EPA-650/2-74-082
SEPTEMBER 1974
Environmental Protection Technology Series
I
55
V
532
UJ
(3
lillliMlillliiiliiKIl
-------
EPA-650/2-74-082
REFINERY CATALYTIC
CRACKER REGENERATOR
SOX CONTROL
PROCESS SURVEY
. by
T. Ctvrtnicek, T. Hughes,
C. Moscowitz, and D. Zanders
Monsanto Research Corporation
Dayton Laboratory
Dayton, Ohio 45407
Contract No. 68-02-1320
Task 1, Phase I
ROAP No. 21ADC-031
Program Element No. 1AB013
EPA Project Officer: Kenneth Baker
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
September 1974
-------
This report has been reviewed by the Environmental Protection Agency
and approved for publication. Approval does not signify that the
contents necessarily reflect the views and policies of the Agency,
nor does mention of trade names or commercial products constitute
endorsement or recommendation for use.
11
-------
ABSTRACT
The report gives results of a survey of conceptual techniques
applicable to fluid catalytic cracker (FCC) regenerator off-gas
sulfur oxide emission reduction, with respect to their applica-
tion both to the FCC system itself and to the regenerator off-
gas. These two control techniques have also been compared with
FCC feedstock desulfurization. The economics for all systems
evaluated are compared. A comprehensive analysis of FCC
operations has produced evidence that sulfur emission control
can most effectively be achieved through steam contacting of
the spent cracking catalyst. This concept is therefore proposed
as the primary subject for further investigation.
iii
-------
TABLE OF CONTENTS
Page
1. INTRODUCTION 1
2. SUMMARY 3
3. THE FLUID CATALYTIC CRACKING PROCESS 5
3.1 PROCESS DESCRIPTION 5
3.1.1 Principal Operations 5
3.1.2 Cracking Catalyst 9
3.1.3 Catalyst Stripping and Regeneration 12
3.2 TYPES OP FLUID CATALYTIC CRACKING UNITS 16
3.3 SIZE AND CAPACITY OF FLUID CATALYTIC CRACKING UNITS 20
3.4 EMISSIONS FROM FCC UNITS 22
3.^.1 General 22
3.4.2 Basis for Comparison 24
3.4.3 Predicting S02 Emission 25
3.5 SUMMARY OF ASSUMPTIONS 28
3.5.1 Technical Assumptions O 29
3.5.2 Economic Assumptions 29
4. FCC FEED DESULFURIZATION 32
4.1 TECHNICAL EVALUATION 32
4.2 ECONOMIC EVALUATION 34
5. PROCESS MODIFICATION 39
5.1 SUMMARY OF CONCLUSIONS 39
5.2 PROCESS MODIFICATION ANALYSIS 43
v
-------
TABLE OF CONTENTS (Continued)
5-2.1 Evidence of Effects of Steam Stripping on
Sulfur Distribution in Fluid Catalytic Cracking 46
5.2.2 Theoretical Aspects of Steam Stripping 5*4
5.3 APPLICATION OF STEAM STRIPPING TO EXISTING FCC UNITS 59
5.3.2 Secondary (Add-On) Separate Steam Stripper 61
5.3.3 Application of Secondary Steam Stripping
(Option 1) 64
5-3.4 Control of Other Pollutants with Steam Stripping 66
6. REGENERATOR FLUE GAS DESULFURIZATION 70
6.1 SUMMARY OF CONCLUSIONS 71
6.2 CANDIDATE PROCESS SELECTION 73
6.3 GENERAL CONSIDERATIONS IN EVALUATION OF THE SELECTED
PROCESSES 83
6-3.1 Technical Evaluations 83
6.3-2 Economic Evaluation 87
6.4 ADSORBENT/ABSORBENT SYSTEMS 88
6.4.1 Westvaco Process 89
6.4.1.1 S02 Production Process Description 90
6.4.1.2 Sulfur Production Process Description 95
6.4.1.3 Experimentation Needed and Proposed by
Westvaco 100
6.4.1.4 Economics 101
6.4.1.5 Comments 105
vi
-------
TABLE OF CONTENTS (Continued)
Page
6.4.2 Shell Flue Gas Desulfurization Process 106
6.4.2.1 Process Description 106
6.4.2.2 Economics 111
6.4.2.3 Comments 111
6.4.3 Union Carbide PuraSiv-S Process 115
6.4.3.1 Process Description 115
6.4.3.2 Economics 118
6.4.3.3 Comments 118
6.5 SCRUBBING SYSTEMS 121
6.5-1 Wellman-Lord Sodium Sulfite Absorption 122
6.5-1.1 Process Description 122
6.5-1-2 Economics 130
6.5-1-3 Comments 130
6.5.2 Magnesium Oxide Scrubbing 134
6.5.2.1 Process Description 134
6.5.2.2 Economics 139
6.5.2.3 Comments 139
6.6 OXIDATION SYSTEM 142
6.6.1 CAT-OX Catalytic Oxidation 142
6.6.1.1 Process Description 142
6.6.1.2 Economics 148
6.6.1.3 Comments 148
vii
-------
REFERENCES
TABLE OF CONTENTS (Continued)
Page
152
APPENDIX A
APPENDIX B
APPENDIX C
APPENDIX D
APPENDIX E
APPENDIX P
APPENDIX G
APPENDIX H
CRACKING CATALYSTS AND THEIR PRODUCERS 162
PREDICTION OF REGENERATOR OFF-GAS S02 166
CONCENTRATION
FEEDSTOCK DESULFURIZATION TECHNIQUES 170
ECONOMIC EVALUATION OF FCC FEEDSTOCK l8l
DESULFURIZATION - DETAILED ESTIMATES
ECONOMIC EVALUATIONS OF CATALYST STREAM 190
CONTACTING (STRIPPING) - DETAILED
ESTIMATES
ECONOMIC EVALUATION OF FLUE GAS 211
DESULFURIZATION SYSTEMS - DETAILED ESTIMATES
EVALUATION FORM FOR ADD-ON FLUE GAS 232
DESULFURIZATION SYSTEMS - DETAILED ESTIMATES
SINGLE UNIT CONVERSION FACTORS, BRITISH TO 236
SI (METRIC)
viii
-------
LIST OF FIGURES
Figure Page
1 Fluid Catalytic Cracking Unit 6
2 Processing Plan for Complete Modern Refinery 10
3 Catalyst Stripper 17
4 FCC Designs 19
5 S02 Concentration in FCC Regenerator Off-Gas 27
6 Hydrodesulfurization of FCC Feedstock, Capital
Investment Cost (February 1973) 35
7 Hydrodesulfurization of FCC Feedstock, Operating
Cost 36
8 Hydrodesulfurization of FCC Feedstock, Incremental
Operating Cost 37
•
9 Distribution of Sulfur in FCC Products 49
10 Effects of Steam Stripping Rate on Sulfur Content
of Coke (Natural Catalyst) 51
11 Improvement of Catalyst Activity and Selectivity
with High Stripping-Steam Rate 53
12 Effect of Steam Stripping of Spent FCC Catalyst on
S02 Concentration in Regenerator Off-Gas 55
13 Block Diagram for a Conceptual Steam Stripping
Facility 63
14 FCC Catalyst Steam Stripping, Capital Investment
Cost (February 1973) 67
15 FCC Catalyst Steam Stripping, Operating Cost 68
16 Westvaco Process - FCC Regenerator Waste Gas
Treatment, S02 Production 91
17 Westvaco Process - FCC Regenerator Waste Gas
Treatment, Conceptual Design Flowsheet 93
ix
-------
LIST OF FIGURES (Continued)
Figure Page
18 Westvaco Process - FCC Regenerator Waste Gas
Treatment, Elevation and Plot Plan 96
19 Westvaco Process - FCC Regenerator Waste Gas
Treatment, Sulfur Production 97
20 Westvaco Process - FCC Regenerator Waste Gas
Treatment, Pilot Plant Program Schedule 102
21 Westvaco Capital Investment Cost (February 1973) 103
22 Westvaco Operating Cost 104
23 Shell Flue Gas Desulfurization Unit 107
24 SFGD Capital Investment Cost (February 1973) 112
25 SFGD Operating Cost 113
26 PuraSiv-S Process Flow Diagram 116
27 PuraSiv-S Capital Investment Cost (February 1973) 119
28 PuraSiv-S Operating Cost 120
29 Wellman Lord, Inc., Sulfur Dioxide Recovery, Sodium
System 124
30 Wellman Lord, Inc., Capital Investment Cost
(February 1973) 131
31 Wellman-Lord, Inc., Operating Cost 132
32 Magnesia Slurry S02 Recovery Process 135
33 Magnesium Oxide Scrubbing Capital Investment Cost
(February 1973)
34 Magnesium Oxide Scrubbing Operating Cost 141
35 CAT-OX Flow Diagram - Flue Gas Reheat System 143
36 CAT-OX Flow Diagram - High Temperature System 144
-------
LIST OF FIGURES (Continued)
Figure Page
37 CAT-OX Capital Investment Cost (February 1973) 1^9
38 CAT-OX Operating Cost 150
xi
-------
LIST OF TABLES
Table Page
1 Capital and Operating Cost Summary for Systems
to Control SO Emissions from Refinery Catalytic
Cracker Regenerator 4
2 Fluid Catalytic Cracking Operating Ranges 8
3 Variations of Particle Distribution in Typical
Fluid Cracking Catalysts 13
4 Design Conditions of Catalyst Strippers 18
5 Use of Various FCC Design Types 18
6 Summary of Refining and Catalytic Cracking Units
in the United States as of January 1, 1973 21
7 Emission Ranges from Fluid Catalytic Cracking Unit
Regenerator, Before and After CO Boiler 23
8 Comparison of Catalytic Cracking Units Operating
Conditions 44
9 Flue Gas Desulfurization Processes 74
10 Flue Gas Desulfurization Processes After Classi-
fication According to Their Availability 82
11 Wellman-Lord S02 Recovery Process Development 123
12 Chemistry of Magnesia Slurry S02 Recovery Process 136
xii
-------
1. INTRODUCTION
To assist in their evaluation of air pollution technology appli-
cable to catalytic cracker regenerator flue gas, the Environmental
Protection Agency has contracted Monsanto Research Corporation
to identify conceptual techniques for reducing petroleum refinery
fluid catalytic cracker regenerator SO emissions to less than
.A.
200 ppm S02 and to perform a feasibility analysis of the techniques
identified. This report contains the outcome of the first phase
of this study.
This study was not aimed at development of sulfur oxides removal
technology for the whole petroleum Industry, or even for a
specific refinery. Instead, our objective was to aid the Environ-
mental Protection Agency in identifying the technology having
the highest potential for application to this specific segment
of the petroleum refining process and thus to allow the EPA to
concentrate its research efforts on this technology, including
eventual full-scale demonstration. Such a demonstration would
then help to establish and define the emission standards for
petroleum refineries. It should be noted that the process
envisioned for demonstration is not intended to represent the
only technology that will have to be applied by petroleum re-
finers. Rather, its primary purpose will be to demonstrate that
the air pollution regulations to be established can be met. The
refiners will be free to select and apply any technology pro-
ducing comparable results.
In order to conduct the study most efficiently and to compare the
selected SOX control systems on an equal basis, FCC operating
conditions had to be carefully defined. The FCC process, the types
of units that have been developed, and the size of these units are
-------
discussed In Section 3. The technical and economic assumptions
applied throughout this study are also presented in that section.
Three approach alternatives to desulfurization of fluid catalytic
cracker (FCC) regenerator flue gas were investigated. The first
alternative process considered for SOX control was desulfurization
of the feed to the cracking operation to a degree that would re-
sult in sulfur emissions of 200 ppm. The second alternative, fluid
catalytic process modification, required investigation and defini-
tion of the variables controlling this operation, especially those
affecting sulfur emission from the FCC regenerator. The third
approach alternative involved the selection and specification of
the most feasible regenerator flue gas desulfurization system.
The systems considered in this alternative were applied as add-on
systems to existing fluid catalytic cracking units. The three
approach alternatives are discussed in Sections 4, 5, and 6, re-
spectively .
A rank ordering of the processing techniques considered has been
established, and the most applicable technique for refinery
catalytic cracker regenerator SOX control considering the last
two approach alternatives has been selected. Technical eval-
uation of the first approach alternative, feedstock desulfurization,
was beyond the scope of this project, but it was evaluated to
provide a basis for economic comparison with the other two alter-
natives. An experimental work plan for the second phase of the
study to fully investigate the most applicable technique and to
supply sufficient data for a pilot plant study has been submitted
to EPA in a separate document.
-------
2. SUMMARY
Conceptual techniques applicable to fluid catalytic cracker re-
generator off-gas sulfur oxide emission reduction have been
evaluated with respect to their application to (1) the fluid
catalytic cracking (FCC) system itself and (2) the regenerator
off-gas. These two techniques of control have also been compared
with fluid catalytic cracker feedstock desulfurization. The
economics for all systems evaluated are compared in Table 1.
A comprehensive analysis of FCC operations has produced evidence
that sulfur emission control can be most effectively achieved
through steam contacting of the spent cracking catalyst. This
concept is therefore proposed as the primary subject for further
investigation in Phase II of this program.
-------
Table 1. CAPITAL AND OPERATING COST SUMMARY FOR SYSTEMS TO CONTROL SO EMISSIONS
PROM REFINERY CATALYTIC CRACKER REGENERATOR
FCC UNIT CAPACITY. BPSD
10.000
50.000
Capital In-
Final Position In vestment Cost
Rank Ordering Desulfurlzatlon Technique $ x 10~6
Cracking Catalyst Typical*
1 Steam Contacting Case
Worst +
Case
2 Westvaco S02 Production
S Production
2 Shell
3 Wellman-Lord
NR FCC Feed Case A+t
Desulfurization
Case Bt+
• Union Carbide
* Magnesium Oxide
« CAT- OX
0.5
0.7
1.0
1.2
1.6
1.7-2.2
5-0
5-0
2.1
1.7-1-9
2.1-2.8
Operating
Cost
t/bbl
13.8
19.8
11.9
13.2
11.7
22.9
95
71
23.1
23.8
25-9
Capital In- Operating
vestment Cost Cost
$ x 10"6 4/bbl
1.2"
*«
2.0
2.8
3.1
1.6
1.9-6.2
11.5
11.5
7.5
1.8-5.1
6.0-8.0
7.6"
**
13-0
6.1
7 1
8.8
9-3
53
39
11.6
10.9
12.1
uapitai In-
vestment Cost
$ x 10-6
2.8
1.5
5-8
7-1
9-5
10-12.5
30.1
30.1
16.3
9.8-11.0
12.5-16.8
uperating
Cost
4/bbl
6.0
11.0
it. 6
6.0
6.8
6.1
17
33
11.8
7.6
8.1
* Processes with Lower Rating
" For 15,000 BPSD Capacity
NR Not Rated
t Typical and Worst Cases are Defined on Page 66
tt Case A wt X S in/out 3-36/0.213
Case B 3.36/2.13
Cases A and B are Defined on Pages 28, 32,& 33
-------
3. THE FLUID CATALYTIC CRACKING PROCESS
3.1 PROCESS DESCRIPTION
3.1.1 Principal Operations
A fluid catalytic cracking (FCC) plant Is composed of three
sections: cracking, regeneration, and fractlonation. The
cracking reactions take place continuously in the reactor, with
the spent catalyst being continuously regenerated and returned to
the reactor. Both the reactor and the regenerator operate on the
fluidization principle, which makes possible the continuity of
flow of catalyst as well as of hydrocarbon feed. The negative
features of fixed-bed designs involving the intermittent shifting
of reactors through cracking, purging, and regeneration cycles are
thus eliminated.
In a typical plant, such as that presented in Figure 1, re-
generated catalyst is withdrawn from the regenerator and flows
by gravity down a standpipe. There a sufficiently high pressure
head is built up to allow catalyst injection into the liquid oil
stream. The resulting mixture of oil and catalyst flows through
a riser (transfer line) where essentially all of the cracking
reactions take place. The vapor stream flows into the reaction
vessel (used primarily for catalyst disengagement). In this
vessel gas velocity is intentionally low so that a high con-
centration of catalyst will result. The cracked product oil
vapors are withdrawn from the top of the reactor after passing
through cyclone separators to free them of any entrained catalyst
particles.
-------
Reactor
Separation
Vessel
Flue gas to particulates removal
and waste heat recovery
Catalyst Stripper
Main
Fractionation
Column
Regenerator
Combustion Air
Slurry
Settler
Raw Oil Charge
Riser Cracker
Gas and Gasoline
to Gas Concentration
-> Light Cycle Gas Oil
> Clarified Slurry
Figure 1. Fluid Catalytic Cracking Unit
-------
The cracking of hydrocarbons that takes place during their con-
tact with the catalyst results in coke deposition on the catalyst.
The catalyst must therefore be periodically regenerated. The
spent catalyst is withdrawn from the reactor and passed through
a steam stripper where residual product vapors are desorbed from
the catalyst. After stripping, the catalyst passes to the re-
generator where the coke deposits are oxidized by air and burned
off. The products of combustion leave the top of the regenerator
and pass through a series of cyclones. Here, the bulk of the
entrained catalyst is recovered. The regenerated catalyst is
withdrawn from the bottom of the vessel to complete the cycle.
The regenerator off-gas is normally sent to an electrostatic pre-
cipitator for removal of catalyst fines. The gases (1000-12000°F),*
containing a high concentration of carbon monoxide, can be sent to
a CO boiler where CO is further oxidized. The CO boiler is located
either upstream or downstream of the precipitator. Gas leaving the
boiler or precipitator (375-800°P) is then discharged to the atmos-
phere. This gas is the source of air pollution. The gas composi-
tion is presented in Section 3.4.
The product vapors leaving the FCC reactor are sent to a fraction-
ation tower where the first separation of products (gas, gasoline,
and cycle gas oil) takes place. These streams are further separated
in accordance with the refinery needs and products. FCC operating
conditions are presented in Table 2.
FCC technology can be applied to the formation of various products
from distillate oils although its major goal for many years was
to convert fuel oil to gasoline. The process is extremely flexible
and can be readily carried out on a wide variety of feedstocks
and over a wide range of temperatures, conversion levels, and
catalysts. These features make the fluid process adaptable to
widely different refinery product yield and quality requirements.
*See Appendix H for British to SI (metric) conversion factors.
-------
Table 2. FLUID CATALYTIC CRACKING OPERATING RANGES*?,25,35,39
Reactor
Regenerator
Temperature, °F
Pressure, psig
Catalyst/Oil Ratio (weight)
Gasoline Yield*, % No Recycle
Recycle
Coke Formation*, % No Recycle
Recycle
Dry Gas Formation*, % No Recycle
Recycle
Conversion*, % No Recycle
Recycle
Volume Hourly Space Velocity
Gas Velocity, ft/sec
Standpipe
Riser
Fluid Bed
Vapor Line
Coke Content of Spent Catalyst, %
Temperature, °F
Pressure, psig
Catalyst Hold-Up Time, min
Gas Velocity, ft/sec
Coke Content of Regenerated
Catalyst, % wt
885-975
9-20
4-20
40-45
Up to 61
4-9
12-14
7-12
13-15
40-60
60-90
1-3
2-7
15-40
1-2
90-115
wt 0.25-2.3
1000-1200
1-13
10-20
1-2
0.05-1.0
* For more complete information, see also Reference 17, p. 768,
and Reference 46, p. 124
-------
The fluid process is capable of processing almost any petroleum
fraction ranging from a naphtha to a reduced crude. Market re-
quirements, however, have made it generally advantageous to
process the medium or high boiling gas oil fractions. The
products that are formed in catcracking include high octane
gasoline, raw materials for alkylate production, petrochemical
raw materials, heating oils, diesel oils, liquefied petroleum
gases, and aromatics such as benzene, toluene, and xylenes (BTX
aromatics).
The location of the catalytic cracking process in a modern re-
finery is shown in Figure 2.
3.1.2 Cracking Catalyst
With minor exception, all commercial cracking catalysts are based
on silica-alumina combinations of one type or another. In general,
they can be divided into three classes: (1) the acid-treated
natural aluminosilicates; (2) the amorphous synthetic silica-
alumina combinations; (3) the crystalline, synthetic silica-
alumina combinations. All of them are high temperature acids
and their catalytic activity is attributed to this activity. It
should be noted that 905? or more of the cracking catalysts
41
currently used fall into class 3.
The earliest cracking catalysts of the silica-alumina type
were the acid-treated montmorillonite clays (natural clays).
These clays are hydrous aluminosilicates containing some base-
exchangeable (zeolite type) ions. During the acid treatment,
these zeolite ions as well as about one-half of the aluminum in
the aluminosilicate structure are removed. These catalysts were
widely used but had two weaknesses. The first of these was that
a certain amount of iron in the crystal lattice became active
9
-------
Dry Gas
Crude Oil
Straiaht-Run Gasoline
Gos
Residuum
>.Motor Gasoline
>*Aviotion Gasoline
•^-Keroslne
>-Light Fuel Oil
and Diesel Oils
^Sulfur
»-Llght Fuel Oil
Fuel Oil
»-Lube Stocks
^•Greases
.^Asphalt
*.Cok«
Figure 2. Processing Plan for Complete Modern Refinery
-------
when high sulfur feedstocks were used. The Iron catalyzes the
formation of coke and hydrogen without contributing to gasoline
formation. During catalyst regeneration, it catalyzes combustion
of coke to C02 rather than to CO, and so liberates unnecessary
quantities of heat.
The second weakness was that they were sensitive to high re-
generation temperatures and would undergo thermal catalyst de-
activation quite easily.
The amorphous silica-alumina (synthetic) combinations were based
on combining silica and alumina gels so that the catalysts con-
tained 10-15% A1203 or 20-30% A1203, which are defined as "Low"
or "High" alumina catalysts, respectively. These catalysts were
iron free, so they could be used with high sulfur oils. They
were also more thermally stable so they could withstand high
regeneration temperatures.
The crystalline, synthetic silica-alumina combinations are
manufactured from the natural mineral faujasite and the synthetic
Linde X and Y molecular sieves. When the base exchangeable ions
are replaced in part by rare earth ions and in part by ammonium
ions, an extremely active cracking catalyst is formed.
There are a large number of cracking catalysts. A list of these
and their manufacturers has been made by Thomas1*1 and is pre-
sented in Appendix A.
The amount of catalytic cracking catalyst purchased in the U.S.
on a daily basis is about 550 tons/day (based upon 5-5 x 106 bpsd
total catcracking capacity and a catalyst attrition of 0.2
Ib/bbl feed). Of this amount, Davison Chemical Division of
W. R. Grace & Co. supplies the largest quantities (approx. 35%).
11
-------
American Cyanamid Co. supplies about 305? of total usage. Nalco
Chemical Co. and Filtrol Corp. each have an estimated 10$ of
the catcracking catalyst market. **3»"* **
The catalysts used in the PCC process are sold commercially in
powdered form. Replacement of catalyst in PCC units is required
because of catalyst deactivation by metals (nickel, vanadium, iron)
contained in FCC feedstocks and "dusting losses" caused by frag-
mentation of catalyst particles during handling.
In general a good catalyst should not lose its activity during
operation, and the loss by attrition should be small. Natural
catalysts are soft and are therefore destroyed more rapidly than
most synthetic catalysts. The zeolite catalysts which came onto
the market during the last decade tend to be harder than the
synthetic catalysts. In some instances, catalyst hardeners have
been used to decrease catalyst attrition.
Obviously, the catalyst for the fluid process must exhibit a size
distribution that fluidizes properly. For most catalysts, the
percentage of up to 80 micron material should be kept above 50$
(preferably 75%) and up to 40 micron fraction above 15?. Data
on size distribution of catalysts are available in the litera-
ture. l7 j1* 5j1*7 An example is shown in Table 3-
3-1-3 Catalyst Stripping and Regeneration
Catalyst stripping is currently used in fluid catalytic cracking
both for product recovery and equipment safety. As was already
indicated, in the FCC process, large quantities of spent catalyst
are continuously circulated back and forth between a conversion
zone where the hydrocarbon feed is cracked and a regeneration
zone where carbonaceous deposits are burned off the catalyst
particles and catalyst is recovered. During the hydrocarbon
12
-------
Table 3- VARIATIONS OP PARTICLE DISTRIBUTION IN TYPICAL
FLUID CRACKING CATALYSTS17
Particle
Size,
Microns
0-10
0-20
0-40
0-80
0-125
0-180
0-500
A
75
95
100
-
-
-
—
B
60
85
95
100
-
-
_
C
0
0
50
100
-
-
—
D
5
15
25
75
-
100
—
E
12
22
41
71
84
93
100
P
1
5
30
75
93
99
100
G
1
3
17
56
-
95
100
H
10
25
50
-
-
100
I
1
12
48
71
86
100
J
13
24
40
-
-
100
K
4
15
33
55-8
-
-
_
L
6
19
44
71
-
-
—
A and B commercially available, but expensive
C theoretically desirable, but expensive
D commercially available at moderate expense
E average of commercially available at average cost
F used catalyst, similar to D sample, when new
G, H, and J samples cited in the literature
I used catalyst, similar to E, when new (now too coarse)
K used synthetic
L used natural
-------
cracking in the reactor, a portion of the feed remains on the
catalyst in the form of coke. The composition of the coke has
been reported as (CaH^) .25>1*2 Additionally, some sulfur
originally present in the feed is deposited on the catalyst. Its
amount varies with the type of feed, rate of recycle, steam
stripping rate, the type of catalyst, cracking temperature, etc.
As indicated by its composition, the coke contains about 10? (by
weight) hydrogen. Additionally, the catalyst and the coke are
exposed to relatively high concentrations of hydrocarbons in the
FCC reactor, which can result in some adsorption of the hydro-
carbons on both the catalyst and the coke. Thus, hydrogen content
of coke has been reported to range from 45? to 16% by weight,26
with the most likely value in range of 7-155&. 17
The amount of hydrogen and carbon that remains on the spent
catalyst after it leaves the reactor is very important for safe
operation of fluid catcrackers. Essentially all coke forming
compounds are removed in the regenerator by air oxidation. The
oxidation reactions are highly exothermic and result in a temper-
ature increase in the regenerator. High temperature in the re-
generator can be detrimental to the catalyst and also can cause
afterburning downstream from the regenerator and result in severe
damage of cyclones and auxiliary equipment. The oxidation of coke
is carried out under lean conditions (insufficient amount of air
for complete combustion) so that not all the carbon is completely
oxidized to carbon dioxide. Usually, the C02/C0 ratio in the re-
generator is maintained between 1.0 and 2.O.17
The heats of combustion (on a weight basis) for the three
oxidation reactions C—»CO, C—»-C02, and H2—»-H20 are 9743-2 Btu/lb,
14,086.8 Btu/lb, and 51,571.4 Btu/lb, respectively. These values
clearly indicate that oxidation of hydrogen produces 5-3 times
more heat than oxidation of carbon to carbon monoxide, and 3.7
times more heat than oxidation to carbon dioxide. The amount of
-------
hydrogen leaving the reactor and entering the regenerator is
therefore very important to proper operation of the FCC units.
Furthermore, since the products that would be burned in the re-
generator are quite valuable, it is economically desirable to
recover them.
When the FCC process was being developed, it was learned that the
use of steam is both an efficient and an economical method of
stripping the hydrocarbons off the spent catalyst for their later
recovery. Various devices and methods have been employed to
remove hydrocarbons from spent fluidized catalyst.
The apparatus which is used to remove entrained hydrocarbon
products from the spent catalyst is commonly referred to as the
catalyst stripper. In present designs, steam is used as the
stripping gas because it can be condensed and easily separated
from the reaction products. Other gases, such as nitrogen, fuel
gas, or oxygen-free flue gas, have been used as stripping agents.25
Their use, however, was not desirable because these gases (1) are
not condensable in existing refinery facilities and would cause
dilution of process streams, and (2) blanketed heat transfer
surfaces and decreased the heat transfer efficiency in condensers.
The catalyst strippers that are in use on existing FCC units are
typically operated in the following manner. Spent catalyst
leaving the reaction zone is passed through a catalyst stripper
where stripping medium passes the catalyst counter-currently.
The stripping gas and stripped out hydrocarbons are discharged
from the stripper either directly into or slightly downstream
of FCC reaction zone.
15
-------
The contacting devices normally used in catalyst stripping are
basically counter-current mass transfer units. Figure 3 depicts
the physical configuration of such units. The baffles, which
appear as having either a herringbone or a chevron configuration,
are used as contacting stages in the stripper. Catalyst, which
enters the top of the stripper, flows in a zigzag fashion down-
ward by gravity over the baffles. The stripping gas (steam) flows
vertically upward through each baffle and around the down-flowing
catalyst particles. The baffles act as stripping gas distribution
plates because each baffle contains many holes over its surface.
Both short circuiting of stripping gas around the baffles and
leakage of catalyst through the holes are avoided by proper design
of the baffles. Details of each stripper design can be found in
any of the patents covering such designs. **8-57 , i 26 Typical
design conditions for catalyst strippers are summarized in Table 4,
3.2 TYPES OF FLUID CATALYTIC CRACKING UNITS
There are several types of FCC units that have been developed by
various companies and that differ slightly in the design of the
major and auxiliary equipment.15 The operating principles,
however, are the same for all of these design types. Table 5
demonstrates the free world market distribution for the six
designs presently used. The number of units based on the Esso
Research & Engineering Co. design was estimated. The company
has designed units presently handling 2 x 106 bpsd or about 25%
of the free world capacity.15 Using an average FCC unit capacity
for the United States (Table 6) and applying this average to
calculate the number of units that can handle the capacity re-
ported by Esso, we have arrived at a total of 59 units. The
same figure represents 25% of 237, the total number of FCC units.
Schematic diagrams of FCC designs appear in Figure 4.66
16
-------
Catalyst from Reactor
Combustion Air
and Catalyst
to Regenerator
Stripper Vapors
to Product
Vapor Line
Stripper Vapors
to Product
Vapor Line
Catalyst from Reactor
Steam Combustion Steam
Injection Air Injection
Perforated Baffles
Internal Riser
Figure 3- Catalyst Stripper
17
-------
Table 4. DESIGN CONDITIONS OF CATALYST STRIPPERS'49 > 5 5, 58
Temperature, °F 900-1000
Pressure, pslg 5-20
Steam Stripping Rate, fi
lb/1000 Ib of catalyst i.a-9
Catalyst Flow, Ib/sq ft/min 200-5000
Superficial Gas Velocity, ft/sec 0.05-25
Stripper Bed Density, Ib/cu ft 15-35
Table 5. USE OF VARIOUS FCC DESIGN TYPES
Design Type (Licensor)
Universal Oil Products Co.
*
Esso Research & Engineering Co.
M.W. Kellogg Co.
Standard Oil Co. (Indiana)
Texaco Development Corp.
Gulf Research & Development Corp.
No. of
Units
125
(59)
31
11
6
5
% of Total
52.5
(25)
13-1
4.6
2.6
2.2
237 100.0
* Estimated
18
-------
Texaco
Regenerator
Cos-oil rtorjt
Gulf
Combustion olr
Up|W ft.d Infliction
Lowr hid injection
Typical UOP
"stacked" FCC
Reactor
flw 901
Early UOP designs
Typical UOP
reactor revamp
(mid-1960'si
Regeneroted-cotolyst
stondpipe
OH feed
Kellogg Orlhoflow riser reactor
pent-catalyst riser
Oil-feed injection
Strip,*,
Riser reactor
Esso Research Pleiicracker
(iser-oocktr configuration "•„,,," tmnsfw-Jin. conjuration
line
WKIor
SUam
RMKtor
factor Kfd iMd -»
Standard of Indiana
Steora
Catalyst
distngoger
Stripper
Rejenerotor
~*£j
Gis-c
A
1
>
x^
.1 (he
Steam
H5
rje •
Riser
reactor
Rwytl
Frocti
!
onator w,l got
HL
RaogoMlim
—
Slorry
1 , Hmvy-cychl oil
1 Decanted oil
Fli» gos
In (0 boiler
UOI' "strnighl riser" PCC
Pressure-
reducing I I Catalyst
chamber y stripper
to aoi-ioixnitniti«n plant
rry settler
New Kellogg design
Twe-stoge regenerator
Riser reactor
Sleom
Figure
FCC Designs
19
-------
3-3 SIZE AND CAPACITY OF FLUID CATALYTIC CRACKING UNITS
Another very important factor in the process alternatives eval-
uation was the size of the FCC unit for which the evaluation would
be made. As of January 1, 1973» there were 246 refineries in
the U.S. on stream.9 Of these, 119 contained FCC units. Table
6 lists United States refineries according to their production
capacity, in the intervals of 5,000 barrels per stream day (bpsd),
excluding refineries larger than 100,000 bpsd. The latter were
broken into two groups, (1) from 100,000 to 150,000 bpsd and (2)
larger than 150,000 bpsd. Presently, the largest refinery in the
U.S. is Exxon's 434,000 bpsd refinery in Baton Rouge, Louisiana.
Table 6 also includes the number and percentage of units that
contain fluid or other catalytic cracking processes and the cumu-
lative percent values for both refining and fluid cracking
capacities. As shown in Table 6, practically any refinery larger
than 5*000 bpsd can operate an FCC unit. The refineries larger
than 30,000 bpsd will almost surely contain some catalytic
cracking operation, with an &5% chance that it will be of the
fluid type (from the total of 140 cat-cracking units, 119 are
fluid). The fresh-feed FCC unit capacity on an average is about
33% of the refinery total capacity in barrels per stream day (bpsd).
The total number of U.S. refineries in 1940 was 461, with a total
capacity of about 4.2 million bpsd.10 As has already been
mentioned, in 1973 the total number of refineries in the U.S.
has been reduced to 246 but these have a total producing capacity
of almost 14 million bpsd. Recently, a million barrels per day
of new capacity within the United States has been announced, with
the smallest unit having a capacity of 150,000 barrels per day.11*12
These facts certainly indicate that the general trend in the
petroleum industry is towards larger and larger capacity, and
these larger refineries obviously contain FCC units. It can
20
-------
Table 6 SUMMARY OF REFINING AND CATALYTIC CRACKING U.IITS II THE UNITED STATES AS OP JANUAB1 1, 1973'
(V)
Number or Rc-
( of Total fineries In FCC Capacity
Refinery Size Nuiber of Re- Refining Capacity Capacity Range with Fresh Total J Units Having
BPSD fineries In Range BFSD (Cumulative) FCC Facilities BPSD BPSD FCC Facilities
1 - 5, 010
5.000-10.000
10,000-16,000
15,000-30, 000
20.000-2S.OOO
25.000-30.000
30,000-35.000
3J.OOO-UO.OOO
no, 000-15. ooo
15. 000-50. 000
50,000-55,000
55,000-60.000
60,000-65,000
65,000-70,000
70,000-75,000
75,000-80,000
30,000-83,000
35.000-90,000
= 1,0(10-95.000
95.000-! )•«, 300
100.000-150,000
.jJ.OOO And Larger
"otol
• Houdrlfl
Note In
15
3"
15
IB
8
18
3
9
2
9
8
*
6
li
2
2
3
6
1
3
17
26
2*6
cm Unit
the source
123.280
215,900
188,100
32', 100
188,800
512.800
101,600
335,900
37.500
in. 500
117,200
233.600
371,000
271,700
117.000
160,000
251,000
536,700
95,000
787.600
2.069.700
6,016,100
13.997.lliO
from which this Info™
2
1
6
7
11
12
It
15
13
21
22
25
27
28
29
31
35
36
11
56
inn
92
67
02
33
68
35
07
17
10
25
23
90
57
5"
59
73
52
36
Oil
66
15
on
nation his been acquired9
0
3
II
6
5
8
3
7
1
3
6
3
5
3
2
1
2
6
1
6
15
21
119
refining
0
7,300
17,700
12.000
12,000
81,700
31,500
90,100
18,000
150,000
100,000
61,700
92,000
73.000
56.000
30,000
33,500
131,000
16,000
221,100
620,100
1,979,300
3,939.100
capacities
1 Thta hBfl
0
12,100
25.300
53.100
18,350
93.100
53,270
102.250
27,000
209,100
112,100
63,710
100,000
103.000
72,000
30,000
39,000
227,900
55,300
261, --00
757,500
2,271,500
1,726,110
In bpsd
0
3 32
26 67
33 33
62 50
11 Lli
100 00
77 78
50 00
88 89
75 00
75 00
83 33
75 00
100 00
50 00
66 «7
100 00
100 00
75 00
83 21
100 00
rlumbor of Re-
fineries In Range
With TCC
Facilities
0
0
1
3
1
6
0
0
1
1
2
0
1
1
0
1
1
0
0
0
1 (1)'
0
21
% Units Having
CC Facilities
T
3 32
33 33
50 00
75 00
77 78
100 00
77 78
100 00
100 00
100 00
75 00
100 00
100 00
100 00
100 00
100 00
100 09
100 00
75 30
100 03
100 00
I of "otal "C
Capacity (Cumulative)
0
0 26
0 80
1 93
2 96
1 91
6 06
8 23
8 80
13 22
15 59
17 05
19 16
21 31
„•? 87
23 50
21 33
20 15
30 32
35 35
51 88
1C3 :i
In a slight difference In the total refining capacity from that reported In the source
Also, only 17 refineries were listed for the state of Louisiana, which Is In disagreement
with the total of IB refineries counted by the source This resulted In the total shown
with _—
of 216 refineries Instead of 217
-------
also be reasonably assumed that the smaller capacity units will
be phasing out. Furthermore, it is apparent from Table 6 that
the refineries in the capacity range from 0 to 30,000 bpsd are
responsible for only 11.35% of total United States production
capacity, and less than 5$ of the total FCC capacity.
Based on these facts we have chosen the range of FCC capacities
between 10,000 and 150,000 bpsd for economic evaluation of the
desulfurization systems for the fluid catalytic cracking operation.
Specifically, our cost estimates were prepared for three capacities:
10,000, 50,000, and 150,000 bpsd of cracked stream. The data
obtained for these capacities have been plotted to obtain the
figures for intermediate sizes.
3.4 EMISSIONS FROM FCC UNITS
3.4.1 General
Atmospheric emissions from catalytic cracking operations have
been established by several investigators. The first relatively
complete study of emissions from the refineries was con-
ducted in the years of 1955 through 1958 in Los Angeles County,
California.1-3
A candid effort by the Environmental Protection Agency to
establish emission standards for different groups of industries
has recently revived interest in petroleum refinery emissions,
specifically, in the emissions discharged to the atmosphere from
fluid catalytic cracking (FCC) operations.1*-7 Nationwide emission
trends have also been established.8 A summary of data obtained
from these sources is presented in Table 7-
These studies were also very helpful in establishing and defining
the operations in petroleum refineries that are mostly responsible
22
-------
Table 7- EMISSION RANGES FROM FLUID CATALYTIC CRACKING UNIT
REGENERATOR, BEFORE AND AFTER CO BOILER
Fresh Feed Rate, bpsd 156,000
Recycle Feed Rate, bpsd 35,000
Stack Discharge Rate,
scfm (60°F, 1 atm, dry basis)
Temperature, °F
Emissions:
Sulfur Dioxide, ppm**
Nitrogen Oxides (as N02) ppm
Carbon Monoxide, % vol.
Carbon Dioxide, % vol.
Oxygen, % vol.
Moisture, % vol.
Nitrogen, % vol.
Hydrocarbons, ppm
Ammonia, ppm
Aldehydes, ppm
Cyanides, ppm
Particulates, grains/scf
Before
CO Boiler
484,000
1000-1200
140-3300
8-394
7.2-12.0
10.5-11-3
0.2-2.4
13-9 - 26.3
78.5-80.3
98-1213
0-675
3-130
0.19-0.94
0.08-1.39
After
CO Boiler*
Up to 30% Volume
Increase
(On Wet Basis)
485-820
Up to 2700f
Up to 500 f
0-14 ppm
11.2-14.0
2.0-6.4
13.4-23.9
82.0-84.2
0.017-1.03
**
Emissions after CO boiler will be affected by the type of
supplemental fuel and operating conditions in the CO boiler
It was reported that up to 60% of sulfur oxides in regen-
erator flue gas may appear as SO3 (see page 85)
Estimated
23
-------
for atmospheric pollution. However, they have also indicated
that the air pollution problem of the petroleum industry is very
complex and varies from refinery to refinery. Specifically, the
type of raw crude oil processed in the refinery as well as the
types of unit operations that a specific refinery utilizes to
produce its final products have a significant effect on the
severity, amounts, and concentrations of atmospheric emissions.
Consequently, a need to define the air pollution problem for each
refinery in the United States individually and individual ap-
proaches to a refinery air pollution abatement was strongly felt
throughout the whole study.
3.4.2 Basis for Comparison
The complete technical and economic comparison of the individual
approach alternatives as well as the individual systems included
in these alternatives could be made only if an equal and uniform
basis for this comparison were well defined. Consequently, we
have established a "typical" FCC operation based on the sulfur
material balance. This "typical" FCC unit may not correspond
exactly to or represent any actual refinery in the United States
but was defined as accurately as possible to represent the present
and future conditions. As such it was used as the basis for
systems comparison, technical as well as economic, for all three
approach alternatives evaluated.
The flue gas desulfurization processes developed primarily for
power generation sources (in our study defined as add-on systems)
were considered applicable for FCC flue gas desulfurization and
were to become a significant part of the study. Applying
operating conditions and economic assumptions similar to those
applied to power plant flue gas desulfurization processes to
the case of refineries would simplify our task without major
-------
applicability restrictions and would result in much more effective
conclusions. The majority of flue gas desulfurization systems
have been studied within the range of sulfur dioxide concentrations
of 2,000 to 2,500 ppm with removal efficiencies ranging from 90
to 95%. All the economic evaluations for these systems presently
available are based on similar assumptions. Consequently, we
have applied similar assumptions (2000 ppm S02 in regenerator
flue gas and 905? removal) to composition and removal efficiencies
of sulfur dioxide from FCC regenerator flue gas. Table 7, which
summarizes the operating conditions and emission ranges of the
FCC units in the U.S., shows that this assumption is well within
the range existing in refineries. The sulfur dioxide emission
level from most U.S. refineries is below the 2000 ppm level.67
However, worldwide demand for petrochemical products indicates
that high sulfur crudes will need to be processed in the future
creating the potential for increased sulfur levels in refinery
atmospheric emissions. Also, the size of the add-on systems
applied to desulfurization of regenerator flue gas is a function
of the gas volume handled by the process and will not change
drastically with sulfur concentrations. (See also Section 6.3.)
The effect of sulfur concentrations on operating cost can be con-
sidered negligible since sulfur credit was excluded from our
cost analysis.
3.4.3 Predicting S02 Emission
As has already been indicated, some sulfur, whether in the form
of sulfur-containing hydrocarbons or as hydrogen sulfide, is
carried with the catalyst into the regenerator. Here, the sulfur
compounds, along with carbon and hydrogen, are oxidized and leave
the regenerator in the form of sulfur oxides. The oxidation
reactions in the regenerator are carried out in an oxygen-lean
atmosphere. The conditions of oxidation are very carefully con-
trolled. Consequently, the sulfur dioxide concentration in the
25
-------
gases leaving the regenerator will be a function of the amount of
sulfur present on the catalyst to be regenerated and amount of
air used for regeneration. Since the amount of air required for
catalyst regeneration (or coke combustion) is proportional to
amounts of individual elements (carbon, hydrogen, sulfur) carried
by the spent catalyst, and since the degree and amount of oxidized
carbon can be represented by the C02/C0 ratio in the off-gas
leaving the regenerator, the following equation was developed
that enables prediction of S02 concentration in the regenerator
off-gas as a function of hydrogen and sulfur content of coke and
C02/C0 ratio.
S02(vppm) = - S x - (1)
2.667 + 5.015 (^) U-S-H) + 27.429H + 2.095 S
where
S02 concentration in volume parts per million (vppm) is
calculated on the dry basis
S = weight fraction of sulfur in coke
H = weight fraction of hydrogen in coke
R = mole ratio of C02 to CO in regenerator off-gas
The full derivation of this equation and necessary assumptions
are supplied in Appendix B.
Based upon the above equation, the weight fraction of sulfur in
coke before combustion, S, would have to be 0.002*13 in order to
obtain a concentration of 200 ppm S02 in the regenerator off-gas
with R = 1 and H = 0.1. This can also be determined from the
diagram in Figure 5, which was calculated using Equation 1. As
it can be seen from Figure 5> sulfur concentration as a function
of sulfur content of coke is practically linear within the sulfur-
on-coke concentration range of 0-3% wt.
26
-------
3000
o_
Q_
O
«*—
O
2000
cz
O
"co
o>
u
c
O
O
1000
200
0.01 0.02 0.03
Sulfur Content of Coke, wt. fraction
Figure 5. S02 Concentration in FCC Regenerator Off-Gas
27
-------
This equation is also very useful in defining the degree of FCC
feedstock desulfurization that would be equivalent to levels of
regenerator off-gas sulfur emissions of 2000 ppm and 200 ppm.
The sulfur content of coke (weight fraction) that would result in
200 ppm sulfur emission was already shown to be 0.002*13, or 0.2*13
wt %. Similarly, the sulfur content of coke that will produce
2000 ppm S02 in the regenerator off-gas was determined to be 2.43
wt %. The ratio of wt % sulfur on coke to wt % sulfur in the FCC
feed ranges from about 0.7 to i.227?29»33>35>61 depending on
FCC operating conditions (primarily sulfur content of the feed,
type of catalyst, steam stripping rate, catalyst-to-oil ratio, oil
recycle, feed hydrodesulfurization, etc). Applying this ratio
one can convert the sulfur content in coke to sulfur content in
the feed or vice versa. In our definition of the degree of FCC
feedstock desulfurization (Section *l) which would be equivalent
to emissions of 2000 ppm and 200 ppm, we have therefore assumed
the sulfur content of feedstock (or the degree to which the
feedstock must be desulfurized) to be 2.**3 wt % and 0.2*13 wt %.
The figures are equal to sulfur contents in the coke since the
ratio of wt % sulfur in coke to wt % sulfur in the FCC feed was
assumed to be 1.
3.5 SUMMARY OF ASSUMPTIONS
Two types of assumptions, technical and economic, had to be made
in order to assure the comparison of all of the processes on the
same basis. The assumptions were made to agree as accurately as
possible with the conditions presently existing in refineries
and were used throughout this whole study. The foundation for
and discussion of the assumptions were presented in previous
sections. When additional assumptions relative to a specific
application were necessary, those are introduced and defined in
the appropriate sections of this report.
28
-------
3.5-1 Technical Assumptions
The majority of the technical assumptions define the operating
conditions and sulfur balance around the fluid catalytic cracker.
These are:
- The regenerator off-gas temperature is 1000°P.
- The regenerator off-gas sulfur dioxide concentration
is 2000 ppm.
- A reducing atmosphere exists in the regenerator off-gas.
*.<- .
- Regenerator off-gas is diluted by 21% in passing through
the CO boiler.
- An oxidizing atmosphere exists in the CO boiler flue gas.
- The CO boiler flue gas temperature is 500°F.
- No sulfur oxides are added to the flue gas from the
supplemental fuel used in the CO boiler.
- The sulfur dioxide concentration in the gases emitted to
the atmosphere is 200 ppm .
3.5.2 Economic Assumptions
The following general economic assumptions were used in the study,
several of which are expressed as percentages of fixed capital
investment (F.C.I.):
A. FCC Unit Size: 10,000, 50,000 and 150,000 bbl feed capacity
29
-------
B. Capital Investment
1. Start-up cost - 10% F.C.I.
2. Working capital - 10.5% F.C.I.
3- Interest on construction loan - construction period of
12 months; financed fixed capital at the rate of 855/yr
for average of half of construction period assumed
4. Does not include sulfur recovery plant capital cost
5- Base period - February 1973
6. Scaling factor - 0.6?
7. CE plant cost index
1968 113-7
1969 119-0
1970 125-7
1971 132.3
1972 137.2
Feb. 1973 140.4
C. Operating Cost
1. Labor - $5-50/manhour
2. Maintenance labor - 2% F.C.I.
3. Maintenance materials - 2% F.C.I.
4. Control laboratory labor - 20$ of operating labor
5. Process water - 30
-------
10. Hydrogen cost - 40(fc/1000 scf
11. Plant overhead - 80% total labor
12. Taxes & insurance - 2% F.C.I.
13. G&A, sales, research - 6% F.C.I.
14. Depreciation - 10% F.C.I.
15. Interest on working capital - 6% working capital
16. Return on investment - 205?
17- Value of steam - 50
-------
4. FCC FEED DESULFURIZATION
Desulfurization of the feed was the first approach alternative
considered for reduction of the sulfur emissions from FCC
operations. Desulfurization technology is very well known to
petroleum refiners and can be applied to almost any petroleum
stock. A large research effort has been directed towards develop-
ing this technology, and a large number of desulfurization
processes applicable to various feedstocks, to desulfurize to
different degrees, and of different designs have been developed
H.l TECHNICAL EVALUATION
In our study we considered the desulfurization of the feed to
the cat cracker as one of the alternatives to reduce the sulfur
emissions in FCC regenerator flue gas. During the study it
appeared that in order to correctly evaluate desulfurization of
the feed to the FCC process, the refinery should not be penalized
for the total cost of the desulfurization, but only for the part
of the desulfurization cost that would result in additional re-
duction of sulfur emissions. If the refinery is willing to de-
sulfurize this feed to a certain degree for whatever reason,
whether to optimize its operation or produce products of needed
quality, but disregarding the sulfur emission in regenerator flue
gas, this cost should not be used for comparison with the other
two approach alternatives. Generally, the sulfur concentrations
of the original gas oils can vary from 0.51 to 3.36 wt % of
sulfur.24
Our theoretical system has been defined in two cases. Case A
will desulfurize the FCC feed from a maximum gas oil sulfur con-
centration of 3.36? to 0.2435? to obtain 200 ppm sulfur emissions
32
-------
in regenerator flue gas. Case B will desulfurize the feed from
3.36% to 2.1*3% sulfur in the feed, which corresponds to 2000 ppm.
The cost difference between Cases A and B thus becomes a measure
of cost needed to desulfurize the FCC feed from 2.^3% to 0.2^3?
equivalent to regenerator flue gas S02 concentrations of 2000 ppm
and 200 ppm, respectively.
The results of such analysis would obviously render a comparison
value even if desulfurization operating conditions of a specific
refinery are considered. For example, if a refinery is desulfur-
izing to a level lower than 2.^3%, the sulfur emissions in the
regenerator flue gas will be lower than 2000 ppm. The expenditure
that a refinery has to make to improve its desulfurization unit
to reduce sulfur emissions to 200 ppm will then be within the
range of figures represented by the cost difference of the two
cases. It should be noted however, that the cost of an add-on
system as it appears in this report will remain basically un-
changed despite the degree of feed desulfurization because
essentially the same volume of flue gas will have to be handled
by this unit.
A modern petroleum refinery is a highly complex operation. In
addition, each refinery operates on its own typical scheme de-
pending on the composition of the crude oil that it is processing,
the type and quality of product produced, and many other variables
Consequently, the benefits that can be realized from the desul-
furization of some of the feedstocks will be primarily a function
of the operations that a specific refinery utilizes to produce
its products. The "real" desulfurization costs should be obtained
from the optimization of the whole refinery operation. The basic
cost estimates presented in this report have been prepared for
the desulfurization of the feed to the catalytic cracking unit
without considering any benefits of this desulfurization on
33
-------
improved conversion efficiencies and reduced operating costs of
the fluid catalytic cracking unit.
This economic evaluation of feedstock desulfurization appeared
to be necessary in order to be able to compare the economics of
all three of our approach alternatives, namely, process modifi-
cation, desulfurization of feed, and add-on flue gas desulfuriza-
tion systems.
Presently, very few desulfurization units are used in the U.S.
The feed to catalytic cracking units is only 5-3% of the total
desulfurization capacity (see Appendix C). When the new air
pollution regulations are enforced and applied fully and the
new energy supply basis is established, it can be expected that
a significant change will occur in petroleum refining, specifically,
desulfurization of some feedstocks. For these reasons, to
enable application of desulfurization technology to air pollution
abatement, and to understand its present application as well as
future potential, we have included in Appendix C a brief discussion
of present desulfurization techniques and air pollution regulations
that will affect the scope to which this technology may be
applied.
4.2 ECONOMIC EVALUATION
The capital and operating costs of FCC feed desulfurization are
presented in Figures 6 and 7 and the incremental cost between Cases
A and B in Figure 8. A detailed breakdown of the cost is summarized
in Appendix D.
Some additional assumptions were made in these economic
evaluations in addition to the assumptions summarized in Section
3.5. These are listed on the following page.
-------
Process Conditions
Reactor pressure, psig 1000
Reactor average temperature, °F 750
Liquid hourly space velocity, hour"1 1.0
• H2/oil ratio, scf/bbl 4000
Hydrogen consumption, scf/lb of
sulfur removed 48
Boiling range of feed, °F 600-1020
Case A Case B
• Sulfur in feed (% wt) 3-36 3-36
• Sulfur in product (% wt) 0.2^3 2.H3
Fixed bed type reactor
Capital Investment Cost
The cost does not include a hydrogen plant
General service facilities - 15% of total process and
utilities cost
Operating Cost
2 operators per shift
Catalyst cost, $1.05/lb
Treatment loss, $0.1?8/bbl of feed
Glaus unit operating cost, $l6.0/long ton of sulfur
produced
35
-------
100
o
i/i
g
1 10-
U)
ON
O
O
a>
I
s
I I
_L
I
Figure 6.
10 100
Plant Capacity, thousand b/sd
Hydrodesulfurization of FCC Feedstock, Capital Investment Cost
(February 1973)
1000
-------
1000
UJ
in
5100
2
o>
10
Case A
CaseB
Wt. % Sulfur
Inlet Outlet
Case A 3.36 0.243
Case B 3.36
2.43
10 100
Plant Capacity, thousand b/sd
1000
Figure 7. Hydrodesulfurization of FCC Feedstock, Operating Cost
-------
looor
o
O
look
U)
00
2
o>
o.
O
10
I
10 100
Plant Capacity, thousand b/sd
1000
Figure 8. Hydrodesulfurization of FCC Feedstock, Incremental Operating Cost
-------
5. PROCESS MODIFICATION
The second conceptual technique evaluated for reducing petroleum
FCC regenerator SOX emissions was FCC process modification. The
purpose of process modification analysis was to determine if a
location in the FCC flow train exists that could be easily modified
and still be competitive with the other two approaches. Justifica-
tion for such an analysis was found and also confirmed later in
the fact that petroleum refiners primarily concentrate on hydro-
carbon production. The sulfur oxides emission reduction from FCC
operations was either of no concern or of secondary importance.
The conclusions of our analysis are summarized below.
5.1 SUMMARY OF CONCLUSIONS
(1) Most of the presently used cracking catalysts are of the
crystalline, silica-alumina type. All three types of
catalysts utilized in catalytic cracking (natural,
synthetic, and synthetic-zeolitic) are built from the
same principal elements and compounds, namely, silicon
dioxide and aluminum trioxide.
(2) Definite differences in sulfur poisoning and activity for
all three catalyst types had been observed, with the
natural catalyst having the highest and the zeolite catalyst
the lowest poisoning properties.
(3) All three catalyst types are known to have demonstrated
affinity to water vapors. Due to the molecular similarities
of H20 and H2S, the catalysts can adsorb H2S also. Some
affinity for sulfur-containing hydrocarbons has also been
demonstrated.62 Presently, however, insufficient data
are available on the catalysts ' affinity to H2S and other
sulfur-containing hydrocarbons at the conditions existing
in fluid catalytic crackers.
39
-------
(H) Although it has been determined that H2S has a definite
poisoning effect on natural catalysts, the mechanism of
poisoning is not fully understood.
(5) Due to the high poisoning properties of natural catalysts,
a means of reducing their poisoning was investigated. It
was determined that contacting of the catalyst with steam
can prevent or reduce this catalyst poisoning.
(6) Because the sulfur poisoning of zeolite and synthetic
catalysts is much less severe, no evidence was available
on the effects of steam contacting these catalysts.
(7) Hydrocarbon cracking using any of the catalysts will result
in deposition of coke on the catalysts. Regardless of the
various mechanisms for the presence of sulfur on the
cracking catalysts (H2S, thiophenic sulfur, adsorption or
chemical reaction with the catalyst or coke, non-thiophenic
sulfur) it has been definitely demonstrated that applying
steam contacting (steam stripping) to spent natural catalyst
will reduce the sulfur transported into catalyst regenerator
and consequently reduce the sulfur oxide emission from the
FCC regenerator.
(8) Steam contacting of natural catalyst has been applied in
three locations in fluid catalytic cracking units:
spent catalyst contacting (stripping steam), regenerated
catalyst contacting (hydration steam), and steam contacting
in the presence of FCC high sulfur feed hydrocarbons
(dispersion steam). Applying steam at all three locations
had a beneficial effect on improved catalyst activity.
conversion, and reduced poisoning. No drawbacks from the
steam contacting on the natural catalyst have been found.
-------
(9) Stripping steam has been sufficiently demonstrated to have a
significant effect on reduction of sulfur oxide emissions in
regenerator off-gas (refer to Section 5.2.1 for details).
(10) At the present time, the refineries use steam stripping to
recover hydrocarbons carried on the spent catalyst to the
regenerator. Consequently, catalyst steam stripping equip-
ment exists in refineries and refiners are fully familiar
with this technique.
(11) Literature search and industry contacts failed to reveal
that steam stripping has been considered and evaluated by
the petroleum refiners as a means of regenerator off-gas
sulfur oxide emission control.
(12) Present steam stripping rates are much lower than those
applied to older (natural) catalysts (see Section 5-3.7).
The rate of steam stripping applied to present catalyst
and required to reduce sulfur oxide emissions from the FCC
regenerator flue gas to the level of 200 ppm has not been
determined. Applying the data available for natural catalyst
and obtained in the late forties indicates that the steam
stripping rate would have to be increased in existing units
about four to twenty times. Considering the technical
developments in the area of catalysts and equipment, this
requirement may be significantly reduced. Even in the first
studies with natural catalyst the steam rates in commercial
units were about 2 to 5 times lower than those measured in
pilot scale units and producing the same effects.
(13) If the sulfur carried by the spent catalyst is in the form
of H2S, stripping should be possible. If the sulfur is in
the form of hydrocarbons, an indication exists that this
sulfur can be replaced using steam and H2S can be formed
over A1203 catalyst. (See Section 5.2.2.)
-------
(1*1) Steam contacting of the zeolitic catalyst at temperatures
below 1100°F will not cause catalyst deactivation. 6I*
(15) Steam stripping will result in increased production of H2S,
a product that is now easily handled by the refineries.
(16) Economics of steam stripping are closely competitive with
other sulfur emission reduction alternatives even though
the most conservative assumptions were applied in our
steam stripping economic analysis.
(17) No special chemicals are needed except steam, which is
available in the refineries and fully compatible with
existing equipment.
(18) Since catalyst contacting occurs at the temperatures
normally existing in FCC operations, no excessive heat
exchange is required. The steam condenser used in steam
stripping represents a much more efficient heat exchange
than the gas-gas heat exchangers so often needed in add-on
system applications.
(19) Although steam stripping will probably result in some
increase in catalyst attrition rates, it offers a potential
solution to reduction of particulate emissions from FCC
regenerators. (See Section 5-3-^.)
(20) Further simplification of the steam stripping technique is
possible so that only relatively minor FCC equipment
modifications may be required (see options 2, 3» and H in
Section 5-3.2).
-------
(21) In some applications, catalyst steam contacting may, through
reducing the amount of coke on the spent catalyst, decrease
the heat load of the FCC regenerator to levels that would
not be sufficient to maintain operating temperatures in
this equipment. Additional investigation of these systems
is needed to determine their present sulfur emissions and
application feasibility of regenerating air preheating.
(22) Steam stripping appears very attractive for refineries
having low sulfur emissions. Applying add-on systems to
these refineries would require essentially the same capital
investments as in the refineries with high emissions.
Steam stripping could be applied by varying the steam
stripping rates.
(23) Steam contacting can control total sulfur oxide emissions
from the FCC regenerator regardless of the form in which
the sulfur is emitted (S02, S03, E2SO^ mist).
5.2 PROCESS MODIFICATION ANALYSIS
The initial steps in our process modification analysis consisted
of comparing the FCC process with moving bed catalytic cracking
also called Thermofor catalytic cracking (TCC) based upon process
operating variables. Analysis of available literature suggested
that the differences in S02 emissions for these two types of
catalytic cracking were not due to a difference in the type of
crude oil processed in these refineries. Sulfur emission data
and the majority of process operating conditions, summarized in
Table 8, have been obtained at the same time and from the same,
relatively small geographical location in the United States (Los
Angeles, California), where presumably the same crude oil is
processed. Also, the comparison was made for an average of 6 FCC
and 9 TCC units, all again from the same geographic area.
-------
Table 8. COMPARISON OF CATALYTIC CRACKING UNITS
OPERATING CONDITIONS1'17'55
FCC TCC
Number of Units 6 9
Fresh Feed Rate, bpsd 156,000 69,000
Recycle Feed Rate, bpsd 35,000 38,000
Fresh Feed to Recycle Ratio 4.5 1.8
Catalyst Circulation Rate, Tons , •. nnn , i.nn
per Hour 14,000 1,400
Stack Discharge Rate,scfm .g, ,
(Dry Basis) 404,000 134,000
Sulfur Dioxide Emission, ppm 308-2190 65-141
Steam Stripping Rate, Ib H20/1000 Ib , R Q , ,-
Catalyst 1'°~9 ^
Catalyst to Oil Ratio* 12 2
Stack Discharge to the Feed Ratio, ~ ,- -\ oc
scfm/bpsd *° J"0
Reactor Temperature, °F** 885-975 780-950
a/Lit
Reactor Pressure, psig 9-20 10-15
Regenerator Temperature, °F** 1,000-1,150 960-1,080
# #
Regenerator Pressure, psig 1-13 Atm. to 1
* Density of oil 7 Ib/gallon, 42 gallons/barrel,
2,000 Ib/ton
** All data have been obtained or calculated from
Reference 1, except the data marked which were
taken from Reference 17 or 55
-------
Several conclusions can be drawn from the comparison shown in
Table 8:
(1) TCC units have significantly lower sulfur dioxide emission
levels than FCC units.
(2) Two and one-half-fold higher fresh-feed-to-recycle ratio
for FCC units would indicate that much heavier feedstocks,
that usually contain high amounts of sulfur,were processed
in TCC units.
(3) The two-fold higher sulfur stack discharge rate per barrel
per day of total feed for the FCC unit indicates that the
absolute amounts of sulfur emitted from the FCC units were
even higher than those indicated by ppm sulfur dioxide
emission data.
(i|) Both of the two previous conclusions indicate that sulfur
oxide concentrations from FCC units were significantly
higher than the sulfur emissions from TCC units.
(5) Although the data for the steam stripping rates and
operational temperatures and pressures were taken from a
different source than the rest of the operating conditions
shown in Table 8, the data nevertheless represent the values
that are generally applied in FCC and TCC units. With
most of the operating conditions (temperature, pressure)
relatively equal and approximately two-fold higher steam
stripping rate for TCC units than that for FCC units, it is
quite evident that the rate of steam stripping can be
directly connected with the differences in sulfur emissions
from the two unit types compared.
-------
Based on these observations, additional evidence was sought
on the effects of steam stripping on sulfur distribution in
catalytic cracking operations to either support or refute this
approach. The steam stripping approach was considered a simple
and economically feasible modification of fluid catalytic cracking
units. It was learned that steam stripping was and still is
used by petroleum refiners to remove hydrocarbons from the spent
catalyst and also to prevent catalyst poisoning to some degree.
However, the technique has never been viewed as a means of
reducing SOX atmospheric pollution from catalytic cracking units.
This statement has been confirmed a number of times during this
study by communication with qualified individuals in the refining
industry to whom this approach was presented. The present use
of steam stripping (although at much lower rates) by petroleum
refiners assures that the technique would be fully compatible
with the petroleum refinery operations. Because of the slightly
different purpose for which steam stripping is presently used
and because essentially no evidence was found of use of this
approach by petroleum refiners to control SOX emissions, more
evidence was needed as to the ability of steam stripping to
control sulfur emissions from FCC regenerator. This is why the
steam stripping approach was fully investigated and analyzed in
this study.
5.2.1 Evidence of Effects of Steam Stripping on Sulfur
Distribution in Fluid Catalytic Cracking
In the late 19^0's, when catalytic cracking started to become a
major source of motor gasoline, many oil companies were in the
process of optimizing the operation of FCC units. Among the
many problems that had to be solved was the problem of catalyst
deactivation by sulfur present in FCC feedstocks and fed into
the reactor. In an effort to understand the mechanism of
-------
catalyst poisoning, several investigators studied the sulfur
material balances around the FCC units using natural and synthetic
catalysts. Their studies involved the determination of sulfur
distribution in the products as a function of process operating
conditions, feedstocks, and catalyst types.
Strong evidence of a correlation between use of steam stripping
and degree of sulfur emissions in the regenerator flue gas is
the comparison of operating conditions for both FCC and TCC units,
presented in the previous section. This evidence is further
supported by a comprehensive review article written by Sittig25
who states, "When a feed stock containing H2S is contacted with
the catalyst, H2S competes with any steam in the feed for locations
in the catalyst Montmorillonite inter-laminar spaces. Controlled
rehydration by steam prevents interaction of the catalyst with
H2S with subsequent poisoning." Work supporting this statement
has been conducted and described by R. C. Davidson.30 This work
also discusses the theoretical aspects of natural catalyst sulfur
poisoning and catalyst dehydration which result in a progressive
increase in coke and gas yields at the expense of gasoline yield
for both natural and synthetic catalysts.
Although most of the work has been carried out on a pilot scale,
some commercial data are presented confirming laboratory findings
that definitely indicated that catalyst hydration reduces sulfur
poisoning and thus improves the catalyst activity, increases
gasoline yield, and reduces the coke yields. In two comparable
runs, one without and one with hydration, the amount of sulfur
emitted from the regenerator was decreased to less than one half
by using 0.5 wt % (based on catalyst) of rehydration steam.
-------
In his tests,Davidson used high temperature steam in two different
locations: (1) in the mixture with feed gas oil (dispersion steam)
and (2) admitting steam near the bottom of the regenerated catalyst
standpipe (hydration steam). About three times as much dis-
persion steam as hydration steam was required to obtain comparable
benefits. Also, the dispersion steam did not prevent the catalyst
from poisoning as well as did the hydration steam.
Additional investigations performed by Healy and Hertwig31 and
Conn and Brackin32 concerning the use of high temperature steam
to prevent sulfur poisoning of the cracking catalyst are of
special interest in this discussion. Healy and Hertwig studied
in detail the distribution of sulfur in the FCC feedstock to four
major products, namely, cracked gas, gasoline, cycle oil, and
coke. They found that about 50% of the feed sulfur remained in
liquid products of fluid catalytic cracking. The rest of the
sulfur was distributed between the gas product (455?) and coke (5?).
Their study was conducted on a pilot scale with feeds of varying
origin and sulfur content. The effect of other significant process
variables, such as temperature, conversion, and pressure on sulfur
distribution was also indicated. Three commercial cracking
catalysts were studied, namely silica-alumina, silica-magnesia,
and natural catalyst. Silica-magnesia catalyst gave the lowest
amounts of sulfur in the coke (about 6.7 times lower than silica-
alumina catalyst and 5 times lower than natural catalyst). This
indicates that catalyst selection can be a significant factor in
reducing sulfur emissions from the PCC regenerator. Similar
observations have been reported by Sittig25 where specially
treated sulfur-resistant (SR) catalysts containing relatively
high amounts of aluminum trioxide have been developed. The data
on sulfur distribution obtained for silica-alumina catalyst at
900°F reactor temperature are presented in Figure 9- Other
investigators have conducted similar studies.33'31*
-------
10
0.2
0>
_c
"o
t/>
(D
.2*
'CD
o
o>
§1.0
2.0
o>
•g
O
1.0
30 35
I
40 45 50
Conversion, Vol. %
55 60
Figure 9. Distribution of Sulfur in FCC Products
-------
In most of the experiments with silica-alumina catalyst the
sulfur content of the coke was between 1.22 and 1.9 wt % depending
on conversion. However, figures as high as 5 wt % (in one case
11/5) have been reported depending on the type of the feed. The
feed consisted of various types of virgin, cracked, or cycle gas
oils with sulfur contents from 0.36 to 1.935?. The significant
effect of steam stripping on the amount of sulfur in the coke has
also been briefly discussed and reported. This effect has been
observed with both good quality and contaminated natural catalysts.
The data showing the variation of sulfur content of coke with
the rate of steam stripping are presented in Figure 10. The
sulfur removed from the coke at high stripping steam rates appears
in the gas as H2S .31
The other work, conducted by Conn and Brackin,32 was a compre-
hensive study on prevention of catalyst sulfur poisoning by use
of steam. The study was done on both pilot and full scales.
Three different locations for steam injection were investigated.
Two of these locations correspond to those studied by Davidson,30
that is, dispersion and hydration steam. The third location
of steam injection is in the spent catalyst stripper; this is
termed stripping steam. All three cases have demonstrated that
steam has a significant effect on sulfur distribution in fluid
catalytic cracking.
The introduction of steam to the regenerated catalyst standpipe
(hydration steam) serves to hydrate the catalyst prior to contact
with oil, thus preventing loss of catalyst selectivity. Experi-
mental work carried out using high sulfur gas oil and 0.5 wt %
hydration steam (based on catalyst) basically confirmed the re-
sults obtained by Davidson30 described earlier. The catalyst
activity declined, the carbon factor practically doubled, and
declines in conversion were observed after the use of hydration
steam was discontinued.
50
-------
3.0
o>
O
*o
2.0
o
O
1.0
234
Stripping Steam, Wt.%ofCoke
Figure 10.
Effects of Steam Stripping
Rate on Sulfur Content of
Coke (Natural Catalyst)
51
-------
Use of dispersion steam has also been investigated. Although the
benefits of dispersion steam in maintaining catalyst selectivity
claimed by Davidson have not been definitely confirmed, a marked
improvement in product yields (0.4# dispersion steam resulted in
more than 5% higher conversion) has been observed in cracking
gas oil of high sulfur content.
Stripping steam was applied in both pilot and commercial units.
In a pilot scale experiment a mixed gas oil containing 1.39?
sulfur was cracked for a 24-hour period. The steam stripping
"rate" was somewhat lower than that employed by commercial units,
0.5-0.8 wt % (based on catalyst). Also, no hydration or dis-
persion steam was used. After 24 hours the stripping rate has
been increased to 4 to 6 wt 55. The immediate effects of this
experiment on increased catalyst activity and lowered carbon and
gas factors are demonstrated in Figure 11 for both pilot and
commercial units.
It should be noted that the steam stripping rates during the
commercial experiment varied between 1.15 to 2.0 wt % and were
considerably lower than the 4 to 6 wt % rates applied in the
pilot plant. This can probably be explained by the better
steam stripping efficiency obtainable in large-scale units.
Additional steam stripping experiments were performed in com-
bination with hydration and dispersion steam. The coke yields
varied from 1.2 to 5 wt %. Mixed gas oil containing 1.45%
sulfur was used. The steam rates varied between 0.68-3-71 wt %
for stripping steam, 0.27-1-5 wt % for dispersion steam, and
0-34-1.86 wt % for hydration steam. This rather complex experiment
revealed that when hydration steam is present to protect the
catalyst from sulfur poisoning, stripping steam can effectively
be used to improve the properties of the catalyst, rejuvenate
52
-------
28
20
LC
ACTIVITY
Jk-
•PILOT PLANT REJUVENATION
OF CATALYST REMOVED FROM
COMMERCIAL UNIT
•TEST IN COMMERCIAL UNIT
1-2 WT. % STRIPPING STEAM
wt t-»
STRIPPING STEAM
„ CARBON FACTOR
I
STRIPPING STEAM
1 i 1
20 4O 60 80
HOURS ON STREAM
100
Figure 11.
Improvement of Catalyst Activity and
Selectivity with High Stripping-Steam
Rate
53
-------
the poisoned catalyst to selectivity levels approximating fresh
catalyst, and reduce formation of coke.
Additional evidence of reduced sulfur emissions from the regen-
erator flue gas due to steam stripping is presented in Figure 12.
One curve represents the data obtained by Conn and Brackin.32
The other curve was obtained from the data measured by Healy and
Hertwig31 by using our mathematical model expressing the relation-
ship between the hydrogen and sulfur content of the coke, C02/C0
ratio in regenerator flue gas, and sulfur emissions , and presented
in 3.4.1. A good agreement between the two measurements
exists. A C02/C0 ratio of 1.2 and a hydrogen content in coke of
0.1 were used in the conversion. As it can be seen from Figure
12, S02 emissions from FCC regenerator can be decreased to 200
ppm or lower by steam stripping the spent catalyst at the proper
rates. The data in Figure 12 show that the quantity of steam
required to achieve a 200 ppm level can vary from 2 to 5 wt % of
catalyst (2-5 Ib H20 per 100 Ib catalyst).
5.2.2 Theoretical Aspects of Steam Stripping
Heating of the silica-alumina catalyst to different temperatures
results in different degrees of catalyst dehydration depending
on heating temperatures. The water, normally part of the catalyst
molecular structure, is driven off. The catalyst can be sub-
sequently rehydrated by cooling. However, two temperatures
appear to cause a significant structural change in the catalyst.
If the natural catalyst is heated above 600°F, the water lost
during the heating will cause a permanent structural change and
the catalyst will not rehydrate by cooling and exposure to humidity.
Heating of the catalyst above 1^50°F will result in the destruction
of the catalyst's montmorillonite structure.
-------
2600
2400
£ 2200
CTJ
2000
1800
£ 1600
o>
c=
o>
1400
c 1200
•£ 1000
CD
| 800
CM
8 600
400
200
Healy & Hertwig 1.46% Sulfur in feed
Conn & Brackin 1.39% Sulfur in feed,
From N\RC Math Model
H = 0.1 R = 1.2
0
1234
Steam Stripping Rate, Wt% of Catalyst
Figure 12. Effect of Steam Stripping of Spent FCC Catalyst
on SC-2 Concentration in Regenerator Off-Gas
55
-------
It is between these two temperatures that the catalytic crackers
operate during the reaction and regeneration cycle. In this
temperature range, usually at around 800°F, the catalyst will
lose additional water which can, however, be resorbed and has
been called by Davidson30 the water of constitution.
The experimental work conducted by Davidson indicates that, owing
to several analogous chemical properties of hydrogen sulfide
and water, these two compounds can compete in sorption on silica-
alumina catalyst. Adsorption of higher amounts of H2S on the
catalyst at 1050°P produces a material almost identical to that
which has been sulfur poisoned and thus possesses lower activity,
selectivity, and higher carbon yield. Many scientists believe
that the poisoning is caused by a reaction of iron in the catalyst
with H2S and formation of iron sulfide. Two other experiments
made by Davidson have definitely proven that presence of water
can prevent the catalyst poisoning and that presence of iron on
the catalyst does not seem to have significant effect on the
poisoning.
Conn and Brackin 32 suggest that similar competitive phenomena
occur between water, hydrogen sulfide, and other sulfur compounds
normally present or formed during catalytic cracking. Since the
concentrations of such harmful compounds are relativly high
in the case of dispersion steam, larger amounts of steam are
needed to protect the catalyst. The improvement of the poisoned
catalyst by intensive steam stripping appears to be a result of
the desorption of sulfur compounds from the catalyst; this has
been demonstrated in Figure 12.
It is not known whether the removal of these sulfur compounds
is due to a simple steam stripping action of hydrogen sulfide
56
-------
and other sulfur compounds from the catalyst or whether it is
brought about by reaction of the steam with the sulfur compounds
to form hydrogen sulfide or some other volatile compounds. Even
though the mechanism of steam stripping has not been fully
determined, there is evidence to show that the sulfur appears
in the stripper off-gas in the form of hydrogen sulfide .31
Some evidence also exists that at constant steam stripping rate
the rapidity of poisoning of natural catalyst and the difficulty
with which the poisoning is eliminated become greater with an
increase of sulfur content in the feed. Thus, it appears that
the net rate of sorption of the poisonous sulfur compounds onto
the catalyst structure is a function of the concentration of
these compounds in the atmosphere surrounding the catalyst, which
tends to cause sorption, and the concentration of steam in this
atmosphere tending to displace these compounds. As such, these
adsorption-desorption phenomena will primarily depend on tem-
perature, partial pressure of steam, and concentration of sulfur
compounds .3Z
Basically, there are not sufficient data to determine the effect
of this type of sulfur compounds on catalyst poisoning and coke
formation. A recently published article33 presents data on
sulfur distribution in the FCC process using the silica-alumina
and zeolite catalysts and suggests that the feed type is the
most important variable affecting sulfur distribution. Other
important factors the authors have observed which may become
useful for better understanding of sulfur distribution in FCC
units are mentioned below.
H2S formation during cracking is kinetically controlled. Even
though zeolite catalysts do display some desulfurization
activity, the desulfurization reactions occur at a much slower
57
-------
rate than do the cracking reactions. It appears that desulfur-
ization of thiophenic compounds proceeds at a slower rate than
for non-thiophenic sulfur compounds and that the H2S is formed
primarily from catalytic decomposition of non-thiophenic sulfur
compounds.
In describing the types of sulfur compounds found in FCC feed,
two terms are used to simplify the discussion: (1) Thiophenic
sulfur is the sulfur that is part of an aromatic ring, as in
thiophene. These compounds may be multi-ring compounds. (2)
Non-thiophenic sulfur is sulfur that is part of straight or
branched chain hydrocarbons, such as mercaptans or thioethers.
Exact distribution of sulfur to H2S, gasoline, cycle oil, and
coke varies with feed, catalyst type, conversion, and process
variables, with space velocity and feed type being most signifi-
cant. Due to the lack of experimental data, especially for newer
catalyst materials ,1*5 '*** and the rather broad range of variables
at which commercial FCC units operate,33 sulfur distribution
phenomena are not well understood.
Some evidence exists35 and some investigators believe36 that
sulfur is present in coke on spent catalyst as multi-ring thio-
phenic sulfur and that this type of aromatic ring structure will
not react with steam to form H2S. However, some other investiga-
tions conducted in this area37*38>125 have produced evidence
to show that a reaction between thiophene and its derivatives
with steam can occur between 350-500°C (662-932°F) over a catalyst
which, in one case, was identified as A1203. ' The following
reaction scheme can be assumed.
58
-------
H,0 ,
H S
2
500°C
5.3 APPLICATION OF STEAM STRIPPING TO EXISTING FCC UNITS
Two possible conceptual methods of applying steam stripping to
existing FCC units have been identified:
(1) An increase of the present steam stripping rates in
existing equipment to the levels needed to achieve an
adequate SO reduction in the regenerator flue gas.
fL
(2) The use of a secondary (add-on) stripping system. The
spent catalyst is removed from the existing equipment and
transferred to a secondary stripper. Sulfur-free catalyst
is then returned to the regenerator.
As explained previously, preliminary calculations based upon the
data collected by Conn and Brackin and Healy and Hertwig (Figure
12) indicate that the steam stripping rate required to achieve a
200 ppm S02 concentration in regenerator off-gas would be approx-
imately 2 to 5 Ib of steam per 100 Ib of catalyst. In our
59
-------
technical and economic evaluations we have assumed 4 Ib of steam
per 100 Ib of catalyst will be sufficient. It should be noted,
however, that the steam stripping rates obtained by Conn and
Brackin and Healy and Hertwig were determined in the late 19^0's
for natural catalysts used in cracking high sulfur feedstocks.
More recent technical developments in the area of catalysts as
well as more efficient equipment66 can have a significant
effect on determining the actual steam stripping rates. Even in
19^7-19^9, when the bulk of experimental work related to steam
stripping was implemented, it was observed that significantly
lower steam stripping rates are required in commercial applications
than those determined in pilot scale units.
The rate of steam stripping will also be a function of the sulfur
concentration in the feedstock processed by a refinery. This
feature can become very advantageous for units that presently
experience relatively low SOX emissions. A slightly increased
steam stripping rate may result in emissions reduced to the
levels desired.
From the facts just presented it can be reasonably expected that
much lower steam stripping rates will be required in practical
applications of steam stripping techniques to reduce sulfur oxide
emissions from regenerator flue gas. The true stripping rates
for each application will have to determined by experimentation.
Lacking more accurate data, we have applied the most conservative
assumptions in our technical and economic analysis of the steam
stripping concepts to produce the worst, most unfavorable con-
ditions possible. This fact should be very strongly considered
whenever the concept of steam stripping as presented in this
report is evaluated.
60
-------
5.3-1 Effect of an Increase of Steam Stripping Rate
Increasing the steam stripping rate would allow the stripping
steam and stripped sulfur compounds to pass through the existing
stripper, reactor, overhead vapor line and FCC fractlonator to
the fractionator overhead condenser where the steam is condensed
and water is removed from the system (see Figure 1). Performing
the SO emission control in this manner would force the sulfur
X
through the present facilities and remove it downstream in existing
sweetening equipment.
Typically, the rate of steam stripping (4 lb/100 Ib of catalyst)
assumed for the technique evaluation would double or triple the
flow rate in the reactor, vapor line, fractionator, and con-
denser. Since refineries are usually built with only a 10% to
15% flow capacity safety factor,27,28,29 the existing equipment
could not handle increased flow rates of this size. Thus major
modifications of present equipment would be required. The types
of modifications necessary are: increased stripper size, reactor
size, vapor lines, and adding a parallel train for fractionation
and condensation.
The bulk estimate for achieving such modifications is stated to
be 1/3 the price of a new fluid catcracker. (A complete unit
consists of reactor, regenerator, fractionators, and all other
supporting equipment.) The price of a complete FCC unit handling
the 25>000 barrel total feed per stream day is approximately
$18,000,000. Thus, use of this method of stream stripping
would require an outlay of about $6,000,000.
5.3.2 Secondary (Add-On) Separate Steam Stripper
The second emission control alternative mentioned above consists
of diverting the spent catalyst stream to a secondary, separate
61
-------
steam stripper. In performing the emission control in this
manner, there are several options available:
(1) Transfer the spent catalyst (using steam as a carrier) to
a second fluidized bed of catalyst where sufficient contact
between steam and catalyst is maintained. The vapors
leaving the catalyst stripper are condensed, and sour
water and H2S-rich gas are separated. The vapors leaving
the separator are sent to a Glaus unit for sulfur recovery
and the sour water is sent to a water treatment plant for
sulfur (H2S) recovery and separation of catalyst fines
from the effluent water. This option is analyzed in detail
later and is schematically presented in Figure 13.
(2) Use essentially the same flow scheme as in option 1
above except that a cyclone would be used instead of
the fluidized stripper. Using a stripping system of this
sort is potentially feasible if sufficient contact between
steam and catalyst can be obtained in the transfer line and
the cyclone.
(3) Place a second catalyst stripper, which is similar in
design to the existing stripper, between the regenerator
and the existing catalyst stripper. This stripper would
be designed to remove the stripper off-gas as a separate
stream so that the bulk of steam would not dilute the
products of FCC reactor. In a sense, this option consists
of increasing the size of the existing catalyst stripper
and operating it under conditions efficient for sulfur as
well as hydrocarbon removal.
62
-------
F C C PRODUCT TO FRACTIONATIOH
CT\
U)
F C C FEED
REGENERATED
CATALYST
EXISTING
CATALYST
STRIPPER
AIR
Figure 13- Block Diagram for a Conceptual Steam Stripping Facility
-------
(4) Increase the steam stripping rate in the existing stripper.
Dilution of FCC product stream can be avoided by proper
modification of the existing stripper similar to that de-
scribed in option 3, so that the excess steam can be removed
as a separate stream.
Of the four options presented, the first option underwent tech-
nical and economic analysis. This option is evidently the most
expensive of the four options because of the large increase in
catalyst inventory and catalyst handling requirements. If any
of the three other options can be applied, a more efficient and
simpler modification and operation will result, and more favor-
able economics can be expected. Additionally, the last three
options at the time of this study appeared to be rather con-
ceptual, and an insufficient technical basis could be established
for their full economic evaluations.
5.3.3 Application of Secondary Steam Stripping (Optional)
In attempting to assess the technical aspects of steam stripping,
the following processing scheme was assumed and presented
schematically in Figure 13-
Catalyst feed for the add-on secondary stripper is withdrawn con-
tinuously from the catalyst standpipe, which connects the
existing catalyst stripper to the regenerator. The entire
catalyst stream is diverted by steam through a transfer line to
a fluidized bed catalyst stripper. The catalyst, after stripping,
is recycled back to the FCC unit for regeneration.
Stripper off-gas primarily consisting of steam plus H2S and
carrying some entrained catalyst particles is sent to a condenser
for condensation. The H2S-rich vapor stream and the sour water
-------
from the condenser are separated in a phase separator. The
rich stream is sent to a Glaus unit for sulfur recovery while
the liquid stream is sent to wastewater treatment for H2S re-
covery and catalyst fines removal. The liquid stream may also
contain condensable hydrocarbons depending on the operating
efficiency of the primary catalyst stripper. However, the extent
of this potential problem cannot be properly evaluated due to a
lack of appropriate data.
The sour water treatment plant consists of an acidifier, a
neutralizer, a clarifier, and a vacuum filter. The acidifier
is fed with waste acid sludge produced in the refinery alkylation
operation. (Practically all refineries larger than 30,000 bpd
having an FCC unit also have alkylation units . ) Acidifying is
needed to lower the pH of the liquor and thereby decrease the
solubility of H2S. The H2S that is driven off in the acidifier
is also sent to a Glaus plant for sulfur recovery.
The acidic water is neutralized with calcium hydroxide before
being sent to a clarifier for settling of catalyst particles.
The neutralized clarified water can be either disposed of or
reused. The sludge leaving the clarifier is dewatered in a
vacuum filter and sent to landfill for disposal.
The application of secondary steam stripping for refinery FCC
regenerator SOV control requires some additional facilities. It
J\i
must be pointed out, however, that many modern refineries already
accommodate some or all of these facilities to handle the wastes
normally produced from petroleum refining. Thus, utilization or
expansion of these facilities to handle the additional load
resulting from steam stripping would have to be evaluated
separately for each specific refinery and application.
Our economic evaluations of the steam stripping technique were
carried out for three nominal capacities of FCC units: 10,000,
65
-------
45,000 and 150,000 bpsd. It was assumed that no facilities to
process waste streams resulting from this approach are available
except for the Glaus plant. It was also assumed that the avail-
able Glaus plant can handle increased loadings of hydrogen sulfide
without expansion.
In our first evaluation for a 45,000 bpsd FCC, it appears that the
cost of catalyst lost due to attrition in the secondary steam
stripper is a major part of the operating cost for this case and
will depend on attrition rates and catalyst/oil ratio. The
attrition rates were assumed to be 0.2 Ib of catalyst per barrel
of feed. This attrition rate is considered typical in existing
FCC operations. In other words, we have assumed that the catalyst
attrition will essentially double (compared to existing conditions)
if steam stripping is utilized.
Realizing that in FCC operation two major pieces of equipment,
the reactor and the regenerator, are involved and participate
in catalyst attrition, application of the attrition rate of 0.2
Ib of catalyst per barrel of feed to an additional piece of equip-
ment, the steam stripper, appeared to be rather pessimistic. Rates
amounting to about 50$ of this rate or 0.1 Ib/ barrel should be
more realistic.
The catalyst/oil ratio can vary between 4 and 20 Ib of catalyst/
Ib of oil. The recent tendency in refineries is towards lower
catalyst/oil ratios. Our cost estimates were prepared for both
0.1 and 0.2 Ib/barrel catalyst attrition rates, and catalyst/oil
ratios of 6 and 12, respectively, and are presented in Figures
14 and 15, as typical and worst cases. The detailed cost estimates
for all three sizes and both cases are summarized in Appendix E.
5.3.4 Control of Other Pollutants with Steam Stripping
As was shown earlier in Table 2, the regenerator off-gas contains
a number of atmospheric pollutants in addition to sulfur oxides.
66
-------
10 r
i/i
L_
E
ts
I
0.1
. . . I
. I
10
100
1000
FCC Capacity, thousand b/sd
Figure 14
FCC Catalyst Steam Stripping, Capital Investment Cost
(February 1973)
-------
100
CO
SlO
CT>
E
0)
Q.
o
Worst Case
Typical Case
I
. . I
10 100
FCC. Capacity, thousand b/sd
1000
Figure 15. FCC Catalyst Steam Stripping, Operating Cost
-------
Assessment of the effects of steam stripping on these additional
pollutants is premature at this time. It is conceivable that
steam stripping may reduce the amounts of nitrogen-containing
hydrocarbons carried to the regenerator, but the effect of this
process on nitrogen oxide emissions is presently unknown.
Obviously, carbon monoxide and some of the hydrocarbons can be
controlled by oxidation in the CO boiler, accompanied with heat
recovery. In the CO boiler the additional introduction of nitrogen
oxides as well as sulfur oxides can be expected, especially if
sulfur-free supplementary fuel is not used.
In the reduction or removal of particulate matter, the steam
stripping may play an important role if proper design modifications
are made. Particulate emissions are caused by catalyst particles
of the size that cannot be removed in existing cyclones in the
regenerator. If the catalyst fines which would not ordinarily
be removed in the regenerator can be removed before reaching
these cyclones, then FCC particulate emissions could be reduced.
The steam stripper and the cyclone in it can be designed to
permit these catalyst fines to be carried to the water treatment
facility downstream of the stripper and prevent their entrance
to the regenerator. This concept is obviously still theoretical
and will require practical verification. Its success will depend
on attrition rates or formation of catalyst fines in various
locations of future FCC units.
69
-------
6. REGENERATOR FLUE GAS DESULFURIZATION
The third approach alternative for reduction of sulfur oxide
emissions from FCC operations comprises processes capable of
removing sulfur oxides from the flue gas leaving the FCC re-
generator. These processes were evaluated as add-on systems
downstream from the cracking catalyst regenerator.
Inclusion of the add-on processes in our study was a result of
noting the apparent similarities in sulfur oxide concentrations
existing in FCC regenerator off-gas and in power plant flue gases,
for which these processes have primarily been developed or are
applicable. The Federal Government, as well as the industry
itself, have been committed to development of flue gas desulfur-
ization technology for several years. Consequently, the application
of power plant flue gas desulfurization technology to reduction of
sulfur oxide emissions from FCC units could not be overlooked.
The development of over one hundred flue gas desulfurization
processes has been actively continuing for several decades and
has been even further intensified in recent years by the in-
creasing effort by the Environmental Protection Agency to pro-
tect our environment by legislative action.
On the other hand, very few such processes have been considered
for refinery application.70*81 In order to include the most
recent developments in flue gas desulfurization we prepared a
comprehensive list of the existing processes as the basis for
further process screening and evaluation. The processes were
then analyzed and screened according to specific selection
criteria to obtain a manageable number of processes for further
detailed evaluation. Six final process candidates have been selected,
70
-------
6.1 SUMMARY OF CONCLUSIONS
(1) Three general categories of add-on gas desulfurization
systems have been selected from an Initial screening of
over 100 processes. The first, the adsorbent system
category, includes the Westvaco process, the Shell Flue
Gas Desulfurization process, and the Union Carbide PuraSiv-S
•
process. The second, the scrubbing system category, in-
cludes the Wellman-Lord sodium sulfite process and the
magnesium oxide slurry process. The third, the oxidation
system category, includes the Monsanto Company CAT-OX
catalytic oxidation process.
(2) The Westvaco, Shell, and Wellman-Lord processes are
presently the most feasible and most applicable add-on
systems for sulfur oxide emission reduction in FCC re-
generator flue gas.
(3) Based on available information, the Westvaco and Shell
processes have been rated equally feasible for FCC
regenerator off-gas desulfurization. Both processes are
at the point of commercial demonstration and would be fully
compatible with the petroleum refinery operations. The
Shell process has been developed by petroleum engineers
for refinery application and certainly should be con-
sidered fully in line with common design and operation
refinery practices. The flexibility of the Westvaco process
may appear very attractive for some refineries, especially
those not having any sulfur treatment facilities at the
present time. This process can produce three types of
sulfur products, allowing adjustment to market conditions.
71
-------
(4) Among the scrubbing systems, the Wellman-Lord process is
well developed for commercial use. The process is very
often applied to Glaus plant tail gas desulfurization in
refineries. However, general operational difficulties
associated with the wet scrubbing media, mist elimination,
and gas reheat, plus the negative effects of the presence
or formation of sulfur trioxide in these systems, place the
Wellman-Lord process below the adsorbent processes. Both
the Westvaco and the Shell processes remove sulfur in its
oxidized hexavalent form and should not have major draw-
backs due to the presence of sulfur trioxide.
(5) None of the add-on systems offers the simplicity, flex-
ibility, and general refinery applicability of steam con-
tacting of the cracking catalyst. Additional testing is
needed for the steam contacting technique, but this is
the case with add-on techniques too.
(6) The oxidation systems, represented by the CAT-OX catalytic
oxidation process, did not appear to be feasible for
refinery applications. The rather poor quality sulfur
product and potential catalyst plugging and poisoning
effects caused by cracking catalyst fines and carbonaceous
fractions, respectively, which are present in FCC regenerator
flue gas, have eliminated this process from further con-
sideration.
(7) On the basis of the conclusions given above, the fluid
catalytic cracking sulfur oxide emission reduction tech-
niques have been ranked as follows:
72
-------
1. Cracking catalyst steam contacting
2. The Westvaco and Shell sulfur oxide adsorbing processes
3. The Wellman-Lord sodium sulfite scrubbing process
6.2 CANDIDATE PROCESS SELECTION
Our initial tabulation of processes yielded a total of 110
potential candidates for desulfurization of flue gases. These
are summarized in Table 9- To reduce this number to a manageable
few for detailed evaluation, a set of criteria by which processes
could be eliminated from further consideration was established.
In order to select a reliable and relatively troublefree process
for refinery application, the processes in Table 9 were screened
four times. The last column of Table 9 shows in which screening
a process was eliminated.
The first screening was based on several factors. Processes were
eliminated from further consideration if they were unable to
reduce the S02 concentration in the effluent to the atmosphere
to less than 300 ppm. They were also eliminated if they were
obsolete, of limited success, abandoned, had corrosion problems,
had an unfavorable U.S. market, were proprietary, would not
license, had technical difficulties, had slow kinetics, were not
versatile, were of limited application, or if insufficient
information was available. Application of these criteria
eliminated 65 processes.
In the next level of screening, each of the remaining ^5 processes
were evaluated on specially prepared forms. An example of the form
used in this evaluation is presented in Appendix G. The form
provided space for listing all technical and economic information
73
-------
Table 9- FLUE GAS-DESULPURIZATION PROCESSES
Process
Name
Developer
Development Raw
To Date Materials
Effluent Eliminated
Concentration in
Byproducts ppm out/ppm In Remarks ScreenlnR
Flue Gas From Boilers
Alkalized
Alumina
Alkalized
Alumina
Amlne
Absorption
Ammonia
Scrubbing
Cal-Sox
Catalytic
Oxidation
Cat-Ox
CO Reduction
Caustic
Scrubbing
Char, Dry
Char, Wet
Char, Wet
Central Elec.
Generating
Board
(Britain)
USBM
Arthur D.
Little Inc.
USSR
Electrlclte
de France
Fulham
Mitsubishi
Showa-Denko
Wade Company
TVA
Hlxson
Poland
Monsanto Co.
Klyoura
Monsanto Co.
Univ. of
Massachusetts
Kureha
Ionics
General
Motors
Bergbau-
Forshung
Kansal
Relnluft
Westvaco
Hitachi
Lurgl
Bench Scale Reducing
Gas
Pilot Plant Reducing
Gas
Bench Scale
Pilot Plant Ammonia
Pilot Plant Ammonia
Pilot Plant Ammonia
Pilot Plant Ammonia
Unknown Ammonia
Pilot Plant Ammonia
Pilot Plant Ammonia
Pilot Plant Ammonia
(TVA)
Unknown Ammonia
Pilot Plant
Pilot Plant Ammonia
Commercial
Laboratory CO
Full Scale NaOH
Pilot Plant NaOH
Full Scale NaOH
Pilot Plant
Full Scale
Pilot Plant
Pilot Plant
Pilot Plant
Full Scale
(2 small
boilers)
H2S
H2S
S02
SO 2, CaSO,,
S, (NH,,)2SOU
(NH,,)2SO,,
(NH^SO,
(NH^jSO,. '
SO 2
SO 2
CaSO,,-CaS03
(NH,)2SOU
H2SOU(78J)
coz; s
Na2S03
M|J.,«>k(3.S)
Na2SOu or
CaSO.,
S02
S02
S02
Sulfur
H2SO,,(15X)
H2SOU(25«)
Discontinued,
high operat-
ing cost,
and high
capital
Abandoned
Abandoned
Corrosion
300/3000
ISO/3000 Abandoned
Corrosion
200/2000
Formation of
Carbonyl
Sulfide and
Hydrogen
Sulfide
Will Not
License
150-300/3000 Electro-
lytic
Regeneration
Carbon Attrition
Claim to
Eliminate NO
0-200/2000
300/3000
1
1
1
1
1
1
3
l
l
3
3
1
3
2
4
1
1
1
3
2
1
2
2
3
-------
Process
Name
H2S Reduction
Lead .Chamber
Process
Lead-Zinc
Ore Residues
Lignite-Ash
Still Process
Lime Inject-
ion, Dry
Lime Scrubbing
General
Lime Scrubbing
Wet
Limestone
Injection
Dry
Limestone
Injection
Dry
Limestone
Injection
Dry
Limestone
Injection
Dry
Limestone
Injection &
Scrubbing
Limestone
Slurry
Scrubbing
Development Raw
Developer To Date Materials
Peter Spence,
Ltd.
Princeton
Chemical
Res. Inc
Tyco
Bulgaria
Flrma Karl
Still (West
Germany)
Germany
Banco
(Sweden)
Bankslde
(England)
Bischoff
Mitsubishi
USSR
Howden 1C I
Process
(England)
Bergbau-
Forschung
(Germany)
Fuel
Research
Institute
(Czechoslovakia)
TVA
Central Re-
search Insti-
tute of Elec-
tric Power
Industries
(Japan)
Rutgers Univ.
Bischoff
(Germany))
Combustion
Engineering
TVA
Commonwealth
Edison
Detroit
Edison
Laboratory
Pilot Plant
Laboratory
Laboratory
Pilot Plant
Pilot Plant
Full Scale
Full Scale
Pilot Plant
Full Scale
H2SO, Plant
Pilot Plant
Full Scale
Laboratory
Full Scale
Steam
Boilers
Pilot Plant
Pilot Plant
(Early Stage)
Pilot Plant
Full Scale
Pilot Plant
Full Scale
Full Scale
H2S
Natural
Gas
Zn-Pb Ore
Residues
Lignite
Ca(OH)2
Ca(OH)2
CaC03
CaC03-
Ca(OH)2
Lime
Lime
Limestone
Limestone
Limestone
NaOH
Lime
Limestone
Limestone
Limestone
Effluent Eliminated
Concentration In
Byproducts ppm out/ppm in Remarks Screening
Sulfur
Sulfur
Nitric Acid
Sulfuric Acid
S02
S02
CaS03-CaSOu
CaSOj-CaSO,,
CaSOj-CaSO,,
CaS03-CaSOi,
CaS03-CaSO,j
CaS03-CaS04
CaSOu
CaS03-CaSOu
CaS03-CaSOM
H2S
CaS03
CaS03
CaSO]
CaS03
NOX
Interference
300/2000
Reaction
Rates Too
Slow - Plant
Shut Down
100/2000
300-600/3000 Limited
Success For
20 years.
Disposal
Process
200-1)00/2000
300/3000 High Cost
300/2000 Disposal
Problem
300/3000 Plant Shut Down,
Scaling Problem
Low Efflc.
Low Effic.
Low Efflc.
1400-200/2000 Expensive
Not Versatile
to Different
Boiler Types
Same as
Combustion
Engineering
Above
2
1
2
2
1
1
1
1
3
3
1
1
2
1
1
1
2
1
1
1
1
1
75
-------
Process
Name
Limestone
Slurry
Scrubbing
Magnesium
Oxide
Process
Manganese
Oxide
Metal Oxide
Slurry
Molten
Carbonate
Na2C03
Absorption
NOSOX
Potassium
Carbonate &
Molten
Potassium
Thlocyanate
Potassium
Formate
Potassium
Phosphate
Potassium
Sulflte
Red Hud
Sea Water
Scrubbing
Sodium
Carbonate
Sodium
Sulflte
Developer
TVA
Zurn Industries
APCO/Key West,
Florida
Procon-UOP
Chemical
Construction
Corp.
USSR
Mitsubishi
Grille
Atomics
International
Division
Preclpitalr
Pollution
Control
Monsanto Co.
Garret R&D
Company
Consolidation
Coal Co.
TVA
Wellman-Lord
Krupp
Mitsubishi
Kanagawa
Univ. of
Calif.
Bollden
Lummus
Wellman-Lora
Development
To Date
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Laboratory
Pilot Plant
Pilot Plant
Laboratory
Laboratory
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Pilot Plant
Laboratory
Pilot Plant
(full scale
to be
built)
None
Full Scale
(H2SOi, Plant)
Effluent
Raw Concentration
Materials Byproducts ppm out/ppm In
Limestone CaSOj
Limestone CaS03
Limestone CaS03
MgO SO 2
MgO SO 2 100-200/2000
MnO NH3 (NHl|)2SOlt 130/1300
MgO SO 2 300/3000
H2S
Na2CO, Na2SO, 600/2000
NaOH S02,Na2S04,
NaOH
S02
H2S, C02
Potassium H2S
Phosphate
H2SOj S02
Red Mud Sludge
From Al
Processing
Red Mud
Sea Water
Sea Water
Sea Water
Na2C03 or Sulfur
CaCO,
Na2S03 S02 200/2000
Eliminated
in
Remarks Screening
Key West
Fla., In
Operation
MnO Air
Pollution
Complex
System
Difficult to
Retro-fit
Need
Baghouse
No Further
Work Planned.
Limited Market
For Fertilizers
For Use On
Existing
Systems
Limited
Application,
Need Al
Source
Must be
Near Cheap
Source of
water
Combustion
Engineering
Claims
Feasibility
3
3
1
4
1
3
1
2
1
2
2
2
1
1
1
1
1
1
1
2
76
-------
Process
Name
Sodium Sulflte-
Zlnc Oxide
Solid Sorptlon,
Dry
Solid Sorptlon,
Uranium
Dioxide
Unknown
Processes
Smelters
Ammonium Sul-
flte, Sulfuric
Acid
Ammonium Sul-
fite, Thermal
Regeneration
Catalytic
Oxidation
Catalytic
Reduction with
Natural Gas
Contact Sul-
furic Process
Citrate Process
Dlmethylanlllne
Absorption
Developer
Johns tone
(patd)
U.S. Bureau
of Mines
Esso &
Babcock &
Wllcox
Houdry
Shell Intern
Research
Brookhaven
National
Laboratories
Universal Oil
Products
United
International
Research
Comlnco
Comlnco
SNPA-Topsoe
ASARCO
Texas Gulf
Sulfur
Allied Chem.
Canada, LTD
Well Estab-
lished, widely
Used
USBH
ASARCO
Effluent
Development Raw Concentration
To Date Materials Byproducts ppm out/ppm In
Pilot Plant Na2S03- CaSOu, SO, 150/3000
(19lO's) NaHS03
Pilot Plant H2, CH,, SO 2
Pilot Plant H2, CH,, SO 2
Pilot Plant H2 or CH,, Sulfur
or CO
Laboratory
Full Scale Sulfur
(Commonwealth
Edison)
Bench Scale H2SOi,
Commercial NH3, H20, S02 150/3000
H2SO,, SO,. (NH,,)2SO,,
NH3, H20 S02 150/3000
SO 1^ \ NHif ) 2^^i*
Commercial 9"X H2SOM 500-1000/17,000
Obsolete CH,, S
Pilot Plant CH,, S
Commercial CHU S
Commercial H20 H2SOU
Pilot Plant CH,,, Citric S 1200-2*00/23,600
Acid
Commercial S02
Eliminated
In
Remarks Screening
Removes
Plyash
50,000 ppm
Complete
Absorption
Used During
WW II
Prototype
of New
Processes
Details Not
Public
Not Suitable for
Low SO 2
Concentration
Developed 25
' Years Ago, 2
or 3 Units
Operating
1
1
1
1
2
1
1
3
l
1
1
1
1
1
2
1
Reduction with Imperial Chem. Pilot Plant
Coke Industries
Coke
Prior WW II
77
-------
Process Development Raw
Name Developer To Date Materials Byproducts
Reduction Bollden Co. Commercial Coke S
with Coke Prior WW II
Unknown ASARCO I Pilot Plant Reducing S
Process Phelps Gas
Dodge
Unknown Inspiration Pilot Plant H2SOU
Process Consolidated
Copper/Golden
Cycle Corp.
Sulfurlc Acid
Plants
Aluminum Hardman-Holden Commercial Aluminum CaSO,, ,
Su irate Ltd Sulfate S02
CaC03
Catalytic USSR Commercial H20 H2SO,,
Oxidation
Process
Catalytic Calgon Lab Scale H20 15J H2SO,,
Oxidation, Char
Solid Sorbent Rohm & Haas Lab Scale S02
Kraft Paper
Mills
Caustic Unknown Unknown White Liquor
Scrubbing Caustic
Other Sulfur Containing Gases-Refineries
Claus Unit Tail J. F. Pilot Plant
Gas, Catalytic Prltchard
IFP Process IFP Commercial H2S S
Beavon Sulfur Ralph M. Pilot Plant Fuel Gas H2S
Removal Parsons &
Process Union Oil
Co. of
Calif.
Direct Pan American Commercial H2S S
Oxidation Petroleum
Corporation
Hydrogen German Commercial H2S S
Sulflde, Sulfur
Dioxide Reaction-
Glaus Process
Scot Shell Commercial Fuel Gas H2S
Others
Aqueous MRC Laboratory Reducing S
Reduction Agent
Effluent Eliminated
Concentration In
ppm out/ppm In Remarks Screening
Corrosion
Scaling
<100/ Ion Exchange
Resin
Absorbent
1500-2500/ CS2 and COS
Not Removed
Removes
CS2, COS,
Requires
Additional
H2S Treat-
ment
>250/
Adlabatlc
1
1
1
1
1
2
2
1
1
1
1
1
1
1
2
78
-------
Process
Name
Double
Alkali
Molecular
Sieve
Developer
PMC
Envlrotech
CM
Zurn Air
Systems
Chemlco
A D. Little
Kureha Chemical
Industry Co.
Shows Denko
Union Carbide
Development Raw
To Date Materials
Pilot Plant Lime,
Soda Ash
CaO, Na2C03
Lime,
Soda Ash
Lime,
Na2C03
Na Salt,
Ca(OH),
Lime,
Soda Ash
Lime,
Soda Ash
Lime,
Soda Ash
Commercial
Effluent
Concentration
Byproducts pom out/pom In
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
CaSO,,
S02(U vol) 15-25/3500
Eliminated
In
Remarks Screening
3
3
3
3
3
3
3
3
i|
79
-------
on each process available to us at the time. This information
included control level achievable, state of development, location
of the control process in the regeneration off-gas flow train,
raw materials needed, products (wastes), marketability of products,
space requirements, problems foreseen to achieve retrofit,
economics, additional pollutants controlled, process reliability,
and finally, applicability of the process for catcracker S02
emission control.
The manner in which some of this information was used to further
narrow our search for the most feasible processes requires
additional discussion, especially with respect to product
marketability, process applicability, and reliability.
The product considered to have the best marketability at the
present time is sulfur. If desulfurization were fully applied
in all areas of industry, the amount of sulfur products produced
could change the market situation drastically. However, sulfur
products will probably always have some favorable properties for
affecting its marketability, e.g., easy handling, storage,
transportation, conversion to other sulfur products, etc.
Products such as S02 and H2S, although they are not readily
marketed, can be converted to sulfur or sulfuric acid. Con-
sequently, they were considered preferable to products such as
(NHil)2S04.
The process applicability criterion was applied to determine
whether the process can or cannot control the sulfur emissions
to the level desired. In the sense used, process applicability
included consideration of the integration of the process with a
petroleum refinery. Important factors considered included process
operating conditions and the ability and effectiveness in doing
its job.
80
-------
Finally, the process reliability criterion was used to determine
how well a process can be applied to refinery FCC unit SOV control,
J^
how well it compares with refinery operations, how easy it is to
control and operate when SOX concentration fluctuates, and how
well the process has been demonstrated in these areas.
The second-level screening was done by classifying the processes
according to their availability. Four categories were established:
advanced first generation, first generation, near first generation,
and second generation. Advanced first generation processes were
those that are commercially available. First generation processes
comprised those demonstrated in an advanced pilot plant stage.
Near first generation processes were considered to be those that
have been through advanced research and development and into the
pilot plant stage. Second generation processes were those that
are still in the bench experimental stage. When basically similar
processes were at different levels of development, the less
advanced ones were eliminated. During this screening, an
additional 19 processes were eliminated, leaving 26 potential
candidates. The remaining 26 have been classified according to
their availability in Table 10.
The third-level screening, which preceded detailed evaluation of
individual processes, was based on the form in which the sulfur
value is recovered. The possible products in order of decreasing
desirability are sulfur, H2S or S02, ammonium sulfate or sulfuric
acid, and a sludge for land disposal. Because of limited land
availability in and around most refineries, nonregenerable sludge-
producing sulfur removal processes are not desirable. Those
products for which there is a limited market, dilute H2S04 and
(NHi^zSO^, are lowest in desirability among the saleable by-products
H2S and S02 are more desirable since they can be used to produce
81
-------
Table 10. FLUE GAS DESULFURIZATION PROCESSES AFTER CLASSIFICATION
ACCORDING TO THEIR AVAILABILITY
Process Name
Advanced First Generation
Ammonium Sulfite
CAT-OX
Caustic Scrubbing
Char, wet
Molecular Sieve
Sodium Sulfite
First Generation
Ammonia Scrubbing
Ammonia Scrubbing
Double Alkali
Developer
Lime/Limestone
Scrubbing
Magnesium Oxide Scrubbing
Near First Generation
Ammonia Scrubbing
Calsox
Char, dry
Manganese Oxide
Consolidated Mining Co.
Monsanto
General Motors
Lurgi (Germany)
Union Carbide
Wellman-Lord
TVA
Hixson
Shell
A. D. Little
Chemico
Envirotech
FMC
GM
Kureha Chemical Co.
Zurn Air Systems
Commonwealth Edison
Bischoff
Mitsubishi
TVA
APCO/Key West, Fla.
Chemical Construction Co.
Mitsubishi
Monsanto
Westvaco
Mitsubishi
Product
SO 2
H2SO,, (78$)
Na2SO(, discharge
H2SO,, (25%)
SO 2
SO 2
S02;H2SOU;(NHU)2SO,,
SO 2
H2S
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
Sludge
SO 2
(NHU)2SO,,
Sludge
Sulfur
82
-------
either concentrated sulfuric acid or elemental sulfur. Similarly,
by-product sulfur is the most desirable product because of its
convenient form and its marketability as a raw material. Twenty
processes were eliminated because they produce the less desirable
by-products, leaving six candidates for detailed evaluation. The
six candidates are representative of three distinct categories:
scrubbing systems, adsorbing systems, and oxidation systems.
The three potential candidate adsorbent/absorbent systems are
processes by Westvaco (two alternatives), Shell, and Union
Carbide. The two scrubbing systems remaining are the Wellman-
Lord process and magnesium oxide scrubbing. The sixth process
is the Monsanto Company CAT-OX® process, which is based on
catalytic oxidation. The complete description of these processes,
including process economic evaluations, is presented in Section
6.4. The applicability of each of the remaining candidates to
catalytic cracking flue gas SOX cleanup is included in the dis-
cussion. Process availability, process description and chemistry,
applicability to catcracker off-gas, physical size requirements,
possible experimental work required, additional pollutants con-
trolled, and economics are also discussed.
The final, fourth-level screening was based on comparison of the
technical soundness and economics of the six processes. After
this comprehensive technical and economic evaluation, three
processes remained: the Westvaco, the Shell and the Wellman-Lord
processes.
6.3 GENERAL CONSIDERATIONS IN EVALUATION OF THE SELECTED PROCESSES
6.3.1 Technical Evaluations
Most of the data for the control processes selected have been
obtained and are based upon burning 3-0-3-5% sulfur coal in a
83
-------
power generation unit. This concentration corresponds to an S02
flue gas concentration of 2000-2500 ppm. In our study we have
assumed that the regenerator off-gas sulfur concentration will be
2000 ppm. This concentration, of course, can be diluted by up
to 30% due to additional volume introduced from combustion in a
CO boiler, assuming that no additional sulfur oxides will result
from the combustion of CO boiler supplemental fuel.
There are several operating schemes possible downstream from the
FCC regenerator. These can lead to alternatives for integration
of an add-on desulfurization system with the existing equipment.
The off-gases leaving the regenerator contain a reducing atmos-
phere with high concentrations of carbon monoxide and a very low
oxygen concentration. Additionally, the gases carry catalyst
particulates. The heat of the carbon monoxide oxidation reaction
can be recovered through its controlled combustion in the CO
boiler.
Obviously, several possibilities appear for integration of a
specific add-on process with existing refinery equipment. The
reducing atmosphere can become very favorable for some of the
processes, especially those in which extensive oxidation
of sulfur dioxide to sulfur trioxide is undesirable. This
oxidation occurring in an oxidizing atmosphere often results in
arduous operating problems, increased consumption of chemical
reactants, more difficult if not impossible regeneration, and
increased operating cost. Scrubber systems would fall into this
category of processes. On the other hand, for some processes an
oxidizing atmosphere and presence of oxygen is essential for
successful process operations. Alternative process locations
for each of the six processes selected are proposed and evaluated
in appropriate sections discussing each process.
8*1
-------
In some applications of fluid catalytic cracking the regenerator
flue gas contains a relatively high fraction of sulfur trioxide
compared to the total concentration of sulfur oxides present in
the stream. Variations of the S03-to-S02 ratio have been
reported, with the percentage of sulfur oxides as S03 ranging
from 10 to 6058-2»3»92 Some investigators feel that the S03
is produced by the oxidation of S02 in the flue gas at elevated
temperature, catalyzed by metals (vanadium, nickel, iron) de-
posited on the cracking catalyst. Equilibrium calculations,
however, show that only 29% of S02 can be converted to S03 in a
stream at 1200°F containing 1000 ppm S02 and 0.75? oxygen.92 If
the oxygen level is 20%, the amount of S02 converted will be 6558.
The metals suspected to catalyze the oxidation of sulfur dioxide
originate in the FCC feed and are deposited on the catalyst in
relatively high concentrations without poisoning the catalyst.
The presence of large amounts of S03 in the treated gases may
have a detrimental effect on the operation of many of the available
flue gas desulfurization processes with results similar to those
caused by the presence of an oxidizing atmosphere, which has been
already discussed.
As was indicated previously, the gases from the regenerator also
contain catalyst fines. Chemically, the catalyst composition is
very different from that of fly ash emitted from stationary
sources. The effect of catalyst fines on the add-on system
operation has not been previously determined and should be fully
investigated if any of the add-on processes are applied to de-
sulfurization of FCC regenerator flue gas. Obviously, if the
process is applied after the electrostatic precipitator, the
effects of catalyst fines may be minimized. Even if the potential
change of catalyst fines properties due to their passing through
the CO boiler is neglected, the effect of the fines on an add-on
85
-------
system may be different in a reducing and oxidizing atmosphere.
It should be realized that the effects of catalyst fines on a
specific process may vary with the type of crude oil processed
and type of catalyst used by a specific refinery.
Effects of potential by-products within a process ought not to be
overlooked either. In addition to the chemical effects of the
catalyst particulates on a specific add-on process, some physical
effects may also result. The catalyst fines may cause plugging
problems and consequently some operating difficulties. They can
accumulate in process equipment and streams requiring their
occasional removal, increased maintenance cost, and in scrubber
systems, increased consumption of scrubbing media. Scrubber
systems, on the other hand, could become a means of very efficient
particulate emission control and possibly replace electrostatic
precipitators.
The regenerator effluent gases leave the regenerator at much
higher temperatures than those usually existing in power plants.
Depending on the specific process operating temperature, this
may require more or less gas cooling before the gas enters the
actual process, and will have some effect on process economics.
The flue gas leaving the CO boiler, however, is much more within
the range of add-on process operating temperatures. Additionally,
scrubber systems will require a reheat of the processed gases to
reduce steam plume formation after the stream is discharged to
the atmosphere.
The effect of the variation of flue gas composition on the
control process chemistry and unit operations requires additional
laboratory investigation. Also, the effects of the high level of
S03, the limited level of 02, the presence of high CO content, and
86
-------
presence of other compounds as indicated in Table 7 would have
to be experimentally determined for each of the processes.
6-3-2 Economic Evaluation
The general assumptions required for the economic analyses of
the add-on processes have been presented in Section 3.5.
Additional assumptions required for a specific process are given
with the discussion of each process. Most of the economic
evaluations for add-on systems have been correlated with megawatt
output of a power plant. In order to apply these data to our
study, a need appeared to convert the power plant megawatt output
to volume of the flue gas treated by a specific add-on process.
This conversion has been assumed to be 2000 scfm/MW.
All of the add-on processes selected [except Monsanto Company's
CAT-OX and the Westvaco process (second alternative)] produce a
concentrated S02 stream. This stream, it was assumed, will be
fed to the Glaus plant. For estimation of the additional cost
credited to the gas treatment in the Glaus plant, the twenty-
third assumption in Section 3.5-2 was applied. The results
obtained from applying this assumption are included in our
detailed cost estimates and they were included in the total
operating cost estimates.
Flue gas desulfurization costs are affected principally by
capacity, S02 content, level of desulfurization, and availability
of required outside services. Different S02 concentrations
would change the economics to some extent. We have arbitrarily
set the regenerator flue gas S02 level at 2000 ppm with a 90%
reduction by the add-on systems (i.e., to 200 ppm). These
criteria will affect the operating costs with regard to raw
material requirements, but since no credit has been taken for
87
-------
sulfur recovered, the effect of S02 level on the operating cost
is minimized. Since the capital equipment is based on volume
treated and not sulfur removed, the required capital investment
for the removal unit will also be minimally affected by S02
effluent concentration. However, lower concentrations of S02
will require smaller regeneration units and consequently lower
capital investment .
The values obtained for capital investment were modified where
necessary to fit the application to catcracking flue gases.
Frequently, heat exchange equipment was required to tailor the
flue gas to the SO removal process . In all cases the most
X
recent information available was utilized and where necessary
updated using the CE index. If values for this index did not
accompany the data, a value for the year in which the publication
appeared was assumed.
The operating cost for the add-on systems has been expressed in
terms of mills/cu ft of treated flue gas . To compare the add-on
systems to the other two alternatives, the values for operating
cost of the add-on systems must be expressed in cents per barrel
of catcracker feed. To obtain these values, the catcracker opera-
tions were assumed to yield 3 scfm of flue gas per barrel per day
of fresh feed. The corresponding cost for a different flue gas
rate can be obtained by inserting the value acquired from the
operating cost graphs of each process in the following equation:
Cost (cents per barrel) = Cost (mills/cu ft) x 144 x p (2)
6.4 ADSORBENT /ABSORBENT SYSTEMS
Three add-on processes remained in this category and have been
evaluated. Each of the processes utilizes a solid acceptor that
-------
collects the S02. The acceptor is subsequently regenerated and
reused. The product of all three processes is a concentrated
stream of S02, although the mechanism for obtaining the stream
differs for each process. The concentrated sulfur dioxide
stream is assumed to be fed to the Glaus plant.
The Westvaco process adsorbs S02 on a carbon acceptor as sulfuric
acid, which is subsequently converted to S02 with a reducing gas.
This process can be modified to directly produce sulfur or to
wash the carbon acceptor and produce dilute sulfuric acid.
Since sulfur is the product considered most desirable, the
process has been evaluated in two alternatives, the first pro-
ducing S02 and the second producing sulfur.
The Shell Flue Gas Desulfurization (SFGD) process absorbs the
S02 as CuSOi! which is later reduced to Cu and S02 with hydrogen.
The Union Carbide PuraSiv-S process adsorbs the S02 on molecular
sieve and recovers the S02 thermally.
6.4.1 Westvaco Process
Both alternatives of this process have been developed by Westvaco
Corporation, Research Center, N. Charleston, South Carolina.
Westvaco has been developing its S02 recovery process in both
bench scale and fluid bed pilot equipment. Initial feasibility
tests were conducted in small fixed bed reactors to determine
the most effective carbon for S02 removal and to evolve the
regeneration sequence for minimum carbon loss. Continuous
operation of the various process steps was evaluated under
simulated conditions in fluidized-bed pilot equipment.
89
-------
Removal of S02 from actual boiler flue gas was first tested In a
1,500 cfh, 6-inch diameter unit operating on a slipstream from a
50 MW oil-fired boiler. This equipment was operated satisfactorily
for continuous periods of up to one week. Scaled-up 18-inch
diameter adsorbers, capable of handling gas rates of 20,000 cfh,
also have been operated satisfactorily on flue gas from an oil-
fired boiler and on simulated Glaus tail gas. Data from these
continuous units showed S02 removal capabilities down to the 50
ppm range.
6.4.1.1 SO? Production Process Description
A simplified block flowsheet of the process for treatment of FCC
regenerator waste gas is shown in Figure 16. The FCC waste gas
is cooled, oxygen is added and the gas is contacted with activated
carbon in a S02 sorber. In the sorber activated carbon catalyzes
the conversion of S02 to sulfuric acid by the reaction:
S02 + 1/2 02 + H20 >- HzSOit (3)
The conversion occurs within the pores of the activated carbon
and the acid formed remains sorbed within the activated carbon.
Recoveries of S02 greater than 90/8 are readily achieved from
gases with concentrations similar to those emitted from FCC
regenerator.
Acid laden carbon flows continuously from the sorber and passes
to the regenerator where it is contacted with H2S from the
refinery to form S02 by the following reactions (4-a) and (4-b)
These two reactions can be combined and yield reaction (5).
90
-------
WASTE GAS FROM FCC
REGENERATOR
300-2000 PPM S02
REGENERATING GAS
FROM REFINERY
-85 % HZS
ACTIVE
CARBON
RECYCLE
MAKE-UP ACTIVE CARBON -
REACTIVATING GAS
FROM REFINERY
STEAM. HYDROGEN
CLEAN GAS TO CO BOILER
OR FLARE
90-160 PPM S02
SULFUR
DIOXIDE
SORBER
S02
f
°2 + H,0
* "2
CARBON
160-300 *F
H2S04
Z 4
REGENERATOR
H2S04 -h H2S
_^ PRODUCT GAS TO REFINERY
CLAUS UNIT, 30-80%S02
REACTIVATOR
OFF-GAS TO FLARE
STEAM.HYDROGEN
STEAM-HYDROGEN TREATMENT
Figure 16.
Westvaco Process - FCC Regenerator Waste
Gas Treatment, S02 Production
91
-------
1/9 H2SOn + 1/3 H2S »• 4/9 S + 4/9 H20 (4-a)
8/9 HgSO,, + 4/9 S »• 4/3 S02 + 8/9 H20 (4-b)
HjjSO,, + 1/3 H2S »• 4/3 S02 + 4/3 H20 (5)
Off-gas from the regenerator will contain 30-35% S02 with an H2S
feed of Q5%• If more concentrated S02 is required as product, a
condenser can be added to remove water vapor and increase the
concentration to 80-90% S02.
After being regenerated the carbon stream is split with 80%
being sent back to the S02 sorber and 20% to a reactivator. In
the reactivator the carbon is treated with steam and hydrogen to
remove any buildup of chemisorbed oxygen or sulfur which lowers
the carbon's activity. Off-gas from the reactivator is sent to
a flare or the CO boiler.
The conceptual design flowsheet for FCC application proposed by
the Westvaco is shown in Figure 17, and includes stream com-
positions and energy balance of the process.
The FCC regenerator waste gas, stream 11, is received from the refinery
at 1100°F at a rate of 30,000 scfm. A part of the gas is used in
the shell side of the Westvaco regenerator to supply process heat
and the remainder is cooled to 500°F in a waste heater boiler
(C-101) which generates 20,000 Ib/hr of 15 psig steam, 22, for
export. The waste gas, cooled to 500°F, is mixed with air, 21,
to raise the oxygen concentration to 3-3%, 3, and then contacted
with activated carbon in a 5-stage, fluid bed S02 sorber (E-101).
In the S02 sorber, which is 15.6 in. dia. x 30 ft long, the S02
is converted to sulfuric acid, which remains adsorbed on the
activated carbon. Water, 19, sprayed into the fluidized activated
carbon evaporates to maintain the desired temperature in the
92
-------
VO
U)
®
"~mss
Figure 17. Westvaco Process - FCC Regenerator Waste Gas
Treatment, Conceptual Design Flowsheet
-------
reactor. The reactor bottom stage is kept at 300°F, below the S03
dew point, until S03 is removed. In the upper stages the tem-
perature is lowered to l60-200°F for efficient S02 removal. After
90/8 SC>2 removal, carbon carry-over is removed in a cyclone (D-101)
and returned to the SOz sorber. Pressure drop requirements are
supplied by a blower (J-101) downstream of the sorber and the clean
gas, 6, is sent to a flare or CO boiler depending on the refinery.
The acid-laden carbon from the 862 sorber is transported to the
5 ft dia. x 36 ft moving bed regenerator (E-201) by a conveyor
and bucket elevator. Here, the carbon is contacted co-currently
in 120 four-inch reaction tubes with HaS-laden gas supplied from
the refinery, 2. The reaction of H£S and sorbed acid produce a
stream of 32? S02, 8, which is returned to the refinery Glaus plant.
Heat to raise the carbon temperature from 250 to 600°P and to supply
the heat of reaction is obtained by indirect heat exchange with a
portion of the FCC regenerator flue gas.
Upon leaving the regenerator the carbon stream, 9, is split, 80%
is returned directly to the SC-2 sorber and the remaining 20% is
treated with a hydrogen-steam mixture, 12, to remove trace sub-
stances which might deactivate the carbon. The reactivator
(E-202) is a 2.5 ft dia. x 8 ft long single-stage fluid bed unit
operating at 1000°F. Both regenerated and reactivated carbons,
**, are reintroduced directly into the top stage of the S02 sorber.
The reactivator off-gas, 14, is sent to a flare or CO boiler.
Carbon makeup, 20, is supplied at a rate of 11.5 Ib/hr from an
8 ft diameter storage vessel capable of holding about 15 tons of
carbon.
-------
Raw Materials and Chemicals
1. Process water for waste heat recovery
2. Activated carbon
3- Hydrogen
*J . Hydrogen sulfide
5. Fuel gas
Physical Size Requirements (Figure 18)
30,000 scfm unit
38 ft x 38 ft x 110 ft
150,000 scfm unit
85 ft x 85 ft x 110 ft
Location in Regenerator Downstream Flow Train
Immediately after regenerator
Regenerator
Westvaco
Control
Process
i
CO Boiler
Electrostatic
Precipitator
r
Stack
— i
6.4.1.2 Sulfur Production - Process Description
Conceptual Design
A simplified process flowsheet of the Westvaco process in which
sulfur is produced as by-product is shown in Figure 19. Plot
areas would be similar to those presented in Figure 18.
95
-------
-120
CONVEVOR
0\
•100
REGENERATOR
•BO
• 60
-40
SCALE, FT
CYCLONE
ADSORBER
Figure 18,
Westvaco Process - FCC Regenerator Waste Gas
Treatment, Elevation and Plot Plan
-------
FCCR HASTE GAS
300-2.000 PPM S02
1000'F
ACTIVE CARBON
HAKE-IP
HYDROGEN FROM
REFI'IERY
CARBON RECYCLE
ACTIVE
CARBON
* 300'F
CLEAN GAS TO CO BOILER OR FLARE
30 - 160 PPS S02
200'F
S02 + 1/2 02 *
M2SUH
3 H2S + H2SOi| —- 1 S +1 H70
»- SULFUR PRODUCT TO STORAGE
270'F
3 H2 + 'I S —- 3 H2S + St
•- TO STACK
Figure 19-
Westvaco Process - FCC Regenerator Waste Gas
Treatment, Sulfur Production
97
-------
Waste gas from the FCC regenerator unit at 1000°P is cooled in a
waste heat boiler to 500°P prior to S02 removal. Process steam
(150 psig) produced in the boiler is exported to the refinery.
Air (not shown) is added to the cooled FCC regenerator waste gas
and the gas is contacted with activated carbon in a 15-5 ft
diameter, 5-stage fluid bed for S02 removal. Greater than 905?
of inlet S02 is converted to t^SOi, and sorbed on the activated
carbon according to the following reaction:
S02 + 1/2 02 + H20
Acid laden carbon from the S02 sorber is then contacted with H2S
in a 6 ft dia. x 25 ft moving bed reactor containing 200 four-
inch diameter reaction tubes. In the upper half of the reactor
(Sulfur Generator) acid is converted to sulfur at 300°F by the
reaction,
3 H2S + HjSO,, - »• 4 S + l\ H20. (7)
Normally} in this reaction conversion of acid to sulfur approaches
99$ and H2S utilization is greater than 95%. In order to insure
that no H2S leaves the process, the off-gas from the sulfur
generator is burned in an incinerator and the incinerated gas
is added to the FCC regenerator waste gases entering the S02
sorber. The lower part of the sulfur generator serves to preheat
the carbon to 880°F by indirect heat exchange with a small part
of the FCC regenerator waste gas prior to sulfur recovery.
The hot, sulfur-laden carbon is then contacted with 30? hydrogen
from the refinery in a 4.25 ft diameter, 8-stage fluid bed unit.
Three-fourths of the sulfur is converted to H2S and one-fourth
is vaporized by the hot gases, i.e.
3 H2 + 1 S - »• 3 H2S + St . (8)
98
-------
The sulfur product is recovered in a conventional condenser such
as is used in a Glaus plant and the H2S gas stream proceeds to
the sulfur generator. Thus the H2S forms an internal recycle
loop within the process. If H2S is available at the refinery,
this step is not required and process costs are reduced.
Raw Materials and Chemicals
1. Process water
2. Activated carbon
3. Hydrogen
4. Fuel gas
Physical Size Requirements
30,000 scfm unit
38 ft x 38 ft x 110 ft
150,000 scfm unit
85 ft x 85 ft x 110 ft
Location in Regenerator Downstream Flow Train
Immediately after regenerator
Regenerator
Westvaco
Control
Process
i *
i
i
i
i
*"
CO Boiler
Electrostatic
Precipitator
»-
Electrostatic
Precipitator
CO Boiler
A"
"""
Stack
99
-------
6.4.1.3 Experimentation Needed and Proposed by Westvaco
a. Current Status
Development of the Westvaco process has reached the
20,000 cfh pilot level. Emphasis to date has been
upon removal of S02 from power boiler flue gases
with the production of sulfur. With support from
EPA all of the process steps have been successfully
tested separately at the 20,000 cfh level. Mechan-
ical integration of the pilot plant is Just being
completed preparatory to long-term cycling tests on
flue gas from an oil-fired boiler. The pilot plant
is well instrumented for data acquisition to eval-
uate process operation.
b. Development for FCC Regenerator Waste Gas
The Westvaco process seems well suited for application
to FCC regenerator waste gases. The gas composition is
similar to streams on which Westvaco extensive SC>2 re-
moval work has been done on the laboratory and 20,000 cfh
pilot levels. In addition, availability of H2S and hydrogen
at the refinery for the production of sulfur or sulfur
dioxide greatly simplifies the process.
Westvaco feels that the process could be readily scaled
from the current pilot scale to 30,000 cfm FCC re-
generator waste gas treatment. Only minor modifications
would be needed to current pilot equipment to evaluate
the FCC regenerator waste gas treatment application.
100
-------
The schedule for the 30,000 cfm process demonstration
program proposed by Westvaco is shown in Figure 20. The
proposed 16 month schedule assumes that rapid development
is needed and that the initial three phases,
1) pilot evaluation of FCC regenerator waste gas
treatment,
2) preparation of process design specifications for
30,000 cfm demonstration, and
3) preliminary design of 30,000 cfm demonstration
unit,
would be in progress simultaneously. Presently, sufficient
pilot information is available to begin preparation of
process specifications and to prepare preliminary
designs. Evaluation of FCC regenerator waste gas
treatment in the current pilot plant would be used to
confirm process applicability and firm up the final design
data.
6.4.1.4 Economics
Capital cost requirements and operating expenses are based on an
evaluation of data provided by Westvaco.91* Values of required
capital investment and operating cost obtained for application
to FCC off-gases appear graphically in Figures 21 and 22. In-
cluded in these figures are values for both S02 production and
direct sulfur production. Detailed operating cost estimates
appear in Appendix F, Tables Fl, F2, F3, P4, F5, and F6.
Additional assumptions used in cost estimates for this process
include:
101
-------
Months
7 8 9 10 11 12 D 14 15 16
20, 000 cfh PILOT UNIT
Modify present unit and operate
at FCC regeneration conditions
Operate as required for demonstration
unit backup information
30, 000 cfm DEMONSTRATION UNIT
Program preparation
Process design
Equipment Specification
Design & construction liaison
Mechanical design
Procurement
Delivery
rnn^trnrtinn
Equipment erection & start-up
Operator training & equip, check-out
Operation for process analysis
^
^
=
!••
-
• ••
-
<»,,
• ••
•»
• •1
...
...
• •1
!••
...
-
J
•
• •1
—
—
o
ro
Figure 20.
Westvaco Process - FCC Regenerator Waste Gas Treatment, Pilot
Plant Program Schedule
-------
10
O)
10'
104
Sulfur Production
\
I I I I I
105
Capacity, scfm
10°
Figure 21. Westvaco Capital Investment Cost
(February 1973)
103
-------
1.0
3 "-1
an
OJ
a.
O
S02 Production
0.01
Sulfur Production
104
105
Capacity, scfm
10°
Figure 22. Westvaco Operating Cost
-------
Carbon cost $0.25/lb
H2S cost $20/ton of sulfur
Carbon lifetime - 1/2 year
Operating labor - 1 man/shift
6.4.1.5 Comments
The Westvaco process seems to be a very attractive option for
the refinery FCC regenerator off-gas desulfurization. The
company is actively working on the process, has previously con-
sidered the FCC application, and has not found any major tech-
nical drawbacks. The process includes basic unit operations
that are common in chemical industry and consequently would not
introduce anything new or "unfamiliar" to the petroleum refiners.
The fact that the process can produce three of the sulfur products
(S02, S, or HgSO^) would offer refineries advantageous flexibility
in integration of the process with existing facilities as well as
future expansion plans. No additional materials that are not
already handled by the refineries are required for the process
operation.
The process will probably operate well when sulfur concentration
fluctuates as well as in the presence of sulfur trioxide. How-
ever, process operation and reliability under existing regenerator
off-gas conditions will have to be fully tested. Specifically,
the process operation in a reducing atmosphere would have to be
determined and evaluated. Although the presence of particulates
is not expected to cause any serious problems, their effects
should also be tested. The process could be applied to gases
leaving the CO boiler as well. However, higher capital and
operating cost may be expected due to the regenerator off-gas
dilution in the CO boiler.
105
-------
6.4.2 Shell Flue Gas Desulfurization Process
This process has been developed by the Shell International Re-
search, Netherlands, and is presently licensed by Universal Oil
Products, Des Plaines, Illinois.
A 300-600 scfra pilot plant began operation in 196? at the
Pernis refinery in the Netherlands. The first full-scale unit
is being constructed at the Showa Yokkaichi Sekiyu refinery in
Japan, with scheduled start-up this year (1973).
6.4.2.1 Process Description
The Shell Flue Gas Desulfurization (SFGD) process uses a dry
acceptor in a static packed bed to accept S02 from gaseous
streams. The acceptor mass, contained in two or more identical
reactors, is subjected to successive stages of acceptance and
regeneration at approximately the same temperature. The net
effect is that S02, free of oxygen and particulates, is obtained
concomitant with the required degree of gas desulfurization.
Figure 23 is a simplified illustration of the SFGD process equip-
ment arrangement. The S02-rich gas passes through the acceptance
reactor(s) for typically 45-60 minutes, until the cumulative slip
(breakthrough) of S02 into the treated gas has reached a designated
S02 concentration. The S02-rich gas stream is then switched to a
reactor containing regenerated acceptor, and the loaded acceptor
is regenerated. Gas from the purging of the reactors between
acceptance and regeneration is treated in the accepting reactor.
106
-------
Treated Flue Gas
o
•si
1
1
f*
. \
Reducing
i Gat
1 i
->
i
i
r
i
<;
'
Timing Device
—
-------
In the "parallel passage" design of SFGD reactor internals, the
acceptor material is contained in a series of packages arranged
in a parallel configuration. The acceptor is contained between
layers of wire gauze with spaces provided between acceptor
packages for gas flow. In this way the gas flows along the
surface of the acceptor packages and not through the acceptor
material. This prevents pressure-drop buildup due to the
deposition of particulate material present in many of the flue
gases. For convenience in fabrication and handling, a number of
layers of acceptor packages, appropriately spaced, are placed
in a container to form a unit cell or module. The required
number of these modules can then be placed in a suitably sized
reactor vessel, according to throughput and S02 removal re-
quirements.
In order to convert the cyclic regenerator reactor gas into a
relatively constant flow of concentrated S02, the off-gas from
the reactor regeneration cycle is charged through a cooler/con-
denser to an absorber/stripper system. Water (or a solvent having
suitable absorption capacity for 802) is used to absorb SC>2 from
the regeneration gas; surge capacity is provided on the absorption
liquid to smooth fluctuations in the S02 rate.
CuO is outstanding as an acceptor in this application in that it
readily forms sulfate with S02 in the presence of oxygen at,
ideally, about 700-750°F, and also can be satisfactorily re-
generated with reducing gas to yield concentrated S02 at about
the same temperature. The stability of the acceptor in cyclic
operation has been demonstrated in a pilot plant in over 20,000
hours of operation.
108
-------
The basic reaction for acceptance of S02 from flue gas is as
follows :
S02 + 1/2 02 + CuO - >- CuSd, (9)
Copper sulfate releases the bulk of the accepted sulfur in the
form of S02 upon regeneration with hydrogen-rich gas in accordance
with the equation:
+ 2 H2 - > Cu + S02 + 2 H20 (10)
Unconverted CuO is reduced to copper.
The following side reaction occurs to a minor extent, but is not
significant when current regeneration procedures are followed:
CuSOn + 3 H2 - »• 1/2 Cu2S + 1/2 S02 + 3 H20 (11)
Since regeneration reduces CuSOi,. to Cu, the following reaction
must take place before the basic acceptance reaction can occur:
Cu + 1/2 02 - >• CuO (12)
The following reaction occurs simultaneously, affecting the
minor amount of Cu2S formed during regeneration:
Cu2S + 2 1/2 02 - »• CuO + CuS04 (13)
109
-------
The small amount of copper reacting in accordance with Eq. 11
undergoes the reaction of Eq. 13, of course, upon reintroduction
of flue gas. The small extent to which this reaction occurs
reflects the exceptional behavior of copper in this service and
continuing improvement of regeneration techniques .
The reaction of Eq. 10 proceeds almost completely, with a sharp
reaction front, when using hydrogen (or CO) as the regeneration
agent. S02 evolution proceeds until the end of the regeneration
period with no H2 slip, and falls off sharply as regeneration gas
slips through.
The very fast reaction of Eq. 12 proceeds with a sharp reaction
front upon reintroduction of flue gas assuring that S02 in the flue
gas will find oxidized copper with insignificant slip as long as
a specified minimum oxygen content of the flue gas is maintained.
The considerable heat release resulting from reactions of Eqs. 12
and 13 results in a temperature peak, which quickly travels through
the reactor. To avoid significant initial slip of SC>2 from thermal
decomposition and to protect the acceptor, the operating conditions
and mechanical design of the reactor must be considered carefully.
A mathematical model has been developed that reflects the kinetics
of the acceptance/regeneration reactions and operating experience.
S02 is accepted according to the reaction described by Eq. 9 with
the reaction front proceeding through the acceptor bed until the
remaining part of unloaded or partially-loaded acceptor at the
exit end of the reactor has become too small to ensure complete
S02 removal. At this point, S02 will start to slip through into
the treated gas. The reactor is switched to the regeneration
mode when the cumulative slip has reached a specified exit
concentration.
110
-------
Physical Size Requirements
Area requirements for the SFGD process have been estimated to
be 37,500 - 52,500 sq ft for a 150,000 bpsd catcracking unit.27,95
Location in Regenerator Downstream Flow Train
After CO boiler
Regenerator
—
CO Boiler
Electrostatic
Precipitator
Electrostatic
Precipitator
CO Boiler
A'
i
SFGD
Control
Process
6.4.2.2 Economics
The capital cost and operating cost estimate, Figures 24 and 25,
are based on data in the literature and contacts with Universal
Oil Products. A complete operating cost summary appears in
Tables F7, F8, and F9, Appendix F. Additional assumptions
applied in the cost estimates include:
Steam cost - 60.2
-------
10
,8.
•w-
*-" 7
s 10'
10'
i i
10'
Capacity, scfm
10"
Figure 24. SPGD Capital Investment Cost (February 1973)
112
-------
1.0
30.1
E
o>
Q.
o
0.01
i i i
10'
105
Capacity, scfm
Figure 25. SFGD Operating Cost
113
-------
atmosphere for the process operation. Ideally, however, the
process would fit better into streams before the CO boiler. The
high regenerator off-gas temperature at this point would avoid a
significant expense for the gas reheat that is necessary if the
process is inserted after the CO boiler. Some CO boilers,
however, operate at temperatures higher than 500°F (see Table 7)
and this problem would not arise. On the other hand, depending
on the operation of a specific FCC regenerator, the off-gas
leaving the regenerator can contain various concentrations of
oxygen that may or may not be sufficient for the proper operation
of the Shell process. The presence of high concentrations of
carbon monoxide in this stream may, however, eliminate these
oxidation effects. Additional experimentation to define the
SFGD process operation under these conditions is required.
If the experimental work were to indicate that the effects of
oxygen concentration prevail above those of carbon monoxide
during the S02 accepting cycle (although this is highly improbable),
an attractive modification of the process operation becomes
apparent. The gas stream after passing through the acceptor
cycle deprived of S02 could be utilized in the regenerator
cycle and no hydrogen would be required. The same option exists
assuming that the necessary oxygen for the acceptor cycle is
supplied by hot air prior to initiation of the regenerator
off-gas feeding to the acceptor reactor. Whether this possibility
really exists will depend on the 02, CO, S02, and acceptor
equilibrium and kinetic conditions in both the acceptor and the
regenerator reactors. It is also conceivable that many of these
problems or possibilities can be resolved by contacting the
process developer or by relatively simple thermodynamic calcula-
tions, and that no experimental work would be required. Our con-
tacts failed to reveal that enough is known about the operation
of the system described above.
-------
Other advantages of the SFGD process include its insensitivity to
particulates loading and simple static bed operation. This would
allow for rather easy automation and trouble-free operation of
this process and make the process more attractive for the refiners,
6.4.3 Union Carbide PuraSiv-S Process
This process has been developed by the Union Carbide Corporation,
Linde Division, of Tarrytown, N.Y., and entered the list of flue
gas desulfurization processes only a year ago (1972).
The PuraSiv-S process has been developed, pilot-plant tested,
and is now ready for sale by Union Carbide Corp. The process
has been tested on a commercial scale at Coulton Chemical Corp-
oration in Oregon, Ohio.
6.4.3-1 Process Description
The PuraSiv-S process, Figure 26, is a fixed bed adsorption
system for removing and concentrating S02 from S02 containing
gases. Owing to competition of H20 with S02 in the adsorption
process on the tailor-made molecular sieve adsorbent, water must
be removed from the flue gas prior to the adsorption step. The
water vapor concentration of catalytic cracker flue gases ranges
from 13.4 to 23.9% vol. Consequently, the flue gas must be cooled
to about 60°F to reduce the water concentration to less than 2%,
a requirement specified by the process developer. The Brink®
demister after the gas cooling is an added precaution to eliminate
SO3 and water mist from the stream entering the molecular sieve
adsorbers.
115
-------
Absorption
^r^
L^^ H SO
60°F
-»» To Stack
500°F
Stack
550°F
Regeneration
Fuel
Furnace
• To Claus Plant
60°F
Figure 26. PuraSiv-S Process Flow Diagram
-------
Upon thermal regeneration of the molecular sieve, the S02 is
recovered and sent to a Glaus unit for sulfur recovery. To
maintain a constant and maximum S02 concentration feed to the
Glaus unit, six adsorption units operating like a three-phase
electrical circuit are utilized. Three units are adsorbing while
the other three are regenerating. Each of the regenerating units is
at a different phase of regeneration so that one of the three
units is always reaching its maximum S02 production. The entire
system is operated by a proprietary timer-control system. The
removal efficiency can be controlled to any desired level by
cycle time and adsorber size.
Raw Materials and Chemicals
The major raw materials and chemicals needed to operate the
PuraSiv-S process are cooling water, fuel oil, and molecular
sieve adsorbent. The major item is the replacement of molecular
sieve which must be done every two years.
Physical Size
For a flue gas rate of 135,000 scfm, the size requirements are
5250 sq ft with a height of 50 ft. Typical dimensions for the
unit of this capacity are 70 ft x 75 ft x 50 ft.
Location in Regenerator Flow Train
The PuraSiv-S process would have to be placed after both an
efficient electrostatic precipitator and a CO boiler. This is
because the'requirements for -low particulates concentration in gas
(essentially free of particles larger than 2-5 microns) to prevent
adsorbent plugging and contamination and low gas temperatures are
very critical.
117
-------
Regenerator
—
CO Boiler
Electrostatic
Precipitator
—
Electrostatic
Precipitator
CO Boiler
e
i
PURASIV-S
Loniroi
Process
C t ifl*
Mack
6.4.3.2 Economics
Figures 27 and 28 represent the capital investment and operating
costs required for application of the PuraSiv-S process to the
off-gases of an FCC regenerator. The data presented are based on
published information and contacts with the process devel-
oper.97*98'99 A complete operating cost summary is presented in
Tables F10, Fll, and F12, Appendix F. Additional assumptions
used for the PuraSiv-S process include:
Temperature of flue gas entering the process is 500°F
with 20% water by volume
Molecular sieve replacement-$l.25/lb
Molecular sieve lifetime-2 years
Operating labor - 1 man/shift
Initial catalyst charge added to fixed capital investment
(50,000 bpsd - $1,451,200)
6.4.3.3 Comments
The PuraSiv-S process is reliable and capable of reducing S02
emission to any desired level. The reliability is attributable
to the simplicity of the process, a straightforward adsorption-
desorption. However, the process is very sensitive to flue gas
composition. The tailor-made adsorbent is designed specifically
for S02, but water competes with S02 for adsorption and must there-
fore be removed. This entails extensive cooling of the flue gas
to 60°F, requiring massive and expensive heat exchange facilities.
The molecular sieve adsorbent is effective in removing S02 in the
presence of 7 to 10 percent C02 , trace quantities of NOX, and CO
concentrations typically found in FCC regenerator off-gas.
118
-------
10
8
•w-
o
CD
c.
I I
104
105
Capacity, scfm
10°
Figure 27. PuraSiv-S Capital Investment Cost (February 1973)
119
-------
1.0
•K 0.1
o
o
2
o>
0.01
104
105
Capacity, scfm
Figure 28. PuraSiv-S Operating Cost
10°
120
-------
Presence of particulates , however, has a detrimental effect on
the adsorption process due to plugging of adsorption reactors.
6.5 SCRUBBING SYSTEMS
Two processes, the Wellman-Lord sodium sulfite absorption process
and the MAGOX magnesium oxide scrubbing process, were evaluated
in detail. In addition to these two scrubbing systems, other
types of scrubbers were also considered and included limestone,
ammonia, and zinc oxide scrubbing systems.
The possibility of applying the scrubber systems to gases with a
reducing atmosphere and minimal oxygen concentrations appeared to
be technically attractive. The reducing atmosphere could prevent
oxidation of S02 to SOs and eliminate some of the operational and
waste disposal problems that the scrubber systems experience when
operated in oxidation atmosphere. All processes, however, had
shortcomings not present in magnesia or sodium sulfite scrubbing.
The shortcomings were primarily associated with scrubbing liquor
regeneration. The ammonia systems produce (NH^^SOjf, an undesirable
product, or thermally decompose the (NHjt)2SOit to NHjfHSOi, and NHa
(see reactions below) in order to recover S02 . But NHa pollution
problems are encountered.
+ S02 + H20 (l4-a)
2 NHjtHSOij + (NHit)2S03 - »- 2 (NH4)2SO£t + S02 + H20
+ NH3 ( 15
Lime/limestone scrubbing is presently viewed as a sludge-producing
process that creates unacceptable waste disposal problems, and
regeneration with recovery of the sulfur value has yet to be
demonstrated. A00 The process application to FCC regenerator
off-gas in comparison with the two processes evaluated in detail
was regarded as less attractive.
121
-------
The zinc oxide process merely uses zinc oxide as an acceptor for
S02 after it has been scrubbed from the flue gas by a sodium
sulfite solution. The transfer involves some rather complicated
chemistry and yet only recovers 70% of the sulfur removed from
the flue gas.88 The remainder is lost as CaS04.
6.5-1. Wellman-Lord Sodium Sulfite Absorption
The process has been developed by Davy Powergas, Inc., formerly
Wellman-Power Gas, Inc., Lakeland, Fla.
The Wellman-Lord scrubbing process was commercialized in 1970.
Table 11 lists the installations currently operating, being
constructed, or on the drawing boards.
6.5-1.1 Process Description
The flow scheme for the basic process is shown in Figure 29. The
S02-rich gas is contacted countercurrently in the absorber by
the sodium sulfite solution and exits the absorber top stripped of
S02. The solution leaving the bottom of the absorber, now rich in
bisulfite, is discharged to a surge tank and then pumped to a
proprietary evaporator/crystallizer in the regeneration section.
Low pressure steam is used to heat the evaporator and drive off
S02 and water vapor. The sodium sulfite precipitates as it
forms and builds a dense slurry of crystals.
The gas stream leaving the evaporator is subjected to partial
condensation to remove the majority of the water vapor before
the product S02 is discharged from the process. The condensate
in the mixture with the sulfite slurry stream withdrawn from the
evaporator is used for re-dissolving the slurry in the dissolving
tank. The sulfite lean solution is then pumped to a surge tank
122
-------
Table 11. WELLMAN-LORD S02 RECOVERY PROCESS DEVELOPMENT
Client
Olin Corporation
Toa Nenryo
Japan Synthetic Rubber Co.
Standard Oil of California
Allied Chemical Corp.
Olln Corporation
Northern Indiana Public
Service Co. (NIPSCO)
Sumitomo Chiba Chemical Co.
Standard Oil of California
Japanese Synthetic Rubber
Kashima Oil Company
Confidential
Chubu Electric
Standard Oil of California
Standard Oil of California
Confidential
Toa Nenryo
Toa Nenryo
Location
Paulsboro, NJ
Kawasaki, Japan
Chiba, Japan
El Segundo Refinery
El Segundo, Ca
Calumet Works
Chicago, 111
Curtis Bay, Md
Gary, Ind
Chiba, Japan
Richmond Refinery
Richmond, Ca
Yokkaichl, Japan
Kashima, Japan
Kawasaki, Japan
Nagoya, Japan
Richmond, Ca
El Segundo, Ca
Kansl, Japan
Arita, Japan
Hatsushima, Japan
Application Offgas Origin
Sulfuric acid
Claus Plant
Oil Fired Boiler
Claus Plants
Sulfuric acid
Sulfuric acid
Coal Fired Boiler - 115 MW
power plant
Oil Fired Boiler
Claus Plants
Steam Boiler Plant (S02 converted
into sulfuric acid)
Claus Plants
Steam Boiler Plant (S02 converted
into sulfuric acid)
Oil Fired Boiler - 220 MW power plant
Claus Plant - 325 TPD S
Claus Plant - 325 TPD S
Power Plant
Claus Plant
Claus Plant
S C F M
Treated
15,000
11,000
121,000
30,210
29,580
78,016
310,000
250,000
37,000
270,000
20,200
112,000
390,000
30,000
30,000
390,000
35,000
9,000
Completion
Date
July 1970
August 1971
August 1971
June 1972
August, 1972
in progress
in progress
in progress
in progress
in progress
in progress
in progress
May 1973
in progress
in progress
in progress
in progress
In progress
-------
ro
Clean Gas 140UF
Evaporator
tvaporaior * »
Crystallizer [
Product
Sulfur Dioxide
110°F
H2O
Heat
Recovery
Dissolving
Tank
Figure 29. Wellman Lord, Inc., Sulfur Dioxide Recovery, Sodium System
-------
and fed back to the absorber. To ensure the selection of the
optimum process design in relation to the overall facility the
pretreatment of feed S02-rich gas requires evaluation on a case-
by-case basis. The type of pretreatment ultimately chosen will
depend on gas temperature, particulate content, organic sulfur
content, sulfur trioxide content, acid mist or vapor content,
and humidity.
The process is based on a sodium sulfite/bisulfite cycle. The
reactions that take place in the process can be abbreviated for
simplicity as follows:
Absorption
S02 + Na2S03 + H20 »• 2NaHS03 (16)
Regeneration
2NaHS03 > Na2S03l + S02 |+ H20 t (17)
Apart from the two major reactions above, sodium sulfate (Na2S01|),
which is nonregenerable in the normal process, is formed in the
absorber as a result of solution contact with oxygen and sulfur
trioxide as follows:
Na2S03 + 1/2 02 »• Na2SOI+ (18)
and
2Na2S03 + S03 + H20 »• NajjSd, + 2NaHS03 (19)
The sodium sulfate so formed is controlled at a level of approx-
imately 5 wt % in the absorber feed stream by maintaining a
continuous purge from the system. High concentration of S03 in
the flue gas, if not removed in the gas pretreatment, would re-
sult in an increased concentration of Na2S04 and proportionally
increased purge of the scrubbing liquor.
125
-------
An additional source of sodium sulfate and thiosulfate is the
so-called disproportionation reaction which takes place in the
regeneration section:
6NaHS03 »• 2Na2SOi( + Na2S203 + 2S02 + 3H20 (20)
Laboratory research and commercial experience have guided the
selection of process operating conditions that minimize all
these sources of sodium sulfate formation.
A makeup of caustic is required to replace that lost in the purge
stream. The caustic makeup solution reacts with the sodium
bisulfite in the absorber solution to form additional sodium
sulfite:
NaOH + NaHS03 >• Na2S03 + H20 (21)
Soda ash (Na2C03) can also be used as the makeup source of sodium.
The simple regeneration scheme of the patented Wellman-Lord process
relies on the favorable solubilities of the sodium system. The
bisulfite form has almost twice the solubility of sulfite at the
temperatures considered for the process. Because of this it is
possible to feed the absorber with a saturated sulfite solution,
or even a slurry, without any fear of additional crystallization
or scale forming despite considerable evaporation of water. This
is because as S02 absorption proceeds, the composition of the
solution is shifted in the direction of increasing solubility
as reaction (16) demonstrates.
There are several advantages to operating the absorber with highly
concentrated solutions. The solubility of oxygen decreases
rapidly as salt concentration increases and so the mass transfer
126
-------
of oxygen into the solution is slowed appreciably. This reduces
oxidation of the sodium salts to sodium sulfate. Also the steam
requirements of the process are directly related to the quantity
of water pumped around the system. Operating .at or near saturation
with respect to sulfite thus reduces the overall steam consumption.
The same solubility effect is taken advantage of in a reverse
fashion in the regeneration section. As SOfc is stripped from
the concentrated solution, the sulfite salt is formed, rapidly
reaching its solubility limit, and precipitates as crystals.
The low SO2 vapor pressure component (sodium sulfite) in the
regenerated solution is continuously removed from the system and
the regeneration operation proceeds with a constant high per-
centage of bisulfite in solution permitting a considerable reduc-
tion in the stripping steam requirements.
The absorber is a two or three stage contacting device. Depending
on the requirements of a specific application, the unit may be
either a tray or a packed tower. For most gases treated, the
absorber will require recirculation around each individual stage,
since the quantity of feed solution is insufficient to adequately
wet the trays or packing.
During operation the recirculation rate can be throttled until
the S02 emissions from the process increase up to the specified
requirements. This will minimize oxidation reactions since the
absorption of oxygen is liquid-film controlled and therefore
proportional to liquid rate. The absorption of S02 is gas-film
controlled and therefore relatively insensitive to liquid rate.
An inherent advantage of the Wellman-Lord process is that the
absorption system can be physically separated by a large distance
12?
-------
from the regeneration system. In some applications, it may be
practical to treat gases from more than one plant by installing
separate absorbers for each source of S02, with all the absorbers
being supplied from a common regeneration system.
The use of solution storage not only enables the process to
operate smoothly during frequent changes of gas flow or S02
concentration, but it also permits a complete shutdown of the
regeneration section to perform scheduled maintenance activities,
without any pause in SC>2 absorption. This feature minimizes the
amount of expensive spare equipment required with no sacrifice
in basic pollution control reliability.
The heart of the regeneration system is a conventional forced-
circulation evaporator/crystallizer . The design parameters of
this unit have been developed such that long-term continuous
operation is assured. The evaporator can be designed to use
very low pressure exhaust steam (30 psig), which might otherwise
be discharged to the atmosphere. In very large plants or in case
of high cost steam it is practical and economical to operate the
regeneration system as a double effect evaporator. This will
reduce steam consumption by about ^Q
The gases leaving the evaporator are subjected to one or more
stages of partial condensation. Existing plants are operating
with both air and water-cooled condensers in this service. The
final product S02 can be delivered at whatever quality is re-
quired for further processing. It is suitable for conversion to
high grade sulfuric acid, elemental sulfur, or liquid S02.
Existing Wellman-Lord plants use the S02 product for either
sulfuric acid or sulfur production.
128
-------
A small amount of the circulating solution oxidized to a non-
regenerable sulfate form and purged from the system can be
dried for sale or disposal, or it can be treated to permit its
discharge as an innocuous effluent. Other process steps are
available which recover the sodium values, thus allowing the
system to operate as a closed loop.
Raw Materials and Chemicals
Sodium as NaOH or Na2C03 - 0.^5 Ib/lb S recovered101
Steam (low pressure 30 psig, 275°P) - 33 Ib/lb S recovered101
Oxidation inhibitors
Physical Size Requirements
The regeneration system does not need to be in the immediate
vicinity of the scrubber. With limited area in petroleum
refineries around catalytic cracking units, this feature could
be advantageous. As can be seen from the listing of commercial
installations, Table 11, several small units in the 30,000 scfm
range have been built to control emissions from refinery Glaus
plants. This suggests other advantages for the process: the
process is known to the refiners, and multiple scrubbers within
a refinery could control several sources of SO emission using a
A.
centrally located regeneration plant.
Location in Regenerator Downstream Flow Train
The Wellman-Lord process should precede the CO boiler to avoid
an oxidizing atmosphere, which would adversely affect the sulfite
concentration by oxidizing it to sulfate, and increase sodium
requirements.
129
-------
Regenerator
Wellman-Lord
Pnnfrnl
Process
i
I
i
i
I*.
CO Boiler
Electrostatic
Precipitator
Electrostatic
Precipitator
CO Boiler
A
i
Stack
6.5.1.2 Economics
The capital investment and operating costs for the Wellman-Lord
scrubbing process are illustrated in Figures 30 and 31 • For
capital and operation cost estimates the results from several
sources of information are presented due to unreconcilable
differences among them.1 ° 1 > i °2 , i o4* , i 05 , i o 9 , 1 1 o
Detailed operating cost estimates are summarized in Tables F13,
F14, and F15, Appendix F. The following assumptions were made
in addition to those Appearing in Section 3-5.2.
NaOH Cost - $60/ton
Operating labor - 3 men/shift
No charge or credit was applied for disposal or use of
Na2SOtt .
6.5-1.3
Comments
The Wellman-Lord sodium sulfite absorption system has several
attractive features that make it desirable for application to
SOX control of fluid catalytic cracking regenerator flue gas.
Several units are presently operating on the tail gases from
refinery Glaus units. As already mentioned, the regeneration
130
-------
10'
,8
OJ
10'
10'
Rochelle.
We 11 man-Lord
NIPSCO Demonstration
Unit
104
Figure 30.
103
Capacity, scfm
Wellman Lord, Inc., Capital Investment Cost
(February 1973)
10e
131
-------
1.0
*-" n i
-------
section may be physically removed from the absorber so that
multiple sources of SOX may be treated using a common regeneration
system. Also, the system can be designed to obtain virtually
any removal efficiency for sulfur oxides.
The process is fully developed. The reliability of the process
can be demonstrated by the operating records of existing commercial
installations. At Japan Synthetic Rubber Company a unit operated
from May 9, 1972 to March 1, 1973 with an on-stream factor of
100?, and operated from June 1971 to March 1973 with an on-stream
factor of 97%. Depending on the size of the surge tanks, the
regeneration system may be shut down for maintenance with the
scrubber continuing operation. The surge tanks will also
minimize any difficulty involved with fluctuation of flue gas
concentration.
Areas in which further study is required include an examination
of possible catcracker catalyst effect on the oxidation of
sulfite to sulfate, which would increase the waste disposal
problem and the process operating cost. The cost for disposal of
a dried salt suitable for landfill could range from $0.75 to $5.00
per ton while deep well injection of a solution compatible with
the geology of the well could cost $1.00 per 1000 gallons or $6.85
per ton (based on 3-5% brine disposal cost). The dried mixture of
Na2SOit and Na2SOa might be a desirable raw material for the paper or
glass industry.
In relation to the waste disposal problem, further study is needed
to investigate possible recovery techniques for the purge stream.
The loss of sodium value consists of approximately 40 Ib Na lost per
106 scf treated.101 This value, as well as the performance of the
scrubber, will be affected by the S03 concentration in the flue
gas. Higher levels of S03 and 02 will reduce the desirability
of this process. Applying the process to gases from the PCC
133
-------
regenerator with its reducing atmosphere and low oxygen con-
centrations is an attractive possible solution. The effects of
other components present in the regenerator off-gas, however,
will need to be investigated. The process would control the
particulates emissions but would require an additional reheat of
the desulfurized stream to prevent plume formation after the
stream has been discharged to the atmosphere.
6.5.2 Magnesium Oxide Scrubbing
The process has been developed by Chemical Construction Company
(Chemico) and Basic Chemicals, New York, New York.
A 425,000 acfm demonstration unit has been built on the No. 6
unit of Boston Edison's Mystic Station. Boston Edison paid for
the absorption system while EPA furnished funds for magnesium
oxide regeneration. Construction was completed in April 1972. 113
To date each of the process steps has been demonstrated with
better than 90% S02 removal efficiency. Further long-term
operation is planned to evaluate system reliability.
6.5-2.1 Process Description
The Chemico/Basic Magnesia Slurry process for recovery of sulfur
dioxide from stack gases is shown schematically in Figure 32.
The process involves operations associated with the following
primary areas: (1) particulate removal
(3) slurry handling and dewatering, (4) solids drying and calcining,
and (5) S02 by-product manufacturing. The overall process
chemistry involved is represented by the reactions shown in Table 12,
Flue gas containing S02 and catalyst fines passes into a particulate
scrubber where particulate matter is removed using recirculated
water as the scrubbing medium. A bleed stream from the scrubber
is thickened to concentrate the fines as a slurry underflow, which
-------
uo
U1
FLUE
GAS
TO STACK
1000°F
HEAT
RECOVERY
WATER
' 1
|310°F 1
i
~~ PARTI
SCRll
i
CULATE
BBER
•^— 1 THICKENER 1
ASH TO POND
t
I75VF
\ \
I
SULFUR DIOXIDE
ABSORBER
SLURRY
i
I OEWATERING SYSTEM 1
LIQUOR
H*
ROVED 1
RECYCLE
POND
WATER
H
AIR
FUEL
WATER MfO
LL
MAKE-UP SYSTEM
MgS03
1
1
RECYCLE
MgO
r-*-
CALCINER
AIR
FUEL
Figure 32. Magnesia Slurry S02 Recovery Process
-------
Table 12. CHEMISTRY OF MAGNESIA SLURRY S02 RECOVERY PROCESS
ABSORPTION
Main Reactions
MgO + S02 + 3H20 -»- MgS03-3H20
MgO + S02 + 6H20 ->• MgS03'6H20
Side Reactions
MgS03 + S02 + H20 -> Mg(HS03)2
Mg(HS03)2 + MgO t- 2MgS03 + H20
MgO + S03 + 7H20 f MgSOi^-THzO
MgS03 + hQ2 + 7H20 -»• MgS01+-7H20
DRYING
Main Reactions
MgS03-3H20 ^ MgS03 + 3H20
MgS03-6H20 ^ MgS03 +6H20
MgSO^-THzO ^ MgSO,t + 7H20
Side Reaction
MgS03 + %Q2 +
REGENERATION
MgS03 MgO + S02
+ hC £ MgO + %C02 + S02
(22)
(23)
(24)
(25)
(26)
(2?)
(28)
(29)
(30)
(3D
(32)
(33)
136
-------
is transported to a disposal area. Overflow from the thickener
is returned to the scrubber circuit for reuse.
Flue gas leaving the particulate scrubber enters a venturi
absorber where it is contacted co-currently with an aqueous
recycled slurry containing magnesium oxide (MgO), magnesium
sulfite (MgS03), and magnesium sulfate (MgSO^). The absorption
reaction takes place between S02 and magnesium oxide and the
magnesium sulfite is formed. Some of the S02 may also react
with MgSOs in the presence of water to form magnesium bisulfite,
which immediately reacts with the excess MgO present to yield
additional MgS03. The quantity of MgO in the absorption slurry
is maintained in excess of the stoichiometric requirement for
reacting with all of the S02 present in the flue gas. A portion
of the sulfur trioxide (S03) contained in the flue gas is absorbed
in the slurry and reacts to form MgSO^. Additional amounts of
MgSOt, can also be formed due to in situ oxidation of a portion
of the magnesium sulfite.
The resulting aqueous slurry is discharged from the absorber and
contains hydrated crystals of MgO, MgS03, and MgS04 as well as
a solution that is saturated with each of these components. A
continuous side-stream of this recycled slurry is diverted to a
centrifuge where partial dewatering produces a moist cake con-
taining crystals of MgS03-3H20, MgS03-6H20, MgSO^-THzO, and
unreacted MgO. The clear liquor concentrate is returned to the
main recirculating slurry stream together with makeup MgO slurry,
and the resulting combined slurry is recycled to the absorber
for further S02 recovery. The wet solids cake is conveyed to a
dryer where free and bound moisture is removed using a direct-
contact drying gas under non-oxidizing conditions. The dryer
product is subsequently calcined to produce MgO, which is reused
137
-------
in the absorption system after having been slaked and slurried
in a makeup tank. The S02-rich effluent gas from the calciner
is then employed in the production of either sulfuric acid,
elemental sulfur, or liquefied S02.
Raw Materials and Chemicals
The process requires magnesium oxide and carbon for its operation.
Amounts of both raw materials will depend on the presence and
formation of MgSO^ in the system as well as the regeneration
capabilities of a specific process. Formation of MgSOi, will be
proportional to the flue gas SQ^ concentration with higher 863
concentration yielding more MgSOt,.
Physical Size Requirements
The area required for a two-stage venturi scrubber system with
fluid bed dryer and calciner was estimated at approximately
1530 cu ft/1000 bpsd.83
Location in Regenerator Downstream Flow Train
Regenerator
Magnesium
Oxide
Scrubbing
i *
i
i
i
CO Boiler
Electrostatic
Precipitator
~*
Electrostatic
Precipitator
CO Boiler
A"
i
— i
Stack
The scrubbing system should be installed upstream of the CO boiler
to minimize oxidation of sulfite to sulfate. This is advantageous
in that it minimizes the operating expense of regeneration
reactions (32) and (33).
138
-------
MgS03 - MgO + S02 (32)
MgSO,, + 1/2 C A MgO + 1/2 C02 + S02 (33)
Regeneration of MgS03 does not require carbon, an added raw
material expense.
6.5.2.2 Economics
The capital investment and operating costs for magnesium oxide
scrubbing appear in Figures 33 and 34. The capital and operating
cost estimates are based on data in published literature.83>88
The detailed operating cost estimates are summarized in Tables
F16, F17, and Fl8, Appendix F. Additional assumptions made
specifically for this process are:
Lime cost - $l6/ton
MgO cost - $102.50/ton
Coke cost - $23-50/ton
Fuel oil cost - $0.09/gal
Operating labor - 3.5 men/shift
6.5.2.3 Comments
Magnesium oxide scrubbing remains a potential candidate for SOX
removal from the FCC regenerator flue gas. The long-term
reliability of the process is yet to be evaluated. As a re-
generable process it is similar to a regenerable limestone
scrubber, but the calcining temperature required for regeneration
is much lower. Magnesium sulfite and sulfate are optimally
calcined between 750 and 1100°F 88 while the calcium system
requires temperatures in the vicinity of 2000°F.100 An additional
139
-------
10C
OJ in'
10'
I/I
0>
10
I I I
105
. Capacity, scfm
106
Figure 33. Magnesium Oxide Scrubbing, Capital Investment
Cost (February 1973)
140
-------
1.0
"fc
tf 0.1
o
o
Ol
2
o>
0.01
105
Capacity, scfm
106
Figure 3^- Magnesium Oxide Scrubbing, Operating Cost
-------
advantage is the physical flexibility of the regeneration system.
The calciner is capable of accepting multiple feeds from many
scrubbers and may be centrally located to accommodate scrubbers
on many SOX emission sources in a refinery.
6.6 OXIDATION SYSTEM
The final process remaining for evaluation differs from the
others in that it does not produce a stream of concentrated
S02 as the product. The product of the catalytic oxidation
(CAT-OX) process is a solution of sulfuric acid of approximately
785? concentration.
6.6.1 CAT-OX Catalytic Oxidation
This process has been developed by Monsanto Company, St. Louis,
Missouri. A 15 MW (24,000 scfm) pilot unit has been operated
in cooperation with Metropolitan Edison Co., Seward, Penn .81
A 100 MW (238,000 scfm) commercial size demonstration unit is being
operated at the Illinois Power Company's Wood River No. 4 unit. It
was financed jointly by the Control Systems Division of the Office
of Research and Monitoring of the EPA and the Illinois Power
Company. It began operation on 4 September 1972 .12° The first
commercial operation of integrated CAT-OX units is estimated to
occur in July 1977- Start-up and operation of a commercial plant
with a reheat alternative could be as early as January 1975-81
6.6.1.1 Process Description
Two flow diagrams, Figure 35 and 36, for the CAT-OX process are
presented. The first is for low temperature (310°F) flue gas
and the second is for high temperature (850°F) flue gas.
1H2
-------
LO
PRECIPITATOR
TO
STACK
850F
CONVERTER
[REHEAT^
BURNERr-i
GAS HEAT
EXCHANGER
CAT-OX
MIST
ELIMINATOR
ABSORBING
TOWER
450°F
REHEAT
BURNER
SULFURIC
ACID
ACID
COOLER
RECYCLE
STORAGE
78% H2S04
Figure 35- CAT-OX Flow Diagram - Flue Gas Reheat System
-------
-DAMPER
DAMPERS
LJ TUBULAR
GAS HEAT
CAT-OX
MIST
ELIMINATOR
ABSORBING
TOWER
COOLANT
COOLANT
ACID
COOLER
RECYCLE
STORAGE
78%
Figure 36. CAT-OX Flow Diagram - High Temperature System
-------
The catalytic converter of the CAT-OX process operates at 850°F.
To treat gases having lower temperatures requires supplemental
flue gas heating, Figure 35- Essentially particulate-free gas
(gas free of particulates larger than 2-5 microns) is required
for optimal operation of the catalytic converter. A 0.005 grain/
cu ft loading of carbonaceous matter creates serious problems of
plugging and poisoning the catalyst bed. Consequently, very
efficient electrostatic precipitators removing in excess of 99.55?
of the particulates are part of the CAT-OX system.
Hot and clean gas is passed through the converter where the
sulfur dioxide is catalytically oxidized on vanadium pentoxide
catalyst to sulfur trioxide according to the reaction:
2S02 + 02 »• 2S03 + 2H20 >• 2^80,, (3*0
The converter is designed so that the catalyst bed can be
emptied onto a conveyor system for transport to a cleaning
process, after which the cleaned catalyst is conveyed back to
the converter. About 2.5? of the catalyst mass is lost during
each cleaning process, which is anticipated to occur about four
times per year. About 48 hours is required for each catalyst
cleaning.
The treated flue gas, now containing S03, passes to a Ljungstrom-
type heat exchanger where about JJOO°F sensible heat is recovered
to heat the incoming untreated flue gas. As a result of heat
recovery in this exchanger, the overall need for fuel usage is
to add 150°F of sensible heat. The temperature of the gas is
maintained well above the dew point. Normal flue gas leakage
in a regenerative heat exchanger of this type will allow about
5/2 12° of the flue gas to by-pass the unit, thereby reducing the
overall efficiency of S02 removal to approximately 85?.
-------
The flue gas is further cooled in a packed-bed absorbing tower,
which operates in conjunction with an external shell and tube
heat exchanger. During cooling, the H20 and S03 combine to
form sulfuric acid, which is subsequently condensed. The tower
brings a cool stream of sulfuric acid into direct contact with
the rising hot flue gas. Exit gas leaves the packed section at
about 250°P while hot acid is constantly being removed from the
bottom of the tower and cooled in the external heat exchanger.
Very fine mist particles of sulfuric acid are formed in the gas
as it is cooled in the absorbing tower. These mist particles
in the flue gas are removed along with some entrained droplets
of circulating acid from the tower by the Brink® mist eliminator
system. The packed section of the absorbing tower and the mist
eliminators are contained within one vessel. The flue gas
leaving the mist eliminator to enter the exit stack contains
approximately 25 ppm of S03 which is less than the amount
normally emitted from the combustion process.
Raw Materials and Chemicals
Annually 105? of catalyst requires replacement. Some sources
list this expense as a raw material, others as a part of the
maintenance cost. We have identified it as raw material.
Physical Size Requirements
CAT-OX Requirements per 1000 bpsd
FCC Capacity122
High Temperature Flue Gas
Flue Gas Reheat
area (sq ft) 186 139
acid storage (60 days) (sq ft) 775 775
146
-------
Location of Process In Regenerator Downstream Flow Train
Rpnpnpratnr
CO Boiler
Electrostatic
Electrostatic
Precipitator
r<1 Dnilar
A
j
CAT-OX
Control
Process
Stack
Precipitator
CAT-OX must follow the CO boiler to ensure an 02/S02 ratio of
10:1 for proper operation of the catalytic reactor. This ratio
is required to obtain adequate oxidation from S02 to S03. There
are three alternatives of process integration with the FCC
equipment depending on flue gas effluent temperature. These
are presented below.
Heat Recovery
>600°F
Precipitator
— *•
-*•
600°F
Heat
Exchanger
Heat
Exchanger
Reh
""*" 450°F
eat to 85
Converter
0°F
I
Absorber
i
Stack
i Heat Recovery
Precipitator
«sn°F
Converter
L
jwn°c
Heat
Exchanger
"—
A*n<%
Absorber
J
Stack
-------
Precipitator
450°F
L
™°F
Heat
Exchanger
»
Converter
I
Absorber
i
Reheat to 850°F
Stack
6.6.1.2 Economics
For economic considerations, the flue gas treated will be
assumed to be available at 310°F, although the gas temperature
after the CO boiler is usually in the range of l»85-820°F. An
actual temperature would result in less gas reheating if com-
pared with the reheat system, but more gas reheating if com-
pared with the high temperature alternative. Additional
assumptions required specifically for CAT-OX include:
Catalyst cost - $4l/cu ft123
Catalyst life - 5 years123
Operating labor - 3 men/shift
The investment and operating costs are shown in Figures 37 and
38. The detailed operating cost estimates are summarized in
Tables F19, F20, and F21, Appendix F.
6.6.1.3
Comments
The CAT-OX S02 removal process was designed for application to
stationary power sources, and extensive study would be required
to optimize the process for application to FCC regenerator
off-gases.
148
-------
5 7
I 10
2
Illinois Power
Q10MW Basis)
104
Rochelle (500MW Basis),
Enviro-Chem (100MW Basis)
105
Capacity, scfm
106
Figure 37. CAT-OX Capital Investment Cost (February 1973)
-------
1.0
ai
0.01
104
io3 -
Capacity, scfm
106
Figure 38. CAT-OX Operating Cost
150
-------
Monsanto Enviro-Chem Systems, Inc. has contacted refiners and
has yet to find a suitable refinery application for CAT-OX.
Reasons for this include:
- size of the application - refineries are too small for
economical operation.
- CAT-OX produces dirty, low-concentration sulfuric acid
not readily usable in a refinery.
- Plugging and poisoning of catalyst due to carbonaceous
material not removed by electrostatic precipitator
(CAT-OX has not been tested and demonstrated on oil-fired
boilers, which would probably produce more carbonaceous
material and increase the catalyst poisoning problem.)
151
-------
REFERENCES
1. Anon., Atmospheric Emissions from Petroleum Refineries, A
Guide for Measurement and Control, U.S. Department of
Health, Education and Welfare, Public Health Service,
Division of Air Pollution, Publication No. 763, I960.
2. Sussman, V. H., Atmospheric Emissions from Catalytic
Cracking Unit Regenerator Stacks, Report No. 4, Los
Angeles County Air Pollution Control District, June,
1957.
3- Danielson, J. A., Air Pollution Engineering Manual, Air
Pollution Control District, County of Los Angeles, U.S.
Department of Health, Education, and Welfare, Public
Health Service, Bureau of Disease Prevention Environmental
Control, National Center for Air Pollution Control,
Cincinnati, Ohio 1967-
4. Shea, E. P., Source Testing - EPA Task No. 8, Standard
Oil Company, Richmond, California, Midwest Research
Institute, Kansas City, Missouri 6*1110, EPA Contract No.
68-02-0228, April 1972.
5- Schulz, E. J., Hillenbrand, L. J., Engdahl, R. B., Source
Sampling of Fluid Catalytic Cracking Plant (Electrostatic
Precipitators and CO boiler) of Standard Oil of California,
Richmond, California, Contract No. 68-02-0230, Task Order
No. 3» Battelle, Columbus Laboratories, Columbus, Ohio,
July, 1972.
6. Schulz, E. J., Hillenbrand, L. J. Engdahl, R. B., Source
Sampling of Fluid Catalytic Cracking, CO Boiler, and
Electrostatic Precipitators at the Atlantic Richfield
Company, Houston, Texas, Contract No. 68-02-0230, Task
Order No. 3 Battelle, Columbus Laboratories, Columbus,
Ohio, July, 1972.
7- Anon., Compilation of Air Pollutant Emission Factors
(Second Edition), U. S. Environmental Protection Agency,
Office of Air and Water Programs, Office of Air Quality
Planning and Standards, Research Triangle Park, North
Carolina, April, 1973-
8. Cavender, J. H., Kircher, D. S., Hoffman, A. J., Nationwide
Air Pollutant Emission Trends 19^0-1970, U. S. Environ-
mental Protection Agency, Office of Air and Water Programs,
Office of Air Quality Planning and Standards, Research
Triangle Park, North Carolina, January, 1973-
152
-------
9. Cantrell, A., U. S. Refineries: where, capacities, types of
processing, The Oil and Gas Journal, April 2, 1973.
10. Anon., Petroleum Pacts and Figures 1971 Edition, American
Petroleum Institute, Washington, D. C., 20006, 1971.
11. Prescott, J. H., Refineries: The Dam Breaks, Chemical
Engineering, August 6, 1973.
12. Anon., Parsons Gets Pact for 200,000 - b/d Louisiana
Refinery, The Oil and Gas Journal, September 24, 1973.
13. Takaoka, S., Petroleum Desulfurization, A private report
by the Process Economics Program, Stanford Research
Institute, Report No. 47, January 1969.
14. Robson, F. L., Giramonti, A. J., Lewis, G. P., Gruber, G.,
Technological and Economic Feasibility of Advanced Power
Cycles and Methods of Producing Nonpolluting Fuels for
Utility Power Stations, United Aircraft Research Laboratories,
December 1970.
15. Anon., 1972 Refining Processes Handbook, Hydrocarbon
Processing, September 1972.
16. Bland, W. P., Davidson, R. L., Petroleum Processing Handbook,
McGraw-Hill Book Company, 1967 -
17- Nelson, W. L., Petroleum Refinery Engineering, McGraw-Hill
Book Company, Fourth Edition, 1958.
18. Gruse, W. A., Stevens, D. R., Chemical Technology of
Petroleum, McGraw-Hill Book Company, Third Edition, I960.
19. Anon., Progress in Petroleum Technology, Number 5 of the
Advances in Chemistry Series, American Chemical Society,
Augus^, 1951.
20. Conser, R. E., The Shell Flue Gas Desulfurization Process:
How Does It Stack Up? Paper Presented at the Conference
on Control of Gaseous Sulphur Compound Emission, University
of Salford, Salford, England, April 9-12, 1973-
21. Anon., Crude Petroleum, Petroleum Products, and Natural-Gas
Liquids: 1971, U. S. Department of the Interior, Bureau of
Mines, Washington D. C., 20240, December 20, 1972.
153
-------
22. Anon., Availability of Heavy Fuel Oils by Sulfur Levels, U. S.
Department of the Interior, Bureau of Mines, Washington D. C.
20240, June 7, 1972.
23. Anon., Burner Fuel Oils, 1971, Ella Mae Shelton Bartlesville
Energy Research Center, Bureau of Mines, U. S. Department of
Interior, Bartlesville, Oklahoma, August 1971.
24. Gary, J. H. Schweyer, H. E., Crude Source and Desulfurization,
Petroleum Refiner, September, 1953.
25. Sittig, M. Catalytic Cracking Techniques in Review, Petroleum
Refiner, September, 1952.
26. Wunderlich, 0. A., Emerson Ivey, F., Jr., Charts Make Cat
Cracker Balance Easier, Petroleum Refiner, February, 1958.
27. Private Communication, Universal Oil Products.
28. Private Communication, M. W. Kellogg Co.
29. Private communication, Brown and Root, Inc.
30. Davidson, R. C., Cracking Sulfur Stocks with Natural Catalyst,
Petroleum Refiner, September, 1947.
31. Healy, J. W., Hertwig, W. R., Sulfur Distribution in
Catalytic Cracking, Significance of Process Variables, Paper
presented before the Division of Petroleum Chemistry,
American Chemical Society, Atlantic City Meeting, September
18-23, 1949-
32. Conn, A. L., Brackin, C. W. Cracking of High Sulfur Stocks,
Use of Steam with Natural Catalyst, Industrial and Engineering
Chemistry, August 1949.
33. Wollaston, E. G., Forsythe, W. L., Vasalos, I. A., Sulfur
Distribution in FVU Products, The Oil and Gas Journal,
August 2, 1971.
34. Fowle, M. J., Bent, R. D., Where Does the Sulfur Go? Petroleum
Refiner, November, 1947.
35. Private Communication, American Oil Co.
36. Private Communication, Monsanto Company
154
-------
37. Jurjew, J. K., Katalytische Umwandlungen Von Heterocyclischen
Verbindungen, II Mitteil: Uberfuhrung Von a-Methyl-Puran In
a-Methyl-Pyrrol Und a-Methyl-Thiophen, III Mitteil: Reziproke
Umwandlungen Der Einfachsten Funfgliedrigen Heterocyclen.
Chemische Berichte, 6_9B, 1002-1004, 1936.
38. Yur'ev, Yu.K., Tronova, V. A., Catalytic Transformations of
Heterocyclic Compounds XV. The Stability of the Catalyst and
Optimum Conditions in the Transformation of Tetrahydrofuran
into Pyrolidine and Thiophene and in that of Furan into
Thiophene, J. Gen. Chem. (U.S.S.R) !LO, 31-4 (1940)
39. Anon., Amoco Oil to Modify Its Cat Crackers, The Oil and Gas
Journal, April 2, 1973-
40. Magee, J. S., Blazek, J. J., Ritter, W. R., Catalyst
Developments in Catalytic Cracking, The Oil and Gas Journal,
July 23, 1973-
41. Thomas, C. L., Catalytic Processes and Proven Catalysts,
Academic Press, 1970.
42. Eberly, P. E., Jr., Kimberlin, C. N., Jr., Miller, W. H.,
Drushel, H. V., Core Formation on Silica-Alumina Cracking
Catalysts, Industrial and Engineering Chemistry, Process
Design and Development, April 1966.
43. Burke, P. D., Catalysts - Part 1, Chemical Week, August
17, 1963-
44. Burke, P. D., Catalysts - Part 2, Chemical Week, August
24, 1963-
45. Dean Edgar, M., Thermal Shock no Problem for Fresh FCC
Catalyst, The Oil and Gas Journal, September 24, 1973-
46. Mills, G. A., Ashwill, R. E., Gresham, T. L., To Evaluate
Cracking Catalysts, Hydrocarbon Processing, September 1967-
47. Nelson, W. L., What Is Good Catalyst, The Oil and Gas
Journal, April 28, 1949-
48. U. S. Patent 2,724,686
49. U. S. Patent 2,587,554
50. U. S. Patent 2,628,157
155
-------
51. U. S. Patent 2,586,705
52. U. S. Patent 2,497,940
53. U. S. Patent 2,496,356
51. U. S. Patent 2,467,850
55- U. S. Patent 2,449,617
56. U. S. Patent 2,487,132
57- U. S. Patent 2,755,231
58. U. S. Patent 2,854,319
59. Sachanen, A. N., The Chemical Constituents of Petroleum,
Reinhold Publishing Corporation, 1945.
60. Sweeney, W. J., Petroleum and Its Products, Twenty-Fourth
Annual Priestly Lectures, The Pennsylvania State College,
Phi Lambda Upsilon and The Department of Chemistry, State
College, Pennsylvania, April, 1950.
61. Private Communication, Gulf Oil Co.
62. Henke, A. M., Stauffer, H. C., Carr, N. L., Molecular
Sieves Regenerate Well, Hydrocarbon Processing, May 1962.
63. Baker, R. W., Maher, P. K., Blazek, J. J., Sieves Make Good
Cracking Catalysts, Hydrocarbon Processing, February 1968.
64. Private Communication, Davison Chemical
65- Conser, R. E., Management of Sulfur Emissions The UOP 1973
Technology Conference, September 1973-
66. Blazek, J. J., Gains From FCC Revival Evident Now, The Oil
and Gas Journal, October 8, 1973-
67- Monsanto Files
68. Private Communication with Struther-Wells Corp.
69. Peters, M. S., Timmerhaus, K. D., Plant Design and Economics
for Chemical Engineers, McGraw-Hill Company, 1968.
70. Elkin, H. F., Constable, R. A., Source/Control of Air
Emissions, Hydrocarbon Processing, October 1972.
156
-------
71. Hyne, J. B., Methods for Desulfurization of Effluent Gas
Streams, The Oil and Gas Journal, August 28, 1972.
72. Draemel, D. C., Regeneration Chemistry of Sodium-Based
Double-Alkali Scrubbing Process, NTIS No. PB-220077, March
1973-
73- Calvert, S., Goldshmid, J., Leith, D., Mehta, D., Wet
Scrubber System Study, NTIS No. PB-213017, July 1972.
74. Calvin, E. L., A Process Cost Estimate for Limestone
Slurry Scrubbing of Flue Gas, Part II - Detailed Estimate
Sheets, NTIS No. PB-219017, January 1973-
75- Anon., Detailed Cost Breakdown for Selected Sulfur Oxide
Control Processes, M. W. Kellogg Company, NTIS No.
PB-209 02H, March 1972.
76. Anon., Applicability of S02- Control Processes to Power
Plants, M. W. Kellogg Company, NTIS No. PB-213421, November
1972.
77. Beers, W. D., Characterization of Claus Plant Emissions,
NTIS No. PB-220376, April 1973-
78. Anon., Cost Nomographs of Selected Sulfur Dioxide Abatement
Methods, Hittman Associates, Inc., NTIS No. PB-207108,
January 1972.
79. Byers, R. L. , Licht, W., Processes for Control of Sulfur
Oxide Emissions, AIChE, 1972.
80. Anon., Removal of S02 from Flue Gas, AVCO Space Systems
Division, NTIS No. PB-177^92, November, 1967.
81. Anon., Summaries of Industrial Processes for Treatment of
Sulfur and Nitrogen Containing Effluent Gases, Prepared for
Office of Air Programs, EPA, by Process Research, Inc.,
June, 1972.
82. Anon., Sulfur & S02 Development, CEP Technical Manual,
AIChE, 1971.
83. McGlamery, G. G., Torstrick, R. L., Simpson, J. P., Phillips,
Jr., J. F., Conceptual Design and Cost Study, Sulfur Oxide
Removal from Power Plant Stack Gas, Magnesia Scrubbing -
Regeneration: Production of Concentrated Sulfuric Acid,
EPA-R2-73-244, May 1973-
157
-------
84. Fogel, M. E., et al., Comprehensive Economic Cost Study of
Air Pollution Control Costs for Selected Industries and
Selected Regions, NTIS No. PB-191054, February 1970.
85. Anon., Removal of Sulfur Dioxide from Waste Gases by
Reduction to Elemental Sulfur, Princeton Chemical Research,
Inc., NTIS No. PB-200071, July 1969-
86. Graefe, A. P., et al., The Development of New and/or Improved
Aqueous Processes for Removing S02 from Flue Gases, Volume
II, NTIS No. PB-196781, October 1970.
87. Gressingh, L. E., et al., Applicability of Aqueous Solutions
to the Removal of S02 from Flue Gases, Final Report, Volume
I, NTIS No. PB-196780, October 1970.
88. Anon., Evaluation of S02 - Control Processes, M. W. Kellogg
Company, NTIS No. PB-204711, October 1971.
89. Slack, A. V., Sulfur Dioxide Removal From Waste Gases 1971,
Noyes Data Corporation, 1971.
90. Nonhebel, G., Processes for Air Pollution Control, The
Chemical Rubber Co., 1972.
91. Anon., Sulphur Emission Control - The Choice of Process
Stays Wide Open, Process Engineering, July 1973-
92. Anon., 1972 NPRA Question and Answer Session on Refining
Technology, The Petroleum Publishing Company, 1973-
93. Anon., Processing Discussed by NPRA Panel, Hydrocarbon
Processing, March 1967-
94. Private Communication, Westvaco Corporation.
95. Dautzenberg, F. M., Nader, J. E., Van Ginneken, A. J. J.,
Desulfurization Process: Shell's Flue Gas Desulfurization
Process, Chemical Engineering Progress, August 1971-
96. Jones, H. R., Pollution Control in the Petroleum Industry,
Pollution Technology Review No. 4, Noyes Data Corporation,
Park Ridge, New Jersey, 1973.
97. Collins, J. J., Fornoff, L. L., Miller, W. C., Lovell, D. C.,
The PuraSiv-S Unit Goes On-Stream at Coulton, Paper
presented at the 66th Annual APCA Meeting, Chicago, Illinois,
June 24-28, 1973.
158
-------
98. Miller, W. C., Adsorption Cuts S02, NOX, Hg, Chemical
Engineering, August 6, 1973.
99. Private Communication, Union Carbide Corporation.
100. Private Communication, EPA Control Systems Laboratory.
101. Private Communication, Davy Powergas, Inc.
102. Schneider, R. T., Earl, C. B., Application of The Wellman-
Lord S02 Recovery Process to Stack Gas Desulfurization,
Paper presented at the Flue Gas Desulfurization Symposium,
New Orleans, Louisiana, May 14-17, 1973.
103- Earl, C. B., Potter, B. H., The Wellman-Lord Sulfur Dioxide
Recovery Process, Paper presented at the Industrial Coal
Conference, University of Kentucky, April 18, 1973.
104. Edmisten, N. G., Bunyard, F. L., A Systematic Procedure
for Determining the Cost of Controlling Particulate
Emissions from Industrial Sources, Journal of the Air
Pollution Control Association, July 1970.
105- Guthrie, K. M., Data and Techniques for Preliminary ...
Capital Cost Estimating, Chemical Engineering, March 24, 1969
106. Anon., W -L S02 Pollution Control, Publication of the
Wellman-Power Gas, Inc., 1973-
107- Anon., Tail-Gas Desulfurization Operations Successful, The
Oil and Gas Journal, February 7, 1972.
108. Anon., Combined S02 Removal Process Set for Testing, The
Oil and Gas Journal, February 5, 1973.
109. Anon., S02 Recovery by Wellman-Lord, Company Publication,
1970.
110. Rochelle, G. T., Economics of Flue Gas Desulfurization,
Paper presented at the Flue Gas Desulfurization Symposium,
New Orleans, Louisiana, May 14-17, 1973.
111. Rochelle, G. T., A Critical Evaluation of Processes for
the Removal of S02 from Power Plant Stack Gas, Paper
presented at the 66th Annual Meeting of the APCA,
June 24-28, 1973-
159
-------
112. Burchard, J. K., Rochelle, G. T. , Schofield, W. R. Smith,
J. 0., Some General Economic Considerations of Flue Gas
Scrubbing for Utilities, Control Systems Division, Envir-
onmental Protection Agency, National Environmental Research
Center, Research Triangle Park, North Carolina, October 1972.
113. Koehler, G. R., Operational Performance of the Chemico
Basic Magnesium Oxide Systems at the Boston Edison Company,
Part I, Paper presented at Flue Gas Desulfurization Symposium,
New Orleans, Louisiana, May 14-17, 1973.
114. Quigley, C. P., Operational Performance of the Chemico
Magnesium Oxide System at the Boston Edison Company, Part
II, Paper presented at Flue Gas Desulfurization Symposium,
New Orleans, Louisiana, May 14-17, 1973.
115- McGlamery, G. G., Magnesia Scrubbing Paper presented at Flue
Gas Desulfurization Symposium, New Orleans, Louisiana,
May 14-17, 1973-
116. Anz, B. M., Thompson, Jr., C. C., Pinkston, J. T., Design
and Installation of a Prototype Magnesia Scrubbing
Installation, United Engineers and Constructors, Inc.,
Philadelphia, Pennsylvania, May 15, 1973-
117. Maxwell, M. A., Koehler, G. R., The Magnesia Slurry S02
Recovery Process Operating Experience with a Large Prototype
System, Paper presented at AiChE 65th Annual Meeting, New
York, New York, November 26-30, 1972.
118. Shah, I. S., MgO Absorbs Stackgas S02, Chemical Engineering,
June 26, 1972.
119- Anon., Papers on Zinc Oxide Slurry Scrubbing compiled by
Tennessee Valley Authority, Division of Chemical Development,
Applied Research Branch, Muscle Shoals, Alabama.
120. Schultz, E. A., Miller, W. E., The Cat-Ox Project at Illinois
Power, Paper presented at the Electrical World Technical
Conference, Chicago, Illinois, October 25-26, 1972.
121. Miller, W. E., The Cat-Ox Project at Illinois Power, Paper
presented at Flue Gas Desulfurization Symposium, New Orleans,
Louisiana, May 13-17, 1973-
160
-------
122. Anon., Air Pollution Control for Electric Utilities, Cat-Ox
Systems by Monsanto Enviro-Chem Systems, Inc., Monsanto
publication, 1970.
123- Opferkuch, R. E., Mehta, S. M., Konicek, M. G., Zanders, D. L.,
Applicability of Catalytic Oxidation to the Development of
New Processes for Removing S02 from Flue Gases, Vol. II,
Monsanto Research Corporation, January 1971.
12^. Conser, R. E., Anderson, R. F., New Tool Combats, S02
Emissions, The Oil and Gas Journal, October 29, 1973.
125- U. S. Patent 3,061,421
126. U. S. Patent 2,637,633
161
-------
APPENDIX A
CRACKING CATALYSTS AND THEIR PRODUCERS
A. CLASS 1. ACID-TREATED NATURAL ALUMINOSILICATES AND
SEMISYNTHETICS
American Cyanamid Company
Aerocat 2000: a semisynthetic fluid catalyst containing 35$
A1203. The ABD is 0.60, the surface area 300 m2gm-1, and
the pore volume 0.50 cm3gm~1.
Davison Chemical Division of W. R. Grace & Co.
Grade SS: a semisynthetic fluid catalyst containing 32%
A1203. It is supplied in two different pore volume grades:
0.58 cm3gm~1 (ABD 0-5D and 0.70 cm3gm~1 (ABD 0.*»7). The
surface area of the latter is 280 m2gm-1.
Filtrol Corporation
Grade 58: 17-5? A1203, fluid. The ABD is 0.65, the surface
area 280-300 m2gm~1, and the pore volume 0.36 cm3 gin-1.
Grade 62: 17-5% A1203 as 3/16 x 3/16 inch pellets. The ABD
is 0.8, the surface area 280-300 m2gm~1, and the pore volume
0.36 cm3gm~1.
Grade 63: 38$ A1203 as 3/16 x 3/16 inch pellets. The ABD is
0.8, the surface area 125-135 m2gm~1, and the pore volume
0.27 cm3gm~1.
Grade 80: 38$ A1203, fluid. The ABD is *0.73, the surface
area 125-135 m2gm~1, and the pore volume 0.27 cm3gm-1.
Grade 100: 51$ A1203 microspheres. The ABD is 0.80, the
surface area 105 m2gm~1, and the pore volume 0.37 cm3gm~1.
Grade 110: pellets.
Grade 110: spherical pellets.
162
-------
Houdry Process and Chemical Company
Kao-Pellets: approximately 3/16 x 3/16 inch. The ABD is 0.77
and the surface area 90-100 m2gm-1
Kao-Spheres: 455? A1203 ca. 0.17 inch in diameter. The ABD
is 0.77 and the surface area 90-100 m2gm~1.
Nalco Chemical Company
Nalcat 783= a semisynthetic fluid catalyst, 335? A1203. The
ABD is 0.50, the surface area 280 m2gm-1 , and the pore
volume 0.65-0.70 cm3gm~1.
B. CLASS 2. AMORPHOUS SYNTHETIC SILICA-ALUMINA, INCLUDING SILICA-
MAGNESIA
American Cyanamid Company
Aerocat: 135? A1203, fluid. The ABD is 0.49 and the pore
volume 0.75 cm3gm-1.
Aerocat Triple A: 255? A1203, fluid. The ABD is 0.43 and the
pore volume 0.89 cm3gm~1.
Aerocat 3C-12: 35? MgO in low alumina, fluid.
Aerocat 3C-20: 3% MgO in high alumina, fluid.
Davison Chemical Division of W. R. Grace & Company
Low Alumina: 135? A1203 fluid. The ABD is 0.43 and the pore
volume 0.77 cm3gm~1.
High Alumina: 285? A1203, fluid, in three different pore
volumes: 0-70 cmSgrrr1 (ABD 0.46), 0.78 cm3gm~1 (ABD 0.43),
0.88 cmSgnr1 (ABD 0.39).
SM-30: 27.55? MgO and 35? P, fluid. The ABD is 0.49 and the
pore volume 0.72 cm3gm~1.
163
-------
Houdry Process and Chemical Company
S-46: 13% A1203, tablets. The ABD is 0.62, the surface area
280-315 m2gm-1, and pore volume 0.61 cm3gm~1.
Mobil Chemical Company
Durabead 1: 10/8 A1203, spheres (beads).
Nalco Chemical Company
Nalcat Low Alumina: 13% A1203, fluid. The ABD is 0.40, the
surface area 520 m2gm~1, and the pore volume 0.80-0.85
cm3 gin"1
Nalcat High Alumina: 255? A1203, fluid. The ABD is 0.40-0.44
the surface area 440 m2gm~1, and the pore volume 0.8-0.9
cm3gm~1.
Universal Oil Products Company
Type FC-2: 135? A1203, fluid.
Type FC-3: 25% A1203, fluid.
C. CLASS 3- CRYSTALLINE SYNTHETIC SILICA-ALUMINA COMBINATIONS
American Cyanamid Company
Aerocat S-4: contains rare earth exchanged "Y" type molecular
sieve in a semisynthetic matrix of 335? A1203 content, fluid.
The ABD is 0.53, the surface area 330 m2gm-1, and the pore
volume 0.57 cm3gm~1.
Aerocat TS-150: contains rare earth exchanged "Y" type
molecular sieve in a matrix of synthetic silica-alumina (155?
A1203), fluid. The ABD is 0.49, the surface area 600 rr^girT1
and the pore volume 0.65 cm3gm-1
]64
-------
Aerocat TS-170 and TS-260: contain rare earth exchanged "Y"
type molecular sieve in a semisynthetic matrix with approx-
imately 33% A1203 content, fluid. The ABD is 0-55 and the
pore volume 0.58 cm3gm~1.
Davison Chemical Division of W. R. Grace & Co.
XZ-15: 13% A1203, fluid. The ABD is 0.40, the surface area
500 m2gm~1, and the pore volume 0.88 cm3gm~1
XZ-25: 3655 A1203, fluid. The ABD is 0.5, the surface area
3*10 m2gm-1 and the pore volume 0.60 cm3gm~1.
XZ-36: 36% A1203, fluid. The ABD is 0.55 and the pore volume
0.55 cm3gm~1.
XZ-40: fluid.
Filtrol Corporation
Grade 800: microspheres, 48$ A1203. The ABD is 0.69, the
surface area 210 m2gm~1, and the pore volume 0.39 cm3gm~1.
Grade 810: pellets.
Houdry Process and Chemical Company
HZ-1: pellets.
Mobil Chemical Company
Durabead 6B and Durabead 8: as spheres (beads); D-5 and D-7:
Fluid.
Nalco Chemical Company
KSF Series: "X" type molecular sieve in a matrix, fluid.
KSG Series: "Y" type molecular sieve in a matrix, fluid.
ND-2: low surface area, "Y" type molecular sieve in a matrix,
fluid.
165
-------
APPENDIX B
PREDICTION OF REGENERATOR OFF-GAS S02 CONCENTRATION
An equation has been developed to predict sulfur dioxide con-
centration in volume parts per million (vppm) in the regenerator
flue gas. The variables affecting the sulfur dioxide emissions
have been integrated into the equation and include C02/C0 ratio
and sulfur and hydrogen content on spent catalyst coke. The
equation has been derived based on the following assumptions:
No 02 is present in the regenerator flue gas
The composition of catalyst coke remains constant over
the whole regeneration period
Air is used for catalyst regeneration
The overall material balance for the catalytic cracker regen-
erator can be written
aC + bH2 + dS + e02 + fN2 — »•
gCO + hC02 + dS02 + bH20 + fN2 (Bl)
The following variables have been used in the equation and
were defined as follows
R _ CO, _ h / moles carbon dioxide ^ (B2)
CO g moles carbon monoxide
K • § • f < »iH carbon > * ln °°*e °n sPent oata1^
Applying equations (B2), (B3), and (B4) the material balance
equation (Bl) can be rewritten
166
-------
aC + QaH2 + KaS + e02 + fN2 >•
gCO + hC02 + KaS02 + QaH20 + fN2 (B5)
Other equations which can be derived from equations (Bl)
through (B4) include
a = g + h = g(l + h/g) = g(l + R) (B6)
1 + R (B?)
h = R a
1 + R (B8)
Substituting the equations (B6), (B7), and (B8) into equation
(B5) we obtain
aC + QaH2 + KaS + e02 + fN2 - »•
CO + R(f)C02 + KaS02 + QaH20 + fN2 (B9)
The amount of oxygen in moles required for oxidation of coke
can be calculated from the right side of equation (B9)
e = l/2(jp) + RCip) + Ka + l/2Qa (BIO)
Hence, the nitrogen amount can also be determined
f = 21 e = 21Cl/2(lfR) + R(ITR) + Ka + 1/2 Qa] (B11)
Sulfur dioxide concentration in vppm of dry flue gas can be
determined from the following equation
6
167
-------
After substituting equations (B3), (B7), (B8), and (Bll) into
equation (B12) we obtain
[S02](vppm) =
R(lTR) + Ka + l^/^ITR5 + R(ITR> + Ka + l/2Qa] (B13)
After expressing the variables K and Q in weight amounts and
cancellation of a terms in equation (B13) we obtain
M
_ 2.667 x 1Q6
[S02](vppm) =
3L]
where M = Pounds S = K 32 ffS/mole S =
wnere n pounds c «• x 12 #c/mole C
T - pounds H9 _ Q 2 #H9/mole H? _ , /fi Q
L -- pounds C - Q x 12 #C/mole C ~ 1/6 x Q
The sulfur and hydrogen content of coke are normally expressed
in weight fractions of coke. By converting variables M and L
according to equations (B17) and (Bl8) below
M = S
1-H-S (B17)
T - H
^ 1-H-S (B18)
and substituting these equations into equation (Bl4), the equation
in final form is
S x 106 .
[S02](ppm) = 2>66? + 5_015(2R+1) (1_H_S) + 27.429H + 2.095S (B19)
RT 1
168
-------
where
S = Weight fraction of sulfur in coke
H = Weight fraction of hydrogen in coke
R = Mole ratio of C02 to CO
169
-------
APPENDIX C
FEEDSTOCK DESULFURIZATION TECHNIQUES
Low Sulfur Fuel Oil Demand
The importance of reducing sulfur content in fuel oil is increasing
every year. Comprehensive governmental restrictions on sulfur
emissions that result in air pollution restrict the use of fuel
containing more than 1% sulfur. Recently, some cities on the
east coast of the U.S., where the sulfur problem seems to be the
most serious, have adopted a regulation by which domestic and
industrial users are limited to fuel oils containing 0.3% sulfur
or less. Additionally, no power plants are to be built or expanded
after July 1, 196? unless there is a guaranteed 20-year supply of
fuel whose emission will not exceed 0.35 pound of sulfur dioxide
per million Btu of heat input. Fuel oil containing 0.3$ sulfur
is needed to satisfy this requirement.
The present situation with respect to availability of crude oil
and low sulfur fuel oil is rather turbulent, vague, and undergoing
various political pressures. This situation has produced numerous
discussions on the subject of desulfurization in connection with
the production of low sulfur fuel oil and reduction of air pollution
problems primarily on the east coast of the United States.
It should be noted that the high sulfur content in fuel oil should
not be blamed on the U.S. refiner alone. In 1971 the total
domestic demand for residual fuel oil amounted to 8.0 x 108
barrels of which 5.6 x 108 barrels were imported.21 As indicated
by Table C-l, these oils contain large amounts of sulfur and are
primarily consumed in the eastern United States.
170
-------
Table C-l
SULFUR CONTENT OF U.S. IMPORTED RESIDUAL
FUEL OILS 22 (January-March 1972)
Sulfur
Content
%
0-0.50
0.51-1.00
1.01-2.00
Over 2.00
Million
Barrels
52.2
49-9
22.4
38.0
%
32.1
30.7
13.8
23- 4
There are six types of fuel oils marketed in the U.S. Grade 2
and grade 6 fuel oils are used most. Their average sulfur con-
tents in the year of 1971 are shown in Table C-2.
Table C-2 GRADES AND SULFUR CONTENT OF U.S. FUEL OILS (197D23
Fuel Oil
Grade
Average Sulfur Content,
Weight %
0.07
5 (Light)
5 (Heavy)
0.24
0.65
1.73
1.42
1.59
A distillate oil intended
for vaporizing pot-type
burners and other burners
requiring this grade of
fuel.
A distillate oil for
general-purpose domestic
heating for use in burners
not requiring No. 1 fuel
oil.
Preheating not usually
required for handling or
burning.
Preheating may be required
depending on climate and
equipment.
Preheating may be required
for burning and, in cold
climates, may be required
for handling.
Preheating required for
burning and handling.
171
-------
Table C-2 shows that the 0.3% sulfur requirement will primarily
affect the production of No.5 and 6 fuel oils. The residual fuel
oil, which is derived from the residue obtained from the distilla-
tion of crude petroleum, is commonly designated as No. 5 or No. 6
fuel oil (Figure 2). The residuum from atmospheric distillation
is further distilled in a vacuum distillation column from which
so called vacuum residue is produced. This residue can be
diluted with light oil (cutter stock) or visbroken to meet the
physical property requirements for No. 6 fuel oil.
The removal of sulfur compounds from petroleum feedstocks has
been practiced since the very beginning of the refining industry.
The sulfur compounds must be removed to improve color, odor, and
other qualities of petroleum products, to improve the life of
sulfur-sensitive catalysts, and to reduce formation of gumlike
substance, corrosion, and air pollution. The trends of the past
and the present capacity of the hydrotreating plants in the
United States are shown in Table C-3. These data show that the
feedstock that was primarily desulfurized in 1973 is naphtha
(57-2%), and distillates (22.7%). The feed to catalytic cracking
units and fuel oil represent only a minor portion, 5.3% and 3-5%
of the total desulfurization feed. The crude oil capacity in
the U.S. in the year 1973 has amounted to 14.0 x 106 b/d, from
which 5-2 x 106 bpd or 37% were desulfurized.
Basically, the same technology is applied to desulfurization of
gas oil. Presently, commercially available processes usually
are categorized as either hydrodesulfurization, which was
originally developed for the middle distillate, or hydrocracking.
These processes are summarized in Table C-4.
172
-------
Table C-3
HYDROTREATING PLANT CAPACITIES IN THE UNITED STATES
BY FEEDSTOCK TYPE 9. 13. (bpsd)
Year
Feedstock
1966
1967
1968
1969
1973
Catalytic Cracking
Feed
Naphthas
Distillates
Gas Oil
Lube Oil
Other
Total
111,100
1,872,710
835,670
60,900
100,900
114,165
3, 095, 7^5
180,300
1,916,100
911,800
52,100
105,700
126,200
3,352,800
186,300
2,128,600
1,009,100
97,900
118,700
115,865
3,656,765
186,300
2,358,600
1,029,100
112,900
118,700
115,865
3,921,765
275,800
2,968,280
1,178,850
181,000
119,910
131,618
5,191,188
5-3
57.2
22.7
3-5
2.9
8.1
100.0
Table C-4 DESULFURIZATION PROCESSES AND THEIR LICENSORS
Processes
Gulfining
Hydrofining
Hydrodesulfurization
Isomax
Trickle Hydrodesulfurization
HGO-Unicracking
Unifining
Licensors
Gulf Research and Development
British Petroleum or Esso
Research and Engineering
Institute Francaise du Petrole
Chevron Research or UOP
Shell Development
Union Oil Co. of California
Union Oil Co. of California
and UOP
Only a very few plants are in operation for desulfurization of
residual oil throughout the world. The air pollution regulations
have led to an effort to desulfurize more and more of the heavier
gas oils.
To meet the regulations for the sulfur content in the fuel oil the
petroleum refiner has three alternative courses of action: 1)
use the low sulfur crude oil; 2) blend the high sulfur fuel oil
with the low sulfur light petroleum fraction; and 3) desulfurize
the high sulfur fuel oil. Today, most of the world's crude oil
supply contains a high amount of sulfur (Table C-5).
173
-------
Table C-5 AVERAGE SULFUR CONTENT OF WORLD CRUDE OILS13
Sulfur
WtS?
Kuwait 2.53
Middle East (includes Egypt) 2.03
Middle East (excluding Kuwait) 1.77
West Texas (below 36 API) 1.77
Venezuela 1.70
Mississippi (low gravity) 1.60
West Texas (all) 1.38
California 1.00
United States average 0.75
Canada (28-45 API) 0.44
East Texas field 0.26
Typical Gulf coastal (U.S.) 0.19
North Africa 0.18
Far East 0.09
The low sulfur crudes are in very short supply. The spec-
ifications for the viscosity and flash point for the fuel oil
pretty much limit the amount of low sulfur petroleum fractions
that can be blended with the high sulfur fuel to reduce sulfur
content. These facts limit the options of the petroleum refiners
and force them more and more to desulfurize.
Reduction of the sulfur content of the fuel can be accomplished
in two ways: direct, or indirect desulfurization. Both methods
are presently considered very effective and technically oriented.
174
-------
The direct desulfurization of residual fuel oil is presently under
development and will require additional research, specifically in
the area of catalyst life and regeneration. The overall catalyst
life is now about 25 to ^5 barrels per pound depending upon the
amount of metal in the residual fuel oil. The degree of desulfur-
ization is limited to about 75%- Above this limit, hydrocracking
becomes controlling and hydrogen consumption increases sharply.
Present difficulties are very similar to those of desulfurizing
the crude oil itself. The problem is the presence of asphaltic
compounds and metallic contaminants that quickly deactivate the
catalyst. Although these compounds are diluted in crude oil, the
absolute amounts are the same, and therefore the catalyst life in
crude oil processing is the same as that of residual fuel oil when
the same amount of residual fuel oil is produced. The operating
conditions for crude oils are also the same. The only difference
is in the liquid hourly space velocity, which will result in
larger plant requirements if desulfurization of the crude oil
is performed.
The indirect hydrodesulfurization of residual fuel oil is con-
sidered most attractive at the present time. In this process as
deep a cut as possible is made in the vacuum gas oil. This stream
is later desulfurized. The desulfurized product is blended back
with vacuum residuals to reduce the sulfur content of the residual
fuel oil. This technology is very well developed, but the degree
of desulfurization is limited to about 40 to H5% even if 97% of
the sulfur is removed from the vacuum gas oil. This is obviously
substantially lower than in the direct desulfurization process.
The desulfurization of vacuum gas oil is also used as pretreatment
of catalytic cracker feedstock to improve this unit operation.
The technology of desulfurization of this stream is well established
and seems to be free of serious problems.
175
-------
Relatively high sulfur removal can be obtained and the catalyst
life is fairly long: 200 to 400 barrels per pound for gas oil,
and 150 to 350 barrels per pound for vacuum gas oil. Hydrogen
consumption is lower — about 484 scf/bbl 13
Variables affecting the rate of desulfurization include type of
feedstock, temperature, total pressure, hydrogen-to-oil ratio,
liquid hourly space velocity, and type of catalyst.
At present insufficient data are available to relate the reaction
rate constant to the properties of feedstock such as gravity,
sulfur content, or boiling range. However, it is generally
observed that nonthiophenic sulfur compounds are removed more
rapidly than the thiophenic compounds, and that the rate of
desulfurization decreases with increasing molecular weight of the
petroleum fraction being processed. Table C-6 presents some
quantitative data on the difficulty of feedstock desulfurization
as a function of its origin. It is interesting to note that the
degree of desulfurization of a gas oil is a function of the
original sulfur content of the gas oil only and is not influenced
by the source of the crude oil from which the gas oil is derived.21*
From this it can be concluded that in order to desulfurize
various stocks or crude oils to the same concentration of sulfur,
conditions of varying severity would have to be applied.
The upper temperature limit for desulfurization is about 800°F.
The temperatures above this point would result in substantial
hydrocracking and higher consumption of hydrogen. Higher hydrogen
pressure favors the desulfurization but higher oil partial
pressure retards the reaction. Generally, if the sulfur content
of the feed increases, and the feedstocks become heavier, the
higher total pressures are necessary to obtain efficient sulfur
removal.
176
-------
Table C-6 REDUCED CRUDE SULFUR CONTENTS/EASE OF DESULFURIZATION20
Reduced Crude
Amposta (Spanish)
Arabian Heavy
Khafji
Kuwait
Kirkuk
Arabian Light
Iranian Heavy
Iranian Light
Qatar
Tia Juana (Venezuela)
Zakum
Murban
Es Sider
Brega
Forcados (Nigerian)
Hassi Messaoud
Sarir
Relative Ease of Desulfurization to
Tia Juana
Iranian Heavy
Khafji
Arabian Heavy
Kirkuk
Iranian Light
Amposta
Kuwait
Arabian Light
Qatar
Zakum
Murban
Sulfur. wt%
6.7
4.4
4.2
4.0
3-7
3-0
2.5
2.4
2-3
2.2
2.0
1-5
0.8
0.4
0.4
0.35
0.25
Difficult
Easy
* Relative ease of direct desulfurization considering
sulfur content, molecular structure, metals content, etc.
Another important variable is hydrogen-to-oil ratio. At con-
stant pressure, temperature, and liquid space velocity, the
hydrogen-oil ratio affects the fraction of hydrocarbons vaporized,
the hydrogen partial pressure, and the catalyst contact time.
As the hydrogen-to-oil ratio is increased, the fraction of hydro-
carbons vaporized increases together with hydrogen partial
pressure, and the catalyst contact time decreases.
177
-------
As the feedstocks become heavier, the lower space velocities are
required, but at low space velocities the higher degree of de-
sulfurization may be accompanied by increased cracking of the
feedstocks and increased coke deposition on the catalyst because
of the longer contact time. This fact suggests that an optimum
hydrogen-to-oil ratio and liquid hourly space velocity exist for
a specific desulfurization case.
Typical operating conditions for desulfurization of different
types of feedstock are summarized in Table C-7.
For catalytic cracking feedstock the desulfurization of the stream
would result in 1) an increase in gasoline yield, 2) elongation
of catalyst life, and 3) improvement of product quality. It has
been reported that with the hydrogenated feed and the conventional
catalyst, the gasoline yield can increase by about 6%. When the
feedstock had been hydrogenated and a zeolite catalyst employed,
the gasoline yield was reported to increase by about 10$ in com-
parison with the non-hydrogenated feedstock and the conventional
catalyst -13
The benefits of desulfurization of feed to the FCC unit are mainly
due to saturation of polycyclic aromatics, which will form car-
bonaceous deposits on the catalyst, and to the removal of metallic
contaminants, which result in shortened catalytic cracking
catalyst life. In addition to benefits of desulfurization on
FCC unit operation and production already mentioned, the quality
of the gasoline and gas oil from a hydrogenated feedstock is
usually superior to that from an untreated feed. Catalytic
gasoline from the hydrogenated feed is higher in octane number
and lower in sulfur content. For the light gas oils, sulfur is
decreased appreciably, diesel index is higher, and color stability
is improved.
178
-------
Table C-7
TYPICAL DESULFURIZATION CONDITIONS FOR VARIOUS
FEEDSTOCKS1 3
Straight run naphtha
Straight run kerosene
Straight run atmospheric
gas oil
Straight run vacuum gas
oil
Residual fuel oil by
fixed bed reactor
Residual fuel oil by
ebullated bed reactor
Crude oil
Temperature
op
tha 690
sene 690
Total
Pressure
psig
350
600
H2/Oil Ratio
scf/bbl
800
800
690
750
765
765
765
875
1000
2000
2000
2000
LHSV
hr-1
8
4
995
4,000
6,3^2
6,3^2
6,342
1-52
1
0.62
1.54
1.03
It can be concluded that the desulfurization of the FCC feed is not
used extensively. The primary reason a refinery desulfurizes is to
reduce the sulfur content of heavier fuel oils to meet air pollution
regulations. This can be done to some degree either by a direct
desulfurization of reduced crude or by indirect desulfurization
of deep heavy gas oil fractions which can be blended with vacuum
residuum. Neither of these techniques interferes significantly
with fluid catalytic cracking and it cannot be expected that
strict air pollution regulations imposed on fuel oil sulfur con-
tent will automatically solve or even affect the sulfur emission
problems of FCC units.
179
-------
However, if we theorize and take Middle East crudes (Kuwait,
Khafji) in our example (this crude would produce a residual fuel
oil containing about H% sulfur), vacuum gas oil of 3.2% sulfur,
and straight run gas oil of 1.2% sulfur, 13 the following can be
said. Assuming that 80% desulfurization of residual fuel oil is
obtained, this would result in 0.8$ sulfur in the stream leaving
the desulfurization unit. If the refinery intends to process
primarily fuel oil, it is very questionable that this stream will
be utilized for gasoline production and fed to a catalytic
cracking unit. The straight run gas oil stream as well as the
desulfurized stream if fed into the catalytic cracking unit
would still result in sulfur emission in the range of 700-1000
ppm in the flue gas leaving the regenerator. In order to reduce
this emission to 200 ppm level either more efficient desulfur-
ization of the feed would be required or additional desulfurization
of the flue gas would be necessary. Thus, in some cases, applying
desulfurization to the FCC feed may not be the final solution to
reduction of SOX emissions from regenerator flue gas.
180
-------
APPENDIX D
ECONOMIC EVALUATION OF FCC FEEDSTOCK DESULFURIZATION
DETAILED ESTIMATES
Table D-l HYDRODESULFURIZATION OF FCC FEEDSTOCK (HEAVY GAS OIL)
TOTAL CAPITAL INVESTMENT
PLANT CAPACITY = 10,000 BPSD AT 90% CAPACITY
Direct Costs
Desulfurization Section $2,425,800
Recovery Section 1,033,200
Total Process and Utilities Cost 3,459,000
General Service Facilities 518,900
Total Fixed Capital Investment $3,977,900
Indirect Costs
Interest on Construction Loan 159,100
Start-Up Costs 397,800
Working Capital 417,700
Total Capital Investment* $4,952,500
* Does Not Include Land
181
-------
Table D-2 OPERATING COST SUMMARY
HYDRODESULFURIZATION OP HEAVY GAS OIL
CAPACITY: 10,000 BPSD AT 90? CAPACITY
FIXED CAPITAL INVESTMENT $4,952,500
SULFUR CONTENT (WT5S IN/ WTJS OUT) 3-36/2.43
Labor
1 Operating $96,400
2 Maintenance 69,200
3 Control Laboratory 19,300
4 Total Labor 184,900
Materials
5 Raw and Process
6 Treatment Loss 300,400
7 Hydrogen 328,100
8 Catalyst 63,900
9 Maintenance 69,200
10 Operating 9,600
11 Total Materials 771,200
Utilities
12 Cooling Water 42,200
13 Process Water 2,200
14 Electricity 66,000
15 Fuel 166,700
16 Total Utilities 277,100
17 Total Direct Operating Cost (4,11,&16) $1,233,200
18 Plant Overhead 147,900
19 Taxes and Insurance 99*100
20 Plant Cost (17, 18, & 19) 1,381,100
21 General & Administrative, Sales, Research 297,200
22 Cash Expenditures (20 & 21) 1,678,300
23 Depreciation 495,300
24 Interest on Working Capital 25,100
25 Total Operating Costs* $2,198,700
26 Cost: (Cents/bbl) 70.91
* Does not include any by-product credit or recovery cost
182
-------
Table D-3 OPERATING COST SUMMARY
HYDRODESULFURIZATION OP HEAVY GAS OIL
CAPACITY: 10,000 BPSD AT 90% CAPACITY
FIXED CAPITAL INVESTMENT $4,952,500
SULFUR CONTENT (WT# IN/ WT!6 OUT) 3-36/0.243
Labor
1 Operating $96,400
2 Maintenance 69,200
3 Control Laboratory 19,300
4 Total Labor 184,900
Materials
5 Raw and Process
6 Treatment Loss 300,400
7 Hydrogen 1,099,500
8 Catalyst 63,900
9 Maintenance 69,200
10 Operating 9,600
11 Total Materials 1,542,600
Utilities
12 Cooling Water 42,200
13 Process Water 2,200
14 Electricity 66,000
15 Fuel 166,700
16 Total Utilities 277,100
17 Total Direct Operating Cost (4,11,&16) $2,004,600
18 Plant Overhead 147,900
19 Taxes and Insurance 99,100
20 Plant Cost (17, 18, & 19) 2,152,500
21 General & Administrative, Sales, Research 297,200
22 Cash Expenditures (20 & 21) 2,449,700
23 Depreciation 495,300
24 Interest on Working Capital 25,100
25 Total Operating Costs* $2,970,100
26 Cost: (Cents/bbl) 94.72
* Does not include any by-product credit or recovery cost
183
-------
Table D-4 HYDRODESULFURIZATION OF FCC FEEDSTOCK (HEAVY GAS OIL)
TOTAL CAPITAL INVESTMENT
PLANT CAPACITY = 50,000 BPSD AT 905? CAPACITY
Direct Costs
Desulfurization Section $7,131,200
Recovery Section 3,037,200
Total Process and Utilities Cost 10,168,400
General Service Facilities 1,525,300
Total Fixed Capital Investment $11,693,700
Indirect Costs
Interest on Construction Loan 467,800
Start-Up Costs 1,169,400
Working Capital 1,'227,800
Total Capital Investment* $14,558,700
* Does Not Include Land
184
-------
Table D-5 OPERATING COST SUMMARY
HYDRODESULFURIZATION OF HEAVY GAS OIL
CAPACITY: 50,000 BPSD AT 90% CAPACITY
FIXED CAPITAL INVESTMENT $14,558,700
SULFUR CONTENT (WT5? IN/ WT5S OUT) 3.36/2.43
Labor
1 Operating $96,400
2 Maintenance 203,400
3 Control Laboratory 19,300
4 Total Labor 319,100
Materials
5 Raw and Process
6 Treatment Loss 901,200
7 Hydrogen 984,100
8 Catalyst 191,700
9 Maintenance 203,400
10 Operating 9,600
11 Total Materials 2,290,000
Utilities
12 Cooling Water 126,700
13 Process Water 6,700
14 Electricity 330,000
15 Fuel 833,300
16 Total Utilities 1,296,700
17 Total Direct Operating Cost (4,11,&16) $3,905,800
18 Plant Overhead 225,300
19 Taxes and Insurance 233,900
20 Plant Cost (17, 18, & 19) 4,365,000
21 General & Administrative, Sales, Research 701,600
22 Cash Expenditures (20 & 21) 5,066,600
23 Depreciation 1,169,400
24 Interest on Working Capital 73,700
25 Total Operating Costs* $6,309,700
26 Cost: (Cents/bbl) 38.94
* Does not include any by-product credit or recovery cost
185
-------
Table D-6 OPERATING COST SUMMARY
HYDRODESULFURIZATION OF HEAVY GAS OIL
CAPACITY: 50,000 BPSD AT 90% CAPACITY
FIXED CAPITAL INVESTMENT $14,558,700
SULFUR CONTENT (WT5? IN/ WT$ OUT) 3-36/0.243
Labor
1
2
3
4
Operating
Maintenance
Control Laboratory
Total Labor
$96,400
203,^00
19,300
319,100
Materials
5
6
7
8
9
10
11
Raw and Process
Treatment Loss
Hydrogen
Catalyst
Maintenance
Operating
Total Materials
901,200
3,298,400
191,700
203,400
9,600
4,604,300
Utilities
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Cooling Water
Process Water
Electricity
Fuel
Total Utilities
Total Direct Operating Cost (4,11,&16)
Plant Overhead
Taxes and Insurance
Plant Cost (17, 18, & 19)
General & Administrative, Sales, Research
Cash Expenditures (20 & 21)
Depreciation
Interest on Working Capital
Total Operating Costs*
Cost: (Cents/bbl)
126,700
6,700
330,000
833,300
1,296,700
$6,220,100
225,300
233,900
6,679,300
701,600
7,380,900
1,169,400
73,700
$8,624,000
53-22
* Does not include any by-product credit or recovery cost
186
-------
Table D-7 HYDRODESULFURIZATION OF FCC FEEDSTOCK (HEAVY GAS OIL)
TOTAL CAPITAL INVESTMENT
PLANT CAPACITY = 150,000 BPSD AT 90S? CAPACITY
Direct Costs
Desulfurization Section $14,887,900
Recovery Section 6,340,800
Total Process and Utilities Cost 21,228,700
General Service Facilities 3,184,300
Total Fixed Capital Investment $24,413,000
Indirect Costs
Interest on Construction Loan 976,500
Start-Up Costs 2,141,300
Working Capital 2,563,400
Total Capital Investment* $30,394,200
* Does Not Include Land
187
-------
Table D-8 OPERATING COST SUMMARY
HYDRODESULFURIZATION OF HEAVY GAS OIL
CAPACITY: 150,000 BPSD AT 90$ CAPACITY
FIXED CAPITAL INVESTMENT $30,394,200
SULFUR CONTENT (WT# IN/ WT5& OUT) 3.36/2.^3
Labor
1
2
3
4
Operating
Maintenance
Control Laboratory
Total Labor
$96,400
424,600
19,300
540,300
Materials
5
6
7
8
9
10
11
Raw and Process
Treatment Loss
Hydrogen
Catalyst
Maintenance
Operating
Total Materials
2,703,600
2,952,400
575,100
424,600
9,600
6,665,300
Utilities
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
Cooling Water
Process Water
Electricity
Fuel
Total Utilities
Total Direct Operating Cost (4
Plant Overhead
Taxes and Insurance
Plant Cost (17, 18, & 19)
General & Administrative, Sales,
Cash Expenditures (20 & 21)
Depreciation
Interest on Working Capital
Total Operating Costs*
Cost: (Cents/bbl)
380,100
20,100
990,000
2,499,900
3,890,100
,11,&16) $11,095,700
432,300
488,300
12,016,300
Research 1,464,800
13,481,100
2,441,300
153,800
$16,076,200
33.10
* Does not include any by-product credit or recovery cost
188
-------
Table D-9 OPERATING COST SUMMARY
HYDRODESULFURIZATION OF HEAVY GAS OIL
CAPACITY: 150,000 BPSD AT 9055 CAPACITY
FIXED CAPITAL INVESTMENT $30,39^,200
SULFUR CONTENT (WTJK IN/ WT5& OUT) 3.36/0.243
Labor
1 Operating $96,400
2 Maintenance 424, 600
3 Control Laboratory 19,300
H Total Labor 540,300
Materials
5 Raw and Process
6 Treatment Loss 2,703,600
7 Hydrogen 9,895,200
8 Catalyst 575,100
9 Maintenance 424,600
10 Operating 9,600
11 Total Materials 13,608,100
Utilities
12 Cooling Water 380,100
13 Process Water 20,100
14 Electricity 990,000
15 Fuel 2,499,900
16 Total Utilities 3,890,100
17 Total Direct Operating Cost (4,11,&16) $17,^98,200
18 Plant Overhead 432,300
19 Taxes and Insurance 488,300
20 Plant Cost (17, 18, & 19) 18,418,800
21 General & Administrative, Sales, Research 1,464,800
22 Cash Expenditures (20 & 21) 19,883,600
23 Depreciation 2,441,300
24 Interest on Working Capital 153,800
25 Total Operating Costs* $22,478,700
26 Cost: (Cents/bbl) 47.39
* Does not include any by-product credit or recovery cost
189
-------
APPENDIX E
ECONOMIC EVALUATIONS OF CATALYST STEAM CONTACTING (STRIPPING)
DETAILED ESTIMATES
E-l. THE CAPITAL INVESTMENT COST
The capital and operating cost estimates of the secondary steam
stripping technique have been prepared for two cases, typical
and worst, as discussed and defined in Section 5.3.3.
The capital investment cost estimate for the typical case and a
45,000 bpsd nominal size FCC unit has been obtained by cost
estimating the major equipment necessary for the process. This
cost is designated as purchased equipment cost. Fixed capital
investment cost has been calculated by applying a fixed capital
investment factor69 of 4.8 to the purchased equipment cost. The
summary of purchased equipment cost determinations and the total
capital investment cost data for this case are summarized in
Table E-5-
The capital investment cost for the worst case and the same size
FCC unit (45,000 bpsd) has been calculated by assuming a scaling
factor of 0.6? and applying it to the fixed capital investment
for the typical case. The cost for the worst case is summarized
in Table E-7. Capital investment cost estimates for both the
typical and the worst case, and for 10,000 and 150,000 bpsd FCC
unit nominal sizes were obtained by applying the scaling factor
of 0.67 to the corresponding fixed capital investment cost
figures determined for the 45,000 bpsd unit. The capital in-
vestment data are summarized in Tables E-l, E-3, E-9, and E-ll.
190
-------
El.l Purchased Equipment Cost for 45,000 bpsd FCC Unit, Typical
Case
a. Catalyst Stripper
(1) Assumptions
- Catalyst-to-oil ratio (C/0) is 6 Ib of catalyst per
Ib of total feed for the typical case, and 12 Ib of
catalyst per Ib of total feed for the worst case
- Steam stripping required to achieve 200 ppm sulfur
emissions in regenerator off-gas is 4 Ib of steam per
100 Ib of catalyst
- Steam line pressure, 125 psig
- Sulfur content of coke before steam stripping, 1.5 wt/?
- Sulfur content of coke after steam stripping, 0.243 wt$
- Stripper operating temperature, 1000°F
- Stripper operating pressure, 40 psia
- Velocity in stripper feed transfer lines, 40 ft/sec
- Velocity in stripper, 2 ft/sec
- Depth of fluidized bed in the stripper was assumed
to be 10 ft with the fluid bed occupying 50% of the
total stripper volume
- Vapor velocity in lines leaving the stripper, 100 ft/sec
- Velocity in stripper standpipe, 7 ft/sec
- Stripper bed density, 15 Ib/cu ft
- Catalyst density in the standpipe, 35 Ib of catalyst
per cu ft
- Hydrogen sulfide produced in the steam stripper will
be fed into existing Claus unit and no additional
cost was assumed to be needed for expansion of this
facility
191
-------
(2) Calculations
Catalyst Circulation Rate (OCR). Ib/hour
CCR (Ib/hr) =
(catalyst to oil ratlo)x(300 Ib per barrel of oll)x(barrel per day of feed oil)
(24 hours per day)
CCR (Ib/hr) = 75 x (45000) = 3-375 x 106
Steam Stripping Rate (SSR). Ib/hour
SSR (Ib/hr) = (Ib of steam per Ib of catalyst) x CCR
SSR (Ib/hr) = 0.04 x 3-375 x 106 = 135,000
Volumetric Flow of Steam (VF), acfs
SSR x 359 cuft/lb mole x (1000+460) x 11.7
VF (ACPS> • 18 IbAb mole x 3600 —
VF (ACFS) = 6.46 x 10~3 x 135000 = 872
Cross-sectional Area (A, ) and Diameter (DT) of Feed
Transfer Lines
VT?
AL (sq ft) = 40 ft/sec = °'025 X 872 = 21'8
DT (ft) = / 4AT = 1.128 x /~AT = 5.27
LI —L LI
T[
Cross-sectional Area (A0) and Diameter (D^) of the
Stripper bs
TTfjl
AS (sq ft) = 2 ft/see = °'5 X 8?2 = *36
DS (ft) = 1.128 x /Ag = 23.6
192
-------
Cross-sectional Area (AE) and Diameter (D ) of Lines
Leaving the Stripper
VT?
AE (S1 ft) ' 100 ft/sec ' °'01 x 872 - 8'72
D (ft) = 1.128 x /"AT" = 3-33
Hi Hi
Cross-sectional Area (Ap) and Diameter (Dp) of Stripper
Standpipe
A ( <*n f 1- 1 = —> , OCR
P v 4 ' 3600 x (bed density) x (velocity)
AP ««« "> • ^SV35°x 7 ' 3'83
Dp (ft) = 1.128 x /Tp~~ = 2.21
Catalyst Inventory (CD in the Stripper
„,. ,. ^ Ap x 10 ft x (stripper bed density)
CI (tons) = -S2000
CI (tons) = 0.075 x 436 = 32.7
The cost of the steam stripper was estimated based on weight of
this equipment. The unit price for 5% Cr, 1/2% Mo steel, which
was assumed to suit this application, was estimated at 53-6
-------
b. Condenser
The area of the condenser (sq ft) was calculated according to
the following assumptions.
- Stripping steam will be cooled from 1000°F to 100°F and
condensed
- The cooling water at 80°F will enter the condenser and
will be evaporated to produce saturated steam at 353°F
- The overall heat transfer coefficient was assumed to be
700 Btu/ft2 hr°F
- The cost of heat exchanger was assumed at $7-75/sq ft67
(5% Cr, 1/2% Mo steel)
Calculation of Heat Transfer Area
Superheated Steam
15 psig, 1000°F
1533 Btu/lb
Saturated Steam
125 psig, 353 °F
1193 Btu/lb
HX
Water 100°F
68 Btu/lb
Process Water
80°F,H8 Btu/lb
Using the heat transfer equation
Q = UA AT,
m
Where
= (1000-353) -(100-80)
In 1000-353
100-80
= 180°F
Q = 135000 x (1533-68) = 197-8 x 106 Btu/hr
A . _Q_ . 197.8 x 1Q6 m 15?0 sq ft
UAT.
m
700 x 180
-------
Cost $ = 7-75 x 1570 = 12,165
Water requirement = 1193^8 1Q& " 176 x 103 Ib/hr
= 351 gpm
The unit will produce 41,000 Ib/hr of 125 psig steam.
The pump delivering this amount of water through the
condenser and superheater operating at 125 psig will
have 50 HP.
c . Steam Superheater
It was assumed that the saturated steam from the condenser will
be used as the feed for the superheater. The cost of the 90 x
106 Btu/hr superheater was estimated at $200, 000. 68 Natural gas
was considered as the fuel.
Heat input required for the superheater was calculated as follows
Q = 135,000 Ib/hr x (1533 - 1193) = ^5.9 x 106 Btu/hr
Superheater scale down:
$ = 200,000 (^2-)0'67 = $127,380
The amount of natural gas was approximated based on 15$ heat
loss and 1000 Btu/cu ft natural gas heating value:
Natural Gas = 1>15 x 9 x 106 x 2** = 1.27 million cu ft/day
195
-------
d. Phase Separator
A 500-gallon stirred tank was assumed to be used for phase separa-
tion. The tank is made of 5% Cr, 1/2% Mo steel. The price of
this tank was estimated at $5,000.67
e. Acidifier
Epoxy-resin-lined, carbon-steel, stirred, 500-gallon tank was
assumed to suit this application. The cost of this tank was
assumed to be $3350.67 7-35 Ib/hr of acid sludge is required to
acidify the contents of the acidifier to pH 3. This amount was
calculated based on the assumption that the sludge will contain
905? sulfuric acid.
f. Neutralizer
A 500-gallon, carbon steel, stirred tank was assumed to suit this
purpose. The cost of this tank was estimated to be $1930-67
The amount of lime needed to neutralize the acid sludge was
calculated to be 5 Ib/hr.
g. Clarifier
The mass flow rate through the clarifier was assumed to be 8^
gal/sq ft hour . 67 The area of the clarifier (sq ft) was de-
termined as follows:
„ 135,000 Ib/hr ^ f
8.3 Ib/gallon x 84 gallons/sq ft hr ±yq sq IT;
From this the diameter of the tank was calculated to be 15-7 ft
^ 16 ft. The depth of the clarifier was assumed to be 10 ft.
The cost of this vessel was assumed to be $5>000 ,67
196
-------
h. Vacuum Filter
It was assumed that catalyst fines can be filtered by a vacuum
filter operating at the load of 15 Ib cake/sq ft hour with 70%
moisture in the cake.
Assuming 0.1 Ib of catalyst per barrel of oil feed will be carried
out from the steam stripper', the weight of filter cake and area
of filter can be determined as follows:
0.1 X JI5000 _ (-^ 1K/V,«,,r.
2*1 x 0.3 ~ 625 lb/hour
and sq ft
The cost of this equipment was assumed to be $13»900. 67
E-2 OPERATING COST
The operating cost estimates were individually calculated for
each case and FCC unit size and are summarized in Tables E-2,
E-4, E-6, E-8, E-10, and E-12.
In addition to the operating cost assumptions summarized in
Section 3-5.2, the following data were applied in operating
cost preparation:
Catalyst loss was assumed 0.1 Ib of catalyst per barrel
of oil feed for the typical case and 0.2 Ib of catalyst
per barrel of oil feed for the worst case
The cost of catalyst was assumed to be $600 /ton
Operating labor - 2 men per shift
197
-------
The cost of raw materials and utilities was assumed
to change proportionally with the size of the FCC
unit
The cost of sulfuric acid was assumed at $37/ton (as
100$
The cost of lime was assumed to be $19-50/ton
198
-------
Table E-l SUMMARY OP CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 10,000 BPSD
TYPICAL STRIPPING OPERATION
Fixed Capital Investment $360,100
i
Initial Catalyst Cost 4,400
Start-Up Cost 36,000
Working Capital 37,800
Interest on Construction Loan 14, 400
Total Investment $452,700
199
-------
Table E-2 SUMMARY OP ANNUAL OPERATING COSTS
FCC UNIT SIZE: 10,000 BPSD AT 90% CAPACITY
TYPICAL STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $360,100
Labor
1 Operating $96,400
2 Maintenance 7,200
3 Control Laboratory 19,300
4 Total Labor 122,900
Materials
5 Raw and Process
6 Acid Sludge 200
7 Lime 100
8 Catalyst Replacement 99,000
9 Maintenance 7,200
10 Operating 9,600
11 Total Materials 116,100
Utilities
12 Process Water 11,000
13 Electricity 700
14 Fuel 31,700
15 Total Utilities 43,400
16 Total Direct Operating Cost (4,11,&15) $282,400
17 Plant Overhead 98,300
18 Taxes and Insurance 7,200
19 Plant Cost (16, 17, & 18) 387,900
20 General & Administrative, Sales, Research 21,600
21 Cash Expenditures (19 & 20) 409,500
22 Depreciation 36,000
23 Interest on Working Capital 2,300
24 Total Operating Costs* (21,22,& 23) $447,800
25 Cost: (Cents/bbl) 13-84
* Does not include by-product credit or recovery costs
200
-------
Table E-3 SUMMARY OF CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 10,000 BPSD
WORST STRIPPING CASE
Fixed Capital Investment $572,900
Initial Catalyst Cost 8,700
Start-Up Cost 57,300
Working Capital 60,200
Interest on Construction Loan 22,900
Total Investment $722,000
201
-------
Table E-4 SUMMARY OP ANNUAL OPERATING COSTS
FCC UNIT SIZE: 10,000 BPSD AT 90$ CAPACITY
WORST STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $572,900
Labor
1 Operating $96,400
2 Maintenance 11,500
3 Control Laboratory 19,300
4 Total Labor 127,200
Materials
5 Raw and Process
6 Acid Sludge 400
7 Lime 200
8 Catalyst Replacement 198,000
9 Maintenance 11,500
10 Operating 9,600
11 Total Materials 219,700
Utilities
12 Process Water 22,000
13 Electricity 1,400
14 Fuel 63,400
15 Total Utilities 86,800
16 Total Direct Operating Cost (4,11,&15) $433,700
17 Plant Overhead 101,800
18 Taxes and Insurance 11,500
19 Plant Cost (16, 17, & 18) 547,000
20 General & Administrative, Sales, Research 34.400
21 Cash Expenditures (19 & 20) 581,400
22 Depreciation 57,300
23 Interest on Working Capital 3,600
24 Total Operating Costs* (21,22,& 23) $642,300
25 Cost: (Cents/bbl) 19-82
* Does not include by-product credit or recovery costs
202
-------
Table E-5 SUMMARY OF CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 45,000 BPSD
TYPICAL STRIPPING OPERATION
PURCHASED EQUIPMENT COST
Fluid Bed Steam Stripper $36,770
Condenser 12,170
Steam Superheater 127,380
Phase Separator 5,000
Acidifier 3,350
Neutralizer 1,930
Clarifier 5,000
Vacuum Filter 13,900
Total $205,500
Fixed Capital Cost (4.8 Factor) 986,400
Initial Catalyst Cost 19,600
Start-Up Cost 98,640
Working Capital 103,570
Interest on Construction Loan 39,460
Total Investment $1,247,670
203
-------
Table E-6 SUMMARY OF ANNUAL OPERATING COSTS
FCC UNIT SIZE: 45,000 BPSD AT 90$ CAPACITY
TYPICAL STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $986,400
Labor
1 Operating $96,400
2 Maintenance 19,700
3 Control Laboratory 19,300
4 Total Labor 135,400
Materials
6
8
9
10
11
Raw and Process
Acid Sludge
Lime
Catalyst Replacement
Maintenance
Operating
Total Materials
1,000
400
445,500
19,700
9,600
476,200
Utilities
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Process Water
Electricity
Fuel
Total Utilities
Total Direct Operating Cost (4,11,&15)
Plant Overhead
Taxes and Insurance
Plant Cost (16, 17, & 18)
General & Administrative, Sales, Research
Cash Expenditures (19 & 20)
Depreciation
Interest on Working Capital
Total Operating Costs* (21,22,& 23) $
Cost: (Cents/bbl)
49,300
3,200
142,500
195,000
$806,600
108,300
19,700
934,600
59,200
993,800
98,600
6,200
1,098,600
7.57
* Does not include by-product credit or recovery costs
204
-------
Table E-7 SUMMARY OF CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 45,000 BPSD
WORST STRIPPING OPERATION
Fixed Capital Investment $1,569,400
Initial Catalyst Cost 39,200
Start-Up Cost 156,900
Working Capital 164,800
Interest on Construction Loan 62,800
Total Investment $1,993,100
205
-------
Table E-8 SUMMARY OF ANNUAL OPERATING COSTS
FCC UNIT SIZE: 45,000 BPSD AT 90>6 CAPACITY
WORST STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $1,569,400
Labor
1 Operating $96,400
2 Maintenance 31,400
3 Control Laboratory 19,300
1 Total Labor 147,100
Materials
5 Raw and Process
6 Acid Sludge 2,000
7 Lime 800
8 Catalyst Replacement 891,000
9 Maintenance •31,400
10 Operating 9,600
11 Total Materials 934,BOO
Utilities
12 Process Water 98,600
13 Electricity 6,400
14 Fuel 285,000
15 Total Utilities 390,000
16 Total Direct Operating Cost (4,11,&15) $1,471,900
17 Plant Overhead 117,700
18 Taxes and Insurance 31,400
19 Plant Cost (16, 17, & 18) 1,621,000
20 General & Administrative, Sales, Research 94,200
21 Cash Expenditures (19 & 20) 1,715,200
22 Depreciation 156,900
23 Interest on Working Capital 9,900
24 Total Operating Costs* (21,22,& 23) $1,882,000
25 Cost: (Cents/bbl) 12.93
* Does not include by-product credit or recovery costs
206
-------
Table E-9 SUMMARY OF CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 150,000 BPSD
TYPICAL STRIPPING OPERATION
Fixed Capital Investment $2,209,900
Initial Catalyst Cost 65,300
Start-Up Cost 221,000
Working Capital 232,000
Interest on Construction Loan 88,^00
Total Investment $2,816,600
207
-------
Table E-10 SUMMARY OF ANNUAL OPERATING COSTS
FCC UNIT SIZE: 150,000 BPSD AT 90% CAPACITY
TYPICAL STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $2,209,900
Labor
1 Operating $96,400
2 Maintenance 44,200
3 Control Laboratory 19,300
Total Labor 159,900
Materials
5 Raw and Process
6 Acid Sludge 3,300
7 Lime 1,300
8 Catalyst Replacement 1,485,000
9 Maintenance 44,200
10 Operating 9,600
11 Total Materials 1,543,400
Utilities
12 Process Water 164,300
13 Electricity 10,700
14 Fuel 475,000
15 Total Utilities 6507000
16 Total Direct Operating Cost (4,11,&15) $2,353,300
17 Plant Overhead 127,900
18 Taxes and Insurance 44,200
19 Plant Cost (16, 17, & 18) 2,525,400
20 General & Administrative, Sales, Research 132,600
21 Cash Expenditures (19 & 20) 2,658,000
22 Depreciation 221,000
23 Interest on Working Capital 13,900
24 Total Operating Costs* (21,22,& 23) $2,892,900
25 Cost: (Cents/bbl) 5-96
* Does not include by-product credit or recovery costs
208
-------
Table E-ll SUMMARY OF CAPITAL INVESTMENT COSTS
FCC UNIT SIZE: 150,000 BPSD
WORST STRIPPING OPERATION
Fixed Capital Investment $3,516,100
Initial Catalyst Cost 130,700
Start-Up Cost 351,600
Working Capital 369,200
Interest on Construction Loan 1*10,600
Total Investment $4,508,200
209
-------
Table E-12 SUMMARY OF ANNUAL OPERATING COSTS
FCC UNIT SIZE: 150,000 BPSD AT 90% CAPACITY
WORST STRIPPING OPERATION
FIXED CAPITAL INVESTMENT = $3,516,100
Labor
1 Operating $96,400
2 Maintenance 70,300
3 Control Laboratory 19,300
4 Total Labor 186,000
Materials
5 Raw and Process
6 Acid Sludge 6,700
7 Lime 2,700
8 Catalyst Replacement 2,970,000
9 Maintenance 70,300
10 Operating 9,600
11 Total Materials 3,059,300
Utilities
12 Process Water 328,700
13 Electricity 21,300
1H Fuel 950,000
15 Total Utilities 1,300,000
16 Total Direct Operating Cost (4,11,&15) $4,545,300
17 Plant Overhead 148,800
18 Taxes and Insurance 70,300
19 Plant Cost (16, 17, & 18) 4,764,400
20 General & Administrative, Sales, Research 211,000
21 Cash Expenditures (19 & 20) 4,975,400
22 Depreciation 351,600
23 Interest on Working Capital 22,200
24 Total Operating Costs* (21,22,& 23) $5,349,200
25 Cost: (Cents/bbl) 10.99
* Does not include by-product credit or recovery costs
210
-------
APPENDIX F
ECONOMIC EVALUATION OF FLUE GAS DESULFURIZATION SYSTEMS
DETAILED ESTIMATES
Table F-l SUMMARY OF ANNUAL OPERATING COSTS
WESTVACO PROCESS (S02 PRODUCTION)
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $866,000
Labor
1 Operating $48,200
2 Maintenance 17,300
3 Control Laboratory 9,600
4 Total Labor 75,100
Materials
5 Raw and Process
6 Makeup Carbon 7,900
7 H2S 2,700
8 Hydrogen 300
9 Maintenance 17,300
10 Operating 7,500
11 Total Materials 35,700
Utilities
12 Process Water 8,300
13 Electricity 27,100
14 Fuel 1,300
15 Total Utilities 36,700
16 Total Direct Operating Cost (4,11,&15) $147,500
17 Plant Overhead 60,100
18 Taxes and Insurance 17,300
19 Plant Cost (16, 17, & 18) 224,900
20 General & Administrative, Sales, Research 51,900
21 Cash Expenditures (19 & 20) 276,800
22 Depreciation 86,600
23 Interest on Working Capital 5,500
24 Charge for Glaus Unit 21,600
25 Total Operating Costs (21,22,23,& 24) $390,500
26 Cost: (Mills/ft3) 0.0275
211
-------
Table F-2 SUMMARY OF ANNUAL OPERATING COSTS
WESTVACO PROCESS (S02 PRODUCTION)
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT = $2,280,000
Labor
1 Operating $48,200
2 Maintenance 45,600
3 Control Laboratory 9,600
4 Total Labor 103,400
Materials
5 Raw and Process
6 Makeup Carbon 39,400
7 H2S 13,300
8 Hydrogen 7,600
9 Maintenance 45,600
10 Operating 10,200
11 Total Materials 110,100
Utilities
12 Process Water 41,400
13 Electricity 135,600
14 Fuel 6,700
15 Total Utilities 183,700
16 Total Direct Operating Cost (4,11,&15) $397,200
17 Plant Overhead 81,800
18 Taxes and Insurance 45,600
19 Plant Cost (16, 17, & 18) 524,600
20 General & Administrative, Sales, Research 136,800
21 Cash Expenditures (19 & 20) 661,400
22 Depreciation 228,000
23 Interest on Working Capital 14,400
24 Charge for Glaus Unit 108,100
25 Total Operating Costs (21,22,23,4 24) $1,011,900
26 Cost: (Mills/ft3) 0.0143
212
-------
Table P-3 SUMMARY OF ANNUAL OPERATING COSTS
WESTVACO PROCESS (S02 PRODUCTION)
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT = $4,620,000
Labor
1 Operating $48,200
2 Maintenance 92,400
3 Control Laboratory 9,600
4 Total Labor 150,200
Materials
5 Raw and Process
6 Makeup Carbon
7 H2S
8 Hydrogen
9 Maintenance
10 Operating
11 Total Materials
Utilities
12 Process Water
13 Electricity
14 Fuel
15 Total Utilities
16 Total Direct Operating Cost (
17 Plant Overhead
18 Taxes and Insurance
19 Plant Cost (16, 17, & 18)
20 General & Administrative, Sales
21 Cash Expenditures (19 & 20)
22 Depreciation
23 Interest on Working Capital
24 Charge for Glaus Unit
25 Total Operating Costs (21,22,
26 Cost: (Mills/ft3)
118,300
40,000
4,800
92,400
15,000
270,500
124,200
406,800
20,100
551,100
4,11,&15) $971,800
120,200
92,400
1,184,400
, Research 277,200
1,461,600
462,000
29,100
324,400
23, & 24) $2,277,400
0.0107
213
-------
Table F-4 SUMMARY OF ANNUAL OPERATING COSTS
WESTVACO PROCESS (SULFUR PRODUCTION)
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $924,000
Labor
1 Operating $48,200
2 Maintenance 18,500
3 Control Laboratory 9,600
4 Total Labor 76,300
Materials
5 Raw and Process
6 Makeup Carbon 23,700
7 Hydrogen 32,000
8 Maintenance 18,500
9 Operating 4,800
10 Total Materials 79,000
Utilities
11 Process Water 8,300
12 Electricity 27,100
13 Fuel 1,300
Total Utilities 36,700
15 Total Direct Operating Cost (4,10,&l4) $192,000
16 Plant Overhead 61,000
17 Taxes and Insurance 18,500
18 Plant Cost (15, 16, & 17) 271,500
19 General & Administrative, Sales, Research 55.500
20 Cash Expenditures (18 & 19) 327,000
21 Depreciation 92,400
22 Interest on Working Capital 5,800
23 Total Operating Costs (20,21, & 22) $425,200
24 Cost: (Mills/ft3) 0.0300
214
-------
Table F-5 SUMMARY OP ANNUAL OPERATING COSTS
WESTVACO PROCESS (SULFUR PRODUCTION)
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT =$2,700,000
Labor
1 Operating $48,200
2 Maintenance 54,000
3 Control Laboratory 9,600
4 Total Labor 111,800
Materials
5 Raw and Process
6 Makeup Carbon 118,300
7 Hydrogen 160,000
8 Maintenance 59,000
9 Operating 4,800
10 Total Materials 337,100
Utilities
11 Process Water 41,400
12 Electricity 135,600
13 Fuel 6,700
14 Total Utilities 183,700
15 Total Direct Operating Cost (4,10,&l4) $632,600
16 Plant Overhead 89,400
17 Taxes and Insurance 54,000
18 Plant Cost (15, 16, & 17) 776,000
19 General & Administrative, Sales, Research 162,000
20 Cash Expenditures (18 & 19) 938,000
21 Depreciation 270,000
22 Interest on Working Capital 17,000
23 Total Operating Costs (20,21, & 22) $1,225,000
24 Cost: (Mills/ft3) 0.0173
215
-------
Table F-6 SUMMARY OP ANNUAL OPERATING COSTS
WESTVACO PROCESS (SULFUR PRODUCTION)
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT =$5,700,000
Labor
1 Operating $48,200
2 Maintenance 114,000
3 Control Laboratory 9,600
4 Total Labor 171,800
Materials
5 Raw and Process
6 Makeup Carbon 354,900
7 Hydrogen 480,000
8 Maintenance 114,000
9 Operating 4,800
10 Total Materials 953,700
Utilities
11 Process Water 124,200
12 Electricity 406,800
13 Fuel 20,100
14 Total Utilities 551,100
15 Total Direct Operating Cost (4,10,&l4) $1,676,600
16 Plant Overhead 137,400
17 Taxes and Insurance 114,000
18 Plant Cost (15, 16, 8= 17) 1,928,000
19 General & Administrative, Sales, Research 342,000
20 Cash Expenditures (18 & 19) 2,270,000
21 Depreciation 570,000
22 Interest on Working Capital 35,900
23 Total Operating Costs (20,21, & 22) $2,875,900
24 Cost: (Mills/ft3) 0.0135
216
-------
Table F-7 SUMMARY OF ANNUAL OPERATING COSTS
SHELL FLUE GAS DESULFURIZATION (SFGD)
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $1,090,000
Labor
1 Operating $48,200
2 Maintenance 21,800
3 Control Laboratory 9,600
4 Total Labor 79,600
Materials
5 Raw and Process
6 Acceptor Replacement 23,000
7 Hydrogen 34,800
8 Maintenance 21,800
9 Operating 4,800
10 Total Materials 81,400
Utilities
11 Steam 900
12 Electricity 12,200
13 Fuel 24,100
Total Utilities 37,500
15 Total Direct Operating Cost (4,10,&14) $201,500
16 Plant Overhead 63,700
17 Taxes and Insurance 21.800
18 Plant Cost (15, 16, & 17) 287,000
19 General & Administrative, Sales, Research 65,400
20 Cash Expenditures (18 & 19) 352,400
21 Depreciation 109,000
22 Interest on Working Capital 6,900
23 Charge for Glaus Unit 16,300
24 Total Operating Costs (20, 21,22, & 23) $484,600
25 Cost: (Mills/ft3) 0.0341
217
-------
Table F-8 SUMMARY OF ANNUAL OPERATING COSTS
SHELL FLUE GAS DESULFURIZATION (SFGD)
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT = $3,230,000
Labor
1
2
3
4
Operating
Maintenance
Control Laboratory
Total Labor
$48,200
64,600
9,600
122,400
Materials
6
7
8
9
10
Raw and Process
Acceptor Replacement
Hydrogen
Maintenance
Operating
Total Materials
115,200
174,100
•64,600
4,800
358,700
Utilities
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
Steam
Electricity
Fuel
Total Utilities
Total Direct Operating Cost (4
Plant Overhead
Taxes and Insurance
Plant Cost (15, 16, & 17)
General & Administrative, Sales,
Cash Expenditures (18 & 19)
Depreciation
Interest on Working Capital
Charge for Glaus Unit
Total Operating Costs (20, 21,
Cost: (Mills/ft3)
4,600
61,100
122,000
187,700
,10,&14) $668,800
97,900
64,600
831,300
Research 193,800
1,025,100
323,000
20,300
81,300
22,&23) $1,449,700
0.0204
218
-------
Table F-9 SUMMARY OF ANNUAL OPERATING COSTS
SHELL FLUE GAS DESULFURIZATION (SFGD)
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT = $6,470,000
Labor
1 Operating $48,200
2 Maintenance 129,400
3 Control Laboratory 9,600
4 Total Labor 187,200
Materials
5 Raw and Process
6 Acceptor Replacement 345,700
7 Hydrogen 522,400
8 Maintenance 129,400
9 Operating 4,800
10 Total Materials 1,002,300
Utilities
11 Steam . 13,900
12 Electricity 183,300
13 Fuel 366,000
Total Utilities 563,200
15 Total Direct Operating Cost (4,10,&14) $1,752,700
16 Plant Overhead 149,800
17 Taxes and Insurance 129,400
18 Plant Cost (15, 16, & 17) 2,031,900
19 General & Administrative, Sales, Research 388,200
20 Cash Expenditures (18 & 19) 2,420,100
21 Depreciation 647,000
22 Interest on Working Capital 40,800
23 Charge for Glaus Unit 244,000
24 Total Operating Costs (20, 21,22,&23) $3,351,900
25 Cost: (Mills/ft3) 0.0157
219
-------
Table F-10 SUMMARY OF ANNUAL OPERATING COSTS
PURASIV-S PROCESS
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $1,650,000
Labor
1 Operating $48,200
2 Maintenance 33,000
3 Control Laboratory 9,600
4 Total Labor 90,800
Materials
5 Raw and Process
6 Molecular Sieve Replacement 145,100
7 Maintenance 33,000
8 Operating 4,800
9 Total Materials 182,900
Utilities
10 Cooling Water 1,300
11 Electricity 52,400
12 Fuel 8,700
13 Total Utilities 62,400
14 Total Direct Operating Cost (4,9,& 13) $336,100
15 Plant Overhead 72,600
16 Taxes and Insurance 33,000
17 Plant Cost (14, 15, & 16) 441,700
18 General & Administrative, Sales, Research 99,000
19 Cash Expenditures (17 & 18) 540,700
20 Depreciation 165,000
21 Interest on Working Capital 10,400
22 Charge for Glaus Unit 16,300
23 Total Operating Costs (19,20,21 & 22) $732,400
24 Cost: (Mills/ft3) 0.0516
220
-------
Table F-ll SUMMARY OF ANNUAL OPERATING COSTS
PURASIV-S PROCESS
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT = $4,820,000
Labor
1 Operating $48,200
2 Maintenance 96,400
3 Control Laboratory 9,600
4 Total Labor 154,200
Materials
5 Raw and Process
6 Molecular Sieve Replacement 725,700
7 Maintenance 96,400
8 Operating 4,800
9 Total Materials 826,900
Utilities
10 Cooling Water 6,300
11 Electricity 262,000
12 Fuel 43,400
13 Total Utilities 311,700
14 Total Direct Operating Cost (4,9,& 13) $1,292,800
15 Plant Overhead 123,400
16 Taxes and Insurance 96,400
17 Plant Cost (14, 15, & 16) 1,512,600.
18 General & Administrative, Sales, Research 289,200
19 Cash Expenditures (17 & 18) 1,801,800
20 Depreciation 482,000
21 Interest on Working Capital 30,400
22 Charge for Glaus Unit 81,300
23 Total Operating Costs (19,20,21 & 22) $2,395,500
24 Cost: (Mills/ft3) 0.0338
221
-------
Table F-12 SUMMARY OF ANNUAL OPERATING COSTS
PURASIV-S PROCESS
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT = $9,6*10,000
Labor
1 Operating $48,200
2 Maintenance 192,800
3 Control Laboratory 9,600
4 Total Labor 250,600
Materials
5 Raw and Process
6 Molecular Sieve Replacement 2,177,000
7 Maintenance 192,800
8 Operating 4,800
9 Total Materials 2,374,600
Utilities
10 Cooling Water 19,000
11 Electricity 786,000
12 Fuel 130,300
13 Total Utilities 935,300
14 Total Direct Operating Cost (4,9,& 13) $3,560,500
15 Plant Overhead 200,500
16 Taxes and Insurance 192,800
17 Plant Cost (14, 15, & 16) 3,953,800
18 General & Administrative, Sales, Research 578,400
19 Cash Expenditures (17 & 18) 4,532,200
20 Depreciation 964,000
21 Interest on Working Capital 60,700
22 Charge for Glaus Unit 244,000
23 Total Operating Costs (19,20,21 & 22) $5,800,900
24 Cost: (Mills/ft3) 0.0273
222
-------
Table F-13 SUMMARY OP ANNUAL OPERATING COSTS
WELLMAN-LORD PROCESS
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $1,500,000
Labor
1 Operating $144,600
2 Maintenance 30,000
3 Control Laboratory 28,900
4 Total Labor 203,500
i
Materials
5 Raw and Process
6 Sodium Hydroxide 3,000
7 Maintenance 30,000
8 Operating 13,000
9 Total Materials 46,000
Utilities
10 Cooling Water 1,100
11 Process Water 700
12 Electricity 13,^00
13 Steam 5,000
14 Total Utilities 20,200
15 Total Direct Operating Cost (*»,9,&14) $269,700
16 Plant Overhead 162,800
17 Taxes and Insurance 30,000
18 Plant Cost (15, 16, & 17) 462,500
19 General & Administrative, Sales, Research 90,000
20 Cash Expenditures (18 & 19) 552,500
21 Depreciation 150,000
22 Interest on Working Capital 9,500
23 Charge for Glaus Unit 12,600
24 Total Operating Costs (20,21,22, & 23) $724,600
25 Cost: (Mills/ft3) 0.0511
223
-------
Table F-14 SUMMARY OP ANNUAL OPERATING COSTS
WELLMAN-LORD PROCESS
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT = $4,420,000
Labor
1 Operating $144,600
2 Maintenance 88,400
3 Control Laboratory 28,900
4 Total Labor 261,900
Materials
5 Raw and Process
6 Sodium Hydroxide 15,200
7 Maintenance 88,400
8 Operating 13,000
9 Total Materials 116,600
Utilities
10 Cooling Water 5,300
11 Process Water 3,300
12 Electricity 66,900
13 Steam 24,300
14 Total Utilities 100,300
15 Total Direct Operating Cost (4,9,&l4) $478,800
16 Plant Overhead 209,500
17 Taxes and Insurance 88.400
18 Plant Cost (15, 16, & 17) 776,700
19 General & Administrative, Sales, Research 265,200
20 Cash Expenditures (18 & 19) 1,041,900
21 Depreciation 442,000
22 Interest on Working Capital 27,900
23 Charge for Glaus Unit 62,900
24 Total Operating Costs (20,21,22, & 23) $1,574,700
25 Cost: (Mills/ft3) 0.0222
224
-------
Table F-15 SUMMARY OP ANNUAL OPERATING COSTS
WELLMAN-LORD PROCESS
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT = $9,080,000
Labor
1 Operating $144,600
2 Maintenance 181,600
3 Control Laboratory 28,900
14 Total Labor 355,100
Materials
5 Raw and Process
6 Sodium Hydroxide 45,600
7 Maintenance 181,600
8 Operating 13,000
9 Total Materials 240,200
Utilities
10 Cooling Water 15,800
11 Process Water 9,900
12 Electricity 200,700
13 Steam 74,300
14 Total Utilities 300,700
15 Total Direct Operating Cost (4,9,&l4) $896,000
16 Plant Overhead 283,300
17 Taxes and Insurance 181,600
18 Plant Cost (15, 16, & 17) 1,360,900
19 General & Administrative, Sales, Research 544.800
20 Cash Expenditures (18 & 19) 1,905,700
21 Depreciation 908,000
22 Interest on Working Capital 57,200
23 Charge for Glaus Unit 188,700
24 Total Operating Costs (20,21,22, & 23) $3,059,600
25 Cost: (Mills/ft3) 0.0144
225
-------
Table F-16 SUMMARY OF ANNUAL OPERATING COSTS
MAGNESIUM OXIDE SCRUBBING
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT = $1,450,000
Labor
1 Operating $154,000
2 Maintenance 29,000
3 Control Laboratory 30,800
4 Total Labor 213,800
Materials
5 Raw and Process
6 Lime 700
7 MgO 3,800
8 Coke 600
9 Maintenance 29,000
10 Operating 15,400
11 Total Materials 49,500
Utilities
12 Fuel Oil 23,300
13 Steam
14 Process Water 22,800
15 Electricity 20,400
16 Total Utilities 66,500
17 Total Direct Operating Cost (4,11,&16) $329,800
18 Plant Overhead 171,000
19 Taxes and Insurance 29,000
20 Plant Cost (17, 18, & 19) 529,800
21 General & Administrative, Sales, Research 87.000
22 Cash Expenditures (20 & 21) 616,800
23 Depreciation 145,000
24 Interest on Working Capital 9,100
25 Charge for Glaus Unit 16,300
26 Total Operating Costs (22,23,24, & 25) $787,200
27 Cost: (Mills/ft3) 0.0555
226
-------
Table F-l? SUMMARY OF ANNUAL OPERATING COSTS
MAGNESIUM OXIDE SCRUBBING
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT = $4,100,000
Labor
1 Operating $154,000
2 Maintenance 82,000
3 Control Laboratory 30,800
4 Total Labor 266,800
Materials
5 Raw and Process
6 Lime 3,700
7 MgO 19,200
8 Coke 3,100
9 Maintenance 82,000
10 Operating 15,400
11 Total Materials 123,400
Utilities
12 Fuel Oil 116,500
13 Steam
14 Process Water 114,000
15 Electricity 101,900
16 Total Utilities 332,400
17 Total Direct Operating Cost (4,11,&16) $722,600
18 Plant Overhead 213,400
19 Taxes and Insurance 82,000
20 Plant Cost (17, 18, & 19) 1,018,000
21 General & Administrative, Sales, Research 246,000
22 Cash Expenditures (20 & 21) 1,264,000
23 Depreciation 410,000
24 Interest on Working Capital 25,800
25 Charge for Glaus Unit 81,300
26 Total Operating Costs (22,23,24, & 25) $1,781,100
27 Cost: (Mills/ft3) 0.0251
227
-------
Table P-18 SUMMARY OF ANNUAL OPERATING COSTS
MAGNESIUM OXIDE SCRUBBING
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENT = $8,390,000
Labor
1 Operating $154,000
2 Maintenance 167,800
3 Control Laboratory 30,800
4 Total Labor 352,600
Materials
5 Raw and Process
6 Lime 11,100
7 MgO 57,600
8 Coke 9,300
9 Maintenance 167,800
10 Operating 15,^00
11 Total Materials 261,200
Utilities
12 Fuel Oil 3^9,600
13 Steam
14 Process Water 341,900
15 Electricity 305,700
16 Total Utilities 997,200
17 Total Direct Operating Cost (4,11,&16) $1,611,000
18 Plant Overhead 282,100
19 Taxes and Insurance 167,800
20 Plant Cost (17, 18, & 19) 2,060,900
21 General & Administrative, Sales, Research 503.^00
22 Cash Expenditures (20 & 21) 2,564,300
23 Depreciation 839,000
24 Interest on Working Capital 52,900
25 Charge for Glaus Unit 244,000
26 Total Operating Costs (22,23,24, & 25) $3,700,200
27 Cost: (Mills/ft3) 0.0174
228
-------
Table F-19 SUMMARY OP ANNUAL OPERATING COSTS
CAT-OX PROCESS
CAPACITY: 10,000 BPSD
FIXED CAPITAL INVESTMENT=$1,890,000
Labor
1 Operating $144,500
2 Maintenance 37,800
3 Control Laboratory 28,900
4 Total Labor 211,200
Materials
5 Raw and Process 51,600
6 Maintenance 37,800
7 Operating 14,500
8 Total Materials 103,900
Utilities
9 Electricity 16,000
10 Total Utilities 16,000
11 Total Direct Operating Cost (4,8,&10) $331,100
12 Plant Overhead 169,000
13 Taxes and Insurance 37,800
14 Plant Cost (11, 12, & 13) 537,900
15 General & Administrative, Sales, Research 113*400
16 Cash Expenditures (14 & 15) 651,300
17 Depreciation 189,000
18 Interest on Working Capital 11,900
19 Total Operating Cost (16, 17, & 18) $852,200
20 Cost: (Mills/ft3) 0.0601
229
-------
Table F-20 SUMMARY OP ANNUAL OPERATING COSTS
CAT-OX PROCESS
CAPACITY: 50,000 BPSD
FIXED CAPITAL INVESTMENT=$5,490,000
Labor
1 Operating $144,500
2 Maintenance 109,800
3 Control Laboratory 28,900
4 Total Labor 283,200
Materials
5 Raw and Process 258,000
6 Maintenance 109,800
7 Operating 14,500
8 Total Materials 382,300
Utilities
9 Electricity 80,000
10 Total Utilities 80,000
11 Total Direct Operating Cost (4,8,& 10) $745,500
12 Plant Overhead 226,600
13 Taxes and Insurance 109,800
14 Plant Cost (11, 12, & 13) 1,081,900
15 General & Administrative, Sales, Research 329,400
16 Cash Expenditures (14 & 15) 1,411,300
17 Depreciation 549,000
18 Interest on Working Capital 34,600
19 Total Operating Cost (16, 17, & 18) $1,994,900
20 Cost: (Mills/ft3) 0.0281
230
-------
Table F-21 SUMMARY OF ANNUAL OPERATING COSTS
CAT-OX PROCESS
CAPACITY: 150,000 BPSD
FIXED CAPITAL INVESTMENTS $11,500,000
Labor
1 Operating $144,500
2 Maintenance 230,000
3 Control Laboratory 28,900
4 Total Labor 403,400
Materials
5 Raw and Process 774,000
6 Maintenance 230,000
7 Operating 14,500
8 Total Materials 1,018,500
Utilities
9 Electricity 240,000
10 Total Utilities 240,000
11 Total Direct Operating Cost (4,8,&10) $1,661,900
12 Plant Overhead 322,700
13 Taxes and Insurance 230.000
14 Plant Cost (11, 12, & 13) 2,214,600
15 General & Administrative, Sales, Research 690»000
16 Cash Expenditures (14 & 15) 2,904,600
17 Depreciation 1,150,000
18 Interest on Working Capital 72,500
19 Total Operating Cost (16, 17, & 18) $4,127,100
20 Cost: (Mills/ft3) 0.019*1
231
-------
APPENDIX G
EVALUATION FORM FOR ADD-ON FLUE GAS DESULFURIZATION SYSTEMS
CONSIDERATIONS IN RANKING S09 EMISSION CONTROL PROCESSES
A. PROCESS NAME
B. DEVELOPER
C. BASIC PRINCIPLE
D. CONTROL LEVEL ACHIEVABLE
1. Claimed (Theoretical)
2. Demonstrated
E. STATE OF DEVELOPMENT
1. Conceptual Development
2. Laboratory Development
3- Bench Scale Development
iJ. Pilot Plant Tested
5. Commercial Installation
6. Demonstration Site
a. Location
b. Conditions, Conclusions, and Recommendations
c. Reference
7- Commercial Availability Date
232
-------
F. PROCESS APPLICABILITY TO CATCRACKER FLUE GAS
G. LOCATION OF THE CONTROL PROCESS IN REGENERATOR OFF-GAS FLOW
TRAIN
H. RAW MATERIALS NEEDED
I. SALEABLE SULFUR PRODUCTS (WASTES)
1. Sulfur
2. Sulfur Dioxide Q] Concentration -
3. Sulfuric Acid Q] Concentration -
4. Other
5. Marketability (According to February, 1973 Market
Conditions)
a. Excellent II
b. Good II
c. Poor
233
-------
J. OTHER BY-PRODUCTS
1. Form
2. Proposed Methods of Disposal
K. SCHEMATIC OP PROPOSED REGENERATOR OFF-GAS FLOW TRAINS
L. SPACE REQUIREMENTS
M. PROBLEMS FORESEEN IN ACHIEVING RETROFIT
N. ECONOMICS FOR CONTROL PROCESS
A. Catcracker Capacity -
B. Investment Costs - $
C. Operating Costs - ($ /year) ; ( d:/bbl) ; (
D. Reference
0. ADDITIONAL POLLUTANTS CONTROLLED
A. Claimed
B. Demonstrated
234
-------
P. PROCESS RELIABILITY
Q. IS THE PROCESS WORTH FURTHER INVESTIGATION FOR CATCRACKER
S02 EMISSIONS CONTROL APPLICATION?
A. Yes
B. No
C. Unable to Determine at Present
D. Comments
235
-------
APPENDIX H
SINGLE UNIT CONVERSION FACTORS. BRITISH TO SI (METRIC)
To convert from
atmosphere (atm)
barrel (for petroleum, 42 gal)
British thermal unit (Btu)
day (d)
degree Celsius (°C)
degree Fahrenheit (°P)
degree Fahrenheit (°F)
foot (ft)
foot3 (cu ft)
foot2 (sq ft)
gallon (gal)
grain (1/7000 Ib)
gram (gm)
hour (h)
inch (in)
inch2 (sq in)
inch3 (cu in)
kilowatt-hour (kwh)
litre (1)
micron (y)
minute (min)
pound-force/inch2 (psi)
pound-mass (Ib)
ton (long)
ton (short)
to
pascal (Pa)
metre3 (m3)
Joule (J)
second (s)
kelvin (K)
degree Celsius
kelvin (K)
metre (m)
metre3 (m3)
metre2 (m2)
metre3 (m3)
kilogram (kg)
kilogram (kg)
second (s)
metre (m)
Q
metre2 (m )
metre3 (m3)
joule (J)
metre3 (m3)
metre (m)
seconds (S)
pascal (Pa)
kilogram (kg)
kilogram (kg)
kilogram (kg)
multiply by
9.80? x 10*
1.590 x 10-1
1.055 x 103
8.640 x 10"
tk=tc+273.15
tc=(tf-32)/1.8
tk=(tf+459.67)/1.8
3.048X10-1
2.832xlOr2
9.290xlO~2
3.785xlO~3
6.480xlO~5
1.000xlO~3
3.600xl03
2.540xlO-2
6.452x10-**
1.639xlO-5
3.600xl06
l.OOOxlO-3
l.OOOxlO-6
6.000X101
6.895xl03
4.536X10-1
1.0l6xl03
9.072xl02
236
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
I REPORT NO
EPA-650/2-74-082
3. RECIPIENT'S ACCESSION>NO.
4 TITLE AND SUBTITLE
Refinery Catalytic Cracker Regenerator SOx
Control Process Survey
5 REPORT DATE
September 1974
6. PERFORMING ORGANIZATION CODE
7 AUTHOR(S) T ctvrtnicek, T. W. Hughes ,
C. M. Moscowitz, and D. L. Zanders
8. PERFORMING ORGANIZATION REPORT NO.
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
Dayton Laboratory
Dayton, Ohio 45407
10. PROGRAM ELEMENT NO.
1AB013; ROAP 21ADC-031
11. CONTRACT/GRANT NO.
68-02-1320
(Task 1, Phase I)
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Phase I Final. 7-11/73
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16 ABSTRACT
The report gives results of a survey of conceptual techniques applicable
to fluid catalytic cracker (FCC) regenerator off-gas sulfur oxide emission
reduction, with respect to their application both to the FCC system itself and to
the regenerator off-gas. These two control techniques have also been compared
with FCC feedstock desulfurization. The economics for all systems evaluated are
compared. A comprehensive analysis of FCC operations has produced evidence
that sulfur emission control can most effectively be achieved through steam
contacting of the spent cracking catalyst. This concept is therefore proposed as the
primary subject for further investigation.
7.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
Air Pollution
Petroleum Refining
Catalytic Cracking
Regeneration
(Engineering)
Sulfur Oxides
Desulfurization
Cost Effectiveness
Air Pollution Control
Stationary Sources
Feedstock
Steam Contacting
13B, 07D
13H, 14A
07A
07B
8. DISTRIBUTION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
237
------- |