AEROSPACE REPORT NO.
TOR-0172(2787)-2
Final Report
An Assessment of the Effects of Lead Additives in
Gasoline on Emission Control Systems which
Might Be Used to Meet the 1975-76
Motor Vehicle Emission Standards
71 NOV 15
Prepared for DIVISION OF EMISSION CONTROL TECHNOLOGY
MOBILE SOURCE POLLUTION CONTROL PROGRAM
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
Office of Corporate Planning
THE AEROSPACE CORPORATION
-------
Report No.
TOR-0172(2787)-2
FINAL REPORT
AN ASSESSMENT OF THE EFFECTS OF LEAD ADDITIVES IN
GASOLINE ON EMISSION CONTROL SYSTEMS WHICH
MIGHT BE USED TO MEET THE 1975-76
MOTOR VEHICLE EMISSION STANDARDS
71 NOV 15
Office of Corporate Planning
THE AEROSPACE CORPORATION
El Segundo, California
Prepared for
DIVISION OF EMISSION CONTROL TECHNOLOGY
MOBILE SOURCE POLLUTION CONTROL PROGRAM
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
Contract No. F04701-71-C-0172
-------
FOREWORD
The Aerospace Corporation has performed (for the Environmental Protection
Agency, Division of Emission Control Technology) an overall assessment of
the effects of lead additives in gasoline on the performance, durability, and
costs of emission control devices/systems which may be used to meet the
1975-76 emission standards for light-duty vehicles. This assessment was
performed in fulfillment of Section 211c(2)B of the Clean Air Amendments of
1970 which states in pertinent part:
No fuel or fuel additive may be controlled or prohibited by the
Administrator pursuant to clause (B) of paragraph (1) except
after consideration of available scientific and economic data
including a cost benefit analysis comparing emission control
devices or systems which are or will be in general use and
require the proposed control or prohibition with emission con-
trol devices or systems which are or will be in general use
and do not require the proposed control prohibition.
This report is the final summary of the overall assessment performed in the
period of June-October 1971. Material related to emission control system
performance, durability, and fuel economy characteristics is contained in
Section 4. A general assessment of the effects of lead additives in gasoline
on emission control system components is presented in Section 5, and a simi-
lar assessment of lead effects on other engine parts is given in Section 7.
Section 6 briefly summarizes the feasibility and implications of lead traps or
exhaust gas scrubber devices. Cost data, in terms of specific system hard-
ware costs and overall consumer costs, are summarized in Section 8.
111
-------
ACKNOWLEDGEMENTS
Appreciation is acknowledged for the assistance and guidance provided by
Mr. G.D. Kittredge and Dr. J. H. Somers of the Environmental Protection
Agency, Office of Air Programs, Division of Emission Control Technology,
for whom this study was conducted. Dr. Somers served as EPA Project
Officer for the study and was instrumental in providing timely access to the
data necessary for the conduct of the study.
The following technical personnel of The Aerospace Corporation participated
in the study and made valuable contributions to the assessment performed
under this contract:
Mr. F.E. Cook
Dr. L.M. Dormant
Mr. J.A. Drake
Mr. L. Forrest
Mr. P.P. Leo
Mr. W.U. Roessler
Mr. M.J. Russi
Mr. W.M. Smalley
Dr. G.W. Stupian
Mr. K.B. Swan
Mr. J.D. Wilson
Approved by
T. lura
Associate Director of Pollution
and Resources Programs
Office of Corporate Planning
M. G. Hinton, Jr.
Manager, Lead Cost-Benefit Study
ir/ctcr of Pollution and
Resources Programs
Office of Corporate Planning
-------
ACRONYMS, TERMS, AND ABBREVIATIONS
A/F
AMOCO
APCO*
Arco
ASTM
CARB
CID
CO
C.R.
CVS
CVS-1
CVS-3
DHEW
Dual Catalytic Converter
EGR
air-fuel ratio
American Oil Company
Air Pollution Control Office
Arco Chemical Company, Division
of Atlantic Richfield Company
American Society for Testing and
Materials
California Air Resources Board
cubic inches displacement
carbon monoxide
compression ratio
constant volume sampling
(test procedure)
single-bag CVS test procedure
(pre-July 1971) using DHEW Urban
Dynamometer Driving Cycle
three-bag weighted average CVS
test procedure (post-July 1971) using
DHEW Urban Dynamometer Driving
Cycle
Department of Health, Education and
Welfare
converter with two beds; one for
oxidizing HC and CO and one for
reducing NO and NO,
exhaust gas recirculation
^Former and current names for EPA air pollution control agencies
Vll
-------
ACROYNMS, TERMS, AND ABBREVIATIONS (cont.)
EPA
Esso
Gasoline Lead Content
HC
HC/CO Catalytic Converter
IIEC
LTR
MBT
NAPCA*
N-O-R
NO
Environmental Protection Agency
Esso Research & Engineering Company
Lead-sterile gasoline is that gasoline
having less than 0.003 gram of lead
per gallon
Lead-free, clear, or unleaded gasoline
as used in this report, is that gasoline
having less than 0.07 gram of lead
per gallon
Low-lead gasoline is that gasoline
having approximately 0.5 gram of
lead per gallon
Fully leaded gasoline is that gasoline
having a normal range of 2 to 3 gramt>
of lead per gallon
hydrocarbons
Converter with single catalyst bed for
oxidizing HC and CO
Inter-Industry Emission Control
Program
Lean Thermal Reactor (air-fuel
ratio >15:1)
Spark advance at maximum torque
National Air Pollution Control
Administration
nitric-oxide-reduction system (by Arco)
oxides of nitrogen (NO plus NO?)
Former and current names for EPA air pollution control agencies
Vlll
-------
ACRONYMS, TERMS, AND ABBREVIATIONS (cont.)
NO Catalytic Converter
OAP
RAM
RON
RTR
SFC
TEL
Tricomponent Catalytic
Converter
UOP
WOT
Converter with single catalyst bed for
reducing NO and NO,
Office of Air Programs
Rapid Action Manifold (thermal
reactor) (by Esso)
research octane number
Rich Thermal Reactor (air-fuel
ratio <15:1)
specific fuel consumption
tetraethyl lead compound (one of
several lead compounds used in
gasoline)
Converter with single catalyst bed for
simultaneously oxidizing HC and CO
and reducing NO and NO-
Universal Oil Products Company
wide-open-throttle (engine operating
condition)
Former and current names for EPA air pollution control agencies
IX
-------
CONTENTS
FOREWORD i"
ACKNOWLEDGEMENTS v
ACRONYMS, TERMS, AND ABBREVIATIONS vii
1. SUMMARY 1-1
1.1 General Conclusions 1-1
1.2 Specific Findings 1-3
1.2.1 Effect of Lead on Performance and Durability
of Emission Control Devices 1-3
1.2.1.1 Catalytic Converter Systems 1-3
1.2.1.2 Thermal Reactors 1-4
1.2.1.3 Exhaust Gas Recirculation (EGR) Systems 1-4
1.2.2 Feasibility of Use of Lead Traps or Exhaust
Scrubber Devices 1-4
1.2.3 Effect of Lead on Other Engine Parts 1-5
1.2.4 General Evaluation of Emission Control
Devices/Systems 1-6
1.2.4.1 Categories of Devices/Systems 1-6
1.2.4.2 Performance of Devices/Systems 1-6
1.2.4.3 Durability of Devices/Systems 1-8
1.2.4.4 Advanced Concepts 1-8
1.2.5 General Cost Summary 1-9
1.2.5.1 Unleaded Gasoline Cost Effects 1-9
1.2.5.2 Emission Control System Cost Effects 1-10
2. INTRODUCTION 2-1
2. 1 Purpose 2-l
2.2 Scope of Study 2-1
REFERENCES 2-3
3 . METHOD OF APPROACH 3-1
XI
-------
CONTENTS (cont.)
4. GENERAL EVALUATION OF EMISSION CONTROL
DEVICES/SYSTEMS 4_1
4.1 Engine Modifications and Operating Considerations 4-1
4.2 Emission Control Devices 4-2
4.2.1 Thermal Reactors 4-2
4.2.1.1 Thermal Reactor Descriptions--General 4-2
4.2.1.2 Engine Modifications 4-8
4.2.1.3 Emission Performance Characteristics 4.9
4.2.1.4 Fuel Economy Characteristics 4-10
4.2.2 Exhaust Gas Recirculation (EGR) Systems 4-12
4.2.2.1 EGR System Descriptions--General 4-13
4.2.2.2 NO Emission Performance Characteristics 4-17
Ji
4.2.2.3 Fuel Economy Characteristics 4-17
4.2.2.4 Drive ability Characteristics 4-22
4.2.2.5 Octane Number Requirements 4-23
4.2.3 Catalytic Converters 4-23
4.2.3.1 Typical Catalysts 4-24
4.2.3.2 Types of Catalytic Converters 4-25
4.2.3.3 Other System Components/Factors 4-30
4.2.3.4 Fuel Economy Characteristics 4-31
4.3 Specific Emission Control Systems 4-36
4.3.1 Catalytic Converter Systems 4-37
4.3.1.1 HC/CO Catalytic Converter Alone (no EGR) 4-37
4.3.1.2 HC/CO Catalytic Converter plus EGR 4-40
4.3.1.3 Dual Catalytic Converter plus EGR 4-41
4.3.1.4 Tricomponent Catalytic Converter (no EGR) 4-43
4.3.2 Thermal Reactor Systems 4-45
4.3.2.1 LTR plus EGR Concept 4-46
4.3.2.2 RTR-Alone Concept 4_53
4.3.2.3 RTR plus EGR Concept 4-57
XII
-------
CONTENTS (cont.)
4.3.3 Combination Systems ........................ 4-62
4.3.3.1 LTR plus HC/CO Catalytic Converter
plus EGR ............................ 4-63
4.3.3.2 RTR plus HC/CO Catalytic Converter
plus EGR ............................ 4_£5
4.3.3.3 RTR plus Dual Catalytic Converter plus EGR .... 4-69
4.3.3.4 RTR plus NO Catalytic Converter plus EGR ..... 4.7 1
2C
4.3.3.5 Stratified Charge Engine .................. 4-72
4.3.4 Summary of Specific Emission Control Systems ...... 4-72
4.3.4.1 Comparison of Emission Levels with
1975-76 Standards ...................... 4-72
4.3.4.2 Lifetime /Durability Effects ................ 4.75
4.3.4.3 Fuel Economy Effects ................... 4_7fc
REFERENCES ................................... 4_79
5. GENERAL ASSESSMENT OF EFFECTS OF LEAD
ADDITIVES ON EMISSION CONTROL
DEVICES/SYSTEMS ............................. .
5. 1 Catalytic Converters ........................... 5_1
5.1.1 Summary of Experimental Data ................. 5-2
5.1.1.1 Laboratory Tests ...................... 5_2
5.1.1.2 Vehicle Tests ......................... 5.9
5.1.2 Maximum Allowable Lead Levels ................ 5-15
5.1.3 Summary ................................ 5-17
5.2 Thermal Reactors ............................. 5-18
5.2.1 Erosion/Corrosion Effects .................... 5-18
5.2.2 Emission Level Effects ...................... 5-22
5.2.3 Summary ................................ 5-22
5.3 Exhaust Gas Recirculation (EGR) Systems ............. 5-24
5.3.1 Relevant Technology Discussions ................ 5-24
5.3.2 1973-74 EGR Systems ........................ 5_26
Xlll
-------
CONTENTS (cont.)
5.3.3 1975-76 EGR Systems 5-27
5.3.4 Summary 5-27
REFERENCES 5-Z8
6. FEASIBILITY AND IMPLICATIONS OF LEAD TRAPS
AND EXHAUST GAS SCRUBBERS 6-1
7. EFFECT OF LEAD ADDITIVES ON OTHER ENGINE PARTS ... 7-1
7. 1 Engine Durability--General 7-1
7.2 Fuel-Sensitive Components 7-2
7.2.1 Exhaust Systems 7-2
7.2.2 Spark Plugs 7-4
7.Z.Z.I Misfiring Mechanism 7-4
7.2.2.2 Life 7-5
7.2.2.3 Cost 7-6
7.2.3 Exhaust Valve Recession 7-7
7.3 Maintenance 7-7
REFERENCES 7-11
8. COST ANALYSIS 8-1
8. 1 Control Device/System Cost Analysis 8-2
8.1.1 Control Device Costs 8-2
8.1.1.1 Engine Modifications 8-2
8.1.1.2 Emission Control System Components 8-3
8.1.1.3 Discussion of Device Costs 8-6
8.2 Overall Costs to the Consumer 8-9
8.2.1 Maintenance Costs 8-9
8.2.2 Operating Costs 8-12
8.2.2.1 Fuel Economy Cost Penalty 8-12
8.2.3 Excluded Costs 8-23
8.2.4 Cost Analyses Results 8-23
REFERENCES 8-27
xiv
-------
CONTENTS (cont.)
APPENDICES
A. VISITS AND CONTACTS A-l
A. 1 Organizations Visited A-l
A.2 Organizations Contacted A-l
B. POSSIBLE CATALYST POISONING MECHANISMS B-l
B.I Summary of Catalyst Poisoning Mechanisms B-l
B.2 Chemical and Mechanical Poisoning Mechanisms B-l
B.2.1 Chemical Poisoning Mechanisms B-l
B.2.2 Mechanical Poisoning Mechanisms B-3
REFERENCES B-5
C. LEAD TRAP DEVICES FOR AUTOMOTIVE VEHICLES
OPERATING ON LEADED GAS C-l
C. 1 Introduction C-l
C.2 Technique for Removing Combinations of Lead
Compound Vapors and Particles from Exhaust
Emissions C-2
C.2.1 Atomics International Molten Carbonate Process C-2
C.3 Techniques for Removing Lead Compound Particles
Only from Exhaust Emissions C-8
C.3.1 Du Pont Cyclone Trap System C-8
C.3.2 Ethyl Corporation Particulate Traps C-9
C.3.3 Dow Chemical Molten Salt Particulate Trap C-10
C.3.4 Cooper Union Molten Salt C-ll
C.3.5.IIT Research Institute Devices C-ll
C.3.5.1 Thermal Packed Bed Device C-ll
C.3.5.2 Sonic Fluidized Bed Device C-12
C.3.6 Houston Chemical Company Particulate Trap C-14
REFERENCES C-l 5
x v
-------
TABLES
4-1. Thermal Reactor Summary 4-5
4-2. EGR Systems Description Summary 4-15
4-3. Catalytic Removal of HC, CO, and NOX--
Laboratory-Scale Experiments 4-26
4-4. Catalytic Removal of HC, CO, and NOX--
Stationary Engine Tests 4-27
4-5. Catalytic Removal of HC, CO, and NOX--
Road Tests 4-28
4-6. Effect of Catalytic Converter System on Fuel
Economy and Performance 4-33
4-7. 1971 Domestic Engine (350 CID)--Normal Choke,
PZ-195 Catalyst 4-39
4-8. 1971 Domestic Engine (350 CID)--Fast Choke 4-39
4-9. Federal Test Results for Some Foreign Vehicles 4-39
4-10. Emission Test Results--1970 Volkswagen with
UOP Catalytic Converter (1972 Federal Test
Procedure) 4-44
4-11. Ethyl Lean Reactor--Emission Data for 1970
Pontiac (Vehicle 766) 4-47
4-12. Ethyl Lean Reactor--Emission Data for 1971
Plymouth (Vehicle 18M-448) 4-47
4-13. Ethyl Lean Reactor—Fuel Economy, Modified
and Standard Cars 4-50
4-14. Ethyl Lean Reactor--Emission Data for Modified
Pontiac (No. 761) Supplied to GARB 4-51
4-ISA, Summary of Emission Control System Emission
Data--Catalytic Converter Systems (Laboratory,
Low-Mileage Tests) 4-74
4-15B. Summary of Emission Control System Emission
Data--Thermal Reactor Systems (Laboratory,
Low-Mileage Tests) 4-74
4-15C. Summary of Emission Control System Emission
Data-- Combination Systems (Laboratory, Low-
Mileage Tests) 4-75
5-1. Effect of TEL on Catalytic Activity 5-4
xvi
-------
TABLES (cont.)
7-1. Spark Plug and Exhaust System Costs 7-9
8-1. EGR System Cost Data 8-4
8-2. Installed Hardware Cost Summary (Cost to
Consumer in New Car) 8-8
8-3. Cost Effects of Use of Unleaded Gasoline 8-19
xvu
-------
FIGURES
1-1. Overall Cost Increase to Consumer Over Lifetime
of Car 1-11
4- 1. Effect of Air-Fuel Ratio on Emission Levels
(Gasoline Spark Ignition Engine) 4-2
4-2. Du Pont Type V Thermal Reactor 4-5
4-3. Air Pump Power Requirements (300-400 CID
Engines) 4_7
4-4. Air-Fuel Ratio and Spark Timing Effects on
NOX Emissions 4-11
4-5. Air-Fuel Ratio and Spark Timing Effects on
Specific Fuel Consumption 4-11
4-6. Arco N-O-R EGR System 4-14
4-7. Toyo Kogyo "Mazda" EGR System (Entry Above
Throttle Valve) 4-14
4-8. General Motors 1972 Design EGR System (Entry
Below Throttle Valve) 4-16
4-9. Effect of Recycle Rate on NO Reduction 4-18
4- 10. Effect of Recycle Rate on SFC 4-18
4-11. Effect of Spark Advance and Recycle--Hot
California Cycle Data 4-19
4-12. Effect of Rich Mixtures on NO Reduction 4-21
4-13. Effect of EGR and Rich Mixture on NO Reduction
and Increase in Specific Fuel Consumption 4-21
4- 14. Reduction in Octane Requirement as a Function
of Recycle 4-23
4-15. Ford Programmed Protection System—Logic
Schematic 4-32
4-16. Ford Programmed Protection System (Vehicle
No. 4)--Catalyst Container and PPS Hardware 4-32
4-17. Converter Flow Development 4-33
4-18. Dual-Bed Axial-Flow Converter 4-34
4-19. Bifurcated Dual-Bed Catalytic Converter 4-34
4-20. Cutaway Sketch of Oxy-Catalyst Converter 4-35
xvm
-------
FIGURES (cont.)
4-21. Exhaust System Backpressures (Road Load
Conditions) 4-35
4-22. Universal Oil Products Mini-Converter
Installation 4-38
4-23. Ford Concept Emission Package "B" 4-42
4-24. Ford Concept Emission Package "C" 4-42
4-25. Tricomponent Conversion vs Air-Fuel Ratio 4-44
4-26. Ethyl Lean Reactor Emissions of Modified
Pontiac (No. 761) Supplied to CARB 4_52
4-27. ESSO Rapid Action Manifold (RAM) Reactor 4.54
4-28. IIEC Type H Exhaust Manifold Reactor (V-8
Engine)--Small Volume with Concentric Core
Design 4.54
4-29. Ford Type J Reactor Durability and Cold Start
Emissions Data 4-55
4-30. General Motors 1975 Experimental Emission
Control System 4-63
4-31. Ford Combined Maximum Effort/Low-Emission
Concept Vehicle 4-66
4-32. Ford Maximum Effort Vehicle--NOX Emissions
vs Fuel Economy 4-68
4-33. General Motors Quick-Heat Manifold and Fast
Choke Configuration 4-70
4-34. NOX vs SFC Increase 4_7g
5-1. Catalyst Life--Leaded vs Unleaded Gasolines 5-2
5-2. Effect of TEL on Catalyst Efficiency for HC
Oxidation 5_4
5-3. NOX Catalyst (CuO/Cr2O3) Deactivation by Lead 5-5
5-4. NOX Catalyst (CuO/Cr2O3) Pellet Composition
Profile (Upon Deactivation with Motor Mix in 5-5
the Fuel)
5-5. Effect of Lead Content in Fuel on Catalyst Type
"BH" Oxidation Activity 5_7
xix
-------
FIGURES (cont.)
5-6. Effect of Lead Content in Fuel on Catalyst Type
"G" Oxidation Activity 5-7
5-7. Effect of Lead Content in Fuel on Catalyst Type
"AJ" Oxidation Activity 5-8
5-8. Effect of Lead Content in Fuel on Catalyst Type
"BI" Oxidation Activity 5-8
5-9. Effect of Lead on Catalyst Life 5-9
5-10. Typical Vehicle Emissions with a Catalytic
Converter 5-11
5-11. Catalyst Efficiency--Cold Start, NDIR HC Data
(Average of Two Vehicles) 5-12
5-12. Catalyst Efficiency--Cruise 30 FID HC Data
(Average of Two Vehicles) 5-12
5-13. Catalyst Efficiency--Cold Start CO Data (Average
of Two Vehicles) 5-12
5-14. Ford 24-Car Fleet Tailpipe HC/CO Emissions
(302 CID Engine) 5-13
5-15. Ford 24-Car Fleet Tailpipe NO Emissions
(302 CID Engine) ? 5-13
5-16. Ford 24-Car Fleet Tailpipe HC/CO Emissions
(390 and 428 CID Engines) 5-14
5-17. Ford 24-Car Fleet Tailpipe NOX Emissions
(390 and 428 CID Engines) 5-14
5-18. Effect of Fuel Additives on Corrosion Weight Loss
ofInconel 601 5-19
5-19. Effect of Fuel Variables on Average Thickness
Losses of OR-1 Alloy During Continuous
Thermal Cycling 5-20
5-20. Weight Change of Test Reactor Cores in Engine
Dynamometer Endurance Test 5-20
5-21. Du Pont Type V Thermal Reactor--HC, CO, and
NO,, Emissions with Leaded and Unleaded Fuel 5-23
JL
5-22. Effect of Leaded Fuel on the Control Efficiency of
Air Cleaner EGR Systems (302 CID Engine) 5-25
7-1. Muffler Lifetime Probability (Operation on
Unleaded Fuel) 7-3
xx
-------
FIGURES (cont.)
7-2. Spark Plug Lifetime Probability (Operation on
Unleaded Fuel) 7_6
8-1. Fuel Economy Penalty due to NO Emission
Reduction 8-14
8-2. Cost of Unleaded Gasoline (Exclusive of
Distribution Costs) 8-16
8-3. Fuel Cost Penalty 8-22
8-4. Increased Consumer Costs Over Lifetime
of Car 8-22
8-5. Breakdown of Increased Consumer Costs over
Lifetime of Car 8-24
C-l. Schematic of an Engine-Compartment-Mounted
Molten-Salt Scrubber C-3
C-2. Fabricated Engine-Compartment-Mounted
Molten-Salt Scrubber C-5
C-3. Installation of Engine-Compartment-Mounted
Molten-Salt Scrubber C-6
C-4. Under-the-Car Lead Trap--Conceptual Design C-7
C-5. Experimental Dual-Anchored Vortex Trap C-10
C-6. Packed Beds for Collection of Submicron
Particles by Thermal Precipitation C-13
-------
1. SUMMARY
-------
SECTION 1
SUMMARY
Based on a review and assessment of data in the open literature and on
discussions with industrial/agency sources, the following conclusions have
been made relative to the effects of lead additives in gasoline on emission
control devices/systems which may be used to meet the 1975-76 emission
standards for light-duty vehicles. For convenience of presentation, some
general conclusions are given first, followed by more specific findings
grouped according to distinctive areas of investigation in the study.
1.1 GENERAL CONCLUSIONS
1. All emission control systems currently planned for use by major
automobile manufacturers and being evaluated by them to meet
the 1975-76 Federal emission standards incorporate or include
a catalytic converter.
2. Lead additives in gasoline are toxic to catalytic materials. Use
of leaded or low-lead gasoline with catalytic converters has
demonstrated that it greatly reduces catalyst activity, thereby
preventing the achievement of a 50,000-mile lifetime at low
enough emission levels to meet the required standards. In addi-
tion to its effect on catalyst activity, lead has been observed to
degrade the structural integrity of NO bulk metal catalysts.
Ji
The scavengers added to gasoline to prevent the accumulation of
harmful lead deposits in the engine also have detrimental effects
upon catalysts. Sulfur and phosphorus have also been noted to
have toxic effects.
3. Lead effects on other major emission control system components,
e.g., thermal reactors and exhaust gas recirculation systems,
although observed to be present to some degree, are such that
materials selection and design techniques can be applied to allow
lead-tolerant systems.
4. Lead traps or exhaust scrubber devices for removal of lead from
the exhaust gases of an engine using leaded gasoline, prior to
passage through a catalytic converter, are not felt to have adequate
lead removal capability nor the technological development status
consistent with other emission system components being con-
sidered for meeting 1975-76 standards.
1-1
-------
Unleaded gasoline should be made available in sufficiently large
quantities to satisfy the demands of vehicles with an installed
catalytic converter unit. The lead content of the gasoline should
be at that level compatible with obtaining a 50, 000-mile useful
lifetime. However, substantive data to precisely establish this
level are not available. Most of the major automakers and catalyst
suppliers have been performing their catalytic converter develop-
ment work with lead levels below 0.06 gm/gal, with most of the
development at levels of 0.02-0.03 gm/gal . At this lead level,
no automobile manufacturer has stated to date that 50,000 miles
of operation at satisfactory emission levels has been achieved.
It is not known whether this durability/lifetime deficiency is
related to the lead level (0.02-0.03 gm/gal), to other trace ele-
ments in the gasoline, or to other catalyst properties. One auto-
mobile manufacturer and one catalyst supplier have stated that
a maximum lead content of 0.03 gm/gal should be an adequately
low level. It should be noted that this value is below the proposed
ASTM specification for unleaded gasoline of 0.07 gm/gal.
With regard to 1975 emission standards, both lead-tolerant sys-
tems (e.g., Esso's rich thermal reactor system; and lead-
intolerant systems (e.g., systems incorporating catalytic con-
verters) have demonstrated approaching the standards on an
experimental basis at low mileage. However, in order to meet
the lower NO levels required in 1976, the lead-tolerant system
would require1 the addition of a NO catalyst (and possibly addi-
tional components) which would render it sensitive to lead.
A general evaluation of emission control devices/systems envisioned
by the automobile industry and ancillary development organizations
has indicated that none of the systems planned for 1976 have demon-
strated the capability of meeting the 1976 NOX emission standard.
Several combination systems incorporating a NOX catalyst have
met the 1976 emission levels on an experimental basis with a new
catalyst. At this time, a 50, 000-mile lifetime has not been demon-
strated. In fact, durability tests over approximately 10-15,000 miles
have not been reported.
At this time, estimated overall costs to the consumer (initial,
operating and maintenance) for emission control systems being
considered for the 1976 Federal emission standards are approxi-
mately $860 above 1970 vehicle costs over an 85,000-mile vehicle
lifetime. This estimate is based on a system incorporating a
dual-bed catalytic converter (with assumed replacement of the
converter unit at 50,000 miles), a "low-grade" rich thermal
reactor, and exhaust gas recirculation.
1-2
-------
1.2 SPECIFIC FINDINGS
1.2.1 Effect of Lead on Performance and Durability of
Emission Control Devices
1.2.1.1 Catalytic Converter Systems
Lead is toxic to catalysts. It can act as a poison through both chemical and
mechanical mechanisms, which are not mutually exclusive. Scavengers
added to gasoline to prevent the accumulation of harmful lead deposits in the
engine also have detrimental effects upon catalysts. Sulfur and phosphorus
have also been noted to have toxic effects on catalysts.
Data on catalyst activity versus lifetime are available on some catalysts for
gasoline lead levels of: (1) 2-3 gm/gal, (2) 0.5 gm/gal, and (3) in the range
of 0.02-0.06 gm/gal. For the lowest levels (0.02-0.06 gm/gal), the exact
amount of lead used is not clearly identified. In general, the catalyst life-
time decreases as lead content is increased. However, at the very low
levels, the data are not sufficient to establish a meaningful correlation. The
data do show that activity and lifetime are drastically affected with lead levels
over 0.5 gm/gal; at levels of 0.02-0.06 gm/gal, the catalyst showed sig-
nificantly better performance than at the higher values tested.
Some catalysts have been designed and tested for operation with leaded gaso-
line; however, test data are not available in sufficient quantity and under the
appropriate vehicle operating conditions to permit an evaluation at this time.
Since catalysts are so adversely affected by lead quantities in leaded gasoline,
a system must be devised to prevent accidental contamination. It has been
stated that a single tankful of regular leaded gasoline can destroy a catalyst
(see Section 4.3.3.2.4). Although this cannot be substantiated, it is apparent
from available data that such quantities could seriously reduce the catalyst's
useful lifetime.
1-3
-------
1.2.1.2 Thermal Reactors
Lead concentrations in gasoline of approximately 0.5 gm/gal should have no
significant detrimental effects on the better oxidation-resistant materials
available. There seems to be no obvious reason (although direct data are
lacking) why such materials could not function with up to 3 gm/gal of lead.
However, the combined presence of lead and phosphorus additives has an
accelerating influence on the corrosive deterioration of a number of different
metallic alloys.
1.2.1.3 Exhaust Gas Recirculation (EGR) Systems
Lead-free or low-lead gasoline is not required for the implementation of
EGR systems, per se. Although the presence of lead additives can result
in deposits in EGR orifices, throttle plate areas, etc., the actual severity
of such deposits would appear to be strongly related to the particular type of
EGR system as well as to control orifice sizes used, and/or to the utilization
of self-cleaning designs (plungers, specially coated surfaces, flexible snap-
rings, etc.) in areas susceptible to deposit buildup.
1.2.2 Feasibility of Use of Lead Traps or Exhaust
Scrubber Devices
Several lead-removal devices are currently under development. One basic
type, requiring cool exhaust gases to enable a particulate form of lead for
collection (e.g., cyclone separators, fiberglass filter devices), is inherently
not suitable for use upstream of a catalytic converter. A second type, capable
of removing lead in the gaseous as well as particle form, is compatible in
principle. The only known and demonstrated device of this type is the molten
carbonate lead trap.
A molten carbonate lead trap device has undergone considerable development
and testing and the results suggest that it might be installed upsteam of a
catalytic converter and have the potential for removing an average of 90 per-
cent of the lead, essentially all the sulfur oxides, and in excess of 80 percent
1-4
-------
of all the particulates. Aside from system design complexities, the need for
adding another component to the already complicated emission control systems,
and the need for periodic chemical replacement, it is felt that the 90-percent
lead removal capability from leaded gasoline will not be adequate for the lead-
sensitive catalysts presently predicted for use by the automobile industry.
Moreover, durability test data on prototype systems are not available to per-
mit the assessment of the decrease in effectiveness versus mileage. There-
fore, it is not felt that this system could be incorporated in 1975-76 model
automobiles.
1.2.3 Effect of Lead on Other Engine Parts
The principal deleterious effect of lead additives in gasoline on engine parts
other than the emission control system, per se, is to reduce the usable life-
time of exhaust systems and spark plugs. Other reported differential effects
(varnish, sludge, rust, wear, etc.), due to unleaded versus leaded gasoline,
do not result in a quantifiable impact on the consumer in terms of operational
considerations or cost. The use of unleaded gasoline can essentially double
the exhaust system life and increase spark plug life approximately 50 percent
in conventional (pre-1971) cars.
Similar spark plug life increases with unleaded gasoline are expected in
1975-76 systems; if long-life exhaust systems (e.g., stainless steel) com-
patible with either leaded or unleaded gasoline are incorporated in 1975-76
systems, no lifetime variabilities would exist for this component.
There is considerable evidence that excessive valve seat wear can occur with
the use of unleaded fuel, particularly at sustained high-speed and high-load
conditions. However, this problem can be solved at very low cost by chang-
ing to induction-hardened exhaust valve seats. One domestic manufacturer
has introduced such valve seats in some 1972 models, with plans for full
implementation by the end of the 1972 model year. Other manufacturers are
also phasing in compatible exhaust valves and seats. All U.S. automakers
plan to market a system compatible with unleaded gasoline by the 1975 model
year.
1-5
-------
1.2.4 General Evaluation of Emission Control Devices/Systems
1.2.4.1 Categories of Devices/Systems
A broad spectrum of emission control devices has been evaluated by the
automotive industry and ancillary development companies. In general, they
fall into the following categories:
a. Catalytic Converter Systems (no form of thermal reactor warmup
device)
1. HC/CO Catalytic Converter Alone
2. HC/CO Catalytic Converter plus Exhaust Gas
Recirculation (EGR)
3. Dual Catalytic Converter plus EGR
4. Tricomponent Catalytic Converter Alone
b. Thermal Reactor Systems
1. Lean Thermal Reactor (LTR) plus EGR
2. Rich Thermal Reactor (RTR) Alone
3. Rich Thermal Reactor (RTR) plus EGR
c. Combination Systems
1. LTR plus HC/CO Catalytic Converter plus EGR
2. RTR plus HC/CO Catalytic Converter plus EGR
3. RTR plus Dual Catalytic Converter plus EGR
4. RTR plus NOV Catalytic Converter plus RTR
X
A summary of the available emission data for these emission control system
concepts is presented in Table 4-15 in Section 4.3.4.
1.2.4.2 Performance of Devices/Systems
Several emission control systems have met, or show promise of meeting,
the Federal 1975 emission standards. These systems are experimental
versions, and emission data do not reflect consideration of any factor to
account for variabilities in production tolerances, testing procedures, or
degradation with mileage.
i-6
-------
The following general observations are pertinent for the performance of
emission control systems:
1. In general, the catalytic-converter-only systems suffer because
of high emissions during the cold start portion of the CVS test
procedure due to slow warmup. In addition, the tricomponent
catalytic converter requires a precision in air-fuel ratio control
not adequately demonstrated to date.
2. LTR plus EGR systems have yet to demonstrate meeting 1975-76
HC and CO standards, although improvements in thermal reactor
design (flameholders, improved mixing, etc.) and fast warmup
choke devices could improve this situation. NOX levels below
approximately 1.3 gm/mi have not been reported for this concept.
3. The most advanced RTR plus EGR systems meet the 1975-76
HC standard and approach the CO standard. NOX levels below
approximately 0.5-0.7 gm/mi have not been demonstrated. At
this NOX level, however, the fuel economy penalty is severe
(approximately 25-30 percent).
4. "Combination systems, " i.e. , various combinations of thermal
reactors, EGR, and catalytic converters, are judged by the
automotive industry to offer the best hope for achieving minimum
emission levels and are under intensive development at this time
for incorporation in 1975-76 model cars. In these combination
systems, the primary function of the thermal reactor is to warm
up a catalyst bed. Therefore, it need not be a "full-size" thermal
reactor, but rather a "low-grade, " less complex type.
5. Thermal reactor (LTR or RTR) plus HC/CO catalytic converter
plus EGR systems are meaningful in terms of 1975 standards;
however, it is generally agreed that a NOX catalytic converter
would have to be added for the lower NOX levels required by the
1976 standards. At this time, NOX catalysts with the required
durability (50,000 miles) have not been demonstrated. In this
case (NOX catalyst), the thermal reactor is restricted to rich
operation, inasmuch as all known NOX catalysts require a
reducing atmosphere.
It is emphasized that the foregoing observations are based on experimental
laboratory data only. If, as the various automakers have suggested, levels
of approximately 50 percent of the 1975-76 standards have to be achieved to
account for the variation of production tolerances, test reproducibility,
degradation with accumulated mileage effects, etc. , then it would appear
1-7
-------
that none of the emission control systems proposed and evaluated to date will
meet the 1975-76 emission standards on a consistent basis.
1.2.4.3 Durability of Devices /Systems
There are no meaningful lifetime or durability data available for any combined
emission control system seriously being considered for implementation by the
U.S. automakers. The approximately 90,000-mile durability test of a thermal
reactor by Ford is certainly significant, but did not include an EGR system or
a catalytic converter. Engelhard has reported a 50,000-mile durability test
for their PTX catalyst; however, it did not include an EGR system for NOX
control. At this point in time, then, emission control system durability or
lifetime remains simply as a goal, with little or no demonstrated capability.
1.2.4.4 Advanced Concepts
A prototype stratified charge engine installed in a one-quarter-ton light truck
was recently tested by EPA, Willow Run, Michigan, and met the 1976 emission
standards at a 3000-pound inertia weight on the dynamometer. However, the
power-to-weight ratio of this vehicle was not sufficiently high to meet all of
the acceleration requirements of the dynamometer driving cycle. This sys-
tem incorporated a thermal reactor, EGR, and a HC/CO catalytic converter.
The Ford Motor Company, developer of this engine, states that it is not suf-
ficiently well developed to permit quantity production in the near future. On
this basis, then, the stratified charge engine concept has not been compared
with the previously identified emission control system concepts which are
compatible with present generation spark ignition engines.
A variety of other advanced engine concepts are in various stages of research
and development. In this category are such concepts as the "lean-burn" and
"prechamber" approaches. There are insufficient data at this time to fully
evaluate the emission control, mass producibility, and costs of these sys-
tems. Therefore, they were not included in this assessment.
1-8
-------
1.2.5 General Cost Summary
It should be noted that in the consumer cost analysis reported herein, the
exhaust system life variability with gasoline lead content is not a factor,
since a long-life exhaust system (e.g. stainless steel) compatible with either
leaded or unleaded gasoline was incorporated into all considered systems in
view of the generally higher temperature levels expected.
The following observations can be made on the estimated costs of unleaded
gasoline and of emission control systems:
1.2.5.1 Unleaded Gasoline Cost Effects
1. If unleaded gasoline is made available, the eventual (circa 1980)
estimated unit cost differential for a single-grade (93 RON)
unleaded gasoline is 0.46 cent per gallon more than the current
conventional two-grade leaded gasoline weighted average price
per gallon. This cost differential is the increase in manufactur-
ing costs for the unleaded gasoline plus a distribution cost
increase necessary for introduction and implementation of a new
(93 RON) unleaded gasoline grade.
2. Utilization of a single-grade (93 RON) unleaded gasoline requires
a reduction in compression ratio (to approximately 8.35:1) from
pre-1971 values (approximately 9.37:1 weighted average). This
reduction in compression ratio increases the specific fuel con-
sumption, thereby increasing the fuel cost over the lifetime of
the average car (85,000 miles) by approximately $130, while the
additional cost per gallon of unleaded gasoline increases the fuel
cost by approximately an additional $30. Both of these fuel costs
are independent of any fuel economy degradation attributable to
the emission control system concept, per se.
3. Implementation of a three-grade unleaded gasoline system, at
the same clear-pool octane number (93 RON), enables an
increase in compression ratio (approximately 8.95:1 weighted
average) over the single-grade (93 RON) system (approximately
8.35:1). This increase in compression ratio reduces the fuel
cost penalty due to compression ratio change from $130 to $50
for this three-grade unleaded gasoline case over the lifetime of
the average car. (A two-grade system would have similar, but
not identical, fuel cost savings.)
1-9
-------
1.2.5.2 Emission Control System Cost Effects
1. Implementation of emission control systems designed to meet
1975-76 emission standards implies very high costs to the con-
sumer. This cost is a strong function of the required NOX
emission level as shown in Fig. 1-1. Only systems incorporating
a NOX catalyst appear to have the potential to meet the Federal
1976 NOX standard of 0.4 gm/mi. As shown in the figure, if a
50, 000-mile NOX catalyst durability is achieved, the estimated
overall consumer cost (initial, maintenance, and operating costs)
over the life of the car is approximately $860 for a single-grade
unleaded gasoline system and approximately $780 for a three-
grade unleaded gasoline case. If the NOX catalyst durability is
only 25,000 miles, for example, additional converter replace-
ments increase these overall costs to $1170 and $1090,
respectively.
2. As most emission control systems which might be employed to
meet 1975 emission standards are the same as, or building-
block portions of, 1976 systems, it is not clearly meaningful to
attempt to compare system costs at the 1975 emission standards
(e.g. , NOX = 3. 1 gm/mi). It would not appear either prudent or
cost effective to attempt to implement one distinctive type of sys-
tem in 1975 model cars and a completely different type of system
in 1976 model cars.
3. For the system currently considered most promising to meet
1976 emission standards (low-grade RTR plus dual catalytic
converter plus EGR), the estimated overall consumer cost with
the single-grade unleaded gasoline supply system previously
noted ($860 above average 1970 vehicle costs over the life of the
car) includes $362 for the fuel cost penalty, $388 for initial
installed hardware costs, and $110 net maintenance cost. A cost
summary of all systems evaluated can be found in Fig. 8-5,
Section 8.2.4.
1-10
-------
1400
1300
1200
1100
1000
900
^ 800
CO
S 700
_i
I 600
t—
500
400
300
200
100
0
,-RTR+EGR + DUALCC*
(25,000 mi LIFE)
• I GRADE UNLEADED GASOLINE
3 GRADE UNLEADED GASOLINE
EGR + HC-COcc* _
-RTR+EGR
+ DUAL cc*
(50,000 mi LIFE)
RTR+EGR
2 GRADE LEADED.
GASOLINE
PRESENT LOWER
LIMIT ON NO.
. EMISSION
* CATALYTIC CONVERTER* EGR SYSTEMS SAME IN COST
EXCEPT FOR $70 DECREASE DUE TO OMISSION OF
LOW-GRADE THERMAL REACTOR
I
I
04 08 12 16
NOX EMISSION LEVELS, gm/mi
20
24
Fig. 1-1. Overall Cost Increase to Consumer Over
Lifetime of Car
1-11
-------
2. INTRODUCTION
-------
SECTION 2
INTRODUCTION
2.1 PURPOSE
The purpose of this report is to present an assessment of the effects of lead
additives in gasoline on emission control devices/systems which may be used
to meet the 1975-76 Federal emission standards for light-duty vehicles in
fulfillment of Section 21 Ic (2)B of the 1970 Clean Air Act (Ref. 2-1). These
standards are:
Standards (gm/mi)
Emission 1975 1976
HC (hydrocarbons) 0.41 0.41
CO (carbon monoxide) 3.40 3.40
NO (oxides of nitrogen) 3.10 0.40
X
when determined by the DHEW urban dynamometer driving cycle using the
Federal CVS (constant volume sampling) test procedure and the three-bag
weighted-aver age technique (CVS-3) for 1975-76 (Ref. 2-2).
2.2 SCOPE OF STUDY
This assessment of emission control system effects encompassed per-
formance, durability, and cost aspects. In the performance area, emission
performance characteristics, fuel economy characteristics, power, drive-
ability, and fuel octane number requirements were examined. In the cost
area, initial, maintenance, and operational cost factors were included.
A primary limitation on the type of and level of assessment in the above
areas is that it was to be based on current knowledge and the state of the art
as could be determined from existing data in the open literature or data
2-1
-------
obtainable from industrial/agency sources. In this regard, it is noted that
the most meaningful and comprehensive data have been made and accumulated
by the automobile manufacturers.
The recent change in Federal test procedures from the single-bag CVS
method to the three-bag weighted-average CVS method (Ref. 2-2) could have
substantial effects on the resultant emission levels of certain systems. At
the time of writing, a large portion of the available CVS emission test data
is single-bag CVS test data. For comparison purposes, therefore, the
present effort is limited in some instances to single-bag CVS emission test
data. The 1975-76 emission standards for the single-bag CVS tests were:
Standards (gm/mi)
Emission 1975 1976
HC 0.46 0.46
CO 4.70 4.70
NO 3.00 0.40
x
These standards are pertinent only when single-bag CVS test data are pre-
sented in the report.
In the cost area, there is a minimal amount of information pertaining to
specific emission control system costs. The cost data reported herein, then,
are limited to engineering estimates based on raw material costs and to com-
parisons with more conventional hardware components.
2-2
-------
REFERENCES
2-1. Public Law 90-148, The Clean Air Act, As Amended.
2-2. "Control of Air Pollution from New Motor Vehicles and New Motor
Vehicle Engines," Federal Register, Vol. 36, No. 128 (2 July 1971),
2-3
-------
3. METHOD OF APPROACH
-------
SECTION 3
METHOD OF APPROACH
In accomplishing the stated purpose in Section 1, within the scope and
limitations noted, the following specific steps were performed in accomplish-
ing the study:
1. A literature survey was made and relevant material obtained.
This material included letters from the automobile manufacturers
to the Administrator of EPA (April 1971) in response to his
request for information regarding progress towards meeting the
requirements of the Clean Air Act.
2. Visits were made to the four domestic U.S. automobile manu-
facturers, various major oil companies active in the emission
control area, catalyst manufacturers, the two major lead
additive manufacturers, and three foreign automobile manufac-
turers for direct discussions of emission control system tech-
nology. These visits not only provided direct comments and
information, but also resulted in published data to supplement
the original literature survey. (See Appendix A for a list of
organizations contacted.)
3. Pertinent emission control system information was compiled
and assessed to provide a comprehensive picture for emission
control systems in general.
4. Specific emission control systems likely to be used for 1975-76
emission standards were then examined relative to the effects of
lead additives on performance, durability, and costs.
The following discussions, then, summarize the significant results of the
assessment in the sectional order of presentation in the report:
1. The basic effect of engine air-fuel ratio variation and engine com-
ponent modifications on uncontrolled spark ignition engine HC,
CO, and NOX emission levels is briefly outlined (Section 4. 1).
2. Specific emission control devices which are useful to control one
or more of the HC, CO, and NOX emission species are defined
and their basic method of operation delineated (Section 4.2).
3-1
-------
3. A wide variety of emission control "systems" (i.e., combinations
of emission control devices to simultaneously control HC, CO,
and NOX emissions) is defined by generic classification. Specific
emission control systems evaluated by various companies to date
are used to illustrate the generic class in terms of emission con-
trol characteristics, fuel economy characteristics, and dur-
ability. Where appropriate, stated effects of lead additives in
gasoline on a particular emission control system are delineated
(Sections 4.3.1 through 4.3.3).
4. The overall spectrum of potential emission control systems is
then compared in emission level capability with the 1975-76
standards, and summarized with respect to their (1) durability,
and (2) fuel economy effects (Section 4.3.4).
5. An overall assessment of the effects of lead additives in gasoline
on the various emission control system concepts is presented in
Section 5.
6. The feasibility and implications resulting from the use of lead
traps or exhaust gas scrubber devices to remove lead additives
from engine exhaust gas are presented in Section 6.
7. The effects of lead additives on engine parts other than the emis-
sion control system, per se, are summarized in Section 7.
8. Finally, the estimated costs of various emission control system
concepts are summarized in terms of (1) initial consumer costs
(as installed in a new car), and (2) overall consumer costs,
which reflect maintenance cost and operating cost in addition
to initial acquisition costs (Section 8).
3-2
-------
4. GENERAL EVALUATION
OF EMISSION CONTROL
DEVICES/SYSTEMS
-------
SECTION 4
GENERAL EVALUATION OF EMISSION CONTROL
DEVICES/SYSTEMS
Gaseous emissions from automobile exhausts may be controlled either by
inhibiting formation of the gases in the engine cylinders or by lowering their
concentration externally. In general, methods of controlling exhaust emis-
sions from automotive spark ignition internal combustion engines to meet the
stringent 1975-76 Federal standards involve certain engine modifications and
the use of a combination of several devices. Multiple methods are necessary
because of the requirement for simultaneous control of the hydrocarbon (HC),
carbon monoxide (CO), and oxides of nitrogen (NO ) constituents in the
a
exhaust gas.
4.1 ENGINE MODIFICATIONS AND OPERATING CONSIDERATIONS
As illustrated in Fig. 4-1, HC, CO, and NO concentrations in the exhaust of
JC
uncontrolled engines are strongly a function of the operating air-fuel ratio.
As can be noted from the figure, at the stoichiometric air-fuel ratio NO
production is very high while HC and CO production is relatively low. For
air-fuel ratios between approximately 17 and 19, levels for all three con-
stituents are reduced considerably from peak values. Currently, engine
operation is restricted to air-fuel ratios below approximately 19 to avoid
excessive power loss and rough engine operation. Operation in the 17-19
range minimizes HC and CO levels, and lowers that of NO , but the concur-
rent reduction of exhaust levels of all three species is far from sufficient to
meet 1975-76 emission requirements. NO formation can be suppressed by
operating in the "rich" regime (air-fuel ratios of approximately 11-13);
however, in this region HC and CO concentrations are very high.
Other factors affecting emissions include spark timing, and induction system
and combustion chamber design. Retarding the spark results in lower peak
4-1
-------
temperatures and less NO formation. Also, the exhaust gas temperature
X:
is higher with a retarded spark, which promotes further combustion of the
HC and CO species in the exhaust system. Induction system modifications
can result in lower emissions by providing a more uniform mixture to the
cylinders and better atomization and vaporization of the fuel. Combustion
chamber design affects the combustion process and, as a result, peak and
exhaust gas temperatures.
o
o
CO
CO
CO
UJ
oc
10
14 16
AIR-FUEL RAT 10
Fig. 4-1. Effect of Air-Fuel Ratio on Emission Levels
(Gasoline Spark Ignition Engine)
4-2
-------
4.2 EMISSION CONTROL DEVICES
4.2.1 Thermal Reactors
A thermal reactor is a chamber (replacing the conventional engine exhaust
manifold) into which the hot exhaust gases from the engine are passed. The
chamber is sized and configured to increase the residence time of the gases
and permit further chemical reactions, thus reducing HC and CO concentra-
tions. In general, the thermal reactor embodies a double-walled and insulated
(between walls) configuration, with port liners to direct the exhaust gases to
its inner core section. In some instances, baffles and/or swirl plates are
used to further promote mixing.
There are two different types of thermal reactors under consideration at
this time: the Rich Thermal Reactor (RTR) and the Lean Thermal Reactor
(LTR).
4.2.1.1 Thermal Reactor Descriptions--General
4.2.1.1.1 Rich Thermal Reactor (RTR)
The RTR is designed for fuel-rich engine operation. As a result of the
associated chemically reducing atmosphere and lower combustion tempera-
tures, the amount of NO formed in the engine cylinders is reduced (Fig. 4-1).
X
If the engine is run rich enough (A/F approximately 11-12), it is possible to
limit the formation of NO to less than 2 gm/mi (Ref. 4-2); however, fuel
economy penalties at these rich mixtures are as high as 20 percent. As the
exhaust from the cylinders contains large quantities of HC and CO, secondary
air supplied by a pump is injected into the reactor to permit further oxidation
of these species.
The thermal reactor should be designed for minimum thermal capacity to
heat promptly to lower emissions for cold start conditions. Since relatively
high temperatures (1700-2000CF) are achieved in the RTR, high-temperature
materials (e.g. , Inconel 601 containing 60 percent Ni, 23 percent Cr, 14 per-
cent Fe, 1-1/2 percent Al) are required for the inner core, baffles, and port
4-3
-------
liners. At these high temperatures, engine misfiring, which produces high
HC levels, could lead to excessive local temperatures and material burnout
conditions in the RTR; therefore, temperature control devices are necessary
to protect it. (Ceramic materials, which could be more tolerant to over-
temperature conditions than metals, have not to date demonstrated the nec-
essary thermal and mechanical shock properties.)
A typical RTR design is shown in Fig. 4-2. The system illustrated is the
Du Pont Type V reactor, one of many experimental versions created by this
company in the course of an evolutionary development program begun in
1962 (Ref. 4-1). The reactor consists of a cast iron outer shell which houses
a tubular core and a shield to reduce the heat loss from the hot core to the
cooler outer shell. Exhaust gases mixed with air supplied by a belt-driven
air pump first enter the tubular core, which is designed to promote mixing
and initiate oxidation. The reacting gases then pass through the core-shield
annulus and the shield-shell annulus where oxidation is completed before
the gases exit into the conventional exhaust system. Sheet metal inserts at
the engine exhaust ports are provided in a number of different reactor designs
to reduce heat loss to the water-jacketed exhaust port surfaces.
Other RTR configurations differ from the Du Pont Type V system principally
in the arrangement of the internal core geometry and in the volume provided
for the shell and core chambers. A summary description of proposed experi-
mental designs for RTR types is presented in Table 4-1. One of these, the
Esso Modified Rapid Action Manifold (RAM) rich reactor system (Ref. 4-Z)
is unique in that the reactor geometry is toroidal rather than cylindrical.
Exhaust gases flow from a manifold collector, around the torus, and exit
through a slot in a central plenum which discharges to an exhaust pipe. This
flow arrangement is said to provide superior mixing of the secondary air with
the exhaust combustibles. Also unique in the RAM system is the use of
flameholders. These devices are designed to produce a stabilized flame at
the outlet of the engine exhaust ports during startup, when the engine is
4-4
-------
EXHAUST GAS
r—OUTER SHELL
RADIATION
SHIELD
TO EXHAUST SYSTEM
Fig. 4-2. Du Pont Type V Thermal Reactor (from Ref. 4-1)
Table 4-1. Thermal Reactor Summary
Reactor
Type
Rich Reactors
Du Pont Type V
Du Pont Type VII (b)
Esso Synchrothermal
Esso Modified RAM
IIEC/Ford Type H
IIEC/Toyo Kogyo
UEC /Nissan
British Small Engine
Lean Reactor
Ethyl Lean Reactor
Induction
Air-Fuel
Ratio
14
11.5-12.5
12. 2
11-13
(a)
(a)
(a)
10-14
17-19
Reactor
Operating
Temperature
(a)
(a)
1600-1900(c)
1600-1750
1600-1850(c)(f)
1600-1800(c)(f)
(a)
1600-1650
1400-1600(c)
Reactor Volume (in. )
Core
60
(a)
103
71(d)
40
61
(a)
70
-
Total
(a)
(a)
(a)
(a)
97
(a)
220
(a)
160
Reactor
Air
Injection
Yes
Yea
Yes
Yes
Yes
Yes
Yes
Yes
No(g)
(a) Not specified (e) Flameholders
Port
Liners
Yes
Yes
No
No(e)
Yes
Yes
Yes
No
Yes
Reference
4-53
4-54
4-12
4-2
4-3
4-10
4-10
4-55
4-5/4-52
(b) Recent Du Pont system (f ) Thermal protection cutoff temperature
(c) With EGR (g) Air injection possibly required during choke period
(d) Torus volume
4-5
-------
choked. The flame serves the purpose of oxidizing CO and HC and accelerates
reactor warmup from cold conditions. Model II RAM reactors were made
from Type 310 stainless steel.
Aside from air-fuel ratio effects, small fuel economy losses are directly
attributable to the thermal reactor. These include the additional backpres-
sure created by reactor flow resistance, and the power required to drive the
air pump for secondary air injection in rich reactor systems. Esso studies
on the RAM reactor indicated that the device added the equivalent of about
one muffler to the total engine exhaust backpressure (Ref. 4-2). An early
Ford/IIEC RTR design (Type D, Ref. 4-3) produced backpressures 8 in. Hg
higher than the 2 in. Hg values obtained for standard production exhaust
systems. However, a later design with revised internal geometry in com-
bination with the use of a dual exhaust system reduced the backpressure to
the production exhaust system level.
Power requirements for the air pump may be gauged from Fig. 4-3. The
data shown are based on current pump designs for V-8 engines ranging from
300 to 400 cubic inches of displacement (CID) (Ref. 4-4). For example, at
2500 rpm (equivalent to a cruise speed of about 65 mph), and estimating a
backpressure of 5 in. Hg, the power requirement indicated is 0.5 hp, which
corresponds to a fuel economy loss of less than one percent.
As mentioned previously, by comparison with these small fuel economy
losses a carburetor calibration change of three A/F units (15 to 12) to mini-
mize NO to less than 2 gm/mi may incur a fuel economy penalty of 15 to 20
X.
percent. And if an exhaust gas recirculation system (EGR), described in
Section 4.2.2, is added to the RTR to further control NO to levels below
x
approximately 1 gm/mi, fuel economy penalties are as high as 20-30 percent.
Although it is instructive to consider these component contributions to the
fuel penalty, the total system loss is more directly useful to an evaluation of
alternate control schemes. These data are discussed in Section 4.3 in
4-6
-------
o
Q_
UJ
CO
CC
O
0
BACKPRESSURE IS SUCH
THAT OVER 3 hp IS
SELDOM REQUIRED
,nnn
buuu
0
5 10
BACKPRESSURE, in.Hg
15
Fig. 4-3. Air Pump Power Requirements
(300-400 CID Engines)
(from Ref. 4-4)
conjunction with the specific combined emission control systems described
in that section.
4.2.1.1.2 Lean Thermal Reactor (LTR)
The LTR is used in conjunction with an engine operated on the lean side of
stoichiometric mixtures, i.e., with excess air. Currently, LTR systems
are limited to air-fuel ratios of approximately 19. As shown in Fig. 4-1,
HC and CO concentrations in the engine exhaust are much lower than in the
case of the RTR (but NO levels are somewhat higher). Therefore, little
X
chemical heat is generated in the reactor and its temperature is governed to
4-7
-------
a large extent by the sensible heat in the exhaust gas. This means that the
oxidation of HC and CO is accomplished within the LTR at lower tempera-
tures than for the RTR, and without the requirement for additional air (i.e. ,
no air pump or mixing limitation). Because of the lower operating tempera-
tures, the durability requirement can be met by less expensive materials
for the construction of the reactor core and baffles; however, careful atten-
tion must be given to minimizing heat losses or conversion is limited by low
reaction rates. On the other hand, more stringent requirements exist for
engine air-fuel mixture control and cylinder-to-cylinder distribution. This
may require utilization of an advanced carburetor or electronic fuel injection.
EGR is generally added for additional NO control. Although little or no
fuel economy penalty is chargeable to the LTR itself, with EGR an approxi-
mate 10-percent decrease in fuel economy is realized for NO levels of
X.
approximately 1.5 gm/mi. Peak power loss due to lean operation causes a
small loss in vehicle performance.
The Ethyl Corporation Lean Reactor is the only known design of a lean oper-
ating system for which specific details of configuration and performance are
available. It is designed for operation at air-fuel ratios of between 17 and
19. As shown in Table 4-1, its operating temperature is 1400-1600°F, or
200-300 degrees lower than those for rich reactor systems. The reactor is
cylindrical and consists of an open-tube liner made of 310 stainless steel
surrounded by a layer of insulation which in turn is enclosed by an outer
casing of 310 or 430 sheet stainless steel (Ref. 4-5).
4.2.1.2 Engine Modifications
Engine modifications required for thermal reactor operation generally differ
for rich and lean systems. For the rich reactor, the modifications are
minor. In addition to the adjustment of carburetor calibration to rich mix-
tures, the timing may be retarded from current production settings to
increase the temperature of the gases leaving the exhaust port. Some form
of overtemperature sensing and control system may be necessary in order to
4-8
-------
prevent excessive reactor temperatures due to possible malfunctions (e.g.,
spark plug misfire) or to sustained high-load operation. Peak temperatures
may be limited by terminating secondary air injection to the exhaust gases
entering the reactor.
Lean reactor operation requires a departure from conventional carburetor
design in order to achieve satisfactory vehicle driveability. Ethyl has
actively pursued this problem and has developed an experimental high-
velocity carburetor which provides the necessary mixture preparation for
satisfactory operation at all operating conditions. Modifications in ignition
timing and carburetor operation during deceleration and idle have also been
made. In addition, lean carburetion involving the use of smaller, dual, or
staged Venturis to provide stronger fuel metering signals and better fuel
mixture preparation is being explored by Chrysler with support from Ethyl
and Bendix (Ref. 4-6). The results of Ethyl tests indicate that overtempera-
ture protection for the lean operating system is not required. In one case,
Ethyl disconnected three spark plug wires and found the reactor temperature
did not increase (however, HC and CO emissions would increase).
4.2.1.3 Emission Performance Characteristics
In general, the performance of thermal reactors relative to the control of
HC and CO is dependent on such configurational factors as the reactor vol-
ume, mass, internal geometry, and heat exchange characteristics. In addi-
tion, the performance may be influenced strongly by engine operating con-
ditions such as air-fuel ratio and spark timing, particularly when these
operating parameters are adjusted to extreme values for the purpose of
achieving concurrent control of NO .
x
Large reactor volume is desirable for good reactor performance because
it provides for longer residence times needed to complete mixing and the
HC/CO oxidation reactions. Appropriate design of the exhaust gas flow path
using internal baffling may mitigate the volume requirement through better
4-9
-------
mixing and control of reactant concentrations. This is the preferred design
route because it minimizes the problem of engine-compartment packaging,
reduces the surface area for heat loss, and tends to provide a low-mass
system. Thin-gauge materials are preferred because they provide for lower
mass designs with low thermal inertia. Rapid response to warmup is an
important reactor design objective because of the heavy discharge of HC and
CO which occurs under cold start engine conditions. Thermal considerations
also dictate close coupling of the reactor to the engine to minimize heat
losses. Frequently, sheet metal liners are provided in such areas as the
water-jacketed exhaust port surfaces and the reactor inlets to conserve heat
in the exhaust gas.
Because of the requirement to control NO emissions to lower levels in
X.
future systems, the development of thermal reactor devices has evolved in
coupling with other emission control devices such as EGR and catalysts.
For this reason, it is not useful to present the emission performance char-
acteristics of thermal reactors in isolation from other emission equipment,
except where a specific system has been actually operated in this mode. The
bulk of the available emissions data concerns the operation of emission con-
trol systems comprised of RTRs or LTRs combined with EGR. Thermal
reactor system emission data are presented in Section 4.3 in conjunction
with the specific combined emission control systems described in that
section.
4.2.1.4 Fuel Economy Characteristics
Generally, the principal factors governing fuel economy losses in thermal
reactor control devices are the selected engine air-fuel ratios and spark
timing adjustments needed to suppress NO emissions. The relationship
between NO emissions, air-fuel ratio, and spark timing is shown in
Fig. 4-4 for an engine operating at constant rpm (Ref. 4-7). The effects of
these engine adjustments on specific fuel consumption (SFC) are shown in
Fig. 4-5 for the same engine operating conditions. The sensitivity of the
4-10
-------
uj ppm
O
AIR 14
FUEL
RATIO 1S
18
tdc tdc
Fig. 4-4. Air-Fuel Ratio and Spark
Timing Effects on NOX
Emissions (from Ref. 4-7)
btdc
AIR
FUEL
RATIO
SPARK
TIMING
tdc
Fig. 4-5. Air-Fuel Ratio and Spark Timing Effects on
Specific Fuel Consumption (from Ref. 4-7)
-------
fuel consumption parameter to air-fuel ratio at spark-retarded conditions
may be noted.
For the sole control of HC and CO, both rich and lean thermal reactor
devices may be operated with lower fuel economy losses than is the case
whenverylow NO values are also of concern. In this mode, the rich
reactor air-fuel setting may be calibrated nearer stoichiometric. Du Pont
has operated its Type V rich reactor under these conditions and has
reported HC and CO emissions of 0.20 gm/mi and 4.50 gm/mi, respectively,
(using the 7-mode test procedure) with a loss in fuel economy of only
1.3 percent (Ref. 4-8). Ethyl quotes little or no loss for its lean reactor
when operated without EGR (Ref. 4-5).
4.2.2 Exhaust Gas Recirculation (EGR) Systems
An EGR system is a means for introducing a portion of the exhaust gas back
into the incoming air-fuel mixture. The amount of NO formed during the
X.
combustion process in the engine cylinder is related to the temperature of
combustion: higher temperatures yield more NO . The temperature of
li.
combustion can be lowered by the introduction into the combustion chamber
of chemically inert substances that absorb part of the heat of combustion.
Exhaust gases from an engine provide a convenient source of such substances.
The specific EGR systems which form the basis of most of the following
discussion were evaluated in the following programs:
1. Arco N-O-R EGR System (Ref. 4-9)
An evaluation by Arco Chemical Company, Division of the
Atlantic Richfield Company, of their nitric-oxide-reduction
(N-O-R) EGR system on 1966 and later model vehicles.
2. Toyo Kogyo "Mazda" EGR System (Ref. 4-10)
An evaluation 1.5 liter "Mazda" vehicle equipped with a thermal
reactor and EGR (accomplished as part of the IIEC Program).
4-12
-------
3. Toyota EGR Test System (Ref. 4-7)
An experimental dynamometer evaluation of the effects of EGR
on three small passenger car gasoline engines.
4. Esso EGR System (Ref 4-11)
An evaluation of EGR system potential for two 1966 vehicles
(Plymouth and Chevrolet). This program was sponsored by
NAPCA.
5. Esso Synchrothermal EGR System (Ref. 4-12)
An evaluation of a Synchrothermal reactor system combined
with EGR.
6. Esso RAM EGR System (Ref. 4-Z)
An evaluation of a rapid action manifold thermal reactor system
combined with EGR.
7. Esso Extended-Use Program (Ref. 4-13)
A durability evaluation of three 1969 Plymouths and three 1969
Chevrolets with EGR systems developed in Ref. 4-11. This
program was sponsored by NAPCA.
8. Arco Fleet Test Program (Refs. 4-14, 4-15, 4-16)
An evaluation by the California Air Resources Board (CARB)
(under Federal Grant No. 68A0605D) of the Arco N-O-R EGR
system on a 120-vehicle fleet test basis.
9. Du Pont Reactor Test Vehicles (Ref. 4-17)
An evaluation by CARB (Project CI) of six 1970 Chevrolets
equipped with the Du Pont thermal reactor and EGR.
4.2.2.1 EGR System Descriptions—General
Many different EGR system designs have been employed by the various
investigators. The location of the exhaust gas pickup, the point of introduc-
tion of the recycled gas into the engine induction system, the metering
devices, and their signal sources have all been varied greatly. For
example, Fig. 4-6 illustrates the Arco N-O-R system (Ref. 4-9) in which
the recycle gas is picked up from the heat riser area of the exhaust manifold
and metered directly (below the carburetor throttle plate) to the intake mani-
fold. Fig. 4-7 illustrates the Toyo Kogyo system (Ref. 4-10) in which the
recycle gas is picked up downstream of the exhaust manifold (and cooled)
4-13
-------
FUEL INLET
DISTRIBUTION
LINE
CARBURETOR
INTAKE
MANIFOLD
IMTAKF £_.
PORT
PYWAIIQT ftAQ
LAnAUol uAo
TO EXHAUST
SYSTEM
"te^As
^>u-i*-Ljr
Vrf
(7
>"
"-L
-*43
1^
1 !W
y
ff \ -•}
-^-^-1
RECYCLE
CONTROL
VALVE
HEAT RISER AREA
OF EXHAUST
MANIFOLD
Fig. 4-6. Arco N-O-R EGR System (from Ref. 4-9)
ACCEL. SWITCH OFF
BELOW 13° THROTTLE ANGLE
1
SPEED SWITCH OFF
ABOVE 50 mph
T
HEAT SENSING VALVE OFF
BELOW 140 °F WATER TEMP.
CARBURETOR
EGR ON-OFF VALVE
SPACER-
SOLENOID
VALVE
INLET MANIFOLD
Fig. 4-7. Toyo Kogyo "Mazda" EGR System (Entry
Above Throttle Valve)(from Ref. 4-10)
4-14
-------
and introduced into a spacer plate above the carburetor throttle plate.
Fig. 4-8 illustrates a below-the-throttle system evaluated by General Motors
(Ref. 4-18). This system is currently installed on some 1972 Buick models
sold in California.
Other significant variabilities in EGR system design include exhaust gas
recycle rate, and restrictions as to when the EGR flow is on or off. For
example, without such restrictions EGR can cause rough idling as well as
loss of power during wide-open-throttle (WOT) operation. Therefore, in
most systems EGR is eliminated at idle or WOT conditions, or restricted
to a lower vehicle speed range. Table 4-2 summarizes the more salient
features of a number of selected systems incorporating EGR with regard to
the foregoing.
Table 4-2. EGR Systems Description Summary
System
Arco N-O-R
Toyo Kogyo
Esso EGR
Esso Synchrothermal
Reactor plus EGR
Esso RAM Reactor
plus EGR
Arco Fleet Test
Program
Esso EGR Extended-
Use Program
Du Pont Reactor
Vehicles
Ethyl Lean Reactor
plus EGR
General Motors
EGR System (see Fig. 4-8)
Tap -off
Location
Heat riser below
carburetor
After exhaust
manifold (cooled)
Upstream of
muffler (cooled)
Upstream of
muffler (cooled)
Upstream of
muffler (cooled)
Heat riser below
carburetor
Upstream of
muffler (cooled)
Upstream of
muffler (cooled)
Near muffler
(cooled)
Exhaust manifold
Injection
Location
Below throttle
Above throttle
Above throttle
Above throttle
Above throttle
Below throttle
Above throttle
Above throttle
Above throttle
Below throttle
Recycle
Rate
(%)
15-22
•MO
varied
•Ml
•M2
0-15
(-10)
9-17
•M5
variable (as
high as 30)
6-20
EGR Shutoff
At
Idle, WOT
varied
Idle, WOT
Idle, below
20-25 mph
cruise
Idle, WOT
Below ~20 mph
cruise
Idle, WOT
Idle, WOT
Idle, WOT
Reference
4-9
4-10
4-11
4-12
4-2
4-14/4-15/
4-16
4-13
4-17
4-5/4-52
4-18
4-15
-------
CLOSED m
VALVE
OPEN
VALVE
DESIGN PRINCIPLE:
EGR IS CONTROLLED BY A VALVE WHICH METERS EXHAUST GAS FROM THE INTAKE
MANIFOLD CROSSOVER AND DISTRIBUTES IT INTO THE INTAKE SYSTEM. A VACUUM
SIGNAL, MODULATED BY THROTTLE BLADE POSITION, ACTUATES THE DIAPHRAGM
EGR VALVE, WHICH IN TURN POSITIONS A CONTOURED SPOOL THAT
REGULATES EXHAUST GAS FLOW.
RECIRCULATION RATE: 6 TO 20%
TESTS HAVE BEEN CONDUCTED ON 95 EGR-EQUIPPED 1972 PRODUCTION ENGINES.
TYPICAL NOX EMISSION RESULTS (1972 FEDERAL TEST PROCEDURE)
WITH EGR WITHOUT EGR
3-1/2-5 gm/mi
6-8 gm/mi
Fig. 4-8. General Motors 1972 Design EGR System (Entry Below
Throttle Valve) (from Ref. 4-18)
4-16
-------
4.2.2.2 NOX Emission Performance Characteristics
Experimental and theoretical data relating NOx reduction to exhaust gas
recycle rate are presented in Fig. 4-9 for engines operating at conventional
air-fuel ratios. The agreement between prediction and test data is good.
For low recycle rates, the reduction of NO is nearly proportional to the
amount of exhaust gas recycled. For higher quantities of recycle, the effect
diminishes. Substantial (approximately 40-80 percent) NOx reductions are
achievable at 10-20 percent recycle rates in the conventional air-fuel ratio
range.
4.2.2.3 Fuel Economy Characteristics
Because of the dilution of the charge and reduced peak combustion tempera-
ture, a reduction in power output occurs (at the same spark advance setting)
which effectively translates into a fuel economy loss. SFC test results for
the vehicle systems previously shown in Fig. 4-9 are shown in Fig. 4-10
and compared with the theoretical prediction of Newhall (Ref. 4- 19). Again,
the correlation is good. The specific data points shown for the Arco N-O-R
system are plotted at the average of the 15-22 percent recycle rates quoted.
Because of the interrelationship of spark timing, cycle temperature, and
power output, it is possible to advance spark timing to avoid or minimize
the effects of EGR on power and SFC. In tests performed by Esso (Fig. 4-11,
from Ref. 4-11), EGR was shown to have a much lower fuel economy penalty
than spark retard for the same NO reduction. It was found possible to
operate with both recycle and some spark advance and obtain some NO
X.
reduction with a slight improvement (approximately 2 percent) in SFC in one
case.
For any given vehicle, then, the fuel consumption penalty would be strongly
influenced by the baseline engine air-fuel ratio and NO emission character-
X,
istics, the amount of NO reduction required to meet a given standard, and
4-17
-------
100
80
o
U.
O
o 60
t—
< j
n
UJ
ui
40
20
= I.O(A/F~I5:|)
NEW HALL (REF. 4-19)
ARCO (REF. 4-9)
O AT 50 mph
• AT 30 mph
ESSO (REF.4-II)
A PLYMOUTH
A CHEVROLET
_L
10 15 20
PERCENT RECYCLE
25
30
Fig. 4-9. Effect of Recycle Rate on NO Reduction
10
s
2 6
Q_
CO
CO
-------
o
o
£
Q
LU
I—
CO
Q
-------
the potential for optimizing spark timing and recycle rate within these
constraints. Therefore, the data of Figs. 4-9 and 4-10 should be considered
as broadly representative only.
Figures 4-12 and 4-13 (from Ref. 4-20) illustrate the separate and combined
effects of air-fuel ratio and EGR recycle rate on SFC and NO reduction.
x
Figure 4-12 shows the dramatic reduction in NO occasioned by extremely
rich air-fuel ratios and the concurrent very high increase in SFC resulting
from such rich operation. Figure 4-13 combines these air-fuel ratio effects
with similar EGR effects to provide a map of EGR flow rate and air-fuel
ratio effects on NO reduction and concurrent SFC increase.
In general, at a given air-fuel ratio, the maximum amount of EGR flow rate
consistent with vehicle "driveability" constraints is required for a minimum
NO level. Therefore, the various air-fuel ratio plus EGR rate lines of
X
Fig. 4-13 combine to indicate an upper limit of NO reduction limit. It
would be expected, then, that emission control systems employing only air-
fuel ratio control and EGR for NO reduction (i.e. , non-NO catalyst sys-
X X
terns) would be characterized by the type of data shown in Fig. 4-13.
Fleet test fuel economy results (with leaded gasoline) for EGR systems are
available in one instance (Ref. 4-16). The prototype N-O-R system-equipped
1967 Comet fleet (approximately 10 percent EGR, EGR off at WOT) had
essentially the same average SFC when tested 4 months (approximately
5000 miles) after installation of the EGR system. After 12 months (about
14,000 miles), this same fleet had an approximate 6 percent SFC increase
compared to a comparable nonequipped Comet fleet. In the case of a sim-
ilarly equipped 1968 Plymouth fleet, EGR-equipped cars had an approxi-
mate 4-5 percent average SFC increase over the nonequipped fleet at
4 months (about 8000 miles) after installation. At 12 months (about 11,000
miles), this increase was approximately 9 percent.
4-20
-------
OPERATING CONDITION
ZUJ
29 20-
00 40-
UQ:
«|eo-
Q08O-
tr
lu
o. IOOL
ENGINE
SPEED
rpm
I6OO
240O
32OO
BRAKE
TORQUE
Kg-m
3.5
4.5
7
SPARK
TIMING
M3T
M3T
MBT- 5'
MBT-IO'
MBT
CODE
o
A
4
A
X
15 14 13 12
AIR-FUEL RATIO
I I
Fig. 4-12. Effect of Rich Mixtures on NO
Reduction (from Ref. 4-20)
A/F-15:!
10%)
( I 5%)
OF EGR RATE
IO
15 2O
Fig.
PERCENT INCREMENT
OF S.F.C.
4-13. Effect of EGR and Rich Mixture
on NO Reduction and Increase in
Specific Fuel Consumption (from
Ref. 4-20)
-------
4.2.2.4 Driveability Characteristics
As mentioned previously, EGR decreases maximum combustion temperature
and pressure. Concurrent with this loss in maximum pressure is an increase
in ignition delay time and a decrease in flame speed, resulting in a retarded
pressure peak. The net effect is a more pronounced cycle-to-cycle pressure
(and torque) variation which affects the smoothness of operation and/or
response ("driveability") which is more pronounced as the EGR flow rate is
increased.
Other noticeable performance effects can be rough idle, stumble during part-
throttle operation, surge at certain cruise speeds, and an increase in full
throttle acceleration time. In general, these effects increase in severity
with increase in EGR flow rate.
Therefore, acceptable driveability effectively sets a limit on recycle flow
rate, particularly for conventional engines operating in the nominal 13-15
air-fuel ratio range. Extensive driveability tests conducted by Esso
(Ref. 4-11) showed the EGR rate limit to be dependent on the vehicle tested.
One vehicle showed nominal driveability at 17 percent recycle, while another
was borderline on acceptability at 15.7 percent recycle.
Driveability evaluations were made in the Arco fleet test program conducted
by CARB (Ref. 4-15). They were based on the impressions of individuals
assigned to drive the cars in the motor pool. The major complaints for
EGR-equipped cars included: "the car sounds noisy," "the car idles too
fast, " "the car hesitates when the throttle is depressed to the floor, " and
"the engine does not idle smoothly when cold. " However, the overall drive -
ability rating, as evaluated by more than 200 drivers in this Arco test,
favored the EGR-equipped vehicles over the nonequipped vehicles. It should
be noted that the EGR rate was only approximately 10 percent.
4-22
-------
4.2.2.5 Octane Number Requirements
Esso data (Fig. 4-14, from Ref. 4-11), indicate that the use of EGR can
result in a reduction in the fuel octane requirement for knock prevention.
This result occurred with EGR present at WOT conditions to prevent high
NO emissions at such periods. But it was previously shown (Table 4-2)
X,
that most EGR system approaches to date shut off EGR at WOT to negate
WOT power loss. Therefore, the true effect of EGR on fuel octane number
requirement is dependent upon the EGR mode of operation at WOT conditions,
S
en
8
s
oc.
10
8
6
4
2
0
O
A PLYMOUTH
O CHEVROLET
1
6 10
PERCENT RECYCLE
14
18
Fig. 4-14. Reduction in Octane Requirement as a
Function of Recycle
(From Ref. 4-11)
To date, the automakers plan to shut off EGR at WOT; therefore, no further
consideration is given herein to lowered octane number requirements
occasioned by the use of EGR.
4.2.3
Catalytic Converters
An automotive catalytic converter is a device containing a catalyst material
which chemically decreases exhaust gas emissions. Three basic catalytic
systems are being considered:
1. Single-bed oxidation catalysts that remove HC and CO
2. Dual-bed devices having one oxidation catalyst bed to remove
4-23
-------
HC and CO and a separate reduction catalyst bed to remove
NO
x
3. Tricomponent or single-bed catalytic devices that simultaneously
remove HC, CO, and NO
H
Both base metal and noble metal catalysts are under intensive evaluation and
development by the automobile industry. Specific configurations of catalytic
converters vary widely. One approach is to use a monolithic coated sub-
strate contained in a cylindrical shell. Another approach is to use a pellet-
ized form of catalyst held in place by interior louvered members, within an
outer container. In general, the specific structural and chemical formula-
tions are considered "trade secrets" by the catalyst suppliers. Necessary
attributes for catalytic converters for automotive use include sufficient
activity, long life, resistance to mechanical shock, and high-temperature
capability.
4.2.3.1 Typical Catalyst s
Literally hundreds of catalyst types have been examined for possible use in
controlling automotive emission of HC, CO, and NO . Usually, these
catalysts were first tested in laboratory-scale experiments, with the more
promising ones then tested in engine dynamometer tests and, finally, in
vehicle road tests. References 4-21 through 4-25 give a good account of many
of the catalysts tested. However, the exact composition of the more prom-
ising types is usually considered proprietary and referred to by a letter
designation or more generally as base (or transition) metal, noble (or
precious) metal, or metallic.
4.2.3.1.1 Base Metal Catalyst
Base metal catalysts employ metals or oxides of metals from the transitional
group (Periodic Table of Elements) which includes vanadium (V),
chromium (Cr), manganese (Mn), iron (Fe), cobalt (Co), nickel (Ni),
copper (Cu), and zinc (Zn). Several metals and their oxides are usually
combined to form a catalyst. Supports such as alumina (A1_O,) and/or
4-24
-------
silica (SiO2) are used to provide structural strength. Tables 4-3 through
4-5 include a small fraction of the formulations that have been tested. It
should be noted that a few base metal catalysts also incorporate trace
amounts of noble metals such as platinum (Pt) or palladium (Pd).
4.2.3.1.2 Noble or Precious Metal Catalysts
The noble or precious metal catalysts that have been tested are primarily
Pt and Pd. They are deposited on A1_O, or SiO, supports and are char-
acterized by relatively low concentrations of active metal (approximately
0.1-0.6 percent by weight). Tables 4-3 through 4-5 list a few of the noble
metal catalysts.
4.2.3.1.3 Metal or Bulk Metal Catalysts
Bulk metal catalysts, as the name implies, are homogeneous metals of
varying shapes that require no support material. Pellets, wires, and honey-
comb structures are some of the shapes that have been used. Monel, copper,
stainless steel, and copper-coated stainless steel also have been used.
4.2.3.2 Types of Catalytic Converters
There are three basic types of catalytic converters that have been tested to
varying degrees and with varying degrees of success.
4.2.3.2.1 HC/CO-Oxidation Catalysts
The catalysts which oxidize HC and CO into carbon dioxide (CO?) and water
(H2O) are referred to as HC/CO catalysts. By far, the greatest effort has
gone into developing this type of catalyst and literally hundreds of combina-
tions have been tested, including base metals, precious metals, and com-
binations of both (Table 4-3). HC/CO oxidation catalysts, as the name
implies, require excess oxygen (air) to convert the HC and CO to HO and
L*
CO . This could be accomplished by operating at lean air-fuel mixtures or
by adding secondary air to the engine exhaust. To date, the latter approach
has been used almost exclusively.
4-25
-------
Table 4-3. Catalytic Removal of HC, CO, and NO --Laboratory-Scale Experiments (from Ref. 4-Z1)
Catalyst Composition
CuO/Cr203(Ba)a
CuO/Cr203(Zn)b
CuO/CR203(Ba)a
6. 8CuO/6. 8 Cr203/86.4
82 CuO/ 17 Cr2O3d
30 CuO/70SiO2
10 Fe2O3/85 Al2O3/5 SiO2
10 CuCrzO4/85 Al2O3/5 SiO2
5 Cu/95 A1203
4 CuO/96 A1203
CuO • Cr2Oj(Ba)a
3MnO2/2CuO
0.6 Pt/99.4 Sl02
0. 2 Pd/99. 8 SiO2
4 Cr203/96 A1203
2CoO/12 MnO,/0. 03 Pd/
76 A12O3
4 Cr,O,/10 CuO/0. 02 Pd/
86 A1203
4 CoO/12 MnO,/0. 02 Pd/
4 Fe2O3/80 *12O3
18 Cr2O3/82 A12O3"
Molec sieve 13X/Cu, Crf
a. Girdle r catalyst G-22:32 (
b. Girdler catalyst G-50- 10 (
c. For O2 < 2 vol %.
Reactive Gas Composition
(ppm)
HC CO NOX
0 60000 4000
0 60000 4000
0 60000 4000
0 60000 4000
0 10000 1500
<3000 400-1400 40-1400
0 12000 20000
0 12000 20000
0 13000 125
250 10000
1000-2500 0-30000
1000-2500 0-30000
2400 40000
2400 40000
2000 20000
1000 38500
1000 38500
1000 38500
1700 24000
1700 24000
Catalytic
Conversion
(vol %)
HC CO N"x
>97
100
70-100 100
91
>90
90
>50 >50
>50 >50
100
>50 >50
>90 >90
>90 >90
>90 90
>90 >90
>90 >80
85 >85
75 >100
60 94
83 93
84 83
:u/25 Cr/11 Ba. d. Harshaw catalyst Cu-0203, pr
~u/26 Zn/31 Cr. e. Catalyst promoted with CsO.
Test Conditions
Space
Temperature Velocity
300-330 15-25
300-330 15-25
340 10
200 10
200-500
300 10
150-400 0. 02
150-400 0.02
7.9
230-450
180-400 5.5-11
180 5.7-11
280 5.7
260 5.7
300-675 10
343 5
343 5
180-450 0. 5
600 10
600 10
Notes
Oxygen reduces catalytic efficiency
Oxygen reduces catalytic efficiency
Oxygen reduces catalytic efficiency,0 attrition
of catalyst with prolonged use
Al2Oj- supported catalyst shows similar properties
to unsupported CuO - Cr-O,
At temperatures
-------
Table 4-4. Catalytic Removal of HC, ,CO, and NO --Stationary Engine Tests (from Ref. 4-21)
Catalyst Composition
(»t ft)
MnOx /CuO /NiO /C rOxe
NiO(Ba)/Al20}
NiO(Ba)/Al203
7CuO/ 09SiO2/Al2O3
30CuO/70SiO2
6CuO/6Cr2O3/AI2O3
CuO/CrzO3'
10CuO/4Cr,O,/ 02Pd/
A1203
62CuO/5Co203/33Al2O3
8CuO/4Co203/ 1 V20j /
20CuO/0 IAg2O/Al2O3
6CuO/0 lPd/65iO2/Al2O3
25-15CuO/ 05- 3Pd/
SiO2/AI2O3
S-10V205/A1203
SV205/7CuO/5S.02/Al203
V205/CuO/Cr203/Al203
50Co304/CaAl204
4-ISMnOx(2-STl)/AI2O3
4U3Og/Al203
MoOx/Al2O3
lPd/IPt/Al2O}
0 !9Pt(Ba)/Al2O3
0 1P1/A1203
0 375PI/0 SF/0 25C1/
A1203
0 IPl/0 5F/A12O3
0 4PI/A12O3
3 ZPd/AljOj
Exhaust Gaa Composition,
Before Converter
(ppm)
HC CO NOX
310-630
1400 29000 155
-
325 10000
1000
418
20000 60000 4000
-
2000 60000 1500
325 40000
1400 3000
140 1750
-
2900
5000-8000 15000-25000
418
50-12000 10000-60000
375 40000
4650
2800
.
.
-
- - -
-
.
-
Catalytic
Conversion3
(vol %)
HC CO NOX
80 >80 87-99
85 77 96
39 45 98
83 95
90
54
88 95
52 56
90
54 72
>BO >50
76 58
69 90
72
90 100
75
77 63
62 68
70
71 -
83 76
69 81
92 81
61
93 80
55
70 95
Test Conditions
Average
_. b d Catalyst Space
TEL Engine Temperature Velocity
(ml/gal) Duration0 Type (°C) (hr"1)
present 34lhr 8 cyl 425-650 1000-20000
0 3 hr CFR 485 10000
3 120 hr CFR - 10000
3 100 hr CFR - 9000
present - 2 cyl 380 6900
present 350 hr 8 cyl
present . 8 cyl 285 3100
present 12000 mi 8 cyl
-1.6 238 hr 8 cyl 480 12000
12 50 hr CFR - 9000
-
3 - CFR 510
12 60 hr CFR
0 1 cyl
25 40 hr 1 cyl 440 13200
2. 2 1 1000 mi 8 cyl -
present 600 hr 1 cyl
12 75 hr CFR - 10000
3 B cyl
2
2 7 188 hr 1 cyl -
3 40 hr 8 cyl 435
3 - - -
present 12000 mi 8 cyl
48 - 8 cyl
3 40 hr
0 2 10000 mi 8 cyl -
Notes
Air added for He and CO conversion
FTC"
Odorous exhausi subsequently
removed by catalytic oxidation
over 1 Pd/Al2O3
O-(7 3 vol %) added to exhaust
FTC"
High thermal stability, high
attrition resistance
FTCd
Air added to exhaust
0 12% S in fuel, multilayer
catalyst
In presence of 3 TEL He conv. =
30 vol %
Air added to exhaust
FTCd, odorous exhaust
Air added to exhaust
FTC , air added to exhaust
FTC . air added to exhaust
FTC . nonuniform distribution
of Pt
FTC"
Class-fiber thread and fiber sup-
port, air added to exhaust
a. At end of test period d CFR Committee on Fuel Research engine, FTC Federal lest cycle
b TEL tetraethyl lead e Mn/Cu/Ni/Cr 4/1/1/6 (mole ratio)
c Hr hours, mi road miles f Cirdler catalyst C- 13
-------
Table 4-5. Catalytic Removal of HC, CO, and NO --Road Tests (from Ref. 4-21)
Catalyst Composition
O2/A12O3
Exhaust Gas Composition,
(ppm)
HC CO NOX
- -
.
-
695 14400
303 11000
264 10700
264
-
6900-7300 40000-50000 -
290 70000
1650 6000 29
289 1240
Catalytic a
(vol ft)
HC CO NOX
>80 >80 >80C
60 90
>80 >90
53 76
66 57
27 0 -
17 32
17
55 76
75 75
75 100
84 100 50
37 40
Test Conditions
Catalyst
TEL Duration Engine Temperature
(ml/gal) (miles) Type (°C)
present 15800 8 cyl >250
present - 8 cyl >250
present 5000 6 cyl >250
3 10000 8 cyl 200-900
2 9 15000 8 cyl 180-460
2 3 12000 8 cyl 550
2 3 12000 8 cyl 590
present 12000 8 cyl 510
3 11300 8 cyl
1 9 1250 6 cyl ~710
present 1000 8 cyl 510
2400 - 250-900
05 18000 8 cyl
(
Notes
Conversion at 50 mph, air injected into
exhaust
Conversion at idling speed, air injected
into exhaust
0. 07 wt % S in fuel
0 04 wt % S in fuel
FTC , catalyst on steel-wool substrate
Odorous exhaust, FTC , catalyst on steel-
wool substrate
FTC .two-stage catalyst, first stage for Pb- removal
Air injected into exhaust gas, conversion
at 60 mph, 0 07ft S in fuel
Engine at idling speed
Air injected into exhaust, conversion
(after 1000 mi on leaded fuel)
2- stroke engine, catalyst stable to
1200°C
catalyst activity sensitive to
Pb-content of exhaust
a At end of test period
b Cr/Mn/Cu/Ni = 6 2/4/1/1 (mole ratio), compare test for details.
c. Data from stationary tests after 341 hours of engine operation, catalyst temperature 515°C, conversion in absence of added air
d. FTC federal test cycle.
it*.
I
to
oo
-------
4.2.3.2.2 NOX and Dual-Bed Catalysts
Efforts to develop a catalyst which will decompose NO in the presence of
excess oxygen have been singularly unsuccessful because the reaction rates
are too slow (Ref. 4-26). It has been found that a number of catalysts will
promote the reduction of NO by the CO and HZ present in an oxygen-deficient
(reducing) atmosphere. This is accomplished by operating the engine at a
rich air-fuel mixture. Under these conditions and in the presence of a
suitable catalyst, NO is converted resulting in nitrogen (N,), CO, and H,O.
Lf L* C,
Some of the catalysts used for the oxidation of HC and CO can also reduce
NO, if operated in the reducing conditions required for a NO catalyst. In
addition, bulk metals such as Monel, copper, or stainless steel have been
used (Ref. 4-24). Since the NOx catalyst requires a reducing atmosphere
whereas the HC/CO catalyst requires an oxidizing atmosphere, if used
together they must be used in series in the exhaust system so that the
exhaust conditions to each can be controlled. The two can be separated by
some distance or be located within the same housing, commonly called a
"dual-bed" catalyst. Warmup time of HC/CO catalysts favors the latter
arrangement.
One of the major problems with NO catalysts has been the formation of
X.
ammonia (NHg) which in itself is an objectionable exhaust product. Most of
the NH3 generated in the NOx catalyst is re-oxidized in the HC/CO catalyst
portion of the dual-bed reactor to NO and H_O, thus defeating the purpose
of the NOx catalyst. A major effort in NOx catalyst research has been to
find one which not only has good NO conversion efficiency but one which
Jt
also produces minimum NH3 and has satisfactory durability.
4.2.3.2.3 HC/CO/NOx-Tricomponent Catalyst
Theoretically, it should be possible to combine the functions of NO reduction
and HC/CO oxidation in a single catalyst. At least one company (Ref. 4-23)
has tested a tricomponent catalytic converter with some success. However,
4-29
-------
the conversion efficiency is very sensitive to the air-fuel mixture and
variations as little as ±0.1 A/F units could substantially affect performance
(Ref. 4-27). Since control of the air-fuel mixture to this level has not been
demonstrated, not much attention is currently being directed to this
approach.
4.2.3.3 Other System Components/Factors
The catalytic converter is only one component of the emission control sys-
tems being considered to meet the 1975-76 standards. The performance
and life of the catalyst are dependent on engine operating conditions and
other emission control components. In turn, the performance and charac-
teristics of the catalytic converter affect the complete system.
The HC/CO catalysts are basically oxidation catalysts and require excess
air to operate efficiently. If the engine operates at a rich mixture ratio
an air pump is required to provide secondary air to the HC/CO catalyst.
Most HC/CO and NO catalysts have operating temperature limitations of
X
approximately 1400°F-1500°F for long-life durability. In the event of an
engine malfunction, such as spark plug misfire, where large quantities
of unreacted fuel can reach the catalysts, higher temperatures (>1800°F)
can be readily achieved. Since most catalytic materials undergo rapid
deterioration at temperatures above 1800°F, a thermal control system may
be required to prevent damage to the converter.
The area of catalytic converter overtemperature protection is very impor-
tant because of: (1) the potential for converter burnup and vehicle fires if
not protected, and (2) the need to preserve the emission control capability
of the converter as an excessive temperature excursion may destroy the
catalyst. The design of such control devices has not been finalized by
industry at this time. In one approach a bypass flow loop is provided around
the catalytic converter. This requires a sensing element and a hot-gas
4-30
-------
valve, perhaps remotely located relative to one another. Another approach
uses a similar overtemperature sensing device to shut off the engine and thus
protect the catalyst. A programmed protection system utilized by Ford in
the IIEC Program (Ref. 4-28) is illustrated in Figs. 4-15 and 4-16; as can
be seen this results in a complex control system.
In view of the complexity and high cost of such overtemperature control sys-
tems, efforts are in progress to develop a catalyst with sufficiently high-
operating-temperature capability to eliminate the need for such a system.
The location of the catalytic converter, relative to the exhaust manifold,
is also important. Because of cold start requirements, it must be close
enough to the manifold exhaust to warm up quickly, but be far enough
removed to prevent overheating of the catalyst.
4.2.3.4 Fuel Economy Characteristics
Fuel economy is affected to some degree by a number of parameters related
to the catalytic converter and its operation, including backpressure buildup,
selected engine air-fuel ratio, and power requirement of the secondary air
pump. The fuel economy penalties resulting from the higher backpressure
and the air pump are generally small.
The backpressure is a function of the catalyst design. Dual-bed catalyst
data shown in Fig. 4-17 (Ref. 4-24) indicate a wide range of pressure drops,
with one design approaching that of a standard muffler. The corresponding
designs are shown in Figs. 4-18 and 4-19, with Fig. 4-19 representing the
later design. The effect on backpressure of a single-bed converter installa-
tion (Fig. 4-20), compared to a standard vehicle without a converter, is
shown in Fig. 4-21 (Ref. 4-22).
In the same program (Ref. 4-22), a dual-bed (parallel) installation was oper-
ated with a special carburetor providing a rich air-fuel ratio of 11.5. A
fuel consumption of 11.8 mpg was observed for the 2500-mile driving schedule.
4-31
-------
LOGIC SENSOR
FUNCTION
Prevents by-pass until water
'temperature indicates a warmed-
up engine
^Activates by-pass at high vehicle
speeds if water temperature
sensor permits
Activates by-pass when engine
load approaches wide-open-
throttle for an extended period of
time if water temperature permits
. Activates by-pass should catalyst
bed reach an over-temperature
condition
PURPOSE
Does not allow by-pass of exhaust
gas during cold (choking) engine
operation
Protects catalyst from high
temperature, high exhaust flow
rates at cross-country turnpike
speeds
Protects catalyst from high
temperature, high exhaust flow
rates occurring during extended
heavy load non-urban operation
Protects catalyst from high
temperature should engine
malfunction. (Shorted plug,
sticking choke)
EXHAUST FLOW
BY PASS VALVE
Fig. 4-15. Ford Programmed Protection System--Logic Schematic
(from Ref. 4-28)
INTAKE VACUUM SENSOR
ACTIVATION MOTOR DRIVE CABLE
BY-PASS EXHAUST LEG
BY-PASS VALVE
117 CU. IN. CONVERTER
(CATALYST "A")
CATALYST BED •*.
OVER-TEMPERATURE SENSOR
Jit • 4
Fig. 4-16. Ford Programmed Protection System (Vehicle No. 4)--
Catalyst Container and PPS Hardware (from Ref. 4-28)
4-32
-------
PHASE I - 465 CU. IN., DUAL BED AXIAL
PHASE III - 360 CU. IN., BIFURCATED DUAL BED
PHASE IV - 360 CU. IN., BIFURCATED DUAL BED
PRESSURE
DROP-IN. Hg
PHASE I
STOCK
MUFFLER
0 100 200 300 400 500 600
FLOW RATE - CFM AT 1100 F
Fig. 4-17. Converter Flow Development (from Ref. 4-24)
This represents a 25. 8 percent loss in fuel economy. Approximately 22 per-
cent of that loss is the direct result of mixture enrichment.
Data from a General Motors single-bed converter system driven 69,000 miles
indicate a 2. 8-5.4 percent loss in fuel economy tests at various driving
conditions (Table 4-6, from Ref. 4-30).
Table 4-6. Effect of Catalytic Converter System on Fuel Economy
and Performance (from Ref. 4-30)
Item
Economy, mpg
City
Highway
30, 50, 70 mph
Performance, sec
0-60 mph
0-1/4 mile
With Standard Exhaust
Without Secondary Air
14.3
16. 1
22.4, 19.6, 16.0
15.4
20.5
With Catalytic
Converter System
13.9
15.6
21.2, 19.0, 15.4
15.9
20.7
Reduction in
Performance
(%)
2.8
3. 1
5.4, 3.0, 3.7
3.2
1.0
4-33
-------
EXHAUST
GAS IN
BAFFLES —
INSPECTION
-HOLES-
SECONDARY
AIR
NOX
CATALYST BED
BAFFLE
EXHAUST
GAS OUT
HC/CO
CATALYST BED
Fig. 4-18. Dual-Bed Axial-Flow Converter (from Ref. 4-24)
SECONDARY AIR
EXHAUST GAS
INLET
HC/CO
BED
HC/CO BED
Fig. 4-19. Bifurcated Dual-Bed
Catalytic Converter
(from Ref. 4-24)
EXHAUST GAS OUTLET
4-34
-------
AIR INTAKE PIPE
ENGINE EXHAUST PIPE
MIXING CHAMBER
Fig. 4-20. Cutaway Sketch of Oxy-Catalyst Converter (from Ref. 4-22)
SINGLE BED-
ASPIRATING SECONDARY
AIR
STOCK MUFFLER
INSTALLATION
10
20 30 40 50 60 70
CARSPEED.mph
Fig.
4-21. Exhaust System Backpressures
Road Load Conditions (from
Ref. 4-22)
4-35
-------
Based on the available information, an estimated 2-4 percent reduction in
fuel economy may be attributed specifically to the catalyst bed installation
and the presence of the secondary air pump.
4.3 SPECIFIC EMISSION CONTROL SYSTEMS
Various combinations of the foregoing emission control devices have been and
are undergoing extensive evaluation by the automakers and other organizations
to explore every possible avenue for meeting the 1975-76 emission standards.
The four U.S. automobile manufacturers stated to the EPA Administrator in
April 1971 (Refs. 4-6, 4-31, 4-3Z, and 4-33) that at that time they could not
meet the standards based on the progress to date, variabilities in production
tolerances, etc. Although a great deal of effort has been expended since that
time, the automakers still have not demonstrated meeting the emission
standards, including the 50,000-mile durability capability.
It is not the purpose of this report to judge the expertise of the auto industry,
but rather to review the existing emission control system technology base,
and make an assessment as to which general approaches appear promising to
meet the 1975-76 standards. To this end, and to provide some order for the
numerous control device combinations possible, the emission control
systems are discussed in the following generic classes:
1 . Catalytic Converter Systems -- those systems primarily based
on some form of catalytic converter and not including special
warmup devices
2. Thermal Reactor Systems -- those systems primarily based on
some form of thermal reactor
3. Combination Systems -- those systems primarily based on some
form of thermal reactor in combination with some form of
catalytic converter
4-36
-------
4. 3. 1 Catalytic Converter Systems
There are four distinctive subclasses of catalytic converter systems:
1. HC/CO Catalytic Converter Alone (no EGR)
2. HC/CO Catalytic Converter plus EGR
3. Dual Catalytic Converter plus EGR
4. Tricomponent Catalytic Converter (no EGR)
4.3.1.1 HC/CO Catalytic Converter Alone (no EGR)
This emission control system concept is characterized by the addition of
HC/CO catalytic converter units and secondary air injection (air pumps) to
conventional engine systems (no EGR). The available data pertaining to this
approach are those provided by Universal Oil Products (UOP) and by
Engelhard Industries. The primary goal of these catalyst manufacturers is
to develop catalysts for supply to the automakers and, consequently, the
bulk of their effort is directed to characterizing catalyst materials with regard
to effectiveness and durability and not to developing emission control systems,
per se. In their characterization activities, then, the catalyst suppliers have
necessarily investigated catalysts with varying amounts of active catalyst
material. It is not known whether the emission levels given below corre-
spond to catalysts under serious consideration for use by the automakers, in
terms of the amount of active material used in the catalyst and its necessarily
attendant cost implications.
In Reference 4-23, UOP described the installation of their "mini-converters"
on a domestic V-8 engine (Fig. 4-22). It was stressed that this was a stock
vehicle purchased from a local dealer and that carburetion, ignition timing,
valve timing, etc. , were as delivered from the factory. Typical emission
results (from Ref. 4-23) are shown in Tables 4-7, 4-8, and 4-9- Table 4-7
shows CVS test values for a large domestic engine with a normal choke,
Table 4-8 shows the same type of results with a faster choke, and Table 4-9
shows results from smaller foreign vehicles. Although CO values are most
attractive, HC and NOX values exceed 1975-76 standards.
4-37
-------
SECONDARY
AIR
ENGINE
SECONDARY «|0jn
AIR
CONVERTER
CONVERTER
2
3
MODIFICATIONS TO VEHICLE
ADDITION OF TWO AIR PUMPS
ADDITION OF TWO MINI CONVERTERS
SECONDARY AIR DELAY (10 sec)
4 NO CHANGE TO CARBURETION, IGNITION TIMING, etc.
Fig. 4-22. Universal Oil Products Mini-Converter
Installation (from Ref. 4-23)
Engelhard (Ref. 4-57) quotes emission levels for their PTX-433 catalytic
converter unit (0. 2 percent Pt) to be (at the end of 50, 000 miles of the AMA
driving schedule):
HC = 0. 70
CO =3.8
NOX =5.0
gm/mi (single-bag CVS cold start test)
Initial emission values have not been reported by Engelhard.
There are, as yet, no reported data pertaining to fuel economy. Engelhard
has reported a 50, 000-mile durability test for their PTX catalyst. The test
was conducted with an unleaded gasoline having a lead content of approximately
0. 03 gm/gal. The catalyst picked up substantial quantities of Pb, Zn, P, and
Ba during the test. The Zn and Ba are contaminants Engelhard associates
4-38
-------
Table 4-7. 1971 Domestic Engine (350 CID)--Normal Choke,
PZ-195 Catalyst (from Ref. 4-23)
Converter
IN Normal Choke
Type 1
Typo 2
Type 2 (Z)
Type 1 (1)
Type 2 (2)
Type 1 (1)
CVS-I Cold Start
HC
2.89
0.62
0.68
-
0.59
CO
17 35
1.45
1.20
-
0.96
NOX
(3.29)
NO Only
3.88
-
-
(1 80)
Eat. 2.11
CVS-1 Hot
HC
-
0. 195
0.21
0.26
0.25
CO
-
0.48
0 20
1. 16
1. 10
NOX
-
3.22
-
1.75
(1.56)
Est. 1.83
Comments
Average of 2 runs
Full air L/R = 1 39/1 33
Full air
Full air (both pumps) to 3rd
reactor only
No air to 5.75 reactors
Air to reactors D 1 and Hi for
1st 100 sec of cold start
Air to reactor #3 only after
100 sec
Table 4-8. 1971 Domestic Engine (350-CID)--Fast Choke
(from Ref. 4-23)
Converter
Type 1 PZ-195
Type 1 PZ-195
Type 3 PZ-195
UOP #2303-41
PTX-5
PTX-5
Type 1 PZ-195
CVS-Cold Start
HC
0. 16
0.42
0.51
0.66
1. 15
1. 13
0.25
CO
1.33
1.21
1.94
3.02
5.77
4.32
2.58
NOx
(Z.82)
NO Only
-
4.74
-
-
5. 32
5.08
CVS-1 Hot
HC
0.035
0.18
0. 14
0 16
0.145
0.26
-
CO
0.25
0.28
0.28
0.45
0.50
0.72
-
NOX
-
-
-
-
-
5.39
-
Comments
Fresh catalyst
Aged 16, 500 miles
Fresh catalyst
Fresh catalyst
Fresh catalyst
Fresh catalyst
Normal choke. 6
cold starts run
Vehicle
Car A
Car B
Car C
Car C
Car C
Car D
Car D
CVS-1 Cold Start
HC
1.41
0.29
0.40
0.57
0.46
0.47
0.74
CO
3.01
0.99
1.20
2.65
3.86
1.22
1.58
NOX
-
1.36
-
1.86
1.77
1.73
2.00
CVS-1 Hot
HC
0.57
0.05
-
-
-
0.35
0 45
CO
4.47
0.36
-
-
-
0.36
0.37
NOX
1.34
-
-
-
-
1.66
1.71
Table 4-9. Federal Test Results
for Some Foreign
Vehicles (from
Ref. 4-23)
4-39
-------
with motor oil. Engelhard's present position is that the most probable
reasons for PTX catalyst deterioration are metal poisons that may be present
in the fuel and lubricating oils.
4.3.1.2 HC/CO Catalytic Converter plus EGR
This emission control system concept is illustrated by the Ford Package "B"
system (Ref. 4-29). Its major components, shown in Fig. 4-23, include:
1. Dual bifurcated axial-flow converters
2. Single rear bifurcated axial-flow converter
3. Programmed protection system to divert exhaust gas around
the first catalyst bed when its temperature exceeds 1350°F
4. Spherical transition metal "pre-attrited" catalyst pellets
5. Secondary air injection just below the exhaust flange
6. Modified distributor with warmup spark retard and reduced
part-throttle advance
7. Spacer-entry EGR
Front and rear converters were used in this concept approach because it was
felt that it may not be possible to accomplish successful development of
transition metal catalysts capable of withstanding temperature excursions
above 1400°F. This configurational approach is dependent upon the ability of
the normal engine exhaust system to provide adequate rapid warmup of the
catalytic converters; hence, the placement of the forward converters is close
to the exhaust flange.
4. 3. 1. 2. 1 Emission Level Characteristics
Typical emission levels for the Ford Package "B" concept are (Ref. 4-29):
HC = 0.8
CO = 11.0 gm/mi (single-bag CVS cold start test)
N
-------
Chrysler (Ref. 4-6) reports emission levels of:
gm/mi (single-bag CVS cold start test)
HC = 0.24
CO = 7.2
NOX = 2.03
for a similar type HC/CO catalytic converter plus EGR system.
4. 3. 1.2.2 Fuel Economy Results
A fuel economy penalty of 8 percent on the chassis dynamometer is reported
for the foregoing Ford Package "B" emission levels.
4. 3. 1. 2. 3 System Lifetime Characteristics
No durability or lifetime test data are reported for this concept.
4.3.1.2.4 Effect of Lead Additives
Concept development of the Ford Package "B" system is being made with
unleaded fuel. See Section 5 for a discussion of leaded-fuel effects.
4.3.1.3 Dual Catalytic Converter plus EGR
This emission control system concept is illustrated by the Ford Package "C"
system (Ref. 4-29). Its major components and operating features, shown in
Fig. 4-24, include:
1. IIEC "bifurcated dual-bed" converter (210 in. 3 NO bed;
150 in. 3 HC/CO bed) (see Fig. 4-19)
2. Engine-driven secondary air pump
3. Integral, below-the-throttle EGR system
4. Programmed catalyst protection system
5. Carburetor enriched 5 percent
6. MBT ignition timing with retard during warmup
This system is being explored by Ford in the IIEC Program because it theoret-
ically has the potential for minimal loss in fuel economy as compared with
RTR plus EGR and HC/CO catalytic converter plus EGR systems. The minimal
dependence on EGR (for NOX reduction) also allows improvements in vehicle
driveability.
4-41
-------
PROGRAMMED PROTECTION SYSTEM
FOR FRONT CONVERTERS
SPACER-ENTRY EGR
REAR
CONVERTER/MUFFLER'
BIFURCATED SINGLE BED
CONVERTERS -100 CU. IN. EACH
CATALYST "AZ"
SECONDARY
AIR PUMP
Fig. 4-23. Ford Concept Emission Package "B" (from Ref. 4-29)
NON-LEADED FUEL
PROGRAMMED PROTECTION SYSTEM
FOR FRONT CONVERTER
SPACER-ENTRY
EGR
POTENTIAL POSITION
OF SECOND DUAL-BED
CONVERTER
DUAL-BED
CONVERTER
SECONDARY AIR PUMP
Fig. 4-24. Ford Concept Emission Package "C" (from Ref. 4-29)
4-42
-------
Unleaded fuel was established as an absolute necessity with this package
because of the observed rapid depreciation in NOX reduction efficiency of
NO catalysts operating on leaded fuels (Ref. 4-29). See Section 5 for a
Ji
discussion of leaded-fuel effects.
4. 3. 1. 3. 1 Emission Level Characteristics
Typical emission levels for the Package "C" concept are (Ref. 4-29):
HC = 0. 85
CO = 10.00
gm/mi (single-bag CVS-1 cold start tests)
NOX = 0. 90
4.3.1.3.2 Fuel Economy Results
A fuel economy penalty of 5 percent was reported for Package "C" (Ref. 4-29)
on the CVS-1 test cycle.
4.3.1.3.3 System Lifetime Characteristics
Durability testing and continued development of the major components in this
package are reported to be under way by Ford (Ref. 4-29). No data in this
regard are reported at this time.
4.3.1.4 Tricomponent Catalytic Converter (no EGR)
In principle, this concept has a three-way catalyst bed for simultaneous
reduction of HC, CO, and NOX and is theoretically extremely attractive. The
principal proponent of this approach has been UOP (Ref. 4-23).
UOP provided a 1970 Volkswagen equipped with its catalytic converter to
APCO for test evaluation. This vehicle was equipped with a 98 CID engine and
automatic transmission. The stock fuel injection was modified to prevent
cutoff of fuel during deceleration and the catalytic unit was installed in place
of the standard muffler. Results of the tests are shown in Table 4-10
(Ref. 4-34).
Further investigations by UOP determined that optimum nitric oxide conver-
sion is obtained (about 90 percent) within a narrow range of air-fuel ratio on
4-43
-------
Table 4-10. Emission Test Results--1970 Volkswagen
with UOP Catalytic Converter (1972
Federal Test Procedure)(from Ref. 4-34)
Emissions
HC
CO
co2
NO/
**
N0x
Emission
Test
2.3
32
444
1.3
0.6
NO box results reported as
##
Saltzman results reported as
Levels (gm/mi)
1 Test 2
1.4
10
431
1. 1
0.8
N02
100
o
LiJ
0
LEAN
STOICHIOMETRIC
AIR-FUEL RATIO
HC
RICH
Fig. 4-25. Tricomponent Conversion vs Air-Fuel Ratio
(from Ref. 4-23)
4-44
-------
the slightly rich side of stoichiometry, as shown in Fig. 4-25. Fig. 4-25
also shows that conversion of HC, CO, and NO can be attained simultaneously.
Chrysler (Ref. 4-6) recently supplied the following tricomponent catalytic
converter data to the EPA Administrator:
Emission Levels (gm/mi)
Cold Start Hot
Emission 7-Mode 7-Mode
HC 0.68 0.27
CO 11.50 1.50
NO,, 0.71 0.33
A.
Chrysler states (Ref. 4-6) that there is not, as yet, an effective three-way
catalyst.
American Motors (Ref. 4-35) has tested a prototype three-control catalyst
supplied by a catalyst manufacturer and has also provided that same manu-
facturer with a vehicle for development of a three-component system. These
programs did not produce a catalyst with the ability to meet three-component
control to the degree necessary for 1975-76. American Motors said that the
variability of air-fuel ratio needed for safe, efficient vehicle operation
proved to be too large for the catalyst to handle. The proper balance between
the "reducing" and the "oxidizing" ability of the engine's exhaust could not be
maintained during normal vehicle usage.
No durability or fuel consumption data are available for this conceptual
approach. The effects of lead additives are addressed in Section 5.
4. 3. 2 Thermal Reactor Systems
There are three meaningful subclasses of thermal reactor systems:
1. LTR plus EGR
2. RTR Alone
3. RTR plus EGR
4-45
-------
(The LTR Alone system is not considered to be a viable system because of its
lack of adequate NOX control. )
4. 3. 2. 1 LTR plus EGR Concept
This emission control system concept is exemplified by the Ethyl lean reactor
design. It consists of a full-size LTR for HC and CO control, an EGR system
for NOX control, and advanced carburetion for engine operation at the selected
lean air-fuel ratio (approximately 17. 5) provided by a specially developed,
high-velocity carburetor. Spark advance characteristics are tailored to pro-
vide the best compromise among fuel economy, driveability, and low emissions.
Ethyl has actively pursued and demonstrated this approach with vehicle tests.
Its work has been aimed at the development of an emission control system that
is not sensitive to fuel additives. Thus all test work reported has been done
with fuel containing approximately 3 ml/gal of tetraethyl lead (TEL). In using
this fuel, the Ethyl lean reactor system avoids fuel economy penalties brought
about by lowering compression ratio to accommodate low-octane fuels. Ethyl
states that the retention of a high-compression ratio also makes it possible to
operate with good driveability at leaner mixtures than otherwise would be the
case, and minimizes problems of EGR with respect to vehicle driveability
effects.
4. 3. 2. 1. 1 Emission Level Characteristics
The most advanced versions of the Ethyl lean reactor system are now
embodied in several Pontiacs and one 1971 Plymouth (Ref. 4-5). Emissions
of two of these cars, based on the single-bag CVS test procedures in use
prior to July 1971, are shown in Tables 4-11 and 4-12. Similar data obtained
with the new (post-July 1971) three-bag CVS test procedure for the 1971
Plymouth are compared with the single-bag data in Table 4-12. As can be
seen, HC emissions are 127 percent and CO emissions are 182 percent of the
corresponding 1975-76 standards. NOX emissions are 342 percent of the 1976
standard, but are well below the 1975 standard (3 gm/mi).
4-46
-------
Table 4-11.
Ethyl Lean Reactor--Emission Data
for 1970 Pontiac (Vehicle 766)
(from Ref. 4-5)
Vehicle Description
1970 Pontiac LeMans
400 CIO Engine
Automatic Transmission
Power Steering
Power Brakes
1972 CVS Procedure
Run Date
4-5-71
4-6-71
4-19-71
4-20-71
4-21-71
4-22-71
6-3-71
6-24-71
Avg.
12-18-70
1970 7 -Mode Procedure
Run Date
4-8-71
4-13-71
4-14-71
Avg.
Equivalent gm/mi
Modifications
3-Venturi Carburetor
EGR System
Exhaust Manifold Reactor
Exhaust Port Liners
Evaporative Loss Controls
Exhaust Cooler Units
Particulate Trapping Device
Air-Injection Pump (Operates During
Choking Period)
Transmission Modifications
(Modulator and Governor)
HC CO NO
X
(gm/mil (gm/mi) (gm/mi)
0.74 7.3 1.40
0.75 7.0 1.60
0.74 5.3 1.70
0.78 6.2 1.70
0.84 6.2 1.48
0.82 5.9 1.45
0.88 6.5 1.45
0.73 6.8 1.40
0.79 6.4 1.52
0. 64 9. 1 1.09
HC CO NO
(ppm) (%) (ppm)
19 0.21 226
20 0.20 200
23 0.21 197
20.7 0.21 208
0.26 5.0 0.81
Table 4-12. Ethyl Lean Reactor--Emission Data for
1971 Plymouth (Vehicle 18M-448)
(from Ref. 4-5)
Vehicle Description
1971 Plymouth Fury III
360 CID Engine
Automatic Transmission
Power Steering
Power Brakes
Air Conditioning
1972 CVS Procedure (Sincle-bae tests)
HC
Run Date em /mi
2-26-71 1.00
3-2-71 0.74
3-8-71 0.92
3-24-71 0.82
4-8-71 1.00
Avg. 0. 89
1975 CVS Procedure (Three-bae testa)
0. 52
Modifications
3-Venturi Carburetor
EGR System
Exhaust Manifold Reactor
Exhaust Port Liners
Evaporative Loss Controls
Exhaust Cooler Units
CO
em /mi
8.0
7.3
7.6
10.0
10.0
8.6
6.2
N0x
em /mi
1.6
1.7
0.86
1. 5
1.23
1.37
1.37
-------
Chrysler (Ref. 4-6) reported similar test results (HC = 0. 7, CO = 7. 0,
NOX = 1. 3) for an LTR plus EGR vehicle using the single-bag CVS test
procedures.
With regard to the single-bag CVS test data, Ethyl states that the first
505 seconds of the 1371-second test (36. 8 percent of the time) contribute
about 78 percent of the HC, 68 percent of the CO, and 48 percent of the NOX
measured in the entire test. Thus, the strong influence of the cold start on
HC and CO emissions is evident for this system.
Further improvements which could reduce HC and CO emissions include:
1. Use of a moderate amount of air injection only during the
first few minutes of warmup operation to increase exhaust
oxidation during the choking period
2. Improvements in heat conservation in the exhaust ports and
exhaust port liners (perhaps by addition of flameholders), and
improvement in the exhaust reactor design to increase
exhaust oxidation
3. Improvements in the intake manifold to promote quicker
warmup and the need for less choking during the warmup
period
4. Alterations in transmission characteristics to accelerate
warmup
5. Use of higher compression ratio to permit still leaner
mixtures and better utilization of EGR, and to produce lower
exhaust gas volumes with consequent reduction in mass
emissions
6. Charcoal absorber traps to reduce HC exhausted during
engine startup
The foregoing are logical technical approaches, but until they are incorpo-
rated and demonstrated the LTR plus EGR emission control system concept
is considered deficient with regard to meeting the 1975-76 HC and CO
standards and the 1976 NX) standard.
4-48
-------
4.3.2.1.2 Fuel Economy Effects
Ethyl has reported fuel economy test results for the aforementioned
Plymouth and Pontiac lean reactor cars (Ref. 4-5). Two test routes were
used to measure fuel economy under consumer driving conditions. Charac-
teristics of the routes are:
1. City and Expressway Route--27. 7-mile loop, 10 stops per
loop; average speed of 36. 7 mph
2. City Route--18. 4-mile loop, 40 stops per loop, average
speed of 23. 4 mph
Table 4-13 compares the results obtained on these test routes with current
lean reactor cars and their nonmodified production counterparts. The noted
economy losses occurred because of the substantial amounts of EGR used;
earlier versions of lean reactor cars without EGR showed little or no loss
in fuel economy in comparison with the corresponding nonmodified car.
4.3.2.1.3 System Lifetime Characteristics
An earlier version of an Ethyl modified 1966 Pontiac was driven throughout
the United States for over 20, 000 miles while being used for a series of
demonstrations. Modifications were similar to those in cars now in use
except that this car was not equipped with EGR and had a less effective
thermal reactor.
Another Pontiac embodying current modifications, among which were EGR
and improved thermal reactors, accumulated over 30, 000 miles in various
types of service including cross-country trips. This car was reported by
Ethyl to have demonstrated excellent durability characteristics and emissions
stability.
One modified 1970 Pontiac was supplied to GARB in November 1970 for testing
and use in general fleet service. Because CARB measured emissions from
this vehicle only by the 1970 test procedure, the emission results cannot be
directly related to data on other cars obtained by the single-bag 1972 CVS
procedure. The California car has the longest uninterrupted history for
4-49
-------
Table 4- 13. Ethyl Lean Reactor--Fuel Economy
Modified and Standard Cars (from
Ref. 4-5)
Item
Average Speed
Stops per Mile
1971 Plymouth Fury III. 360 CID
Standard Car
Modified Car
Car A
Car B
Economy Loss
Avg
1970 Pontiac LeMans, 400 CID
Standard Car
Modified Car
Economy Loss
Avg.
City Route
23 mph
2.17
11.1 mpg
11.0 mpg
11.1 mpg
0.5%
6.6%
11.5 mpg
10.6 mpg
7.8%
8.6%
City and
Expressway
36 mph
0.36
16. 7 mpg
14. 5 mpg
14.7 mps
12.6%
14. 9 mpg
13.5 mpE
9.4%
Ethyl lean reactor vehicle modifications without changes or updating. The
total mileage of this vehicle at the last test point was 12, 000 miles: the
California test accounts for 8000 miles, and an additional 4000 miles were
accumulated during its trip to the West Coast and its preliminary testing
period. Emissions of this car have good stability; results are shown in
Table 4-14 and plotted in Fig. 4-26.
The basic nature of a lean reactor system would predict less difficulty in
obtaining satisfactory durability than would be the case with a rich reactor
system. This is because exhaust gas entering the reactor from the lean
engine contains about 100 ppm HC, 0. 1-0.4 percent CO, and approximately
2-4 percent O2 (without an air pump). Therefore, little chemical heat is
generated in the reactor and its temperature is governed by the degree to
4-50
-------
Table 4-14. Ethyl Lean Reactor—Emission Data for Modified
Pontiac (No. 761) Supplied to GARB (from
Jlef._4-5)
Date
11-19-70
1-26-71
4-1-71
5-18-71
Approximate
Miles in California
0
3,000
6, 000
8, 000
Emissions (gm/mi)
HC
0.60
0.47
0.51
0.42
CO
8.34
7.87
8. 11
8.8
NO
X
0.72
0.48
0.77
0.79
Note: GARB Laboratory measurements by 1970 equivalent mass
method
which the sensible heat in the exhaust gas is conserved. This means that
the lean reactor operates in a temperature range of 1400°F to 1600°F, even
under high-speed turnpike conditions, which is a range that good-quality
stainless steels should tolerate well. Moreover, tests by Ethyl (Ref. 4-5)
indicate that the lean reactor is not subject to destructive temperature excur-
sions, even with a continuously misfiring spark plug. Thus, durability
should not be seriously decreased by situations in which engine malfunctions
should occur.
Harmful deposits are also a consideration in system lifetime. However, the
more advanced modulating EGR system now used on lean reactor cars was
found to be free of such deposits after 12, 000 miles of service on the car in
GARB service, and was tested successfully for 30, 000 equivalent miles on
the dynamometer.
Therefore, even though a system life of 50, 000 miles has not yet been
demonstrated, there appears to be no fundamental reason why it could not be
achieved by an LTR plus EGR emission system concept.
4-51
-------
co
CO
UJ
2.0
1.5
1.0
0.5
0
30
20
10
0
2.0
1.5
1.0
0.5
( 1971 LIMIT 2.2 )
1975 CALIF PROPOSED LIMIT
1971 LIMIT 23
1975 CAUF PROPOSED LIMIT
(1971 LIMIT 4.0)
1975 CALIF. PROPOSED LIMIT
3000
6000
8000
MILES
Fig. 4-26. Ethyl Lean Reactor Emissions of Modified Pontiac
(No. 761) Supplied to GARB (Measurements by 1970
Equivalent Mass Method) (from Ref. 4-5)
4-52
-------
4. 3. 2. 1. 4 Effect of Lead Additives
As mentioned above (4. 3. 2. 1), all tests performed by Ethyl with lean
reactor cars have been with fully leaded fuels and no adverse effects have
been observed on the thermal reactor, per se. Ethyl does recognize that
deposits in the EGR system can be expected to result from the decomposition
of fuel and lubricant additives, from tars and carbonaceous matter produced
during combustion, and from ferrous oxides from exhaust system parts. In
addition, water condensate could be an important factor in promoting deposits.
Ethyl found (Ref. 4-5) that the utilization of self-cleaning EGR orifice designs
(plungers, specially coated surfaces, flexible snap-rings, etc. ) in areas
susceptible to deposit buildup was a practical method to negate deposit plugging
and loss of EGR effectiveness.
4.3.2.2 RTR-Alone Concept
This emission control system concept is exemplified by the Esso Modified
Rapid Action Manifold (RAM) thermal reactor design (Ref. 4-2) and the Ford
Type J thermal reactor test program (Ref. 4-3). Other RTR performance
and durability test programs by Esso, Du Pont, Ford, and other IIEC
members are more suitably related to the RTR plus EGR subclass and are
discussed in the next section (4. 3. 2. 3).
The RTR-Alone system controls NOX by fuel-rich carburetion and (in some
instances) spark retard. CO and HC, derived from the fuel-rich engine, are
mixed and burned with injected secondary air in the reactors and exhaust pipe.
The Esso RAM thermal reactor (Fig. 4-27) consists of a torus made of
Type 310 stainless steel. Connecting arms lead exhaust gases from the
engine to the torus. The gases flow around the torus and exit through a slot
into a central plenum and then into the exhaust pipe. The slot is positioned so
that the gases must flow at least half-way around the torus before they can
leave, and so that a portion of the circulating gases goes all the way around
to mix with the entering engine exhaust. Air is injected into each engine
exhaust port and is aimed toward the valve. Most of the thermal reaction
4-53
-------
CLEAN
EXHAUST
EXHAUST
EXHAUST
EXHAUST
FLAME
HOLDER
EXHAUST
Fig. 4-27. Esso Rapid Action Manifold (RAM)
Reactor (from Ref. 4-2)
EXHAUST GAS INLET
EXHAUST
GAS
OUTLET
Fig. 4-28. IIEC Type H. Exhaust Manifold Reactor (V-8 Engine)--Small
Volume with Concentric Core Design (from Ref. 4-3)
4-54
-------
takes place as the gases swirl through the reactor. Any CO or HC not
burned in the reactor continues to react in the exhaust pipe after it is heated.
Flameholders are located at the exit of each engine exhaust port; they act
to stabilize the flame at the exhaust port outlets during startup, when the
engine is choked. Once the choke is open there is insufficient fuel to maintain
a flame at the flameholders, but by this time the reactor proper is hot and
the flame is held there.
The Ford Type J thermal reactor, although not fully described in the
literature (Ref. 4-3), is stated to be essentially equivalent to the Type H
smaller volume (97 in. ) series of IIEC exhaust manifold reactors (Type H,
Ref. 4-3) shown in Fig. 4-28. It is stated that a one-piece shell reactor
core was used with thermal growth provisions at the core inlet neck areas
(Ref. 4-3). The core was constructed of Inconel 601 material.
4. 3. 2. 2. 1 Emission Level Characteristics
Esso test results for the Modified RAM system (Ref. 4-2) are shown below:
Emission Levels (gm/mi)
Results /Standards CO HC NOX
Modified RAM Results 4. 2 0. 07 1. 89
1975 U.S. Standards 4.7 0.46 3.00
These are single-bag CVS test data. As indicated, HC and CO values are
lower than 1975-76 standards, and NOX values are higher than 1976 standards
(0. 4 gm/mi).
The Ford Type J thermal reactor test data (Ref. 4-3) are shown in Fig. 4-29.
All emissions (HC, CO, NOX) are above 1975-76 standards, except for NOX
which is below the 1975 standard of 3 gm/mi (running approximately 0. 5 to
0. 7 gm/mi).
4.3.2.2.2 Fuel Economy Results
Several tests to measure fuel economy were made by Esso with the Modified
RAM system (Ref. 4-2). A 2-hour city driving cycle was used with stop-and-go
driving, 28-mph average speed, and cruises up to 45 mph. Some turnpike
4-55
-------
VEHICLE DESCRIPTION
VEH.: 302-2V AUTO. GALAXIE AIR PUMP: HI CAPACITY
DIST.: 70F37 W/DIST-0-VAC REACTOR PROTECTIVE SYSTEM
CARB.: DOAF-U MOD. RICH LINER MATERIAL: INCONEL - 601 (.03 IN. THICK)
FUEL: (0/.5 GM/GAL TEL)
HC-
GM/MI
CO-
GM/MI
NOX-
GM/MI
MILES
.80
.40
20
0
20
15
10
5
0
1.0
.9
.8
.7
.6
.5
.4
IEC
(1)
(2)
(3)
NEC GOAL - 7.1 GM/MI
NEC GOAL-0.68 GM/MI
20,000
40,000
60,000
(1) CARB. CLEANED & TEST RE-RUN (2) INSUFFICIENT SECONDARY AIR -
AIR PUMP REPLACED. (3) INSUFFICIENT SECONDARY AIR - AIR PUMP REPLACED.
Fig. 4-29. Ford Type J Reactor Durability and Cold Start
Emissions Data (from Ref. 4-3)
tests were also made, lasting 1 hour, at an average speed of 58 mph. The
results, showing fuel economy compared with the base car with the original
carburetor, are given below:
% Fuel Economy Debit
City Driving 16. 9
Turnpike 9. 5
No fuel economy test results were reported (Ref. 4-3) for the Ford Type J
thermal reactor tests.
4.3.2.2.3 System Lifetime Characteristics
Esso tests of the Modified RAM concept were demonstrative only; no dura-
bility tests have been made. If this concept were tested for durability, a better
material than the 310 stainless steel used in the demonstrator model would be
required.
4-56
-------
The Ford Type J reactor was installed in vehicle for durability testing.
As reported in Ref. 4-3, 62,000 miles of heavy-duty operation were
accumulated on the reactors at the time of writing (January 1971). Although
failures were experienced on ancillary components as mileage was
accumulated, the thermal reactor system continued to control emissions
to essentially IIEC target levels when all emission component subsystems
were operational (see Fig. 4-29 for notation of carburetor and air pump
problems). This vehicle was equipped with a reactor overtemperature
protection system which limited peak temperatures to 1850°F.
A recent communication (Ref. 4-36) with Ford indicated that 90, 000 miles
of durability operation has been completed. At test termination, a number
of heat cracks in the liner were found and a hole had developed in one small
area. Although a loss of performance was observed, the system was still
performing relatively well.
4.3.2.3 RTR plus EGR Concept
This emission control system concept is exemplified by the Du Pont thermal
reactor system (Ref. 4-8), the Esso RAM thermal reactor (Ref. 4-2), and
the Ford Package "A" system (Ref. 4-29). HC and CO emissions are con-
trolled as described in Section 4. 3. 2. 2 for the RTR-Aloneconcept; EGR is added
for NOX control (see Section 4. 2. 2) over and beyond that afforded by rich
engine operation.
The configurational aspects of the Esso RAM reactor and Ford-type reactors
were described in Section 4. 3. 2. 2. The Du Pont-type reactor is of conventional
configuration (as shown in Fig. 4-2).
In the case of the Esso RAM system (Ref. 4-2), EGR in the amount of approxi-
mately 12 percent of engine intake air was used for maximum NOX control.
It was taken in the vicinity of the muffler, passed through finned tubing for
cooling, and introduced into the carburetor above the throttle plates. EGR was
not used at throttle positions below 20-25 mph cruise because it increased CO
and HC emissions. Also, it was not used during warmup because it prevented
4-57
-------
a flame from being established quickly. The use of EGR resulted in about
a 50-percent decrease in NOX emissions from the level achieved by fuel-
rich engine operation.
The EGR system utilized in the Du Pont thermal reactor system concept
(Ref. 4-8) is basically similar to that used by Esso. Recirculation is shut
off at (1) idle to give smooth engine operation, and (2) at WOT conditions to
prevent loss in maximum vehicle performance. The recirculation rate
employed has been varied, with the most recent Du Pont system employing
an approximately 18 percent recycle rate.
The EGR system utilized on the Ford concept emission Package "A" design
(Ref. 4-29) is a below-the-throttle recycle injection system, with recycle
pickup taken before the muffler. The particular rate of recycle is not given.
4. 3. 2. 3. 1 Emission Level Characteristics
Esso test results for the RAM system (Ref. 4-2) are shown below:
Emission Levels (gm/mi)
CO HC NOX
3.7 0.08 0.72
These are single-bag CVS test data. As indicated, the CO value is slightly
below the 1975-76 standard, the HC value is considerably below the 1975-76
standard, and the NOX level exceeds the 1976 standard.
More recent single-bag and three-bag CVS tests were reported by EPA
(Ref. 4-37) to give the following emission value ranges for a RAM-equipped
1971 Ford LTD:
Emission Levels (gm/mi)
Test Type CO HC NOX
Single-bag tests 3.80-5.90 0.14-0.20 0.60-0.65
Three-bag tests 3.19-4.76 0.10-0.11 0.67
4-58
-------
The most recent Du Pont thermal reactor system is quoted by Du Pont
(Ref. 4-8) to have the following emission levels (single-bag CVS test
procedure):
HC = 0.05
CO = 9.20
NOX = 0.52
gm/mi
The Ford Package "A" single-bag CVS test results were (Ref. 4-29):
gm/mi
HC = 0. 30
CO = 9.00
NOX = 1.40
Chrysler Corporation (Ref. 4-6) reported single-bag CVS laboratory test
data emission values for the RTR plus EGR concept of:
HC = 0.23
gm/mi
CO
NO,
= 13.80
= 0.45
4.3.2.3.2 Fuel Economy Results'
Tests were made by Esso to measure fuel economy for the RAM system.
They were run as described for the Modified RAM system in Section 4. 3. 2. 2. 2.
Specific results, showing fuel economy compared with the base car with the
original carburetor, are given below:
% Fuel Economy Debit
City Driving 22.4
Turnpike 17.4
The recent Du Pont thermal reactor system (described above) is stated
(Ref. 4-8) to have a 21-percent fuel economy loss under city-suburban driving
conditions.
The Ford Concept "A" system (Ref. 4-29) had a 20-percent fuel economy loss
on the chassis dynamometer, and an 18-percent loss under city-suburban
driving conditions.
4-59
-------
Chrysler (Ref. 4-6) estimated a fuel economy penalty of approximately
30 percent for the RTR plus EGR concept evaluated by them.
4. 3. 2. 3. 3 System Lifetime Characteristics
Esso tests of the RAM concept were demonstrative only; no durability tests
have been made.
The Ford Package "A" concept also was demonstrative only. However, the
durability test program for the Ford Type J reactor (Section 4. 3. 2. 2. 3) would
be equally applicable to the reactor portion of the Package "A" concept. As
mentioned, the final results of this durability program have not been
published or released by Ford.
Du Pont supplied six cars to CARB in the fall of 1970 for its evaluation in a
two-year program. These cars were 1970 Chevrolets with 350 CID engines
and automatic transmissions, and were equipped with the Du Pont particulate-
trapping system as well as thermal reactor and EGR. The six test cars,
along with six production vehicles for comparison, were assigned to the State
Motor Pool in California for normal driving service by state employees. In
June 1971, the average of the odometer readings of the six vehicles was
17, 954 miles. As the vehicles had about 3000 miles of operation prior to
incorporation of the emission control system, about 15,000 miles of durability
testing of the emission control system were actually logged.
Near the end of August 1971, a failure of a timing chain occurred in one of
the six test vehicles. The failure was described as an elongation of the timing
chain, which eventually caused a hole to be rubbed in its cover. None of the
six vehicles in the control fleet was affected.
Du Pont (Ref. 4-38) states that similar wear was observed in three of the six
CARB test cars. Symptoms of similar wear had been previously detected in
three reactor vehicles tested by Du Pont. Timing chain pins, cam followers,
rocker arms, and valve guides were affected. Du Pont is convinced that the
wear problem is due to the lapping action of small (0. 02-0. 05 micron) metal
4-60
-------
oxide particles mixed in the engine oil. These small particles come from
the reactor core and find their way through the EGR line to the lubrication
system (presumably by entering the intake valve ports and then passing
through the piston rings and/or exhaust valve guides).
Severe oxidation of the reactor core 310 stainless steel material was
demonstrated in Du Pont tests of two reactors which lost 0. 5 pound of core
weight (23 percent) after 20, 000 miles of testing. Du Pont feels that the
wear problem could be overcome by using a material such as Inconel 601
in the reactor core. Since oxidation of any part of the exhaust system is a
potentially similar hazard, a more complete solution would be to use an EGR
gas source upstream of the thermal reactor.
Because of these problems, the GARB test program has been discontinued.
Du Pont plans (Ref. 4-39) to concentrate on the development of an improved
thermal reactor emission control system, rather than retrofit current
devices. New systems may include (1) Esso-type RAM features, (2) air
injection modulation, (3) spark advance adjustment, and (4) fuel injection
(for more precise air-fuel ratio control).
Durability tests of EGR alone for a similar type EGR system were conducted
by Esso (Ref. 4- 13) for NAPCA. In this test program, the EGR system was
evaluated in three 1969 Plymouths and three 1969 Chevrolets over 52, 000 miles
under city-suburban driving conditions simulated on a tape-controlled mileage
accumulation dynamometer. No major problems were reported. Engine wear
and cleanliness were considered normal for the mileage and driving regime.
These results appear to bear out Du Font's contention that the CARB test
program failure was related to the thermal reactor core oxidation process.
4.3.2.3.4 Effect of Lead Additives
As noted in Ref. 4-29, Ford changed to the use of a low-lead (0.5 gm/gal) fuel
because of concurrent supporting program efforts showing severe corrosion of
4-61
-------
material specimens when exposed to high-temperature exhaust gas from an
engine operated on fuel containing 3 gm/gal of TEL (more fully reported in
Ref. 4-29).
Du Pont studies (Ref. 4-8) indicate that the presence or absence of lead
has no effect on corrosion or oxidation of high-temperature materials, such
as Inconel 601, at the temperatures at which RTR's normally operate;
i.e., between 1700°F and 1900°F.
The presence of lead in gasoline, and particularly the combination of lead
and phosphorous, causes deterioration of the reactor core at localized points
where the exhaust gases coming in from the engine impinge on the interior
surfaces. This deterioration, termed erosion, was reported in Refs. 4-40
and 4-41. Erosion was shown to be caused by lead, and it was accelerated
when lead and phosphorous were combined. It was also shown that erosion
is subject to partial control by changing the reactor geometry to minimize
gas impingement. Further, it was shown (Ref. 4-41) that a nickel-chromium
alloy such as Universal Cyclops Uniloy 50/50 (50 percent Ni, 50 percent Cr)
has exceptionally good resistance to erosion. Since erosion is quite localized,
Du Pont concludes that small patches of Uniloy 50/50 could be inserted at the
erosion points to protect the core, which would be of a less expensive material
such as Inconel 601.
4. 3. 3 Combination Systems
There are four meaningful subclasses of combination systems:
1. LTR plus HC/CO Catalytic Converter plus EGR
2. RTR plus HC/CO Catalytic Converter plus EGR
3. RTR plus Dual Catalytic Converter plus EGR
4. RTR plus NOX Catalytic Converter plus RTR
(The LTR plus Dual Catalytic Converter plus EGR concept is not considered
a feasible approach as a reducing atmosphere is required for all known
NOX catalysts.)
4-62
-------
4.3.3.1 LTR plus HC/CO Catalytic Converter plus EGR
This emission control system concept is exemplified by the General Motors
"1975 Experimental System" (Ref. 4-31). The major new components of this
low-emission concept vehicle (shown in Fig. 4-30) include:
1. Improved carburetor with altitude compensation and power
choke (electronic fuel injection may be used in some models)
2. Exhaust gas recirculation (into intake manifold)
3. HC/CO catalytic converter
4. Air injection pump
5. Unitized ignition system
Although not specifically illustrated in Fig. 4-30, it is understood (Ref. 4-42)
that the system includes the General Motors Air Injection Reactor (A. I. R. )
system, wherein slightly lean (A/F approximately 15-16. 5) carburetion plus
IMPROVED CARBURETOR
WITH ALTITUDE COMPENSATION
AND POWER CHOKE
EXHAUST GAS
RECIRCULATION
AIR INJECTION
PUMP
CATALYTIC
CONVERTER
UNITIZED
IGNITION
Fig. 4-30. General Motors 1975 Experimental Emission Control System
(from Ref. 4-31)
4-63
-------
air injection into the exhaust manifold serves as a "low-grade" LTR to
provide rapid warmup of the catalytic converter. After warmup, the air
injection is directed to the catalytic converter.
4.3.3.1.1 Emission Level Characteristics
Emission test values for the General Motors proposed 1975 system
(Ref. 4-31) are:
HC = 0.54
CO = 9.20
NOX = 1.00
gm/mi (CVS cold start single-bag tests)
More recent three-bag CVS data reported by General Motors (Ref. 4-43)
for this type of system are:
HC = 0.40
CO = 5. 50
NOX = 0.95
gm/mi
American Motors (Ref. 4-35) is currently testing, and has in the past two
years tested, three basic catalyst types for the control of HC and CO in a
configuration similar to the General Motors approach delineated above.
Typical baseline (zero vehicle and system miles) emission levels are:
HC
0.04
0. 11
0.45
CO
2.35
3.29
2.96
Emissions (gm/mi)
Catalyst
Platinum- monolithic
Base metal-bead
Platinum-be ad
EGR was used to give NOX levels of 3 gm/mi.
4. 3. 3. 1. 2 Fuel Economy Results
General Motors has not indicated the fuel economy characteristics of their
proposed 1975 system (Ref. 4-31), except to indicate a goal of approximately
10 percent loss in fuel economy.
Vehicle
3000-lb Javelin;
single-bag CVS tests
3500-lb standard car;
three-bag CVS tests
3500-lb standard car;
single-bag CVS tests
4-64
-------
4. 3. 3. 1. 3 System Lifetime Characteristics
General Motors (Ref. 4-31) has emphasized that its advanced emission control
concepts are experimental, and that durability and/or lifetime characteristics
are not well defined at this point in time. Although the Corporation is attempting
to develop catalytic converters for useful lifetimes of 50,000 miles, initial
converters placed in service in some 1974 models may have a recommended
replacement interval of approximately 25, 000 miles (Ref. 4-43).
4.3.3.1.4 Effect of Lead Additives
Previous comments with regard to the effect of lead additives on thermal
reactors and EGR systems were given in Section 4.3.2.2.4. With respect
to HC/CO catalytic converters, the following position was recently taken by
General Motors (Ref. 4-44):
Lead seriously affects catalyst life; all of some
300 catalysts tested in cars were affected by
lead; there is some regenerative property, but
very little.*
4.3.3.2 RTR plus HC/CO Catalytic Converter plus EGR
This emission control system concept is exemplified by the Ford "Combined
Concept Emission Package" (Ref. 4-29). In this type of system, the thermal
reactor acts as a "preheater" for the HC/CO catalytic converters. Carbure-
tor enrichment and EGR are utilized for NOV control.
J\.
The major components and features of this Ford combined "maximum
'
^This refers to the case of using unleaded gas after the catalyst has been
exposed to leaded fuels
4-65
-------
effort" low-emission concept vehicle (the A-B system), shown in Fig. 4-31,
include:
3
1. Two 97 in. UEC Type H reactors (with center core)
2. Two noble metal catalytic converters
3. Reactor inlet and outlet sheet metal liners
4. Modified cylinder heads with exhaust port liners
3
5. One engine-driven secondary air pump (16 in. displacement
6. Below-the-throttle EGR system
7. Production-type carburetor with richer calibration
8. Production distributor with modified curve
9. More spark retard during warmup (until engine water
temperature reaches 120°F)
10. Modified crankcase ventilation
11. Prototype reactor protective system to limit maximum
core temperature to 1850°F
12. Unleaded fuel requirement
HC/CO
CONVERTERS
SPACER-ENTRY EGR ON-OFF VALVE
IIEC REACTOR TYPE H
SECONDARY AIR PUMP
IIEC REACTOR TYPE H
Fig. 4-31. Ford Combined Maximum Effort/Low-Emission
Concept Vehicle (from Ref. 4-29)
4-66
-------
4. 3. 3. 2. 1 Emission Level Characteristics
The average emission data from ten separate cold start CVS single-bag tests
of the Ford combined "maximum effort" vehicle at low mileages are given
below along with the low and high value ranges:
Emissions (gm/mi)
Test Data
Average
Low
High
HC
0.28
0. 11
0. 53
CO
3.4
1.7
6.7
NOX
0. 76*
0. 51
1.02
*NOX emission levels were measured using a nondis-
persive infrared instrument for NO and a nondispersive
ultraviolet instrument for NO2.
More recent data from Ford (Ref. 4-45) for a similar system with a higher
EGR flow rate are:
HC = 0.25
CO = 2.95
NO = 0.55
gm/mi (three-bag CVS tests)
4.3.3.2.2 Fuel Economy Results
At the level of emissions shown, the Ford "maximum effort" test vehicle
had a 27-percent loss in fuel economy over baseline vehicles for a city-
suburban driving schedule. Limited testing was conducted on this vehicle
to minimize the significant fuel economy losses. When the fuel economy
loss (on the CVS chassis dynamometer test) was reduced from approximately
25 percent to about 10 percent by running less rich, HC and CO emissions
increased only slightly, and NOX emission levels increased from the
0. 72-gm/mi level to the 1. 3-gm/mi level. Figure 4-32 shows NOX emission
levels from a series of CVS cold starts as they relate to the fuel economy
loss (on the chassis dynamometer) over a baseline vehicle. For all tests,
HC and CO emission levels were below 0. 4 and 4. 0 gm/mi, respectively.
4. 3. 3. 2. 3 System Lifetime Characteristics
Ford also emphasizes that their advanced emission control concepts are
4-67
-------
experimental and that durability and/or lifetime characteristics are not
well defined at this point in time (Ref. 4-33).
4.3.3.2.4 Effect of Lead Additives
Previous comments with regard to the effect of lead additives on thermal
reactors and EGR systems were given in Section 4. 3. 2. 3. 4. With respect
to HC/CO catalytic converters, the following statements were recently
made by Ford (Ref. 4-46):
Lead-free gasoline is mandatory; one tankful of
gasoline containing 3 gm/gal of lead seriously
impairs catalyst performance (although some
catalysts have some recuperative power); trace-
lead (0. 015 gm/gal) is probably tolerable;
phosphorus, chlorine, bromine, and sulfur are
also detrimental.
VEHICLE DESCRIPTION
- 302-2V AUTO. GALAXIE
- PORT LINERS
- THERMAL REACTORS
- NOBLE METAL CATALYSTS
- AIR CLEANER EGR
- OTHER MODIFICATIONS
1.5
1.0
N.OX-GM/MI.
(DETERMINED ON
CVS TEST) 0.5
0 10 20 30 40
FUEL ECONOMY - % LOSS FROM BASELINE CVS
Fig. 4-32. Ford Maximum Effort Vehicle-- NOX Emissions
vs Fuel Economy (from Ref. 4-29)
4-68
-------
4.3.3.3 RTR plus Dual Catalytic Converter plus EGR
This emission control system concept is exemplified by an experimental
General Motors system (Ref. 4-31) and by experimental test systems of
Ford (Ref. 4-45) and the American Oil Company (AMOCO) (Ref. 4-47).
The General Motors system consists essentially of the 1975 proposed control
system plus a quick-heat manifold and fast choke, as well as the addition of a
NOX catalytic converter for NOX control (Fig. 4-33, Ref. 4-31). In addition,
the engine is run rich (A/F approximately 14-15) to provide the necessary
reducing atmosphere for the NOX catalytic converter. Emission test results
are:
HC =0.2
CO =4.0
NO.. = 0. 6
gm/mi (single-bag CVS tests)
More recent three-bag CVS test results reported by General Motors
(Ref. 4-43) indicate emission bands (composites of several tests) as follows:
HC = 0.2-0.3
CO = 2.5-6.0
NO.. = 0.35-0.85
gm/mi
No specific details are available on the Ford test vehicle, although it is
presumably similar to the Ford combined concept package (Fig. 4-31)
except for the use of a dual catalytic converter instead of an HC/CO catalytic
converter. Emission test results reported at low mileage with a fresh catalyst
are:
HC
CO
NO,
0. 27
2.24
0.60
gm/mi (three-bag CVS tests)
4-69
-------
CARBURETOR
INTAKE MANIFOLD
RISER EXTENSIONS
Fig. 4-33. General Motors Quick-Heat Manifold and Fast
Choke Configuration (from Ref. 4-31)
Similar emission data are reported by AMOCO (Ref. 4-47). In this case,
two different dual catalytic converter configurations were evaluated in vehicle
tests. The three-bag CVS test emission data are (based on two tests of each
configuration):
Emission Levels (gm/mi)
Configuration
Pelletized NO catalyst plus
pelletized HC/CO catalyst
(fresh catalyst; no mileage)
Monolithic NO catalyst
plus monolithic HC/CO
catalyst (at 100 miles)
HC
0.26
0.38
CO
1. 72
2.07
NOX
0. 55
0. 68
Although only laboratory, low-mileage data on this concept are available,
it is a logical approach to achieving lower NOX levels at reasonable fuel
consumption penalties (<10 percent) and, in principle, is merely the replace-
ment of a single-bed HC/CO catalytic converter with a dual-bed HC/CO/NO
.X
converter or the addition of a NOX catalytic converter to a system already
4-70
-------
incorporating an HC/CO converter. Although not openly reported, it is known
that this emission control system concept is under intensive evaluation by the
automotive industry with respect to its potential for meeting the stringent 1976
NO standards.
Comments regarding the effects of lead additives in gasoline are deferred to
Section 5.
4.3.3.4 RTR plus NOX Catalytic Converter plus RTR
This emission control system concept is exemplified by one American Motors
(Ref. 4-35) experimental system. American Motors tests of this system showed
it to meet the required levels of HC, CO and NO simultaneously, at zero mile
conditions . It utilizes an exhaust manifold reactor operating rich, with sufficient
secondary air injection to increase the sensible energy content of the exhaust
gas and remain "reducing." This exhaust is fed through an NO catalyst followed
X.
by additional secondary air and another thermal reactor. Zero-mile emission
levels as tested in a 4500-pound Jeep vehicle with a 360 CID engine were:
HC = 0.01
CO = 2.44 gm/mi (CVS 3-bag data)
NO.. = 0.37
.A.
The nominal air-fuel ratio was 12:1, and no EGR was employed. The first
thermal reactor was unbaffled with a stainless steel liner. The second thermal
reactor is located in a compartment directly behind the NO catalyst bed. It
A.
is stated that the exhaust gas temperature from the second RTR is approximately
1800 F, which is felt to be too hot for safe vehicle operation.
It is believed that the fast warmup intake manifold with a timed choking mecha-
nism could possibly lower CO to provide increased margin. Lack of system dura-
bility experience, severe installation problems, very high fuel consumption
(estimated 25 percent SFC penalty) are current problem areas and these preclude
serious consideration of this system by American Motors at this time .
4-71
-------
4.3.3.5 Stratified Charge Engine
A prototype stratified charge engine installed in a one-quarter-ton light truck
was recently tested in the EPA test center in Willow R
-------
test procedures for emissions testing. The 7-mode, 7-cycle procedure (FTP)
defined in the Federal Register of June 1968 (Ref. 4-49} was changed to the
single-bag CVS technique (CVS-1) described in the Federal Register of
July 15, 1970 (Ref. 4-50). Following this change, a new three-bag CVS
weighted-aver age procedure (CVS-3), applicable to 1975-76 systems, was
defined in July 1971 (Ref. 4-51). These changes are found to have a sub-
stantial impact on the resultant emission levels measured for certain systems.
In general, it is not possible to convert FTP to CVS-1 data because of the
differences iii emphasis on the cold start emission contribution, because of
differences in the driving cycle, and because of differences in the test instru-
mentation. A similar set of constraints applies to the conversion of CVS-1
to CVS-3 data.
As only CVS data is regarded as representative of system performance in
relation to the goals presently established for 1975-76 systems, the available
CVS data previously given in Section 4.3 are summarized in Table 4-15.
Except for the RTR plus NO Catalytic Converter plus RTR concept
Jt
(Table 4-15C), no emission control system has demonstrated meeting 1976
concurrent (HC, CO, and NO ) emission standards, whether by CVS-1 or
Ji
CVS-3 test procedures. With regard to the data in Table 4-15, the following
observations can be drawn.
1. In general, the catalytic-converter-only or catalytic-converter-plus
EGR systems do not appear meaningful for meeting 1975-76 standards
This apparently results from CVS cold start effects and the lack of
provision for rapid "warmup" capability of the catalyst converter.
There are some instances, however, where catalytic-converter -
only systems meet the HC and CO values with a fresh catalyst.
2. The LTR plus EGR system has yet to demonstrate meeting 1975-76
HC and CO standards . Proposed changes to the thermal reactor
(flameholders, etc.) might help, and the addition of an HC/CO
catalytic converter would certainly enhance the HC and CO picture
for the lean thermal reactor approach. As lean operation precludes
the use of an NOX catalyst, this approach is limited to EGR NOX
reduction levels and is not capable of meeting 1976 NOX emission
levels on this basis.
3. RTR systems (noncatalyst), e.g., the Esso RAM system, are well
below 1975-76 HC standards and approach (met in one case) 1975-76
4-73
-------
Table 4-15A. Summary of Emission Control System Emission
Data--Catalytic Converter Systems (Laboratory,
Low-Mileage Tests)
. CVS Cold-Start Emissions
_ (gm/mi)
System Type Fed Std 1975-76
HC/CO Catalytic Converter Only (No ECR)
UOP Tests
U S 1971 Domestic V-8 (Normal Choke)
US 1971 Domestic V-8 (Fast Choke)
Some Foreign Vehicles
Engelhard Tests
PTX-433 Catalyst (0 2«i Pi)
HC/CO Catalytic Converter + ECR
Ford Package "B"
Chrysler
Dual Catalytic Converter -f EGR
Ford Package "C"
Tncomponcnt Catalytic Converter
APCO Tests of UOP System (1970 VW)
Single-Bag (CVS-1)
(Pre-July 1971)
HC
0 46
0. 59 to
0 68
0 16 to
0 51
0. 29 to
1.41
0 70
0 60
0 24
0 85
1 4 to
2 3
CO
4 7
0. 96 to
1.45
1 21 to
2 58
0 99 to
3.86
3 80
11 0
7 2
10 0
10 to
32
NOX
3 Vs^
/<* 4
2 11 to
3 88
4.74 to
5 08
1 36 to
2.0
5 0
1 3
2 03
0 90
0 6 to
1 3
Three-Bag (CVS- 3)
(Post-July 1971)
HC
0 41
CO
3.4
NO,
^ 0.4
Reference
4-23
4-57
4-29
4-6
4-29
4-34
Table 4-15B.
Summary of Emission Control System Emission
Data--Thermal Reactor Systems (Laboratory,
Low-Mileage Tests)
"— CVS Cold-Start Emissions
_____(gm/mi )
System Type Fed Std. 1975-76
LTK plus EGR
Ethyl Corporation
Pontiac
Plymouth
Chrysler
RTR Alone
Modified RAM
Ford Type J Reactor
RTR plus EGR
RAM
Esso Tests (Chev )
EPA Tests (1971 Ford LTD)
(a)
(b)
(c)
(d)
(e)
Recent Du Pont System
Ford Package "A"
Chrysler
Single-Bag (CVS- 1)
(Pre-July 1971)
HC
0 46
0 64
0 89
0 70
0 07
0 1-0 3
0 08
0 20
0 20
0 14
0 10
0 14
0 05
0.30
0 23
CO
4 7
6 4
8 6
7.0
4 2
6-12
3 7
5.9
3 8
4 8
4 54
4 77
9 2
9 0
13 8
NO,
3 '*sf
1 52
1 37
1 30
1 89
0 5-0 7
0 72
0.65
0 60
0 60
0.67
0 63
0 52
1 40
0 45
Three Bag (CVS- 3)
(Post-July 1971)
HC
0.41
0 52
0 II
0 10
CO
3 4
6 2
4 76
3 19
NO,
3 l^*'
1 37
0 67
0 67
Reference
4-5
4-5
4.6
4-2
4-3
4-2
4-37
4-8
4-29
4.6
4-74
-------
Table 4-15C. Summary of Emission Control System Emission Data--
Combination Systems (Laboratory, Low-Mileage Tests)
' ' . _ CVS Cold-Start Emissions
Igm/mi)
System Type Fed Sid • 1975-76
LTR plus HC/CO Catalytic Converter plus EGR
C M "1975 Experimental Syatem"a
American Motors
Platinum-Monolithic. Air-injection Reactor
Base Metal- bead Air-injection Reactor
Platinum-bead, Air-injection Reactor
RTR plus HC/CO Catalytic Converter plus EGR
Ford Combined Concept Package
"Maximum Effort" Tests
"Improved Fuel Economy" Tests
High- rate EGR System
RTR plus Dual Catalytic Converter plus EGR
C M " 1975 System" plus Quick -heat Manifold
and Fast Choke plus NO Catalytic Converter"
Ford Dual Bed Catalyst System
AMOCO Vehicle Tests
Pelletned Catalysts
Monolithic Catalysts
RTR plus NOi Catalytic Converter plus RTR
American Motors (Jeep with 360 CID, no ECR)
Single- Bag (CVS-1)
(Pre-July 1971)
HC
0 46
0 54
0 40
0 45
0 28
~0 3
0 2
CO
4 7
1 2
2 35
2 96
3 4
~3 5
4 0
NOX
3 *S
sS**> 4
1 0
3 0
3 0
0 76
1 }
0 6
Three-Bag (CVS- 3)
(Post-July 1971)
HC
0 41
0 40
0 11
0 25
0 2to
0 3
0 27
0 26
0.38
0 01
CO
3 4
5 5
3 29
2 95
2 5lo
6 0
2 24
1 72
2 07
2 44
N0x
3 \^S
sS7 04
0 95
3 0
0 55
0 35to
0 85
0 60
0 55
0 68
0 37
Reference
4-31/4-43
4.35
4-29
4-29
4-45
4-31/4-43
4-45
4-47
4-35
aGeneral Motors A I R System, i e , "low-grade" LTR. lean (A/F = 15-16 5) operation
General Motors A 1 R System, i e , "low-grade" RTR. rich (A/F = 14-15} operation
CO levels . NOX levels are determined by combined air-fuel
ratios and EGR reduction effects, and although significantly low
(approximately 0.7 gm/mi), do not meet 1976 NOX standards.
Combined systems, incorporating some form of thermal reactor
(whether it be a. "full-size" reactor or a "low-grade" reactor,
e.g., General Motors A.I.R. system) for catalyst bed warmup
under cold start conditions, appear to offer the means for eventu-
ally meeting the 1976 standards for all three emission constituents
(HC, CO, and NOX). Even in this case, a NOX catalyst bed would
be required to meet NOX standards.
Without a NOX catalyst, both the RTR plus EGR and "low-grade"
RTR plus HC/CO catalytic converter plus EGR concepts, offer a
somewhat similar potential for concurrent emission reduction
(approaching 1975-76 HC and CO standards and achieving NOX
levels higher than the 1976 standard) .
4-75
-------
It is emphasized that the foregoing observations are based on experimental
laboratory data only. If, as the various automakers have suggested, levels
of approximately 50 percent of the 1975-76 standards have to be achieved to
account for the variation of production tolerances, test reproducibility,
degradation with accumulated mileage effects, etc., then it would appear that
the emission control systems proposed and evaluated to date will not meet
the 1975-76 emission standards.
4.3.4.2 Lifetime/Durability Effects
As detailed in Section 4.3 for each specific emission control system dis-
cussed, there are no meaningful lifetime or durability data available for any
combined emission control system seriously being considered for implementa-
tion by the U.S. automakers.
The approximately 90, 000-mile durability test of a thermal reactor by Ford
(Ref. 4-36) is certainly significant, but did not even include an EGR system,
let alone a catalytic converter.
Engelhard (Ref. 4-57) has reported a 50,000-mile (AMA driving schedule)
durability test of their PTX-433 catalytic converter unit; again EGR was
not incorporated.
At this point in time, then, overall emission control system durability and
lifetime remain simply as goals, with little or no demonstrated cabaility.
4.3.4.3 Fuel Economy Effects
As indicated in Sections 4.2 and 4.3 there is a wide variability in the SFC
values reported for the various emission control systems. The primary
factor, however (as shown in Section 4.2.2.3), is the combined effect of air-
fuel ratio and EGR flow rate utilized to control NOX to different levels. Over
and above this basic effect is the variation in method of reporting SFC effects.
For the same emission control system, different results are obtained with
different driving cycles and/or dynamometer test procedures.
4-76
-------
A general correlation for a number of the systems examined in Section 4. 3
is presented in Figure 4-34, where the SFC increase (over the baseline
vehicle without the specific emission control system) is shown as a function
of the NOx level achieved. This general correlation is nearly the same as
the Ford/IIEC estimate of Ref. 4-29 (presented herein as Fig. 4-32,
Section 4. 3. 3. 2. 2).
As can be seen in Fig. 4-34, both non-catalytic-converter systems (LTR +
EGR, RTR + EGR) and HC/CO catalytic-converter systems (HC/CO catalytic
converter + EGR, RTR + EGR + HC/CO catalytic converter) comprise the
general correlation. This is as would be expected, since the NO level
realized is related only to conditions present in the engine cylinder (i.e.,
air-fuel ratio and percentage EGR) and not to any external device or condition.
As these data points represent test configurations employing operating param-
eters (air-fuel ratio, EGR rate) selected to produce minimum NO levels,
the shape of the general correlation curve conforms to (again, as would be
expected) the limiting envelope of SFC versus NO reduction capability
previously shown in Fig. 4-13.
Also shown in Fig. 4-34 that relationship of SFC increase versus NO level
estimated to occur if a NO catalyst at 75-percent conversion efficiency were
added to a system characterized by the general correlation line and if this
addition did not influence other parts of the system (75 percent selected as a
typical number; representative values for NO catalysts now under develop-
ment are not available).
As can be noted, extremely high fuel consumption penalties occur if NO
levels below approximately 1 gm/mi are achieved, unless an NO catalyst is
used. Even then, SFC increases of approximately 7-8 percent are envisioned
at NO levels of approximately 0.4 gm/mi.
4-77
-------
T
SYSTEM AND SOURCE
oc.
o
o
oc.
Q.
CO
H 3
CT>
i 2
o
O
d
A
D
Cf
0
LTRtEGR (ETHYL PLYMOUTH)
LTRtEGR (ETHYL PLYMOUTH)
LTRtEGR (ETHYL PONTIAC)
LTR+ EGR (ETHYL PONTIAC)
RTRtEGR(DUPONTCHEV)
RTRtEGR (RECENT DUPONT SYSTEM)
RTRtEGR(ESSORAM)
RTRtEGR(ESSORAM)
RTR + EGRtHC/CO CAT CONV
(FORD "MAXIMUM EFFORT" VEH)
RTRtEGR+HC/COCAT CONV
(FORD "MAXIMUM EFFORT" VEH)
RTR + EGR+HC/CO CAT. CONV
(FORD "MODIFIED MAX. EFFORT" VEH ]
RTR tEGR(CHRYSLER I
HC/CO CAT CONV t EGR
(FORD PACK "B")
DUAL CAT CONV + EGR
(FORD PACK "C")
DRIVING SCHEDULE
CITY
CITY-EXPRESSWAY
CITY
CITY-EXPRESSWAY
CARB CAR POOL
NOT SPECIFIED
TURNPIKE
CITY
CITY-SUBURBAN
CVS CHASSIS DYNA
CVS CHASSIS DYNA
NOT SPECIFIED
CVS CHASSIS DYNA
CVS CHASSIS DYNA
— GENERAL CORRELATION
— ESTIMATED FOR ADDITION OF NOx
CATALYST BED AT 75 PERCENT EFFICIENCY
5 10 15 20 25 30
PERCENT SFC INCREASE (OVER UNCONTROLLED VEHICLE )
35
Fig. 4-34. NOX vs SFC Increase
4-78
-------
REFERENCES
4-1. Exhaust Manifold Thermal Reactor Development at Du Pont,
Petroleum Chemicals Division, Du Pont de Nemours & Co. ,
Wilmington, Delaware (20 January 1971).
4-2. R. J. Lang, "A Well Mixed Thermal Reactor System for Automotive
Emission Control," SAE Paper No. 710608 (June 1971).
4-3. A. Jaimee, D. E. Schneider, A.I. Rozmanith, and J. W. Sjoberg,
"Thermal Reactor Design, Development and Performance," SAE
Paper No. 710293 (January 1971).
4-4. Air Injection Pump Horsepower Requirements for V-8 Engines,
Drawing No. 3210665, American Motors Corporation (4 September 1969).
4-5. The Ethyl Lean Reactor System, Ethyl Corporation Research
Laboratories, Detroit (1 July 1971).
4-6. Chrysler Corporation, Letter in response to a request from the
Administrator, Environmental Protection Agency, Subject: Regarding
Emission Control (1 April 1971).
4-7. K. Matsumoto and H. Nohira, "Oxides of Nitrogen from Smaller
Gasoline Engines," SAE Paper No. 700145 (January 1970).
4-8. Effect of Lead Antiknocks on the Performance and Costs of Advanced
Emission Control Systems. Du Pont de Nemours & Co. , Wilmington,
Delaware (15 July 1971).
4-9. W. F. Deeter, et al.,"An Approach for Controlling Vehicle Emissions,"
SAE Paper No. 680400 (May 1968).
4-10. Y. Kaneko, et al., "Small Engine Concept Emission Vehicles," SAE
Paper No. 710296 (January 1971).
4-11. W. Glass, et al., "Evaluation of Exhaust Recirculation for Control of
Nitrogen Oxides Emissions," SAE Paper No. 700146 (January 1970).
4-12. W. Glass, D.S. Kim and B. J. Kraus, "Synchrothermal Reactor
System for Control of Automotive Exhaust Emissions," SAE
Paper No. 700147 (January 1970).
4-13. G. S. Musser, et al. , "Effectiveness of Exhaust Gas Recirculation
with Extended Use," SAE Paper No. 710013 (January 1971).
4-79
-------
REFERENCES (cont.)
4-14. Demonstration of the Technological Feasibility of Controlling Oxides
of Nitrogen from Vehicular Exhaust, Progress Report No. 2,
California Air Resources Board (May 1969) Federal Grant No. 68A0605D).
4-15. Demonstration of the Technological Feasibility of Controlling Oxides
of Nitrogen from Vehicular Exhaust, Progress Report No. 5,
California Air Resources Board (April 1970) (Federal Grant
No. 68A0605D).
4-16. Demonstration of the Technological Feasibility of Controlling Oxides
of Nitrogen from Vehicular Exhaust. Progress Report No. 6,
California Air Resources Board (July 1970) (Federal Grant
No. 68A0605D).
4-17. Pro.ject CI: Du Pont Reactor Vehicles. 4th Progress Report,
California Air Resources Board (18 June 1971).
4-18. F. W. Bowditch (Director, Automotive Emission Control, General
Motors Corporation), Letter to the Administrator, Environmental
Protection Agency, Subject: Comments on Notice of Proposed Rule-
Making for NOY on 1973 New Motor Vehicles, Federal Register of
27 February 1971. Vol. 36. No. 40 (28 April 1971).
4- 19. H. K. Newhall, "Control of Nitrogen Oxides by Exhaust Recirculation
--A Preliminary Theoretical Study," SAE Paper No. 670495 (1967).
4-20. S. Ohigashi, et al. , "Heat Capacity Changes Predict Nitrogen Oxides
Reduction by Exhaust Gas Recirculation," SAE Paper No. 710010
(January 1971).
4-21. "Study of Catalytic Control of Exhaust Emissions for Otto Cycle
Engines," Stanford Research Institute (April 1970).
4-22. G. J. Nebel and R. W. Bishop, "Catalytic Oxidation of Automobile
Exhaust Gases," SAE Paper 29-R (January 1959).
4-23. T. V. De Palma, "The Application of Catalytic Converters to the
Problems of Automotive Exhaust Emissions," Paper presented at the
Interpetrol Congress, Rome, Italy (24 June 1971).
4-24. G. H. Meguerian, "NO Reduction Catalysts for Vehicle Emission
Control," SAE Paper No. 710291 (January 1971).
4-25. "IIEC, A Cooperative Research Program for Automotive Emission
Control," SAE Publication SP-361 (1971).
4-80
-------
REFERENCES (cont.)
4-26. M. Shelef, K. Otto and H. Gandhi, "The Heterogeneous Decomposition
of Nitric Oxide on Supported Catalysts. Atmosphere Environment,
Vol. 3, Pergamon Press (1969), pp 107-122.
4-27. Personal discussion with representatives of the Universal Oil
Products Company, July 1971.
4-28. E. E. Hancock, R. M. Campau and R. Connolly, "Catalytic Converter
Vehicle System Performance: Rapid Versus Customer Mileage,"
SAE Paper No. 710292 (January 1971).
4-29. R. M. Campau, "Low Emission Concept Vehicles," SAE Paper
No. 710294 (January 1971).
4-30. H. W. Swochert, "Performance of a Catalytic Converter on Nonleaded
Gasoline," SAE Paper No. 690503 (May 1969).
4-31. Progress and Programs in Automotive Emissions Control, Progress
Report to the Environmental Protection Agency, General Motors
Corporation (12 March 1971).
4-32. American Motors Corporation, Letter in response to a request from
the Administrator, Environmental Protection Agency, Subject:
Regarding Requirements of the Clean Air Act (2 April 1971).
4-33. H. L. Misch (Vice President--Engineering, Ford Motor Company),
Statement to the Environmental Protection Agency (6 May 1971).
4-34. J. C. Thompson, Exhaust Emissions from a Passenger Car Equipped
with a Universal Oil Products Catalytic Converter, Air Pollution
Control Office, Environmental Protection Agency (December 1970).
4-35. C. E. Burke (American Motors Corporation), Letter to The Aerospace
Corporation (29 September 1971).
4-36. B. Simpson (Ford Motor Company), Letter to The Aerospace
Corporation (15 September 1971).
4-37. Telephone conversation with H. Gompf, Environmental Protection
Agency (12 October 1971).
4-38. R. C. Butler (Du Pont de Nemours & Co.), Letter to Office of Air
Programs, Environmental Protection Agency (30 August 1971).
4-81
-------
REFERENCES (cont.)
4-39. Telephone conversation with J. Mikita, Du Pont de Nemours & Co.
(September 1971).
4-40. E.N. Cantwell, et al., "A Progress Report on the Development of
Exhaust Manifold Reactors," SAE Paper No. 690139 (January 1969).
4-41. W. J. Earth and E.N. Cantwell, "Automotive Exhaust Manifold
Thermal Reactors--Mate rials Considerations," Presented before the
Division of Petroleum Chemistry, Inc. , I6lst Meeting of the
American Chemical Society, Los Angeles, California (28 March - 2
2 April 1971).
4-42. Telephone conversation with F. Bowditch, General Motors
Corporation (September 1971).
4-43. Personal discussion with representatives of the General Motors
Corporation (September 1971).
4-44. Personal discussion with representatives of the General Motors
Corporation (30 June 1971).
4-45. Telephone conversation with B. Simpson, Ford Motor Company
(October 1971).
4-46. Personal discussion with representatives of the Ford Motor Company
(1 July 1971).
4-47. J. H. Somers (Office of Air Programs, Environmental Protection
Agency), Letter to the Aerospace Corporation, Subject: Discussion
with Dr. G. Meguerian (7 October 1971).
4-48. "Army Motor Vehicle Meets 1976 Emission Levels," Environmental
News (Press Release). Environmental Protection Agency,
Washington, D. C. (24 September 1971).
4-49. "Standards for Exhaust Emissions, Fuel Evaporative Emissions, and
Smoke Emissions Applicable to 1970 and Later Vehicles and Engines,"
Federal Register, Vol. 33, No. 108, Part II (4 June 1968).
4-50. "Control of Air Pollution from New Motor Vehicles and New Motor
Vehicle Engines, "Federal Register, Vol. 35, No. 136
(1 July 1970).
4-82
-------
REFERENCES (cont.)
4-51. "Control of Air Pollution from New Motor Vehicles and New Motor
Vehicle Engines," Federal Register, Vol. 36, No. 128 (2 July 1971).
4-52. Consequences of Removing Lead Antiknocks from Gasoline, A Status
Report, No. AC-10, Ethyl Corporation, Detroit (August 1970).
4-53. E. N. Cantwell, et al. , "Recent Developments in Exhaust Manifold
Reactor Systems, " Presented at a meeting of the Automobile Division,
Institution of Mechanical Engineers, Paper No. ADP-13, London,
England (11 May 1970).
4-54. E. N. Cantwell (Du Pont de Nemours & Co. ), Letter to The Aerospace
Corporation (12 August 1971).
4-55. S. Lawrence and J. Wisdom, "Emission Control by Exhaust Manifold
Reactor--An Initial Study for Small Engines, " Presented at a meeting
of the Automobile Division, Society of Mechanical Engineering,
Paper No. ADP-13(A)/70, London, England (11 May 1970).
4-56. D. Hirschler (Ethyl Corporation), Letter to The Aerospace
Corporation (August 1971).
4-57. Engelhard Industries, Inc., Letter to The Aerospace Corporation
(13 October 1971).
4-83
-------
5. EFFECTS OF LEAD
ADDITIVES ON EMISSION
CONTROL DEVICES/SYSTEMS
-------
SECTION 5
GENERAL ASSESSMENT OF EFFECTS OF LEAD ADDITIVES
ON EMISSION CONTROL DEVICES/SYSTEMS
In Sections 4.2 and 4.3, various comments, statements, and/or positions
relative to the effect of lead additives in gasoline were reported where they
were pertinent to a particular device/system under discussion. The pur-
pose of this section is to examine the relevant data and make a general
assessment of the effects of lead additives on the various emission control
devices/systems.
5.1 CATALYTIC CONVERTERS
Degradation of the performance of catalytic converters employed as pollu-
tion control devices on automobiles run on leaded and unleaded gasoline is
observed to occur much more rapidly with leaded gasoline . Degradation
may occur either by loss of catalytic activity, or physical attrition, or both.
The lead component of gasoline thus clearly constitutes a catalyst "poison,"
which acts through a variety of chemical and mechanical toxicity mechanisms
that are not mutually exclusive.
Even though numerous theoretical and laboratory investigations have been
performed on catalyst poisoning, the complex composition of exhaust gas,
the wide range and number of engine operating parameters, and the many
types and configurations of catalytic materials, make it very difficult to
arrive at generalizations regarding the most likely mechanisms. Neverthe-
less, a review of these mechanisms has indicated that lead, sulfur, and
phosphorus compounds would have a deleterious effect on catalysts. Experi-
mental data with prototype catalysts, run with actual automotive exhausts
under realistic operating conditions, are therefore most meaningful in assess
ing the effects of lead. These are discussed in this section. A brief discus-
sion of possible catalyst poisoning mechanisms can be found in Appendix B.
5-1
-------
5.1.1 Summary of Experimental Data
5.1.1.1 Laboratory Tests
The available data on lead effects are primarily for HC and CO oxidation
catalysts. To date, NOx reduction catalysts have been studied much less
fully.
Tests conducted by the Studebaker-Packard Corporation (Ref. 5-21) on an
HC/CO catalytic converter, using leaded and unleaded gasoline, indicate
rapid deterioration of catalyst effectiveness with leaded gasoline (Fig. 5-1)
100
80
35
o 60
UJ
o
t 40
20
UNLEADED GAS
* LEADED GAS
10
20
TIME, hr
30
40
Fig. 5-1. Catalyst Life--Leaded vs Unleaded Gasolines
(from Ref. 5-21)
At the Bureau of Mines, Hofer, et al., studied an alumina (A1,O ) catalyst,
£• j
and chromia (Cr2O3), manganic sesquioxide (Mn2O_), and urania (U-OR)
5-2
-------
catalysts supported on alumina, using gasoline containing 3 ml/gal/ of TEL
(Ref. 5-1). Test results from that study are listed in Table 5-1. As a
criterion for loss in catalytic activity, they used the rise in catalyst
temperature T required for oxidation of 80 percent of four selected
cl
hydrocarbons tested individually. In these tests the catalysts were exposed
to engine exhaust for periods of approximately 340 hours, except for the
A^O, catalyst which was exposed for 126 hours. The lead deposits were
in the form of the sulfate (PbSO.), oxysulfate (PbSCL • PbO), and chloro-
bromide |Pb(Cl,Br)2 . About half the lead contained in the fuel appeared
to be deposited on the catalyst. The data show no increase in activation
temperature with unleaded fuel, whereas with leaded fuel a significant rise
occurs. Effects of lead poisoning on the HC oxidation efficiency of platinum
(Pt) and vanadium pentoxide (V^Cv) catalysts are summarized in Fig. 5-2
(Ref. 5-2). The data indicate that these catalysts are adversely affected by
lead compounds; V^C" is more resistant to lead poisoning than Pt.
Catalyst deactivation as affected by TEL, motor mix (TEL plus scavengers)
and bromoethane (C2H,-Br, similar to a scavenger) is illustrated in Fig. 5-3
for a copper oxide-chromia (CuO/Cr,O ) NO catalyst (Ref. 5-3). Deactiva-
^ j X
tion with TEL was very rapid initially, but after about 20 hours it decreased
at a lower rate. Deactivation with bromoethane was very fast, and after
60 hours the catalyst was almost inactive. The effect of motor mix was
intermediate between that of TEL and bromoethane. This shows that lead
and scavengers are detrimental to this catalyst.
Composition changes in this copper oxide-chromia catalyst upon deactivation
with motor mix in the fuel, as determined by electron-probe microanalyses,
are shown in Fig. 5-4 (Ref. 5-3). Lead was concentrated about 15 microns
thick at the pellet surface . Copper appeared to remain immediately behind
lead, whereas chromium tended to migrate toward the center of the pellet.
No such segregation was observed in fresh catalysts or catalysts aged in
the absence of lead. Because of this change in composition during deactiva-
tion by lead in the fuel, restoration of catalyst activity appears unlikely. Thus,
5-3
-------
Table 5-1. Effect of TEL on Catalytic Activity (from Ref. 5-1)
Catalyst
M2°3
Mn?O_/Al_O,
U308/A1203
Cr203/Al203
Cr203/Al203
T, (°C) Before
a
Exposure To
Exhaust
506
375
420
325
325
T (°C) After
a
Exposure To
Exhaust
529
600
600
430
325
Concentration
of TEL
(cc/gal)
3
3
3
3
0
Note. Period of exposure of catalyst to exhaust was -340 hours, except
for plain Al O_ which was 126 hours
100
8?
l
0 80
O 60
O
O
O
O
cc.
O
40
20
I I I
Pt, UNLEADED FUEL
V205, UNLEADED FUEL
V205. LEADED FUEL
V
\
0 50 100 150 200 250
RUNNING TIME -hours
Fig. 5-2. Effect of TEL on Catalyst Efficiency for HC
Oxidation (from Ref. 5-2)
5-4
-------
(jn
o. a
0.8
0.7
0.6
[NOx]
0.4
0.3
0.2
0.1
(
^0-0
0 ' C2H5Br
/ 0 ^
1 /
-
/ /
| /TEL 4- SCAVENGERS
|0 / (MOTOR MIX)
O *
I/. . -
/° 0-2^°^ °
^^ TEL ONLY
„ £l*^°
-------
NO catalysts of this type will require the use of unleaded gasolines
Ji
exclusively.
A comparative study of a number of catalysts (Ref. 5-2) showed that in the
case of a palladium/alumina (Pd/Al9O_) catalyst, the efficiency for HC
& j
oxidation was 62 percent for an unleaded catalyst and 42 percent for the
leaded catalyst (equivalent to approximately 10,000 miles of road use). For
CO oxidation, however, there was no deleterious effect of lead; that is, the
conversion figure remained at 97 percent. This is inconsistent with other
data denoting the effects of lead on catalyst conversion efficiency.
Laboratory test data for noble metal and transition metal oxide catalysts are
presented in Figs. 5-5 to 5-8 (Ref. 5-4). Catalyst BH (Fig. 5-5) has been
evaluated with low-lead (0.5 gm/gal) fuel. As indicated, this catalyst
retained sufficient activity to meet IIEC goals for at least 50,000 miles. It
should be noted that the IIEC goals represent higher emission levels than
the Federal standards for 1975-76.
Transition metal oxide catalyst data are shown in Figs. 5-6 through 5-8.
Catalyst G is similar to catalyst BH except it is smaller in diameter. It has
been evaluated with leaded, low-lead, and unleaded fuels. The data and
model predictions indicate that this catalyst will stay below IIEC HC and CO
emission levels for 50, 000 miles with unleaded fuel, about 15,000 miles with
low-lead fuel, and about 6000 miles with fully leaded fuel.
Catalyst AJ is an improved, extruded version of the composition used in
Catalyst G. Again, increased performance is noted with decreasing lead
level.
Catalyst BI is also a transition metal oxide catalyst similar to Catalyst AJ;
however, an additional transition metal oxide component was added to
further improve catalyst stability. As shown in Fig. 5-8, similar trends
are observed.
5-6
-------
0.5 gm/GAL LEAD
100
KCO AT 50
750 T-
SEC •'
10
0
MINIMUM CO OXIDATION
ACTIVITY FOR NEC GOALS
KC3H6 AT
750 °F-
SEC 0
MINIMUM HC OXIDATION
ACTIVITY FOR DEC GOALS
20,000 40,000 60,000 0 20,000 40,000 60,000
ACCUMULATED MILES
Fig. 5-5. Effect of Lead Content in Fuel on Catalyst Type "BH"
Oxidation Activity (from Ref. 5-4)
LEAD CONTENT, gm/GAL: oQ aO.5 -3.0
MINIMUM CO OXIDATION
ACTIVITY FOR NEC GOALS
SEC
Xo
\ MINIMUM HC OXIDATION
_ \ ACTIVITY FOR IIEC GOALS
'\
>» I I [
0 20,000 40,000 60,000 0 20,000 40,000 60,000
ACCUMULATED MILES
Fig. 5-6. Effect of Lead Content in Fuel on Catalyst Type "G"
Oxidation Activity (from Ref. 5-4)
5-7
-------
LEAD CONTENT, gm/GAL ° 0 a 0.5 • 3.0
200
CO,
750°F-
SEC1
OXIDATION
• ACTIVITY FOR IIEC GOALS
MINIMUM HC OXIDATION
ACTIVITY FOR IIEC GOALS
I I I I I
20,000 40,000 60,000 0
ACCUMULATED MILES
20,000 40,000 60,000
Fig. 5-7. Effect of Lead Content in Fuel on Catalyst Type "AJ1
Oxidation Activity (from Ref. 5-4)
LEAD CONTENT, gm/GAL • oQ a 0.5 0.0
Kco AT
750 °F-'
SEC"1
20
10
MINIMUM CO OXIDATION
ACTIVITY FOR IIEC GOALS
Kc3H6 AT
750 °F-
SEC'1
20,000 40,000 60,000 0
ACCUMULATED MILES
MINIMUM HC OXIDATION
ACTWTYJOJHIJC GOALS
20,000 40,000 60,000
Fig. 5-8. Effect of Lead Content in Fuel on Catalyst Type "BI"
Oxidation Activity (from Ref. 5-4)
5-8
-------
Some catalyst manufacturers are pursuing the development of lead-tolerant
catalysts. Although some success has been reported, test data are not
available in sufficient quantity and under the appropriate vehicle operating
conditions to present an evaluation of these systems at this time.
It is emphasized that the foregoing are projections of catalyst performance;
vehicle tests which simulate typical customer vehicle usage are necessary
to verify these predictions and establish catalyst durability.
5.1.1.2 Vehicle Tests
Catalyst half-life data obtained by Ford from a fleet test program are shown
in Fig. 5-9 (Ref. 5-5). As indicated, catalyst half-life decreases with
increasing lead content in the gasoline. American Cyanamid fleet tests
(Ref. 5-6) showed similar trends.
40,000
E 30,000
-------
Data obtained from Universal Oil Products Company (UOP) are shown in
Fig. 5-10 (Ref. 5-7). Catalyst conversion efficiency decreases very rapidly
with leaded fuel and very little when unleaded fuel is used.
Ford (Ref. 5-5) has conducted a road test program on a CuO/V^O,. catalyst
to determine the effects of TEL on HC and CO oxidation effectiveness.
About 18,000 road miles were accumulated for four pairs of cars, using
0.05, 0.5, 1.5, and 3 .0 ml/gal of TEL, respectively. HC conversion was
adversely affected by the presence of TEL in the fuel, and the half-life for
HC conversion was estimated to be 33,000 miles for unleaded fuel and
7500 miles for fuel having a concentration of 3 ml/gal of TEL. However,
CO conversion efficiency was the same for leaded and unleaded fuels.
The average efficiency based on a cold start is shown in Fig. 5-11. The
curves clearly illustrate the detrimental effect of TEL in the fuel. Similar
results were obtained at 30-mph cruise (Fig. 5-12). Because these data
were taken with a hot catalyst bed, the initial HC conversion efficiency of
the catalyst was higher.
The average amounts of CO removed during cold start tests are indicated
in Fig. 5-13. Although the efficiency gradually deteriorated from the initial
80 percent value, there are no trends to indicate a lead effect on the rate of
deterioration.
Data published by Ford (Ref. 5-8) are presented in Figs. 5-14 and 5-15 which
show the results obtained on the 302 CID engine group, in which two vehicles
were operated on fuel containing 3 ml/gal of TEL and two on commercially
available unleaded fuel over similar driving cycles with base metal catalysts
on alumina support.
Figures 5-16 and 5-17 show the results obtained on the 390-428 CID engine
groups, in which two vehicles were operated on leaded fuel and two on
unleaded fuel. Both engine groups, when operated on unleaded fuel, have
5-10
-------
300
200
CO
CO
100
08
CO
CO
CO
,-06
04
0.2
I
10,000 20,000 30,000 40,000 50,000
MILEAGE
Fig. 5-10. Typical Vehicle Emissions with a Catalytic Converter
(from Ref. 5-7)
5-11
-------
O INDOLEME CLEAR
INDOLENE 5
oiNDOLENE 15
OINDOLENE 30
9 12
WILES x I(T3
Fig. 5-11. Catalyst Efficiency--
Cold Start NDIR HC
Data (Average of Two
Vehicles) (from
Ref. 5-5)
18 FULL
CONVERTER
100
I
O INOOLENE CLEAR
A INDOLENE 5
o INOOLENE 15
OINDOLENE 30
9 12
MILES xlO'3
Fig. 5-12,
15 18
FULL
CONVERTER
Catalyst Efficiency--
Cruise 30 FID HC
Data (Average of
Two Vehicles) (from
Ref. 5-5)
T
T
INDOLENE CLEAR
INDOLENE 5
INDOLENE 15
INDOLENE 30
9 12
MILESxIO'3
Fig. 5-13. Catalyst Efficiency-
Cold Start CO Data
(Average of Two
Vehicles) (from
Ref. 5-5)
18
FULL
CONVERTER
5-12
-------
NON-LEADED FUEL VS LEADED CONTROL FUEL
-o NON-LEADED FUEL x x LEADED CONTROL FUEL
200
HC-PPM
CO-MOLE %
3,000
9,000
12,000
6,000
MILEAGE
* CALCULATED BY LEAST SQUARES AVERAGING TECHNIQUE
( ) FIGURES INDICATE NUMBER OF VEHICLES INCLUDED IN AVERAGE
Fig. 5-14. Ford 24-Car Fleet Tailpipe HC/CO Emissions*
(302 CID Engine) (from Ref. 5-8)
NOx-PPM
1400
1200
1000
800
600
400
200
0
NON-LEADED FUEL VS LEADED CONTROL FUE
NON-LEADED FUEL o— o un XH^< LEADED CONTROL FUEl
N0*
0
2) (2) (2) (2) (2)
— — —2— _L^L. ^j"
12) (2) (2) (2) £}
o
m
t i 11 i
L
2000 4000 6000 8000 10,000 12,000
MILEAGE
* CALCULATED BY LEAST SQUARES AVERAGING TECHNIQUE
( ) FIGURES INDICATE NUMBER OF VEHICLES INCLUDED IN AVERAGE
Fig. 5-15. Ford 24-Car Fleet Tailpipe NO Emissions* (302
CID Engine) (from Ref. 5-8)
5-13
-------
NON-LEADED FUEL VS LEADED CONTROL FUEL
LEADED CONTROL FUEL x---*<
HC-PPM
CO-MOLE %
2000
10,000 12,000
4000 6000 8000
MILEAGE
* CALCULATED BY LEAST SQUARES AVERAGING TECHNIQUE
( ) FIGURES INDICATE NUMBER OF VEHICLES INCLUDED IN AVERAGE
Fig. 5-16. Ford 24-Car Fleet Tailpipe HC/CO Emissions*
(390 and 428 CID Engines) (from Ref. 5-8)
NOX - PPM
NON-LEADED FUEL VS LEADED CONTROL FUEL
NON-LEADED FUEL o D
LEADED CONTROL FUEL *--X
1000
800
NO,
(2)
GOAL
(2)
(2)
X
(1)
2000 4000 6000 8000
MILEAGE
10,000 12,000
* CALCULATED BY LEAST SQUARES AVERAGING TECHNIQUE
( ) FIGURES INDICATE NUMBER OF VEHICLES INCLUDED IN AVERAGE
Fig. 5-17. Ford 24-Car Fleet Tailpipe NOX Emissions*
(390 and 428 CID Engines) (from Ref. 5-8)
5-14
-------
better emission control compared with the vehicles on leaded fuel. The
findings are essentially in agreement with previous Ford fleet testing
(Ref. 5-5). It should be noted (Fig. 5-17) that NO reductions for leaded
2C
gasoline show anomalous results because of unscheduled carburetor enrich-
ment due to filter clogging.
Other evidence for the deleterious effects of lead is indirect. Schwochert
(Ref. 5-9) reports excellent results with a supported noble metal catalytic
converter tested with unleaded fuel. At the end of a 50,000-mile road test,
HC and CO conversion efficiencies of approximately 70 percent were obtained.
This represents better performance than noted elsewhere for leaded gasoline.
5.1.2 Maximum Allowable Lead Levels
The available data with respect to lead additives in gasoline and their effect
on catalytic converter performance and durability indicate that lead levels of
0.5 gm/gal and greater would have deleterious effects and that vehicles with
such emission control devices would be unable to meet emission standards
after extremely short operational times. It has been stated by some of the
automakers that one tankful of fuel containing 3 gm/gal of lead would severely
harm the catalyst system. A catalyst manufacturer has stated that, based
on its test experience, one tankful of 3-gm/gal leaded fuel would be very
deleterious if the catalyst had been operated for extended mileage, but that
this effect would not be immediately apparent when the catalyst was fresh
(zero mileage conditions). In any event, it is clear that even one tankful of
gasoline could greatly shorten the durability capability of catalysts for
meeting emission standards. Similarly, all automakers and domestic
catalyst manufacturers state that the use of 0.5-gm/gal leaded fuel on a
continuous basis is unacceptable.
All the available data indicate that unleaded gasoline is required for emission
control systems using catalytic converters that would also have acceptable
durability characteristics to meet the emission standards of 1975-76.
5-15
-------
Unleaded gasoline, however, does contain very small amounts of lead
resulting from current (and foreseeable) refinery and distribution practices.
Therefore, it is very important to have specifications that limit the maximum
permissible amount of lead in unleaded gasoline. The American Society for
Testing and Materials (ASTM) has under consideration a proposed revision
to their standard specifications for gasoline (ASTM Designation: D 439-70)
in which they have selected the value of 0.07 gm/gal of lead for unleaded
gasoline. This value, which is not official at this time, was influenced in
part by the level given in Interim Federal Specification VV-G-001690
(Army-MR). This specification defines unleaded gasoline as follows:
Unleaded gasoline shall be defined as gasoline to which the
addition of lead compounds is not permitted. Lead com-
pounds present shall not exceed that amount which results
from contamination when good refinery and distribution
practices are followed, and shall not exceed 0.07 gm/gal.
This Federal specification was developed by Army representatives and a
task force which included representatives from petroleum and automotive
companies .
It should be noted, however, that with respect to catalyst performance and
durability, most of the data available for unleaded gasoline did not identify
the precise level of lead contained in the gasoline. The automakers and
catalyst manufacturers have stated that their experimental work has been
done with unleaded gasoline containing lead in the range of about 0.02-
0.06 gm/gal, and that most of the development work was with lead levels of
0.02-0.03 gm/gal. It was difficult to ascertain that any data are available
for lead levels between 0.06 and 0.5 gm/gal.
Since the proposed ASTM maximum level of 0.07 gm/gal of lead was apparently
based on refinery and distribution considerations, rather than catalyst life
considerations, it would appear that this specification bears further investiga-
tion to ensure that the lead content is reduced to that amount which is
5-16
-------
compatible with obtaining a 50, 000-mile useful lifetime . In this connection,
many of the automakers and catalyst manufacturers have stated that the
ASTM value of 0.07 gm/gal is probably too high. Some of the automakers
have expressed opinions that on the basis of limited data 0.05 gm/gal may be
a reasonable limit. Engelhard (Ref. 5-22) has stated that, based on the
performance data with its catalyst, the maximum lead content in gasoline
should be 0.03 gm/gal. Ford (Ref. 5-23) stated that although they did not
have an accurate quantitative answer with respect to the maximum lead level
for acceptable catalyst life, the Engelhard limit of 0.03 gm/gal might be
reasonable.
It should also be emphasized that at the lead levels of unleaded fuels used,
no automaker has stated to date that 50, 000 miles of operation at satisfactory
emission levels has been achieved. It is not known whether this durability/
lifetime deficiency is related to the lead level (0.02-0.03 gm/gal), to other
trace elements in the gasoline, or to other catalyst properties.
As can be seen from the above, substantive data to precisely determine the
maximum permissible lead content compatible with catalytic converters are
not available. In order to establish a meaningful maximum lead level, the
characteristics of durability versus lead content must be established for
catalysts capable of meeting the 50,000-mile requirement. The maximum
lead level selected should also consider the feasibility and economics of
providing gasoline at this lead level.
5.1.3 Summary
Fleet and laboratory test data show that HC and CO conversion efficiencies
of base metal catalysts are adversely affected by the presence of TEL in the
fuel.
It was found that scavengers in the fuel contribute significantly to deactivation
via depletion of the active component from the surface of the catalyst. The
rate of deactivation increases with increasing amounts of TEL, in the fuel.
5-17
-------
Substantive data to establish an upper lead level compatible with catalytic
converters are not available.
5.2 THERMAL REACTORS
This section summarizes the relevant experimental data illustrating the
effects of lead additives on thermal reactors. The data base for the dis-
cussion includes vehicle testing by Du Pont and by Ethyl, and laboratory
material studies conducted under the IIEC Fuel Composition Effects Project
and the NASA/Lewis Materials Evaluation program.
5.2.1 Erosion/Corrosion Effects
Thermal reactor durability or effective lifetime may be significantly affected
by erosive and/or corrosive deterioration caused by the presence of lead
compounds in the exhaust gas. The problem has been the subject of recent
intensive investigations. The symptoms of erosion are generally exhibited
as a deterioration of the baffles and reactor core surface in localized areas
opposite the valve ports. Du Pont (Ref. 5-10) has analyzed the erosive
behavior of a number of thermal reactor material candidates with various
fuels . Its study has shown that erosion is chemical rather than mechanical
in nature and is affected by TEL. Of the alloys tested, Uniloy 50/50 (50 per-
cent Cr, 50 percent Ni) was found to be the most resistant to attack. Uniloy
50/50 erosion rates are regarded by Du Pont as being acceptable for normal
passenger car service. Since the thermal reactor erosion tends to be
localized, Du Pont proposes to insert small patches of Uniloy 50/50 at the
erosion points and use a less expensive material for the reactor core.
Du Pont {Ref. 5-10) concludes that the mechanism of corrosion is primarily
oxidation and that the presence of phosphorus in the fuel accelerates corro-
sive attack (Fig. 5-18). Inconel 601 and Armco 18 SR were determined to
be promising corrosion-resistant materials for reactor applications.
Fuel composition effects on thermal reactor durability were investigated in
an IIEC materials evaluation program (Ref. 5-11). The materials tested (as
core specimens in a vehicle reactor) included ferritic (nickel-free) and
5-18
-------
10
6
-2
I
— • — FUELS A THROUGH F
X ci ici r
rUtL u
* FUEL H
Pb = gm/gal
P = mq/ liter
S = Wt %
FUEL
A
B
C
D
E
F
G
H
Pb P S
0 NIL 0.003
0 21 0.03
0.5 NIL 0.002
0.5 3.6 0.02
3 NIL 0.004
3 NIL 0.04
3 3.1 0.003
3 17 002
1700 1800
REACTOR TEMPERATURE, °F
1900
Fig. 5-18. Effect of Fuel Additives on Corrosion Weight Loss of
Inconel 601 (from Ref. 5-10)
austenitic stainless steels, high-nickel alloys, and various coatings on
low-cost materials. Tests of OR-1, a low-cost nickel-free alloy candidate,
showed that with leaded fuel, halides and phosphorus contribute heavily to
metal deterioration at elevated temperatures (Fig. 5-19). Leaded gasoline
without phosphorus showed considerably less erosion. Thus, the halides
and phosphorus are major contributors to material loss. The OR-1 material,
as well as other low-cost candidates, exhibited good corrosion resistance
with low-lead (0.5 gm/gal) or unleaded fuels. On the basis of these and
other similar results, the study concluded that operation with unleaded or
low-lead fuels was required to achieve satisfactory reactor life. Accordingly,
0.5-gm/gal fuel was used in all of the Ford/IIEC durability vehicle test work.
The low-cost alloy steels, however, were ultimately rejected for reactor use
5-19
-------
BARE SANDBLASTED SPECIMENS
THICKNESS
4.5
4.0
3.5
3.0
LOSS-
MILS/50 HRS. 2 °
1.5
1.0
.5
0
'3 gm/GAL LEAD, HALIDE
(MM OR AM), PHOSPHORUS
3 gm/GAL. LEAD
.0.5 gm/GAL LEAD,
HALIDE (MM),
OR NON-LEADED
1720 1760 1800
MAX. CYCLE TEMP.-°F
1840
Fig. 5-19. Effect of Fuel Variables on Average
Thickness Losses of OR-1 Alloy
During Continuous Thermal
Cycling (from Ref. 5-11)
+50
E
UJ
o
o
01
o
o
cr
o
-50
-100
-150
-200
-250
-ARMCO I8S/R
-INCONEL60I
CORE WEIGHT
(LOSS AISI 651)
NBSA-4I8A
GLASS -
CERAMIC/
651
25%
200 400 600 800
ENDURANCE TEST HOURS
1000
1200
TEST CYCLE
FOR EACH I00hr45hr AT
70mph (I900°F); 25hr AT
35mph(l500°F), 20hr
ACCELERATION, DECELERATION.
AND IDLE; 10hr COOL TO LESS
THAN 500°F, 2500N-OFF
CYCLES; 130 CYCLES TO 70mph
Fig. 5-20. Weight Change of Test Reactor Cores in Engine
Dynamometer Endurance Test (from Ref. 5-12)
5-20
-------
because of poor high-temperature strength. The most promising reactor
materials were all intermediate-cost nickel-alloy steels.
The NASA/Lewis Research Center has been conducting a thermal reactor
materials evaluation and development program for EPA (Ref. 5-12) . Coupon
screening, core testing, and full-scale thermal reactor endurance testing
(by Tele dyne-Continental) have led to the identification of two materials
which appear to be suitable for rich reactor operation with leaded fuels.
These are GE 1541 (15 Cr-4Al-lY) and Inconel 601. Armco 18 SR may also
be a candidate •
Recently, the fuel used for endurance testing in the NASA program was
switched from leaded to unleaded gasoline. No change in the weight loss
of GE 1541 or Inconel 601 was observed (Ref. 5-12). However, lead effects
may have been encountered with Dow-Corning 9458, a glass-ceramic core
material candidate which blistered and spalled in coupon testing using
leaded gasoline. When this material was tested in a core configuration using
unleaded gasoline, no deleterious effects were observed. Endurance test
results for various thermal reactor material candidates are shown in
Fig. 5-20.
The available data on materials testing with leaded fuels indicate that corro-
sion effects due to lead halides and/or phosphate compounds in the exhaust
are temperature related. Figure 5-19, for example, suggests that, at
temperatures approaching 1700°F, corrosive weight loss rates are not
sensitive to fuel composition. There is, therefore, a rational basis for the
Ethyl claim that the lead composition of fuel has no impact on the Ethyl Lean
Reactor, which operates at temperatures below 1700 °F even under high-speed
turnpike conditions (Ref. 5-13). Ethyl has found that 430 stainless steel (with
zero nickel content) has a useful life in their lean reactor of about 30, 000 miles
of road service. The same material has a life of only 17 hours when tested on
the dynamometer at 100-mph vehicle speed with retarded spark to increase
temperature. Under the same dynamometer conditions, a duplicate reactor
5-21
-------
fabricated of 310 stainless steel (20 percent nickel) showed no deterioration
for more than 200 hours (equivalent of 20,000 miles). Therefore, Ethyl con-
cludes that 310 stainless steel should provide a tenfold improvement over the
30,000-mile road service obtained with 430 stainless steel.
5.2.2 Emission Level Effects
Test data published by Du Pont (Ref. 5-10) show that lead has no effect on the
emissions of the Du Pont reactor. A 1970 Du Pont reactor vehicle (350 CID
Chevrolet of the type supplied to CARB) with 17,000 miles of operation using
conventional leaded gasoline was cleaned (combustion chamber deposits
removed) and equipped with new plugs and points. The car was then operated
for 6000 miles with unleaded gasoline on a chassis dynamometer. Emission
tests conducted before and after operation with unleaded fuel produced the
results plotted in Fig. 5-21. Du Pont regards these data as proof that opera-
tion with or without leaded fuel has no significant effect on thermal reactor
emission levels. This is supported by additional test data from a Du Pont
Type I reactor which was operated for 100,000 miles with leaded gasoline.
A small increase in HC emissions (10 percent) and CO emissions (20 percent),
compared with the initial values, was observed. Diagnostic tests by Du Pont
indicated that these minor effects were due primarily to changes in carburetor
metering. It is noted that data reported by CARB in the Du Pont reactor
vehicle fleet test program show slightly increasing HC, CO, and NO emissions
Ji
with mileage accumulation.
5.2.3 Summary
Lead has been shown to enhance, by varying degrees, the corrosion of a number
of alloys. The effect of lead in thermal reactors is that of being one more
detrimental factor over and above the already severely corrosive environment.
Based on laboratory data, leaded fuels with a TEL content of up to 0.5 gm/gal
do not significantly increase the rate of corrosion of the better oxidation-
resistant materials above that obtained with unleaded fuels .
5-22
-------
E
§.
co"
z
o
CO
0
10
0
2|-
0
O
LEADED
FUEL "
UNLEADED
""FUEL
0 5000 10,000 15,000
ODOMETER, mi
20,000 25,000
Fig. 5-21. Du Pont Type V Thermal Reactor--HC,
CO, and NOX Emissions with Leaded
and Unleaded Fuel (from Ref. 5-10)
Some alloys should be able to meet the 100,000-mile durability requirements,
However, as this degree of durability has not yet been achieved without
material failures, with or without lead, firm conclusions must await further
testing.
It appears that the durability requirements could be met with the Ethyl Lean
Reactor using leaded fuel.
5-23
-------
Fuel composition may have a significant effect on reactor material
durability at the temperature levels associated with rich reactor operation.
High-nickel-content alloys appear to be required in order to achieve satis-
factory durability. The combined presence of lead and phosphorus additives
has an accelerating influence on the corrosive deterioration of a number of
different metallic alloys.
5.3 EXHAUST GAS RECIRCULATION (EGR) SYSTEMS
With the exception of the RTR Alone, RTR plus NO Catalytic Converter plus
RTR, and Tricomponent Catalytic Converter concepts, all generic emission
control system approaches described in Section 4.3 utilize EGR for some
measure of NO control.
x
All of the available data relating to potential effects of lead additives in
gasoline on EGR systems are experimental in nature. Therefore, this sec-
tion will briefly summarize the more significant findings to date and the pro-
jected plans for EGR systems in this regard.
5.3.1 Relevant Technology Discussion
Nearly all of the reported test data were based on the use of fully leaded
gasoline (exceptions noted below). No low-lead (0.5 gm/gal) data were
available.
The most significant data base for EGR system durability effects is the
extended-use program conducted by Esso (Ref. 5-14) for NAPCA. In this
test program, an EGR system previously developed (Ref. 5-15) was evaluated
on three 1969 Plymouths and three 1969 Chevrolets over 52,000 miles under
city-surburban driving conditions simulated on a tape-controlled mileage
accumulation dynamometer. No major problems were reported. There was
no decrease in NO reduction efficiency over the 52,000 miles. Engine wear
j£
and cleanliness were considered normal for the mileage and driving regime.
In addition, the EGR system was found to be compatible with commonly
5-24
-------
employed air injection and engine modification systems used for HC and
CO control.
In the aforementioned Esso EGR extended-use program with a fully leaded
fuel, it was noted that throttle plate deposits occurred. Analysis of the
deposits indicated the main components were lead chlorides and lead bro-
mides. Despite the throttle plate deposits with leaded fuel, however, the NO
control effectiveness was unaffected. Control tests of a vehicle with unleaded
fuel indicated essentially no throttle plate deposits.
Ford (Ref. 5-16) has published data (shown as Fig. 5-22) which indicate that
the use of leaded fuel does lead to decreased EGR system efficiency. It is
stated that deposits affect flow characteristics by changing critical dimensions
or by preventing control valve seating (thereby altering the programmed rate
of recycle flow and changing the NO reduction efficiency) .
50
cT-
>-
40
30
o
CCL
8 20
§ 10
0
/-UNLEADED
(2) /
UNLEADED FUEL
(2)
LEADED FUEL
(6)
THE NUMBERS IN PARENTHESES REPRESENT
THE NUMBER OF VEHICLES USED IN CALCULATING
THE PLOTTED AVERAGE VALUE
0
3000
6000
9000
MILEAGE
Fig. 5-22. Effect of Leaded Fuel on the Control Efficiency of Air
Cleaner EGR Systems (302 CID Engine)
(from Ref. 5-16)
5-25
-------
The Esso EGR tests utlized above-the-throttle recycle injection with rather
large (approximately 1/2 inch) flow orifices for approximately 15-17 percent
EGR flow rate. The orifice size in the Ford air-cleaner injection systems
(Fig. 5-22) is not stated in the reference.
The limited amount of test data in Ref. 5-14 pertaining to EGR NO reduction
Ji
in the same vehicle, using both leaded and unleaded gasoline, indicates the
same nominal level of NO reduction for both fuels.
x
5.3.2 1973-74 EGR Systems
During visits by The Aerospace Corporation to General Motors, Ford,
Chrysler, and American Motors (Refs . 5-17 to 5-20), all four companies
stated that EGR was necessary and would be employed to meet the 1973-74
emission standards for NO (3 gm/mi).
There was a general consensus that lead could cause deposit and life problems
in the EGR system. The degree of the effect would be dependent upon the EGR
tap-off location and injection orifice size used, i.e., the smaller the hole
size the more likely clogging could or would occur.
Ford referred to previously published data (Fig. 5-22, from Ref. 5-16) which
indicated a decrease in effectiveness to 0.15 at 6000 miles from 0.35 initially
when a leaded fuel was used (7 to 10 percent EGR flow rate).
American Motors expects to use approximately 10 percent EGR in 1973 for
most models, particularly with V-8 engines and/or large vehicles. Its cur-
rent test information shows that this results in a substantial increase in intake
manifold, EGR system, combustion chamber, and exhaust valve seat deposit
buildup. The deposit is presumed to be largely lead, although its exact com-
position has not been identified. The deposit may be responsible for the
increased tendency to burn exhaust valves, increased valve leakage, and
EGR system plugging. It is felt that these problems are proportional to the
amount of lead additive in the fuel at a given recirculation rate. These severe
problems of deposit buildup were only identified by American Motors.
5-26
-------
5.3.3 1975-76 EGR Systems
It appears that in general (Refs. 5-17 to 5-20), the 1975-76 EGR systems
will be similar to those finally selected for the 1973-74 cars, except for
possible orifice size changes to accommodate increased EGR flow rates and
associated necessary EGR control modifications.
5.3.4 Summary
The foregoing brief discussion and assessment of relevant EGR technology
indicate the following, with respect to lead additives:
1. The presence of lead additives in gasoline does not, per se,
significantly affect the NOX reduction performance or basic
durability of EGR systems, based on the very little test data
available.
2. The presence of lead additives in gasoline can result in deposits
in EGR orifices, throttle plate areas, etc. The actual severity
of such deposits, in terms of NOX reduction efficiency, etc. ,
would appear to be strongly related to the particular type of EGR
system employed as well as to control orifice sizes used, and/or
to the utilization of self-cleaning designs (plungers, specially
coated surfaces, flexible snap-rings, etc.) in areas susceptible
to deposit buildup.
3. Lead-free or low-lead gasoline is not, therefore, required for
the implementation of EGR systems, per se.
5-27
-------
REFERENCES
5-1. L.J.E. Hofer, J.F. Schultz and J. J . Feenan, Bureau of Mines
Report No. RI 6243 (1963).
5-2. R.S. Yolles, H. Wise and L.P. Berriman, Study of Catalytic Control
of Exhaust Emissions for Otto Cycle Engines, Final Report SRI
Project PSU-8028, Stanford Research Institute (April 1970).
5-3. G.H. Meguerian, "Nitrogen Oxide--Formation, Suppression and
Catalytic Reduction," PD 23, Paper 3, American Oil Company,
Research and Development Dept., Whiting, Indiana.
5-4. K.I. Jagel and F.G. Dwyer, "HC/CO Oxidation Catalysts for Vehicle
Exhaust Emission Control," SAE Paper No. 710290 (January 1971).
5-5. E.E. Weaver, "Effects of Tetraethyl Lead on Catalyst Life and
Efficiency in Customer Type Vehicle Operation, " SAE Paper No.
690016 (January 1969).
5-6. R.M. Yarrington and W.E. Bambrick, "Deactivation of Automobile
Exhaust Control Catalyst," APCA Journal 20. No. 6 (June 1970), p. 398.
5-7. R.R. Allen and C-G. Gerhold, "Catalytic Converters for New and
Current (Used) Vehicles," Paper presented at the Fifth Technical
Meeting, West Coast Section of the Air Pollution Control Association,
San Francisco, California (8-9 October 1970).
5-8. E. E. Hancock, R.M. CampauandR. Connolly, "Catalytic Converter
Vehicle System Performance: Rapid Versus Customer Mileage," SAE
Paper No. 710292 (January 1971).
5-9- H.W. Schwochert, "Performance of a Catalytic Converter on Non-
leaded Gasoline," SAE Paper No. 690503 (May 1969).
5-10. W.J. Barth and E.N. Cantwell, "Automotive Exhaust Manifold Thermal
Reactors--Materials Considerations," Presented before the Division
of Petroleum Chemistry, Inc., 161st Meeting of the American Chemical
Society, Los Angeles, California (28 March-2 April 1971).
5-11. A. Jaimee, "Thermal Reactor—Design, Development and Perfor-
mance," SAE Paper No. 710923 (January 1971).
5-28
-------
REFERENCES (cont.)
5-12. R.E. Oldrieve and P.L. Stone, Letter Report on the Current Status
of the NASA/Lewis Program for Evaluation and Development of
Materials for Automobile Thermal Reactors, Materials and Structures
Division, NASA/Lewis Research Center, Cleveland (30 April 1971).
5-13. The Ethyl Lean Reactor System, Ethyl Corporation Research Labora-
tories, Detroit (1 July 1971).
5-14. G.S. Musser, et al., "Effectiveness of Exhaust Gas Recirculation
with Extended Uses," SAE Paper No. 710013 (January 1971).
5-15. W. Glass, etal., "Evaluation of Exhaust Recirculation for Control of
Nitrogen Oxides Emissions," SAE Paper No. 700146 (January 1970).
5-16. H.L. Misch (Vice President--Engineering, Ford Motor Company),
Testimony before the California Air Resources Board, Sacramento,
California (4 March 1970).
5-17. Personal discussion with representatives of the General Motors Corpo-
ration (30 June 1971).
5-18. Personal discussion with representatives of the Ford Motor Company
(1 July 1971).
5-19. Personal discussion with representatives of the Chrysler Corporation
(29 June 1971).
5-20. Personal discussion with representatives of the American Motors
Corporation (29 September 1971).
5-21. D.L. Davis and G.E. Onishi, SAE Paper No. 486F (March 1962).
5-22. Engelhard Industries, Inc., Letter to The Aerospace Corporation
(13 October 1971).
5-23. Personal discussions with representatives of the Ford Motor Company
(30 September 1971).
5-29
-------
6. FEASIBILITY/IMPLICATIONS
OF LEAD TRAPS/EXHAUST
SCRUBBERS
-------
SECTION 6
FEASIBILITY AND IMPLICATIONS OF LEAD TRAPS
AND EXHAUST GAS SCRUBBER DEVICES
Although lead is not compatible with catalytic converters, it is not clear that
it has to be removed from gasoline for this reason alone. A variety of chem-
ical scrubbers which efficiently convert volatile lead halides to nonvolatile
compounds that are nontoxic to catalysts are being developed and might be
used. However, most of these scrubbers require periodic maintenance that
involves the replacement of certain chemicals. This shifts the maintenance
problem from more frequent replacement of the catalyst, which would be
required if no scrubber were used, to replacement of chemicals.
An investigation conducted to determine the feasibility and cost of chemical
or mechanical lead traps upstream and downstream of emission control
devices culminated in the collection of information from seven organizations
with a total of eight devices carried to various stages of development. The
techniques employed by these devices were found to fall into two basic cate-
gories. The first technique is capable of removing lead in the gaseous as
well as particulate form at exhaust manifold temperatures, so that the
exhaust is held sufficiently hot for further catalytic treatment to control NO ,
Jfc
HC, and CO emissions. The second technique requires that exhaust gases be
cool enough for a sufficient amount of lead to be in particulate form for col-
lection. A detailed treatment of lead traps and exhaust gas scrubber tech-
nology is presented in Appendix C.
A study of the physical characteristics of exhausted lead compounds at the
efficient operating temperatures of catalytic reactors indicated that lead
would exist in the vapor state; thus, only those devices utilizing the first
technique were considered practical for use in this application. With the
exception of the Atomics International molten carbonate lead trap device, all
techniques required that lead exist in the particulate form for collection and
6-1
-------
removal. As a consequence, the Atomics International device was the only
lead trap candidate that could be considered as practicable for use upstream
of catalytic reactors .
The Atomics International lead trap device was developed on the basis that
lead halides which are mildly acidic can be expected to react chemically with
a basic alkali carbonate. Thus, a molten carbonate salt was evolved for
scrubbing gaseous and particulate lead compounds from automobile exhausts .
This device has undergone considerable development and testing and the
results suggest that a molten carbonate lead trap design could be installed
upstream of a catalytic reactor and have the potential for removing 90 per-
cent of the lead, essentially all the sulfur oxides, and in excess of 80 percent
of all particulates over the entire operating range of the automobile. In addi-
tion, the molten carbonate device compares favorably to standard mufflers
with regard to engine noise attenuation and backpressures in the exhaust sys-
tem. During 7-mode cycling tests, a prototype ^asign of the molten carbon-
ate process achieved operating temperatures of 1000 F in slightly more than
2 minutes, with gas temperature drops across the device never exceeding
20 F. A detailed cost estimate of an under-the-car design capable of fitting
the available space on all currently operational automobiles was conducted by
Atomics International, and it was indicated that the cost to the user would
range from $35 for a factory installation to $45 for retrofit units . It is
presently estimated that the salt will require changing every 15,000 to 20,000
miles at a cost of $10.
Aside from system design complexities and the need for adding another com-
ponent to the already complicated emission control systems evolved to date,
it is felt that the 90-percent lead removal capability from regular leaded
gasoline predicted by Atomics International for their device is not adequate
for the lead-sensitive catalysts presently predicted for use by the automobile
industry. If one considers the use of low-lead gasoline (0.5 gm/gal), the
Atomics International system might be compatible with catalyst maximum
lead requirements. However, there have not been sufficient hardware tests
6-2
-------
to establish reliable removal efficiencies, and durability tests on prototype
systems are not available to assess the decrease in effectiveness versus
mileage. Therefore, it is not felt that this system could be incorporated in
1975-76 model automobiles.
The removal of particulate lead compounds in the downstream portion of
automobile exhaust systems can be accomplished by cyclone trap devices.
However, such inertial separation schemes would only be applicable to
exhaust systems in which catalytic reactors were either not present or were
insensitive to lead compounds. Development and testing by both Du Pont and
Ethyl indicate that 65 to 85 percent of the particulate lead compounds can be
removed by this type of device.
6-3
-------
7. EFFECT OF LEAD
ADDITIVES ON OTHER
ENGINE PARTS
-------
SECTION 7
EFFECT OF LEAD ADDITIVES ON OTHER ENGINE PARTS
This section addresses the effects of lead, per se, on the basic engine
components and systems. General factors related to engine durability are
briefly discussed and detailed information is provided on certain components
which are significantly affected by lead additives. In addition, fuel-associated
maintenance costs are examined and their cost considerations identified. The
effects with respect to emission levels and emission control devices are
separately discussed elsewhere in this report.
7.1 ENGINE DURABILITY--GENERAL
The lead salts formed by the halogens added to leaded fuels for scavenging
purposes would be expected to cause an increase in engine wear. As early as
the 1940's, laboratory and proving ground tests were run with engines operating
on leaded and unleaded fuels in order to determine this effect. Although the
results from these tests showed slight but consistently greater wear with
leaded fuel operation, it was concluded (Ref. 7-1) that "the increased wear,
although clearly existent in practically all parts of the engine, was not con-
sidered to be of sufficient amount to be of practical importance."
A number of test programs have been run in more recent years which were
also directed toward determining the effect on engine durability of leaded and
unleaded fuel operation (Refs . 7-2 through 7-23). These investigations pro-
vided comparative data on varnish/sludge formation, piston ring gap increase
and ring weight loss, cylinder bore wear, bearing wear, oil contamination,
and rust. Although many conflicts were noted, the data generally confirmed
a degradation in engine durability due to the effects of lead and its associated
scavengers, as reported in the earlier tests.
Although these conditions which cause wear might be expected to affect the
time until mechanical repair would be required, this effect was not quantified
7-1
-------
in any of the data sources. As a result, these data are of little value for
predicting the maintenance cost differential between operation with leaded
or unleaded fuel. In retrospect, this is not surprising as evidenced by the
extensive and costly controlled fleet test programs more recently instituted
by American Oil (AMOCO), Du Pont, and the Ethyl Corporation in order
to assess this effect.
7.2 FUEL-SENSITIVE COMPONENTS
From the results of the controlled fleet test program, exhaust systems and
spark plugs were noted to be particularly cost sensitive to the effects on
leaded operation. Since they have a major influence on the fuel-related
maintenance costs, they are separately discussed in some detail. In addition,
the valve recession with unleaded fuel is also covered since it too has cost
implications.
7.2.1 Exhaust System
Reference 7-2 states that the cost to the motoring public is in excess of $0.5
billion per year for muffler and exhaust pipe replacements . The high cost in
this area is attributed to the corrosive nature of the exhaust products and the
incapability of the materials used in exhaust systems to withstand this corro-
sive attack. Although the halogen acids present in the exhaust products with
operation on leaded fuel contribute to this condition, there are little data
which quantify this effect. Based on limited test conditions, it was reported
(Ref. 7-3) that muffler life would be approximately doubled with operation on
unleaded instead of leaded fuel. These limited data apparently have gained
general acceptance and reference to this source is made in much of the
literature which reflects this order of improvement in muffler life for unleaded
fuel operation.
Although muffler life data for leaded fuel operation can be derived from the
replacement numbers stated in the controlled fleet test programs, these data
are not necessarily valid since the replacement interval could be quite sensi-
tive to the test mileage cutoff time. An assessment of muffler life, therefore,
was obtained based on data from Ref. 7-4. From these data, a muffler life
7-2
-------
of 37,500 miles was indicated. Considering the life of the original
equipment provided, this would result in a replacement of 1.67 mufflers for
100, 000 miles of operation with leaded fuel.
The expected muffler life for unleaded operation would be even more sensitive
to the test mileage cutoff time. Since the data in Ref. 7-4 are not applicable
to this case, a statistical analysis was made of data provided by Ethyl for
their fleet located in the Detroit area. Results of this analysis, presented
in Fig. 7-1, show a median life of 77, 500 miles. This would result in a
replacement of 0.29 muffler for 100, 000 miles of operation with unleaded
fuel.
Exhaust system costs are a particularly difficult item to estimate because of
the many, many differences in configuration and associated component costs.
180
ISO
140
120
o
o
o
in
a so
60
40
20
1 I I
MEAN LIFE = 77,288 mi
STANDARD DEVIATION = 21.427 mi
MEDIAN LIFE = 77,500mi
i I I I i
001 OOSQI 02 09 I 2 5 |Q 20 30 40 5Q 60 70 80 90 95 98 99 998 999 9999
PROBABILITY OF FAILURE, %
Fig. 7-1. Muffler Lifetime Probability (Operation on Unleaded
Fuel)
7-3
-------
This is further compounded by differences in replacement charges made by
car dealers, garages, service stations, and specialty shops. Based on
prices obtained from a number of sources, which included Chilton's Manual,
specialty shops, and a large mail order/retail firm, the cost for a typical
exhaust system replacement (muffler, pipes, fittings) is estimated to be $45.
Based on the cost and number of replacements required for 100, 000 miles of
operation, then, the exhaust system expense would be 0.075 cent per mile
for leaded fuel and 0.013 cent per mile for unleaded fuel.
7.2.2 Spark Plugs
7.2.2.1 Misfiring Mechanism
Spark plug misfiring is generally caused by one of three conditions: gap
bridging, gap erosion or a reduction in shunt resistance (Ref. 7-5). The latter
condition is particularly sensitive to leaded fuel operation where the lead
oxides and lead chlorobromide deposits form on the internal ceramic and
provide a conductive path from the center electrode to the spark plug base.
In general, the conductivity of these deposits increases with temperature.
The level of conductivity, however, is higher for deposits which are formed
at the low temperature associated with low-power operation. Hence, under
conditions of high acceleration, misfiring would be promoted since the gen-
eration of high temperatures is attained faster than the change in deposit
composition. At sustained high-power operation, however, the initial deposit
composition would change to a composition with lower conductivity and the
conditions promoting misfiring would be alleviated.
The influence of lead and its accompanying chlorobromide scavengers on the
gap bridging and gap erosion mechanism for spark plug misfiring is not as
evident. Although heavy deposit accumulation on and around the electrodes
with leaded fuel operation has been reported (Ref. 7-6), the composition of
these deposits is not identified. A detailed understanding of both the mecha-
nism for formation and the mechanical/chemical characteristics of these
deposits is required to separate the effects due to lead additives on this
7-4
-------
misfiring mechanism. It has been reported (Ref. 7-7) that lead additives
have only a slight effect on the corrosion of nickel-alloy spark plug electrodes,
and that the effects of sulphur contamination in a reducing atmosphere (rich
mixture) are of much greater significance. Again, more data are required
to quantify the effects of lead additives for this type of misfiring mechanism.
7.2.2.2 Life
Although a complete understanding of the misfiring mechanisms could provide
a basis for establishing spark plug life for either leaded or unleaded fuel
operation, it is doubtful that this information would be particularly meaningful
except for establishing a car manufacturer's recommended-mile-change inter-
val. Other driver-related factors such as driving characteristics, sensitivity
to engine malperformance, and maintenance habits could well be overriding
effects.
Again, for the same reason given for the determination of muffler life,
spark plug life for leaded fuel operation was obtained from annual spark plug
sales. From Ref. 7-4, spark plug life was determined to be approximately
13,000 miles per set. This figure was based on an assumed average annual
vehicle mileage of 10,000 miles and 7.3 spark plugs per vehicle. If the set
of spark plugs that was provided as original equipment is included, the spark
plug life of 13,000 miles per set translates to a replacement of 6.7 spark plug
sets for 100, 000 miles of operation on leaded fuel.
For the unleaded fuel case the only data available for the determination of
spark plug life are those reported by the controlled fleet test operations.
Again, a statistical analysis was performed on data provided by Ethyl; the
analysis considered the actual replacement mileage as well as the mileage of
operation subsequent to replacement. The results of this analysis, presented
in Fig. 7-2, show a median life of 20, 350 miles for the Detroit fleet and
25,050 miles for the Baton Rouge fleet. The average of these values (22, 700
miles) was used to establish a replacement of 3.4 spark plug sets for 100,000
miles of operation on unleaded fuel.
7-5
-------
120
100
80
o
o
o
- 60
to
40
20
BATON
DETROIT ROUGE
MEAN LIFE 22,979 27,652 mi
SB ».™ 16,474 ™
MEDIAN LIFE 20,350 25,050 mi
BATON ROUGE•
0
0.01 0.1
10 40 80 95
PROBABILITY OF FAILURE, %
99 999 9999
Fig. 7-2. Spark Plug Lifetime Probability
(Operation on Unleaded Fuel)
7.2.2.3 Cost
Spark plug replacement costs are also quite variable . Replacement cost
per the Chilton Manual averages about $17; for the do-it-yourself owner, the
cost would be around $5, assuming discount-priced spark plugs; and the ser-
vice station charge would be somewhere between these values. A replacement
cost of $10 per set was somewhat arbitrarily assumed for this study.
With the replacement of 6.7 and 3.4 sets for 100,000 miles of operation as
derived above, spark plug costs would be 0.067 cent per mile for leaded
fuel and 0.034 cent per mile for unleaded fuel.
7-6
-------
7.2.3 Exhaust Valve Recession
There is considerable evidence that exhaust valve seat wear increases with
the removal of lead additives (Refs . 7-8 through 7-13). At operating condi-
tions of sustained high speed and load, exhaust valve recession rates have
been reported as high as a 0.010 inch per 1,000 miles (Ref s. 7-10 and 7-12).
Not all engine types, however, are similarly affected. Volkswagen, for
example, anticipates no exhaust valve seat problems with use of unleaded
fuels. It is also of interest to note that this wear mechanism was not apparent
from the data obtained by the controlled fleet test programs for vehicles
operating with unleaded fuel.
The mechanism for valve recession with operation on unleaded fuels is
reported in Refs. 7-8, 7-9, 7-12, and 7-13; it is attributed to localized welding
and subsequent pull-out of fragments with accompanying wear due to abrasion.
This condition is apparently alleviated in leaded fuels by the lead chlorobro-
mide deposits which act as high-temperature solid-film lubricants. Although
a number of chemical and/or metallurgical approaches could be applied to
reduce the valve wear condition apparent with the use of unleaded fuels, the
change to induction-hardened exhaust valve seats, as recommended in Ref.
7-14, appears to be an immediate and reasonably low-cost solution to the
problem.
7.3 MAINTENANCE
Fuel-related maintenance costs have been collected in three controlled fleet
test programs for comparable vehicle groups operated on leaded and unleaded
fuel. The results are of particular interest since, as it was concluded above,
they represent the only source of data for the determination of such costs.
Unfortunately, the results obtained in the three programs are not in agree-
ment. AMOCO (Ref. 7-15) reported an average fuel-related cost differential
of 0.418 cent per mile in favor of unleaded fuel. The fuel-related mainte-
nance costs reported by Du Pont (Ref. 7-13) resulted in a differential of
0.052 cent per mile in favor of unleaded fuel. Ethyl (Ref. 7-16) reported a
7-7
-------
cost differential of 0.078 cent per mile, also in favor of unleaded fuel. In
supplemental information provided by Ethyl, this differential was reduced to
less than 0.016 cent per mile by elimination of some of the fuel-related
maintenance items previously reported.
Since there is a significant difference between AMOCO results and those
reported by Du Pont and Ethyl, an evaluation was made to establish the
reason for the disparity. Unfortunately, the only data available from AMOCO
were total fuel-related maintenance costs, and the number of spark plug and
muffler changes. Further, these data were also compromised by a limited
mileage accumulation which averaged less than 25,000 miles per vehicle.
Nevertheless, it was found that the data were quite useful in the evaluation
since spark plug and muffler expenditures represent a sizeable percentage of
fuel-related maintenance costs.
Examination of AMOCO data showed that spark plug and muffler replacement
costs represented approximately 40 percent of fuel-related maintenance costs
for leaded fuels and approximately 31 percent for unleaded fuels. Expressed
on a cents-per-mile basis, these percentages reflect a cost of 0.251 cent per
mile for leaded fuels and 0.067 cent per mile for unleaded fuels. These num-
bers were derived on the basis of a spark plug replacement cost of $10 per
set and an exhaust system replacement cost of $45 per replacement.
From examination of Du Pont data, spark plug and exhaust system replace-
ment costs represented approximately 29 percent for leaded fuels and 14 per-
cent for unleaded fuels. On a cents-per-mile basis, these percentages
reflect a cost of 0.123 cent per mile for the leaded fuels and 0.052 cent per
mile for the unleaded fuels.
From examination of Ethyl data, spark plug and exhaust system replacement
costs represented approximately 38 percent for leaded fuels and 22 percent
for unleaded fuels. These percentages reflect a cost of 0.134 cent per mile
for leaded fuels and 0.062 cent per mile for unleaded fuels.
7-8
-------
It should be noted that the Du Pont and Ethyl data were used to directly
compute the percentages and costs given above. A check using the derived
costs that were applied to the AMOCO spark plug and exhaust system replace-
ments, however, showed an insignificant difference in results by the use of
these derived costs.
On the basis of percentage of spark plug and exhaust system cost to the total
fuel-related maintenance cost, the three data sources are in reasonable
agreement. Further, on a cents-per-mile basis the costs reported for the
unleaded fuel vehicles were in surprisingly close agreement. For the leaded
fuel vehicles, however, AMOCO costs were approximately double that
reported by Du Pont and Ethyl. These values are summarized in the table
below:
Table 7-1. Spark Plug and Exhaust System Costs
Source
AMOCO
Du Pont Company
Ethyl Corporation
Leaded Fuel
% of Total
40
29
38
f/Mile
0.251
0.123
0.134
Unleaded Fuel
% of Total
31
14
22
f/Mile
0.067
0.052
0.062
To provide a basis for judging which costs were more reasonable, the pre-
viously derived exhaust system and spark plug life-and-cost data for the leaded
case were used to arrive at a cost of 0.142 cent per mile for these replace-
ments. Since this cost is in close agreement with the 0.123 and 0.134 cent
per mile reported by Du Pont and Ethyl, it generally substantiates the reason-
ableness of their cost data. The very much higher costs reported by AMOCO
apparently reflect a condition which for some reason is not typical.
If the fuel-related maintenance costs as reported by both Du Pont and Ethyl
are accepted as valid, it is of interest to note that the difference in mainte-
nance cost between the leaded and unleaded vehicles is almost exclusively
7-9
-------
that associated with the difference in spark plug and exhaust system expense.
For example, if spark plug and exhaust system costs are subtracted from the
Du Pont data, the difference in fuel-related maintenance cost would be 0.019
cent per mile in favor of leaded fuel. Similarly for the Ethyl data, the dif-
ferential would be 0.006 cent per mile, but in favor of the unleaded fuel.
On the assumption that spark plug and exhaust system replacement costs
reflect the essential difference in fuel-related maintenance costs, the
derived life-and-cost data for both the leaded and unleaded cases can be used
to independently determine the difference in fuel-related maintenance costs.
As given above, the cost would be 0.142 cent per mile for the leaded case;
similarly for the unleaded case the cost would be 0.047 cent per mile. A
differential fuel-related maintenance cost of 0.095 cent per mile is thus
indicated in favor of unleaded fuel operation. This would amount to approxi-
mately $81 over the lifetime of an average automobile.
7-10
-------
REFERENCES
7-1. W.S. James, "Piston-Ring and Cylinder-Bore Wear," SAE Journal
(August 1961), pp. 33-40.
7-2. R.A. Heath, "Muffler Corrosion--It's Cause and Control," SAE
Transactions (1959), p. 553.
7-3. F.J. Cordera, et al., "TEL Scavengers in Fuel Affect Engine Per-
formance and Durability," SAE Paper No. 877A (June 1964).
7-4. 1968/Automobile Facts/Figures Report, Automobile Manufacturers
Association, Inc., Detroit (1968).
7-5. H.P. Julien and R.F. Neblett, "Spark Plug Misfiring--Mechanism
Studies," SAE Preprint 123-T (October 1959).
7-6. Fuel Composition--Its Relationship to Emission Control on Future,
Present, and Past Model Year Vehicle Systems, Ford Motor Company
(7 March 1970).
7-7. C.M. Heinen (Chrysler Corporation), Letter to The Aerospace
Corporation (14 September 1971).
7-8. W. Giles, "Valve Problems with Lead-Free Gasoline," SAE Paper
No. 710368 (22 October 1970).
7-9. D. Godfrey and R.L. Courtney, "Investigation of the Mechanism of
Exhaust Valve Seat Wear in Engines Run on Unleaded Gasoline," SAE
Paper No. 710356 (January 1971).
7-10. H.W. Schwochert, "Performance of a Catalytic Converter on Non-
leaded Gasoline," SAE Paper No. 690503 (May 1969).
7-11. Consequences of Removing Lead Antiknocks from Gasoline, A Status
Report, No. AC-10, Ethyl Corporation, New York (August 1970).
7-12. A.E. Felt and R.V. Kerley, "Engines and Effects of Lead-Free
Gasoline," SAE Paper No. 710367 (October 1970).
7-13. Effect of Lead Antiknocks on the Performance and Costs of Advanced
Emission Control Systems, Du Pont de Nemours & Co. , Wilmington,
Deleware (15 July 1971).
7-14. W. Giles, "Induction Hardening Makes Exhaust Valve Seats Wear Less
with Nonleaded Fuel, " Presented at SAE International Mid-Year
Meeting, Montreal, Canada (7-11 June 1971).
7-11
-------
REFERENCES (cont.)
7-15. H.R. Taliaferro, L.T. Wright and R.C. Mallatt, "Gasoline for
Reducing Automobile Pollution, " An address before the American
Association for the Advancement of Science Symposium, 137th Meeting,
Chicago, Illinois (26 December 1970).
7-16. Car Maintenance Expense When Using Leaded and Nonleaded Gasoline,
Ethyl Corporation, Detroit (2 July 1971).
7-17. A.J. Pahnke and J.F. Conte, "Effect of Combustion Chamber Deposits
and Driving Conditions on Vehicle Exhaust Emissions," SAE Paper
No. 690017 (January 1969).
7-18. L.G. Pless, "Some Effects of Experimental Vehicle Emission Control
Systems on Engine Deposits and Wear," SAE Paper No. 710583 (June
1971).
7-19. L.G. Pless, "The Effects of Some Engine, Fuel and Oil Additive
Factors on Engine Rusting in Short-Trip Service," SAE Paper
No. 700457 (May 1970).
7-20. G.S. Musser, et al., "Effectiveness of Exhaust Gas Recirculation with
Extended Use," SAE Paper No. 710013 (January 1971).
7-21. R.R. Allen (Universal Oil Products Company), Letter to The Aerospace
Corporation (14 September 1971).
7-22. W.R. Epperly, "Future U.S. Emission Control Systems," Presented
to the Environmental Protection Agency, Washington, D.C. (6 May 1971)
7-23. W.G. Agnew, "Gasoline Changes Affecting Emission Control," Pre-
sented to the Oil Companies, Office of Science and Technology, and
HEW at the General Motors Technical Center (January-March 1970).
7-12
-------
8. COST ANALYSIS
-------
SECTION 8
COST ANALYSIS
The cost implications arising from the incorporation of emission control
systems in cars to meet the 1975-76 emission standards encompass the
areas of (1) initial investment costs, (2) maintenance-related costs, and
(3) operating costs. In view of the fact that meeting the 1975-76 standards
has not been adequately demonstrated by vehicle manufacturers and that
no final selection of system components has been made, the cost estimates
presented herein are limited to engineering estimates based on projections
of current designs.
Initial investment cost, to the car purchaser, is the increase in initial
purchase price occasioned by the installation of the emission control system
hardware in the new car. Maintenance-related cost is that increase in
maintenance cost directly attributable to the inclusion of the emission
control system. Operating cost, similarly, is that increase in operating
cost (i.e., fuel cost) directly related to changes in either the vehicle fuel
economy resulting from the emission control system, or an increase in
basic fuel cost per gallon as required by the nature of the emission control
system.
The approach followed in this section is to select a baseline vehicle having
performance and fuel economy characteristics typical of cars not employing
sophisticated emission control systems; i. e. , 1970 cars designed to operate
within the constraints of the normal two-grade leaded gasoline supply
system. Then, selected generic emission control systems (as previously
defined in Section 4. 3) are added to the baseline vehicle and typical fuel
economy penalties associated with each generic class are assessed. The
fuel penalty consists of the control device effects and any compression
ratio limits occasioned in the case of generic classes of emission control
8-1
-------
devices constrained to operation on unleaded gasoline. Finally, for each
generic class overall costs to the consumer (initial, maintenance, and
operating costs) are summed up based on an average vehicle operational
lifetime.
8. 1 CONTROL DEVICE/SYSTEM COST ANALYSIS
8. 1. 1 Control Device Costs
Specific control devices and their associated engine- or vehicle-related
requirements were grouped in the general areas of (1) engine modifications,
(2) emission control system components, and (3) exhaust system initial
costs. The cost estimates presented are based on cost of materials used
in the device, difficulty of manufacturing, comparisons with existing
automotive components, and discussions with automotive vehicle and
equipment manufacturers.
8. 1. 1. 1 Engine Modifications
8. 1. 1. 1. 1 Carburetion
All auto manufacturers have stressed the need for improved carburetion
(and/or fuel injection) for improved air-fuel ratio control in engines
employing complex emission control systems. Informal estimates of hard-
ware cost increases vary widely. For purposes of the present study, a
differentiation was made between "lean" and "rich" systems. A nominal
increase of $13 was assessed for improved carburetion for "rich" systems.
For the more difficult "lean" systems, which require better mixture
preparation and distribution as well as air-fuel ratio control, this cost
increase was selected to be $25. Although fuel injection systems may be
utilized on some car models, fuel injection system costs were not included
as a variable since this system has not been identified as a requirement
for any generic emission control system concept.
8.1.1.1.2 Ignition/Distributor/Control Systems
All automakers have stressed the need for some form of "unitized" ignition
system for longer life and better control of ignition spark to prevent
8-2
-------
degradation of emission levels. Some have also expressed the opinion that a
new engine control system may be required to provide more precise control
of the variables and coordinate spark timing and carburetion. For purposes
of the present study, it was assumed that all advanced emission control
system concepts would incorporate features of this'general type, and an
estimated cost penalty of $37 was assessed all generic system classes.
8. 1. 1. 1. 3 Long-Life Exhaust Systems
Exhaust system components become part of or are replaced by emission con-
trol system components in nearly all advanced emission control system con-
cepts. Also, exhaust gas temperature levels are generally increased
(severely, in some cases) to provide more optimum operating conditions for
certain control system components (e.g., catalytic converters) or because of
the operation of certain components [e. g., rich thermal reactor (RTR)].
Because of the postulated requirement for 50, 000 maintenance-free miles of
operation, and the interaction between control system components and normal
exhaust system components, a long-life exhaust system (e.g., stainless
steel) was assumed to be included in every generic class of emission control
system. An initial cost of $60 was estimated for this system, and a cost
credit of $28 allowed because of the deletion of the normal exhaust system,
resulting in a net additive cost penalty of $32 for every generic class.
8.1.1.1.4 Exhaust Valves and Seats
As discussed in Section 7, the use of lead-free gasoline, for most cars,
will result in the requirement for valves and/or valve seat modifications to
prevent valve recession effects. Any generic class of emission control
systems operating on unleaded gasoline was assessed at $3-per-car cost
penalty for such modifications.
8. 1. 1. 2 Emission Control System Components
8.1.1.2.1 Exhaust Gas Recirculation (EGR) Systems
A wide variety of EGR system designs exists, as was shown in Section 4. 2. 2.
Similarly, there is a rather wide variation in available cost estimates for
8-3
-------
EGR systems, per se (see Table 8-1). For purposes of the present study,
a representative value of $25 was assessed all systems incorporating EGR.
Table 8-1. EGR System Cost Data
Source
Chrysler Corporation
Du Pont (EGR System)
Du Pont (In-line filter)
Esso Research b
Engineering
Universal Oil Products
Research Triangle
Institute (1970)
Research Triangle
Institute (1971) . EGR
system
Materials
Cost
(dollars)
Z.96
0.35
Manufactured
Cost
(dollars)
3.70
0.44
~10.00
Cost to
Consumer
(on new car)
(dollars)
30.00
7.40
8.28
0.88
Z5.00
25.00
20.00
Maintenance
Costs
(dollars)
8.00/year
Description
1973-74 system (Ref. 8-1)
Cooled gas, above -throttle
injection (Ref. 8-2)
Cooled gas, above-throttle
injection (Ref. 8-3)
(Ref. 8-4)
(Ref. 8-5)
Fixed-orifice system (Ref. 8-6,
unpublished draft)
8. 1. 1. 2. 2 Catalytic Converters
Considering the state of development for catalytic converters, there is an
understandable lack of cost data and a reluctance on the part of the catalyst
manufacturers and the automobile companies to estimate what the cost will
be. However, some speculative costs have been obtained from available
reports and by visits to certain sources as shown in the following table:
Consumer Cost per Car, Dollars
Source
HC/CO
UOP
Engelhard
Esso
RT1
SRI
150
50 50
75 75
130 (avg)
84
92
Tricomponent
120
Reference
8-4
8-7
8-8
8-6
8-9
8-4
-------
Since there was a great deal of disagreement on catalyst converter costs and
a general lack of cost data, an independent estimate was made. This esti-
mate considered: (1) catalyst cost, (2) converter material cost, (3) manu-
facturing labor, overhead, and profit, (4) installation costs, and (5) sales
profits. The resulting values are:
Converter Type
HC/CO Catalytic Converter
Base or Noble Metal
Dual-Bed Converter
Base or Noble Metal
Tricomponent Converter
Noble Metal
Original Equipment Cost to
Consumer, Dollars
98/car
129/car
98/car
It is interesting to note that base metal and noble metal catalyst costs per
vehicle are about the same even though the cost per pound is much greater
for the noble metal catalyst. The reason of course is that a smaller amount
of noble metal catalyst is required. With the cost of the base metal and
noble metal catalyst equalized, the remaining cost for metal fabrication,
profit, installation, etc. , also tends to equalize such that cost to the
consumer is about the same for the two types of catalytic converter.
8.1.1.2.3 Thermal Reactors
Du Pont (Ref. 8-2) estimated that the cost of their RTR is approximately
$48 (two times the estimated manufacturing cost of approximately $24).
Ethyl (Ref. 8-13) estimated the cost of their lean thermal reactor (LTR) to
be approximately $100 (including an upgraded exhaust pipe).
Again, because of these substantial variances and general lack of cost data
from the automakers, an independent estimate was made. This estimate
considered volumetric and materials differences between the LTR and RTR
approaches and included considerations of materials cost, manufacturing
8-5
-------
labor and profit, installation costs, sales profits, and credit for standard
exhaust manifolds. The resulting values are:
Reactor Original Equipment Cost to
Type Consumer, Dollars
RTR 125/car
LTR 110/car
In a similar manner, the cost of a "low-grade" thermal reactor, rich or lean,
was estimated to be $70 per car. Such "low-grade" reactors would be
smaller in volume, have less insulation, might not have a core liner, etc.,
and would approach an oversize standard exhaust manifold in configuration.
8. 1. 1.2.4 Air-Injection Pump
A nominal cost of $29 was assessed each configuration incorporating an air
pump. Additional costs for plumbing, etc. , were assumed to be accounted
for either in the thermal reactor or catalytic converter cost estimates.
8. 1. 1. 2. 5 Over temperature Protection System
As mentioned in Section 4, RTR and/or catalytic converter systems may
require overtemperature protection systems. Their need is not established
at the present time, as automakers are searching for catalysts with higher
temperature capabilities to avoid the system complexity introduced by the
addition of such a protection system, the exact details of which are
ill-defined at the present time.
Informal cost estimates for such systems range from $25 for the simpler
to approximately $100 for the more complex type. For purposes of the
present study, a cost penalty of $50 was assessed any generic concept
incorporating either a catalytic converter or an RTR.
8. 1. 1. 3 Discussion of Device Costs
The foregoing engine modification and emission control system component
cost penalties, as noted, are based on consideration of a variety of
8-6
-------
assumptions and evaluation techniques. Every effort has been made to
ensure that cost levels are "comparable," both as to variation in devices
within a given component class (e. g. , within thermal reactors) and from
component class to component class (e.g. , from thermal reactors to
catalytic converters).
In this manner, although the cost values used herein may not exactly
coincide with those eventually forthcoming from the automakers, the
relative cost levels between the various generic classes of overall emission
control system concepts made up of these components are meaningful on a
comparative basis.
Informal cost estimates seen to date generally support the component cost
assessment levels made herein.
8. 1. 2 Emission Control System Initial Hardware Costs
As previously noted in Section 4, meaningful complete emission control
system concepts fall into discrete generic classes. All of the various
experimental emission control system concepts evaluated by industry to
date can be reasonably identified within this generic system of classification.
As a primary purpose of the cost analysis effort was to provide a measure
of the cost differences between the various conceptual approaches, all
generic classes were assessed component or hardware cost penalties on a
common basis by use of the component costs described above
(Section 8. 1. 1).
Table 8-2 summarizes the initial hardware costs for the generic classes
considered in the cost analysis. Identified are the discrete components
(and their costs) of each generic class and a summation of the initial total
hardware cost to the consumer, as installed in a new car.
It should be noted that systems incorporating both a thermal reactor and
catalytic converter have been selected to use a "low-grade" thermal
8-7
-------
Table 8-2. Installed Hardware Cost Summary
(Cost to Consumer in New Car)
^\
^^"v^Concept
^^^^
Cost Item ^"^^^
Initial Costs (dollars)
Engine Modifications
Carburetion
Ignition/Diatrlbutor
L.L. Exhaust System
Valves, Seats
Emission Control
Components
ECR
Thermal Reactor
Catalytic Converter
(C. C )
Air-injection
Pump
Overtemp. Protection
System
Exhaust System Credit
Total Installed
Hardware Cost (dollars)
Thermal Reactor Systems
LTR t EGR
25
37
60
-
25
110
.
.
.
-28
229
RTR f EGR
13
37
60
-
25
125
.
29
50
.28
311
Combination Systems0
Low -Grade
LTR + EGR
+ HC/COC.C
25
37
60
3
25
70
98
29
50
-28
369
Low-Grade
RTR t EGR
t HC/COC.C.
13
37
60
3
25
70
98
29
50
-28
357
Low-Grade
RTR + EGR
+ Dual C. C.
13
37
60
3
25
70
129
29
50
-28
388
Use Leaded Gasoline
Catalytic Converter + ECR are the same except for $70 decrease due to omission of Low-grade
Thermal Reactor.
reactor, since the primary purpose of the reactor is to warm up the
catalytic converter. Therefore, it is felt that the full-size thermal
reactor is unwarranted in this case.
Catalytic-converter-only systems were not treated as a separate item
because of their poor cold start characteristics to date. However, with
regard to their costs, they would be identical to the RTR + EGR + HC/CO
catalytic converter and RTR + EGR + dual catalytic converter concepts of
Table 8-2, except for the deletion of the RTR cost ($70). Similarly, their
SFC- versus NO -level characteristics are considered the same as their
x
thermal reactor plus catalytic converter counterparts and, therefore, the
overall consumer costs hereinafter developed are equally applicable to
8-8
-------
them, except for the $70 cost differential. In addition, the tricomponent
catalytic converter concept has not been treated since it requires a
precision in air-fuel ratio control not yet demonstrated. As the required
air-fuel ratio control system has not been identified, it has not been possible
to provide a reasonable cost estimate for this function.
Similarly, the RTR plus NO Catalytic Converter plus RTR concept of
X,
American Motors was not costed as American Motors regards it as a
laboratory experiment only and not a viable contender for 1975-76 systems.
As shown in the table, initial installed hardware costs range from $229 to
$388. Thermal reactor systems are lowest in cost, and the dual-bed
catalytic converter system is the most expensive, as would be expected.
8. 2 OVERALL COSTS TO THE CONSUMER
To the foregoing initial hardware costs must be added the various mainte-
nance and operating costs, as pertinent to each generic emission control
system class, in order to determine the total or overall cost to the new
car buyer. The various cost determinations and assumptions in this regard
are delineated in the following sections.
8. 2. 1 Maintenance Costs
Several maintenance cost areas affected by the emission control system
concept and/or the use of leaded or unleaded gasoline were identified. These
are spark plug life, maintenance of the overtemperature protection system,
catalytic converter system replacement, and exhaust system replacement.
An average lifetime mileage of 85, 000 miles and an average automobile age
of 8. 4 years are used in the analysis. These values are based on the
statistics of the percentage of cars still registered as a function of the car
age (Ref. 8-11) and on the average miles per years as a function of the age
of the car (Ref. 8-12).
As treated in Section 7, it was determined that spark plugs operated with
unleaded gasoline have an average longer lifetime than when operated with
8-9
-------
leaded gasoline. Although the exact lifetime levels in each case are not
a priori determinable when installed in the various emission control
system concepts (particularly with improved ignitor systems), repre-
sentative values of 13, 000 miles for leaded gasoline and 20,000 miles for
unleaded gasoline were selected to illustrate typical maintenance cost
differences for spark plug changes in the two cases.
Each spark plug change was estimated to cost $10 (representing a combi-
nation of installations at a garage and home replacement). For the 85, 000-
mile baseline car life used herein, this resulted in a cost savings of $23
for cars using unleaded gasoline.
The next area of maintenance cost considered is in regard to the over-
temperature protection system. As noted previously, if needed and incor-
porated, this system is a critical part of the overall emission control
system. A nominal cost value of $5 per year has been assessed for
inspection only of this system. No cost for actual repair has been included.
Again, based on the average 8. 4-year lifetime used herein for the baseline
car, this results in a total inspection cost of $42 for the concepts incorpo-
rating overtemperature protection systems.
The goal of the automakers is to develop a catalytic converter unit capable
of meeting a 50, 000-mile maintenance and/or replacement-free require-
ment. Demonstration of this capability at 1975-76 emission levels has
not been made to date. However, for purposes of the present cost
assessment, it is assumed that this requirement will be met for HC/CO
catalysts. Both 25, 000- and 50, 000-mile replacement intervals are con-
sidered for illustrative purposes for NO catalytic devices since to date
•X
the demonstrated life is quite low.
The exact method of catalytic converter replacement or refurbishment is
a matter of current debate. Potential users of converters incorporating
pelletized catalyst beds project the eventual possibility of being able to
8-10
-------
withdraw used catalyst pellets from the converter (by vacuum means, etc.)
and insert new or fresh pellets. Potential users of monolithic catalyst
beds similarly envision converter designs which enable simple cartridge-
type replacement of the monolithic bed. If these replacement/refurbishment
techniques become a reality, they might reduce converter replacement
costs.
In both cases, these are mere projections at the moment, with demon-
strated capability lacking. Therefore, for purposes of the present cost
analysis it has been assumed that the total catalytic converter unit will be
replaced. Discrete costs for replacement were estimated based on
original manufactured cost, wholesaler's costs, dealer installation cost,
and nominal profits. These values are:
Converter Type
HC/CO Catalytic Converter
Dual-Bed Catalytic Converter
Replacement Cost, Dollars
123/car
156/car
As mentioned in Section 8. 1. 1. 1. 3 above, a long-life exhaust system was
a basic part of each generic class of emission control system considered.
It was assumed that this system would last the life of the car. As the con-
sumer has been penalized for this added cost, the exhaust system replace-
ment costs normally anticipated in the baseline car case must be subtracted
from the overall costs to the consumer to maintain the relative cost effect
of the car equipped with the new emission control system.
As discussed in Section 7, a representative average value of $45 per
exhaust system replacement and an average life of approximately 37, 500
miles are assumed. This results in a total cost credit of $60 for each
emission control system concept for an average car lifetime of 85, 000
miles.
8-11
-------
8. 2. 2 Operating Costs
Operating costs, as utilized herein, are restricted to fuel economy cost
penaltie s.
8. 2. 2. 1 Fuel Economy Cost Penalty
Fuel economy cost penalty is defined as the cost of fuel for the car over
its lifetime when equipped with the emission control system being evalu-
ated minus the car lifetime fuel cost of an average 1970 car. This defi-
nition gives the increase in fuel costs due to the emission control system.
When added to the emission control system initial hardware and mainte-
nance costs, the total cost of the emission control system is obtained.
8.2.2.1.1 1970 Car
The characteristics of the 1970 car are:
Weighted average compression ratio: 9. 37:1 (Refs. 8-13 and 8-14)
Miles/gallon: 13. 5 (Ref . 8-14)
Lifetime mileage: 85, 000 miles (as noted in Section 8. 2. 1.)
Weighted average price of fuel (leaded) for 1970 car: 37.40
(based on regular at 35. 69 £/gal (Ref. 8-14), premium at
39. 69 £ /gal, price spread from Ref. 8-15, and regular
gasoline sales 57.4 percent of total gasoline sales;
from Ref. 8- 14 )
Lifetime fuel cost for the 1970 car from the above: $2355
To account for the fact that some 1970 cars are designed to use regular
gasoline and some premium, a weighted average compression ratio has
been used for the baseline 1970 car. Similarly, since some drivers of
cars designed to use regular buy premium, an average price of gasoline
based on the percentage of sales of each grade has been selected as the
baseline gasoline cost per gallon.
8-12
-------
8.2.2.1.2 Fuel Economy Effects of Emission Control Devices — General
As discussed in Section 4, the fuel economy of a car equipped with an
emission control system depends primarily on the method of reducing
NO emissions, the NO emission level achieved (see Fig. 4-34), and
xx
whether or not the emission control system can tolerate lead in the engine
exhaust (catalytic systems are lead intolerant). HC/CO emission reduction
devices have only secondary effects on fuel economy, excluding unleaded
gasoline effects, since engine air-fuel ratio is set by the NO emission
reduction device for low NO emission levels. These secondary effects
are due to increased engine exhaust backpressure, and to power for a
secondary air pump (in some cases), and are much the same for most
systems. For the reasons discussed in Section 6, lead scrubbers in the
engine exhaust will not be considered and all systems incorporating cata-
lytic devices are assumed to use unleaded gasoline. Fuel economy effects
considered, therefore, consist of those due to NO emission reduction
j£.
devices and the use of unleaded gasoline for systems containing catalytic
devices. These are considered separately and then combined to obtain
an overall fuel cost penalty for several basic classes of systems which
encompass the major types of systems proposed for 1976 cars.
8.2.2.1.3 NO Emission Reduction System Effects on Fuel Economy
As discussed in Section 4, NO emission reduction systems fall into two
3£
broad categories: (1) those depending only on the use of a combination of
engine air-fuel ratio change and EGR, and (2) those utilizing NO catalysts
3k
in conjunction with a smaller change in engine air-fuel ratio and a lower
EGR rate than in the first category.
Typical engine fuel economy changes with the addition of emission control
systems are shown in Fig. 8-1 which was developed from information in
Section 4. For the two categories shown, the fuel penalties differ. All
engines are assumed to have the same compression ratio so there are no
unleaded gasoline effects reflected in the values. The curve for EGR plus
air-fuel ratio change applies to thermal reactor systems and those
8-13
-------
E
o»
CO
i
UJ
CO
CO
X
o
•EGR + A/F ONLY
CATALYST
0 5 10 15 20 25 30
PERCENT FUEL ECONOMY PENALTY
Figure 8-1. Fuel Economy Penalty due to NO
Emission Reduction x
incorporating HC/CO catalysts, without an NO catalyst in the system.
The NOx catalyst curve applies regardless of the method of HC/CO emis-
sion reduction used.
It is evident from the figure that, excluding the cost effects of their use
of unleaded fuel, NO catalyst systems are attractive from a fuel economy
standpoint. It is also evident from the figure that decreasing the required
NOx emission level for any type of system reduces fuel economy. This is
due to the necessity of increasing the EGR rate and/or decreasing the
air-fuel ratio as the NO emission level is reduced.
8-14
-------
8. Z. 2. 1.4 Fuel Economy Cost Effects of Using Unleaded Gasoline
Engine efficiency improves as compression ratio increases. In general,
as the compression ratio increases, however, the gasoline octane number
must also increase. The addition of lead to the fuel has proven to be the
cheapest way of providing high octane number fuels. Engines with emission
control systems which cannot tolerate lead in the fuel (e. g., catalytic
systems) must, therefore, either operate at lower compression ratios
with poorer efficiency than engines equipped with emission control systems
which are lead tolerant, or use more expensive fuel than is required for
the engines equipped with lead-tolerant emission control systems. A com-
promise between compression ratio reduction and higher cost fuel for the
engines equipped with lead-intolerant emission control systems is of
course another alternative. As will be shown later, the latter is the mini-
mum cost approach for engines equipped with lead-intolerant emission
control devices.
The determination of a cost optimum compression ratio-fuel octane
number combination is, of course, dependent upon the assumptions made
in the analysis, particularly those for fuel price and the percent of
"knock-free" satisfaction. As a result, it can be expected that different
investigators would arrive at different optimums, depending upon their
particular selection of the input variables. Reference 8-16 has determined
such a cost optimum compression ratio-fuel octane number combination.
Unfortunately, the analysis was based on providing a gasoline with which
98 percent of all cars would be knock-free when tested by the Coordinating
Research Council method (based on technical or "trace-knock" satisfaction).
This is a much higher percentage of knock-free cars than has been typical
of pre-1971 cars (Ref. 8-17 indicates 80-percent knock-free, based on
technical satisfaction, is more typical of cars with leaded fuel; this is
equivalent to approximately 95-percent customer satisfaction). The analy-
sis of Ref. 8-16 has therefore been recalculated with the only changes
being that 80 percent of all cars are considered knock-free (based on
8-15
-------
technical satisfaction) and a slightly different variation of gasoline cost
with octane number is used. The increase in average price of unleaded
(clear) gasoline used, versus octane number, exclusive of distribution
costs, is shown in Fig. 8-2.
cc
-------
costs. It should be noted that there are three data points (from Ref. 8-18) in
the region of most interest that have zero lead in the total pool:
Total Added
Clear -Pool Refining Cost,
Schedule Year RON
L 1980 93.5 0.34
L 1976 94.4 0.60
M 1980 94.2 0.43
Also, Schedule N, having very low lead content in the total pool for year
1980, indicates an added refining cost of 0. 10 £/gal at a clear-pool RON
of 92.6.
To the foregoing added refining cost of unleaded gasoline (Fig. 8-2) was
added 0. 26 cent per gallon to reflect the cost of a third pump and associ-
ated storage tanks for the distribution of unleaded gasoline (Ref. 8-18).
It should be noted that this value does not incorporate any costs associated
with segregated pipeline and distribution systems to ensure against con-
tamination by leaded products, i.e., only "normal" precautions were con-
templated. If the trace-lead-level content is required to be substantially
lower than that obtainable by normal precautions, such pipeline and dis-
tribution costs would increase, perhaps substantially. The exact trace
lead level at which this change in requirements exists is presently
unknown to any degree of certainty.
The computed optimum RON for unleaded fuel (based on the foregoing
assumptions) is -between 94 and 95 (the solution is characteristically
rather shallow in slope in the optimum region). A 93 RON gasoline, how-
ever, was selected for the fuel cost penalty assessment to be consistent
with concurrent studies (EPA-funded) to determine unleaded gasoline
investment and manufacturing costs (Ref. 8-18). The choice of 93 RON
gasoline, rather than the optimum RON gasoline determined herein,
8-17
-------
increases the fuel cost penalty only slightly, and reduces the refinery
capital investment required.
The 93 RON chosen is, of course, different from the 91 RON with which
the U.S. automobile manufacturers have indicated their cars will be capable
of operating, at least in the immediate future. However, several automo-
bile manufacturers in informal discussions have indicated that they do not
consider 91 RON optimum and may very well increase their RON require-
ment with time. Their choice of 91 RON was apparently heavily influenced
by their desire to specify a fuel which could be more easily made available
during the sudden transition to the use of unleaded gasoline. The long-
term, case is believed more appropriate for the purposes of this analysis
and, hence, 93 RON has been chosen.
Both single- and three-grade (three different ON unleaded fuels sold)
unleaded gasoline cases have been analyzed since it is anticipated that, at
the initial introduction of unleaded gasoline, a single grade will be offered
because of need to retain service station pumps to sell leaded gasolines
for older cars. However, as the older cars disappear from the road the
leaded fuel pumps could gradually be converted to dispense additional
grades of unleaded gasoline.
A multigrade unleaded gasoline system is advantageous to the consumer.
The reason is that different engines of the same design do not all have the
same octane number requirements because of manufacturing tolerances
and variations in operating conditions. With a single-grade system, the
fuel octane number must be selected to satisfy the cars with the highest
requirement, and all cars must use this expensive fuel. With a multigrade
system, those cars capable of using a cheaper fuel of lower octane number
may do so, lowering the average price of gasoline; or conversely, if the
average price of gasoline is held constant, the compression ratio and
fuel economy may be increased as the number of grades is increased.
8-18
-------
The results of the analysis are shown in Table 8-3. The fuel cost shown
is that for 93 RON from Fig. 8-2 (0. 20 cent per gallon) plus 0. 26-cent-per-
gallon distribution cost, as noted above. Again, there have been varying
Table 8-3. Cost Effects of Use of Unleaded Gasoline
Item
Calculated Optimum RON
Pool RON Used in Analysis
Compression Ratio (93 RON)
Percent Change in Fuel Economy due to Compression Ratio
Change from 1970 Car3
Fuel Price-- Af/ gal over Average Fuel Price for 1970 Car
Fuel Cost Penalty due to Price/Gallon of Unleaded Fuel, 85,000
MUes.c No SFC Loss
Unleaded Gasoline Fuel Cost Penalty over 85,000 Miles0
(Compression Ratio plus A£/gal Effects)
1 -Grade
94+
93
8.35:1
-5.4
0.46b
$30.00
$160.00
3 -Grade
94+
93
8.95:1
-Z
0.46b
$30.00
$80.00
aConstant car performance (acceleration, power)
Equivalent to 2. 16 £/gal above leaded regular grade gasoline
Approximate, varies with NOX emission level, these values assume no NOX SFC related cost
estimates from different sources as to the eventual increased cost of
unleaded gasoline. Although there may be differences of opinion regarding
manufacturing cost increase, distribution cost effects, etc., the overall
(circa 1980) cost increase of 0. 46 cent per gallon used in this study for
calculation purposes should be sufficiently representative to illustrate
unleaded gasoline cost effects.
Traditionally, refiners, distributors, and service stations have made
larger profits from the manufacture and sale of premium grade leaded
gasoline than from regular grade leaded gasoline. Premium grade has
accounted for approximately 42. 6 percent of gasoline sales, but with the
8-19
-------
introduction of unleaded gasoline, the sales of premium grade will
decrease. When only unleaded gasoline is available, to make the same
average profit per gallon the cost of unleaded gasoline should be the
present weighted-aver age price of leaded gasoline plus the increased cost
of manufacturing unleaded gasoline (over that of the weighted-average
leaded gasoline) plus any associated increase in distribution costs for
unleaded gasoline. This is the 0. 46-cent-per-gallon figure derived above.
However, since the price of leaded regular grade gasoline is less than the
weighted-average sales price for leaded gasoline, this 0.46 cent per
gallon is equivalent to a 2. 16 cents per gallon increase above the price
of leaded regular grade gasoline.
It was assumed that the three-grade system would have the same clear-
pool RON (93) as the single-grade system and that the overall manufactur-
ing plus distribution cost effects (0.46 cent per gallon over conventional
leaded gasoline weighted-average price) would be the same. Although the
incremental manufacturing costs for the three-grade system might be slightly
different (even at the same 93 RON pool) from the 0.20 cent per gallon of the
single-grade case, it is not felt that this difference would be large enough to
significantly alter the results (e.g. , 0. 1-cent-per-gallon gasoline cost increase
is equivalent to less than $8 over 85,000 miles of operation.)
As can be seen from Table 8-3, the fuel cost penalty associated with the
higher cost per gallon of unleaded fuel is not large (approximately $30).
The major fuel cost penalty with the use of unleaded gasoline (over leaded
gasoline) is due to the lowering of engine efficiency associated with reduced
compression ratio required by the lower octane number of the unleaded
gasoline. For the single-grade system, there is an additional $130 fuel
cost penalty due to the necessity of reducing compression ratio to 8. 35:1
(from 9. 37:1 for the leaded fuel case; 80-percent knock-free technical
satisfaction) with its attendant loss in fuel economy of 5.4 percent. In this
case, then, the total fuel cost penalty attributable to the use of unleaded
gasoline is $160.
8-20
-------
The most significant effect of the three-grade 93 overall RON system is
the ability to increase compression ratio (to 8. 95:1 at 80 percent knock-
free technical satisfaction). This compression ratio-octane number
satisfaction relationship was determined (as was the single-grade case)
in a manner similar to that developed in Ref. 8-16 (constant car perfor-
mance, acceleration, power).
As shown in Table 8-3, the increase in compression ratio made possible
by the three-grade system reduces the fuel economy penalty loss to $50
(for a 2-percent fuel economy loss compared to the leaded fuel reference
case) and gives a total fuel economy penalty (compression ratio effect
plus increased price-per-gallon effect) of $80. This represents a savings
of $80 over the single-grade 93 RON case shown in the table. Although
not shown, a two-grade unleaded system (at the same overall 93 RON)
would be expected to provide similar (but not identical) cost savings over
the single-grade system.
8. 2. 2. 1. 5 Fuel Economy Cost Penalty Results
As discussed in Section 4, the change in engine fuel economy (excluding
unleaded gasoline) due to the addition of an emission control system, is a
function of the NO emission level and whether or not an NO catalyst is
X X
used. Figure 8-1 gives these relationships. They have been combined
with the previously described compression ratio and higher costs of
unleaded gasoline effects to obtain the total fuel cost penalty for emission
control systems as a function of NO emission level shown in Fig. 8-3.
The following observations may be made from the figure:
1. The fuel cost penalty is sizable at the lower NOX emission
levels, regardless of the type of emission control system,
because of the necessity of using low air-fuel ratios and/or
high EGR rates.
2. Emission control systems incorporating NOX catalysts have
more potential for lower fuel costs than other systems,
particularly at low NOX emission levels, because of their
ability to use higher air-fuel ratios and less EGR. It should
8-21
-------
1000
800
-CO-
£]~600
z
UJ
0.
I—
8 400
d
=>
Lt_
200
CATALYST
"SYSTEMS WITHOUT
NOX CATALYST
SYSTEMS
INCORPORATING
NOX CATALYST
GRADE
GRADE UNLEADED GASOLINE
-3 GRADE UNLEADED GASOLINE
THERMAL REACTOR PLUS EGR
SYSTEMS USING LEADED GASOLINE
I
0.4 0.8 1.2 1.6 2.0
NOX EMISSION LEVELS, gm/mi
Fig. 8-3. Fuel Cost Penalty
1400
1300
1200
1100
1000
900
* 800
£
8 700
_i
I 600
500
400
300
200
100
0
I
I
]
I-RTR *EGR » DUAL C.c*
(25,000 mi LIFE)
-1 GRADE UNLEADED GASOLINE
- 3 GRADE UNLEADED GASOLINE —
EGR * HC - CO c.c* -
'///, PRESENT LOWER
LIMIT ON NO,
EMISSION
2 GRADE LEADED.
GASOLINE
•CATALYTIC CONVERTER* EGR SYSTEMS SAME IN COST
EXCEPT FOR $TO DECREASE DUE TO OMISSION OF
LOW-GRADE THERMAL REACTOR
I
I
I
Fig. 8-4. Increased Consumer
Costs Over Lifetime
of Car
0.4 0.8 1.2 1.6
NOX EMISSION LEVELS, gm/mi
20
24
8-22
-------
be noted, however, that no one has yet demonstrated
reasonable life with a high-performance NOX catalyst.
3. If a durable NOX catalyst does not become available,
emission control systems which do not incorporate HC/CO
catalysts are superior from a fuel cost standpoint because
of their ability to use leaded gasoline and a higher com-
pression ratio engine.
4. For emission control systems which incorporate catalytic
converters and must use unleaded gasoline, multigrade
gasoline systems offer significant fuel cost advantages
over single-grade systems.
8. 2. 3 Excluded Costs
No costs have been included for:
1. Research and development activities
2. Compliance emission testing after car purchase
3. Production emission testing
The required tests are undefined at present, and research and development
costs are not yet all accrued and the rate of their amortization unknown.
8.2.4 Cost Analyses Results
The overall sum of the emission control system initial hardware cost,
maintenance cost, and operating cost is shown in Fig. 8-4 for the selected
generic systems as a function of the NOx emission level. The cost is
displayed as a function of the NOX emission level, since the various systems
are not all capable of the same NOX emission reduction and the operating
costs (fuel costs) are highly dependent on the NOX emission reduction.
Also shown on the chart are the presently demonstrated lower limits on
NO emissions for the various systems as discussed in Section 4.
A.
A breakout of the fuel costs from the total cost at selected NOY emission
H
levels is shown in Fig. 8-5. The emission levels selected correspond to
those values which represent the presently demonstrated lower limit
NO emission level for one or more systems.
8-23
-------
Q FUEL COST PENALTY 3
1 C-5
m HARDWARE AND ^ ^
MAINTENANCE COSTS + 8
1200
MOO
1000
900
800
700
•co-
fe" 600
o
o
500
400
300
200
100
n
—
—
—
—
—
—
ce i
UJ I
-»- +
LlJ tt! ^
T ° 5 •— «-> Q 5c
^^ ^^ ^" (J /^« ^^
ceo ^'uj.jj CDQ
LLJ + cc
ro
CC r^ i O
S + 5[ ce uj
M^N Q^ — J J^— T
10 o Q ^ ce
ii i + 1. 1 i
— ,jj , —
+ cc o ce
cc uj ce uj
i~ ^. o Q
fy* ^j
UJ ^ ~ S
UJ f—~ C^j
Q CC
-------
The following cost observations and conclusions are based on the engineering
cost estimates made herein:
1. All systems have high overall costs. The cost increases
rapidly as NOX emission level decreases. If the cost of an
emission control system over the car lifetime were $1000,
the national annual cost with all cars on the road using
this system would be of the order of $10 billion.
2. Systems incorporating an NOX catalyst are the only systems
with the potential to meet the Federal standard of 0. 4 gm/mi
by 1976. In addition, at low emission levels they are the
minimum cost systems provided that a 50,000-mile catalyst
life can be achieved. The total cost of ownership in this
case is approximately $860. However, severe problems in
developing a durable NOX catalyst have been encountered,
and such a catalyst may not be available by 1976. If only a
25, 000-mile lifetime NOX catalyst is available, the total
cost of ownership is increased by approximately $300.
3. The increased cost of unleaded gas is 0. 46 cent per gallon
compared to the average leaded gas cost. This amounts to
about a $30 increase over the life of the car. If the lowering
of the average compression ratio -- and the associated fuel
economy penalty --is attributed to lead removal, another
$130 can be added to the cost of unleaded gasoline.
4. If the durability problems of the NOX catalyst system cannot
be resolved and its use is precluded, the RTR system, which
can tolerate leaded gasoline, would be lowest in cost. There-
fore, there is no cost advantage in unleaded gasoline unless
a durable NOX catalyst can be developed.
5. The lead-tolerant LTR system is attractive from a cost
standpoint, but its NOX emission levels are high.
6. In general, the lead-intolerant systems (catalyst plus
thermal reactor systems) have higher initial hardware costs
than those systems tolerating lead in the gasoline. This is
because the catalyst plus thermal reactor systems include
most of the parts of the straight thermal reactor systems as
well as the catalyst. The thermal reactor in these systems
is needed to provide fast warmup of the catalyst bed. If a
fast warmup catalyst (of equivalent cost) is developed, the
initial cost for the catalyst system is reduced by only $70,
because of the omission of the low-grade thermal reactor.
8-25
-------
7. Lead-intolerant systems (catalyst systems) have higher
maintenance costs than those systems tolerating lead because
the cost of replacement of the catalyst bed is greater than
the cost savings on spark plugs with unleaded gasoline.
Muffler cost savings discussed earlier were precluded by
the use of a long-life exhaust system on all 1976 systems.
8. A system incorporating a NOX catalyst has the lowest fuel
cost because it has the lowest SFC which more than offsets
the higher cost effects of unleaded gasoline.
9. A lead-intolerant HC/CO catalyst system without a NOX
catalyst has higher fuel costs than a lead-tolerant thermal
reactor system. This is due primarily to the cost effects
of unleaded gasoline and compression ratio, as both systems
have about the same SFC for a given NOX level.
Of course, as more definitive data become available on emission control
system technology (e. g. , emission levels, durability, costs), the fore-
going conclusions may be subject to appropriate modification.
8-26
-------
REFERENCES
8-1. Personal discussion with representatives of the Chrysler Corporation
(29 June 1971).
8-2. Effect of Lead Antiknocks on the Performance and Costs of Advanced
Emission Control Systems, Du Pont de Nemours & Co. ,
Wilmington, Delaware (15 July 1971).
8-3. Personal discussion with a representative of the Esso Research and
Engineering Company (6 July 1971).
8-4. R.R. Allen and C.G. Gerhold; "Catalytic Converters for New and
Current (Used) Vehicles," Paper presented at the Fifth Technical
Meeting, West Coast Section of the Air Polution Control Association,
San Francisco, California (8-9 October 1970).
8-5. "Chapter 3: Mobile Sources," The Economics of Clean Air, Report
of the Administrator of the Environmental Protection Agency to the
Congress of the United States, Senate Document No. 92-6 (March 1971).
8-6. Personal discussion with a representative of the Research Triangle
Institute (unpublished data)(6 August 1971).
8-7. Personal discussion with representatives of Engelhard Industries, Inc. ,
(July 1971).
8-8. L.S. Bernstein, A.K.S. Raman and E.E. Wigg, "The Control of
Automotive Emissions with Dual Bed Catalyst Systems," Presented to
the 1971 Central States Section of the Combustion Institute
(23 March 1971).
8-9. R.S. Yolles, H. Wise and L.P. Berriman, "Study of Catalytic Control
of Exhaust Emissions for Otto Cycle Engines," Stanford Research
Institute, Final Report, SRI Project PSU-8028 (April 1970).
8-10. The Ethyl Lean Reactor System, Ethyl Corporation Research
Laboratories, Detroit (1 July 1971).
8-11. Automotive News--1971 Almanac, 35th Review and Reference Edition,
Slocum Publishing Co. , Detroit (26 April 1971).
8-12. Relationships of Passenger Car Age and Other Factors to Miles
Driven, U.S. Department of Transportation, Bureau of Public Roads,
Washington, D.C.
8-27
-------
REFERENCES (cont.)
8-13. Consequences of Removing Lead Antiknocks from Gasoline, A Status
Report, No. AC-10, Ethyl Corporation, Detroit (August 1970).
8-14. "Passenger Cars: Oil, Automotive Trends," 1971 National Petroleum
News Factbook Issue.
8-15. Personal discussion with representatives of Du Pont de Nemours &
Co., regarding their winter and summer gasoline surveys
(October 1971).
8-16. E.S. Corner and A.R . Cunningham, "Value of High Octane Number
Unleaded Gasolines in the United States," Presented before the
Division of Water, Air, and Waste Chemistry, American Chemical
Society, Los Angeles, California (28 March-2 April 1971).
8-17. W.E. Morris, et al. , "1971 Cars and the 'New' Gasolines, " SAE
Paper No. 710624(June 1971).
8-18. Economic Analysis of Proposed Schedules for Removal of Lead
Additives from Gasoline, Bonner & Moore Associates, Inc.,
Houston, Texas (25 June 1971).
8-28
-------
APPENDICES
-------
A. VISITS AND CONTACTS
-------
APPENDIX A
VISITS AND CONTACTS
A.I ORGANIZATIONS VISITED
American Motors Corporation
American Oil Company
Atomics International
Bayerische Motoren Werke (BMW), Germany
Bonner & Moore Associates
California Air Resources Board
Chrysler Corporation
Degussa, Germany
E.I. Du Pont de Nemours 8t Company
Engelhard Minerals & Chemicals Corporation
Esso Research & Engineering Company
Ethyl Corporation
Ford Motor Company
General Motors Corporation
Kali-Che mie, Germany
Mobil Oil Company, Research Laboratories
NASA/Lewis Research Center
NSU-Audi, Germany
Research Triangle Institute
Universal Oil Products
Volkswagen (VW), Germany
A.2 ORGANIZATIONS CONTACTED
(Response by Phone or Letter)
Cooper Union
Daihatsu Kogyo Company, Japan
Dow Chemical Company
W.R. Grace Company
Houston Chemical Company
Illinois Institute of Technology, Research Institute (IITRI)
Monsanto Company
Nissan Motor Company, Japan
Toyo Kogyo, Japan
University of California at Los Angeles
A-l
-------
B. POSSIBLE CATALYST
POISONING MECHANISMS
-------
APPENDIX B
POSSIBLE CATALYST POISONING MECHANISMS
Degradation of the performance of catalytic converters employed as pollution
control devices on automobiles run on leaded and unleaded gasoline is
observed to occur much more rapidly with leaded gasoline. Degradation may
occur either by loss of catalytic activity, or physical attrition, or both. The
lead component of gasoline thus clearly constitutes a catalyst "poison" which
acts through a variety of chemical and mechanical toxicity mechanisms that
are not mutually exclusive .
B.I SUMMARY OF CATALYST POISONING MECHANISMS
Among the possible chemical poisoning mechanisms are:
1. Chemisorption of toxic species at active surface sites, thereby
rendering the catalyst inactive
2. Chemical conversion of the catalyst to an inactive, nonvolatile
compound
3. Chemical conversion of the active component of the catalyst to
a volatile compound, thereby reducing the quantity of catalyst
4. Reaction of toxic species with the catalyst support, resulting in
a decrease in the structural stability of the catalyst and sub-
sequent decline in surface area or mechanical attrition
Among the possible mechanical poisoning mechanisms are:
1. Deposition of a coating on the catalyst surface, rendering the
latter inaccessible to reacting species
2. Deposition of poison at the mouths of catalyst pores, thereby
also reducing the availability of much of the catalyst surface
B.2 CHEMICAL AND MECHANICAL POISONING MECHANISMS
B.2.1 Chemical Poisoning Mechanisms
Substances which are capable of forming much stronger bonds with the
catalyst surface than the bonds normally formed by reactants and products
B-l
-------
are termed chemisorptive poisons. A poison, because of its high bonding
energy, resides on a catalyst surface for times very much longer than the
necessarily brief residence times characteristic of reactants. Criteria for
predicting the toxicity of various compounds toward metallic catalysts have
been developed by Maxted (Ref. B-l). Although much of Maxted's work was
done with catalysts suspended in liquids, the conclusions may be applicable
to gaseous media as well, and should also be applicable to metal oxide cata-
lysts. Two of Maxted's categories of catalyst poisons are relevant to the
leaded gasoline problem:
1. Poisons containing nonmetallic elements of groups Va and Via of
the Periodic Table, notably sulfur and phosphorus
2. Poisons containing toxic metal ions
Nonmetals, such as sulfur or phosphorus, tend to be poisons in compounds
because of their lower oxidation states, whereas lead is a chemisorptive poison
for metallic catalysts in essentially any oxidation state. Because the mode
of bonding metal ions to a metal surface differs from that of nonmetals, the
toxicity criteria for metal ions are different. Maxted notes that a toxic metal
ion must have a particular electronic structure (at least five d electrons) and
lead meets this criterion. G.C. Bond (Ref. B-2) suggests a theoretical
explanation of Maxted's observations.
The toxic species present in automobile exhausts could react chemically with
the catalyst to form a layer consisting of a new compound or compounds that
do not display catalytic activity. Possible reactions are presented in Ref. B-3.
Roth (Ref. B-4) has reported that a copper catalyst volatilized when exposed
to a stream of chlorobenzene at high temperatures. This was interpreted as
occurring by conversion of the copper to copper chloride, which was sufficiently
volatile to leave the remaining catalyst material. Since the vapor pressure of
most transition metal halides is exceptionally small at automotive catalyst
operating temperatures, it is difficult to understand how the halides of these
metals could volatilize into streams of higher pressure. Nevertheless, the
complexity of exhaust gases from the combustion of leaded gasoline is such
B-2
-------
that volatilization of a portion of the catalyst is possible, regardless of the
nature of the mechanism.
The catalyst poison may chemically attack not only the active catalyst mate-
rial, but the support as well, resulting in structural degradation of the cata-
lyst and in subsequent loss of catalyst by attrition. Lead has also been
observed to attack preferentially along the grain boundaries of a Monel NO
X
catalyst, thereby causing agglomeration (Ref. B-5).
B.2.2 Mechanical Poisoning Mechanisms
The deposition of a catalytically nonreactive layer of material on the catalyst
surface is another possible poisoning mechanism for automotive pollution
control catalysts. This coating need not be chemically bonded to the catalyst
surface•
X-ray diffraction analysis confirms the presence of lead sulfate, lead oxysul-
fate, and lead oxyhalides on the surface of spent catalysts (Refs. B-4, B-6,
and B-7). Hofer, et al., also report the presence of lead halides (Ref. B-6).
All these authors agree that lead sulfate is the principal lead compound.
Yarrington and Bambrick found pyromorphite IsPb-CVOJ, • PbCl9] (Ref. B-7).
L j *t £• £»*
The presence of the compound 3Pb,(PO.)_ • PbCl? has been reported
(Ref. B-8).
Deposition of lead compounds may occur preferentially at catalyst pore mouths,
which will destroy catalytic activity by making the internal surface of the cata-
lyst unavailable. The rate at which catalyst activity declines is an indication
of whether poisoning is by blockage of pore mouths or by uniform deposition
over the entire catalyst surface (Ref. B-9). Yarrington and Bambrick con-
clude that poisoning occurred by uniform deposition over the active surface
for the catalysts which they tested (Ref. B-7).
The complex composition and huge variety of exhaust gas constituents and
the wide range and number of engine operating parameters, together with the
many types and configurations of catalytic materials, make it very difficult
B-3
-------
to arrive at generalizations regarding the most likely mechanisms.
Nevertheless, a review of these mechanisms has indicated that lead would
have a deleterious effect on catalysts. Experimental data with prototype cata-
lysts, run with actual automotive exhausts under realistic operating conditions,
are therefore most meaningful in assessing the effects of lead.
B-4
-------
REFERENCES
B-l. E.G. Maxted, in Advances in Catalysis III, W.C. Frankenberg,
V.I. Komarewsky, andE.K. Rideal, editors, Academic Press,
New York (1956).
B-2. G.C. Bond, Catalysis by Metals, Academic Press, New York (1962).
B-3. W.H. Page II, U.S. Patent 3,072,458, Universal Oil Products
Company (8 January 1963).
B-4. J.F. Roth, Paper presented at the 161st Meeting of the American
Chemical Society, Los Angeles, California (April 1971).
B-5. Esso Research and Engineering Company (private communication).
B-6. L.J.E. Hofer, J.F. Schultz and J.J. Feenan, Bureau of Mines Report
No. RI 6243 (1963).
B-7. R.M. Yarrington and W.E. Bambrick,(to be published).
B-8. D. Bienstock, et al., Bureau of Mines Report No. RI 6323 (1963).
B-9. A. Wheeler, in Advances in Catalysis III; W.C. Frankenberg, V.I.
Komarewsky, and E.K. Rideal, editors, Academic Press, New York
(1951).
B-5
-------
C. LEAD TRAP DEVICES
FOR AUTOMOTIVE
VEHICLES
-------
APPENDIX C
LEAD TRAP DEVICES FOR AUTOMOTIVE VEHICLES
OPERATING ON LEADED GAS
C.I INTRODUCTION
Devices that would allow the use of leaded gasoline with catalytic converters
are under development. These devices would have to remove lead from the
engine exhaust upstream of the catalyst. Techniques for collecting lead from
exhaust emissions of automobiles operating on leaded fuel fall into two basic
categories. The first technique removes lead in the gaseous, as well as par-
ticulate, form at exhaust manifold temperatures so that the exhaust is held
sufficiently hot for further catalytic treatment to control NO , HC and CO
n
emissions. The second technique requires that exhaust gases be cool enough
for a sufficient amount of lead to be in participate form for collection. Thus,
in order to utilize the second technique prior to catalytic treatment, additional
equipment will be required to first condense the lead into particulate form for
collection upon exiting the exhaust manifold and then to reheat the exhaust gas
for efficient catalytic reaction. In addition to the extra complexity and cost
for such a device, the limited available space in present automobile designs
would appear to make this technique impracticable for use with catalytic
reactors. However, such systems are considered appropriate for removal of
lead from exhaust emissions subsequent to catalytic treatment, or in a system
where catalytic converters either are not present or are insensitive to lead
compounds. The various lead removal techniques will be identified and dis-
cussed under the two classifications described above, namely, (1) combined
lead compound vapors and particles, and (2) lead compound particles only.
A literature search and telephone contact with knowledgeable industry repre-
sentatives for past and current activities directed toward devices for collecting
lead from automobile exhaust emissions revealed that the extent of such work
seems to have been primarily limited to Atomics International, Cooper Union,
C-l
-------
Dow Chemical, Du Pont, Ethyl, Houston Chemical Company, and Illinois
Institute of Technology (IIT) Research Institute. These seven organizations
have been responsible for a total of eight devices in various stages of devel-
opment. With the exception of Atomics International, all techniques required
that lead be in a particulate form for collection and removal. The Atomics
International approach has been the subject of considerable development that
has included laboratory, as well as automobile road and engine dynamometer
testing, whereas four of the seven particulate-only collection techniques pro-
posed by Cooper Union, Dow Chemical, and IIT Research Institute have not
progressed beyond exploratory testing in the laboratory. The cyclone-type
traps proposed by Du Pont and Ethyl for collecting lead and other participates
have apparently undergone a fair amount of design and testing effort. Addi-
tional details of each system are summarized in the following paragraphs.
C.2 TECHNIQUE FOR REMOVING COMBINATIONS OF LEAD
COMPOUND VAPORS AND PARTICLES FROM EXHAUST
EMISSIONS
As indicated in the foregoing discussion, Atomics International has developed
the only approach for removing lead in the vapor as well as particulate form.
A description of the early development work is presented in Refs. C-l and
C-2. For purposes of the present discussion, pertinent information relative
to the concept, current design approach, efficiency, costs, and compatibility
with catalytic reactors is summarized.
C.2.1 Atomics International Molten Carbonate Process
This approach is based on the fact that lead halides and sulfur dioxide, which
are mildly acidic, can be expected to react chemically with a basic alkali
carbonate. Thus, a molten carbonate salt was evolved to scrub gaseous.
liquid, or solid lead compounds from automobile exhausts. The carbonate,
the choice of which resulted primarily from melting point and cost consider-
ations, consists of roughly equal parts by weight of lithium, sodium, and
potassium carbonate, and melts at a temperature of 750°F. The lead and
C-2
-------
sulfur compounds react with the carbonate and are removed from the exhaust
stream within the scrubber.
Laboratory tests and road tests of a breadboard device installed in an auto-
mobile suggested that a muffler replacement device could potentially remove a
large portion of the lead from the exhaust. An under-the-hood molten carbonate
device was designed, fabricated, and installed on an automobile for road tests.
A schematic of this device is illustrated in Fig. C-l. The gas from the exhaust
manifold is brought into the device through the inlet and maintains the salt in a
molten state. The flow of hot gas is accelerated through the venturi tube, across
the venturi throat, out the venturi recovery tube, and into a wetted-mesh
reaction zone. The venturi action provides the pumping power to lift the molten
salt up through the salt intake tube and aspirate it into the gas stream. As the
VENTURI
INLET-
TO
VENTURI
RING
BAFFLE
IMPACTION
PLATE
•VENTURI
BYPASS
INLET
DEMISTER
WETTED
MESH
MOLTEN
SALT
RESERVOIR-
M/ENTURI
SUCTION
TUBE
OUTLET
SAMPLE
PORT
EXHAUST
OUTLET
Fig. C-l. Schematic of an Engine -Compartment •
Mounted Molten Salt Scrubber
(from Ref. C-2)
C-3
-------
gas stream passes through the wetted mesh, removal of participates is
accomplished by absorption on the mesh. After the gases pass through the
reaction zone, they pass over a baffle into the demisting zone where final
removal of the entrained salt is accomplished. The absorbed particles are
carried with the demisted melt into the melt pool where the heavier particles
of lead and corrosion products form a slurry at the bottom of the molten salt
pool. From the demister zone, the exhaust gases flow through the outlet port
with a maximum reduction in temperature of 10-20°F. To reduce corrosion,
an air-cooled bypass line is provided at the device inlet to permit temperature
control of exhaust gases passing the molten salt pool; this prevents the melt
from exceeding temperatures of 1200°F. The exhaust passed through the
bypass line is combined with gas flowing through the venturi in the wetted-
mesh reaction zone.
A fabricated under-the-hood molten carbonate device is shown with a catalytic
reactor in Fig. C-2, and installed in an automobile engine compartment in
Fig. C-3. Eight thousand miles of random over-the-road operation, and
subsequent California 7-mode cycling tests (Ref. C-3) on an engine dynamom-
eter for an additional 7500 equivalent miles, indicated that this device could
be expected to remove over 90 percent of the lead, essentially all the sulfur
oxides, and in excess of 80 percent of all participates over the entire operating
range of an automobile, in addition to attenuating engine noise as well as
standard mufflers. It was also observed that backpressures in the exhaust
system did not exceed those experienced in standard mufflers.
The device contains 10-12 pounds of carbonate and requires an engine com-
partment space of 10 inches in diameter and 14 inches in height. When ser-
vicing is required, the salt is removed from the device and fresh carbonate
is added through a fill tube. The salt must be changed periodically to main-
tain the melting point of the mixture and to remove accumulated waste particles
from the system. Oxidation of sulfite to sulfate in the carbonate melt results
in an increase of about 25 F in the melting temperature in about 10,000 miles
of operation. Lead and other metals form a sludge at the bottom of the salt.
C-4
-------
Fig. C-2. Fabricated Engine-Compartment-Mounted Molten-Salt
Scrubber (Courtesy of Atomics International)
It is presently estimated by Atomics International that the salt will require
changing every 15,000 to 20,000 miles at a cost of $10. By extrapolating the
deepest corrosion penetrations linearly, Atomics International predicts that
the life of a device fabricated from aluminized steel will be a minimum of
50,000 miles.
Atomics International has recognized that the under-the-hood prototype device
cannot be accommodated in all automobile engine compartments, and is there-
fore, currently in the process of developing an under-the-car device which
may be capable of retrofit onto all currently operational automobiles. A con-
ceptual design of such an under-the-car device is shown in Fig. C-4. The
Pricing and Estimating Division of Atomics International has estimated that
the production cost of this device will be $19.58 per unit based on one million
units per year. At typical markup and profits plus installation, the cost to
C-5
-------
n
Fig. C-3. Installation of Engine-Compartment-Mounted Molten-Salt Scrubber
(Courtesy of Atomics International)
-------
DEMISTING
/MESH
,VENTURI
TOP COVER
FORWARD
END PLATE
BYPASS VALVE
WETTED SCRUBBER
'MESH REAR ENO PLATE
EXHAUST
GAS
INLET,
EXHAUST
GAS
OUTLET
\
/ PICKUP TUBE
'•PERFORATED
SHEET
\ FLOW HOLES XMELT
SCREEN x FILL HOLE
DRAIN PUJG
SHEET \
'2nd BAFFLE
Fig. C-4. Under-the-Car Lead Trap--Conceptual Design (from Ref. C-2)
the user is reported to range from $35 for a factory-installed device to $45
for retrofit units.
The molten carbonate lead trap device will not be compatible for leaded fuel
operation in conjunction with a catalytic reactor unless the lead is removed
in sufficient quantities and durability of removal capability is sufficiently long.
The lead trap should not provide any significant delay in attaining efficient
operating temperatures for the catalytic reactor. During the 7-mode cycling
tests (Ref. C-3), the lead trap achieves operating temperatures of 1000°F in
slightly over 2 minutes, with temperature drops across the device never
exceeding 20 F. As previously mentioned, the molten carbonate process
removes practically all SO? from the exhaust gases and thus permits con-
sideration of catalytic reactors which are susceptible to SO;? poisoning.
Finally, the lead trap device has not exhibited backpressures in excess of
those found in conventional exhaust systems and attenuates the engine noise
as well as standard mufflers.
C-7
-------
C.3 TECHNIQUES FOR REMOVING LEAD COMPOUND PARTICLES
ONLY FROM EXHAUST EMISSIONS
This section summarizes the seven known trap devices that have been
suggested for collecting lead particles from automobile exhaust emissions.
With the exception of the IIT Research Institute's thermal packed bed and
Cooper Union's molten salt device, each system includes inertial separation
with varying degrees and types of coagulation and/or agglomeration of the
particles to facilitate separation. With the exception of the Du Pont and
Ethyl devices, the techniques suggested have not progressed beyond limited
exploratory testing in the laboratory.
Dow Chemical (Ref. C-4) and Ethyl (Ref. C-5) appear to be in general agree-
ment that lead in the automobile exhaust system remains in a partially
vaporized state above 650°F. To ensure temperatures below this value in the
exhaust system for speeds up to 60 mph, it would be necessary to locate lead
participate collection devices near the end of the tailpipe. Du Pont (Ref. C-6)
indicates that to achieve maximum effectiveness in collecting lead particles
from auto emissions by inertial separation, the exhaust gases should be
cooled below 550°F.
C.3.1 Du Pont Cyclone Trap System
A relatively simple exhaust particulate trapping device which replaces the
normal exhaust system has been devised by Du Pont. It consists of a dual
fluted pipe to cool the exhaust, and a cyclone trap to separate and collect the
solid particles (Ref. C-6). The combination of the fluted pipe and the dual
system increases the surface area and cools the exhaust gas below 550°F
under normal driving conditions. (This temperature limitation is considered
necessary in order to achieve maximum effectiveness for inertial separating
devices such as cyclones.) After being cooled in the fluted pipe, the gases
flow into a mesh-lined cyclone trap box. The exhaust gas first passes through
wire mesh to agglomerate the particles and then through a cyclone to separate
the particles from the gas. The cyclone boxes have sufficient capacity to
store all the separated lead salts for the expected life of the car.
C-8
-------
A 64,000-mile test of this device on a 1966 car resulted in a 70-percent
reduction by weight in lead participate emissions compared with an equivalent
vehicle with a conventional exhaust system.
The simple cyclone device was further developed to provide additional cooling
and agglomeration by wire mesh-lined pipes and a mesh-filled box placed
ahead of the two cyclone traps. Lead participate exhaust emissions from two
cars equipped with the advanced trap system operating for an equivalent
70,000 miles was compared with those from two similar cars with conventional
exhaust systems over the same operating range. The reported results indi-
cated that the improved trap system reduced lead participate emissions by
80 to 85 percent.
C.3.2 Ethyl Corporation Participate Traps
Ethyl reports exploration of a wide variety of trapping devices and concludes
that for a simple, low-cost, muffler-type device, the principal of inertia!
separation is the most promising approach (Refs. C-5, C-7, C-8, and C-9).
Ethyl has also discovered that the greatest hindrance in the use of inertia!
trapping systems is the high temperatures (ranging from 1000°F to 1500°F) of
the gas exhausting from the manifold. Since Ethyl considers lead halides to
be partially vaporized above 650°F, it has been necessary to design and test
several heat exhangers to obtain one that is capable of giving temperature
drops of up to 900°F.
Advancement of the inertial separation principle has led to the development of
a unit (Fig. C-5) composed of two inertial elements referred to as anchored
vortex tubes. Particulates separated at the walls are rejected at high energy
through slots near the base of the closed-end tube. This unit is currently
undergoing tests and has been subjected to 24,000 miles of operation to date,
resulting in a 65-percent reduction in exhausted lead. On the basis of present
test mileage, the capacity of this device is considered to permit a life of at
least 50,000 miles. Costs are estimated to be in the range of standard-type
mufflers.
C-9
-------
Fig. C-5. Experimental Dual-Anchored Vortex Trap (fromRef. C-9)
Ethyl believes that more complex devices which add an interceptor ahead of
an inertia! device are more promising than the plain inertia! device. The
interceptor is a chamber of loosely packed or loosely woven material on which
particles can impinge and grow, and then migrate on through to be collected
by the inertial device. Although preliminary testing of this approach is
reported to have given reductions in exhausted lead in the 70 to 90 percent
range, Ethyl considers that further development work is necessary to make
such devices simple and practical.
C.3.3 Dow Chemical Molten Salt Participate Trap
In a limited effort by Dow Chemical, the approach selected for design of a
trapping device was to wet the exhaust particles with a suitable liquid medium
which increases the mass of the particles and facilitates inertial separation
(Ref. C-4). Prototypes were constructed and subjected to a restricted number
of laboratory tests. Sufficient data have not been obtained to evaluate this
system at this time.
C-10
-------
C.3.4 Cooper Union Molten Salt Approach
Cooper Union also conducted an exploratory study on the removal of lead
particulates from automobile exhausts by means of molten salt (Ref. C-10).
The approach was based on the assumption that lead particles, once cap-
tured and wetted by molten salts, would be prevented from being re-entrained
into the gas stream because of the relatively high surface tension of molten
salts. Exhaust gases from an idling 1964, 6-cylinder Falcon were passed
through a molten salt kettle and the resulting lead content was compared with
that of the untreated gas. Limited data from these tests showed a reduction
of lead participate emissions which varied from 32 to 72 percent.
C.3.5 ITT Research Institute Devices
IIT Research Institute has conducted research efforts to establish the feasi-
bility of developing two collection devices for the removal of particulate
matter from internal combustion engine exhausts (Ref. C-ll). The first
approach considered thermal precipitation which makes use of the phenome-
non of particle migration and deposition in a temperature gradient. The
second approach was the fluidized bed, which makes use of the phenomenon of
high-velocity gradients .
C.3.5.1 Thermal Packed Bed Device
The thermal deposition approach is based on the fact that particles move from
hot-gas streams to cold surfaces under the influence of thermal force. A
particle in the thermal gradient can be expected to be hotter on one side than
the other. On the average, gas molecules striking the hot side will rebound
at a higher temperature and, hence, at a higher velocity than those striking
the cold side. This imbalance creates a force on the particle directed toward
the colder end of the temperature gradient.
On the basis of this principle, an experimental Laboratory setup of a packed
bed made up of high-heat capacity material was subjected to aerosols approxi-
mating lead basic constituents found in automobile exhausts. The collection
efficiency of the thermal packed bed device was determined by simultaneous
C-ll
-------
sampling of the gas both upstream and downstream of the bed. Initial tests
were conducted using a packed bed of 8-mm glass helices at a gas-packing
temperature difference of 1112 F. The collection efficiency of this device
was found to be 86 percent.
Experimental results from tests indicated that the most important variable
affecting the collection efficiency of the thermal packed bed device is the gas-
packing temperature difference. It was found essential to maintain temperature
differentials in excess of 390 F to operate the device with a high efficiency.
Thus, it becomes necessary for the device to be located downstream of the
manifold, where most of the lead is in participate form and where the exhaust
temperature difference will be sufficient to ensure high collection efficiency.
Since the bed will heat with time on the passage of hot exhaust gases, a tech-
nique is required to permit operation of the device while maintaining a large
gas-packing temperature differential at all times. One method proposed for
accomplishing this result is shown schematically in Fig. C-6. Two packed
beds would be used intermittently to clean the gas. As the exhaust gas flows
through one bed, the other is cooled by a flow of relatively cold outside air.
When the temperature sensor indicates that one bed is becoming too warm
for efficient operation, the butterfly valve changes to channel the flow to an
alternate bed. Another approach to cool the bed internally was also suggested
in order to require only one packed bed.
Again, insufficient effort has been directed towards developing the thermal
packed bed device to permit prototype designs for automobile installation and
estimates of frequency of replacing packing material, maintenance, and life
of device as well as costs .
C.3.5.2 Sonic Fluidized Bed Device
The second technique studied by IIT Research Institute to control particle
emissions from auto exhausts was the fluidized bed. The mechanism of aero-
sol removal in fluidized beds is related to the high-velocity gradients in a gas
flow around the particles in the bed. The gas particles do not follow the flow
C-12
-------
VENT
CLEAN GAS
A (COOLING)
PACKED BED
B (COLD)
PACKED BED
SCREEN
HOT
DIRTY
GAS
Fig. C-6. Packed Beds for Collection of Submicron Particles
by Thermal Precipitation (from Ref. C-ll)
stream lines because of their inertia and impact on the fluidized bed particles
adhering to them. The velocity gradients between the bed and aerosol parti-
cles can be enhanced by superimposing a sound field.
Collection efficiency of the fluidized bed was found to increase sharply with
power input to the sound driver units when standing waves were used. In
these tests, 210-500-micron glass beads were used as the fluidized material
and efficiencies of up to 90 percent were obtained with the generation of
standing sound waves for inputs of 125 watts to each of four loudspeaker sound
driver units. Since natural sound levels in an auto exhaust are not sufficient,
an auxiliary acoustic unit would be needed to generate an additional ZOO watts
of acoustic energy.
C-13
-------
IIT Research Institute envisions that such a device designed for installation
in automotive systems would occupy about 2.5 cubic feet of space and weigh
approximately 45 pounds. The sonic generator would probably be an air-
driven siren supplied by a belt-driven air compressor. The cost for such a
device was estimated to be somewhat less than $100 at the retail level. Main-
tenance is considered to be minimal and the material for refilling the fluidized
bed would cost approximately $2.50. Although this particular approach has
not progressed beyond the conceptual stage, the dissipation of acoustic energy
from such a device installed in an automobile could result in significant
design problems.
C.3.6 Houston Chemical Company Particulate Trap
In a news release (Ref. C-12), PPG Industries' Houston Chemical Company
disclosed the development of a new particulate trap that it claimed would
remove 99 percent of all particulate matter from automobile exhausts and
might protect catalytic reactors from lead poisoning. This system is expected
to replace the normal exhaust system of an automobile, to be only slightly
more expensive, and to have a life expectancy of at least 50, 000 miles .
Preliminary discussions, subsequent to this news release, indicate that the
system consists of a fiberglass filter system designed to be used after a
cyclone separator. The system will operate only at exhaust temperatures
below 700°F to avoid decomposition of the fiberglass . Also, it is necessary
to control the exhaust to low temperatures (approximately 500 F) to ensure
that the lead particulates have condensed from the vapor phase. As such,
this system would only be compatible with a lead-sensitive catalyst that could
function at low temperatures. However, such catalysts have not been identi-
fied to date. Mileage accumulation to 10,000 miles indicates removal capa-
bility of 99 percent (of particulates above 0. 3 micron) and a backpressure
increase equivalent to the standard muffler, according to the Houston
Chemical Company.
C-14
-------
REFERENCES
C'1> Development of the Molten Carbonate Process to Remove Lead and
Other Participates from Spark Ignition Engine Exhausts. Report
No. AI-70-47, Atomics International, Canoga Park, California
(8 July 1970) (HEW Contract No. CPA 70-3).
C-2. L.F. Grantham and S.S. Yosin, "Removal of Lead Gases with Molten
Alkali Metal Carbonates, " Paper presented at the American Chemical
Society Symposium on Current Approaches to Automotive Emission
Control, Los Angeles, California (28 March - 2 April 1971).
C-3. "Standards for Exhaust Emissions, Fuel Evaporative Emissions, and
Smoke Emissions Applicable to 1970 and Later Vehicles and Engines,"
Federal Register. Vol. 33, No. 108, Part II (4 June 1968).
C-4. J.B. Moran, et al. , Development of Particulate Emission Control
Techniques for Spark Ignition Engines. Report No. EHS70-101 . Dow
Chemical Company, Midland, Michigan (July 1971)(EPA Contract
No. EHS 70-101).
C-5. H.D. Coffee, Jr., et al., "Clean Air Car," Ethyl Corporation,
Detroit (Unpublished report)(24 July 1970).
C-6. K. Habibi, et al. , "Characterization and Control of Gaseous and
Particulate Exhaust Emissions from Vehicles," Paper presented to
the Air Pollution Control Association, San Francisco, California
(October 1970).
C-7. D.A. Hirschler and F.J. Marsee, Meeting Future Automobile
Emission Standards, Report No. AM-70-5, Ethyl Corporation
Detroit (April 1971).
C-8. G.L. TerHaar, et al. , "Composition, Size and Control of Automotive
Exhaust Particulates," Paper presented to the Air Pollution Control
Association, Atlantic City, New Jersey (June 1971).
C'9' Consequences of Removing Lead Antiknocks from Gasoline. A Status
Report, No. AC-10, Ethyl Corporation, Detroit (August 1970).
C-10. Shang-I Cheng, et al. , "Removal of Lead from Automobile Exhausts
by Molten Salts, " Environmental Science and Technology. Vol. 5
No. 1 (January 197TT
C-15
-------
REFERENCES (cont.)
C-ll. S. K. Sood and Richard Karuhn, Development of Particulate Emissions
Control Techniques for Spark Ignition Engines, Report No. C6186-5,
Illinois Institute of Technology Research Institute (February 1971) (EPA
Contract No. CPA-22-69-134).
C-1Z. Chemical Engineering News (20 September 1971).
C-16
------- |