17-2.04
1.60-2.48
0.00009
0.9 ppm
0.037-0.955
99.5+
0. 15-06-. 03
0.88
0.65
0
N. Carolina
27///T
Ml
Bright Sulfur
Single
1565
97.14
7.0-8.0
98.0
2500
33.5
0.56-6.15
.0004-. 0059
4-59
98.3
5-10
N. Carolina
270/T
M2
Dark Sulfur
Single
1565
95. 6%
7.0-8.0
98.0
1300-3300
17.4-44.2
1.0-30.55
0.17-0.30
1700-3000
0.03-0.06
99.9
20-80
N. Carolina
270/T
H
Dark Sulfur
Dual
1500-1800
99.75-99.85
85-105
9.0
99.7
40-300
1.0-4.0
1.0-3.0
97.7
0
Florida
100 /T
-------
TABLE 2 (continued)
SUMMAKi! "T EMISSIONS FROM VARIOUS TYPES OF CONTACT SULFUR1C ACID PLANTS SURVEYED DURING NORMAL CONDITONS
Code Letters
Type of Sulfur
Type of Plant
H.S04(iOO%) Produc-
tion Rate, tons/day
Conversion Efficiency
Percent
Stack.Gas Rate, scmf
x 10J
SO. Entering Converter
Vol. %
S02 Removal Eff. %
SO. Emitted, ppm by
Vol.
Ib. per ton s
Ib./acf x 10~°
SO.Emitted, ragSO»/scf
Vol. % S03
ppm (vol.; S0_
MiM: H2S04 Emitted
Ib./ton Acid
H,SO, Mist Removal
Efficiency %
mg/acf
Average SO-, ng/scf
Average S03 mg/scf
Opacity %
Local Standard
SO, (new)
i.
A
Bright Sulfur
Dual
1500-1510
99.00-99.9
9.3-9.4
99.7
170-400
10
Mississippi
500 ppm (n)
2000 ppm (e)
X
Bright & Dark
Single
1100-1150
97.6-98.0
74.3
8.0
97.9
1600
34.96
99. 5+
3.10
Florida
1C«
Y
Sludge Acid & Sulfur
Single with sodium
scrubbing
1000
98.0
8.0
90-92
30-200
.200-. 224
.068-. 087
99.5
Maryland
2000 ppm
J(1)
Copper Ore
Roaster S0_
Dual
500
92.7-99.9
100.0
2-19
99.7
0.03-3110.2
Texas
25#/T
(1)
DAP feed from smelter.
-------
TABLE 2 (Continued)
SUMMARY Of EMISSIONS FROM VARIOUS TYPES OF CONTACT SULFURT.C ACID PLANTS SURVEYED DURING NORMAL OPERATIONS
u*
Code Letters
Type of Sulfur
Type of Plant
H SO.(IOOZ) Produc-
tion* Rate, tons/day
Conversion Efficiency
Percent
5 Lack Gas Rate, scfm
x 10
SO, Entering Converter
Vol. Z
SO, Removal Eff. Z
SO, Emitted, ppm by
Vol.
Ibi/ per ton ,
Ib./scf x 10
SO. Emitted, mgSO./scf
Vol. * SO. J
ppm (vol.7 SO,
Mist H,SO, Emitted
Ib./ton Acid
H.SO, Hist Removal
Efficiency
mg/scf
Average S02» mg/scf
Average S03 , mg/scf
Opacity Z
Local Standard
SO, (new)
-------
The single absorption plants located in Maryland ("Y") use
sodium scrubbing systems. The process descriptions of the
ammonia scrubbing system and the sodium scrubbing system are
found in Section IV, "Process Description".
In the dual absorption plants (A,H,Q,J) the SO, emission con-
trol is inherent in the process itself. Control of acid mist
emissions is accomplished through mist elimination devices
incorporated in plant design. Section IV has further infor-
mation on the emission control system found in the dual absorp-
tion plants.
EMISSIONS DATA OBTAINED
The emissions and operating data obtained during this study
were supplied by most of the major sulfuric acid manufacturers
that are located in the United States. Emissions data was
received from more than 65 percent of the new dual absorp-
tion plants that were started up during 1974.
Table 2 shows operating data and emissions data from all plants
visited by the contractor plus 3at«: on two plants (A,X) not
visited by the contractor. Production rates are shown for
each plant and the SO2 emitted is shown in pounds of SO2 per
ton of 100 percent H-SO, produced. The acid mist emitted is
listed in terms of pounds of (100 percent) H2SO4 mist per
ten of 100 percent H2S04 produced. The data in Table 2 were
recorded during steady-state operations and serves as a basis
for comparison to data obtained during upsets.
44
-------
Table 3 lists data that were recorded at a dual absorption
acid plant during a new source performance standards test
conducted in 1974. This test was observed by EPA personnel
and all emissions were well below the EPA standard.
Figure 7 shows the relationship between the EPA standard emis-
sion limit for SO- expressed as "pound per ton of 100 percent
acid produced" and "parts per million by volume" in the stack
gas for various SO, concentration inlet to the converter. The
standard of "four pounds per ton" is marked. A similar rela-
tionship for acid mist is presented in Figure 8 comparing
"pound per ton" and "mg per scf". This basic format will be
used to present much of the data from operating plants col-
lected in the study.
The steady state data in Table 2 can be used for comparison
to information obtained during plant malfunctions. A variety
of malfunctions were observed and appropriate data concerning
these malfunctions are presented in Section VI - "Analysis
of Process Variables and Emissions". Malfunction data ob-
tained included many start-up conditions for single absorp-
tion acid plants equipped with either ammonia scrubbers or
sodium scrubbers, and for dual absorption acid plants. Data
were obtained on a dual absorption acid plant (plant Q) show-
ing the effect on S02 emissions of catalyst ageing, burning
bright and dark sulfur, loss of absorber acid concentration,
and loss of sulfur feedstock. Data from plant "Jn shows the
effect of inlet SO- variations on SO- emissions, and converter
45
-------
TABLE 3
EPA NEW SOURCE PERFORMANCE STANDARDS TEST DATA
DUAL ABSORPTION CONTACT ACID PLANT
(PLANT "Q")
Traverse ^(Acid Mist) S02
scfm Ib/cf Ib/Ton Acid Ib/cf lb/ton Acid
Sample 01 106,593 5.59 x 10~7 0.055 1.66 x 10~5 1.64
Sample 92 99,472 4.25 x 10~7 0.037 1.61 x 10"5 1.42
Sample §3 96,262 5.60 x 10~7 0.046 2.48 x 10~5 2.04
46
-------
9'?. 92
Sulfur Conversion - Percent of Feedstock Sulfur
99,7 99.0 98.0 96.0 92.9
10,000
LUl
EPA SULFUR DIOXIDE EMISSION STANDARD ,
FOR NEW CONTACT ACID PLANTS !
100
1.0 2.0 3.0 5.0 10.0 15 20 30 50 100 200 300 500 1000
S02 Emissions - Lb Per Ton of 100% t^SO^ Produced
Figure 7
-------
EPA ACID MIST EMISSION STANDARD FOR NEW CONTACT ACID PLANTS
oo
u.
u
WJ
9)
ft.
8*
g
o
u
10
2
•o
SULFURIC ACID PLANT VOLUMETRIC AND
j MASS EMISSIONS OF ACID MIST AT VARIOUS'
4 INLET S02 CONCENTRATIONS BY VOLUME
...I i I j_4_u.. i i-4_>_ .
MB UI
• •uliii III,!. U*iu£nbaU^>i i
.01 .02 .03 .05 .10 .15.20 .30 .50 1.0 2 3
Acid Mist Emissions - Lb H2SO^ Per Ton of 100% HgSO^ Produced
Figure 8
-------
malfunctions, on a dual absorption plant. Data were obtained
from a single absorption plant equipped with an ammonia scrub-
ber (plants K and U) showing the effecc on SO- emissions of
pH control, specific gravity variations, effluent concentra-
tion variation from the ammonia scrubber, and deterioration
of ths scrubber packing. A single absorption acid plant with
a sodium scrubber (plant Y) was studied during a loss of pH
control in the sodium scrubber, specific gravity variations,
and evaporator plugging in the reclaiming system. The effect
of these malfunctions on SO- emissions is illustrated by the
data.
An extensive study was made of the effect of downtime duration
on SO- emissions during subsequent start-up. These data will
be further analyzed in the next section to establish a criterion
for converter reheat requirements with various shutdown periods.
49
-------
SECTION VI
ANALYSIS OF PROCESS VARIABLES AND EMISSIONS
NORMAL PLANT OPERATIONS
The primary emphasis of this study is the relationship be-
tween process variables and emissions from contact sulfuric
acid plants operating in upset conditions. The upset condi-
tions have been classified as start-ups, shutdown, and mal-
function or misoperation. In order to understand completely
the nature of plant upset conditions and how the process
variables effect emissions, it is necessary to review data
obtained from plants in normal operation. The normal operating
data were used to establish a basis for comparison with para-
meters during upset conditions.
Converter Temperature Control
The most important parameter for controlling sulfuric acid
converter efficiency are temperatures in the catalyst bed.
In normal operation the temperature of each bed is controlled
by adjusting the heat transfer in the heat exchangers between
converter stages. In normal operation a range of temperatures
of one to three percent is permitted for the inlet temperature
to each bed. Deviation from this range of temperatures will
reduce the conversion efficiency of the catalyst. When temp-
eratures are controlled within this range a specific converter
efficiency is expected for each stage. The summation of the
efficiencies of each stage represents the overall conversion
for the plant. Typical converter temperature setpoints and
50
-------
TABLR 4A
NORMAL STEADY STATB OPERATING TEMPERATURES AHD CONVERSIONS IN EACH COHVEEtTER STAGE OF CONTAft ACID PLAHTS
Standard Single Absorption Operating Conditions
(Plants U, and U.)
Teoperature Conver- Range of Temp.
location
x. so: °c
Gas Entering 410.0
First Pass
Gas Leaving 601.8
First Psss
Rise in leap. 191.8
Gas Entering 438.0
Second Pass
Gas Leaving 485. 3
Second Pass
Rise In Temp. 47.3
Cuculatlve Conversion
Gas Entering 432
Third Pass
Oas Leaving 443
Third Pass
Rise In Temp. 11
emulative Conversion
Caa Entering 427.0
Fourth Pass '
Gas Leaving 430.3
Fourth Pass
Rise in Temp. 3.3
emulative Conversion
s ion
°F X
770
iliS
345 74.0
820
906
86 18.4
92.41
810
830
20 4.3
96.71
800
806
6 1.3
98.0X
oc
415-52*
600
175-185
440-450
500
50-60
435-440
450
10-15
425-430
430-435
5
°F
779-797
1112
315-333
824-852
90-108
815-824
§42
18-27
797-806
806-815
9
Single Absorption
With Air Quenching
°C °F Conv.
425
612
187
438
500
62
439
448
9
423
427
4
797
113*
337 72.0
620
y>32
112 21.0
93T5X
822
838
16 3.8
96.8X
793
801
1 A.
EPA-NSPS Performance Test on Dual Absorption (Plant Q)
Temperature Ranges
°F °F X Dual Absorption Test Runs<*>
Set Temp. Conver-
Polnts Range alon 1234 5
820° 770-830
1100- 1100
320-270
800° 000-830
900-950
100-150
800° 800-830
830-860
30-60
790° 800-830
830-850
30-20
795
1098
75 307
828
1001
IS 173
90X
829
884
6-7 55
832
863
3.7-2.9 31
99. 7-99. 9X
797
1104
307
830
1001
179
832
886
54
834
866
32
800
1113
318
843
994
151
822
865
43
820
S4?
27
803
1113
310
845
988
143
816
859
43
815
«42
27
806
1118
312
845
992
1A7
821
863
42
821
•46
25
*The high value Is based on a Reich Test with an expected error of +11, therefore the
value for actual plant operation should be 98.5X.
-------
ranges are given in Table 4A. This table also presents the
expected conversion for each stage. Three major types of
plants are included: single absorption, single absorption
with air quenching, and dual absorption plants.
As shown in Table 4A an overall conversion efficiency of 98
percent is anticipated from a single absorption plant. The
addition of air quench to a single absorption plant will in-
crease the conversion efficiency to 99.5 percent. A dual
absorption plant can obtain between 99.7 and 99.9 percent
overall conversion efficiency. A series of five test runs
on the dual absorption plant described in Table 4A shows the
effect of catalyst bed inlet temperatures on the temperature
rise across each bed and upon the conversion efficiency for
each stage. This series of runs illustrates the need for
close temperature control on catalyst bed inlet temperatures.
The data presented in this table were obtained from previous
EPA test.*4*
A study was made to determine a quantitative relationship be-
tween converter temperatures and SO emissions in a dual ab-
sorption acid plant. A multiple linear regression analysis
program was run using temperature and emission data from
plant "J" during upset conditions to provide a wider range of
temperature variations. An analysis was also performed on
data taken from normal plant operations. The relationships
developed in the analysis will be presented in a later section.
52
-------
Several models were used in the regression analysis, with SO^
as the dependent variable and converter temperatures as in-
depent variables. Complete computer results of this analysis
is contained in Appendix C. The analysis of normal operating
conditions was run on 30 dat^ points selected at random from
the plant "J" operating data. Six different regression models
were used in an effort to establish correlation between SO2
and temperatures. None of the correlations was significant.
It is believed that the lack of significant correlation
resulted from: (1) interdependence" between temperatures, and
(2) process parameters affecting S(>2 emissions that were not
considered in the regression study. These variables include
inlet SOp concentrations, sulfur furnace temperatures, and
catalyst ageing. Because plant "J" is fed from a copper
smelter and therefore has a highly variable imput concentra-
tion and flow rate, an additional regression analysis was per-
formed on 31 random data points selected from normal operations
from plant "Q"- It was thought the more stable conditions of
inlet concentration of SO2 and other process variables might
improve the correlation from the regression analysis. The
correlations from this analysis were also included. After
performing these two analysis it was concluded that a regres-
sion model of SC>2 emissions in terms of converter temperatures
cannot be developed without additional data and study.
A base case was established for operation of double absorption
acid plants by running standard statistical analysis on 64
53
-------
random data points selected from normal operation of plant
11J". This study was performed in an effort to confirm the
information presented in Table 4A derived from a variety of
sources of data. The results of the study on plant "J" in-
dicates that if the eight converter temperatures are controlled
within a specified range, the SO- emissions will be between
zero and 307 ppm. The results of this study were highly
significant and compared favorably to data presented in Table
4A. The results of this study are presented in Table 4B.
To further confirm the converter temperature ranges for nor-
mal SO2 emissions 13 data points were selected from the data
for plant "J" during times of high SO- emissions. Standard
statistical analysis was performed on these data points and
the mean temperatures calculated. Table 4C compares the mean
temperatures from the normal run analysis and those from the
off-normal analysis. The converter temperatures during up-
sets were consistently below the mean operating temperature
for normal operations and generally fell outside of the
acceptable range.
The statistical analysis comparing converter temperatures and
S02 emissions confirms the data presented in Table 4A and
clearly establishes the important relationships between these
parameters. Some variation in acceptable temperature range
will occur between plants because of variation in catalyst
activity and types of catalysts. These data should be a good
indication, however, of the acceptable ranges for modern dual
54
-------
TABLE 4B
CONVERTER TEMPERATURE AND S02 OUTLET CONCENTRATION
PLANT "J" - UPSET CONDITIONS
Normal Run
Ul
Variable
Y (ppm) SO- Exhaust Stack
Xj^ (°F) Inlet - 1st Converter Bed
X, (°F) Outlet - 1st Converter Bed
X3 (°F) Inlet - 2nd Converter Bed
X^ (°F) Outlet - 2nd Converter Bed
X, (°F)Inlet - 3rd Converter Bed
X, (°F) Outlet - 3rd Converter Bed
X7 (°F) Inlet - 4th Converter Bed
Xg (°F) Outlet - 4th Converter Bed
Start-Up
Ov/n
127
Emissions (ppm)
1678
Mean Temperature ( F)
824.0
1040.9
837.0
892.0
786.0
791.0
762.0
762.0
315.0
1006.5
811.5
841.0
732.5
729.0
702.0
707.0
-------
TABLE AC
CONVERTER TEMPERATURES AND S02 OUTLET CONCENTRATION
PLANT "J" NORMAL OPERATION
in
Variable
(Y-l) ppc «0 Exhaust Stack
(XI) Temp ( F) Inlet -
(X2) Temp (°F) Outlet
(X3) Temp (°F) Inlet -
(XA) Temp (°F) Outlet
(X5) Temp (°F) Inlet -
(X6) Temp (°F) Outlet
(X7) Temp (°F) Inlet -
(X8) Temp (°F) Outlet
1st Converter Bed
- 1st Converter Bed
2nd Converter Bed
- 2nd Converter Bed
3rd Converter Bed
- 3rd Converter Bed
Ath Converter Bed
- Ath Converter Bed
127
Temperatures I
82A
10A1
837
892
786
791
762
762
0-307
810-838
9A8-1133
891-855
811-973
73A-838
737-8A5
70A-820
702-822
-------
NORMAL RANGE OF OPERATION OF SINGLE ABSORPTION AND
DUAL ABSORPTION ACID PLANTS AS FOUND IN THIS STUDY
10.000
I
3
I . L J
Single Absorption Acid
Plant Operating Range
New Dual Absorption
Operating Range
I I I I I I I
.01 .02 .01.04.OS 1.0 2.0 J.Ct.OS.O 10.0 20 30 *0 V> 100
r*U«lon. Lb P«t Ton of 100X HjSO. Produced
Figure 9
57
-------
TABLE 5
PLANT "X" TEST DA1A
Conversion Tons Average
R'->n. wt. Z S02 Wt. % S07 Efficiency Per Day Opacity
No. Inlet -
-------
absorption plants equipped with the newer types of catalyst.
Single Absorption Plants
The basic sulfuric acid producing plant for many years has
been the single absorption type. This type plant has a long
history of operation in contrast to the dual absorption plant
and consequently a large amount of data are available. Even
though the single absorption plant will not meet EPA standards
(see Figure 9) it is importaat as a basis for studying the
application of tailgas cleaning systems to single absorption
plants and for supplementing data obtained from dual absorp-
tion plants.
Plant X shown in Table 2 of Section VI was selected as a typ-
ical single absorption plant to illustrate the SO, emissions
and converter efficiency for this type of plant. The S02 con-
version data, daily production rate, and opacity of the vent
gas for plant X is shown in Table 5. This 1,000 ton per day
acid plant was operating far in excess of allowable SO- emis-
sion standards and at times in violation of opacity standards.
The SO, emissions from plant X as shown in Table 5 fall within
the range of SO- emissions for all single absorption plants
studied. The SO- emissions ranged from 30 to 60 pounds per
ton of 100 percent sulfuric acid produced.
Acid mist emissions originate from three separate sources.
Because of the different form of the acid mist from these
three sources, special provisions must be made for measuring
59
-------
the quantity of each type of mist emitted. Table 6 illustrates
the amount of acid nist occurring in each of these types as
measured by three separate analytical procedures. In the
series of five tests shown, the EPA standard measurement
method measures only the entrained acid carried over from
the absorber. The results from this method are lower than
those obtained using the special Monsanto method that measures,
in addition to absorber carry-over the fine particulate acid
mist formed in the absorber and SO, passing through the ab-
sorber. A comparison of the standard and special methods
will show the standard method vill always yield a result that
is lower than the special method. In this particular single
absorption acid plant located in the Northeastern United States,
the acid mist standards were not violated, even when applying
the special analysis method. These data were furnished from
an EPA performance test.
Though the single absorption acid plant shown in Table 6 could
meet the acid mist standard, data from plant "U" covered in
this study indicate that this plant cannot meet the acid mist
standard. Data plotted in Figure 10 shows monthly averages
of acid mist emissions over a period of six years. None of
the points plotted fall within the performance standard. This
data illustrates the need for special equipment to control
acid mist emissions regardless of the type of plant considered.
As a contrast to data shown in Figure 9, Table 7 shows a sum-
60
-------
TABLE 6
Test
Stack Gas Ib H.SO,
Flow Rate per scf
scfm x
ACID MIST TEST
(DATA FURNISHED BY EPA)
Production
mg H.SO, Tons/Day
per scf (H2S04)
(5)
Emission
Ibs H2S04/Day
Emission
Ibs H2S04/ Ton of Acid
Standard Special
Method
Kathod
1A 20,737
IB
1C
2A 20,354
2B
2C
3A 20,958
3B
3C
4A 20,506
4B
4C
5A 20,549
SB
5C
5.37
2.29
4.60
0.90
1.18
4.40
0.78
0.94
4.45
0.82
0.97
4.57
0.94
0.90
0.224 185
•0.096
0.208 175
0.041
0.053
0.200 175
0.036
0.043
0.202 193
0.037
0.044
0.207 193
0.043
0.041
16.04
6.84
22.88
13.48
2.64
3.46
19.58
13.28
2.35
2.84
18.47
13.14
2.42
2.86
18.42
13.52
2.78
2.66
18.96
0.087
0.12
0.077
0.11
0.076
0.11
0.068
0.10
0.070
0.10
-------
Monthly Variations in Acid Mist Emission Over tb Past Six Years
(Plant Uj, Bright Sulfur Burned In a Single Absorption contact Acid Plant)
to
U.
u
v>
01
cu
f
CM
s
o
•H
U
2
CO
•H
SULFURIC ACID PLANT VOLUMETRIC AND
MASS EMISSIONS OF ACID MIST AT VARIOUS
INLET S02 CONCENTRATIONS BY VOLUME
.01 .02 .03 .05 ' .10 .15 .20 .30 .50 1.0 2 3
Acid Mist r-miesions - Lb H2SO^ Per Ton of 1002 HjSO^ Produced
Figure 10
-------
TABLE 7
ACID MST OBSERVATIONS (6)
01
bl
Location
Faulsboro, H. J.
Linden, N. J.
Faulsboro. R. J.
Elizabeth, H. J.
Hevarv, N. J.
Linden, N. J.
Clbbstovn, N. J.
Deepvatnr, H. J.
••veil. Pa.
Cooganj
Olln Corp.
American
Cyanaaid
Corporation
Olln Corp.
Allied Chcalcal
Corporation
Essex Chealcal
DuPont, Grasaelll
Work*
DuPont, Rapauno
Works
DuPont, Chambers
Worka
Allied Chealeal
Corporation
Feedstock
Spent
Acid. S
Dark S
Spent
Acid, S
Spent
Acl
-------
mary of acid plants tested by EPA that will generally meet
the acid mist performance standards. All these plants were
burning spent acid or dark sulfur, contributing to an increased
acid mist evolution. A review of the acid mist emissions and
opacity for these plants shows that the application of a packed
fiber filter (Brink) will eliminate all acid mist carry-over
contributing to opacity. In two cases, plants equipped with
filters packed with pebbles or Intalox saddles had a vent gas
stream with opacity of 30 to 40 percent. This type of packed
bed has a low removal efficiency for the very small particle
size of the acid mist. The Newell, Pennsylvania plant was
equipped with a Brink filter and had an opacity reading of
10 to 20 percent. The high opacity reading in the vent gas
probably indicated the Brink filter was leaking. An important
conclusion that can be drawn from the data of Table 7 is that
a single absorption acid plant equipped with a vent gas clean-
ing system must also be equipped with an efficient filter for
eliminating acid mist and high opacity in the vent gas.
Single Absorption Plants with Gas Cleaning Systems
The need for acid mist filters on a single absorption plant
equipped with a vent gas cleaning system was indicated in
the previous paragraph and in Table 7. It has also been pre-
viously indicated that the single absorption plant cannot
reach the SO, emissions standards without provisions for re-
moving SO- from the vent gas.
When a single absorption plant, however, is equipped with
--•' 64
-------
proper acid mist filters and an efficient vent gas cleaning
system, both the S02 and acid mist performance standards can
be met. Data obtained in our study from plant "Y" is shown
in Table 8. Plant "Y" is a single absorption acid plant
equipped with a sodium scrubber system and acid mist entrain-
ment filters. As indicated in Table 8 this plant in normal
operation can reach an SO- emission level of 2.59 pounds per
ton and an acid mist emission level of .043 pounds per ton.
In a series of three tests, Test 1 was discarded because of
possible sampling equipment failure. Test 3 was run at a
higher gas flow rate than Test 2 and produced acid mist emis-
sions equal to the performance standard of .05 pounds per ton.
Although the tendency toward increased acid mist emissions at
higher gas rates is expected, the cause and effect relation-
ship of this data cannot be substantiated.
Dual Absorption Plants
The primary difference between the single absorption acid
plant and the dual absorption plant is the installation of a
primary absorber following the third catalyst bed. A primary
absorber in this location reduces the SO3 concentration of
the gas entering the fourth catalyst stage. Removing SO3
from this gas permits a higher driving force in the fourth
converter stage to shift the conversion equilibrium in the
direction of more SO3 formation. The effect of the primary
absorber on overall conversion efficiency is illustrated in
Table 9.
65
-------
TABLE 8
EPA PERFORMANCE TEST RESULTS FOR S02 AND ACID
MIST EMISSION CONTROL EFFECTIVENESS FROM SODIUM SCRUBBER
(PLANT "Y")
SO, + Acid Mist
Test No.
2
3
Gas Flow
(SCFM)
35,141
36,292
Production
Rate
(Tons/Day)
730
730
SO, (Outlet)
PPMZ Lb/Ton
250
227
2.59
2.85
(ppmj
2.26
7.79
J (Outlet)
H SO, H SOA
(lb7SCF) (L5/Toi
0.279
0.965
0.043
0.162
-------
TABLE 9
EFFECT OF PRIMARY ABSORBER IN THE DUAL ABSORPTION PROCESS
(PLANT "Q")
(Basis 100 raola Feed Gas at 9.0% S0£ and 12%
Converter Feed Cos Vol. % SO, 9.00 12.00 0 79.0 100.00
97% Conversion -8.73 -4.365 +8.73 0 -4.365
To First Absorber Vol. % SO, 0.27 7.635 8.73 79.0 95.635 2700
SO, Absorbed 0 0 -8.73 0 -8.730
toJFourth Catalyst Vol. % SO, 0.270 7.635 0 79.0 86.905 2700
97% Conversion -0.262 -0.131 +0.262 0 -0.131
To Final Absorber V.,i. % SO, O.OOR 7.504 0.262 79.0 86.774 81
SO, Absorbed 0 0 -0.262 0 -0.262
ToHSxit Stack 0.008 7.504 0 79.0 86.512 81
Volume, % S02 .0081
Volume, ppm SO- 8_1
Overall Conversion Efficiency 99.91%
-------
As indicated in Table 9, the SO2 to SO., conversion in the
first three sections of the catalyst converter reaches a total
of 97 percent. If this gas were sent directly to the fourth
catalyst bed an additional conversion of 1.3 percent could be
expected as indicated in the review of the single absorption
plant presented in Table 4A of Section VI. This will give an
overall conversion efficiency for a single absorption plant
of 98.3 percent.
The addition of the primary absorber after the third catalyst
stage, however, will reduce the SO, concentration in the gas
entering the fourth catalyst bed to essentially zero. The
remaining SO- can now be converted in the fourth catalyst
stage '_o an additional extent of 3.8 percent (based on origi-
nal SO- present) providing an overall conversion efficiency
99.91 percent. This high overall conversion efficiency reduces
the SO2 concentration in the vent gas to less than 100 ppm.
The result of this increased SO- conversion efficiency can be
seen in test data obtained from plant "Q"- These data presented
in Table 10 show results of three performance test runs mea-
suring S02 and acid mist emissions. During these tests the
plant was operating somewhat higher than design capacity, as
indicated by the emission results based upon actual production
rates and design production rates. Comparison to the perfor-
mance standards indicates the plant was operating well below
the standards at all times.
-------
TABLE 10
SUMMARY OF TEST RESULTS FROM DUAL ABSORPTION
PLANT "Q1
S0? Emissiors, Ib/ton
Based on actual production
Based on design rate
Based on actual production
adjusted for higher in-
stack monitor readings
Test No. 1 Test No. 2 Test No. 3 NSPS
1.66
1.69
2.17
1.43
1.52
1.86
2.06
2.28
2.85
4.0
4.0
4.0
Acid Mist Emissions. Ib/ton
Based on actual production 0.056
Based on design rate 0.057
0.037
0.039
C.046
0.051
0.15
0.15
69
-------
Plant "Q" was typical of all dual absorption plants in normal
operation. The range oi" SO2 emissions from all plants studied
was presented in Figure 9 (Page 57) in the preceding section
on single absorption plants. This figure also shows the range
of emissions from dual absorption plants. Plant "Q" falls
approximately mid-way in this range. This figure also illu-
strates that essentieliy all the data obtained from dual absorp-
tion plants for SO- emissions fall below the performance stand-
ard limit.
Figure U further illustrates the SO- emissions from single
and double absorption plants operating under normal conditions.
In addition to showing the minimum, maximum, and average emis-
sions for the single absorption and double absorption plants,
Figure 11 also indicates the range of emissions experienced
during initial operation of new plants. During the initial
operation of a new plant in good operating order, the SO- emis-
sions will fall between 1.5 and two pounds per ton of IOC per-
cent sulfuric acid produced.
Another important fact illustrated Lv Figure 11 is the effect
of higher concentration cf S02 entering the converter of the
double absorption plant (oight to ten percent). For a single
absorption plant the maximum S07 concentration in the feed
gas to the converter was eight percent. High SOj feed concen-
trations and higher conversion efficiencies both promote in-
creased production from the dual absorption acid plant and
more efficient and economical overall operation.
5 70
-------
Sulfur Conversion - Percent of Feedstock Sulfur
99.92
10,000
96.0 92.9
2,000
o 1,500
NORMAL OPERATING RANGE FOR
DUAL ABSORPTION AND SINGLE
ABSORPTION ACID PLANTS
-------
SHUTDOWN OPERATION
The operation of shutting down a contact sulfuric acid plant
does not necessarily produce SO2 or acid mist emissions in ex-
cess of performance standards. If the proper shutdown procedure
is followed, most emissions associated with the shutdown ori-
ginate from a malfunction or upset that is also the cause of
shutdown. The most important aspect of the shutdown procedure
is preparation of the plant for subsequent start-up to minimize
start-up emissions. In this section we disruss shutdown
operations as they affect emissions on the start-up of a dual
absorption acid plant. Our primary attention is applied
to the length of the shutdown period compared to emissions on
restarting the plant.
The data obtained from most of the plants studied did not con-
tain a sufficient number of shutdov.*n operations to provide a
basis for statistical analysis of shutdown duration compared
to start-up emissions. Data from plant "J", however, was
collected over a long period of time, including many shutdown
and start-up operations. The monitoring program for plant "J"
fraa which the data was extracted covered a period of 5,038
hours or 212 days. During this period of time the acid plant
was in operation for a total of 190 days or 90 percent of the
time period. After accounting for periods of inoperation of
the plant and emission monitoring instrumentation, data were
collected during 86 percent of the period of the monitoring
program. The monitoring instrumentation recorded one reading
72
-------
for each process parameter every three minutes. At the end
of each 15 minute interval the average of the previous five
readings was computed. These fifteen minute averages were
used as the basic data points for analyzing the performance
of the plant. During this period of operation, data for 23
start-ups were obtained after plant shutdowns ranging from
30 minutes to five days.
To simplify the analysis for the plant shutdowns, shutdown
periods were divided according to duration of the shutdown
in categories of less than one hour, one to two hours, two
to six hours, six to ten hour, 10 to 15 hours and over 15
hours. The sulfur dioxide emissions from the plant during
start-up, from the beginning of the start-up operation through
the sixth hour after start-up operations began, were plotted
against time using a linear regression analysis program on
a computer. The sulfur dioxide emissions were used as the
dependent variable and operation time as the independent
variable. Regression curves were plotted by the computer
averaging all the data points in each category on each hour
of operation during the start-up period. A summary of these
regression plots is presented in Table 11.
A review of Table 11 shows that data from two to six start-
up operations were included in each of the six categories.
In the "less than one hour", category where only two start-ups
were observed, the confidence level of the results is quite
low. For those categories containing larger numbers of
73
-------
TAfiLZ 11
THE EFFECT OF DOWNTIME SHUTDOWN DURATIONS UPON
EFFLUENT S00
CONCENTRATIONS (VOL. ppm) DURING START-UP
(FOR PLANT "J")
Shutdown Duration Peak S0_; ppm Hours to reach Number of
Hours During Startup 300 ppm SO,, Startup Cases
<1
1-2
2-6
6-10
10 - 15
15 !•
185
520
1,920
1,600
2,250
2,970
— •• 2
5
4.4 3
3.50 3
4.80 4
5.70 6
Reading at 4th
Hour (SO^-ppm)
64
123
350
275
400
749
-------
start-ups a greater confidsnce level can be assigned. Con-
sidered as a total analysis the results have high significance.
Comparison of the "hours to reach 300 ppm" for each category
indicates the operation time to compliance with the standards
increases with increased duration of shutdown. This time
curve flattens, however, at approximately six hours, indicating
that with normal start-up operations all plants should be able
to come within compliance within six hours. It is also apparent
that the longer the shutdown period, the higher the peak SO-
emissions during the start-up operation. Thi.s analysis shows
that most of the start-up operations approached 300 ppm at
the fourth hour, indicating that most plants can be started
up within four hours with SO- concentration reduced to the
performance standards regardless of the length of the shut-
down duration.
The start-up analysis provided here must be evaluated in terms
of the type of plant providing the operational data. Plant
"J" is a sulfuric acid plant using the dual absorption design
with SO, feed from a copper smelter. Because of the nature
of the SO2 feed, concentration of S02 entering the converter
varies from 1.6 to 19 percent. Also, the start-up of the acid
plant is dependent upon start-up problems within the smelter,
giving a much wider variation in start-up conditions than would
be expected from the same type of plant operating with elemental
sulfur feed. The peak SO- concentration reached during start-
up is probably the parameter most affected by the variation
in start-up conditions of the smelter.
75
-------
DUAL ABSORPTION AVERAGE EMISSIONS AFTER START-UP
n KT "j-
2.
LIX.CKD
IKIKATIOX OK ..•MltDiK.'M LESS TIIAS
OKI: iKiri:.
DUKvncs or Mii'iiviu: srTW.ni
ONF AMI) TWO HOUkS.
!)UKAT10X OK SIIUIDOVS SEIVEtN
T»l AND SIX II.IL'R>.
tll'RATIHN OF SHl'tlO'.1:) RC.TUEJ3I
SIX ANP. TtN HOtKS.
UUKATIOX OF SlftTlKIWM U^VKEN
IP AND 15 IKVKS.
PURA.-10H OK SliUTIXlVX C!(EATER
THAN IS IlilUIIS.
TOTAL
KUK-.FK
or CASK
*
6
23
2 IS 4 5
READING HOURS AFTL:P START-UP
Figure 12
76
-------
-4
-J
Fig. No.
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
TABLE 12
SUMMARY OF START-UP OPERATIONAL DATA ON DUAL (DAP) AND SINGLE (SAP)
ABSORPTION ACID PLANTS FOR FIGURES 25 THROUGH 41
Peak
so2
Emissions
(Vol. ppm)
Type
of
Plant
DAP
DAP
DAP
DAP
DAP
DAP
DAP
SAP
Downtime
Duration
(Hours)
12.5
J1.5
6.5
6.0
16.5
9.0
5.5
12.0
Time
Required
To Reach
Normal
Operating
Level
(Hours)
6.5
2.5
5.5
7.0
5.0
6.0
3.0
—
Time
SO.. Emissions
Exceeded
EPA Standard
of 300 ppm
(Hours)
3.0
2.0
4.5
5.0
4.0
2.0
2.0
—
DAP
DAP
DAP
DAP
DAP
DAP
DAP
DAP
DAP
37.0
16.0
13.5
16.5
10.5
9.0
Incomplete
Data
1.0
160.0
7.0
15.0
3.5
7.0
6.0
5.0
2.0
1.5
5.2
5.5
11.0
2.5
6.
4.
.5
,5
Comments
3.0
Incomplete
Data
1.5
4.0
1800
1500
1600
1200
1500
700
2000
4000
2900
2800
2400
3100
3100
1500
950
2200
3000
Scrubber Down for 28
hours - Plant not in
Compliance at Anytime
-------
To provide a more realistic presentation of the start-up data
for the 23 cases presented in Table 11, arithmetic average-j of
SO- concentration tor each hour of start-up within each cate-
gory of downtime were plotted on Figure 12. The use of arith-
metic averages and time plots show the cyclic nature of most
plant start-up operations and the increase in maximum SO- con-
centration with increased shutdown duration period. This
plot also shows that most shutdown and start-up cycles will
provide SO- emissions below EPA standards within six hours.
With £fctt ANO^tlon of ver,y long shutdowns,, exceeding ten houafc,
the plant can be brought to compliance within about four hours.
The four hour shutdown-period-breakpoint for heat conservation
and restart without reheating the plant is also illustrated
by this curve. The EPA standard of 360 ppm for a 10 percent
inlet SO- concentration is shown. This is typical for dual
absorption acid plants but may be high for a plant fed from
a copper smelter.
Concentrations of SO- during start-up were obtained for most
of the dual absorption plants studied, providing data from a
variety of plants in a variety of shutdown and start-up
conditions. SO- concentration was plotted against time at
the beginning of the start-up operation for each of these data
sets. This information is presented of Figures 25 through 41
included in Appendix B. These figures are summarized in Table
12. The data, taken from plants burning elemental sulfur and
using th« dual absorption process, generally confirm the in-
formation obtained from the regression analysis on plant "J"-
78
-------
A general summation of all of the plant start-up data indi-
cates most plants can reach compliance within four hours
regardless of the shutdown period, and essentially all plants
can be below 300 ppm after six hours of start-up operations.
Also verified is the four-hour maximum shutdown duration be-
fore reheat of the converter is required to insure SO- emis-
sions of less than 300 ppm during the start-up operations.
These data and conclusions apply to typical sulfuric acid
plants in normal operation with no special precautions for
minimizing SO- emissions (such as maintaining low production
rates during initial operation periods or using heat and hot
concentrated acid from associated plants to assist in the
start-up operation). This data also include the occurrence
of normal upsets and equipment failures during the start-up
operation resulting in higher SO2 emissions during some opera-
tions.
START-UP OPERATIONS
Single Absorption Plants
During the course of this study data were not collected on
single absorption contact acid plants unless they were equipped
with a vent gas cleaning system permitting them to meet emis-
sions standards. Understanding of the start-up procedure on a
single absorption plant, however, is important to the evalua-
tion of the operation of the vent gas cleaning systems. Start-
up data from two identical single absorption acid plants (Plants
'Jj and U2) 'is shown in Tables 13 and 14, and 15 and 16 respec-
tively in the next section (page 81).
79
-------
A review of converter temperatures tabulated for these plants
indicates that all temperatures were controlled acceptably
during the entire start-up operations. One variation in temp-
eratures, however, occurred that might be considered a cause
of high S02 emissions. As shown in Tables 15 and 16, the
peak emissions of SO2 from plant U2 occurred at a point where
the inlet temperature of the catalyst mass B was also at a
peak. The outlet temperature for mass B in the preceding
f
reading was also hic-h indicating thit. temperature was some-
what out of control. Since insufficient data were available
from these plants 1-.o run a statistical analysis, conclusive
relationships between catalyst bed temperatures and SC>2 emis-
sions cannot be established. A general evaluation of this
plant operation, however, indicates that the peak SO2 occurred
approximately two hours after start-up, which is normal for
a single absorption plant of this type.
A review of furnace temperatures for plant U^ presented in
Table 13 indicates full sulfur feed to the furnace was not
established until approximately five hours into the start-up
operation. A reduction of SO2 entering the converter contri-
buted to the relatively low concentration of SO2 emissions
fron this plant during this start-up. If full sulfur feed
had been applied to the furnace at the beginning of the start-
up operation, SO2 concentration exit the absorber tower would
have reached approximately 4,000 ppm rather than the 3,000
ppm experienced.
80
-------
TABLE 13
START-DP TEMPERATURES FOR PLANT "l^"
SINGLE ABSORPTION ACID PLANT WITH AMMONIA SCRUBBER
1
Tin
2:43
3:00
3:15
3:30
3:45
4:00
4:15
4:30
4:45
8:00
5; 15
5s 30
5:45
6:00
6:15
6:30
6:45
7iOO
7:30
7:45
8:00
8:15
8:30
8i45
9:15
Furnace
IMP SF
1710
1710
1712
1704
1710
1710
1728
1732
1738
1748
1744
1738
1731
1721
1728
1728
1750
1800
1805
1807
1806
1900
Boiler
Exit
602
602
60S
606
607
607
609
610
610
611
614
612
609
607
607
610
610
610
625
627
6^9
629
630
A - Maaa
Ent* -iicr
836
834
8i4
832
834
830
829
832
6:2
831
832
828
824
824
820
820
824
822
822
830
841
836
831
829
A - Haas
Exit
1093
1092
1091
NT2
1092
1091
1090
1092
1093
1095
1096
1094
1C90
1089
1086
1084
1082
1083
1081
1091
1094
1100
1100
1096
B - Mass
Entrance
815
819
820
821
823
825
824
826
826
826
627
828
826
826
877
827
831
829
826
826
839
839
832
837
B - hacs
Exit
869
899
911
920
926
931
931
932
934
936
936
937
939
939
939
939
939
939
936
939
945
949
949
952
C - Maaa
Entrance
824
832
836
834
828
828
827
829
830
833
834
835
834
834
834
834
834
834
834
835
840
'840
845
849
C - Haea
Exit
862
865
870
874
877
873
•71
870
874
677
879
879
879
879
879
879
879
879
877
882
892
895
896
901
D - M«38
Entrance
804
823
811
602
829
841
818
801
793
797
799
800
800
800
800
800
800
800
800
800
810
8kO
812
816
D - Maaa
Exit
716
740
803
825
817
830
841
846
829
812
811
811
813
813
313
813
813
813
812
812
807
821
824
829
9:.* ^ld
Inlet Temp.
17«
175
176
180
183
185
187
187
189
100
190
191
192
192
192
193
194
195
198
204
206
208
208
204
Acid Teoperature
Exit Absorption
Tower
200
203
201
206
210
212
213
213
215
217
216
216
216
218
219
218
219
227
229
229
233
232
IWTBi All Temperature! are In F.
-------
V)
to
TABLE 14
START-UP PLANT "U"
AMMONIA SCu'JBBLR CONDITIONS AND EMISSIONS DURING ACID PLANT START-UP
% SO, 2 S02
Entrance to Absorbing Tower
Time Converter Exit
2:30 PM .24
2:45 .22
3rOO .208
3:15 7.6 .30
3:45 7.9 .335
4:00 8.8 .2;
4:15 .245
4:45 9.2 .235
5:00 8.6 .205
5:15 .20
5:30 8.4 .195
5:^5 .205
6:(.0 8.4 .19
-------
00
Ul
10.000
u
8
2 5.000
x
.0
e
a.
a.
3.000
2.000
o 1.500
1,000
o
0)
a
8
500
Sulfur Conversion - Percent Feedstock Sulfur
99.7 99.0 98.0 96.0 92.9
' ffRgfegffifitft finffr i •
- EXAMPLE OF SO, EMISSIONS DURING *•
START-UP OF SINGLE ABSORPTION
ACID PLANT VS. SINGLE ABSORPTION
ACID PLANT WITH COMIHCO-TYPE
AMMONIA SCRUBBER PLANT U,
10.0 15 20 30
50
100
200 300 500
1000
Figure 13
-------
The only operating parameter for plant U, that seems to be
abnormal is the temperature of the acid in the absorber.
This temperature is somewhat higher than normal, possibly in-
dicating a shortage of cooling water for the acid coolers in
the absorption system. This high temperature possibly con-
tributed to an increase in acid mist emissions although con-
firmation of this fact is not available in the data.
Single Absorption Plant with Vent Gas Cleaning Systems
Plants U, and U2 described in the previous section are equipped
with ammonia scrubber vent gas cleaning systems. Extensive
data were obtained from these two plants in various modes of
operation. The temperature profile across plant U, is pre-
sented in Table 13, while absorber operating conditions are
presented in Table 14. The performance of the ammonia scrubber
applied to plant IK is illustrated in Figure 13.
A significant feature of Figure 13 is the illustration of con-
stant SO, concentration from the ammonia scrubber in terms
of parts per million by volume over a range of scrubber inlet
SO2 concentration of 1,900 ppm to 3,350 ppm. This comparison
illustrates the lack of precision in defining S02 emissions on
the basis of ppm in the vent gas. When calculated in terms
of pounds of SO, per ton of acid produced the output from the
ammonia scrubber varied from i.5 to 10 pounds per ton. SO2
emissions presented in mass units much more clearly illustrate
the effect of variable input concentrations of S02 to the
ammonia scrubber. Although plant U2 showed a performance of
84
-------
5.5 to 10 pounds of S02 per ton of acid produced, this plant
is capable of producing vent gas containing less than four
pounds per ton during periods of careful operation and no
malfunction.
Start-up data for plant U2 is presented in Table 15 and 16, with
ammonia scrubber performance in Figure 14. Plant U2 is identi-
cal to plant U,, and comparison of temperature profile across
the plant and the SC>2 concentration differential across the
converter indicates the two acid plants were operating in
very similar manner. Comparison of the S02 emissions from
the ammonia scrubber, however, indicates a great difference in
operation between the scrubbers on plants V^ and plant U2.
During the start-up operation of plant U~, the vent gas booster
blower used in conjunction with the Brink filter in the scrub-
ber outlet was not in operation. This blower was shutdown
because of lack of steam for driving the turbine. With the
blower shutdown and the Brink filter bypassed, a very dense
plume of particulate matter was being emitted from the ammonia
scrubber. Since all process parameters for the two scrubber
systems are similar except for the SO.^ emitted, the only
explanation for the difference in SO2 emissions is error in
the S02 analysis. The SO2 was analyzed in this start-up
using the Reich test. If the gas sample collected for the
SO2 analysis were insufficiently filtered it would contain
quantities of ammonium bisulfite and ammonium sulfite. parti-
culate, that interferes with the S02 analysis producing a
85
-------
TABLE 15
START-UP TEMPERATURE CONTACT ACID PLANT U» SINGLE ABSORPTION PLANT WITH AMMONIA SCRU2BER
MASS MASS MASS MASS
Boiler A B C D Inlet Acid Inlec
Time Exit Entrance Exit Entrance Exit Entrance Exit Entrance Exit To Scrubber Temperature
00
6:00
6:15
6:30
6:45
7:00
7:15
7:30
7:45
8:00
8:15
8:30
8:45
9:00
NCvrv.
503
550
563
576
590
602
6C4
609
612
615
615
619
/•TT TEX
629
640
684
710
729
760
781
802
814
814
819
314
817.
pPDAfUO^C
810
878
955
1050
1118
1050
993
1090
1095
1116
1132
1147
1149
1 *RE ™
760
746
773
768
742
733
762
822
869
868
856
852
849
p
734
793
845
872
975
1000
878
803
906
916
986
982
971
738
771
822
859
880
900
843
795
851
882
890
883
886
684
757
828
855
878
975
1010
983
871
849
889
926
925
714
733
775
812
833
878
905
908
843
806
811
856
850
490
604
738
786
793
846
883
926
924
884
840
856
871
130
139
142
162
162
180
188
195
200
203
206
204
200
159
184
184
197
202
208
212
215
215
208
203
• i
-------
TABLE 16
AMMONIA SCRUBBER CONDITIONS AND EMISSIONS DURING ACID PLANT START-UP ON PLANT "U2"
(COMINCO SCURBBER WITH MIST ELIMINATOR)
CO
Time
6:00 PM
6:15
6:30
6:45
7:00
7:15
7:30
7:65
8:00
8:15
8:30
8:45
9:00
9:15
9:30
9:45
10:00
10:15
10:30
10:45
11:00
11:15
11:30
Entrance to
Converter
% SO,
6.45
7.5
8.8
3.3
8.55
•
8.55
8.65
8.5
9.0
Absorbing
Tower Exit
% SO.,
.278
.34
.255
.255
.24
.245
.255
.33
.33
.255
.18
.205
.21
.185
.19
.235
.183
Effluent
Outlet
% SO*
.118
.091
.121
.190
.152
.128
.128
.128
.110
.103
.094
.110
-------
100,
An Example of a Single Absorption Plant "U2" Entering and Leaving
Malfunction Operating Conditions
SULFUR CONVERSION - PERCENT OF FEEDSTOCK SULFUR
99.7 99.0 98.0
97.0 96.0 95.0
S?;LFURIC ACIO PLANT FEEDSTOCK SULFUR.
CONVERSION VS. VOLUMETRIC AND MASS
S02 EMISSIONS AT VARIOUS INLET S02
CONCENTRATIONS BY VOLUME
Input to Comincc
Output of Cominco Emissio
I I I I I I I I i I
1.5
2.0 2.5 3CO4.0 5.0 6.0 7.0 9.0
15
20 25 30
40 50 60 70 80 90100
S02 EMISSIONS LB. PER TON OF 100% HjSO^ PRODUCED
Figure 14
Note: Numbers represent clock time sequence.
88
-------
higher result. This explanation for the high SO. emissions
cannot be verified at this point, but appears to be the most
logical explanation. In addinion the SO, emissions from the
scrubber from plant U_ do not show the same degree of varia-
tions with changes in scrubber input SO- concentrations as
seen in data from plant U^. This fact may be an indication
the particulate matter is masking the true SO, concentration
in the vent gas.
Dual Absorption Plants
It has been previously stated that the dual absorption plants
can be started up without exceeding SO, emissions by careful
attention to procedures and process control, and if malfunc-
tions of equipment and control instruments do not occur. This
type of start-up, however, requires up to five days of prepara-
tion time, and process equipment and instrumentation do not
always perform properly after a shutdown period. A preheat
cycle for a dual absorption acid plant not equipped with
auxiliary heaters is shown in Figure 15. During this preheat
cycle all of the heat for preheating the converter is obtained
by firing fuel in the sulfur furnace. Since the maximum temp-
erature of the sulfur furnace is 1,500F, fuel must be burned
in the furnace in a cyclic manner to prevent overheating the
furnace. The off-on control of heating in the furnace permits
adjustment of the heating rate of each catalyst bed. Because
of the thermal lag throughout the converter, the first catalyst
bed is heated in a cyclic manner reflecting the off-on cycles
89
-------
:> a r>o -
COLD START-UP OF TYPICAL DUAL ABSORPTION ACID PLA
RJRNACE
HEAT-UP
CYCLE
PED
1742V
LEGEND
FURNACE TEMP, F
•CONVERTER TEMP. °F
4 5 6
TIME IN DAYS
5
Figure 15
-------
in the furnace, while the fourth catalyst bed heats at a slow
rate continuing for a period up to five days. If sulfur feed
to the furnace is started before the fourth catalyst bed
reaches operating temperature, incomplete conversion will
take place in this bed causing high emissions of SO., from the
absorber. Because of lost production resulting from the slov:
preheating procedure, preheat is often reduced and sulfur
feed begun before proper temperature is obtained in the fourth
bed. This is the primary cause of high S02 emissions from a
sulfuric acid plant during the initial one to two hours of
start-up operations. Once SO- is admitted to the converter
the exothermic oxidation in the converter will quickly bring
the catalyst bed to operating temperature, reducing the SO2
emissions to a more normal level.
Another cause of high SO2 emissions during ste.rt-up operations
is operation of the sulfur furnace at full sulfur feed rates
from the beginning of the start-up procedures and before
catalyst bed temperatures have stabilized. These emissions
on start-up can be reduced by initiating sulfur flow at re-
duced rates until proper operating temperatures in all catalyst
beds have been obtained.
A typical dual absorption plant start-up will produce SO2 emis-
sions with a peak value of up to 3,000 ppm at one to two hours
after start-up. Emissions will normally be in excess of per-
formance standards for a period of two to four hours after
the beginning of sulfur feed. This range and duration of
91
-------
emissions will normally occur during start-up operations fol-
lowing a shutdown period of more than four hours. This is
primarily due to the lack of adequate preheating time when
converter temperatures have fallen below permissible limits.
These start-up relationships are shown in Figure 12, discussed
in the previous section.
The complete history of SO2 emissions for a dual absorption
plant (Plant "Q") considered typical during start-up opera-
t~i$ns is shown in Figure 16. This curve indicates that the
plant was started up w:.th maximum SO? concentration feed to
the absorber, with a p6=ak SO, emission reached in approxi-
mately two hours after start-up began. At this point the sul-
fur feed concentration was reduced, causing the 802 em^ss^on
to fall from 1,500 ppm to 600 ppm. The SO- feed concentration
was again slowly increased as the converter stabilized, and
emissions continued to drop to 200 ppm at normal SO2 concen-
tration in the feed. At this point the unit was operating
at a total emission of 2.4 pounds per ton of ^^Q4 produced.
An attempt was made to establish a statistical model for start-
up emissions in terms of converter temperatures using data
from plant "J" start-ups. A standard statistical analysis
of these same thirteen data points was presented in Table 5B
as compared to normal data. The regression analysis gr.ve
correlations that could not be considered significant, for
the same reason discussed in the section on normal operating
data, in addition, the inability of the catalyst masses to
92
-------
An Example of a Dual Absorption Pl*nt ("Q") Entering and Leaving Malfunction
Operating Conditions—During Start-Up
SULFUR CONVERSION - PERCENT OF FEEDSTOCK SULP'R
i
g
H
m
<
u
10.000
1.000
i.OOO
7.000
•,000
3.0JO
4.000
1.000
I.MO
2.090
I.MO
I.OOO
MO
•00
700
«00
MO
400
MO
IM
200
100
1
99.7
FERFORXAHCF. STANDARD
99.0 98.0 97 C 96.0 95.0 92.9
tICBD
innnit liaUf t»taalooa During ShuUm
S^aV.riSS'Starl-vV' fjalasloa D«cr**aUic TTI
1400 In. 4 14JO li 1*40
1410 7 IJOO 12 ISM
1420 I 1510 11 1620
14M » ISJO 14 1700
14(0 10 15)0
/
/&
/
^/
/
fy
/
/
t
if
S
/
y
^
'/
/
/
y
t
'
/
y
y
/
'
/I
*?
/
/
X
SULFURIC ACID PLAI?T FEEDSTOCK SULFUI
CONVERSION VS. VOLUMETRIC AND MASS
S02 EMISSIONS AT VARIOUS INLET SO,
CONCENTRATIONS BT VOLUME
. Tnod
YOt
y
•d
yT sj?
VT
fXi/i
^
J J.
f
/
/
/
y*
>
!/
(
9
/
1
f
/
/
'
^
/
/
{
A
.
$.
* f
s
i
/
(* y'
T'
S A
i /X /
/
s
s
s
/
/
/
j
/\
/
^
y
/
y
' .
/
/
y
A
s
/
S
fe^y
y^
^
j
y
/
^
s
f
'
/
f
J
j
'
s
t
*
/
t
/
/
^
^
y
/I
w
4
Ijr
y
s
/
j
y
/
j
f
s
/
y
'
\s
/
j
1
v
j
9
,
/
/
'
'
>
^
/
/
11 20 2J 10 40 M 45 70 H N100
SOj EMISSIONS - LB. PER TON Ot '.OOX H2S04 PRODUCED
Hote: NunKer valuea represent clock tlse Intervals froa a stable steady-state dual
absorption plant approaching excessive emissions and returning back to normal
operating conditions within 1400-1740 or three hours 40 minutes tlae. Violation
time less than two hours.
Figure 16
93
I
it
-------
follow sudden changes in SO- inlet concentrations causes a
displacement in the temperature—SO- emissions relationship
during the transient conditions of start-up. An acceptable
model of the catalyst operation during start-up must include
factors involving this lag in conversion efficiency that could
not be determined from the data avaiable.
The 13 data points considered in the regression analysis were
also analyzed statistically for SO- emissions in terms of
time after the beginning of start-up operations until 300
ppm S02 emissions is reached. It was calculated that the
average period was 6.34 hours, with 95 percent of all start-
up periods falling between 3.14 and 9.54 hours. These results
were in agreement with the graphical data presented previously
in this section. Details of the statistical analysis are in-
cluded in Appendix "C"-
MALFUNCTION CONDITIONS
Single Absorption Plants
Single absorption acid plants cannot meet performance stand-
ards for SO2 emissions without the addition of vent gas clean-
ing systems. Acid mist emissions from single absorption plants
are also frequently above the performance standards unless the
plant is equipped with special means for eliminating this mist.
Although single absorption plants are not specifically included
in this study, data were obtained from two single absorption
acid plants (Plant U, and U2) for acid mist emissions when
using both dark and bright sulfur. These data serve to illu-
94
-------
strate the problems encountered when burning dark sulfur in a
single absorption plant that is not equipped with appropriate
filters to remove the acid mist. As illustrated on Figure
17, the emission of acid mist for dark sulfur feedstock is
approximately three times that encountered with bright sulfur.
The emissions from dual absorption acid plants equipped with
norm?! mist eliminators is plotted on the same curve for com-
parison. The burning of dark sulfur in the modern dual ab-
sorption acid plant does not normally affect the acid mist
emissions significantly, since these plants are equipped with
efficient mist removal systems. The single absorption acid
plants plotted on this figure are equipped with high efficiency
Brink filters, and when these are in operation acid mist emis-
sions are below performance standards.
The range for acid mist emissions for all plants studied,
burning dark sulfur and bright sulfur, are shown in Figure 18.
The plants described in Figure 17 fall within the range of
all the plants studied.
A statistical study was made of Plants l^ and U2 in an effort
to show the effect of catalyst aging on SO2 emissions and acid
mist. The results of the study for S02 emissions were in-
conclusive, since factors other than catalyst aging mask
the change of S02 with time. One of the major causes of S02
variation in time was the introduction of dark sulfur into
the unit during some periods of operation. The specific effect
of dark sulfur on S02 emissions was not clear, however, since
95
-------
0\
u
CO
01
a.
C?
«/)
CM
oo
s
c
o
£
u
I
Acid Mist Emission Average Operating Levels for Both Dual (Plant Q) and
Single Absorption (Plant U-^ & U2) Contact Acid Plants
.92 99.7 99.0 98.0
Plane Uj-
Dark Sulfur
IL 25.0 mc/'SCF
Plant
Bright Sulfur
r
SULFURIC ACID PLANT VOLUMETRIC AND
MASS EMISSIONS OF ACID MIST AT VARIOUS
INLET SOo CONCENTRATIONS BY VOLUME
.01 .02 .03 .05 ' .10 .15.20 .30 .50 1.0 2 3
Acid Miat Emission* - Lb H2S04 Per Ton of 100Z HjSO^ Produced
Figure 17
-------
NORMAL RANGE OF OPERATION FOR ACID MIST EMISSIONS
OF NEW DUAL ABSORPTION AND SINGLE ABSORPTION
ACID PLANTS BURNING BRIGHT AND DARK SULFUR
10.000
3.000
1.000
2.000
1.300
1.000
500
100
a
8
e
1
IIIIUI
LEGEND
New Dual Absorption Plants
'•- Single Absorption Acid Plants
^^
1
Burning Bright Sulfur -•
Single Absorption Acid Plants j '
Burning Dark Sulfur
.02 .01 .04 M .1 .1 .1 »S U> V V> IP V>
AC10 HIST EMISSION*. LB KjSO^ PE* TOM Of lOOt HjSO4
Figure 18
97
-------
other factors were also involved. The effect on SO- emissions
of catalyst aging probably cannot be determined using this
type of evaluation, since the addition of new catalyst during
each cleaning cycle will tend to replace the aged catalyst over
a period of approximately 10 to 12 years, thus maintaining the
overall activity of the catalyst bed at a point slightly below
that of a new plant.
Correlation of operating time with acid mist for those plants
however show a very significant change during the period when
dark sulfur was being burned. When bright sulfur was burned
a very constant acid mist emission rate was experienced. When
dark sulfur was introduced to the plant the acid mist emis-
sions increased (approximately 20 times) until the dark sulfur
feed was discontinued and bright sulfur was again fed to the
furnace. With the introduction of bright sulfur, acid mist
immediately dropped to the normal level. A plot of the acid
nrst emissions with respect to time is shown in Figure 19.
Single Absorption Plants with Tail Gas Cleaning
Operating data were obtained from an extended period of opera-
tion from a single absorption acid plant burning bright sul-
fur and equipped with an ammonia scrubber. Three operating
periods for this plant (Plant "K"), containing a series of
malfunctions is shown in Figures 20, 21, and 22, A study of the
total operating data from this plant indicates that the important
operating parameters are scrubber gas flow, ammonia concentra-
tion in scrubber liquid, and scrubber liquid pH. In the opera-
98
-------
COMPUTER PLOT OF ACID MIST VS. TIME
SO,
Plant M,
---•%«(. Ml — -««lUl
».» -1.*
5J} -|'l
;:( :S:J
I.I -O.I
».? ;o.>
H.« 0.7
.J 0.1
.1 O.I
!• o!e
'} I'D
t'l hi
•J !•?
*1 «
:t J:|
In ?!<
l!» lln -
Sit !i.
ii! lit
0.* J.«
oU «Ii>
J:? ?:?
8:j |:S
8 it ?*•
.; ».o
S'J S'«
!l? rlj
»•>'.« «' i
l*.« «.s
»!• i'. i
!A.| -V.,
••0 «./
b-i •••
>.•>..' V.N
till »!l
; «.« iri.u
«.» lu.l
• — ii.j - io./.
i I.* 10. I
|.» 1U.«
1 . 7 \ r . *
• I.i- . Ulf
1.0 lll.i
9.5 III. 1
(.> ll.l
i!« "no
»•• II. •
«.' II. «
t
1
1
- 1
t
1
1
t
1
t
1
..A
f
• r
1
• A
* A
• A
• •1
-A
-1
• 1
-1
• 1
- -s
•-1
-\
•S
. A
\i
A
11
'1
l»
1 1
l«
1 •
i:
i *
1 1
- 1
- 1'
• i *
v !'"«
1/II/A4
•/?>/*<
1A/ )/««
ii/ «/'*
I/ ^/&J
>/ J/*»J
!/!>/»*
1/1 '/c9
*/ 1(//AO
O/ 1 '/6-J
10/70/6:
llii/ii
*/ */r*i
|!/ o/r.i
l>f ?//j
|?/|«/7Q
n{ i^ti
— !/Jv/"
•/ii/"
^/ I/"
• /to.'/'
ii/'//»>
tt 0" I
i/'l/M
«/'//r»
A/ I// 1
V/11//I
.... «/;«//!
. IO/IM/JI
'•lo.co "" -J.on "" i
l
j
ii
u
10
t'
!i.
?6
lu
II
)» ..
Jv
J
n
5u
•*«
jj
«*
/i
/i
'S
/b
/I
'9
«• 1
4 /
ii
«•
16
Ji
101
Hi
i D
1 A
1 t
I 1
1 i
\ t
I A
1 A
1 A
I 1
1 A
"""{""
[\ .
'I
J«
... .
J.i..
I/-.
o*t*nroiMS di lifO iut
.00 i«.oa J7.00 )c.*
—
— •
• - -
A 1
A
*. *
jzijVLi i-
A
1
A
A
...... - I
A
S v
A -
Dark Sulfur
Feed
A
>
A
\ —
v i
\ i
I •
I • • i
1 A
\ <
\- '
'4 ;.-••..;; •
.
Reproduced from
best available copy.
FIG. 19
99
-------
MALFUNCTION OF SINGLE ABSORPTION. ACID PLANT K WITR AMMONIA SCRU35ER
1200-j
o
o
000-
JCO
20O
100-
PPM. SOtOUT
•NH3 CONCENTRATION
Te 10
12
z'4
4 T .'o1 fe1
4 ' 4 '
n
TIME 'IN HOURS
Figure 20
-------
MALFUNCTION OF SINGLE ABSORPTION ACID PLANT ;<
WITH AMMONIA SCRUBBER
•- z
int A % M Lj MI mtti MI i mil i tu i iMmttm/rri'
CONTROL* POINT
LEGEND
T:;:
RRM. 50? OUT \ :-2.'. z
NH5 CONCENTRATIOfJ ;j
4 '
• fe ' i • r^i^r^
TIME 'IN HOURS
^t ^^ i • 4 • 4 ' *
Figure 21
-------
rt II
o
to
IL'OO-
I !'_•(*•
vCC-i
sccH
H-
13
O
c?1
co •-
c! i
c: 1
tttntmtitttn
HQO-
300
20C-J
-*J
MALFUNCTION OF SINGLE ABSORPTION ACID PLANT K
WITH AMMONIA SCRUBBER
I
i Z
i ?
U- -
! ~-
N! H rwMiMi«»tiuit«»miumtMi
CONTROL POINT
-4 '^
PRM. S0t OUT
NH3 CONCENTRATION
r-
Z
' ' ' ' '
i
' 4 ' i ' ft) ' iW i ' 4
i
' A
' A ' .'o '
TIME hN- HOURS
Figure 22
-------
tion of the plant observed, the last two parameters were gener-
ally related. Ammonia concentration was selected for plotting
against sulfur dioxide emissions, with periods of abnormal gas
flow noted on the graphs.
Figure 20 contains one period of 24 hours when SO- emissions
were below 300 ppm. During this period of time the ammonia
concentration was steady at six percent. Prior to this opera-
tion the ammonia concentration was out of control, generally
running lower then six percent, causing the S02 emissions to
maintain a constant 800 ppm. At the end of the stable oper-
ating period the flow through the scrubber was increased
beyond the scrubber capacity for a short period, causing the
SO- emissions to increase drastically from 300 ppm to approx-
imately 1,000 ppm. During this time the ammonia concentration
remained generally high but was unable to control the emission
of SO2 because of high gas flows resulting from hignacid pro-
duction. A reduction of gas flow for a ohort period again
reduced the emissions to below 300 ppm, followed by another
period of high gas flow with 1,000 ppm emissions. This curve
shows the effect of exceeding the scrubber capacity for gas
flow, resulting in a large increase in S02 emissions. Simply
maintaining high concentrations of ammonia in the solution
during these high gas flow periods will not maintain S02
emissions at the appropriate level.
Figure 21 further illustrates the effect of ammonia concen-
tration on SO- emissions. During the initial part of this
103
-------
operating period the gas flow through the scrubber was below
the normal level, permitting control of SO- at 250 ppm with
an ammonia concentration of only five percent. When the ammo-
nia concentration was lost because of ammonia evaporator prob-
lems, SO2 momentarily increased to 1,000 ppm before the ammonia
concentration was raised reducing the S02 emissions. Continued
upset in the control of ammonia concentration eventually re-
sulted in the shutdown of the unit.
In Figure 22 the ammonia scrubber was operating at a gas flow
above its normal rates during the initial part of this period.
During this 12 hour period the S02 emission was 500 ppm with
an ammonia concentration of seven percent. Significant in-
creases in the gas flow through the scrubber resulted in loss
of ammonia concentration control, with the ammonia concentra-
tion dropping to the set point of four percent. During this
period of time the SO- emissions increased to 800 ppm. This
portion of the curve illustrates the inability of the scrubber
to maintain SO, emissions at the 300 ppm level with an ammonia
concentration equal to the four percent setpoint. After con-
tinuing upsets in flow rates and ammonia concentrations during
the remainder of the two and a half day period shown on this
curve, the ammonia concentration was again established at approxi-
mately seven percent,and with gas flow returned to normal, S02
emissions dropped below 300 ppm.
These three figures show the need for careful ammonia concen-
tration control at a level of approximately seven percent
104
-------
and gas flow rates within the proper limits of scrubber design
if SO, is to be maintained below the 300 ppm standard. Con-
trolling the anunor.ii at a concentration of four percent will
reduce the amount of ammonia consumed in the scrubber, but
will always result in an emission level of approximately 800
ppm when gas flow rates are within design limits.
Data were collected on one single absorption acid plant equipped
with a sodium scrubber. This plant is designated as Plant "Y".
Figure 23 shows the response of the sodium scrubber to a sud-
den drop in gas flow rates through the scrubber. The design
flow of this scrubber is 5,000 scfm but normal operation is
set at 6,300 scfm. During the period of operation shown on
the graph, the gas flow through the scrubber was dropped to
3,200 scfm causing the SO2 emissions to drop from 940 ppm to
50 ppm. During this same interval the scrubbing solution
temperature increased and the pH dropped slightly. These
other parameter changes were primarily results of the drop in
flow rates.
The sodium process is capable of reducir" S02 in the vent gas
to 250 ppm in normal operation, as derao.n ..rated by EPA performance
test. The data from plant "Y" indicates the high sensitivity
of the process to flow rates, and the need to restrict gas flows
to within design limits.
105
-------
1 00 it
81?
£3
75*-
65 &-
HALFUWCTION OF SINGLE ABSORPTION_PLANT_"Y" WITH SODIUM SCRUBBER
*< s *«
H M «/>
I2O
no
too
9O-
eo-
fl
5uO-
I
7000
6000
5000
4000
3000
o
u
in
8
1200-
900 -
600
300-
2 ' 4 ' 6 ' 0 ' 10 12 14 ' IS 18 20 22 21 26 28
TIME IN HOURS
Figure 23
106
-------
Dual Absorption Plants
A series of operating data for plant "Q" covering a five hour
operating period is presented in Figure 24. Plant "Q" is a
dual absorption acid plant burning bright sulfur, with a
capacity of 1650 tons per day of 100 percent sulfuric acid.
At the time the data were taken the plant was still in its
shake down period after its initial start-up, and all the
initial construction problems had not been eliminated. The
five hour period shown contains a series of malfunctions
affecting a variety of process parameters and causing excess
•
emissions of SO- and acid mist.
The basic malfunctions that occurred were the loss of tempera-
ture control in the fourth bed of the converter, loss of
dilution water to the acid tanks feeding the absorbers, and
loss of sulfur flow to the sulfur furnace. The result of the
combination of these malfunctions was an increase in SC^ con-
centration from less than 100 ppm to a peak of 1,300 ppm and
an increase in opacity from zero to 40 percent.
The sequence of events during the five hours of operation can
be followed by referring to the large numbers of Figure 24, and
is described in the legend on the second page of the figure.
During the first hour of operation the plant was operating
with normal emissions and with all parameters in normal range.
At the one hour point temperature of the fourth bed of the
converter began to show loss of control. This decrease in
catalyst bed temperature began to change the SO2 concentration
1C7
-------
TRANSIENT POCESS CONDITIONS DURING DUAL ABSORPTION PLANT MALFUNCTIONS
IhR
2HR.
5 HP
100
• CONTROL POINT
• — 7O MIN.
100
-LOSS OF ACID CONCENTRATION CONTROL-
• CONTROL POINT
2 96
u.
2 MR.
3HR.
4HR.
5HR.
ZI400
UJ
I'200
' 2lOOO
5600
2600
400
200
0
MO MIN. EXCESSIVE sO EMISSION
75 MIN. LOSS OF
10 CONCENTRATION CONTROL
ERX STANDARD 10% OPACITY
EPA. STANDARD 300RP.M. SO
3HR.
/Of
TIME IN HOURS
FIG.24
4HR.
5HR.
-------
FIGURE 24 (CONTINUED)
DESCRIPTION OF SIGNIFICANT OPERATING POINTS
Point
No.
1 Dilution water off with resulting rapid rise in
l^SO, concentration and SO, Emissions
2 Beginning of excessive SO. Emission do* to loss of
converter temperature control
3 Sulfur gun plugged
4 Loss of 50Z sulfur flow
5 Loss of 75% sulfur flow
6 25Z sulfur flow
7 50Z sulfur flow
8 75Z sulfur flow
9 100Z sulfur flow
10 Temperature recovery point - start of conversion improvement
11 Stabilized converter bed temperatures
12 S0_ Emissions back within EPA standard compliance due to
stable converter temperature levels
13 98.5Z H,SO, concentration point at which high (30%-40Z)
opacity plume disappears
14 Dilution water malfunction
15 96.5Z acid in final absorber - poor 503 absorption into
high opacity plume
16 Opacity due to loss of H7SO. concentration control
17 Subcooling of converter temperature
18 Lost 4th bed conversion because of interpass absorber failure
108
-------
in the vent. The SO2 concentration increased slowly at first,
crossing the 300 ppm line at point 2. A short time after this
point, dilution water flow to the acid pump tanks feeding t/io
absorbers became inadequate causing the concentration of acid
in the absorbers to increase. The concentration of acid to
the primary absorber crossed the control point of 98.5 percent
acid concentration at point 1. The increased acia concentra-
tion was seen in the final absorber at point 14. The reduction
in dilution water flow resulted from a makeup water line of
insufficient size to provide the required water. When the
insufficient water flow became apparent from the control charts,
water hoses were added to the pump tanks to dilute the acid
and restore concentration control.
As the concentration of acid in the primary absorber began to
increase the amount of SO-, absorbed in the acid was reduced,
increasing concentration of S03 in the fourth bed of the con-
verter. This increased SO., reduced the efficiency of the
fourth bed, causing an increase of SO, emissions from the vent.
The result of this reduced conversion efficiency in the fourth
bed is seen in the converter temperatures at point 18. As
the SO- emitted from the stack began to increase, the opacity
readings began to climb at about the third hour.
While the SO, emissions were increasing and the opacity was
rising, the operators attempting to re-establish dilution
water flow failed to notice the sulfur furnace temperatures
were dropping. Low sulfur furnace temperatures resulted from
109
-------
the failure of one sulfur gun at point 3, reducing the flow by
25 percent, and the loss of a second sulfur gun at point 4, re-
ducing the sulfur flow to 50 percent of normal. A third sul-
fur gun failed at point 5,leaving only 25 percent of normal
sulfur flow at point 6 on the furnace temperature curve.
During this period of loss of sulfur flow, two actions were
occurring that partially counter-acted each other. The loss
of S02 flow -to the converter reduced the temperatures in the
converter, which in turn reduced conversion efficiencies.
During the same period the reduced SO- concentration in the
feed reduced the need for conversion, permitting the excess
SO- and unabsorbed SO- to be purged from the primary absorber
and fourth catalyst bed. As the SO- was removed from the
fourth catalyst bed the efficiency of this bed increased,
reducing the amount of S02 emissions from the stack. The
SO- emissions reduction begin at about 3.5 hours.
When the dilution water flow was re-established at point 15,
opacity began to fall at point 16, reaching zero at point 13.
After it was determined that the three sulfur guns had failed,
they were immediately restarted (at points 7, 8, and 9), re-
storing the flow of S02 to the converter and starting reheat of
the catalyst beds. After re-establishing the sulfur flow the
catalyst beds reached stable temperature at point 11, one half
hour after starting the last sulfur gun. The short time re-
quired for recovery of normal converter temperatures illustrates
the large heat imput from the SO2~S03 reaction compared to the
110
-------
small heat loss (illustrated by the slow cooling of the con-
verters when the flow of S02 to the converter was reduced).
The SO2 concentration from the stack reach 300 ppm at point
12, 4.5 hours into the operating period.
A very interesting sequence of events is presented in Figure
24, pointing up one of the major differences in operation
between single absorption and dual absorption acid plants.
In a single absorption plant, Iocs of dilution water to the
absorber increases the SO, ^.mission, causing acid mist, but
does not appreciably change the SO, concentration in the
vent stack. In a dual absorption plant however, loss of
dilution water to the primary absorber reduces the absorption
of SO,, increasing the concentration in the fourth catalyst
bed reducing its conversion efficiency. This will result in
an SO, emission increase. Continued loss of dilution water,
causing an increase in concentration of acid in the final
absorber, results in high SO, emissions and increased opacity.
In the case of the plant "Q" experience, an acid concentration
of 103 percent is reached in the primary absorber and about
100 percent in the final absorber. It is also interesting
that in the case of the operation of plant "Q" the S02 emis-
sions from the plant decreased as the sulfur feed to the fur-
nace was being lost and catalyst temperatures were falling.
This sequence indicates the SO2 emissions under the operating
circumstances encountered in this event are more sensitive to
S02 inlet concentrations than to catalyst temperatures. The
111
-------
curves also show the slow decrease in catalyst temperatures
when sulfur is lost, and the very rapid increase in catalyst
temperatures .vhen sulfur flow is reinstated. It appears from
this sequence of events that if loss of sulfur guns is detected
sufficiently early to re-establish flows within a reasonable
time, the converter temperatures can be re-established in a
short period without causing excessive SO, emissions from
the plant. Considering the size of Plant "Q", the transient
conditions exhibited during this sequence of operations are
surprisingly fast. This illustrates the need for close temp-
erature and flow control of all the major process parameters.
112
-------
SECTION VII
DISCUSSION
EMISSIONS DURING NORMAL OPERATION
The normal emissions for steady state operation of sulfuric
acid plants consist of sulfur dioxide (S02), acid mist and
opacity. Acid mist and opacity are related and under certain
conditions, are different methods of measuring the same pol-
lutant.
The SO2 emission from a sulfuric acid plant results from a
single cause. This cause is the inability of the catalyst
in the converter to convert 100 percent of the SO, generated
in the sulfur furnace into SO, for eventual ahsorpbion in the
absorber. The effect of conversion efficiency is illustrated
in Table 2, Page 41 of Section VI. Unconverted SO2 is un-
affected by passing through the absorber and is emitted with
the vent gas.
The acid mist, on the other hand, can rer.ult from three dif-
ferent sources, all of which generate sulfuric acid, the
ultimate form of the acid mist emitted. The most direct
source of acid mist is c-rry-over of liquid particles of sul-
furic acid from the top of the absorber. These particles
range in size above three microns. A second source of acid
mist emission is the formation of fine particles of sulfuric
acid in the system by the reaction of SO3 from the converter
with water vapor contained in the gas. This water vapor is
present because of insufficient drying of combustion gas or
113
-------
the generation of water vapor by the combustion of organic
material contained in the sulfur. The effect of dark sulfur
on emissions can be seen in Figure 17, Page 96 of Section VI.
This source also produces sulfuric acid particles in the range
of three microns and larger. The third source of acid mist
emissions is the failure of the absorber to fully absorb all
S03 fed to the bottom of the tower. Any SO3 leaving the top
of the absorption tower will immediately react with moisture
in the ambient air to form very fine particles of sulfuric
acid. These particles range in size below one micron.
Measurement of the opacity of the vent gas stream from a sul-
furic acid plant is to a large extent the measure of the amount
of sulfuric acid emitted from the process. Because of the
fine particle nature of the acid mist and the effect of atmos-
pheric conditions on the emissions, the measurement of opacity
may be greater in one situation for the-same amount of acid
mist emitted than in another. Atmospheric effects on opacity
may be seen in Table 7, Page 63 of Section VI. Acid mist
resulting from acid carry-over from the absorber or reaction
between S03 and water vapor in the absorber produces a dense
white cloud in a cool, moist atmosphere, while the reaction
of S03 gas and atmospheric moisture produces a blue cloud of
a less dense nature. In addition to acid mist, the oxides
of nitrogen (NO ) may also contribute to opacity. The pre-
jC
sence of NO and SO- in the g?s stream produces acid mist
X «
by formation of a NO -S00 complex similar to the reaction in
"
114
-------
the chamber process. Nitrogen oxides are generated by com-
bustion in the sulfur furnace or by the release of nitrogen
oxides from combustion of organic material contained in the
sulfur.
Steady State Emissions and Equipment Limits
As shown in Table 2, Page 41 of Section III a typical single
absorption sulfuric acid plant emits SO2 in a concentration
between 1,500 and 4,000 ppm. This S02 concentration results
from operation at normal design capacities depending upon the
efficiencies of the converter to oxidize S02 to SO,. The
major variables influencing converter efficiency are the
amount and condition of the catalyst contained in the converter,
The catalyst becomes dirty with use and loses efficiency. Ef-
ficiency can be increased by cleaning. The claaned and par-
tially replaced catalyst will again perform at higher effi-
ciencies ; with proper cleaning and operation a catalyst
bed will serve for a period of 15 to 20 years. During the
cleaning process, six to ten percent of the catalyst is lost
due to screening of the material in the first bed and must be
replaced by new catalyst. The new catalyst is generally
placed in the last bed and the catalyst from succeeding beds
is used to replenish the material in the first bed. Misopera-
tibn of the converter or the use of sulfur containing large
quantities of organic material can increase the deterioration
rate of the catalyst. The effect of burning dark sulfur on
catalyst activity is shown on Figure 19, Page 99 of Section VI.
115
-------
The use of dirty sulfur feedstock will increase the frequency
of cleaning of the catalyst as well as hasten the time of re-
placement. Overheating of catalyst to the range of 1200F
will reduce useful life of the catalyst to three or four
years. Careful operation and maintenance of the converter
will assure SO- emissions in the range specified in the design
of the plant.
The most important operational parameters effecting SO- emis-
sions from a single absorption acid plant are the temperatures
of the converter bed?. Converter temperature ranges are shown
in Table 4A, Page 51 of Section VI. Operation within a re-
latively narrow range of temperature for each bed is required
to achieve optimum conversion. These temperatures are con-
trolled by adjustment of by-pass dampers on the heat exchangers,
by careful regulation of the SO- concentration in the feed
gas to the converter, and the temperature of the sulfur fur-
nace. The allowable range of converter temperatures is shown
in Table 4C on page 56 of Section VI, Analysis of Process
Variables and Emissions. Mean operating temperatures for
the plant observed are also shown.
Each constructor of sulfuric acid plants employs individual
details of design for the converter. These variations in
converter design cause small variations in the efficiency of
the catalyst and require different amounts of catalyst to
obtain the appropriate conversion efficiencies. Some of
the important design details in catalytic converters are the
116
-------
features reducing the wall effect or channeling of cj2 conversion. Improved efficiency is also ob-
tained by more careful control of temperatures in the con-
verter and the use of higher catalysts loadings. Most of the
process parameters in the dual absorption plant affect SD2
emissions in the same manner as for the single absorption
plant.
The use of vent gas cleaning systems in conjunction with single
absorption sulfuric acid plants provides an alternate to the
dual absorption sulfuric acid plants for reducing S02 emis-
sions. Under normal steady-state operations a concentration
of 250 parts per million of SO2 is obtainable using the sodium
scrubber system (Table 8, Page 66). Concentrations as low
as 200 parts per million have been demonstrated by the ammonia
scrubber system (Table 16, Page 93) . The most efficient vent
gas cleaning system studied is the molecular sieve systen>.
This system will produce normal operating SO2 concentration
of less than 10 parts per million.
117
-------
With any vent gas cleaning system, the concentration of SO-
in the gas leaving the single absorption plant absorber in-
fluences the concentration of SO2 in the vent gas from the
cleaning system. However, this influence is not felt to the
same extent as with a single absorption plant operating with-
out a cleaning system. The buffering effecc of the vent gas
cleaning system somewhat reduces the effect of temperature
control and the catalyst condition in the acid plant. With
a vent gas cleaning system of sufficient capacity the single
absorption plant can be operated up to 100 percent of the
original capacity without excessi'* • SO, emissions.
Operating parameters in the cleaning system, however, often
strongly influence the concentration of SO- in the vent gas
and become the primary parameters that must be controlled
to maintain the concentrations within standards. In the
sodium and ammonia alkali scrubbers, the most improtant pro-
cess parameters are concentration and pH of the scrubbing
fluid, temperature, and the liquid/gas ratio flowing through
the absorber. These control points must be established for
a specific scrubber design and must be maintained within
narrow limits to insure proper SO2 absorption.
The molecular sieve system is less sensitive to both SO2 con-
centration in the input and other operating parameters. Suf-
ficient detail is not available on the operating conditions
of the system for a full analysis of the effect of process
parameters. Because of the simplicity of the system the pri-
118
-------
mary control of adsorption efficiency is the condition of
the.molecular sieve packing in the adsorption tower.
Acid mist resulting from carry-over of sulfuric acid from
adsorption towers is controlled primarily by the installation
of demisters in the outlet section of the absorber. In the
case of the double absorption plant a demister is installed
in the top of the primary absorber to reduce the carry-over
of sulfuric acid into the converter bed and to protect the
heat exhangers following the absorber. When properly equipped
with demisters and operated to minimize generation of acid
Tfe- - :
mist in the unit, the normal dual absorption acid plant will
emit between three and 15 milligrams of sulfuric acid per
standard cubic foot of vent gas. These concentration levels
apply when burning clean elemental sulfur as feedstock. If
dirty sulfur containing organics or acid sludges are burned,
additional acid mist is generated within the process, seen
in Figure 18, Page 97 of Section VI. Part of the acid mist
will carry through the demisters, increasing the concentration
of acid mist emitted. Most sulfuric acid plants constructed
during the last ten years are equipped with appropriate de-
misters for controlling acid mist emissions. However some
older single absorption plants still in operation cannot meet
the standards because of lack of appropriate mist eliminator
installation.
The liquid alkali vent gas cleaning systems cannot of them-
selves remove sufficient quantities of acid mist to permit
119
-------
standards to be met. When sodium or ammonia scrubbing systems
are installed, demisters capable of roducinc the acid mist
carry over to within standards, must be installed before or
after the scrubbers. In some cases a high energy venturi
scrubber is operated in conjunction with the S02 vent gas ab-
sorber to eliminate the acid mist emissions.
Adsorption of acid mist as well as SO- gas is inherent in the
nature of the molecular sieve vent gas cleaning system. With
this system it is claimed no additional mist elimination equip-
ment is required; however, data on the one plant evaluated to
date indicate that the acid mist concentration before the
molecular sieve adsorber will meet the standards without
further cleanup and thus the claim cannot be verified as yet.
Since opacity in sulfuric acid plant vent stacks results pri-
marily from acid mist, any of the previously discussed remedies
for reducing acid mist will also reduce opacity. In addition
to acid mist, NOV is also a contributor to opacity. The pre-
ii
sence of NO in the absorber feed will not only add a yellow
Ji
color to the vent plume but will generate acid mist by forma-
tion of a complex between NO and-^O- as in the chamber pro-
X »
cess. Proper operation of the furnace can reduce the NOx
generated from this source. However, if sulfur containing
organics must be burned, more efficient filters are needed to
remove the small sized particulate formed by reaction of water
vapor and SO-. Addition of high efficiency filters after the
absorber of a single absorption plant will permit burning
120
-------
less expensive dark sulfur without excessive opacity.
Feedstock Changes
The effect of sulfur containing organic material has previously
been discussed. In plants burning sludge acid a large variety
of contaminants are encountered in the feedstock, making con-
trol of emissions from the plant difficult. Sulfur dioxide
emissions vary with such feedstock because of the variability
in the concentration of S02 fed to the. converter. Sudden
increases in the inlet SO^ concentration increases the outlet
SQ2« Because the operation of the sulfur furnace is nore
difficult, burning sludge acids can cause upset in the temp-
eratures in the furnace, resulting in sublimation of sulfur
i
and eventual deposition on the catalyst. A return to normal
operating conditions will then oxidize the deposited sulfur
greatly increasing the S02 load in the converter. This results
in large concentrations of SO2 over a short period of time
in the plant vent stack.
With fresh catalyst in a dual absorption plant the design sul-
fur feed rate should not cause the S02 concentration in the
stack to exceed the design limits. However as catalyst
deteriorates or becomes dirty, feed rates must be reduced
to maintain the acceptable concentration of SO2 in the vent.
Some modern plants are designed with sufficient catalyst to
permit feed rates in excess of the design quantity when the
catalyst is new and has maximum activity. The intention of
the excess catalyst in such a plant is to provide for some
121
-------
deterioration of catalyst activity without exceeding emission
standards at designed operating rates. Every effort must be
made to control the feed rate and composition of the burner
exit gas and conditions in the sulfur furnace as consistently
a
-------
tank levels are used, arrangements must be made to isolate
the tank from other product inputs or products outputs during
the period of the test. If storage capacity is limited this
isolation requirement may be a disadvantage in using this
method of production measurement. If a flow meter is avail-
able for continuously measuring the flow of product, the read-
ings from this meter can be used for establishing daily pro-
duction rate. Appropriate calibration procedures must be
followed to assure accuracy of the meter, and periodic test
must be run to prove this accuracy. With either production
rate measurement method, analysis of the product must be per-
formed on a periodic basis to permit calculation of production
on the 100 percent sulfuric acid basis.
During the initial performance test of a new acid plant, stack
y.T- analysis is normally run using the stan^.-rd EPA impinge-
ment train methods to establish the gas :<_-.sity, gas velocity,
and concentration of SCK and acid mist. In order for this
test to be performed, proper sample points must be installed
in the stack in compliance with the regulation. Some acid
plant desiqns have provided stacks of insufficient neight to
obtain the? necessary lengths of straight run before and after
the sample points to permit compliance with this test proce-
dure. Full compliance with sample point configuration must
be assured for tho performance test or any follow up testing
that must be clone usinq tne standard EPA methods.
The regulations also require a continuous SO2 monitor installed
123
-------
on the vent stack. The instrument most commonly applied at
this time is the DuPont model 460 analyzer. Data obtained
from plants using this instrument indicate that the S02 con-
centration readings are reliable if the instrument is proper-
ly maintained. Improper installation and maintenance of this
instrument.however, can cause erroneous results, generally
in the direction of low concentration measurements. If main-
tenance records are kept on this instrument a better evalua-
tion of its performance can be obtained.
A variety of instruments using the principle of light trans-
mission have been used for measurement of opacity in sulfuric
acid vent stacks. These instruments generally appear to have
difficulty in maintaining reliable operation, and visual com-
parison of the opacity of the vent gases is normally relied
upon. Since visual opacity estimates include water vapor
contained in the stack gases, complete analysis of the gases
including sulfuric acid mist must be available before corrected
opacity measurements can be obtained.
/
The only reliable methods available for obtaining acid mist
concentration in the vent stacks are the impinger analysis
methods. Since these test are difficult and time consuming,
requiring two hours to run, they are not normally run in rou-
tine operation. Most plant operators will use visual measure-
ment of opacity as an indicator of possible problems with acid
mist emissions. When visual opacity estimates indicate the
possibility of the presence of excess acid mist the impinger
124
-------
tests are run on the stack to provide information necessary
for correcting the problem. Impinger tests must also be run
during any period of special performance test and evaluation
required by EPA.
SHUTDOWN EMISSIONS
Most plant shutdowns can be classified as either planned or
unplanned shutdowns. The shutdowns resulting from catastrophic
failure are beyond our concern :'.n this study. Shutdowns can
further be classified by short term or long term depending
upon the length of time the plant is expected to be shut down.
For purposes of our discussions a short term shutdown can be
considered as one lasting less than four hours and a long term
shutdown as one lasting more than four hours. When a planned
shutdown is properly handled no excess emissions should be ex-
perienced from the sulfuric acid plant. The primary importance
of shutdown method used is the effect of the shutdown on the
subsequent start-up emissions. A properly prepared plant,
shutdown for four hours or less, can be restarted without re-
heating the plant equipment. Figure 12, Page 76 of Section
VII shows SC- emissions from a dual absorption plant during
start-up after shutdown of various durations. If the shut-
down extends for more than four hours, reheat of plant equip-
ment is required to minimize emissions on start-up.
Planned Shutdown
The most important feature of a planned shutdown is prepara-
tion of the plant for eventual start-up to minimize start-up
125
-------
emissions. The first action in preparing for a planned shut-
down is to have available a supply of strong acid for a transfer to
the absorber and the drying tower. This increase in acid
strength will help to offset the effect of moisture leakage
into the plant and the accumulation of excess moisture result-
ing in dilution of the acid during the start-up operation.
Stronger acid in the absorber and drying tower during start-
up will reduce the acid mist emissions from the plant.
After the plant is prepared for shutdown by increasing the
concentration of the acid, the sulfur feed to the furnace is
simply shut off and the generation of sulfur dioxide stopped.
After a very few minutes of continued air flow to remove ex-
cess SO2 from the furnace and move SO-j from the converter
to the absorber the air supply is shutoff and the dampers iso-
lating the heat exchangers and converter are closed. The
closed dampers will help to reduce the influx of humid air
and conserve heat contained in the equipment. With normal
heat losses from the plant, four hours downtime should not
reduce the temperature below the point for start-up without
preheat.
If the plant shutdown will extend for a prolonged period and
maintenance must be performed on the process equipment, the
unit must be cooled sufficiently to permit entrance. With
this type of shutdown, air flow through the system is continued
until all SO, is transported from the furnace to the converter
and all SO, is removed from the converter to the absorber to
126
-------
make sulfuric acid. After most of the SO3 has bean stripped
from the catalyst and transported to the absorber, the drying
tower is bypassed to prevent dilution of the drying acid that
would make start-up more difficult. The undried air contain-
ing very small quantities of SO-j is vented to the atmosphere
until the converter temperatures approach a dew point of ap-
proximately 100C. At this temperature dry air is again in-
troduced to the system to prevent condensation of moisture
and formation of sulfuric acid in the equipment, causing ex-
cess corrosion. The dried air containing the residual amounts
of SOj is ac^ain vented to the atmosphere until the converter
has reached the desired temperature. If only the furnace and
the waste heat boiler must be cooled, undried air can be used
for cooling if it is vented before entering the converter.
Unplanned Shutdown and Equipment Malfunctions
Many problems can arise in a contact sulfuric acid plant that
cause emissions in excess of -standards arid often require an
unplanned shutdown. Only a few of the major causes of shut-
down are discussed here. Related shutdown causes can be
grouped together in the following sections:
1. Sulfur feed system (Figure 24)
2. Combustion air (Figure 24)
3. Heat exchanger and converter temperature control (Figure 24)
4. Absorber and strong acid systems (Figure 24)
5. Vent gas cleanup systems (Figures 20, 21, 22, and 23)
127
-------
One of the most frequent causes for unplanned shutdown is
failure of the molten sulfur pumps and burners. When sulfur
pumps or burners are lost and the S02 to the converter is
reduced, excess emissions froui the absorber do not necessarily
occur. If the flow of sulfur can be reinstated before the
converter temperatures have dropped below the acceptable range,
the SO2 converter efficiency can be maintained and SO- emis-
sions controlled within limits. However if the converter
temperature drops below 750F, the converter efficiencies
decrease as the temperatures decrease and emissions increase
when flow of molten sulfur is reinstated. A complete sequence
of plant operation following loss of sulfur feed is shown in
Figure 24, Page 108 of Section VI.
The combustion air blower in a sulfuric acid plant is one of
the most critical items in the plant. This blower is normally
driven by a steam turbine supplied with steam from the waste
heat boiler. The speed of the air blower is controlled to
provide sufficient air for com.-lete combustion of the sulfur
to sulfur dioxide and dilution air for subsequent catalytic
oxidation of the sulfur dioxide to sulfur trioxide. If in-
sufficient air is provided by the blower because of malfunc-
tions of the blower control system or operation below the
normal operating speed, incomplete conversion of S02 to SO^
will take place in the converter. Failure to supply sufficient
air to the system is probably the most frequent cause of ex-
cessive SO2 emissions in sulfuric acid.plants. When the air
supply is dropped further, insufficient air is available for
128
-------
the combustion of sulfur to SO2. In this situation the high
temperatures in the sulfur furnace cause sublimation of sul-
fur and subsequent redepositing of sulfur in the catalytic
converter.
Since the deposition of sulfur in the converter takes place
during periods of low temperatures, resulting from insuffi-
cient sulfur combustion, a return to normal operating tempera-
tures will oxidize the sulfur providing a large increase in
the concentration of sulfur dioxide in the converter. This
situation results in a large increase in SO. emissions from
the plant. An added danger of subliming sulfur in the fur-
nace is the deposition of the sulfur on boiler tubes causing
failure of the tubes. Failure is caused by the burning
&f sulfur on the tube surface, resulting in localized
melting of the tubes.
If the combustion air blower shuts down suddenly removing
the flow of air to the furnace, no transfer medium will be
available for sweeping the excess S02 and SO3 from the system.
In these situations higher levels of emissions will occur during
the subsequent start-up operation.
The control of acid mist emissions depends upon feeding dry
air to the sulfuric acid system to prevent reaction of SO3
and water vapor to form the acid mist. If the concentration
of acid in the drying to^er drops below the operating limits,
or if the quantity is reduced below that required by the air
129
-------
passing through the tower, excessive acid mist emissions will
occur. This situation is usually detected first by the appear-
ance of a heavy white plume from the vent stack.
Heat exchangers in sulfuric acid plants are subject to excess-
sive corrosion because of the presence of sulfuric acid and
high temperatures. If a leak occurs in a water cooled heat
exchanger the addition of water to the S02~S03 system results
in a dense white plume of steam and sulfuric acid emerging
from the vent stack. This situation will require a plant
shutdown and excessive acid mist emissions will occur during
the period of operation while the plant is being shutdown.
In a dual absorption plant using gas-to-gas heat exchangers,
a leak in a heat exchanger can result in gas with a high con-
centration of SO- by-passing a portion of the converter
catalyst beds. When this occurs the third or fourth catalyst
bed can be overloaded with SO- and the concentration of SO2
from the vent stack will increase. With this type of failure
the plant must be shut down for repairs and excessive SO2
will occur until the plant is completely purged of SO2.
Failure of temperature control systems for the converter is
one of the most minor malfunctions in terms of hardware but
can be a major cause of excessive SO2 emissions. Many of
the older plants relied heavily on manually adjusted dampers
bypassing the heat exchangers to regulate the inlet and outlet
temperatures of the catalyst beds. Most modern acid plants
130
-------
depend more heavily on automatic control, introducing the pos-
sibility of temperature control failure. In most cases the
significant temperatures in the converter are recorded on a
multipoint recorder that gives a profile of the temperatures
across the converter beds, providing a powerful tool for trouble
shooting temperature control problems. If such a failure occurs
manual control can usually be assumed to return the temperatures
to proper operating points. Emissions resulting from this
type of failure should not last long since the cause is easily
corrected.
Plant shutdowns resulting from failures in the absorber system
generally will arise from failure of the concentrated acid
pumps supplying scrubbing acid to the absorber. If the con-
centration of the strong acid flowing to the absorber is re-
duced below operating requirements, SO, gas will be emitted
from the absorber because of the inability of the lower con-
centration acid to absorb the SO.,. This reduced absorption
results from the increase in surface tension of the low con-
centration acid, reducing the transfer of SO^ into the absorber
liquid. If the concentration of the absorber acid becomes
too high the emissions of SO3 will also increase because the
reduced solubility of SO., in acid with concentration above
99.2 percent will increase the amount of S02 emitted. The
effect of absorber acid concentration on SO emissions is
shown very graphically in Figure 24 of Section VII for both
the primary and secondary absorber. Temperature of the ab-
131
-------
sorbing acid must be maintained between 185F and 212F to
maintain the vapor pressure of the sulfuric acid sufficently
low to permit absorption of the SO.,.
Efficient absorption of S03 and sulfuric acid in the absorp-
tion tower requires good distribution of the acid flow down
the tower and intimate contact with the SO., gases flowing
up the column. If the volume of acid fed to the column is
reduced, distribution of liquid in the column will be uneven
and incomplete absorption results. Also, there is a risk of
increasing the acid concentration in the tower beyond the
99.2 percent, thus producing SO., emissions. Complete failure
of the concentrated acid pump feeding the absorber will
result in immediate shutdown of the plant and large concen-
trations of SO, emissions for a short period of time until
the S0_ content of the plant has been vented.
All modern sulfuric acid plant absorption towers are equipped
with a knitted wire mesh demister in the top of the absorption
tower. In normal operation this demister collects most of
the entrained acid particulates and prevents them from being
carried into the vent gas stream. If this demister becomes
partially plugged for any reason, the velocity of gas flow
through the remaining portion of the demister vrill be too high
to permit collection of the acid mist in the demister. This
acid mist will be carried through the demister and emitted
from the vent stack as entrained 'acid mist.
132
-------
Vent gas cleaning systems are subject to many of the same
problems affecting acid absorbers. The proper liquid/gas
ratio in the absorber must be maintained to assure high ab-
sorption efficiency. If the flow of liquid is reduced or the
flow of gas is increased beyond normal operating ranges, poor
absorption of SO- will take place and SO2 emissions will in-
crease. •"•"•• ^ •: - • ; .-.' • . r • • '
The alkali scrubber systems operate with close pH control on
the absorbing solution flowing down the column. If the con-
trol system fails and the pH is allowed to deviate from the
established range, poor absorption of the SO, from the tail-
gas will result, in the case of the ammonia scrubber an
excessive amount of sub-n.icron particulates of ammonium sul-
fite will be generated, possibly overloading the packed bed
filter (Brink) normally installed to eliminate particulate
emissions from the ammonia scrubber. Plants have been observed
in operation where the Brink separators were plugged or the
booster blowers for transporting the gases through the Brink
separators were inoperative. When the filter system was by-
passed a large cloud of particulate matter was emitted from
the ammonia scrubber. The emission is not classified as acid
mist but is objectionable none the less and must be controlled
to meet opacity standards.
Cooling water is used in a sulfuric acid plant for cooling
product and recirculating acid streams. Failure or cooling
133
-------
water in these systems will permit the acid temperatures to
rise beyond the range of satisfactory operation resulting in
various amoutns of acid vapors being emitted from the pro-
cess. These emissions may occur from storage tanks or other
areas where the hot acid is accumulated rather than from the
vent gas stack.
The failure or impropoer operation of many other small pieces
of equipment will cause similar problems in operation of the
plant increasing emissions. Most of these, however, will
fall into the same categories as those discussed and result
in the same type of situations occurring.
Emissions from Misoperation of Plant
Frequently, high emissions from the acid plants result from
misoperation of the plant. One of the major reasons for
emissions is an attempt to operate the plant at a production
rate higher than permitted b^ the amount or condition of the
catalyst. Full recognition of the condition of the catalyst
and its reduced capacity due to ageing must regulate the amount
of production expected from the plant. If more production is
demanded than is permitted by catalyst condition, excessive
S0_ emissions will occur.
In some cases it is necessary to burn sulfur containing
quantities of organic matter. Burning this sulfur will cause
emissions above the established limits, that may not be con-
trollable with the existing plant equipment under any oper-
ating conditions. The only solution for eliminating these
134
-------
emissions is to prefilter the sulfur to remove the organic
material or to obtain a supply of pure sulfur.
Improper attention to control of catalyst temperatures is
one of the most common causes of excessive S09 emissions,
but also one of the easiest problems to solve. Careful and
constant operator attention to this control area will yreatly
reduce the probability of SO- emissions.
Both laboratory and process instrumentation information is
available to assist in controlling the acid concentration
and temperature in the absorber and drying towers. These
process parameters are normally easy to control with proper
attention and must be kept within proper range if acid mist
standards are to be maintained.
Improper attention to or misadjustment of many of the process
parameters in acid plants or vent gas cleaning systems will
cause one or more of the plant pollutants to exceed standards.
Careful and skillful control of the process at all times is
required to comply with regulations.
START-UP PROCEDURES
Contact sulfuric acid plant start-up can be classified into.
three categories. These are initial plant or new catalyst,
hot start-up aft^r short term shutdown, and cold start-up
after long term shutdown. Each type of start-up will require
a particular start-up procedure to compensate for the condi-
tions existing in the plant. The start-up procedure
135
-------
applied to toe plant will also depend upon the availability of
start-up heaters in the equipment and upon the availability
of an associated sulfuric acid plant to provide the required
heat and concentrated acid.
New Plant and New Catalyst Start-Up
When a new sulfuric acid plant is constructed the catalyst is
received packed in polyethylene bags to prevent adsorption of
excessive atmsopheric moisture. During installation of cata-
lyst in the converter, however, considerable atmospheric
moisture is adsorbed in the catalyst, requiring predrying of
the catalyst bed before introduction of sulfur dioxide into
the converter. During the start-up procedure the catalyst
must also be activated to complete the reaction of the cata-
lyst and sulfur to provide maximum activity.
The initial operation in the start-up procedure * ' is to pass
hot air at approximately 250F through the catalyst bed at
such a rate to permit increase in catalyst temperature at a
rate of 50°F per hour. This air is dried by passing through
the air drying tower. The flo\T of dry air through the cata-
lyst is continued until the temperature has been increased
in the catalyst beds to approximately 357F and the inlet and
outlet bed temperatures are approximately equal, indicating
the catalyst is now dry. After the catalyst is dry, heating
is continued until the first catalyst bed inlet temperature
reaches 825F. At this temperature the catalyst is capable of
converting SO0 to SO,. Sulfur feed to the furnace is started
•w O
136
-------
at a low rate to provide low concentrations of SO- flowing
through the converter. A low concentration of SO- is required
to limit the maximum temperature in the catalyst bed to below
1,OOQF. Temperatures not exceeding 1150F must be maintained
to prevent damage to the catalyst. Control of these maximum
temperatures in the first catalyst stage will provide lower
temperatures in the succeeding stages.
Tho SO2 concentration should be gradually increased, maintain-
ing these maximum temperatures, until full sulfur input rates
are achieved. The total activation period for new catalyst
will range between two and three weeks. The SO- emissions
during catalyst activation should remain below the permissible
concentrations if the inlet SO- is controlled at the appropriate
low concentrations relative to the increasing activity of the
catalyst. Once completed this procedure is not repeated un-
less the converter is opened for catalyst cleaning or addition
of new catalyst, permitting the adsorption of water vapor.
Cold Plant Start-Up - Dual Absorption
The single absorption acid plant cannot obtain erissions within
the permissible limits of the standards in normal operations
without the addition of a vent gas cleaning system. The dual
absorption acid plant, however, can achieve standard emissions
i p \
limits. Discussion with plant engineers during plant surveysv
indicated dual absorption plants can be started from a cold
condition without exceeding standard S02 emissions. To accom-
plish a start-up without exceeding standards, the start-up
137
-------
procedure must be performed very carefully maintaining close
control on all process parameters, it is also necessary to
provide a complete preheat of the converter to obtain maximum
conversion efficiency before S02 is emitted to the system.
Hot concentrated sulfuric acid is required in the drying tower
and absorption tower to limit the amount of acid mist genera-
tion in the system. Although a clean start-up of a cold
plant is possible, it is seldom feasible because of time and
production constraints.
This type of s»-.art-up is best accomplished in a plant associ-
ated with another sulfuric acid plant that can provide hot
concentrated acid for charging the air dryer and absorber.
Although this type of start-up is ideal and has been demon-
strated, a short period of SO- emissions exceeding standards
is more frequently experienced. Routine difficulties in
establishing temperature control around the catalyst beds,
will generally cause the emission of S0_ to exceed the
standards for several hours. Difficulties in controlling
sulfur feed at a satisfactory rate to control temperatures of
the sulfur furnace and concentrations of fiO2 in the converters
will also cause S02 emissions.
A plant that is not equipped with preheaters Must depend upon
the reaction of the S0_ in the converter to heat the converter
to final reaction temperatures. With this arrangement hioh
concentrations of SO2 will be encountered during the rather
long period of time required to heat the converter to peak
138
-------
efficiency. This procedure will usually produce emissions
in excess of standards for two or more days. A cold start-up
procedure for a dual absorption plant is shown in Figure 15,
Page 90 of Section VI. Most new sulfuric acid plants being
designed today are equipped with adequate preheaters.
Table 12, Page 77 of Section VT summarizes start-up data for
plants of all types with various periods of downtime prior
to restarting. These figures clearly illustrate the difficulty
in obtaining low S02 emissions after a shutdown of four hours
or more, requiring reheating of the converter.
Warm Plant Start-Ups - Dual Absorption Process
If a plant shutdown has not extended beyond a four hour period
and proper heat conservation has been practiced in the plant-
catalyst temperatures should be sufficiently high to permit
restarting of plant without reheating. In these situations
restarting the sulfur flow to the furnace and careful read-
justment of dampers to maintain temperatures in the catalyst
converter beds will permit restarting the plants without
exceeding SO, or acid mist standard emissions. The S02 emis-
sions during hot start-ups are summarized in Table 12.
Vent Gas Cleaning Systems - Single Absorption
The star:-up of vent gas cleaning systems applied to a single
absorption acid plant requires very simple operating procedures.
Tha emissions from the vent gas cleaning system will depend
largely upon the start-up procedures used in the single ab-
139
-------
sorption plant, since the outlet SO2 is dependent upon the
concentration cf S02 in the incoming gas to the clean-up
system.
Figures 13 and 14, Pages 83 and 88 of Section VII show this
relationship with an ammonia scrubber. As previously stated,
the vent gas cleaning systems provide a buffer between the
single absorption plant and the vent gas that will reduce
the intensity of concentration swings in S02 emitted from
the single absorption plant. In general terms it appears the
sodium and ammonia scrubbing systems can be started up with
peak SO- emissions one half of those normally encountered
in a single absorption plant (Figure 13, Page 83 of Section
VI).
Start-up of the ammonia and sodium scrubbing systems requires
only the establishment of proper sodium or ammonium concen-
trations and liquid flow rates through the scrubbers and
appropriate pH of the circulating solutions. These condi-
tions are established before starting the sulfur flow to the
furnace. The primary operating precautions required to pre-
vent excessive SO, emissions from the scrubbers are careful
control of temperature, the pH, and concentration of the cir-
culating solutions during the period of increase in S02 con-
centrations from the single absorption plant absorber. Emis-
sions from an ammonia scrubber system during typical start-up
are shown in Figure 14 of Section VI.
140
-------
In an ammonia scrubber system, opacity is most likely to
occur in excess of standards during start-up. The quantity
of the ammonium sulfite particulate generated in an ammonia
scrubber of the types now in use is very sensitive to control
of pH and relative concentrations of NH3 and SO-. It is pos-
sible that the equipment installed to remove this particulate
will be overloaded by the excessive amounts of particulate
formed during the upset condition experienced during the
start-up. In some situations sufficient steam is not avail-
able from waste heat boilers during start-up to operate both
the air blower and the booster blower required to overcome
the pressure drop in the ammonia scrubber particulate filter.
In this situation it is routine practice to bypass the parti-
culate filter so the steam driven booster blower is not required.
When the filter is bypassed large quantities of particulate
matter are emitted from the ammonia scrubber causing a heavy
blue plume far in excess of the opacity standards. These
emissions can largely be eliminated if arrangements are made
to place the particulate filter in operation before SO2 is
introduced to the system.
No data were available on start-up conditions for the Union
Carbide molecular sieve system, but according to literature
previously published*9* this system is very tolerant to large
surges in SO2 and acid mist concentrations from the absorber
and should emit very low levels of these pollutants during a
normal start-up of the single absorption plant equipped with
the PuraSiv system.
141
-------
SECTION VIII
REFERENCES
1. Environmental Protection Agency, Standards of Performance
for New Stationary Sources. (40-CFR - Part 60, Sub Part
H) Federal Register Vol. 36, No. 245, December 23, 1971.
2. Environmental Protection Agency, Standards of Performance
for New Stationary Sources Emissions During Start-Up,
Shutdown, and Malfunction, (40-CFR - Part 60) Federal
Register, Volume 39, No. 39 872, Nov. 12, 1974.
3. Evaluation of H2S04 Plant Emissions at the Coulton Chemical
Corporation which is Controlled by the PuraSiv S Process,
York Research Corp. for Control Systems Laboratory, Office
of Research and Development, EPA Research Triangle Park,
Report No. Y-8479-2, Contract No. 68-02-1401. May 6, 1975.
4. NSPS Compliance Test Report, Feb. 19-20, 1974, EPA Region
IV file D-10.9.3.
5-6 EPA Continuous Source Monitoring Program Test Results,
Files of Performance Standards Branch, Standards Develop-
ment and Implimentation Division, EPA Research Triangle
Park.
7. Continuous Monitoring of a Copper Smelter Double Contact
Process Acid Plant, Scott Research Laboratories Inc. for
EPA Research Triangle Park Report No. SRL 2113 15 0574,
Contract No. 68-02-0233, May 31, 1974.
8. Towse, C. F. Private Communications with F. D. Kodras,
Catalytic, Inc., November 11, 1974.
142
-------
9. Collins, J. J., L. L. Fornoff, K. D. Manchanda, W. C.
Miller, D. C. Lovsll. The PuraSiv S Process for Removing
SO2 from Sulfuric Acid Plant Tail Gas, Union Carbide
Corp. (Presented at the 66th Annual AICHE Meeting,
Philadelphia, November 15, 1973).
10. Private Communication from R. S. Twichill of the Volunteer
Army Ammunition Plant, Chattanooga, Tennessee with
F. D. Kodras
143
-------
SECTION IX
GLOSSARY OF TERMS, ABBREVIATIONS,
AND SYMBOLS
The following listing is a glossary of terms and abbre-
viations that are used in this report. Also included is a
list of chemical symbols that are found in the report.
GLOSSARY OF TERMS
Absorber
A device utilized to extract selectively one or more
elements of a gas stream from others by absorption in a
liquid medium. Usually the process is performed in cylindri-
cal towers packed with an inert material thus providing a
large surface area for intimate contact between the rising
gas and the falling liquid. (The process may also be carried
out in a tower containing perforated trays in which the rising
gas bubbles through the layer of liquid on the trays.)
Absorption
A process in which one or more constituents are removed
from a gas stream by dissolving them in a selective liquid
solvent. This may or may not involve a chemical change.
Acid, New
Sulfuric acid made from elemental sulfur or other sul-
fur-bearing materials, but not from spent acid strengthened
by addition of sulfur trioxide.
144
-------
Acid, Regenerated
High-purity sulfuric acid made from decomposition or
regeneration of spent acid from petroleum refineries or other
chemical processes.
Acid Sludge
The residue left after treating petroleum oil with sul-
furic acid for the removal of impurities. It is a black,
viscous substance containing the spent acid and impurities
which have been separated from the oil.
Adsorption
A reaction in which one or more constituents (adsorbates)
are removed from a gas stream by contacting and adhering to
the surface of a solid (adsorbent). Periodically the adsor-
bent must be regenerated to remove the adsorbate.
Aerosol
A colloidal system in which particles of solid or liquid
are suspended in a gas. There is no clear-cut upper limit
to the particle size of the dispersed phase in an aerosol,
but as in all other collodial systems, it is commonly set
at 1 micro-meter. Haze, most smoke, and some fogs and clouds
may be regarded as aerosols.
Agglomeration
Groups of fine particles clinging together to form. a
larger particle.
Ambient Air
That portion of the atmosphere, external to buildings,
to which the general public has access.
145
-------
Boiler
A closed pressure vessel in which the liquid, usually
water, is vaporized by the application of heat.
British Thermal Unit (Btu)
The mean British thermal unit is 1/180 of the heat required
to raise the temperature of one pound of water from 32°F to
212°F at a constant atmospheric pressure. It is about equal
to the quantity of heat required to raise one pound of water
1°F. A Btu is essentially 252 calories.
Burner
A device for the introduction of fuel and air into a fur-
nace at the desired velocities, turbulence, and concentration
to establish and maintain proper ignition and combustion of
the fuel.
Carryover
The chemical solids and liquid entrained in the steam
from a boiler or effluent from a fractionating column, absorber,
or reaction vessel.
Catalyst
A substance capable of changing the rate of a reaction
without itself undergoing any net change.
The catalyst in a chamber plant is gaseous nitrogen oxides.
In the contact process the catalyst is a solid, consisting of
vanadium pentoxide and various promoters deposited on a highly
porous siliceous carrier.
Glaus Elemental Sulfur Plant^
A process for converting S02 to elemental sulfur by
catalytic reaction with H2S.
146
-------
Condensed Fumes
Minute solid particles generated by the condensation of
vapors from solid matter after volatilization from the molten
state, or generated by sublimation, distillation, calcination,
or chemical reaction when these processes create airborne par-
ticles.
Converter
The vessel that houses the solid vanadium catalyst. The
i
catalyst is placed in several horizontal trays or stages located
in series, with means for qas cooling between the various stages.
Demister (Collector)
1. A mechanical device used to eliminate finely divided
liquid particles from process streams by impaction and agglom-
eration.
2. Apparatus made of wire mesh or glass fiber and used
to eliminate acid mist as in the manufacture of sulfuric acid.
Economizer
•
A heat recovery device designed to transfer heat from the
products of combustion to a fluid, usually feedwater for a
steam boiler. The water flows through a bank of tubes placed
across the flue gases and is heated by these gases prior to
entering the boiler.
Effluent
Any waste material (solid, liquid, gas) emitted by a pro-
cess.
147
-------
Emission
The total amount of a solid, liquid, or gaseous pollutant
emitted into the atmosphere from a given source in a given
time, and indicated in grams per cubic meter of gas, pounds
per hour, or other quantitative measurement.
f t
Entrainment
The process of particulates or other materials being
carried along by a gas stream.
Excess Air
Air supplied for combustion in excess of that theoretically
required for complete combustion, usually expresses as a per-
centage of theoretical air, such as "130 percent excess air."
Exothermic Reaction
A -reaction which produces heat.
Feedstock
Starting material used in a process. This may be raw
material or an intermediate product that will undergo additional
processing.
Flue
Any duct, passage, or conduit through which the products
of combustion are carried to a stack or chimney (see also
breeching).
Flue Gas
The gaseous products of combustion passing from the fur-
nace into the stack.
Fuel
Any form of combustible matter—solid, liquid, vapor, or
gas, excluding combustible refuse.
148
-------
Fume
Fine solid particles predominately less than 1 micro-meter
in diameter suspended in a gas. Usually formed from high-
temperature votilization of metals, or by chemical reaction.
Impingement
In air sampling, impingement refers to a process for the
collection of particulate matter in which the gas being sampled
is directed forcibly against a surface. 1. Dry impingement:
the process of impingement in the gas stream where particulate
matter is retained upon the surface against which the stream
is directed. The collecting surface may be treated with a
film of adhesive. 2. Wet impingement: the process of im-
pingement in a liquid which retains the particulate matter.
Impingeme.it Separators
Devices using the principle that when a gas stream carry-
ing particulate matter impinges on a body, the gas is deflected
around the body, while the particles, because of their greater
inertia, tend to strike the body and be collected on its sur-
face. The bodies may be in the form of plates, cylinders,
ribbons, or spheres.
Material Balance
An accourting of the weights of material entering and
leaving a process.
149
-------
Mosh " " "
The number of holes per linear unit in a sieve or gauze,
or the space between the wires of the sieve expressed in inches
or millimeters.
Metric Ton
2204.6 pounds or 1000 kilograms.
Mist
A suspension of any finely divided liquid in a gas.
Mist, Sulfuric Acid
Extremely small acid particles that are true aerosols.
No exact range of particle size is available.
Net Ton
2000 pounds (sometimes known as a "short ton").
Nitrogen Oxides
A general term pertaining to a mixture of nitric oxide
(NO) and nitrogen dioxide (N02>.
Olaum (Fuming Sulfuric Acid)
A heavy, oily, strongly corrosive liquid that consists
of a solution of sulfur trioxide in anhydrous sulfuric acid.
It fumes in moist air and reacts violently with water.
Onstream Time
The length of time a unit is in actual production.
Opacity
The degree to which emissions reduce the transmission of
light and obscure the view of a distant object.
150
-------
Orsat
An apparatus used for analyzing flue gases volumetrically.
Oxidation
The act or process of combining oxygen with a substance,
with or without the production of a flame.
Oxides, Sulfur
As used in this report, sulfur oxides include sulfur diox-
ide and sulfur trioxide, and/or sulfuric acid mist or spray.
Packed Column (Packed Scrubber or Packed Tower)
A vertical column used for distillation, absorption, and
extraction, containing packing; e.g., Raschig rings, Berl
saddles, or crushed rock, which provide a large contacting
surface area between phases. Normally, gas flow is counter-
current to liquid flow.
Particulate Matter
*•
Any dispersed matter, solid or liquid, in which the indi-
vidual aggregates are larger than single small molecules (0.0002
micro-meters) but smaller than 500 micro-meters.
Performance Test
Measurements of emissions used for the purpose of determin-
ing compliance with a standard of performance.
Plume
The path taken by the continuous discharges of products
from a chimney or stack. The shape of the path and the con-
centration distribution of gas plumes is dependent on turbu-
lence of the atmosphere.
151
-------
primary Air
In incineration, air which is introduced with the refuse
into the primary chamber.
Primary Standard
The national primary ambient air quality standard which
defines levels of air quality which are necessary to protect
public health.
Ringelmann Chart
A standardized chart giving shades of gray by which the
densities of columns of smoke rising from stack may be com-
pared .
Scrubber
A device used to remove entrained liquids and solids from
a gas stream by parsing the gas through wetted "packing" or
spray (see absorber).
Secondary Air
Air introduced into a combustion chamber beyond the point
of fuel and primary air introduction for the purpose of achieving
more complete oxidation.
Smelting
Any metallurgical operation in which metal is separated
by fusion from impurities with which it may be chemically com-
bined or physically mixed, such as in ores.
SO2 Gas or 803 r.as
A gas in which S02 or SO3 is present with other constituents
such as oxygen or nitrogen.
152
-------
spray, Sulfuric Acid
Large acid particles introduced into the gas by mechanical
entrainment. If emitted to atmosphere, they are invisible and
fall rapidly to the ground.
Stack or Chimney
Any flue, conduit, or duct arranged to conduct an effluent
to the open air.
Sulfur, Crude
A low-sulfur-content raw material consisting of a mixture
of elemental sulfur and inert material.
Sulfur, Elemental
Any sulfur in elemental form, regardless of source.
Sulfur, Recovered
An extremely high-purity sulfur containing no organic
matter, less than 0.005 percent ash, and no free acid or water
unless exposed to the atmosphere. (If shipped molten, it may
also contain traces of hydrogen sulfide.)
Surge Tank
A storage reservoir at the downstream end of a feeder
pipe to absorb sudden rises of pressure and to furnish liquid
quickly during a drop in pressure.
Tail Gas
The exhaust or waste gas from a process.
153
-------
Vapor
The gaseous phase of a substance that generally exists as
a liquid or solid at room temperature.
Vapor Plume
The stack effluent consisting of flue gas made visible by
condensed water droplets or mist.
venturi Scrubber
A type of high energy scrubber in which the wasta gases
pass through a tapered restriction (venturi) and impact with
low-pressure water. Gas velocities at the restriction are
from 15,000 to 20,000 fpra and pressure drops from 10 to 70
inches water gage.
Visible Emission
An emission of air pollutants greater than 5 percent
opacity of 1/4 Ringelmann.
Waste Heat Boilers
Boilers which utilize the heat of exhaust gas or process
gas to generate steam or to heat water.
154
-------
ABBREVIATIONS
o
C temperature, degrees Celsius
ft feet
2
ft square feet
ft , cf cub?c feet
°F temperature, degree Fahrenheit
Ibs pounds
Mscfm thousands of standard cubic feet per
minute
ppm parts per million by volume
psig pounds per square inch gauge
PSIA pounds per square inch absolute
scfm cubic feet per minute measured at
standard conditions: (70°F and 760
mm (29.92") Hg
SCFH standard cubic feet per hour
sp. gr. specific gravity—compared to water at
60°F
155
-------
CHEMICAL SYMBOLS
CO,
H20
NH-,
SO,
SO-
H2S04
NaHSO
Na2S04
(NH4)2S04
carbon dioxide
water
ammonia
nitrogen
total nitrogen oxides in a mixture
oxygen
sulfur
sulfur dioxide
Sulfur trioxide
sulfuric acid
sodium sulfite
sodium bisulfite
sodium sulfate
ammonium sulfate
ammonium bisulfite
ammonium sulfite
156
-------
UNITS OF MEASURE
Conversion Table
Degrees Fahrenheit (°F) = 9/5 Degrees Celsius (°C) + 32
Inches (in) = 215 Centimeters (cm)
Feet (ft) = 30 Centermeters (cm)
2 2
Square Inches (in ) =6.5 Square Centimeters (cm )
Cubic Feet (ft ) (cf) = .03 Cubic Meters (m )
Pounds (Ib) = .45 Kilograms (kg)
Ton (short) = -9 Tonnes (t)
Gallons (gal) =3.8 Liters (1)
Pounds Per Square Inch (psi) = .06 Kilograms Per Square Centi-
meter (kg/cm2)
157
-------
APPENDIX A
..-'--"'• •'
PROCESS DESCRIPTIONS
SULFURIC ACID PRODUCTION
AND
TAIL GAS CLEANING PROCESSES
Table of Contents
Page
Theroretical Operating Data for Sulfuric Acid Plants . . 159
Single Absorption Process for Sulfuric Acid
Production 162
Dual Absorption Process for Sulfuric Acid Production . . 169
Sodium Scrubbina Process for Sulfuric Acid Tail Gas
Cleaning . . \ 176
Ammonia Scrubbing Process fo.: Sulfuric Acid Tail Gas
Cleaning 184
Molecular Sieve Process for Sulfuric Tail Gas
Cleaning 191
158
-------
TABLE 17
OPERATING DATA FOR THE SULPURIC ACID PLANT (THEORETICAL)
Pressure, in. H20
vocation in
acid plant
Enter blower
Enter dry tower
Leave dry tower
Enter sulfur burner
Leave sulfur burner
Leave boiler #1
Enter filter
Leave filter
Enter pass #1
t->
(j\
vo Leav<_ pass f*l
Air enter pass #2
Leave pas? .S.2
Air entet pass *3
Leave pass S3
Enter pass #4
Leave pass #4
Enter economizer
Leave economizer
Enter 98 absorber
Leave 98 absorber
Pressure drop (knock out)
Pressure drop demister
Leave burner
Leave boiler
Enter hot gas filter
Enter "A" layer #1
Leave "A" layer SI
316 tons/day
-5
95
89
84
78
76
67
66
60
60
56
46
37
26
29
28
19
16
7
6
4
1
3
3/4
1/2
1/2
1/4
3/4
3/4
3/4
1/4
1/4
1/2
1/4
1/4
1/4
3/4
3/4
1/2
3/4
3/8
150 tons/day
-1
28
26
25
24
24
22
22
19
17
15
12
12
9
9
4
3
1
—
—
—
1/2
1/2
3/4
3/4
1/4
1/4
3/4
3/4
3/4
1/4
3/4
1/2
1/4
1/4
1/2
Temperature, °C
316 tons/day
0-(32°F)
29.0-(85)
60-(35)
972-U782)
424-(795)
434-(813)
—
426-(799)
476-(889)
—
612-U134)
450-(842)
510- (750)
439-(822)
447-(837)
422-(792)
426- (799)
— —
259-(498)
240-(464)
—
—
—
971-U780)
423-(793)
434-(813)
425- (797)
476-(889)
150 tons/day
14-57°F
948-U738)
293-(559)
427-(80l)
—
405- (761)
476- (889)
__
603-(1117)
440-(824)
491-(916)
434-(813)
439-(822)
401-(754)
400-(152)
—
198-(3S8)
—
—
—
—
948-U738)
294-(561)
430-(806)
405-(761)*
476-(889)*
*See Table 5
-------
TABLE 17 (Continued)
Location in
acid plant
Leave "B" layer U
Enter layer #2
Leave layer #2
Enter layer $3
Leave layer $3
Enter layer #4
Leave layer #4
Leave economizer gas
Pressure, in. H^O Temperature
316 tons/day 150 tons/Jay 316 tons/day
612-U134)
438-(820)
500- (932)
439- (822)
448-(838)
423-(793)
427-(801)
259-(498)
, °C
150 tons/day
603-U117)*
440-(825)
491-(916)
434-(813)
439-(822)
401-(?54)
400-(752)
198-(388)
2 *See Table 5.
Parameter
Blower rpm
% S02 after boiler #1
% SO- into reactor
% S02 out stack
% Conversion i
Stack appearance
Boiler #1 bypass valve
Boiler #2 bypass valve
Superheater #1 valve
Superheater #2 valve
Saturated stean bypass
Sulfur consumption
Production rate @ 100% H2S04
316 tons/day
4225
10.3
9.5
0.25
97.55
good
28° open
closed
full open
1/2 open
closed
212,550 Ib/day
632,250 Ib/day
150 tons/day
2400
10.3
8.5
0.19
98.0
good
32° open
3/4 open
full open
1/2 open
closed
2 Ib/hr
-------
TABLE 17 (Continued)
Parameter
Temperature of sulfur in pit, °C
Temperature of sulfur entering burner, °C
Saturated steam, psig
Superheated steam, psig
Superheated steam, °C
110 psig steam to plant, °C
Rate of flow to plant, Ibs/hr
Feed water entering economizer
Feed water leaving economizer
Temperature of acid entering acid absorber, °C
Temperature of acid leaving acid absorber, °C
Strength leaving acid absorber
Acid level tank-o-meter , in.
Temperature of acid entering dry tower, °C
Temperature of acid leaving dry tower, °C
Acid level tank-o-meter, in.
Temperature of acid leaving pumps, °C
(66°Be')
316 tons/day
140°-(284)
140°-(284)
338
322
307°-(585)
254°-(489)
31,800
105°-(221)
208°-(406)
82-(180)
120-(248)
99.17%
8 1/4
68-(154)
66-U51)
7-{45)
110-(230°F)
150 tons /day
140°-(284)
133°-(271)
325
319
315°-(599)
285°-(545)
15,400
104°-(219)
213°-(415)
80-(116)
91- (198)
99.1%
8.2
60-U40)
—
7-(75)
98-(208)
-------
PART 1
SINGLE ABSORPTION PROCESS
FOR
SULFURIC ACID PRODUCTION
162
-------
MAJOR EQUIPMENT LIST
Figure A-201
SINGLE ABSORPTION SULFURIC ACID PLANT
Equipment Number Description
K-101 Sulfuric Furnace
F-102 Air Filter
C-103 Main Air Blower
V-104 Drying Tower
B-105 Waste Heat Boiler No. 1
X-106 Economizer After Fourth Bed
V-108 Converter with Four Catalyst Beds
V-109 Secondary Absorber Tower
V-110 Secondary Absorber Acid Cooler
B-lll Waste Heat Boiler No. 2
V-112 Secondary Absorber Acid Pump -nd Tank
X-113 Product and Storage Tank
X-114 Product Acid Cooler
X-115 Drying Tower Acid Cooler
X-119 Product Acid Cooler
163
-------
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, SO,
Air
Sulfuric Acid, H2S04
Sulfuric Acid Mist
Water
Total (tons per day)
Temp (°C)
Pressure (inches water)
Corposition
Conversion
Total (SCFM)
Total (gal/min)
Figure A-201
MATERIAL BALANCE SINGLE ABSORPTION PLANT
(1)
Sulfur
Feed
0.331
(2)
Air to
Drying
Tower
2.988
51.9
(3)
Air to
Sulfur
Burner
2.988
0.331
135
0
3
.030
.018
50
85
2.988
50
80
51.4
(4)
Gas Leaving
Sulfur
Bur nor
0.663
2.656
3.319
850
75
10.20 S02 (Volume)
50.9
(5)
Gas
Entering
Converter
0.663
2.656
3.319
410
60
50.9
*Flow Basis: One ton per day 100 percent H2S04 production.
-------
(Continued)
Stream Nc.
Description*
(6)
Gas
Leaving
Converter
0.010
0.816
2.493
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, SO.
Air
ijlfuric Acid, HjSOj
Sulfuric Acid Mist
Water
Total (tons per day)
Trap. (*C)
Pressure (inches water)
Composition
Conversion SB.5% of SOj to
Total (SCFH) 48.0
Total (gal/min)
3.319
445
20
10-f 303 (Volume)
(7)
Gas Entering
Oleum Tower
(8)
Oleum Product
Leaving Oleum Tower
150
10
Depends on SO3 absorbed
251 Oleum
48.0
(9)
Gas Leaving
Oleum Tower
75
3
(10)
Gas Entering
Absorption Tower
(No Oleum Production)
0.010
0.816
2.493
3/319.
150
10
48.0
•Plow Easis: One ton per day 100%
production.
-------
(Continued)
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SO.,
Sulfur Trioxide, S03
Alt
Sulfuric Acid, 82S04
Sulfuric Acid Hist
Hater
Total (toas per day)
Temp. (»C)
Pressure (inches water)
Composition
Conversion
Total (SCFM)
Total (gal/min)
(11) (12)
Acid Product Gas Leaving
Absorption
Tower
.010'
-
2.493
1.000
.020 2.503
1.020
70 75
3
3995 PP» SO2
43.04
.091
(13)
Hater for K2SO4
Absorption For
Absorption Tower
.183
.183
25
(14)
H.so. for
Oleum Tower
(15)
Circulation
to Absorption
Tover
(16)
Circulation to
Drying Tower
70
10.000
.200
10.200
70
9.03*
9.0S4
50
.031
.994
.AZ2
•Flow Basist One ton per day 100%
production.
-------
! t i • • {. i*«ft
« ".....L:l..
—~i—ui • TI?.'.- -t-i ;
y- I I h ^n^te
i » I I *cio '/
*—»-»-J r i '-
A iO I
-------
PART 2
DUAL ABSORPTION PROCESS
FOR
SULFURIC ACID PRODUCTION
168
-------
MAJOR EQUIPMENT LIST
Figure A-202
DUAL ABSORPTION 3ULFURIC ACID PLANT
Equipment Number Description
K-201 Sulfur Furnace
F-202 Air Filter
C-203 Main Air Blower
V-204 Drying Tower
B-205 Waste Heat Boiler
X-206 Economizer After Fourth Bed
X-207 Economizer After Third Bed
V-205 Converter With Four Catalyst Beds
V-209 Secondary Absorption Tower
X-210 Secondary Absorber Acid Cooler
X-211 No. 1 Heat Exchanger
V--212 Primary Absirber Acid Purcp and Tank
V-213 Product Acid Storage Tank
X-214 Product Acid Cooler
X-215 Drying Tower Acid Cooler
V-216 Drying Tower and Secondary Absorber
Acid Pumps
X-217 No. 2 Heat Exchanger
X-218 Steam Superheater
X-219 Product and Cooler
X-220 Primary Absorber Acid Cooler
V-221 Primary Absorber Tower
169
-------
Stream No.
Description*
Sulfur, S
Sulfur Dioxide/ S02
Sulfur Trioxide, S03
Air
Sulfuric Acid, H2S04
Sulfuric Acid Mist
Water
Total (tons per day)
Temp. (°C)
Pressure (inches water)
Composition
Conversion
Total (SCPM)
Total (gal/min)
Figure A-202
MATERIAL BALANCE DUAL ABSORPTION PLANT
(1)
Sulfur
Feed
.327
(2)
Air to
Drying
Tower
2.345
51.2
(3)
Air to
Sulfur
Burner
2.945
.327
135
0.030
2.945
50
78
2.945
50
76
51.2
(4)
Gas Leaving
Sulfur Burner
.653
2.618
3.271
850
75
10.1 S02 (Volume)
50.2
*Flow Basis: One ton per day 100% HjSO. production,
-------
(Continued)
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SO.,
Sulfur Trioxide, S03
Air
Sulfuric Acid, R2SO4
Sulfuric Acid Mist
Hater
Total (tons per day)
Temp. (°C)
Pressure (inches water)
Composition
Conversion
Total (SCFH)
Total (gal/nin)
(5)
Gas
Entering
Converter
.653
2.618
3.271
410
70
50.2
(6)
Gas Leaving
3rd Stage
of Converter
.021
.790
2.460
3.271
445
55
10.41 803 (Volume)
96.71 Conversion
of SOj to S03
47.4
(7)
Gas
Entering
Oleum Tower
(8)
Oleum Product
Leaving
Oleum Tower
150
44
Depends on S02 absorbed
251 Oleum
•Flow Basis: One ton per day 100% HjSOj production.
-------
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SOj
Sulfur Trioxide, S03
Air
SulCuric Acid, H2SO4
Sulfuric Acid Hist
Water
Total (tons per day)
Temp. (*C)
Pressure (inches water)
Composition
Conversion
Total (5CFM)
Total (gal/min)
Gas Leaving
Oleum
Tover
(Continued)
(10)
Gas Entering Primary
Absorption Tower
(Mo Oleum Production)
.021
.790
2.460
(11)
Acid Product
75
40
3.271
ISO
44
47.4
1.000
1.000
(12)
Gac Leaving
Primary
Absorption Tower
.021
2.460
2.481
75
40
8700 Ppn S02
42.5
0.0908
•Flow Basisi One ton per day 100% HjS04 production.
-------
Stream «o.
Description*
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, SOj
Air
Sulfuric Acid, H2S04
Sulfuric Acid Mist
Water
>
Total (tons per day)
Temp. (*C)
Pressure (inches water)
Composition
Conversion
Total (SCFM)
Total (gal/Din)
(13)
Hater for H,SO.
Absorption for
Absorption Towers
(Continued)
(14)
H,S04 for
Oleum Tower
.184
.184
25
70
H.SO. Circulation
to Primary
Absorption Tower
10.953
10.953
70
(15b)
H,SO. Circulation
to Secondary
Absorption Tower
8.532
8.532
70
0.030
0.994
0.775
•Flow Basisi One ton per day 100% RjSOj production.
-------
(Continued)
Stream Ho.
Description*
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, SOj
Air
Sulfuric Acid, H2S04
Sulfuric Acid Miat
i Hater
Total (tons per day)
Temp. CO
Pressure (inches water)
Composition
Conversion
Total (SCFH)
Total (gal/rain)
(16)
HjSO.
Circulation to
Drying Tower
9.054
(17)
Gas Entering
4th Stage
of Converter
.021
2.460
42.5
(18)
Gas Leaving
4th Stage
of Converter
.0005
.02C
2.435
42.4
(19)
Gas Entering
Secondary
Absorption Tower
.0005.
.026
2.455
42.4
(20)
Tail Gas frcm
Secondary
Absorber
.0005
2.455
9.054
50
2.481
425
40
2.481
430
25
2.481
150
15
2.455
75
3
229' Ppn
42.2
0.822
•Mow Basis: One ton per day 100% HjS04 production.
-------
us
tuuns
-------
PART 3
SODIUM SCRUBBING PROCESS
FOR
SULFURIC ACID TAIL GAS CLEANING
176
-------
MAJOR EQUIPMENT LIST
Figure A-203
SODIUM SCRUBBING PROCESS
Equipment Number Description
B-301 Booster Fan
V-302 Prescrubber
P-303 Prescrubber Circulation Pump
D-304 Demister •.''',.
'••-'f
V-305 SO2 Scrubber
D-306 SO2 Scrubber Demister Pad
P-307 SO2 Scrubber Circulation Pump
V-308 Brine Holding Tank
A-309 Brine Holding Tank Agitator
P-310 Brine Feed Tump to Evaporator -
Crystallizer
V-311 Evaporator - Crystalliser
X-312 Heat Exchanger for Evaporator -
Crystallizer
P-313 Evaporator-Crystallizer P«icirculatJ.on
Pump
P-314 Evaporator-Crystallizer Discharge Pump
S-315 Centrifuge
V-316 Mother Liquor Tank
A-317 Mother Liquor Tank Agitator
P-318 Mother Liquor Discharge Pump
C-319 Conveyor for N32SO, Crystals to
Dissolving Tank
V-320 Dii-«5olving Tank
177
-------
Equipment Number Description
A-321 Dissolving Tank Agitator
P-322 Dissolving Tank Pump
X-323 ' Condenser
V-324 Separator Tank
P-325 Separator Tank Condensate Pump
V-326 50 Percent NaOH Storage Tank
P-327 50 Percent NaOH Pump
V-328 Alkali Make-Up Tank
D-329 Alkali Make-Up Tank Discharge Pump
178
-------
Figure A-203
MATERIAL BALANCE SODIUM SCRUBBING PSQCESS
Stream No. (1) (2) (3)
Description* Tail Gas Entering Cleaned Gas dodium Hydroxide
Packed Tpwar Gas Leaving Packed Make-Up to
Scrubber Tower tfas Scrubber Packed Ga» Scrubber
Sulfur, S
Sulfur Dioxide, SOj 0.00994 O.OOOSO
Sulfur Trioxide, SO.,
Air
Sulfurlc Acid 2.822 2.822
Sulfuric Acid Mis-.
Hater .001316
Sodium Bisulfite, NaHSOj
Sodium Sulfite, Na2SO3
Sodium Sulfate, Na2SO4
Sodium Hydroxide, NaOH .001316
Total (tons per day) 2.83194 2.8225 .002632
Total (SCFM) 51.38 51.30
Votal (Gallons per minute) .000322
Temp. (*C) 75 75 30
Pressure (inches of water)
Composition 1560 ppm 802 86 ^>n ^°2
Conversion . 951 Scrubber Efficiency
•Flow Basis: One ton per day 100% HjSOj production.
-------
(Continued)
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SOj
Sulfur Trioxide, SO.
Air
Sulfuric Acid
Suifuric Acid Mist
Hater
Sodium Bisulfite, NaHS03
oo Sodium Sulfite, Na,SO,
o * •*
Sodium Sulfate, NajSOj
Sodium Hydroxide, NaOR
Total (tons per day)
Total (SCFH)
Total (Gallons per minute)
Temp. («C)
Pressure (inches of water)
Composition
Conversion
(4)
Scrubber Brine
Bleed to Holding
Tank
(S)
Scrubber Brine
Recirculation to
Packed Gaa Scrubber
(6)
Brine Feed from
Holding Tank to
Evaporator-Crys ta11i zer
0.3552
0.01238
0.00323
0.00269
0.05382
0.00896
5.6054
1.9534
0.5096
0.4246
8.493
1.414
0.3552
0.01238
0.00323
0.00269
0.05382
0.00896
(7)
Gas Exiting
the Bvaporator-
Crystallizer
0.00381
0.03810
0.04191
1.0851
1.9851
•flow Basisi One ton per day 1001 HjSOj production.
-------
(Continued)
Stream No.
Description*
Sulfur, s
Sulfur Dioxide, SOj
Sulfur Trioxide, SO}
Air
Sulfuric Acid
Sulfuric Acid Mist
Hater
Sodium Bisulfite, NaHSOj
Sodium Sulfite, NajSOj
Sodium Sulfate, Na-jSO^
Sodium Hydroxide, NaOH
Total (tons per day)
Total (SCFM)
Total (Gallons per minute)
Temp. (8C)
Pressure (inches of water)
Composition
Conversion
(8)
Gas Entering
the Separator
Tank
O.C0381
0.03810
0.04191
1.9851
(9)
SO. Recycle
Gas to
Acid Plant
0.00381
(10)
Condensate from
Separator Tank
Recycled to Dissolving Tank
(11)
Concentrated Liquor
Feed to
Centrifuge
0.03810
0.00381
0.02968.
0.03810
0.00634
0.00152
O.OJ899
0.00465
0.01516
•Plow Bc.sisi One ton per day 100% B2S04 production.
-------
(Continued)
Stream No. (12) (13) (14) (1$) (16)
Description* Na-SO, Crystals Na-SO, Slurry Na-SO, and Na,SO. Na-SO. and Na,SO, Na,SO. Purge
to Dlosolvin? TarJcs Recycle to Slflrry to Notner Recycle to Evaporator
Packed Gas Scrubber Liquor Tank Crystallizer
Sulfur, S
Sulfur Dioxide, SO^
Sulfur-Trioxide, SOj
Mr
Sulfuric Acid
Sulfuric Acid Hiat
Water .000745 .01824 .000771. .000775 .000388
Sodium Bisulfite, NaltSO,
Sodium Sulfite, N»jSO3 .00546 .00546 .00353 .00353 .00177
Sodium Sulfate, Na2S04 0.00465 0.00465 .002:3
Sodium Hydroxide, NaOH
Total (tons per day) ..006205 .0237 .00896 .00b9o .00449
Total (SCFM)
Total (Gallons per minute)
Temp. CO
Pressure (inches of water)
Composition
Conversion
•Flow Basisi One ton per day 100% H-SO. production.
-------
•••?• .^ *'£• ' I
(*, i ft\ "TTlfttM, (."^i""
1 '*"*• -!•»$• A
"i""! f" •" —————— •' a)^, ^
1 ! jacaaft) . ' .'r,;1'
fl f^ll]
— 4 M '•;•••
fi r-;r:J?^__
n _?
«i.i •>
'«**«
j";n'.
-J 3
-All fwr-—L
\ BXB
,/ I"*-:
T
sii'T''1 I
-------
PART 4
AMMONIA SCRUBBING PROCESS
FOR
SULFURIC ACID TAIL GAS CLEANING
184
-------
MAJOR EQUIPMENT LIST
Figure A-204
AMMONIA SCRUBBING PROCESS
Equipment Number Description
V-400 Packed Tower Gas Scrubber - 1
P-401 Ammonia Scrubber Brine Circulation
Pump
V-402 Packed Tower Gas Scrubber - 2
V-503 Ammonia Pcrubber Brine Circulation
Pump
V-504 I.D. Exhaust Ian
V-405 Holding Tank
V-406 Holding Tank Mixer
V-407 Ammonia Brine Preparation Tank Mixer
F-408 Brink Filter
P-409 Ammonia Brink Preparation Tank Pump
X-410 Ammonia Vaporizer
V-411 Ammonia Brine Preparation Tank
185
-------
Figure A-204
MATERIAL BALANCE AMMONIA SCRUBBING PROCESS
Stream No. (1) (2) (3)
Description* Tail Gas Entering Cleaned Gas Leaving Ammonia Mafcj-Up
Packed Tower Ges Packed Tower Gas to Packed Tower Gas
Scrubbers ' Scrubbers Scrubber
Sulfur, S
Sulfur Dioxide, S02 i 0.00994 0.00050
Sulfur Trioxide, SOj
Air 2.822 2.022
Sulfuric Acid, H2S04
Sulfuric Acid Hist
M Hater
en Ammonium Bisulfite, NH4HSO3
Ammonium Sulfite, (NH4>2S03
Ammonium Sulfate, (NH4>2SO.
Ammonia (Anhydrous) 0.00189
Total (tons per day) 2.83194 2.8225 0.00189
Total (SCFM) 51.38 51.30 0.05544
Total (gal/min)
Temp. (»C) 75 75 25
Pressure (inches water)
Composition 1560 ppa SOj 86 ppm SOj
Conversion 95% Scrubber Efficiency
•Plow Basis: One ton per day 100% H-sO.- production.
-------
(Continued) > ',
Stream Mo. (4) (Sa) <5b) (6)
Description* Scrubber Brine Scrubber Brine scrubber Brine Brine Feed to
Bleed to Holding Tank Recirculation Circulation Regeneration Column
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, S0}
Air
Sulfuric Acid, H2SO4
Culfuric Acid Hist
Nater 0.02641 \ 2.633 2.633 C.02641
Ammonium Bisulfite, NH4HS03 0.01039 0.9S93 0.9595 0.01039
M Ammonium Sulfite, (NH4)2S03 0.00377 0.4035 (>( 0.4035 . 0.00377
3 Ammonium Sulfate, (NH4)2S04 0.00204 0.2505 •" ' 0.2505 ": v . 0.00204
Ammonia (Anhydrous)
Total (tons per day) 0.04261 4.2465 4.2465 0.04261
Total (SCFM)
Total (gal/min) 0.00710 0.707 0.707 0.00710
Temp. (°C)
Pressure (inches water)
Composition
Conversion
•Now Basisi One ton per day 100% «2SO4 production.
-------
(Continued)
Stream No.
Description*
Sulfur
Sulfur Dioxide, S02
Sulfur Trioxide, S03
Air
Sulfuric Acid, U2S04
Sulfuric Acid Mist
Hater
Ammonium Bisulfite, NH4HSO}
Aeunonium Sulfite, (NH^ljSO-j
Ammonium Sulfate, (NH4>2S04
Ammonia (Anhydrous)
Total (tons per day)
Total (SCFM)
Total (gal/min)
Temp. (»C)
Pressure (inches water)
Composition
Conversion
(7)
Gas Exiting the
Regeneration
Column
0.003355
0.000945
0.004300
0.0523
(8)
.Gas Entering the
Separator Tank
0.003355
0.000945
0.004300
0.523
(9)
SO- Recycle
Gas to
Acid Plant
0.003355
0.003355
0.0261
(10)
Feed to Vertical
Tube Evaporator
0.02641
0.01325
0.03966
(11)
Ammonium Sulfate
Solution Peed to
Centrifuge
0.00129
0.01325
0.01454
•Flow Basis: One ton per day 100% HjS04 production.
-------
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SO.,
Sulfur Trioxide, S03
Air
Sulfuric Acid, H2S04
Sulfuric Acid Hist
Water
Ammonium Bisulfite, NH4H£03
Ammonium Sulfite, (NH4>2S03
Ammonium Sulfate, (NH4)2S04
Ammonia (Anhydrous)
Total (tons per day)
Vital (SCFM)
Total (gal/min)
Temp. (°C)
Pressure (inches water)
Composition
Conversion
(12)
Ammonium Sulfate
Feed to
Dryer
(Continued)
(13)
Ammonium
Sulfate Recycle to
Vertical Tube Evaporator
(14)
Ammonium
Sulfate Exiting
the Dryer
0.00036
0.00265
0.00301
0.00093
0.01060
0.01153
0.00003
0.0026S
0.0026B
(IS)
Ammonium
Sulfate to
Storage
0.00003
0.00265
0.00268
•Plow Basis* One ton per day 100%
production.
-------
".*!' *•'">•
[Z::,r.
1 ;•*> /
l-Bl '*!•»• », ••SH
{ /i;, / \ "u*t3 fcvvr»
" , . ! SS3
I
•• T/
• -, •• f
; T i
•' /^Jl M
,./"'" ^-T'T
*ir • ^
1 - i .:*;---'•?'•
.1 ; !
; lrr~#.
! •
«
L..
1
T
i...
{..
1j
./ \
f
j
\/
/
^.s
r ' 2
A «= •{'
fAf»CO tOw(6 A. /fi\ AT\
^ ! S*
(y i i izza
\/
Jvr ' «*C
y'
*l /^
^J
i P^ j'J- v * Jp* ;| r t D
"CES
^na
.^. 1... , ^^-^--»_,_-.v .1 LI/ «"-'P""-'"
•r" | -j' \. Mlf.'l.'ft
0
«UN>» vu;gyc»
-------
PART 5
MOLECULAR SIEVE PROCESS
FOR
SULFURIC ACID TAIL GAS CLEANING
191
-------
MAJOR EQUIPMENT LIST
Figure A-205
MOLECULAR SIEVE PROCESS
Equipment Number Description
X-500 -Tail Gases Cooler
X-501 Tail Gases Demister
V-502 Acid Circulation Pot
P-503"'"' Acid Circulation Pot Pump
V-504 SO2 Absorbers
P-505 I.D. Fan for Tail Gases
F-506 Air Filter
X-507 Condenser
V-508 Air Dryers
X-509 Furnaces
p-510 I.D. Fan for Air
D-511 Drier Regeneration Fan
X-512 SO2 Gas Cooler
192
-------
Stream No.
Description*
Sulfur, S
Sulfur Dioxide
Sulfur Trioxide, SO-
Sulfuric Acid Mist
Water
H1
5 Air
Total (tons per day)
Total (SCFM)
Temperature (°F)
Pressure (inches of 1
Composition
Removal Efficiency
Figure A-205
MATERIAL BALANCE - MOLECULAR SIEVE PROCESS
(1)
Tail Gases Entering
Demister
0.00994
0.00032
1560 ppm S02
(2)
Demioted Gases
Entering
SO- Adsorbers
0.00994
0.000082
35
2.822
2.833
51.38
170
in. H20
2.822
2.832
51.38
170
28 in. H20
1560 ppm SO,
(3)
Collected Mist
Flow from Demister
To Acid Circulation Pot
0.000738
90% Acid Mist Removal
*Flow Basis: One ton per day 100% H2SO4 production,
-------
(Continued)
Streair. No.
Description*
Sulfur, S
Sulfur Dioxide, S02
Sulfur Trioxide, S03
Sulfuric Acid Mist
Water
Air
M
VO
*• Total (tons per day)
Total (SCFM)
Temperature (° F)
Pressure (inches of H20)
Composition
Removal Efficiency
(4)
Clean Gases
Out from Adsorbers
to Stack
0.0000994
2.8220
2.8221
51.38
100
18 in. H20
15-25 ppm S02
99% SO-
(5)
SOg Containing
Desorbed Gases
Recycled to Acid
Plant
0.00994
(6)
Air Flow to
Air Dryers
0.5644
0.56434
10.5
200
6000 ppm S02
0.0115
0.5650
0.5765
10.5
60
*Flow Basis: One ton per day 100% H2SO. production.
-------
(Continued)
VO
en
Stream No.
Description*
Sulfur, S
Sulfur Dioxide, SO-
Sulfur Trioxide, S03
Sulfuric Acid Mist
Water
Air
Total (tons per day)
Total (SCFM)
Temperature (° F)
Pressure (inches of
Composition
Removal Efficiency
(7)
Dry Air Flow
From Air Dryers
To Furnaces
0.565
0.565
10.5
60
(8)
Dry Heated Air Flow
From Furnaces
To Adsorbers
0.565
0.565
10.5
200
*Flow Basis: One ton per day 100% production,
-------
^
**»22>J
" f
MB~\' ;••>;
'
® "tiiaek ©
u--u* «.• \^f"'--.
ixzM/ l2j '
• •••-•" ~
f r^H**
(^ '"' i , *. /•^'*(;
V--, '[IT II" •""''• iV * /
i-wr:) --J xi( • i
liJ . Ill I) <-' : -•**-!
• mi." i n
t , .. ...F . ' , .. ,, II
— •> ^^r^ U
| r~i s*.
\ /-e » /^ ? :• ™':"
\/« i/r^i
w r& \/ r^1 / — i
/~~- 4i V * Y * rti ":
"" j « j «*«--- ^ / Af 'U
'fT /\F! A !* f £
i V— / \ ^~ / \ ffiN I lA_—
^ *r ' 4"1" I ' Cr-O-^— '
*• ' ' * ''V-i J i
,T". , 1 -f • • • . . !
< i , — ft J •<*• i t- - ... • rflL
?E i '- 2 X .Js?
i ft^-n*— l-c- .; t-i <"J2I3 J
& MOLECULAR SIEVE
^tOUPE A-2C5
M*
^
/'
^
j
•
—i
j
T
f
c
»<
Tl
;>
?1
J
5
i
:
I.,.
^-^.
V
A
/\
1\
L-r-"^
1!
^
6
n • o
tiBU
'TT3^
XT-.
1 \^i/
^
'
Q
i
i
i.r.
.^-s
V
A
/\
/\
^-— •"•"
i!
™*
* r7
0
J (*
7
I*
-------
APPENDIX B
PLANT START-UP DATA
197
-------
No.
LIST OF FIGURES
Appendix B
25 Dual Absorption Plant Start-Up 8 June 1974 Three Hours
to Compliance
26 Dual Absorption Plant Start-Up of 29 June 1973 - A
Series of Two Malfunctions With Each Start-Up Requiring
Less Than One Hour to Reach EPA Standard
27 Single Absorption Plant Start-Up 20 August 1973 Unstable
Start-Up of Less Than Two Hours After Two Hour Shutdown
28 Dual Absorption Plant 6 September 1973 Shutdown
29 Dual Absorption Plant 7 September Shutdown
30 Dual Absorption Plant 11 September 1973 Start-Up
31 Dual Absorption Plant 16 September 1973 Start-Up
32 Single Absorption Plant 17 September 1973 Start-Up
33 Dual Absorption Plant 2 October 1973 S-art-Up
34 Dual Absorption Plant 23-24 October 1973 Start-Up
35 Dual Absorption Plant 31 October 1973 Start-Up
36 Dual Absorption Plant 20 November 1973 Start-Up
37 Dual Absorption Plant 27 November 1973 Start-Up
38 Dual Absorption Plant 29 November 1973 Start-Up
39 Dual Absorption Plant 1 December 1973 Start-Up
40 Dual Absorption Plant 30 April 1974 Start-Up
41 Dual Absorption Plant 20 October 1974 Start-Up
198
-------
vo
vp
ppra SO-
(Exit Stack)
3000 -
2500 -
DUAL ABSORPTION PLANT
START-UP 8 JUNE 1973
3 HOURS TO EPA COMPLIANCE
Figure 25
EPA Standard
300 ppm
Percent Efficiency
Conversion
100
EPA Standard
Percent
Efficiency
10
-------
ppm S0_
(Exic Stack)
DUAL AfcSORPTION PLANT
START-UP OF 29 JUNE 1973 A SERIES 0" TWO MALFUNCTIONS WITH
EACH STAKT-UP REQUIRING LESS THAN ONE HOUR TO REACH EPA STANDARD
Figure 26
i-Ct
3000 -:.,-
t
2500 -
2000
Percent Efficiency
SOo Conversion
100 EPA Standard
Percent
" Efficiency
500 —T
A Standard I
0 ppm
-------
SINGLE -ABSORPTION PLANT
START-UP 20 AUGUST 1973
UNSTABLE START-UP OF LESS THAN
"&"
•••>;'J&i*
ppm SO,
(Exit Stack)
3000 — •
2500 — .
2000 — •
N)
O
h->
150C —
1000 —
500 —
EPA Standard
300 ppm
~~
(
Figure 27
)
•
~T 1
\ •
T
/
.
S
' *4
° t I
hs
2
«•*
1
I
I
(
i
m
'[
\
\
\
1
S
!lkJ__
c
IV
/
/
ki
'
j
S"
^<
ft
•
S,
<•«"
4
k
(
_ ..
m*fV %•*'
"I
•X
•^
s.
5
•ti
^
Jlf^HI
1
1
ta«
«
*n
i
s
••
•M
"*
«•
U.
7
tm
v
tilt
^
1
8
••*
1
"t
1
1
9
.1
1
< g Percent Efficiency
';- ";- , S02 Conversion
•'Iff- '•*
i
1
t
JLUU /ir/\ acanaara
" yy firnciency
no
' y /
rt c
— — y j
Q/
"
— 92
• — 91
n A
— go
•
Time Hours
-------
DUAL ABSORPTION PLANT
6 SEPTEMBER 1973
SHUT-DOWN
rjLKuru «.w • .
ppm SO, " ..
(Exit Stack) . _;.-..>•• [:
3000 —
2500 —
N)
O
to
1500 —
1000 —
500 •" "
)0 ppm
0 —
(
"f
''**
.J
1
!
; —
ii
s
x
'»^ \
\
D
S
?
T
—
'
1
/
i~|- -i-
1
t
f
-
?i
*••
iS^-
1 —
T
t
1
I"*"*
/
>
)
f
,
,
/
r
,
»
\
1
\
V
\
k
*"•
\
**,
y
N
•f
K
in
*i
rt"1
^
t
^
nt*
^
•^
k
^
.
s
s
J"1
"
•
I1*
III
mi
IW
Nil
1 1 II 1 !
2345678
Time Hours i
I
.«•••
•
9 3
—
_1
1
i
Percent 1
S0£ Conv
100
00
ii
| nl
1
1
.
i
,'
!j
•;
1
--
yi
— 96
— 95
— 94
— 93
— 92
— 91
— 90
•
EPA Standard
Percent
Efficiency
-------
NJ
O
DUAL ABSORPTION PLANT
7 SEPTEMBER 1973
Figure 29
ppra SO.
(Exit Stack)
3000
2500 —
2000
150C
1000
500
EPA Standard"
300 ppm
0 —t
Percent Efficiency
S02 Conversion
.— 100
99
98
97
96
95
94
EPA Standard
Percent
Efficiency
— 93
— 92
— 91
— 90
10
Hours
-------
DUAL ABSORPTION PLANT
11 SEPTEMBER 1973
START-UP
I I
ppm SO*
(Exit Stacl
3000 —
• M '
..
. i-il
2000 —
0
4k
_.
150C
1 1
1000
i.
i
.'{
i
->00
FPA Sfanderd
300 ppm
n
1
j
\
\ ,
I
V
•
\
•
>
/
f
-
j
/
L
\
\\
"
/
,
L
\
\
!
/
1
7
f
• ;
!
J
*
V
(
i
t
•
-
1
k
1
.i
.rt1
•h,
;
i
»••
N
i
.
*••
'
f
1
i
i
"ii
>
|
!
i
i
i
/
r
•.„
•
/
.1.
,/
^
(
i
/*
,
•
S|
^
ni
%
fri
••o
IM
fc.
ir
in*
S,
.
^
3G
Ut
^
~
I;
..^*
^*
•^ -
t
•
i
i
^
1
1
•<
r
hi
P"
IW
dB
|
i '
•
•i
'
*H
*T
i •
1
•r! -•
l*N
g<
^
«•
X
N*
!-.-
'( ' '
I '
"f
1"*
i
t
1
'
1
^
'
S02 Conversion
— 99 Efficiency
,— 98
1 — 97
96 .
5
i
94
.
93
: 92
, — 91
;— 90
.
i
.
10
Tine Hours
-------
to
o
Ul
,ppm S0«- -.1 \
(Exit Stack)
3000
2500 —7
2000
1500
EPA Standard
300 ppm
DUAL' ABSORPTION 'PLANT
16 SEPTEMBER 1973
START-UP
Figure 31
Percent Efficiency
S02 Conversion
|— 100 EPA Standard
Percent
-------
SINGLE ABSORPTION PLANT
START-UP
17 SEPTEMBER 1973
ppm SO, (Exit Stack)
to
o
4000
3500
3000
2500
2000
1500
1000
Efficiency
' WELLMAN-LORD SODIUM
SCRUBBING SYSTEM DOWN
12 16 20 24 28 32 36 40 44 48 52
Time (Hours)
-------
EPA
300 ppm
ppm SO,
(Exit Stack)
3000 —;-
2500
DUAL ABSORPTION PLANT
START-UP
2 OCTOBER 1973
Figure 33
Percent Efficiency
S02 Conversion
-i- i— 100 EPA standard
Percent
• 9? Efficiency
98
10
-------
to
o
oo
DUAL ABSORPTION PLANT
START-UP
23-24 OCTOBER 1973
Figure 34
ppm SO, (Exit Stack)
4000 -
3500 -h-rr
EPA S-andard
300 ppm
8 10 12 14 16 18 20 22 24 26
Percent Efficiency
o2c
100
S0« Conversion
EPA Standard
-------
DUAL ABSORPTION PLANT
START-UP
31 OCTOBER 1973
Figure 35
O
VO
ppm SO.
(Exit Stack)
300 ppm
Percent Efficiency
S0 Conversion
— 100
EPA Standard
Percent
Sfticiency
Time Hours
-------
10
»-•
o
DUAL ABSORPTION PLANT
START-UP
20 NOVEMBER 1973
Figute 36
ppm SO,
(Exit Stack)
3000 —
2500 —
2000 —•;
Percent Efficiency
S02 Conversion
t---l~M— 100
EPA Scd.
•JOO ppm
10
-------
DUAL ABSORPTION PLANT
START-UP
27 NOVEMBER 1973
Figure 37
ppm S02
(Exit Stack)
3000 —!_,
2500 -
2000 —;—
150C
1000 —'•-
EPA Standard
300 ppm
10
Percent Efficiency
SO- -Conversion
100
EPA Standard
Percent
Efficiency
-------
DUAL ABSORPTION PLANT
START-UP
29 NOVEMBER 1973
Figure 38
ppm SO.
(Exit Stack)
i
_.„_
!
•
1
r~
.
,
1
ro
"
•
150C i,
'•
ji
1 1
i
'
.
300 • ;
jrp A S tr\ nd n rrf , Mi , „,„„_ _
300 ppm
°^
0
-••
r
\
5
/
t
y
_f
!
1_
-
r
t
-•
r/
\
-
VI
V
... 4
111
* V
1
1
-
L
*»•
in
L.
IK
a«
ii*
•**
PI
^
»^
11
,.»x
*'
s
11*
^
||»
\
>
V
•*••
_
„. rjii.!'"!'"!
I'! 1
2345
Time Hours
j
I
1
7
_
_
—
•-
_
—
— ,
..*
«
_
„_
1
891
1
_
Percent Efficiency
S02 Conversion
^ 10° EPA Standard
"99 Jfertent
i Conversion
; qft
97
Ofi
yo
nl
93
h — 92
— 90
j
i
-------
DUAL ABSORPTION PLANT
START-UP
1 DECEMBER 1973
Figure 39
ppra SO,
(Exit Stack)
innn
JUUW
*3 e f\r\ __
OUU
">rtrtrt ._.
ZUUU •"••
.j
w
1500 — '
1000 —
500 —
jrpA Standard _ _ __
300 ppm
0 -
1
D
I
i
g
•
S
:
*"*
(
;?
f
\
\
s
•
5
:
1
1
\
\
Y
k
r
/'
i
s
s
j
•
£
\
\
»»•'
"
1^
/
/
I/
/
/
J
/
• f
^•J[
-j
•
>1>
<,
.•J
>
s
,»v
>
I*1"
s
..
^
'i
"*
•r
"•H
••
^
sv
.•r.
SK
"t
}
2 3 A 5 6
Time Hours
7
1
8
1
9
Percent Efficiency
S&2 Conversion.
1 ftfl ™. . r.
J,
~^
--~ nfA bianaara
"J Conversion
' 98
95
i
9'i
; — 93
— 92
— 91
— 90
•
-------
to
DUAL ABSORPTION PLANT
START-UP OF 30 APRIL AFTER ONE HOUR ,
SHUTDOWN WITH ONE HA^F HOUR TO EPA STANDARD
.r ,;ure 40
ppm SO-
(Exit Stack)
3000
2500 —
2000 —
150C —;
1000
500 —
Percent
Efficiency
100
-------
DUAL ABSORPTION
to
(-•
Ul
START-UP
20 OCTOBER 1974
Figure 41
ppm SO,
(Exit Stack)
3000 -f-4
2500 —
2000
150C
EPA Standard
300 ppm
Percent Efficiency
SO- Conversion
— 100 EPA standard
Percent
Efficiency
-------
APPENDIX C
STATISTICAL ANALYSIS OF PROCESS
PARAMETERS AND EMISSIONS
Table of Contents
Regression Analysis Model of Dual Conversion Acid
Plants - Mehta .................... 217
Statistical Analysis of Dual Conversion Acid Plants -
Nixon ......................... 296
216
-------
REGRESSION ANALYSIS MODEL
OF
DUAL CONVERSION ACID PLANTS
Catalytic, Inc.
Project 42466
EPA Contract 68-02-1322
by
Dr. B. M. Mehta
217
-------
An attempt was made to explain S0_ emission in terms
01 process control parameters in H2S01| plant, particularly
temperatures of the four catalyst beds in which S02 to °>0
conversion is carried out. Data were collected from two,
dual absorption sulfuric acid plants: i) Sulfuric acid plant
of ASARCO at ElPaso, Texas, and ii) TexasGulf sulfuric
acid plant. Each data point constituted 9 observations.
They are: i) parts per million (ppm) of SO- - exit
stack and 11 to ix (X) Inlet and outlet temperature in
°P of four catalyst beds. For easy interpretation data
are divided in two parts, i) start-up and shutdown of
plant and ii) normal plant run.
Several models were tried using multiple regression
with ppm of SO- as the dependent variable and process
parameters and their different transforms selected as
independent variables. In the following pages each model
is described ir- brief. Complete computer repults are
attached.
ASARCQ Sulfuric Acid Plant
Normal run: L'^ta set (ASAP.H consist of 30 data
points selected randomly f~on normal plint run. Different
statistical models were then tried on this set.
218
-------
Run 1
S02 emission as a function of inlet temperature of the
four catalyst beds.
Yl = o^ + «X, xl + °<2_x3 + °<3 x5 + °<^ x7
The regression was not significant at the 5 per cent level.
Run 2
S02 emission as a function of inlet and outlet tempera-
tures of the four catalyst beds.
Yl - <, +
-------
Run 5
SO^ emission as a function of inlet temperature at the
four catalyst beds and temperature drop across the three inter
coolers.
xH) + <>< (x7-x6)
The regression was not significant at the 5 per cent level.
Run 6
SO- emission as a function of temperature rise across
the four catalyst beds and equilibrium constants at the inlet
temperature of each bed.
Yl = +
where C = EXP (8929/CI60 +xl) - ft
F = EXP (8929/C<60 +x3) -'^.678)
I = EXP (89297(^60 +x5) - 4.678)
L = EXP (89297(^60 +x7) - ^.678)
The regression was not significant at the 5 per cent level.
For this data none of the models satisfactorily e.-.plain
SO- emissions in terms of the process parameters, from point
of view of 6 significant regressions. Several explanations
for lack of fit are: i) variables on the right-hand side
are not independent of each other and are highly correlated,
ii) there are other process parameters (measured and not
measured) which affect the SO^ emission which were not
taken into account; e.g., concentration of I^SOjj used
in absorption tower, initial concentration of S02, furnace
temperature and catalyst coins.
22Q
-------
It is known that S02 to SO., conversion is favored by
lowering the temperature at which reaction takes place and
it is also known that reaction rate is favored by higher
reaction temperature. This sets a definite range on the
temperature of Catalytic oxidation zones. Thus, to have
S02 emission below certain set level, the eight temperatures
should be controlled within a certain range. To find out
this range, 6*1 data points were randomly selected from the
normal plant run (ASAR 3) and standard statistics were cal-
culated. Summary of the calculations is as follows:
Variable Mean 95* range for xi
Xi Xi + 26
PPM SC2 Exist Stack 127 0-307
Temp. (°F) Inlet 1st 821 810-838
Converter Bed
Temp. (°P) Outlet 1st 10H1 9^8 - 1133
Converter Bed
Temp. (°F) Inlet 2nd 837 819 - 855
Converter Bed
Temp. (°P) Outlet 2nd 892 811 - 973
Converter Bed
Temp. (°F) Inlet 3rd 786 73* - 838
Converter Bed
Temp. (°P) Outlet 3rd 791 737 - 8U5
Converter Bed
Temp. (°P) Inlet Hth ?62 704 - 820
Converter Bed
Temp. (°P) Outlet Uth 762 702 - 822
Converter Bed
221
-------
It has been found that in this ASARCO plant, SCL emission
is below EPA limits, if the 8 temperatures stay within above
ranges. There are not enough data on the upset condition in
order to define the effect of deviation from above tempera-
ture range on S02 emission from stack.
Start up of ASARCO acid plant: ASARCO plant is not able to
control the SO,, emission during start-up. SOp emission
appears dependent on length of shutdown time to a certain
extent.
It was observed that if shutdown time of the plant is
less than two hours, i'0_ emission will not exceed the EPA
limit. Above two hours of shutdown time, S02 emission
is above the EPA limit during start-up. There are 13 data
points (STA1) for start-up time and shutdown time.
Start-up time is defined as the time in hours which a plant
will take to come under the EPA compliance. An attempt
was made to find a linear relation between start-up time
and shutdown time.
Start-up time - o^0 + c<| (shutdown time)
The regression was not significant at the 5 percent level.
Mean and standard deviation were calculated based on this
13 data points. It is calculated that average start-up time
is 6.311 hours. For 95 per cent of the time the start-up
time should be between 3-1'* and 9-5^ hours. Not a single
data point was more than nine hours. Details of above two
runs can be found at the end of this report.
222
-------
Thirteen data points were selected (ASAR2) having high
emission and standard statistics were calculated from
these and are compared with one of normal run.
Variable Mean
Normal Run " Start Up
Yl (ppm) 127 1678
XI (°F) 821.0 815.0
X2 (°F) 1010.9 1006.5
X3 (°F) 837.0 811.5
XI (°F) 89?-o 811.0
X5 (°F) 786.: 732.5
X6 (°F) 791.1 729.n
X7 (°F) 762.0 702.0
X8 (°F) 76?.0 707.0
Above table clearly shows that four converter beds were
not hot enough for proper S0_ and SO, conversion.
TexasGulf Sulfuric Acid Plant
Data set (T^O^^011315^3 of 31 data points selected
randomly from large set of normal plant run. Different
statistical models were then tried on this set.
Run 1
SO- emission as a function of inlet temperature of the
four catalyst beds.
Yl »
-------
Run 2
S02 emission as a function of inlet and outlet temperatures
of the four catalyst beds.
Y1 = *<• + °
-------
Ranp;e
x for xi
Variable Mean Mean + 26
PPM SO Exit Stack 128 56 - 200
2
Temp. (°P) Inlet 1st 802 702 -
Converter Bed
Temp. (°F) Outlet 1st 1117 1095 - 1139
Converter Bed
Temp. (°P) Inlet 2nd 841 817 - 865
Converter Bed
Temp. (°P) Outlet 2nd 977 903 - 1051
Converter Bed
Temp. (°P) Inlet 3rd 812 784 - 840
Converter Bed
Temp. (°F) Outlet 3rd 848 797 - 899
Converter Bed
Temp. (°F) Inlet 4th 8l6 788 - 844
Converter Bed
Temp. (°P) Outlet 4th 841 809 - 873
Converter Bed
Ifc-has been found that SO- emission will be below the EPA limita-
tions if the eight process temperatures remain within the
respective ranges.
The plant has a capability of heating the catalyst beds
to a proper temperature before passing S02 through catalyst;
because of this they are able to control the emission during
start up.
RECOMMENDATION FOR FUTURE DATA COLLECTION AND ANALYSIS:
At every point of analysis it was felt that there is a
lack of total number of complete data points. Large number
of data points were inconp^ete, meaning one or more process
variables were not recorded. There were no data to compare
225
-------
the two plants of the same design. To make any conclusive
statements data should be collected from at least three plants
of same design. Record of each higher S0« emission should he
kept. This record of upset conditions will help us narrow the
temperature range for eight variables (XI, X2-X8) and define
the exact variable with its magnitude which caused the higher
emission. It will be easier for the regulatory agency to pin-
point whether a particular upset condition was legitimate cr
not. Reliability of SO- analyzer should be checked very often
to ensure us of correct data.
Nomenclature Used in Regression Analysis
Yl = PPM S02 Exit Stack
XI = Temp (°P) Inlet 1st Converter Bed
X2 = Temp (°F) Outlet 1st Converter Bed
X3 = Temp (°F) Inlet 2nd Converter Bed
X4 = Temp (°F) Outlet 2nd Converter Bed
X5 = Temp (°P) Inlet 3rd Converter Bed
X6 = Temp (°P) Outlet 3rd Converter Bed
X? = Temp (°F) Inlet 1th Converter Bed
X8 = Temp (°F) Outlet 4th Converter Bed
225
-------
MULTIPLE REGRESSION ANALYSIS
ASARCO PLANT NORMAL OPERATION
227
-------
ASARCO
Data Set 1
ASAR1
HEADY
LIST ASARI
ASAM 1U07FST 11/25/74
iOO 191.825* 1 100.850/975*835*860.825,795
101 145.810. 1030,830*890,785.800.765.760
102 274,815, 1 100,840*945,820*850.825*820
103 P28,81R, 1058,842,934,826*810,798,782
104 156,830, 1 100,830*900, 760*780*755*780
105 701,825* 955* 820, 845* 730, 725* 705* 710
106 95. 830. 1 000. 830. 860* 795. 780, 740, 730
108 1 38, 830, 1 070, 845, 930, 805, 840, 785, 775
109 239,825, 1 120,840*940*800*835, 795*810
110 145*820* 1075*845*945*835*840*825*815
111 1 1 2* 8 1 5* 1 1 00* 850, 955* 820* 850* 825* 820
112 129*830* 1000*850*900, 790, 790, 745, 750
113 149,825, 1080,845,950,835*840,820,825
1 1 4 222, 825, 1 070, 830* 955* 825* 790, 770, 755
115 259*820* 1055, 835* 9 15*780* 780* 760, 755
116 173*815* 1075,835*895*775*800*765*770
117 331*830* 1060*830*910, 770,790, 770* 780
118 327,830, 1010,835,870,765,785,750,765
1 19 388, 830* 1 1 1 0, 850, 930, 300. 8 1 5, 765, 770
120 229,810* 1000.840,890,825,840,815,800
121 255,825, 1080,820*885,720, 690, 675,685
122 257,835, 1035,840,865,780,790,770,760
123 223, 825, 1 090* 840* 895, 785, 805, 770, 780
124 297.820, 1010,835,940,770,780,765,750
125 374,815, 1065,835*910*780*790*760,770
126 353, 8 1 4, 960, 834, 883, 796, 804, 78R, 770
127 181,812* 1 1 18*834,936*780*800*752*754
128 182, 82.4, 1114, 834, 922, 768, 775* 758* 764
129 102,824* 1 136*840*962*800*830*782*,782
223
-------
RUN 1
MULTIPLE" KSGRESSI0N PK0GRAM
XMFAN
8.224828F+02
1.061241E+03
8.373793F.+ 02
.9.148966F+02
7..91S517F+02
8.022069E+02
7.732069E+02
7.717931E+02
YMFAN
2.363793F+02
0BSERVFD
1.9100000F-»-f'?
I.4500000E+02
2.7400000E+02
2.2800000E+02
J.5600000F+02
7.010COOOE+02
9.5000000E+01
1.3800000E+02
2.3900000E+02
1.4500000F.+ 02
1. 1200000F+02
1.2900000E+02
1.4900000E+02
2.2200000F+02
2.5900000F+02
1.7300000F+02
3.3100000F+02
3.P700000F+02
3.B800000F+02
P.P90CCOOF+0?
2.b500COCF+Of
2.S7000COF+0?
?..?300COCF+02
2.9700000F+02
3.740UOOOE+02
3.5300000F.+ 02
1.8100000F+02
1.8200000E+02
J.0200000F+02
1
2
2
1
3
3
2
1
2
2
2
3
2
1
1
3
2
?
?
2
2
2
2
2
CALCULATED
4841 158F+02
2.6274889E+02
2. 1089986F+02
1.6604196F+02
1621091E+02
6632825E+02
2258276F+02
9546666F+02
6267472E+02
7914977E+02
7519773E+02
5949214E+02
8181375E+02
5662385F+02
71 12709F+02
0795304E+02
842207 1F+02
7173433E+02
8R37R31F+02
4-020267E+02
5517351F-I-02
4062902F*0?
8405968F+02
5501 194F+02
4848755E+02
4954862E+02
2.8634020E+02
2. 1785139F+02
EST VAR
1.9074241F+03
2.3663184F+03
2.1450056F+03
I.5383148F+03
2.4381770F+03
2.93015^3E+03
4.0136465F+03
1.3826847F+03
1.0300517F+03
1.5902827E+03
2.7250538E+03
5.5225457E«-03
1.8744414E+03
6.4185410E+03
6.5617222E+02
1.4858165F+03
3.1945885F+03
1.3731419F+03
3.5124348F+03
2.2762290F+03
4.2575563F+03
P.4709R90F+03
7.354P339F+02
1.2105110F+03
1.3776373F+03
1.3649071F*03
2.4464908E+03
9.6137556F+02
5.3500893F-«-02
229
-------
RUN 1 (Continued)
• , (.DFGRFF OF FREE. SUM OF SULIAhFS.
. REGRESSION.
. REMAINDER .
. TOTAL
4
24
28
9.752241F+04
. 3. 155564F+05 .
4.130768E+05 .
VARIANCE F.STIMATF .
2.438060F+04
1 .314818F+04
MULTIPLF C0RRF.LATI0N COEFFICIENT = 4.858875F 01
CBNSTANT TFKMCINTFHCFPT) = 3.6R0666F*03
C0FFFICIFNTS
3.2238233E-01
•3.1750097E+00
•2.2641340E+00
9.58897A4E-01
FST OF SD
3.5627722E+00
4.4162024E+00
.1.7134936E+00
1.4836531F+00
F FvATI0< 4, 24 DEGREES 0F FKEED0M>= 1.854294F+00
UPPER RT HAND P0KTI0N 0F MATRICES 0NLY
INVERSE ADJ DATA MOMENT MATRIX
K0WC 1) 9.6541E-04 -4.6081F-04 9.7809E.-07
M)U( 8) 1.4833F-03 -1.5673F-04 -1.5615E-04
h0WC 3) 2.2331E-04 -1.4045E-04
R0VX 4) J.6742E-04
EST COEFFICIENT VAR-C0VAR MATRIX
R3W< U 1.2693E+01 -f.0588F*00 1.2R60E-02
f«W( 2) 1.9503E+01 -2.0608F.+ 00 -2.0530E*00
R0VC 3) ?.93flE+00 -1.8467E+00
R0W< 4> 2.2012E+00
ADJ DATA CORRELATION MATRIX
MJV:C 1> l.OOOOE + 00 -3.8508E-01 P.1066E-03
hJ3W< 2) 1.0000F*00 -2.7233F-01 -3.1334E-01
R0W< 3) l.OOOOF+00 -7.2642F-01
F0VX A) l.OOOOF+00
1.3341E-04
1.7541E+00
3.31B4E-01
230
-------
RUN 2
MULTIPLE RFGRFSSI0N PROGRAM
XMFAN
8. 2?4828E+02
1.06124 IF*03
8.373793F.' 02
9. 148966F+02
7.915S17E+02
8.022069F+02
7.7320f9E+02
7.717931F+02
YMF AN
2.363793F+02
NlWREtO INDICES PF VAFIATFS DFSIf4KF-» OP
2.5136933F+OP
3.3747058F+0?
I . 73 19^33F+0?
?.39Pl095F+Oy
FST VAh
5. 1931016F+03
3. 1793207F+03
2.2841035F+03
3.^3979#>8F + 03
3.6382258E+03
4.2805878E+03
5.965544PF+03
4. 1270S49F+03
2.7757467F+03
2.3800124F+03
P.9965314F+03
7.R335695E+03
4.2692028E+03
7.3538857F+03
I. 1255771F + 03
2. !9727b«F + 03
3.4R8AAPSF+03
2. 9fOP4bPF+C3
3.7527lRf>F+03
4.04lf.?90F+03
<"•. f 574fr 1 7F+03
f-. 9751 4R3E+03
?. 01 1 1006F+03
6.ft555059F+03
P.030PPR8F+03
3.5187125F-t-03
4.3945007F+O3
?.5«f6?04F+03
1 . OPOOOOOF+ 02 1.5301
3.0770284F+03
231
-------
RUM 2 (Continued)
. DFf-hFF OF f-KFF. SI'M OF St;llAKFii. VAKIANCF ESTIMATF .
8 . 1.575476F+05 . 1.969345F+04
. RFMAINDFR . 20 . P.555312F + 05 . 1.277656E+04
. TOTAL . PR . 4.130788E+05
MULTIPLE COhRELATION COEFFICIENT « 6.175746F-01
CONSTANT TFWICINTFF-CFPT) = 2.512162F+C3
COEFFICIENTS FST OF SD
8.1945439F-01 3.6111681F+00
- 1.39S715IF+00 H.3710714F-01
- 1 .3P02066F+00 5.0597B7&F+00
I.0498617F+00 1.3460SR6F+00
-P.6366182E+00 2. 00 1 7645E+00
- 7.b009734E-01 &•2222179F+00
7.0603361E-01 3.4613877F+00
=.316F*00 P.6R75709F + 00
F KATI0C K* 20 DFGhEES 0F FKEFDOM)= 1.541373F+00
I'PPFK KT HAND POKTIOM OF MAThlCFS
ADJ DATA MOMENT MATRIX
Mn.'C 1) 1 .CP07E-03 - I.RS54F-0? -5.2963F-0* 4.08J8E-05 -6.R396F-OS
7.7021F-05 1.940BF-04 -1.0449F-04
KC'.-'C 2) 5.4846E-05 3.6579F-05 -6.7828E-05 I.I669F.-05 -3.5251E-05
1.1361F-04 -8.86RBE-05
Km-'< 3) 2.003RF-03 -R.745SF-05 -8.4696E-05 -4.2333E-04 1.8356E-04
-3.1409E-05
Fi?'-.'C 4> 1.4181E-04 -4.0494F-05 4.3142F-05 -1.3490F-04 9.4492E-05
»^V< 5> 3.1363F-04 -8.3R43K-05 -2.57R2F-04 1.6258E-04
hfl'.-'{ ^) 3.RT51E-04 -2.1693E-04 -4.79ROF-OS
9.377tF-04 -5.7385E-04
8> 5
FST COEFFICIENT VAK-C'^VAK MAI I-IX
inV( I) 1.3041E*01 -2.3706F-OI -6.7669F*00 5.^152E-01 -8.7386E-01
9.H406F-01 P.4797F+00 -1.33bOr+00
WV.C L> 7.007bE-01 4.^735F-Ol -S.6661F-OI 1.4910E-01 -4.5038E-01
I.'iblbE*00 - 1 . 1331E-t-00
K«V.'( 3) P-56C1E + 01 -1.1174F+00 -1.0821F+00 -5.4087F«-00 2.3453E+00
-4.0UUF.-G1
icTWC 4) 1.K119F + 00 -';. 173K=--01 5.5121F-01 -1.7235F+00 1.2073F+00
i\M'< b) .,.007JF + Of -1.0712F*00 -3.2941F+00 9.0772F+00
hfl'.-.t 6> 4.9383F+00 -2.7716F*00 -6.1302F-01
Ji):-.'C 7> 1.1981F + 01 -7.3319F + 00
i\3!-'C 8) 7.P230E+00
232
-------
RUN 2 (Continued)
ADJ DATA COKKFLATIflN MAT MX
KO'.-'< 1)
Kfl'.-C V)
S.OOVf-K-O
M lv < 3 )
-^'.951 IK-Or
NHV< A)
I.OVC 5)
KOV<
t\OV<
7)
8)
.OOOOF+00 -7.R4POF-0? -3.703SF-01
1.983HF-01 -1.3755F-01
.OOOOF+00 1.1034F-01 -7.i
-S.OSf^-F-Ol
.OOOOF.+ OO -1.6406F-01 -l.i
1.07fc>9F-Ol -1.2089F-01
01 1.339 IF-01
.OOOOF-MIO -1.9fJ01F-t.M I.K4P7F-01 -3.699PF-01 3.337PF-01
.000()F*OO -P.^OR1F-01 -^.7541F-01 3.8611F-D1
.OOOOF+On -3.fi033F-01 -l.i
.OOOOF-KlO -7.8814F-01
.OOOOF+00
233
-------
RUN 3
It.LT JrLr . r ' ;.F ; .' I"V
/..••;[•»"••: re
7.9lbb)7F*6?
7. 7:^'C'A9F-"•<•)'«?
I.910000CF+02
1.4bCOOOOE+02
2.7400000E+02
2.2800000E+02
7.0100000F+02
9.500000CE+01
I .3SOOCOOE+02
2.3900000E+02
1.4500000E+02
i- 1200000E + 02
1.2900000E+02
I.4900000E+02
g.5900000E+OC
».7300000E+02
3.'3100000E*02'
3.270COOOE*02
3.88COCOOE+C2
2.2900000E+02
2.5500000E+02
2.5700000E+02
2.?300COOE*02
2.9700GOOr+02
3i530GOOOF+G«?
1.8IOOOOOF+02
1.8200000F.+ 02
1.0200000E+02
CALCULATED
3405FCH7E+02
5402068F+C2
8I26160F+-02
8290439F+02
7816805E+02
4280216E+02
198913<^F + 02
1. 7R55RJ/-F + 02
1F+G2
2.662J9b»t*02
3.295073GF*02
2.252383fF+02
3. M7b50^F + 02
2.522790fE+02
1.955291-SF+02
2. 39 21 09 SF+ 02
1.5301 l-i^
FST VAh
5. 19309-.VP-+03
3. 1793200E + 03
3.2397977F+03
5. 9655-6 f-RF + 03
4. 1270b3PE+03
2.380013fiF*03
7.8335^ 9fF+03
F «-03
I. 1255772F + 03
2.
3.75?71> fF + 03
A.Q< 292F+03
2.01 1 IC06F+03
(• .fbbbO/7F + 03
H.03022SOF+03
3.0770275F+03
234
-------
RUN 3 (Continued)
.DFGfcFE TF FrcEE. M» OF SLfAF:FS. VAhlANCF ESTIMATE
. hEGKFSSIGiM. & . 1 . 57547.'-?+05 . I . 9/£9345F+04
. REMAINDER . £0 . e". 55531PF + OS . 1.277656E+04
. TflTAL . J'fi . 4.i3078nF+05 .
.vJLLTIPLI- CPMxELATICN CTF f 1 CI F.^l ~ ^. 1 75V'f-F- '• I
Cr>vJbTA,MT TFKY(INTFKCFr'T) = p.5!2I£P
COFFFICIF'vITS FIT TF i.T
-5.792f-065F-01 3.
-?.. 7034557^-01 5.i
-3.:lAThIX
i-n'.-< 1) 1-0384F-03 -5.2006E-04 -1.4956F-O5 I.145IF-0*
-P.701C---05 4.1770F-05 -1.9318E-04
ivf' C 2J 1.9707F-03 -5.0537F-04 1.1174E-04 -3.1249F-05 5.4357F-05
-3.8CI8F-04. 6.3082E-05
hOV.'C 3) 5.3245E-04 -3.6015F-04 -2.3581F-05 2.£477F-06 3.0267E-04
;;•". ( 4} 3.5538F-04 2.4922F-05 -4.040^^-05 -2.6491F-04 -i
KCV< 55 5.48^^F-05 -6.7828E-05 -3.5251F-05 -8.
in.'< f) 1 ./.181E-04 4.314J>E-05 9..
r.f.-C 7) 3.8f51E-04 -/..7980F-05
i .%'. ( S ) 5 .
:fV.:< 1) l.3C-'7F + OI -f .'-'•< .'E+i:*.: -1.91O9F-C1 1.4f31F+00 4.A3'
-:«.4bKF-01 5.33foF-01 -f».«^KIF + 00
:,rv( 2) P.5179F + 01 - f. ^5^9^ + 00 1.4277F + 00 -3.9926F-0 1 6.9450E-01
-^.rt57^P+00 R.C597F-01
rC.;(. 3) 6.KOPVE + GO -A.6015F. + OO -3.01P9E-01 3.3829r-02 3.8670F + 00
:'.-< ^) 4.5405F + GO 3.1-.42F-01 -5.1625F-G1 -3.3847E + 00 -1.0881E-01
".-.< b) 7.0)75r-01 -tf.'*MF-01 -4.5038F-01 -1.1331F+00
''-•( f) 1.8119r 00 5.51P1F-01 1.2073F + 00
". < 7) 4.9383r+00 -f.!30?F-Cl
" C -<) 7.2C3GF + UO
235
-------
RUN 3 (Continued)
I . 3^51 F-0 1 1.52C8F-01
AUJ I..A'JA CPhfvELAl ICN >iAlhlX
uO-. C 1) l.OCOOF + QO -3.f2S5F-Jl -!
-7.Oocf.F-02 6.5933F-0? - :-•. b? 1 3F-01
.^'.-C 2) l.OOOGF+00 -/1.9336F-01 1.3353F-C! -9.5C52F-02 U0282F-01
S.9765F-02
r.f ;( 3) .OOOCF + 00 -B.:-'795F-01 -1.3799F-01 9.63S6E-03 6.671BF-01
'c. • OH'i^ fc" - U
.-•"••..( 55
. - . C • )
(:•-•< 7 )
i.^vc -;>
.OOOOF + OC) 1 .7.-.f- 1?-01 -1.7999F-01 -7.1473F-01 -I.900IF-0?
.OOCOF + 00 -7.AV09F-01 -2. 4g 1 IF-O I - '_>. 03f C>F- 0 I
.GOGOF+OC l.-5^.r'7F-Ol 3.3372F-0!
.OGUCE-i-CO - l.OJ-V.^E-Cl
236
-------
RUN 4
MULT1H.F KFGKELMON PhPGrAM
XMFAV
P.3H7586F+02
1.Of5bl7F+Ol
-1.413793E+00
YMFAN
2.363793F+0?
-1.233448F+02
-2.90COCGE+01
NLMBFfo INDICES 0r VAhlATES DEiIt\ED?7» 1, 2, 3* 4* 5, 6* 7
KANT TO SFF PREDICTED VALUES, tES OR N07YES
n&SFKVFC
1.910000GF+02
1.45000GOF+02
2.740COOOF+02
2.2800000E+02
1.56CCOOCF+02
7.0100000E+02
9.50000COF+01
1.380000CE+C2
2.3900000E+02
1 . 4SGCOOCF+02
1. 12GOQOOE+02
1 .29GOOOOF-»-02
1 . 4900UUOF*02
2.2200000F+02
2.59GOCOGF*0?
1.7300000F+02
3.3100000E+02
3.2700000F+02
3.8800000E+02
2.2900000E*02
2.5500000F+0?
2.5700000F*02
R.P300000F+0?
2.9700000r«-02
3.7400000F+02
3. 5300000F+02
1.31000COE+02
t.8200000F*02
1.0200000E+02
CALCULATED
1 . 4576747E+02
2.2822065E+02
1.7170387E+02
1.8901328F+02
2.79I6687F*02
4.3310248F*02
2. 1394675F+02
.8950993E+02
.8618751F+02
.8374593F*02
•8582b7SF+C2
2.90161R3E+C?
.9572399E*02
.708R799F+0'.J
2.666751 1F+G2
2. 1915545F*C2
3.2208684F+02
3.3443362E+0?
1 . 53066^8F+0?
2. 1 197438F+02
3. 1239980F*Ot>
2.7538743E+0?
2.0064470F+02
3.8245955F+0?
2.4I78304E+02
3.2762712F*0?
1.5179771F+0?
2.4451697E+OP
t . 48027b2F-»-02
4
9
1
3
3
3
5
3
2
2
2
5
3
6
1
1
3
2
2
3
6
5
1
6
1
3
2
2
2
FST VAR
4.5900233E+03
9.7321996E+02
1.9210613E+03
3.0153719E*03
3.5150346E*03
8430605E+03
5.6574145F+03
8960092F+03
2.66903J1E*03
2.I291872E+03
2.7313220E+03
7376433E*03
8199624E+O3
6.9119391F+03
1.0877826F*03
6943443F*03
3717892E+03
7861410E+03
4790886F+03
3542331F+03
4202791F+03
0633582F+03
R623307E+03
4164783F*03
6738419F+03
0974757F+03
8075017F+03
2-4126092F+03
2.B972714F+03
237
-------
RUN 4 (Continued)
. DEGRF.F OF Ff6F-0 1
CONSTANT TFKMCINTERCEPT) = 4.f41007F+0?
COEFFICIENTS
7. 832 109 IE- 01
8.9553I01F-01
1.0?74l?OF+00
4.0431 1 I2E-01
I.8165000E+00
2. 1242252E+00
EST 3F 1-D
3. 102634/iF+Ou
1.6S50263E+00
1.6031329F+00
3. 1277579fc>00
1.3935297E+OG
1.9998195E+00
F RATI0C 7» 21 DEGREES 0F FREEDOM)= 1 . 77631 5E+00
UPPEK r.T HAND F0KTia\ 9F -SATKICES 3NLY
INVERSE ADJ
RBV.'C
1)
7
-J.4820E-04
R0WC
2>
2
DATA MOMENT
.7915E-04
- 1.
MATRIX
9313E-04
-1.
978BE-04
9.
5005E-05
7.5807E-04
-2. 1 16SE-06
.2l7Or-04
2.
871-2E-05
I.
9845F.-04
-I.
2721E-04
1.1857E-0-
-2.5587E-05
K3WC
K0VK
R0WC
fKM-K
h3W<
FST
R3 W <
3)
4)
5)
6)
7)
COEFF
i >
2
5
7
1
3
1C
9
-1.8310E+00
ROW<
-3. 1
H3'/(
hOWC
HOIJ{
F3W<
K3WC
ADJ
R0W(
-4.2
W.:<
-9.5
R0W<
R0WC
ROW<
F0l-,'(
K0W(
2>
613E-
3>
4)
5)
6)
7)
DATA
1 )
2)
2
01
2
6
9
1
3
-0802E-04
.6496E-04
.9182E-04
•5718E-04
.2370E-04
-5.
1.
-1.
1.
7725E-05
8359F-04
5006F-04
3330E-05
-I.
1 *
-2.
8788E-0'.
0412E-04
6126E-05
2.
-5.
6509E-05
0821E-06
•
I.OI79E-04
IFNT VAH-C0VAK MATRIX
. f nf3E+00
-2.6153E-
.7391E+00
.5700F+00
.9801E+00
.7829F.+ 00
.9419E+00
.9993E+00
C0RRELATION
0
.OOOOE+00
-2.
02
3.
-7.
2.
-1.
1.
38ME+00
5474E-01
1319E-01
2683E+00
8540F+00
A469E-01
-2.
2.
-?.
1.
-3-
^4^8E+OO
4519F+OO
3213F+00
2864F+OO
2278F-01
1.
-1.
3.
-6.
1738E+00
5716E+00
2751F-01
2789F-02
*
9.36fOF+00
I.4649C+0
1.2577E+Ort
MATRIX
-4.
6468E-01
-4.
9153F-01
I.
4320E-01
9.6514E-Ot
-4.2151E-03
,OOOOE*00
1.
3370E-01
5.
607/iE-Ol
-3.
0360E-01
6.3517E-C.
514E-0?
3)
4>
5)
6)
7)
.OOOOE+00
.OOOOE+00
.OOOOE+00
•OOOOE+00
.OOOOE+00
-1.
2.
-4.
5.
6838E-01
7450E-01
2536E-01
9095E-02
-4.
3.
-5.
6294F-01
^941E-Ol
1604F-02
1.
-1.
4fr60F-01
lb84F-OP
3.9229E-0
238
-------
RUN 5
•MULTIPLE KEOnESSION PR0yKAM
AMFAM YMEAN
8.373793E+02
7.915517E+02
7. 732069E+02
-2.238621E+02
-I.233448E+02
-2.900000E+01
OPSFKVFD
1.9100000E+02
1.4500000E+02
2.7400000E+02
2.2800000E+02
1.5600000E+02
7.0IOOOOOE»-02
9.50000COE+01
1.38GOOOOE+02
2.3900000E+02
1.4500000E+02
1. 1200000E+02
1.2900000E+02
l./:900000F + 02
P.2POOOOOF+02
2. b900000?+02
1.7300000E+02
3.310GOOOE+02
3.2700000E+02
3.R30000GF*02
2.2900000E+02
2. 5500000E+02
2.570GCOOE+02
2.230UUOOE+02
2.9700000F-I-02
3. 7400000E*02
3. 5300000E+02
I.8100000E+02
I.8200000F+02
1.0200000E+02
CALCULATED
1 . 5^28 175F+02
2.5583495F+02
1.813I787E+02
1.8 44400 5E+ 02
2.6756554E+02
4.4090838E+C2
2.2602637F+02
.9336302E+02
.7761720E+02
.7333390F+02
.7742534F+0?
2.5273370E+02
. 67f 7840F+0?
.7A29530F+0?
2. 69£f9 1 bE*02
2.3298209P+G2
3. 1633318F+02
• . 3. 1899459E + 02
1 .3324874F+02
2.25447R2F+02
3.2072278E+02
2.7093^27E>02
1.94577f 1E+02
3. 85251 64E+02
2. 41 31 63 6E+ 02
3. 4106736E + G2
1.703
2. 1924171E+03
3.09537*8E>03
2.8058874F+03
4.0869005E+03
5.4963314F+03
3. 4286303E+03
1.9572336E+03
2.2289837E+03
2.8684380E+03
6. 4435959F-»-03
2. 086599 IF+03
7.0191212F+03
1.0077 5 16E+03
2.0812487F-I-C3
3. 1956933F+03
2. 1666915E+03
3. 599037 1F+03
3.6791603F+03
6. 1728024E+03
4.5704200E+03
1.9243626E+03
6.2185045E+03
1.3317221F+03
3.29851f>3F*03
A. lf.93351E+03
2.3717083E+03
2.894Q653E+03
239
-------
RUN 5 (Continued)
.DEGREE
. REGKFSSI0N.
. r.EMAI-NDEK •
. T9TAL
£F FKFE.
7
21
28
ii.-V 55 F SQUARES.
1.555389E+05 .
2.575399E+05 .
4. 130788E+OS .
VARIANCE ESTIMATE .
2.221985E+04
I. 22638 1E+04
MULIIrLF L0KHELATI0N COEFFICIENT - 6.136250F-01
CONSTANT TERM(INTERCEPT) - 2.442241F+03
tr?FFICIENTS
1.01£4109E+00
49P5453E+00
0713257E+00
I280M6E+00
231b423E+00
7174977E-01
5965652E-01
EST flF SD
3.5043322F+00
5.0824137F+00
1.7772377F.+ 00
1.5910638E+00
7.0851822F-01
1.2431693E+00
2.1656701E+00
F RATIOC 7* 21 DEGREES 0F FRF.EDOM>= 1 .81 1823E+00
L^PER RT HAND P0RTIQN 0F MATRICES 0MLY
INVEhSE ADJ DATA MOMENT MATRIX
R5WC 1) 1.0013E-03 -5.7038E-04
-5.8283E-05 -6.8153E-05
ROisfC 2) 2. 10*63E-03 -1.7370E-04
4.6877E-04
h0W( 3) 2.5755E-.04 -1.5066E-04
R0VK 4) 2.0642E-04 1.9192E-05
R0WC 5> 4.0933E-05 -5.3005E-05
R0W( 6) 1.2602E-04 5.1162E-05
R0WC 7) 3.8244E-04 1
FST C0EFFICIF.\'T VAh-CBVAR MATRIX
M5'.-'< 1) 1.&PBOF+01 -6.9951E+00
-7.1477E-01 -8.3581E-01
FOWC 2J 2.5831E+01 -2. 1302E+00
5. 7489E+00
R0VK 3) 3. 1586E+00 -1.8476E+00
K0W< 4) 2.5315E+00 2.3536E-01
R0WC 5) 5.0200E-01 -6.S004E-01
R0WC 6) 1.5455E+00 6.2744E-01
R0W< 7) 4.6901E+00
ADJ DATA C0RRELATI0N MATRIX
R0WC 1) l.OOOOE+00 -3.9275E-01
-1.64Q7E-01 -1.1013E-OI
R0WC 2) .OOOOE+00 -2.3583E-01
. 5.2230E-0
R0WC 3) .OOOOE+00 -6.5341E-01
R0W( 4) .OOOOE+00 2.0878E-01
R0W( 5) .OOOOE+00 -7.3800E-01
R0WC 6) .OOOOE*00 2.3305F-01
R0W< 7J .OOOOE*00
1.9936F-05
-2.9351E-04
1.5830E-05
-1.2179E-05
-4.2778E-05
2.4449E-01
-3.5995E+00
1.9414E-01
-1.4936E-01
-5.2462E-01
*
3.9257E-02
-4.4513E-01
1.5417E-01
-7.5512E-02
-3.4190E-01
1.5617E-04
-7.2585E-05
-5.8350E-05
-1.1680E-04
•
1.9152E+00
-8.9017E-01
-7.1560E-01
-1.4324E+00
3.4350E-01
.
-2.4720E-01
-3.2389E-01
-4. 1571E-01
3.4946E-OS
1.352IE-04
1.8883E-05
4.2857F-01
1.6582E+00
2.31S8F.-OI
•
I.7261E-01
2.6244E-01
6.0168E-02
240
-------
RUN 6
MULTIPLE REGRESSION PROGRAM
XMFAN
YMF AN
.363793E+0?
9.8P7349E+00
9.074578E+00
1. 18747CF+01
1.333941F+01
2.387S86E+02
7.751724E+01
1.065517F.+ 01
-1.413793E+00
OBSERVED
1.9100000F+02
1.4500000E+02
2.7400000F.+ C2
2.2800000E+02
1.5600000E+02
.7.O100000E+02
9.5000000E+01
1.3800000E-I-02
2.'3900000E+02
1.450COOOE+02
1. 1200000E+02
1.2900COOE+02
1.4900000F.-»-02
2. 22CCOOOE+02
2.5900COOE+02
1.7300000E+02
3.3100000E+02
3.27COCnOE+02
3.8800000F+02
2.2900000F. + 02
2.5500000F+C2
2.5700000E+02
2.2300COOE*02
2.9700000E+02
• 3. 7400000E+02
• 3.5300COOE+02
1.8100000F+02
1.8200000F+02
1.0200000E+02
CALCL-LATED
1.4357352F+02
2.51005C7E+02
1.7048439E+02
I.8943520E+02
2.8295581E*02
4.6035979E+02
1.9938140E+02
I.8419168E+02
1.7830961E+02
1.7628336F+02
1.7648867E+02
2.6030870E+02
I.8120280E+02
1.74I7364E+02
2.6743936F+02
2.3688920F+02
3. 1849516E+02
3.29407-86E+02
1.3771054E*02
2.2283218E*02
3. 1639798E+02
2.5733025E+02
1.9565390E+02
3.8880613E+02
2*. 49 7931 7E+ 02
3.31S8829E+02
I.7504447E+0?
2. 482321 5E+02
1.5122622E+02
EST VAR
5.2*50875F+03
2.9747375E+03
1.8271849E+03
3.5001995E+03
3.7033857E+03
4.4900138E+03
6. 336781 6E+03
3.9995152E+03
2.7319487E+03
1.7780305E+03
2.50608S8E+03
7.3608817E+03
3.4371625E+03
7.0632825E+03
1. 04260 12E+03
2.0657528E+03
3.3541844E+03
2.8196238E+03
3. 1960301E*03
3.6016420E+03
7.8090645E+03
6.8128232E+03
2.0078595E+03
6.0070475E+03
1 .8984281E+03
3.3300662E+03
3.4877042F*03
2.9347674E+03
2.8208440E+03
241
-------
RUN 6 (Continued)
.DEGREE" 0F FREE.
REGRESSION. 8
fcF'VAINDFR . 20
T0TAL . 28
SL'v flF SCARES. VARIANCE ESTIMATE
1.6B2727F+05 . 2.103409E+04
2.448061F+05 . 1.224030E+04
4. 130788E+05 .
MULTIPLE CC»-.KFLATI0N COEFFICIENT = 6.382494E-01
CONSTANT TFrvMC INTERCEPT) = 8. 7475A9F+01
CeEFr IC
5.6432fOfF+01
•2.3488160E+01
•1.31885R5E+00
9.9253QJ7E-OI
•7«8814243E-01
7. Ol 4^8 ME- 01
EST OF SD
6.5450923F+01
1.0449889E+OP
3.7f26343F+01
2.4915874E+01
8.3700842H-01
1.3187024E+00
2.2304178E+00
2.6216908E+00
F RATI0C 8* 20 DEGREES 0F FREED3M>= 1.718429E+00
UPPER RT HAND PORTION OF MATRICES 0NLY
INVERSE ADJ DATA MOMENT MATRIX
haW( 1> 3-4998E-01 -2.0495E-01
8.0006E-03
4.6639E-04 -1.3135E-03 3.8213E-03
K0MC 2) 8.9214E-01 -1.6692E-OI
8.2054E-03 -I.3965E-03
K0WC 3) I.1566E-01 -6.5065E-02
-1.9209E-03
R0WC 4) 5.07I8E-02 -4.5072E-04
K0WC .5) 5.7236E-05 -7.0654E-05
R0WC 6> 1.4207E-04 5.5719E-05
R3UK 7) 4.0-C-42E-04 -3.7772E-05
R9l-.:< 8) 5.6I53E-04
EST C0FFFICIFVJT VAR-C0V/AR MAThlX
K3KC 1) /4.2833F + 03 -2.5086E+03
3.9091E-02
5.367IE-04
5.9846E-04
-4.5807E-05
1.O5C1E-04
9.7930F+01
5.7083E+00 - 1.6073E+01 4.6774E+01
K0W< 2) 1.0920E+04 -2.0431E+03
I.0044E+02 -1.7094E*01
RSWC 3> 1.4157E+03 -7.9641E+02
-2.3512E+01
R3WC 4) 6.2080F+02 - 5. 5169E>00
K0W< 5) 7.0058E-01 -8.6483E-01
K0WC 6) 1.7390E+00 6.8202E-01
F«0K< 7) 4.9748E+00 -4.6234E-01
R0WC 8) 6.8733E+00
4.7843E+02
6.5695E+00
7.3253E+00
-5.60f9E-01
1 .2879E+00
1.4299E-02
5.5051E-04
•3.Q573E-04
3.3340F-03
•9.2ll'7F-05
6.8523F.-04
•7.8406F-04
•4.5553E-03
4.5525E-04
I.7503E+02
6.7384E+00
•3.7422E+00
4.0809E+01
•1.1275E+00
•8.3874E+00
•9.5971E+00
•5.5758E*01
5.5725E+00
242
-------
RUN 6 (.Continued)
ADJ DATA
F0WC 1)
6.6143E
R0W( 2)
4.3092E
R0WC 3)
-2.3835E
F0WC 4)
K0WC 5)
FOKC
R0WC
CBRRELATI0N MATRIX
I.OOOOE+00 -3.6678E-01
02-1.1013E-OI
.OOOOE+00 -5
-0 -6.2394E-02
OOOOE+00 -8.4951E-01
-0
R0VK
6)
7)
8)
3.9766E-02
2.7259E-01
1962E-01 1.8377E-01
1.0733E-OI -1.531 Of-01
7.7039E-02 -6.9644E-02
2.0860E-01 -7.5421E-02 -4.644QE-01
.OOOOE+00 -2.6454F-01 2.2295F-01
.COOOE+00 -7.8353E-01 -3.0034E-01
.OOOOE+00 2.3188E-OI 3.7251E-01
.OOOOE+00 -7.9067E-02
.OOOOE+00
7.3433E-01
-5.1383E-01
8.5308E-02
243
-------
„;.- ASARCO
Data Set 3
... ASAR 3 ' '
ASAK3 09137EST 12/02/74
100 131.820, 1070,820.850. 730.770.690.700
101 139,820. 1030, 8?0, 880, 770,770.735*730
102 24*825,1080.840.920.800,815.795,800
103 48,820, 1040,840,910,800,815,745,800
104 51.825, 1000,820,860,790,780,750,765
105 I 56,830,I 100,830.900.760,780, 755,780
106 95,830, 1000,830.860,795.780.740.730
107 30,830, 1015,830,870,770,770,750,755,
108 138,830, 1070,845.930.805.840.785,775
109 47,825, 1045,845,930,825,830,820,815
HO 259,825. 1 120,840,940,800,835,795,810
111 145.820.1075.845.945.835.840,825.815
112 47.820.1030.840.910*815,815,810,810
113 DELETE
114 38,820.II05.H40.930.790.800.740.760
115 37,820. 1000.835,880,780,800,775,760
116 41.8^0, 1055.845,9'0,810.820,795,780
118 112,615.1100,850,V55.820,850, 825, 820
119 43, 815, 990, 840, 89 3, 820, 800, 735, 800
I2O 129,830,IOOO,850,°00. 790.790,745. 750
121 164,804.914.822.852. 736. 722. 684, 678
122 43, 835, 1110, 830, t:65, 720, 735, 720, 7 ~5
123 60. f(?5, 1000.810.840.735,735,720.720
124 27,830.1 100.820.855.725,735,725, 725
125 100,835, 1055,825.850,745,755,730,730
126 100, 8?5, 1070.840,915,790,780,760,765
127 43,835, 1015,830,860.780,785,765,755
128 222,825, 1070,830,955.825.790.770,755
13O 259.820, 1055.835.915. 780.780, 760.755
131 128,815,960. 835,885,790. 780. 770. 760
133 173,815, 1075,835,895,775,800,765,770
134 102*810, 1075,845,940,810,800,780,775
135 311,830, 1060,830,910,770,790,770*700
136 48* 8 15, 980. 835, 890, 790, 790, 775, 770
137 3?7, 830, 1010, 835, 870* 76 i, 785, 750, 765
138 4ts8, 830, 1110, 850* 930* 800* 815* 765, 770
139 229* 8 10,1000,840,890, 825,840, 815, ROO
140 92* 8 15, 1070, 840, 925, 800, 805. 785. 795
141 71,830,1070,820,865,750,760,730, 730
142 95,820. 1070,840,940,810,770,775,765
143 I 17,820, 1070,840,890,825,830,790,785
144 67*820*970,830.855,795,800,790*800
145 105, 830, 9?5, P35, 860, P. 10, 805, 760, 7tv
147 107*835*1005, 845*885*760*765*740*745
148 60.835. 1 100.845.950.810.800.765* 755
149 33*835. 1030, 850,900, 790, 815, 790, 7«5
150 48,830, 1030,850,900, 785,785,760,765
151 ?23,8?5,1090,840.895.785.b05,770,7HO
152 IO7.8?0.1005,840.b90.790.frOO, 785, 800
153 48*830,1030.«40,890,765,T(5,735,735
154 202,825, I095,R40,955.825,825,785, 775
155 102,8?0, 1065,840* 920.800.800.7KO,790
156 146.820.980.830,870.770.775,765. 755
. 157 67.820, 1000,835,875,770.775,765,765
158 297,820.1010.835.740.770,7&0. 765. 750
159 1 17,815,1030. K40.73S, 785,790, 760. 750
160 164,835, 1045, B40, K9Q, 760.775, 740,745
161 122,835,1035. H 50. 945. til 5. hi 5. 760. 755
162-85.8?0,lOftO.K40.900.790.815,7RO,780
163 399,820,1065.835,P90*765,750,715,710
164 89,825, I 050,K50,9?0.810.hOO,7K5, 775
165 228, 8?S, I Of-5, 840, 910. 785, 775,750, 750
166 140, 82S. 1000,845, B75»7hO, 785. 725. 725
167 294,835.1030.82b.R60. 755.790.750. 740
168 9?. 835.955, 840, 865,780, 775,730,725
244
-------
»TATiimi ft ft ri
IW^OLf* Mr *
MAN • KPk
WADJUSTr'D
SUM<(KCt>.
IMADtJI '5TCD
"
(BIASED) VARIANCE •
XlflJt)**ff)M
AwJUSTCO STANDARD DEVIAT18N • SChT(AVR) • *•••
KURT3SI3 •
•
• .
CMriDEMcr
LIVD.
»O.OI
• 5.0*
99.01
1
•
10
II
• te
13
14
15
1*
IT
18
1*
20
21
22
23
24
f S
26
67
28
2*
30
31
32
33
34
35
36
',f
38
• 3* •
40
41
42
43
•44
45
46
47
'48
49
50
SI
52
S3
54
55
56
57
58
5*
•60
61
62
63
64
SIMC tXCI >-XBR>**4)/(N*UVK»*2>
t-SIOED
LCUEH
LIMIT
108
104
• 7.
.
131
13*
24
48
SI
156
• 5
30
138
47
83*
145
47
38
37
41
112
43
12*
164
43
60
27
100
100
43
«22 ,
259
128
173
1152
331
48
327
388
229
92
71
95
117
67
105
107
60
33
48
223
107
48
202
102
I4A
67
297
117
1 * *
172
85
3*9
8*
22«
1*0
294
*2
CCNriDCNCr INTFhVALS 8N
Mr<«
UPPER
LIMIT
.40 146.32
.66 ISO. 06
18* . IST.53
24 1.538
27 3.077
30 4. 615
33 6.154
37 .7.6*2
38 ».23I
"41 IO.76*
43 IS.3B5
43 15.385
43 IS. 385
47 18.462
47 18.462
48 24.615
48 24.615
48 24.615
48 24.615
51 26.154
60 29.231
60 29.231
67 32.308
67 32.308
71 33.846
85 35.385
8* 36.923
98 40.000
92 4O.OOO
' 95 . 43.077
•3 . 43.077
ICO 46.154
100 46.154
IC2 49.231
102 49.231
105 50.76*
107 53.846
.107 53.846
112 55.385
117 58.462
117 58.462
122 60.000
128 61. SDR
12* 63.077
131 64.615
138 66.114
139 67.'692
I4O 69.231
145 70.769
146 72.308
156 73.846
164 76.923
164 76.9F3
173 78.462
202 BO. 000
222 B 1.538
223 83.077
228 84.613
229 86.154
23* 87.692
259 89.231
294 90.76*
2*7 92.308
327 93.8*6
331 75.383
388 *6.9?3
39* 98.462
127.36
103.50
375.00*
8128.4
•0. 158
Bf ST. 4
• O'CTO
I .fiSST
4.0028
MFAN t STANDARD DFVIMIHN
STANDARD
L0WFR
LIMIT
7*. 394
T7.403
73.748
POR A NBRHAL
12.768
13.470
14.19*
14.954
16.002
16.271
17.096
17.661
17.661
17.661
18.626
18.826
19. 124
1*. 124
19.124
1*. 1?4
20.037
22.927
22.9P7
25.327
25.327
S6.756
32.055
33.646
34.859
34.859
36.0BB
36.088
38.168
38.168
3*. 009
39.009
40.282
41.136
41.136
43.289
45.462
45.462
47.648
50.2HI
50.720
51.598
54.6*1
55.097
55.532
57.696
SB. 127
62.369
65.661
65.661
69.226
79.429
85.118
85.371
86.597
86.833
89.038
92.628
96.666
96.90*
*8.39*
98.7*9
99.794
•9.8*0
DFVIATI'N
UPPCR
LIMIT
106.64
no. os
117.25
OlSTKIF'. TIHN
11.22*
10.3*4
*.5B4
8. BOO
8.310
T.040
6.327
2.277
2.27T
2.277
0.364
0.364
•5.491
-5.491
-5.4*1
-S.4*l
-6.117
-6.304
-6.304
-6.*8I
-6.981
-7.090
-3.329
-3.277
-5.141
-5.1*1
-6.989
•6.*f*
-T.9B6
•7.986
-10.221
-10.221
-10.487
-12.710
-12.710 '
-12.09*
-13. COO
-13.000
.-12.352
- 11.057
-12.357
-13.017
-11.493
-IP. 546
-13.699
-13.073
-I4.IRI
-11.477
-11.262
-11.262
-*.236
•0.571
3-579
2.2»4
I.9HI
O.A79
1.346
3.39(1
5.897
4.5*6
4.753
3.364
P. Ml
1.799
0.1531 • KPLwnf-CKPV-SrtlhNBV STATISTIC
- ••—-—245
-------
DfSCKIPTIVE STOUSTICS
XI
IN»TJ'.ISTE3 fPlA^ED) VAMANCl »
S' -i. **2>N • t'VH
• «»*•••*••»••
ADJUSTED STANDARD DEVIATION » SGKTCAVR) ....<
NEWNESS . sm-xefi>*>3>/cN*uvK**i.s> ..
•KURT8SIS « SUN((XCI>-XPR)»*4>/(N*UVR**2> ....
B-SICED C6WF«ENCE INTERVALS ON
MEAN
CONFIDENCE LOWEh UPPER
L. VEL LIM;T LIMIT
90.01 828.68 825,72
9S.OX BC3.38 826. OS
99.01 821.78 826.62
t 820 804 I.S38
e 820 CIO 4.615
3 823 810 4.6IS
4 820 8]S 15.385
S 825 BIS 15.385
6 830 BIS IS.3BS
7' 830- 815 IS. 385
B ' 830 BIS 15.385
9 830 SIS 15.385
IO 825 015 IS.38S
II 825 820 44.615
12 820 820 44.6IS
13 820 880 44.615
14 820 320 44.615
1} 820 820 44.61 S
16 820 8*0 44.615
. 17 815 820 44.615
18 815 82O 44.415
19 830 820 44.615
CO 804 820 44.615
21 635 820 44.615
£2 825 820 44.615
23 830 820 44.615
24 83S 020 44.615
BS 825 . 820 ' 44.615
£6 835 820 44.615
27 825 B£0 44.615-
28 820 820' 44.615
£9 815 B2O 44.615
30 115 . 825 63.077
Jl 810 625 63.077
32 830 825 63.077
33 815 82S 63.O77
34 83O 825 63.077
35 83O • 825 63.077
36 810 885 63.077
37 815 825 63.077
38 830 825 *3.O77
39 8^0 825 (3.377
40 820 Si 5 63.077
41 820 8£5 S3.077
42 830 430 B3.C77
43 835 830 83.077
44 335 830 83.077
45 835 830 83.077
46 830 830 R3.077
47 825 830 P3.077
48 P20 830 83.077
49 830 630 83.077
50 825 830 (0.077
51 S^O 830 M.077
52 8?0 • 830 83.077
53 6PO 830 83.077
54 820 830 83.O77
55 815 1*35 98.462
56 835 B35 98. 4'?
57 835 ••••»•••••••
. 824. PO
. 875.00
«. ArtA
. JI.UUU
52.349
. S3. 180
'-0.23593
O-4OOA.
MFAN * STANDARD DFVIATIBN
STANDARD
LOWER
. UIHIf
6.3715
6.2117
5.9184
FOS A NORMAL
0.28O
2.57J
£.573
1C. 3*7
10.347
10.347
10.347
10.347
iO.347
1C. 347
28.218
PS. 218
88.218
28.218
e-j.eie
28.218
28.218
28.218
28.218
C8.2I8
28.218
28.?I8
P8.2IB
P.B.f.18
28.PIB
E8.2IB
P.8.2IB
28.218
28.218
S4.3SI
54.351
• 54. 351
54.351
54.351
54.351
54.J51
54.351
54.351
54.351
54.351
54.351
78.667
78.667
78.667
78. 667
78.667
78.667
78.667
1C. 667
7R.667
78.667
78.667
78.667
7R.667
93.064
93.064
93.064
93.064
93.P64
93.064
93.O64
93.CI64
93.064
V3.0f 4
DEW! ATI -N
UPPF*
LIMIT
8.SS84
8.8321
9.4098
DISTRIFVTIPN
-1.259
-2.042
-2.048
-5.O37
-5.037
-5.037
-5.037
-5.037
-5.037
-5.037
-16.397
-16.397
-16.397
-If. 397
-I6.3«7
-16.397
-J6.397
-16.397
-16.397
-16.397
-16.397
-16.397
-16.397
-16.397
-16.397
-16.397
-16.397
-If.S^V
-I6.3V7
-S^Jl'6
-8. 726
*ft.7?6
-4.7?6
-0.726
-8.7P6
-8.7?6
-0.7?6
-8.7?6
-R. 7?^
-8.7P6
-8.7P6
-4.4IC
-4.410
-4.410
-4.410
-4.410
-4.4IO
-4.410
-4.4IU
.4.410
-4.4IC
-4.AIO
-4.41O
-4.410
-5.39H
-5.39S
-5.39H
-5.3°^
-5. 31*
•v.5.3'"!
-• 39R
-S. .t***1
-5.3»8
-5.3'H
246
-------
DESCRIPTIVE STATISTICS FOR X2
NUMBER OF OBSERVATIONS ................................................ 64
MEAN - XBR = SUM(X(I))/N .......................................... 1040.9
MEDIAN [[[ 1052.5
RANGE [[[ 206 . 1
UNADJUSTED (BIASED) VARIANCE
SUM ((X(I)-X3R)**2)N = UVR ................................... 2077.2
UNADJUSTED STANDARD DEVIATION = SORT( JVR) = S ..................... 45.57
ADJUSTED (UNBIASED) VARIANCE -UVR*N/(N-1) = AVR .................. 2143.7
ADJUSTED STANDARD DEVIATION SORT(AVR) ............................ 46.31
MEAN STANDARD)EiATION
CONFIDENCE LOWER UPPER LOWER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.0% 1031.4 1050.4 41.3 55.9
95.0% 1029.5 1052.3 40.4 57.62
99.0% 1025.9 1055.9 38.35 61.7
-------
DESCRIPTIVE STATISTICS FOR X3
NLMBEK 0F OBSERVATIONS .'....' 64
MEAN = XBR = SUMCXCI»/N , 836.75
MEDIAN 840.00
RANGE 40.000
UNADJUSTED CBIASED) VARIANCE =
SUMCCXCI>-XB;0**2)N = UVR 78.875
UNADJUSTED STANDAKD DEVIATION = SCF.TCUVF;) = S 8.8812
ADJUSTED (UNBIASED) VARIANCE = UVR*N/(N-1> = AVR 80.127
ADJUSTED STANDARD DEVIATION = SORTCAVR) 8.9514
i - '
SKEWNESS = SUM«XCI>-XBR>**3>/CN*UVR**1.5> -0.67846
KURT0SIS - SUM-XBR)**4>/ 3.1771
•
2-SIDED CONFIDENCE INTERVALS GN MEAN * STANDARD DEVIATION
MEAN STANDARD DEVIATI0N =
CONFIDENCE • L0WER UPPER L0WER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.OA
95.OZ
99.0%
834.88
834.51
833.78
838.62
838.99
839.72
7.8209 10.505
7.6248 10.841
7.2647 11.550
FOR A N0RMAL DISTRIBUTION
****************************************
ID
HO
r>o
^o
50
X3
INPUT
DATA
8/»5
8f?P
835
840
850
0.2042 =
Or.DERED
DATA
630
835
840
840
845
850
CU-XIT-ATIVF
PF;.CF<\!TM-F
27. A9P
/(3.077
75.385
87.692
98.462
CUMULATIVE
NORMAL
22.540
^2.250
64.172
64.172
8£.164
93.059
STATISTIC
248
DIFFERENCE
-5.152
-0.827
-11.212
-11.212
-^.528
•-.402
-------
DESCRIPTIVE STATISTICS FOR X4
NU"rER 0F OBSERVATIONS 64
MEAN = XBR = SUM-XBR>**2)N = UVR - 1689.2
^'ADJUSTED STANDARD DEVIATION = SGRTCUVR) = S ' 41. 10C
ADJUSTED CUNBIASEDJ VARIANCE = UVfv*N/CN-1 > = AVR 1716.1
ADJUSTED STANDARD DEVIATION = SQRT 41.425
NEWNESS = SIJM-XBR)**3)/(N*UVK**1.5> -L377S
KUhTOSIS = SUMtCXCI>-XBR>**4>/CN*UVK**2> 6.9509
C-SIDED CONFIDENCE INTERVALS 0N MEAN * STANDARD DEVIATION
MEAN STANDAHO DEVIATION
CONFIDENCE LOWER UPPER LOl-'ER UPPER
LEVEL LIMIT LIMIT LIMIT . LIMIT
90.07 883.26 900.55 36.194 48.61f
95.07 881.56 . 902.25 35.286 50.171
99. OX 878.15 905.66 33.620 53.453
FC-R A NORMAL DISTRIBUTION
^****************W ********************
i
X4
INPUT BRDF.f^ED Cl.WLATI VF C-iWl'LATIVF
DATA DATA HEi\CE\TAGF NORMAL DIFFEI-FM'.F
1O 930 8^-0. 'PC. CCO P2.059 ?
20 852 375 32.30« 34.159 l
:
-------
DESCRIPTIVE STATISTICS F0R XT
NUMBER flF 0BSERVATIONS 64
MEAN = XBR = SUM-Xl?rO**n>N = UVPv ... 699.11
INADJUSTF.D STANDARD DEVIATION = Sl.HTCl'VF.) = S ' 26.441
ADJUSTED (UNBIASED) VARIANCE = UVK*N/CN-1> = AVR 710.20
ADJUSTED STANDARD DEVIATION = SGHTCAVIO 26.650
SXEWNESS = SUMCCX-XBR>**3>/CN*l;t.f<**1.5> -0.46017
KLIRTOSIS = Si:M(CX(I)-XPR)**4)/(N*l'V.i:.**£>) 2.7862
2-SIDED CONFIDENCE IWTFF.^'ALS RN MEAN & STANDARD DEVIATION
MFAN STANDARD DEVIATION
CCNFIDFNCE LOUFR UPPEK LOWER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.0%
95.07
99.07
780.39
779.30
777.10
791.SI
79&.C-!
79/i. SO
23.284 31.276
22.700 32.276
21.628 34.387
F0R A NORMAL DlSTRIPUTIBN
^ ********** **<•*
10
£0
30
'•0
ISO
73'.
775
810
OKDFKED
DATA .
760
770
7':5
795
810
825
',:• XILATIVF
U-.923
fH.077
o3.077
9f- .923
CiMULATIVF
NORMAL
16.506
27.471
48.573
63.287
81.656
92.857
0.091A = KflLMOGORflV-SMlF.Nf!'- STATISTIC
250
DIFFFF.FNCE
-0.417
-3.PS8
2.420
0.210
-1.4£1
-4.066
-------
DESCRIPTIVE: STATISTICS FOK X6
NUM6ER OF OBSERVATIONS 64
MEAN = XBR = SUMCXCI»/N 791.36
MEDIAN '. 790.00
^
RANGE 128.00
INADJUSTED (BIASED) VARIANCE =
SUM((X(I)-X8K)**2)N = UVR 727.82
INADJUSTED STANDARD DEVIATION SORT(UVR) = S 26.978
ADJUSTED (UNBIASED) VARIANCE = UVF.*N/(N-1 > = AVR " 739.38
ADJUSTED STANDARD DEVIATION = SOF.T(AVR) 27.191
NEWNESS = SUM((X(I>-XBR)**3)/(N*UVN**1.5) -0.17163
KURTOSIS = SUMC(XCI)-XBR)**4)/(.\'*llVFv<*2) 3.0185
2-SIDFD CONFIDENCE INTERVALS ON MEAN « STANDARD DEVIATION
MEAN STANDARD DEVIATION
CONFIDENCE L0WER UPPER LOWER UPPER •
LEVEL LIMIT LIMIT LIMIT LIMIT
9O.07
95.07
99.07
785.69
784.57
782.33
797.0?
798.15
300.39
23.757 31.912
23.162 32.932
22.063 35.086
FOR A NORMAL DISTRIBUTION
10
bC
X6
INPUT
DATA
830
800
830
825
800
0.0785
DATA
770
LATIVF
790
80C
£15
835
K0L»nGOROV-i.;-'Ii'
251
20.000
38. ^-62
53.846
CI.'Ml'LATIVF
NORMAL
?.\ .608
4R
62
80
94
006
467
769
575
STATISTI C
DIFFl'RENCE
1 .608
-4.655
-5.840
-6.764
-3.8^6
2.267
-------
DESCRIPTIVE STATISTICS F0R X7
NUMBER 0F OBSERVATIONS - .64
MEAN = XBR ~ SUMCXCI»/N 76~ll94
MEDIAN *. « . i .' , . . 765. 00
RANGE » 141. 00
UNADJUSTED (BIASED) VARIANCE =
Si!MC-XER)**3)/CN*UVR**I.5> -0. 1S11 1
KURT0SIS = SUMCCXCI)-XER)**4)/CN*UVR**2>. «. 3.1532
2-SIDED CONFIDENCE INTERVALS 0N MEAN ^ STANDARD DEVIATION
•
MEAN. STANDARD DEVIATION
C0NFIDENCE LOWER UPPER LOWER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.07
95.0%
99.02
755.78
754.57
752.15
768.09
769.30
771.73
25.765 ' 34.608
25.119 35.715
23.933 3t5.051
FOR A NORMAL DISTRIBUTION
% *********************************
10
20
30
40
50
f-C
X7
INPUT
DATA
820
68 *
765
790
7«5
785
DATA
730
750
760
770
785
310
PERCE'NTAGF
15.385
36.923
47.692
64.615
31.538
92.308
NSfxMAL
13.940
34.281
47.381
60.773
78.291
94.843
DIFFERENCE
-1.445'
-2.642
-0.312
-3.842
-3.247
2.535
0.0332 = KOLMOGOROV-SMlRNaV STATISTIC
252
-------
DESCRIPTIVE STATISTICS FOR X8
NLME-EK CF OBSERVATIONS ' 64
MEAN = XBR = SUM(XCI))/N 761.92
MEDIAN 762.50
RANGE -» i .*.' 142.00
UNADJUSTED (BIASED) VARIANCE =
Sl'M((X(I)-XBR)**2)N = UVR 920.51
INADJUSTED STANDARD DEVIATION = SGRT(UVR) = S '..... 30.340
ADJUSTED (UNBIASED) VARIANCE = UVR*N/CN-1) = AVK 935.12
ADJUSTED STANDARD DEVIATie.N = SQRT(AVR) 30.580
SXEWVESS = SUMC(X(I)-XBR)**3)/(N*UVR**I.5) .. -0.27644
KURT0SIS = .SUM((X(I)-XBn)**4)/CN*UVR**2) * 2.8667
I
2-SIDED CONFIDENCE INTERVALS C?N MEAN & STANDARD DEVIATI0N
MEAN ' STANDARD DEVIATION
CONFIDENCE LR'.-.'ER UPPER L0VF.R UPPFR
LI.VEL LIMIT LIMIT LIMIT LIMIT
90.0% 755.54 768.30 26.'718 35.888
95.0% 754.28 769.56 26.048 37.036
99.07. 751.77 772.07 24.818 39.458
FOR A N0RMAL DISTRIBUTION
****************************************
X8
INPUT CKDERED CWL-LATIVE CUMULATIVE
DATA DATA PERCF.N'TAGF. NORMAL DIFFERENCE
1C- 815 730 20.000 14.827 -5.173
r'o 678 750 0£.3C^ 34.832 C.5&4
'0 770 760 49.?M /)7.494 -1.736
'•0 785 770 63.077 60.417 -2.660
f.-i 775 755 78. '.f 2 77.478 -0.983
(•':: 775 810 93.P^.A 94.205 ' 0.359
0.0827 - KOLMHCOKOV-SMIKNGV STATISTIC
253
-------
MULTIPLE REGRESSION ANALYSIS
ASARCO PLANT - START-UP DATA
254
-------
100
10J
102
103
104
JOS
106
107
108
109
110
11 1
112
5.0, 12.5
6.0, 51.5
5.0,6. 5
7.0,16.5
5.0,9-0
9.0, 5-5
8.0,37
6.5, ICO
9. 0, 1 {•
4.0, 13.5
7.0, 16.5
6.0, 10.5
5.0,9.0
ASARCO
Data Set
6TA 1
MLLTIPLF REUKESSI7IN PR0GKAM
XtfEAN
2.300000E+OI
YMEAN
6.346154E+00
OBSERVED
5.0000COOE+00
6.0000000E+00
5.0000000E+00
7.0000000E+00
5.00000GOE+00
9.0000000E+00
8.0000000E+00
6.5000000E+00
9.000COOOE+00
4.0000000E+00
7.00000COF*00
6.0000000E+00
5-OOOOOOOE+OO
CALCULATED
6.3005722E+00
6.4152616E+00
6.2829276E+00
6.3123352E+00
6.2902795E+00
6.2799869E+00
6.3726206E+00
6.7343335E+00
6.3108648E+00
6.3035129E*00
A.3VC3352F+00
6.2946907E*00
6.2902795E+00
EST VAR
2.452320AE-01
2.8661375E-01
2.7467671E-CI
2.30907f.OE-01
2.6124755E-01
2.8051260F-0!
2.2411012E-C1
2.5243762E+00
2.4125303E-01
2.3090760F-01
2.5398586F-01
2.612A755E-01
255
-------
REGRESSION.
REMAINDER .
TOTAL
.DEGREE 0>F FREE.
1
1 1
12
SSJM 0F SQUARES. VARIANCE ESTIMATE
1.808564E-01 . 1.808564E-01
3.05I145P+01 . 2.773768E+00
3.069231F+01
MULTIPLE CORRELATION C0EFFIC1ENT = 7.676304E-02
TFRMCINTERCEPT) 6.263813E+00
EST ?,F SD
1.1516670E-02
COEFFICIENTS
lr.-Fi .-.I H»=\L.
DEGREfS OF
IrtN CF MATRICES ONLY
IMVf :-£E AL.J DATA M^MFNT MATRIX
i.rv< 1) /I.7B17F-OS
F6T CflFFFICIENT VA»,-C3VAK MATRIX
\-C.V < 1> 1.3263E-OA :.
ADJ DATA CeF'-f^ELATIOM MATRIX
r.r.VC 1) l.OOOOF+00
DESCRIPTIVE STATISTICS FOR Y1
NUMBER 9F 0E-SFRVATI0NS 13
MEAN = XER = SUM-XER>**?>N = UVR 2.3609
UNADJUSTED STANDAHD DEVIATION SLRT(LJVK) = S 1.5365
ADJUSTED (UNBIASED) VARIANCE = I Vr.*N/**4)/(N*l'V'h**2> 2.0771
256
-------
CONFIDENCE
LEVEL
90.OX
9 5. OX
99.0%
2-SIDED C0NFIDENCF INTERVALS ON MEAN * STANDARD DEVIATION
MEAN STANDARD DEVIATION
L0WEH
LIMIT
5.5556
5.3797
4.9913'
UPPER
LIMIT
LOWER
LIMIT
UPPER
LIMIT
7.1367
7.3126
7.7010
1.2082 2.4234
1.1463 2.6400
1.0414 3.1599
FOR A N9RMAL DISTRIBUTION
****************************************
1
2
3
4
5
6
7
8
9
10
1 I
12
13
Yl
INPUT
DATA
5.0000
6.0000
5.0000
7.00CO
5.0000
9.0000
SiOOOO
6.5000
0000
0000
OOOO
0000
0000
ORDERED
DATA
4.0000
5.0000
5.0000
5.0000
5.0000
6.0000
6.0000
6.5000
.0000
.0000
8.0000
9.0000
9.0000
7,
7,
CUMULATIVE
PERCENTAGE
7. 143
35.714
35.714
35.714
35.714
50.000
50.000
57.143
71.429
71.429
78.571
92.857
92.857
CUMULATIVF
NORMAL
7. 1 19
19.997
19.997
19.997
19.997
41.432
41.432
53.832
65.867
65.867
84.946
95.148
95.148
DIFFERENCE
-0.0?4
-15.717
-15.717
-15.717
-15.717
-8.568
-8.568
-3.311
-5.561
-5.561
6.375
2.291
2.291
0.1846 = KOLM0G0R0y-SMIRNOy_SJAJISTI_C_
DESCRIPTIVE STATISTICS F0R XI
NUMBER 0F OBSERVATIONS ..
MEAN = XBF. = SL'M =. S ...
ADJUSTED (L'VEIASFD) VARIANCE - UVR*N/C\'-1> = AV'R.
ADJUSTED STANDARD DEVIATION = SCRTCAVF.1
NEWNESS = SL'M((X(I)-XBR)**3)/(N*HVh**1.5>
KURT0SIS = SUM(CXCI)-XER)**4)/CN*L'VR**2>
13
28.000
13.500
154.50
1608.7
40.109
1742.7
41.746
2.6964
9.0626
257
-------
C0NFIDENCE
LEVEL
90.0%
95.0%
99.02
2-SIDED CONFIDENCE INTERVALS ON MEAN I STANDARD DEVIATI0N
----T--
MEAN STANDARD DEVIATION
LOWER
LIMIT
7."3641
2.7730
-7.3665
UPPER
LIMIT
48.636
53.227
63.366
L0WER
LIMIT
UPPER
LIMIT
31.538 63.259
29.936 68.912
27.184 82.48A
FOR A NORMAL DISTRIBUTI0N
****************************************
1
2
3
4
5
6
7
8
9
10
11
18
13
X t
INPUT
DATA
12.500
51.500
6.5000
16.500
9.0000
5.5000
37.000
160.00
1£.000
13.500
16.500
10.500
9.0000
flhDF.RED
DATA
5.5000
6.5000
9.0000
9.0000
10-500
12.500
13.500
16.000
16.500
16.500
37.000
51.500
160.03
CUMULATIVE
PEt-.CENTAGF
7.143
14.286
f?S. 57 I
2«. 57 1
35.714
42.357
50.000
57. 1^3
71.429
7 1 . 4^9
78.571
85.714
92.S57
CUMULATIVE
NORMAL
29.495
30.327
32.^51
32.451
33.754
35.521
36.417
38.688
39.148
39. 148
58.535
71.326
99.922
PJFFEr.ENCF
22.353
16.041
3.379
-1.961
-7.336
-13.583
-IP.455
-32.281
-32.2R1
-20.037
-14.389
7.065
0.3778 = K0LM3G0RCV-SMIRN0V STATISTIC
258
-------
100
101
102
103
104
105
106
107
108
{09
110
111
112
'
ASARCO
Data Set
ASAR2
-:« 09.-51FST 11/27/74
859* 820, 1 1 20, 850, 950, 835* 860* 800* 780
2463,835, 1 010*830* 870* 790*. 780* 760* 765
P. 145* 81 6* 824* 770* 772* 704* 7 10* 674* 664
1863* 666* 822* 720* 754* 684* 692* 638* 640
1695*825* 1030,830*875* 765*775*745*740
131 1*830* 1075,825*870* 745*740,700,695
'l 577*830, 1030,845*920,790*765,735,725
101 £, 840* 1 055* 790* 750* 6 1 5, 625, 700* 820
3 107* 8 I 0* 1030, 730, 755* 635* 640* 640* 645
3092. 825* 9SO* 800* 820* 700* 690* 650* ' "0
995*835* 1065,840,890*765*755*710* . iO,
1 088* 860* 1090* f 30* 345* 725* 71 5, 690.. 68S
1009* 810* 955* 840, 86O* 770* 730, 685, 680
Reproduced from
beil available copy.
DESCRIPTIVE STATISTICS F0R
NUMBER 0F 0BSEJWAT.*0NS .,
MEAN * XBR = SUM(X(I»/N
MEDIAN
i
RANGE
UNADJUSTED (BIASED) VARIANCE
SUM((X(I>-XBR)**2)N = UVR .
... 13
•; 1678.2
1311.0
•2248.0
0.59019E+06
UNADJUSTED STANDARD DEVIATI0N = SQRT(UVfc) = S , 768.24
ADJUSTED (UNBIASED) VARIANCE = UVR*N/(N-)> = AVR 0.63937E+06
ADJUSTED STANDARD DEVIATION SQRTCAVR) 799.61
SKEWNESS - SUM((X
KURT0SIS - SUM((X(I)-XBR)+*4)/(N*UVR**2) J«.:1775
259
-------
C0NFIDENCE
LEVEL
90.OX
95. OX
99.01
MEAN
L0WFK
LIMIT
1
1282..9
1 195.0
100J.7
UPPER
LIMIT
2073.4
2161.4.
2355.6
•*, 0N MEAN * STANDARD DFVIATI0N
STANDARD DEVIATION '
LBWER
LIMIT
UPPER
LIMIT
604.07 1211.7
573.39 1319.9
520.6.9 1579.9
F0R A N0RMAL DlSTRlBUf I0K.
****************************************
1
2
3
A
5
6
7
8
9
10
11
12
13
Yl
.INPUT
' DATA
859
2463
2145
1863
1695
1311
1177
1012
3107
3092
995
1088
10O9
• 0.215-4
0RDERED
DA'iA
859
995
1O09 v
1012.
1088
1 177
131 ]
1695
1863
2145
2463
3092
3107
= K0LM0G0R0V-SMI
CUMULATIVE
PERCFNTAGE
7.143
14.286
2 1 . 429
26.571
35.714
42.857
50.000
57.143
64.286
71.429
78.571
85.714
92.857
CUMULATIVE
N0RMAL
15.281
19.645
20. 134
20.239
23.024
26.541
32.306
50.840 .
59. 141
72.:034
83.684 •
96.148 -
96.303 ;•:'
RN0V STATISTIC ''"-.«&?J
DIFFERENCE
8.138
5.359
-1.295
-8.C32'
-12.690
- 16*316
-17.694
-6.302
-5.145
0. 60S
S.,112
10.434
260
-------
DFLCRIPTI'F STATIST ICi- Fdh XI
MUM&FK PF PbbFRVAIIONi; 13
iMFAN = XBR - SIW(X< I) )/N 815.54
.MEDIAN 825.00
KANCE 194.00
LNADJL'iTFLr CBIASfD) VARIANCE
Sl'MCCXC I)-XFih>**?)N UVK 2028.7
UNADJUSTED STANDAI-D DEVIATION = IJlkTCUVR) = L 45.041
ADJUSTED (UNMASFD) VARIANCE = l'VK*N/(,M-1 ) = AVK ?197.fl
ALJIISIFD STANDARD DFVIATION - SQ.HT-XPK) **3)/CN+UVR** 1 .5) -R.7IOO
KUKT0SIS -• i>IJM«X(I )-XBh)**4)/) 9.-»331
2-Sl.DFJJ CONFIDENCE INTFKVALS f)N MEAN * STANDARD DFVIATI0N
MFAiM . STANDARD DEVIATION
C0NFIDFNCE LOV.F I-. UPf'FK L0KFH UPPFR
LFVFL ^IMIT MMIT LIMIT LIMIT
90.05
9b.OZ
99.07
792.36
787.?!
775.62
638.71
8/i3.K7
855.25
35.416 . 71.039
33.617 77.387
30.528 92.62P
FPK A NORMAL DISTRIBUTION
1
2
3
4
b
f
7
8
9
1C.
II
I"
13
XI
INPUT
I'ATA
820
R35
H\f
K30
830
H10
825
835
860
HiO
0.3761 -
PI DFRFO
DATA
(• f f.
H 10
KIO
hU
830
830
835
835
840
860
Cl'Ml'LATIVF
HFKOFNTAPF
7. 143
'f 1 . 4f>9
2 1 . A^>9
2b.S71
35.71 .'.
50.0CC
bv.OOO
f4.p«f
f4.2H6
78.571
7R.571
8b.71/i
92.8S7
Cr*H'LATIVF
NORMAL
0.071
45.298
/•S.298
50.393
53.791
b7.997
57.997
62. 1 14
f 2 . 1 1 4
6f.098
6^.098
A9.909
82.854
DIFFFRFNCF
-7.072
23.869
23.869
21.821
18.077
7.997
7.997
-2.172
-2.172
- 12.474
- IP.474
-15.805
- 10.003
STATISTIC''
261
-------
DFSCKIHTIVF STATISTICS FOK X?
Nl'MBFK OF OBSERVATIONS 13
MFAN XPK - SUM VAKIANCF =
Sl'M«X(I)-XF:**2>N = UVR 7899.3
INADJUSTFD STANDARD DFVIATI0N = SGRTCUVFO = S 8R.R7B
ADJUSTED (UNMAStfD) VARIANCF = liVR*N/ = AVR 8557.6
ADJUSTED STANDARD DFVIATI0N = SORTCAVR) . .. 92.507
SKFWNFSS = SUMCCXU >-XFR>**3>/-XBR>**4>/(N*UVK**2> • 3.1248
2-SIDED CONF1DFNCE INTFRVALS 0N MFAN * STANDARD DFVIATI0N
MEAN STANDARD DFVIATI0N
CONFIDc.NCF l.OWFK UPPE.R L0WFH • UPPFR
LF.VFL LIMIT LIMIT LIMIT LIMIT
90.07
95.07
99.OX
960.89
950.71
928.25
10S2.3
I0f2.5
1085.0
69.886 140.18
66.336 152.70
60.239 182.78
FOR A NORMAL DISTRIBUTION
****************************************
1
2
3
A
5
f
7
8
9
10
11
12
13
X2
INPUT
DATA
1 120
1010
824
822
1030
1075
1030
1055
'030
980
1065
1090
955
OKDFRFD
DATA
822
824
9bS
980
1010
1030
1030
1030
1055
1065
1075
1090
1 120
CUMULATIVF
PFRCFNTAGF
7.143
14.286
21.429
PR.571
35.714
57-. 1 43
57.143
57.143
64.286
7 j.429
78-571
85.71/1
92.857
CUMULATIVF
N0RMAL
2.298
2.419
28.844
38.678
51 . 159
59.978
59.978
59.978
69.953
73.602
77.012
81.631
88.984
0.2152 = KPLMPinOKOV-C-MKNOV STATISTIC
DIFFFRFNCF
-4.844
-11.867
7.415
10.107
15.745
2,835
2.835
2.835
5.667
2. 174
-1.560
-4.083
-3.873
-------
DFSCKIP1 M'F STATISTICS FC7K X3
NUME-.FK PF OFSERVATI'.'NS 13
MFAN XBR - SUMCX(I))/N 811.54
wlFDl AN 830.00
KAN OF '. 130.00
INAHJUSTrn V'AMANCF =
Sl*l«X-XF«IO**f?>N = I!VI-. 1316.9
IN/MUl'STFD STANpAM.) DFVIAT10.M SCKTU'»'K> = S 36,.?R9
ADJUSTEU (IHMFIASFD) VAi.'IANCF = UVI\*N/ CN- 1 ) AVR 14?6..6
ADJUSTED STANDARD DEVIATION - bOKT(AVK) 37.770
SKEl-'NEiS = SI.'M(CX(I )-XpK>**3)/CN*l'VF:**l .5) -1.1587
Kl'KTPSIS = SL'M«X( I )-XF'K'.**4)/(N*UV'K**P) 3.5?R6
?-MDFU CONFIDFNCF I\'1FKVALS ON MFAN t, STANPAhD PFVIATION
MFAN STANDARD DFVIATIRN
COMF It.F-MCF LDVFK I'HHFf. LOWFK I'PPFK
LFVFL LI.XIT LIMIT LIMIT LIMIT
90.07 79?.87 R30.?l f>8.534 57.P34
95.07 7K8.71 R34.36 P7.085 f?.349
99. Q'/ 779.54 843.54 24.595 74.6?P
FOR A NRRMAL DI SThI PI'TION
X3
INK'T OKDFRFD CIMI-LATIVF n"Ml'LATl"F
I'ATA DATA PFKCFNTAPF VPF.MAL DIFFFRFNCF
«bO 7^'0 7.143 0.76.8 -6.374
:• ^30 770 14.?86 13.57? -0.714
720 790 i'K.571 ?8.4?f, -0.146
b 830 hOO 3S.7M 38.000 ?.?85
t *?5 K^5 '!;••>. 8b7 A3.9?3 ?.l,
7 K4S t<3() //».?h6 68.750 4,
n 79(! 830 f A.Z'nf 68.750 4.465
V 7SO r.714 81.? 17 -4.497
13 840 Hbi, 9P.K57 84.57^ -8.?84
-iMll.MPV STATISTIC
263
-------
DFSCMP1 IVF. STATISTICS FOh X4
Nl.'MFFK OF 0&SFKVATIONS 13
MEAN = XBh - SUM(XU))/N 840.85
MEDIAN 860.00
KANGF 200.00
INADJl'L-TFD = AVK 4348.8
ADJUSTED STANDARD DEVIATION = SQKTCAVR) 65.945
SKEUWFSS = SLMCCXC1 )-XPR)**3J/CN*UVR** 1 . 5) -0.10921
KUhTOSIS - SUMCCXCI>-XPh)**4>/CN*UV.c.**2) 1.8710
2-SIDFD CONFIDENCE INTERVALS PN MEAN f STANDARD DEVIATION
MEAN STANDARD DEVIATION
CONFIDENCE LOWER UPPER L0WER UPPEh
LEVEL LIMIT LIMIT LIMIT LIMIT
90.0?
95.0?
99. OS
808.25
601.00
784. 9B
873.44
880.70
896.71
49.819 99.929
47.289 108.86
, 42.942 130.30
F0K A NORMAL DISTRIBUTION
I
£
3
4
5
6
7
8
9
10
II
12
13
X4
INPUT
DATA
950
870
77?
754
875
870
920
750
755
8«?0
890
845
860
OhDEKED
DATA
750
754
755
772
82C
K45
860
870
H70
890
920
950
CUMULATIVE
PERCENTAGE
7.143
14.286
? 1.429
28.571
35.714
42.857
50.000
64.286
6/«.286
71.429
78.571
85.714
92.857
CUMULATIVE
NORMAL
8.416
9.393
9.650
14.825
37.596
52.511
61.426
67.079
67.079
69.774
77.198.
88.499
95.106
DIFFERENCE
1.274
-4.893
-11.779
-13.747
1.882
9. ',54
11.426
2.793
2.793
-1.655
-L374
2.784
2.249
0.1594 = KOLM0G0h:OV-i>MlHN0V STATISTIC
264
-------
STAIiblICi, FOK Xb
.\UML-F h OF PFibh KVA1 I 0Mb 13
MRAN = XF
MlM«X-XF r.)**^)/(N*t'V-'K+*fc) P.3599
P-L-IUFt) CfMF IDFNUF IMFl.VALi. ON MF/YN! f. S1ANDAKD PFVIATI^N
•MFAN tTANDAKD DFVIA1ION
CONFIUFNCF LO'.-Ff tUVFK LP'/F}. I'HPFK
LM/EL LI^lIl LIiYll L-IVfll LIMIT
90.0'x
9b.U5f
99 .OT-
• 1 S
.17
7f3.93
770.91
78'.33
47.970
4b.533
41.3/18
FCIh A
104.8?
l?5.4f
Dli-TI-.IH TI0N
*-t- ******* * *
y
HJ
11
xs
Ix'H 1
I'ATA
704
7/.b
79O
f I >
700
77d
d. 1 c-(\-
Oi.l-FLFP
DATA
Mb
/••3S
/• K4
700
704
7 r- '•
7/ib
770
79O
790
CI v(-LAI IV F
I'Fi.CFNIAPF
7. 143
f S . 7 1 •'•
. O'lM
( 4. P
7 1 . /if V
:-: b . 7 1 4
-. 5.714
NPi-.MAL
3.POK
30.417
3P. f*-'
4'..f7ci
',7.77V
7P.P39
81 .7P5
L-IFFFI-.FMCF
-3.935
-K.Of G
O.RO?
1 . K4*
-3.0S9
? . 4 1 8
7.779
5.P5S
^.P5S
O.K11
-3.9B9
-3.989
1 . HI T>
15 1 1 C
265
-------
DFSCMPTIVF STATISTICS FOh X6
MUMF-FK OF C&SFJA'ATIONS 13
MFAN - XPK = SUM VAKIANCF =
SUMCCXU >-X£K>**2>N ^ HVK 3550.5
tNADJl'STFL STANDAKD DFVIATION = SCKTUJV'K) = S 5?. 586.
ALJl'STFD CliNMASFD VAKIANCF IIVh*\'/; 3R4f.3
ADJL'STFD STANDAKD DEVTATION = SCKT(AVB> 6P'.019
SKFV.'NFSi = £IM(CX(I)-XPH)**3)/CN*HVK**1.5) 0.2!^07
KUKT0S1S = Sli^CCXd >-XFK>**4>/CN*1'VM=*2> , 3.00P2
P-LIDFD CPNFIDFNCF INlFKVALi- ON MFAN « STANDAhD DFVIATION
MFAN STANDAKD DFVIATION
CPNFIPFNCF LOV.FK I'PPFK LOWFh UPFFR
LF*0
690
CI'Ml'LATIVF
710
715
730
740
755
765
775
780
H6Q
7. 1-03
21.429
PH. 571
35.714
42.857
50.000
57 . 1 ^3
64.286
71.429
7ft. 571
85.714
92.857
CHMl'LATIVF
NOKMAL
4.678
7.56*
26.473
27.539
37.967
41.070
50.643
57.039
66.248
71.920
77.087
79.456
98.267
PIFFFI.FNCF
-2.465
-6.722
5.044
-1.032
2.252
-1.787
0.643
-0.104
1.962
0.492
- 1.485
-6.259
= KOLMOGOKOV-5'VllKNOW STATISTIC.
266
-------
DESCRIPTIVE STATISTICS FOR X7
NUMBER OF.OBSERVATIONS 13
MEAN = XBR = SUMCX-XBR>**2>N - UVR PI 41. 5
tNADJUETFD STANDARD DEVIATION = SOKTO'V'iO - S 46. P76
ADJUSTED (1'NBIASFD) vAKlANCF - UV'K*N. -X&R>**3)/CN*t'VR**l .5) 0.447f3
KURTPSIK = StM> 8.4753
2-SIDFD CRNFIDFNCF INTFfcVALS ON MFAN * STANDARD DEVIATION
MFAN STANDARD DEVIATION
CONFIDFNCF LOWER UPPER LOWFR UPPFR
LEVEL LIMIT LIMIT LIMIT LIMIT
90.07
95. 07,
99.07
678.P7
67P.97
661.?7
725.89
731.18
•»42.8B
36.387 72.986
34.539 79.508
31.364 95.167
FOR A NORMAL DISTRI Pt'TION
****************************************
t
P
s
(•
7
K
9
10
1 1
\','
K<
X7
INPUT
DATA
800
760
67-a
700
73S
700
640
^50
710
PRDFRFD
PATA
638
640
650
674
685
690
700
700
710
735
745
76O
8OO
C! vi
PFKCFNTAC-F
7. 143
U.13P<-- =
?«.571
35.714
A?.857
57.143
57. 1/.3
6/4.
71 .
78.571
R5.71A
9P.857
'*• STATISTIC
Cl'MULATIVF
NORMAL
9.170
9.873
13.980
P7.997
36. l/i6
40 . 1 0 1
48. PRO
56.53?
75.PK7
81 .358
RR.543
97.898
niFFFKFNCF
P.OP7
-4.413
-7. /|/ip
-0.574
0.43?
-P.756
-8.863
-8.8*3
-7.753
3.858
P. 786
P.8P9
5.041
267
-------
DFSCF.IPTIVF. STATISTICS F0K-X8
NlMF'Fh OF OB5.FKVATIONS 13
MFAN XBK SUM**i?>N = UVK P993.2
INADJPSTED STAMDAKD DFVIATION = SGKTCUVR) = S 54.710
ADJUSTFD t.UNBIASFD) VARIANCE = UVh*N/CN-l> = AVK 3242.6
ADJUSTED STANDARD DFVIATION = SChT'AVF;) 56.944
SKEMVESS = SUMC(X
-------
MULTIPLE RFGRESSION ANALYSIS
TEXAS GULF PLANT NORMAL OPERATION
269
-------
Texas Gulf
Data Set
TGX2
TGX2
13:27EST
1 1/18/74
18,845, 991, 8k> 1,863, 821,846
1 5, 846, 992, 822, 863, 822, 846
18,845,992,821,^63,821,846
1 8, 843, 994, 822, 865, 820, 847
1 3, 845, 988, 8 1 6, 859, 8 1 5, 842
17,840,991,818,860,818.842
104,830, 1001,832,886,834,866
098,828, 1001,829,884,832,863
097, 826, 1 000, 826, 883, 830, 862
125,849,984,81 1,849,823,850
1 1 7, 843, 982, 808, 848, 820, 849
1LO, 845, 981,808,845,818,848
128, 848, 980, 800, 830, 8 1 0, 838
134,855,983,808,842,821,847
1 30, 853, 986, 8 1 3, 851, 827, 85 1
121,842,979,816,852,826,849
1 1 0, 840, 92.0, 730, 730, 735, 755
120, oL.0, 950, 7/63, 763, 769, 786
132,801, 1 128,840,968,790,808,795,822
120,803, 1 125,840,969,800,830,801,830
,808, 1 130,841,970,800,830,803,831
70,802, 1 103,842,977,81 1,846,801,828
80,800, 10^ 1,839, 975, 81 1,845,805,827
03, 828, 960, 804, 834, 806, 8 IS
109,846,999,826,869,821,849
1 13,849,999,826,868,821,846
117,850,999,826,867,821,846
1 26, 791,791, 800, 8 1 2, 793, 821
1 20, 848, 980, 803, 8 '. 5, 8 1 8, 837
18,847,983,817,848,833,850
128, 834, 972, 8 1 4, 850, 830, 855
100
101
102
:03
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
135,806,
120,805,
134,806,
132,800,
140,803,
135,803,
222,797,
188,795,
200,795,
120,806,
127,803,
130,805,
140,808,
141,81 1,
127*809,
100,799,
425*805,
125,799,
132,801, 1
120,803, 1
1 10 ,808,
70,802, 1 1
80,800, 1C
60,778, 1
1 18,805,
1 10,806,
1 10,809,
200,793,
130,800,
95,798, 1
100,803,
270
-------
RUN 1
MULTIPLE RFGRESSI0N PR0GRAM
XMF.AN
YMEAN
1.37935SE+02
8.019677E+G2
1. 1 16581F+03
8.409032E.+ 02
9.753871E+02
8.094194E+02
8.438064E+02
8. 132258E+02
8.385161E+02
OBSERVED
.3500000F+02
.2000000F+02
.3400000F+02
.3200000E+02
.4000000F+02
.3SOOOOOE+02
2.2200000E+02
1.8300000E+02
2.0000000E+02
1.2000000E+02
.2700000E+02
.3000000E+02
.4000000E+02
.4100000E+02
.2700000Ev02
.OOOOOOOE+02
4.2500000E+02 {
I.2500000E*02
1.3200000F+02
.• .2000000F->02
1. 1000000F + 02
7.0000000f*01
8.0000000E+01
6.0000000F+01 «
1. 1800000E+02
1. 1000000F+02 l
1. lOOOOOOEi-02
2.0000000F+02 !
1.3000000F+02
9.5000000F+01 «
l.OOOOOOOE+02
/
CALCULATED
.2078900F.+ 02
. 1261 185E+02
.2078900E+C2
.0160772F+02
. 1758125E+02
.2855834E+02
. 1 I50338E+02
. 1500999E+02
.2596205E+02
•3082470E+02
.41381 18E+02
.4250045E+02
.586454^E-i-02
.3809330E+02
.2826169E+02
. 1518727E+02
>.8460406F+02
. 797521 OE+02
.6967646F+02
.5894531E+02
.7S52371E+02
.2717273E+02
.2959251E+02
).3029634F+01
I.0406^96F+02
?.9424274E+01
I.0780532E+02
2.5528C15E + 02
I.2574157E+02
?.81 17317E+01
1.57965S8E+02
EST VAR
.7629593F+02
.72161 12F+0£
.7629593E+02
2. 1430334E+02
.5935394E+02
.2542261E+02
2.8636440E+02
3. 1667466E+02
3.3377359E+02
2.2583242E+02
2.0372619F+02
1.6594186F+02
2.2959603E+02
3.7531740E+02
3.5310062E+02
2.0482986F+02
1.5403200F+03
7.7819038E+02
1.7183188F+02
1.3067693F+02
2.4079640E+02
4.5954791F+02
2.6824281F+02
1.6558976F+03
3.0912896F+02
3.5481437E+02
4. 1230586E+02
1.9507907F. + 03
3.4419073F+02
4.7251964E+02
5.5681231E+02
271
-------
RUN 1 (Continued)
.DEGREE
. REGRESSION.
. REMAINDER .
. T0TAL
OF FREE.
4
26
30
SUM 0F SQUARES.
5.330958E+04 .
6.949829E+04 .
1.228079F+05 .
VARIANCE ESTIMATE .
1.332740F+04
2.673011E+03
MULTIPLE C0RRFLATI0N C0EFFICIENT = 6.588545F-01
C0NSTANT TERMCINTERCEPT) = 8.442381E+02
COEFFICIENTS
3.7227268F+00
•2.7871378E+00
•2.0432343E+00
3.7595600E-01
EST 0F SD
I.9468376E+00
1.0914522E+00
1.3717363E+00
1.3798760E+00
F KATI0C 4* 26 DEGRFES 0F FREED0M)= 4.985911E+00
UPPEfc RT HAND P0RTI0N 0F MATRICES 0NLY
INVERSE ADJ DATA M0MENT MATRIX
R0WC 1) I.4179E-03 -4.9918E-04 -9.6133E-05
R0W( 2> 4.4567E-04 1.6484E-04 -1.7670E-04
R0V:< 3) 7.0395E-04 -6.6529E-04
R0W( 4) 7.1233E-04
1.C495E-04
EST C0EFFICIENT VAR-C0VAR MATRIX
R0WC I) 3.7902E + 00 -1.^343E+00
R0V.'C 2> 1. 1913E+00 . 4.4061E-01
R0WC 3) 1.8817F*00 -1.7783E+00
K0WC 4) 1.9041E+00
-2.5697E-01
-4.7233E-01
2.8052E-01
ADJ DATA
R0W( 1)
M3W< 2>
K0W< 3)
R0W< 4)
C0RRELAT10N
l.OOOOE+00
I.OOOOF+00
l.OOOOE+00
l.OOOOE+00
MATRIX
-6.2795E-01 -9.6222F-02
2.9429F-01 -3.1362E-01
-9.3950E-01 ,
1.0442E-01
272
-------
RUN ?.
MULTIPLE REGRESSION PROGRAM
XMEAN
8.019£77E+02
1. J 16581F+03
8.409032E+02
9.753871E+02
8.43P064E+02
8. 132258L+C2
8.385161E+02
OBSERVED
.3500000E+C2
,2000000?+02
.3400000E+02
, 3200000E+02
-4000000F.+ 02
.3500000F+02
.220000GE+02
.8800000F.+ 02
.OCOOOOOE+02
.2000000E+02
,2700000E+02
•3000000E+02
. 4GOOOOOE+02
, 4IOOOOOE+C2
YMEAN
1.379355F+02
.OOOOOOCE C2
.2500000F+02
.2500000E>02
*32000GGF+0£
• 200UOOOF. + 02
. lOOOOOOF + Oi'.
.OOOOOOOF.»01
.OOOOOOOE+OI
1000000E+02
6.OOCOOOOF+01
I,
1
2
1.
9,
1.
OUOOOOOE+02
3000000F+02
500000CE+01
OODOOOOE+02
i
1
f
«
(
<
t
CALCULATED
.2065665E+02
. 1030036E+02
.2104080E*C2
J.25139P6E+OJ
.2426820E+02
.38650t6F*02
.49b389bE+02
. 739869 1E+02
>.OJ85824E»-02
.2962520E+02
. 6646784Z+02
.51 10239E+02
.5410184E+02
.2018i!86E+02
.2256832E+02
.0040894F+02
J. 5849109e + u2
.4777413E+02
. 3669044F+0?
.42l0728F-r02
. 676029''F + 02
.273695RE+02
.63I8821F+02
'!o05?K^Elo2
J.64ft9852E<-01
K 1H48912F + 01
I .832b24?E+02
?.8300663F+01
r »9!r03340E+Gl
I. 6-:-'30497E+02
EST VAK
1.9785551F*G2
1.7.'J32858E+02
2.0352449E+02
3. 1 108493E+02
r .2289126F+02
3.8737159E+02
3. 834064 JF+02
4. 7998 1 7IiE*02
6. '076563E+02
2.9235762E+02
3.6230898E+02
3.4323903E+02
2.0027279E*02
3.8303051E+02
4.087b623E+02
1.6786530E+G2
1 .4481389E+03
6. 1707298F+02
6.5819022E+02
3.3319^^^-^02
3.9892946E+C2
4.286.8723F+02
7. 1413675F+u«£
2.6847986F+02
2.6215510E+02
3.5&00741F+02
1.7646602E+03
1. 1 139235E+03
5.6266279E+02
6.97349I5F+0^
272A
-------
.DEGREE
. REGRESSION.
. REMAINDER .
. TOTAL
OF
8
22
30
RUN
FREE.
2 (Continued)
SUM OF SQUARES.
8.332108E+04 .
3.948679E+04 .
1.228079E+05 .
VARIANCE ESTIMATE .
1.041514E«04
1.794854E+03
MULTIPLE CORRELATION COEFFICIENT 8.P36911E-01
CONSTANT TERMCINTERCEPT) = 3.377379E+03
COEFFICIENTS
6.9265897E+00
• 1.9615544E+00
•4.9321602E*00
3.8414428E-01
•8.0901479E*00
2.6P23307E+00
3.3243019E+00
•1.4300937E+00
EST OF SD
2.9622758E+00
1.3438454E+00
2.6892745E+00
7.8405521E-01
2.86b3007E*00
1.9660347E+00
3.6216309E+00
3.4853787E+00
273
-------
RUN 3
0E-SF J-A'FD
.3500COOE+02
.2000000E+02
.3400000E+02
.3200000E+02
.4000000E+02
.3500000E1+02
2.2200000F+02
1.8800000E+02
2.COOOOOOE+02
.2000000E+02
.27000COE+02
.3000000E+02
.4000000E+02
.4100000F+02
.2700000F+02
.OOOOOOOF+02
4.2500000E+02
.2500000E*02
.3200000E+02
.200COOOE+02
.1000000E+02
7.000OOOOE+01
8.0000000E+01
6.0000000F+01
1.1800000F+02
1.1000000E+02
1.1000000F+02
2.0000000E+02
1.3.000000F+02
9.5000000F+01
J.OOOOOOOF+02
8
CALCULATED
.2020103E+02
.1019730F+02
.2063533E+02
.2685916E+01
EST VAK
1.9600022E+02
1.7226513E+02
.2360301E+02
.3945266F+02
.5329643E+02
.7854978E+02
.90663I9E+02
.2883658E+02
.6709222E+02
.5I01917E+02
.5333083E+02
.1790353E+02
.2097887F+U2
.0182105E+02
3.6049421E+02
.4720183E+02
.3748979E+02
.4178868E+02
.6679512E+02
.2717058F+02
.6560886E+02
7.6136909E+01
1.0048923E+02
9.0594877E+OJ
1.8387648E+02
8.1264380F+01
8. 1622675E*01
1 . 6949627E+02
2.8238817F+02
2. 1 576.39 5F+02
3. 946036.1E + 02
3.8050038F+02
4.8585695F+02
5.6866451E+02
2.8990100F+02
3.7909062E+02
3.5555361E+02
2.0519335F.*02
3.6860193E+02
3.9176983F+02
1.709B269E+02
U4479093F+03
6.I584507E+02
6.4252438E+02
3.2926337E+02
3.9416342E+02
4.2381024E+02
6.9543617E+02
1.3874716E+03
2.7134381E+02
2.6150071E+02
3.5628143E+02
1.7540974F+03
1.I005696E+03
5.7953856E+02
6.9552187E+02
274
-------
RUN 3 (Continued)
.DEGREE
. REGRESSION.
. REMAINDER .
. TOTAL
0F FREE.
8
22
30
SUM 0F SQUARES.
8.3663I2E+04 .
3.914475E+04
1.228079E+05 .
VARIANCE ESTIMATE .
1.045789E+04
1.779307E+03
. MUL11PLE CORRELATION COEFFICIENT = 8.253800E-01
I
CONSTANT TrKf!< INTERCEPT) = 3.690478E+O4
COEFFICIENTS
•3i-l»b3336E+03
•6.7058524E+03
6*3034923E+OR
3.2799301F+03
4.4362473E+03
• 1.9235075E+03
EST 0F SD
3.5086377E+03
2.0939614E+03
3.4049073E+03
1.0446127E+03
3-7574227E+03
2.5669906E+03
4.6816896E+03
4.6394966E+03
F HATIOC 8* 22 DEGREES 0F FREED0M>= 5.877508E+00
UPPER RT HAND P0RTI0N 0F MATRICES ONLY
INVERSE ADJ
R0W< 1) 6.
-8. 3859E*02
R0WC 2) 2.
6.1970E+01
R0WC 3) 6.
5.4034E+03
4) 6.
DATA MOMENT MATRIX
9187E+03 -6.6065E+02 -5.0060E+03
6.0114E+03 -5.6340E+03
4643F.+ 03 -1
-1.4519E+03
5157F+03 -1.7610E+03 -1.3368E+03
1.0083E+03 4.6869E+02
2062E-»03 5.0306E+02 1.0187E+02 6.9088E+02
1.1S55F*03 -4.9030F+03
9.7282E+02
RfJWC
R0WC
R0WC
ROW<
b)
6>
7)
8)
1328E+02 9.3397E+02 -6.6186F+02
9347F+03 -4.7959E+03 -3.9ft22E+O3
7034F+03 2.0486E+03 -2.2883F+03
231RE+04 -I.I230E+04
20V7E+04
EST COEFFICIENT VAh-COVAR MATRIX
R0WC 1) 1.
-1.49.?1E+06
RRKC 2) 4,
1.1026E+Ob
R0l-.'< 3) 1.
9.M43E+06
KDWC
ROWC
R0W<
4)
b)
6)
7)
S)
1.0912E+06
1.41 18E+07
6.5894E+06
2. 1918E+07
2. 1 525E+07
6.5211E+02
2.6049E+03
2311F+07 -1.17b5E+06 -8.9073E+06 1
1.0fr96F+07 -1.0025F+07
3847E+06 -2.1461E+06 8.95IOF+05 1
-2.b833E+06
lb93E+07 -3.1334E+06 -2.3786E+06
1.6618E+06 -1.1777E+06
-8.5334E+06 -7.0501E+06
3.6451E+06 -4.0715E+06
- 1.9981E+07
1.7941F+06
1.8125F+05
2. 1094E+06
1. 1603E+06
4.6349F+06
8.3395F+Ob
1.2?93F+Ofi
-8.7239E+06
-1.7309E+06
275
-------
ADJ DATA
R0WC I)
-1.6567E-
R0WC 2)
:. 1248E-
H8WC 3)
6.0861E-
R0W< 4)
S)
6)
7)
8)
RUN 3 (Continued)
C0RRELATI0N MATRIX
l.OOOOE+00 -1.6COOE-OI -7.4559E-01
4.8950E-01 6.3257E-02
01
1
02
1
01
R0W<
R0W<
R0WC
6.51 I6E-01 -6. 1583E-01
-OOOOE+00 -3.0101E-01 4.0921E-01 2.3037E-02 -2.2870E-01
-2.6591E-01
.OOOOE+00 -8.8097E-01 -1.8592E-01 2.4134F-01 -5.4727E-01
.OOOOE+00 4.2339E-01
.OOOOE+00 -8.8472E-01
.OOOOE»-00 3.0331E-OI
•OOOOE+00 -9.1990E-01
.OOOOE*00
-4.3918E.-01
-4.0077E-01
-3.4187E-01
2.3725E-0.
2.6588E-01
-3.5716E-01
hEADY
7ST0P
14.7/i UNITS
0000.37 TCH
OFF AT 14:20EST 11/21/74
USED
BYE
0014.89 CHU
0009.14 KC
276
-------
Texas Gulf
Data Set
TGX4
12/09/74
10U
101
102
103
104
105
106
107
108
109
no
111
112
113
114
115
116
117
lift
119
120
If 1
122
123
1?4
!25
126
127
128
129
130
135,b06,
120,805,
134,806,
132, ROC*
1*0,803,
135,803,
222,797,
188,795,
200,795,
120,806,
127,803,
130, H05,
140,808,
141,81 1,
127,809,
100,799,
125, 799,
132,801,
120,803,
1 10 ,808
70,802, 1
80,800, 1
60, 778,,1
1 18,805,
1 10,806,
1 10,809,
200,793,
130,800,
95,798, 1
100,803,
1 18,845*991, 821,863, 821,8/if
1 ' b. 8 46, 992, K£-2,8?3,822,8/.6
1 18,845,992, 821,863,821,H46.
1 18,«43, 994, 822, 865, 8^0, 8*7
; 13,845,988,816,859,815,842
1 17,840,991,818,860,818,842
104,830, 1001,632,886,834,866
098,828, 1001, 8^9,884,832, 863
097, 8?*, 1 000, 8f 6, 883, 830, 86?
125,849,984,81 1,849,823,850
1 1 7, 843, 982, 808, 8*8, 820, 849
1 20, 8^5, 98 1, 808, 845, 8 1 8, 848
1 28, 848, 9RO, 800, 830, 8 1 0, 838
1 34, 855, 983, 808, 842.. 82 1 , 847
130,853,986,813,851,827,851
121,842,979,816,852,826,849
120,850,950,763,763,769, 786
128, 840, 96b, 790, 808, 795, 822
125, 840, 969, 800, 830, 801,830
I 130,841,970,800,830*803,831
03,842,977,81 1,846,801,828
391,839,975,81 1,845,805,827
03, B28, 960, 804, 834, 806, 819
1 09, 846, 9V9, 826, 869, 82. 1 , 849
1 13,849,999,826,868,821,846
1 17,850,999,826,867,821,846
126,791,791,800,812, 793,821
1 20, 848, 980, 803, 8 1 5, 8 1 8, 837
18,847,983/817,848,833,850
128,834,972,814,850,830,855
277
-------
STANDARD STATISTICS
DFSCMP1IVF STATISTICS FOR YI
NUMtFK OF 0F>SthVATI0Ni> 30
MFAN = XBh = SIMM-XBK>+*2>N = UVfc 1255.2
tNADJUSTFD STANDARD DFVIATION = SGKTCUVfO = S ;. . . . 35.428
ADJUSTFD (LHNBIASFD) VAKIANCF = HVh*N/(N-l) - AVK ' 1298.4
ADJUSTFD S1ANDAKD DEVIATION = SQHT(AVR> 36.034
SKFkNFSS = SUM 3.8945
2-SIDFD C0NFIDFNCE INTFHVALS 0N MFAN « STANDARD DFVIATI0N
_.^ STANDARD DFVIATI0N
C0NF1DFNCF """L0WFh UPFFh L0WFR UPPFR
LFVFL LIMIT LIMIT LIMIT LIMIT
90.0?
9 5. OX
99.OX
117.19
114.91
110.23
139.55
141.82
146.50
R9.746 46. 1 13
28.698 48.441
26.823 53.570
A NPIRMAL PI STRIPUTI0N
278
-------
Yl
INPUT
DATA
1
2
3
4
5
6
7
8
9
10
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
?8
29
30
135
120
134
132
140
135
222
188
200
120
127
130
140
141
127
100
125
132
120
110
70
80
60
1 18
110
1 10
200
130
95
100
0RDFKFP
DATA
60
70
80
95
100
100
1 10
110
110
1 18
120
120
120
125
127
127
130
130
132
132
134
135
135
140
140
141
188
200
200
222
0.2296 = KfiLMflGPkOV-SMIHNPlV STATISTIC
Cl'MULATIVF
PFkCFNTAGF
3.226
6.452
9.677
12.903
19.355
19.355
29.032
P9.032
29.03?
32.258
41.935
41.935
41.935
45. 161
51.613
51.613
58.065
58.065
64.516
64.516
67.742
74. 194
74. 194
80.645
80.645
83.871
87.097
93.548
93.548
96.774
CUMULATIVF
NORMAL
2.890
5.264
8.976
17.723
21.558
21.558
30.513
30.51?
30.513
38.679
40.820
4O.820
40.820
46.278
48.487
48.487
51.808
51.808
54.016
54.016
56.212
57.303
57.303
62.659
62.659
63.705
95.103
97.659
97.659
99.532
DIFFFKFNCF
-0.336
-1.188
-0.702
4.820
2.203
2.J?or>
.48 1
481
481
6.421
16
16
16
17
-3. 26
-3. 26
-6.257
-6.257
-10.500
-10.500
-11.530
-16.891
-16.891
-17.986
-17.986
-20.166
8.006
4. 1 1 1
4. 1 1 1
2.758
279
-------
DFSCR1PT1VE STATISTICS FOK XI
t
OF OhSFHVAIIONS ........................................... 30
~ XBR = SUM(X(I))/N ................................. 801.87
MEDIAN ............ .. ......................... . ............ 803.00
• ......... 33. 000
^ADJUSTED (BIASED) VARIANCE =
SllM((X(I)-XSK)**2)N = UVK ............................... 39.116
INADJUSTED STANDAKD DEVIATI0N = -SOKT ............... -1.7507
KUK10SIS - SUM«XCI>-XBfO**4>/ ................. 7.6792
V
2-SIDFD CPINFIDENCE INTERVALS 0N MEAN I STANDARD DFVIATI0N
MEAN STANDARD DEVIATION
C0NFIDE.NCE L0WER UPPER LOWER UPPER
LEVFL LIMIT LIMIT LIMIT LIMIT
90.0? 799.89 803.84 5.2511 8.1404
95.0% 799.49 804.24 5.0661 8.5514
99.07 798.67 805.07 4.7352 9.4569
FOR A NORMAL DISTRIBUTION
280
-------
XI
-INPUT 0RDFKFD CUMULATIVE CUMULATIVE
DATA DATA PFRCFNTAGF N0HMAL DIFFFRFNCF
1 806 778 3.226 0.009 -3.217
2 805 793 6.452 8.168 1.716
3 806 795 12.903 14.019 1.116
4 800 795 12.903 14.019 1.116
5 803 797 16.129 22.212 6.083
6 803 798 19.355 27.164 7.809
7 797 799 25.806 32.612 6.8C6
8 795 799 25.806 32.612 £.806
9 795 800 '35.484 38.459 2.975
10 806 800 35.484 38.459 2.975
11 803 800 35.484 38.459 2.975
12 805 801 38.710 44.581 5.872
13 808 802 41.935 50.836 8.901
14 811 803 58.065 57.070 -0.994
15 809 803 58.065 57.07O -0.994
16 799 803 58.065 57.070 -0.994
17 799 803 58.065 57.070 -0.994
18 801 803 58.065 57.070 -0.994
19 803 805 67.742 68.884 1.142
20 808 805 67.742 68.884 1.142
21 802 805 67.742 68.884 1.14R
22 600 806 80.645 74.208 -6.437
23 778 806 80.645 74.208 -6.437
24 805 806 80.645 74.208 -6.437
25 806 806 80.645 74.208 -6.437
26 809 808 87.097 83.252 -3.844
27 793 808 87.097 83.252 -3.844
28 800 809 93.548 86.894 -6.655
29 798 809 93.548 86.894 -6.655
30 803 811 96.774 92.447 -4.327
0.1374 = K0LM0G0R0V-SMIRN0V STATISTIC
281
-------
DESCRIPTIVE STATISTICS F0H X2
NUMBER OF OBSERVATIONS 30
MEAN = XBh - SUMCX(I»/N 1116.8
MEDIAN 1118.0
KANGF 43.000
«
INADJUSTED CBIASED) VARIANCE =
SUMC(X-XBR>**2>N = UVK .. 111.43
UNADJUSTED STANDARD DEVIATION = SGRT = S 10.556
ADJUSTED (UNBIASED) VARIANCE = UVR*N/CN-1> = AVR 115.27
ADJUSTED STANDARD DEVIATION = SQRTCAVR) • 10.736
SKEWNESS = SIW«XCI>-XBFO**3>/(N*UVR**1.5> -0.66769
KUhT0SIS s SUM«X(I)-X&ft)**4)/CN*UVh**2) 2.7854
. • <*•-• >;. . • • •
2-SIDFD CONFIDENCE INTERVALS RN MEAN R STANDARD DEVIATION
MEAN STANDARD DEVIATION
C0NFIDFNCE L0WEK UPPEk L0lvEK UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.07
95. OZ
99.07
1113.5
1112.8
1111.4
I 120.1
1120.8
1122.2
8.8628 13.739
8.5505 14.433
7.9920 15.961
F0K A f.URMAL DISTRIBUTION
282
-------
X2
INPUT
DATA
1
2
3
A
5
6
7
8 1
9 1
10
1 1
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
18
15
18
18
13
17
04
098
097
125
1 17
120
123
134
130
121
120
128
125
130
103
091
103
109
113
1 17
126
120
1 18
128
OKDFKFD
DATA
1091
1097
1098
1 103
1 103
1 104
1109
1113
1 1 13
11 15
1117
1 117
1117
11 18
1118
11 18
1 1 18
1120
1120
1 120
1121
1125
1125
1 126
1128
1128
1 128
1130
1130
0.1741 = KOLM0G0&0V-SMIRN0V STATISTIC
CUMULATIVE
PERCENTAGE
• 3.226
6.452
9.677
16. 129
16. 129
19.355
22.581
29.032
29.032
32.258
41.935
41.935
41.935
54.839
54.839
54.839
54.839
64.516
64.516
64.516
67.742
74.194
74. 194
77.419
87.097
87.097
87.097
93.548
93.548
96.774
CUMULATIVE
NORMAL
0.813
3.258
3.997
9.933
9.933
1 1 .659
23.376
36.169
36. 169
43.3^3
50.743
50.743
50.743
54.450
54.450
54.450
54.450
61.717
61.717
61 .717
65.217
77.750
77.750
80.425
85. 157
85. 157
85.157
89.055
89.055
94.543
DIFFERENCE
-2.413
-3. 19/i
-5.681
-6.196
-6.196
-7.696
0.796
7. 137
7. 137
1 1 .085
8.808
8.808
8.808
-0.389
-0.389
-0.389
-0.389
-2.799
-2.799
-2.799
-2.525
3.556
3.556
3.006
-1.940
-1.940
J--.1.940
-4.493
-4.493
-2.232
283
-------
DESCRIPTIVE STATISTICS'
xa
NUMBER 0F OFSE KVATI3NS 30
MFAN = XBR SUMCXCI»/N 840.93
MEDIAN 844.00
HANGE 64.000
INADJUSTED (BIASED) VARIANCE =
SUM((X(1>-XBR>**2)N = UVh 137.60
INADJL'STFD STANDARD DF.VIATI0N = SGRT(UVR) = S 11.730
ADJUSTED (UNBIASED) VARIANCE UVk*N/CN-l) - AVK 142.34
ADJUSTED STANDARD DEVIATION = SORTCAVR) 11.931
NEWNESS = 'SUM((X(I)-XBR)**3>/(N*UVR**l.5) -2.5867
KUkTOSIS = SUM(CXU>-XBK>**4>/CN*UVK**2> 11.321
2-SIDED C0NFIDE.MCE INTERVALS 0N MEAN « STANDARD DEVIATION
MEAN STANDARD DEVIATION
CONFIDENCE LOWFk UPPER LOl-'ER UPPER
LEVEL LIMIT L'IMIT LIMIT LIMIT
90.07
95.07
99.07
837.23
836.48
834.93
844.63
845.39
846.94
9.8487 15.268
9.5016 16.039
8.8810 17.737
FOR A NORMAL DISTRIBUTION
284
-------
X3
INPUT 0RDFRFD CUMULATIVF CUMULATIVF
DATA DATA PFRCFNTAGF 'NORMAL DIFFFKF.NCF
1 . 845 791 3.226 0.001 -3.224
2 846 826 6.452 10.534 4.083
3 845 828 12.903 13.917 1.014
4 843 828 12.903 13.917 1.014
5 845 830 16.129 17.973 1.844
6 840 834 19.355 28.057 8.703
7 830 839 22.581 43.563 20.983
8 828 840 32.258 46.882 14.624
9 826 840 32.258 46.882 14.624
10 849 840 32.258 46.882 14.624
11 843 841 35.484 50.223 14.739
12 845 842 41.935 53.562 11.627
13 848 842 41.935 53.562 11.627
14 855 843 48.387 56.876 8.489
15 853 843 48.387 56.876 8.489
16 842 845 61.290 63.340 2.049
17 850 845 61.290 63.340 2.049
18 840 845 61.290 63.340 2.049
19 -840 845 61.290 63.340 2.049
20 841 846 67.742 66.446 -1.296
21 842 846 . 67.742 66.446 -1.296
22 839 847 70.968 69.445 -1.5?3
23 828 848 77.419 72.318 -5.101
24 846 848 77.419 72.318 -5.101
25 849 849 83.871 75.05? -R.819
26 850 849 83.871 75.052 -8.819
27 791 850 90.323 77.636 -12.687
28 848 850 90.323 77.636 -12.687
29 847 853 93.548 84.409 -9.140
30 82* 855 96.774 68.081 -8.693.'
0.2356 = KPLMOGOfcOV-SMlKNOV STATISTIC
285
-------
DFSCKIH7I\'F STATISTICS FPK X4
NUM&FK OF ObSFRVATIONS 30
MEAN = XFR = Sli>J)/N 977.?3
MFDIAN 983.00
KANGE '. •> 210.00
INADJUSTFD (BIASED)'VARIANCE =
SIJM((X(I)-XPF.)**?>N = UVK .. 1349.5
LNADJUSTFD STANDAKD DFVIATI0N = SOKT(l'VH) = S 36.736
ADJUSTFD (I'NRIASFD) VAMANCE = l.'Vh*N/(N-1 ) = AVf% 1396.0
ADJUSTED STANDAKD DFVIATION = SOHT(AVK) 37.364
SKFV'NFSS = SlW<(X -4.3015
= SlM((X(I)-XBK)**4)/-(N*UVK**f!) P2.063
2-SIDEt) CONFIDENCE INTFKVALS ON MFAN £ STANDAKD DF.VIATI0N
MFAN STANDAKD DEVIATION
CONFIDENCE LOVFh UPPFK LOKFR I'PPFR
LFVEL LIMIT LIMIT LIMIT LIMIT
90.07 9*5.64 .988.82 30.844 47.815
95.07, 963.28 991.19 29.757 50.229
99.07 958.43 996.04 P7.813 55.547
F0f< A NOKMAL DISTKIPUTION
286
-------
X4
INPUT OKDFRFD CUMULATIVF CUMULATIVE
DATA DATA PFKCFMTAGF NOHMAL DIFFFRFNCF
1 991 791 3.2P6 0.000 -3.226
2 992 950 6.452 23.304 16.852
3 992 960 9.677 32.232 22.554
4 994 968 12.903 40.241 27.338
5 988 969 16.129 41.280 25.151
6 991 970 19.355 42.325 22.970
7 1001 972 22.581 44.430 21.850
8 1001 975 25.806 47.617 21.810
9 1000 977 29.032 49.751 20.719
10 984 979 32.258 51.886 19.628
11 , 982 980 38.710 52.951 14.242
12 381 980 38.710 52.951 14.242
13 980 981 41.935 54.015 12.079
14 983 981 45.161 55.076 9.914
15 986 983 51.613 56.133 4.520
16 979 983 51.613 56.133 4.520
17 950 984 54.839 57.186 2.347
18 968 986 58.065 59.275 1.211
19 969 988 61.290 61.339 0.048
20 970 991 67.742 64.373 -3.369
21 977 991 67.742 64.373 -3.369
22 975 992 74.194 65.366 -8.828
23 960 992 74.194 65.366 -8.828
24 999 994 . 77.419 67.319 -10.100
25 999 999 87.097 71.991 -15.106
2f 999 999 87.097 71.991 -15.106
27 791 999 87.097 71.991 -15.106
2« 980 1000 90.323 72.885 ' -17.438
29 983 1001 96.774 73.764 -23.010
30 972 1001 96.774 73.764 -23.010
•0.3024 = KOLM0G0H0V-SMIkN0V STATISTIC
287
-------
DESCRIPTIVE STATISTICS FOR X5
NUMBFR OF OBSERVATIONS 30
MEAN = XBK = SUM-XBFO**2>N = UVtt 186.33
UNADJUSTED STANDARD DEVIATION = SORTCUVR) = S 13.650
ADJUSTED (.UNBIASED) VARIANCE = UVR*N/(N-1) = AVR 192.7,5
ADJUSTED STANDARD DEVIATION = SQRT(AVR) $ 13.884
SKEWNESS = SUMCCX'-XBR>**3>/ -1.4408
KURT0SIS = SUM«X(I)-XBR)**4)/(N*UVR**2) 6.2938
2-SIDED CONFIDENCE INTERVALS 0N MEAN & STANDARD DEVIATION
MEAN STANDARD DEVIATION
_________________________ _________________________
CONFIDENCE LOWER UPPER LOWER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.05?
' 95. 03?
99. 0*
807.76
806.88
805.08
816.37
817.25
819.05
11.461 17.767
11.057 18.664
10.335 20.640
FOR A NORMAL DISTRIBUTION
288
-------
X5 ....... _. _.__
INPUT PIKDF.KFD CUMULATIVF CL'MULATIV/F
DATA DATA PFKCENTAGF NORMAL DIFFFKF.NCF
1 8?1 763 3.226 0.020 -3.205
2 822 790 6.452 5.598 -0.853
3 821 800 19.355 19.239 -0.116
A 822 800 19.355 19.239 -0.116
5 816 800 19.355 19.239 -0.116
6 818 800 19.355 19.239 -0.116
7 832 803 22.581 25.686 3.106
8 829 804 25.806 28.061 2.255
9 826 , 808 35.484 38.479 2.996
10 811 808 35.484 38.479 2.996
11 808 808 35.484 38.479 2.996
12 808 811 45.161 46.938 1.777
13 800 811 ' 45.161 46.938 1.777
14 808 811 45.161 46.938 '1.777
1= 813 813 48.387 52.680 4.293
16 ' 816 814 51.613 55.537 3.925-
17 763 816, 58.065 61.153 3.088
18 790 816 ' 58.065 61.153 3.088
19 800 817 61.290 63.883 2.593
20 800 818 64.516 66.544 2.028
21 811 321 70.968 74.003 3.036
2? 811 821 70*968 74.003 3.036
23 804 822 77.419 76.284 -1.135
24 826 822 77.419 76.284 -1.135
25 826 826 90.323 84.221 -6.102
26 826 826 90.323 84.221 -6.102
27 800 826 90.323 84.P21 -6.102
28 803 826 90.323 84.221 -6.102
29 817 829 93.548 88.870 -4.678
30 814 832 96.774 92.446 -4.328
0.1257 = K0LM0G0K0V-SMIKNOV STATISTIC
289
-------
DESCRIPTIVE STATISTICS FOR X6
NUMBER OF PBSERVATIPNS i 30
MEAN = XPH = SUMCXCI»/N * 847.60
MEDIAN 849.50
RANGE 123.00
UNADJUSTED (BIASED) VARIANCE. =
SUM«X(M-XBR)**2)N = UVR 628.57
UNADJUSTED STANDARD DEVIATION = SQRT(UVR> = S .«. 25.071
ADJUSTFD (UNBIASED) VARIANCE = UVK*N/(N-1) = AVR 650.25
ADJUSTED STANDARD DEVIATION = SQRTCAVFO .*.... 25.500
SKEWMESS = SIW«XU>-XPK>**3>/CN*UVR**1.5> -1.209f>
KURTnSIS - SIW((X(I)-XPR>**4)/(N*UVK**2) 5.3209
2-SIDED CONFIDENCE INTERVALS ON MEAN f. STANDARD DEVIATION
MEAN STANDARD DEVIATION
CONFIDENCE LOWFR UHPFK LOWER UPPER
LEVEL LIMIT LIMIT LIMIT LIMIT
90.07 839.69 855.51 21.050 32.6J2
95.07 .838.08 857.12 20.308 34.280
99.07 834.77 860.43 18.982 37.910
FOht 4 NORMAL DISTRIBUTION
290
-------
X6
INPHT (IKDFKFD Cl'MULATIVF CUMULATIVF
DATA DATA PERCFNTAGF NORMAL DIFFFRFNCF
1 863 763 3.226 0.045 -3.1RO
P. 063 808 6.452 6.02? -0.430
3 863 812 9.677 R.135 -1.543
4 865 815 12.903 10.055 -2.848
5 859 830 22.581 24.504 1.923
6 860 830 22.581 24.504 1.923
7 886 830 22.581 24.504 1.923
8 884 834 25.606 29.690 3.884
9 883 842 29.032 41.309 12.277
10 849 ' 845 35.484 45.939 10.456
11 848 845 35.484 45.939 10.456
12 845 846 38.710 47.498 8.789
13 830 848 45.161 50.626 5.46'4
14 842 848 . 45.161 50.626 • 5.464
15 851 849 48.387 52.189 3.802
16 852 850 51.613 53.749 2.136
17 763 851 54.839 55.304 0.465
18 808 852 58.065 56.850 -1.215
19 830 859 61.290 67.258 5.968
20 830 860 64.516 68.661 4.145
21 846 863 74.194 72.705 -1.488
22 845 863 74.194 72.705 -1.4R8
23 834 863 74.194 72.705 -1.488
24 869 865 77.419 75.249 -2.170
25 868 867 80.645 77.661 -2.984
26 867 868 83.871 78.814 -5.056
27 812 869 R7.097 79.933 -7.164
28 815 883 90.323 91.747 1.424
29 848 884 93.548 92.328 -1.221
30 850 88ft 96.774 93.395 -3.379
0.159/1 = KOLMOGfiR0V/-SMIKNOV STATISTIC
291
-------
DESCRIPTIVE STATISTICS FOR X7
NUMBER OF 0BSEKVATI0NS 30
MFAN = XF>h = SUMCX-XF>R>**P>N = MV/R 193.87
INADJUSTFD STANDARD DEVIATION = SOf-.T(UVh) - S 13.9JM
ADJIiSTFD CUNtUASFD) VARIANCE = UVh*N/CN-l> = AVk J>00.5f
ADJl'STED 5.TANDARD DFVIATION = SQF.TfAVK) 14.162
*xn-'NES£ = SIM((X(I>-XBR)**3)/**4)/CN*UVK**2> 5.14P9
P-SIDED CONFIDENCE INTFKVALS 0N MFAN & STANDARD DFVIATI0N
MFAN STANDARD DEVIATION
CONFIDENCE LOt-.'FK UHPFK LOl-.'FR UPPER
LFVFL LIMIT LIMIT LIMIT LIMIT
90.07
95.07
99.0^
811.44
810.55
808.71
820.23
821.1?
822.96
11.690 18.123
11.279 19.038
10.542 21.054
F0R A NORMAL DISTRIBUTION
292
-------
X7
INPUT ORDERED CUMULATIVE CUMULATIVE
DATA DATA PERCENTAGE NORMAL DIFFERENCE
1 821 769 3.226 0.047 -3.179
2 822 ' 793 6.452 5.345 -1.107
3 821 795 9.677 7.063 -2.614
4 820 '801 16.129 14.745 -1.384
5 815 801 16.129 14.745 -1.384
6 818 803 19.355 18.242 -1.113
7 834 805 22.581 22.215 -0.366
8 832 806 25.806 24.373 -1.433
9 830 810 29.032 34.020 4.988
10 823 815 32.258 47.654 15.396
11 820 818 ' 41.935 56.080 14.144
12 818 818 41.935 56.080 14.144
13 810 818 41.935 56.080 14.144
14 821 820 48.387 61.570 13.183
15 . 827 820 48.387 61.570 13.183
16 826 821 67.742 64.238 -3.504
17 769 821 67.-742 64.238 -3.504
18 795 821 67.742 64.238 -3.504
19 801 821 67.742 64.238 -3.504
20 803 821 67.742 64.238 -3.504
21 801 821 67.742 64.238 -3.504
22 805 822 70.968 66.838 -4.130
23 806 823 74.194 69.359 -4.834
24 821 826 77.419 76.359 -1.061
25 ^821 827 80.645 . 78.480 -2.165
26 821 830 87.097 84.143 -2.954
27 793 830 87.097 84.143 -2.954
28 818 832 90.323 87.318 -3.004
29 833 833 93.548 88.728 -4.821
30 830 834 96.774 90.022 -6.752
0.2275 = K0LM0G0R0V-SMIRN0V STATISTIC
293
-------
DESCRIPTIVE STATISTICS FOR X8
NUMBER OF OBSERVATIONS .' 30
MEAN = XBR SUMCX(I»/N 841.30
MEDIAN * 846.00
RANGE 80.000
•INADJUSTED (BIASED) VARIANCE =
SUM(CXU)-XBR)**2)N = UVR 244.21
UNADJUSTED STANDARD DEVIATION = SQRT(UVR) = S 15.627
ADJUSTED (UNBIASED) VARIANCE = UVR*N/(N-1) = AVR 252.63
ADJUSTED STANDARD DEVIATION = SQRTCAVR) 15.894
NEWNESS = SUM((X(I)'XBR)**3)/**4)/
-------
XP
INPUT ORDERED CUMULATIVE CUMULATIVE
DATA DATA PERCENTAGE NORMAL DIFFERENCE
1 846 786 3.226 0.025 -3.201
2 846 819 6.452 . 8.031 1.579
3 846 821 9.677 10.077 0.399
4 847 822 12.903 11.232 -1.671
5 842 827 16.129 18.414 2.285
6 842 828 19.355 20.136 0.781
7 866 830 22.581 23.856 'l.275
8 863 831 25.806 25.848 0.042
9 862 837 29.032 39.337 10.305
10 850 838 32.258 41.776 9.518
11 849 842 ' ' C8.710 51.756 13.047
12 848 842 38.710 51.756 13.047
13 838 846 54.839 61.627 6.788.
14 847 846 54.839 61.627 6.788
15 851 846 54.839 61.627 6.788
16 8/1? 846 54.839 61.627 6.788
17 786 846 54.839 61.627 6.788
18 822 847 61.290 64.006 2.716
19 830 847 61.290 64.006 2.716
20 831 848 64.516 66.332 1.816
21 828 849 74.194 68.597 -5.597
22 • 827 849 74.194 68.597 -5.597
23 819 849 74.194 6R.597 -5.597
24 849 850 80.645 70.794 -9.852
25 846 850 80.645 70.794 -9.852
26 846 851 83.871 72.916 -10.955
27 821 855 87.097 80.564 -6.533
28 837 862 90.323 90.360 0.038
29 850 863 93.548 91.391 -2.157
30 855 866 96.774 93.991 -2.783
0.'2163 = K0LM0G0R0V-SM1RN0V STATISTIC
295
-------
STATISTICAL ANALYSIS
OF
DUAL CONVERSION ACID PLANTS
Catalytic Project 42460
EPA Contract 68-02-1322
by
Dr. D. E. Nixon
296
-------
1.1 Introduction
This report is concerned with analyses of and comments
on several data sets containing information on sulfur
dioxide emission at sulfuric acid producing plants.
Data sets, figures and other computer results contained
in the report and are divided into three sections:
A, B, and C. Each section represents a different type
of analysis using the five data sets provided. The data
sets examined were as follows:
Data Set 1: Data from ASARCO sulfuric acid plant. (A-1,2)
Data Set 2: Data from ASARCO sulfuric acid plant. (A-3)
Data Set 3: Data from Texas Gulf H-SO. No. 1 plant.
Data Set 4: Data from Texas Gulf H-SO. No. 2 plant.
Data Set 5: Data relating to a start-up analysis des-
cribed in Figure VI-1, page VI-6 Appendix
VI: "Analysis of Dual-Absorption Acid Plant
Continuous SO- Monitoring Data."
These data sets are referenced in this section according
to the above numbers (l)-{5).
By way of further introduction it should be reported
that the writer of this section has been completely un-
biased in presentations herein from the point of view
of his lack of knowledge of the operations of sulfuric
acid plants. Thus, some of these results may be refuted
from theoretical considerations. Some additional insights
297
-------
DATA SET 1 AND 2
COMPUTER RUN A-l
1
2
1
4
5
A
7
8
9
10
11
l 2
13
U
is
16
17
18
19
20
'1
22
23
24
25
26
27
?8
29
30
31
32
33
34
35
3A
37
38
39
40
41
47
43
44
4S
46
47
48
49
50
51
52
8?5
810
8*0
875
830
895
820
8*5
870
616
8»5
870
650
840
fljS
810
816
874
874
894
897
8?Q
894
890
870
875
870
825
825
810
830
630
814
854
828
878
815
870
870
875
825
895
850
870
870
890
810
810
870
890
315
780
0
0
n
0
0
n
0
0
n
1016
1100
osn
665
1055
1 ooo
1055
1094
i 004
1114
1078
10?A
107*
1136
1070
1030
1080
1040
1000
955
1000
1015
1070
83(^
92?
104?
1044
1060
1100
1090
0
0
O
0
0
0
0
0
0
0
0
0
0
850
845
91S
850
840
850
850
850
870
8?6
850
fl7Q
675
79Q
floo
825
812
818
834
816
Slfl
832
84Q
820
87Q
84Q
84r>
8?0
8?o
610
*!0
845
71ft
754
8?4
81%
875
810
840
850
840
fJSS
87Q
840
84S
810
810
840
840
840
I 810
\ «40
955
680
850
895
870
90S
870
880
840
880
975
«ss
700
750
07*;
905
912
876
922
014
886
«oo
962
850
880
920
910
860
845
860
87o
930
730
744
864
834
845
87Q
835
840
870
920.
805
88Q
910
870
835
840
910
880
88Q
745
770
7*0
7fiO
76S
70o
780
800
7*5
740
754
835
775
690
615
770
700
760
7Aft
768
709
7*9
768
800
710
770
800
8no
70Q
730
795
770
805
710
700
710
714
715
76S
745
860
610
8 AS
7SS
755
7*0
7*0
695
695
800
770
770
745
755
815
770
765
795
R1S
825
785
740
750
860
795
700
625
770
805
804
762
774
782
770
794
822
710
770
815
815
780
725
780
770
840
800
800
758
778
735
765
745
855
835
86%
740
735
770
770
700
710
800
760
790
760,
715
775
_Z60.
750
775
7flS
785
780
7*8
75?
8?5
7f 0
705
700
70S
775
764
74?
758
760
7s«
754
78?
690
715
795
745
750
705
740
750
7*5
60?
700
66?
706
605
715
710
700
600
89S
67Q
705
740
760
690
700
770
770
755
7?5
715
785
760
760
765
795
790
790
74?
754
f95
7AO
745
820
690
765
796
74?
764
760
767
762
782
700
730
800
800
765
710
730
755
775
688
698
686
702
685
735
740
760
820
825
675
705
740
760
720
735
765
780
760
735
2130
23
SS
894
435
n?
175
33
76
93
191
110
11
1012
19S
26
48
11
182
28
*8
74
102
111
139
124
48
51
701
94
30
136
16
209
57
80
260
48
61
127
416
6S7
2
123
117
19
111
268
29
258
_Z8
1197
7.21
1.95
i.ai
5461
5.28
4.77
4*18
2.29
1.10
0«00
0*00
A. 00
0*00
0400
o.oo
o.oo
o»oo
0*00
0*00
0«00
0*00
0*00
0*00
4.58
4*95
4.74
1.0.1
4.38
5*34
4.34
5*64
5*08
Q.lO
3*81
3*50
1.56
4.70
4.77
2*47
?.30
2*94
6*65
0.32
4.15
6.64
3.32
1*91
6.56
4.31
7.22
?«27
7.24
298
-------
DATA SET 1 AND 2
COMPUTER RUN A-2
53
54
55
56
57
56
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
90
81
82
83
84
85
86
87
88
89
90
9l
92
93
9«
95
96
97
98
99
100
101
102
103
104
105
106
107
108
700
700
800
810
890
810
815
890
890
810
8*5
820
815
810
875
7*8
818
0
8(]4
815
815
815
«35
895
895
806
816
744
666
815
895
«95
89.0
740
790
8^0
815
896
810
8lO
550
710
810
805
8*5
895
695
810
860
8*5
8a5
840
760
810
815
850
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
796
818
n
91«
1110-
1000
1015
lOtO
107ft
1015
850
824
840
82?
955
1080
1030
99S
660
760
1075
950
990
1030
945
760
690
1030
980
1065
IftIO
1050
900
10*0
1090
1075
1080
750
955
955
1040
8f>0
800
800
810
840
835
850
840
840
845
810
845
845
850
855
687
702
0
892
310
810
810
830
81Q
765
604
77Q
764
7-»0
785
890
810
810
670
670
895
815
816
845
810
610
615
780
800
840
845
850
89Q
810
845
845
855
69Q
840
845
865
960
845
89Q
845
355
145
9?.0
870
850
905
875
865
865
69Q
895
766
694
0
852
865
840
860
87o
955
820
784
772
760
754
755
885
875
860
78Q
6VQ
870
855
888
920
°30
760
635
755
820
890
915
920
860
845
910
890
920
705
860
860
915
810
7SO
7
-------
have been gained toward the end of the project period and
these are summarized under recommendations Subsection 1.6.
1.2 Data Set (1)
This data (A-1,2) was extracted from a rather large data
set from the ASARCO Plant and essentially represents data
points associated with upset conditions, during start-up,
shutdown and points before and after these occurrences.
Altogether, there were 108 observations on the following
variables:
X, = First Converter Bed Inlet Temperature in °F. This
is also referenced as BMI1 on some of the computer
print-outs in tae appendix.
X2 = First Converter Bed Outlet Temperature in °F. This
is reftirenced as BMO1.
X- = Second Converter Bed Inlet Temperature in °F. This
is referenced as BMI2.
X. = Second Converter Bed Outlet Temperature in °F. This
is referenced as BMO_.
X_ = Third Converter Bed Inlet Temperature in °F. This
is referenced as BMI3.
Xg = Third Converter Bed Outlet Temperature in °F. This
is referenced as BMO3.
X_ = Fourth Converter Bed Inlet Temperature in °F. This
is referenced as BMI4.
Xg = Fourth Converter Bed Outlet Temperature in °F. This
is referenced as BM04.
300
-------
X_ = Percent 50~ Entrance to Converter. This is refer-
enced as PSO2E.
Y = Parts per million S02 - exit stack. This is refer-
enced as PPMSC>2.
In A-l, 2 these variables are listed for each observation
in the above order with the exception that X- and Y are
interchanged. Data values with a value of 0. denote
missing observation (these were not recorded on original
charts). Of the 108 points, only 54 contained complete
data.
As a first attempt to explain the dependent variable Y
in terms of these processes parameters simple correla-
tions were computed. Also scatter plots such as A-4
were obtained. Following are the simple correlations.
Correlation
of Y with Correlation N
Xl -0.0214 107
X2 0.0265 70
X3 0.0147 107
X4 -0.1112 107
X5 -0.1356 107
Xc -0.1995 107
D
X? -0.3390 107
Xg -0.3390 107
X 0.4030 94
301
-------
JOB
PflOc
DAT* OCr> IMvf
*N*Lrsis« REGieS
DAT* SET" |
• A
N.
P*OC N • 108» flEJ. 1 •
0.
SUM NT • 106
•Jrt
501
— — — • 3200.00 -
— - — =p^
'. *
!,00 540.00 510.00 420.00 « P
H
0 1200.00 -
640.00 -
• /
0.00 •
50
-
1
0
1
... -
I 1
k f* i i i j**^
• 13 « « (?
'
t 1 1
— 1
1
i
i
—
i*
(•> <$> G>
• •
___- —
•
•
•
•
«
•
•
•
• • *•
•
•
•
••
i* •
»_ .
> •
>• • • •
>• •
»>>*« — *
»•**••
.00 540.00 580.00 620.00 640.00 700.00 740.00 ,80.00 820.00 860.00 900.
0
0
San,
cOT-Tgfl PLOT
.___ASAECC
COMPUTER RUN A-4
-------
/.WALV'IJ" MlTUT 1?
DATA SET- i "EAD *• IOB» PROC N • toe*
MULTIPLE LINEAR R E 0 R
4flOU» NO 1
NO nr tNIEPfuoesT V»ai»BLeS g
CPS" 9*0000«-11 TOR SINGULAR MATRIX TEST
DEPENDENT VARIABLE a PPMJQ?
MEAN) ANO STlNOARO OFVIATION9
VARIABLE MEAN JT.OfVlATlON
RMI1 610.6607 46*7637
BrfOl 979.1071 113*1216
>)MI2 40A.049) 99.9Q90
u RMOZ 639.4107 70*r<922
w QMI3 7)7.69A« 77.7620
RM03 740OS7J Al*7^23
flHO* 7llil764 49*1660
PPMS02 479*;0^~ 77|.|671
NORMAL NATRIV*
12345
ROM 1
1)0764*9937 194031*0)96 114400. A9A) 9213?*40)7 667)4*2)19
ROM 2
0*0000 701407*1972 32399?*4M| 14692^*9)99 219401*421)
1)9976*9000 I3t<94>9267 6?ni*)997
ROM )
0*0000 0.0000 194746*9939 900774*9469 |52466*91?6
8o9U.?500 67774*1072 42)4*7614
RON 4
0*9000 0*0000 0*0000 274079.99)7 17)791.9619
124169*7)00 1?0771*4931 4A17.rp0A
ROM 5
0*0000 0*0000 0*0000 0*0000 326)17*6)91
46'34*f900 6)987*0)99 X81.497]
0*4000 0*0000 0*0000 0*0000 0*0000
26174(0000 6'tft6**?4' 2&9A.0497
RON 7
0*4000 0*0000 GiOOOO 0*0000 0*0000
l»l»01*9000 106148*9002 1691*7000
RON 6
p«ac i
BASIS 4.0
RtJ. N • 0* SUM NT • 106
"••
ESSION A N A L T S I S
6
59914*7656
199957*6972
122976*9144
141)22*7696 A,<
COMPUTER RUH A-5
)13303«071)
)67966*A972
0*0000
-------
0.1000 0*0000
0*0000 112198.9144
AO If 9
' 0*0000
0*0000
CQ«(RCL»T!OS MAI
ROM i
1 .AOOO
0*4483
ROM 2 '.
0.4*630
ROM 3
0
-------
JOt OEF" ••••*•
PROC OEF*> NIXON
DATA OEF« n*VF
TJTJTTJT
t
PAGE 2
BASIS 4,0
OAT* SET"
106*
PROC N »
108*
RCJ» N •
0»
SUM NT •
J08
1/26/75
VARIABLE COEFFICIENT PET* COEFFICIENT
CONSTANT 82
RN01
8HI2
RH02
HMI3
RH03
IJHI4
10,9130
4.3215
o.ono
4>6*OA
1.6510
-5.74S3
1 ^5938
9*6670
PS02E 218.3635
STANDARD ERROR OF FST1 MATf ............
COEFFICIENT OF OETER^INATION (*oj)«,,.
01 MULTIPLE CIRQELATION COETICIFNT CAOJ>
-0*2731
•0,0016
0,3581
•0.15M
-0,0750
•0*1690
•0*1627
•0* J04?
0*5677
•
a
•
•
•
499.8371
0,4421
0*3330
0*66(9
0*5770
ST*NO*««0 OEVHTtHNS «NO T VALUES Of
HEORESSION CTCFFICIENTS-
VARIABLE
BUM
BMOt
BM02
8HT4
BM04
STO OEvUTION
3.3081
318597
6,8677
T
-1,31
•0*01
0*84
•0*43
•0*22
•0*40
-0*76
STO OEV RET*
0*209?
0.9730
a!3527
0.33*9
0*32*9
0.402?
F
1.71
0.00
0*71
0.18
0<05
0*16
0*57
P507E
00,4606
3»61
13*04
HIGH OtOFR PlRTItl CORREL ATlnu CnFFF AND R?-f)FLCTF
VARIABLE PARTIAL CORR CnCF RJ-OELETE
COMPUTER RUN A- 7
MM
RH01
HMI2
0«U9t
'0*0009
0
-------
tfliiAPC n.r.
-------
Jot oer-
Pant
(^ DAT* OFT"
( ANALYSIS*
OAT* SET-
Q1VF
MllLTR
1
READ N«
BASIS 4.0
1/26/75
ioe» P*QC N • ice» REJ. N • o» SUM NT • IOB
— — •
Tifti r nr -fiPstnim <
DBS NO*
1
2
1
4
5
6
7
8
10
11
S 13
•J 14
15
16
ir
18
19
20
22
23
94
25
96
27
28
29
10
11
12
11
14
35
16
37
ia
, 39
40
41
42
4}
44
45
ACTUAL
131*0000
119*0000
121*0000
48*0000
51*0000
701*0000
94*0000
30*0000
118*0000
16*0000
209*0000
17.0000
60*0000
260*0000
48.0000
61 ,0000
7.0000
199.0000
164.0000
41*0000
60*0000
4 3 -• 0000
9461*0000
229.0000
1.0000
419*0000
9145.0000
873*0000
1863*0000
255*0000
1695*0000
65.0000
?1>.0000
16.0000
o.oooo
1311 .0000
21*0000
54*0000
1177*0000
18.0000
49.0000
21jQOOO
H07.QOOO
1099.0000
995.0000
PREDICTED
993*0955
613*3399
1 -247. 1463
247.9943
1091 .0495
46A.7005
614.0999
19A.7111
•340.6139
401 *01*1
791.99*8
1083.3*97
691 .9917
25*. 11 13
131 .2092
1959.5003
7?9.7*39
205.2*00
500,0131
219, MB9
•444,0467
601*0601
1239,99*8
694.6**2
505,3111
1294*4496
196*7098
"279.8*90
•127.9064
990,4999
59*. 5*91
•95.0112
84**7056
-4*. 0142
609.9972
593, 3971
161A.9017
1371,1*11
64A.618)
RESIDUAL
•852.0255
•474.3399
169.6514
295*1461
•196.9243
•390.0425
•374.7005
•584*0299
1 1.9S45
356*6139
•192.0161
•737,9985
•503*0775
•893*3897
•573.9J57
•197,3113
•194,9029
•1123,5001
•565*7839
•1«2*?600
•440,0131
250*0121
1448,7164
-10*9389
46710467
•171*0601
915.0019
2*8^9114
986*5412
•250.1111
400*5574
•111*709"
•172.9964
286*8699
I91i'*9064
190*5771 COMPUTER RON A-9
•507*1891
149*011?
374*2944
66*0142
•551.997?
•502.1971
1490*0901
1790*8169
Ki6*14l4
46
211.0000
.1674
•175.1674
-------
I 46
} 50
51
53
54
69.0000
J?«0000
970(0000
?28(0000
21(0000
1009(0000
^n.nnno
? • ?^49
971 (3< 99
631 100)3
• 01 * • AAAA
7M*.1A73
29l«03l
•403I&931
102(412?
54
M.0004
•NiTSOs 0
1i3T3o
o
CO
COMPUTER RON A-10
-------
Only the correlations between Y and X? and Xg are sign-
ificant at the five percent level. Since the difference
between X7 and Xfl is insignificant (a fact no doubt
characteristic to the upset conditions which the data
represent), these two variables are regarded as the same.
It is felt that X_ is not influencing Y but rather this
correlation is significant because of the following
reason: When the fourth converter bed is finally brought
up to temperature, SO- emission is lower, the nature of
start-up. It should be observed that almost all of the
above correlations are based on the full 108 observations,
unfortunately not the case on omitting missing data in
fitting the following model:
K-i
Complete results of this fit are contained in A-5,10.
While this regression equation gave a significant fit
to the data at the one percent level and the model
accounted for approximately 44 percent of the variabil-
ity in Y, its usefulness in prediction is questionable.
The model does seem to predict "rather close" to the
high SO- emissions, in order of magnitude, but is some-
what more erratic in predicting low to medium SO- emis-
sions. The writer would have liked an additional data
set having similar characteristics as (1) in order -to
verify the reproduceabil.ity of the model. Since this
309
-------
set was not available, the 108 points were split down
the middle and the model refitted to the last 54 points
of which 40 were complete. These results are contained
in A-ll, 15 and they appear to be comparable to the
original fit. In each case PS02E was the most signifi-
cant contributing variable to the regression and, in
fact it should be noted that the correlation 0.4030
between Y and X_ compares with the 0.413 reported on page
VI-13 of the appendix referenced in the description of
Data Set (5). It is understood that the latter correla-
tion was obtained from the larger ASARCO data set with
upset conditions removed.
In summary, the high variability in S02 emission and its
erractic behavior during these upset conditions seems
to preclude a satisfactory explanation in terms of the
process parameters considered. It is felt that the main
reason for the inadequacy of the model in explaining
S02 emission during upset conditions is due to the fact
that data was reported at two-hour intervals and, in fact,
since it is possible for a plant to shut down and start
up in just a few hours, the few data points recorded does
not adequately report the changing nature of SO, emission
and the process parameters during these times.
1.3 Data Set (3)
This data consisted of 54 observations on-the variables
described in Section 1.1 except that in this case the
310
-------
OAT* ocr* invr BOSIS 4«o
ANALYSIS- KULYR 1/26/75
DATA SET" 1 *EAO N« 94* PROC N • 94* REJt N • 0» SUM NT • 94
MULTIPLE LINEAR REGRESSION ANALYSIS
,*•—;— '
CROUP NO I
NO pr iNoEPEqnrNT tf<9t*aics 9
EPS* 5.0000«-11 FOR SINGULAR MATRIX TEST
DEPENDENT VA914BLC *PPHS02
Mfiut lur^ ^T.urnop nruttTtflue
VARIABLE MEAN STtOrvIATION
hi
t-"
NORMAL
Mil
9M01
BUT9
MO?
MIJ
riM03
MI4
RM04
PSO?E
PPMS02
MITRIV-*
804.3250
999,3500
834.0000
730.0500
707.4750
705.9750
9.9467
616.325Q
16.3780
191,8369
76.2849
88.0590
91.0989
50.3915
45.9439
2.1432
872.3052
18«?77.4500 HH13.6?50 91«3«OOOO 63463. 3501 43)29*9900
ROM 2
0.0^00 97*925.0999 976390.7SOO 9889(1.0000 183349.3000 167903«0»99
108101«3499 963 Q9O4.99 6599.«755
ROM 3
O'OOOO 0<0000 173994.3750 173043.0000 135667.9500 110911.9300
67931.3750 30638.4790 3541.1538
HOT .4
0.0000 0.0000 0.0000 226936.0000 147384(0000 126036,0000
101927.0000 91989.0000 1714.S7QH ; COMPUTER RUN A-11
ROM 9
O.flOOO
24108(11500
MOM 6
0(QOOO
4879*8499
ROM; 7
0,0000
98797(9749
0.0000
38982.0500
0.0000
37038.8499
0.0000
81956,4749
0.0000
2934,3669
o.nooo
?1S9, 8605
o.oooo
1 149.9111
0.0040
0.0000
0,0000
302393,9001
0*0000
0*0000
289166,3000
323661.0999
0.0000
RON
-------
10.0000
0<0000
, ROM 9
0*0000
0*0000
CORRCL4TION
1
ROM 1
1*0000
0.4740
ROM 2
O.fl"954
0*4521
ROM 3
0»»575
0.5158
ROM •
0.5«??
0.6808
ROM 5
0*J278
0*1395
ROM 6
0.2273
0*8273
ROM 7
0*4740
1 n)000
ROM 8
Oi30«7
0««0*9
ROM 9
0.5*92
0*2734
ROM |0
0.0360
•Oi3217
0.0000
82322*9749
0.0000
0.0000
•UTHIK"
2
0«495»
OOO*7
1 .0000
0>Mt2
O.S719
0.«2)8
0«'970
0.4824
O'MJO
0>?426
0>1879
0>'269
O.«5?l
0.^089
0««412
1*0000
O.A481
0»?t9S
0.1785
•0*3112
0*0000
144.1768
0*0000
179(1465
1
0*H75
O.M9?
0.*T19
0.6481
l *onoo
0.^350
9*^718
0*SA57
0*^921
0*3987
044679
0*9836
0*5158
0.7734
0.4738
o*'i9a
0.6350
1*0000
0*1136
0*472b
0*0000
0.0000
•
o«n*52
0*0360
0.7070
0.1785
0*8718
0.1136
1 .0000
•0.0437
0*563«
•0*0488
0.4650
•0.0872
0.6408
•0.3?17
0.482*
•0*3812
0.5857
0»47?8
•0.0*37
1*0000
0*0000 OtOOOO
0*0000 0*0000
5 «
0*3278 0*9973
0.4*30 0*1879
0*5921 0*4679
0.5634 0*4650
1*0000 0*9243
0*9243 1*0000
0.1395 0*0273
0*?426 0
-------
JOt ^t*" ••»•••
PROC orr« NTKON
DATA DCr> n»\/F
ANALYSIS* MULTR
DATA SET* 1 READ N« 94»
— •—-
< 'f '
REGRESSION COEFFICIENTS-
VARIABLE COEFFICIENT RET* COEFFICIENT
CONSTANT 9101*994«j
UN 11 -6.6094 -OM27*
*>««0l ' '-«M* p. 1979
SMI? 4.7)0? 0.9199
(H02 -S.OOJ1 •0«4ir«
flMTl -9.79?« -0.9)1*
HMD] •O.?2*0 *0«0?34
DMI4 -1.941? "O.O'OI
HHC« "S.Q9B5 -0.2664
PJ02C 281.131? 0.69Q7
COEFFlCtfNT OF OFTFR>M4| 0.5519
BNIJ -0*0997 0*9634
IJM03 •0-0094 0.9679
HM!« •0>0?A9 0.5*74
•)H04 *0*04«0 0*5649
P50?E a«91»« 0*3391
, HOE 2
BASIS 4.0
1/26/75
PROC N • 94* REJ* N • 0» SUM »T • 54
«9J. 7521
0*9679
0.4303
0*7934
0*6621
OEV BETA f
0*243A 3*00
0.1100 0.40
0*4941 1>09
0*4146 1*11
Oi421t 0.10
0*4537 0*00
0*5709 0*02
0*9444 0.24
0>i'J3 19*89
fLETE
COMPUTER RUN A-13
-------
ANALYSIS Of VARIANCE TA«JL£
SOURCE o.r. VJM SQUARTS
ERROR 10
TOTAL 3
LPVEL of ft <». 30) • 99.901
COMPUTER RUN A-14
-------
JOH DEM
PROC Off"
DATA REr»
' ANALYSIS'
DATA SET*
••••••
NIXON
otvr
MULTR
1
READ „•
PAOC 3
BASIS 4.0
1/26/75
54* PROC N • 54* RCJ* N • 0* SUN HT • 54
TtR| r or nEstniiti <
OBS NO*
1
2
4
5
A
7
8
9
10
11
13
14
5 :;
ia
19
20
21
?2
?3
94
25
S7
20
29
10
31
32
33
34
35
16
37
38
39
40
ACTUAL
7*0000
129*0000
41*0000
60*0000
•i.nnnn
9463*0000
222*0000
1.0000
432*0000
2145*0000
87i.nnoo
1863*0000
255*0000
45*0000
212*0000
I AiOOOO
0*0000
1311*0000
21.0000
. 54*0000
1177*0000
18*0000
49*0000
21*0000
iior.oooo
3099. OCOO
995*0000
911.0000
89,0000
32.0000
970*0000
228*0000
95.0000
21*0000
t009*0000
350*0000
71*0000
PRCnlCTfO
59*9019
1319.2146
429.5530
1494.7935
556.0416
1379.9045
AA';*8-.90
1005.3818
765*1948
401.6012
f
I "161 .5950
/ 120«>9191
619*5970
I '241.1945
1095*1547
•145.7956
767.5104
578.0504
' 2219.5975
1093*1494
607.1949
74*7455
143*1099
811 .5415
1950.3745
100»»7621
•919.A177
•278.3445
999.7096
77*. 14*0
760.3483
RESIDUAL
•45,9019
•1183.9144
•5A9.ni09
•3'9.5530
•653.646?
1016*2065
-74.1514
118.9ft.AO
•1?4.0436
765.0955
901.1 460
857.618?
•510.1941
159.1975
•316.A019
•396.4439
At 1 .<484
161.5950
101*0409
•598.5?70
295,1945
151.8451
143.795*
•716,5104
•557,0504
«47.*725
•96il
-------
1
2
1
4
5
6
r
a
9
10
tt
12
13
— 14
15
16
17
IB
19
20
21
22
2)
2«
u, 25
s *•
o» ?T
28
29
10
31
32
33
34
33
)A
37
30
39
40
41
42
»3
4«
45
• A
47
46
49
40
9l
32
33
94
806
844
803
883
603
81)8
696
611
811
609
60*
799
800
M3
607
f99
601
*3
80S
603
802
80*
800
680
7*8
ao3
001
795
795
a«j
800
880
600
641
601
ano
640
603
842
803
683
602
842
803
803
889
809
602
793
6*0
796
603
803
603
ll?3
1120
1117
1117
1120
1123
11?8
1134
111*
1130
831
11?1
10?4
1071
11?9
0
1121
11?5
1130
1123
1103
1104
1091
1099
1103
1133
1 1 16
1178
1105
1118
1111
1116
1101
1100
1101
1109
1104
1105
1100
1106
1102
1100
1100
1110
1100
1113
1117
107
126
1?0
116
126
127
1123
849
846
84)
84Q
815
646
848
855
851
653
854
84?
«»2
610
448
850
840
0
641
.819
442
9«2
819
818
898
816
851
1*2
84Q
844
««1
84Q
444
842
6*3
84t
«42
6*2
841
843
641
640
84Q
613
1«6
849
850
812
7<»l
648
Mr
834
6)4
614
984
944
082
90Q
981
963
960
963
986
906
985
974
149
870
919
950
968
969
976
911
977
975
975
971
960
960
469
964
960
960
979
979
965
969
966
987
967
967
965
986
966
966
986
960
999
9»?
999
994
766
960
983
972
972
971
811
614
801
603
8n«
808
'800
80*
611
813
614
6l«
755
745
741
763
7«6
Bon
8no
60?
611
611
611
808
604
609
619
so*
604
60Q
616
619
610
81?
810
809
810
609
809
809
609
610
610
800
896
896
696
811
804
001
617
614
6tO
806
849
649
MS
646
645
646
830
842
650
651
650
652
767
779
740
762
606
624
630
620
846
646
645
641
634
840
840
636
831
640
650
849
848
630
649
M8
846
649
840
847
850
850
850
833
669
866
867
658
812
813
346
650
874
642
693
893
690
618
614
691
610
89|
696
69 f
897
676
779
770
750
769
795
841
613
803
641
601
603
602
606
612
815
61?
608
646
610
604
600
807
603
604
605
805
645
605
605
805
605
795
89J
891
091
--646
793
616
613
630
627
824
630
830
849
94<)
946
649
636
847
850
851
650
649
608
795
769
766
812
830
831
837
826
826
827
123
819
826
830
828
824
829
831
830
124
630
626
4?6
129
829
129
830
630
830
830
122
149
146
646
841
421
137
150
855
650
646
190
122
1?7
130
130
1)0
140
141
139
197
131
1QO
45
0
0
123
132
190
110
106
70
90
60
0
60
70
75
55
65
60
80
60
90
90
92
75
95
80
75
80
too
96
67
0
116
110
110
102
200
130
95
too
too
93
4.67
OtOO <
0*65
0*00
Ot89
OtOO
0.91
0;74
0>00
0*71
0*00
0«75
0<00
0.00
4.00
4tOO
0*73
0.00.
0.74
0*00
0.69
0.00
"0.63
0.00
0.83
0400
0.84
0.00
0.93
0.00
0.69
0.00
0.90
o.oo
0.67
0.00
0«6'2 DATA SET 3
0400
0.91 COMPUTER RUN A- 3
0.00
0.61
0.00
-------
paoc
04T4 OEF«
vixov
Oivr
PACE 1
HAS1S (.0
AHALTSIS"
DATA SCT*
READ «•
94*
PROC N «
94*
RCJ« K •
0*
SUN NT •
94
1/24/75
MULTIPLE LINEAR REGRESSION
A N A I T S I S
OROU» NO i
HO nr
EPS* 9*oooo«-n FOR sisouiA* MATRIX TEST
DEPENDENT VA41ARIE • PPHSO?
HCANt ANf) STlMDAfln D'VlATION^
VARtlBLE
NEA*< ST.OfVlATION
0MO|
1114.1930
qM02
9^9.61)4
809.60"
39.2113
?.OM«
»)Mt4
P50?E
0.81 3l
16.4530
0*0847
p
104
30.4C33
NORMAL MlTRI
o* t
1100««152
•4.7192
76 4.0 769 793*6924
ROM 2
0*0000
1A64.9307
4173.3S46
1491. «4«7
3?0>1930
'5.9A71
ROM 1
COOOO
0.0000 34*4,4A14 10903.6192 823.3076 1921*7693
ROM 4
0.4000
0.0300
0*0000 38439*1938 2907.0769 8201*6924
COMPUTER RUN A-16
ROM
0.0000
1429.4614
0.0000
1171.9221
OiOOOO
•0.1946
0.0000
1230.0386
2130*8462
•
ROM »
0*0000
2424.1916
0.1000
?6?9.Y07«
OlflOOO
-4.5919
0.0000
0*0000 9396.6199
irai—r
0*1000
>767.«J383
TTOU—B
o.oooo
1649.0769
o.oooo
9.0494
o.nono
0*0000
OiOOOO
-------
ROM
0.1000
0*0000
9
0*1000
0*0000
COBRCL4TTON "i
1
ROM
. ROM
ROM
ROM
ROM
£ ROM
ROM
ROM
ROM
ROM
1
1*0000
0*1/0*
2
0*31/3
3
0*5073
0*1360
4
0*4056
0*>660
5
0*23*1
0*4941
6
0*3190
0*46/3
r
0*1/04
1*0000
8
0*5436
QtM/4
9
•0*3465
to
0.3391
0*8024
0*0000
3522*1538
0*0000
0*0000
fj*j f X"
2
0.11144
0*^4 )^
I * DOftO
0*1690
0*0943
-0.19?3
0.3491
-0«t /?/
0*5630 .
0*11/3
0./4/4
1*0000
•0«3|43
0*5/06
0«'421
0*0000
•7.9000
0*0000
0*1/9*
J
0*5123
•0*346)
0.0443
1*0000
0.9111
-0.2B86
0*1994
•0*0131
0*4451
•0*1453
0*1)60
-0*»596
•0*3143
•a. 9791
1*0000
•0*35/9
•0*1630
0*0000 0*0000 0*0000
0*0000 1*0000 0*0000
• 56
0*4056 0*9341 0*3l»0
0.33°1
•0*1993 "OO/27 •0*21/3
0.5/06
0.0115 0*3994 0*4451
•0.15/2
1.0000 Oi42?8 0.5695
•0.5468
0.422A 1*0000 0«<2/1
•O.J531
0.5695 0*82/1 1*0000
•0.1?/2
0.9660 0.4941 0*46/3
0*0024
0*1*91 0*5630 0*6490
0*7421
•O.?1 16 "0*0131 •0*1453
•0*1630
•0.5468 -0.2531 '0*32/2
1*0000
COMPUTER RUN A- 17
-------
PRQC off* iixnN
OAT*, Offo DAVE
ANALYSIS* MOLTR
DATA SET" 1 "CAO N« 94»
VARIABLE COEFFICIENT BETA COEFFICIENT
CONSTANT -1164.5826
pMOi 0.1998 0.0147
BMI2 -». 1128 -0.8151
RMP2 fl.08'6 0.1101
SB :$j|j 3|K]|
£ P$02E -30.8526 "O.Q85B
vO
COEFFICIENT OF DETERMINATION fAOJ)...,
MULTIPLE CORRELATION COE»FICIFI?419 9.99
0.1158 0.55
EL£l£
COMPUTER RON A-18
t
-------
Of VARIANCE TABU
n«r. <\JH
RCORCSSION 9 |«6»J«9«M 2jflfl.??19 9.79
16 '
TOT»L ?5 21200O4A?
>r r( 9/ 16) > ioo»oot
COMPUTER RUN A-19
-------
J01 Of" ••••••
•••njc OEF« mxON
tmr» nrr* mvr
ANALYSIS" MUITR
OAT» SET* t HEAD N* 54*
PACE )
CROC N • 54* RfJ* N • 0' SUN NT • 94
TAOI r nr prcfmt.i e
DBS NO.
1
3
4
5
6
a
9
to
II
J >3
15
i-r
90
'2
94
?6
DUfqilN-MlTSOs
ACTUAL
120.0000
127.0000
1 JO. 0000
140.0000
141.0000
127. 0000
100*0000
139*0000
1 1 ft . nof)9
70.0000
80*0000
60.0000
75*0000
65*0000
80.0000
90*0000
92*0000
95.0000
75*0000
100*0000
87.0000
118.0000
110.0090
900.0000
95.0000
100*0000
0 STATTSTIC
PBCOICTCO
121*1367
127.9907
195.5090
197.0902
1?R<4990
96.8331
117.6582
( 1A.7014
99.8039
51.0956
84*8518
9| .1559
79.5977
89.9416
84*3802
94.2*60
94.8*45
105*8397
105*5764
109i43?5
• 1.5653
RFSTOUAL
•6ilJ67
•0*9907
4.4980
12.9701
12.5080
9.7585
3*1669
14*3418
•96.9014
•19.8039
•8.5561
6.9744
"9.8518
•15*8247
•11.1559
21*0560
12*4723
19.0584
•9.380?
5.7040
•7*8845
12*1603
3.7699
4. 1977
•10.5764
•2*4)25
•
COMPUTER RUN A-20
-------
plant was considered in normal operating mode. The same
model as in Section 1.1 was fitted to the 26 points
having complete data. The results are given in A-16,
20. In this case the regression is significant at the
one percent level and accounts for approximately 85 per-
cen™ of the variability in SO- emission. In addition
(See A-20), the model very nicely predicts SO- emission.
Estimates of the coefficients in the model are given at
the top of A-18. No attempt was made to reduce the set
of independent variables to a smaller set and this may
be possible through additional analyses. In view of
the interdependence between the process parameters it
is quite reasonable not to expect signs of particular
coefficients to have some a'priori value (+ or -). In
fact, the usual interpretation that a given regression
coefficient measures the change in the dependent variable
with the ether variables held constant is not appropriate
for highly correlated independent variables. The model
simple presents the best linear fit over the range of
values observed.
1.4 Data Sets (3) - (4)
These data, obtained from two similar plants of Texas
Gulf, offered the opportunity to examine the effect of
catalyst aging over time. Each set has one to three
pieces of data for most months from March 1968 through
December, 1973. In addition, these data contain values
322
-------
of SO,, not previously considered in this section.
Graphs of SO~ and SO- versus time (Bl, B2, B3) suggest,
for the SO- data at least, that there are definite trends
starting at what appears to be the catalyst cleaning
month, CCM. In order to test the hypothesis that SO,
is linearly related to time the following model was
used:
Where Y = mg S02/SCF and T = month since CCM. It was
assumed that the CCM for Data Set (3) occurred in June,
1968, August, 1969, and August, 1972, and for Data Set
(4) occurred July, 1968, August, 1969, and June, 1972.
Two comparable samples were extracted from each of the
data sets and all observations for a particular month
were averaged to obtain one value for each T. The re-
lationships are shown in Bl, B2, B3, and the results
are summarized in the following table:
S02 Versus T (from CCM)
From (3)
Frcm (4)
Sample 1 Sample 2 Sample 1 Sample 2
13
7.9
121.1
.79
24.7
14
17.9
36.3
.87
47.4
12
8.0
127.8
.72
29.0
17
15.7
33.1
.78
72.4
correlation
std. error
Each of these regressions is significant at the five
323
-------
percent level. Interestingly, there is no significant
difference between comparable samples from Data Sets (3)
and (4) indicating that this measure of catalyst effect
is the same for the two similar plants. The -fact that
the slopes for Sample 2 were about twice that for Sample
1 was apparently due to the fact that these plants burned
"dirty sulfur" during the period October, 1972 through
August, 1973. Unfortunately, this fact masks any conclu-
sion of a catalyst effect difference over years. The
SO, data for Sample 1 from (3) and (4) did not correlate
with T; moreover, no trend was examined for Sample 2 trom
the two data sets due to any trend line being obscurred
by "dirty sulfur" burning. A comparison of SO, means
for the "dirty sulfur" period versus the other months
for each of the data sets showed an extreme difference
(for Data Set (2) these means were 22.7 and 1.4 respect-
ively) . A comparison of overall SO- and SO, means between
Data Sets (2) and (3) showed no significant difference.
Analysis of variance of the SO- data (B4, B5, B6, B7)
illustrates significant month and year effect for each
of the data sets except for the month effect for Data
Set (3) . It has been suggested that a reason for this
is the fact that "there was a less controlled operation
of Plant Number 1. The writer understands that the
latter statement will be verified and the burning of
"dirty sulfur" will be investigated on further contact
324
-------
5$ 2ft: '"&
nr C»TII.»ST rrnci
1E40 «.
107,
r N « »07. HEJ. N . 0, SUM
CONTACT ACID PLANT
SINGLE ABSORPTION (CHEMICO)
*T •
107
e»m »,o
U/ 5/74
so?
UTu*L
V I/I
2S9.0
16}!oJ
1 41,0
17|.0
I2S.1
i'«!o
214. O-
|93lo
117. n
117:!)
1)4.0
i IJ«.0
IS7 »0
18l.O
t AM *fl~
' 1 41* 0
-ttj-r
ilj-2
1 ?>*L ft
243^0
Sil:S~
ST ft
»3*0
- ?ii-
isjto
Jl4>0
119.0
3s!o
— 17(1. « —
7.011.0
JSU
1*7.0--
/OA.O
— ll]If»—
zii.o
16ft. 0
?ȣ. O
9 Va f|
2?0.0
ZlAln
18»:S-
— 207. O—
1 ?4 .O
237.0
1— -207.0 --
9 0.0
rn
. u
i \ OA .*>-•-
J1 I 3».1
Sjlu*
20| ,n
1 31 .0
133.0
- 140.0
|7|.0
144.0
1'«!n
'I?:!!
— JOA.O -
-4t>iS—
IS 3*0
_L»O:O
»^*>.«« w i LIA r* wi1
1 41*111 nir-Jl I.irr K..C
— VALUE. rmjoy U4TC URS.
NO.
16?. rt 9A ^
IA3ln "o"^ 4/lV/e4 2
{,;•, •<'?•} s/ 1/61 4
— 1*47 i2< S/16/A4 b
!/;•; '*•' s/71/ftn 6
14?:s .Jl;^ J' j.l '6" 7
~l*6'e -?s'i I/'o/JS IU
1 A 7 * 1 i i ^/ti/ft^ |i
- u'-j :r:i 3/'»;sH {?
>»"'5 -s?:S s/iS/jj i;
_ »*:5 ^!»:^ li/u/A? }*
ILO r _I ? * • tO/?J/6*$ 1 (J
••*•* i^«)* I1/&/A4 lO
/n • 4*3 I 1 / " o 3 l»
'8:* JS:* • ti/o/6i 20
7?'» "^J'J 17/10/A4 22
171 A .it'I !/ 0/A» ?J
I7>*ll ii*i - 1/7'/6» 74
-»|:S H: ''ii;" "
171 i T?*. 1' J/A9 2/
I7J*4 "I!"* J/H/67 2«
— 173:3 -?2 a l/l'/69 7«
174 1 1V» " ' *7 V/|S' '"
I74« 2» < 4/30/69 31
i75:S *J:. 5/7e/*'' ??
~l7il4 ?5:? " "" «/!«/*» 34
176's -Ta'J 9/11/A9 36
17 O *19.S 9/17/A9 37
I?/;' liJ;' lo/li"** J"
- ifa'i "J'i 11/PO/A9 41
17? 5 .jj, 10/3U/6V 42
•'_"••* r^4>^ 11/7b/AO A i
I/O 3 • 1« w i|F^i*rOV «J
I7« » *?!*U l/IS/70 4b
i'v.6 i«.« _ l/2'/70 «»
- ISS:2 ;j::^ ''JJ^' ;•
!»}'s -TJi'i */i5/'I! I*
«2':» «:? {?i?siEs H
!sS:i ?s^: '«'«« ?i
!"*'« "'* !?/!i/J« "
I"!'? T.'t*^ 1/'*/?V S7
I«» I i7,*2 4/74/71 SB
!s?:,j . ?:; «ii;;{ 23
!& , S:i -'''J''! 8J
IB?, t, , I/ 4/77 64
j"!i5 i?:» i/w/7? A*
Us:* ^'' j'iy'' **
(•"I" 77 » 1/7C/77 67
I?;.? ':« J/U/7? j;
149'i "f?'i S/ 1/77 70
M»IJ oi S/l ~'j*j *^ / »f / /? 7ft
!«}'? !;/'> °/i :/" 7/
Iga'I .1''. "'1//77 7B
I"?!* -CH|j - n/U/77 /V
lyj.' -"'ili' !?'^'^I2 ""I
|9u's " l?/?6/7? 9J
I")*'? .»i ' '/ »/'» 4'
19?:? -.4'5 7/20/7J ^>
n.s:« -s»:« y >"i 10
1 4 A i i«.- t/tt/7) d/
mil I?i' 1/l»/7i f.e
.* -I*'* »/ <('7» nv
197 , .*o', »/l»-'»l 9U
197:1 i'? •". •>/ CNOPOINTS Or*?RilNU \.l*r
22 JL':2° I60'0» Z40.00 320.00 400.00
»"""""_!""
* 4
A
A
A *
* ! *
» « ••- -
A
t
»*
* - • -.„ . — . _ 4 . _
4 - "
A
*
"A • — ~ -_- _- — • -,-^ — . _ „
t *
.*»
•M
4 1
. *
| - - A -• .. ^_ . » .T.
; •
A
A
A •
A
A
A
!. ..
. •
A A * *
." i •
* . --
' A
A
A '
' . • —
A
*
! . '
> A *.._
A
t f
A
. A ....
A -.--.--.
A ^
A
1 • •
A
4
A
• .
. f
•
^ p
4
* " A* »""?
A
A
4
i 4
*
A
A
A
A
A
A
»°
325
-------
JO" OFF* ••••••
'"DC Qlt* Oiyr
D»Y* OFr« il<
itoor or CATALYST trrccr
flEAO N« • 44t
REJ. N • -- o.
9J -
MCC «
BASIS 4.C
It/ )//«
A EMISSIONS RBM.
22S.O
151.0
1*9.0
*•»••«
1*1.0
—20*.O •
217.0
III*0
74.0
I««.D-
20.0
}36.g
!«»•?
—170.S
-I??:*
'SO?
istsn
VALUE
14V.«
IA1.3
IA0.4
161.3
141.0
4?.3
,62.8
43.J
63.9
4*. 3
.44.8
44.3
4S.4
166.)
166.a
147.3
167.«
16(1.3
I6ll.ll
169.3
149.0
170.3
170.4
171.3
171.8
17?.)
17?.d
173.3
173.8
174.3
1/4.l»
174.3
J7S.4
17o.«
176.4
177.4
177.9
170.4
176.9
1/4.4
79.9
*.•
I44.4
15-J.4
76*.0
:— 33*.O
201. S
202..1
207.%
• J01.0
203. i
2l)4.4
204.4
20S.O
- «».*
?06.a
Reproduced from
bcsl available
COMPUTER
IU1E
J/7//6A
/17/AK
S/ i?/ Aft
4./I4/64
S/ 10/64
4/ 9/44
4/1 J/»4
4/76/64
4/ 4/64
4/77/A4
9/11/64
9/|e/«4
9/24/64
10/17/64
11/?2/*4
lt/l«/64
1 7/1 ^ /^ *i
I/ 7/69
j/74/64
3/ 4/69
3/1 4/69
4/1*1/69
l/2'/69
A/ 4/69
It 3/69
-• - 7/J3/64
107 V/69
10/74/69
It/26/69
3/31/70
• 7
-------
STUDY or CATALYST rrrici
RCiO «• 9*. PUCK N •
9S.
RCJ. U r -0. IUM »T-» IS-
f»«E
"fiJ
"J01
ACID MIST SO^
COMPUTER RUN B-3
ACTUAL TREND TBni'l
vAluE VALUE raarn
1.4
-1.0
i;!
-».4
•l.n
•ill
-1.2
-I.o
-o.v
-0.7
2.' -0.4
1.8 -0.1
-».e —-o.i
2.2 S.2»-02
0.6
O.V»
1.7
7l*
9\*
ill
7.1
9,*
7.7
1.1
0.9
1.1
• 0.7
-0.5
•I.O
-*',•,
-n*4
-«.4
:;:?--
!!•'
--ii* -—
-I.*
••ii
•ii*
-4.'
-J^S
•-«.»
-si;
-sli
I.I ft.5 -«i
I.? 6.7 -*
-6.5 6.(I - -*
«.l
IU.1
7».0
-11.8 -
11.9
-*.»
-*.»
-ft.4
-ft.7
-ft.*
-7.1
-ftlft
•7.0
-*.?
•7.1
*.*
7.S
A.S
inls
17.«
l».;
lft.4
-11^»
15l'
*::!
!i:5
-9.0
-9.4
•in.1
-II.>
-11.7
-II.*
-II.*
UAIE
1/27/68
4/1//68
^/10/«8
*/lJ/»«
6/26/64
«/ 0/68
4/ln/AK
1/7//A8
•J/|0/k«l
»/?*/68
IO/ 1/68
0/22/68
!/!•/*»
1/21/6*
2/IV/6B
J/ '/69
/2«/69
I/ A/ft«
1/14/69
4/IU/69
*/2'/6»
«/ 4/69
7/22/69
9 /« J/69
1/26/»«
2/10/70
!;*}«!
?^J«8
*/20/ro
4/21/TO
8/75/70
10/1V/70
n/io/ro
it/i'/ro
11/25/70
I7/I//TO
1/U/71
0/20/71
2/ 6/71
1/15/77
1/21/72
4/14/72
•/75/77
^/l»/77
7/21/77
T/7S/77
7/J1/77
H/16/7?
K/IO/77
10/ 5/77
??'??«
I/ V/7?
1/20/77
|7/ «/77
5Jij«j
1/1J/71
4/U/7I
4/25/71
5/10/71
S/2J/71
«/ll/7J
7/-2/7J
«/l*//l
«/^8/-J
9/lV/Tl
IO/ J/71
10/24/7]
11/21/71
{2/il/ri
O'CNDrOINTS Or TBtSD LlMt
QMS.
" *-iO.OO 0.00 10.00 20. CO 10.00 *0.«0
]
t
ll
15
: ' • i
.»»» »»»***i^lMWi»l»Wto(^Wto>MVM'^MM.WMA|M>»*>--
.*»*«•« I/^IMIM^C^O *«j»\f »ww**O4oa*«9i/*h»K.»-c«a^<<
| i
??
?i
54
55
54
5<.
59
60
61
62
6J
64
-• •>— <
66
67
6<
A9
70
fi
71
7«
7S
r&
7f
7V
8U
— 41 -
»2
til
«4
85
2*
89
90
91
92
91
- «4 -
95
°ll * *
I *
i »
1 *
;\;-
. 1 L • — . ._
I'**
: ' *
' A
A
A
< •
4
A
\
1
; \.
-« I »".,..« • • ».- •
•• \ • _... . ... .
. ,
\
. i
i* i
' 8 ' V_
' \. •.:".«... _ .._
\ ' -,.«
t A
" 1 A
> . •
.......i...L..:....l™.™.^.^^:^::-
«' I
""' 1 i
Reproduced from
best available copy.
127
-------
pf.f
OAT
ANilYSi
DATA SC
Off" •••«••
mi °tfi
si-jo* or C»I»LTST crrtci SAMPLE i
tor. PHC.C N • 107.
RCJ. N •
SUM «T •
107
PACE i
BASIS «.?
»l/ 5/7*
SOI
AC 10 MIST SO,
ACTUAL THCNO
—»ALUt —
IV:
^V/nMI^VIfclA 1
DATE ORS.
""'-10.
4/IV/fiA J
4/76/AK J
S/ l/Ci« 4
S/IC/A1 S
S/?1/6H *
S/U/61 '
7/7//MI a
r/)i>/6« V
»/ 0/>« Id
A/tS/AK 11
»/7e/61 12
9/ «/6A IJ
9/lV/ftA 14
9/7S/A* !S
in/ !/»« 16
|n/ll/6ft If
I"/ 6/?n IV
11/lS/hit 20
1 1 /7S/ frit 21
t7/ 1 d /Alt 22
I/ e/69 21
1/2'/o9 24
7/11/69 2S
1/13/69 28
3/I//69 2V
*/ V/69 - il>
•/1U/AO 31
S/ 6/69 32
*/20/*9 3J
r/ :/»9 is
9/11/A9 16
o/l'/ftl 17
Q/^6/ls9 Jtt
10/10/09 IV
1D/IO/69 4U
10/JU/A9 4^
1I/2S/A* 4J
t/|S/rO 44
l/2*/7Q 4)
7/IV/7II »J
4/7d/7C ill
S/t2//«t 4V
AY 4//o su
1 0/7U/ ' •) SI
11/1<(/7'1 Si
II/I6/7O ^J
17/ 2/70 S4
1 '/ 14/70 S6
4/7J/7I S9
9/ J/M SV
0/16/fl 6U
9/24/7 1 61
M/ 1/71 A2
»/ i/7> 6J
I/12//7 •••
7/21/77 S
t/1«/7? 66
4/11/77 6n
4/7S/77 04
^ 1/77 JO
S/lO/?7 7?
A/10//7 fi
*//^/7J r«
in/ tl~ ' 7e
in/ 1 I/O /f
-lo/!j/;> '*
{•''/Jo/7? JJ
17/76/7? Ai
7/ 6/7) V4
7/2«/?l •»
»/ I/7J f.»
t/ *»/7 J 47
1/1"/M AO
*/ «'/71 Av
4/|7/7» VO
S/ ?/'J ^1
S/l'/7) 92
A/ 7/7J 91
*/'(/') 4<
7/75/7J 96
«/ 1/71 «7
•*/72/ri va
9/IJ/7) 49
9/7«/7) IJU
0/fltlt 101
in/iu/fi iu/
10/11/M HI
11/71/71 I'jb
I?/ «/7) |06
17/?a/71 107
A«.^ w -r, OOOPOINTS Br "*c(]B i. NE
00 -2.00 6.00 14.00 72.00 10.80
f A!
\ * * -------
1 A
I A
1 A
\ A
\ A
\ i
.1 A .
\ A
\ A
1 A
\ j
1 J
A
*A
k
»T * >
\ '
i
A4 i
l\
1
V ... _. - _ _ . - _,.,-.. _. — .
A <
> •
v
• • .
A 1
» A
A \
A \
A I
1 \ ;zi:z'::. L
,\
\ -- - •-
i
\
\
A \ - - - -
A I
A \
• \
A* \
A' A. '
* \
" *j ~" \
*. - 1
A* \
A \
.... .» . ... I .
; \ '„.!_.__•..*
! \ ~ " A
\ A
1 *
1 — A
1 1 A -
I ' *
4 ..»
* I »
••* i *
I 4
;- \- •••
»
A i
A 1
A 1
A • '
» 9
Reproduced front
PCS) zvailable copy.
328
-------
STUDY OF YEAR EFFECT SAMPLE 1
PflOC DEE" OAyr
DATA OEM NIK
ANALYSIS" AKOVAJ
ONE-MAY ANALYSIS OF
USING *T1*
SUMMARY OF EACH OF
CROUP VAvUE.OF
NO,»J *T1*
! »
5 I
BETMEEN
GROUPS
B mm
to
TOTALS
BASIS 4,0
READ N« 107, PROC N • 10T> REJ. N • 0, SUM MT • 107
VARIANCE OF VAB- S02
AS THE SELECT (TREATMENT) VARIABLE,
THE 6 TREATMENT GROUPS.
NtJJ HEANtJ) SDEVtJ) VARCJ]
51 ||?!^66^ la* ^f"* IflS ''1$
20 158.0000 86.0918 7411.7695
:? 222.6750 68.3885 46T6.9637
9U13U U.FfCmVi Rw Hr*| •* *| f
62943, 10 5.00 12588.62 3.07
405869.44 99.00 4099,69
466612.53 104*00
CONFIDENCE LEVEL OF F( 5> 99)« 90.72S.
-------
DEE" DAVE
OAT* Rcr« xis
ANALYSIS' ANOVA1
DATA SET* 1
STUDY or YEAR Errrcr SAMPLE 2
95.
PRCC
95.
KEj.
SUM MT •
95
BASIS 4.0
ll/ 5/M
SUMMARY OF EACH Or THE 6 TREATMENT GROUPS.
GROUP VALUED, NtJJ „„,„,
\ \ \l I?f:!iH
1 — a 5 "" " 275.6000
5 1 1» 131.2272
6 6 20 26J.1SOO
-
5UH5O i). FH
'. BETWEEN
GROUPS 232729.80
MI iHIN
GROUPS 545228.56
J^ lul*L3 ff/vio«JO
SDEVCJ) VARCj;
2l:^2g |o^'!«li
tl.24d$ 6*0*. 5897
70.8069 SR99.JOOO
I3d.93^a 19303.5948
39.7065 1576.7658
EFni" " ' ' MEAN "53 T
5.00 46545.96 7.60
89.00 6126.16
:
CONFIDENCE LEVEL OF F( 5/ 89).100.001.
COMPUTER RUN B-5
-------
04VE
NALVSTS* ANOVAl
DATA SET- I
STUDY OF MONTH CrrtCT SAMPLE 1
READ N»
107*
PROC N •
107,
PA8E
REJ. N • 0*
SUM NT
107
ONC-MAY ANALYSIS OF VARIANCE Or VAR» S02.
USING *T1. AS THE SELECT (TREATMENT) VARIABLE.
SUMMARY Or EACH Or THE 12 TREATMENT GROUPS.
ggoi
- NO.i
u
wJ
M
13 mu5i
|
1
9
I
r
BETWEEN
GROUPS
•ITHIN
GROUPS
TOTALS
r?« N(J1 MEANtJl SOEVtJl VAR[JJ
I 7 216.0000 78.4241 At50.3333
2 A 1*5.1667 49.5.73.0 1646.1667
3 10 205.4000 68.3381 . 4670,1000
5 12 180.0000 38.7764 1503.6364
$ 6 1'5'}«6? i'«$506 1572.1667
? 5 4)9.6000 84.7720 7186.3000
9 12 186.6667 74.5146 5552.4942
10 15 147.9333 7V. 5447 6327.3524
3 *? !J'8888 J?'5i?S 55or'2"f
SUMSO 0« FRtCDOH H£AN SO F
4?3176.97- 93.00 4550*29
469612. S3 104.00
CONFIDENCE LEVFL
T NOT COMPUTED. r$1.0.
COMPUTER RUN B-6
-------
OAT
DAVE
t
* SET*
STuor or MONTH ErrECT SAMPLE 2
BEAD i> 95, PROC N • 95,
PACE 1
REJ. H •
SUM NT •
ONE»4Ar ANALYSIS OP VARIANCE Or VAH« SO?
USING «Tt« AS THE SELECT {TREATMENT) V«HlAaLE.
SUMMARY Or EACH Or THE 12 TRiATME'JT GROUPS.
VALuc.or
10
44-
NtJ)
MEANfJI
s
A
—«-
9
10
11
—f-
9
n
n
76J. 1000
—105.818?
l5ol6364
—i««!eooo
SOEV(J)
VARCJ)
440.
43
97
.447?
.192?
435
«<»
l?48r.95?4
7SJ*»«6J8-
6680.5714
5084.4545
76.4059 5637,8545
••I99»3fl5»—i«740.?00«-
Ul
BETWEEN
MlTHIN
GROUPS
TOTALS
.••AVW»«WVW«~|^«f
SUMSO
1 9 JA2i5 * 1 w
564332.76
777958.36
D. FREEDOM
1 1 .00
83.00
94.00
HCAN so r ;
iroOz.Jr 1.30
7040.15
•
CONFIDENCE LEvrL nr r( It. H3)« 99.091.
COMPUTER RON B-7
-------
with Texas Gulf.
1.5 Data Set (5)
No attempt was made to find a model to fit the smooth
curves shown in the referenced report (see (5) Section
1.1); however, a two-way analysis of variance was per-
formed (B4 to B7) using the six length of shutdown
groups as one category and time into start-up as the
other category. As a B4 to B7 shows there is a signifi-
cant group and a significant time effect with SO2 emis-
sion treated as the response variable. Regressions were
obtained for each of the separate groups (Cl to C6) and
the results are summarized below. The model Y =d
for each group.
Group
1
1-2
2-6
6-10
10-15
15
- 10.3
- 12.6
- 92.4
- 79.8
-229.2
-287.6
106.8
178.07
757.0
627.1
1417.3
1969.0
N
12
30
18
18
24
36
Correlation
-0.430
-0.154
-0.266
-0.372
1
-0.512*
-0.515*
*Significant at five percent level.
The use of these regression equations to predict the
time after start-up at which the plant reaches the com-
pliance limit is questionable. Since this time is the
important variable, it was suggested that means and
333
-------
K
OA
m mi
ATA 01f'
PAtit 3
READ Hm 12, CROC N • 12* REJ. N •
SUM MT • 12
»i> «.0
II/IV/M
0
10.00
TIME
ScAITCR PLOT OiREGRESSlON LINE ENQPCINtS
COMPUTER RUN C-l
-------
READ N. JO, PROC N • 30* REJ. N •
SUM MT •
PAKE 4
600.00 •
402.00 -
'l.00~ "" 2.00' 3.00««00S.OO 6.00 7.00 6.00 9 00
364.00 -
24).00 -
lotoo
TINE
COMPUTER RUN C-2
-------
OEF* ******
UATA^SEfl RC6Rct RCAO N, la, PROC N . ie, REJf N . 0, SUN HT . lt
0,09 1.00 2,00 3. CO 4.00 9.00 «.00 7.00 0.00 9.00 10. I
2000,00 »
1606,00 -
.
1212,00 -
£
y 010,00 -
424,00 -
30,00 -
*
^^7 TO PPt*.
^"•^
SftPftA. <
c
,
^
^"-^.
!• (
• • j^
*<*^1Si^
V
)
-
"^k^-
i>
L
-
1
Q. A i. <
*
J
^
>^S
f
^«*^
# « ^^r~^»f»i
10
0.00 t.Oa 2.00 3.00 4.00 £. (5,00 \ 6,00 7.00 T. 00 9*00 iQ.OO
T IHE
SCATTER PLOT
O.RCQRESSION LI"E END?OINIS
COMPUTER RUM C-3
-------
U
H
DA
r
— RCAO •«• - 18. -' — »»ROC-K-»- ••— 18r
8UH KT • \9- —
B«»IS 4.0
11/1V/H
0.00 t OU
3.00 4.00 ?«2°
f
0.00 1.00 2.00 3.00
6.00
TIME
SCATTER PLOT o.REGRESSION LINE ENOPOINTS
COMPUTER RUN C-4
-------
OB OEF» ••••••
JOB OEF»
sm BEF:
READ 1* it, HROc N • 24*
0.
SUM NT • 24
ems 4,o
11/J9/M
2500,00 »
2006,00 -
1512.00 -
c
N
0 lOiA.Ou •
5*4,00 -
1.00 1.00 2.00 3.00 4.00 9.00 6.00 7.00 8.00 9«00 10.
. £
•
>X.
[ /x* l)
1
300 fP*
I
1
3t).DO •
i
1 1
l •
rjs>2S0P/
'<
^
^>s.
X
1
1
«
>
X.
t
f ^
) \
. 1
-
(
K.
"x.
^s^
\ "^
f
<
•V
Kw
^X
'
•, ^V^ f
x I
i
- -
_
i or 9 IT V>J.. * *
0
- •'
}.00 1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00 10.00
SCATTER PLOT
TIME
fl.Rrr.RESSlOh LINE ENOPQ1NTS
COMPUTER RUN C-5
-------
JOB OEF» ••••«•
8tf: ?
"""
READ N«
36*
HROC N •
36*
RCJ. N •
SUM NT •
J6
Ibl* «.0
U/19/ffl
0.00 100 6.00 9*00 10.iO-
0.00
l.OU
2.00
3,00
4.00
5.00
7.00
8.00
9.0U 10.00
SC*'TCR PLOT
TIME
0«RCGf)CSS10N
COMPUTED RON C-6
-------
variance of this quantity be computed
1.6 Problems and Recommendations
Major problems associated with this study were:
(a) The complicated nature of the process and relation-
ships studied made full investigation of the acid
plant system too -time consuming for the type of
task approved.
(b) The number of plants from which data was gathered
was insufficient to discover plant to plant vari-
ability.
(c) All of the data looked at represented data gathered
after the fact so that no experimentation in manipu-
lating the process parameters was accomplished.
Any additional studies which may be performed on the SO,
emissions should consider the need for expanded data
collection mentioned above with the exception of (c) as
it may be impossible to find plants willing to manipulate
process parameters to force high emissions. At this
writing it is felt that any future analysis of the S02
emissions as a function of process parameters should
begin with data at approximately five minute intervals.
The availability of this data would provide the ability
to study a time-lag effe-ct in explaining SO- emission.
The high interdependence between the process parameters,
and the observations of several continuous time charts
on these parameters suggest a time-lag model might be
appropriate.
340
------- |