EPA-450/2-78-021
April 1978
COST OF BENZENE
REDUCTION IN GASOLINE
TO THE PETROLEUM
REFINING INDUSTRY
~
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Waste Management
Office of Air Quality Planning and Standards
Research Triangle Park, North Carolina 27711
-------
-------
This report is issued by the Environmental Protection Agency to report
technical data of interest to a limited number of readers. Copies are
available free of charge to Federal employees, current contractors and
grantees, and nonprofit organizations - in limited quantities - from the
Library Services Office (MD-35), U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina 27711; or, for a fee, from the
National Technical Information Service, 5285 Port Royal Road, Springfield,
Virginia 22161.
This report was furnished to the Environmental Protection Agency
by Arthur D. Little, Inc. , Acorn Park, Cambridge, Massachusetts,
in fulfillment of Contract No. 68-02-2859. The contents of this report
are reproduced herein as received from Arthur D. Little, Inc. The
opinions, findings, and conclusions expressed are those of the author
and not necessarily those of the Environmental Protection Agency.
Mention of company or product names is not to be considered as an
endorsement by the Environmental Protection Agency.
Publication No. EPA-450/2-78-021
11
-------
ABSTRACT
This report assesses the cost to the U.S. petroleum industry
of removing benzene from the two largest contributors to the benzene
levels in the gasoline poolrefinery reformates and FCC gasoline.
Predictions were made of the 1981 gasoline pool composition and the
benzene content of gasoline component streams. A process route was
selected for each stream and the benzene removal costs in 1977
dollars were developed. Removal of 94.5% of benzene from reformates
and FCC gasoline would reduce U.S. average benzene content from 1.37%
to 0.26%. This would require an investment of $5.3 billion, and total
costs of $2.5 billion per year, including capital recovery, or 2.2
cents per gallon of gasoline. Costs for some small refineries would
be up to 7 cents per gallon of gasoline, or three times the U.S.
average costs. These costs are for benzene removal only, and do not
include costs of octane replacement, volume replacement or the effect
on the chemical industry. When these other factors are considered, it
is roughly estimated that the total costs, including capital recovery,
would be $3.8 billion per year, or 3.3 cents per gallon of gasoline.
iii
-------
TABLE OF CONTENTS
CHAPTER 1
EXECUTIVE SUMMARY
Page
1-1
1.1 Objectives 1-1
1.2 Approach and Results 1-1
1.3 Conclusions . 1-20
CHAPTER 2 GASOLINE POOL COMPOSITION 2-1
2.1 Assessment of Current Capacity 2-2
2.2 Reformate and FCC Gasoline Pool Contribution . . . 2-10
2.3 1981 Gasoline Pool Composition 2-17
CHAPTER 3 BENZENE CONTENT OF GASOLINE 3-1
3.1 1977 U.S. Pool Benzene Level . . 3-3
3.2 1977 Regional Pool Benzene Level 3-8
3.3 Projected 1981 U.S. Pool Benzene Level 3-15
3.4 Selective Removal of Benzene from
Blend Components 3-31
CHAPTER 4 TECHNOLOGICAL OPTIONS FOR BENZENE.REMOVAL
FROM GASOLINE 4-1
4.1 Reformate Benzene Control Technologies 4-2
4.2 FCC Gasoline Benzene Control Technologies .... 4-7
4.3 Other Technological Alternatives 4-15
4.4 Processing Routes for Other Gasoline Streams . . . 4-16
CHAPTER 5 ECONOMICS OF BENZENE REMOVAL FROM REFORMATES
& FCC GASOLINE 5-1
5.1 Basis For Economics 5-3
A. Process Route Description 5-3
B. Investment Costs 5-6
iv
-------
TABLE OF CONTENTS (Cont.)
Page
CHAPTER 5 (Cont.)
C. Manufacturing Costs 5-9
D. Effect of Variables on Economics . . . . '. . 5-11
E. Scale-up of Economics 5-15
5.2 National Cost of Benzene Removal from
Reformates and FCC Gasoline 5-17
A. Total U.S. Cost 5-17
B. Regional Differences . . . 5-20
C. Sensitivity to Refinery Hydrogen Costs . . . 5-20
D. Sensitivity to FCC Hydrogenation Step .... 5-22
5.3 Impact on the Small Refiner 5-25
5.4 Effect of Assumptions on Costs 5-34
CHAPTER 6 OTHER ECONOMIC ISSUES ASSOCIATED WITH
BENZENE REMOVAL 6-1
6.1 Octane Loss 6-1
6.2 Volume Loss 6-6
6.3 Impact on the Chemical Markets 6-11
6.4 Estimated Cost of Other Economic Issues
Associated with Benzene Removal 6-17
6.5 Other Items that Warrant Further Study 6-22
A. Evaluation of Economics of Benzene
Removal from Other Streams 6-22
B. Evaluation of the Effect of Crude Oil
Quality, & Naphtha Cut Point
Changes on Benzene Removal . 6-23
7.0 REFERENCES 7-1
-------
LIST OF TABLES
Table No. Page
1.1 Estimated 1981 U.S. Pool Blend 1-3
1.2 Component Benzene Content Estimated from
Refinery Survey 1-5
1.3 U.S. Pool Octane Loss 1-11
1.4 National Cost of Benzene Removal from Reformate
& FCC Gasoline 1-13
1.5 Rough Cost of Benzene Removal from Reformate &
FCC Gasoline, Including Octane, Volume &
Chemical Market Loss 1-19
2.1 1977 U.S. ReformingBTX Extraction Capacity 2-3
2.2 1977 U.S. Cat CrackingAlkylation Capacity 2-4
2.3 1977 U.S. Cat Hydrocracking Capacity 2-5
2.4 1977 U.S. Thermal Processing Capacity 2-6
2.5 Designations of Refining Districts 2-7
2.6 1977 Gasoline Reforming Capacity by PADD 2-9
2.7 1981 Total Reforming Capacity by PADD 2-11
2.8 1981 Gasoline Reforming Capacity by PADD 2-12
2.9 1981 Catalytic Cracking Capacity (Fresh Feed)
by PADD 2-13
2.10 1981 U.S. Pool Gasoline Blend by PADD 2-19
3.1 Illustrative Ranges of Benzene Content of
Blending Components 3-4
3.2 U.S. Pool Benzene Content Survey Results 3-5
3.3 Refineries Surveyed for Gasoline Benzene Data .... 3-6
3.4 Results of Refinery Benzene Survey on Total
Gasoline Pool 3-7
3.5 Results of Refiner Benzene Survey on Gasoline
Blend Components 3-9
3.6 Estimated Benzene Level of U.S. Gasoline Pool .... 3-10
3.7 Distribution of duPont Survey Gasoline Samples .... 3-11
3.8 Effect of Reformer Parameters on Reformate
Benzene Content 3-19
vi
-------
LIST OF TABLES (Cont.)
Table No. Pc.ge
3.9 Comparison of Benzene Content of Reformate
From Selected Naphthas 3-20
3.10 Sources of Benzene in Reformate 100 RON Severity . . . 3-21
3.11 Illustrative Effect of Reformer Initial Boiling
Point on Product Benzene Level 3-24
3.12 El Segundo Aromatics Reformer Effects of
Changeover to Rheniforming Catalyst 3-27
3.13 Estimated Benzene Level of 1981 Gasoline Pool .... 3-32
3.14 Benzene Contained in Gasoline Pool - 1981 3-33
3.15 Pool Benzene Levels Achieved by Extraction of
Blend Components 3-35
4.1 Illustrative Reformate Properties 4-3
4.2 Illustrative FCC Gasoline Properties 4-9
4.3 Pool Octane Loss Due to Benzene Removal from
Reformates and FCC Gasoline 4-13
5.1 Removal of Benzene from Reformate Investment
Cost - U.S. Gulf Coast 1977 5-8
5.2 Removal of Benzene from FCC Gasoline Investment
Cost - U.S. Gulf Coast 1977 5-10
5.3 Base Case Manufacturing Costs for Removal of
Benzene from Reformates 5-12
5.4 Base Case Manufacturing Costs for Removal of
Benzene from FCC Gasoline 5-13
5.5 Reformate Fractionation Investment 5-15
5.6 FCC Gasoline Fractionation Investment 5-16
5.7 FCC Gasoline Hydrogenation Investment 5-16
5.8 Reformate & FCC Gasoline Extraction Investment .... 5-17
5.9 National Cost of Benzene Removal from Reformate
and FCC Gasoline 5-18
5.10 Summary of Regional Results 5-21
5.11 Hydrogen Costs 5-22
vii
-------
LIST OF TABLES (Cont.)
Table No. Page
5.12 National Cost of Benzene Removal from Reformate
and FCC Gasoline (Hydrogen at Fuel Value) 5-23
5.13 Effect of Hydrogenation Step on FCC Gasoline
Costs 5-24
5.14 National Cost of Benzene Removal from Reformate
and FCC Gasoline (Without Hydrogenation
of FCC Gasoline) 5-24
5.15 Costs for 10,000 B/SD Refinery vs. U.S. Average . . . 5-29
5.16 Total U.S. Costs of Removal of Benzene from
Gasoline Reformate by Capacity Range 5-32
5.17 Costs of Removal of Benzene from FCC Gasoline
U.S. by FCC Gasoline Capacity Range 5-33
6.1 U.S. Pool Octane Loss Associated with Benzene
Removal 6-2
6.2 Effect on U.S. Pool Octane of Benzene Removal
from Reformates 6-3
6.3 Effect on U.S. Pool Octane of Benzene Removal
from FCC Gasoline 6-4
6.4 Benzene Removal from Gasoline Octane Blending
Values 6-5
6.5 National Octane Loss Penalty Range of Possible
Costs 6-8
6.6 Range of Values of Benzene for Alkylation to
Cumene for use as a Gasoline Blending
Component 6-10
6.7 Volumetric Penalty for Benzene Removal from
Reformates and FCC Gasoline 6-12
6.8 Benzene Supply & Demand 6-13
6.9 Range of Loss of Chemical Benzene Values Accompany-
ing Benzene Reduction in Gasoline 6-15
6.10 Value of Toluene as Gasoline vs. Benzene 6-16
6.11 Rough Cost of Other Economic Issues Associated
with Benzene Removal from Reformate
and FCC Gasoline 6-21
viii
-------
LIST OF FIGURES
Figure No. Page
1.1 Gasoline Pool Composition & Benzene Contribution . . . 1-4
1.2 Petroleum Administration for Defense (PAD)
Districts 1-7
1.3 Benzene Removal by Component 1-8
1.4 Regional Cost of Benzene Removal from Reformate
and FCC Gasoline 1-15
1.5 Cost of Benzene Removal vs. Gasoline Production
Using Refinery Produced Hydrogen 1-16
2.1 PADD's I-IV FCC Unit Capacity/Demand 2-16
2.2 PADD's I-IV Reformer Capacity/Demand 2-18
3.1 Distribution of PADD Gasoline Samples by Benzene
Content - Summer 1976 3-12
3.2 Distribution of Gasoline Samples by Benzene
Content - Summer 1976 . 3-14
3.3 Cumulative Percent PADD Gasoline Samples by
Benzene Content - Summer 1976 3-16
3.4 Typical Reforming Reactions ..... 3-17
3.5 Simplified Representation of Processing for
Gasoline Pool 3-23
3.6 Potential Equilibrium Yield of Benzene from
Its Precursors at 950°F 3-25
3.7 Effects of Pressure & Benzene Severity on Yields
of Benzene from 130-310°F Arabian Naphtha 3-26
4.1 Flow Diagram for Benzene Removal from Catalytic
Reformate 4-5
4.2 Flow Diagram for Benzene Removal from FCC
Gasoline 4-11
5.1 Flow Diagram for Benzene Removal from Catalytic
Reformate 5-4
5.2 Flow Diagram for Removal of Benzene from FCC
Gasoline 5-5
ix
-------
LIST OF FIGURES (Cont.)
Figure No. Page
5.3 Cost of Benzene Removal vs. Refornate &
FCC Gasoline Capacity 5-26
5.4 Cost of Benzene Removal vs. Gasoline Production
Using Hydrogen Plant Hydrogen 5-27
5.5 Cost of Benzene Removal vs. Gasoline Production
Using Ref ineryProduced Hydrogen 5-28
6.1 Cost of Adding Octane Number as a Function of
Pool Octane 6-7
6.2 Current Distribution of Benzene Producers 6-18
6.3 Current Distribution of Benzene Consumers 6-19
6.4 Distribution of Benzene Producers with Benzene Removal
from Reformates & FCC Gasoline 6-20
x
-------
APPENDIX A
GASOLINE POOL COMPOSITION
TABLE OF CONTENTS
Section No. Page
A.I Summary A-l
A. 2 Methodology of Study A-2
A. 3 FCC Unit Results A-4
A.4 Catalytic Reforming Results A-19
List of Tables
Table No. Page
A.I Cluster Model Output Summary for FCC Units A-5
A. 2 PADD I Gasoline Production A-4
A. 3 PADD I FCC Gasoline Yields, LV% A-7
A.4 PADD I FCC Utilization A-7
A.5 PADD II FCC Gasoline Yields, LV% A-8
A.6 PADD II FCC Utilization A-10
A. 7 PADD III FCC Gasoline Yields, LV% ....-.- A-ll
A.8 PADD III FCC Utilization A-ll
A.9 Tentative Estimates of Percentages FCC Gasoline
in PADD IV Gasoline Pool A-13
A. 10 PADD IV FCC Capacity Utilization A-15
A. 11 PADD V FCC Gasoline Yields, LV% A-16
A. 12 PADD V FCC Utilization A-16
A. 13 PADD's I-IV FCC Capacity, MB/SD A-18
A. 14 PADD's I-IV FCC Feed, MB/CD A-19
A.15 Cluster Model Output Summary for Gasoline Producing
Catalytic Reforming Units A-21
A.16 Typical Low Pressure Reforming Yields 160/380°F
Arabian Light Naphtha A-22
xi
-------
APPENDIX A
List of Tables (Cont.)
Table No. Page
A. 17 Percentage of Reforming Units Containing
Bimetallic Catalysts A-22
A. 18 1976 Reformer Feed Rate, MB/D . A-23
A.19 PADD I Reforming Yields, LV% A-24
A. 20 PADD I Reformer Utilization A-26
A.21 PADD II Reforming Yields, LV% A-27
A.22 PADD II Reformer Utilization A-27
A.23 PADD III Reforming Yields, LV% A-28
A. 24 PADD III Reformer Utilization A-31
A.25 PADD IV Reformate in Gasoline Pool A-33
A. 26 PADD IV Reformer Utilization A-33
A.27 PADD V Gasoline Production, MMB/CD A-34
A.28 PADD V Reformer Yields, LV% A-34
A. 29 PADD V Reformer Utilization A-35
A.30 PADD's I-IV Reformer Capacity, MB/SD A-37
A.31 PADD's I-IV Naphtha Reformer Feed, MB/CD A-37
List of Figures
Figure No. Page
A.I PADD I FCC Unit Capacity/Demand A-6
A. 2 PADD II FCC Unit Capacity/Demand A-9
A.3 PADD III FCC Unit Capacity/Demand A-12
A. 4 PADD IV FCC Unit Capacity /Demand A-14
A. 5 PADD V FCC Unit Capacity/Demand A-17
A.6 PADD's I-IV FCC Unit Capacity/Demand A-20
A. 7 PADD I Reformer Capacity/Demand A-25
A. 8 PADD II Reformer Capacity/Demand A-29
xii
-------
APPENDIX A
List of Figures (Cont.)
Figure No. Page
A. 9 PADD III Reformer Capacity/Demand A-30
A. 10 PADD IV Reformer Capacity/Demand A-32
A. 11 PADD V Reformer Capacity/Demand A-36
A.12 PADD's I-IV Reformer Capacity/Demand A-39
xiii
-------
APPENDIX B
RANGE OF CONTENT OF GASOLINE COMPONENT STREAMS
TABLE OF CONTENTS
»
Section No. Page
B.I Gasoline Blend Survey Request B-2
List of Tables
Table No. Page
B.I Refineries Surveyed for Gasoline Benzene Data .... B-5
B.2 API/NPRA Refinery Gasoline Survey B-6
List of Figures
Figure No. Page
B.I Information Needs for Benzene Removal From
Gasoline Study B-4
xiv
-------
APPENDIX C
ECONOMICS OF BENZENE REMOVAL
FROM REFORMATES & FCC GASOLINE
TABLE OF CONTENTS
Section No. Page
C.I Benzene Removal Study'- Utility Costs C-l
C.2 Refinery Hydrogen Manufacturing Costs C-2
C.3 Regional Cost of Benzene Removal from
Reformates & FCC Gasoline C-5
C.4 Energy Cost for Benzene Removal from
Reformates & FCC Gasoline C-18
C.5 Cost of Benzene Removal for a 10,000 B/D
Refinery . C-23
List of Tables
Table No. Page
C.I
C.2
C.3
C.4
C.5
C.6
C.7
C.8
C.9
PADD I Costs of Benzene Removal from
Re formate
PADD II Costs of Benzene Removal from
Re formate
PADD III Costs of Benzene Removal from
Reformate
PADD IV Costs of Benzene Removal from
Reformate
PADD V Costs of Benzene Removal from
Reformate
Total U.S. Costs of Benzene Removal from
Reformate
PADD I Costs of Benzene Removal from
FCC Gasoline ,
PADD II Costs of Benzene Removal from
FCC Gasoline ,
PADD III Costs of Benzene Removal from
FCC Gasoline ,
, . . C-6
, . . C-7
, . . C-8
, . . C-9
, . . C-10
. . . C-ll
. . . C-12
. . . C-13
. . . C-14
XV
-------
APPENDIX C
List of Tables (Cont.)
Table No. Page
C.10 PADD IV Costs of Benzene Removal from
FCC Gasoline C-15
C.ll PADD V Costs of Benzene Removal from
C.12
C.13
C.14
C.15
Total U.S. Costs of Benzene Removal from
FCC Gasoline
Reformate Energy Requirements
FCC Gasoline Energy Requirements
Hydrogen Energy Reauirements ,
. . . C-17
- . . C-20
, . . C-21
. . . C-22
APPENDIX D Nomenclature D-l
xvi
-------
CHAPTER 1
EXECUTIVE SUMMARY
1.1 Objectives
The primary objective of this study is a preliminary assessment of the
cost of removing benzene from refinery reformates and fluid catalytic cracked
(FCC) gasoline, the two principal sources of benzene in gasoline. Other issues
associated with benzene removal such as octane loss and chemical market impact
are discussed in a general vane.
In order to develop the national impact, the work is divided into five
tasks which are discussed in detail in the following chapters:
CHAPTER 2 - Gasoline Pool Composition
CHAPTER 3 - Benzene Content of Gasoline
CHAPTER 4 - Technological Options for Benzene Removal
from Gasoline
CHAPTER 5 - Economics of Benzene Removal from Reformates
and FCC Gasoline
CHAPTER 6 - Other Economic Issues Associated with
Benzene Removal
1.2 Approach & Results
The first step in this study was the development of the projected gasoline
blend for a future year when gasoline quality, in terms of octane and lead
content, will have stabilized. The year 1981 was selected in order to allow
the lead phase down regulation to take effect, and sufficient time for construc-
tion of facilities to remove benzene from gasoline.
Gasoline production was based on the Carter energy goal forecasts, which
projected gasoline demand increasing to 7,450 MB/D for 1980 through 1982, and
decreasing after 1982. Thus, the base year 1981 would be during the period of
maximum demand and reflect maximum volumes of gasoline to be treated.
1-1
-------
The U.S. pool blend was estimated from the Arthur D. Little Lead Phase-
(2)
Down Study , and a production capacity assessment of current and projected
gasoline-producing equipment. Current gasoline-producing unit capacities were
categorized by size and region. Only firm, announced capacity increases were
included between 1977 and 1981, and no capacity increases were projected beyond
1981, due to decreasing gasoline demand. Projected 1981 unit capacities were
also the basis for scale-up of reformate and FCC gasoline benzene removal costs.
Projected unit capacities, estimated unit yields, and refinery utilization
factors were used to calculate the estimated blend composition on a regional
basis for 1981. The results are shown in Table 1.1.
The second step in this study was to develop the benzene content of the
gasoline component streams. The benzene content of the gasoline components was
/
examined from published data, and a survey of 34 refineries sponsored by the
American Petroleum Institute (API), and National Petroleum Refiners Association
(NPRA). The most noticable characteristic of benzene levels in gasoline is the
wide variation of benzene content of gasoline components and gasoline blends.
Gasoline blend contents varied from 0.15% to 4.26% in a recent survey of U.S.
gasoline. The survey of 34 refineries indicated a variation of 0.2% to 4.0%
benzene in the gasoline pool.
Similar to the range of benzene pool contents reported, the benzene content
of the individual refinery streams shows a significant variation. Typical benzene
levels reported in the API refinery survey were used to determine an average
gasoline component benzene content. These averages were applied to the U.S.
pool blend, shown in Table 1.1, to get an average U.S. pool benzene content. The
results are shown in Table 1.2. The average pool benzene concentration of 1.30%
falls in the range of 1.0 and 2.0 volume percent reported in previous studies.
Refinery reformates and FCC gasoline are the largest volume components in
the gasoline pool, and the major benzene contributors (see Figure 1.1). Whereas
reformate comprises only 30% of the pool, it makes up 66% of the benzene content.
FCC gasoline accounts for 34.5% of the pool, and 20% of the benzene content.
Complete removal of benzene from these two components would control 86% of the
benzene in the gasoline pool.
1-2
-------
TABLE 1.1
ESTIMATED
1981 U.S. POOL BLEND
Stream
Reformats
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Natural Gasoline
Light Hydrocrackate
Isomerate
St. Run Naphtha
Total
MB/CD
2232
2568
1016
104
473
93
188
137
101
538
7450
Vol. %
30.0
34.5
13.6
1.4
6.4
1.2
2.5
1.8
1.4
7.2
100.0
1-3
-------
Figure 1.1
Gasoline Pool Composition and Benzene Contribution
(Volume Percent)
Gasoline Pool
Makeup
Contribution to
Benzene Content
All Other
14.7%
St. Run Naphtha
7.2%
Alkylate
13.6%
FCC Gasoline
34.5%
Reformate
30%
1-100%
- 80%
- 60%
- 40%
- 20%
- 0%
All Other
7%
St. Run Naphtha
7%
FCC Gasoline
20%
Reformate
66%
-100%
- 80%
60%
40%
20%
J- 0%
SOURCE.;, Arthur D. Little calculations
-------
TABLE 1.2
COMPONENT BENZENE CONTENT
ESTIMATED FROM REFINERY SURVEY
Typical Average,..,. Range of Benzene Content
Benzene Content Report in Survey
Vol. % Low Vol. % High Vol. %
Reformate 2.8(2)/3.0(3) 0.5 10.0
FCC Gasoline 0..8 0.2 2.5
Alkylate 000
Raffinate 0.2 0 1.0
Butanes 000
Coker Gasoline 1.4 0.5 2.5
Natural Gasoline 1.5 0.1 3.5
Lt. Hydrocrackate 1.1 0.5 2.0
Isomerate 0.4 0 1.0
S.R. Gasoline 1.4 0.5 3.0
SOURCE: ADL Calculation
(2)1977 operation
(3)1981 operation
Typical Average . . ( .
Gasoline Pool 1.30V '/1.37(> '
Range of Pool Content 0.2 4.0
1-5
-------
The considerable variation in component benzene content (as shown in
Table 1.2), is because of the differences in operation at each location. The
effect of operating variables on benzene content was investigated for the two
largest benzene contributors in the gasolinerefinery reformates and FCC
gasoline.
The primary variables affecting reformate benzene content are the level
of benzene precursors in the naphtha feedstock to the reformer and the overall
process severity. The amount of benzene precursors in the feed is a function
of the origin of the crude oil and the naphtha feed to the reformer. If the
naphtha precursor content and process severity are known, the benzene level can
be predicted more accurately.
Although less is known about the impact of process variables on the
benzene content of FCC gasoline, the primary variables are, once again, the
benzene precursor content of the feed and the severity of operation. Refiners
have not been able to develop definitive trends of the effect of FCC process
variables on FCC gasoline benzene content.
Benzene levels were investigated on a regional, as well as total U.S.
basis. The U.S. was divided according to Petroleum Administration For Defense
Districts as shown in Figure 1.2. PAD District's I and II had similar benzene
distributions to the U.S. pool. PADD III benzene levels were lower, as a
result of high levels of reformates extraction on the Gulf Coast. PADD V benzene
levels were higher than the U.S. because of higher levels of reformate in the
pool.
Reformate and FCC gasoline are the two largest contributors to the benzene
pool level. The effect of removing benzene by gasoline component on pool benzene
content and benzene production, are shown in Figure 1.3. Removing 94.5% benzene
from reformate would lower 1981 pool content from 1.37% to 0.52%, and increase
benzene production .970 billion gallons per year. Removal of 94.5% benzene
from reformates and FCC gasoline would further reduce the pool benzene content
to 0.26%, and increase benzene production 1.27 billion gallons per year. Control
of all gasoline components would reduce average benzene content to 0.08%, and
increase benzene production 1.48 billion gallons per year, or approximately
equal current supply.
1-6
-------
Figure 1.2
Petroleum Administration for Defense
(PAD) Districts
(Inct. Alaska
and Hawaii)
SOURCE; Bureau of Mines 1976 Annual Petroleum Statement
1-7
-------
Figure 1.3
Benzene Removal By Component
(1)
1.3 -
O)
C
S i.o -
(O
CD
4-
O
C
QJ'
c 0.5 -
o
0
Benzene
o
1.37
0.52
0.26
0.165
0.13
0.08
f t t t t 1
Plus Plus Plus Plus
Uncontrolled Control Control Control Control Control
1981 of of of of of
Pool Reformate FCC S.R. Natural All Other
Gasoline Gasoline Gasoline Streams
£ 1 1 * * * *
S 1.5-
C
o
r
1 i.o -
QJ
i.
(O
g 0.5 -
E
cx
c
QJ
M
QJ
CQ 0 -
-K- 1976 U.
S. Demand
0.97
1.27
1.38
1.41
1.48
SOURCE: Arthur D. Little calculations
' ^Based on 94.5% removal efficiency.
1-8
-------
The third step in this study was to investigate the technological options
for benzene removal from gasoline. The main emphasis was to select processing
routes for the two main contributors to the gasoline pool.
Selection of the processing route for reformates and FCC gasoline was
based on a survey of the literature for current benzene removal processes, dis-
cussions with industry sources and a qualitative evaluation of processing route
economics. Although other processing routes were discussed, only commercially
proven processes were considered for final processing route selection.
The processing route selected for refinery reformates was as follows:
STEP 1: Two-tower fractionation of the full boiling range
refinery reformate from gasoline reformers to
concentrate 95% of the reformate benzene in a C
fraction (C, heart cut)
b
STEP 2: Extract 99.5% of the benzene from the C heart
cut to give an overall benzene removal from
reformate of 94.5%.
The naphtha feed to typical gasoline reformers was cut at 170 to 180°F,
True Boiling Point (TBP) with commercial fractionation. The Cr cut on reformate
o
product from gasoline reformers was 160 to 200°F TBP with commercial fractiona-
tion in order to remove 95% of the benzene.
Although several processes were considered, the Sulfolane extraction process
was selected as representative of the extraction process for benzene. The
Benzene extraction process was designed to remove 99.5% of the benzene. The
Aromatic extract was fractionated in a benzene tower to remove traces of toluene.
The benzene was treated in a clay tower for color to produce chemical grade
benzene.
Other commercially feasible processing routes for the removal of benzene
from reformate were rejected on the basis of the level to which benzene could
be reduced, the likely cost of benzene removal, the effect on gasoline octane
and the production of a low quality benzene product.
1-9
-------
The processing route selected for FCC gasoline was as follows:
STEP 1: Two-tower fractionation of the full boiling range
FCC gasoline to concentrate 95% of the benzene in
a C fraction (Cf heart cut).
b o
STEP 2: Hydrogenate the C, heart cut to remove olefins,
di-olefins, and sulfur.
STEP 3: Extract 99.5% of the benzene from the C- heart cut
to give an overall benzene removal from FCC gaso-
line of 94.5%.
The hydrogenation step is required for the FCC gasoline to remove olefins, di-
olefins, and sulfur that may interfere with the extraction process. Although
some sources indicate that extraction of a mixture of olefins and aromatics
may be possible, this has not been commercially demonstrated. Also, a mixture
of olefins and aromatics is not a chemical grade product and would present
additional disposal problems. This route, if feasible, would have the advantages
of lower costs and reduced gasoline octane loss.
As shown in Figure 1.1, the only other significant contributor to the 1981
gasoline pool benzene content is light straight run gasoline. This stream could
also be fractionated to form a C- cut, mildlyhydrotreated to remove sulfur,
followed by sulfolane extraction to remove benzene. The remaining benzene
containing gasoline streams could be handled by similar processing sequences
outlined for reformates and FCC gasoline if further reduction in the gasoline pool
content was desired.
Benzene is a highly desirable, high octane gasoline blending component.
Removal of benzene from the gasoline pool results in a decrease in gasoline
octane. In addition, conversion processes such as the hydrogenation of FCC gaso-
line cause octane losses in other hydrocarbons which further affect the gasoline
pool octane. We have estimated the effect of removing benzene from reformates
and FCC gasoline on an octane barrel basis. The results are shown in Table 1.3.
The key element of the development of the national impact of benzene
removal from reformates and FCC gasoline was the development of the economics
for the selected processing option for each of these streams.
1-10
-------
TABLE 1.3
U.S. POOL OCTANE LOSS
RON MON R+M/2
Refinery Reformates 0.13 0.06 0.10
FCC Gasoline (Hydrogenation) 1.12 0.48 0.80
FCC Gasoline (Extraction) 0.04 0.02 0.03
Total (FCC Gasoline) 1.16 0.50 0.83
Total (Reformates &
FCC Gasoline) 1.29 0.56 0.93
1-11
-------
The economics were first developed on a 1977 Gulf Coast basis for a
base case for reformates and FCC gasoline. The main variable affecting the
economics of benzene removal from reformates and FCC gasoline was determined
to be the total volume to fractionation, hydrogenation, and extraction. Although
extraction costs are somewhat dependent on aromatics content, because of the
greater dependence on total volume to extraction, the economics were assumed
to be independent of aromatics content. The base case economics were scaled
up on a regional basis by capacity in order to get the national impact of
benzene removal in 1977 Gulf Coast dollars.
The details of the calculations of base case economics and scale up are
shown in Chapter 5. The national impact of benzene removal from reformates
and FCC gasoline is shown in Table 1.4.
As can be seen in Table 1.4, the capital requirement in 1977 dollars for
benzene removal from reformates is $2.0 billion. The capital requirement for
removal of benzene from FCC gasoline is $3.3 billion and the investment
required to remove benzene from both reformate and FCC gasoline is $5.3 billion.
There would be some potential savings from economies of scale through combining
the reformates and FCC gasoline streams prior to extraction.
The manufacturing costs to remove benzene from both reformates and FCC
gasoline are $2.5 billion per year. About 52% of these costs are capital
related, 37% variable costs, and 11% for labor and maintenance.
The main component of variable operating costs is energy requirements
for steam, fuel, and utilities. The total energy requirements are 54 million
Crude Oil Equivalent (COE) barrels per year of $648 million per year. Energy
requirements amount to 70% of variable costs or 26% of total operating costs.
The costs of removing benzene from gasoline were converted to costs per
barrel of gasoline using the 1981 estimated gasoline production of 7.45 million
barrels per day. The cost of removing benzene from reformates is 0.82 cents
per gallon of U.S. gasoline, and the cost of removing benzene from FCC gasoline
is 1.37 cents per gallon of U.S. gasoline. The cost of benzene removal from
these two streams is 2.19 cents per gallon.
1-12
-------
TABLE 1.4
NATIONAL COST OF BENZENE REMOVAL
FROM REFORMATS & FCC
GASOLINE
Investment Costs: $ Billion Reformates
Process 1
Offsites 0
Total Plant 1
Other Capital 0
Total Capital 1
Manufacturing Costs: ($M/SD (345 SD/Yr)
Variable Costs
Labor & Maintenance
Fixed Costs 1
Total Manufacturing ($M/SD) 2
Total Manufacturing ($MM/Yr) ^
( 2)
Total Manufacturing (c/Gal) v '
Energy Costs: (Fuel @ $12.00/FOEB)
COE: MB/Yr 21
$MM/Yr
.009
.404
.413
.584
.997
801
329
,592
,722
939
0.82
,930
263
FCC
Gasoline
1.746
0.699
2.445
0.845
3.290
1,886
433
2,207
4,526
1,562
1.37
32,086
385
Total
2.755
1.103
3.858
1.429
5.287
2,687
762
3,799
7,248
2,501
2.19
54,016
648
Based on 345 SD/Yr
.Based on 7,450 B/D gasoline
1-13
-------
These costs are only for removal of benzene from reformates and FCC
gasoline, and do not include the costs of removing benzene from other streams,
or the costs associated with replacing lost octane, gasoline volume, and
benzene disposal.
There are regional differences in gasoline blend and unit capacities
that will cause the impact of benzene removal to be higher in some regions
than the national average. The costs of benzene removal on a regional basis
are shown in Figure 1.4. The most noticable feature of Figure 1.4 is the
higher cost for benzene removal in PADD IV, because of the smaller average
unit sizes in that district. Also, PADD V costs for removing benzene from
reformates are above the national average because of the somewhat higher
concentration of reformate in the PADD V pool.
The national cost of benzene removal from FCC gasoline was based on
producing hydrogen plant hydrogen at all locations with FCC unit capacity.
Some locations may have sufficient reformer hydrogen available at fuel value.
Since a detailed hydrogen balance at each location was beyond the scope of
this study, the sensitivity to hydrogen cost was developed. If all locations
were able to use refinery produced hydrogen at fuel value, the total cost of
benzene removal would drop from 2.2 to 2.0 cents per gallon of U.S. gasoline.
If the hydrogenation step were not required in the removal of benzene from
gasoline, the total cost of benzene removal would drop from 2.2 to 1.6 cents
per gallon of U.S. gasoline.
The most important variable affecting the economics of benzene removal is
the unit capacity. The effect of capacity on benzene removal costs from
reformates and FCC gasoline is shown in Figure 1.5.
The increased costs with decreasing size results in a cost of benzene
removal of up to 7 cents per gallon of gasoline produced for the small refiner,
as compared with the U.S. average of 2.19 cents per gallon. In addition, the
removal of benzene from gasoline would have a greater affect on the small
refiner's ability to blend gasoline because of less operational flexibility and
fewer blending stocks. It is likely that some small refiners may not be able
1-14
-------
I
h-1
Ln
-------
C 4.0
o
c
g 3
2.0
1.0
40 50 60
Gasoline Capacity: MB/SD
70
80
0 10 20 30
Source: Arthur D. Little, Calculations.
Figure 1. 5 Cost of benzene removal vs. gasoline production using refinery-produced hydrogen.
90
100
-------
to remain in the gasoline business when removing benzene from reformates and
FCC gasoline due to the high costs associated with meeting gasoline lead phase-
down requlations and the increased demand for unleaded gasoline.
This study is primarily concerned with developing the level of benzene in
gasoline, the technological options for benzene removal, and the costs of
removing benzene from the two principal sourcesreformate and FCC gasoline.
In addition, the costs associated with restoring the volume and octane quality
of the gasoline pool prior to benzene extraction, and the impact on the chemical
industry of the large increase in benzene supply are key economic issues.
Simple methods were used to suggest the possible magnitudes that these octane
losses, volume losses and chemical market impacts might reach.
The effect on gasoline pool octane of removing benzene from reformates and
FCC gasoline is summarized in Table 1.3. These octane losses can be restored
through some combination of new investment and processing conditions at refineries.
A rough assessment of these costs was made using octane replacement cost data
developed in the Arthur D. Little Lead Phase-Down Study, literature sources,
and other industry studies. The octane replacement penalty is expected to range
from 0.33 to 0.66 cents per gallon of gasoline.
The benzene produced by the extraction of reformates and FCC gasoline
amounts to 82.7 MB/D, or about 1.1% of total gasoline. The idea of converting
the benzene back to a high octane gasoline blending component has obvious attrac-
\
tions since it restores the volume and the octane loss. Rough economics were
developed for alkylating benzene with propylene, which is generally more available
and less expensive than ethylene. The value of benzene as cumene is approximately
equal to benzene fuel value, depending on propylene feedstock and manufacturing
costs. Based on the difference between benzene value as unleaded gasoline and
the alternate values as cumene, the volumetric loss penalty is expected to range
from 0.08 to 0.23 cents per gallon of gasoline.
The removal of benzene from reformates and FCC gasoline will approximately
equal the current benzene supply of about 100 MB/D. Although benzene demand
is projected to increase to about 130 MB/D by 1981, and 170 MB/D by 1985, much
1-17
-------
of this increase is expected to be supplied by benzene extracted from pyrolysis
gasoline as a by-product of olefin manufacture. Even if current imports are
discontinued (5 MB/D), and all toluene hydrodealkylation units are shutdown
(29 MB/D), we project an excess benzene production of about 35 MB/D in 1981,
and 23 MB/D in 1985. With the large excess in benzene supply, prices would
drop to very low levels and stimulate chemical demand for benzene and benzene
derivatives. This could eventually balance the benzene demand with supply,
but the economic dislocations in the chemical industry would be enormous.
With the supply of benzene in excess of the traditional markets, the
price would likely drop, and an alternate use for benzene as fuel or conversion
to cumene for gasoline blendstock would arise. Based on 1976 average prices
and volumes, the potential range of loss of chemical value that the chemical
industry would seek to recover through other chemical products would be 0.50 to
0.61 cents per gallon of gasoline.
The total national cost of benzene removal octane loss, volume from
reformates and FCC gasoline, including octane loss volume loss and chemical
market loss are summarized in Table 1.5. The total cost would range from 3.10
to 3.69 cents per gallon of gasoline.
Although the costs for octane replacement, volume replacement and chemical
market loss are based only on rough calculations, we feel that the range of
costs shown is representative of the likely costs. These costs range from
0.91 to 1.50 cents per gallon and are of the same order of magnitude as benzene
removal costs alone. Because of the magnitude of the costs, these other economic
issues are areas that would warrant additional development in future studies.
Some other areas that were not considered in detail in this study and
would warrant further study are:
New technologies available for benzene removal that
are not now commercially proven.
Evaluation of economics for benzene removal from
light straight run gasoline and other gasoline
streams.
1-18
-------
TABLE 1.5
ROUGH COST OF BENZENE REMOVAL
FROM REFORMATS & FCC GASOLINE INCLUDING OCTANE,
VOLUME & CHEMICAL MARKET LOSSES
Cents per Gallon of Gasoline Low High
Benzene Removal Cost 2.19 2.19
Octane Loss Penalty 0.33 0.66
Volume Loss Penalty 0.08 0.23
Chemical Market Loss 0.50 0.61
Total Cost 3.10 3.69
Million Dollars per Year Low High
Benzene Removal Cost 2,501 2,501
Octane Loss Penalty 379 758
Volume Loss Penalty 95 259
Chemical Market Loss 574 701
Total Cost 3,549 4,219
1-19
-------
Evaluation of the effect of crude oil quality and cut
point changes on benzene level and removal costs.
A more detailed analysis of the small refiner impact.
A more detailed analysis of octane replacement and
chemical market loss.
1.3 Conclusions
a_. The removal of benzene from reformates and FCC gasoline would have
high costs to the petroleum industry.
b_. Costs to small refiners would be up to three times higher than
average costs.
c.. The large volume of benzene produced would present disposal problems
as benzene most likely could not be absorbed in the traditional
chemical market.
d.. Although an in-depth analysis of the chemical market dislocations
and costs of replacement of benzene octane and volume has not been
attempted, the overall rough cost of benzene removal from reformates
and FCC gasoline would be as shown in Table 1.5.
1-20
-------
CHAPTER 2
GASOLINE POOL COMPOSITION
To estimate the benzene content of the United States gasoline pool and
its removal costs, the U. S. gasoline pool composition must be estimated in
terms of principal blend components. Such an estimate of the gasoline pool
composition is developed in this chapter; the benzene level of the U. S. pool
is estimated in Chapter 3; and the cost of benzene removal is projected in
Chapter 5.
Because of regulations requiring the phase-down of gasoline lead levels
and the increasing demand for unleaded gasoline, the benzene content of motor
gasolines marketed in the United States is changing year by year. A base year of
1981 was chosen for this study, by which time these changes should have sta-
bilized. The Carter Energy goal forecasts project peak gasoline production
in 1981, with an absolute decline in volumetric consumption thereafter. There-
fore, selection of a 1981 base year is also advantageous in that it provides
adequate, installed processing capacity for reduction of the benzene content
of gasoline after 1981 as well. Finally, considering the probable time interval
before implementation of regulations governing the benzene content of gasoline,
and the time requirements for engineering design and plant construction of
benzene removal facilities, it is unlikely that benzene regulations could be
implemented before 1981.
Industry-wide statistics to provide a basis for estimating the U.S. gasoline
(2)
pool composition are not available. However, a recent study by Arthur D. Little,
Inc., on the impact of lead additive regulations on the petroleum refining industry
provides a basis for this estimate. That study involved a detailed calibration and
simulation of the U. S. petroleum refining industry, and used a "cluster model"
linear-programming methodology. Through extensive cooperation with the American
Petroleum Institute (API) and the National Petroleum Refiners Association (NPRA),
a model was developed to represent the behavior of the petroleum refining industry
in general and the gasoline pool composition in particular.
2-1
-------
Since the development of this model in 1975, the refining industry has
undergone many changes. Additional processing units have been installed or
announced, actual gasoline growth rates have differed slightly from projections,
and crude slates have varied from projected estimates. Therefore, evaluations
of actual changes were undertaken and the gasoline pool composition was updated
to ensure reliability of these estimates for the current study. As this study
focuses on the impact of benzene removal from catalytic reformate and FCC
gasoline, and as these two streams represent nearly two-thirds of the U.S.
gasoline pool, the composition analysis was directed principally at these two
streams.
An analysis is presented here of the current installed gasoline producing
capacity. Then, assessments of new, announced capacity are made which, combined
with gasoline demand forecasts, allow an evaluation of any further gasoline
capacity requirements to meet 1981 gasoline demand. From this capacity avail-
(2)
ability, we can use the ADL cluster models to estimate the gasoline pool
composition in 1981.
2.1 Assessment of Current Capacity
A tabulation of existing gasoline-producing capacity in U.S. refineries
as of January 1, 1977, is shown in Tables 2.1 through 2.4. Because economies
of scale of benzene removal units are an important component of the economic
impact assessment (see Chapter 5), these tabulations are presented as a function
of processing unit size. Also, to allow an update of the cluster model gasoline
blends, our tabulations are also categorized by Bureau of Mines Refining
District. The refining district classifications and ADL cluster models used
to represent these are presented in Table 2.5 for reference purposes.
Table 2.1 shows the installed reforming capacity in each refinery, by
size range of reforming capacity and by refining district. The table indicates,
for example, that 3,661.8 MB/SD of total reforming capacity is installed in a
total of 179 refineries in the United States. The largest number of refineries
containing reformers is in the Texas Gulf refining district category (44 refin-
eries) , with a combined reformer capacity of 1,163.2 MB/SD. The most common
reformer capacity in each refinery is the 20 to 50 MB/SD range, consisting of
installations in 52 refineries, with a total capacity of 1,582.6 MB/SD.
2-2
-------
TABLE 2.1
1977 U.S. REFORMING - EXTRACTION CAPACITY
Capacity Range
Bast Gout
Saxll Mid-Continent
Uri« Mld-Veat
Innlllaa. Call
Texaa Golf
Uaat Coact
Bocky Hountaina
TOTAL 0.8. A.
N» MB/SD
U KB/CD
X Total
ritar of Baflnerlaa:
%aat Coaat
Saall Mid-Continent
LarM Mid-Meat
loo.Ua. Coif
Vblt Coaat
neky Mountain*
TOTAL U.S.A.
lEFOlMIBG CAPACITT (MB/SD)
0 - 1.9 2.0-4.9 3.0-9.9 10.0-19.9 20.0-49.9 SO. 0-99. 9 iffo TOTAL
4.4 10. 5 9.5 33.0 255.7 112.0 - 423.1
1.2 ; 8.5 41.4 73.3 119.7 - . 244.1
4.0 11.6 23.9 116.8 286.3 263.4 - 706.2
2.2 ' 6.3 - 36.5 256.8 85.0 . - 386.8
6.0 15.0 61.2 59.5 307.8 379.7 334.0 1163.2
2.6 15.6 19.5 84.0 356.1 150.0 - 627. B
2.6 . 21.9 44.1 40.0 - - . - 108.6
23.0 89.4 199.6 443.1 1582. 6 990.1 334.0 3661.8
0.61 2.51 5.51 12.11 43.21 27.01 9:11 100.01
i
34 1 3 7 2 -20
12 6 5 5 19
343 10 9 4 33
22- 2 7 1 .-14
253 6 12 2 ,- 30
367 3 - - - 19
19 28 29 34 52 14 3 179
10. 6X 15.61 16.21 19.01 29.11 7.81 !.'?! 1001
1TX PRODUCT I0» CAPACITT (MB/SD)
0- 0.5- 1.25- 2.5- 5.0- 12.5- fc
0.49 1.249 2.49 4.9 12.49 24.9 250 TOTAL
0.6 - 10.7 5.3 - - 16.6
3.4 - - - - 3.4
10.7 - ... 10.7
- 18.1 ... 18.1
0.4 1.8 4.1 32.0 47.8 41.3 41.0 168.4
3.5 - 2.5 - - 6.0
0.0
0.4 2.4 21.7 60.8 55.6 41.3 41.0 223.2
0.21 1.1J 9.7t 27. 2t 24. 9X 18. 5X 18.41 100.0X
X TOTAL RByORM. « 6. IX
1 - 3 1 - - 3
..]....]
--3----J
2 - - 2
2 - 1 - - 3
------ 0
1 3 9 12 10 4 -39
2.6X 7.7X 23. IX 30.81 25. 6X 10.21 100.01
BTX
X
I«for»
3.9
1.4X
1.5X
4.7X
14.51
l.OX
OX
Hydro-
Dealkyl-
Izatlon
KB/SD
1.6
1.0
2.5
3.4
12.9
0.0
0.0
23.4
0.6
1
1
1
1
0
0
»
CDICIi Oil aad Caa Journal. Batch 21, 1977
-------
TABLE 2.2
1977 U.S. CAT CRACKING - ALKYLATION CAPACITY
KM: elkrlatloo
CAFACITT RAttCE
Capacity: MB/SO
Reat Coeet
tall MidCont
Urge Mldveet
U Gulf
1 Texaa Gulf
Heet Coeet
Rocky Mte.
Total U.S.A.
KB/SO
KB/CD
I Total
ram or REFINERIES
Seat Coaet
Smell M.C.
Large H.H.
LA Gulf
Texaa Gulf
Heet Coaet
Rocky Mte.
Total U.S.A.
Z Totel
SOURCI:
OIL I GAS
0-4.9 5-9.9
2.4 45.3
31.5
_
3.4 38.7
_
9.1 46.2
14.9 161.7
0.31 3.3Z
1 6
4
1 5
3 5
5 20
3.71 14.81
JOURNAL! CATCRACKIJIC CAPACITY'RUBB
10-19.9 20-39.9
74.4
68.3
90.8
26.5
66.0
56.0
72.2
454.2
9.3Z
3
5
6
2
5
4
5
30
22.21
Oil t Cae Journal. Kerch 28
105.0
201.0
329.4
49.0
287.0
160.9
1132.3
23.31
4
8
11
2
10
5
_
40
29. 6Z
, 1977;
40-79.9 2 80
182.0 220.0
85.0
306.4 326.0
78.0 471.8
348.0 686.0
400.0
1399.6 1703.8
28.81 35. OZ
3 2
2
6 3
1 4
6 5
8
26 14
19.31 10. 4Z.
Bureau of Minee,
FEED (MB/SO)
Total
581.4
402.0
1084.3
625.3
1429.1
616.9
127.5
4866.5
100.01
12
22
30
9
32
17
13
135
100.01
Fetrolem
Recycle
71.3
88.9
92.0
29.8
7155.5
123.7
41.4
NR for@
12
22
30
9
17
_li
135
Reflnere
OGJ: Alleviation Production Capacity MB/SD)
' 0-0.9 1-1.9 2-3.9
3.1 11.1
6.2 23.1
4.7 9.1
3.0
6.2 19.5
1.8 5.7
1.7 1.2 15.2
1.7 23.2 86.7
0.2X 2.71 9.9Z
2 4
4 8
3 3
1
1 2
215
2 15 30
1.7Z 12.41 24.81
4-7.9 8-15.9 16
51.9
39.3 28.3
71.0 36.7 82.5
9.6 27.4 91.4
37.1 83.4 88.3
42.5 70.4
199.5 298.1 262.2
22.9S 34.2Z 30. Lt
5
7 2
12 4 4
223
7 7
:z_ =_ =
35 28 11
28. 9 J 23. IX 9. IX
Total
66.1
96.9
204.0
131.4
234.5
120.4
18.1
871.4
100.01
11
21
26
8
17
8
121
ioo. ox
(KB/CD)
Charge
135.9
125.1
245.5
175.5
338.4
142.2
27.5
1190.1
10
20
26
i
17
8
120
Output
61.2
88.5
193.9
125.9
236.1
111.7
17.5
34.1
10
20
It
1
17
_i
120
Annual
-------
TABLE 2.3
1977 U.S. CAT HYDROCRACKING CAPACITY
CAPACITY RANGE
Capacity; MB/SD
East Coaat
Small NldCont
Large Midwest
LA Gulf
Texas Gulf
N) West Coast
1
01 Rocky Hts
Total U.S.A.
MB/SD
Z Total
OMBER OF REFINERIES
East Coast
Small MidCoot
Urge Midwest
LA Gulf
Texas Gulf
Hast Coast
Rocky Mts
Total U.S.A.
X Total
SOURCES Oil
Oil 4 Gas Journal: Distillate Hydrock. (KB/SO) Residual Hydxocracklna
0-4.9 5-9.9 10-19.9 20-49.9 - 50 Total 0-4.9 5-9.9 10-19.9 20.49.9
17.0 30.0 47.0
7.7 7.7
11.0 81.5 92.5 _
29.5 49.0 78.5
5.5 31.7 115.0 68.0 220.2 ~ ~
3.0 62.7 267.2 332.9
4.9 4.9 1.0 ~
21.1 151.9 542.7 68.0 783.7 1.0 '
2.7Z 19.4X 69.2X 8.7X 100.01 100.01
_ 1 1 2
2 2
_ _ 1 3 __ 4 ~ _ ~ _
_ _ 2 2 4
2 2 4 1 9
1 4 9 14 ~
1 - 1 1 ~~ ~~ ~~
610 19 1 36 1
16.7Z 27.8Z 52.7Z 2.8Z 100.0Z 100.0X
. and Oas Journal. Kerch 28. 1977
(MB/SD)
- 50 Total
0
_ 0
_ 0
0
0
_- 0
1.0
1.0
100. OX
0
0
0
0
o
_ 0
1
100. OX
100. OX
*~ufrfl on/o^ff RT4frp
-------
TABLE 2.4
1977 U. S. THERMAL PROCESSING CAPACITY
OIL & CAS JOURHAL: COKING CAPACITY (MB/SD)
Other Thermal Process Capacity (MB/SD)
CAPACITY RANGE
Capacity; MB/SD
East Coast
Small MidCont
Large Midwest
LA Gulf
Texas Gulf
West Coast
Rocky Mts.
i Total U.S.A.
CTi
MB/SD
MB/CD
Z Total
NUMBER OF REFINERIES
East Coast
Small MidCont
Large Midwest
LA Gulf
Texas Gulf
West Coast
Rocky Mts.
Total U.S.A.
Z Total
0-4.9 5-9.9 10-19.9
15.0
7.8 19.5 67.8
59.3
16.0 34.0
8.8 23.9 24.0
7.0 10.5
4.4 14.8 10.0
21.0 81.2 220.6
2.0Z 7.6Z 20.6Z
^^ ^M «
2 3 4
4
2 2
232
1 1
1 JL 1
5 11 15
9.4Z 20.81 28.3Z
20-29.9 30-39.9 - 40 Total
23.7 44.0 82.7
95.1
92.6 34.0 185.9
28.0 50.0 128.0
54.0 61.0 171.7
53.6 67.0 237.8 375.9
29.2
251.9 162.0 331.8 1068.5
23.61 1S.2Z 31. OZ 100. OZ
113
_ _ _ 9
4 19
116
2 2 11
2 2 5 11
_ $
10 5 7 53
18.9Z 9.4Z 13.2Z 100. OZ
C-4.9 5-9.9 10-19.9
14.4
7.1 6.7
7.9 10.0
7.0 13.3
6.7 17.0 30.0
8.8 23.3 29.8
3.4 12.0
33.9 66.0 97.5
7.6Z 14.8Z 21. 8Z
1
21
3 1
1 1
32 3
33 2
2 2
13 9 8
36. 1Z 25. OZ 22.3X
20-29.9 30-39.9 - 40 Total
14.4
13.8
21.0 38.9
47.8 68.1
20.0 85.0 158.7
20.0 55.0 136.9
15.4
61.0 187.8 446.2
13. 7Z 42. 1Z 100.0Z
W. 1--T 1
_ _ 3
1 5
. 1 3
1 1 10
1 1 10
__ __ 4
3 3 36
8.3Z 8.3Z 100. OZ
SOURCE: Oil and Gas Journal, March 28, 1977
-------
TABLE 2.5
DESIGNATIONS OF REFINING DISTRICTS
ADL
Cluster Model
East Coast
Small Mid-Continent
Large Mid-West
Louisiana Gulf
Texas Gulf
Rocky Mountains
West Coast
(1)
PAD District
I
II
II
III
III
IV
V
Refining District
East Coast, Appalachian No. 1
Oklahoma - Kansas - Missouri
Indiana - Illinois - Kentucky
Appalachian No. 2
Minnesota - Illinois - Kentucky
Louisiana Gulf Coast
Texas Gulf Coast
Texas Inland
Arkansas - Louisiana Inland
New Mexico
Rocky Mountains
West Coast (Incs. Alaska & Hawaii)
(1)
Not represented by a specific cluster model.
2-7
-------
Not all of the reformate from these reformers enters the gasoline pool.
Particularly in PADD III, a substantial fraction of reformer capacity is dedicated
to the production of benzene, toluene and xylenes (BTX) for petrochemical sales.
Therefore, a further segregation of reformer capacity is required to ascertain
the reformate level entering the gasoline pool.
Estimates of BTX production were obtained from the 1977 Stanford Research
(3)
Institute Directory of Chemical Producers. These data were compared with
Oil and Gas Journal data on production capacity (Table 2.1) to confirm the source
of this BTX. Individual refiners were also contacted to confirm the Oil and Gas
Journal data, supply missing BTX extraction data, determine the source of BTX
production (e.g., reformate vs. ethylene crackers), and to determine whether any
of this BTX reformate was also blended into the gasoline pool.
At some locations, all available benzene is recovered by segregation of
benzene precursors into one reformer, and the light reformate product is extract-
ed. At other locations, full-range reformate is produced on one or more reformers,
separated into a light reformate cut and extracted. In either case, although not
all reformate is actually extracted, all available benzene is removed and no new
extraction capacity would be required. At other locations, only a portion of
the naphtha feed is segregated for benzene production or only a portion of full-
range reformate is extracted; the remainder is blended directly into gasoline,
so some new extraction capacity would be required for the control of gasoline
benzene content. At many locations, no current extraction capacity exists and
extraction would have to be added for all reformate production. The result of
our analysis was the segregation of reformer capacity between benzene and gasoline
production. Individual refiners were contacted to confirm this segregation of
capacity, and a final assessment of total reforming capacity, BTX reforming
capacity, and net gasoline reforming capacity was obtained.
The resulting net reforming capacity, which produces reformate directed
only into the gasoline pool, is presented in Table 2.6. In addition, 100 MB/D
of heavy reformate from BTX reformers is directed to the PADD III gasoline pool,
but this heavy reformate contains negligible levels of benzene. Negligible
quantities of heavy reformate from BTX reformers enter the gasoline pool in
other PAD Districts.
2-8
-------
TABLE 2.6
1977 GASOLINE REFORMING CAPACITY
BY PADD
REFORMING CAPACITY
RANGE (MB/SD) 0-1.9
2.0-4.9
5.0-9.9
10.0-19.9
20.0-49.9
50.0-99.9
TOTAL USA 19- 28
SOURCE: Arthur D. Little Calculations
26
36
42
TOTAL
PADD
PADD
PADD
PADD
PADD
I 4.4
II 5.2
III 8.2
IV 2.6
V 2.6
N> TOTAL USA 23.0
VO
NUMBER OF
LOCATIONS
PADD
PADD
PADD
* PADD
PADD
I 3
II 4
III 7
IV 3
V 2
10.5 9.5
20.1 65.3
21.3 40.7
21.9 44.1
15.6 19.5
89.4 179.1
4 1
6 9
7 6
6 7
5 3
34.6
217.9
108.5
40.0
84.0
485.0
3
17
7
3
6
211.0
303.9
377.8
366.1
1258.8
6
10
13
13
60.0
233.4
204.0
90.0
587.4
1
4
3
'
1
330.0
845.8
760.5
108.6
577.8
2622.7
18
50
43
19
30
160
-------
Incremental reforming capacity estimates after January 1, 1977 are based
(4)
upon firm, announced capacity additions from the Oil & Gas Journal and Hydrocarbon
Processing. ' The 1981 total reforming capacity, including BTX and gasoline
reformers, is shown in Table 2.7; the gasoline reforming capacity projected
from firm announcements is shown in Table 2.8. In the latter tabulation, we
assumed that all new capacity additions were allocated to gasoline production,
to provide a conservative estimate of benzene removal costs. However, new
incremental BTX capacity is negligible and will have little impact on cost
assessment as discussed in Chapter 5.
As FCC gasoline is not extracted, Table 2.2 is directly useful in calcu-
(4 5)
lating the gasoline pools. Firm announcements ' of new capacity additions
were used to project 1981 FCC unit capacity, as shown in Table 2.9. Projected
reformer and FCC yields were applied to the 1981 reformer and FCC unit capacities
on a PADD basis to get the projected volumes of reformate and FCC gasoline pro-
duction in 1981, as shown in Tables 2.8 and 2.9.
The benzene contribution of each of the other blend components of the U.S.
gasoline pool is relatively small. Hence, alkylation capacity (Table 2.2).hydrocracking
capacity (Table 2.3), and thermal processing capacity (Table 2.4) were used in
(2)
updating the U.S. gasoline pool composition from the earlier study, without
augmentation by announced capacity additions. It was assumed that this capacity
would adequately supply the gasoline component volumes indicated in Section 2.3.
The validity of this assumption, however, does not significantly influence
Chapter 5's impact analysis.
With these capacity assessments available, gasoline yields from reformers
and FCC units can be updated to reflect technological advances and crude slate
(2)
trends since the last study. An analysis of the supply and demand character-
istics of these two processing unit categories can then be used to verify and
update their contribution to the U.S. gasoline pool.
2.2 Reformate and FCC Gasoline Pool Contribution
Using historic and projected unit capacities of gasoline reformers and
FCC units, we can analyze gasoline production rates from these units. This
analysis has two purposes; (1) by comparing historic gasoline production rates
and unit capacities to the cluster model output, the cluster gasoline blends
2-10
-------
TABLE 2.7
1981 TOTAL REFORMING CAPACITY
BY PADD
REFORMING CAPACITY
RANGE (MB/SD) 0-1.9
2.0-4.9
5.0-9.9
10.0-19.9
PADD I 4.4
PADD II 6.2
PADD III 9.7
PADD IV 2.6
PADD V 2.6
TOTAL US 25.5
NUMBER OF
LOCATIONS
PADD I 3
PADD II 4
PADD III 8
PADD IV 3
PADD V 2
10.5 9.5
16.5 53.3
20.3 73.7
23.9 44.1
15.6 19.5
86.8 200.1
4 1
5 8
7 11
7 7
5 3
33.0
216.2
95.0
40.0
103.2
487.4
3
16
7
3
7
20.0-49.9
255.7
436.7
553.1
TOTAL US
20
28
30
36
7
15
19
12
53
50.0-99.9
112.0
263.4
530.2
2
4
8
£100
TOTAL
425.1
992.3
439.0 1721.0
110.6
647.0
439.0 3896.0
16
20
52
63
20
31
186
SOURCE: Oil and Gas Journal, Hydrocarbon Processing
-------
r-o
TABLE 2.8
1981 GASOLINE REFORMING CAPACITY
REFORMING CAPACITY
RANGE (MB/SD) 0-1.9
PADD I 4.4
PADD II 6.2
PADD III 9.7
PADD IV 2.6
PADD V 2.6
TOTAL US 25.5
POTENTIAL REFORMATS
YIELD (MB/SD) 20.5
NUMBER OF
LOCATIONS
PADD I 3
PADD II 4
PADD III 8
PADD IV 3
PADD V 2
2.0-4.9
10.5
16.5
20.3
23.9
15.6
86.8
69.7
4.
5
7
7
5
BY PADD
5.0-9.9 10.0-19.9
9.5 34.6
53.3 244.0
53.2 107.5
44.1 40.0
19.5 103.2
179.6 529.3
144.1 425.5
1 3
8 18
8 7
7 3
3 7
20.0-49.9
211.0
334.4
366.3
366.1
1277.8
1029.5
6
11
13
13
50.0-99.9
60.0
233.4
374.5
1
4
5
>100 TOTAL
330.0
887.8
931.5
110.6
597.0
0
0
TOTAL US
20
28
27
38
43
11
2856.9
2297.4*
18
50
48
20
31
167
*Does not include 100 MB/SD heavy naphtha from BTX reformers.
SOURCE: Arthur D. Little calculations
-------
TABLE 2.9
ISJ
I
U)
1981 CATALYTIC CRACKING CAPACITY (Fresh Feed)
BY PADD
Catalytic Cracking
Capacity Range
(MB/SD) 0-4.9
PADD I
PADD II 2.4
PADD III 3.4
PADD IV 9.1
PADD V
TOTAL U.S. 14.9
Potential FCC
Gasoline Yield 8.4
Number of Locations
PADD I
PADD II 1
PADD III 1
PADD IV 3
PADD V
5.0-9.9
-
77.3
43.7
39.7
-
160.7
91.1
_
10
6
4
-
10.0-19.9
74.4
147.1
92.5
72.2
56.0
442.2
249.7
3
10
7
5
4
20.0-39.9
105.0
560.4
322.0
23.5
160.9
1171.8
660.7
4
20
11
1
5
40.0-79.9
182.0
391.6
555.0
-
400.0
1528.6
857.4
3
8
9
-
8
>80.0
220.0
326.0
1157.8
-
-
1703.8
968.6
2
3
9
-
-
Total
581.4
1504.8
2174.4
144.5
616.9
5022.0
2835.9
12
52
43
13
17
TOTAL U.S.
20
29
41
28
14
137
SOURCE: Arthur D. Little calculations
-------
can be verified and adjusted, giving an adjusted blend for use in the present
study, and (2) with this additional model calibration, firm announced capacity
additions can be compared to projected gasoline demand to determine whether
the announced capacity is adequate for future production levels; this analysis,
in turn, will identify the capacity required for benzene removal from these
blend components.
To examine future demands on the available processing units, we used the
Carter Energy plan gasoline forecast:
Year Gasoline Demand. MB/D
1977 7,250
1981 7,450
1985 7,000
The supply/demand analysis for FCC units and catalytic reformers indicates
that existing plus announced, firm capacity additions will provide adequate gaso-
line capacity for the indefinite future. Specifically, the gasoline demand pro-
jections peak in about 1981, with a continuous decline thereafter. Although
capacity will be tight in 1981, it will meet 1981 demand and become increasingly
surplus thereafter. It is not surprising that the industry has announced con-
struction plans adequate to meet projected demand for the next three years;
however, the decline in gasoline demand after 1981 is unusual from an historical
viewpoint. Of course, limited expansions could occur after 1981, because of an
individual refiner's lack of access to the excess unit capacity owned by other
refiners.
Methodology
(2)
The FCC unit yields from the cluster model should reflect the changing
impact of crude slate, FCC unit feed hydrogenation, and the lead phase-down
requirements between the individual years studied in the EPA lead phase-down
study, 1973, 1977, 1980 and 1985. These yields were reviewed and revised as
necessary to reflect changing crude slates.
A tabulation of historic levels of FCC capacity and actual Bureau of Mines
gasoline production was made yearly from 1970 through 1976. Various percentages
of FCC gasoline in the pool were assumed in the vicinity of the cluster model
2-14
-------
predictions. From the gasoline production figures, assumed percentage of FCC
gasoline, FCC unit gasoline yield, and FCC unit capacity, a stream-day utiliza-
tion factor could be calculated. The assumed percentage of FCC gasoline in
the pool, which gave about 90% of stream-day utilization during periods of
significant FCC capacity growth, was taken as the best estimate of FCC gasoline
percentage in the total pool. Although individual refiners indicated that FCC
gasoline percentage can vary over a considerable range, our calculated percentage
was confirmed with selected refiners as reasonable.
A similar procedure was followed in estimating the percentage reformate
in the gasoline pool. However, as noted earlier, a substantial fraction of the
published reformer capacity is dedicated to BTX production, which is not directly
applicable to gasoline pool calculations, and reformer capacity was therefore
segregated between BTX production and gasoline production. After reconfirming
this segregation with individual refiners, "gasoline reformer capacity" assess-
ment could be made, which ranged from 100% of total reforming capacity in PADD
IV, to about 50% of total capacity in PADD III. Reformate percentage estimate
in the gasoline pool was then calculated by the same technique used for FCC units,
and was then confirmed as being in a reasonable range by discussions with indi-
vidual refiners.
Results
As discussed in detail in Appendix A, each PAD District was analyzed
separately, using historic FCC unit capacity and representative FCC unit gasoline
yields for that district. Under conditions for which FCC unit capacity was
expanding rapidly, we assumed that the units operated near 90% stream-day utiliza-
tion. This provided FCC gasoline production estimate which, when compared to
the total gasoline production of the District, allowed an estimate of the percen-
tage FCC gasoline in the total pool. Summary of the results is illustrated in
Figure 2.1; the histogram indicates total installed stream-day capacity in PADD's
I-IV, whereas the solid line represents estimated FCC feed rates. When the solid
line reaches 90% of the stream-day capacity, the units are fully utilized. The
dashed lines indicate expected demands on FCC capacity that will be required to
meet projected gasoline demand in the future.
2-15
-------
Figure 2.1
PADDs I-IV FCC Unit Capacity/Demand
4,500 !
4,000
p
u
PQ
3
P
Q 3,500
I
V
H
n
U
<
5J
u
u
Pn
3,000
?. snn
/
/
/
t
,
/
(
A
t
.
*
y
>
. ; '
x
(
,
'"'.' Capacity
i '' "'/
i
_^ * **. '
^ . ''>. Demand
! S.
\ 'j
X
' i
1 : i 1
: , ;
1
; ; ;
.1 '
i i :
.'I',! ! ' i .
''; ' ] , ' . ^
70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
YEAR
2-16
-------
From this analysis by PAD District, we concluded that the installed
FCC capacity indicates FCC gasoline to be 34.5% of the U.S. gasoline pool
in 1981. This estimate confirms the validity of the independent estimate of
:he
(6)
(2)
33.5% FCC gasoline in the pool from the cluster model results and corresponds
to another published estimate of 38%.
A similar analysis was conducted to determine the reformate in the gasoline
pool of each PAD District consistent with the gasoline reformer capacity as
developed in Section 2.1. In this analysis, reformer yields were used that
reflect substantial operation with bimetallic-catalysts and recent trends in
U.S. crude slates. Results are presented in Figure 2.2. The top histogram
shows total reformer capacity; diagonally lined portions indicate estimated BTX
capacity. When the solid line, representing historic naphtha feed rate, or
the dashed line, representing naphtha feed projected to supply future gasoline
demand reaches 90% of the open histogram, the gasoline reformers are near
full utilization.
From this analysis by PAD District, we concluded that the installed gaso-
line reformer capacity indicates reformate to be 30% of the U.S. gasoline
(2)
pool in 1981. This estimate agrees with the cluster model results of 25.7%
and independent estimates of 33% reformate in the U.S. pool.
2.3 1981 Gasoline Pool Composition
The supply/demand analysis described in Section 2.2 provides an estimate
of the contribution of reformate and FCC gasoline to the gasoline pool produced
in each PAD District, which can in turn be projected to provide a U.S. gasoline
pool. As shown in Table 2.10, the reformate percentage varies from a low of
25% in PADD III to a high of 43% in PADD V. In PADD III, of course, substantial
reformate fractions are recovered for BTX production, thereby being excluded
from the gasoline pool. Because of substantial reformer capacity, the initial
boiling of the naphtha charge is lower in PADD V than in the rest of the nation;
this wider-boiling reformate, then, constitutes a larger fraction of the gaso-
line pool. As noted in Chapter 3, this factor, when combined with significant
Alaskan North Slope Crude runs, will likely result in a somewhat higher benzene
content of PADD V gasolines than is experienced in the rest of the nation.
Because of the preponderance of reformate in the PADD V pool, the FCC gasoline
fraction of the pool is the lowest of any PAD District.
2-17
-------
3500
Figure 2.2
PADDs I-IV Reformer Capacity/Demand
o
n)
nl
O
0)
Q
OJ
o
3000
2500
2000
1500
BTX Reformer
Capacity
Mogas Reformer
Capacity
\
-Capacity
1000
Year 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85
-------
TABLE 2.10
1981 U. S. POOL GASOLINE BLEND (BY PADD)
PADD I
PADD II
PADD III
PADD IV
PADD V
Total U.S.
Component :
Reformats
FCC Gasoline
Alkylate
Raffinate
Butanes
to
jL Coker Gasoline
VO
Natural Gasoline
Lt. Hydrocrackate
Isomerate
St. Run Gasoline
Total
MB /CD
244
309
108
10
48
21
12
0
16
45
813
Vol %
30.0
38.0
13.3
1.2
6.0
2.6
1.5
2.0
5.5
100.0
MB/CD
.670
758
346
10
130
31
62
0
27
198
2232
Vol %
30.0
34.0
15.5
.4
5.8
1.4
2.8
1.2
8.9
100.0
MB/ CD
757
1090
408
79
218
30
100
67
58
221
3028
Vol %
25.0
36.0
13.5
2.6
7.2
1.0
3.3
2.2
1.9
7.3
100.0
MB/CD
78
96
38
0
14
3
5
1
0
17
252
Vol %
31.0
38.0
15.0
5.6
1.2
2.0
.4
6.7
100.0
MB/CD
483
315
116
5
63
8
9
69
0
57
1125
Vol %
43.0
28.0
10.3
.4
5.6
.7
.8
6.1
5.0
100.0
MB/CD
2232
2568
1016
104
473
93
188.
137
101
538
7450
Vol %
30.0
34.5
13.6
1.4
6.4
1.2
2.5
1.8
1.4
7.2
100.0
-------
Gasoline pool rafflnate fraction was established from the BTX extraction
analysis discussed in Section 2.1. Coker gasoline, hydrocrackate and isomerate
(2)
levels were established by revising the cluster model results as necessary
to bring them into agreement with the capacity statistics of Tables 2.3 and 2.4.
Alkylate levels were established by correcting the cluster results for the
revised FCC gasoline production estimates. Butanes were blended for approximate
vapor pressure control and straight run gasoline was obtained by difference,
while ensuring that substantial agreement with the cluster result for this com-
ponent was maintained.
2-20
-------
CHAPTER 3
BENZENE CONTENT OF GASOLINE
To evaluate the ambient benzene associated with benzene-containing gaso-
line and to examine the cost of benzene removal to the petroleum industry,
it is necessary to specify the current and projected gasoline benzene content.
The primary focus of this study is the cost of removing benzene from two major
contribution streamscatalytic reformate and FCC gasoline; thus, these streams
were examined in more detail. As will be indicated in Chapter 5, however, the
cost of benzene removal from these streams is not heavily dependent on their
benzene content.
It is well known that wide variations in benzene content exist in
gasolines marketed in the United States. A survey of 34 refineries represent-
ing the petroleum industry was made with the cooperation of the Benzene Task
Force of the American Petroleum Institute (API) and the support of the National
Petroleum Refiners Association (NPRA). The 34 refineries reported benzene
contents of their gasoline pool and individual blending components. These
were coded in order to maintain confidentiality and furnished to ADL. From
the survey data we estimated that the current benzene content of gasoline
ranges from 0.2% to 4.0%, with a nationwide average of 1.3%.
The survey data was not entirely adequate to estimate regional variations
in motor gasoline benzene content; therefore the results of a recent 211 sample
duPont survey were used to indicate regional trends. We found that PADD V
exhibits somewhat higher benzene content in gasoline than does the rest of the
nation. This is believed to be due to three factors: (1) the reformer feedstock
is fractionated in such a fashion that more benzene-forming precursors are
included than in the rest of the nation, (2) the crude oil processed probably
inherently contains more benzene precursors than the crude oil processed in the
rest of the nation and (3) little benzene is extracted for petrochemical use.
3-1
-------
It is necessary to project the 1981 benzene content of the gasoline pool
to evalute future trends relevant to any regulation of gasoline benzene content.
An evaluation of the important refinery variables influencing gasoline benzene
content can assist in this effort.
Good predictive ability of reformer variable influence was achieved,
because these variables are well understood as a result of historic commercial
benzene production for the petrochemical industry. Because of increased octane
requirements of unleaded gasoline alone, the 1977 benzene content of 1.3% is
expected to increase to 1.37% by 1981. Additional changes will occur because
of changing U.S. crude slates; although the impact on reformate benzene level of
for changing crude sources can be predicted in principle, inadequate crude
assay data precluded quantitative projections. It is noteworthy, however, that
the Alaskan North Slope crude has a great propensity for contributing to high
gasoline benzene level.
The ability to project changes in benzene level because of future trends
in FCC unit operation is currently lacking. However, we believe that these
variables are of secondary importance in projecting future gasoline benzene trends,
because of counter-balancing effects of variables necessary to maintain unit
heat balance and the relatively small changes in operations required between
1977 and 1981.
Finally, after projecting the 1981 benzene content in gasoline, we can
determine the hypothetical levels of benzene in the gasoline pool by removing
benzene from each of the individual blend streams. Control of benzene in
catalytic reformate will reduce the U.S. pool benzene level to approximately
0.52 Vol. %, and the control of benzene in both reformate and FCC gasoline will
reduce the U.S. pool benzene level to approximately 0.26 Vol. %. These calcula-
tions only indicate achievable results, as they were not made while maintaining
constant pool octane. The estimated cost of such benzene removal is presented
in Chapter 5; possible consequences of benzene extraction on the petrochemical
industry are considered in Chapter 6.
3-2
-------
3.1 1977 U.S. Pool Benzene Level
Every study of gasoline benzene content to date has indicated wide varia-
tions in measured benzene content of gasoline samples. In fact, wide variations
exist in the benzene content of gasolines sold in the U.S. because of differences
in feedstock quality, processing configurations, and operating conditions.
For example, a refiner processing Arabian Light crude oil and feeding a 180° to
400°F naphtha to the reformer will observe less than 0.5% benzene in the refor-
mate. A refiner processing North Slope crude oil and feeding a 140° to 310°F
naphtha to the reformer will observe more than 10% benzene in the reformate
(see Section 3.3). A refiner extracting benzene for petrochemical sales may
market gasoline containing less than 0.5% benzene, whereas a refiner blending
pyrolysis gasoline for octane enhancement may market a gasoline exceeding
4% benzene. Ranges of benzene content of individual gasoline blend :components
from previous studies are shown in Table 3.1; reported ranges of U.S. pool
benzene content are shown in Table 3.2.
This study will determine not only the benzene content of the average
gasoline pool, but also estimates of the benzene content of the several gasoline
blend components of Table 3.1 to examine the effect of benzene content control
of each stream. The survey of the 34 U.S. refineries represented about two-
thirds of U.S. gasoline production capacity. The refineries surveyed comprise
25 refineries owned by 8 major refiners, and 9 owned by independent refiners
(see Table 3.3). In the survey and subsequent portions of the study, ADL
cooperated extensively with representatives of the Environmental Protection Agency,
members of the API and the NPRA, and representatives of the API Benzene Task
Force; this collaboration resulted in enhanced accuracy in representating
current industry operations and projecting future trends. The survey form
used and the survey data obtained are presented in Appendix B along with a dis-
cussion of the criteria for evaluating and compilating the survey data.
The survey results, abstracted in Table 3.4, show that the average typical
pool benzene level of the refiners surveyed was 1.25%, with a range of typical
concentrations from 0.6% to 2.5%. Note, however, that this is not a production-
weighted average and, therefore, is not representative of the U.S. pool benzene
content. Also, the 34 individual refineries produce gasoline containing benzene
ranging on the average from 0.8% to 1.8% and occasionally as high as 4%.
3-3
-------
TABLE 3.1
ILLUSTRATIVE RANGES OF BENZENE CONTENT OF BLENDING COMPONENTS
Range
Component Vol. %
Reformate 1.0-7.0
FCC Gasoline 0.1-3.0
Alkylate 0
Raffinate 0-1.0
Butanes 0
Coker Gasoline 0.5-2.5
Natural Gasoline 0.1-3.5
Light Hydrocrackate 0.5-2.0
Isomerate 0
S.R. Gasoline 0.5-3.0
Pyrolysis Gasoline 0.5*-15.0
*Pyrolysis gasoline after benzene extration
Sources:
(1) PEDCO Environmental, "Atmospheric Benzene Emissions", prepared
for the Environmental Protection Agency, October 1977.
(2) R.K. Burr, "Benzene Extraction from Motor Gasolines," internal
memorandum to J.F. Durham, Environmental Protection Agency,
June 21, 1977.
(3) Arthur D. Little estimates
3-4
-------
TABLE 3.2
U.S. POOL BENZENE CONTENT SURVEY RESULTS
Range Typical
Source Vol. % Vol. %
NIOSH 1976(1) 0.88-2.00 1.24
Gulf Oil, Oct. 1976(2) 0.54-2.39 1.25
duPont, June 1977 0.15-4.26 1.00
(4)
PEDCO Environmental, Aug. 1977V --- 2.00
Sources:
(1) Hartle, R., & Young, "Occupational Exposure to Benzene at
Service Stations," National Institute for Occupational Safety
and Health, Cincinnati, Ohio.
(2) Runion, H.E., "Benzene Gasoline II," Gulf Science and
Technology Co., American Institute Hygiene Association,
38(3), August 1977.
(3) E.I. duPont de Nemours & Co., "Hydrocarbon Distribution in
Commercial Gasolines-Summer 1976," June 1977.
(4) PEDCO Environmental, "Atmospheric Benzene Emissions", prepared
for the Environmental Protection Agency, August 1977
3-5
-------
TABLE 3.3
REFINERIES SURVEYED FOR GASOLINE BENZENE DATA
PAD District
I
I
I
I
II
II
II
II
II
II
II
II
III
III
III
III
III
III
III
III
III
III
III
III
III
III
V
V
V
V
V
V
V
V
Refiner
Arco
Exxon
Gulf
Witco Chemical
Amoco
Amoco
Delta Refining
Gulf
Indiana Farm Bureau
Mobil
Shell
Union
Amoco
Arco
Chevron
Exxon
Exxon
Gulf
Gulf
Louisiana Gloria
Marion Co.
Mobil
Shell
Shell
South Hampton
Union
Arco
Beacon
Chevron
Chevron
Mobil
Petrochem
Union
U.S. Oil & Rfg.
Location
Philadelphia, PA
Bayway, NJ
Philadelphia, PA
Bradford, PA '
Sugar Creek, MO
Whiting, IND
Memphis, TENN
Toledo, OH
Mt. Vernon, IND
Joliet, ILL
Wood River, ILL
Lemont, ILL
Texas City, TX
Houston, TX
Pascagoula, MS
Baton Rouge, LA
Baytown, TX
Belle Chasse, LA
Port Arthur, TX
Tyler, TX
Theodore, AL
Beaumont, TX
Houston, TX
Norco, LA
Silsbee, TX
Beaumont, TX
Carson, CA
Hanford, CA
El Sequndo, CA
Richmond, CA
Torrance, CA
Ventura, CA
Los Angeles, CA
Tacoma, WASH
3-6
-------
RESULTS OF REFINERY BENZENE SURVEY
ON TOTAL GASOLINE POOL
Benzene Content,
Reported Average Pool
0.9
1.0
1.4
0.5
2.0
0.6
1.8
0.35
1.5
0.6
0.7
0.8
1.0
2.5
1.1
1.1
1.5
0.9
0.8
1.3
0.8
1.5
0.9
1.0
1.1
2.4
1.39
1.75
0.8
1.37
1.6
0.8
3.4
1.2
1.25
0.6-2.5
% (Vol.)
Reported Range
0.7-1.5
0.2-2.5
1.2-1.6
0.4-0.8
0.4-3.1
0.5-1.0
1.2-2.5
0.3-0.4
1.3-1.8
0.6-1.2
0.3-1.6
0.7-0.9
0.5-1.6
0.9-4.0u;
*
0.7-1.4
0.2-2.0
0.7-0.9
*
0.4-1.8
0.2-2.5
0.2-2.5
0.6-1.3
*
0.8-2.0
*
1.0-2.4
1.6-1.8
0.6-1.0
1.26-1.59
*
0.7-0.9
3.0-4.0
1.0-1.4
0.8-1.8
0.2-4.0
Refinery Code
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
Arithmetic
Average
Range
*
Not reported
Includes pyrolysis gasoline which is normally extracted
(2)
Arithmetic average of samples, not weighted by volume production
3-7
-------
Estimated average benzene concentrations in gasoline blend components
from the survey are shown in Table 3.5. Clearly, the stream exhibiting the
highest concentration of benzene is catalytic reformate, with an average concen-
tration of 2.8% and a highest reported observation of 10%. As indicated in
Chapter 2, reformate also comprises 30% of the gasoline pool, so its contribution
to gasoline pool benzene is very large. Although FCC gasoline contains only
0.8% benzene, FCC gasoline comprises about one-third of the pool; hence, FCC
gasoline also contributes significantly to the pool benzene level. Pyrolysis
gasoline is shown in Table 3.1 to have a very high benzene concentration. The
quantities of unextracted pyrolysis gasoline blended into the U. S. pool are
believed to be small, but this conclusion should be verified in any subsequent
work.
Estimates of the volume-weighted average benzene content of each stream
component and the 1977 U. S. gasoline pool from this study are shown in Table 3.6.
As indicated earlier, the average U. S. pool benzene content is estimated to
contain 1.3% benzene, two-thirds of which is due to catalytic reformate. The
combined contribution of reformate and FCC gasoline represents more than 85%
of the U. S. pool benzene content. Obviously, efficient control of these two
sources would substantially reduce the pool benzene content.
3.2 1977 Regional Pool Benzene Level
Because the sampling of 34 refineries is inadequate for defining regional
variations in the benzene content of gasolines, benzene data collected in a 211
sample duPont survey in 1976 were used to indicate directional trends. This
survey was based on gasoline samples taken from 16 major U. S. cities. The cities
sampled, the distribution of samples among these cities, and the average benzene
content by PAD District are shown in Table 3.7. The average of all U. S. samples
is shown in Table 3.7 to be 1% benzene; however, this average is not weighted
by volumetric production, and is indicated below to generally agree with the
estimate in Table 3.6.
The distribution of the duPont data by benzene content is shown in Figure
3.1, plotted by PAD District. Also plotted in Figure 3.1 are the 34 typical
pool gasoline data taken from the refinery survey (Table 3.4). PADD IV has
not been plotted because of the small number of samples and the single sample
location.
3-8
-------
TABLE 3.5
RESULTS OF REFINER BENZENE SURVEY
ON GASOLINE BLEND COMPONENTS
Benzene Content. % (Vol.)
Blend Stream Estimated Average Reported Range
Reformate 2.8 0.5-10.0
FCC Gasoline 0.8 0.2-2.5
Alkylate 0 0
Raffinate 0.2 0-1.0
Butanes 0 0
Coker Gasoline 1.4 0.5-2.5
Natural Gasoline 1.5 0.1-3.5
Light Hydrocrackate 1.1 0.5-2.0
Isomerate 0.4 0-1.0
Straight-Run Gasoline 1.4 0.5-3.0
^ 'Based on 56% C^ ISOM capacity with estimated 0% benzene in
Cc Isomerate and 1% benzene in C/- Isomerate.
3-9
-------
TABLE 3.6
ESTIMATED BENZENE LEVEL OF U.S. GASOLINE POOL
1977
Volume Percent
Pool Composition
Component
Reformate
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Natural Gasoline
Lt. Hydrocrackate
Isomerate
S.R.Gasoline
TOTAL
Estimated Average Benzene Content
MB/D
2175
2500
986
102
464
87
181
131
102
522
%
30.0
34.5
13.6
1.4
6.4
1.2
2.5
1.8
1.4
7.2
%
2.8
0.8
0
0.2 -
0
1.4
1.5
1.1
0.4
1.4
Pool Contribution, %
0.84
0.28
0
<0.01
0
0.02
0.04
0.02
<0.01
0.10
7250
100.0
1.30
3-10
-------
TABLE 3.7
DISTRIBUTION OF DUPONT SURVEY GASOLINE SAMPLES
City
Atlanta, Ga.
Jacksonville, Fla.
Miami, Fla.
Philadelphia, Pa.
Newark, N. J.
Total PADD I
Detroit, Mi.
Chicago, 111.
Kansas City, Ka.
Wichita, Ka.
Oklahoma City, Ok.
Tulsa, Ok.
Total PADD II
Houston, Tx.
New Orleans, La.
Total PADD III
Denver, Co.
Total PADD IV
Los Angeles, Ca.
San Francisco, Ca.
Total PADD V
TOTAL U.S.
PAD District
I
I
I
I
I
II
II
II
II
II
II
III
III
IV
V
V
No. of
Samples
68
81
20
11
31
211
Average
Benzene
Content
Vol. %
1.00
0.99
0.96
0.92
1.09
1.00
3-11
-------
40 -,
OA
30 -
20 -
10 -
0
PADD I:
I
0
68 Samples
in 5 Cities
1
. ,
_o> hrr-* 1 o ~n . .. _.,_i
30 _
20 -
OJ
a 10
E
& n
u 0
"o
i/»
c3 30
u
I 20
o
«,0
Q
£ o
03
30 _
20
10
0
1
n
0
8
8
o
8
PADD II:
81 Samples
in fi PitipQ
o ' ' ft *- n i 1 i
1 i 1 i i i i
PADD III:
20 Samples
in 2 Cities
O O O 0 1 0 1
n-r
0
ft
o
o
o
o
PADD V:
31 Samples
in 2 Cities
o J i i o r i
i i i ii i
~>
o
o
0 0
o o
n o o O
0?5
1 o
o
8
1.0
o o
O I O
0 ft X t ft f ft
T i
1.5 2.0
USA:
21 1 Samples
in 16 Cities
O Refinery Pool
Benzene
0 Concentrations
i X i
2.5 3.0
3.5
Benzene Concentration (Volume %)
Source: E.I. du Pont de Nemours & Co., "Hydro-Carbon Distribution in Commercial Gasolines
Summer 1976". June 1976.
Figure 3.1 Distribution of PADD gasoline samples by benzene content Summer 1976
3-12
-------
In general, the distributions indicate probable benzene content between
0.5% and 1.2% for PADD's I through III and the total U. S. , with a substantial
tail in the distribution extending to very high benzene levels. This tail could
be from the variation in crude type with the PAD Districts, but is more likely
because of few refiners operating their catalytic reformers with a low initial
boiling point naphtha or poor fractionation in the naphtha prefractionator.
PADD III may indeed have a bimodal distribution as suggested by the
duPont data, because of the large numbers of BTX extraction facilities there.
The peak at low benzene concentrations may represent refiners with extraction;
the high peak may indicate refiners without extraction. However, because the low
peak is no lower than evident for PADDs I and II, the bimodal distribution is
more likely due to inadequate sampling.
PADD V shows a probable benzene level between 0.8% and 1.5%, and is believed
to be somewhat higher than the rest of the nation because of crude differences
and differences in reformer operations (see Section 3.3).
The distribution of refinery data from the ADL survey generally agrees
with the duPont data, considering the small number of samples in both surveys.
For example, as indicated above for PADD III, the 20-sample duPont survey should
not be interpreted to suggest that the benzene contents in PADD III contain no
elements in the 1% to 2% range. Rather, the 34 locations surveyed in the
current study could indicate a similar population to the duPont survey, with
the differences arising from random sampling. We conclude, without rigorous
statistical analysis, that the gasoline benzene populations from which both the
ADL and duPont surveys were drawn are probably similar, if not identical. The
number-average benzene content from the duPont survey of 1% benzene is probably
identical to the volume-weighted average benzene content of 1.3% from the ADL
survey. Differences arise from sampling errors and weighting functions, and
are influenced heavily by PADD III, a major volume contribution to the U.S. pool.
In Figure 3.2, the U. S. distribution is shown on an expanded scale, again
illustrating the long tail resulting from few samples of high benzene concentration.
3-13
-------
u>
25
20
_
Q.
E-
§ 10
Distribution on a 0.2%
Interval
Distribution on a 0.1%
Interval
0 0.5 1.0 1.5
Benzene Concentration (Volume %)
Source: E.I. du Pont de Nemours & Co., "Hydro-Carbon Distribution in Commercial Gasolines -
Summer 1976." June 1976.
Figure 3.2 U.S. distribution of gasoline samples by benzene content Summer 1976
-------
In Figure 3.3, the duPont data are plotted as a cumulative frequency
distribution for PADDs I through IV, PADD V, and the total U. S. Again, the
higher benzene concentration in PADD V is quite evident. For example, the benzene
content in 50% of the gasoline samples of PADDs I through IV exceed 0.75%; in
PADD V, over 80% of the samples exceeded 0.75% benzene. Insufficient data are
available to ascertain any differences for cumulative distribution for PADDs I-IV.
3.3 Projected 1981 U. S. Pool Benzene Level
The primary variables expected to affect the benzene content of the U. S.
gasoline pool between 1977 and 1981 are the changing U. S. crude slate (particu-
larly the penetration of Alaskan North Slope crude oil) and the changing reformer
severity due to lead phase-down. A description of the influence of the important
operating parameters on reformate and FCC gasoline benzene content is discussed
below. These variables are then evaluated as to their likely impact on U. S.
pool benzene level in 1981.
Reformer Variables
The conditions under which catalytic reformers produce benzene are well
understood because (a) the reactions are relatively well-defined and (b) the
production of benzene from BTX reformers has been an important petrochemical
process for many years. As indicated in Figure 3.4, the main benzene-producing
reactions are dehydrogenation and dehydrocyclization, although isomerization and
dealkylation also play a role. The fundamental variables are the benzene precursor
levels in the naphtha feedstock to the reformer and the overall process severity
(or degree of conversion of these precursors to benzene).
The amount of benzene precursors in the feed is a function of crude oil
origin and the boiling range of the naphtha feed to the reformer. For example,
Alaskan North Slope naphtha has the highest inherent concentration of benzene
precursors of any naphtha examined in this study, exceeding even ftiat of Nigerian
naphtha. In contrast, naphtha from Arabian Light crude oil contains low concen-
trations of these precursors. Furthermore, the concentration of these precursors
depends upon the initial boiling point of the naphtha feed to the reformer.
3-15
-------
100
90
S 80
a
£ -m
Cumulative Percent Gasoline
S 8 8 o S S S
A
PADD SI-IV / ^
(180 Samples)/^ /^ /
,
A
o
/ X /
/ /^
^ /
r /
/Q
f-\
j-j
PADD
V
(31 Samples)
n ^^-i
|>/;
Y rf\
//
/H
]
xx Ub/3
tf^n
r
(21 1 Sample;
A
\
>l
A
Q PADD I-IV
0 PADDV
^ USA
0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5
. ._ Benzene Concentration (Volume %)
Source: E.I. du Pont de Nemours & Co.
Figure 3.3 Cumulative percent PADD gasoline sampta by benzene content - Summer 1976
-------
Figure 3.4
TYPICAL REFORMING REACTIONS
^ ^.9.
Dehydrogenation - -]
L_ R.CC-C-C ^T-*"
-T
Dehydrocydiiation R-C-C-C-C .___
*****
CR C C C C ,i_ .*
/VR
C
Hydrocrotking R-C-C-C i H, *
Dcmclhylalion R-C C-C-C 1 M, »-
Dculkylalion l^iJ *" ^' *
Transalkylalion - 2 K jT *
^^ H,
V
R ^
R-C-C = C-C 4 H,
w + H>
R"
(?J + H,
C .
RCCC
R'
C
RH i C-C-C
H
R C.C CH + CH,
(O) ^ CH,
(e.g.)
©cl!'©
t-vg.) W)
SOURCE: Oil and Gas Journal, p. 86, November 15, 1976.
3-17
-------
For example, operation with a North Slope naphtha boiling above 180°F can contain
fewer precursors than operation with an Arabian naphtha boiling above 140°F.
Finally, many commercial fractionators do not operate with efficient fractiona-
tion, which also affects the precursor content of the feedstock.
The effects of these variables are shown in Table 3.8. The Alaskan North
Slope naphtha always produces two to three times more benzene upon reforming
than Arabian Light naphtha, all other conditions being constant. However, the
benzene content of each naphtha can vary by a factor of two, depending upon the
boiling range of the naphtha feed. The further effect of process severity, as
measured by the reformats octane number, is shown in Table 3.8 to be comparatively
minor, particularly when it is recognized that most commercial reformers operate
between 95 and 100 RON.
We expect that most straight-run naphthas fed to U. S. reformers will
fall between the extremes shown in Table 3.8, although such stocks as heavy
hydrocrackate may exceed these levels. Because of the importance of the few
specific chemical compounds which are benzene precursors, however, an excellent
reformer feedstock for gasoline production is not necessarily a high benzene
-I-.- < v Y . > . T ,. >,. . ,. 3,,; 4.,.. , ..,.,;. i _ _ ^. L. ^..
producer. For example, Table 3.9 shows the predicted benzene content of refornate
for four naphthas ranked in decreasing order of gasoline yield. Note that
predictions of benzene content of the reformats from the five principal precursors
agree with measured benzene levels for Alaskan North Slope and Arabian naphthas.
However, the predicted benzene content does not rank in the same order as the
gasoline quality of the reformer feed.
An indication of the effect of the naphtha initial boiling point on benzene
yield is shown in Table 3.10. Here, the principal benzene precursors are ranked
in the order of increasing boiling points, along with their contributions to
benzene yield from the reformer. If a perfect fractionation of 140°F were made
for the reformer feed, all these precursors would be fed to the reformer and the
benzene yield would be as indicated, either 8.63% or 3.50%, depending on the
crude oil source. By contrast, if the cut were made at 170°F, the hexanes and
methycyclopentane would remain in the light straight-run gasoline, thereby
eliminating 35% to 45% of the reformats benzene yield with 140°F reformer feed.
3-18
-------
u>
I
TABLE 3.8
.EFFECT OF REFORMER PARAMETERS
ON REFORMATS BENZENE CONTENT
Naphtha Source
TBP Feed Boiling Range, °F
Reformats Octane, RONC
Benzene Content of Reformate, %
-ALASKAN NORTH SLOPE-
160/380 160/380 140/310 140/310
90 100 90 100
4.8 5.6 8.4 10.2
ARABIAN LIGHT ; *
160/380 160/380 140/310 140/310
90 100 90 100
1.6 2.0 3.8 5.3
-------
N)
o
Nigerian Medium
Alaska North Slope
Nigerian Light
Arabian Light
TABLE 3.9
COMPARISON OF BENZENE CONTENT OF REFORMATE
FROM SELECTED NAPHTHAS
160/380°F
100 RON
ESTIMATED BENZENE IN REFORMATE DUE TO NAPHTHA
N+2A(1)
93.7
76.8
76.5
50.0
Cf Paraffins
G
0.14
0.33
0.20
0.24
Cyclohexane
2.09
2.95
3.15
1.03
Methycyclopentane
0.78
1.28
1.40
0.21
PRECURSOR, LV %
Benzene
0.35
1.49
0.37
0.14
Estimated
C + Total Benzene
0 3.36%
0 6.05%
0 5.12%
0.42 2.04%
Measured
Total Benzene
-
5.83%
-
2.10%
(1) Naphthene -I- 2 x Aromatics level in naphtha feed, an indicator of high gasoline yield naphthas.
-------
TABLE 3.10
SOURCES OF BENZENE IN REFORMATS
100 RON SEVERITY
BENZENE
YIELD TO FEED OF
140-310°F NAPHTHA
BENZENE PRECURSORS BOILING POINT. "F ALASKA NORTH SLOPE ARAB LT.
i-Hexane 140.5 0.34 0.29
n-Hexane 155.7 0.72 0.71
Methylcyclopentane 161.3 1.95 0.57
Benzene 176.2 2.50 0.60
Cyclohexane 177.3 3.12 1.03
C?+ 180+ 0 0.26
0
3-21
Total 8.63 3.50
o
-------
Further increasing the initial boiling point to 180°F would, in theory, eliminate
all of the benzene from the North Slope reformate and more than 90% of the
benzene from the Arabian Light reformate. These results would only directionally
be achieved in practice because (a) the fractionation is not perfect and some
precursors would be contained at all cut points and (b) alterations in reformer
severity would be required as the cut point is changed in order to maintain
gasoline pool octane.
As the cut point is increased, temperatures exceeding 177°F would result
in inclusion of benzene in the straight-run gasoline stream. The simplified
processing route in Figure 3.5 yields the resulting C,. - 400°F gasoline benzene
content of Table 3.11. The benzene content of the reformate and the blend of
straight-run gasoline and reformate are shown (Table 3.11) as a function of
idealized cut point between the streams. Note that this blend is only straight
run naphtha and reformate and excludes butanes as well as other major gasoline
blending components. The blended C - 400°F gasoline from North Slope naphtha
ranges from 1.7% to 6% benzene, but cannot fall below 1.7% because of the
benzene content of the straight-run component. The Arabian Light gasoline
varies between 0.6% and 2.6% benzene, with contributions from both the straight-
run gasoline and dealkylation in the reformer regardless of cut point.
As discussed in Appendix A, we believe that PADD I through IV refineries
operate at a 170°F to 180°F cut point, whereas PADD V refineries are well
represented by a 140°F or 150°F cut point. As a result, this factor alone should
cause the benzene content of PADD V gasolines to be higher than in the rest of
the nation (Figure 3.3). Furthermore the benzene content of PADD V reformate
should increase further as North Slope crude fully penetrates West Coast
markets, as indicated by a comparison of Tables 3.6 and 3.11.
Secondary effects on benzene production occur from new bimetallic catalysts
in catalytic reformers. For example, these catalysts allow lower pressure
&
reformer operation than possible with platinum catalysts. As i%dicated in
Figure 3.6, lower pressure operation favors the equilibrium yields of benzene
from its precursors. As indicated in Figure 3.7, the benzene yield is slightly
increased because of lower operating pressures. Another means of exploiting bi-
metallic catalysts is to increase the reformer severity resulting in increased
benzene yield (see Table 3.12).
3-22
-------
Figure 3.5
SIMPLIFIED REPRESENTATION OF NAPHTHA
PROCESSING FOR GASOLINE PRODUCTION
u>
N>
CO
Cs - 400°F
Naphtha
D
I
S
T
I
L
L
A
T
I
0
N
C. - TBP
Straight Run Gasoline
REFORMER
REFORMATS
C5 - 400°F
Gasoline
TBP - 400QF
-------
TABLE 3.11
ILLUSTRATIVE EFFECT OF REFORMER INITIAL_
BOILING POINT ON PRODUCT BENZENE LEVEL
ALASKAN NORTH SLOPE
ARABIAN LIGHT
u>
ro
TBP CUT POINT. °F
140
150
160
170
180
BENZENE IN C5-400°F
GASOLINE POOL
6.0%
5.7%
5.1%
3.7%
1.7%
BENZENE IN
400 °F E.P. REFORMATS
6.4%
6.1%
5.5%
4.1%
0
BENZENE IN C5-400°F
GASOLINE POOL
2.6%
2.2%
1.6%
1.2%
0.6%
BENZENE IN
400° F E.P. REFORMATE
3.0%
2.6%
2.0%
1.5%
0.2%
-------
Figure 3.6
POTENTIAL EQUILIBRIUM YIELD OF
BENZENE FROM ITS PRECURSORS AT 950°F
to
I
NJ
U1
100 |
90
80 I
70 I
60 I
50 U
40 L.
30
20
10
0
0
Cyclohexane
Methylcyclopentane
100
200
300
400
500
600
Hydrogen Partial Pressure, psia
SOURCE: Chevron Research Co., "Production of Aromatic Hydrocarbon by Low-Pressure Rheniforming"
NPRA Meeting, San Antonio (1976)
-------
Figure 3.7
EFFECTS OF PRESSURE AND SEVERITY
ON YIELDS OF BENZENE FROM
130-310°F Arabian Naphtha
CO
H
0)
oT «
-------
TABLE 3.12
EL SEGUNDO AROMATICS REFORMER
EFFECTS OF CHANGEOVER TO RHENIFORMING CATALYST
Feed
Boiling Range
P/N/A, LV%
Catalyst Type
Pressure, psig
LHSV, Hr-i
H2/HC, Mole Ratio
Run Length, Months
Average Yield Decline, LV%
MCP Conversion, %
Yields from Feed, LV%
Benzene
Californian
Straight Run
C6-C7
47/49/4
Pt
250
1.65
3.0
2-3
5-6
60-80
6.7
Rheniforming
250
2.0
2.5
7
<1
90-95
9.4
SOURCE: Chevron Research Co., "Production of Aromatic Hydrocarbon by Low-
Pressure Rheniforming", NPFA Meeting, San Antonio (1976)
3-27
-------
FCC Unit Variables
Substantially less is known about the impact of process variables on
the benzene content of FCC gasoline, because it has not been an historic source
of benzene for the petrochemical industry. In general, however, it is known
that catalytic cracking reactions proceed with a substantial preservation of
ring structure. Therefore, the benzene content of the gasoline will be a func-
tion of benzene precursor content in the feed (e.g., substituted single-ring
aromatics) and severity of operation.
In general, the origin of the gas oil should influence gasoline benzene
content, with the paraffinic gas oils providing the lower benzene levels in
the FCC gasoline. As such, Alaskan North Slope gas oil will likely produce
higher benzene levels than Arabian Light gas oils, if operated at the same
conversion level. Similarly, FCC feed hydrotreating should directionally
reduce FCC gasoline benzene content. Because FCC units must be operated in
heat balance, however, paraffinic and hydrotreated feeds are usually processed
at higher conversion levels, thereby increasing the conversion of the precursors
that are present and moderating the reductions in benzene content otherwise
expected.
Similarly, if all other variables are held constant, increasing conver-
sions will increase the gasoline benzene level. The use of zeolitic catalysts,
which allow higher conversion levels, should directionally increase the gasoline
benzene content. Again, because it is necessary to maintain heat balance on
the unit, other process variables are often changed simultaneously.
During the course of this study, several refiners examined their process
data on FCC gasoline benzene content and were unable to develop definitive
trends of the effect of process variables. Due to the counter-balancing effects
of the process variables, however, it is not likely that the benzene content of
FCC gasolines will change significantly by 1981.
1981 Pool Composition
Any projection of future benzene content of U.S. gasolines must reflect
future changes in crude slate, expansions of processing units to meet future
3-28
-------
gasoline demands, and changes in the operational mode of these units to meet,
for example, lead phase-down requirements. In approaching such a problem, it
must be recognized that many complex interactions exist in the petroleum
refining industry, and compromises among various processing routes will be
utilized to respond to these changes while minimizing capital investments and
manufacturing costs. To estimate future benzene levels carefully, therefore,
linear programming techniques would be useful. Because insufficient data on
crude oil benzene precursors are currently available, an alternate approach
was used, which exploited an extensive data base on industry-wide operations
(2)
from earlier linear programming analyses, and which was augmented by an
independent analysis of the effects of these variables on gasoline benzene
content.
(2)
The results of this earlier lead phase-down analysis provide a reasonable
representation of likely future refinery operation in 1981. From these data,
the average FCC unit conversion is projected to increase only from 72% in
1977 to 73% by 1981. Also, with revised estimates of gasoline demand growth,
Chapter 2 shows that the current percentage of FCC gasoline in the U.S. pool of
34.5% is expected to remain approximately constant through 1981.
Finally, as discussed above, changes in crude slate between 1977 and 1981
are expected to have only a secondary effect on FCC gasoline benzene concentra-
tion, because of counter-balancing changes required to maintain heat balance
on FCC units. Therefore, we believe that a reasonable approximation to the 1981
benzene content of FCC gasoline is to maintain the estimated 1977 level of
0.8% benzene of Table 3.5.
Although the reformate percentage in the gasoline pool will not change
significantly by 1981, reformer severity will certainly increase in the refin-
ing industry, because of the completion of lead phase-down and the market
(2)
demand for unleaded gasoline. An earlier study indicated that average U.S.
reformer severity will increase from 96 to 99 RON between 1977 and 1981. The
predictive ability of benzene content due to reformer severity is good, as
indicated in Table 3.9. Whether the changes in severity are between 94 and
3-29
-------
97 or between 96 and 99 RON, the same changes In benzene content are predicted,
(2)
due to the approximate linearity seen in Figure 3.7. Hence, the earlier study
was used only to indicate that a 3 RON increase in reformer severity is expected
by 1981; the results presented here do not depend on the absolute level of current
reformer severity. That is, it is irrelevant whether the current severity is
indeed 96 RON.
To estimate the change in benzene content because of reformer severity,
then, the effect of reformer severity changes on several typical U. S. reformer
feedstocks was projected. We found that the benzene content of the reformate
increased by about 2.7% for each octane number change of the reformate, depend-
ing slightly upon naphtha source, naphtha boiling range, and octane level. It
is estimated that reformate benzene content will increase by 8.1% between 1977
and 1981, or from an absolute level of 2.8% (Table 3.5) in 1977 to 3.0% by
1981, because of reformer severity changes alone.
The impact of changing U. S. crude slate could also significantly influence
reformate benzene levels, as suggested in Table 3.9: In particular, increasing
proportions of Middle East crudes could diminish the benzene content of PADD I
through III reformates, whereas increasing proportions of Alaskan North Slope
crude could significantly increase PADD V reformate benzene levels. Estimating
the impact of this variable quantitatively, however, requires crude assay data
on the principal benzene precursors of Table 3.9 for the important domestic and
foreign crudes. Because these data are not currently available, further cor-
rections in benzene content of the reformate due to crude slate changes are not
now possible. Such a compilation must be completed before further analysis on
benzene content-.of motor gasoline is undertaken.
i
i
Pyrolysis gasoline is not expected to increase significantly in the gasoline
pool by 1981. Chapter 6 does indicate that pyrolysis gasoline production will
/
expand dramatically over the next decade; however, we expect that economic con-
siderations will favor the extraction of the high benzene content in pyrolysis
gasoline for petrochemical sales. Any regulation specific only to benzene levels
in reformates and FCC gasoline would invalidate this conclusion.
3-30
-------
The other blend components of Table 3.6 contribute only in a minor way
to pool benzene concentration, so their percentage in the 1981 gasoline blend
was held constant. Nominal changes will occur in these percentages, but such
changes are beyond the level of precision of the present study, being much less
than the changes due to crude slate.
The revised benzene content for 1981 gasolines is presented in Table 3.13.
We estimate that the average benzene content of U. S. gasolines will be approxi-
mately 1.37%, without correction for changes in the probable U. S. crude slate
by 1981.
3.4 Selective Removal of Benzene from Blend Components
Having determined the benzene content of the individual gasoline blend
components in 1981 and their volumetric contribution to the total gasoline pool,
we can rank the major blend components in terms of their contribution to the
pool benzene level. This ranking is shown in Table 3.14.
With an estimated 1.37% benzene, the total gasoline pool in 1981 is
projected to contain about 102 MB/D of benzene, or about 1.6 billion gallons per
year. For the purposes of comparison, the total petrochemical demand for benzene
in 1981 is projected to be about 2 billion gallons per year. The supply/demand
consequences of benzene removal from U. S. gasolines will be discussed in
Chapter 6.
On a volumetric basis, reformate and FCC gasoline represent about two-thirds
of the total gasoline pool. However, because of their high benzene content,
these two streams account for 85% of the total benzene in the pool. The major
source of benzene by far is catalytic reformate, representing nearly two-thirds
of the benzene in U. S. gasolines.
The majority of the benzene in U. S. gasolines is represented by only a
few blend components: More than 90% of the benzene is contained in the first
three entries of Table 3.14, and more than 95% is contained in the first four
entries. It should be noted, however, that for tabulation purposes, these four
blend components represent a grouping of many individual refinery streams. For
example, the four components are split into many subclasses for individual product
3-31
-------
TABLE 3.13
ESTIMATED BENZENE LEVEL OF 1981 GASOLINE POOL
Volume Percent
Pool Composition^ Typical Average Benzene Content
Component
Reformate
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Natural Gasoline
Lt. Kydrocrackate
Isomerate
S.R. Gasoline
TOTAL 7450 100.0 1.37
MB/D
2235
2571
1014
104
477
89
186
134
104
536
%
30.0
34.5
13.6
1.4
6.4
1.2
2.5
1.8
1.4
7.2
%
3.0
0.8
0
0.2
0
1.4
1.5
1.1
0.4
1.4
Pool Contribution, %
0.90
0.28
0
<0.01
0
0.02
0.04
0.02
<0.01
0.10
3-32
-------
TABLE 3.14
BENZENE CONTAINED IN GASOLINE POOL
Volume
Contribution to
Component U.S. Poolj MB/D
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Re formate
FCC Gasoline
S.R. Gasoline
Natural Gasoline
Lt. Hydrocrackate
Coker Gasoline
Isomerate
Raffinate
Alky late
Butanes
2235
2571
536
186
134
89
104
104
1014
JLLL
1981
Benzene
Contribution to
U.S. Pool, %
0.899
0.275
0.100
0.037
0.033
0.017
0.006
0.003
0
0
Contained
Benzene
MB/D
67.0
20.5
7.5
2.8
2.5
1.3
0.4
0.2
0
0
Cumulative
Percent
Contained
Benzene
65.6
85.6
93.0
95.7
98.1
99.4
99.8
100.0
100.0
100.0
TOTAL
7450
1.370
102.2
100.0
3-33
-------
sales, such as solvents and naphtha-jet fuel, and for octane blending flexibility.
Therefore, control of the benzene content of these streams by the addition of
extraction facilities will vary with location rather than simply adding a
single unit to each of four refinery streams.
With this simplification in mind, the hypothetical reduction of benzene
in gasoline achievable by extracting these gasoline blend components is shown
in Table 3.15. The original, unextracted pool is shown to contain 1.37%
benzene. The result of progressively adding extraction facilities to each blend
component is then shown, assuming 95% recovery in fractionation to obtain the
extraction plant feed and 99.5% extraction efficiency for an overall control
of 94.5% (see Chapter 5 for additional details).
If only reformate is controlled the gasoline pool benzene content is
reduced to 0.52%. Compared to an uncontrolled pool containing 1.37% benzene,
this results in a 62% reduction of benzene content and produces 63.3 MB/D of
extracted benzene.
If both FCC gasoline and reformate are extracted, the resulting pool
benzene level is 0.26%, an 81% reduction in the uncontrolled benzene level,
with an associated production of 82.7 MB/D of benzene. To obtain about 90%
reduction in pool benzene would require extraction of the first four blend
components, providing a pool benzene content of 0.13%. Because of the efficiencies
assumed, the lowest achievable pool benzene content is 0.078%, or 94.3% reduction
(see Table 3.15). Diminishing reductions in pool benzene and rapidly increasing
costs, although not examined in this study, are expected.
No attempt was made in Table 3.15 to maintain pool octane constant. Hence,
even the benzene content after reformate extraction, 0.52%, should only be viewed
as an indicative level, because readjustment of the pool composition would be
required to maintain pool octane. For example, the FCC gasoline contribution
to the pool may be increased to maintain octane levels, thereby increasing pool
benzene levels above 0.52%. Rough costs for octane replacement were developed
in Chapter 6. More definitive projections of pool benzene level would require
linear programming runs to predict new gasoline blends upon extraction of each
component in turn.
3-34
-------
TABLE 3.15
POOL BENZENE LEVELS ACHIEVED
BY EXTRACTION OF BLEND COMPONENTS
1981
Uncontrolled
Gasoline Pool
Streams Extracted
(Cumulative)
Pool Benzen
Content, %
1.37
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
Re formate
FCC Gasoline
S.R. Gasoline
Natural Gasoline
Lt. Hydrocrackate
Coker Gasoline
Isomerate
Raffinate(2)
Alky late
Butanes
0.520
0.260
0.165
0.130
0.102
0.083
0.078
0.078
0.078
0.078
Percent Reduction
of Uncontrolled
Pool
Cumulative Extracted
Benzene, MB/D
0.520
0.260
0.165
0.130
0.102
0.083
0.078
0.078
0.078
0.078
62.0
81.0
88.0
90.5
92.6
93.9
94.3
94.3
94.3
94.3
63.3
82.7
89.7
92.3
94.6
95.8
96.3
96.3
96.3
96.3
(1)
(2)
Benzene removal based on 95% recovery in fractionation to produce C, cut
and 99.5% removal in extraction.
Already an extraction product. No further removal assumed.
3-35
-------
CHAPTER 4
TECHNOLOGICAL OPTIONS FOR BENZENE
REMOVAL FROM GASOLINE
A wide variety of technological options can be contemplated for benzene
removal from refinery gasoline blending streams. Ideally, all of these options
would be made available in a linear-programming algorithm and would afford
selection of the option or combination of options which serve to minimize manu-
facturing costs. Because benzene-related information for such model runs is not
currently available in sufficient detail to warrant such an approach, we made
preliminary evaluations of candidate processing routes and selected a route which
is believed to provide representative costs of benzene removal from motor gasolines.
As shown in Chapter 3, the two major benzene contributors to the gasoline
pool are refinery reformates and FCC gasoline. Thus, the primary emphasis in
this chapter is the selection of a processing route for each of these streams.
Cost assessments associated with these routes are presented in Chapter 5. However,
to reduce the benzene content of gasoline below 0.26%, it is necessary also to
remove benzene from blend streams other than reformates and FCC gasoline.
Technological options for processing other blend streams are also discussed, but
their costs of benzene removal have not been assessed in this study.
In selecting processing routes, a primary criterion used was that the
process be commercially proven. Secondly, the processing route must be highly
efficient, removing at least 90% of the benzene in the process stream under
consideration, without adding significantly to the benzene content of other
process streams. Thirdly, for candidates passing both of these tests, a qualita-
tive evaluation of likely costs of benzene removal was made, and the least expensive
option was chosen. In this context, it was recognized that a process which
greatly reduced pool octane levels would incur substantial economic penalties
for octane upgrading.
4-1
-------
We concluded that extraction of benzene-containing reformate heart cut
was the preferred option for benzene removal from refinery reformates. For
benzene removal from FCC gasoline, extraction of a hydrogenated heart cut was
selected, with the hydrogenation step being required to remove olefinic and sulfur
bearing compounds to preserve characteristics related to normal commercial practice.
In each of these processing routes, substantial optimization and energy integra-
tion would be expected for any application in a specific refinery. Because of
the generalized approach taken in this study, the economics of these processing
routes, which will be discussed in Chapter 5, could not reflect such optimization.
The above processing routes for reformate and FCC gasoline, although not
criteria used in process selection, result in the production of chemical-grade
benzene. Other possible processing routes are discussed herein which do not
produce such a high-purity benzene. In particular, direct extraction of an FCC
gasoline heart cut (without pre-hydrogenation), which would produce an extract
containing perhaps 90% benzene and 10% olefins and sulfur compounds, could greatly
reduce costs of FCC gasoline extraction. Although this route is feasible, it
was not judged to be commercially proven. Although this option was not examined
in detail in the present study, it clearly warrants further examination, includ-
ing disposition alternatives of the extract. As evident from Table 3.11,
adjustments of the reformer feed cut point can reduce reformate benzene content.
However, such results are highly crude-specific and do not generally result in
substantial removal efficiencies. Further study of this route is also warranted
to define the degree of control achievable at lower cost, particularly for small
refiners, as noted in Chapter 5.
Finally observations are provided in this chapter on the likely technological
routes useful for controlling benzene in other gasoline blend components.
4.1 Reformate Benzene Control Technologies
Reformates are generally characterized by a boiling range from about 100°F
to 400°F and a high concentration of total aromatics. Although, as suggested
in Chapter 3, the benzene content of reformate can vary widely, processing '
characteristics of catalytic reforming result in a stream virtually free of
olefinic and sulfur-bearing compounds (see Table 4.1). The initial boiling point
4-2
-------
TABLE 4.1
ILLUSTRATIVE REFORMATE PROPERTIES
Gravity, °API
ASTM Dist, °F
IBP
10%
30%
50%
90%
EP
Naphtha Charge
'I 52.6
F
198
222
242
264
340
376
ane, Clear 54.6
ane +3cc TEL 76.2
(Vol. %)
42.7
0.9
s 37.8
18.8
0.5
Reformate
43.8
110
180
222
252
342
415
100.0
104.2
20.5
1.5
1.5
76.5
0.2
Parafins
Olefins
Naphthenes
Aromatics
Sulfur, ppm
SOURCE: Hydrocarbon Processing, 55, No.9, September (1976)
4-3
-------
of the naphtha feed to the reformer can range from 130°F to 200°F, and the
reformate product is often separated into several fractions for gasoline blending
flexibility. Also, some of these fractions may already be extracted for petro-
chemical benzene, toluene and xylene (BTX) production.
Processing Route Selected
The processing route selected for the study is shown in Figure 4.1 to
consist of two primary steps:
STEP 1: Fractionate the full range refinery reformate to produce a
C, heart cut.
o
The full range reformate product from gasoline reformers if fractionated
in two new towers to produce a C, heart cut. In the first tower, an iso-hexane
o
and lighter cut is removed from the full range reformate and sent to gasoline
blending. In the second tower, a C, cut is recovered for extraction and the
o
C7+ reformate is directed to gasoline blending.
STEP 2: Extraction to remove benzene in a new sulfolane unit.
The Udex, Arosolvan, and Sulfolane processes are all employed for aromatics
extraction and would be applicable for removal of benzene from the C, reformate
heart cut. As the process economies for these processes are similar and the
Sulfolane process is widely used in the industry, the Sulfolane process was
selected as representative of extraction processes for benzene. In this step,
the benzene is extracted, treated in a clay tower, and recovered as chemical
grade benzene in a benzene tower.
From economic analyses, we determined that the cost of benzene removal by
gasoline extraction is primarily dependent upon the volume to be extracted and
not upon the stream benzene concentration for the range of concentration encount-
ered in reformates and FCC gasoline (although the cost per unit of benzene
extracted does depend heavily upon benzene concentration). Obviously, a very
expensive fractionation train could be built, giving efficient fractionation, a
narrow-boiling range (small volume) heart cut, and a small extraction plant
investment. Alternatively, a less expensive fractionation train could be consid-
ered, requiring a wide boiling range heart cut to recover all the benzene and
a large extraction plant investment. After consideration of these trade-offs,
4-4
-------
Catalytic
Reformats
13,330 B/SD
Benzene
400 B/SD
i
Ln
11,330 B/SD
Deisohexanizer (New)
C6 Heart Cut
2000 B/SD
Benzene
380 B/SD
Sulfolane
Extraction
Unit (New)
C, Fractionator (New)
o
Figure 4.1
FLOW DIAGRAM FOR
BENZENE REMOVAL FROM CATALYTIC REFORMATE
Overhead
2000 B/SD
Benzene
378 B/SD
Raffinate
1622 B/SD
Benzene
2 B/SD
C + Gasoline
9330 B/SD
Benzene
20 B/SD
-------
it was concluded that fractionation should be designed to provide a 160 to 200°F
C, heart cut, about 15% of a typical reformate stream. This is expected to
provide about 95% recovery of the reformate benzene. Furthermore, a benzene
tower was required on the extract stream to recover toluenes for blending into
the gasoline pool, thereby minimizing pool octane losses. Obviously, these
trade-offs must be considered in detail for any given refinery, and the design
parameters could change significantly between locations.
Very high extraction efficiencies also increase extraction plant investment.
As only 95% recovery was assumed for the fractionation columns, 99.5% efficiency
was judged to be adequate for the extraction plant efficiency. Hence, overall
benzene recovery from the reformate becomes 94.5%.
In addition to the selected processing route for benzene removal from
refinery reformates, several other processing routes were investigated. Process-
ing routes rejected for reformates were as follows:
Aromatics Extraction of Total Reformate
Although the extraction economics are largely independent of aromatics
concentration at the 10% level, increased aromatics concentration can decrease
operating costs at higher aromatics levels. For aromatics levels above 80%,
different processing routes can become attractive, such as extractive distillation.
Therefore, aromatics extraction from the total reformate was evaluated to determine
whether cost advantages would accrue from higher aromatics concentration. It
is not surprising that this method was more costly than the selected route, for
this method is not even used for current BTX production.
Split Light and Heavy Reformate, followed by BTX
Extraction of Light Reformate
Splitting the total reformate into two fractions and extracting the light
reformate fraction (160 to 310°F) is the current commercial method of BTX
production. Although marginal operating cost benefits exist for this approach
as compared to the selected route, resulting from higher concentrations of
aromatics in the feed, this approach would be uneconomic relative to extraction
of a C,. cut because of the increased volume extracted and the associated large
extraction plant investment. Also, this approach would require additional
4-6
-------
fractionation facilities to separate benzene, toluene and xylenes products after
extraction, in order that toluene and xylenes may be blended into the gasoline
pool to minimize octane losses.
Prefractionation of Naphtha Feed Without Extraction
of Benzene from Light Naphtha
A benzene reduction in the gasoline pool could be obtained by increasing
the reformer naphtha feed initial boiling point to about 200°F TBP and eliminating
most C- precursors from the reformer charge. This would not recover naturally-
occurring benzene in the naphtha, which would remain in the fraction bypassed
around the reformer. With the increase in naphtha feed cut temperature, adjust-
ment of reformer severity to maintain pool octane levels would be necessary (see
Chapter 3).
This option was not selected because of the relatively smaller reduction
in the benzene content of the gasoline pool and the paucity of relevant crude
assay data. However, this could be an attractive alternative for small refiners,
where the size of required extraction facilities for a C,. cut is very small and
o
thus very costly per barrel of gasoline sold.
C, Heart Cut of Reformate & Deep Hydrogenation to
Remove Benzene
Complete hydrogenation of a C, heart cut from the reformate has the advantage
of eliminating extraction costs, providing a product which can be blended directly
into the gasoline pool, and eliminating the benzene disposal problem. However,
the costs of hydrogenation of the C, cut from reformate would largely, if not
completely, offset the savings in extraction costs. In addition, complete hydro-
genation would result in a 20 octane number loss in the C& heart cut, which would
have serious implications for the gasoline pool. Finally, deep hydrogenation
of refinery reformates is not judged to be a commercially practiced process.
4.2 FCC Gasoline Benzene Control Technologies
Fluid catalytic cracking (FCC) units convert high molecular weight gas
oil fractions principally into gasoline, but also produce by-product fuel oil
fractions. As the cracking of high molecular weight hydrocarbons occurs in the
absence of hydrogen, the products are highly unsaturated and contain substantial
4-7
-------
levels of olefins and aromatics. Olefins and aromatics exhibit high octane ratings;
thus these unsaturated compounds are highly desirable components of FCC gasoline.
Because no hydrogen is present during cracking reactions, sulfur is substantially
retained in the products from the FCC unit. Indeed, the FCC gasoline sulfur
<»
content is, by far, the highest of any gasoline blend component used in the U.S.
today. Since only small quantities of H2S were produced in the hydrogenation step,
no facilities were provided to remove
Full boiling range FCC gasoline is generally characterized by a boiling
range from 100° to 400°F and a high concentration of olefins, aromatics and sulfur,
as shown in Table 4.2. This gasoline is commonly separated into at least two
fractions, a light FCC gasoline and a heavy gasoline, for octane blending flexi-
bility.
Process Route
The processing route selected for this study is shown in Figure 4.2 to
consist of three primary steps:
STEP 1; Fractionation of full range FCC gasoline
This fractionation is accomplished in two new towers. The full range FCC
gasoline is deisohexanized in the first tower. The deisohexanized FCC gasoline
is then fractionated to obtain a C, heart cut from the second tower. The C_+
o /
gasoline stream is then directed to the gasoline pool.
STEP 2 ; Hydrogenation to remove olefins, di-olefins and sulfur
The C, heart cut from fractionation is hydrogenated in a two-stage system
o
to saturate olefins, di-olefins and remove sulfur prior to the extraction step.
STEP 3 ; Benzene removal by Sulfolane extraction
The hydrogenated C£ heart cut is extracted to obtain a chemical grade
o
benzene product. The process is the same as for extraction of benzene from the
C, cut on reformates.
o
The fractionation and extraction steps of Figure 4.2 are similar to the steps
for benzene removal from refinery reformates. Although FCC gasoline fractionation
facilities exist in many refineries today, these facilities are not designed
for the functions of Figure 4.2, and are assumed to be available for further
4-8
-------
TABLE 4.2
ILLUSTRATIVE FCC GASOLINE PROPERTIES
(1) Full Light
General Properties Range Gasoline Gasoline
Aromatics (Vol. %) 26.0 21.0
Olefins (Vol. %) 30.0 35.5
Sulfur, ppm 600 200
Research Octane, Clear 91.5 93.8
Research Octane +3g. TEL 99.0 99.9
(2)
Light Gasoline Hydrocarbon Analysis, Wt.%
Isobutane 0.01
Isobutene 0.02
Normal butene 0.06
Butane (trans) 0.09
Butane (cis) 0.14
3 Methyl 1 Butene 0.24
Isopentane 5.47
Butadiene .01
1 Pentene 1.36
2 Methyl 1 Butene 1.73
Normal Pentane 1.73
Isoprene 0.12
2 Pentene (trans) 2.88
Pentadiene .05
2 Pentene (cis) 1.61
2 Methyl 2 Butene 4.23
1, 3 Pentadiene (trans) 0.24
Cyclopentadiene 0.09
2, 2 Dimethyl Butane 0.03
1, 3 Pentadiene (cis) 0.08
Cyclopentene 1.27
Hexenes 21.19
Cyclopentane 0.60
2, 3 Dimethyl Butane 1.52
Hexane 0.43
2 Methyl Pentane 7.41
Benzene 2.38
3 Methyl Pentane 4.59
Cyclohexane 1.09
Normal Hexane 2.99
Cyclohexene 0.44
Methycyclopentane 6.12
2, 4 Dimethyl Pentane 1.13
2.2, 3 Trimethyl Butane 0.25
Heptenes 8.79
3, 3 Dimethyl Pentane 0.10
2 Methyl Hexane 2.78
Dimethyl Cyclopentanes 2.57
4-9
-------
TABLE 4.2 (Cont.)
Light Gasoline Hydrocarbon Analysis, Wt.
(2)
Light
Gasoline
Normal Heptane
Methyl Cyclohexane
Toluene
C 's
A^'s
C°+
07
48
68
4.91
1.90
2.12
SOURCES: (^
(2)
Hydrocarbon Processing, 55, No. 9, September (1976)
Personal Communication from Leo Hollein, Exxon Co.,
to J.R. Felten, November 16, 1977
4-10
-------
FCC Gasoline
30,000 B/SD
Benzene
240 B/SD
V
24,000 B/S5
Deisohexanizer (New)
Hydrogen
Recycle
Hydrogen
Recycle
Compressor
C, Fractionator (New)
b
Figure 4.2
FLOW DIAGRAM FOR
REMOVAL OF BENZENE FROM FCC GASOLINE
Overhead _
6000 B/SD
Hydrogen '
2475 Mscf/d *~
Cg Heart Cut
4500 B/SD
Benzene
998 R/
-------
separations of the C + gasoline stream of Figure 4.2. As discussed in Section 4.1
95% benzene recovery was assumed in the fractionation columns. This required
15% of the FCC gasoline in the C, heart cut. Extraction efficiency was set at
Because of the olefin, di-olefin and sulfur content of this C heart cut,
o
two-stage hydrogenation was also included in the processing route of Figure 4.2.
This provided an extraction plant feedstock whose composition was well within
the range of commercial extraction experience. However, the hydrogenation step
also substantially degrades the octane rating of the raffinate, leading to sub-
stantial costs of pool octane recovery. Substantial incentive exists to define
the conditions under which direct extraction of the heart cut is possible,
thereby reducing the associated octane degradation, hydrogenation plant invest-
ment, and hydrogen generation requirements. Such evaluations have not been made
in this study.
To indicate the octane impact of hydrogenation, the effect on gasoline pool
octane of the selected processing routes for reformates and FCC gasoline is shown
in Table 4.3. As shown there, a large fraction of the total octane debit is from
hydrogenation.
The results of this analysis agree with industry data, which indicate a pool
loss of 0.2 octane for benzene removal from reformate and 0.8 octane for benzene
( 8)
removal from reformate and FCC gasoline. The details of the octane loss calcu-
lations are shown in Chapter 6. Approximate cost estimates of octane replacement
are also presented in Chapter 6.
Processing routes investigated and rejected for benzene removal from FCC
gasoline are as follows:
Aromatics Extraction of Total FCC Gasoline
Including the hydrogenation step as outlined for the preferred route,
extraction of the full range FCC gasoline would be much more costly than extraction
of a Cf heart cut because of the increased volume extracted and the octane
D
degradation of the total FCC gasoline.
4-12
-------
TABLE 4.3
POOL OCTANE LOSS DUE TO BENZENE REMOVAL
FROM REFORMATES & FCC GASOLINE
Refinery Reformates (Extraction)
U. S. Pool Octane Loss
RON MON R+M/2
0.13 0.06 0.10
FCC Gasoline (Hydrogenation)
FCC Gasoline (Extraction)
FCC Gasoline - Total
1.12
0.04
1.16
0.48
0.02
0.50
0.80
0.03
0.83
Refinery Reformates & FCC
Gasoline - Total
. 1.29
0.56
0.93
4-13
-------
Split Light & Heavy FCC Gasoline followed by
Extraction of Light Gasoline
This route has been eliminated because of the same reasons as outlined
above for full-range FCC gasoline. In addition, current fractionation for a
light and heavy FCC gasoline cut for ^gasoline blending at many locations is
inefficient, resulting in considerable benzene in the heavy FCC gasoline.
Fractionation of C, Heart Cut and Direct Extraction
o
As indicated above, this option is highly appealing, but lacks adequate
commercial demonstration to be used as a primary route. It warrants further
investigation, because of its economic attributes as discussed in Chapter 5.
Further studies should include evaluations of the volume of olefin-aromatics
mixture which would be extracted and of the economic disposition of an extract
stream containing 90% benzene and 10% olefinic and sulfur-bearing compounds.
Sulfur Treating of FCC Gasoline & Extraction
of Aroma tics
(9)
Mild hydrogenation, such as Amoco's selective Ultrafining , could remove
sulfur compounds but would not remove olefinic compounds from an FCC gasoline
heart cut. This process, if substituted for the two-stage deep hydrogenation
unit of Figure 4.2, may provide a useful compromise between the selected proces-
sing route and the above direct extraction route, should the latter prove unfeasible.
This compromise would be expected to offer intermediate investments and octane
losses. However, neither the Amoco process nor the extraction of its effluent
was judged to have experienced adequate commercial demonstration to be selected
as the primary processing option of this study.
Fractionation of a C£ Heart Cut followed by Deep
Hydrogenation to Remove Benzene
This approach is unattractive relative to the chosen route, because of
increased octane loss and inadequate commercial demonstration (as discussed for
the deep hydrogenation route of a heart-cut from refinery reformates).
4-14
-------
4.3 Other Technological Alternatives
In addition to the processing routes discussed in Section 4.1 and 4.2,
several other techniques exist that may aid in the removal of benzene from the
gasoline pool. These processing options have not been identified above either
because they were new technologies that are far from commercial fruition, or
because they appeared to offer advantages over the processing option selected
only in special circumstances. These options may, in the future, provide an
economic means for removing benzene from gasoline.
Both the processing route for reformate and the route for FCC gasoline
involve extraction of a C, heart cut stream. To improve economies of scale,
o
these C cuts could be combined prior to the extraction step if both streams
0
were to be extracted. This would be especially important in small refineries,
where the C, cuts are small and involve extraction facilities of less than normal
D
commercial size. In some cases, it may be economical to consider combining the
C, streams from several small refineries for extraction at one common location.
o
A possible alternative to Sulfolane extraction for benzene removal from
the C, heart cut of either reformate of FCC gasoline is extractive distillation.
D
As extractive distillation requires high aromatic contents in the range of 80%
to be economically competitive, Sulfolane extraction is the preferred route below
50% aromatics. As the benzene content of the C, cut from reformate would be
b
only 20% benzene, and the C, cut from FCC gasoline would be about 5% benzene,
aromatics recycle may be useful in special circumstances in order to use extractive
distillation. A second requirement for extractive distillation is the presence
only of trace amounts of other aromatics. In the processing routes selected,
sufficient toluene was present that a benzene tower was included in the design
to remove about 1% toluene contained in the C- heart cut; this minimized pool
o
octane losses. The toluene could interfere with the extractive distillation
process.
The concept of increasing the naphtha initial boiling point to bypass
benzene precursors around the reformer directly into gasoline was discussed in
Chapter 3 and merits further evaluation. Although this may be a viable option
for some small refiners, insufficient crude quality data are available to
adequately assess the general utility of the approach. To properly assess this
4-15
-------
option, it would be necessary to obtain detailed benzene precursor data on all
major crudes processed in the United States.
Certain xylenes are currently recovered by a crystallization process. It
is possible that a similar crystallization process could be developed to remove
benzene from gasoline. For example, aromatic hydrocarbons may be removable by
absorption processes, using molecular sieves.
A possible option for replacing pool octane lost by benzene removal is
alkylation of benzene with propylene to form cumene. Another possible processing
option for replacing pool octane is the alkylation of benzene with ethylene to
form ethyl-benzene.
Commercial processes exist, such as the Pyrotol process, for complete hydro-
cracking of all non-aromatics from a heart cut to produce benzene along with a
light hydrocarbon by-product stream. This process was not considered in our
development because of the decrease in gasoline volume associated with hydrocracking
the olefins and saturates to light hydrocarbons and the increased benzene formed
through hydrodealkylation reactions. In addition, although no detailed investiga-
tion has been undertaken, the Pyrotol process is expected to require high hydrogen
levels and expensive technology relative to the processing system chosen for
this study.
A final option for benzene removal is burning the Cf benzene heart cut. It
b
may be possible to burn a C- cut containing only 5% to 20% benzene without encounter-
b
ing the combustion problems associated with burning pure benzene. Although
this option has the disadvantage of consuming a large volume of gasoline at fuel
value, it would have some merit if benzene alternate disposal values approached
fuel value.
4.4 Processing Routes for Other Gasoline Streams
Light Straight Run Gasoline
As noted in Section 3.4, a significant contributor to 1981 pool benzene
content, after reformates and FCC gasoline, is light straight run gasoline. This
stream could be fractionated to form a C, cut, hydrotreated to remove sulfur,
b
and extracted to remove benzene. Although economic analyses of the costs of
4-16
-------
removing benzene from light straight run gasoline are beyond the scope of this
study, general observations on probable economics are contained in Chapter 5.
Coker Gasoline
The coker gasoline would require similar processing to FCC gasoline. Forma-
tion of a C, heart cut would be followed by moderate hydrogenation to remove
b
olefins and di-olefins, and finally, extraction to remove benzene.
Pyrolysis Gasoline
For the purposes of this study, all pyrolysis gasoline was assumed to be
extracted in 1981 because of the economics of BTX recovery for petrochemical
sales. However, if benzene were to be removed from pyrolysis gasoline, the proces-
sing route would require an initial mild hydrogenation step to remove di-olefins
and eliminate gum formation problems prior to further processing. This would be
followed by a second moderate hydrogenation step to remove olefins and di-olefins,
and then by extraction to remove aromatics.
Natural Gasoline
The processing sequence for natural gasoline would be similar to light
straight run gasoline. The natural gasoline stream would be fractionated to form
a C, cut, mildly hydrogenated, and extracted.
Light Hydrocrackate
As the light hydrocrackate has already been hydrogenated, no further hydro-
genation would be required, and the processing route would become similar to the
route for refinery reformates. The light hydrocrackate would be fractionated to
form a C, cut and extracted to remove benzene.
b
Isomerate
The small benzene contribution to the U.S. gasoline pool from C, isomerate
could be removed by a processing route similar to that proposed for light straight
run.
4-17
-------
Alkylate, Raffinate and Butanes
Alkylate and butanes contain no benzene, whereas raffinate is the by-product
of an extraction process. The small amounts of aromatics contained in raffinate
are the result of design extraction recoveries of about 99.5% aromatics. No
further processing is recommended for these three streams.
4-18
-------
CHAPTER 5
ECONOMICS OF BENZENE REMOVAL FROM
REFORMATES AND FCC GASOLINE
The primary purpose of this study is to determine the costs of
the removal of benzene from refinery reformates and FCC gasoline in the U. S.
The key elements in this development were to establish a basis for economics
and to scale up the base case economics to determine the national impact.
The economics, developed using the processing routes selected in Chapter 4,
were first developed for a 1977 base case for reformates and FCC gasoline.
Investment costs were based on 1977 U. S. Gulf Coast costs and included factors
to account for offsites, interest during construction, startup, royalty and
working capital. Process manufacturing costs were based on Arthur D. Little and
industry estimates of process requirements and 1977 Gulf Coast prices. Manufac-
turing costs included variable operating costs, labor, maintenance and capital
charges.
Because a detailed refinery hydrogen balance was beyond the scope of this
study, economics for removal of benzene from FCC gasoline were based on new hydrogen
plant hydrogen at each location. This assumption results in a maximum cost for
hydrogen in the base case economics. The sensitivity to hydrogen price was
considered by also developing economics with refinery hydrogen at fuel value.
The main variable affecting the economics of benzene removal from reformates
and FCC gasoline was the total volume to fractionation, hydrogenation and extrac-
tion. The projected 1981 volumes of reformates and FCC gasoline requiring
extraction were developed on a capacity and regional basis in Chapter 2. The
basic scale-up methodology used was to scale up the base case economics on a
volume basis and to add processing capability to remove benzene from reformates
and FCC gasoline at each location.
5-1
-------
The economics were developed separately for reformates and FCC gasoline.
Some economies of scale would be possible by combining reformate and treated
FCC gasoline heart cut streams prior to extraction. However, this would require
detailed process information on an individual refinery basis, which is not
generally available, and an individual refinery scaleup, which was beyond the
scope of this project.
The result of the scaled-up base case economics was the national cost of
benzene removal from reformates and FCC gasoline. Removal of benzene from
reformates would require an investment of $2.0 billion with annual manufacturing
costs of $0.9 billion per year, or 0.8 cents per gallon of gasoline produced.
Removal of benzene from both reformates and FCC gasoline, excluding H~S recovery,
would require an investment of $5.3 billion and annual manufacturing costs,
including capital recovery, of $2.5 billion per year, or 2.2 cents per gallon of
gasoline produced.
Because of differences in refinery size, the impact of benzene removal is
more severe in some regions than in others. This is particularly true in
PADD TV, where the costs for benzene removal are about 50% higher than the
national average.
As the requirement for a hydrogen plant at each location is a worst-case
analysis, the national economic impact of benzene removal with hydrogen at
fuel value was also developed. Using internally-produced refinery hydrogen at
fuel value would reduce investment by $0.5 billion and decrease annual manufactur-
ing costs by $0.2 billion per year, or 0.2 cents per gallon of gasoline produced.
A possibility exists that the hydrogenation step would not be required in
benzene removal from FCC gasoline; therefore, we determined the national impact
of benzene removal without FCC gasoline hydrogenation. The elimination of
the hydrogenation step would reduce investment by $1.5 billion and decrease
annual manufacturing costs by $0.7 billion per year, or 0.6 cents per gallon of
gasoline produced.
The cost of benzene removal in the small refinery was estimated, because
the most important variable affecting the costs of benzene removal is the
volume to extraction. Thus the cost of benzene removal to the small refinery
5-2
-------
will be greater than the national average, both from economic and operational
flexibility standpoints. In the case of a 10,000 B/D refinery, the total
manufacturing costs of benzene removal from reformates and FCC gasoline would
be about 6.9 cents per gallon of gasoline produced, or three times the national
average. In addition, it is likely that many small refiners would be unable
to meet projected lead and MMT phasedowns with benzene removed from their
gasoline pool. The severe economic impact, coupled with the loss of operational
flexibility, could cause small refiners to withdraw from the gasoline market.
5.1 Basis for Economics
A. Process Route Description
The selected processing routes were discussed in detail in Chapter 4. A
simplified flow diagram for the base case for removal of benzene from reformates
is shown in Figure 5.1. The full range catalytic reformate (13,330 B/SD) is
first deisohexanized to remove isohexane and lighter components (2,000 B/SD).
The deisohexanizer bottoms (11,330 B/SD) is then fractionated to obtain a C,
cut (2,000 B/SD). The C.,+ bottoms from the C, fractionator (9,330 B/SD) are
/ o
returned to gasoline blending, and the C, heart cut is sent to sulfolane extrac-
D
tion, where 378 B/SD of benzene is recovered and 1,622 B/SD raffinate is
released to gasoline blending or petrochemical feed. The Sulfolane extraction
process uses a clay tower and benzene fractionation tower to remove the small
amount of toluene (approximately 1%) in the C, heart cut to make chemical grade
o
benzene.
A simplified flow diagram for the base case for removal of benzene from
FCC gasoline is shown in Figure 5.2. The full-range FCC gasoline (30,000 B/SD)
is first deisohexanized to remove isohexane and lighter components (6,000 B/SD).
The deisohexanizer bottoms (24,000 B/SD) is then fractionated to obtain a
C- cut (4,500 B/SD). The C..+ bottoms from the C, fractionator (19,500 B/SD),
o / o
is returned to gasoline blending, and the C, heart cut is sent to the hydrogena-
tion section.
The C, heart cut is hydrogenated at 650 psig pressure in a two-stage hydro-
b
genatlon with interstage cooling and hydrogen recycle. Total hydrogen usage
is 550 SCF/barrel of C, feed.
b
5-3
-------
Catalytic
Reformats
13,330 B/SD
Benzene
AGO B/SD
11,330 B/SD
Deisohexanizer (New)
C& Heart Cut
2000 B/SD
Benzene
380 B/SD
Sulfolane
Extraction
Unit (New)
C, Fractionator (New)
o
Figure 5.1
FLOW DIAGRAM FOR
BENZENE REMOVAL FROM CATALYTIC REFORMATS
Overhead
2000 B/SD
Benzene
378 B/SD
Raffinate
1622 B/SD
Benzene
2 B/SD
C7+ Gasoline
9530 B/SD
Benzene
20 B/SD
-------
FCC Gasoline
30,000 B/SD
Benzene
240 B/SD
i
Ln
24,000 B/SD
Deisohexanizer (New)
Hydrogen
Recycle
Hydrogen
Recycle
Compressor
C. Fractionator (New)
O
Figure 5.2
FLOW DIAGRAM FOR
REMOVAL OF BENZENE FROM FCC GASOLINE
Overhead
6000 B/SD
Hydrogen
2475 Mscf/d "
05 Heart Cut
4500 B/SD
Benzene
798 R/sn
Two Stage
Hydrogenation
(New)
&
Sulfolane
Extraction
(New)
Benzene ^_
227 B/SD
Raffinate
4273- B/SD
Benzene
1 B/SD
C-+ Gasoline
l4,500 B/SD
Benzene
12 B/SD
-------
The hydrogenated C cut (4,500 B/D) is sent to the Sulfolane extraction
o
plant, where 227 B/SD of benzene is recovered and 4,273 B/SD of raffinate is
released to gasoline blending or petrochemical feed.
B. Investment Costs
Investment costs are based on 1977 U.S. Gulf Coast costs for onsite battery
limits investment. The onsites, or process investment, is the equipment required
to carry out the chemical reactions and physical separations for benzene removal
from gasoline. In addition to the process investment, offsites investment is
required to provide supporting facilities for the operation of the battery limits
plant. Offsites investment includes such items as sufficient storage facilities
for raw materials and final products; supplying required utilities such as
steam, cooling water, electrical power, instrument air and inert gas; and addi-
tions to the infrastructure for a processing plant, such as maintenance facilities,
warehouses, administration buildings, laboratory facilities and waste disposal
facilities.
The process investments for this study are based on data developed by
Arthur D. Little in previous studies, data available in the literature, and data
received through extensive discussions with engineering contractors, process
licensors and refinery processors.
The offsites for this study are based on 40% of onsite process investment
and include capital requirements for all utilities production except power gener-
ation, which is on a purchased basis. The offsites factor is also derived from
previous Arthur D. Little studies and discussions within the industry.
In addition to total fixed plant investment (the sum of process and offsites
investment) the investor will incur additional costs for interest during construc-
tion, startup costs, working capital and royalty payments.
The construction of facilities to remove benzene from gasoline would normally
take about three years to complete from initial engineering to startup. Interest
during construction is based on a normal rate of expenditure for such a project
20% of totajL plant investment in the first year, and 40% of investment in the
second and third years with a 10% interest charge.
5-6
-------
An investment charge of 5% of total plant investment was used to cover
startup costs, for hiring and training process operators, special startup crews
brought in for initial operation, inefficient use of utilities and off-spec
product during startup, and wear and tear and damage to equipment and facilities
that can not be attributed to normal operation.
Working capital is based on five days' supply of feedstock and benzene
product inventories necessary to allow for typical interruptions in normal
operation. All unfinished feedstock was valued at $15.90/B, the 1977 disposal
value as unleaded gasoline. In addition to unfinished feedstock, working capital
included the initial Sulfolane charge required for the Sulfolane extraction
process.
Royalty payments were required for the Shell Oil Sulfolane process.
Royalty payments were based on paid-up royalty charges for aromatlcs recovery
as follows:
Aromatics Recovery Paid-Up Royalty
Million Gals/Year $/Gallon/Yr Aromatics
1-15 0.0125
15 - 30 0.0100
30 - 200 0.0075
Over 200 0.0025
Benzene recovery from all FCC gasoline streams and 156 of 167 total reformate
streams falls into the 1 to 15 million gallons per year category. The 11 largest
reformate streams fall into the 15 to 30 million gallons per year category.
The summation of total plant investment, interest during construction,
startup costs, working capital and royalty payments is the total project invest-
ment. Because of the allowances for interest during construction, startup costs,
etc., the total project investment equals the instantaneous capital investment
at startup. Thus capital charges are based on recovery of total project invest-
ment at startup.
The design basis and base case investment costs for removal of benzene from
reformates are shown in Table 5.1.
5-7
-------
TABLE 5.1
REMOVAL OF BENZENE FROM REFORMATS INVESTMENT COSTS
U.S. GULF COAST - 1977
Design Basis:
Reformer Change
80% Reformate Yield
15% C, Cut Yield
D
3% Benzenein Reformate
95% Benzene Recovered in C, Cut
o
99.5% Benzene Extracted
16,666 B/SD
13,333 B/SD
2,000 B/SD
400 B/SD
380 B/SD
378 B/SD
Investment: M$
Process
Offsites @ 40%
Total Plant Investment
Interest During Construction
Startup Costs
Working Capital
Royalty
Total Project Investment
Fractionation Extraction
2,830
1,132
3,962
3,170
1,268
4,438
Total
6,000
2,400
8,400
1,596
420
1,346
77
11,839
5-8
-------
The design basis and base case investment costs for removal of benzene from
FCC gasoline are shown in Table 5.2.
C. Manufacturing Costs
Process manufacturing costs are based on Arthur D. Little and industry
estimates of process requirements and 1977 Gulf Coast prices.
Variable costs include fuel, power, steam, water, and chemical costs and
vary directly with process throughput. These costs are incurred only when the
process is on-stream. Thus variable costs are based on process throughput on a
stream day, rather than a calendar day, basis. Variable operating costs were
converted to an annual basis by assuming 345 stream-days operation per year.
Utility costs are calculated on a variable cost basis for fuel, steam and
cooling water. Necessary generation and distribution facilities are included in
offsites. In the case of electric power, costs were based on purchases,
with the cost of electric generation facilities included in the electricity
cost. The costs of power distribution, however, are included in offsites.
Since a refinery steam balance was beyond the scope of this study, steam
costs were based on producing incremental 600 pslg steam with no low pressure
steam recovery, and should be viewed as maximum steam costs. Maximum waste heat
recovery would reduce energy costs about 10 to 12%.
For the purpose of this study, hydrogen costs are based on new hydrogen
plant hydrogen at all locations. The hydrogen cost includes a capital recovery
cost on the required capital investment for hydrogen manufacture. A detailed dis-
cussion of the implications of hydrogen balance and hydrogen cost basis is found
in Section 5.2C of this chapter and in Appendix C.
Labor, supervision and maintenance costs are included in semi-variable
costs. Semi-variable costs are incurred as a result of manning and start up of
a process unit, but do not vary with throughput. Semi-variable costs are not
incurred if a process unit is permanently shut down. Labor charges are based on
unit manning requirements for a 24-hour daily operation and a 40-hour week salary
supervision.
5-9
-------
TABLE 5.2
REMOVAL OF BENZENE FROM FCC GASOLINE
INVESTMENT COST
U.S. GULF COAST 1977
Design Basis:
FCC Gasoline
15% C. Cut
o
0.8% Benzene In FCC Gasoline
95% Benzene Recovered in C, Cut
0
99.5% Benzene Extracted
30,000 B/SD
4,500 B/SD
240 B/SD
228 B/SD
227 B/SD
Investment; M$
Process
Offsites @ 40%
Fractionation
3,500
1,400
Total Plant Investment 4,900
Interest During Construction
Startup Costs
Working Capital
Royalty
Total Project Investment
Hydro-
genation
5,000
2,000
7,000
Extraction
7,830
Total
14,090
5,640
19,730
3,749
987
2,690
43
27,199
5-10
-------
Labor costs are based on $12.00 per manhour, including benefits, for operating
manpower. Supervisory costs are based on $27,000 per year, including benefits,
for salary supervision. Maintenance charges are based on typical overall
refinery maintenance costs of 4% per year of total plant investment.
Fixed operating costs occur as a result of capital investment and are
incurred regardless of whether the process unit is on-stream, shut down for
routine maintenance, or permanently shut down. Capital charges of 25% of
total investment are used to provide annual cash flow necessary to repay debt
financing, pay income taxes, and earn a return on project equity of about 15%
DCF. Local taxes and insurance are based on 2.5% per year of total investment.
The base case manufacturing costs for removal of benzene from reformates
are shown in Table 5.3; the base case manufacturing costs for removal of benzene
from FCC gasoline are shown in Table 5.4.
Utilities are calculated on a variable cost basis, with the exception of
electric power, which is on a purchased basis. Utility cost calculations are
shown in Appendix C.
Hydrogen costs are based on new hydrogen plant use at each location and
include a return on hydrogen plant investment. Hydrogen cost calculations are
shown in Appendix C.
D. Effect of Variables on Economics
The investment cost for benzene removal from reformates and FCC gasoline
is a function of total gasoline production, volume of C- heart cut, aromatics
content and percent aromatics recovery. Fractionation investment costs are
primarily dependent on total volume of reformate fractionated and, to a lesser
extent, the volume of C. cut.
b
The most important variable affecting Sulfolane extraction investment
costs below 50% aromatics content is the total volume to extraction. This is
because the solvent circulation rate and rotating disk extractor size are more
dependent on the total feed to extraction than the aromatics content. For high
5-11
-------
TABLE 5.3
BASE CASE MANUFACTURING COSTS
FOR REMOVAL OF BENZENE FROM REFORMATS
Fractionation Extraction Total , ,
Units /SD Units /SD Units/ SD $ /Units $/SD M$/Yr/ '
MANUFACTURING
Variable Costs:
Sulfolane: Ibs
Fuel : FOEB
Power : Kwhr
Cooling Water: Mgals
V Steam: Mlbs
M
"^ Sub-Total Variable Costs 4,652 1,605
Semi-Variable Costs:
-
-
1,680
8,120
950
25
-
1,500
400
430
25
-
3,180
8,520
1,380
1.50
12.00
0.025
0.030
3.10
38
-
80
256
4,278
13
-
28
88
1,476
Operating Labor: Man/ Shift 1.5 1.5 2 * 913 315
Man Hr.
Supervision: Foreman/Day 1 27, 000 / ?9 2y
Year
Maintenance: 4% of Plant Investment/Year 974 336
Sub-Total Semi-Variable Costs 1,966 678
Fixed Costs:
Capital Charge: 25% of Capital Investment /Year 8,579 2,960
Taxes & Insurance: 2.5% of Capital Investment/Year 858 296
Sub-Total Fixed Costs . 9,437 3,256
Total Manufacturing Costs 16,055 5,539
$/B Reformate (13,333 B/SD) . 1.20
$/B C Cut ( 2,000 B/SD) 8.03
(1)345 SD/Year
-------
I
(->
u>
TABLE 5.4
BASE CASE MANUFACTURING COSTS
MANUFACTURING COSTS
Variable Costs:
Hydrogen: MSCF
Catalystic Chemicals
Fuel: FOEB
Power: Kwhr
Cooling Water: Mgals
Steam: Mlbs
Sub-Total Variable Costs
Semi-Variable Costs:
Operating Labor: Man/Shift
Supervision: Foreman/Day
Maintenance: 4% of Plant
Sub-Total Semi-Variable Costs
Fixed Costs:
Capital Charge: 25% of Total Investment/Year
Taxes & Insurance: 2.5% of Total Investment/Year
Sub-Total Fixed Costs
TOTAL Manufacturing Costs
FOR REMOVAL
Frac t ionat ion
Units /SD
-
-
865
6,860
1,540
1.5
OF BENZENE FROM
Hydro genat ion
Units /SD
2,475(2)
-
138
18,900
785
0
'l.5
FCC GASOLINE
Extraction Total
Units/ SD Units/ SD
2,475
-
138
3,375 23,140
900 8,545
934 2,474
1.5 4.5
1
$/Unit
3.07
-
1200
0.025
0.03
3.10
12. OO/
Man Hr.
27.000/
Year
$/SD
7,598(2)
285
1,656
579
257
7,689
18,044
1,371
78
2,288
M$/Yr(1)
2,621
98
571
200
89
2,646
6,225
473
27
789
3,737
21,681
43,462
1,289
19 * 710 6,800
1,971 680
7,480
14,994
(1)
(2)
345 SD/Year
Capital recovery on hydrogenation included in hydrogen variable cost
-------
concentration of aromatics, the aromatics content becomes more significant
than for low concentration due to increased size for solvent recovery and
aromatics separation facilities.
Overall investment cost increases with decreasing aromatics content,
increases with increased percent recovery of aromatics, and decreases with
decreasing aromatic carbon number. However, these effects are all secondary
to the volume to extraction variable.
Hydrogenation investment costs are primarily dependent upon total volume
to hydrogenation. Investment costs for the hydrogenation step depend, also to
a lesser extent, on the olefin-di-olefin content of the FCC gasoline.
Variable operating costs are also a function of total unit throughput,
aromatics content and percent aromatics recovery. For low aromatics concentra-
tions, variable operating costs depend most on total volume throughput.
An important variable in determining benzene removal costs is the percent
removal of benzene desired. The percent benzene removal dictates the width of
the C, cut required, and thus the volume of C cut to extraction or hydrogena-
D D
tion and extraction. As we discussed above, the volume processed is the
most critical variable affecting economics. For the purposes of this study,
we have selected 95% benzene removal in the primary fractionation step as a
reasonable target removal. To obtain 95% removal, it is necessary to make a
true boiling point (TBP) heart cut from about 160°F to 200°F. This will
typically result in about a 15 volume percent C, cut from reformates and FCC
gasoline and contain about 1 volume percent toluene. Higher than 95% benzene
removal could be obtained by fractionating for a wider TBP cut, but this change
would greatly increase investment costs.
The percent benzene recovery in extraction is also an important variable
for benzene recoveries above 99.5%. A target level of 95% benzene removal was
used for the fractionation step; therefore, increased costs to obtain a high
benzene recovery in extraction are not warranted, and a 99.5% recovery was assumed.
5-14
-------
E. Scale-Up of Economics
The volume of reformates and FCC gasoline to fractionation, hydrogenation
and extraction is the most critical variable affecting benzene removal economics;
the other variables were assumed to have minimal impact on the economics.
Thus the base case economics were scaled up only on the basis of volume.
The basic scale-up methodology for this study was to add processing capa-
bility to remove benzene from reformates and FCC gasoline at each location.
The projected 1981 volume of reformates and FCC gasoline requiring extraction
was developed by capacity range in Chapter 2 and was shown in Tables 2.8 and 2.9.
Investment costs were scaled up from the base case costs to the regional
*f
and U.S. costs by capacity distribution using the equation I = A (C) ; where
I = Total Plant Investment: M$
A = A Constant Calculated from the Base Case
C = Capacity: B/SD
x = Exponential Investment Factor
The investment cost scale-up factors for reformate fractionation invest-
ments are shown in Table 5.5.
TABLE 5.5
REFORMATS FRACTIONATION INVESTMENT
Total Plant
Investment
$/B/SD
1,366
709
449
297
238
189
*Base case
The investment cost scale-up factors for FCC gasoline fractionation and
hydrogenation investment are shown in Tables 5.6 and 5.7.
5-15
Reformate
Capacity (B/SD)
750
2,750
6,000
13,333*
28,000
60,000
Constant
A
5.134
5.134
5.134
5.134
5.134
5.134
Exponent
X
0.80
0.75
0.72
0.70
0.70
0.70
-------
FCC Gasoline
Capacity (B/SD)
1,400
4,300
8,500
16,900
30,000*
45,200
*Base case
FCC GASOLINE
Constant
A
3.599
3.599
3.599
3.599
3.599
3.599
TABLE 5.6
FRACTIONATION INVESTMENT
Exponent
X
0.80
0.75
0.72
0.70
0.70
0.70
Total Plant
Investment
$/B/SD
845
444
286
194
163
144
Hydrogenation
Capacity (B/SD)
210
645
1,275
2,535
4,500*
6,780
*Base case
TABLE 5.7
FCC GASOLINE HYDROGENATION
Constant
A
12.741
12.741
12.741
12.741
12.741
12.741
INVESTMENT
Exponent
X
0.80
0.80
0.75
0.75
0.75
0.72
Total Plant
Inves tment
$/B/SD
4,373
3,949
2,132
1,796
1,556
1,078
The investment scale-up factors for extraction investment for both reformates
and FCC gasoline are shown in Table 5.8.
5-16
-------
TABLE 5.8
REFORMATS & FCC GASOLINE
EXTRACTION INVESTMENT
Total Plant
Extraction Constant Exponent Investment
Capacity (B/SD) A x $/B/SD
112.5 21.700 0.80 8,438
412.5 21.700 0.75 4,815
900 21.700 0.72 3,231
2,000* 21.700 0.70 2,219
4,200 21.700 0.70 1,776
9,000 21.700 0.70 1,413
*Reformate base case
Variable operating costs were scaled up linearly with fractionation, hydro-
genation and extraction capacity according to the volume of reformate and FCC
gasoline requiring extraction, as developed in Chapter 2 and shown in Tables 2.8
and 2.9.
Labor and supervision costs were assumed constant regardless of unit capacity.
The base case costs were scaled up to a regional U. S. basis from the number
of units in each size category.
Capital-related costs were scaled up on a regional and national basis by
applying the appropriate percentages to the scaled investment costs.
The base case economics were scaled up on a regional basis by capacity to
determine the national cost of benzene removal. The regional economics of
benzene removal from reformates and FCC gasoline are shown in Appendix 5.3.
5.2 National Cost of Benzene Removal from Reformate & FCC Gasoline
A. Total U. S. Cost
Using calculations based on the scale-up procedure of the previous section,
we added benzene removal processing capacity to all projected 1981 unextracted
reformate and FCC gasoline capacity. The result of this scaleup is the national
economic impact of removing benzene from reformates and FCC gasoline as shown
in Table 5.9. 5-17
-------
TABLE 5.9
NATIONAL COST OF BENZENE REMOVAL
FROM REFORMATES & FCC GASOLINE
Investment Costs: Billion $
Process
Offsites
Total Plant -
Other Capital
Total Capital -
Manufacturing Costs: (M$/SD (345 SD/Yr)
Variable Costs
Labor & Maintenance Costs
Capital Related Costs
Total Costs: (M$/SD)
Total Costs: (MM$/Yr)
Total Costs: (c/Gal)(3)
(4)
Energy Costs: (Fuel @ $12.00/FOEB)
COE: MB/Yr
MM$/Yr
Reformates
1.009
0.404
1.413
0.584
1.997
801
329
1,592
2,722
939
0.82
21,930
263
FCC
Gasolinev '
1.446
0.579
2.025
0.744
2.769
1,090
433
2,207
3,730
1,287
1.12
26,573
319
Hydrogen
0.300
0.120
0.420
0.101
0.521
796(2)
0
0
796
275
0.25
5,513
66
Total
2.755
1.103
3.858
1.429
5.287
2,687
762
3,799
7,248
2,501
2.19
54,016
648
(1)
(2)
(3)
(4)
Excluding hydrogen costs
Includes hydrogen plant capital recovery costs
Based on 7,450 BD gasoline
Included in variable manufacturing costs
5-18
-------
The removal of benzene from all refinery reformates would result in a total
capital cost of $1.997 billion, and an annual cost of $939 million per year.
We can translate this annual cost to a gasoline basis by dividing the annual
operating costs by the projected annual gasoline production in 1981 of 7.45
million barrels per calendar day, or 114.2 billion gallons per year. On this
basis, the cost of removing benzene from reformates is 0.82 cents per gallon
of gasoline produced.
Removal of benzene from refinery reformates is an energy intensive process.
With fuel at $12.00 per FOE barrel, energy accounts for $263 million per year, or about
28% of the total manufacturing cost of removing benzene from reformates. The
energy cost amounts to 0.23 cents per gallon of gasoline produced. Details of
the energy cost calculations are shown in Appendix C.
Estimations of the national impact of removing benzene from FCC gasoline,
based on hydrogen plant hydrogen, are also shown in Table 5.9. The investment
and operating costs required for hydrogen production have been shown separately
to isolate the effect of hydrogen costs.
The removal of benzene from all FCC gasoline would result in total capital
costs of $2.769 billion for FCC gasoline processing investment, plus $0.521
billion for hydrogen plant investment for a total investment cost of $3.29 billion.
Annual manufacturing costs would be $1.562 billion for removal from FCC gasoline,
including hydrogen costs. Assuming an annual gasoline production of 7.45 million
barrels per calendar day, this equates to 1.37 cents per gallon of gasoline
produced.
With fuel at $12.00 per FOE barrel, energy accounts for $385 million per year,
or 25% of the total cost of removing benzene from FCC gasoline using hydrogen
plant hydrogen.
The overall cost of benzene removal from refinery reformates and FCC
gasoline using hydrogen plant hydrogen would be a total capital investment cost
df $5.287 billion and an operating cost of $2.501 billion per year. These costs
translate into an overall cost of 2.19 cents per gallon of total gasoline produced.
The overall energy costs of $648 million per year amount to 0.57 cents per gallon
of gasoline production.
5-19
-------
The largest component of the cost of removing benzene from reformates and
FCC gasoline is capital-related. Capital-related costs (excluding hydrogen
plant capital recovery costs) are $1.311 billion per year, or 1.15 cents per gallon
of gasoline produced. If hydrogen plant capital recovery is included, total
capital-related costs increase to $1.453 billion per year, or 1.27 cents per
gallon of gasoline produced. Thus, total capital-related costs account for about
58% of the total benzene removal cost.
B. Regional Differences
Significant regional differences exist in the industry that will cause
the cost of benzene removal to be higher in some regions than average across
the nation. A regional impact summary of benzene removal is shown in Table
5.10.
The high cost for benzene removal in PADD IV is apparent in Table 5.10.
Total costs per barrel of reformate, per barrel of FCC gasoline, and per barrel
of total gasoline produced are all higher than for any other PAD District.
These increased costs result from the smaller average unit sizes in PADD IV.
The costs for removal of benzene from reformates in cents per gallon of
gasoline produced are far lower in PADD III than any other region. The difference
results from the large amount of current benzene extraction in PADD III and the
smaller increase in new capacity required by 1981.
The costs in cents per barrel of gasoline in PADD V are the highest for
reformates (with the exception of PADD IV), and the lowest for FCC gasoline. This
is a result of the higher percentage of reformate and the lower percentage of
FCC gasoline in the gasoline pool for PADD V relative to the other PAD Districts.
A more detailed breakdown of the total national costs for benzene removal from
reformates and FCC gasoline by PAD District is shown in Appendix C.
C. Sensitivity to Refinery Hydrogen Costs
Hydrogen availability and the cost of refinery hydrogen is a function
of refinery processing configuration, crude type and product specifications.
5-20
-------
TABLE 5.10
SUMMARY OF REGIONAL RESULTS
Reformate PADD I PADD II PADD III PADD IV PADD V TOTAL
Investment: MM$ 214 577 699 116 391 1,997
Total Cost: $/B Refor-
mate 1.13 1.13 1.24 1.73 1.11 1.18
(21
-------
As a detailed analysis of individual refinery hydrogen balance is beyond the
scope of this study, we assumed for our base case economics that a hydrogen
plant would be required at each location. Because the requirement for a hydrogen
plant at each location is a worst-case analysis, we have also developed economics
for the best case in which hydrogen is normally available at fuel value in Table 5.11.
TABLE 5.11
HYDROGEN COSTS
Hydrogen Plant Hydrogen as Fuel
Average U.S. Cost: $/MCF 3.40 0.65^
Range of U.S. Cost: $/MCF 2.70-12.04
Total U.S. H2 Cost: $/SD 795,534 152,076
Million $/Yr 274 52
Total U.S. Cost: C/Gal Gasoline 0.25 0.05
at $12.00/FOEB
Using the above analysis, we can value all hydrogen at fuel value rather
than hydrogen plant value, and thus reduce overall costs of removing benzene from
reformates and FCC gasoline from 2.19 cents to 1.99 cents per gallon. The
effect of valuing all hydrogen at fuel value on the national impact of benzene
removal is shown in Table 5.12.
D. Sensitivity to FCC Hydrogenation Step
The selected processing route for FCC gasoline includes a hydrogenation
step to remove olefins and sulfur prior to extraction. Although it has not
been commercially proven, some sources indicate that Sulfolane extraction of an
olefin/aromatic mixture may be possible. This route would have the advantage of
reduced pool octane loss and savings in hydrogenation costs. If the hydrogena-
tion step were eliminated, overall investment costs and total processing costs
would be reduced considerably, as shown in Table 5.13.
5-22
-------
TABLE 5.12
NATIONAL COST OF BENZENE REMOVAL FROM
REFORMATES & FCC GASOLINE
(Hydrogen
Investment Costs: Billion $
Process
Offsites
Total Plant -
Other Capital
Total Capital -
Manufacturing Costs: M$/SD (345 SD/Yr)
Variable Costs
Labor & Maintenance Costs
Capital Related Costs
Total Costs: (M$/SD)
Total Costs: (MM$/Yr)
Total Costs: (o/Gal)
Energy Costs: (Fuel @ $12.00/FOEB)
FOE: MB/Yr
MM$/Yr
at Fuel Value)
FCC
Reformates Gasoline
1.009
0.404
1.413
0.584
1.997
801
329
1,597
2,722
939
0.82
21,930
263
1.446
0.579
2.025
0 . 744
2.769
1,090
433
2,207
3,730
1,287
1.12
26,573
319
Hydrogen
0
0
0
0
0
152
0
0
152
52
0.05
4,372
52
Total
2.455
0.983
3.438
1.328
4.766
2,043
762
3,799
6,604
2,278
1.99
52,875
634
(1)
(2)
Excluding hydrogen costs
Based on 7,450 B/D gasoline
5-23
-------
TABLE 5.13
EFFECT OF HYDROGENATION STEP ON FCC GASOLINE COSTS
Investment:
Total Costs:
MM$
MM$/Yr
C/Gal Gasoline
$/B FCCs
Fractionation
& Extraction
1,786
880
0.77
0.90
Hydrogenation
1,504
681
0.60
0.70
Total
Costs
3,290
1,561
1.37
1.60
Eliminating the hydrogenation step, the costs for removal of benzene from
FCC gasoline become slightly lower than the costs of benzene removal from
reformates. The total national impact of benzene removal from reformates and
FCC gasoline without the hydrogenation step is shown in Table 5.14.
Investment:
Total Costs:
TABLE 5.14
NATIONAL COST OF BENZENE REMOVAL FROM
REFORMATES & FCC GASOLINE
(Without Hydrogenation of FCC Gasoline)
MM$ 1,997
MM$/Yr, 939
C/Gal Gasoline 0.82
FCC
Gasoline
1,786
880
0.77
Total
Costs
3,783
1,819
1.59
In addition to the benzene removal costs, extraction of an unhydrogenated
FCC gasoline stream would result in an olefins and aromatics mixture that would
present a disposal problem.
5-24
-------
5.3 Impact on the Small Refiner
The most important variable affecting the costs of benzene removal is the
volume to extraction. Thus, removal of benzene from reformates and FCC gasoline
will have more sever economic and operational impact on the small independent
refiner than the large major refiner.
The economics of benzene removal from reformates and FCC gasoline were
developed as a function of unit size in Section 5.1 of this chapter. The total
operating costs in $/SD of reformates and FCC gasoline have been plotted against
reformate and FCC gasoline production capacity in Figure 5.3. As illustrated
by the chart, operating costs in $/B/SD increase dramatically for the small
refiner.
It was shown in Section 5.2 that the average national cost of benzene removal
from reformates and FCC gasoline was 2.19 cents per gallon of gasoline. To estimate the
effect of benzene removal with size for total gasoline production, we have assumed
that the gasoline blend at each location was the same as the national average.
Using the average U.S. blend of 30% reformate and 34.5% FCC gasoline (this assumes
both reformate and FCC capacity at each location), we have shown the total operat-
ing costs in cents per gallon of total gasoline as a function of gasoline capacity
in Figure 5.4 (with hydrogen plant hydrogen) and Figure 5.5 (with hydrogen as
fuel). From Figure 5.4, the cost in cents per gallon of gasoline for a 10,000 B/D
refinery producing 5,000 B/D gasoline with hydrogen plant hydrogen would be 6.9
cents per gallon of gasoline, or more than three times the average U.S. refinery
costs.
To illustrate, we have determined the costs for a 10,000 B/SD refinery.
Most refineries of this size would not have an FCC unit, so we have developed
costs for a 10,000 B/SD refinery with reforming both with and without an FCC unit.
The calculations are shown in Appendix C. A summary of the results is shown in
Table 5.15. In the case of reforming only, we have estimated gasoline produc-
tion at 25% of crude charge, with 60% reformate in the gasoline pool. On this
basis, costs of removing benzene from gasoline are 5.28 cents per gallon of
gasoline, or about six times the national average cost of removing benzene from
reformates of 0.82 cents per gallon. Total investment costs are estimated at
$3.872 .million or $1,550/SD gasoline. This compares with average investment costs
of $268/B/SD gasoline for removal from reformates.
5-25
-------
: 4
O
O
2 T
iM «
a>
cc
FCC Gasoline
(Hydrogen Plant Hydrogen)
FCC Gasoline
(Hydrogen as Fuel)
15 20 25 30 35
Reformate/FCC Gasoline Capacity - MB/SD
Source: Arthur D. Little, Calculations. , ,
Figure 5.3 Cost of benzene removal vs. reformate and FCC gasoline capacity.
-------
I
NJ
Reformates
T) FCC Gasoline
-A Reformates and FCC
Gasoline
40 50 60
Gasoline Capacity: MB/SD
Source: Arthur D. Little, Calculations.
Figure 5.4 Cost of benzene removal vs. gasoline production using hydrogen plant hydrogen.
100
-------
Ln
I
N>
00
Reformates
FCC Gasoline
Reformates and FCC
Gasoline
0 10 20 30 40 50 60 70 80
Gasoline Capacity: MB/SD
Source: Arthur D. Little, Calculations.
Figure 5.5 Cost of benzene removal vs. gasoline production using refinery-produced hydrogen.
90
100
-------
TABLE 5.15
COSTS FOR 10,000 B/SD REFINERY vs. U.S. AVERAGE
A. Reforming Capacity Only: 2,500 B/SD Gasoline
Remove Benzene from Reformats
Manufacturing Cost: £/Gal Gasoline
Investment Cost: $ Million
Investment Cost: $/B/SD Gasoline
10,000
B/SD Refinery
5.28
3.872
1,550
U.S. Average
0.82
2,769
268
B. Reforming plus FCC Capacity; 5,000 B/SD Gasoline
Remove Benzene from Reformate
Manufacturing Cost: C/Gal Gasoline
Investment Cost: $ Million
Investment Cost: $/B/SD Gasoline
10,000
B/SD Refinery
2.64
3.872
775
U.S. Average
0.82
2,769
268
Remove Benzene from Reformate &
FCC Gasoline
Manufacturing Cost: c/Gal Gasoline
Investment Cost: $ Million
Investment Cost: $/B/SD Gasoline
10,000
B/SD Refinery
6.83(1)6.09(3)
9.227(2)
1,845
U.S. Average
4,766
640
(1)
(2)
(3)
Includes return on hydrogen plant investment
Excluding return on hydrogen plant investment
Using refinery produced hydrogen at fuel value
5-29
-------
In the case of a 10,000 B/D refinery with both reforming and FCC cracking,
we have estimated gasoline production at 50% of crude charge, with 30% reformate
and 34.5% FCC gasoline in the gasoline pool. On this basis, costs of removing
benzene from reformates only are 2.64 cents per gallon of gasoline, or about
three times the national average. Total investment costs are $3.872 million
or $775/B/SD gasoline, or about three times the average cost of $268/B/SD.
The costs of removing benzene from reformates and FCC gasoline for a
10,000 B/D refinery are estimated at 6.83 cents per gallon of gasoline, or about
three times the average U.S. cost of 2.19 cents per gallon of gasoline. Similarly,
total investment costs are $9.227 million or $1,845/B/SD gasoline, or over three
times the average cost of $640/B/SD gasoline.
In the case of a small refiner with both reforming and FCC cracking, the
hydrogen produced on the reformer could possibly be used in the FCC gasoline
hydrogenation step. This is likely because most refineries of this size would
not include a hydrocracker and if the run on a sweet crude would have excess
hydrogen available. With the excess hydrogen valued as refinery fuel, the costs
of benzene removal from reformates and FCC gasoline would drop from 6.83 cents
per gallon to 6.09 cents per gallon gasoline. This compares to average national
costs of 1.99 cents per gallon of gasoline with hydrogen priced as refinery fuel,
or 2.19 cents per gallon with hydrogen plant costs.
It is obvious from our analysis that the removal costs of benzene would
have a more severe impact on the small refinery. These costs could be as high as
6 to 7 cents per gallon of gasoline, or $1.50/B of crude.
In addition to the removal cost, the removal of benzene from gasoline
would have a greater effect on the small refiner's ability to blend gasoline
because he has less operational flexibility and fewer blending stocks. Our
projections of total reforming and FCC units in 1981 indicate 167 locations with
reforming and only 137 locations with FCCU capacity. Most of the 30 locations with
reforming capacity, but not FCCU capacity, are small refineries under 20,000 B/D.
These refineries will have a higher percentage of reformate in their pool than the
U.S. pool and will tend to have a higher percentage benzene.
5-30
-------
In Chapter 2 we discussed the API benzene survey of 25 major company
refineries and an NPRA survey of 9 small refineries. The average benzene content
at the small refineries was 1.63%, as compared with an average of 1.11% benzene
from the 25 major company refineries. The average range of benzene content
(1.31 to 1.87%) was also higher than the major refinery average (0.59 to 1.79%).
These limited data tend to support the hypothesis that many small refineries
will have relatively high gasoline pool benzene levels. As these small refiner-
ies are more dependent on reformate for pool octane, removal of benzene from
reformates would also have a greater effect on their ability to blend gasoline.
It is likely that many small refineries would be unable to meet projected lead
phase down and possible elimination of MMT with the removal of benzene from
their gasoline pools, which could cause refinery shutdowns or withdrawal from
the gasoline market.
Economics were developed by region and capacity range in this study. The
total national costs of benzene removal from reformate and FCC gasoline by
capacity range are shown in Tables 5.16 and 5.17. These tables could be used
to estimate the number of refineries that would have substantially higher costs
than the national average. For example, 48 refineries would have costs double
the national average for removal of benzene from reformates. Similarly, about
25 refineries would have double the national average costs for benzene removal
from FCC gasoline. Thus, many refineries would experience a much more higher
cost than would be indicated on a national average.
5-31
-------
TABLE 5.16
TOTAL U.S. COSTS OF REMOVAL OF BENZENE FROM
GASOLINE REFORMATE-BY REFORMATS CAPACITY RANGE
REFORMER CAPACITY RANGE (MB/SD)
REFORMATE CAPACITY RANGE (MB/SD)
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/Start-up Costs
5. Working Capital and Royalty
6. Total Investment
y MANUFACTURING COSTS ($/SD) ^
i^3 Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9. Maintenance
10. Total Semi-Variable Operating Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Operating Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
Number of Gasoline Reformer Locations
Total Capacity-Reformate i(MB/SD)
0-1.9
0-1.5
27,990
25,940
53,930
12,944
2,187
69,061
7,153
19,840
6,252
26,092
55,049
88,294
$4.31
$30.5
20
20.5
2.0-4.9
1.6-3.9
49,423
50,336
99,759
23,942
7,437
131,138
24,318
26,784
11,566
38,350
104,464
167,132
$2.40
$57.6
28
69.7
5.0-9.9
4.0-7.9
64,747
69,817
134,564
32,294
15,375
182,233
50,276
26,784
15,601
42,385
145,258
237,919
$1.65
$82.1
27
144.1
10.0-19.9
8.0-15.9
139,525
154,673
294,198
70,608
45,401
410,207
148,457
37,696
34,108
71,804
326,977
547,238
$1.28
$188.8
38
425.5
20-49.9
16.0-39.9
262,250
293,753
556,003
133,441
109,847
799,291
359,193
42,656
64,463
107,119
637,115
1,103,427
$1.07
$380.7
43
1029.5
50.0-99.9
40.0-79.9
129,640
145,215
274,855 1,
65,965
64,545
405,365 1,
212,166
10,912
31,867
42,779
323,117 1,
578,062 2,
$0.95
$199.4
11
608.1
TOTAL
673,575
739,734
413,309
339,194
244,792
997,295
801,563
164,672
163,857
328,529
591,980
722,072
$1.18
$939.1
167
2297.4
(1)
345 Stream Days per .year (SD/YR)
-------
Ul
U)
U)
TABLE 5 .17
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE
U.S.A. BY FCC GASOLINE .CAPACITY RANGE
FCC UNIT CAPACITY RANGE (MB/SD)
FCC GASOLINE CAPACITY (MB/SD)
INVESTMENT ($000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Const. /Start-up Costs
6. Working Capital & Royalty
7 . To tal Inves tmen t
OPERATING COSTS ($/SD) ^
Variable Costs:
8 . Hydrogen
9. Other Variable Costs
10. Total Variable Costs
Semi-Variable Costs:
11. Labor
12. Maintenance
13. Total Semi-Variable Costs
Fixed Costs:
14. Total Fixed Costs
15. Total Operating Costs
TOTAL OPERATING COSTS ($/B)
TOTAL OPERATING COSTS ($ Mill ion/ Year)
Number of FCCU Locations
Total Capacity - FCC Gasoline (MB/SD)
0-4.9
0-2.8
7,100
5,509
9,384
21,993
5,275
766
28,038
8,346
3,227
11,573
7,245
2,550
9,795
22,349
43,717
$ 5.20
$ 15
5
8.4
5-9.9
2.9-5.6
40,485
47,742
58,841
147,068
35,297
8,299
190,664
50,804
35,001
85,805
28,980
17,051
46,031
151,978
283,814
$ 3.12
$ 98
20
91.1
10-19.9
5.7-11.2
71,345
79,861
109,756
260,962
62,631
22,748
346,341
95,797
95,936
191,733
42,021
30,255
72,276
276,068
540,007
$ 2.16
$ 186
29
249.7
20-39.9
11.3-22.5
128,183
177,953
204,817
510,953
122,629
60,190
693,772
207,678
253,841
461,519
59,409
59,241
118,650
553,006
1,133,175
$ 1.72
$ 391
41
660.7
40-79.9
22.6-45.1
140,030
200,065
223,782
563,877
135,331
78,110
777,318
217,154
329,413
546,567
40,572
65,377
105,949
619,600
1,272,116
$ 1.48
$ 439
28
857.4.
^45.2
139,885
156,565
223,523
519,973
124,794
88,239
733,006
215,755
372,136
587,891
20,286
60,285
80,571
584,279
1,252,741
$ 1.29
$ 432
14
968.6
TOTAL
527,028
667,695
830,103
2,024,826
485,961
258,351
2,769,138
795,534
1,089,554
1,885,088
198,513
234,759
433,272
2,207,280
4,525,640
$ 1.60
$ 1561
137
2835.9
Stream Days per Year (SD/Yr)
-------
5.4 Effect of Assumptions on Costs
In order to facilitate development of the economics in this study,
there were assumptions made that would tend to make the removal costs higher
or lower than a more detailed analysis.
Assumptions Leading to Higher Removal Costs
1. Hydrogen plant required at each location
2. Steam costs based on 600 psig steam with no low pressure steam
recovery
3. No by-product credit for H~S or light gas produced in hydrogenation
step
4. Assumed no volume gain in hydrogenation step
5. Separate extraction of reformates and FCC gasoline
Assumptions Leading to Lower Removal Costs
1. No facilities provided for H S recovery
2. Costs based on U.S. Gulf Coast location
3. Costs based on constant 1977 dollars
4. No cost included to meet clean air act restrictions
Other Uncertainties
1. Assumed typical average crude quality and cut point at all locations
2. Assumed typical process configurations and processing routes at
each location
5-34
-------
CHAPTER 6
OTHER ECONOMIC ISSUES ASSOCIATED WITH BENZENE REMOVAL
The preceding chapters have dealt with the sources of benzene in gasoline
and the technology and costs of removing benzene from the two principal
sourcesreformate and FCC gasoline. It was beyond the scope of this study to
look quantitatively at the costs associated with restoring the volume and octane
quality of the pool to the pre-benzene extraction levels, or to assess the impact
on the chemical industry of throwing large volumes of benzene on the market.
These are, however, key issues which need to be resolved to render a final
judgement on the economic impact of controlling benzene at the refinery level.
The volume and octane loss impacts associated with benzene removal can
best be analyzed using linear programming techniques. By incorporating benzene
related information in the process and stream data of a refinery model, runs
can be executed to assess the total economic impact of benzene removal to any
desired level in a manner similar to that employed for evaluating the economic
impact of lead removal and lead phase down. This is a major study which could
be undertaken later if circumstances warrant. In this chapter, simple methods
have been used to suggest possible magnitudes that these octane loss, volume
loss, and chemical market impacts might reach. Also contained in this chapter
are other items which were not previously discussed in detail, but would warrant
further study.
6.1 Octane Loss
The effect on gasoline pool octane of removing benzene from reformates and
FCC gasoline is summarized in Table 6.1. For reformate, the octane loss is
because of the higher blending value of benzene, relative to the average pool.
These calculations are shown in Table 6.2. In the case of FCC gasoline, there
is a large additional octane loss due to the hydrogenation step. The hydrogena-
tion octane loss was based on data contained in the 1976 Arthur D. Little Lead
Phase-Down study. These calculations are shown in Table 6.3. The octane loss
calculations are based on the blending values shown in Table 6.4. The results
6-1
-------
TABLE 6.1
1981 U.S. POOL OCTANE LOSS
ASSOCIATED WITH BENZENE REMOVAL
RON MON R+M/2
Refinery Reformates 0.13 0.06 0.10
FCC Gasoline (Hydrogenation) 1.12 0.48 0.80
FCC Gasoline (Extraction) 0.04 0.02 Q.Q3
FCC GasolineTotal 1.16 0.50 0.83
Refinery Reformates and
FCC Gasoline 1.29 0.56 0.93
6-2
-------
TABLE 6.2
EFFECT ON 1981 U.S. OCTANE POOL
OF BENZENE REMOVAL
Total Reformate
Less Benzene in Heavy Reformate
Total Gasoline Reformer Reformate
Separation Efficiency: .95
Extraction Efficiency: .995
Benzene Removed from C, Cut
D
Effect on U.S. Octane Pool
MB/CD
U. S. Pool^ 7,450
(2)
Less Benzene Removed ( 62.9)
Less Butane(2) ( 9.5)
Net U.S. Pool 7,377.2
U. S. Pool Octane Loss
FROM REFORMATES
MB/CD Vol.
2,232 3.
( 100) (0.
2,132 3.
RON MON
91.00 83.00
106.50 89.80
92.00 89.00
90.87 82.94
0.13 0.06
% BZ MB/CD BZ
0 67.0
5) ( 0.5)
11 66.5
.95
.995
62.9
R4M/2 RVP
87.00 10.50
98.45 3.20
90.50 59.00
86.90 10.50
0.10
Assume benzene backed-out of 91 RON/83 MON pool gasoline on an unleaded basis
blending values (Table 6.4)
6-3
-------
TABLE 6.3
EFFECT ON 1981 U.S. OCTANE POOL
OF REMOVAL OF BENZENE FROM FCC GASOLINE
Volume
1981 Basis MB/SD RON MON R4M/2 RVP
U.S. Pool 7,450 91.00 83.00 87.00 10.50
Hydrogenation of FCC gasoline C, Cut 385(1) (21.6 ) ( 9.2 ) (15.4 )(2)
b
Intermediate U.S. Pool 7,450 89.88 82.52 86.20 10.50
(3)
Benzene Removal from C, Cutv ' - 19.4 106.50 89.90 98.15 3.2
O
Less Butane(3) . - 2.9 92.00 89.00 90.50 59.0
Net U.S. Pool 7,427.7 89.84 82.50 86.17 10.50
U.S. Pool Octane Loss 1.16 0.50 0.83
gasoline C cut = 2,568 MB/SD x 0.15 = 385 MB/SD
Octane loss due to hydrogenation from ADL Lead Phase-Down Study, Appendix H,
dated May 1976, Pgs. 29, 30:
Hydrogenated FCC RON =71.4 MON =71.3
Unhydrogenated FCC 93.0 80.5
Change in ON 21.6 9.2
Change in R-Hi/2 15.4
ADL blending values (Table 6.4)
6-4
-------
TABLE 6.4
BENZENE REMOVAL FROM GASOLINE
OCTANE BLENDING VALUES
RON MON R+M/2
Compound Cl Cl Cl
Benzene 106.5 89.8 98.15
Cumene 106.9 88.0 97.45
Toluene 113.9 102.1 108.0
Cyclohexane 83.0 77.2 80.1
Butane 92.0 89.0 90.5
Ethylbenzene 115.6 98.6 107.1
RVP
3.2
0.2
1.0
3.3
59.0
0.4
(1)
ADL estimates blended into 91 RON/83 MON unleaded gasoline
6-5
-------
(8)
in Table 6.1 agree well with Exxon data which indicate a loss of 0.2 octane for
benzene removal from reformate only, and 0.8 octane for benzene removal from all
sources in gasoline.
The octane loss shown in Table 6.1 can be restored through some combina-
tion of new investment and processing conditions at refineries. Alternatively,
motor vehicle designs can be modified to use lower quality gasolines (with the
related vehicle efficiency implications). At the refinery level, the cost of
adding clear octanes increases rapidly at higher pool octane numbers. Figure 6.1
shows a rough assessment of the cost of adding octanes derived from previous
ADL work on lead phase-down combined with other sources drawn from the literature.
Contacts were also made with the industry on this issue. Responses range
from 15 to 50 cents per octane number barrel (.36 to 1.19 cents per octane
number gallon) depending on pool octane and individual refinery constraints.
At the 91-92 RON unleaded pool level, the average appeared to be about 30 cents
per octane barrel (.71 cents per octane gallon), which is a little on the high
side compared to Figure 6.1. Table 6.5 shows that the octane penalty might
range from one-third of a cent per gallon to two-thirds of a cent per gallon for
a total national impact of between $380 and $760 million/year, depending on the
cost of replacing the lost octanes. About 85% of the penalty is because of the
hydrogenation step associated with the extraction of FCC gasolines. If this
step could be avoided, the octane penalty would be relatively small and, in
addition, the costs and investments associated with the hydrogenation step could
be avoided. Clearly, it is important to determine the technical and economic
practicality of eliminating this hydrogenation step.
6.2 Volume Loss
The benzene produced by the extraction of reformates and FCC gasoline
amounts to about 1.1% of U.S. gasoline production, or about 82.7 MBPD (63.3 MBPD
from reformate and 19.4 MBPD from FCC gasoline). In addition to these volumes,
the industry would be called upon to provide additional refinery fuel and feed-
stock for hydrogen manufacture (the costs for which, however, have been accounted
for in developing the direct cost of benzene removal).
6-6
-------
c
o
to
O
0)
c
n)
4J
o
o
t-i
0)
CL
c
0)
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
84 85 86 87 88
Clear Octane Number: R4M/2
SOURCE: Arthur D. Little and Industry Data
89
Figure 6.1 - Cost of Adding Octane Number as a
Function of Pool Octane: 1977 $
6-7
-------
TABLE 6.5
NATIONAL OCTANE LOSS PENALTY
RANGE OF POSSIBLE COSTS
Octane^ '
Loss
Cents per Octane Number Barrel
(2)
Cents per Gallon of Gasoline
Reformate .10
FCC . 83
Total .93
Of Which Due to
Hydrogenation (.80)
Million Dollars per Year
Reformate .10
FCC . 83
Total .93
Company
Lows
15
.036
.296
.332
(.286)
41
338
379 .
Figure ^
6-1
19
.045
.375
.420
(.362)
51
429
480
Company
Average
30
.071
. .593
.664
(.571)
81
677
758
Of Which Due to
Hydrogeration
(.80)
(327)
(413)
(652)
(1)R+M/2
(2)
(3)
Based on 7,450 MB/D
At 91 RON/87 R4M/2
6-8
-------
If the benzene withdrawn from the gasoline pool is worth more than gasoline
(i.e., for chemical markets), the differences should be credited to the costs
estimated in the proceeding chapters. On the other hand, if the benzene is
worth less than gasoline, the differences should be debited. In the next
section, it will be shown that the volumes are large, relative to chemical
markets, and new uses for benzene will have to be found. Among the possibilities,
are conversion to an acceptable gasoline blending component, refinery fuel, or
exports (for use as a gasoline blending component or for chemical manufacture).
Incineration could be a last resort for small quantities at remote locations.
The idea of converting the benzene back to a high octane gasoline blending
component has obvious attractions, particularly from an analytical point of
view, since it restores both the volume and the octane loss. Benzene could be
alkylated with propylene to cumene or with ethylene to ethylbenzene; both of
which have octane blending values about the same as benzene or better (see
Table 6.4). Table 6.6 shows a range of values for producing cumene as a gaso-
line blendstock. To produce one pound of cumene takes about .69 pounds of
benzene and .38 pounds of propylene. If these components were valued at their
1976 average value, as reported by the U.S. Tariff Commission, the calculated
value added is about 2 cents per pound of cumene, which can be taken as one
measure of the cost of manufacture. At this cost of manufacture with propylene
at its 1976 chemical value, benzene would only be worth 12.1 cents per gallon
(see Table 6.6). Even if the manufacturing costs were halved to 1 cent per
pound of cumene, the value of benzene would only be 22.8 cents per gallon.
Both of these are below benzene's fuel value (26.5 cents per gallon). Only
if propylene were available at fuel value would it be possible to realize a
value for benzene above its fuel value and then only if the cost of manufacture
were less than 2 cents per pound. It appears that the conversion of benzene to
cumene is unlikely to be economical, although further analysis would be required
to give a definitive answer. Similarly, it is likely that conversion to
ethylbenzene (which requires more costly and less readily available ethylene)
will be unattractive. Conversion to cyclohexane could also be explored,
possibly in the context of severe hydrogenation of the FCC heart cut to convert
the contained benzene to cyclohexane, thereby avoiding subsequent extraction.
6-9
-------
TABLE 6.6
RANGE OF VALUES OF BENZENE FOR ALKYLATION TO CUMENE
FOR USE AS A GASOLINE BLENDING COMPONENT
A. 1976 Reported Values*
LB/Gal.
Benzene 7.37
Propylene 4.35
Cumene 7.21
C/LB
10.6
7.4
12.1
C/Gal.
78.1
32.2
87.3
B. Calculated Value Added for Cumene Manufacture
.69 Benzene @ 10.6 cents per
pound
.38 Propylene @ 7.4 cents per pound
Total
1.00 Cumene @ 12.1 cents per
Calculated Value Added
C. Value of Benzene for Conversion
pound
to Cumene
Propylene at
Fuel Value (4.2C/LB)
Unleaded Gasoline Value $/B
Cumene Blending Premium**
Cumene Gasoline Value $/B
C/LB
Cumene Manufacturing Cost
Propylene Cost (.38 LB/LB Cumene)
Cumene Total Cost Ex. Benzene
Value of .69 LB Benzene
Value of Benzene C/LB
C/Gal.
$/B
15.90
2.10
18.00
5.94
1.00
1.60
2.60
3.34
4.84
35.7
14.99
15.90
2.10
18.00
5.94
2.00
1.60
3.60
2.34
3.39
25.0
10.50
7.31
2.81
10.12
12.10
1.98
Propylene at
Chemical Value (7
15.90 15.
2.10 2.
18.00 18.
5.94 5.
1.00 2.
2.81 2.
3.81 4.
2.13 1.
3.09 1.
22.8 12.
9.55 5.
1976
4C/LB)
90
10
00
94
00
81
81
13
64
1
07
*U.S. International Trade Commission Synthetic Organic Chemicals
**At 19c/octane barrel and 11.2 octane (R+M/2) premium
6-10
-------
Tentatively, it seems unlikely that benzene can be converted to gasoline
blending components at a value much above fuel value. The possibility of export
might be considered, particularly as a gasoline blending component, since
foreign governments would presumably seek to protect their chemical manufactur-
ing enterprises from U.S. dumping of surplus benzene. The possibility of exports
requires further study.
Table 6.7 summarizes the volumetric loss penalty ranging from fuel value
down through incineration. Value reduction to fuel value amounts to about
$150 million per year (a little over 0.1 cents per gallon of gasoline). If,
because of high aromatic content of the flue gas, benzene can not be utilized
as a fuel and had to be converted to cumene, the volume penalty could be in the
order of $250 million and a little over .2 cents per gallon of gasoline.
Finally, the maximum possible loss is in the order of .4 cents per gallon and
$500 million per year, if the benzene had to be incinerated.
6.3 Impact on the Chemical Markets
Large volumes of surplus benzene will depress the chemical benzene price
to the levels of its alternative disposal value as outlined above. This loss
of benzene value, to the extent that it reflects capital and operating costs
associated with existing benzene producing facilities, will have to be recovered
in the price of gasoline and/or other chemical products. Some chemical centers
such as Peurto Rico, which are heavily dependent on aromatics, might be
particularly hard hit.
Table 6.8 indicates the general magnitude of the problem from a benzene
supply/demand point of view. In 1976, benzene demand was in the order of
100,000 barrels per day, of which about two-thirds were derived from naturally
occuring benzene extracted from reformates, the pyrolysis gasoline from olefins
plants based on heavy liquids, and from coal. The remainder came from imports
and toluene hydrodealkylation. Toluene hydrodealkylation is generally considered
to be the marginal source of benzene, since tolune, which would otherwise be
use as a high octane blending component, is converted to benzene at relatively
low yield. As more and more benzene became available from a program to reduce
benzene in gasoline, the hydrodealkylation units would shut down and the toluene
6-11
-------
TABLE 6.7
VOLUMETRIC PENALTY FOR BENZENE REMOVAL
FROM REFORMATES & FCC GASOLINE
Cumene Manufacturing
Cost @ 1.5C/LB &
Propylene at:
Loss in Benzene Value
Value as Gasoline C/Gal.
Alternate Value c/Gal.
Less in Value C/Gal.
$/B
Cents per Gallon Gasoline
Reformate
FCC Gasoline
Total
Million Dollars per Year
Reformate
FCC Gasoline
Total
Fuel
Value
37.9
26.5
11.4
4.79
0.097
0.030
0.127
111
34
145
Fuel
Value
37.9
30.4
7.5
3.15
.063
.020
.083
73
22
95
Chemical
Value
37.9
17.5
20.4
8.57
.173
0.053
0.226
198
61
259
Incineration
37.9
0
37.9
15.90
.322
0.098
0.420
367
113
480
^Based on 7,450 MB/D
6-12
-------
TABLE 6.8
BENZENE SUPPLY AND DEMAND
(Thousands of Barrels per Day)
1981 1985
Refinery Reformates
FCC Gasoline
Subtotal
Olefin Plants
Coal
TOTAL
Required Additional
Supply From
Toluene Hydrodealkylation
Imports (Exports)
TOTAL
TOTAL SUPPLY
References :
49.4 112.7
19.4
49.4 132.1
10.7 33.2
3.9 3.9
64.0 169.2
29.3 0 0
5.1 (43.0) (26.3)
34.4 (43.0) (26.3)
98.4 126.2 142.9
J.E. Fick, Chemical Purchasing, p. 21, Sept. 1977.
P.E. Baggett, Chemical Institute of Canada, Chemical
Research Assoc. , Montreal
Schoeffel, et al., Chemical
, November 3, 1977.
Engineering Progress, p
1976 Low High Low High
DEMAND 98.4 126.2 142.9 158.5 180.1
SUPPLY
Direct Availability
112.7
19.4
132.1 132.1
46.6 65.2
3.9 3.9
182.6 201.2
0 0
(24.1) (21.1)
(24.1) (21.1)
158.5 180.5
Oil and Gas Journal, p. 45, Feb. 21, 1977.
Chemical Engineering News, March 31, 1975.
6-13
-------
would be diverted back to gasoline. Imports would also cease under such a
program.
Table 6.8 shows that, based on the range of benzene demands projected
in the literature, even with all hydrodealkylation units shut down, a substantial
surplus would exist in 1981 (reflected in Table 6.8 as a requirement to export
benzene) and that by 1985, the surplus would persist. Growth in benzene demand
is not adequate to absorb the additional benzene extracted from reformate through
a benzene reduction in gasoline program as well as the additional volumes
which will arise from the new heavy liquid olefin plants. (Note, as in the case
of reformate and FCC gasoline, the pyrolysis gasoline from the olefin plants
would also have to be extracted before it could be blended back to gasoline.)
As indicated in Table 6.8, which reinforces some of the conclusions made earlier,
benzene values can drop to very low levels. This could also stimulate chemical
demand for benzene and accelerate the time when the surplus might be taken up
to an earlier date than implied in Table 6.8. However, during the transition
period, the chemical industry could be severely disturbed,as companies producing
benzene derivatives gain market advantage over those producing derivatives
with similar end uses from other raw materials. The effect of a benzene reduction
program might not be confined simply to the more obvious loss of benzene values.
Table 6.9 indicates the potential range of loss of chemical values which
the industry would seek to recover through other chemical products and/or
gasoline. If benzene fell to fuel value, the loss of chemical value, based on
1976 average prices and volumes, would be a little over $600 million, equivalent
to 0.54 cents per gallon of gasoline. If benzene is converted to cumene, the
loss is higher. In the extreme, if benzene fell to zero value, the loss would be
0.76 cents per gallon. These losses are of the same order of magnitude as
the estimated cost of benzene removal from reformate, estimated in Chapter 5
(.82 cents per gallon). These losses are based on directly available U.S.
benzene and benzene derived from toluene hydrodealkylation (HDA). The loss of
value because of imports would be zero. The loss of value associated with toluene
HDA is less than on directly available benzene (amounting to the value associated
with toluene as a gasoline blending component after adjustment for yield, as
shown in Table 6.10). Table 6.9 shows that the loss of value for the
benzene directly available is about .4 cents per gallon, with benzene at fuel
6-14
-------
TABLE 6.9
RANGE OF LOSS OF CHEMICAL BENZENE VALUES
Alternate Value
Loss
1976 Volumes (MBPD)
Direct Availability
Hydrodealkylation
Imports
Total
Loss in Value
Cents per Gall
Direct Availability
HDA
Imports
Total
Direct Availability
KDA
Imports
Total
ACCOMPANYING BENZENE REDUCTION
IN GASOLINE
Cumene
Cost
Manu facturing
@ 1.5C/LB &
Propylene at :
Fuel
Value
.ue C/Gal. 78
C/Gal. 26.5
C/Gal. 51.5
>D)
Lty 64
\ 29
5
98
of Gasoline
Llity .442
.094
0
.536
Der Year
Llity 505
107
0
Fuel
Value
78
30.4
47.6
64
29
5
98
.409
.094
0
.503
467
107
0
Chemical
Value
78
17.5
60.5
64
29
5
98
.520
.094
0
.614
594
107
0
Incineration
78
0
78
64
29
5
98
612
574
701
.670
.094
0
0.764
765
107
0
872
(1)
Based on 7,450 MB/D gasoline
6-15
-------
TABLE 6.10
VALUE OF TOLUENE AS GASOLINE vs. BENZENE
$/B C/Gal.
Unleaded Gasoline Value 15.90 37.9
Toluene Blending Premium* 2.86 6.8
Toluene Gasoline Value 18.76 44.7
Barrels Toluene per Barrel Benzene 1.21 1.21
Toluene Gasoline Values Benzene 22.70 54.0
Benzene Sales Value 32.76 78.0
Loss of Chemical Value 10.06 24.0
*At 19 cents per octane barrel and 15.1 octane (R+M/2) premium
6-16
-------
value, up to .7 cents per gallon is benzene had zero value.
In addition to the cost of disposal of the large increase in volume in
the chemical market, there is the problem of major dislocations of benzene supply
relative to traditional sources.
The current distribution of benzene producers is shown in Figure 6.2. This
corresponds well with the distribution of benzene consumption plants shown in
Figure 6.3.
With the removal of benzene from all refinery reformates and FCC gasoline,
benzene becomes much more widely scattered over the U.S.,as shown in Figure 6.4.
The feasibility of bulk transporting benzene from many of these producing loca-
tions to the existing benzene consumption plants is questionable. Most likely
the benzene in many of these locations would have to be shipped in small tank
truck cargoes which would make the costs excessive.
Another alternative would be to locate new benzene consuming plants in the
vicinity of the new benzene producing plants. This also would be a questionable
approach, because of the small volumes of benzene produced and dislocations from
the traditional areas of demand for benzene chemical derivatives.
The costs of handling and transporting the benzene produced in many PADD IV
locations may be such that alternate disposal of benzene through conversion
to cumene for gasoline blending or as fuel would be more economical than trying
to reach traditional benzene chemical markets.
6.4 Estimated Cost of Other Economic Issues Associated
with Benzene Removal from Reformates & FCC Gasoline
The total cost of octane loss, volume loss and chemical market loss
are summarized in Table 6.11. These losses range from a low of 0.9 cents per
gallon of gasoline, to a high of 1.5 cents per gallon of gasoline. Annual costs
range from $1,048 million to $1,718 million per year.
6-17
-------
00
***N>-
/ !
.'' \
°»>s*"--! S
/ c
j '\
I. >.
N. DAKOTA \
KANSAS
^.MISSOURI \
V
\
C. .L..-SCENTUCKY \^
^oy--.^ /
/ urA-ffi
I i- . !
j I COLORADO"1
\ / A / f
\ ' / j
^ """T^^^]4?r°--^Ws^^
I i j I "*
/ Li ./
j --«^-j. f
/ j
.J..j~-^ j
I'V^rr^u.
y !" ;^v.
I f \N.VA./
>W / VA.
\
V.
Refinery Producers
A Other Producers
SOURCE: Arthur D. Little estimates
Figure 6.2
Current Distribution of Benzene Producers
-------
SOURCE: PEDCO Environmental, "Atmospheric Benzene Emissions", August 1977
Figure 6.3 - Current Distribution of Benzene Consumers
-------
hO
O
|
-------
TABLE 6.11
ROUGH COST OF OTHER ECONOMIC ISSUES
ASSOCIATED WITH BENZENE REMOVAL FROM REFORMATES & FCC GASOLINE
Cents per Gallon of Gasoline Low Medium High
Octane Los Penalty 0.33 0.42 0.66
Volume Loss Penalty 0.08 0.13 0.23^
Chemical Market Loss 0.50 0.54 0.6
Total 0.91 1.09 1.50
Million Dollars per Year
Octane Loss Penalty 379 480 758
Volume Loss Penalty 95 145 259^
Chemical Market Loss 574 612 701^
Total 1,048 1,237 1,718
Based on cumene manufacturing costs, propylene at chemical
value. Incineration losses are unrealistic
6-21
-------
If these losses are added to the benzene removal costs for reformates
and FCC gasoline, we get a total national impact of 3.1 to 3.8 cents per
gallon of gasoline. This amounts to annual costs of $3.5 to $4.2 billion per
year.
6.5 Other Items that Warrant Further Study
A. Evaluation of Economics of Benzene Removal
from Other Streams
The economics of benzene removal were only developed for reformates and
FCC gasoline in this study. The economics of removing benzene from other
gasoline pool streams was beyond the scope of this study. The economics could
be developed for the other gasoline pool streams, however, through the processing
routes discussed in Chapter 4. The procedure would be analogous to the proce-
dure for reformates and FCC gasoline. First, develop base case economics for
each stream. Second, scale the economics according to capacity. Third, apply
the scaled economics to the projected production capacity distribution of each
of these streams on a regional basis. Finally, sum the costs of benzene removal
on a capacity and regional basis to get the national impact.
Light straight run gasoline would be the next stream recommended for
evaluation, since it is the third largest benzene contributor to the pool. For
initial evaluation, the same 180°F overpoint would be assumed for naphtha feed
to reformers. The volume and benzene content of light straight run would be
developed from the gasoline pool data of Chapter 2, and the benzene content data
of Chapter 3. The costs for fractionation to obtain a C, cut and mild hydro-
genation would have to be developed for light straight run independently,
however, the sulfolane extraction costs could be used directly based on the
volume of C cut to be extracted.
b
For a detailed analysis of the costs of removing benzene from light
straight run, further work would be required. A complete analysis would include
evaluation of the effect of crude quality and naphtha cut point changes. This
would require LP runs to determine the optimum naphtha cut point in order to
minimize total benzene removal costs from reformates and light straight run.
6-22
-------
B. Evaluation of the Effect of Crude Oil Quality & Naphtha
Cut Point Changes on Benzene Removal
The most important variables affecting reformate benzene level are
the levels of benzene precursors in the naphtha feedstock to the reformer
and reformer severity. The effect of reformer severity was developed in Chapter 2.
The level of benzene precursors in the naphtha feed is a function of crude oil
quality and naphtha cut point.
The effect of benzene precursor level was handled in this study, by
assuming a typical naphtha cut point to gasoline reformers of 180°F and an
average benzene content in reformate of 3.0 volume percent. Sufficient crude
data were not available to make an in-depth analysis of benzene precursor levels.
In order to make an in-depth analysis of the effect of crude oil quality
and naphtha cut point, it would be necessary to obtain data on all major crudes
processed and detailed information on individual refinery crude slate. It
would then be necessary to make LP runs to determine the light straight run and
reformate benzene content with various naphtha cut points.
6-23
-------
7. REFERENCES
(1) "Fact Sheet on President Carter's Energy Program", Petroleum Intelligence
Weekly, Special Supplement, P. 4, April 25, 1977
(2) "The Impact of Lead Additive Regulations on the Petroleum Refining
Industry", Vol. I and II, Arthur D. Little, Inc., EPA-450/3-76-016-2,
May 1976
(3) "1977 Directory of Chemical Producers", Stanford Research Institute,
1977, Pgs. 430, 913, 914, 933, 934, 935
(4) "Worldwide Construction", Oil and Gas Journal, 75 (No. 41), Pgs. 110 - 113,
October 3, 1977
(5) "Worldwide H.P.I. Construction Boxscore", Hydrocarbon Processing,
Pgs. 3 - 16, October 1977
(6) Sterba, M. J. and Haensel, V., "Catalytic Reforming", Ind. Eng. Chem.,
Product Research Development, L5_ (No. 1), Pgs. 2 - 17, 1976
(7) "Hydrocarbon Distribution in Commercial Gasolines-Summer 1976",
E. I. duPont de Nemours and Company, Inc., June 1977
(8) "OSHA Benzene Exposure Regulations", Exxon Company U.S.A. to U.S. Depart-
ment of Labor (OSHA), August 27, 1977
(9) McBride, W.L. and Mosby, J.F., "Low-Pressure Heavy Distillate Ultrafining",
Amoco Oil Company to NPRA Annual Meeting, April 1973
(10) Gerth, R., "Sulfolane Royalty Costs", Shell Oil Company, October 1977
7-1
-------
APPENDIX A
GASOLINE POOL COMPOSITION
U.S. GASOLINE PRODUCTION CAPACITY
SUPPLY/DEMAND ANALYSIS
A.1 Summary
An analysis was made of the present and projected production rates and
capacities of the major U.S. gasoline producing unitscatalytic reformers
and fluid catalytic cracking (FCC) units. The primary purpose of the analysis
was two-fold: (a) by comparison of historic gasoline production rates and
capacities of these units to the cluster model output, verification and adjust-
ment of the cluster gasoline blends can be achieved, giving an adjusted blend
for use in the benzene removal study; (b) with this additional model calibra-
tion, firm announced capacity additions can be compared to projected gasoline
demand to determine if the announced capacity is adequate for future production
levels; this, in turn, will identify the capacity required for benzene extraction.
The conclusions of the study indicate that the cluster model output does
indeed require adjustment in order to be consistent not only with existing
gasoline producing capacity, but also with the limited industry perspectives
of the pool blend composition. Using the results of the present analysis, the
1981 U.S. gasoline pool is expected to be comprised of:
Reformate
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Natural Gasoline
lit. Hydrocrackate
Isomerate
S.R. Naphtha
Total -
Present
Study
30.0
34.5
13.6
1.4
6.4
1.2
2.5
1.8
1.4
7.2
Cluster
Model Results
25.7
33.5
13.3
3.3
6.9
1.2
3.5
2.9
1.8
7.9
U.O.P. '
for 1972
33
38
13 (with polymer)
-
-
4 (thermal)
-
-
-
12 (with natural
and butanes)
100.0
100.0
100.0
(1)
M.J. Sterba and Vledimir Haensel, IEC Prod. Res. Dev., 15, No. 1, 2 (1976)
A-l
-------
(The U.O.P. estimates are quoted for illustrative purposes
only, and were not considered to be authoritative. Details associated
with the above calculation are not included herein.)
Considerably larger deviations were found for individual PAD Districts,
with the poorest agreement exhibited by PADD V:
Cluster
Present Study Model Results
Reformate 43% 26.1%
FCC Gasoline 28% 28 %
The supply/demand analysis for FCC units and catalytic reformers indicates
that presently existing plus announced, firm capacity additions will provide
adequate capacity for the indefinite future. Specifically, the gasoline
demand projections exhibit a maximum in about 1981, with a continuous decline
thereafter; although capacity will be tight in 1981, it is adequate to meet
demand and then becomes increasingly surplus thereafter. Indeed, it is not
surprising that the industry has announced construction plans adequate to meet
projected demand for the next three years; the unusual characteristic from
an historical viewpoint is the absolute decline in gasoline demand after 1981.
Hence, it is recommended that these announced capacity figures be used to
estimate benzene extraction costs, assuming no further additions will be necessary.
Furthermore, it is observed that, although gasoline-producing unit margins should
improve through 1981, they should not be adequate to support new investments
in these units after 1981. Of course, limited expansions could occur after 1981,
because of an individual refiner's lack of access to the excess unit capacity
owned by other refiners.
A. 2 Methodology of Study
The FCC unit yields from the cluster model should reflect the changing
impact of crude slate, FCC unit feed hydrogenation, and the lead phase-down
requirements between the individual years studied in the EPA lead phase-down
study, 1973, 1977, 1980 and 1985. These yields were reviewed and discussed with
industry sources; the yields were judged to be reasonable for every cluster
other than the East Coast cluster, which was adjusted downwards slightly.
A-2
-------
The percentage FCC gasoline in the cluster pools was also examined,
and observed to fluctuate erratically from year-to-year, probably due to the
L.P. optimization undertaken for each year being relatively insensitive to the
percentage FCC gasoline in the pool. Furthermore, a simplifying assumption was
made in the cluster model runs that no new downstream capacity could be added
in existing cluster refineries. Since substantial additions have, in fact, been
made, this also biases the cluster model pool composition. Further cluster
model differences are attributable to projected gasoline growth rates differing from
present estimates.
A tabulation of historic levels of FCC capacity and actual B.O.M. gaso-
line production was therefore made by year from 1970 through 1976. Various
percentages of FCC gasoline in the pool were assumed in the vicinity of the
cluster model predictions. From the figures on gasoline production, assumed
percentage of FCC gasoline, FCC unit gasoline yield, and FCC unit capacity,
a stream-day utilization factor could be calculated. The assumed percentage
of FCC gasoline in the pool which gave about 90% of stream day utilization during
periods of significant growth of FCC capacity was taken as the best estimate
of FCC gasoline percentage in the total pool. Although individual refiners
have only a range of guesses of the correct percentage, this figure was
checked with selected refiners and confirmed to be reasonable.
A similar procedure was followed to estimate the percentage reformate in
the gasoline pool. However, with catalytic reformers, a substantial fraction of
the capacity is dedicated to BTX production, which is not directly applicable to
gasoline pool calculations. Therefore, estimates of BTX production were obtained
from a Stanford Research Institute report on this topic. These data were
compared to Oil and Gas Journal data on extraction capacity to confirm the likely
source of this BTX. Individual refiner discussions were also conducted to deter-
mine the source of the BTX production (e.g., reformate versus ethylene crackers)
and to determine if a fraction of this BTX reformate was also blended into the
gasoline pool. This allowed estimates of the segregation of reformer capacity
between BTX production and gasoline production. After reconfirmation of this
segregation with individual refiners, an assessment could be made of the "gasoline
reformer capacity", which ranged from 100% of total reforming capacity in PADD IV
A-3
-------
to about 50% of total capacity in PADD III. An estimate of reformate percentages
in the gasoline pool was then calculated by the same technique as used for FCC
units, and then confirmed as being in a reasonable range by discussions with
individual refiners.
A.3 FCC Unit Results
The cluster model output for FCC unit gasoline yields and percentage of
FCC gasoline in the gasoline pool are shown in Table A.I. The yields are generally
reasonable, reflecting feedstock variations and feed hydrotreating, although
the PADD I yields appear to be a few percentage points too high. The percentage
FCC gasoline in the total gasoline pool appears to be too low on average.
Conversations with a major eastern refiner indicate that they have 35 to 40%
FCC gasoline in their pool. Haensel of U.O.P., (IEC Prod. Res. Dev., 15, No. 1,
P.2, 1976)' reports 38% as a U.S. average. Also, the model results are somewhat
erratic in certain years, notably 1977 from PADD I and PADD V. The following
subsections report reasonable averages to be used in each PAD District.
PADD I
Total FCC capacity (fresh feed basis) for PADD I is shown in the histogram
of Figure A.I. No new capacity additions have been announced. Historic data
from the Bureau of Mines on PADD I gasoline production is shown below:
TABLE A.2
PADD I GASOLINE PRODUCTION
(BOM), MB/CD
1970 1971 1972 1973 1974 1975 1976
661 697 713 748 702 680 760
A-4
-------
TABLE A.I
CLUSTER MODEL OUTPUT SUMMARY
FOR FCC UNITS
1973 1977 1980 1985
PADD I
FCC Unit Yield, % 54.7 60.0 59.7 59.5
FCC Gasoline in Total Pool, % 37.8 30.6 33.2 34.6
PADD II
FCC Unit Yield, % 55.5 54.0 56.6 55.8
FCC Gasoline in Total Pool, % 32.0 32.8 34.8 35.7
PADD III
FCC Unit Yield, % 53.5 53.2 53.6 56.8
FCC Gasoline in Total Pool, % 32.3 34.0 34.3 34.6
PADD I through III
FCC Unit Yield, % 54.4 54.3 55.5 56.8
FCC Gasoline in Total Pool, % 33.0 33.0 34.3 34.5
PADD V
FCC Unit Yield, % 53.0 53.2 53.8 55.0
FCC Gasoline in Total Pool, % 27.4 21.5 28.5 29.9
TOTAL U.S.
FCC Unit Yield, % 54.2 54.2 55.3 56.6
FCC Gasoline in Total Pool, % 32.1 31.4 33.6 32.7
The yield of gasoline in all PAD Districts is projected to increase slightly
(Table A.I) over time, due to increased paraffinicity and additional feed hydro-
treating. Therefore, for PADD I, the following yields were assumed, with a
linear interpolation from Table A.I.
A-5
-------
700
6OO
a
2
U)
D
t
o
a.
g
8
LL
400
300
Figure A.I
PADD I
FCC UNIT CAPACITY/DEMAND
A
/
1O II 12 IT, 14 IS 76 77 IS 79 8O 8l 61 83 84 85
A-6
-------
TABLE A.3
PADD I FCC GASOLINE YIELDS, LV %
1970
54.7
1971
54.7
1972
54.7
1973
54.7
1974
55.6
1975
56.6
1976
57.5
1980
57.5
1981
57.5
1985
57.5
On this basis, the percentage utilization of FCC capacity for the indicated
percentage of FCC gasoline in the total pool, becomes:
TABLE A.4
PADD I FCC UTILIZATION
FCC Capacity, MB/SD
FCC Feed, MB/CD
% Utilization
FCC Feed, MB/CD
% Utilization
FCC Feed, MB/CD
% Utilization
FCC %
37
37
40
40
38
38
1970
622.4
447.1
72
483.4
78
459.2
74
1971
637.6
471.5
74
509.7
80
484.2
76
1972
653.4
482.3
74
521.4
80
495.3
76
1973
626.3
506.0
81
547.0
87
519.6
83
1974
573.8
467.2
81
505.0
88
480.0
84
1975
560.0
444.5
79
480.6
86
456.5
82
1976
565.4
489.0
86
528.7
94
502.3
89
It would be expected from Figure A.I that the capacity utilization would
be low between 1970 and 1975, for absolute decreases in capacity took place.
Also, it would be expected that utilization should approximate 90% by 1976, for
increases in capacity are taking place, but no new announcements have been made.
Usage of 38% FCC gasoline in the pool meets these requirements, and is recom-
mended. In addition, it is in good agreement with the 1973 figure (Table A.I)
used in the model calibration, and would not be expected to vary significantly
in the study period. The solid line in Figure A.I shows the resulting FCC feed
rate on an historical basis, in MB/CD. When this solid line reaches 90 to 93%
of the histogram, it is expected that the FCC capacity is fully utilized.
A-7
-------
If a total PADD I through IV production of 6.325 MMB/D is projected for
1981 and 5.89 MMB/D is used for 1985, and if these estimates were prorated
among PAD Districts on the basis of 1976 production, the PADD I gasoline produc-
tion in 1981 would be 813.1 MB/D and 1985 would be 757.8 MB/D. Using 38% gaso-
line in the pool and 57.5% yield, the gas oil feed rate required would be 547.4
MB/D in 1981 and 500.4 MB/D in 1985. The dashed line in Figure A.I represents
these projections, indicating a capacity utilization of 92% in 1981. Since
this small amount of additional capacity can be met by debottlenecking, by
transfers from other PAD Districts, or by adjustments in the gasoline pool
composition, there is no need for significant further additions to PADD I
FCC capacity if the 1985 demand projection is correct.
PADD II
Total FCC capacity for PADD II is shown in the histogram of Figure A.2.
The capacity has shown a consistent increase over the current decade, although
only marginal new additions have been announced. The yield patterns of Table A.I
for PADD's II and III should follow consistent patterns. Therefore, the 1977
yield point of Table A.I for PADD II was not used, and the following yields are
recommended:
TABLE A.5
PADD II FCC GASOLINE YIELDS, LV %
1970 1971 1972 1973 1974 1975 1976 1980 1981 1985
55.5 55.5 55.5 55.5 55.8 56.1 56.3 56.6 56.6 56.6
On this basis, the percentage utilization of FCC capacity, for the indicated
percentage of FCC gasoline in the total pool becomes:
A-8
-------
Figure A. 2
PADD II
FCC UNIT CAPACITY/DFJtfAND
g
^
| (400
UJ '
p
j£
^
o.
vJ 1)2.00
LL
j
/
'
/-
-
/
X
X'
r ~
«- «
+, ^ 1)6M4*JO
r X.
X
70 71 12 73 74- 75 76 77 78 79 8O 8*
YEAR
83 84
A-9
-------
TABLE A.6
PADD II FCC UTILIZATION
FCC % 1970
1971
1972
1973
1974
1975
BOM Gas. Production, MB/SD
FCC Capacity, MB/SD
FCC Feed, MB/CD 32
% Utilization 32
FCC Feed, MB/CD 38
% Utilization 38
FCC Feed, MB/CD 34
.% Utilization 34
78
80
84
87
81
82
1204.4 1248.9 1309.1 1366.0 1328.9 1362.9
92
95
100
103
96
97
1077.6 1117.4 1171.3 1222.2 1188.2 1219.4
83
85
89
92
86
87
1976
1759 1824 1912 1995 1950
1304.4 1311.8 1315.6 1326.4 1387.9
1014.2 1051.7 1102.4 1150.3 1183.3 1147.7 1185.7
2012 2086
1404.8 1434.2
83
1408
98
1259.8
88
As indicated in Table A.6, percentages of FCC gasoline in the total pool as
high as 38% for PADD II as suggested by Haensel are unlikely, for the FCC yields
cannot be in error by a sufficient magnitude to make the FCC utilization figures
reasonable. The model results(Table A.l)of 32% are quite reasonable, and 34%
is recommended.
The historic FCC gas oil feed on this basis is shown as a solid line in
Figure A.3,and the projections on a 1976 prorata basis are shown as dashed lines.
Even in 1981, the utilization rises only to 89%, so little additional need for
expansions in FCC capacity is foreseen.
PADD III
Total FCC capacity (fresh feed basis) for PADD III is shown in the histo-
gram of Figure A.3and new firm capacity additions which have been announced are
shown as the dashed extension of the histogram. A strong continuing growth
in capacity is evident in PADD III, suggesting high utilization factors over
the current decade.
The yields assumed for PADD III are tabulated below. Since crude and
processing differences between 1976 and 1980 cannot account for the yield differences
between PADD II and PADD III in Table A.I,the 1985 yield point from Table A.I was
also used for 1980:
A-10
-------
TABLE A.7
PADD III FCC GASOLINE YIELDS, LV %
1970
53.5
1971
53.5
1972 1972
53.5
53.5
1974
54.3
1975
55.2
1976
56.0
1980
56.8
1981
1985
56.8 56.8
Using these yields, the percentage utilization of FCC capacity is shown
below, for several assumed percentages of FCC gasoline in the total pool:
TABLE A.8
PADD III FCC UTILIZATION
FCC % 1970
1971
1972
1973
1974
1975
1976
BOM Gasoline Production,
MB/CD
FCC Capacity, MB/SD
FCC Feed, MB /CD
% Utilization
FCC Feed, MB/CD
% Utilization
FCC Feed, MB/CD
% Utilization
FCC Feed, MB/CD
% Utilization
-
32
32
35
35
37
37
36
36
2329.0
1718.9
1393.0
81
1523.6
89
1610.7
94
1567.2
91
2473.
1818.
1479.
81
1617.
89
. 1710.
94
1664.
92
0
2
2
9
3
1
2608.0
1867.1
1559.9
84
1706.2
91
1803.7
97
1754.9
94
2692
1829
1610
88
1761
96
1861
102
1811
99
.0
.7
.2
.1
.8
.4
2631
1895
1550
82
1695
89
1792
95
1744
92
.0
.4
.3
.9
.8
.3
2729.0
1943.3
1582.0
81
1730.3
89
1829.2
94
1779.8
92
2830.0
1972.8
1617.1
82
1768.8
90
1869.8
95
1819.3
92
It would appear that the percentage FCC gasoline in the pool is between
35% and 37%, and 36% is recommended. It is noteworthy that, if the PADD II yields
were used, the utilization factor listed above at 37% would be similar to those
listed for 35%. Hence, a 2% band of uncertainty is the best that can be achieved.
The value of 6% does give high utilizations expected for the continuing growth
in capacity evidenced in Figure A.3. Finally, since the percentage of reformate
will be shown later to be the lowest of all the PAD Districts, it would be expected
that the percentage of FCC gasoline would be high, due to its octane contribution.
A-ll
-------
I
2,200
2.jODO
A
f
g"
i
Vv LftOO
A \fiUU
<&
f
>^
H
g
^
y
LL
I / r\o
UfoOC'
\,400
Figure A. 3
PADD III
FCC UNIT CAPACITY/DEMAND
r
C.fli^«?.in
i
I
I
r4
;
/
/
f"
/
/
/
/
/
V
/
s
/
s
s
.. *-; _.
" - J)E>^4^ D
, ' \
*^
>
s
s
\
V
\
»
1O II 72
14 T5 16 17 18 19 80 ©I 81 83 84- 8s
YEAR
A-12
-------
The projections for future years are shown as a dashed line on Figure A.3,
where the 1981 and 1985 gasoline demands are distributed among PAD Districts
on a 1976 prorata basis (leading to 1919 MB/CD gas oil feed in 1981 and 1787
MB/CD in 1985). The 1981 utilization factor thus becomes 87%, leading to no
expected shortage of FCC capacity in PADD III for the foreseeable future. It
is, of course, not surprising that construction plans have already been
announced which provide adequate capacity in 1980. The unusual characteristic
of the current demand projection is the maximum in absolute gasoline demand in
1980, making that total capacity adequate for all future years.
PADD IV
Total FCC capacity (fresh feed basis) for PADD IV is shown in the histo-
gram of Figure A.4; new firm capacity additions are shown as the dashed extension
of this histogram.
Since no cluster models were developed for PADD IV, it is assumed that
the yields are identical to those of PADD II, which has the most similar crude
slate. These yields were reported in Table A.5.
Since PADD IV has shown a relatively strong rate of growth of FCC capacity
over the current decade, let us tentatively assume 90% capacity utilization and
calculate the resulting percentage of FCC gasoline in the total pool. The
results of this calculation are shown in Table A.9.
TABLE A.9
TENTATIVE ESTIMATES OF PERCENTAGE FCC GASOLINE
IN PADD IV GASOLINE POOL
1970 1971 1972 1973 1974 1975 1976
FCC Capacity, MB/SD 153.5 156.6 157.9 152.3 176.5 178.1 181.2
FCC Feed <§ 90%, MB/CD 138.2 140.9 142.1 137.1 158.9 160.3 163.1
FCC Gas. Prod., MB/CD 76.7 78.2 78.9 76.1 88.6 89.9 91.8
BOM Tot. Gas., MB/CD 203.0 214.0 221.0 229.0 226.0 230.0 236.0
% FCC in Pool - 37.8 36.6 35.7 33.2 39.2 39.1 38.9
A-13
-------
200
160
I
1
^
LlJ
D IfeO
A
g
t
u
o
LL I4°
110
Figure A. 4
PADD IV
/
/
/
«
/
/
\J
FCC
f
UNIT
X
CAPACITYy
'
X
X"
DF24AND
*"^ *.
-" v
" X
X' v^
^
\
\
\
10 Ti 1Z 73
IB 16 17 78 79 8O 0\ B2.
84
A-14
-------
It is likely that the capacity utilization exceeded 90% in 1973, anal-
ogously to PADD III. Hence, the percentage FCC gasoline must exceed 33.2%
of the total pool. The capacity utilization probably approached 90% in 1972
and 1976, suggesting between 36 and 39% FCC gasoline in the total pool. Since
all other estimates from Table A.9 are on the higher side of the range, an
estimate of 38% is reasonable. Hence, the capacity utilization in PADD IV
becomesr:
TABLE A.10
PADD IV FCC CAPACITY UTILIZATION
1970 1971 1972 1973 1974 1975 1976
FCC Capacity, MB/SD 153.5 156.6 157.9 152.3 176.5 178.1 181.2
BOM Gas. Prod., MB/CD 203.0 214.0 221.0 229.0 226.0 230.0 236.0
FCC Feed, MB/CD 139.0 146.5 151.3 156.8 153.9 155.8 159.3
% Utilization 91 94 96 103 87 87 88
Although 38% would thus appear reasonable after 1974, it is indicated that
.the percentage in the pool drifted below this level due to shortages in FCC
capacity in 1972 and 1973; overall averages are inadequate for such a small
number of FCC units as are present in PADD IV.
The solid line of Figure A.4, then, represents the estimated FCC feed rate
for PADD IV, taken from Table A.10 except for 1973, which was calculated assum-
ing 96% utilization. The prorata projections for 1981 and 1985 are shown as
dashed lines in Figure A.4. It is not immediately apparent why the 17 MB/D
capacity increment is needed (Little America Refining Co., Casper, Wyo.). Since
it is reported that construction is to be completed in 1978, verification of
this expansion would be straight-forward. However, such verification was not
attempted due to the lack of importance to the overall U.S. balance. Further-
more, whether the expansion is completed or not, FCC capacity will be adequate
for the foreseeable future. Finally, if this capacity is installed and run at
90% uitlization, it would increase the percentage FCC gasoline in the total
PADD IV pool only to 40%, slightly above the recommended estimate of 38%.
A-15
-------
PADD V
Total FCC capacity for PADD V is shown in the histogram of Figure A.5.
No new capacity additions have been announced. Since the total capacity has
been nearly constant at about 580 MB/SD over the current decade, it is expected
that this capacity is underutilized. Furthermore, since the reformate in the
PADD V pool will be shown later to be the highest of any PAD District in the
U.S., the low pool percentages of Table A.I are not unreasonable. It is thus
only possible to check that these percentages provide utilization factors of
less than 90% and little further adjustment can be made. For this calculation,
the following yields were interpolated from Table A.I:
TABLE A.11
PADD V FCC GASOLINEVYIELDS. LV %
1970 1971 1972 1973 1974 1975 1976 1980 1981 1985
53.0 53.0 53.0 53.0 53.1 53.1 53.2 53.8 53.8 55.0
The calculated capacity utilization thus becomes:
TABLE A.12
PADD V FCC UTILIZATION
FCC % 1970 1971 1972 1973 1974 1975 1976
FCC Capacity, MB/SD - 574.6 588.8 578.7 577.9 585.0 591.2 591.2
BOM Gas. Prod., MB/CD - 815.0 827.0 885.0 919.0 895.0 908.0 965.0
FCC Feed, MB/CD 28 430.6 436.9 467.6 485.5 471.9 478.8 507.9
% Utilization 28 75 74 81 84 81 81 86
FCC Feed, MB/CD 30 461.3 468.1 500.9 520.2 505.7 513.0 544.7
% Utilization 30 80 80 87 90 86 87 92
A-16
-------
ICO
6O0
2
f
01
\
a
a
u
8
500
4oo
300
Figure A.5
PADD V
FCC UNIT CAPACITY/DEMAND
11 71 73 74 75 76 77 78 7
-------
From these calculations, it is difficult to believe that the PADD V
percentage of FCC gasoline in the total pool could exceed 30% without prompt-
ing more capacity additions in the 1973 to 1977 period, a time of significant
additions of reforming capacity. Also, it is difficult to believe that the
percentage in the pool would be much below 28%, for the amount of underutilized
capacity would be excessive. It is concluded, therefore, that the cluster
model estimate of 28% is reasonable.
The solid lines of Figure A.3 show the FCC feed data of Table A.12,
and the dashed lines present the future projections, based upon 1.08 MMB/D
total gasoline production in 1981 and 1.03 MMB/D in 1985 (equivalent to 562.1
MB/CD FCC feed in 1981 and 524.4 MB/CD in 1985). This provides a maximum
capacity utilization of 91% in 1981, or no need for new FCC capacity other than
marginal increments to meet the specific needs of individual refiners.
PADD's I through IV
Since substantial product movement between these PAD District routinely
takes place, assessment of the overall FCC balance is warranted. The total FCC
capacity, abstracted from Tables A.4, A.6, A.8, and A.10 is shown below.
PADD
I
II
III
IV
Total
1970
622.4
1304.4
1718.9
153.5
3799.2
TABLE A.13
PADD's I-IV FCC CAPACITY, MB/SD
1971
637.6
1311.8
1818.2
156.6.
3924.2
1972
653.4
1315.6
1867.1
157.9
3994.0
1973
626.3
1326.4
1829.7
152.3
3934.7
1974
573.8
1387.9
1895.4
176.5
4033.6
1975
560.0
1404.8
1943.3
178.1
4086.2
1976
565.4
1434.2
1972.8
181.2
4153.6
Similarly, the total gas oil feed to the FCC units can be abstracted:
A-18
-------
PADD
I
II
III
IV
Total
1970
459.2
1077.6
1567.2
139.0
3243.0
TABLE A.14
PADD's I-IV FCC FEED, MB/CD
1971
484.2
1117.4
1664.1 .
146.5
1972
495.3
1171.3
1754.9
151.3
1973
519.6
1222.2
1811.4
146.2
1974
480.0
1188.2
1744.3
153.9
% Utili-
zation 85
3412.2
87
3572.8
89
3699.4
94
3566.4
88
1975
456.5
1219.4
1779.8
155.8
3611.5
88
1976
502.3
1259.8
1819.3
159.3
3740.7
90
The results are plotted in Figure A.6; the overall capacity utilization
rises only to 89% by 1981, indicating again no need for FCC unit additions
other than those required for specific situations for individual refiners.
This conclusion is not particularly surprising, for capacity announcements
have already been made which will serve the FCC capacity requirements through
1980; the gasoline projection, in turn, provides an absolute decline in gasoline
demand after this time. The obvious conclusion is that no new capacity is needed
ever, beyond current announcements; the obvious uncertainty is the gasoline
demand projection.
A.4 Catalytic Reforming Results
The cluster model output for catalytic reforming unit yields (associated
with gasoline production and not BTX production) and percentage reformate in the
gasoline pool are shown in Table A.15. Although the yield patterns are intended
to represent the effect of changing crude types through time and changing
severity with the progression of lead phase-down, the yields past 1973 appear
too low. For example, the substitution of bimetallic catalysts often allows
reforming at lower pressures. The poorest quality naphtha reformed in significant
quantities in the U.S. is Arabian Light naphtha; as shown in Table A.16, even
this naphtha gives higher yields than reported in Table A.15. The data of
A-19
-------
Figure A.6
4,500
4,000
Q
cf
UJ
₯ **
t
^
o
o
o
LL
3jCOO
2,500
*..*
PADD's I- IV
FCC UNIT CAPACITY/DEMAND
/
'
/
f
/
A
-
x--
y
s
<
''
( ~ ~ ~_
^ ' . ^ DEM40A
X
X
.
70 II 11 13 14 IS 16 17 IB 19 80 S( 81 83 84 85
YEAR
A-20
-------
TABLE A.15
CLUSTER MODEL OUTPUT SUMMARY FOR
GASOLINE-PRODUCING CATALYTIC REFORMING UNITS
PADD I
Reformer Yield, %
Reformate in Total Pool, %
Refomer Severity, RON
PADD II
Reformer Yield, %
Reformate in Total Pool, %
Reformer Severity, RON(D
PADD III
Reformer Yield, %
Reformate in Total Pool. %
Reformer Severity, RON(*)
PADD I - III
Reformer Yield, %
Reformate in Total Pool, %
PADD V
Reformer Yield, %
Reformate in Total Pool, %
Reformer Severity, RON
TOTAL U. S.
1973
84.7
29.3
95.4
88.8
27.4
90.0/91.4
1977
81.3
33.0
97.0
1980
77.5
30.9
98.0
83.7 77.7
27.6 25.0
96.5/90.7 100/98.7
87.8 76.9 75.6
25.6 24.2 25.5
90.0/90.0 95.1/99.3 100/99.7
Reformer Yield, %
Reformate in Total Pool,
87.7
26.8
83.9
33.1
92.6
87.0
27.7
80.1
26.7
81.5
26.1
93.8
80.4
26.7
76.7
26.1
79.4
24.4
96.9
77.1
25.9
1981
74.3
30.9
100.0
75.1
27.5
100/100
74.9
27.6
100/100
74.9
28.1
74.1
26.2
100
74.7
27.8
(1)
(2)
The first entry refers to the Large Midwest Cluster and the second
entry to the Small Midcontinent Cluster
The first entry refers to the Lousiana Gulf Cluster and the second
entry to the Texas Gulf Cluster
A-21
-------
TYPICAL LOW
160/380°F
Severity, RON
C + Yield, %
TABLE A. 16
PRESSURE REFORMING YIELDS
ARABIAN LIGHT NAPHTHA
90 100
82.5 77.0
._ ,._..,._ .. . . , , _
TABLE A. 17
PERCENTAGE OF REFORMING UNITS
CONTAINING BIMETALLIC CATALYSTS
U. S. AVERAGE
PADD I
Pennsylvania
New Jersey
PADD II
Illinois
Indiana
Ohio
Oklahoma
PADD III
Texas
Louisiana
PADD IV
Wyoming
Montana
PADD V
California
Washington
SOURCE: Oil & Gas
1972
25.7
37.1
23.6
34.3
28.8
28.4
15.5
10.9
86.6
18.1
11.9
20.7
28.8
0
56.7
32.1
34.6
14.5
Journal
1973
34.3
43.3
25.9
35.9
47.1
43.7
54.5
17.7
87.6
26.5
24.6
19.3
40.8
17.7
55.5
26.6
28.2
16.4
1974
39.7
45.3
32.6
37.3
46.8
37.8
76.5
15.8
88.1
30.4
25.7
32.4
42.4
29.7
51.2
48.0
45.6
62.4
1975
43.6
46.9
29.5
53.8
49.7
55.3
60.3
9.7
89.9
34.5
31.9
32.5
50.6
34.2
59.8
53.7
50.2
73.9
1976
49.9
49.9
29.2
63.5
52.3
53.3
66.6
16.6
89.5
42.8
45.5
27.5
63.3
51.6
65.9
61.0
57.4
72.9
1977
56.1
53.4
37.7
61.9
53.0
55.3
63.4
16.3
84.9
52.6
55.5
35.4
56.0
38.0
64.3
70.9
69.1
75.9
A-22
-------
Table A.17 indicate that substitution of bimetallic catalysts is continuing
rapidly, and will probably continue for several years. Furthermore, discus-
sions with a major PADD V refiner indicates their yields expected to remain
in the 80 to 85% range for the next five years, due to bimetallic catalysts
and high quality North Slope crude. Although this problem with reformer
yields was recognized in the lead phase-down study and parametric runs were
executed to confirm that the overall study results were not greatly influenced
by this factor, adjustments in the yields of Table A.15 will be required for
the present study.
The percentage of reformate in the gasoline pool of Table A.15 varies
substantially from year-to-year, probably for the same reasons as discussed
for the FCC unit. The figures are substantially below that reported by Haensel,
which indicated reformate was 33% of the U.S. pool. Furthermore, in 1976, the
PADD V gasoline production from the Bureau of Mines was 965 MB/CD. In order
to bracket the possible ranges of reformer charge stock associated with this
gasoline production, let us assume the figures of Table A.15 are applicable
for PADD V. Assuming the figures for 1973, 1977 and 1980 are, in turn, appli-
cable to 1976, the following PADD V reformer feed rates can be calculated:
TABLE A.18
1976 REFORMER FEED RATE, MB/D,
BASED UPON MODEL RESULTS OF;
1973 1977 1980
381 309 297
Since the 1976 reformer capacity in PADD V was 605 MB/SD and only 40 to
60 MB/SD was dedicated to BTX production, these reformer charge rates are
obviously too low. Indeed, a simple comparison of gasoline reforming capacity
(about 550 MB/SD) to gasoline production (965 MB/CD) would indicate that the
percentage reformate in the PADD V pool from the cluster model must be nearly
50%, instead of the 25 to 35% cluster model result of Table A.15.
A-23
-------
The following subsections, therefore, report improved estimates of the
reformer yields and percentage of reformate in the gasoline pool.
PADD I
Total historic reformer capacity for PADD I is shown in Figure A.7. Based
upon aromatics production levels, it is estimated that in 1976 the BTX reformer
capacity was 95 MB/SD. BTX reformer capacity'in earlier years was taken to
be in proportion to total reformer capacity, since the error involved in this
approach probably does not exceed 20 MB/CD. In later years, it was assumed
to be constant.
The PADD I reformer yields and percentage reformate in the gasoline pool
from the lead phase-down model study were shown in Table A.15.
This decline in yield is due to increasing reforming severity and poorer
crude quality. For reference, the reforming yield on Arabian Light at 100 RON
and 225 psi is 77% and on Alaskan is 83.6%. Since PADD I has about 53.4% bimetal-
lic catalysts in 1977, it is likely that no more than one-half of the units
have low pressure operating capability. Also, over 50% of PADD I crude has yield
performance similar to Alaskan North Slope. Hence, it is felt that the yield
decline of Table A.15 is too severe, and the following yields were used:
TABLE A.19
PADD I REFORMING YIELDS^ LV %
1970 1971 1972 1973 1974 1975 1976 1980 1981 1985
84.7 84.7 84.7 84.7 83.9 83.0 82.2 80.0 80.0 80.0
To evaluate possible percentages of reformate in the PADD I gasoline pool,
two levels were selected from Table A.15 for consideration, one from PADD I
and one from the PADD I through III composite:
A-24
-------
Figure A.7
PADD I
REFORMER CAPACITY/DEMAND
100
70 71 72 73 74 75 76 77 78 79 8O 81 81 85 84 85
YEAR
A-25
-------
TABLE A.20
PADD I REFORMER UTILIZATION
Reformate % 1970 1971 1972 1973 1974 1975 1976
Reformer Capacity, MB/SD - 260 258 288 290 290 308 325
BOM Gas. Production, MB/CD - 661 697 713 748 702 680 760
Naphtha Feed, MB/CD 26.8 209 221 226 237 224 220 248
% Utilization 26.8 80 85 78 82 77 71 76
Naphtha Feed, MB/CD 30.0 234 247 253 265 251 246 277
% Utilization 30.0 90 96 88 91 87 80 85
As shown in Figure A. 7, continuous additions of PADD I reforming capacity are
apparent over the present decade. If the reformate percentage in the pool were
as low as 26.8%, the capacity utilization would be so low that these new additions
would not be needed. By contrast, use of 30% reformate in the pool provides very
reasonable capacity utilizations, so a figure of 30% should be adopted for the
present study, with the yields of Table A. 19.
The solid line in Figure A. 7 shows the demand for gasoline reformer capacity,
expressed as naphtha feed in MB/CD. Demand will be limited when it reaches 90 to
93% of the stream day capacity. The future demand projection for gasoline in
PADD's I - IV is 6.325 MMB/CD in 1981 and 5.89 MMB/CD in 1985. If this were
distributed among the PAD Districts in proportion to 1976 production, the dashed
projection line in Figure A.7 would be obtained.lt is apparent that, under these
conditions, there is no need for additional gasoline reforming capacity in PADD I
for the foreseeable future, for calendar day demand reaches only 92% of stream
day capacity in 1981.
PADD II
Total historic reformer capacity for PADD II is shown in Figure A.8. In 1976,
the PADD II BTX reformer capacity was estimated to be 105 MB/CD. BTX capacity, in
earlier years, was taken to be proportional to total reformer capacity and, in
later years, was taken to be constant.
A-26
-------
The PADD II reformer yields from the model runs are shown in Table A.15.
The PADD II percentage of bimetallic catalyst was 53.0% in 1977, indicating
that at about half of the units may be operable at lower pressure, thereby improv-
ing yields. Also, the crude slate should not become appreciably poorer than used
for PADD I. Hence, the following reforming yields are reasonable:
TABLE A.21
PADD II REFORMING YIELDS, LV %
1970
88.8
1971
88.8
1972
88.8
1973
88.8
1974
87.1
1975
85.4
1976
83.7
1980
80.0
1985
80.0
The PADD II percentage reformate in the pool is shown in Table A.15 to be
about 27.5%. Since the national average has been reported to be about 33% and
since PADD II is one of the major gasoline producers, three levels of reformate
in the .pool were evaluated, as shown below:
TABLE A.22
PADD II REFORMER UTILIZATION
Reformer Capacity, MB/SD
BOM Gasoline Production, MB/CD
Naphtha Feed, MB/CD
% Utilization
Naphtha Feed, MB/CD
% Utilization
Naphtha Feed, MB/CD
% Utilization
Reformate %
:D
27.5%
27.5%
30 %
30 %
33 %
33 %
1970
625
1759
545
87
594
95
654
105
1971
695
1824
565
81
616
89
678
98
1972
715
1912
592
83
646
90
711
99
1973 1974 1975 1976
775
1995
618
80
674
87
741
96
803
1950
616
77
672
84
739
92
840
2012
648
77
707
84
778
93
870
2086
685
79
748
86
822
95
A-27
-------
Again, with the growth in PADD II reforming capacity observed in this decade,
it is probable that the utilization approximated 90% and hence, the percentage
reformate in the pool is about 30%. The solid demand line in Figure A.8 repre-
sents the naphtha feed rate at this level.
If the 1981 and 1985 gasoline demand is distributed among the PAD Districts
in accordance with 1976 production, the dashed line in Figure A.8 is obtained.
Hence, assuming that PADD II supplies no more gasoline to other PAD Districts
than its historic proportion, reformer capacity will become tight around 1980.
However, since the projections indicate that this need for capacity is only
transitory, it will likely be met by debottlenecking or imports and transfers
from other PAD Districts, rather than by a major reformer expansion. Of
course, an individual refiner may become short of capacity, even though other
refiners have ample capacity, leading to individual cases possibly deviating
from this generalization.
PADD III
Total historic refining capacity for PADD III is shown in Figure A.9.
It was estimated that, in 1976, the BTX reforming capacity was 775 MB/SD and
that, in addition, 100 MB/CD of by-product heavy reformate enters the gasoline
pool from locations having only BTX reformers. Hence, this reformate must
be deducted from the gasoline pool reformate before evaluating the contribution
of gasoline reformer. On Figure A.9, the 775 MB/SD of BTX capacity is shown
for 1976, prorated on total capacity in prior years and held constant in later
years. It is apparent that the precise definition of BTX capacity is more
critical for PADD III, so the results of the analysis will probably be less
accurate for this PAD District. As for PADD II, the percentage of bimetallic
catalyst is 52.6%, indicating significant potential for further substitution.
Since the crude slate and operating severity will, in the long run, be generally
similar between the districts, similar yields to Table A.21 are taken from
1976 through 1985. The yields used, therefore, were:
TABLE A.23
PADD III REFORMING YIELDS, LV %
1970 1971 1972 1973 1974 1975 1976 1980 1981 1985
87.8 87.8 87.8 87.8 86.4 85.1 83.7 80.0 80.0 80.0
A-28
-------
Figure A.8
PADD II
REFORMER CAPACITY/DEMAND
7O -7| 72 73 74 75 76 17 76 79 3O 61 82L
-------
Figure A.9
PADD III
REFORMER CAPACITY/DEMAND
1600
\600
§
4.
ftr
e
ul
600
600
400
1O 11
73 74. 15 76 77 78 79 SO 61 BZ 83 64 S&
YEAR
A-30
-------
The percentage reformate in the gasoline pool is shown in Table A. 15 to approxi-
mately 25.5% for PADD III, based upon the cluster model runs. As indicated in
Table A.24, reasonable capacity utilizations are obtained with this model result:
TABLE A. 24
PADD III REFORMER
Reformer
Reformate %
Capacity, MB/SD
BOM Gasoline Production, MB/CD
Naphtha
%
Feed, MB/CD 25.5
Utilization 25.5
1970
644
2329
563
. 87
UTILIZATION
1971
653
2473
604
93
1972
707
2608
644
91
1973
722
2692
668
93
1974
740
2631
661
89
1975
758
2729
700
92
1976
764
2830
743
97
Apparently, the reformate percentage in the PADD III pool is markedly lower than
the other PAD Districts because of the substantial BTX production level. Hence,
in PADD III, a gasoline pool comprised of 25% reformate is recommended. The
required naphtha feed for this case is shown in Figure A.3 as a solid line. With
future gasoline demand prorated on 1976 production levels by PAD District, the
dashed projection of Figure A.9 is obtained. As with PADD .II, capacity will be tight
around 1980, but there is no long term need for new reformer expansions beyond
those already announced.
PADD IV
The historic reformer capacity in PADD IV is shown in Figure A.10. There is no
BTX capacity in PADD IV. Since PADD IV was not simulated in the cluster model,
reformer yields are assumed to be equal to those of PADD II (Table A.21), which has
the crude slate most closely approximating that of PADD II.
If a 90% utilization factor is assumed, because of the steady addition of
reforming capacity in PADD IV, the following percentages of reformate in the pool
are determined:
A-31
-------
IZO
110
90
uJ
a
70
60
50
X
Figure AilO
PADD IV
REFORMER CAPACITY/DEMAND
X
Dti^AdO
10 7( 72 13 "74- 15 76 77 78 7<* 8O 81 81 83 84 65
A-32
-------
TABLE A. 25
PADD IV REFORMATS IN GASOLINE POOL
1970 1971 1972 1973 1974 1975 1976
Reformer Capacity, MB/SD 77 82 88 91 98 103 110
Reformate Produced, MB/CD 61.5 65.5 70.3 72.7 76.8 79.2 82.9
BOM Gas. Production, MB/CD 203 214 221 229 226 230 236
% Reformate in Pool 30 31 32 32 34 34 35
The reformate production, and hence the percent reformate in the pool, in 1974 and
1975 are probably overstated, because reformer capacity was generally underutilized
due to lower gasoline demand than anticipated. Therefore, a 31% reformate level
in the pool was assumed for PADD IV, resulting in the following reformer capacity
utilization:
TABLE A.26
PADD IV REFORMER UTILIZATION
1970 1971 1972 1973 1974 1975 1976
Reformer Capacity, MB/CD 77 82 88 91 98 103 110
Naphtha Feed, MB/SD 70.9 74.7 77.2 79.9 80.4 83.5 87.4
% Utilization 92 91 88 88 82 81 79
Prorating 1981 and 1985 demand based upon 1976 production levels gives the
dashed projection in Figure A.10. Unless demand growth occurs preferentially on the
small base for PADD IV relative to other PAD Districts, additional reforming
capacity will not be required.
PADD V
As noted earlier, PADD V reforming capacity suggests substantial deviation
from the cluster model percentages of reformate in the gasoline pool. Consequently
discussions were initiated with selected PADD V refiners. They indicated that the
percentage reformate in the pool is about 45% and that the reformer yields are in
A-33
-------
the 80 to 85% range. Furthermore, they indicated that the yields are not
expected to decline substantially through the 1980's, because improved feedstock
quality and lower reformer pressure will offset the higher severity operation for
lead-free gasoline. They felt the 1985 reformer yield from the cluster model
runs, 74.1%, was substantially too low. They also confirmed the ADL estimate
of 40 - 60 MB/D of BTX reformer capacity. They cautioned that the Oil and Gas
Journal reformer capacity was too low, in that the capacity figures do not reflect
recent and potential debottlenecking capacity. Finally, it was indicated that
substantial amounts of 130 - 180°F naphtha is fed to PADD V reformers, as well as
the more traditional 180 - 400°F naphtha.
Taking 50 MB/SD as BTX capacity in 1976, keeping this figure constant in
years after 1976, and ratioing it to total capacity in years before 1976, the
PADD V gasoline capacity is shown in the histogram of Figure A. 11.
Historic gasoline production from the Bureau of Mines and projected gasoline
production in PADD V is shown below:
TABLE A.27
PADD V GASOLINE PRODUCTION, MMB/CD
1970 1971 1972 1973 1974 1975 1976 1980 1985
0.815 0.827 0.885 0.919 0.895 0.908 0.965 1.08 1.03
Taking the average PADD V reformer yields to be as given by the model in 1973 and
1977, and to decline no further after 1977 gives:
TABLE A.28
REFORMER YIELDS. LV %
1970 1971 1972 1973 1974 1975 1976 1980 1985
83.9 83.9 83.9 83.9 83.3 82.7 82.1 81.5 81.5
A-34
-------
Constant yields were assumed from 1970 - 1973 because of relatively constant
crude quality and pre-lead phase-down. Constant yields were also assumed post-1977
due to the trade-offs between increasing crude quality and lower reformer pressure
versus higher severity reforming.
With these figures, utilization of PADD V gasoline reformers can be determined:
TABLE A.29
PADD V REFORMER UTILIZATION
Reformate 5
-
-
40
40
43
43
45
45
I 1970
383
815
388.6
101
417.7
109
437.1
114
1971
422
827
394.3
93
423.9
100
443.6
105
1972
493
885
421.9
86
453.6
92
474.7
96
1973
500
919
438.1
88
471.0
94
492.9
99
1974
503
895
429.8
85
462.0
92
483.5
96
1975
502
908
439.2
87
472.1
94
494.1
98
1976
554
965
470.2
85
505.4
91
528.9
95
Reformer Capacity, MB/SD
BOM Gas. Production, MB/CD
Naphtha Feed, MB/CD
% Utilization
Naphtha Feed, MB/CD
% Utilization
Naphtha Feed, MB/CD
% Utilization
An average percentage reformate in the PADD V gasoline pool during the current
decade is about 42% to 43%, and 43% is used in the present study. The percentage
reformate in the pool was probably lower in 1970 and 1971, due to the exclusion of
light naphtha from the reformer feedstock. If 93% utilization were assumed for these
years, the percentage reformate in the pool would have been 37% and 40% for 1970
and 1971, respectively. In any event, more precise estimates are not possible
of the reformate percentage in the gasoline pool.
The solid line of Figure A.11 represents the naphtha feed rate to PADD V
gasoline reformers, taken from Table 29 except for 1970 and 1971, which was assessed
at 93% utilization. The dashed line of Figure A.11 represents the anticipated
reformer demand for future years, assuming 43% reformate in the total pool. Although
reforming capacity will be tight in the late 1970's, it is more likely to be met by
minor debottlenecking, imports, or temporarily diminished percentages of reformate
in the pool (with octanes provided by FCC gasoline or alkylate or slight octane
A-35
-------
Figure A.11
600
Q
^
a
400
UJ
LU
30O
250
PADD V
REFORMER CAPACITY/DEMAND
7O 11 12. 13 74 15 76 77 18 79 8O 61 8^ 8*5 64 65
A-36
-------
erosion in the finished gasoline) rather than general capacity expansion. Hence, the
outlook for long-term strength in naphtha/gasoline margins in PADD V is unfavorable.
PADD's I - IV
Since substantial product movement between these PAD Districts routinely
takes place, assessment of the overall reformer balance is warranted. The total
gasoline reforming capacity, abstracted from Tables A.20, A.22, A.24, and A.26 is
shown below:
TABLE A.30
PADD's I-IV REFORMER CAPACITY, MB/SD
PADD 1970 1971 1972 1973 1974 1975 1976
I 260 258 288 290 290 308 325
II 625 695 715 775 803 840 870
III 644 653 707 722 740 758 764
IV 77 82 88 91 98 103 110
TOTAL - 1606 1688 1798 1898 1931 2009 2069
Similarly, the naphtha charge for gasoline production can be abstracted from these
tables:
TABLE A.31
PADD's I-IV NAPHTHA REFORMER FEED, MB/CD
PADD 1970 1971 1972 1973 1974 1975 1976
I 234 247 253 265 251 246 277
II 594 616 646 674 672 707 748
III 552 592 631 655 648 686 728
IV 71 75 77 80 80 84 87
1840
UTIL. - 90 91 89 88 85 86 89
1970
234
594
552
71
1451
90
1971
247
616
592
75
1530
91
1972
253
646
631
77
1607
89
A-37
1973
265
674
655
80
1674
88
1974
251
672
648
80
1651
85
1975
246
707
686
84
1723
86
-------
The results are plotted in Figure A.12; although capacity will be tight
in 1980, there is no significant need for reforming capacity other than unique
situations for individual refiners.
A-38
-------
Figure A.12
PADD's I-IV
REFORMER CAPACITY/DEMAND
CAPACITY
70 ' 7 I ' 72 ' 75 74 75 76TJ
-------
APPENDIX B
RANGE OF CONTENT OF GASOLINE
COMPONENT STREAMS
On October 3, 1977, Arthur D. Little, Inc., met with representatives of the
EPA, API, NPRA and oil industry to discuss benzene removal from gasoline.
At that meeting, Figure B.I was designed for the data information needs of
the benzene removal from gasoline study. Through the efforts of the Benzene
Task Force of the API and the NPRA, the data request was sent to 34 U. S. refin-
eries, as shown in Table B.I.
All refineries contacted responded to the API and NPRA questionaires and
were quite cooperative with follow-up discussions of their submissions. Based
on discussions with the individual refiners, the benzene component data were
accumulated according to the blend component designations developed in Chapter 2.
The data were coded to maintain confidentiality of individual refinery inputs.
The coded benzene survey data are presented in Table B.2.
As can be seen from the data in Table B.2, the benzene content data sub-
mitted by the refiners indicate a considerable range of possible benzene content
for most components. Variations in feedstock quality, processing configuration,
processing severity or special blending requirements can account for this
range. Through our discussions with the various refiners, we were able to sort
out most of these differences and arrive at a reasonable assessment of benzene
content of each of our blend components, as shown in Table B.2. We estimated
the U. S. pool benzene content as 1.30 volume %. This figure is based on the
1977 benzene content data from Table B.2 and our projected gasoline pool com-
position from Table 2.10 in Chapter 2. The estimated U. S. pool content compares
favorably with the available data on current pool content shown in Table B.2
in Chapter 3 and falls within our projected pool content range of 1.0% to
1.5 volume %.
B-l
-------
Minencan i*etroi«um insiiiuie
2101 L Street Northwest
Washington, D.C. 20037
202-457-7000
EDWARD P. CROCKETT
(202)457-7084
October 5, 1977
Dear t . . ; ,. '.; ;;
. - - '.. <*'
You are aware, I believe, that the Environmental Protection :
Agency has commissioned Arthur D. Little to evaluate.the impact <...!
on the U. S. refining industry of reducing benzene levels in the
U. S. gasoline pool. EPA is considering this as an alternative
to vapor recovery as a means of reducing benzene levels in the
ambient air.
Representatives from the Environmental Affairs Department's
Stationary Source and Economics Committees, EPA, and the Arthur
D. Little (ADL) case team met recently to discuss this study. A
copy of the ADL Technical Proposal is enclosed. They have a
period of four months to complete the study.
A major portion of this work involves the assessment of the
likely benzene content of the U. S. gasoline pool. Current
information in this area is limited. There is a range of report- .
ed benzene contents in gasoline but little specific data on :
typical current benzene levels. v
In order that this study be based on the best current infor- ;
mation available, we request your assistance in providing infor- : '
mation on the typical benzene content of your gasoline pool and
gasoline blending components. The data requested is to be "typical*
as it is not intended that extensive effort be made to compile
data from each refinery. This data will be used to develop
typical levels of benzene in gasoline on a regional basis for*
scale up to the U. S. pool. Results of the study will not include
data on a refinery-specific basis but on a combined regional
basis.
An equal opportunity employer B *
-------
- 2 -
Attached is a copy of the data requested for this study and
a list of all refiners and refineries to be surveyed. You will
note that a company contact is requested on the form. This
individual would be contacted in the event that data from one
source appears to be significantly disparate from the typical
data from the other refineries. Possible errors can be checked
or reasons for the variation determined through individual follow-
up. Due to the short period of time ADL has to complete this
study, your best estimate of current benzene in gasoline levels
is requested by October 31, 1977. Replies should be directed to
me with a copy to: .<.'
John R. Felten
Arthur D. Little, Inc.
35 Acorn Park
Cambridge, Mass. 02140
(617)864-5770 x 3108
I appreciate your assistance in providing this information
for this important study.
Cordially,
Edward P. Crockett
EPCrmvt
Enclosures
B-3
-------
FIGURE: B.I
INFORMATION NEEDS FOR BENZENE
REMOVAL FROM GASOLINE STUDY
Company
Company Contact:
Name
Title
Address
Telephone No.
Refinery;
Gasoline Pool:
Typical Current Benzene Content: Vol. %
Range of Benzene Content: Vol. %
Gasoline Blending Components: (List all components)
»
Typical Current Benzene Content: Vol. %
Range of Benzene Content: Vol. %
B-4
-------
TABLE B.1
REFINERIES SURVEYED FOR GASOLINE BENZENE DATA
PAD District
I
I
I
I
II
II
II
II
II
II
II
II
III
III
III
III
III
III
III
III
III
III
III
III
III
III
V
V
V
V
V
V
V
V
Refiner
Arco
Exxon
Gulf
Witco Chemical
Amoco
Amoco
Delta Refining
Gulf
Indiana Farm Bureau
Mobil
Shell
Union
Amoco
Arco
Chevron
Exxon
Exxon
Gulf
Gulf
Louisiana Gloria
Marion Co.
Mobil
Shell
Shell
South Hampton
Union
Arco
Beacon
Chevron
Chevron
Mobil
Petrochem
Union
U.S. Oil & Rfg.
Location
Philadelphia, PA
Bayway, NJ
Philadelphia, PA
Bradford, PA
Sugar Creek, MO
Whiting, IND
Memphis, TENN
Toledo, OH
Mt. Vernon, IND
Joliet, ILL
Wood River, ILL
Lemont, ILL
Texas City, TX
Houston, TX
Pascagoula, MS
Baton Rouge, LA
Baytown, TX
Belle Chasse, LA
Port Arthur, TX
Tyler, TX
Theodore, AL
Beaumont, TX
Houston, TX
Norco, LA
Silsbee, TX
Beaumont, TX
Carson, CA
Hanford, CA
El Sequndo, CA
Richmond, CA
Torrance, CA
Ventura, CA
Los Angeles, CA
Tacoma, WASH
B-5
-------
0)
Code No.:
Reformats
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Nat. Gasoline
Lt. Hydrocrackate
Isomerate
S. R. Gasoline
Pool
Pool Range
TABLE B-..2
A.P.I./NPRA REFINERY GASOLINE SURVEY
ADL BLEND BENZENE CONTENT: VOL. %
10
0.8
0.6
0
0.1
0
3.9
1.0
-
1.3
0.9
0.7-1.5
1.8
0.3
0
0
0
-
2.1
0.3
2.1
1.0
0.2-2.5
1.6 Ov '
1.2 1.1
0 0
0 <0.1
0 0
-
1.4
1.4 <0.1^ '
1.4 0.5
1.2-1.6 0.4-0.8
3.8 2.7
0.9 0.5
0.6(1) <0.1
0
0 0
0.9
11.4(1) 1.0
-
_ _
2.0 0.6
0.4-3.1 0.5-1.0
4.6
0.8
0
0
0
-
0.9
-
2.1
1.8
1.2-2.5
0.7VA' 1.4V" 3.3
(2)
0.2 1.3 . 0.8
<0.1 - 0
<0.1 - 0
<0.1 - 0
_ _ -
0.4 - 1.4
0.9
0.8 0.6(2) 1.1
0.35 1.5 0.6
0.3-0.4 1.3-1.8 0.6-1.2
(1)
(2)
Excluded from U. S. average as a typical
Excluded from U. S. average due to incomplete data
-------
TABLE B.2(Cont.)
A.P.I./NPRA REFINERY GASOLINE SURVEY
ADL BLEND BENZENE CONTENT; VOL. % (Cont.)
Code No.
11
12
13
14
15
16
17
18
19
(1)
Excluded from U. S. average as a typical
Includes pyrolysls gasoline which is normally extracted
20
Reformate
FCC Gasoline
Alkylate
Raffinate
Butanes
Coker Gasoline
Nat. Gasoline
i Lt. Hydrocrackate
Isomerate
S. R. Gasoline
Pool
Pool Range
6.5 0.8(1) 1.2 4.5 4.3
0.4 0.6 0.3 1.6 0.5
0 0 <0.1 <0.1 0
0.3 0.2 1.9(1) 0.3
00-00
- 0.7 - 18. 7 -
2.5 - 1.3
1.0 - - 1.5 -
_
2.5 0.4 - 1.2 3.0
0.7 0.8 1.0 2.5 1.1
0.3-1.6 0.7-0.9 0.5-1.6 0.9-4.0(3)
1.3 4.3 1.4 1.4 2.2
0.8 1.2 1.2 0.9 0.9
0 <0.1 000
0.2 - 1.1(1) 0.4
00000
0.3
, (1)
2.7 - - 3V '
1.9 -
-----
0.9 2.0 0.4 1.5
1.1 1.5 0.9 0.8 1.3
0.7-1.9 0.2-2.0 0.7-1.4 - 0.4-1.8
-------
TABLE 'B-2 (Cont.)
A.P.I./NPRA REFINERY GASOLINE SURVEY
ADL BLEND BENZENE CONTENT: VOL. % (ContO
(2)
.(2)
tri
Reformate
FCC Gaso
Alkylate
Raffinate
Butanes
Coker G
Nat. Ga
Lt. Hyd
Isomerate
S. R. i
Pool
Pool Range
Code No . :
1.0
.ine 1.0
0
0
ioline 1.0
dine 0.5
icrackate 0.5
oline 0.5
0.8
;e 0.2-2.5
22 . 2J 24
2.0 2.1(2) 0.5(1)
0.7 0.9(2) 1.0
0 - 0
- - 0
0 - -
1.5 - 2.6
0.5 1.7(2) 1.1
1.5 - 1.0
1.5 0.9 1.0
0.2-2.5 0.6-1.3
25
1 2.0
1.0
0
0
1.8
1.0
1.1
0.8-2.0
26 x" 27 x"' 2_8
2.1 2.7
1.0
0
0
0.8 2.0
2.4 1.39 1.75
1.0-2.4 1.6-1.8
29 30
0.9 1.8
0
0.2
0.5
0.6 1.8
0.8 1.3
. 0.6-1.0 1.26-1
(2)
(1)
(2)
Excluded from U. S. average as non-typical
Excluded from U. S. average due to incomplete data
-------
VO
TABLE B-2 (Cont.)
A.P.I./NPRA REFINERY GASOLINE SURVEY
ADL BLEND BENZENE CONTENT: VOL. % (Cont.)
(2) (2)
oi \f- 1 1\') "1 "4 "*A
Reformate 2.8 0.9 5.0 2.5
FCC Gasoline 0.5 - - 1.6
Alkylate 0
Raf f inate
Butanes - - 0 0
Coker Gasoline - - - -
Nat. Gasoline - - - -
Lt. Hydrocrackate -
Isomerate -
S. R. Gasoline - 0.3 0.5 0.1
Pool 1.6 0.8 3.4 1.2
Pool Range - 0.7-0.9 3.0-4.0 1.0-1.4
Reported
Average
2.8
0.8
0
0.2
0
1.4
1.5
1.1
0.4(4)
1.4
1.25
0.8-1.8
Reported
Range
0.5-10.0
0.2- 2.5
0
0 - 1.0
0
0.2- 2.5
0.1- 3.5.
0.5- 2.0
0 - 1.0
0.5- 3.0
0.6- 2.5
0.2- 4.0
(2)
(4)
Excluded from U. S. average due to incomplete data
Based on 56% C5 ISOM capacity/44% C, ISOM capacity with estimate of 0% benzene in C,. Isomerate
A« J 1 °/ 1. A« » n« A -? « /~* T *«. A_*. . u 4- . '
and 1% benzene in C., Isomerate
o
-------
APPENDIX C
ECONOMICS OF BENZENE REMOVAL
FROM REFORMATES AND FCC GASOLINE
APPENDIX C.1
BENZENE REMOVAL STUDY - UTILITY COSTS
Pricing Basis: September 1977 Gulf Coast
Fuel
0.5% Sulfur No. 6 Fuel 6,000 MBtu/B
Steam
1,275 Btu////.85 efficiency*
Steam Fuel Cost - 1,500 MBtu/M// x $2.00/MMBtu
Electricity: (0.5 Kwhr/M//) ($0.025/Kwhr)
Boiler F.W. : (1 M# BFW/M// Stm) ($0.07/M#) @ 60°F
Other Variable Costs: (Maintenance, labor, etc.)
600 // Stm. , 640°F*
Power
Purchased Power (Fuel @ $2.00/MMBtu)
Energy Requirement: 10,000 Btu/Kwhr
Power Fuel Cost - 10,000 Btu x $2.00/MMBtu
Cooling Water
Fuel: 0.008 MMBtu/Mgal x $2.00/MMBtu
Electricity: 0.4 Kwhr/Mgal x $0.025/Kwhr
Other: Chemical, etc.
Total
Energy Costs
Fuel:
Steam:
Power:
Cooling Water:
Unit
FOE B
M#
Fuel
Electricity (0.5 Kwh/M#)
Boiler F. W. (60°F)
Total
Kwhr
MGallons
Fuel
Electricity (0.4 Kwhr/M#)
$ 12.00/B, or $2.00/MMBtu
1,500 Btu/#
$ 3.00/M//
$ 0.01/M//
$ 0.07/M#
$ 0.02/M$
$
$
$
3.10/M//
0.025/Kwhr
0.020/Kwhr
0.016
0.010
0.004
$ 0.030/Mgal
Btu/Unit
6,000,000
1,500,000
5,000
1,440
Total
C-l
1, 506,440
10,000
8,000
4,000
12,000
$/Unit
12.00
3.000
0.010
0.010
3.020
0.020
0.016
0.008
0.024
-------
APPENDIX C.2
REFINERY HYDROGEN MANUFACTURING COSTS
The cost of manufacturing hydrogen was based on data obtained from published
literature and three major manufacturers;
Foster-Wheeler Corporation, Livingston, New Jersey
C & I/Girdler, Louisville, Kentucky
Howe Baker, Tyler, Texas
Cost data was obtained for hydrogen manufacturing capacities from 100 MSCF/
day to 10 MMSCF/day. Investment costs varied exponentially with unit capacity,
whereas variable costs were directly proportional with unit capacity. Labor
requirements were constant at one man per shift for all hydrogen plants.
s
The cost bases were as follows:
Investment Costs
yi2 - eve/-5
(C) Capacity (I) Investment*
MMSCF/Day $ Millions
0.100 0.6
1.000 1.0
5.000 3.9
10.000 ' 5.5
*1977 battery limits installed plant
C-2
-------
Variable Costs
Naphtha (Feed plus Fuel) B
Electricity (KWh)
Cooling Water/Boiler Feed
Water (M Gallons)
Export Steam (M Ibs)
Catalyst & Chemicals
Total Variable Cost
Usage
Per MCF
0.098
1.0
0.4
(0.062)
$
$
$.
$
Price
14.00/B
0.025/KWh
0.030/M Gal
3.10 /M#
Cost
$/MCF
1.37
0.03
0.01
(0.19)
0.01
1.23
Capital charge factors were calculated on the same basis as for plant
investment for facilities to remove benzene from reformates and FCC gasoline.
A sample calculation of the hydrogen plant hydrogen cost for required
benzene removal from a 30,000 B/SD FCC gasoline are as follows:
Hydrogen Costs: $/MCF
FCC Gasoline: B/SD
H2 Required: MSCF/D
Process Investment: M$
Offsites @ 40%
Total Plant: M$
IDC @ 19%
Start-up @ 5%
Working Capital
Total Capital: M$
30,000
2,475
2,740
1,096
3,836
728
192
17
4,773
(cont.)
C-3
-------
Operating Cost:
Variable (including Naphtha) 1.23
Labor 0.12
Maintenance @ 4% Plant Investment 0.18
Capital @ 25% Total Investment 1.40
Tax, Insurance & Miscellaneous
@ 2.5% Total Investment 0.14
Total Cost: $/MCF 3.07
C-4
-------
APPENDIX C.3
REGIONAL COSTS OF BENZENE REMOVAL
FROM REFORMATES & FCC GASOLINE
The cost of benzene removal was determined by capacity range for each
PAD District in this study. The results of these calculations are shown in
Tables C.I through C.6 for refinery reformates, and Table C.7 through C.12
for FCC gasoline.
C-5
-------
TABLE C.I
PADD I
COSTS OF REMOVAL OF BENZENE FROM REFORMATE
REFORMATE CAPACITY RANGE (MB/SD)
NUMBER OF LOCATIONS /CAPACITY
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/Start-up Costs
5. Working Capital & Royalty
o> 6. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9 . Maintenance
10. Total Semi-Variable Operating Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
0-1.5
3
3.5
4,779
4,429
9,208
2,210
373
11,791
1,221
2,976
1,068
4,044
9,399
14,664
$ 4.19
$ 5.1
1.6-3.9
4
8.4
5,956
6,066
12,022
2,885
896
15,803
2,931
3,968
1,394
5,362
12,597
20,890
$ 2.49
$ 7.2
4.0-7.9
1
7.6
3,415
3,682
7,097
1,703
811
9,611
2,652
992
823
1,815
7,661
12,128
$ 1.60
$ 4.1
8.0-15.9
3
27.7
8,231
9,220
17,451
4,188
2,956
24,595
9,665
2,976
2,023
4,999
19,605
34,269
$ 1.24
$ 12.0
16.0-39.9
6
168.8
40,151
44,973
85,124
20,430
18,011
123,565
58,894
5,952
9,869
15,821
98,494
173,209
$ 1.03
$ 59.8
40.0-79.9
1
48.0
9,083
10,175
19,258
4,622
5,069
28,949
16,747
992
2,233
3,225
23,075
43,047
$ 0.90
$ 14.9
18
264.0
71,615
78,545
150,160
36,038
28,116
214,314
92,110
17,856
17,410
35,266
170,831
298,207
$ 1.13
$ 102.9
-------
TABLE C . 2
PADD II
COSTS OF REMOVAL OF BENZENE FROM REFORMATE
REFORMATE CAPACITY RANGE (MB/SD)
NUMBER OF LOCATIONS/CAPACITY
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/Start-up Costs
5. Working Capital & Royalty
6. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9. Maintenance
10. Total Semi- Variable Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Mill ions /Year)
0-1.5
4
5.0
6,827
6,327
13,154
3,157
534
16,845
1,745
3,968
1,525
5,493
13,427
20,665
$ 4.13
$ 7.1
1.6-3.9
5
13.2
9,360
9,533
18,893
4,534
1,408
24,835
4,605
4,960
2,190
7,150
19,796
31,551
$ 2.39
$ 10.9
4-7.9
8
42.6
19,141
20,640
39,781
9,547
4,545
53,873
14,863
7,936
4,612
12,548
42,942
70,353
$ 1.65
$ 24.3
8-15.9
18
195.2
58,004
64,974
122,978
29,515
20,828
173,321
68,105
17,856
14,258
32,114
138,154
238,374
.$ 1.22
$ 82.2
16-39.9
11
267.5
63,628
71,270
134,898
32,376
28,542
195,816
93,331
10,912
15,640
26,552
156,085
275,968
$ 1.03
$ 95.2
40.79.9
4
186.7
35 ,.329
39,575
74,904
17,977
19,716
112,597
65,140
3,968
8,684
12,652
89,751
167,543
$ 0.90
$ 57.9
Total
50
710.2
192,289
212,319
404,608
97,106
75,573
577,277
247,789
49,600
46,909
96,509
460,155
804,453
$ 1.13
$ 277.5
-------
TABLE C . 3
PADD III
COSTS OF REMOVAL OF BENZENE FROM REFORMATE
o
REFORMATE CAPACITY RANGE (MB/SD)
NUMBER OF LOCATIONS/CAPACITY
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/Start-up Costs
5. Working Capital & Royalty
6. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9 . Ma int enanc e
10. Total Semi-Variable Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
0-1.5
8
7.8
10,650
9,870
20,520
4,925
832
26,277
2,721
7,936
2,379
10,315
20,945
33,981
$ 4.36
$ 11.7
1.6-3.9
7
16.2
11,487
11,699
23,186
5,565
1,729
30,480
5,652
6,944
2,688
9,632
24,230
39,514
$ 2.44
$ 13.6
4-7.9
8
42.6
19,141
20,640
39,781
9,547
4,545
53,873
14,863
7,936
4,612
12,548
42,942
70,353
$ 1.65
$ 24.3
8-15.9
7
86.0
38,642
41,667
80,309
19,274
9,176
108,759
30,005
6,944
9.311
16,255
86,692
132,952
$ 1.55
$ 45.9
16-39.9
13
293.0
87,065
97,528
184,593
44,302
31,263
260,158
102,228
12,896
21,402
34,298
207,372
343,898
$ 1.17
$118.6
40-79.9
5
299.6
71,263
79,822
151,085
36,260
31,967
219,312
104,530
4,960
17.517
22,477
174,814
301,821
$ 1.01
$104.1
Total
48
745.2
238,248
261,226
499,474
119,873
79,512
698,859
259,999
47,616
57,909
105,525
556,995
922,519
$ 1.24
$ 318.3
-------
TABLE C . 4
PADD IV
COSTS OF REMOVAL OF BENZENE FROM REFORMATS
REFORMATS CAPACITY RANGE (MB/)
NUMBER OF LOCATIONS /CAPACITY
0-1.5
3
2.1
1.6-3.9
7
19
.1
4-7.9
7
35
.3
8-15.9 16-39.9
3 0
32
.0 0
40.79.9 Total
0 20
0 88
.5
INVESTMENT ($000)
0
i
1.
2.
3.
4.
5.
Fractionation Plant
Extraction Plant
Total Plant Investment
Interest During Construction/Start-up Cost
Working Capital & Royalty
2,867
2,657
5,524
1,326
224
13,
13,
27,
6,
2,
544
794
338
561
038
15,
17,
32,
7,
3,
861
103
964
911
767
9,
10,
20,
4,
3,
509
652
161
839
414
41
44
85
20
9
,781
,206
,987
,637
,443
7,074 35,937 44,642 28,414
MANUFACTURING COSTS ($/SD)
Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9. Maintenance
10. Total Semi-Variable Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
733
2,976
640
6,664 12,316 11,165
6,944
3.170
6,944
3.£22
3,616 10,114 10,7766
5.639
9,988
$ 4.76
$ 3.4
28.645
45,423
$ 2.38
$ 15.7
35.584
58,666
$1.66
$ 20.2
2,976
2.337
5,313
22.649
39,127
$ 1.22
$ 13.5
116,067
30,878
19,840
9.969
29,809
92,517
153,204
$ 1.73
$ 52.9
-------
n
i
TABLE C . 5
PADD V
COSTS OF REMOVAL OF BENZENE FROM REFORMATE
REFORMATS CAPACITY RANGE (MB/SD)
NUMBER OF LOCATIONS/CAPACITY
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/ Start-up Cost
5. Working Capital & Royalty
6. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9. Maintenance
10. Total Semi-Variable Operating Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Mill ions /Year)
0-1.5
2
2.1
2,867
2,657
5,524
1,326
224
7,074
733
1,984
640
2,624
5,639
8,996
$ 4.28
$ 3.10
1.6-3.9
5
12.8
9,076
9,244
18,320
4,397
1,366
24,083
4,466
3,968
2,124
6,092
19,196
29,754
$ 2.32
$ 10.3
4-7.9
3
16.0
7,189
7,752
14,941
3,586
1,707
20,234
5,582
2,976
1,732
4,708
16,129
26,419
$ 1.65
$ 9.1
8-15.9
7
84.6
25,139
28,160
53,299
12,792
9,027
75,118
29,517
6,944
6,179
13,123
59,877
102,517
$ 1.21
$ 35.4
16-39.9
13
300.2
71,406
79,982
151,388
36,333
32,031
219,752
104,740
12,896
17,552
30,448
175,164
310,352
$ 1.03
$107.0
40-79.9
1
73.8
13,965
15,643
29,608
7,106
7,793
44,507
25,749
992
3,433
4,425
35,477
65,651
$ 0.89
$ 22.7
Total
31
489.5
129,642
143,438
273,080
65,540
52,148
390,768
. 170,787
29,760
31,660
61,420
311,482
543,689
$ 1.11
$ 187.6
-------
TABLE C.6
TOTAL U.S. COSTS OF REMOVAL OF BENZENE FROM
GASOLINE REFORMATE--BY PADD
PADD
INVESTMENT ($000)
1. Fractionation Plant
2. Extraction Plant
3. Total Plant Investment
4. Interest During Construction/Start-up Costs
5. Working Capital and Royalty
6. Total Investment
_ MANUFACTURING COSTS ($/SD) ^
M Variable Costs:
i i
7. Total Variable Operating Costs
Semi-Variable Costs:
8. Labor
9. Maintenance
10. Total Semi- Variable Operating Costs
Fixed Costs:
11. Total Fixed Operating Costs
12. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
Number of Gasoline Reformer Locations
Total Capacity-Reformate (MB/SD)
I
71,615
78,545
150,160
36,038
28,116
214,314
92,110
17,856
17,410
35,266
170,831
298,207
$1.13
$102.9
18
264.0
II
192,289
212,319
404,608
97,106
75,573
577,287
247,789
49,600
46,909
96,509
460,155
804,453
$1.13
$277.5
50
710.2
III
238,248
261,226
499,474
119,873
79,512
698,859
259,999
47,616
57,909
105,525
556,995
922,519
$1.24
$318.3
48
745.2
IV
41,781
44,206
85,987
20,637
9,443
116,067
30,878
19,840
9,969
29,809
92,517
153,204
$1.73
$52.8
20
88.5
V
129,642
143,438
273,080
65,540
52,148
390,768
170,787
29,760
31,660
61,420
311,482
543,689
$1.11
$187.6
31
489.5
TOTAL
U.S.A.
673,575
739,734
1,413,309
339,194
244,792
1,997,295
801,563
164,672
163,857
328,529
1,591,980
2,722,072
$1.18
$939.1
167
2297.4
(1)
345 Stream Days per Year (SD/Yr.)
-------
Table C.7
PADD I
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE: 345 SD/Yr
FCC UNIT CAPACITY RANGE: MB/SD 0-4.9
FCC GASOLINE CAPACITY: MB/SD 0-2.8
NUMBER OF LOCATIONS/CAPACITY - / -
INVESTMENT ($000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Construction/Start-up Costs
n
jl, 6. Working Capital & Royalty
ro
7. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
8. Hydrogen Variable Costs
9. Other Variable Costs
10. Total Variable Costs
Semi- Variable Costs:
11. Labor
12. Maintenance
13. Total Semi-Variable Costs
Fixed Costs:
14. Total Fixed Costs
15. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
5-9.9 10-19.9
2.9-5.6 5.7-11.2
- / - 3/42.8
12,229
13,688
18,813
44,730
10,735
3,899
59,364
16,420
16,444
32,864
4,347
5,186
9,533
47,319
89,716
$ 2.10
$ 31
20-39.9
11.3-22.5
4/60.4
11,718
16,268
18,724
46,710
11,210
5,502
63,422
18,986
23,206
42,192
5,796
5,416
11,212
50,554
103,958
$ 1.72
$ 36
40-79.9
22.6-45.1
3/104.6
17,083
24,407
27,301
68,791
16,510
9,529
94,830
26,492
40,187
66,679
4,347
7,976
12,323
75,589
154,591
$ 1.48
$ 53
>80
>45.2
2/126.5
18,269
20,448
29,192
67,909
16,298
11,524
95,731
28,178
48,601
76,779
2,898
7,873
10,771
76,307
163,857
$ 1.30
$ 56
Total
12/334.3
59,299
74,811
94,030'
228,140
54,753
30,454
313,347
90,076
128,438
218,514
17,388
26,451
43,839
249,769
512,122
$ 1.53
$ 176
-------
TABLE C.8
PADD II
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE: 345 SD/Yr
FCC UNIT CAPACITY RANGE: MB/SD
FCC GASOLINE CAPACITY: MB/SD
NUMBER OF LOCATIONS/CAPACITY
0-4.9 5-9.9 10-19.9 20-39.9 40-79.9 >82
0-2.8 2.9-5.6 5.7-11.2 11.3-22.5 22.6-45.1 >45.2 Total
1/1.4 10/43.8 10/83.3 20/317.2 8/221.6 3/184.5 52/851.8
o
INVESTMENT ($000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Construction/Start-up Costs
6. Working Capital & Royalty
7. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
8. Hydrogen Variable Costs
9. Other Variable Costs
10. Total Variable Costs
Semi-Variable Costs
11. Labor
12. Maintenance
13. Total Semi-Variable
Fixed Costs:
14. Total Fixed Operating Costs
1,183
918
1,564
3,665
880
128
4,673
1,391
538
1,929
1,449
425
1,874
3,725
19,465
22,954
28,290
70,709
16,970
3,990
91,669
24,426
16,828
41,254
14,490
8,198
22,688
73,069
23
26
36
87
20
7
115
31
32
63
14
10
24
92
,800
,642
,615
,057
,894
,589
,540
,958
,004
,962
,490
,093
,583
J>9 7
61
85
98
245
58
28
333
99
121
221
28
28
57
265
,540
,435
,332
,307
,874
,897
,078
,705
,868
,573
,980
,441
,421
,497
36
51
57
145
34
20
200
56
85
141
11
16
28
160
,192
,708
,838
,738
,977
,188
,903
,125
,139
,264
,592
,897
,489
,140
26
29
42
99
23
16
139
41
70
111
4
11
15
111
,645
,823
,577
,045
,771
,808
,624
,097
,885
,982
,437
,483
,830
,294
168,825
217,480
265,216
651,521
156,366
77,600
885,48-7
254,702
327,262
581,964
75,348
75,537
150,885
705,822
15. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
7,528 137,011 180,642 544,491 329,893 239,106 1,438,671
$5.38 $ 3.13 $ 2.17 $ 1.72 $ 1.49 $ 1.30 $ 1.69
$3 $ 47 $ 62 $ 188 $ 114 $ 82 $ 496
-------
TABLE C.9
PADD III
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE: 345 SD/Yr
FCC UNIT CAPACITY RANGE: MB/SD
FCC GASOLINE CAPACITY: MB/SD
NUMBER OF LOCATIONS/CAPACITY
0-4.9 5-9.9 10-19.9 20-39.9 40-79.9 >80
0-2.8 2.9-5.6 5.7-11.2 11.3-22.5 22.6-45.1 >45.2 Total
1/1.9 6/24.8 7/52.5 11/182.9 9/315.2 9/657.6 43/1234.9
INVESTMENT ($000)
1.
2.
3.
4.
5.
? 6.
M
* 7.
Fractionation Plant
Hydrogenation Plant
Extraction Plant
Total Plant Investment
Interest During Construction/Start-up Costs
Working Capital & Royalty
Total Investment
1,606
1,246
2,123
4,975
1,194
173
6,342
11
12
16
40
9
2
51
,021
,997
,018
,036
,609
,259
,904
15
16
23
54
13
4
72
,001
,791
,076
,868
,168
,783
,819
35
49
56
141
33
16
192
,485
,262
,699
,446
,947
,662
,055
51,478
73,549
82,267
207,294
49,751
28,715
285,760
94,971
106,294
151,754
353,019
84,725
59,907
497,651
209,562
260,139
331,937
801,638
192,394
112,499
1,106,531
MANUFACTURING COSTS ($/SD)
Variable Costs:
8.
9.
10.
Hydrogen Variable Costs
Other Variable Costs
Total Variable Costs
1,888
730
2,618
13
9
23
,830
,528
,358
20
20
40
,142
,171
,313
57
70
127
,491
,270
,761
79,831
121,100
200,931
146,480
252,650
399,130
319,662
474,449
794,111
Semi-Variable Costs:
11.
1 12.
13.
Fixed
14.
15.
TOTAL
TOTAL
Labor
Maintenance
Total Semi-Variable Costs
Costs:
Total Fixed Costs
Total Manufacturing Costs
MANUFACTURING COSTS ($/P)
MANUFACTURING COSTS ($ Millions/Year)
1,449
577
2,026
5,055
9,699
$. 5.10
$ 3
8
4
13
41
78
$
$
,694
,642
,336
,373
,067
3.15
27
10
6
16
58
114
$
$
,143
,361
,504
,044
,861
2.19
40
15
16
32
153
313
$
$
,939
,399
,338
,087
,186
1.71
108
13,041
24,034
37,075
227,779
465,785
$ 1.48
$ 161
13,041
40,929
53,970
396,678
849,778
$ 1.29
$ 293
62,307
92,942
155,249
882,016
1,831,376
$ 1.48
$ 632
-------
TABLE C.10
PADD IV
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE: 345 SD/Yr
FCC UNIT CAPACITY RANGE: MB/SD
FCC GASOLINE CAPACITY: MB/SD
NUMBER OF LOCATIONS /CAPACITY
INVESTMENT ($ 000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Construction/Start-up Costs
o
, 6. Working Capital & Royalty
i-1
01 7. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs :
8. Hydrogen Variable Costs
9. Other Variable Costs
10. Total Variable Costs
Semi-Variable Costs
11. Labor
12. Maintenance
13. Total Semi-Variable Costs
Fixed Costs
14. Total Fixed Costs
15. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($Milllions/Year)
0-4.9
0-2.8
3/5.1
4,311
3,345
5,697
13,353
3,205
465
17,023
5,067
1,959
7,026
4,347
1,548
5,895
13,569
26,490
$ 5.19
$ 9
5-9.9
2.9-5.6
4/22.5
9,999
11,791
14.533
36,323
8,718
2,050
47,091
12,548
8,645
21,193
5,796
4,211
10,007
37,536
68,736
$ 3.05
$ 24
10-19.9
5.7-11.2
5/40.9
11,686
13,081
17,978
42,745
10,259
3,726
56,730
15,691
15,714
31,405
7,245
4.956
12,201
45,219
88,825
$ 2.17
$ 31
20-39.9 40-79.7 >80
11.3-22.5 27.6-45.1 >45.2
1/13.3 - / - - / -
2,581
3,582
4,123
10,286
2,469
1,212
13,967
4,181
5,110
9,291
1,449
1,193
2,642
11,133
23,066
$ 1.73
$ 8
Total
13/81.8
28,5~7
31,799
42,331
102,707
24, -51
7,4_>2
134,810
37,487
31,428
68,915
18,837
11,908
30,745
107,457
207,117
$ 2.53
$ 72
-------
IABLE C.ll .
PADD V
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE: 345 SD/Yr
FCC UNIT CAPACITY RANGE: MB/SD
FCC GASOLINE CAPACITY: MB/SD
NUMBER OF LOCATIONS /CAPACITY
INVESTMENT ($ 000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Construction/Start-up Costs
i 6. Working Capital & Royalty
H
°" 7. Total Investment
MANUFACTURING COSTS ($/SD)
Variable Costs:
8. Hydrogen Variable Costs
9. Other Variable Costs
10. Total Variable Costs
Semi-Variable Costs:
11. Labor
12. Maintenance
13. Total Semi-Variable Costs
Fixed Costs:
14. Total Fixed Costs
15. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Million/Year)
0-4.9 5-9.9 10-19.9
0-2.8 2.9-5.6 5.7-11.2
- / - - / - 4/30.2
8,629
9,659
13,274
31,562
7,575
2,751
41,888
11,586
11,603
23,189
5,796
3,659
9,455
33,389
66,033
$ 2.19
$ 23
20-39.9
11.3-22.5
5/86.9
16,859
23,406
26,939
67,204
16,129
7,917
91,250
27,315
33,387
60,702
7,245
7,792
15,037
72.735
148,474
$ 1.71
$ 51
40-79.9 >80
22.6-45.1 >45.2
8/216.0 - / -
35,277
50,401
56,376
142,054
34,093
19,678
195,825
54,706
82,987
137,693
-
11,592
16,470
28,062
156.092
321,847
$ 1.49
$ 111
Total
17/333.1
60,765
83,466
96,589
240,820
57,797
30,346
328,963
93,607
127,977
221,584
24,633
27,921
52,554
262,216
536,354
$ 1.61
$ 185
-------
TABLE C.12
COSTS OF REMOVAL OF BENZENE FROM FCC GASOLINE
U.S.A. BY PADD
PADD
INVESTMENT ($000)
1. Fractionation Plant
2. Hydrogenation Plant
3. Extraction Plant
4. Total Plant Investment
5. Interest During Construction/Start-up Costs
6. Working Capital and Royalty
7. Total Investment
oMANUFACTURING COSTS ($/SD) (1'
i
!^ Variable Costs:
8. Hydrogen
9. Other Variable Costs
10. Total Variable Costs
Semi- Variable Costs:
11 . Labor
12. Maintenance
13. Total Semi-Variable Costs
Fixed Costs:
14. Total Fixed Costs
15. Total Manufacturing Costs
TOTAL MANUFACTURING COSTS ($/B)
TOTAL MANUFACTURING COSTS ($ Millions/Year)
Number of FCCU Locations
Total Capacity - FCC Gasoline (MB/SD)
I
59,299
74,811
94,030
228,140
54,753
30,454
313,347
90,076
128,438
218,514
17,388
26,451
43,839
249,769
512,122
$ 1.53
$ 176
12
334.3
II
168,825
217,480
265,216
651,521
156,366
77,600
885,487
254,702
327,262
581,964
75,348
75,537
150,885
705,822
1,438,671
$ 1.69
$ 496
52
851.8
III
209,562
260,139
331,937
801,638
192,394
112,499
1,106,531
319,662
474,449
794,111
62,307
92,942
155,249
882,016
1,831,376
$ 1.48
$ 632
43
1234.9
IV
28,577
31,799
42,331
102,707
24,651
7,452
134,810
37,487
31,428
68,915
18,837
11,908
30,745
107,457
207,117
$ 2.53
$ 72
13
81.8
V
60,765
83,466
96,589
240,820
57,797
30,346
328,963
93,607
127,977
221,584
24,633
27,921
52,554
262,216
536,354
$ 1.61
$ 185
17
333.1
TOTAL
U.S.A.
527,028
667,695
830,103
2,024,826
485,961
258,351
2,769,138
795,534
1,089,554
1,885,088
198,513
234,759
433,272
2,207,280
4,525,640
$ 1.60
$ 1561
137
2835.9
(1)
345 SD/Year
-------
APPENDIX C.4
ENERGY COSTS FOR BENZENE REMOVAL
FROM REFORMATSS & FCC GASOLINE
Benzene removal from refinery reformates and FCC gasoline is an energy-
intensive process. Energy is required for direct fuel burned, steam, electric
power, cooling water and hydrogen production. The energy requirement for hydrogen
is slightly higher for hydrogen plant hydrogen than for refinery-produced hydrogen
because of hydrogen plant fuel requirements. Energy requirements are shown in
detail in Tables C.13 through C.15. The energy requirements for benzene removal
from reformates and FCC gasoline are summarized as follows:
Hydrogen Plant Hydrogen
FCC
Reformates Gasoline* Hydrogen Total
FOE MB/Yr 21,930
MM$ @ $12/FOEB 263
Refinery-Produced Hydrogen
26,573
319
5,513
66
54,016
648
FOE MB/Yr 21,930 26,573 4,372 52,875
MM$ @ $12/FOEB 263 319 52 634
The energy costs for benzene removal represent about 70% to 90% of variables
costs for benzene removal, and about 25% to 28% of total operating costs, including
capital charges.
Expressed in terms of FOEB and energy costs per barrel of benzene removed,
energy requirements are as follows:
Hydrogen Plant Hydrogen
FCC
Reformates Gasoline* Hydrogen Total
FOE B/B
Benzene Removed
$/B Benzene Removed
0.96
11.52
3.76
45.12
0.78
9.36
1.80
21.60
*Excluding hydrogen requirements
C-18
-------
Refinery-Produced Hydrogen
Reformates
0.96
11.52
FCC
Gasoline
3.76
45.12
Hydrogen
0.62
7.44
Total
1.76
21.12
FOE B/B
Benzene Removed
$/B Benzene Removed
As can be seen from the above analysis, energy requirements to remove
benzene from gasoline are nearly double the volume benzene removed. Thus, replac-
ing the benzene and the expended energy would require crude runs of up to 84
million barrels per year, or 230 MB/D. This represents about a 1.4% increase
in total U.S. crude runs and would require construction of at least one large
grass-roots refinery or several smaller expansions to meet energy needs.
C-19
-------
TABLE C.13
REFORMATS ENERGY REQUIREMENTS
Energy Use
Steam
Electric Power
Cooling Water
Total Base Case
FOEB/D
Btu/Unit
1,506,440/M#
10,000 Btu/KWh
12,000 Btu/MGal
Base Case Usage
Units/Day MM Btu/D
1,380
3,180
8,520
2,078.9
31.8
102.2
2,212.9
368.9
Reformer Charge:
Base Case: 13,333 B/SD
Total U. S.: 2,297,400 B/SD
Total Energy Usage
Annual Usage; 345 CD/Yr
Annual Energy Cost @ $12.00/FOEB
63,565 FOEB/SD
21,930 M FOEB/Yr
263 MM$
C-20
-------
TABLE C.14
FCC GASOLINE ENERGY REQUIREMENTS
(1)
Base Case Usage
Energy Use
Direct Fuel
Steam
Electric Power
Cooling Water
Total Base Case
FOEB/D
Btu/Unit
6,000,000/FOEB
1,506,440/M//
10,000/KWh
12,000/MGal
Units /Day
138
2,474
23,140
8,545
MM Btu/D
828
3,727
231
103
4,889
814.8
FCC Charge;
Base Case: 30,000 B/SD
Total U. S.: 2,835,900 B/SD
Total Energy Usage
Annual Usage: 345 CD/Year
Annual Energy Cost @ $12.00/FOEB
77,023 FOEB/SD
26,573 M FOEB/Yr
319 MM$
(1)
Excluding hydrogen
C-21
-------
TABLE C.I5
HYDROGEN ENERGY REQUIREMENTS
A. Hydrogen Plant Hydrogen
Energy Use
Naphtha Feed
Direct Fuel
Steam
Electric Power
Water
Total Unit Energy Cost
Total Requirements: Units/Day
Total Requirements: MM Btu/Day
Annual Requirements: 345 SD/Yr
Annual Cost @ $12/FOEB:
Hydrogen Requirements
Units/MCF Btu/MSCF
(1)
Btu/Unit
18,850 Btu/Lb. 24.9'
6,000,000 Btu/FOEB
1,506,440 Btu/M# (0.062)
10,000 Btu/KWh 1.0
12,000 Btu/MGal 0.400 MGal
488,215
( 93,399)
10,000
5,000
409,816/MSCF
233,962 MSCF/Day
95,881 MM Btu/Day
5,513 MFOEB/Yr
66 MM$/Yr
B. Refinery-Produced Hydrogen
Total Hydrogen Requirements: Units/Day
Total Requirements: MM Btu/Day
Annual Requirements: 345 SD/Yr
Annual Cost @ $2.00/FOEB
233,962 MSCF/Day
76,038 MM Btu/Day
4,372 MFOEB/Yr
52 MM$/Yr
Includes fuel requirement
(2)Hydrogen at 325,000 Btu/MSCF
C-22
-------
APPENDIX C.5
COST OF BENZENE REMOVAL FOR A
10.000 B/D REFINERY
REFINERY CAPACITY: 10,000 B/SD
A. Reformer Only; 2,500 B/SD Gasoline
1,500 B/SD Reformate (60%)
Removal of Benzene From Reformats;
Cost: $/B/SD Reformate 3.70
$/B/SD Gasoline 2.22
C/Gal/SD Gasoline 5.28
$/B/SD Crude 0.525
Total Investment Cost: MM$ 3,872
$/B/SD Gasoline 1,550
B. FCCU or Reformer plus FCCU:
5,000 B/D Gasoline
1,500 B/D Reformate (30%)
1,725 B/D FCC Gasoline (34.5%)
Removal of Benzene From Reformate;
Cost: $/B/SD Reformate 3.70
$/B/SD Gasoline 1.11
C/Gal/SD Gasoline 2.64
$/B/SD Crude 0.56
Total Investment Cost: MM$ 3,872
$/B/SD Gasoline 775
C-23
-------
Removal of Benzene From FCC Gasoline:
Hydrogen Cost:
Cost:
$/MCF
$/B/SD FCC Gasoline
$/B/SD Gasoline
C/Gal/SD Gasoline
$/B/SD Crude
Total Investment Cost: MM$
$/B/SD Gasoline
Hydrogen Plant
Hydrogen
12.00
5.10
1.76
4.19
0.88
5,355
1,071
Refinery Hydrogen
at Fuel Value
0.65
4.20
1.45
3.45
0.73
Removal of Benzene From Reformates & FCC Gasoline:
Hydrogen Cost: $/MCF
Cost: $/B/SD Reformate
$/B/SD FCC Gasoline
$/B/SD Gasoline
C/Gal/SD Gasoline
$/B/SD Crude
Total Investment Cost: MM$
$/B/SD Gasoline
12.00
3.70
5.10
2.87
6.83
1.44
9,227
1,845
0.65
3.70
4.20
2.56
6.09
1.28
C-24
-------
APPENDIX D
NOMENCLATURE
B/SD
Bbls/SD
BTU
COE, FOE
g/gal
gm/gal
LV %
MB
Mbbls
MB/CD
MB/SD
MKWH
Mlbs
MMB
MMB/CD
MMB/Yr
MSCF
MMSCF
PPM
SCF
$/B/SD
MM$
M$
C/Gal
RON
MON
(R-W/2)
Cl
Barrels per Stream Day
British Thermal Unit
Crude Oil Equivalent (6,000,000 BTU per COE)
Grams per Gallon
Liquid Volume Percent
Thousands of Barrels
Thousands of Barrels per Calendar Day
Thousands of Barrels per Stream Day
Thousands of Kilowatt Hours
Thousands of Pounds
Millions of Barrels
Millions of Barrels per Calendar Day
Millions of Barrels per Year
Thousands of Standard Cubic Feet
Millions of Standard Cubic Feet
Parts per Million
Standard Cubic Feet
Dollars per Barrel per Stream Day
Millions of Dollars
Thousands of Dollars
Cents per Gallon
Research Octane Number
Motor Octane Number
(Research Octane plus Motor Octane Number)/2
Clear Octane (without Lead)
D-l
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-450/2-78-021
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Cost of Benzene Reduction in Gasoline to the
Petroleum Refining Industry
5. REPORT DATE
April, 1978
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
F.C. Turner, J.R. Felten, J.R. Kittrell
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Arthur D. Little, Inc.
Acorn Park
Cambridge, Massachusetts 02140
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-02-2859
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Air Quality Planning and Standards
Emission Standards and Engineering Division
Research Triangle Park, North Carolina 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
200/4
15. SUPPLEMENTARY NOTES
16. ABSTRACT
This report assesses the cost to the U.S. petroleum industry of removing
benzene from the two largest contributors to the benzene levels in the gasoline
pool - refinery reformates and FCC gasoline. Predictions were made of the 1981
gasoline pool composition and the benzene content of gasoline component streams.
A process route was selected for each stream and the benzene removal costs in
1977 dollars were developed. Removal of 94.5 percent benzene from reformates
and FCC gasoline would reduce U.S. average benzene content from 1.37 percent
to 0.26 percent. This would require an investment of $5.3 billion and total
costs of $2.5 billion per year including capital recovery, or 2.2 cents per
gallon of gasoline. Costs for some small refineries would be up to 7 cents
per gallon of gasoline or three times the U.S. average costs. These costs
are for benzene removal only, and do not include costs of octane replacement,
volume replacement or the effect on the chemical industry. When these other
factors are considered, it is roughly estimated that the total costs including
capital recovery would be $3.8 billion per year or 3.3 cents/gallon of gasoline.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Benzene
Gasoline
Gasoline Blending Component
Gasoline Pool
Catalytic Cracked
Gasoline
Reformate
Unleaded Gasoline
Lead Phase-Down
Octane Number
18. DISTRIBUTION STATEMENT
Unlimited
cassie
(This Report)
21. N
PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (Rev. 4-77)
PREVIOUS EDITION I S OBSOLETE ..
------- |