PB-201 410
EMISSION REDUCTION USING GASEOUS FUELS FOR VEHICULAR PROPUL-
SION
Institute of Gas Technology
Chic ago , 11 li noi s
June 1971
Distributed . , .'to foster, serve and promote the
nation's economic development
and technological advancement.'
NATIONAL -ECHNICAL INFORMATION SERVICE
U.S. DEPARTMENT OF COMMERCE
This document has been approved for public release and sale.
-------
PB 2O1 410
INSTITUTE OF GAS TECHNOLOGY
INSTITUTE OF GAS TECHNOLOGY
IIT CENTER
CHICAGO, ILLINOIS 60616
IGT
EMISSION REDUCTION
USING GASEOUS FUELS
FOR VEHICULAR PROPULSION
Final Report
on
Contract No. 70-69
for
AIR POLLUTION CONTROL OFFICE
ENVIRONMENTAL PROTECTION AGENCY
Cincinnati, Ohio 45227
June 1971
-------
8929
SUMMARY
This study was undertaken to accomplish lour objectives:
1. To provide a comprehensive assessment of the state-of-the-art
and the potential for reducing vehicular pollution emissions by
using gaseous fuels in commercially available engine systems.
i. To examine the logistics of gaseous fuels to assess the avail-
ability, cost, storage, and handling methods and safety require-
ments of gaseous fuels.
3. To determine the feasibility of using natural gas as a supple-
mentary fuel in a two stroke cycle, dieSel powered intercity
bus as a means of reducing exhaust pollution while operating
in urban areas and bus terminals.
4. To recommend specific research and development programs
necessary to confirm or establish the low emission character-
istics and economic feasibility of selected gaseous fueled
vehicular engine systems.
The findings of the study as they relate to each objective are summarized
below.
1. Reduction of Vehicular Pollution Emissions
The greatest benefit from conversion to gaseous fuels may be realized
in the immediate future while pollution control devices are being perfected
for use on gasoline-powered vehicles.
Four fuels were examined that can be injected into an internal combus-
tion engine in gaseous form. These were propane, natural gas, ammonia,
and hydrogen. Both ammonia and hydrogen have serious deficiencies from
the point of view of harmful emissions as well as price and availability and
cannot be recommended for motor fuel use.
Propane and natural gas have similar characteristics as far as emission
reductions are concerned, particularly if no distinction is made between
the reactivity of various hydrocarbons in producing photochemical smog.
The use of either of them in a conventional gasoline engine will require the
addition of pollution control devices in the exhaust system as well as adjust-
ments in the engine itself in order to meet 1980 Federal Emissions Standards.
-------
8929
When the required ancillary equipment is added and engine adjustments ari.
made, the levels of emissions from propane and natural gas engines are
not greatly different than from a gasoline engine using the same equipment.
The emissions of NO are reduced somewhat more in the case of the gast.3us
fuels.
Z. Logistics Consideiations
The rapidly increasing demand for clean fuels for stationary applications
which has been created by air pollution control regulations, is beginning to
tax the ability of the natural gas industry to supply both propane and natural
gas.
The amount of energy consumed as motor fuel in the United States is
about 80$ of the amount of energy distributed by the natural gas industry.
Thus, there is no hope of converting all or most of the automotive vechicle
population to natural gas fuel since this would require the production of
natural gas to be almost doubled — at a time when the industry is straining
to maintain its normal growth of 6£ per year.
For similar reasons, we cannot consider the complete conversion of
motor vehicles to propane, for the entire propane industry is one tenth the
size of the gasoline industry, and its principal source of supply is tied to
natural gas production.
Since both ammonia and hydrogen are currently made in large measure
from natural gas, they do not offer acceptable alternatives and their current
prices do not permit them to serve as supplementary fuels if they must
compete on a price basis. They have other technical limitations as well.
However, some natural gas and propane can be made available for use in
the immediate future. The study shows that the diversion of natural gas
from boiler fuel uses, such as electric generation, to motor fuel use results
in a net reduction of harmful pollutants - at least in New York City. Divert-
ing one half of the natural gas presently used for generating electricity in
New York would furnish enough fuel to operate all of the commercial fleet
vehicles in the city on natural gas.
INSTITUTE
»v
r t
8929
Of the two forms of natural gas — liquid and gas, compressed natural
gas is more readily available and currently more popular. Imported LNG
from overseas may become quite attractive,however, because of its pro-
jected prices and the longer vehicle range that it provides.
At the same time.thc production of lead-free gasoline results in an increase
in the yield of propane from re.ineries. This could provide an additional
source of propane for vehicle use although the amount to be expected from
this source is difficult to determine at present because the conversion to
lead-free gasoline is just beginning.
Nevertheless,we believe that sufficient supplies of natural gas and propane
can be made available to fuel all of the vehicles that are likely to be con-
verted to gaseous fuels in critical pollution areas before 1975 — barring
some legislative edict thai would require their use.
The handling and safety problems of both propane and natural gas are
well known since these fuels have been used in motor vehicles for many
years either in this country or in Europe. They are necessarily highly
volatile, flammable fuels comparable to gasoline. While some individual
characteristics serve to favor one fuel or the other, both have established
acceptable safety records and should not be disqualified on this basis.
3. Conversion of Two-Stroke Diesel Engine
The methods currently available for converting the 6V-71N and 8V-71N
diesel engines to the part-time use of natural gas cannot be recommended
in view of their ineffectiveness in improving exhaust emission control and
their high costs of conversion and inconvenience of operation. Except for
some reduction in odor, none of the conversions promise reduction in
regulated pollutants below those now reported for the engines when properly
equipped with available pollution control devices offered by the manufactuer.
4. Recommendations
The findings and conclusions of the study indicate that further development
of the use of natural gas and LP gas as motor fuel Is justified wherever logisti-
cal and economic conside-ations permit their use. A number of specific
recommendations are made and these are summarized below in order of their
importance.
-------
89Z9
Further information is needed on the reactivity of various typeb of hydro-
carbon emissions from engines in order to clarify the extent to which various
gas engine emissions are harmful.
It is recommended that an optimized design of an 1C engine to use gaseous
fuels be built and tested to demonstrate the performance attainable as well
as the emissions produced by such an engine.
Additional research and development is needed to better determine the
effectiveness of exhaust catalytic reactors and afterburners in handling the
lower concentration of pollutants from natural gas and LPG engines.
Further research is needed on the automatic duel-fuel concept in order
to determine how to obtain the maximum number of clean vehicle Rules with
a minimum amount of gaseous fuels.
A need exists for odorants that can be used in LNC in the same manner
as in natural gas.
It is recommended that more experimental work be done in measuring the
pollution level from gas vehicles as a function of engine characteristics and
modifications.
It is recommended that a controlled test program be undertaken to document
the history of a fleet of gas-fueled veKTcles aa far as emissions, fuel costs,
maintenance expense, and other significant items over an extended period of
operation.
It is recommended that a survey be made of European experience in gas-
fueled vehicles, particularly in France and Italy.
It is recommended that additional urban areas besides New York City
be examined to determine the potential of gaseous fuels to reduce overall
pollution levels in other local urban areas.
There is a need to examine in more detail the possibility and potential
price of liquefied petroleum gas imported into the United States from
overseas.
I N S T I 1 II 1
69 ZS
It is rfcommcndcd thai further studies be made of the feasibility of
increasing the ynOd of propane from oil refineries at Ihe expense of gasoline
production and the resultant effects upon cost of both fuels and upon their
distribution expense.
-------
8929
TABLE OF CONTENTS
Page
1. INTRODUCTION 1-1
2 Vehicular Pollution Characteristics of Caseous Fuels 2-1
2-1 Combustion Character.sties of Caseous Fuels 2-1
2. 1 1 Basic Properties 2.2
2 1.2 Combustion Characteristics 2-2
213. Formation of Exhaust Pollution Products 2.7
2-1.3.1. Hydrocarbons (HC) 2-7
2 1 3.2 Nitrogen Oxides (NO*) 2.14
2.1.3 3. Carbon Monoxide (CO) 2.27
2 2 Exhaust Emission Characteristics in SI-Engines 2.32
2.2.1. General Discussion 2-32
2.2.2. Determining Factors of Exhaust Emission 2-32
2.2.3. Emission Characteristics of Propane 2-34
2.2.3 1. Fuel-Air Ratio 2-34
2 2.3.2 Ignition Timing and Engine Speed 2-38
22 3.3. Throttle Position 2-41
2.2.3.4. Compression Ratio ' 2-46
2.2.4. Emission Characteristics of Methane 2-47
2 Z.4.1. Air-Fuel Ratio 2-47
2 2 4.2. Throttle Position 2-49
2.2.4.3 Ignition Timing 2-50
2 2. 5. Emission Characteristics of Ammonia 2-52
2.2 5.1. Combustion of Ammonia 2-52
2.2.5.2. Effect of Engine Variables 2-53
2.2.6. Emission Characteristics of Blended Gas 2-54
2.2.7. Emission Characteristics of Hydrogen 2-55
-------
8929
TABLE OF CONTENTS. Cont.
Page
2.3. Vehicular Exhaust Emissions 2-54J
2.3. 1. Exhaust Emission Testing Procedure 2-58
2.3.2. Effect of Testing Procedure on Exhaust Emission Data 2-60
2.3.3. Exhaust Emission of Gasoline-Fueled Vehicles 2.63
2.3.4. Exhaust Emission of Gaseous Fuel Powered Vehicles 2-71
'2.4. Operating Characteristics of Gaseous Fuels in SI-Engine 2-7A
2.4.1. Propane 2-76
2.4. 1.1. Power Capability 2-76
2.4.1.2. Fuel Consumption 2-78
2.4.1.3. Engine Maintenance 2-79
2.4.2. Methane 2-79
2.4.2.1. Power Capability 2.79
2.4.2.2. Fuel Consumption 2-81
2.4.3. Ammonia 2-81
2.4.3.1. Power Capability 2-81
2.4.3.2. Fuel Consumption 2-M
2.4.3.3. Engine Maintenance Z-4Z
3. EFFECT OF GASEOUS FUELED VEHICLES ON
POLLUTION LEVELS IN NEW YORK CITY 3-1
3.1. New York City Traffic Patterns 3-1
3.2. Emissions From Present Vehicle Traffic 3-5
3.3. Comparison With Gaseous Fuel Use in Central
Electric Generating Stations 3-12
4. LOGISTICS OF GASEOUS FUELS 4-1
4.1. Natural Gas 4-1
4. I.I. Natural Gas Industry 4-1
4.1.1.1. Residential Sales 4-1
4.1.1.2. Commercial Sales 4-2
4.1.1.3. Industrial Sales 4-2
I N S T I T u 1 I
8929
TABLE OF CONTENTS. Cont.
-'. 1.1.4. Other Sales
4. 1. 1.5. Capital Structure
4. 1.2. Natural Gas Supply
4. 1. 3. Demand
4.1.4. Sources of Supplemental Natural Gas Supply
4.1.4.1. Imported Canadian Gas
4.1.4.2. Imported Mexican Gas
4.1.4. 3. Gas From Alaska
4. 1.4.4. Imported LNG
4. 1.4.5. North Atlantic Continental Shelf
4.1.4.6. Synthetic Pipeline Gas
4. 1.5. Supply Deficiency
4.1.6. Compressed Gas
4.1.7. LNG
4. 1.8. Price of Natural Gas for Motor Fuel Use
4. 1.8. 1. Compressed Natural Gas
4.1.8.2. LNG
4.2. Propane
4.2.1. Field Production of Propane
4.2.2. Refinery Production
4.2.3. LPG Importation
4.2.4. Domestic Supply Outlook
4.2.5. Demand Outlook
4.2.5. 1. Residential and Commercial Markets
4.2.5.2. Petrochemical Market
4.2.5.3. Engine Fuel Market
4.2.5.4. Other Markets
4.3. Ammonia
4. 3. 1. Product Characteristics
Page
4-2
4-2
4-3
4-5
4-6
4-7
4-8
4-8
4-9
4-12
4-13
4-14
4-14
4-16
4-17
4-17
4-21
4-23
4-23
4-25
4-26
4-27
4-33
4-34
4-34
4-35
4-38
4-44
4-44
XI
c * s
TFCMMOLOGV
-------
89Z9
8929
TABLE OF CONTENTS, Cent.
4.3.2. Ammonia Supply
4.3.3. Demand
4.3.4. Price
4.4. Hydrogen
4.4.1. Hydrogen Supply
4.4.2. Prices of Hydrogen
5. STORAGE AND HANDLING OF GASEOUS FUELS
5. 1. Natural Gas
5.1.1. Bulk Storage
5.1.2. Compressed Gas
5.1.3. LNG Storage
5.1. 4. Vehicular Storage
5. 1.4.1. Compressed Natural Gas
5.1.4.2. LNC
5. 1. 5. Handling Procedures and Practices
5.1.5.1. Compressed Natural Gas
5.1.5.2. LNG
5.1.6. Safety
5.1.6.1. Compressed Natural Gas
5.1.62. LNG
5.2. LPG
5.2.1. Bulk Storage
5.2.2. Bulk Transportation
5.2.3. Vehicular Storage
5.2.4. Handling Practices
5.2.5. Safety
5.3. Ammonia
5.4. Hydrogen
Page
4-' I
4-47
4-52
4-54
4-54
4-57
5-1
5-1
5-1
5-1
5-2
5-3
5-3
5-4
5-6
5-6
5-9^
5-12
5-12
5-12
5-14
5-14
5-14
5-15
5-16
5-16
5-17
5-20
5.4. 1.
5.4.2.
6.
6.1.
6. 1. 1.
6.1.2.
6.1.3.
6.1.4.
6.1.4. 1.
6.1.4.2
6.1.4.3.
6.1.5.
6.2.
6.2. 1.
6.2. 1. 1.
6.Z. 1.2.
6.2. !.3.
6.2.2.
6.2.3.
7.
7.1.
7.1.1.
7.1.2.
7.2.
7.3.
7.4.
TABLE OF CONTENTS. Cont.
Storage and Handling
Safety
ECONOMIC IMPACT OF THE USE OF GASEOUS FUELS
Industrial Impact Based in Large-Scale Conversion
Effect Upon the Natural Gas Industry
Effect Upon the LPG Industry
Effect Upon the Oil Industry
Effect Upon Equipment Suppliers
Engines
Tanks
Other Components
Effect Upon Vehicle Users
Industrial Impact Based on Commercial Fleet
Vehicle Operation
Vehicle Conversion Costs
Compressed Natural Gas Conversion Kit
LNG Conversion Costs
LPG Conversion Costs
Economic Incentives
Dual-Fuel Operation
COMPARISON OF GASEOUS FUEL SYSTEMS
Emission and Performance
General Discussion
Summary of Gaseous Fuel Emission Levels
Logistics Considerations
Operating Costs
Potential for TSarly Commercialization
Page
5-20
5-21
6-1
6-1
6-1
6-1
6-2
6-3
6-3
6-3
6-4
6-4
6-7
6-7
6-8
6-9
6-9
6-10
6-16
7-1
7-1
7-1
7-1
7-5
7-7
7-10
-------
8929
89 Z9
TAIll.KOJ (.ONTENI.S. Conl.
8. ANALYSIS OF THE US EOF NATURAL GAS
IN AN INTERCITY BUS
8. 1. Objective
8.2. Introduction
8.3. Emission Control Regulations
8.3.1. Current Status
8.2.2. Fugure Regulations
8.4. Current Equipment and Emissions Produced
8.4. 1. Emission Related Features
8.4.2. Emissions Produced
8.4. 3. Anticipated Effects of Conversion to Caseous Fuels
8.5. Engine Modifications Required
8.5.1. Conversion to Spark Ignition Engines
8.5.2. Dual-Fuel Conversion
8.5.2.1. Dual-Fuel Engine Performance
8.5.2.2. Emissions From Dual-Fuel Engines
8.5.2.3. Engine Modifications Required for Dual-Fuel Operation
8.5.3. Power Boost Systems
8.5.4. Stratified Charge Gas-Fueled Diesel Engine
8.6. Alternatives Available
8.7. Conclusions and Recommendations
9. RECOMMENDATIONS
10. REFERENCES CITED
Page
8-1
8-1
8-1
8-2
8-2
8-2
8-5
8-5
8-6
8-7
8-8
8-9
8.9
8-10
8-11
8-12
8-12
8-14
8-14
8-15
9-1
10-1
2. 1.
2. 1.
2. 1.
2. 1.
2. 1.
2.1.
2.1.
2. 1.
2. 1.
2. 1.
2. I.
2. 1.
2. 1.
2.1.
2. 1.
2.2.
.3.
,3.
3.
3.
3.
3.
3.
3.
3.
3.
3.
3..
3. .
3. .
3..
3..
I.IS'I OP ILLUSTRATIONS
Figure No. Page
-1 TEMPERATURE PROFILE OI QUENCH ZONE 2-8
-2 QUENCH SURFACE AND PISTON-RING CREVICES
OF AN SI-ENGINE COMBUSTION CHAMBER 2-9
-3 COMPUTED EQUILIBRIUM COMPOSITION FOR
CONST ANT-VOLUME ADIABATIC COMBUSTION OF
ri-OCTANE 2-15
-4 TEMPERATURE VERSUS EQUILIBRIUM NITRIC
OXIDE CONCENTRATION 2-19
-5 KINETIC PROGRESSION OF NITRIC OXIDE
FORMATION UNDER VARIOUS TEMPERATURES 2-19
-6 DURATION OF ENGINE EVENTS FOR A PRODUCTION
318 CID V-8 ENGINE 2-20
-7 RATE OF NITRIC OXIDE FORMATION IN HIGH
PRESSURE HYDROGEN-AIR FLAME 2-21
-8 THEORETICAL PEAK OTTO CYCLE TEMPERATURE
OF VARIOUS FUELS 2-23
-9 THEORETICAL PEAK OTTO CYCLE PRESSURE OF
VARIOUS FUELS 2-23
-10 PREDICTED EQUILIBRIUM PEAK CONCENTRATION
OF NITRIC OXIDE 2-24
-11 NITRIC OXIDE CONCENTRATION VS. AIR-FUEL RATIO
IN A STANDARD GASOLINE-FUELED SI-ENGINE 2-25
•12 COMPARISON OF EXHAUST NO CONCENTRATIONS
OF TWO TYPES OF CI-ENGINES 2-26
-13 COMPARISON OF NO EMISSION OF STRATIFIED
CHARGED AND NORMAL SI-ENGINES 2-27
• 14 CARBON MONOXIDE CONCENTRATION DURING
EXPANSION CYCLE- THEORETICAL VS. MEASURED
EXHAUST LEVELS 2-29
-15 AIR-FUEL RATIO VS. CARBON MONOXIDE EMISSION 2-31
1 EFFECT OF AIR-FUEL RATIO ON EXHAUST HYDRO-
CARBON CONCENTRATION OF PROPANE-FUELED
SI-ENGINE 2-34
I H 5 T
M M n i
-------
8929
figure No.
2 2.3.
2 2.3
2.2.3,
2.2 3.
I 2.3.
2.2.3.
-2
-3
-5
-6
-7
2.2.3.-8
2.2.3.
2.2.3
2.2.4.
-9
.10
.1
2.2.4..2
2.2.4.
2.2.5.
2.2.5.
2.2.7.
2.3. l.-l
LIST OF ILLUSTRATIONS. Cent.
Page
EFFECT OF AIR-FUEL RATIO ON EXHAUST
HYDROCARBON CONCENTRATION '-35
EFFECT OF AIR-FUEL RATIO ON EXHAUST
CARBON MONOXIDE CONCENTRATION 2-36
EFFECT OF AIR-FUEL RATIO ON EXHAUST
CARBON MONOXIDE CONCENTRATION 2-37
EFFECT OF AIR-FUEL RATIO ON EXHAUST NITRIC
OXIDE CONCENTRATION OF PROPANE 2-38
EFFECT OF IGNITION TIMING ON HYDROCARBON
EMISSION AT CONSTANT POWER 2-39
EFFECT OF IGNJTION TIMING ON HYDROCARBON
EMISSION OF PROPANE-FUELED ENGINE AT
IDLING SPEED 2-40
EFFECT OF IGNITION TIMING ON NITRIC OXIDE
EMISSION OF PROPANE-FUELED ENGINE AT
CONSTANT POWER AND VARIOUS AIR-FUEL RATIOS 2-41
POWER CONTOURS OF PROPANE-FUELED CFR
ENGINE 2-42
EXHAUST EMISSION VERSUS POWER LEVEL FOR
PROPANE-FUELED CFR ENGINE 2-44
EFFECT OF AIR-FUEL RATIO ON EXHAUST CARBON
MONOXIDE AND HYDROCARBON CONCENTRATION
OF NATURAL GAS-FUELED ENGINE 2-48
EFFECT OF AIR-FUEL RATIO ON EXHAUST
NITROGEN OXIDES CONCENTRATION OF NATURAL
GAS-FUELED ENGINE 2-49
COMPARISON OF LEAN MISFIRE LIMITS 2-51
THEORETICAL PEAK CYCLE NITRIC OXIDE
CONCENTRATION OF AMMONIA COMBUSTION 2-52
NITROGEN OXIDES IN ENGINE EXHAUST 2-53
EMISSION OF NITROGEN OXIDES OF A HYDROGEN-
FUELED SI-ENGINE 2-56
DRIVING SCHEDULE OF CALIFORNIA 7-MODE
CYCLE TESTING PROCEDURE 2-58
I n s T I T u
8929
1.IS I1 Of IL1.US TUATIONS. I onl
Figure No.
232-1 EXHAUST EMISSIONS OF A GASOLINE-FUELED
VEHICLE UNDER. VARIOUS MODES OF DRIVING
232-2 EFFECT OF ACCELERATION CHANGE ON
EXHAUST EMISSIO, 1 MEASUREMENT
233 1 IMPROVEMENTS MAi'E IN EXHAUST EMISSION
REDUCTION OF GASOLINE-FUELED VEHICLES
2.3.4.-1 EXHAUST EMISSION OF VEHICLES CONVERTED TO
GASEOUS FUELS
241-1 EFFECT OF IGNITION TIMING ON POWER OUTPUT
' " ' OF PROPANE-FLELED ENGINE AT VARIOUS
AIR-FUEL RATIOS
241-2 EFFECT OF IGNITION TIMING ON FUEL
CONSUMPTION OF PROPANE-FUELED ENGINE
AT VARIOUS AIR-FUEL RATIOS
2.4.2. -L POWER CONTOURS OF METHANE-FLELED CFR
ENGINE
3. l.-l COMPREHENSIVE DRIVING CYCLE OF THE CITY
OF NEW YORK
3. 1. -2 QUICK DRIVING CYCLE OF THE CITY OF NEW YORK
3 2 -1 COMPARISON OF NEW JERSEY ACID DRIVING CYCLE
' ' DRIVING CYCLE AND NYC QUICK DRIVING CYCLE
4 1.5.-1 UNITED STATES GAS SUPPLY
4.^.4. -1 PROPANE SUPPLY AND DEMAND
4.2.5.-1 SALES OF PROPANE FOR VEHICLE APPLICATIONS
4. 2. 5.-2 HISTORICAL PROPANE PRICE PATTERNS
4. 2. 5. -3 LOCATION OF MAJOR LP-GAS PIPELINES AND
TERMINALS
4.3. 2. -1 PRODUCTION AND CONSUMPTION OF ANHYDROUS
AMMONIA IN U.S.
4. 3. 3. - I SEASONAL FLUCTUATIONS IN ANHYDROUS
AMMONIA PRODUCTION
4. 4. 1. - I ANNUAL MERCHANT HYDROGEN PRODUCTION.
high purity
Page
2-61
2-62
2-67
2-73
2-77
2-78
2-80
3-2
3-3
3-7
4-15
4-28
4-37
4-39
4-42
4-46
4-49
4-55
-------
8929
8929
LIST OF ILLUSTRATIONS, Cont.
Figure No.
6.2.2.-1
6.2.2.-2
6.2.2.-3
8.3.2..1
8.5.3.-1
Page
EMISSIONS OF HYDROCARBONS AND OTHER ORGANIC
GASES IN TONS PER DAY FROM MOTOR VEHICLES
IN LOS ANGELES COUNTY
EMISSIONS OF CARBON MONOXIDE IN TONS PER DAY
FROM MOTOR VEHICLES IN LOS ANGELES COUNTY
EMISSIONS OF OXIDES OF NITROGEN IN TONS PER DAY
FROM MOTOR VEHICLES IN LOS ANGELES COUNTY
13.MODE DIESEL EMISSIONS TEST CYCLE
THE CENTURY "TORQUE TOPPER" SYSTEM
Table No.
6-12
6-13
6-14
8-4
8-13
2.
2
2.
2.
2.
2.
2.
1. l.-l
1.2.-1
1.3.-1
1.3. -2.
1.3. -3
3.3.-1
3. 3. -2
3.2..2
3.2. -3
3.3.-1
3.3.-2
4. 1. l.-l
4. 1.2.-1
4.1.3.-1
4.1.4.-1
4. 1.4.-2
4.1.7.-1
LIST OF TABLES
PaBe
BASIC PROPERTIES OF GASEOUS FUELS 2-3
FUNDAMENTAL BURNING VELOCITIES FOR
HYDROCARBONS 2-4
EXHAUST HYDROCARBON COMPOSITION OF
GASOLINE-FUELED AUTOMOBILES 2-12
EXHAUST HYDROCARBON COMPOSITION OF
PROPANE-FUELED SI-ENGINE 2-13
ADIABATIC FLAME TEMPERATURES AND NITRIC
OXIDE CONCENTRATIONS 2-17
PROGRESSION OF AUTOMOTIVE EMISSION
STANDARDS 2-64
AVERAGE EMISSIONS OF 1970 VEHICLES TESTED
ACCORDING TO THE NEW JERSEY ACID CYCLE
PROCEDURE 2-71
CHARACTERISTICS OF VEHICLE POPULATION
USED FOR CALCULATION OF EMISSION LEVELS 3-4
COMPARISON OF EMISSION DATA OBTAINED WITH
NYC CYCLES AND 1968 FEDERAL PROCEDURE 3-5
EMISSIONS PRODUCED BY NYC VEHICLES 3-9
EMISSION FACTORS FOR GASEOUS FUELED VEHICLES 3-11
EMISSION CONVERSION FACTORS FOR ELECTRICAL
POWER GENERATION 3-13
ELECTRICAL GENERATING EMISSION SUMMARY 3-14
QUANTITY OF GAS SOLD AND REVENUES BY SALES
CLASSIFICATION 4-3
ECONOMICAL RECOVERABLE U.S. GAS SUPPLY 4-4
TOTAL PRIMARY CONSUMPTION IN U.S. AND
SHARE PROVIDED BY NATURAL GAS 4-5
CANADIAN GAS AVAILABLE FOR EXPORT 4-7
LNG COSTS AT LOS ANGELES 8 500 MILLION
SCF/DAY 4-12
LNG FACILITIES IN THE U.S. AND CANADA 4-18
T I T U T E
1 F f H N C
M O I O C V
-------
8929
LIST OF TABLES, Cent.
Table No.
4. 2. 4. -1 REFINERY PROPANE PRODUCTION
4.2.5.-1 PROPANE SOLD FOR VEHICLE FUEL BY STATE
4.2. 5. -1 PHYSICAL CHARACTERISTICS OF ANHYDROUS
AMMONIA
4. 3.3. -1 MAJOR AMMONIA PRODUCING STATES (CAPACITY)
4.3.3. -2 MAJOR AMMONIA CONSUMING STATES
4. 4. -1 PHYSICAL PROPERTIES OF HYDROGEN
4. 4. 1. -1 COMPARISON OF THE MANUFACTURING COSTS
OF HYDROGEN FROM SEVERAL PROCESSES
4. 4. 2. -1 MERCHANT HYDROGEN COST
5.1.5. -1 CNG COMPRESSION COSTS USING ENGINE-DRIVEN
COMPRESSOR
5. 1. 5. -2 COMPOSITION OF LNG SAMPLES WITHDRAWN
FROM STORAGE (625 X 10' CF)
5. 3. -1 COMPARATIVE COST OF AMMONIA TRANSPORT
5. 3. -2 LOCATION OF MAJOR AMMONIA STORAGE
LOCATIONS
7. 1. 2. -1 RELATIVE EFFECT OF MANUFACTURER'S
MODIFICATIONS AND USE OF GASEOUS FUELS ON
VEHICLE EXHAUST EMISSIONS
7. 1. 2. .2 ESTIMATED EMISSIONS IN CM/MILE FOR GASOLINE
AND GASEOUS FUELS COMPARED WITH PROPOSED
1980 FEDERAL STANDARDS
7. 3. -1 COMPARISON OF GASOLINE AND GASEOUS FUEL
COSTS
8.3.2. -1 MODE DIESEL EMISSIONS CYCLE
8.4. 2. -1 EXHAUST EMISSIONS FROM A 6V.71 (2V. ) DETROIT
DIESEL DIV. ENGINE
Page
4-3
4-40
4-45
4-50
4-51
4-54
4-56
4-58
5-8
5-11
5-18
5-18
7-Z
7-4
7-9
8-3
8.7
INSTITUTE
XX
c
TECHMOLOCr
8929
1. INTRODUCTION
The exhaust emissions from automotive vehicles comprise the greatest
single source of air pollution in our urban atmosphere. The growing concern
for the quality of our environment has caused a great deal of effort to be
expended in finding ways to reduce the harmful emissions from automotive
exhausts. In general, there are five methods to reduce these emissions
from automobile exhausts. They a -e —
1 Remove the pollutants from the engine exhaust stream
2. Modify the engine so that pollutants are not formed
3. Use a new kind of engine that inherently has a cleaner exhaust
4. Change the fuel in some ways so that formation of pollutants will be
minimized
5. Use a completely new fuel that produces less harmful emissions
Most of the effort to date has been expended in modifying the existing
internal combustion engine system or in developing new engines such as the
gas turbine, steam engine, or other external combustion engines.
This report is concerned with the fifth method listed. It is specifically
limited to those fuels that are gaseous in form when introduced into the
engine, although they may be handled and stored in liquid form prior to
injection.
Of the five methods listed only the first four have been thoroughly
examined to date. None of these first four methods has produced a
completely acceptable solution to the problem although there is a possibility
that a combination of 1, 2, and 4 may produce an acceptable solution to the
overall problem by the year 1975 or 1976. In the meantime, the lack of an
acceptable solution has led many people to investigate the use of certain
gaseous fuels that have inherently cleaner burning characteristics than gasoline.
These fuels include natural gas, liquefied petroleum gas (LPG), ammonia,
and hydrogen. Natural pas and LPG are, of course, the most practical ones.
Ammonia and hydrogen have been included in order to complete an objective
1-1
-------
8<)Z9
analysis of all the possible contenders. These particular gaseous fuels were
selected because their combustion chemistry is relatively simple and they
are leas likely to produce the heavy hydrocarbons and the partially com-
pleted reactions that cause so much of the problem with conventional gasoline.
There were a number of reservations about including ammonia,but it was
decided to include it in order to document the information that is available
on this relatively unknown fuel.
LPG has been used as an automotive fuel for many years. The number
of LPG vehicles estimated to be in operation total approximately 300.000.
While thia represents a very small portion of the automotive population, it
does mean that the technical and economic problems associated with its use
in competition with gasoline fueled vehicles have in some measure been
resolved. However, its use has been dictated by considerations other than
air pollution control and it is necessary to examine the emissions from LPG-
fueled engines and the logistics of LPG supply to assess its potential as a
means of air pollution abatement.
Although natural gas has not been used extensively in the United States
as an automotive fuel, it has an extensive history of operation in Italy and
France dating back to World War I. Recently the Pacific Lighting Corporation.
a gas utility serving the Los Angeles area, made a number of experimental
installations of compressed natural gas fuel systems in conventional automo-
biles in order to demonstrate their practical value in reducing automotive
pollution levels in the Los Angeles area. At about the same time the San
Diego Gas & Electric Company made several experimental installations
of fuel systems using natural gas in liquid form (LNC). These experimental
installations created considerable interest in gas fueled vehicles both in
California and elsewhere. There are currently more than ZOOO natural-gas—
fueled vehicles in experimental operations throughout the country. Most of
these vehicles use compressed natural gas and many of them are operating
in California. Thus far. they are almost entirely experimental operations.
1-2
8929
One might dismiss gas-fueled vehicles from consideration as a means
of relieving our air pollution problem simply because of the tremendous
logistic problem that exists in converting our automotive population from
gasoline to either or both natural gas and LPG. Such large-scale conversion
is impractical not only from the point of view of its impact on the automotive
industry, but also because of its effects upon the natural gas industry as well.
However, much of our air pollution problem is centered around our large
urban areas where fleets of commercial vehicles comprise a greater portion
of the pollution problem than in the general case. We believed, therefore,
that consideration can and should be given to the use of gaseous fuels for
many fleet vehicle operations as a means of alleviating the pollution problem
in our major cities.
At about the same time that this study was proposed to the Air Pollution
Control Office (APCO), the Institute of Gas Technology had also presented a
proposal to the Greyhound Bus Company to study the feasibility of using
gaseous fuels in intercity buses during the period of time that they were
operating in urban areas and enclosed bus terminals. As a result of conver-
sations among all of the parties involved, we agreed to include that study as
Task IV of the subject study. The work was, therefore, carried out in the
course of the study and reported in Section 8 of this report.
1-3
-------
89Z9
Z VEHICULAR POLLUTION CHARACTERISTICS OF GASEOUS FUELS
Gaseous fuels by nature of their lower molecular weights and Lesser
amounts of carbon tend to produce frwer of the heavy hydrocarbons that con-
tribute to the formation of photochemical smog. In addition, the combustion
of these lighter hydrocarbon fuels within the engine cylinder can proceed
more quickly because no evaporation of liquid droplets is required before
the fuel is brought to the ignition temperature. Thus, the combustion pro-
cess can proceed more nearly to completion during the brief power stroke
of the cycle carrying less unburned fuel into the exhaust.
Some gaseous fuels such as ammonia and hydrogen contain no carbon
at all and therefore cannot produce the ingredients necessary to form
photochemical smog. Unfortunately, they have other drawbacks that detract
from their use as automotive fuels.
Z. 1. Combustion Characteristics of Gaseous Fuels
The combustion chambers of an automotive engine can be visualized as a
group of batch reactors connected in parallel, and the exhaust system as a
flow reactor. The combustion of fuel in an engine represents a large number
of simultaneous and sequential nonsteady-state chemical reactions at widely
varying temperature and pressure condtions, with the exhaust emission as
their final reaction products. This analogy has prompted attempts at develop-
ment of mathematical models to accurately predict engine performance and
exhaust emission compositions from inputs of engine design, operating para-
meters, and fuel composition. Although these modelling efforts have not
been completely successful, they have contributed toward a better under-
standing of the engine emission phenomenon. Since a change of engine fuel
represents a drastic change in the combustion reactions, its effect on ex-
haust emission of engines is partially a reflection of the combustion char-
acteristics of the fuel involved.
2-1
-------
89Z9
2.1.1. Basic Properties
There are several basic physical and chemical properties of fuels
relevant to the combustion reaction. Some of these properties of common
g:=eoua fuels, together with those of a few liquid fuels that are representa-
tive of gasoline in molecular complexity and volatility, are presented in
Table Z.l.l.-l.
2. l.Z. Combustion Characteristics
In the internal combustion engine cycle, power and efficiency would be
maximum if complete combustion were to occur instantaneously at TDC
(top dead center) ignition timing. The power then is proportional to the
total rise in pressure that occurs at constant volume, and both power and
efficiency increase proportionally to the amount of expansion that occurs in
combustion.
In actual practice, however, a fuel-air mixture does not react instant.
aneously but exhibits a finite mass rate of burning which varies with the
nature of the fuel and with engine conditions. Thus, the kinetics of the com-
bustion process must be coupled with the timing of the cycle by advancing or
retarding the spark timing to optimize pressure rise, efficiency and power at
varying rpm. It is apparent, therefore, that the fundamental burning velo-
city of the fuel is an important parameter affecting the coupling of fuel to
engine design. Table 2.1.2. -1. shows that the burning velocity is more or
less constant for ethane and higher aLkanes.
The unsaturated hydrocarbons have higher values at low molecular
weights, but decrease rapidly as the gasoline range is approached. Benzene
and cyclohexane also have burning velocities about the same as those of the
alkanes. Therefore, the burning velocity of gasoline, which is a mixture of
alkanes, cycloalkanes, alkenes, and aromatics in the range of four to ten
carbon atoms, should be only slightly higher than the i nd-alkane lewl of
about 39-40 cm/a. Thus, the fundamental burning velocity of propane is
close enough to that of gasoline go that a minimum of engine modification
2-2
INSTITUTf •>«• -it 1f'-"1C'-~
8929
W
3
b,
U?
i
(O
u
u,
0
in
w
at
u
0.
o
S
u
2
n
i
~
rM
3
00
C
(4
co
c
O
P3
C
u
I
n
o
(d
V
O
s
«
X
c
1
1«
•o
X
g
B
0,
>
Id
o
B
o
•
1
o
U
u
d
o
o
J
k
U
i,
"3
u
1
U
|
D
U
•
o
3
2
S
^v
u
E
1
^
*—
O
(fj
Oj
u
^
s
n
1
I
rfV
rM
00
in
d
T
o
a^ S
O 0*
o
00
^"
xO
00
I-"
T
v
—
£
(J
Methane
o
nl
T
NO
CD
?
i-
0
_
e>
r4
S
o
T
*n
o
Xs
H*
U
Ethane
0-
CO
T
m
f
CO fM
O fM
in in
o -T
-^
i
00
f
•*o
rg"
5
A
u"
Propane
.,
00
in
f
J
O
-V.'
•O
2
- 2
T
m
rM
S
u
N-Butane
-M C^
o* P- r*
in o r-
T T '
«n •*
r\J f\J m
rvj -* 1
1 1
3
Of* I— 00
O O^ NO tT>
o o do
in tj< o
eC » «
•o
•W 0
O «0 -i
7 T
i i
i 0 i
. a «s e
X 0 * <
«
T *
•/> i
en ^
™ fM
rt 43 rt
^» *n 01
a* f o
t^ o r-
000
o o o
m o oo
5 £ 5
e a o
f f f
in m
«r» IM
«*l ^ (VI
B
* -. S
X
£ '• £
(j y
'. S
1 is 2
is 1 3 S
a** 5* 1
m
-H
O
GO
rt
^a-
oo
o
p"
?
g
2;
_H
00
1^
5
Benzene
o
T
u
^
T
JO
O*
oo
• 3
o
J.
JT£ \$- fct t?. fcS. fcS.
O f> in o rt
-------
8929
8929
Table Z. 1. 2. -1. FUNDAMENTAL BURNING VELOCITIES FOR HYDROCARBONS
115
Fuel
Methane
Ethane
Propane
Butane
Pentane
Hexane
Heptane
2-Methylpropane
2, 2-Dimethylpropane
2-Methylbutane
2, 2-Dimethylbutane
2, 3-Dimethylbutane
2, 2, 3-Tnmethylbutane
2-Methylpentane
3-Methylpentane
2, 3-Ehmethylpentane
2, 4-Dimethylpentane
Ethene
Propene
"1 -Bute-Tie
1-Pentene
1-Hexene
2-Methyl- -propene
2-Methyl- -butene
3 -Methyl- .butene
2. Ethyl- 1 -butene
2. Methyl- -pentene
4 -Methyl- -pentene
Propyne
1-Butyne
1-Pentyne
1-Hexyne
4-Methyl-l-pentyne
2-Butyne
3-Hexyne
Cyclohexane
Benzene
Maximum Uo,t
cm/s
84. 5
102.8
99.5
96.2
98.0
98.0
98.3
87. 5
83.0
92.5
90.0
91.7
90. 5
93.0
92.7
92.2
89.9
184. 5
113.4
1TIT5-
110.0
108. 5
95.0
99. 5
106.9
100.3
101.2
104.0
189. 1
155.0
140.0
127.0
116.9
135.6
118.0
98.4
104. 5
Maximum Uf.
cm/s
33.8
40. 1
39.0
37.9
38. 5
38. 5
38. 6
34.9
33. 3
36.6
35. 7
36. 3
35.9
36.8
36. 7
36. 5
35.7
68. 3
43.8
*JT*
42.6
42.1
37. 5
39.0
41.5
39.3
39.6
40.5
69.9
58. 1
52.9
48. 5
45.0
51.5
45.4
38.7
40.7
Volume %
* Fuel at
Maximur< Uf
9.96
6.28
4. 54
3. 52
2.92
2. 51
2. 26
3.48
2.85
2.89
2.43
2.45
2.15
2.46
2.48
2.22
2.17
7.40
5.04
3.87
3.07
2.67
3.83
3. 12
3. 11
2.65
2.80
2.62
5.86
4. 36
3. 51
2.97
2.87
4.36
3.05
2.65
3.34
Calculated from using the average values A = 5.07 Bq cm and A{ = 11.25
sq cm.
T U0 = observed linear flame velocity. Uf = fundamental flame velocity.
should be required Methane, although Us burning velocity is dislmcly
lower, should offer no major problem
The fundamental burning velocity of hydrogen is about seven times as
fast as that of propane or gasoline and efficient engine-fuel coupling might
require major engine medications A more serious difficulty is associated
with hydrogen's poor antiknock propc. rties.
Ammonia presents even more difficult problems since its burning velo-
city is only 1/38 that of propane. Coupling such a slow combustion process
to the internal combustion engine is quite difficult. The spark timing must
be advanced from the 20-40 deg ETC (before top center) required for gaso-
line to 100-120 deg ETC. Even then, complete burning is approached only
at low rpm At high speeds, complete burning can be accomplished only if
Ihe area of the flame front is increased by increasing turbulence or the com-
bustion characteristics are improved by prcmixmg with faster burning hydro-
gen
The ignition of a fuel-air mixture is dependent not only on fuel-air ratio,
but on a number of other factors as well. These include temperature, pres-
sure, concentration of inert diluents, initiating or inhibiting species, chamber
diameter, surface to volume ratio, and characteristics of the wall surfaces.
Changes in engine variables that affect any of these factors will also affect
the ignition characteristics of the system. Failure to match engine para-
meters to the ignition characteristics of the fuel can produce misfiring and
thus result in the loss of power and efficiency together with increased
emission of unreacted or partially reacted hydrocarbons and carbon mon-
oxide. A complete understanding of the ignition characteristics of a parti-
cular fuel-oxidant system, therefore, is essential to the effective coupling
of that system to the engine cycle
Autoigmtion is defined as the self-ignition of a homogeneous mixture
01 a fuel oxidant. In an int.rnal combustion engine the autoigmlion pheno-
menon associated with knock proceeds in three stages" -
2-4
2-5
-------
8929
1. Peroxide stage — alkyl peroxides are produced and reactions
are not exothermic.
2. Cool names stage — excited formaldehyde produces a blue
radiation and exothermic reactions are present.
3. Hot flame stage — reactions are rapid and highly exothermic,
producing intense radiation and an audible knock.
According to Walsh, "* knock ratings are a sensitive function of mole-
cular structure. He associates knocl< resistance with cool flame formation
and says that cool flames only arise when the net chain branching factor is
sufficiently great. Chain branching increases with the presence of CHj
groups and since neither methane nor ethane contain CH2 groups autoigm-
tion is not a problem with these fuels.
The current trend toward elimination of lead additives from gasoline
makes it both difficult and expensive to meet octane number requirements of
modern cars. The gaseous fuels exhibit a distinct advantage in regard to
antiknock performance. The octane numbers of nonleaded gasoline,
methane, propane, and ammonia are 93, 120, 97.4, and 111 respectively.
One of the major problems associated with ignition and internal combustion
engines is that of misfire limits. Misfire tends to occur either on the rich
or lean side of stoichiometric ratio, but particularly on the lean, probably
because the volumetric rate of heat release falls below that of the heat loss.
Emission of both carbon monoxide and hydrocarbons is minimized by
operation further on the lean side and this can be obtained without misfire
by use of gaseous fuels.
One major difference between gaseous hydrocarbons and gasoline is
that the latter involves droplet combustion whereas gases are combusted in
a homogeneous state. Physical processes in droplet combustion strongly
influence burning velocities so that the rate of heat release per unit volume
can be much less than that of homogeneous combustion. Also in the gasoline-
fueled engine poor mixing in the carburetor-manifold system leads to varia-
tions in cylinder to cylinder air fuel ratios which, in turn, leads to cylinder
to cylinder variations in temperature and pressure.
2-fc
T ' T II T r
8929
2. 1.3. Formation of Exhaust Pollution Products
2.1.3.1. Hydrocarbons (HC)
Z. 1. 3. 1. 1. Sources of Exhaust Hydrocarbons
There are certain fuels, such as hydrogen and ammonia, that do not
involve hydrocarbons at all in their combustion reaction. Therefore, a
discussion on exhaust hydrocarbons would be completely irrelevant to these
types of fuel. On the other hand, the exhaust hydrocarbons are closely re-
lated to the combustion of fossil fuels, notably hydrocarbon fuels.
In an ideal combustion process of hydrocarbon fuels such as gasoline, in
which thermodynamic equilibria are attained, the combustion products
should contain no hydrocarbons. The presence of hydrocarbons — either
fuel type or nonfuel type — in the exhaust is a positive indication of incom-
plete combustion.
One obvious cause of incomplete combustion in an engine is the insuffi-
cient oxygen supply which occurs at extreme engine operating conditions
such as rich fuel-air ratio, poor fuel mixing, and at light load. However, in
normal operation of a well-adjusted engine, such extreme conditions are
rarely encountered, the fuel air ratio is generally maintained close to the
stoichiometric value. Therefore, the hydrocarbons found in the exhaust of
well-adjusted engines must be contributed by other factors.
One major source of exhaust hydrocarbons is the unburned gas in the
quench zone of the combustion chamber".'"' "' 78»'** It is well known that when
flame approaches a cool surface, such as engine head, chamber walls, and
piston face — the flame temperature is drastically reduced and the flame is
thus extinguished, resulting in incomplete oxidation of fuel. This drastic
temperature reduction due to wall quenching is illustrated in Figure 2. 1. 3. -1.
The temperature difference between chamber wall and flame center, can
be so large (usually on the order of 3000*-4000'F) that fuel-air charge near
the wall may fail to ignite. Although some of the unburned gases may be
2-7
-------
8929
; AVERAGE WALL
I TEMPERATURE,Tw
cr
UJ
a.
Z
—-QUENCH ZONE —
DISTANCE FROM CYLINDER WALL*
Figure 2. 1. 3.-1. TEMPERATURE PROFILE OF QUENCH ZONE
oxidized later during the expansion and exhaust processes, a sufficient
amount remains unchanged. One investigation estimated that almost 50%
of exhaust hydrocarbons came from the quench zone and about 30% of them
entered the atmosphere.
Another major source of exhaust hydrocarbons is the unburned fuel
trapped in combustion crevices — such as the space around the piston,
space between the engine head and block, and crevices around valve seats80' lrl
(Figure 2. 1. 3. -2). Due to the narrow passage of these crevices, the flame
is prevented from propagating into the regions beyond the restriction. One
such region is the crevice below the top piston ring. It has been estimated
that, under some conditions, the piston-ring crevices alone contribute up
to 50% of the exhaust hydrocarbons.177
A complete combustion cycle of an Si-engine consists of many processes,
thus the hydrocarbons from quench zones and chamber crevices are sub-
jected to the influence of these processes before being expelled into the
N S T I T u T E
TECHNOLOGY
8929
VALVE
COMBUSTION
CHAMBER'VOLUME
Figure '2. 1. 3. -2. QUENCH SURFACE AND PISTON-RING
CREVICES OF AN SI-ENGINE COMBUSTION CHAMBER
atmosphere. The major factors are the extent of postcombustibn oxidation
of unburned gases in the combustion chamber and the extent of oxidation in
the exhaust system." Both factors are influenced by some of the engine
variables to be discussed later.
Another potential source of unburned hydrocarbons from an engine is the
blowby of unburned charge. In a conventional four-stroke cycle Si-engine,
the blowby loss is usually to the crankcase and only a small portion of un-
burned mixture is lost through the exhaust valve and during the valve over-
lap period. However, the blowby loss could be significant in a two-stroke
cycle compression ignition engine and in a rotary engine.
INSTITUTE
T F C H N O
-------
8929
2. 1. 3. 1. 2. HC Concentration and Engine Operation Variables
A positive approach of eliminating exhaust hydrocarbons is to eliminate
the sources of unburned gases. In an SI-engine, this approach would mean
the elimination of quenching and a crevices' effect during combustion. How-
ever, due to practical limitations in engine design and modification, both
the wall quenching and the crevices' effect can only be minimized not elimi-
nated. A significant reduction in exhaust hydrocarbons has been achieved by
minimizing the quench surface area and the size of piston-ring crevice"'79'1**
Further reduction in exhaust hydrocarbons may be realized by increasing
the amount of oxidation occurring in the combustion chamber and exhaust
system.
Since the mass of unburned fuel resulted from wall quenching of the flame
and the crevices'effect is directly determined by the volume and density of
fuel-air charge in the quench zones and combustion chamber crevices, it is.
therefore, under the influence of some engine variables, such as fuel-air
ratio, ignition timing, air-flow rate (percent of throttle), engine speed, and
compression ratio. However, their effect on fuel-air charge volume and on
fuel-air charge density are not similar. The volume change predominates in
two cases: fuel-air ratio and ignition timing. The density change predomi-
nates in the other three cases: air-flow rate, engine speed, and compres-
sion ratio. Furthermore, the effect of engine variables on the amount of
unburned gases due to wall quenching does not always correlate with that
due to the crevices'effect The net effect on exhaust hydrocarbons depends
on the relative magnitude of hydrocarbon contribution from these two sources.
In addition to their effect on the volume and density of fuel-air charge in-
side the combustion chamber, some of the engine variables can also affect
the temperature and residence time of the combusted gases. Therefore,
the engine variables permit further oxidation of the unburned gases to pro-
ceed inside the chamber and exhaust system prior to being ejected into the
atmosphere. For example, retarding the ignition timing increases the total
T u T E
2-10
CAS
TECHNOLOGY
8929
volume of quench gas (increasing hydrocarbons) while reducing the density
uf the crevice Kascs (reducing hydrocarbons), the net effect of ignition re-
tardation on the source of unburned hydrocarbons may be insignificant.
However, retarding the ignition timing in an Sl-enginc increases the tempera-
ture of combustion products during the expansion and exhaust cycles, thus
allowing additional oxidation and resulting in reduced exhaust hydrocarbons1.8'1'
The magnitude of the effect of spark timing on exhaust hydrocarbon con-
centration depends on the setting of other engine variables, notably fuel-air
ratio.
The fuel-air ratio affects the flame speed and the amount of excess oxy-
gen available to post-flame reactions, thus affecting the exhaust hydrocarbon
concentration. The availability of excess oxygen promotes the post-flame
reactions. However, excess air tends to reduce flame velocity, resulting in
lower peak temperature and pressure, and therefore retarding the post-
name reactions. The optimum air-fuel ratio for minimum exhaust hydro-
carbon emission varies among fuels and depends on other variables such
as combustion chamber volume and surface area and ignition timing.
The approach of reducing exhaust hydrocarbons emission through im-
proved post-flame reactions has been found to be more effective in engines
having large cylinder displacement, low compression ratio, and small bore-
, 60,99,112
stroke ratio.
Z. 1. 3. 1. 3. Composition of Exhaust Hydrocarbons
Gasoline consists of many hydrocarbons, varying among the grades sold.
More than 40 individual hydrocarbons have been identified among different
grades of gasoline.1 Therefore, it is not at all surprising to detect various
fuel hydrocarbons in the exhaust of gasoline-fueled engines knowing that un-
burned fuel is a source of exhaust hydrocarbons. One investigation, using a
gas chromatography technique, identified more than 25 various hydrocarbons
in the exhaust of gasoline-fueled automobiles, as shown in Table 2.1. 3. -l.
INSTITUTE
2-11
OF CAS
TECHNOLOGY
-------
89Z9
8929
Table 2. 1.3-1. EXHAUST HYDROCARBON COMPOSITION OF GASOLINE-
FUELED AUTOMOBILES (Average of 62 Vehicles) '"
Table 2. 1. 3-2. EXHAUST HYDROCARBON COMPOSITION OF PROPANE-
FUELED SI-ENGINE*
Hydrocarbons
Me thane
Ethylene*
Acetylene
Propylene
n- Butane
Isopentane
Toluene
Benzene
n- Pentane
m + p-Xylene
1-Butene
Ethane
2-Methylpentane
Mole %
1 A
1 O.
14.
14.
6.
5.
3.
3.
2.
i.
I.
1.
1.
1.
Total
f
5
1
3
3
7
1
4
5
9
8
8
5
Hydrocarbons
2, 2,4-Trimethylpentane
o-Xylene
Isobutane
2, 3-Dimethylpentane
2, 4-Dimethylpentane
Methylacetylene
Ethylbenzene
1 -Methyl-4-ethylbenzene
Trans-2-Butene
2, 3-Dimethylbentane
1, 3, 5-Trimethylbensene
1, 2, 4-Trimethylbenzene
84.9 moles
Mole %
1
I.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
2
0
9
8
9
3
9
7
7
6
5
4
4
Hydrocarbon
Ethane CjH^
Methane CH4
Propylene C}H6
Acetylene C2HZ
Ethylene C2H4
Propane C3H,
At air-fuel ratio
Data presented
Volume
Percent ol
f Total Hydrocarbons
Combustion Chamber Exhaust
0.6
2. 6
2.9
3.9
6.9
83. 1
of 15. 5 and ignition timing of 20
1.
11.
7.
10.
19.
48.
degBTC.
6
7
2
8
8
9
in Table 2. 1. 3-2 indicate that approximately half of the
exhaust hydrocarbons are of original fuel while the
products. However, the composition of
means constant.
varied from 16 to
The investigation with
86 % (on carbon basis)
other half are combustion
the exhaust hydrocarbons is by no
38
propane
found that the
fuel portion
of the total exhaust hydrocarbons.
Products of engine combustion.
The data presented in Table 2. 1. 3-1 indicate that roughly 60% of the
total exhaust hydrocarbons are products of engine combustion, consisting
of partially oxidized and/or cracked gasoline constituents. The remaining
40% of the identified exhaust hydrocarbons resemble the original fuel.
even in the relative composition ratio. This 40% portion represents the
unburned and unchanged fuel. These data indicate that the composition of
exhaust hydrocarbons does vary with that of the fuel, provided the differ-
ence in fuel composition is significant enough.
Laboratory investigation using 99%-pure propane in a single-cylinder
spark-ignition engine also showed the relationship between the composition
of exhaust hydrocarbon and that of fuel (Table 2. 1. 3-2),
depending upon the experimental setting of engine variables. In general.
fuel-air ratio and ignition timing exert the most significant influence on the
composition of exhaust hydrocarbons.
The composition of exhaust hydrocarbons is important in the fact that
some hydrocarbons could contribute to the manifestations of photochemical*
smog more than others.' In general, olefinic hydrocarbons (particularly
Cj, C4, Cs olefins) have much higher reactivity — smog contributing- than
paraffiruc and aromatic hydrocarbons.' ' Therefore, the absence of higher
olefins and the relatively lower C2 and Cj olefins concentration contribute to
the significantly lower reactivity of the propane-fueled engine exhaust (Table
2. 1. 3-2) as compared to that of gasoline-fueled engine exhaust (Table
2. 1. 3-1). Further discussion of hydrocarbon reactivity will be presented
in the sections dealing with various gaseous fuels.
2-12
I N S T I T U T F
TECHNOLOGY
INSTITUTE
2-13
CAS
TECHNOLOC. V
-------
89 Z9
2.1. 3.1. Nitrogen Oxides (NOX)
The exhaust of automotive engines — both spark-ignition engines and
diesel-type engines — is known to be a significant source of nitrogen oxides.
Among the many existing species of nitrogen oxides, only nitric oxide (NO)
can be found in appreciable quantities at the combustion temperatures of
fuels, as shown by both thermodynamic data of equilibrium constants of
various nitrogen oxides and actual measurement of engine exhaust. It
has also been thought that, although nitric oxide can be readily oxidized to
form nitrogen dioxide (NOj), the short time that combustion gas remains
inside an engine and its exhaust system does not allow the completion of
oxidation. Therefore, the only nitrogen oxide concerned in engine exhaust
emission is the nitric oxide.
2. 1. 3. 2. 1. Reaction M echamam of NO Formation
The exact formation mechanism of nitric oxide in a combustion engine is
not well understood. Traditionally, the formation of nitric oxide is repre-
sented by the bimolecular mechanism for nitrogen fixation,
NI + Oz = 2 NO (1)
which is an expression of conservation of mass rather than of an elementary
reaction.
Other investigators proposed the incorporation of some elementary re-
actions involving atomic species such as N and O to explain the unusually
high nitric oxide concentration found in the combustion of ammonia149and
hydrogen. Some of the suggested possibilities of an atomic reaction mech-
anism are111 —
O2 + N - NO + O (2)
N4 + O - NO + N (3)
N + O + M - NO + M (4)
INSTITUTE
2-14
OF CAS
TECHNOLOGY
8929
However, in view of their small atomic dissociation in air (Figure 2. 1. 3-3),
especially nitrogen, under engine temperatures and pressure, the contribu-
tion of atomic oxygen and nitrogen in overall nitric oxide formation in engine
combustion may be quite small. Others have suggested an atomic scheme
m
involving N2O, where the final results are identical to those of the bimolec-
ular mechanism expressed by equation (2).
O
FUEL-AIR EQUIVALENCE RATIO
(F-A RATIO/STOICHIOMETRIC F-A RATIO)
20
FUEL MICH
Figure 2. 1. J-3. COMPUTED EQUILIBRIUM COMPOSITION
FOR CONSTANT-VOLUME ADIABATIC COMBUSTION OF n -OCTANE1"
2-15
-------
8929
2. 1.3.2.2. Equilibrium Theory uf NO Formation
Numerous attempts were made to correlate the amount of nitric oxide
found in engine exhausts to that obtained through theoretical calculations.
These attempts involved studies of not only the formation mechanism at tne
combustion of fuel but also the change of nitric oxide concentration under
other conditions of an engine cycle, namely, expansion, compress-.un, and
exhaust.
Earlier studies with gasoline-fueled Si-engines' indicated that the nitric
oxide concentration in the exhaust corresponds approximately to the chemical
equilibrium concentration calculated for the peak combustion temperature.
This observation led some investigators to conclude that chemical equilibrium
is reached inside the combustion chamber. However, it was also observed
that the approximate concentration correlation exists only with a fuel-rich
mixture and not at all with a fuel-lean mixture. It was also observed that
NO concentration found in exhausts is several orders of magnitude greater
than the equilibrium concentration calculated on the basis of exhaust tempera-
ture.
The attainment of equilibrium was also suggested by another more recent
investigation" in which the concentration of nitric oxide under normal opera-
tion of a CFR engine was found equal to that under a large excess of nitric
oxide added to the carbureted air-fuel mixture. The equilibrium theory was
further strengthened by the approximate correlation of the nitric oxide con-
centration between the observed values from the exhaust of an SI-engine and
those predicted for constant-volume adiabatic combustion of various hydro-
carbon fuels (Table 2. 1. 3-3). The information presented in Table 2. 1. 3-3
indicates that the equilibrium nitric oxide concentration is closely related
to some combustion properties inherent to the fuel involved.
INSTITUTE
2-16
CAS
TECHNOLOGY
8929
Table 2. 1. 3-3.
ADIABATIC FLAME TEMPERATURES AND NITRIC
OXIDE CONCENTRATIONS140
Constant Pressure Combustion
Fuel
Methane
Ethane
Propane
it-Octane
Eth>lene
Acetjlene
* Air-fuel
Adiabatic
Flame Temp. ,
•F
3619
3691
3707
3727
3930
4305
equivalent ratio
Nitric Oxide
Concentration,
pom
1 577
.927
2030
2161
3312
6812
= 1, T = 77'F, P
Constant Volume Combustion
Adiabatic
Flame Temp. ,
•F
4346
4398
4443
4465
4675
5082
= 10 atm
Nitric Oxide
Concentration,
ppm
4, 314
5,051
5,270
5, 551
7, 597
13, 320
However, more recent investigations found that nitric oxide, once formed
124,125il56
in combustion, does not decompose during the expansion process. It was
observed that the kinetics of decomposition reactions, as expressed by
Equations (1), (2), and (3), are not rapid enough for the temperature-pressure
path imposed on the combusted gases by the engine operation. In fact, the
concentration of nitric oxide was found to reach a peak at "peak flame tem-
perature" and remain constant throughout the expansion and exhaust proc-
esses. A term of "frozen equilibrium" was used to describe this observed
phe nome non.
The attempts to predict the NO concentration in the exhaust by the equilib-
rium theory has not been completely successful. The difficulties may be
related to the invalid assumption that a single flame temperature prevails
throughout the combustion chamber, which is known to be nonexisting as
illustrated by the temperature profile in the quench zone of a combustion
chamber (Figure 2. 1. 3. -1). Also mixing and fuel distribution may be
significant factors in this phenomenon. Furthermore, the engines, in
reality, are operated unJer conditions that have neither constant volume nor
constant pressure. The extreme temperature sensitivity of equilibrium
2-17
-------
8929
concentration of nitric oxide — concentration increases exponentially with
the increase in temperature as shown in the kinetic equations (Equations
5 and 6) of nitric oxide formation (Equations 3 and 4) and Figure 2. 1. 3-4.
Experimental difficulties in temperature measurement and sampling tech-
nique further complicate the study.
where
r = k L
Ae
E_
'RT
(5)
(6)
r = reaction rate
k = specific rate constant
c. = concentration of component j
y. = order of reaction with respect to component j
A = frequency factor for the reaction
E & activation energy
R = universal gas constant
T = absolute temperature
2. 1. 3.1. 3. Kinetics of NO Formation
It has been suggested recently that the nitric oxide formation may never
reach an equilibrium in engine operation and that the equilibrium theory
may be irrelevant to the calculation of NO concentration in the exhaust. It
has been shown that equilibrium in nitric oxide formation could not be
reached in the engine due to the insufficient time. Using the reaction
mechanism represented by Equation (l), the minimum times required to
reach equilibrium NO concentration under various temperatures, while
keeping pressure, volume, and concentration constant, are shown in
Figure 2. 1.3-5.
Z-18
INSTITUTE
8929
7OOO
8 6000
5 5000
a:
l-
U 4000
u, 300°
o
O 2000
y
5 IOOO
3000 4OOO
TEMPERATURE.*F
9000
A-II3
Figure 2. 1. 3-4. TEMPERATURE VERSUS EQUILIBRIUM NITRIC OXIDE
CONCENTRATION '"
(At 600 psi and fuel-air equivalence ratio of 1:1)
075
050
0.25
Figure 2. 1. 3. -5. KINETIC PROGRESSION OF NITRIC OXIDE FORMATION
UNDER VARIOUS TEMPERATURES"' "
2-19
-------
8929
The data presented in Figure Z. 1. 3-5 indicate that the attainment of
90% of equilibrium nitric oxide concentration at 3000"K will require 4
milliseconds - which may well exceed the total period of combustion of ar
Si-engine operating a! 3000 rpm. Furthermore, as combustion proc-eds,
temperature and concentration vary appreciably, and in all cases su'licient
time is not likely to ba available to attain the equilibrium concentration
values. The time scale of events in an operational engine is illustrated in
• Figure 2. 1. 3-6. which shows that the completion of the power stroke -
which includes the combustion of fuel and expansion of burned gas - requires
a time duration of approximately 8 milliseconds at 3000 rpm.
ZOO
1800 240O SOOO 360O
RPM
600 1200
Figure 2. 1. 3-6. DURATION OF ENGINE EVENTS FOR A PRODUCTION
318 CID V-8 ENGINE «
The time-dependent nature of the nitric oxide concentration is further
illustrated in the combustion of hydrogen and air (Figure 2. 1. 3-7).
Z-20
N S T I T U T F
8929
Figure 2. 1. 3-7. RATE OF NITRIC OXIDE FORMATION IN HIGH PRESSURE
HYDROGEN-AIR FLAME '"
-------
8929
for the peak cycle temperature. However, the attainment of equilibrium
nitric oxide concentration on a local basis cannot be ruled out.
2. 1.'3. 2. 4. NO Concentration and Engine-Operation Variables
Nitric oxide concentrations in engine exhaust are affected by some oper-
ating parameters in both spark-ignition and compression ignition engines.79
For Si-engines, the major parameters are fuel-air equivalence ratio and
spark timing. Manifold air pressure, compression ratio, and engine speed
are also important. The fuel-air ratio and spark timing are particularly
important to nitric oxide concentration.
The fuel-air ratio governs the amount of heat that can be released by the
fuel and the rate at which the enthalpy of fuel is being released during com-
bustion. Thus, the fuel-air ratio is closely related to the peak temperature
and pressure that can be attained by a specific fuel in a given combustion
chamber.
Since the enthalpy value that can be released during combustion (heat of
combustion) is an inherent property of a fuel, the effects of the fuel-air
ratio on'peak combustion temperature and pressure vary among different
fuels (Table 2. 1. 1. -1). The effects can be illustrated by the theoretical
peak Otto-cycle temperature and pressure of various types of fuel, assum-
ing that flame speed (heat-release rate) is not affected (Figures 2. 1. 3. -8
and 2.1.3. -9).
Due to its temperature and pressure dependent nature, the nitric oxide
formation in an engine is, therefore, closely related to the fuel-air ratio.
Such relationships can be indicated by the effect of the fuel-air ratio on the
calculated equilibrium peak cycle concentration of nitric oxide for various
types of fuel, assuming, of course, that the equilibrium in nitric oxide
formation is attained in the engine combustion (Figure 2. 1. 3-10).
2-22
N <, T I
8929
2.500
08 09
FUEL
L£*N #- EQUIVALENCE RATIO
ISO-OCTANE
PBOmNE
BBS?-
METHAMX
REFORMED
HEUANC
AMMONIA
14
FUEL
RICH
Figure 2. 1.3-8. THEORETICAL PEAK OTTO CYCLE TEMPERATURE
OF VARIOUS FUELS «»
so •
- EQUIVALENCE RATIO
Figure 2. 1. 3-9 . THEORETICAL PEAK OTTO CYCLE PRESSURE
OF VARIOUS FUELS'5'
2-23
-------
J O.OO3 -
8929
« EQUIVALENCE RATIO • t
Figure 2. 1. 3-1Q. PREDICTED EQUILIBRIUM PEAK CONCENTRATION
OF NITRIC OXIDE15'
It should be remembered, however, the nitric oxide formation is also
affected by other engine combustion variables, notably flame speed and
residence time in the combustion chamber (determined by ignition timing
and other engine variables), the actual nitric oxide concentration in the
engine exhaust is not exactly represented by the calculated equilibrium con-
centrations. A typical plot of nitric oxide concentration in the engine ex-
haust against air-fuel ratio of a gasoline-fueled standard spark-ignition
automobile engine is shown in Figure 2. 1. 3-11. in which the effect of
spark timing is also indicated.
2-24
SZOO
48OO
4400
4000
3600
3200
2800
"00
tdoo
ieoo
1200
eoo
400
o
- KE r
O TOC
6iO! BTOC
D 20' BTOC
tf 30* BTOC
_ O 40'BTOC
v-8 ENGINE
4*Hg INTAKE DEPRESSION
JiOO RPM
I I I I
8929
II 14 15 16
ENGINE AIR-FUEL RATIO
Figure 2. 1. 3-11. NITRIC OXIDE CONCENTRATION VS. AIR-FUEL RATIO
IN A STANDARD GASOLINE-FUELED SI-ENGINE"
In a compression-ignition engine, the fuel-air ratio also has a predomi-
nant effect on the nitric oxide concentration but in a different way compared
to that in an Si-engine. Since the fuel-air ratio in a Cl-engine generally
increases with the increase in load, so does the nitric oxide concentration.
The exact point of fuel-air ratio where peak NO concentration occurs de-
pends on some engine characteristics.
In a prechamber type Cl-engine, NO concentrations rarely exceed a value
of 1000 ppm, and the peak concentration generally occurs near the middle of
931 98
full torque value. On the other hand, high NO concentrations exceeding 2500
ppm have been recorded in the exhaust of direct fuel-injection type of engines,
in which the fuel injection timing becomes a significant factor — comparable
to the effect of spark timing in an Si-engine. A general comparison of NO
concentration in the exhaust of the two types of Cl-engines is presented in
Figure 2. 1. 3-12.
2-25
-------
8929
ZOOO
I9OO
IOOO
800
DIRECT-INJECTION
ENGINE
x' PRECMAMBER
ENGINE
02 04 06
FUEL-AIR EQUIVALENCE RATIO
08
Figure 2. 1. 3-12. COMPARJSON OF EXHAUST NO CONCENTRATIONS
OF TWO TYPES OF Cl-ENGINES **
The reduced exhaust nitric oxide concentration of Cl-engines compared
to that of Si-engines has been attributed to the high heterogeneity of fuel in
106, IM
the combustion chamber. The nitric oxide emission was reduced by the
locally fuel over-rich conditions in the combustion zones. The prechamber
type engines have greater fuel-air heterogeneity than the direct fuel-injection
type engines, thus lower NO.
The effect of fuel-air heterogeneity on NO emission can also be seen in
the combustion of a stratified Si-engine. A stratified charge engine, such
as the Baudry engine, brings a fuel-rich mixture to the vicinity of the
spark plug while the normal intake manifold is fed either with pure air or
with a fuel-lean mixture. Thus, a high heterogeneity in fuel distribution is
achieved. Furthermore, the stratified charge engine meets the load vari-
ations by adjusting the strength of the fuel-rich mixture instead of using
throttle adjustment as in the case of normal Sl-engincs. Therefore, the
fuel heterogeneity is maintained, resulting in lower nitric oxide emission
2-26
8929
over a wide range of load, even though having a higher localized peak NO
emission (Figure 2. 1. 3-13).
400O
3000 •
20OO -
IOOO -
i
4 6 B 10 12
BRAKE HORSEPOWER
A-II6
Figure 2. 1. 3-13. COMPARISON OF NO EMISSION OF STRATIFIED
CHARGED AND NORMAL SI-ENGINES15
The engine operation variables affect the nitric oxide emission through
their effects on the temperature-pressure-Ume diagram of the combustion
process. These variables and their effect on exhaust emission in different
types of engines using various types of fuels will be discussed in more de-
tail in later parts of this report.
2. 1. 3. 3. Carbon Monoxide (CO)
2. 1. 3. 3. 1. Sources of Exhaust Carbon Monoxide
It is generally accepted that carbon monoxide in an engine exhaust,
similar to nitric oxide, originates in the high-temperature region associ-
ated with the flame in the combustion chamber. In a fuel-rich mixture,
the formation of carbon monoxide is largely due to an insufficient supply
of oxygen for the complete oxidation of fuel. This is believed to be true
BO
regardless of the homogeneity of the fuel-air mixture. However, in a
2-27
-------
8929
fuel-lean mixture* the controlling mechanism of the carbon monoxide
formation ib nut afa clear, and the homu^cncity (if the fuel-air mixture may
become bigmficanl.
It has been observed that carbon monoxide concentration in the exhaust
of an Si-engine nearly corresponds to the equilibrium concentrate- pre-
dicted on the basis of peak cycle temperature, but does not decrease ac-
cording to equilibria as exhaust gas temperature decreases during expan-
sion. At the end of the expansion cycle, the exhaust concentration is much
greater than that predicted on the basis of equilibrium kinetics (Figure
2.1.3-14).
The reaction mechanism believed to be responsible for carbon monoxide
oxidation can be represented by the elementary reaction:
CO + OH ^ CO2 + H
(I)
This reaction has been shown to be fast enough to reach equilibrium in the
post-flame gases'?1 Therefore, the kinetics of the reaction are dependent
on the concentration ol various constituents. It is also known that recom-
binations of hydrogen
H + H = H2
H + OH = HjO
(2)
(3)
are let-molecular reactions which do not proceed rapidly enough to
reach equilibrium in post-flame gases. Thus, there is an excess of
atomic hydrogen, and the hyperequilibnum concentrations of atomic
hydrogen tend to shift the carbon monoxide oxidation toward the left,
resulting in large concentrations of carbon monoxide in the exhaust.
2-28
N S T I T U T E
8929
100,000
50,000
60,000
40,000
Predicted Equilibrium
Concentration
Measured Exhaust
Concentration
Fuel-Air Equivalence
Ratio
0.2 0.4 0.6 0.8
Extent of Expansion
1.0
Figure 2. 1. 3- 14 . CARBON MONOXIDE CONCENTRATION DURING
EXPANSION CYCLE- THEORETICAL VS. MEASURED
EXHAUST LEVELS154
2-29
-------
8929
1. 1. 3. *. t. CO dim rnlrulmn anil Knijiiu' Variableb
The- unly encn": ojioralinu vanalili lli.il hat. .1 si|;nifi> anl rffcil on the
exhaust concentration of carbon monoxide is the air-fuel ratio. Others,
such as ignition timing, throttle position, engine speed, air-fuel charge,
and coolant temperature, have little or no influence on the exhaust carbon
monoxide emission?0'"* Some engine design variables, on the other hand,
may affect the carbon monoxide emission through their effect of air-fuel
ratios. In gasoline-fueled engines, the fuel-air mixture, vaporization
and distribution are known to affect the carbon monoxide emission. Such
problems, however, are minimized when gaseous fuels are used. With
gaseous fuels, the fuel-air mixing will be much more efficient, and there
will be no vaporization and distribution problems.
The high carbon monoxide emissions are primarily associated with
fuel-rich operations. As the air-fuel ratios are varied from fuel-rich to
stoichiometric, the carbon monoxide emission decreases rapidly. Once
the air-fuel ratio reaches stoichiometric or leaner, very little carbon mon-
oxide will be produced. A typical air-fuel ratio versus a carbon monoxide
emission relationship is shown in Figure 2.1. 3- 15.
2-30
N S I I T U T
8929
c
o
O
o
12
10
fuel
rich
0.8 0.9 1.0 1.1 1.2
Air-Fuel Equivalence Ratio
1. 3
fuel
lean
Figure 2. 1. 3-15 . AIR-FUEL, RATIO VS. CARBON MONOXIDE
EMISSION »*
2-31
-------
89Z9
Z. Z. Exhaust Emission Characteristics in Si-Engines
Z. 2. I. General Discussion
A realistic comparison of the exhaust emission control potential of var--
ous types of fuels can only be made on an overall basis using the type of
engines and operating conditions that are designed lor the specific fuel in-
volved. Unfortunately, such a comparison is not possible because most of
the available automotive engines are designed to operate on gasoline or
diesel fuel, and are usually designed for maximum power output rather than
optimum exhaust emission control. Therefore, emission studies on gaseous
fuels using engines designed for gasoline may not reveal the true emission-
control potential of the gaseous fuel involved.
However, if the interest is in the gaseous fuel's potential to reduce ex-
haust emissions of existing automotive engines, then emission studies have
to be made on the simple basis of switching the fuel on the currently avail-
able engines. The attentions are then focused on finding the optimum oper-
ating conditions of the engine and necessary modifications required to
achieve the full potential in the exhaust emission reduction of suitable gas-
eous fuels. Since most of the existing automotive engines are spark-ignition
types, emphasis is thus placed on the engine factors relevant to this type of
engine. However, other types of engines that are used in increasing num-
bers in vehicles are also discussed and presented under separate sections.
Z. Z. Z. Determining Factors of Exhaust Emission
The level and composition of the exhaust emissions of an engine are known
to be influenced by many engine variables. These variables can be divided,
according to the magnitude of their effect on emission, into two groups:
Primary Variables
a Type of Engine
• Type of Fuel
Z-3Z
8929
• Operation Variables
Fuel-Air Ratio
Ignition-Timing Schedule
Engine Mode or Duty Schedule
Secondary Variables
• Fuel-Air Charge Distribution
• Fuel-Air Charge Temperature
• Compression Ratio
• Engine Deposit
• Ambient Conditions
• Ignition System Design
• Fuel Composition
In all emission studies, the type of engine and fuel are usually fixed.
Thus the primary variables are engine operation variables - fuel-air ratio,
ignition timing, and engine mode. In the case of laboratory daw. using re-
search engines, the engine mode is usually expressed in terms of engine
speed and power output. Furthermore, the power output of the engine mode
ma> be expressed in various ways depending upon the type of engine involved.
With CFR or other research engines, the power output may be expressed in
terms of throttle position which is usually defined as the ratio of actual air-
flow rate to the maximum attainable airflow rate at wide open throttle (WOT).
In the case of a production engine, the power levels may be expressed in
terms of brake horsepower or percent of maximum power.
The effect of these variables on the emission characteristics of various
gaseous fuels will be discussed on the basis of available information from
laboratory investigations in which emission studies are made by isolating the
variables. Exhaust emission characteristics based on a road test or a
dvnamometer driving-cycle test to simulate traffic patterns will be dis-
cussed under separate sections.
2-33
-------
8929
2. -. 5. Emission Characteristics of Propane
2.1. 3. 1. Fuel-Air Ratio
Ir. general, the exhaust hydrocarbon concentration of a propane-fueled
engine decreases from both extremes of fuel-rich and fuel-lean operations
and reaches a minimum at a fuel-air ratio slightly leaner than stoichio-
mei.-ic. The exact location of the minimum and the magnitude of hydrocar-
bon concentration are affected by other variables, notably ignition timing
anc throttle opening (Figures 1.1. 3. -1 and i. 1. 3. -Z).
4OO
300
ZOO
i
g 100
PROPANE ,99 X pun
20 d*8 BTC
Z d*« BTC
I
07 08
FUEL
RICH
09 10 II 12 13 14 19
AIR- FUEL EQUIVALENCE RATIO
16
17
FUEL
LEAN
A-II5
Figure Z.Z.3.-1. EFFECT OF AIR-FUEL. RATIO ON EXHAUST HYDROCARBON
CONCENTRATION OF PROPANE-FUELED SI-ENGINE18
(Single-Cylinder Engine, Speed 1360 rpm. Compression Ratio 9:1, Air-Flow
Rate 8. i SCFM, Inlet Charge Temperature 135*F. Hydrocarbon Analysis
by Gas Chromatography)
2-34
8929
«OOO|
u
E
3OOO -
2000 -
1000 -
07 08 O9 10 ii 12 13 14 IS 16 IT
FUEL RICH AIR-FUEL EQUIVALENCE RATIO FUELLEAN
A-1129
Figure 2.Z.3.-2. EFFECT OF AIR-FUEL RATIO ON EXHAUST HYDROCARBON
CONC ENTRATION M
(CFR Engine, Engine Speed 1000 rpm. Compression Ratio 8:1, Ignition
Timing 30 Deg. BTC, Inlet Charge Temperature 160*F, Hydrocarbon
Analysis by FID)
We observe a substantial difference in the magnitude of the exhaust hydro-
carbon concentrations between the two sets of research-engine data presented
in Figures 2. 2. 3. -1 and 2. 2, 3. -2. The difference may be attributed to the
different settings of other engine variables — such as airflow rate, engine
speed, and compression ratio — and the difference in fuel composition as
well as in the analytic method used in the investigation.
The composition of exhaust hydrocarbons of a propane-fueled SI-engine
has not been found to vary with the air-fuel ratio."'" The major components
include propane, ethylene, methane, and acetylene (Table 2. 1. 3. -1 ). The
concentration of propane in the exhaust tends to decrease from fuel-rich to
2-35
-------
8929
stoichiometric and then increase as more air is added. The amount of
methane and acetylene decreases rapidly from fuel-rich to stoichiometric
and remains low. EthyJene and propylcne content also decreases from fuel
rich to stoichiometric but rises slightly in fuel-lean operations. Such tr< ids
were found to be true at various settings of ignition timing.
The air-fuel ratio also has a significant bearing on the exhaust concentra-
tion of carbon monoxide. As in the case of gasoline, the carbon monoxide
concentration decreases rapidly from fuel rich to stoichiometric and re-
mains low as more air is added (Figure i. 2.3 .3)
ZOO
o
O 100
z
o
2
I
5 zo
u
0
I
\
\
I
«
0
2
•
— a
j
ul
a
w
1 i
i
i
i
i
i i
i
i
i
h I
1 a
, J
1 »-
1 —
1 »
I 3
\ u
1 K
1 Ik
'r. |i
GASOLINE
PROPANE.
99% pure
1 |
08 09 i0 ll 12 IS 14
AIR-FUEL EQUIVALENCE RATIO
Figure 2. 2. 3-4 EFFECT OF AIR-FUEL RATIO ON EXHAUST CARBON
MONOXIDE CONCENTRATION ' s
(CFR Engine, Engine Speed 1000 rpm. Compression Ratio 8:1, Inlet
Temperature 125*F, Spark Timing 5 deg ETC, 30% Throttle Setting)
The fuel-air ratio is also the most significant single variable affecting
the exhaust nitric oxide concentration in propane-fueled engines. In general,
the nitric oxide concentration increases from both fuel-rich and fuel-lean
settings and reaches a maximum at a fuel-air ratio slightly leaner than
stoichiometric, which follows a. similar pattern to that of gasoline (Figure
2.2.1.-5)
2-37
-------
8929
7000
6000 —
0.7 0.8 0.9 1-0 I.I 1.2 1.3 1.4 1.3 1.6
AIR-FUEL EQUIVALENCE RATIO
Figure 2,2.3..-6 EFFECT OF AIR-FUEL. RATIO ON EXHAUST NITRIC
OXIDE CONCENTRATION OF PROPANE 56
(CFR Engine, Engine Speed 1000 rpm, Compression Ratio 8:1 Ignition
Timing 30 deg BTC, Inlet Charge Temperature 160°I',Nitrogen Oxides
Analysis by NDIR )
Other-engine variables such as ignition timing, throttle opening, and engine
speed, also have a significant effect on nitric oxide concentration.
2. Z. 3. 2. Ignition Timing and Engine Speed
As in gasoline-fueled engines, the concentration of exhaust hydrocarbons
of propane-fueled Si-engines decreases at all air-fuel ratios, as the ignition
timing is retarded (Figure 2.2.3.-6). The effect of ignition timing retardation
on hydrocarbon emission is most pronounced at low engine speed. Increas-
ing engine speed generally decreases hydrocarbon emission.
2-38
I MS T I TU TE
8929
o- 5.000
Z
O
(D
-------
8929
4.OOO
o
E
3.000
4
U
o
g 2.000
I/I
3
Z t ,000
I I
-10 -SO S lO IS
IGNITION TIMING, ce« BTC i
20
Figure 2.Z.3.-7. EFFECT OF IGNITION TIMING ON HYDROCARBON
EMISSION OF PROPANE-FUELED ENGINE AT IDLING SPEED "
(Production V-B Engine Maintained at Road Load and at Each A-F
Ratio by Varying Throttle Position)
At constant power output and ignition timing, increasing engine speed in-
creases nitric oxide emission slightly. The increase is more pronounced
with lean mixture ratios and is believed to be caused by the change in throt-
tle aetting.
INSTITUTE
2-40
s » s
TECHNOLOGY
8929
l.OOO
I
(A
2 ',000
10 o n n w n «o « M » 10 u ro » w » «o 45 x> »
IGNITION TIMING. de« BTC
• nit
Figure2.Z.3.-8.EFFECT OF IGNITION TIMING ON NITRIC OXIDE EMISSION
OF PROPANE-FUELED ENGINE AT CONSTANT POWER AND VARIOUS
AIR-FUEL RATIOS **
(Production V-8 Engine Maintained at Road Load and at Each A-F Ratio
by Varying Throttle Position)
2. 2. 3. 3. Throttle Position
In addition to the air-fuel ratio and ignition timing, another engine vari-
able that is closely related to the exhaust emission level as well as engine
power output at an Si-engine is the throttle position which determines the
maximum power that can be extracted from the fuel.
However, the actual power obtainable is dependent on the fuel-air ratio
of the mixture and igi.ition timing. Spark ignitions that are too early or too
late yield less than the maximum power obtainable. And the maximum-
power ignition, in turn, is a function of virtually every other engine
2-41
-------
8929
8929
parameter, both fixed and variable. Therefore, if some of the engine
parameters — auch as engine speed and fuel-air charge temperature — are
fixed, the power output of an Si-engine may be determined by the combina-
tion of the three major engine variables — fuel-air ratio, ignition timing,
and throttle position. The throttle posibon-fuel-air ratio-power output
relationship of a propane-fueled CFR engine is illustrated in Figure 2. 2J. .9
in which the power output is expressed in terms of percent of maximum in-
dicated mean effective pressure (1MEP) obtainable with gasoline under test
conditions- (Maximum 1MEP of gasoline is 133 calculated on the basis of
combustion chamber pressure versus time relationship over the compres-
sion and expansion portions of the engine cycle. )
08 09 10 II IZ 13 14 19
AIR-FUEL EQUIVALENCE RATIO
Figure 2.2.3.-9.POWER CONTOURS OF PROPANE-FUELED CFR ENGINE "
(At Maximum Power Ignition Timing, Engine Speed 1000 rpm. Compres-
sion Ratio 8:1)
It is indicated in Figure2.2.3.-9that various combinations of air-fuel
ratios and throttle position will yield the same power level. For example,
at the maximum-power ignition timing, the propane-fueled CFR engine will
yield 40% IMEP at air-fuel equivalence ratio of 1.0 and 40% throttle.
The same power level can be produced at an air-fuel equivalence ratio of
1. 25 and 55% throttle, or at 1. 375 air-fuel equivalence ratio and 60% throttle.
The range of air-fuel ratio- throttle position combinations is, as indicated
by Figure2.2.3.-9 widest at low and medium power levels, and becomes
narrower as power level is increased. The power contours presented in
Figure 2.2.3.-9 a re established under certain given conditions. If these con-
ditions, such as engine speed, compression ratio, and ignition timing, are
changed, the shape of the power contours as well as the range of misfire
limit will be changed also.
Since the exhaust emission level of an Si-engine is very much influenced
by air-fuel ratio, it is obvious that various combinations of air-fuel ratios
and throttle position will yield different emission concentrations. Further-
more, power contours for minimum hydrocarbon emission are different
from those of minimum nitric oxide emission. An example of such a plot
relating power levels to emission levels is presented in Figure 2. 2.3.-10
in which emission-power level relationships for minimum HC and NO of
both gasoline (premium grade, leaded) and propane are presented.
The data presented in Figure 2.2.3.-10 are based on maximum-power igni-
tion timing for power levels greater than 25% IMEP (30% in the case of gas-
oline) and 5" ETC retarded ignition timing for power levels below 25% IMEP.
The nitric oxide emission levels are obtained on the basis of the leanest possible
air-fuel ratios for minimizing specific fuel consumption (not closer than five
percentage points to the lean misfire limit). The hydrocarbon emission
levels are obtained under the conditions selected for the nitric oxide emis-
sion, and are found to be quite close to that obtained on the basis of air-fuel
ratios yielding the lowest hydrocarbon emission under the given test conditions.
Therefore, neither the nitric oxide nor the hydrocarbon curves represent the
lowest emission level obtainable at a given power level.
2-42
2-43
-------
8929
s
X 6
U
GASOLINE (L«03ed)
PROPONE 99 % our.
10 20
RETARDED
SPARK
:K> 4O 50
M*i POWER
SPARK
7O 80 90 IOO
. IMEP < HI4
Figure 2.2.3.-10. EXHAUST EMISSION VERSUS POWER LEVEL FOR
PROPANE-FUELED CFR ENGINE "
(CFR Engine, Engine Speed 1000 rpm, Compression Ratio 8:1.
Inlet Charge Temperature 125*F)
As shown in Figure 2.2.3.-10 the highest nitric oxide emission of propane,
13 g/ihp-hr, occurs at 90% IMEP, and is obtained, according to informa-
tion revealed by the reference, on the basis of leanest air-fuel equivalence
ratio (1. 11) capable of producing 90% IMEP at 100% throttle. However, the
same power level can also be produced, according to Figure 2. 2.3.-9 by
other combinations of air-fuel ratio and throttle setting, such as 0.95 air-
fuel equivalence ratio and 95% throttle which would yield a nitric oxide level
of only 5. 2 g/ihp-hr.
2-44
INSTITUTE
8929
The reduction of nitric oxide as a result of fuel-rich operation has been
clearly indicated in Figure 2. 2.3.-5 which also shows that the effect of throt-
tle position on nitric oxide emission is much less pronounced when the air-
fuel ratios are less than stoichiometric. As the air-fuel ratio becomes
leaner than stoichiometric, increasing the throttle opening is accompanied
by a significant increase in nitric oxide emission.
The slightly fuel-rich operation at high power levels has another advantage
in ignition retardation. It has been shown that retarding the ignition timing
reduces nitric oxide emission. However, retarding the ignition timing from
that of maximum power, reduces the engine power output. The amount of
power reduction resulting from ignition retardation is much less at air-fuel
ratios ncher than stoichiometric.(See Section 2. 4. 1. ) Therefore, further
reduction in nitric oxide emission can be realized by retarding the ignition
timing.
However, enriching the fuel-air charge at high-power operation increases
the hydrocarbon and carbon monoxide emissions. The amount of increase in
emission is proportional to the amount of fuel enrichment. The carbon mon-
oxide emission, in particular, will be increased significantly if the fuel-air
ratio is made richer than stoichiometric. Therefore, a compromise may
be necessary if all three pollutants are to be kept within certain allowable
limits. Furthermore, power output and fuel consumption also have to be
considered and a compromise in engine performance may also be necessary.
Data presented in Figure 2.2.3.-10 also indicate that if the engine is oper-
ated in medium and low power levels, the pollution emission level can be
kept low. Therefore, one way to reduce emissions is to use a power-
ful engine and operate it well below its maximum power output.
2-45
-------
8929
2.2.3.4. Compression Ratio
Increasing the compression ratio of an Si-engine increases the surface-
to-volume ratio of the combustion chamber and the cylinder pressure.
Theoretical calculations indicate that at compression ratios greater than Si
the exhaust hydrocarbon concentration of a propane-fueled engine increases
with increasing combustion ratioi and the effect of compression ratio is
more pronounced in engines with small cylinder volume.1" Experimental
results using a research engine also indicate that increasing the compression
ratio from 7 to 11 increases the exhaust hydrocarbon concentration by 67%."
INSTITUTE
2-46
CAS
TECHNOLOGY
8929
I. 2.4. Emission Characteristics of Methane
2. 2. 4. 1. Air-Fuel Ratio
The effects of air-fuel ratio on exhaust hydrocarbons and carbon mon-
oxide concentrations of methane-fueled engines follow a similar pattern to
that of a propane-fueled engine. The hydrocarbon emissions decrease from
both fuel-rich and fuel-lean settings and reach a minimum at air-fuel ratios
slightly leaner than stoichiometric (air/methane = 16.9). The exact magni-
tude of hydrocarbon emission at each air-fuel ratio depends on the setting of
other engine variables, notably ignition timing and throttle position.
As in the propane-fueled engines, the exhaust hydrocarbons of methane
or natural gas powered engines have been found to consist primarily of un-
burned fuel.'5'107 The olefinic hydrocarbon content in the exhaust is consider-
ably lower than that of gasoline-fueled engines. The magnitude and com-
position of the exhaust hydrocarbons are influenced by the composition of
the fuel— whether it is pure methane, liquefied natural gas (L.NG), or pipe-
line natural gas of certain compositions1.07
In a methane-fueled engine, the air-fuel ratio is also the single most im-
portant factor in determining the exhaust concentration of carbon monoxide.
The carbon monoxide emission is associated primarily with fuel-rich engine
operations and decreases very rapidly as the air-fuel ratio nears stoichio-
metric. Once the air-fuel ratios are leaner than stoichiometric. very little
carbon monoxide emissions are observed. Once the air-fuel ratio is fixed,
other engine variables have a relatively insignificant effect on carbon mon-
oxide emission. The effect of air-fuel ratio on hydrocarbon and carbon
monoxide emissions of a methane-fueled Si-engine is illustrated in Figure
2. 2.4-1 in which natural gas rather than pure methane was used and air-
flow rate (throttle position) was varied.
The effect of air-fuel ratio on the exhaust nitrogen oxides concentration
of a methane.fueled Si-engine also follow a similar pattern to that of a
propane-fueled engine, increasing from both fuel-lean and fuel-rich settings
2-47
-------
8929
8929
a
o
Airflow, 0.291b/mm
Airflow, 0. SO Ib/min
B
|
s
o
Itf
I
o
0.7 0.8 0.9 1.0 1.1 l.Z 1.3 1.4 1.5 1.6 1.7
Air-Fuel Equivalence Ratio
Figure 2. 2.4.-1.EFFECT OF AIR-FUEL RATIO ON EXHAUST CARBON
MONOXIDE AND HYDROCARBON CONCENTRATION OF
NATURAL GAS-FUELED ENGINE*4
(CFR Engine, Engine Speed 1000 rpm, Compression Ratio 8:1.Ignition
Timing 30 deg BTC, Inlet Charge Temperature 160* F, - Hydrocarbon
Analysis by FID)
and reaching a maximum at an air-fuel ratio slightly leaner than stoichio-
metric. The exact magnitude of nitrogen oxide emissions at various air-
fuel ratios is affected, similar to propane and gasoline cases by
the settings of other engine variables, noticeably ignition tuning and throttle
opening. The effect of air-fuel ratios on nitrogen oxides emission of a
methane-fueled engine is illustrated in Figure 2. 2.4.-2. Again, natural
gas instead of pure methane was used.
5000
I
. 4000
e
v
£? 2000
Airflow, 0.29 Ib/rr
Airflow, 0. 50 lb/rr|in
0.9 1.0 1.1 1.2 I.} 1.4 l.s
Air-Fuel Equivalence Ratio
Figure 2. 2.4.-2. EFFECT OF AIR-FUEL RATIO ON EXHAUST NITROGEN
OXIDES CONCENTRAION OF NATURAL GAS-FUELED ENGINE5'
(CFR Engine, Engine Speed 1000 rpm. Compression Ratio 8:1, Ignition
Timing 30 deg BTC, Inlet Charge Temperature 160*F,Nitrogen Oxides
Anal/sis by NDIR)
2.2.4.2. Throttle Position
The effect of throttle position on exhaust emissions of a methane-fueled
Si-engine is also very similar to that of a propane-fueled one. In general.
increasing the throttle opening of an engine increases the fuel consumption
and power output, thus affecting the exhaust emissions. The throttle posi-
tion has a very significant effect on the level of nitric oxide emissions,
particularly at air-fue> ratios leaner than stoichiometric (Figure 2.2.4.2).
2-48
2-49
INSTITUTE
-------
8929
Increasing the throttle opening increases the filling of the engine cylinder
with an air-fuel mixture and, therefore, the nitric oxide emission increases.
At the same time,opening the throttle increases the total amount of air and
improves the fuel combustion, thus reducing the hydrocarbon emissions.
Since the amount of fuel indited into the engine is also increased as a re-
sult of increasing the throttle opening at constant air-fuel ratios, the reduc-
tion of hydrocarbon emission is not as pronounced as the increase in nitric
oxide emission.19 The carbon monoxide emission, on the contrary, is not
significantly affected by throttle position even at air-fuel ratios much richer
It has been indicated that the low exhaust emission of an SI-engine i«, associ-
ated with engine operation at air-fuel ratios leaner than stoichiometric.
Therefore, the potential of an engine fuel to reduce exhaust emission de-
pends on its ability to burn in lean mixtures. Both methane and propane
have shown that they are superior to gasoline in this important respect.
In addition to the lower overall emission levels than that of gasoline, both
methane and propane can be operated leaner than gasoline at any given
throttle position without misfire (Figure 2.2.4.-3). Furthermore, both
methane and propane, as indicated :n Figure 2. 2.4. -3 can be operated at
leaner air-fuel ratios and at lower throttle settings than is possible with
gasoline. These features enable an engine to be operated at lower load
without producing significant carbon monoxide emission — a feature of sig-
nificant importance in normal urban traffic and indoor vehicular engine
applications.
2. 2. 4. 3. Ignition Timing
Although detailed information is not available, it is believed that ignition
timing has a significant effect on the exhaust emissions of a methane - fueled
Si-engine. Furthermore, it is believed that the relationship between igni-
tion timing and exhaust emissions of a methane-fueled engine is quite similar
to that of a propane-fueled engine. Therefore, it can be surmised that
2-50
8929
60
so
40
10
ZO
10
Casoli
Methane
Propane
0.7 0.1 0.9 1.0 I.I I.Z 1.3 1.4 1.5
Air-Fuel Equivalence Ratio
Figure 2.2.4-3. COMPARISON OF LEAN MISFIRE LIMITS"
(CFR Engine. Engine Speed 1000 rpm. Compression Ratio
8:1, Maximum Power Ignition Timing, Inlet Charge Tem-
perature, 12 5* F)
exhaust emissions of methane-fueled engines can be substantially reduced
by retarding the ignition timing but at a sacrifice of engine performance.
Due to the lack of detailed information on the effect of engine variables
on exhaust emissions it is not possible at this time to make an accurate
comparison of the emission-reduction potentials of methane and propane.
However, available information does indicate that these two gaseous fuels
are quite similar in many respects, and methane is slightly better in terms
of engine exhaust emissions — producing lower levels of carbon monoxide,
hydrocarbons, and nitric oxide under identical conditions. 56'"
2-51
-------
89Z9
2. 2. 5. Emission Characteristics of Ammonia
2. 2. 5. 1. Combustion ot Ammonia
The stoichiometric equation of ammonia combustion can be represente-
by:
4 NHj * 3OZ - 6 HjO + 2 Nj
With the absence of carbon in us structure, the combustion of ammonia
yields no carbon monoxide and hydrocarbons. Oxides of nitrogen are the
only combustion products considered to be harmful. However, in an engine
fueled with ammonia, unburned ammonia is also exhausted. This could
present a pollution problem.
In contract to theoretical predictions based on the equilibrium concentration
at peak cycle temperatures and pressures, ammonia has been found to pro-
duce more nitrogen oxides than gasoline,"'10Tas shown in Figures 2.2.5.-1
and 2.2.5-2.
si "_
Z
a
h
II
I
«
H
u
5"
a
«
u
O
2
H
11 -
x
- I
a
• o
B
O
U
•S
8
0.9 1.0 1.1 1.2 I.J
Air-Fuel Equivalence Ratio
Figure 2.2.5-1. THEORETICAL PEAK CYCLE NITRIC OXIDE
CONCENTRATION OF AMMONIA COMBUSTION145
2-S2
I N 5 T I T U
8929
§
S2
e
Ammonia
\ Iso-Octane
*
\
\
0.« ('.9 1.0 1,1 1.1 I.J 1,4 l.S I.I
Air-Fuel Equivalence Ratio
Figure 2. 2. 5. -2. NITROGEN OXIDES IN ENGINE EXHAUST145
(CFR Engine, Compression Ratio 10:1. Engine Speed 1800
rpm. Manifold Pressure 30 in. Hg)
2.2. 5.2. Effect of Engine Variables
The exhaust nitric oxide concentrations of an ammonia-fueled Si-engine
are found to increase when thermal efficiency of the engine increases. At
15% thermal efficiency the nitric oxide concentration wan found to be 200
ppm, which increased to 1200 ppm at 30% thermal efficiency." This re-
lationship between nitr.c oxide concentration and thermal efficiency of the
engine has been observed in gasoline-fueled engines.
The thermal efficiency of an Si-engine fueled with ammonia is dependent
upon, similar to the gaiolme-fueled engine, many engine variables, iiclud-
ing air-fuel ratio, spark-ignition timing, and compression ratio1.7'n> IJS
In all cases, the adjup'.ment of engine variables that increase the engine's
2-53
-------
89Z9
thermal efficiency will also increase the exhaust nitric oxide concentrations.
Hydrogen has also been used to improve the combustion ol ammonia, which,
in turn, increases the nitric oxide emission.
The slow-burning characteristics of ammonia causes a large quantity of
unburned fuel to be expelled into the exhaust.37'158 An ammonia concentration
81,000 ppm, calculated on dry basis, has been found in the engine exhaust.
The addition of 2% hydrogen reduces the nitric oxide concentration by 94%,
down to 5300 ppm. The effect of hydrogen addition in ammonia is quite
similar to that of increasing the air-fuel ratio on carbon monoxide emissions
of gasoline-fueled engines. In addition to adding hydrogen, other potential
methods of reducing exhaust ammonia include catalytic conversion of am-
monia to hydrogen and nitrogen, chemical conversion, and ammonia adsorp-
tion. Supercharging the ammonia has also been found effective in reducing
unburned ammonia.*"1'"11*
In fact, nitric oxide concentrations as high as 1. 5 to 4. 5% have been
measured in the cylinders of an ammonia-fueled Si-engine. The nitric
oxide concentrations in the exhaust are substantially reduced from that
found in the cylinder, but are still significantly higher than that of gasoline-
fueled engines.
The high nitric oxide concentrations produced by ammonia combustion
led some investigators to believe that a part of the nitric oxide is formed
through pyrolytic reactions in which nitric oxide may be evolved from HNO
radicals which, in turn, are evolved from successive kinetic steps in the
oxidation of ammonia.1"'145'15'
2. Z. 6. Emission Characteristics of Blended Gas
Lee and Wimmer" tested the emission characteristics of a fuel gas
having the following composition:
2-54
89Z9
Methane
Hydrogen
Carbon Munuxidc
Carbun Dioxide
Water
35. 3% (mole basis)
JO. 0%
J.J%
16.4%
15.0%
It was found that this fuel gas. produced by the steam reforming of
hexane, provides satisfactory performance as an engine fuel and produces
lower exhaust emissions than either propane or methane. The reduced
emissions were attributed to the presence of hydrogen which extends the
lean misfire limits of methane. However, in view of the composition of
the fuel gas, the presence of inerts — carbon dioxide and water — can also
be credited with the reduced emissions, particularly in nitric oxide emis-
sions.
The ability of hydrogen to extend the lean misfire limit of hydrocarbon
fuels is well known. However, due to hydrogen's high burning velocity,
only a limited amount of hydrogen can be incorporated without experiencing
knock or other engine performance problems. However, recent investi-
gations into ihe combustion of pure hydrogen in an SI-engine indicate that
hydrogen can be successfully used by utilizing special fuel-injection tech-
niques.140 Moreover, it was found., in contrast to predictions on the basis
of peak cycle temperature, that a research Si-engine fueled with hydrogen
produces surprisingly low emissions of nitric oxide.140 Thus, there is no
reason to doubt that hydrogen cannot be blended with methane to further
reduce the level of exhaust emissions.
Z.Z.7 Emission Characteristics of Hydrogen
As in the case of ammonia, the combustion of hydrogen yields no carbon
monoxide or hydrocarbons. The only stoichiometric product of hydrogen
combustion is water. Hydrogen's wide limits of flammability when mixed
with air (4. 1 to 74{ hydrogen in air) enable the combustion to take place
at extremely fuel-lean and fuel-rich conditions, greatly reducing the
Z-55
-------
89Z9
possibility of incomplete combustion in an engine Therefore, as an engine
fuel, the only pollution concern of hydrogen, particularly in view of its high
heat of combustion, is its emission level of nitrogen oxides.
Recent studies sponsored by the Air Pollution Control Office, Environm. ital
Protection Agency, and conducted at Oklahoma State University indicati J that
with the use of suitable fuel injection, a spark-ignition internal corrju..iian
engine can be fueled with hydrogen. "" One of the test engines involved in the
study was a modified four-stroke, air-cooled, single cylinder 3. 5 horsepower
Clinton Model 494-0301 engine equipped with a special gaseous hydrogen
injection system. The operation of the test engine was found to be quite
satisfactory and the emission level of nitrogen oxides surprisingly low. A
comparison of nitrogen oxides emission of this engine to that of similar
gasoline- and benzene-fueled research engines is presented in Figure 2.2. 7. -1.
a
a
U
°
1
— Gasoline (5b)
Benzene (178)
^Hydrogen
os i-o i-s 2-0 «• s i-o j-s
Air-Fuel Equivalence Ratio
Figure 2. 2. 7. -1. EMISSION OF NITROGEN OXIDES
OF A HYDROGEN-FUELED SJ-ENGJNE1'8
(Spark timing 7*BTC, Compression Ratio 8 1.
Engine speed 2000-4000 rpm, wide-open throttle)
I H S T I T U T
2-56
c »
8929
Thr results indicate that the peak nitrogen oxides emission occurred at
peak rngmi- power output - which is synonymous with fuel-rich operation
iind*r the IPBI conditions. The magnitude of nitrogen oxides emission of the
hydrogen-fueled engine at peak power was found, however, to be much lower
than that of other common gaseous fuels. Further reduction of this magnitude
can be obtained by increasing th« air-fuel ratio. As in the case of hydrocarbon-
fueled engines, increasing the air. fuel ratio of the hydrogen-fueled engine
decreases the power output, hut the amount of reduction in engine power
output, due to hydrogen's high flammabihty is considerably less than that
of other fuels.
The data presented in Figure 2 2. 7. -1 also indicate the extended lean
misfire limit of hydrogen is greater than 3. 0 (as compared to 1. 4 for gaso-
line and 1. 7 for propane anc" methane), which indicates that the nitrogen oxides
emission can be significantly reduced if an automotive engine can be suitably
converted to hydrogen.
It should be noted, however, that the NOX emission shown in Figure
2 2. 7. -1 are much low., r than we have any reason lo expect in view of the
relatively high combust.on temperature of hydrogen. This leads to speculation
as to the possibility tha* some of the NOX formed may not have been measured
in the exhaust gas. It i- at least possible that some nitric oxide may have
combined with the water present and was separated out before the remaining
exhaust gas was analyzed for NOX.
These data demonstrate the attractions of hydrogen as a clean motor fuel
from the combustion vie.vpoint However, the broad air/fuel range thai
makes it attractive also makes it more hazardous from a safety standpoint.
It is possible by the use of hydrogen mixtures, such as hydrogen and methane
for example, tc reduce the explosive mixture range and still provide suitable
misfire limits. Much more work needs to be done with such mixtures however
before a suitable fuel miMure can be found that will meet the safety and logis-
tics requirements of motor fuels in addition to the combustion requirements.
2-57
-------
8929
2. 3. Vehicular Exhaust Emissions
2. 3. 1. Exhaust Emission Testing Procedure
The first regulation on automotive exhaust emissions was established by
the California State Department of Public Health in 1959 in which standards
for exhaust concentrations of hydrocarbons and carbon monoxide were
adopted.10 This early regulation was later expanded to cover crankcase,
fuel tank, and carburetor emissions and to include detailed testing proce-
dures for measuring automotive exhaust emissions, including the 7-mode
driving schedule (Figure 2.3.1.-1).
o.
Q
B
U
a
U
41 M> n 104
Time, seconds
129 117
Figure Z.3.1.-1 DRIVING SCHEDULE OF CALIFORNIA 7-MODE CYCLE
TESTING PROCEDURE"
The California testing procedure was later modified and adopted by the
Federal Government and became the 1970 Federal automotive exhaust
emission testing procedure. >"•»»
Z-58
N 5 T I T U I
8929
The early automotive exhaust emission regulations were expressed in
terms of concentration of pollutants in the exhaust. Such regulations allow
cars with higher exhaust flow rates to discharge a greater weight of emis-
sions into the atmosphere. Thus, attempts were made to make the emission
measurement more equitable, and resulted in regulations based on emission
weight.
Various mass emission measurement techniques have been proposed,
including the proportional sampler, fuel How with exhaust concentrations,
total bags, exhaust flow rate with concent rations, and the constant volume
sampler recently proposed by the Air Pollution Control Office (APCO). 2*
However, each of the proposed techniques had, at the time they were pro-
posed, certain limitations although most of these techniques have been used
by some investigators Dver the years. Therefore, the 1970 Federal standards
were adopted on the basis of calculated mass emissions.110
This calculation requires measurement of the emission concentrations in the
exhaust and then assumes an average exhaust volume flow rate for each
weight class of vehicle. While this procedure on the average requires equal
mass emission control for different size vehicles, it does not recognize the
variation in the exhaust volume flow rate for different vehicles in the same
weight class.
Since it was first proposed, changes and improvements have been made
in the constant volume sampler technique that appears to have solved many
of the problems found in the earlier mass emission techniques. The con-
stant volume sampler (CVS) technique measures a constant volume of a mix-
ture of exhaust gas and dilution air. A small sample of the exhaust flow is
stored in a plastic bag and the concentration is then measured. Since the
total air flow and samplt flow are known, the actual mass of each constituent
of the exhaust can easil) be measured. This emission test procedure has
been proposed by APCO to be used for 1972 and later new vehicle certifica-
tions. >'«•»>
2-59
-------
8929
2. 3. 2. Effect of Testing Procedure on Exhaust Emission Data
During the development of a suitable testing procedure for measuring
exhaust emission, many changes were made in practically all parts of th'
procedure, including vehicle preparation, driving schedule, test cyc'e,
weighting factors, method of calculations, and instrumentation. TJa>-n
change made in the test procedure invariably affects the final emission
data. For example, the effect of driving schedule and weighting factors on
emission measurement can be seen in Figure 2.3.2.-1, which illustrates the
emission levels of a gasoline-fueled vehicle under various modes of driving.
The data presented in Figure 2.3.2.-lclearly indicate that the level of
exhaust emissions vary markedly among the various driving modes. From
the concentration point of view, the two deceleration modes have the highest
levels of hydrocarbons and carbon monoxide while the two acceleration modes
are highest in nitrogen oxides. However, after the introduction of assigned
weighting factors, it becomes evident that the two acceleration events are
the largest emission contributors. The significance of these two events can
be further illustrated by the effect of changes made in the acceleration mode
on emission measurement (Figure 2.3.2.-2).
In addition to the driving schedule, there are other factors related to the
test cycle that are equally important in determining the final emission levels
These include the starting condition of the engine —hot or cold, soak time
of the engine, and whether the test cycle is open or closed. Therefore,
unless the driving schedule, test cycle, and many other related factors are
exactly alike, the emission measurements of a vehicle made under different
test conditions can be significantly different.
The analytic instrumentation as well as sampling technique are also known
to have intimate influence on emission measurement. For example, two of
the most commonly used techniques for determining exhaust hydrocarbon
emission are nondispersive infrared(NDIR) analysis and flame ionization
detector (FID). The NDIR technique has fast response capable of following
2-60
I N S T I T U T r
Cumulative Time, seconds
10 34 49 60 75 '0«
8929
ill 137
10 34 49 60 75 104 1Z9 137
Cumulative Time, seconds
Figure 2.3.2.-1EXHAUST EMISSIONS OF A GASOLINE-FUELED VEHICLE
UNDER VARIOUS MODES OF DRIVING163
2-61
-------
8929
2 sec
Faster Acceleration
Exhaust Volume, 46.28 CF/mile
Exhaust Emission
CO = 1.24% (19.2 gm/mile)
HC = 318 ppm (2.60 gm/mile)
NO = 1560 ppm (3.80 gm/mile)
2 se
Slower Acceleration
Exhaust Volume, 43. 72 CF/nule
Exhaust Emission
CO = 1. 52% (22.0 gm/nile)
HC = 307 ppm (2. 37 gm/rrule)
NOX = 965 ppm (2. 35 gm/rrule)
Figure 2.3.2.-2 EFFECT OF ACCELERATION CHANCE ON EXHAUST
EMISSION MEASUREMENT10*
(4-Cylinder Gasoline-Fueled Vehicle,Federal 7-Mode Hot Cycle)
rapid exhaust concentration changes from the cyclic operation of a vehicle;
but it is known to have unequal response to various hydrocarbons. 11 n-
hexane is used as sensitizer in NDIR analysis, only paraffins will respond
properly, while olefins, acetylenes, and aromatics, that are structurally
different from n-hexane, will give lower response values. The FID, on the
other hand, eventually gives the same response, on a total carbon basis, to
all hydrocarbons but requires proper flame conditions, i. e., a proper ratio
of hydrogen-oxygen-nitrogen for accurate measurement. The FID is found
to be slower than NDIR in response time and vulnerable to oxygen interfer-
ence.71'144 The significant differences between these two techniques prevent
2-62
I M S T I T U T E
N O L O f *
8929
the direct comparison of hydrocarbon emission measurements. In general,
the FID technique gives a substantially higher level of hydrocarbons than the
NDIR analysis. An accurate conversion or comparison of the emission data
using these two techniques requires accurate correlation data.
A similar situation is also found in a nitrogen oxide determination in which
various techniques — such as phenoldisulfonic acid (FDA) methods, Saltzman
and modified Saltzman methods," mass spectrometric technique,"'12* nitrous
fume analyzer, nitrogen dioxide colorimeter, and nondispersive infrared
analysis — have been proposed and used in many investigations. Although a
comparison of measurement data can be made more accurately than in the
case of hydrocarbons, it may also require suitable correlation information
particularly at low concentration levels.
2. 3. 3. Exhaust Emission of Gasoline-Fueled Vehicles
Since the gasoline-fueled spark-ignition internal combustion engine was
first invented, efforts in the automotive industries have been aimed primarily
at improving its performance. During the last decade, however, increasing
attention has been placed upon the control of pollution emissions. In 1966,
the first legislative regulation was established to curtail the automotive emis-
sions. As a result, changes were made in the design of gasoline-fueled en-
gines to meet the immediate needs. Additional efforts in research and develop-
ment are also being made toward the realization of a "pollution-free" vehicle.
The challenge that the automotive industry faces can be illustrated by the forth-
coming Federal regulations of automotive emission control (Table 2.3.3. -1).
To meet the emission standards established in 1966 and 1970, some changes
and innovations in engine design have been made primarily to reduce the hydro-
carbon and carbon monoi ide emissions. These changes include the following:
• Closed positive crankcase ventilation (PCV) system to prevent crankcase
exhaust emission.'11*1'7*
2-63
-------
8929
8929
in
1
Z
H
K
z
o
2
u
U
>
S
o
H
a
h
O
1
ca
U
od
U
O
£
1
m
m
01
V
2
"
i
Standard
~s
0
•o
•o
V
a
D
a
0
14
a.
1 3
P
a*
O
— •o
IM ? !;
r- *
i IQ
ij -g
~" "• 1
0 1
•* w
!'
r- t. n
e> u -S
-"•08
« S
H I- U -°
Z 0, U U »
0 F
2-64
c « «
• Fuel evaporation control system for stopping evaporative emission.
System includes thermal-expansion fuel tank, venting system for fuel
vapor, carbon vapor -adsorp'ion canistor and air-purging system, and
carburetor vapor control.41'1*"1"
• Exhaust air injection system — injecting air into exhaust port or manifold
for oxidizing exhaust hydrocarbons and carbon monoxide. System in-
cludes engine-driven air pump, antibackfire and check valves.26'144'147
• Modified ignition system— retarded ignition timing at idle and low engine
speed, transmission controlled vacuum advance at low engine speed for
maintaining retarded timing, and temperature ovende for advancing
spark timing at high engine temperatures.
• Modified carburetion — heated intake manifold for improving uniformity
of air-fuel charge and its distribution among cylinders, increased idle
speed at leaner air-fuel ratio, deceleration throttle control for reducing
"decel" hydrocarbon emissions, leaner air-fuel ratio at power mixture
for reduced hydrocarbon emission during acceleration, and heated intake
air to maintain near-constant air temperature
lo"1'144tl4T
• Modified combustion chamber design— reduced surface-volume ratio,
sealed piston-ring orifices, and contoured cylinder head gasket. "•"•uiiiu
With these engine modifications, the automotive industry has been able
to produce vehicles meeting the 1970 Federal standards for exhaust hydro-
carbon and carbon monoxide. However, since further reduction in both
hydrocarbon and carbon monoxide emissions and control of nitrogen oxide
emission were proposed for 1975 and 1980, it has been apparent that the
engine modifications listed above alone would not meet the requirement.
One example is the exhaust manifold air injection system whose effective-
ness is dependent upon engine operation because of the variations in the
exhaust temperature, exhaust port static pressure, flow rate, and hydro-
carbon concentration during the exhaust cycle. Thus, with the air injection
system alone, there exists a limitation in the maximum possible reduction
in hydrocarbon and carbon monoxide emissions. The continuing research
and development efforts to further reduce the exhaust emission of a gasoline-
fueled Si-engine include the following:
• Exhaust recirculafon for reducing nitrogen oxide emission. Recycling
up to 15% of exhaust gas has been proposed.
2-65
-------
8929
• Modified carburetion for enriched fuel-air mixture at high power oper-
ation."41""
• Exhaust thermal reactor to improve the effectiveness of air injection
system and to allow enriched operation for lowering nitrogen oxide
emission without sacrificing the hydrocarbon and carbon monoxide
,.u.itrol. Ul«"*"4'
• Development and optimization of catalytic coverter systems for reducing
hydrocarbon and carbon monoxide emissions.147
• Development of catalytic converter systems for reducing nitrogen oxide
emissions.••9>7S'ISO
The results of research and development and improvements made in
exhaust emission control of gasoline-fueled vehicles are summarized and
tabulated in graphic form in Figure 2.3.3.-2, in which all data are on the basis
of either California 7-Mode or 1970 Federal 7-Mode testing procedures, and
were converted to weight basis using a vehicle weight of 4000 Ib. Some of
the data are hot-cycle values while others are composite-cycle values. The
hydrocarbon emissions arc based on NDIR analysis.
Recent results of research and development efforts have indicated the
approach that automotive and oil industries are taking to meet the immediate
needs in exhaust emission control of gasoline-fueled vehicles. It has been
indicated that, prior to the development of emission-free engines, the vehicles
to be produced in the next few years most likely will have catalytic converters
for reducing exhaust emissions to the levels of proposed 1975 and 1980 stand-
ards. The vehicles may also have exhaust recirculation, enriched carbure-
tion during power-mode operations, and retarded ignition timing for reducing
the nitrogen oxide output. There will be an air injection and exhaust thermal
reactor system for post-exhaust oxidation of hydrocarbons and carbon monoxide
emissions that are increased as a result of engine adjustment for controlling
nitrogen oxide. The pollutants coming out of the exhaust thermal reactors
will be reduced by the catalytic converters. In order to maintain the effec-
tiveness of catalytic converters, the use of unleaded gasoline may become
mandatory.
2.66
1
1
I
1
1
to
•S
r,t •*
1 °
o £
H g>
U i
Q Z
U
n!
z w
S j
S u 1
as
H > 1
3 2 1
< j
*3 .
™ U. -o
Pi
2 ° g
•it o
w o •£
H ,-
§ -3
s
u
>
o
5
a
Figure 2.3.3.-1.
•
••"
„
-'•
^
-
-"
0 ,
V
°''
^
O '
o
2'
«n ,
o '
«
e
N
0
„
0
*
drocarbone, g/mile
S
o
^"
o
*
e.
0
o
o
0
o
•a"
e
""
2
o
e
e
PI
o
"•
•
"
°
«
a
V
-o
1
S
«
d
U
»
**
M
T,
-s,'
3
o
N
—
_*
-a
"•
-
J
5
•A
,
e
m
0
*
Lrogen Oxides, g/rrule
Z
91 U U
£SE
00 00 00
o* CM m •»
r*^ r^ in c
f^ Is- (
..I'i-
1
1
--t-f-
1
1
1
1
1
1
Average of 63 1964 and
1965 vehicles using
leaded gasoline. 5
.*
u a o t
Egg 1
3D Ofi OO C
r in m c
in *
0 r i
i
i
i
I
1
-I---
i
(
i
i
Average ot Vj IVoi ana
1965 vehicles using un-
leaded gasoline.
M" r
1 0 0
ig i
10 OO C
) 00 <4
t f*l O
Y
—
i
V-8 vehicles. "
*>
tut
111
a u t
3 0 C
J M U
...
'I
1
C
itoi proaucuon venicie
with exhaust air injectic
«•
«
» -
! ^
• ,
> c
1 r>
,
r~
-
system lor n«-->-<_i
control."
i
i j
l| J
"no ^
1 •* 0
1 in _
7 ,
•1 —
T
1
i
i
i
•
i
i
j j
with 15% exhaust re cir-
culation."
n %
a
J! "
ig-^
1 "^^ t
wtf
) >M tr
< «• a
? 1
1 i
•*"!
i
equipped with leaner
o'
i
«
t
)
3
ignition."
>•
JO
10
c
0
I*
il u
C Q V
te: l. me vertical lines
2. Some of the emisi
assuming 4000 Ib
o
5
2-67
I N S T I T U
-------
Figure 2.3.J.-1. Continued. IMPROVEMENTS MADE IN EXHAUST EMISSION REDUCTION
OF GASOLINE-FUELED VEHICLES
Hydrocarbons•
Carbon Monoxide — — —
Nitrogen Oxides
Hydrocarbons, g/nule • "•' ":l ••.' 0>.4
Carbon Monoxide, g/nule o
Nitrogen Oxides, g/mile o o.z 0-4 0.6 0.6
7. Same vehicle as (6) with
modified carburetion and
ignition timing for reduc-
ing NOX emission."
8. Same vehicle as (7) but
with 15% exhaust recir-
culation added. "
9. 1966 V-8 air injection
vehicle with optimized
jets and enriched tar-
buretion, hotter spark
plug, retarded timing,
and exhaust recycle.
10. 1967 V-8 air-injection
vehicle equipped with ex-
hauat thermal reactor
and enriched carbure-
tion. "
11. Above vehicle equipped
with synchronized air in-
jection and further en-
riched carburetion durin)
1
.0 12.0 11.0 14.0
1.0 1,2 1.4 1.6 1. 8 2.0 2,2 2.4 2,4 2. a
_
r
_|
_
izmi;
i-
~
_ i
— — i
- — 19.0 g/nrule
- ••• 4. 74 g/nule
-—28.9 g/mile
<
— »• 2 1 . 1 g /mile
- fc.23. 2 g/nule
Figure 2.3.3.-11 Continued. IMPROVEMENTS MADE IN EXHAUST EMISSION REDUCTION
OF GASOLINE-FUELED VEHICLES
Hydrocar. ns Carbon Monoxide Nitrogen Oxides
Carbon Monoxide, g/nile • '•° J/° J;° 4;° s.-° *;° 7;° 8>.° *•.' ::".'.
Nitrogen Oxides. R /mile °- ' "• 4 "•' °-' ' ' ':' ':* '-' '•-' ':8 *:J *:' *'•' *:"
12. Vehicle 11 with 11% ex-
haust recycle during
partial throttle operation
and temperature control-
led enriched carburetion
during acceleration."
13. Vehicle similar to (6)
equipped with HC-CO
catalytic converter. At
start of catalyst dur-
ability test using un-
leaded gasoline. Hot
cycle data.149
14. Above vehicle^ter
1 5. Vehicle similar to (4)
equipped with NC-CO
catalytic converter. At
start of catalyst dura-
bility test using unleaded
gasoline. Hot cycle
data '«
1
_
» ^_ ^_ __ *^
_
I
i_ i__ — — — ^
|
I
~H I
1
1
1
1
1
1
I
1
i I
1 1
1
1
1
|
I
I
1
S
-------
O
H
U
Q
U
I
Si
u >
"
o
O
I
u •*"*
25-^
•
892?
The vehicles described above are rapidly becoming a reality and the
reduction in exhaust emission can be envisioned. However, there are oome
questions concerning the durability of the emission-control subsystems.
Looking at the standards proposed lor 1975 and 1980. one can easily realize
the narrow allowable margin for maladjustment or malfunctioning of the
future engine and its emission control subsystems. This situation can be
illustrated by the results of vehicular emission tests conducted on 1970
vehicles by the New Jersey State Department of Health (Table 2.3.3.-2).
Table 2.3.3.-2. AVERAGE EMISSIONS OF 1970 VEHICLES
TESTED ACCORDING TO THE NEW JERSEY ACID
CYCLE PROCEDURE"
Exhaust Emission
Hydrocarbons Carbon Monoxide
—^——— gm/miie ————
New Jersey ACID Cycle
1970 Federal Standard
4. 5
2.2
50
23
2-70
Since all 1970 vehicles are assumed to have passed the emission standards
established for 1970, and the cited reference has indicated that the New
Jersey ACID cycles results correlate with the Federal procedure results,
the difference in exhaust emission indicated in Table 2. 3. -2 can only be
attributed to the deterioration of emission-control engine adjustment and
systems. Under the stringent standards proposed for 1975 and 1980, the
problem of maintaining emission control systems will be much more diffi-
cult.
2. 3. 4. Exhaust Emission of Caseous Fuels Powered Vehicles
Although some of the gaseous fuels, particularly propane or LPG.
have been used as motor fuel for some time, the exhaust emission of these
vehicles has been taken for granted and has not been investigated in depth
2-71
I N S T I 1 II I [
-------
8929
until recently. Therefore, information on the exhaust emission of vehicles
powered by gaseous fuels as a function of engine design and operation are
very limited. Even propane or LPG - the best known gaseous fuel, has
only been investigated on the basis of simple fuel conversion on existing i ie-
oline.fueled vehicles for the purpose of either single-fuel or dual-fue oper-
ation. The conversion usually involves the installation of a gasec-i; -uel
storage tank and fuel line system, a pressure regulator, a fuel-air mixer.
and a vaporizer if LPG or liquefied natural gas (LNG) is used. Some
engine adjustments, such aa ignition retardation, blocked manifold heat
riser, and/or disconnection of vacuum ignition advance, may also be in-
volved. The emission data of some of the converted vehicles are presented
in Figure 2.3.4.-1.
Comparing the data presented in Figures 2.3.3.-1 and 2.3.4.-1, it is apparent
that a simple conversion of existing vehicles from gasoline to propane or
methane results in a significant direct reduction in exhaust emissions,
particularly in hydrocarbons and carbon monoxide, although not enough to
meet the 1975 Federal standards. The available information also indicates
that a simple switch of fuel from gasoline to gaseous fuels alone, or to-
gether with some simple adjustment of engine variables, will not meet the
requirement. However, when propane or methane is used, the degree of
engine adjustment required for exhaust emission reduction is leas than that
of gasoline. According to information derived from studies made for gaso-
line-fueled vehicles, it is apparent that many innovations made for emission
reduction are also applicable to propane and methane due to their similar
combustion characteristics. For example, there is no reason to doubt that
in a propane or methane-fueled vehicle, the nitric oxide emission would
be reduced fay exhaust recirculation, or the hydrocarbon and carbon mon-
oxide emissions would be reduced by the installation of exhaust thermal
reactors. It has also been known that catalytic converters designed for
gasoline would work well with propane and methane. In fact, the absence
of lead and the relatively simple composition and clean combustion of
2-72
Q
bl
H
tC.
W
O
u
a
j
u
G? U]
a<
So
u
f-
X
U
c o «
3 — 5 e
'.
2-73
-------
Figure 2.3.4,-!. Continued. EXHAUST EMISSIONS OF VEHICLES CONVERTED TO
GASEOUS FUELS
Hydrocarbons
Carbon Monoxide •
Nitrogen Oxides —
Hydrocarbons, g/rmle ° °-' V V "•* V
Nitrogen Oxides, g/mile •
4. Average of 7 vehicles
(model unknown] converted
to propane, tested accord-
ing to 1972 Federal 9 -CVS
procedure,
5. One of the 7 vehicles in 4,
with 6-eylinder engine."7
(,. A Ifl70 V-B vehicle con-
verted to propane, equipped
with variable venturi car-
bureter, lean air-fuel mi*
ture, limited distributor
advance, disconnected
vacuum advance, and in-
creased idle speed. «s
7. Above vehicle with a pair
ol catalytic HC-CO con-
verters added.
8. Vehicle of 7 with enlarged
catalytic chamber and
slightly lowered comprea-
Bion ratio, hot-startdata.
9. Above vehicle cold start
data."
D 6.0 T.O S.O 9.0 10.0 11. 0 J2.0 12.0 il.O
I
I
t
^_
1
__|
1
1
1
j
E=r J
-4
— _ »
1
=.-d_..'..
-i
1 4. 16 g/nule
1
w
1
I
-------
8929
2. 4. Operating Characteristics of Gaseous Fuels in Si-Engine
2. 4.1. Propane
2. 4,1.1. Power Capability
With a simple fuel-system conversion, the maximum power output of a
propane-fueled engine has been found to be 82 to 95$ of the same engine
fueled with gasoline.1' "• 19> "' 77-'2''5- 10S ""'The observed reduction in power
output is believed to be the result of reduced volumetric efficiency of gaseous
fuels. The exact amount of power reduction depends on the changes made on
the engine during the conversion.
The power output of a propane-fueled engine is affected by several major
engine variables. One of these variables is the air fuel ratio. Using a CFR
engine fueled with propane, the maximum torque was found to occur at an
air-fuel ratio slightly richer than stoichiometric.'5 Changing the air-fuel
ratio from this setting results in a power loss. The power loss is more
significant when the air-fuel charge is made leaner. In addition to greater
power loss, leaning the air-fuel ratio also increases the cycle-to-cycirc
variations in peak pressure, burning time, and rate of pressure rise that
result in inferior operation of the vehicle.21 The effect of air-fuel ratio
on power output of_a._propane-fueled engine has been illustrated by
Figure 2.2.3.-9 in which the power outputs at various air-fuel ratios and
throttle positions are indicated.
Ignition timing is another engine variable affecting the power output of
a propane-fueled engine. Switching from gasoline to propane changes the
value of the least advance for best torque (MET) of an engine. The exact
change in MET varies among various sources. The MET value of a propane
fueled engine also varies with the air-fuel ratios, advancing with the leaning
of air-fuel charge (Figure 2. 4.1.4l).Varying the ignition timing from the
MBT value results in loss of engine power; the loss is particularly significant
at lean air-fuel ratios (Figure 2. 4. 1..1. } The magnitude of power loss varies
with engine speed, but the engine "speed has no significant effect on the shape
of power contours at various air-fuel ratios."5'
2-76
I N S T I T II T F ~r r. . * T r - u .• - - -
8929
a
,c
o? 28
U
is
o
0.
24
20
16
tn
H
D
D,
H
D 12
O
2 8
to
to
1 4
-------
8929
carburetion system showed performance comparable to that of a gasoline-
fueled vehicle, and experienced no wintertime freeze-up by using underhood
air during winter operations. "°
2.4.1.2. Fuel Consumption
On the basis oi Drake specific fuel consumption (BSFC, Ib/bhp-hr), an
engine has shown lower fuel consumption with propane than that with
gasoline, varing from 9 to 12^ among various sources. •••••• The
reduced fuel consumption of a propane-fueled engine, as compared to
gasoline, is greater in part-throttle, low-speed operations than in wide-
open-throttle, high speed operations - a feature of some importance in
urban vehicular uses. However, due to the lower specific weight of propane,
and the richer setting of the air-fuel ratio for maximum power, a propane-
fueled engine will require more fuel, on volume basis, than a comparable
gasoline-fueled engine.
The fuel consumption of a propane-fueled engine is dependent on the
setting of many engine variables, -articularly air-fuel ratio and ignition
timing, as shown in Figure 2. 4.1. - 2.
O
2
en
O
u
SI
u
u
a.
V)
1.3
1.2
1.1
1.0
0 9
0.8
0.7
0 6
0. 5
21.0
1500 RPM
SO
40
3D
20
1C
IGNITION TIMING. 'ETC
Figure 2. 4. 1.-2. EFFECT OF IGNITION TIMING ON FUEL
CONSUMPTION OF PROPANE-FUELED ENGINE
AT VARIOUS AIR-FUEL RATIOS*6
(Production V-8 engine maintained at road load
and constant intake air-flow rate)
2-78
INSTITUTE r r " ' - - - u .. - . -
H929
At a given ignition timing schedule, increasing the throttle position of a
propane-fueled engine allows a leaner air-fuel charge, thus lowering the
specific fuel consumption.99 However, the specific fuel consumption is
increased if the air-fuel mixture is leaned near the misfire limit,
particularly at high throttle positions
Data presented in Figures 2. 2. 3-9. 2. 4. 1. -1. and 2. 4. 1. -2 indicate that
from the point of view of performance a propane-fueled engine has an
optimum ignition timing (about 30 to 35'BTC) and air-fuel ratio (approximately
17 1) that provide lowest fuel consumption and power loss resulting from
retarding the ignition timing. As the air-fuel ratio is made leaner, the
fuel consumption will not be reduced if power output is kept constant, and
a greater amount of power will be lost if ignition timing is retarded from
the optimum setting for reducing exhaust emissions.
2.4.1.31 Engine Maintenance
The propane-fueled engines have shown certain advantages over gasoline-'
fueled engines in the area of engine maintenance. The cited advantages
include the following-
• Much less scale and gum buildup inside the combustion
chamber and engine cylinders, thus requiring less frequent
engine overhaul.'5-1"'"'
• Less contamination of lubricating oil, thus allowing longer
service between oil changes and leas frequent replacement
of oil filter.:"- '».»'.'».'»
• Improved spark plug life.1S411"
• Improved service life of exhaust system.
2.4.2. Methane
2.4.2.1. Power Capability
The potential maximum power output of a methane-fueled engine is
quite similar to that of a propane-fueled engine. Due to the reduced
volumetric efficiency of gaseous fuel, an approximate reduction of 10%
in peak power output occurs in an Si-engine when the fuel is changed from
2-79
-------
89Z9
gasoline to methane.'5"07 If liquefied natural gas is used, the utilization of
latent heat of vaporization to cool the intake air charge will improve the
volumetric efficiency of the fuel, thus increasing the power output. In-
creasing the compression ratio further increases the power output.184
Methane or natural gas has also been satisfactorily used in a compression
186
ignition type engine using spark ignition.
Using a CFR engine and operating at maximum-power spark timing, the
power output-air-fuel ratio-throttle position relationship of a methane-
fueled engine has been found to be similar to that of propane (Figure 2. 4.1. -l).
The effect of spark timing on power output of a methane-fueled engine has
not been thoroughly investigated.but is believed to be similar to that of
propane. Vehicle, fueled with natural gas have been satisfactorily operated
with retarded timing and disconnected vacuum advance.'" Similar vehicular
uses have been reported in trade journals.
80 88'9(Max) ,100
0.8
Q9 1.0 1.1 1.2 1.3 1.4 1.5 1.6
AIR-FUEL EQUIVALENCE RATIO
1. 7
Figure 2.4. 2-1. POWER CONTOURS OF METHANE-FUELED
CFR ENGINE"
(At Maximum Power Ignition Timing, EngineSpeed lOOOrpm,
Compression Ratio 8:1)
2-80
M 5 T
T U T f
89Z9
2. 4. 2. 2. Fuel Consumption
On the basis of brake specific fuel consumption (BSFC, Ib/bhp-hr), a
13$ reduction in fuel consumption has been observed on CRF engines as the
fuel is switched from gasoline to methane." Testing with vehicles con-
verted to dual-fuel operation inder urban driving conditions, an average of
10$ reduction - on the basis of 3tu/mile - in fuel consumption has been
recorded with methane.107
Like propane, methane allows leaner air-fuel ratios than gasoline at
all throttle positions, thus permitting further potential reduction in fuel
consumption, particularly at greater throttle openings (Figure 2. 4.2 -1).
However, like propane, the specific fuel consumption of a methane-fueled
engine will be increased if the air-fuel ratio is made too lean —near the
misfire limit.
In addition to air-fuel ratio, ignition timing also affects the fuel con-
sumption of a methane-fueled engine. Although detailed information is
not available, it is believed that the effects of ignition timing may be
similar to those of apropane-fueled engine (Figure 2. 4. 1. -2). Retarding
the ignition timing lowers the power output,thus increasing the specific
fuel consumption.
2.4.3. Ammonia
2.4.3.1. Power Capability
On the basis of the ideal Otto cycle with partial decomposition of
ammonia into hydrogen and nitrogen during the compression stroke, the
theoretical maximum power output of an ammonia-fueled Si-engine was
found to be about 77$ of that of a gasoline-fueled engine.1M Experimental
measurements of power output of ammonia-fueled engines vary widely,
ranging from 17. 5 to 70$ of the peak power of gasoline-fueled engines,
depending upon the engine modifications made.37' "• 145,158 Adding hydrogen
or utilizing catalytic decomposition of ammonia increases the power
output.37'158 Increasing the compression ratio and supercharging at higher
2-81
-------
8929
engine speeds further increase the power output of an ammonia-fueled engine
and can raise it up to that of a gasoline-fueled engine.37
The power output of an ammonia-fueled engine is critically affected by
the airrfuel ratio. The leanest air-fuel ratios for best torque (LET) were
found to be slightly leaner than stoichiometric (6. 06:1),but are comparatively
much richer than that of a gasoline engine.
Changing the air-fuel
ratio from the LET value reduces the engine power output rapidly, and the
effect varies with the change in engine speed.IH
The minimum spark advance for best torque (MET) of ammonia-
fueled engines was found to be much greater than that for gasoline,
indicating the relatively slow burning rate of ammonia. MET values of
around 100'BTC have been recorded with ammonia (20°to 40°BTC in
gasoline engine) and were found to be essentially constant despite the
change in engine speed.37
2. 4. 3. Z. Fuel Consumption
Theoretical calculations indicated that the specific fuel consumption
of an ammonia-fueled engine is Z. 5 to 3. 5 times that for gasoline.1S8
Experimental results have shown that an engine will consume twice as
much or more ammonia than i so-octane regardless of whether the engine
is operated at peak-power or peak-economy conditions.1M
2.4.3.3. Engine Maintenance
Due to its alkalinity and the large quantity of water and nitric oxide it
produces in combustion, ammonia has been thought to be corrosive to
engine materials. Experimental results indicate that certain materials
such as rubber, neoprene, and cast copper-lead bearings have shown
definite deterioration as a result of handling ammonia." Common engine
metals such as cast iron and aluminum have shown slight weight gain and
surface discoloration after exposure to ammonia combustion. The effect
2-82
8929
of ammonia combustion is particularly significant on aluminum and other
copper-containing alloys, rendering them less compatible.72 The com-
bustion of ammonia has been found to present no particular problem to
lubricating oil. "
2-83
-------
8929
3. EFFECT OF GASEOUS FUELED VEHICLES ON POLLUTION
LEVELS IN NEW YORK CITY
This section assesses the possible impact OB the pollution level using the
gaseous fueled vehicles in a major urban area. New York City was chosen
because of its severe automotive pollution problem and because its traffic
patterns differ from Los Angeles, which has already been given close attention
by automotive pollution analysts.
Since it may be impractical to convert the entire motor vehicle population
including private automobiles to gaseous fuels because of logistics problems
alone, this question was subordinated to the more practical one of converting
commercial fleet vehicles only. The analysis attempts to answer the following
questions:
1. If all of the commercial fleet vehicles in New York City were converted
to gaseous fuel, how much would the automotive pollution level be
reduced?
2. Does it matter whether the fuel is natural gas or propane?
3. How much fuel would be required in each case and how could it be
obtained?
4. From the point of view of pollution control, is it more desirable to
burn gaseous fuels in automobiles or in electric generating stations?
3. 1. N'ew York City Traffic Patterns
In order to determine the extent of motor vehicle pollution in New York
City, the State of New York, Department of Health, conducted a survey in
1968 in which the NYC vehicle populatiDn and traffic patterns were charac-
terized. Vehicle emission test procedures were then developed based on
regional traffic patterns.127 The areas included in the survey were most of
Manhattan, and the adjacent areas of Brooklyn, Queens, and the Bronx. The
traffic patterns were classified according to three general classifications of
NYC streets: 1) expressways, 2) arteries, and 3) capillaries. The vehicular
mileage (a two-way street was counted as twice the equal length of a one-way
street) of the selected areas were-r
N S T I T U T F
-------
8929
Expressways. 96.1 miles
Arteries, 383. 1 miles
Capillaries, 986. 5 miles
Traffic counts were also made at selected points to determine the number
of vehicles using each of the three categories of streets at any given time of
day. In addition to this, the average speed of the vehicles using each type
of street was determined. It was found that NYC vehicles spend approxi-
mately 34«g of their time idling and travel at a weighted average speed of 13
miles per hour.
Subsequent to the establishment of NYC traffic patterns and operating modes,
a comprehensive dynamometer driving cycle was developed for measuring
emissions from NYC automobiles (Figure 3. 1. -1).
Time, seconds
O. T. = open throttle.
Figure 3.1..1. COMPREHENSIVE DRIVING CYCLE
OF THE CITY OF NEW YORK1 "
Based on the comprehensive driving cycle, the test procedure was reduced
to a quick cycle by mathematical techniques to provide a more efficient method
of emission measurement. The quick cycle developed is shown in Figure
3.1.-2.
3-2
I N 5 T I T U T F
8929
X
a
S
Q
W 20
U
a,
w
jj,.
w
C. T.
C. T.
0 10 20 30 40 50 60 70 SO
Time, seconds
C. T. = closed throttle
Figure 3.1. -2. QUICK DRIVING CYCLE OF THE CITY OF NEW YORK
The NYC vehicle population was broken down into the following categories:
passenger vehicles, medallion taxicabs, rental cars and private taxi cabs,
and commercial vehicles.1 ** The commercial category includes all trucks
without a further breakdown according to gross vehicle weight or engine
displacement. The 1969 NYC registrations were as follows:
Passenger Vehicles 1,558,754
Medallion Taxicabs 11,760
Rental cars and private cabs 45,618
Trucks 115,018
The principal variable used in the determination of emission levels is the
number of vehicle miles driven. Average national figures were used to
obtain the number of vehicle miles driven within each category of registrations.
The basis for this method and its explanation is found in Reference 1'4 The
total number of vehicle miles is then determined by the number of vehicles
registered in each category. The total number of vehicle miles is then
divided by 300 to arrive at an average daily value of emissions. Three
hundred was chosen as a means of allowing for the time during which a vehicle
may not be in use as much as on business days. The number of vehicle miles
and the factors used in their calculation are presented in Table 3.1. -1.
3-3
-------
r^i *Z
H
z
o
£ «'
< _:
-^ rj-
la
« 0
U $
v>
a
o
H
c
.Q
^
^
c
3
to
v
*£ -S
*4
National
Miles D
OS
"oo o
CO O
Tt* 0
cr» m
•—«—-.
n
_Q
S
£
S S
r- r-
00 -<
m
^
01
n)
Q
oc
0)
fl)
o
O w
., ^
0) C *
OD 2
C -^ •-
0 "^ x
co * nj
£ s"
o
O
o
o
00
^
in
^
C
nJ
»4
«}
u
3
1
^
td
*-*
V
rt
h
(X
8929
The privately owned passenger vehicle is not regarded as one that could be
converted to a gaseous fuel. This is primarily due to the handling characteris-
tics of the fuel itself, the conversion expense, and the variance of the vehicle
operation. The same may apply to rental cars,with some exceptions.
Rental cars are subject to intercity use, but because of servicing procedures,
such as refueling at central garages, th^y may be easily operated as a dual-
fueled vehicle. With the dual fuel system the rental car could be driven using
gasoline in rural areas and gaseous fuels in urban driving. Therefore, for
discussion in this report they are considered as a gaseous fueled vehicle.
3.2. Emissions From Present Vehicle Traffic
Both the comprehensive and quick driving cycles were used by New York
City for measuring exhaust emissions of passenger automobiles on a chassis
dynamometer. A 198-car fleet, representative of NYC car population in 1968,
was tested with the developed cycles as well as with the 1968 Federal
procedure. Good correlations in hydrocarbon, carbon monoxide,and carbon
dioxide emissions were found between the two cycles. A comparison of the
emission measurements using the NYC quick cycle and the 1968 Federal
procedure is shown in Table 3.2.-1.
Table 3.2. -1. COMPARISON OF EMISSION DATA OBTAINED WITH
NYC CYCLES AND 1968 FEDERAL PROCEDURE127
# Vehicle
Cumulative
10
25
50
75
90
Mean
NYC
Com]
460
630
850
1110
1540
1152
Cycles
3 Quick
460
660
840
1090
1500
1108
Federal
Hot Cycle
270
370
500
650
890
602
NYC
Comp
1.50
1.97
2.52
3. 15
3.98
2.78
Cycles
Quick
1.12
1.67
2.21
3.00
4.05
2.65
Federal
Hot Cycle
0.97
1.43
2. 10
3. 00
3.90
2.34
3-4
The hydrocarbon emissio-.s as determined by NYC cycles are distinctly
higher than would be indicated by the 1968 Federal cycle. This difference is
most likely caused by the prominence of the idle mode in the NYC cycles. The
3-5
I N S T I T U T
-------
8929
carbon monoxide levels as determined by NYC cycles are also higher than
indicated by the Federal hot cycles, but the difference is not as pronounced as
in the case of hydrocarbon emissions. The difference in emission levels
produced by the NYC and 1968 Federal cycles reflects the basic difference in
traffic pattern between the City of New York and the West Coast.
In the New York survey, the emission measurements were made by using the
variable-dilution'bag sampling technique and NDIR hydrocarbon analysis.
The Federal test procedures for motor vehicle emissions have been changed
and new procedures based on the constant-volume sampling technique and FID
hydrocarbon analysis have been put in effect. For estimating the potential benefit
in emission reduction with gaseous fuels, it is necessary to have emission
information in terms of a mass unit, such as grams per mile. The NYC
emission data provided'by the cited reference127 ( Table 3.2. -l) are in
concentration units. Although a mass conversion formula was included, we
found that the emission data converted to mass units according to the given
formula are unreasonably low. For example, the mean hydrocarbon emission
produced by NYC quick cycle shown in Table 3.2. -1 is 1108 ppm, which
corresponds to 0. 109 Ib/hr, or 3. 74 g/mile, according to the given mass
conversion formula, since the NYC quick cycle has a time duration of 0.021
hour and a distance of 0. 18 mile. Comparing this value with other available
emission data based on Federal procedures, it was noticed that the value of
3. 74 g/mile is exceptionally low in view of the fact that NYC quick cycle
produces twice as much hydrocarbon emission as the 1968 Federal hot cycles.
Examining the procedures used in the NYC survey, we found that detailed
information on sampling technique, such as volume of sample gas collected
and the ratio of total volume of gas through a variable-dilution sampler to bag
sample, is not available, 'thus mass conversion formula could not be veri-
fied.
However, we noted that subsequent to the development of NYC test cycles,
the State of New Jersey embarked a similar survey for the purpose of
developing an emission testing procedure for its urban motor vehicles.
8929 .
The result is a 4-mode short driving cycle (ACID cycle) that closely resembles
the NYC quick cycle. A comparison of the two driving cycles is presented in
Figure 3.2.-1.
\ N. Y. C. Quick
\ Cycle
10 20 30 40 50 60 70 80
Time, seconds
Figure 3.2. -1. COMPARISON OF NEW JERSEY ACID
DRIVING CYCLE AND NYC QUICK DRIVING CYCLE*9
The New Jersey ACID driving cycle was developed through the use of
techniques and procedures similar to those used in the NYC survey, aimed
at simulating the urban traffic pattern of the state. The developed technique
utilizes a chassis dynomometer with fixed inertia and proportional road load,
and the emissions are measured by using a constant-volume sampling technique,
thus directly yielding the total mass of pollutants emitted during the cycle.
The hydrocarbon emissions are determined by the flame ionization detector
(FID) technique, which has been specified in the new Federal specifications.
The ACID driving cycle has been used by the New Jersey Department of
Health in measuring the exhaust emissions (hydrocarbons and carbon
monoxide only) ofover 1000 New Jersey vehicles, and has been found to
correlate well with the 1970 Federal procedure. In view of the fact that
suitable .NYC emission data are not available, the New Jersey data-were .
3-6
3-7
-------
chosen for estimating the effect of selected gaseous furl on exhaust emission
of the NYC motor vehicles.
The emission factors for gasoline-fueled vehicles were compiled by IGT to
be as follows:
Hydrocarbons 6.8 p/vehicle mile^
Carbon Monoxide 78.0 g/vehicle mile
Nitrogen Oxide 4.0 g/vehicle mile
The above factors and the data summarized in Table 3.2. -1 were used to
determine the emissions from the listed classifications of vehicle registrations.
In Table 3.2. -2, it can be seen that the passenger vehicle contributes approxi-
mately 73< of the total vehicle emissions. Due to the unlikelihood of converting
vehicles in this classification to a gaseous fuel, the emissions from this
category are not shown to be reduced in Table 3.2. -2 and are carried over
into Columns 2 and 3 of the table.
Emission factors for gaseous fueled vehicles are presented in Table 3. 2.-3
from work done in Section 2. The same factors would apply to both propane
and natural gas. It was found that propane and natural gas can reduce vehicle
emissions by comparable amounts. Two values for gaseous fuel emissions
are given, a high and low value. The low value is dependent upon various
degrees of sophistication employed in the actual engine conversion.
From Table 3. 2.-2, it can be seen that through conversion of fleet vehicles,
taxi cabs, trucks, rental cars, and private cabs, the level of emissions can
be reduced from 3682 tons/day to 2756 tons/day (using an average value of
the high and low emission figures). This is equivalent to an approximate
reduction of 24«g in the number of tons emitted into the atmosphere daily.
* The average of 1968, 1969, and 1970-model sample vehicles of the
State of New Jersey. Emission data were obtained by using New Jersey
ACID test procedure, which closely resembles the NYC quick-cycle
procedure.
+ Estimated from available vehicular emission data.
3-.8
I N S T I T U I I
8929
o
3
Q£
s
0
^
3
U,
(A
3
O
01
" 2
« ,3
*£
|
^3
O
U
o
X ' —
M h
> 3
oc
o a
b -a
Z oo
> &
Q 1
W U<
u
^ w
Q 8
° »
ft (3
z
g
pg (M *P O
m -H i
V ^
3 o o: ~*
•3 H •*!
O
w
w
2 «
o
CM m
'. O
3
3
U
O^fOt^l'}' O^Ni)Ox^l'-- CT^mo^^f^* COf- COlfO CO m xO
%i3inr'iirtco ^ -H -^ — 4 (\i o^oo^^m i/^o^ IA^ -^ ^ o
O(^ao—*^o r-o^r— •'t r«i — • rj r— o*m o^lm oo oo
NO CO °°
tr--*oro-3' inc^inco'* -^trimrac- roi~-r~vOui^po
oo^^m" tX"»rtt\r(M c*— «r-oo coevlvOco **"
vjao^TTO ^1/1 — £2 5 S »
c
C
Ifl
r^ MM 0000
°°-a. <£-a, >^ >.o "^ ^* M "rt^*?J *rt^*'2^*
C3O -M^I o "ort™00 ort ° "o *3 w '3
S H Q ^ ^ H Q * 3 HO « H Q | Q
« -^ fS "^ ^ ~ ^~* f 1
"""Vb O U-(JOO "rt^UOO ^000 _
SCOZ -JKUZ -^KUZ SKOZ 2
nj 0) HO
H O! H H
3-9
-------
vehicles of the State of New Jersey. Emission data
lure»which closely resembles the NYC quick-cycle
0
1
rt
m
V
•a
0
o
r-
•0
it
s
cr
00
vC
o
average
.
J3
•o
e obtaine
V
cedure.
o
p.
•a
§
0]
ehicular emi£
s.
available '
o
imated fr
0]
W
XI
Lg the installation of gaseous fuel system, pressure
id some minor adjustments of engine variables —
, increased idle speed, and disconnected vacuum
K
c
c
c
imple conver:
>-,
X
bo
*
O
rt
1
3
s
3 S
X rt
u >
n rt
1
111
m
X
oc
C
p
a
»>.
imated bi
n
W
•0
in
O
id
14
*J
X
O
W
u
c
<4
J3
"a
X
u
1
U
uri carbureto
c
T3
,
W
13
C
rd
;actor system
few years) .
V
thermal r
n the next
• rl
1-8
1-3
3z
i a
.a «
(i) XI
89Z9
89Z9
Tabl« 3. 2. -3. EMISSION FACTORS FOR GASEOUS FUELED VEHICLES
1. High Figures.
Hydrocarbons
Carbon Monoxide
Nitrogen Oxides
2. Low Figures:
Hydrocarbons
Carbon Monoxide
Nitrogen Oxides
0.2
1.0
0.8
Estimated on the basis of simple conversion involving the installation of
gaseous fuel system, pressure regulator, air-gas mixer, blocked manifold
heat, and some minor adjustments of engine variables — such as lean
air/fuel ratio, slightly retarded timing, increased idle speed, and
disconnected vacuum advance.
Estimated by assuming the use of —
4 More sophisticated venturi carburetor, which allows lean air to fuel
mixture at low power levels and enriched mixture at high-power operations
• More refined adjustment of engine variables and an engine equipped
with exhaust recirculation, exhaust air-injection thermal reactor
system, and catalytic HC-CO converter (simulating automobiles
that will be produced in the next few years).
3-10
3-11
t N S T I T U 1
-------
8929
3. 3 Comparison With Gaseous Fuel Use in Central Electric
Generating Stations
For comparative purposes the emissions were calculated for central
electric generating stations. Data are based on the amount of electricity
generated by Consolidated Edison of New York for 1969*.JVith the use of
emission factors reported by Esso Research1'and Stanford Research Institute,
the total emission level for 1969 was calculated. The electrical generating
data and the emission factors for each type of fuel are summarized in Table
3.3. -1.
Six cases were developed with the intent of determing how CO, HC, and
NO2 emissions would change by fuel substitution on an equivalent Btu basis.
These cases are summarized in Table 3. 3.-2.
Case I shows actual fuel consumption in 1969 and the emissions resulting.
Ca.se II shows how the emissions would change if the coal consumed was
replaced by an equivalent amount of natural gas. The greatest decrease is
found in the amount of CO emitted, while the next major change is found in
the amount of NO2 emitted. Hydrocarbons show the least change.
Case III was developed to determine how the emission pattern would change
if the amount of natural gas necessary to fuel fleet vehicles was used in
electrical generation, decreasing the amount of coal by an equivalent amoun^.
The amount of natural gas consumed would increase from 89. 1 to 119. 1
trillion Btu. The end result, although less dramatic than in Case II, is a
reduction of all the emission components.
Case IV is intended to show the same as Case III with the substitution of
natural gas for fuel oil. It can be seen that there is little change in the
total value of emissions. However, the amount of CO and HC emitted is
reversed. That is, in Case III CO is 429 tons less than in Case IV, while
the HC in Case IV is 325 tons less than the HC in Case II. The overall
effect is a net reduction of 104 tons when Case III is compared to Case IV.
Both represent a decrease of approximately 29 tons or 8.0$ in the overall
daily emissions based on Case I.
3- 12
INSTITUTE C F
8929
Table 3.3. -1. EMISSION CONVERSION FACTORS FOR
ELECTRICAL POWER GENERATION
Fuel
CO1" NO;11 HC1" SOZ14Z Part.1
lb/106 Btu-
Coal (2$ sulfur) 0.0192 0.769 0.00769 2.52 0.61
Fuel Oil (24, sulfur) 0.00027 0.695 0.0239 2.1 0.067
Natural Gas -N- 1.390 -N- -N- -N-
N = negligible
Btu conversion factors —
Coal: 13,000 Btu/pound, 26 million Btu/ton
Fuel Oil: 150,000 Btu/gal
Natural Gas: 1000 Btu/CF
1969 Fuel Consumption for Electric Generation
Consolidated Edison of New York
Coal: 3, 758, 047 tons 97. 7 trillion Btu
Fuel Oil: 1, 148, 330, 710 gallons 1 72. 7 trillion Btu
Natural Gas: 79, 076, 044.000 CF 79. 1 trillion Btu
kWhrGenerated 1969= 30,987,890,278
Peak Load: 7, 266, 000 kW
System Load Factor: 51. 8<
Cases V and VI were developed in light of the current natural gas supply
situation. With natural gas supplies being extremely limited, would diversion
of the amount of natural gas required to fuel fleet vehicles in the New York
area from electrical generation increase or decrease pollution levels' The
amount of gas required (approximately 40 billion CF) could be replaced by
coal (Case V) or fuel oil (Case VI). There is little change in the overall
emisssion level when Case V is compared to Case VI. However, noticeable
change is found when the HC and CO leve's are compared. When comparing
Cases V and VI with Case I there is an overall increase in emissions of
approximately 15.24 or 10<, respectively.
At this point a comparison can be made between vehicle emissions and
those resulting from electrical generation. With the natural gas required
for vehicle fuel supplied by diverting from electrical generation, an increase
3-13
-------
,0
*(
U
*
BJ
^j ^
1 0
3
CO
2
O
CO •*
2 "c
W u
i
i n
W o
U c_J
_i
y
H
U M
3 3
.
1
ro
.0 o
H
0
CO
a
U
*"
ca
O
£;
01
CO
u
a
Z a!
.3*1
5 o u
" -3 «
X rt ~i'
CU £j 2
3 1 <*
00 j= n
S* •
•a-3 2
<*J u z
i
c
-S61
| 8 o
Q> y '^
3 «> a
h 3 S
C
3
Js
,3o
i- « o o §§;? ~ n ? ^S 2^
O (M >O O 00
aorjoo oorocom
'H -H CO CO — *
o o so co in
1-
-------
8929
4. LOGISTICS OF GASEOUS FUELS
4. 1. Natural Gaa
4. 1. 1. Natural Gaa Industry
The natural gas industry plays a major role in the energy economy of the
United States, supplying more than 32 f. of the energy annually consumed in
the United States. In 1969 this amounted to approximately 22 trillion CF
of natural gas. Of this, about 15 trillion CF was distributed and sold by the
natural gas transmission and distribution industry. The remainder was used
internally by the industry or sold directly by producers.
The gas transmission and pipeline industry operates a vast network of
over 900,000 miles of underground pipelines that reach into every state in
the Union and serve virtually every city of any size. It is a nationwide
distribution system designed to take natural gas from the wellhead at a value
of approximately 16-20#/million Btu and to transport it throughout the
country selling it to various users at prices dependent upon the distance of
transmission and the quantities and terms of sale to each individual type of
customer. The cost of the gas itself is only a small portion of the total
price paid by the user. The retail price must cover the cost of transmission,
distribution, metering, service and other expenses associated with the
operation of this pipeline network. Thus the user pays from 2 to 10 times the
cost of the gas at the wellhead depending upon the expense involved in pro-
viding him with gas service.
There are tour general headings under which the sales of natural gas is
classified. By definition these headings are—
4.1.1.1. Residential Sales
Generally consisting of customers using natural gas in household quantities
for space heating, water heating, cooking and other uses. Because of the
small quantities and high service costs, this category of use has the highest
rate structure of all.
4-1
-------
8929
4.1.1.?. Commercial Sales
Generally consisting of customers primarily engaged in the wholesale
or retail trades, apartment dwellings of five or more families, and any
application not related directly to a manufacturing process. The natur -• of
the customers' primary business or economic activity at the location rjirved
usually determines this classification although classification practices vary
throughout the industry. Retail rates are lower than residential sales, but
higher than large industrial users.
4.1.1.3. Industrial Sales
Customers engaged primarily in industrial activities calling for large
quantities of fuel for heating or processing are classed as industrial users.
Many utilities further classify these customers by SIC numbers. Energy sold
under interruptible or off-peak rates is generally placed under this category.
It has the lowest retail rate structure because of the large quantities purchased.
4. 1.1.4. Other Sales
This classification includes municipalities or divisions of Federal or
local governments under special contracts or agreements applicable only to
public authorities using the energy for general or institutional purposes.
Table 4. 1. 1. -1. summarizes the quantity of gas sold and revenues derived
from each sales classification during 1968.
4. 1.1.5. Capital Structure
The gas industry requires very large amounts of capital to provide the
service that it renders. In 1968 approximately $890 was invested for each
customer served. The extensive pipeline network in the United States is
now worth approximately S35 billion. If the industry continues its present
growth rate of 6«f! per year, by 1980 an additional $30 billion worth of facilities
will be required. In order to obtain these facilities, the industry must spend
approximately $3 billion annually for additional equipment and facilities. One
4-2
INSTITUTE
E C H M
8929
of the major problems facing the industry is how to raise this amount of
capital in Ihc face of rising construction costs and high interest rates.
Table 4. 1. 1. -1. QUANTITY OF GAS SOLD AND REVENUES
BY SALES CLASSIFICATION
Class of Service
Residential
Commercial
Industrial
Other
Total
Trillions of Cubic Feet Revenues, $million
4.5
1. 7
7.5
0.6
14.3 8,646.0
4.1.2. Natural Gas Supply
The projected use of natural gas as a motor fuel must be tempered by
consideration of the available supplies. This subject is discussed in some
detail because of current concern for future gas supplies.
Clearly, the price of natural gas at the wellhead has been regulated at a
level too low to generate sufficient new supplies to meet the heavy demands
created by the low regulated price. This is evident from the fact that during
the last two years, less gas was added to the inventory of "proved reserves"
than was removed by deliveries to consumers.
Historically, additions have been made to reserves that were greater than
the amount removed from the ground and consumed; the long-term trend
had been to gradually increase the quantities of natural gas reserves. Much
of the reserve additions have come as a byproduct of oil exploration. However,
during the past decade the attention of the oil industry has shifted to fields
outside of the United States and the rate of drilling within the country has
declined to its lowest level in 27 years. This has resulted in smaller additions
to reserves; since 1968, the additions were considerably less than the amount
of gas produced. This is the principal cause of concern and has stimulated
the discussion regarding the potential supply shortage.
4-3
-------
89 29
The FPC is currently reviewing its pricing regulations in order to
alleviate the present supply situation. It has shown signs of willingness to
improve the economic incentive thai would, in turn, improve the supply
situation. Nevertheless, the amount that the wellhead price must be raised •
is difficult to determine as well as the impacl on the reserve situation.
The amount of proved reserves supporting U.S. natural gas production
represents less than 20.< of the total amount of potential gas reserves that
are estimated to lie under the United States and its offshore areas. Table
4.1.2.-1 presents estimates of the potential U.S. gas supply situation.
Table 4. l.Z-1. ECONOMICAL RECOVERABLE U.S. GAS SUPPLY*7
Potential Gas
Committee, IGT,
December 31, 1968 December 31, 1969
• trillion CF-
275
660-883
935-1108
370*
1305-1478'
Proved Reserves (including storage)
Future Discoveries
Remaining Recoverable
Produced to Date (excluding
stored gas)
Ultimately Recoverable
* Includes over 400 trillion CF of Alaskan gas.
f Includes 47-58 trillion CF of pre-1946 field waste.
The primary problem is one of providing additional incentives to prove
out those gas reserves which are now classified as future discoveries. We
should note that if sufficient drilling and exploration efforts were made to
transfer half of the 660 trillion CF shown as Future Discoveries, to the
category of Proved Reserves, it would more than double the Proved Reserves
that currently limit annual production. Thus our ultimate reserves appear
ample but we must pinpoint their locations so that they may be connected to
the-pipeline network.
4-4
-INSTITUTE Of <-. • ' -rr-u-JoiocY
8929
4,1.3" Demand
The use of natural gas in the United States will continue to grow, but at a
slower rate than it has in the past. Table 4.1.3-1 summarizes the forecast
U. S. natural gas requirements and the market share of natural gas as a
percent of total U.S. primary energy consumption.
Table 4. 1.3. -1. TOTAL PRIMARY CONSUMPTION IN
U.S. AND SHARE PROVIDED BY NATURAL GAS9*
Year
L'.S. Natural Gas
Requirements,
trillion CF
21.5
22.4
23.3
26.5
Share of Primary
Energy Consumption
Supplied by Natural Gas
3Z.l,j
33. 0
-------
8929
These are major factors to consider in the industrial sector. In many
situations the use of natural gas has proved to be the best economic alternative
in the search for nonpolluting fuels.
Industrial gas sales patterns have shown a substantial increase in r .cent
years when compared to past patterns increasing 5. 4£ in 1967, 8.?^ i.i 1968,
and 9.5$ in 1969. In addition, consumption of natural gas for central power
generation has shown even sharper increases. The Federal Power Commission
reported increases of 5. 3
-------
8929
It is reasonable to expect that future growth of gas reserves will be found
in the northern frontier areas as present exploration activities continue.
Early in 1969. the Canadian Petroleum Association published its estimate
that Canada's ultimate recoverable natural reserves would be-725-trillion CF
of which the Western Canada's sedimentary basin is expected to yield 270
trillion CF. Thus, the Northern frontier areas might be expected to yield
455 trillion CF. On this basis, another estimate was made by the Canadian
Energy Board'"of the quantities of additional gas that could be exported to
the United States if the reserves expected in the Northern frontier areas
materialize. These additional quantities are shown on line 3 of Table 4. 1.4. -1.
4.1.4.2. Imported Mexican Gas
Imports from Mexico are expected to remain at the level of 0.04 trillion
CF reported in 1968. Mexico has estimated reserves of approximately 12
trillion CF. Its annual production is about 575 billion CF including gas
exported to the U.S. The domestic consumption in Mexico is about 440
billion CF. Thus, the reserve-to-production ratio including export gas is
about 21, and even excluding exported gas, the reserves-to-production ratio
is about 27. Thus, Mexico has a supply of gas that is barely more than
.adequate for its own domestic needs; unless substantial new reserves are
uncovered, we can.expect little relief for our domestic gas shortage from
below the Rio Grande' River.
4.1.4.3. Gas From Alaska
The recent discoveries of oil at Prudhoe Bay in Alaska have revealed
substantial quantities of natural gas as well. While estimates of the total
quantities vary, we can probably count on 200 trillion CF of natural gas
reserves in Northern Alaska. Since about 25 trillion CF of natural gas
reserves are needed to back up one 48-in. pipeline, there appear to be
ample reserves to serve several pipelines from Alaska to the United States.
However, it will require about 4 or 5 years to build even the first pipeline
from Prudhoe Bay to the United States. Thus we cannot expect to obtain
4-8
I N S T ! T
8929
gas from Alaska until about 1975. The first such pipeline might have a
capacity of about. 3 billion CF/day.
4.1.4.4. Imported LNG
Thus far four projects for the importation of LNG into the United States
have been announced — all of them to the East Coast. Several additional
projects are known to be in the planning stage. It is possible that by the
late seventies, LNG importation could exceed 3 billion SCF per day if a
sufficient number of LNG tankers are built.
Of the 33 LNG tankers that have been built or are known to be on order,
only 11 will be available for East Coast shipments. These ships will be able
to import LNG at a rate of about one billion SCF/day by 1975. To increase
the importation rate to three billion SCF/day will require about 20 additional
large LNG tankers. These must be ordered during the next year or two if
all of the projected importation plans are to be in full operation by the late
1970's. There are about 12 shipyards with LNG tanker experience, (all of
them European) so the problem is one of fitting new LNG tanker orders into
busy shipyard schedules.
4. 1.4.4. 1. El Paso Project
The largest LNG import project announced thus far is that of the El Paso-
Natural Gas Co. In July of 1969, El Paso announced a 25-year agreement
with SONATRACH, a national Algerian gas and oil producing company, to
import 1 billion SCF/day of Algerian natural gas into the East Coast of the
United States beginning in 1973. The size of the project has recently been
increased to 1.5 billion SCF/day. The price of this gas delivered to the
U.S. will be about 58. 5 #/million Btu. This represents a sizable increase
over the price of about 50#/million Btu reported earlier.
The project will require from 12 to 14 large LNG tankers. Contracts
for two of them have been announced. These will be 750, 000 bbl capacity
ships and are scheduled for delivery in late 1974 and late 1975. The
4-9
-------
8929
estimated cost of the original project was S900, 000, 000, but has probably
increased somewhat due to increased tanker costs.
4.1.4.4.2. Philadelphia Gas Works Project
Philadelphia Gas Works and Corporacion Venezolana del Petroleo (CVP)
have signed a 20-year contract to allow PCW to import 500 million Ci'/day
of Venezuelan LNG. Total investment for the project, which should begin
in the fall of 1974, is expected to be S400 million: $200 million for
Venezuelan liquefaction plant and port facilities, $102-$160 million for
three or four 680,000-bbl capacity tankers, and about $40 million for the
Philadelphia port and vaporization facilities. The investment is thus 80#/
million CF of daily capacity.
4.1.4.4.3, Distrigas Project
Distrigas Corp. plans to bring Algerian LNG into the New England area,
primarily for peakshaving purposes. Recently the company asked the
federal Power Commission for permission to import up to approximately
6 billion CF/yr of natural gas in liquefied form over a 20-year period
starting November 1, 1971. The LNG will be bought from Alocean, Ltd.,
a joint organization of SONATRACH and GazOcean International, and will
come from the liquefaction plant at Skikda, Algeria.
To transport the LNG, Distrigas chartered the Descartes which has a
capacity of 1 billion CF of natural gas (300, 000 bbl) and should be in service
in the spring of 1971. Contracts call for the delivery of two shiploads between
August 1, 1971 and March 31, 1972; four shiploads in the 12-month period
starting April 1, 1972; and six shiploads each year thereafter. Shipyard
space has also been reserved for a 440, 000-bbl (1.47 billion CF of natural
gas) tanker for 1973 and a 600, 000-bbl (2 billion CF of natural gas) tanker
for 1974.
4-10
I N S T I T
8929
4. 1.4.4.4. Columbia —Standard Oil Project
Recently, Columbia Gas System, Inc. contracted with subsidiaries of the
Standard Oil Company to import LNG at a rate equivalent to 425 million CF
of gas per day from Venezuela. The gas will be liquefied in a plant to be
built by Creole Petroleum Corporation in a joint venture with the Venezuelan
Government. The plant is estimated to cost $154 million.
The LNG will be delivered to Columbia Gas at its planned facility at Cove
Point, Maryland on the Chesapeake Bay for a price of about 53. 5#/million
Btu which is subject to future escalation. Columbia will buy the supply from
Esso LNG under a 20-year contract for 40. 5#/million Btu and Esso Oceanic
will deliver to Cove Point for an additional 13#/million Btu. It is estimated
that this gas can then be delivered in the Washington D. C. area at prices
around 60^/million Btu.
4.1.4.4.5. Importation on the West Coast
Very serious consideration is currently being given to the movement of
LNG to the West Coast of the U. S. , particularly the Los Angeles area from
the South Coast of Alaska. This does not necessarily involve gas from the
Prudhoe Bay area. There are estimated reserves of about 6 trillion CF
available in Southern Alaska that are presently used to supply the Phillips
Marathon LNG project transporting LNG to Japan. Additional gas could be
supplied to the Los Angeles area by pipelining gas from Northern British
Columbia to the Pacific Coast and liquefying it for transport by sea to
Los Angeles. Both of these alternate sources have been examined. The
Canadian alternative, of course, has the advantage of avoiding the compli-
cations of the Jones Act. Studies made by Arthur D. Little of comparative
costs of shipping LNG from Alaska to Los Angeles indicate that the Jones
Act results in an increase of about 8#/1000 CF.
A comparison of the cost of importing LNG to Los Angeles on three
different bases is presented in Table 4. 1.4. -2. We can see that LNG
4-11
-------
8929
transported from the Cook inlet in Alaska to Los Angeles could be vaporized
at a Los Angeles terminal for about 71 f/ 1000 CF, and LNG transported from
British Columbia to the sea coast at Stewart, B.C. and liquefied there for
transportation to Los Angeles would also cost about the same price. If the
Jones Act could be set aside, LNG could be brought into Los Angeles for
about 63#/1000 CF and would gain a competitive advantage over importation
from British Columbia.
Table 4. 1.4. -2. LNG COSTS AT LOS ANGELES & 500 MILLION SCF/DAY
LNG From
Cook Inlet
(Under Jones
Act)
15.0
--
28.
22.9
5.0
70. 9 '•
LNG From
Cook Inlet
(No Jones
Act)
15.0
--
28.
14.8
5.0
62.8
LNG From
British Columbia
18.0
6.9
28.
12.6
5.0
70.5
Gas Cost
Pipeline Transport to Coast
Liquefaction Cost
Sea Transport
Terminal Costs
Total Cost of LNG
(Vaporized)
4.1.4.5. North Atlantic Continental Shelf
In recent years there has been increasing interest in exploration for oil
and gas off the East Coast of the United States and Canada in areas ranging
from Labrador to as far south as Norfolk, Virginia. The rapid evolution of
deepwater drilling and underwater pipeline technology makes it possible to
give serious consideration to the development of any oil and gas reserves that
may exist in this area. The Continental Shelf is defined as the submerged
part of the continent out to a water depth of 600 ft. Thr North Atlantic
Shelf is approximately 100 miles wide and extends approximately 750 miles
4-12
I N S T I T
89E9
from Norfolk. Virginia to the outer tip of the Georges Bank east of Boston.
This area is larger than the Continental Shelf off Louisiana and Texas.
The geophysical exploration that has been carried out has been primarily
limited to aeromagnetic surveys and no data are currently available as to
the possible yield of gas from the region. Drilling activity in this offshore
area has been delayed because of increasing concern over control of drilling
and well completion techniques. The Canadians have moved more rapidly
and a number of exploratory wells have been drilled offshore Nova Scotia
in the vicinity of Sable Island. These initial efforts have been sufficiently
promising to encourage further exploration. This potential source of
additional gas supplies is a particularly important one for it is the one area
from which gas could be supplied to the large energy markets of the Eastern
Seaboard at prices that would be competitive with present-day gas prices.
4.1.4.6. Synthetic Pipeline Gas
The technology required to make a synthetic gas that is completely inter-
changeable with natural gas is quite different from the technology used to
make the low Btu manufactured gas that has been used by the gas industry
for many years before the advent of natural gas. To produce a high Btu
gas comparable to methane, the waste material must be removed and enough
hydrogen added to increase the hydrogen of the product gas about 25^. This
hydrogen is added chemically and does not appear as free hydrogen.
At the present time there are three pilot plants under construction to
demonstrate the technology and the economics of producing synthetic pipeline
gas from coal and lignite. The projected development schedules will permit
the manufacture and sale of commercial quantities of synthetic gas by about
1978. However, it will be 1980 before large quantities of gas from this
source could begin to make a significant contribution to the nation's gas
supply. It is presently projected that synthetic gas can be produced and
delivered to major market areas for prices ranging from 60-70^/tnillion
Btu depending upon transportation distances and local coal costs.
4-13
-------
8929
4. 1. 5- Supply Deficiency
If the gas available from the various supplemental gas supply sources arc
added to the projected natural gas supply, the total projected supply of gas
still falls short of the estimated future requirements that have been projected
by the Future Requirements Committee. This is shown in Figure 4. 1. 5. -1.
This figure does not include any projection of additional natural gas pro-
duction which could result from an increase in wellhead prices. However,
since it will be several years before such increases in prices will result in
additional production due to the time required for exploration and completion
of production wells, the supply deficiency of natural gas is expected to
gradually increase until 1975. If additional supplies of natural gas can be
brought to market in addition to the projected alternate supplies, the deficiency
can be reduced beyond that date. It is unfortunate that this period of gas
supply deficiency coincides with the period during which it would be most
desirable to have an alternative fuel available for use in automobiles to
serve until such time as pollution control devices and clean engines can be
developed to operate on conventional motor fuels.
4,1."6. Compressed Gas
.Natural gas used to fuel those motor vehicles designed for operation on
compressed natural gas must first be compressed to the very high supply
pressures required for this service.
Normally natural gas is transmitted by pipeline at pressures ranging from
perhaps 300 psi up to 1000 psi or more. The distribution utility will reduce
these gas pressures in the course of distributing gas through its pipeline
network Bo that gas supplied to the ultimate consumer may range in pressure
from a few inches (water column) up to a few psi although some commercial
and industrial users may be supplied with gas at pressures as high as 50 or
60 psi.
In order to store a useful quantity of gas in a motor vehicle, it is necessary
to store it at pressures on the order of 2000 to 3000 psi. Therefore it.is
4-14
INSTITUTE nr r. < ' - r - u .. t. , • .- •
8929
45
40
35
30
s25
5
z
o
20
10
Gas Production
from
Lower 48 States
70 75 80 65 90
VEAR
Figure 4. 1. 5. -1. UNITED STATES GAS SUPPLY
4-15
-------
8929
necessary to compress the gas available at a refueling point from whatever
the local distribution pressure may be to a level of about 3000 psi. This
compressed gas is either stored in a manifold of high pressure tubing con-
tainers or is delivered directly to the high pressure fuel tank installed in
the vehicle.
In supplying compressed natural gas for motor vehicle use it is desirable
to have the compression as close to the point of use as possible in order to
avoid the cost and complications of distributing high pressure gas. For
this reason, relatively small compressors are used at the refueling station
rather than at some central point away from the vehicle; the compressors
used are relatively small in size and compression costs are relatively high.
On the other hand, it is a simple matter to add this compression equipment
to the facilities of a normal automotive fuel service station in order to
permit the servicing of a vehicle with either gasoline, natural gas, or both
fuels.
4.1.7. LNG
The problem of supplying LNG for motor vehicle use is quite different
however. It cannot be liquefied economically in small quantities close to
refueling stations. The liquefaction process is sensitive to the economics of
scale and it must be done in large scale plants in order to achieve the cost
levels necessary to compete with automotive fuels. It must then be distri-
buted by truck or rail transportation in much the same manner as gasoline
motor fuel.
Favorable economics can be achieved in coastal areas using imported
LNG which is produced in very large scale plants and delivered in bulk
liquid form at points close to many large automotive fuel markets. Most of
these large projects will not get underway until about 1975. Nevertheless,
the Distrigas Corporation has already requested permission from the FPC
to import some LNG for this use by 1971.
4-16
8929
However, in the interior of the country, we must rely upon LNG from
domestic liquefaction plants. At the present time there are 13 such plants
in operation as listed in Table 4. 1.7 -1 along with others under construction
or planned. These are gas utility peakshaving plants that are designed to
liquefy and store natural gas during the summer months for revaporization
and use during periods of peak demand in the wintertime.
This is a relatively new activity, for until 1965 there were no such plants
at all. Now there are over forty companies that have built or are planning
to build LNG facilities, and twelve are expanding or duplicating their existing
plants. The average peakshaving plant has a liquefaction capacity on the
order of 5,000,000 CF/day which is rather small to achieve the cost levels
needed to provide LNG at prices competitive with other motor fuels. Those
utilities having LNG available offer it for sale at prices ranging from 8# to
15# /gal. As the use of LNG increases, larger liquefaction plants will be
built and the cost of liquefaction can be reduced. Nevertheless, the peak-
shaving application of LNG is not growing rapidly enough to provide adequate
sources of LNG for motor fuel use in the near future. Such growth would have
to be stimulated by a demand for LNG as a motor fuel.
4.1.8. Price of Natural Gas for Motor Fuel Use
4. 1.8.1. Compressed Natural Gas
Since compressed natural gas for automotive use is obtained by com-
pression of locally available supplies in small compressor service stations,
it is difficult to provide general price data on gas in this form.
Generally a large commercial or small industrial rate will apply to the
gas used and these may range from 35# to 70#/million Btu in various areas.
The compression costs will also vary depending upon the volume handled, the
pressure ratio of compression, and the duty cycle used. Compression costs
on the order of 70# to 90#/million Btu have been reported in some installa-
tions- " Compressor costs cannot be considered alone. For example.
4-17
-------
8929
Table 4.1.7.-1. Part 1. LNG FACILITIES IN THE U.S. AND CANADA
Cor
1.
2.
3.
4.
5.
npany & Plant Site
Alabama Gas Corp.
Birmingham, Ala.
(Sccuii'l tank)
Atlanta Gas Light Co.
Riverdale, Ga.
Baltimore Gas &
Electric Co.
Spring Gardens, Md.
Boston Gas Co.
Boston, Mass.
(Second tank) >
British Columbia Hydro
& Power Authority,
Status 1C6
On-Stream
On-St ream
Under
Construction
Under
Construction
.On-Stream
Under
Construction
Under
. Construction
Capacity.
CF/day
4.65
10. 0
2.75
6.0
-.
2.50
Moragc,
10* CF
625
1000
500
1000
1000
625
10* CF/day
127
--
150
(plus 2 plug-
in units)
187.5
..
150
10.
Vancouver, B. C.
Brooklyn Union Gas Co.
Green Point, N. Y.
(Second tank)
- Citizens Gas * Coke
Utility
Indianapolis, Ind.
Commonwealth Natural
Gas Corp.
Tidewater, Va.
Consolidated Edison
of New York
New York City, N. Y.
Fall River Gas Co.
Fall River, Mass.
On-St ream
Gaz Metropolitatn On-Stream
Riviere des Prairies, Que.
(Seco'nd tank) Under
Construction
Linde Div. of Union
Carbide Corp.
a) Sacramento, Calif.
b) Ontario, Calif.
On-Stream
5.68
Under
Construction
Under 5.0
Construction
Under 5.0
Construction
Under 0.5
Construction
625
1000
1000
150
1000
1000
NA
10
4-18
8929
Table 4.1.7.-1. Part 2. LNG FACILITIES IN THE U.S. AND CANADA
Company & Plant Site
13.
14.
15.
16.
Long Island Lighting
Co. , Holbrook,
L. I. , New York
Lowell Gas Co.
Tewksbury, Mass.
Memphis Light, Gas
ft Water Division
Memphis, Term.
NEGEA-Air Products
Hopkinton, Mass.
Liq
Status 1 0*
Under
Construction
On-Stream
On-Stream
Planned
CF/day
3.0
4.5
5.0
18.5
10* CF
600
1000
1000
NA
10* CF/day
150
60
200
248
(Subject to FPC approval
17.
18.
19.
20.
21.
of purchase from Tenn,
Gas Pipeline Co. )
New England Electric
System, Lynn, Mass.
Northern * Central Gaa
Co. , Hager, Ont.
Northern States Power
Co. . Eau Clair, Wise.
Northwest Natural Gas
Co. , Portland, Ore.
Philadelphia Electric
Co. . West
Planned
On-Stream
On-Stream
On-Stream
Under
Construction
7.85
2.5
2.0
2. 5
6.0
1000
625
270
625
1200
57.6
85
24
120
300
Conshohocken, Pa.
22. Philadelphia Gas Works
Philadelphia, Pa.
(Completed except
for storage tank)
23. San Diego Gas *
Electric Co.
Chula Vista, Calif.
(Addition)
Under 16.0
Construction
On-Stream
Under 7.0
Construction
625
1200
120
120
4-19
-------
8929
8929
Table 4.1.7.-1. Part 3. LNG FACILITIES IN THE U.S. AND CANADA
Under
Construction
On-Stream
Company A Plant Site
<*•!. Southern Conn. Caa
Co., Bridgeport,
Conn.
25. Texas Eastern Trans.
Company, Statcn
Island, N.-Y.
26, Transcontinental Gas On-Stream
Pipe Line Corp.
Carstadt, N. J.
(Second tank) Planned
27. Union Light. Heat * Under
Power Company Construction
(Cincinnati Gas &
Electric Co. Subsidiary)
Erlanger, Ky.
Liq Capacity. Storage, Vaporization.
'0 CF/dav 10* CF 10* CF/day
6. 6
10.08
2H, Wisconsin Natural
Gas Co.
Oak Creek, Wise.
On-Stream
1.0
1.0
• 1200
20-10
1000
1000
13. 7
252
50
4-ZO
a small compressor operated continuously can pump a large reservoir to
the required pressure over a Z4.hr period and serve to refuel a vehicle
daily. Alternatively a larger compressor can pump up the vehicle tank
pressure directly without using a reservoir. The choice of either system or
some combination of the two is a function of the characteristics of the individ-
ual application; the number of vehicles, refueling schedules, tank capacities,
etc.
Regular gasoline costs about 12 to 14^/gal at the refinery level and perhaps
14^ to 16#/gal to fleet vehicle operators before taxes. Each gallon contains
about 120, 000 Btu of energy which is equivalent to about 115 CF of natural
gas. Actual experience has shown that differences in engine fuel economy
permit the use of 100 CF of gas (or 1 therm) as a rough comparison to 1
gallon of gasoline. This means that compressed gas must sell for 12^/therm
or $1.20/million Btu in order to be competitive with regular gasoline in
many areas.
4.1.8.2. LNG
Since LNG is vaporized before injection into an engine, its performance
is comparable to that of compressed natural gas. The San Diego Gas &
Electric Company has reported automobile fuel consumption on LNG of 13
miles/gallon compared to 15 miles/gallon using gasoline in the same
vehicle.*1 On this basis the cost of LNG equivalent to gasoline at
12^/gal is also about $1. 20/million Btu.
It has already been mentioned that imported LNG is projected to cost less
than SI. 00/million Btu and in large quantities it may be available for as low
as 65^/million Btu. This leaves sufficient margin to permit LNG to be sold
in a rather broad area before the distribution expense adds sufficient costs
to exceed comparable prices for gasoline.
The cost of liquefying domestic LNG is another matter, however. Very
large liquefaction plants such as those built in Alaska and Algeria would
permit domestic natural gas to be liquefied at costs of perhaps 35 ^/million Btu.
4-21
I N S T I T U T
-------
8929
Using gas that itself cost 35#/million Btu would produce LNG for about
70^/million Btu — well below the $1. 20/million Btu equivalent to gasoline
at 12#/gal.
However, LNG peakshaving plants.produce LNG at much higher costs
because of their relatively small size and adverse utilization .'actors. These
plants only operate about 200 days per year and have relati/ely high storage
costs because they must hold the total annual production to meet the winter
demand.
A large demand for LNG in the motor fuel market would permit an
expansion of these plants both in size and in utilization factors that would
permit them to approach the large base load importation plants in performance.
However to liquefy gas domestically on the same scale as importation plants
would require quantities of domestic gas comparable to those being imported.
This would tend to defeat the purpose o£ LNG importation.
While this procedure might be feasible economically, it could only be
justified where the use of LNG in motor vehicles is shown to be a more
desirable way of using our natural gas resources than current use patterns.
4-22
8929
4.2. Propane
The supply of domestically produced propane results from two sources.
In 1967, 70£ of propane was produced by natural gas processing plants
(NGL). The balance, 30^ was produced by petroleum refineries (LRG).
In 1969, these figures changed to show that 72£ of propane was produced in
NGL plants and 28
-------
8929
484 in available input supply for propane manufacturing. However, as volumes
of natural gas are increased, a greater percentage will come from the southern
Louisiana fields.
This gas is leaner than the west Texas casing head gas that it will replace.
In the period 1955-1960, the propane and heavier material content of natural
gas was 1.5 gal/1000 CF. In 1966 the content dropped to 0.66 gal/1000 CF
and is expected to be 0.534 gal/1000 CF by 1975. This represents a ZOg
decrease in liquids available per 1000 CF.
Since ethane and propane are not completely extracted in present pro-
cessing, it is reasonable to expect deeper extraction to offset the declining
liquid content. It should be noted that this is not true for butane, whose
current recovery is quite high. The ethane content of natural gas is twice
that of propane. At 30«g ethane recovery, nearly all propane is recovered.
The generally accepted economical limit of ethane recovery is 60$. Since
ethane is considered to be the more desirable feedstock for ethylene, further
extraction is foreseen, with a resultant slight increase in propane yield.
The extent of extraction is limited, however, by the Btu values imposed
by contract price schedules. It is expected that industry averages will
fall far short of 50<{ ethane extraction, and some indications are that a 25o{
level is a realistic figure. For example, the recommended heat value for
southern Louisiana is 10Z5-1075 Btu/CF. After 90<{ propane extraction, die
gas has a value of 1006 Btu/CF; after 30< ethane recovery, the value is only
999 Btu/CF. The penalties on the price of natural gas and the market
value of propane and ethane will determine the economically feasible level
of extraction. The effect on propane pricing can be illustrated as follows:
If natural gas sells for 15rf/mm BTU at the wellhead and has a heating value
in excess of the required minimum, the propane it contains is worth approxi-
mately 1.4#/gal as a component of the natural gas. If the propane in excess
of that required to maintain minimum heating value is removed and sold
separately it has a value of approximately 3.0#/gal or twice its value
4-24
N s i i r
89Z9
as a natural gas component. Thus the producer has an incentive to remove
excess quantities of propane where its cost of extraction is less than the
difference between these two values.
Since the normal price range of propane sold is 3-5i/gal, a sharp
increase in cost of input natural gas could seriously disrupt the economics
of merchandising propane, for the use of propane as a feedstock by the
petrochemical industry is highly dependent on the price and availability of
product.
However, the validity of this argument is dependent on the present existence
of excess quantities of propane in the residual natural gas streams that are
not being charged to the user. The argument loses strength if the gas sold
has a Btu basis for pricing, or if the residual propane is not economically
extractable. The establishment of pricing guidelines along Btu lines is more
common today because of the availability of equipment developed to measure
thermal content with a volume-measuring orifice meter.
4.Z.2. Refinery Production
In refinery operations, production propane and other light hydrocarbons
may be sold as products or used internally as fuel. Thus, production depends
greatly on price because propane must economically justify its alternative
uses. To the extent that natural gas is often the substitute fuel when propane
is sold to the market, the price of natural gas will determine the minimum or
heating value of propane and set the floor level price of propane. The final
limiting factor on refinery production is the physical capacity for producing
propane.
Production of propane from refineries, unlike production from natural
gas, is less dependent on input source volume. Thus, while the volume of
crude is a determinant, the type of processing step plays a more significant
role in determining production quantities.
In 1969, LRG production was 20,000 bbl/day and the amount of propane
produced is on the order of 1. 5g of the crude feedstock.
4-25
-------
8929
4.2.3, LPG Importation
In 1968 the United States imported a total of 236 million gallons of propane,
virtually all of it from Canada. This was an increase of 34^g over 1967.
During the same year 107 million gallons of propane was exported. T-ius,
net imports of LPG are quite small — less than 130 million gallon? arc.ually.
One reason for this is that oil import quotas apply to LPG as well as
heavier oils and serve to restrict importation of LPG. Recent concern for
fuel supplie.s in the United States, however, have caused some revisions in
these oil import quotas and the climate is now more favorable to large scale
importation of LPG.
The most probable sources of propane aside from Canada are those
countries such as Venezuela, Algeria, and Libya that are potential suppliers
of LNG, since LPG is a by-product of natural gas production. Thus, its
availability depends upon the rate of development of the markets for foreign
natural gas — such as importation of LNG into the United States.
The LNG importation programs planned for the U.S. alone could yield as
much as 400 to 500 million gallons of propane annually if maximum amounts
are recovered.
Large quantities of LPG can be imported into the United States if American
buyers can contract for excess LPG from foreign gas fields. There is
currently little history of negotiations for such gas both because domestic
supplies have so far been adequate and because import quotas have discouraged
them. Nevertheless the growing shortage of LPG has stimulated interest
in this question and it is anticipated that such importations will be made in
another year or so provided suitable pricing can be arranged.
The price of propane in New York Harbor ranged from 7. 25^/gal to
8.75rf/gal during 1968 or about 0. 50# to 1.00#/gal less than in 1967. Current
prices are 9# to9.5#/gal.
4-26
8929
4.2.4. Domestic Supply Outlook
In previous supply studies it had been estimated that demand would outstrip
supply by the early 1970's, most likely during the 1972-73 heating season.84
Because of the growing fuel, shortage, this situation will develop earlier than
anticipated. Industry forecasts indicate that we are now on the brink of a
supply shortage. Once establish^d.this imbalance will result in increased
propane prices. This will tend to curtail the consumption of propane in the
petrochemical industry as other users outbid them for available supplies.
This trend is being accelerated by the growing shortage of natural gas and
other fuels that will increase the demand for propane as an alternative to
natural gas or low sulfur oil.
There is general agreement in industry sources that demand will exceed
supply. The area of disagreement lies in the anticipated magnitude of the
situation.
Supplies of LP-gas have increased 25. 5$ over 1966 levels in 1969. Some
industry forecasters feel that supply will increase 10-15< in 1970. up from
the 5-7.5£ increase for 1969. After the 1970 increase, supply rate of
growth is expected to stabilize at 2.5
-------
8929
Q
2
<
W
Q
Q
2
>•
0,
0,
in
U
O
a!
0.
8929
The removal of lead from gasoline would greatly improve propane supply
from refinery sources. Lead removal results in a lower octane fuel which
can only be burned in engines of lower compression ratios than those of today.
The amount of increased propane supply is related to the degree of lead
r ernoval.
A Federally indorsed program (proposed by HEW) provides for the gradual
reduction of lead in gasoline. Following these recommendations California
has passed a bill in the Senate (now before the House) calling for the following
provisions:
1. As of July 1, 1971 each gasoline marketer in California must malfe~—
available one grade of low-lead regular gasoline, probably 91 octane
rating, maximum 0. 5 g lead/gal.
2. July 1, 1974 each marketer must supply one lead-free regular grade
of gasoline, probably 91 octane rating.
3. July 1, 1977 no lead can be contained in any gasoline, premium or
regular.
Before discussing the potential impact of each of the above provisions
certain assumptions must be drawn. These are —
• The product nature of propane will not change — that is, it will remain
a by-product in other refinery processes
• No allowance is made for imports and exports, all production is domestic
e Major uses of propane, by market, will remain relatively unchanged
• This discussion is principally concerned with the engine fuel market
C?lifornia has been the most active in terms of antipollution legislation.
In its role of precedent setting legislation, it is feasible that the program
outlined above will be adopted on a national basis.
Provision (1) presented above is the mildest in terms of change from
present conditions and is the least effective in terms of changing the supply
picture. Low-lead gasoline requires the least deviation from current
production methods and can be made at an additional cost of a fraction of
4-28
4-29
-------
8929
a cent per gallon. Cost at the pump would increase about 1^ or 2# per
gallon reflecting additional distribution costs and added pumps.
The use of 91 octane fuel in a car requires a reduction of compression
ratio from about 9. 0 to 8. 2:1. This results in a reported loss of appr -xi-
mately S< in gasoline economy or between 1 and 2 miles per gallo".. ".iowever,
lead is still present and the car woxild require a thermal reactor to meet air
pollution requirements. A catalytic reactor could not be used since it
would be poisoned by any trace of lead. Due to the pressure drop from the
required air recirculation an additional 15# penalty would be imposed on
fuel economy, resulting in a total decrease in efficiency of 23$. To move
a vehicle the same distance as with standard regular grade gasoline approxi-
mately 23^ more fuel would be required.
Provision (2) requires a gasoline completely free of lead. A recent
Government panel established the cost of producing a completely lead-free
gasoline at 0. S# to 1.5l/gal. The retail price has been found to average
3# above Ike price of regular grades of gasoline, again reflecting additional
marketing and distribution costs. The same loss of fuel economy (8
-------
8929
generally more expensive than alternative methods. Secondly, the price of
propane would be greatly affected.
For propane to become a primary product it would have to carry a greater
share of refinery operating and investment costs. The best way to break
this down would be on a ^/Btu basis in comparison with gasoline. The
following assumptions would have to be made:
• A regular grade of gasoline* contains 113.200 Btu/gal and sells for
13.0rf/gal excluding transportation and taxes.
• A gallon of propane* contains 91.800 Btu and costs 6.25#/gal excluding
transportation and taxes
• All other factors remain the same
.At 13.0#/gal the price of gasoline is equal to $1. 15/million Btu, the price
of propane is equal to SO. 68/million Btu. Assuming the cost burden to be
carried by propane is equal to that of gasoline, the net effect would be to
increase propane prices to about 10<# to ll<*/gal excluding transportation and
taxes. However, a propane price of 8.5#/gal would be equivalent to a
refinery price of 12#/gal considering differences in density and heating value.
Thus, this projected price would represent a significant increase in the
price of propane.
Other important factors also enter into the picture. At this increased
price much of the propane consumed by the petrochemical industry would be
freed for other applications. Ethane would become a more favorable
feedstock. An increase in propane cost may raise the price of other
hydrocarbons, but it is felt the increase would not be proportional to that
of propane.
In 1969 the amount of propane produced by NGL processing plants averaged
535,000 bbl/day. Propane is the largest single by-product of natural gas
processing. Industry forecasts indicate that the amount of propane produced
* Price plus or minus 0.25(1 f. o. b. refinery or terminal (Texas West Coast).
4-32
I N S T I T U ] t
89Z9
will increase approximately 2
-------
8929
approximately 213,000 bbl/day was consumed by the petrochemical industry.
If the price of propane is increased it is expected that petrochemicals would
seek other feedstocks, thus freeing approximately 27$ of domestically
produced propane for consumption in other market segments.
4.2.5 I. Residential and Commercial Markets
The largest sales category is the residential and commercial market
which in 1968 accounted for 41. 5^ of the total sales. This represents an
increase of approximately 7$ over sales in 1967. Industry consensus is
that sales in the residential and commercial area will tend to decline in the
next 10 years.
The pressures causing this trend include natural gas encroachment,
electrical competition for cooking and space heating, and pricing practices
that limit development of this market. The growth pattern of this market
-has tended to follow the supply situation, as is evidenced by 7.2<£ growth
in 1963, 2. 5£ in 1964, 3. 2$ in 1965, and 6. 2£ in 1966.
4.2.E.2. Petrochemical Market
Chemical feedstocks represent the second largest sales category of
LP-gas sales, and 1968 estimates of sales placed it at 38. 7
-------
89Z9
Hope for improvement in this market lies in engine development, steadier
prices, available supplies, and sales based on advantages and performances
rather than price. The demand for air pollution control also enhances the
position of.propane. Industry forecasters expect this market segment to show
the largest increase during this decade. While little or no change is antici-
pated in other segments, the mobile fuel market is expected to increase 200<{
by 1980 or a total of 3. 5 billion gallons. The largest sales increases are
expected to come from urban fleets, fork lift trucks, stationary engines, farm
tractors, and highway fleets. In general, equipment sales are expected to
increase 60< and fork lift sales in the 30-40^ range. See Figure 4.2.5. -1.
In recent years this market has shown the best increase in terms of sales
and has made a turnaround from the sales decline experienced in the mid-1960's.
This decline in sales has been attributed to several major factors: variable
product quality, unsuitable equipment, and unstable prices.
The problem of product quality is related to composition of propane. HD-5
is the accepted industry specification for propane to be used in internal com-
bustion engines. This specification covers a wide variation of product com-
position. The variable products are methane and propylene. The propylene
content can vary 5 percentage points from 0 to 5
-------
8929
This is primarily due to the nonavailability of equipment specifically designed
to run on propane and the CTA does not wish to install engine conversion kits.
Propane price has also been an important factor in its adoption for vehicle
fuel use. The price of propane is very unstable and historically has been
subject to wide variations. As previously mentioned, at low price leve.s large
amounts are consumed by the petrochemical industry. Thus, less product is
available for consumption in other segments. When price begins to increase
larger amounts are made available to other markets, but then the major cost
advantage of propane is lost when compared to other fuel prices. The historical
pattern of propane price is shown in Figure 4. 2 .5 .-2 .
Table 4.2.5:. -1 lists the breakdown by state of propane sales used for
vehicular fuel as reported by the National LP-Gas Association. The states
consuming the largest amounts of propane are either major producers -of
LP-gas or located near pipeline transport depots, indicating the significance
of transport and storage costs in its price structure. Figure 4.2.5 .-3 shows
major pipeline and terminal locations.
On a national basis, propane's use as a fuel for internal combustion engines
can be broken down as follows: industrial fork lift trucks 46.8^, farm
tractors 25.0^, buses and trucks 15. 64, stationary engines 8.4£ , automotive
2.9<
g
I
U
U
3
M
Z
at
U
H
H
W
o
2
0.
I
£
§
H
4-39
N S T I T U T E
-------
8929
Table 4.2.5. -1. PROPANE SOLD FOR VEHICLE FUEL BY STATE
1965 1968
Table 4.2.5. -1. Cent.
State
Alabama
Alaska
Arizona
Arkansas
California
Colorado
Connecticut
Delaware
Florida
Georgia
Hawaii
Idaho
Illinois
Indiana
Iowa
Kansas
Kentucky
Louisiana
Maine
Maryland 81 D. C.
Mas sachusetts
Michigan
Minnesota
Mississippi
Missouri
Montana
Nebraska
Nevada
6,241
21
5,750
62,904
30, 192
11,727
727
423
19,362
10,030
508
617
46, 105
7,580
7,916
42, 738
5, 148
23,027
232
1,416
3,196
4,883
5,397
33,694
6,822
1,926
14,284
1,273
1000 C3.1
10.527
--
6,043
59,098
25,896
17,333
1,256
423
20,293
8, 164
1,067
826
64,480
12,252
9.696
52,598
9,804
3 1 , 3 79
245
1,768
2, 195
8,524
15,469
42,441
12,038
7,907
28,567
739
8929
PROPANE SOLD FOR VEHICLE FUEL BY STATE
1965 1968
New Hampshire
New Jersey
New Mexico
New York
North Carolina
North Dakota
Ohio
Oklahoma
Oregon
Pennsylvania
Rhode Island
South Carolina
South Dakota
Tennessee
Texas
Utah
Vermont
Virginia
Washington
West Virginia
Wisconsin
Wyoming
TOTAL
204
5,789
29.472
5,525
2,488
3.234
13,574
70.860
511
8.654
558
1,805
3,543
7.881
664,403
1.804
136
3,408
2,994
522
5,249
7,065
1,193,818
1000 gal
386
7,651
25,534
6,769
3,563
7,659
17,908
64, 147
1,090
8,287
688
2,530
6,972
13.364
558,440
1,274
54
3,329
2,821
100
9.325
7.579
1,228.498
4-40
4-41
I N S T . I T U T E
-------
8929
8929
increases. Propane usp in the production of synthetic rubber represents
a relatively stable application and can be expected to maintain a gradual
growth rate.
W
H
a
z
in
W
3
u
pi.
A
i
s
sj
o
2 2^
u
i
CO
4-42
4.43
-------
8929
4.3. Ammonia
Although ammonia is not usually regarded as a fuel, it does in fact have
all of the necessary qualities and has been used as a rocket fuel for many
years.
4.3. 1. Product Characteristics
Anhydrous ammonia consists of one part nitrogen and three parts hydrogen
by volume. It is a liquid at —28°F and atmospheric pressure. At higher
temperatures it is a colorless gas with a pungent odor and is lighter than air.
Ammonia has a flammability limit of 16-27«J by volume in air at atmos-
heric pressure. It is completely soluble in water. In a closed system
ammonia is a liquid at pressures at or above its vapor pressure.
Ammonia will vigorously attack copper, silver, zinc, and many alloys,
particularly those containing copper. Iron and steel are generally not
affected in anhydrous ammonia service.
Anhydrous ammonia has various effects on personnel that are dependent
upon its concentration and physical state. Concentrations of 5000 ppm and
above of the gas in air can cause immediate death; 700 to 5000 ppm can
cause burns, blisters, eye damage, and damage to respiratory system
membranes.
However, ammonia is easily detectable at concentrations of 40 to 50 ppm
due to its pungent odor. Since its odor can be readily detected at concen-
trations well below its toxicity level, individuals tend to vacate areas of
vapor concentrations before harmful levels are reached.
The physical characteristics of ammonia are summarized in Table 4. 3.1-1.
4. 3.?» Ammonia Supply
The production of ammonia ia-the.past has utilized a variety of raw
materials as a feedstock. Any substance capable of being converted into
free hydrogen without introducing unfavorable side reactions or impurities
4.44
8929
can be used as a feedstock for the production of ammonia.'4 to the past, coke
from coal was reacted with steam to produce a hydrogen-rich gas mixture.
Some ammonia plants have been built to utilize refinery off-gases which were
rich in hydrogen, and others utilize heavier hydrocarbon feedstocks.
Table 4. 3. J.-l. PHYSICAL CHARACTERISTICS OF ANHYDROUS AMMONI/
Molecular weight
Boiling point
Vapor pressure
Freezing point
Specific gravity
Pounds per gallon
Specific gravity of liquid
CF of gas per pound
CF of gas per gallon liquid
Btu/CF HHV
Btu/CF LHV
Flammability limits. £ Volume in air
17.032
-ZS.O'F
212
-107.9'F
0.6173 (air = 1)
5.14
0.5880 (water =1)
22.28
114.7
434
359
15.5-27.00
The use of natural gas as a raw material for making the ammonia syntbesd
gas mixture has gained wide acceptance throughout the country since its
introduction in the early 1940's. It has become the principal feedstock for
ammonia production in the United States and is used exclusively in recently
constructed plants. Approximately 93£ of the ammonia produced today is
synthesized from natural gas, while the remaining 7{ is produced from coke
oven gas, natural organics, or imported materials. The large scale use of
ammonia as a motor fuel would require other raw material sources than
natural gas since it is already in short supply.
The production of anhydrous ammonia in the United States has continued
to increase following a general expansion trend established during the mid-
4-45
-------
8929
1950's. A reduction in the rate of growth was experienced during 1957 and
1958. The annual rate of growth was between 7-9«g/year. A dramatic
increase took place from 1965 to 1968 when production capacity went from
a little over 10 million tons/year to more than 18 million tons/year in 19&t>.
An expansion rate exceeding 25cg/year was experienced. Figure 4.3.2. -1
depicts the growth pattern of the U. S. anhydrous ammonia industry. The
rate of growth experienced during the mid-601 s will not continue, but will
probably decrease to a more modest 5-6^/year.
02
5 2
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1 -
PRODUCTION
CAPACITY
(No Allowance
For Plant
Shutdown)
PRODUCTION
. Estimated Capacity if
Marginal Producers
C.ease Operations
58 59
60
—I—
61
62
—i—
63
I
64
YEAR
65
66
67 68
69
Figure 4.3.2. -1. PRODUCTION AND CONSUMPTION OF
ANHYDROUS AMMONIA IN U.S.
4-46
I N S T I T I'
8929
There are several reasons for this growth pattern. Among them are oil
industry diversification and production by the chemical industry. Petroleum
companies were seeking ways to diversify and to upgrade refinery products.
'The easiest way to achieve this was to switch to low-cost ammonia production
utilizing refinery by-product hydrogen. This, coupled with the government's
new rapid amortization programs allowing faster tax write-offs enhanced
ammonia production to the industry.
The chemical industry is characterized by strong competition for particular
products, one of them being ammonia. Large capacity plants reduce the
capital investment cost per unit of production. The lower production cost
results in higher profit margins as well as the ability to meet strong com-
petition.
The high rate of growth will not continue. Newer, more efficient plants
are forcing less efficient producers of smaller capacity out of production.
Approximately 3 million tons of yearly production will be lost in this manner.
However, the net increase does not lose its significance when overall capacity
is concerned.
4.3.3. Demand
The amount of ammonia produced is dependent upon agricultural demand.
Approximately 80«J of the ammonia produced in the United States is used for
agricultural purposes. In 1969 approximately 12. 7 million short tons of
ammonia was produced; 10.2 million tons were consumed by agriculture. 1T^Df
this 10.2 million tons approximately 50% is applied to the soil as anhydrous or
aqua ammonia with the remainder applied in combined form with other ferti-
lizers. The rate of increased demand for agricultural application is expected
to stabilize at approximately 7-8$ /year. Figure 4. 3.2.-1 shows total
anhydrous ammonia production through 1969.
The remaining 20jf of ammonia produced is consumed by industries
other than agriculture. Approximately Z500 industrial outlets for ammonia
exist consuming 2. 5 million tons of product. The bulk of this output is
4-47
INSTITUTE
-------
8929
directed toward captive end uses and the remainder is sold for further
processing or direct-vise as ammonia.
Direct uses account for approximately Ijg of all ammonia produced. Largest
direct uses consist of applications in steel treating, sulfite pulping, petro-
leum refining, and fermentation. These outlets are well established and
are expected to grow at the same rate as the general economy.
Ammonia is a raw material used for the manufacture of four major
W
industrial intermediates: nitric acid, amines, hydrogen cyanide, and urea.
Growth in this segment is expected to be the same rate as for direct uses
above.
Production patterns for anhydrous ammonia reflect the seasonal nature of
its principal market, agriculture. Agricultural demands are dependent upon
the field crop and its growing season. The seasonal fluctuation is not aa
severe when observed on a national basis due to the balancing effect of the
variety of crops and seasons. Certain regions, in particular the northern
section of the country, are more affected by the seasonal nature of the
market. Figure 4.3.3. -1 describes the national seasonal fluctuation of
anhydrous ammonia production.
It is-expected that the agricultural market will continue its dominance in
the demand for anhydrous ammonia, assuming there will be no radical
change in the Government's soil bank policy. The Government's liberal
attitude toward high-yield crops on limited acreage provides an incentive
to the farmer to increase the yield per acre, resulting in stimulation of
the fertilizer market. In addition, research tests have shown that nitro-
genous fertilization of forest and grazing lands increases the yield per
acre. It is expected that the growth rate for consumption of fertilizer in
this application will increase and eventually exceed consumption for field
crops. This will broaden the ammonia market and aid in the utilization of
current surplus production capacity.
4-48
1 N <, 1 I I li
ETHNOLOGY
8929
-I—I 1 1 1 T
Sept Oct Nov Dec Jan Feb Mar Apr May Jun Jul
Month
Aug
Figure 4.3.3. -1. SEASONAL FLUCTUATIONS IN
ANHYDROUS AMMONIA PRODUCTION1**
Industrial consumption, although relatively small when compared to the
agricultural market, is looked upon as a stabilizing factor by the ammonia
industry. The market is characterized by a stable rate of growth and it
is not subject to wide variations in demand. It can act to stabilize the
seasonal variation and it is hoped that it will help to utilize surplus pro-
duction.
Anhydrous ammonia production facilities are located in every geographical
region of the United States. However, the percentage of capacity located
within each region varies to a great extent from approximately 0.5^ in New
England to 30. 0$ in the West South Central region. The concentration of
ammonia capacity is directly related to the availability of natural gas as a
4-49
-------
8929
feedstock and ammonia consumption within the region. The natural gas in
this area is relatively inexpensive when compared with other areas and there
IB the second advantage of shipment via water routes. Table 4.3.3. -1,shows
ammonia production capacity by state and Table 4.3.3. -2, shows ammoniJ.
consumption by major consuming states. When considered together it tan be
readily observed how feedstock source and transportation relate to o;ie
another.
Table 4. 3. 3.-1. MAJOR AMMONIA PRODUCING STATES (CAPACITY)
1968
State
Texas
Louisiana
California '
Mississippi
Arkansas
Iowa
Perm sylvania
Nebraska
Illinois
Ohio
Other
1968
Thousand
Short
Tons/Yr
3, 150
3, 150
1,300
1,250
850
800
625
600
525
450
--
4 of u.s.
Production
18
18
8
7
5
5
4
4
3
2
--
17.250
I N S T I T U 1 1
4-50
8929
Table 4. 3. 3. -2. MAJOR AMMONIA CONSUMING STATES
-.162
1967
State
Iowa
Illinois
Nebraska
Texas
Indiana
Kansas
California
Minnesota
Missouri
Washington
Agricultural Ammonia Consumption
Direct Application,^
16.60
14.73
10.25
9.66
5.94
5.83
5.64
4.Z3
3.32
2.51
4-51
-------
89Z9
4.3.4. Price
The actual price paid for ammonia is largely determined by consideration
of the following factors:
• Nature of the ammonia purchaser's business (e.g. distributor.
processor, wholesaler).
• Quantity of product purchased annually
• Season or time of year when purchase is to be delivered
• Prevailing product supply-demand situation at time of purchase.
The ammonia industry is largely oriented toward agriculture. Thus, at
the present time there is no way of determining what effect a nonagricultural
market would have on the industry. An almost universal method of price
adjustment in this industry is the "coproducers discount. " This is the
discount customarily offered by one producer to another. Interpretation of
the term coproducer is very liberal and includes virtually all bulk ammonia
customers. The universal use of discounting procedures minimizes its
competitive effectiveness. However, it frequently is connected to one or
more of the other pricing factors.
The discount may range from $3 to $9/ton reduction in the price of
ammonia. The actual value of the discount is difficult to estimate since it
is related to the other price factors. The seasonal discount is most significant
during the late fall-early winter period. Ammonia is normally produced
on a year-round basis. During the off-season, the product is stored in
producer-owned facilities. In order to increase the overall storage capacity
available to a producer, a seasonal-discount is offered as an incentive to the
distributor or processor to purchase ammonia during the off-season.
This practice is universal throughout the industry. The normal seasonal
discount ranges from $5 to $ll/ton, and in some cases this may be higher.
The prevailing product supply-situation is the most important pricing
factor.
4-52
INSTI! • r " ^ GSS T F ' H H 0 ' '
8929
The quantity of product purchased plays an important role in the price
determination. Almost all bulk purchases involve tank car volumes of product.
Therefore, differentials in delivery costs are very slight. The quantity
discount is a very important consideration in dealing with cooperative asso-
ciations or purchasing groups representing smaller volume users.
The peak demand period for product is from early March to late June.
During this time prices are at their highest level and seasonal discounts
are not employed. If there is not enough product to meet demand, the
discounts discussed take on less meaning; if supply exceeds demand, they
would play a more important part in price determination.
The price of anhydrous ammonia moves over a wide range dependent on
the aforementioned factors. Reported prices range from $50 to $95/ton
and in some instances higher up to $102/ton. This is equivalent to approxi-
'Tnately 13^ - 29rf/gal. However, this price can be misleading. As discussed
in Section 2.4, if ammonia were used as a fuel, 2 to 2-1/2. times the volume of
ammonia would be required (on a per gallon basis) to travel the same dis-
tance as with gasoline. This would more than double the cost of fuel per
mile of travel in comparison to gasoline. A second factor is that the above
price is based on the use of natural gas as a feedstock. Increased production
from other feedstocks would result in higher prices.
4-53
-------
8929
4.4. Hydrogen
Hydrogen is a gaseous fuel like methane or natural gas. It becomes a
cryogenic liquid at—423°F, about 160°F lower than the temperature of LNG
It is the lightest fuel we have, weighing only 4. 37 Ib/CF in liquid form, about
one sixth the density of LNG.
Hydrogen has been used for military purposes both as a rocket engine fuel
and experimentally as a jet engine fuel. For such applications its great heat
energy of 60,957 Btu/lb is an asset. For motor vehicle use, however, its
low density, high cost, and hazardous nature present various objections to
its use.
The physical properties of hydrogen are presented in Table 4.4-1.
Table 4.4-1. PHYSICAL PROPERTIES OF HYDROGEN
Molecular Weigl^ ..
Boiling Point
Freezing Point
Critical Temperature
Critical Pressure
Gas Density @ 68"F Si 1 atm
Liquid Density ® — 423° F
Liquid/Gas Expansion Ratio
Gas Constant
Specific Heat (cp @ STP)
High Heat Value @ 60° F & 30" Hg
Limits of Flammability in air
Max Flame Temperature (air)
Auto Ignition Temperature
2.016
- 423°F
-434°F
-398.8°F
12.8 atm
0.005234 Ib/CF
4.371b/CF
865
66.8 ft/°R
3.42 Btu/lb/°F
319 Btu/SCF
4 to 15%
3990'F
1060-F
4.4, 1- Hydrogen Supply
The hydrogen industry in the United States can be divided into two categories:
I) proprietary production and 2) production for resale. The amount of hydrogen
4-54
|NST|TUTE <- F P AS TECHNO '-COY
8929
produced on a proprietary basis cannot be readily ascertained due to inte-
gration with other processes. Large quantities of hydrogen are produced and
used internally by several industries for annealing, production of chemicals,
treating fats and oils, and hydrocracking. It is assumed that hydrogen
produced internally is not availabl e for resal e and would not enter the gaseous
fuel market.
Hydrogen produced for resale is termed "merchant hydrogen. " Sales in
this category are to small users, whose requirements are not large enough
to support a plant internally, and sales to the Federal Government. There
are four major suppliers of merchant hydrogen in the United States. They
are National Cylinder Gas Corp. ; Air Products; Airco; and Linde, Division
of Union Carbide.
Production of merchant hydrogen in the past 3 years has been on the
decline. This is attributable to the cutback in the aerospace program
resulting in smaller government requirements for rocket fuel. Thirty-one
billion CF of high purity hydrogen was produced in 1969, while production
reached a high of 34. 7 billion CF in 1968; anticipated 1970 production is 27. 6
billion CF. Figure 4.4. 1. -1 shows annual projected production up to 1975.
?s
I 8 - 1
*•" 1^
fr ^
I960 1969 1970 1971 1972 1973 1974 1975
Year
Figure 4.4. 1. -1. ANNUAL MERCHANT HYDROGEN
PRODUCTION, high purity
4-55
-------
8929
The largest hydrogen producing facilities are in Florida and California
indicating the influence of the aerospace industry on production patterns.
Much of the hydrogen produced is made from natural gas since this offers
the lowest raw material and process costs of the several methods available.
It takes about 500 SCF of methane to produce 1000 SCF of hydrogen. Stated
somewhat differently it requires 500,000 Btu of methane energy to produce
319,000 Btu of hydrogen energy. Since we have inadequate supplies of methane
already, it does not appear wise to use them to produce hydrogen for engine
fuel. Other sources would have to be used. The cost of hydrogen from these
119
sources along with methane is presented in Table 4.4.1.-1.
Table 4.4.. 1. -1. COMPARISON OF THE MANUFACTURING COSTS
OF HYDROGEN FROM SEVERAL PROCESSES
Cost,(t/lO'SCFH,
Raw materials
Utilities .
Labor at $4/hr
Overhead, 60$ of labor
Maintenance and operat-
ing supplies, 4g of
capital investment
Fixed charges at 9^
Total, #/103SCF
Steam Steam
methana naphtha
Partial Coal gasification
oxidation .using steam and
reforming reforming of fuel oils oxygen
12.4
4.0
1.2
0. 7
1.7
3.8
23.8
20.6
4.2
1.2
0.7
2.0
4.5
33.2
13.8
5.1
1.4
0.8
2.1
4.6
27.8
Capital investment, $
16.0
3.8
2.8
1.7
4.6
10.3
39.2
millions
5.8
Based on hydrogen production, 40 million standard cubic feet; hydrogen
delivery pressure, 1700 psig; fixed charges, 9«£
Natural gas at $0.25/IO6 Btu.
c Naphtha at $20/ton
dFueloil at $2/bbl; 150,000 Btu/gal.
' Bituminous coal at $5/ton; 12,700 Btu/lb.
Includes electricity, cooling water, and boiler feedwater.
8929
The most probable source is the partial oxidation of fuel oil. It should be
noted that the cost of 27.8^/1000 SCF is based on a price of $2.00/bbl for
fuel oil. Current prices are $3. 00 and may well go to $3. 50/bbl. At 3. SO/
bbl the above cost of hydrogen would increase to about 38.2^/1000 SCF. On
this basis hydrogen from natural gas would still be competitive at a gas cost
of 54#/million Btu based on the above data. Thus, unless gas prices to this
market are increased to above 54^/million Btu.hydrogen can continue to be
made from methane as the lowest cost method. As long as gas is used as the
raw material for making hydrogen, the increased use of hydrogen for motor
fuel will drain larger quantities of natural gas from the supplies available
than would the use of natural gas directly.
4.4.Z. Prices of Hydrogen
The above problem need not be of great concern since the price of hydrogen
in any case is much too high to be competitive with other alternative clean
fuels.
The above production cost of 38.2^/1000 SCF is equivalent to $1.20/million
Btu for gaseous hydrogen at the production plant. It must still be liquefied or
further compressed and distributed to the user. Because of the low density of
hydrogen, the distribution cost is relatively high and results in high prices to
the user when compared with other fuels on a cost/million Btu basis.
The price paid for merchant hydrogen varies inversely with the amount used
over a specified period of time, usually one month. All prices are f. o. b.
storage terminal or plant. The price of liquid hydrogen ranges from 36^/100
CF when purchased in tank wagon lots of 7800 gallons, to $1. 70/liter (29.65
CF) in small volume containers. Gaseous hydrogen ranges from $2.50/100
CF, in high-pressure cylinders to 37. 5#/100 CF in 80, 000 CF tube trailers.
The price of 36#/100 CF is equivalent to about $11. 25/million Btu for liquid
hydrogen while the 37. 5#/100 CF for gaseous hydrogen is slightly higher.
Other volume price schedules are presented in Table 4.4.2. -1.
4-56
4-57
' N 5 T I T U T F
-------
8929
Obviously the cost of distributing hydrogen to small users becomes
prohibitive whon compared to other alternative gaseous fuels.
Table 4.4.2.-1. MERCHANT HYDROGEN COST
99.9 percent purity
113.4 CF/gal
Liquid hydrogen*(delivered in 7800 gal 36<*/100 CF
tank trucks to customer storage (40.8#/gal)
Gaseous hydrogen delivered in tube type trailers
(80,000 CF, average capacity)
0-20,000 CF/month $2. 00/100 CF
next 20, 000-80, 000 CF/month 0. 75/100 CF
next 80,000-3000 CF/month 0.45/100 CF
next 300,000-600,000 CF/month 0.38/100 CF
Note: Customer is allowed to unload gas as used, no charge for trailer
storage.
Gaseous hydrogen delivered in high pressure
steel cylinders
(56" high, 9"dia. 191 CF/cylinder 2000 Psi)
0-2000 CF/month
next 2000-5000 CF/month
next 5000-10,000 CF/month
next 10,000-20,000 CF/month
Price does not include cost of customer storage.
Plus monthly cylinder rental fee-
$2.50/100 CF.
1.50/100 CFt
1.20/100 CF
1.05/100 CF
8929
5. STORAGE AND HANDLING OF GASEOUS FUELS
5.1. Natural Gas
5. 1.1. Bulk Storage
Most of the bulk-storage facilities of the natural gas industry are under-
ground. The large gas holders ti:at can still be found in many large cities were
usually built as accessories of a loi~al manufactured gas plant. Since the gas
industry has now converted almost completely to natural gas supply, these
manufactured gas plants are no longer in operation. The gas holders them-
selves are no longer needed, but many of them are kept in service as a means
of maintaining steady gas pressures under the fluctuating loads that may be
experienced during the daily load cycle. Few, if any, of these gas holders
have been built in recent years, and reliance is placed upon line pack and
underground storage. In 1968, the gas industry stored approximately 1.4
trillion CF of natural gas underground, usually near the downstream end of
gas transmission pipelines, in order to be able to serve the winter heating
loads without overtaxing the capacity of transmission lines themselves. This
quantity is about 10$ of the total amount of gas distributed by transmission and
distribution utilities. These underground storage facilities usually consist of
either depleted gas wells or other underground formations that are capable
of holding gas under pressure for indefinite periods of time. A relatively
small but increasing amount of gas is stored in liquid form as LNG, usually
in aboveground metal tanks.
5.1.2. Compressed Gas
The gas industry stores small amounts of gas under pressure either by
packing transmission lines or by filling high-pressure pipe and tubing mani-
folds where the gas can be held as a cushion against load fluctuations. Line
pack is a technique of building up gas transmission line pressures by pumping
gas into the line faster than it is used by customers at the burner tip. As the
load fluctuates during a daily cycle, the line pressure may fall as the gas
consumption exceeds the pumping capacity of the gas compres sors and
4-58
5-1
-------
89 Z9
recover when consumption drops below pumping capacity. Where the
transmission line or the distribution pipeline network itself does not provide
sufficient storage capacity to take care of such fluctuations, additional
volume can be added by building piping or tubing manifolds that are connected
to the piping system and pumped to high pressure during periods of low
demand for use when the demand exceeds the capacity of the pipeline distri-
bution system.
5.1.3. LNG Storage
During the last 6 years, the gas utility industry has begun to build
liquefied natural gas storage facilities. The first of these was placed in
operation in 1965 by the Wisconsin Natural Gas Company at Oak Creek,
Wisconsin. Since then, approximately 13 LNG peakshaving facilities have
been placed in operation; over 40 gas utilities at the present time either have
in operation or have planned some form of LNG storage facilities. A tabu-
lation of these facilities is presented in Table 4. 1. 7.-1.
This is an aspect of the gas utility industry that is growing quite rapidly.
Of the 13 utilities having facilities in operation, 12 of them have announced
plans for additional facilities or for additions to existing ones. The purpose
of these plants is to hold natural gas from summer to winter. During the
summer months when the demand for gas is low and pipelines have excess
capacity, natural gas is liquefied at a relatively slow rate in a liquefaction
plant that might be operated for a period of perhaps 200 days during the year.
This LNG is accumulated in a storage tank for use during the very cold days
of the winter when the demand for natural gas increases sharply. Generally
speaking, there are from 10 to 20 days per year when unseasonally cold
temperatures create exceptionally high demands for natural gas. By using
these peakshaving LNG plants, the gas demands can be met without
requiring the additional pipeline capacity necessary to serve this relatively
infrequent load.
Geographically,these peakshaving facilities may be located anywhere
there is a sufficient fluctuation in demand from summer to winter to justify
5-2
8929
their use. Table 4. 1.7. -1 shows such plants in operation in Birmingham,
Alabama,and Memphis, Tennessee,as well as in Wisconsin, New York, and
Massachusetts. Since they contain LNG in some amount at all times, they
provide a nucleus of a distribution system for LNG for other purposes such
as motor fuel use.
At importation points for base-load.LNG along the coastline, it is necessary
to install large receiving tanks to hold the LNG delivered by tankers. The
tanks must have 1-1/2 to 2 times the capacity of the LNG tanker serving the
facilility. This amounts to 1 million barrels of liquid or more. At the
present time one such facility is in operation, two are under construction.and
it is anticipated that six or eight more unloading facilities will be built on the
East Coast during the next several years. Possibly two more will be built
on the West Coast as well.
These unloading facilities will have a much greater capacity to supply
LNG in the coastal areas of the country than peakshaving plants. However,
their construction and financing is justified by long-term contracts to supply
LNG to utilities for normal utility purposes. Any large supplies of LNG
from these facilities for motor fuel use will require either expansion of the
presently projected facilities or diversion of LNG from planned utility uses.
5.1.4. Vehicular Storage
5.1.4. 1. Compressed Natural Gas
Compressed natural gas is stored in high-pressure steel cylinders
identical to those used for the storage of other compressed gases such as
oxygen, nitrogen, argon, etc. These cylinders are designed to hold gas
at about 2200-psi pressure and to withstand the rigors of handling in normal
industrial use.
The ICC Type3AA cylinder is used in many current vehicle installations.
At 2000-osi pressure, a Type-3AAA2265 tank will hold only 250 CF of oxygen
but. because of the compressibility characteristics of natural gas, it will hold
about 312 CF of natural gas. This is. roughly equivalent to 3 gallons of
5-3
ft S T I T I T f
-------
8929
gasoline and will provide a range of about 45 miles; thus, more than one
such cylinder is usually installed. Each one weighs about 130 Ib when
filled and occupies over 2 cubic feet of space.
It is necessary to use most of the trunk space in a passenger-type a-'to-
mobile to install sufficient capacity for a 100-mile range. However, f-is is
not a serious problem in short-range vehicles,which ordinarily do not utilize
the trunk space. In commercial vehicles more space is usually available
and cylinders can be buried in inaccessible locations since no access is
required once they are installed. It is general practice to leave the cylinders
installed in a-vehicle and to refuel by connecting high-pressure lines from
a service station supply reservoir or gas compressor.
These cylinders may cost in the neighborhood of $60.00 each and comprise
a major portion of the cost of converting a vehicle to natural gas. Since the
tanks are already made in significant quantities, it is not likely that mass
production of natural-gas-fueled automobiles would result in substantial
reductions in this cost unless radically new types of pressure vessels are
developed for this service.
5. 1.4.2. LNG
5.1.4,2.1. Transportation Vehicles
LNG is transported in vacuum-jacketed steel tanks mounted on
trucks and semitrailers. They are similar in most respects to the liquid
oxygen and liquid nitrogen trucks and trailers that have been in use for many
years. Two sizes have been popular: 6000-gal and 11,000-gal capacities
The larger size represents the maximum-size semitrailer that can be used
in most states. In some instances, this size trailer is weight limited as
well as size limited.
These trailers are designed to hold an internal pressure of about
70 psi before a relief valve will open to prevent further pressure increases.
Normal transfer operating pressures are about 30 psi.and liquid is usually
transferred by vaporizing the product liquid to increase tank pressure to
5-4
8929
30 psi to force liquid from the vehicle. The vacuum jacket is designed to
provide a sufficiently low boil-off rate to permit the trailer to stand for
48 hours without venting gas through the relief valves.
LNG has been transported between Philadelphia and Boston for many
months on regular daily schedules,and less frequently throughout many other states
of the country. At least four companies are presently engaged in the business
of transporting LNG. To date,there have been no reported highway accidents
resulting in LNG spillage, although LNG trailers have been involved in a
half-dozen highway accidents,which is not unreaonable for the mileage they
have accumulated.
Railroad tank cars and river barges can be designed to carry LNG in
the same manner as highway trailers; nevertheless, the economics and
distribution patterns of the industry have not yet required them. However,
tank cars have been designed and at least one barge has been ordered for
operation on the East Coast.
5.1.4.2.2. Vehicle Fuel Tanks
Vacuum-jacketed fuel tanks are also built in small sizes for use as
motor-vehicle fuel tanks. A tank of 15 - 2C gal capacity can be installed
in the trunk of a passenger car to provide a range comparable with con-
ventional gasoline fuel tanks. Like the larger trailer tanks they have low
boil-off rates and can be left unattended for up to 72 hours without venting;
however, they must be vacuum-jacketed to achieve these low boil-off rates.
Plastic foam or perlite insulation alone does not provide sufficient thermal
protection to prevent pressure buildup during a 72-hour storage period. The
72 hour period is sufficient to allow a vehicle to be parked over a weekend or
during a servicing period without the necessity of draining the fuel tank.
At least two manufacturers produce vacuum-jacketed cryogenic tanks,
suitable for use as motor-vehicle fuel tanks. These tanks presently cost
about $400 or more in small quantities. Manufacturers have estimated that
this price could be reduced substantially if production rates were increased
sufficiently to justify hard-tooling and continuous production.
5-5
-------
8929
5. 1.5. Handling Procedures and Practices
5.1.5. 1. Compressed Natural Gas
One of the primary attractions in using compressed natural gas is its
ready availability. Virtually every industrial plant, commercial garage, and
potential refueling station has natural gas service nearby. It is only
necessary to install a compressor and storage reservoir to provide fueling
service. The price of the compressor depends on its rated capacity. A
compressor with a capacity of 8 CF/min is available for approximately $2900,
and the cost of one that would compress 100 CF/min is approximately $12,000.
The equipment required is quite simple and consists of a high-pressure
compressor, a bank of high-pressure storage cylinders,and a valve, fitting
regulators, and hoses to make j.m the system. Operations may be conducted
in one of two ways:
a. When a few vehicles are operated on a regular schedule and can be
refueled overnight, the high-pressure storage cylinders are not necessary
since the natural gas is available directly from distribution lines and the
compressor can be connected to a manifold which feeds the vehicle fuel tanks
Hirprtly, Thp pressure in the vehicle tank is built up overnight,much like
chax-g-wg-.a battery. This system is less convenient,but is more economical
for small fleets.. The slow-charge system can refuel more than one vehicle
at a time if a manifold system is used. The number of vehicles that can be
charged at one time is determined by the size of the compressor. The cost
of such a system capable of refueling 25 vehicles at one time can vary between
$2000 and $3000 .excluding the compressor.
b. When larger fleets are operated or when vehicle refueling schedules
are unpredictable, the compressor is used to maintain a supply of high-
pressure gas in a bank of gas cylinders. The vehicles are refueled from
the high-pressure bank as required.
Some combination of the two systems may also be used.
5-6
I N S T I T
8929
The compressor presents an unusual requirement because of the extremely
high pressure ratio required. Gas must be accepted at less than 1 psi.and be
compressed to 2500 or 3000 psi. A few compressors are available commer-
cially that can operate over this pressure range, but the selection is limited
and costs are high. This is the major cost item in providing compressed
natural gas for motor fuel purposes.
Comparative costs for compressing gas by the slow-charge system are
presented in Table 5. 1. 5. -1. Because of the nature of the systern.no high-
pressure storage facilities are required. The calculations are made using
the following assumptions:
• 20 vehicles in fleet
• Each vehicle driven 50 mile/day
• Gas consumption is 9 mile/100 CF
• 600 CF/day required for each vehicle
• Each vehicle is operated 260 day/yr
• Each vehicle requires 156,000 CF/yr
• Entire fleet requires 12,000 CF/day
• Entire fleet requires 31.2 million CF/yr
• Cost of natural gas is $0. 72/1000 CF
• Fuel taxes not included
• Compressor is driven by natural gas engine
The cost as calculated in Table 5.1.5.-1 will vary according to the type of
application. It is intended to show only the approximate cost based on the
assumptions stated above. The cost of the compression equipment is based
on information obtained from 2 suppliers of CNG equipment. Other costs are
subject to change according to the type of accounting system used by indivi-
dual operating company. Some firms may elect to adopt a longer deprecia-
tion life which would result in a reduction of annual cost. The maintenance
expense is based on the purchase of an annual maintenance contract from the
seller of the equipment. The price of such a contract was found to vary from
SI. 00 to $0.25 per running hour. The cost of the contract is based upon the
size of the installed system. The fleet operator may elect to perform his own
maintenance work and the cost of the system would change accordingly.
5-7
-------
8929
Table 5. 1. 5. -1. CNG COMPRESSION COSTS USING
ENGINE-DRIVKN COMPRESSOR
Capital Cost
Compressor Cost
Manifold Piping & Installation
Operating Cost
Depreciation, 10 yr straight
line, ho salvage value
Return on investment before
tax (15< of investment with
1 0 yr average year-end
amount)
Compressor Fuel Cost
(6CF/100CF compressed.gas
required - cost $0.00432)
* Maintenance expense, on
contract basis per running
hour ® $1.00/hr
Total Cost of Compression
Cost/lOOCF Compressed Gas
Cost of Gas/lOOCF
Total Cost of Gas
Varies according to type of contract purchased.
100 CFM
$12000
3000
S15000
$ 1500
2025
135
520
S 4180
13.4'
7.2
25 CFM
$ 5500
2500
$8000
$ 800
1080
135
1300 @$0.50/hr
$ 3315
10.6(*
7.2
8929
The refueling operation does not require full time personnel. It is based on
the assumption that drivers will connect and disconnect the vehicles to the re-
fueling system. The cost of the natural gas itself will vary in different geogra-
phical locations throughout the country. Other factors which would affect com-
pression cost include on-site storage of pressurized gas and electrically
driven compressor. A storage 'ank capable of holding 2800SCFof gas at a
pressure in excess of 2250 psi is ivailable for $3200 excluding installation,
labor and material.
The high pressure used, from 2000 to 3000 psi, requires some indoctrina-
tion and training on the part of operating personnel. Gas utilities have dis-
couraged broad and indisriminate use of high-pressure equipment by the gen-
eral public and have generally confined their interest to commercial and
industrial fleet vehicle operations where personnel can be properly trained and
supervised.
5.1.5.2. LNG
Unlike compressed natural gas, LNG must be delivered to the vehicle re-
fueling point from some distant location by LNG t a nk trucks. The distribu-
tion pattern is more comparable to the current distribution of gasoline or LPG.
This practice of distributing LNG is already established in the gas utility
industry. A number of utilities operate satellite peak shaving facilities lo-
cated remotely from their main liquefaction plants. LNG is regularly trans-
ported by trailer to these satellite tanks for storage until needed locally for
peakshaving. Although the practice is not yet wide spread throughout the
country, it has been demonstrated to be a safe, practical means of distribu-
ting and storing natural gas.
The refueling of vehicles is similar to the refueling of compressed gas
vehicles except that no high pressures are involved. Some liquid is flashed
to vapor upon filling warm tanks and hoses, but this need not be hazardous for
vented gases are either returned to the system or piped to points from which
they can be safely discharged.
5-8
5-9
I N S T I T U I !
-------
8929
At the present time there are no suitable odorants for use in LNG. The
odorants normally used in natural gas are removed before liquefaction and
replaced only during revaporization of the liquid into a utility system. These
odorants do not function properly over the extreme temperature range of
LXG use. Liquid leaking from an LNG vessel would contain a disproportionate
amount of odorant compared to vapor being vented from the same tank. Thus
the amount of odor may have little relationship to the amount of leakage.
However, some research is being carried on to develop both new odorants
and methods of using present odorants to maintain odorant concentration in
both liquid and vapor phases over the range of temperatures involved. Until
such odorants are available it may not be desirable to park LNG-fueled
vehicles indoors for any length of time that might permit unodorized leakage
or vented gas to accumulate.
The quality of LNG is generally superior to the quality of conventional
natural gas and much higher than LPG. The higher hydrocarbons such as
ethane, propane, pentane, etc. , which are present in natural gas are
largely removed during the liquefaction process. Thus LNG is closer to
pure methane and has a higher octane rating and more consistent physical
properties than most supplies of natural gas. However, during storage, LNG
is continually boiling and selectively thstills the methane vapor out of the
liquid, increasing the proportion of any hydrocarbons which may be present.
This process is called aging, and if carried to extremes could result in
lowering of the octane rating of the fuel.
This is not a serious problem in normal operations because the continued
use and turnover of supplies tend to keep the supply fresh. Also, the octane
rating of LNG is about 115 to 117. This is well above the 90-95 range
presently being advocated for automotive fuels. Thus the LNG would have to
age very severely before encountering trouble of this kind. No reports of any
difficulty due to LNG quality have been found in automolive use. Data
supplied by the San Diego Gas h Electric Company (Table 5. 1.5. -2) shows
there is no appreciable change in the composition of LNG over a period of
5-10
8929
t,
O
O Q,
O
o o o o o o o o o o o o
w 2
o £
2
2 »
w 2
2 T
d, ^ "". ™ "". . "I
z ^
^ o
2 e
Q
- i
5 Ho<>^lnoN-ln=OMo~=o
^ s *• "• °- ^ ~- ^ ^ ^ ^- " ^ °-
(fl "S
o E
I I
2 O
O *
II? z ••-......,..
g OOOOOOOOOMOO
S 8
O B
U
(NJ
Ul
"". ^ o- o~ o-
-H ^, xc >o *o o i^ r^ r^ r*- ^ r— f^ f^
in Q 5'^2ir^oln~~"~~'~~"^~""~"
„ (Mlvjrg — IVJOOO — OOO
n) • -1 —
H
5- 11
-------
8929
time greater than 1 year. The data are based on monthly analyses of the
composition of LNG stored at their liquefaction plant in Chula Vista.
In aircraft use of LNG some concern has been expressed for the amount
of inert material that may be present. Some species of LNG contain signifi-
cant quantities of nitrogen (up to 10$ or so) which are not always re-no- ed.
It is normally of no consequence since gas is sold on a heat energy oasis.
However, it does increase the weight of fuel carried in an aircraft. For
this reason NASA and those aircraft companies experimenting with LNG for
aircraft have taken care to limit the amounts of such inert materials as
nitrogen that may be present.
5.1.6. Safety
5.1.6.1. Compressed Natural Gas
The use of compressed natural gas in a motor vehicle may be safer than
the use of gasoline. The high-pressure cylinder is not likely to be ruptured
in any collison that would not completely destroy the vehicle anyway.
Experience with high-pressure gas cylinders in the industrial gas supply
industry has demonstrated a high level of safety. The major area of concern
is neither the gas cylinder nor the flammability of its contents but the possi-
bility of breaking or accidentally disconnecting a high-pressure line or fitting.
This can be minimized by sound engineering and careful installation. There
is ample experience in the compressed gas industry to provide this knowledge.
However, it is not a problem that should be turned over to the average auto-
mobile mechanic.
5.I.6.Z. LNG
Even LNG may be a safer motor fuel than gasoline. The LNG tank is a
double-walled vessel of much heavier construction than a gasoline tank and
will usually be installed in a better protected location on the vehicle.
Large-quantity spillage should be less hazardous for several reasons:
LNG is more difficult to ignite than gasoline. It is lighter than air and will
5-12
8929
evaporate and rise into the atmosphere rather than settle into pits and sewers.
When it burns the flame is less luminous and radiates less heat to surrounding
metal structures. Once extinquished it is less likely to reignite because of
lower metal temperatures and its own high ignition temperature.
LNG and CNG fuel systems have been found to be as safe as conventional
gasoline fuel systems by insurance underwriters. No incidents of increased
insurance premiums have been reported to IGT.
5-13
-------
8929
5.2. LPG
5.Z.I. Bulk Storage
The seasonality of the demand for propane presents a storage problem to
suppliers in their effort to provide supply in peak-usage months. The ratio
of winter to summer demand may range from 1-1/2:1 to 5:1. Because of
this wide variation, production in excess of demand must be stored during
the slack season. In 1966 stored propane ranged from 368 million to 1.265
billion gal.
Underground storage capacity increased by 13.6.J in the year following the
1965-66 supply shortage. During this time storage capacity was 141 million
bbl (5.922 billion gal). In addition, aboveground storage is increasing, but
this figure — amounting to 200 million gal in 1968, and increasing to 295
million gal in 19 70-can be considered unavailable as it represents pipeline
fill, terminal and bulk plant storage, and industrial worKing stocks. The
increase of 95 million gal in the 3-year period is due largely to additional
pipelines, increased underground storage, and additional capacity of other
types of storage. The bulk of storage will remain underground where
cost/gallon is 5-15# as opposed to 40-50# for aboveground storage.
However, the storage areas are concentratei-at the production end of the
distribution line. For example, the top three states in underground storage
in the beginning of 1969 were Texas, 51. 63 (3, 044, 870 gal); Kansas, 13.6<$
(800,650 gal); and Louisiana, 12.8
-------
8929
5.2.4. Handling Practices
Propane fuel is transported in truck trailers and delivered to pressure
storage tanks at points of u.s_e. in much the same manner as gasoline, LNG,
ammonia, and other volatile liquids. The practice is widespread and long
established. Since propane is capped and sealed during storage and tr-nsport,
there are no vapor losses or emissions from propane fuel systems.
Vehicle fuel systems for propane use are manufactured by a number of
suppliers; natural gas vehicles have drawn heavily from the equipment
made for LFG vehicles. Virtually all of the natural gas vehicles use LPG
carburetors, for example.
Of all the gaseous fuels, propane is probably the most convenient fuel to
-handle because it-can be compressed to liquid at reasonable pressures and
therefore does not require venting provisions or exceptionally high pressure
equipment.
5.Z.5. Safety
While LPG has been used for many years and has established an adequate
safety record.it may suffer somewhat from comparison with both gasoline
and natural gas.
Since it is a compressed liquid it will, upon the release of its pressure,
immediately vaporize more quickly than either gasoline or LNG. An LPG
container will release more combustible vapor than the same size compressed
gas container and is more often used in larger sizes. The consequences of
this have been observed in a number, of propane railroad car accidents in
recent years which prompted a review of transportation practices relating
to propane shipment.
The problem is not as serious in vehicle fuel tanks, however, because
the smaller tanks are less apt to rupture in an accident both because of
their heavy construction relative to gasoline tanks and their more protected
position in a vehicle.
5-16
8929
5.3. Amnfeonia
Ammonia is moved by truck, barge, train, and pipeline. Large quantities
are transported by barge and train shipment from the major producing area
to storage and distribution terminals in major consumption areas. The basis
for selection of a transport method is proximity to water terminals, availability
of equipment,and distance. The least expensive form of transport is by barge.
Barges can be brought from Brownsville, Texas, as far north as Minneapolis
and St. Paul. A typical refrigerated barge has a capacity of 2500 tons
(972,500 gal). Some considerations regarding barge shipment include size
of receiving terminal and accessibility to receiving terminals. Some inland
waterways cannot accommodate fully loaded barges.
Rail shipment of ammonia is the second most economical form of ship-
ment. Many variables affect the cost of rail shipment. Among them are
size of tank car, storage tank size, and degree of equipment utilization. The
seasonality of ammonia use requires high utilization of equipment for a
short period of time. Many leasing arrangements are for long-time periods
and the possibility of idle equipment for a long period may not make rail
shipment attractive to the producer. The expanding LPG markets limits
the availability of rail car equipment. Both commodities can be shipped in
the same equipment after the car is purged.
Truck shipment of ammonia is limited to small quantities for a relatively
short distance, usually 100 miles or less. It is the most expensive form of
transport and is usually restricted to transfers from distributor to end user.
Recent work has been completed on an ammonia pipeline generally
paralleling the Mississippi River to Iowa. It is believed that the cost of
pipeline transport will be one-half of the barge cost. A second advantage of
pipeline transport would be greater dependability and ability to meet seasonal
peak demands. Table 5.3.-1 summarizes the costs of the different types of
transport.
5-17
-------
8929
Table 5. 3. -1. COMPARATIVE COST OF AMMONIA TRANSPORT
Transport Type Costi mills/ton-milej
4
35-60
20-30
Barge
Rail Car — Standard
Jumbo
Truck
.Pipeline (estimated)
60
2
Storage of ammonia is found to be concentrated in the areas of utilization.
Table 5. 3. -2 shows the percentage of storage found in the major consuming
states.
Table 5; 3. -2. LOCATION OF MAJOR AMMONIA STORAGE LOCATIONS
State Percent
Illinois
Iowa
Texas
Nebraska
Louisiana
California
Mississippi
Missouri
Arkansas
Ohio
18.06
13.80
10.07
8.31
7.98
4. 73
4.30
2.98
2.79
2.63
Louisiana is not a major consuming state. Storage here is used to com-
pensate for fluctuations in inventory between barge shipments.
Anhydrous storage is made up of the following types: truck trailers, rail
cars, barges, and stationary tankage. Stationary tankage accounts for approxi-
mately 85< of the volume stored. Approximately 75$ of the agricultural
ammonia produced is sold during the spring fertilizer season of about 1-
I N S T I T I'
5- 18
r. a s
ECHNOLO
8929
month duration. Therefore, approximately 50< of all ammonia produced
must be stored somewhere in the distribution network for 11 months. The
size of stationary storage tanks ranges from 30,000 to 100,000 gallons.
Size selection is determined by sales and demand requirements within the
geographical area.
Ammonia is readily available in the United States, based upon the present
production capacity and demand situation. Any increased demand would not
produce a need for increased production capacity. New applications and uses
are continually being sought, particularly in nonagricultural areas, to absorb
excess capacity and reduce cyclical fluctuations in production.
The extent of its adoption as a motor fuel cannot be readily determined.
Limited research has been carried out to determine the feasibility of fueling
an internal combustion engine with ammonia. Conclusions from this research
indicate that the use of ammonia as a fuel will not become popular.
Equipment required to use ammonia in an internal combustion engine does
not vary a great deal from the equipment required for propane. However, as
discussed in Section 2.4 of this report, the efficiency of the engine is reduced
appreciably and the amount of fuel required (approximately 3 times the amount
of gasoline) makes it impractical for vehicular fuel. No investigation has
been made regarding the practical application of ammonia as a fuel.
5-19
I N 5 T I
-------
8929
5.4.' Hydrogen
5.4.1. Storage and Handling
Merchant hydrogen accounts for the greatest share of transport. Pro-
prietary hydrogen is produced onsite and does not require transport over any
appreciable distance. Large quantities of liquid hydrogen are transported
over the road in trailer-tractor combinations. Pipelines have been proposed,
but the volume requiring transport and its reliability do not justify such a
scheme over very long distances. Other methods of transport, which can be
justified only by larger volumes, are not required. The cost/unit volume
of transporting hydrogen is very high due to its low density and the extremely
low temperature requirements.
Large quantities of hydrogen are distributed by the manufacturer from his
plant or bulk storage facility. For example, in the Chicago area, liquid
hydrogen, produced and shipped from California, is stored in bulk storage
facilities. Gaseous hydrogen is generally produced in the marketing area.
Liquid hydrogen is distributed directly from the bulk storage plant.
The cost of the gas includes delivery charges within 100 miles. Large
quantities are delivered in cryogenic 7800-gal (884,520 CF gross cap. ) tank
trucks with smaller quantities shipped in special vessels with specially
designed flatbed trucks. Large deliveries of liquid are then placed into the
customer's storage facility. In the case of small quantities, the customer
may rent the shipping vessel at an additional monthly charge.
Gaseous hydrogen is transported in one of two ways. Large quantities
are shipped in high-pressure tube trailers with an average capacity of
80,000 CF. The trailer is delivered to the customer and left until it is
empty. Generally, there is no transfer of the gas into storage and no rental
fee for placement of the trailer. Small quantities of gaseous hydrogen are
shipped in high-pressure cylinders. A 56-in.-high, 9-in.-diameter cylinder
has a capacity of 191 CF at 2000-psi pressure. Cylinder gas is purchased
5-20
8929
through a distributor,with prices and shipping arrangements established by
each individual distributor.
Except for certain experimental military applications of hydrogen for
engine fuel, there are no reported uses that provide much insight in handling
and use of hydrogen as an engine fuel. However, the precautions that must
be taken in such industrial applications •> s cooling of electric generators
with hydrogen suggest that it will be a difficult and expensive fuel to handle
where there is close contact with or involvement of persons not trained to
handle hydrogen.
5.4.Z- Safety
The wide limits of combustibility and the high flame velocity of hydrogen
present more severe handling problems than either LPG or natural gas.
While air can be premixed with natural gas, for example, the premixing of
hydrogen and air is more apt to produce explosive mixtures. The flame
velocity of hydrogen is much higher than methane and therefore the hazards of
flashback and flame propagation upstream in flowing gas systems are much
greater. These potential hazards require that hydrogen fuel systems be
carefully purged with nitrogen before filling and precautions must be taken
to prevent air from entering the system under any circumstances.including
emergency repairs or replacement of parts.
Since hydrogen is colorless and odorless.it will require the addition of
odorants as in the case of natural gas. The development of odorants for
liquid hydrogen is probably even more difficult than for LNG.
The above discussion applies to the use of hydrogen alone. When it is
blended with other fuels,it may become much easier to handle. The manu-
factured gas that was distributed by gas utilities for many years contained
as much as 50< hydrogen and yet was distributed and used safely by the
general public.
Until more definitive information is made available on possible blends
of hydrogen and other fuels for automotive use, it is impossible to discuss
their handling and safety problems in any detail.
5-21
-------
B9Z9
6. ECONOMIC IMPACT OF THE USE OF GASEOUS FUELS
The impact of using gaseous fuel upon the petroleum industry, the auto-
motive industry, and the transportation industries will vary depending upon
the scope and manner in which these fuels are used. For this reason, this
subject is discussed in two different phases. First, the impact is dis-
cussed on the basis of large-scale conversions that would be necessary to
provide substantial relief to the overall air pollution problem. In this case,
it is necessary to include consideration of the private automobile. Second,
the impact is discussed on the basis of local commercial fleet vehicle oper-
ations,which would cause less disruption of the fuel supply patterns in the
country, but might provide substantial relief to the air pollution control
problem in congested urban areas.
6. 1. Industrial Impact Based on Large-Scale Cpnversiuns
6. 1. 1. Effect Upon the Natural Gas Industry
In 1968 the United States consumed about 83 billion gallons of motor fuel.
Of this, 58 billion gallons were consumed by passenger cars alone. This
total amount of motor fuel is equivalent to approximately 97 billion therms
of natural gas, or about 80% of the total amount of natural gas distributed
by the gas utility industry. Therefore, if we wished to convert the motor
fuel industry to natural gas, it would be necessary to almost double the
production of the natural gas industry. This is obviously impractical in
view of the problems which the natural gas industry faces in finding suffi-
cient gas to serve its present markets.
fe. 1.2. Effect Upon the LPC Industry
It was explained in an earlier section that the LPG industry is growing
more rapidly than the available supplies of LPG. It is projected that some-
time during the next few years, the demand for LPG will exceed the avail-
able supplies and the increasing use of LPG for pollution control may hasten
the day when this occurs. However, it was also noted that a few cents
6-1
-------
89
-------
8929
are presently being made and used in industry in sizable quantities. There
appear to be no serious engineering problems involved, but there will be
many production problems if the scale of this industry is expanded to pro-
duce the quantities required for large-scale conversion of the automotive
industry to any gaseous fuel.
6. 1. 4. 3. Other Components
The other components such as carburetors, heat exchangers, valves,
regulators, etc., are available commercially in small production quanti-
ties. The production of these components by the various industries that
manufacture them could be expanded to the quantity of production required
for conversion to any of the gaseous fuels.
6. 1. 5. Effect Upon Vehicle Users
The large-scale conversion of automobiles to LPG would nessarily in-
volve the private automobiles. It would cause little change to the vehicle
user except for the increased cost of ownership and operation. He might
experience some difficulty obtaining LPG supplies during the period of
transition, but the range of the vehicle and the refueling and service prac-
tices would be similar. Maintenance costs for the engine and catalytic ex-
haust reactors would be reduced.
The cost of LPC fuel will necessarily be several cents higher per gallon;
however, the quality, availability, and price will be more consistent than
presently experienced in the LPG industry and should be comparable to
present gasoline fuel practices.
The major factor will be the increase in the first cost of the vehicle. If
such vehicles are produced at the factory,the net increase in cost will be
considerably less than the costs of conversion that are discussed in Section 7
of this report. No new components of any consequence are added,and the
major factor will be the net increase in cost of the fuel tank. Large-volume
production will permit some reduction in cost of present LPG fuel tanks,and
6-4
B9Z9
credit can be taken for the cost of present tanks and fuel pumps that are
deleted. The increased cost is difficult to predict without detailed esti-
mates of component costs in volume production, but it might be on the
order of $100' to $150.
This represents an increased cost to the user without any personal bene-
fit other than the resultant reduction in air pollution. Thus he may resist
buying such a vehicle unless legislation requires its use as in the case of
other antipollution devices. This raises the question of how such a conver-
sion program might be implemented. Although that is considered to be
beyond the scope of this study, some observations may be worthwhile.
It does not appear that any large segment of the automobile population
could be converted to any gaseous fuel at an earlier date than the present
automotive industry program to provide a "clean" car by 1975 or 1976 using
lead-free gasoline and exhaust reactor devices.
If that program is successful or promises to be successful in meeting
air quality goals, then there will be little justification for gaseous fueled
vehicles except where they offer economic advantages apart from their low
pollution qualities. If, however, the automotive industry is only partially
successful in meeting the air pollution goals, it may be desirable to con-
sider the use of propane fuel as an additional step necessary to clear the
air.
The information gathered in this study suggests that conversion of the
private automobile to propane will be more expensive than the measures
presently being planned by the automotive and oil industries. The modifica-
tion of the oil refinery industry to produce sufficient propane and the cost
of revising the distribution system to deliver propane in place of gasoline
are much more drastic than making and distributing lead-free gasoline.
Thus, the conversion to propane will be difficult to justify unless the pres-
ent efforts of the automotive industry are demonstrated to be inadequate.
6-5
-------
8929
However, the conversion to lead-free gasoline is a step in the direction
of increasing the yield of propane from refineries. Also, the use of ex-
haust system devices may still be necessary with propane-fueled vehicles
in order to achieve projected pollution levels. Thus the present efforts of
the automotive industry would not be wasted if it became necessary to con-
vert to propane.
6-6
I N S T I T U T t
8929
6. 2. Industrial Impact Based on Commercial Fleet Vehicle Operation
The preceding section discussed the possible conversion of the total motor
vehicle population to gaseous fuels.pointing out the limitations imposed by
the logistics of supply. This section discusses the conversion of commer-
cial fleet vehicles only. This is a much simpler problem for it avoids the
private automobile and its special problems of consumer cost, servicing,
etc. , as well as reducing the magnitude of the logistics problem.
Since about 70% of the motor fuel consumed is burned in passenger cars,
the logistics problem is reduced by 70%. Of course, the benefits from re-
duced pollution using gaseous fuels will apply to only 30% of the total vehicle
population. Thus, this alternative represents only a partial solution to the
problem. Nevertheless, there is considerable interest and activity on the
part of gas and LPG suppliers as well as fleet vehicle operators and govern-
ment air pollution control agencies in applying gaseous fuels to fleet vehicle
operations.
At the present (ime, about 85,000 barrels/day of propane is consumed
in the vehicle engine market. Approximately 48
-------
8929
suited to the fuel used. All of the systems also require installation of one
or more pressure-regulating devices to control the flow of gas between
the tank and the engine.
Some components required for conversion are not the same in all cases
and can be a major cost item in the total conversion cost. An example of
this is the fuel tank required for each of the three forms of fuel. The com-
ponents required for conversion are examined in the following paragraphs.
•'t
Another factor influencing the conversion cost is thedegre-e pf sophistica-
tion employed. Some vehicle owners feel that merely installing the conver-
sion kit without the use of instrumentation is adequate. Others prefer to
modify the engine and adjust it to obtain maximum performance. This is an
additional expense contributing to the conversion cost.
6.2. 1. 1. Compressed Natural Gas Conversion Cost
Conversion kits are available for compressed natural gas for approxi-
mately $300. These kits include the fuel line, two pressure regulators, a
gas-air mixer which replaces the air cleaner, and solenoid valves to control
the flow of gas in the vehicle. Fuel tanks are not included in the conversion
kit. The number of tanks vary in each application according to the distance
each vehicle is required to travel. The cost of a tank of 312 CF capacity at
2265 Ib pressure is approximately $60. One tank will allow the operator to
travel 30 to 40 miles between refills. Approximately 4 hours is required to
install the conversion kit, excluding testing.
The cost of CNG conversion,including one vehicle tank, but excluding com-
pression costs and storage, is approximately $360 per vehicle, although it
may run as high as $500 if additional tanks are used and the vehicle is
modified to improve performance.
6-8
8929
6.2.1.2. LNG Conversion Cost
The equipment necessary for converting an engine for LNG use is the
same as that required for CNG with the addition of a vaporizer, since the
gas must-be transformed from liquid to gaseous form. The cost of a kit,
including the vaporizer, is approximately $350. As in the case of CNG,
the labor and fuel tank are not included. The time required for installation
is the same as for CNG.
The major cost factor in the LNG conversion is the fuel tank itself.
Prices of the tank can vary depending on its construction and capacity. The .
cost, as stated in Section 5. 1. 4. 2. 2, , for a vacuum-jacketed tank is approxi-
mately $400 for a 14-gal tank and higher for larger tank sizes.
The use of LNG as a fuel may require the installation of onsite storage
facilities. This is dependent on the number of vehicles to be fueled and the
proximity of the location to LNG supplies. The cost of onsite storage can
vary a great deal due to tank construction and capacity. San Diego Gas &
Electric Company has installed onsite refueling stations for the California
State Highway Department and the San Diego Zoo for use in its LNG
vehicle operations. The tanks used are rebuilt liquid nitrogen tanks of
1700-gal capacity. The initial cost of these tanks was $2500, excluding in-
stallation costs.
The cost of converting one vehicle to LNG is approximately $750,exclud-
ing onsite storage requirements and labor.
6.2.1.3. LPG Conversion Costs
The equipment required for LPG conversion is much the same as that
for LNG. Estimates of conversion range from $350 to $5pO depending on
the sophistication of the equipment used. Some companies report the use
of a dynamometer in timing the engine after conversion. The added instru-
mentation and time required account for the price variance.
6-9
-------
8929
The cost of the fuel tank is included in the conversion price. However,
the price of the tank can vary from $104 to $200 depending on capacity and
arrangement. A tank mounted in the vehicle trunk with a 19-gal capacity
costs $104, while a saddle-type truck tank is approximately $200,ex' luding
installation.
In summary, the cost of converting a vehicle to LPG fuel is $350 mini-
mumfbut can be more depending on the installation method and type of tank
used. The conversion of an existing vehicle to compressed gas or propane
involves an expenditure of $350 to $500,while conversion to LNG may cost
from $800 to $1000.
6, 2. 2. Economic Incentives
To be economically attractive the user must obtain a reduction in other
vehicle operating expenses sufficient to offset these costs during the life of
the vehicle. Commercial fleet vehicle operators have indicated several
factors that help to make gaseous fuel operations attractive economically.
• Lower fuel costs have been reported on compressed natural
gas and LPG,and JLNG costs promise to be attractive when
imported supplies become available.
• Reductions in fuel taxes in vehicles meeting state pollution
control requirements increase fuel cost savings and help to
offset conversion costs.
• Federal Government purchasing regulations encourage pur-
chase of pollution-free vehicle systems even at a premium
price relative to conventional vehicles.
• In the case of compressed natural gas, elimination of fuel
inventory costs provide additional economic advantages to
large fleet operators.
• Reductions in maintenance expense of 7% have been reported.
• Transfer of conversion equipment items to successive vehicles
reduces conversion cost per vehicle.
6-10
N S T I T II T
8929
A recent development found in the propane industry is the leasing of the
equipment used in vehicle conversions. Adoption of a lease arrangement
is, of course, evaluated on an individual basis. Jn some cases, leasing of
the equipment for a monthly fee over a specified time period can offer the
lessee a tax advantage he could not obtain by purchasing the equipment out-
right. The same applies to acquisition of onsite storage facilities.
If the lease arrangement is selected, the only out-of-pocket cost to the
lessee is the cost of labor involved in the conversion. The monthly cost is
then as low as $8 per vehicle for equipment used in the conversion, depending
on the vehicle.
These factors serve to stimulate interest in gaseous fuel vehicle opera-
tion apart from the resulting low pollution emissions. However, it is diffi-
cult to predict how rapidly the conversion of fleet vehicles to gaseous fuels
will grow. Three factors will govern this rate; all of them are difficult
to predict.
• Shifts in the relative prices of gasoline, compressed natural gas,
L.NG, and propane.
• Additional tax incentives and other legislated inducements
offered by government bodies.
• Progress of the automotive industry in providing a "clean"
car using gasoline fuel.
In the absence of some significant changes in these factors, the rate of
conversion of commercial fleet vehicles is not likely to tax the ability of
fuel suppliers or equipment manufacturers to serve the market.
Even the complete conversion of commercial vehicles in an urban area
to gaseous fuels can reduce the overall pollution level by less than 30^ be-
cause of the limited number of vehicles involved. It can be estimated that
there would be a 15$ to 25% reduction in HC, CO, and NOx emissions during
the decade from 1973 to 1983. These reductions are shown by the shaded
areas in Figures 6.2.Z. -1, -2, and -3, for HC, CO, and NOX respectively.
These shaded areas have been added to figures presented in Reference 188
for Los Angeles County.
6-11
-------
8929
2
I 2000
WW II /
\y
•Vilh no
trol program
^ With present
control progrltr
1950 1»60 I9"0
As of end of vi-tr
1°80 1990
A -51616
Figure 6.2.2. -1. EMISSIONS OF HYDROCARBONS AND OTHER ORGANIC
GASES IN TONS PER DAY FROM MOTOR VEHICLES IN
LOS ANGELES COUNTY
6-12
INSTITUTE
8929
ww 11/
f\y
r
'JViihno •
conrrol program -*•
\
With prvtiiM
"comrol program
i960 1970
As o( cat of VL>«r
Figure 6.2.2. -2. EMISSIONS OF CARBON MONOXIDE IN TONS PER DAY
FROM MOTOR VEHICLES IN LOS ANGELES COUNTY
6-13
-------
8929
Figure 6.2.2. -3. EMISSIONS OF OXIDES OF NITROGEN IN TONS PER
DAY FROM MOTOR VEHICLES IN LOS ANGELES
COUNTY
6-14
8929
It can be seen that the total emissions in tons per day of hydrocarbons
and carbon monoxide from motor vehicles have already peaked and are now
declining as the proportion of controlled vehicles in the total automobile
population increase. Although the emission of nitrogen oxides has increased
as a result of the steps taken to reduce hydrocarbon emissions, it is expected
to peak in 1971 and will then begii, to decrease as nitrogen oxide emission
controls are incorporated into the existing automobile population. Since the
last of these peaks occurs in 1971, it will be impossible for gaseous fueled
vehicles to have any impact upon the maximum amount of emissions pro-
duced and they can only serve to increase the rate of decline in pollution
emissions levels.
The upper dotted curves represent the projected decline in emissions
from conventional vehicles under the present program of emission controls.
The lower (dash line) curves defining the shaded area represent the reduc-
tion in emissions obtained by converting the commercial vehicles to natural
gas or propane.
While the data are presented for Los Angeles County, the form of the
curves is representative of what can be expected in other urban areas since
the uniqueness of Los Angeles is its ability to accumulate emissions in its
atmosphere, not in the rate at which they are generated. Since Los Angeles
controls were imposed about two years ahead of Federal Control Regulations
the time scale must be shifted to be representative of other urban areas.
The ultimate level of emissions that is finally achieved will be deter-
mined by the emission control standards that are established and enforced
rather than by the means of achieving those standards. Since additional
ancillary equipment is needed on gaseous fuel vehicles in order to meet the
projected standards, there is an economic incentive to provide only enough
equipment to reach the required level and not to exceed it. Thus, the two
curves level off at the sa.ne emissions level which is determined by the
governing control regulations.
6-15
-------
8929
These curved assume that all of the commercial vehicles are converted
to gaseous fuel over a period of several years during which the shaded area
widens. If only a portion of the commercial vehicle population is converted,
the band will be correspondingly narrower. It would be difficult to estimate
whether sufficient fuel is available to serve all of the vehicles represented
by the shaded area in each of our urban areas. However, sufficient fuel
could be made available by reallocation to serve certain urban centers. Each
one must be examined individually because of the variations in energy logis-
tics in various regions of the country.
One method of reducing the amount of gaseous fuel required is the dual-
fuel vehicle.
6.Z.3. Dual-Fuel Operation
Most of the experimental cars convert£>-
incorporated a dual-fuel capability that permits the operator to switch to
normal gasoline fuel even while the vehicle is in motion. This was originally
conceived as an emergency procedure to avoid being stranded away from a
source of gaseous fuel.
However, the switching has been demonstrated to be simple enough to
consider a dual-fuel vehicle in which natural gas is automatically used under
high emission driving conditions such as idle and acceleration while switch-
ing to lead-free gasoline under low emission conditions such as normal cruis-
ing speeds. To date no vehicles have been built to explore this technique for
improving the effectiveness of utilizing the limited supplies of natural gas
that will be available for motor vehicle operation. The amount of time such
a vehicle would operate on each fuel may vary widely due to individual driving
habits as well as traffic conditions and highway patterns.
However, by assuming that the vehicle is operated according to a known
pattern of operation, for example, the New Jersey Cycle, it is possible to
estimate the amount of time the vehicle will utilize gaseous fuel and gasoline.
It is known that transient modes of operation contribute the major portion of
vehicle emissions. Therefore, operating with gaseous fuel only in those
modes would contribute to a significant reduction in emissions at a savings
in gaseous fuel consumption. Based on the transient time indicated in the
above cycle, approximately 47Jt of the vehicle operations will be with gaseous
fuels.
Since the modes of the cycle using gaseous fuel will involve higher than
average fuel consumption, the quantity of gaseous fuel burned will be
greater than 50< of the total.
A precise estimate requires a knowledge of the fuel consumption for each
type of fuel during each mode of the cycle, a knowledge of the emissions from
each fuel during each mode, a knowledge of the optimum combing tiwarofex-
haust emission accessory equipment tha*^^— '•• y*. ^-^t^Iled, and knowledge
of the explicit means of "^l^i******"*^ • Thus a rather involved study is re-
not found in the survey of the literature.
Addi t
^ needed together ^^ further study of the
most favorable vehicle configuratiX^.,^ Ught Qf ,he data accumulated in
this report. This was considered to beVathe scope of the study.
6-16
6-17
I N S T I T U T F
-------
89Z9
7. COMPARISON OF GASEOUS FUEL SYSTEMS
7.1. Emission and Performance
7.1.1. General Discussion
The information contained in Section I ' shows that the slow
burning rate characteristic of ammonia and the knock rating of hydrogen
make both of these fuels much less desirable for automotive use than natural
gas and propane. In the case of amn.onia, slow combustion results in
emission of unreacted ammonia which is an offensive pollutant. With a hydro-
gen-fueled engine,nitrogen oxide emission levels will be high and cannot be
reduced with proposed catalytic exhaust reactor systems. Therefore only
natural gas and propane are discussed in detail.
The study also reveals that the differences in emissions and performance
between natural gas and propane are slight. Therefore no distinction is
made between them in comparing gasoline engine emissions and performance
with those of these two fuels. If future standards recognize differences in
reactivity of hydrocarbon emissions in causing photochemical smog, then
natural gas will have an advantage over propane.
Simply converting to either propane or natural gas does not reduce
exhaust emissions to the levels established by the 1980 Federal Standards.
To achieve those levels, additional engine adjustments must be made and
exhaust control devices must be added.
The reductions in emissions to be expected from the various steps that
can be taken are summarized below. For comparison purposes, gasoline
emission data from a 1967 Chevrolet equipped with PCV fuel vapor control
and air injection were used as a reference18 and all emissions are presented
as percentages of that level.
7.1.2. Summary of Gaseous Fuel Emission Levels
Case 1. The simple conversion of gasoline-powered vehicles
to gaseous fuels including propane, L*NG, or CNG will
reduce emissions of the reference vehicle defined
above to the following relative levels:
50$ f.i HC emissions
5056 of CO emissions
95$ of NO emissions
7-1
-------
8929
to
Z
2
«m
OQ
, ft
§8
Q i-i
02
S kJ
U) H
2s
ss
Sa
r,
W >*>*V!. WV -«vt^«. -*.
S°3 'US? SSS 22° ^oo i^oo o « .?
z
, c
US rsf f.
. z\ |||t »;
Si .5- £-,cC'"C -^ T o
B'p C '" u S u jj 0 ^ a ir
If IJ |l|{ 51 ; \ \
S~ >
-------
8929
15$ of HC emissions
10$ of CO emissions
20$ of NO emissions
^ x
Case 6. The addition of both an exhaust reactor and an exhaust
gas recirculation system combines to permit better
performance than Case 5, lower hydrocarbon and CO
emissions, and minimum nitrogen oxides emissions a?
well. In this case, the emission levels will be:
15$ of HC emissions
10$ of CO emissions
10$ of NO emissions
Case 7. The farther addition of a catalytic exhaust reactor system
permits some further reduction in emission levels as
follows:
10$ of HC emissions
8$ of CO emissions
5 £ of NO 'emissions
x
A vehicle equipped with all of these devices would still
suffer a power loss comparable to the other two engine
conversions involving an exhaust recirculation system.
These emissions are converted to grams per mile and compared with
the 1S80 Federal Standards presented in Table 2.3.3-1 and also shows the
corresponding emissions from- a gasoline-fueled vehicle using the same
equipment installation's*.
Table 7.1.2.-2. ESTIMATED EMISSIONS IN CM/MILE FOR
GASOLINE AND GASEOUS FUELS COMPARED WITH
PROPOSED 1980 FEDERAL STANDARDS
Emission
Reference Vehicle per Table 7. 1.Z.-1. Case 7
Gasoline Fuel Gaseous Fuel Federal Standard
Hydrocarbons
Carbon Monoxide
Nitrogen Oxide
0.26
3.0
0.5
0.26
1.34
0.25
0.25
4.7
0.4
N S T I I
7-4
G .
89 Z9
It can be seen that although the nitrogen oxide emissions are reduced
to within the 1980 Federal Standards, the differences in emissions between
gas-fueled and gasoline-fueled vehicles using all of this accessory
equipment is not great. We can therefore conclude that in the time period
around 1975 when the automotive industry has met or is close to meeting
the pertinent Federal Standards that the incentive for use of natural gas or
propane will not be very large. The gains to be obtained in using natural
gas fuels are much greater in th^ immediate future when the emissions
from gasoline fueled vehicles are relatively higher. However, the logistics
of supplying natural gas fuels will prevent the use of significant quantities
of natural gas or propane for several years, approximately paralleling
the time period of development of gasoline powered clean automobiles.
7.Z. Logistics Considerations
Of all the gaseous fuels studied only LPG and natural gas are sufficiently
attractive to warrant serious consideration for motor fuel use. Both
ammonia and hydrogen can be eliminated because of their relatively high
cost and limited availability as well as their technical limitations in engine
use. Since both of them currently use natural gas as a raw material to
achieve their present market prices they can be expected to increase in cost
if produced from other raw materials in order to avoid depleting natural
gas reserves.
Since LPG and natural gas are produced from the same gas fields.they
are subject to the same limitations as far as future increases in supply are
concerned. Both the natural gas industry and the LPG industry are being
hard pressed to supply our current needs for these gases in existing markets.
Any large additional quantities of gaseous fuels for motor vehicle use must
come from some other source. Fortunately we have other sources to consider.
Both LPG and LNG can be imported from countries overseas such as
Algeria, Venezuela. Libya, Nigeria.and the Middle Eastern countries which
have large reserves of natural gas and natural gas liquids. However, the
transportation facilities to move large quantities of these gases into the
United States are currently nonexistent. The natural gas importation projects
presently being plai.ned will provide little excess LNG for motor fuel usage.
In order to obtain financing,each o£ these projects has obtained commitments
7-5
-------
B929
for delivery of its total supply before initiating the project. These commit-
ments have been made to transmission and distribution utilities to supply
existing gas markets. The importation of additional LPG from foreign
countries is currently limited by a shortage of tankers. However, given a
firm market demand for gaseous fuels in the United States, commercial
projects could be organized to deliver significant quantities of both LPG
and LNG into the coastal areas of the United States in a period of four or
five years.
There are other domestic sources for both of these fuels, however,
at the present time about 5 trillion CF or approximately 20$ of our ..
annual natural gas production is sold with the understanding that it will be
delivered only when it is available and that the supply may be interrupted
during winter months in order to supply firm heating customers. Much of
the gas sold for industrial boiler fuel uses is sold on this basisi However,
not all of the 5 trillion CF of gas is actually interruptible. Many utilities
have sold interruptible gas to customers who have not provided alternate
fuel burning equipment and they have never been interrupted. Nevertheless,
many industrial plants and electric utilities are equipped to burn coal, oil,
or gas and can switch fuels depending upon their relative cost and availability.
It was shown in Section 3 that half of the natural gas presently burned by
the Consolidated Edison Company in New York City would be sufficient to
fuel all of the commercial fleet vehicles in New York City with natural gas.
It was also shown that the conversion of such fuel use from boiler fuel
consumption to motor fuel consumption would result in a decrease in the
overall pollution level. Thus the diversion of interruptible gas supplies
represents a potential source- of natural gas for motor fuel use that should
be adequate to supply the foreseeable growth in this area during the next
several years.
The possibility of increasing the yield of propane from oil refinery
production was discussed in Section 4, This appears to be the only source
of gaseous fuels that has the potential capability of supplying a major share
of the motor fuel requirements of the United States. It would, of course,
take many years to convert from the production of gasoline to the produc-
tion of propane. However, increasing the production of lead-free gasoline
necessitates an increase in the yield of propane because of the shift in
refinery processes necessary to produce higher octane species in lieu of
7-6
N S T I I
89Z9
the addition of tetraethyl lead. Thus some increase in production of propane
can be expected immediately and additional propane could be produced if
the demand for this fuel continues to grow.
It does not appear likely that the operation of gas fuel vehicles will
grow at a pace that will outstrip available supplies of natural gas and LP
gas during the next few years unless some stimulus is provided by legisla-
tive action at the state or federal level. Such legislative action seems
unlikely unless, or until, it becomes apparent that the present program
of converting to lead-free gaioline will not achieve its objectives. Thus
there is a period of perhaps 3 years in which to further develop our
experience and knowledge of the use of gaseous fuels in various types of
motor vehicles before it becomes necessary to make a decision on whether
such fuels must be used more widely in order to solve the pollution problem.
7.3. Operating Costs
A comparison of gasoline, diesel oil, and gaseous fuel costs is pre-
sented in Table 7.3-1. The market prices presented are representative of
the fuel prices available to very large fleet operators before taxes. These
prices are shown in cents per gallon and also in cents per million Btu.
In addition the price of each fuel that would be equivalent to gasoline at
$1.04 per million Btu(12< /gal) is also shown. These prices are generally
lower on a cents per gallon basis than gasoline because of differences in
heat content and density.
Using the gasoline data, and assuming a vehicle travels 15 miles per
gallon of gasoline using regular gasoline at 121/gal, it can be calculated
that it will burn 7700 Btu per mile and will require 80<5 worth of fuel to
travel 100 miles. The comparative cost of traveling 100 miles is also shown
for each of the gaseous fuels; first, on the assumption that the same number
of Btu is required and secondly on the basis of reported experience in fuel
consumption with each of the gaseous fuels. On the basis of 7700 Btu per
mile it can be seen from the table that operation on LPG would cost
bbl to 99£ for the range of fuel prices used instead of 80^. LNG would
cost from 54£ to $1.05 per 100 miles for a range of LNG costs from bt to
\U per gallon (70< to $1.37 per million Btu.) Compressed natural gas
7-7
-------
8929
would cost approximately the same amount since its cost in cents per
million Btu is closely comparable to that of LiNG. However, reported experi-
ence with gas-fueled vehicles indicates that actual fuel consumption is lover
than a comparison on an equivalent energy basis shows, A new fuel cost
per 100 miles was therefore calculated for the comparable fuel consumptions
presented in the table. At 13 miles per gallon of LNG it would ccjt -.jjproxi-
mately 46|* to 90|! per 100 miles based on L.NG costs ranging from 6^ to \2i
per gallon. Also, compressed natural gas consumption at 15 miles per 100
SCF would result in a cost of 62rf to 90^ per 100 miles for the price range
shown. A vehicle operating on LPG and obtaining 14 miles per gallon would
incur a fuel cost-of 57 to 86 cents per 100 miles. In the case of each gaseous
fuel the cost per 100 miles of travel ranges from near the cost of diesel
fuel (54^) to somewhere above the cost of gasoline (80)i).
However, these possible savings in fuel cost must be sufficient to off-
set the additional cost of conversion of the automobile to gaseous fuel in
order to be economically attractive. Where the cost of gaseous fuel is near
the low end of the price range as shown in the cable, this can be done. However,
it is much more difficult-to justify corrve.' ^Ic-ns of diesel engines to gaseous
fuel because of the relatively low price of diesel fuels.
Because of variations in patterns of operation, installation costs and othe
other factors it is difficult to generalize on the over-all costs of operation
of gaseous fuel vehicles in comparison to gasoline powered cars. However,
there are literally thousands of LPG vehicles in operation that have been
justified on the basis of economic advantages alone. Similarly there are
over 2000 natural gas-fueled vehicles, most of them using compressed natural
gas, that are presently being evaluated to assess their operating cost in
comparison to gasoline powered cars. Many of these vehicles have shown
economic advantages as well as cleaner exhaust emissions. The cost of
operation of gaseous fuel vehicles is sufficiently close to the cost of operat-
ing gasoline-powered cars that a small change in relative costs can easily
swing the balance from one fuel to the other.
7-8
8929
Table 7.3-1. COMPARISON OF GASOLINE AND GASEOUS FUEL COSTS
Gasoline Diesel Oil LPG LNG CNG
Item
Physical Properties
Sp gr, liquid/water 0.7'. 0,82
Density, lb/gal 6.0 6.8
Btu/gal
Prices
t. /gal-market price* 12 10
-------
8929
7.4. Potential for Early Commercialization
LPC vehicles have been in cornmerical operation for many years and
represent a viable business even though the total population of 250,000 LPG
vehicles represents an almost negligible proporation of the 105 million motor
vehicles in operation in the United States. Natural g?E-fueled vehicles have
only been given serious consideration in the United States in the past three
years, but at the present time, over 2000 vehicles are in operation. These
are largely experimental vehicles being operated by utility companies,
government agencies,and interested equipment manufacturers. However,
because of the economic incentive to use gas-fueled vehicles in order to
comply with pollution regulations, it is anticipated that the market will grow
very slowly until there is some change either in the economic incentive or
in regulatory requirements.
7-10
8929
8. ANALYSIS OF THE USE OF NATURAL GAS IN AN INTERCITY BUS
H. 1. Objective
The objective of this section is to determine the feasibility of using
natural gas as a supplementary fuel in two-stroke cycle, diesel-powered
intercity buses as a means of reducing exhaust pollution while operating in
urban areas and bus terminals. This involves primarily two engines, a
six cylinder 6V-71 engine and an 8 cylinder 8V-71 engine both made by the
Detroit Diesel Division of General Motors Corporation.
8,2, Introduction
Because of fuel supply logistics problems, the most logical user of gas-
fuel-supplemented dicscl vehicles would be a large fleet owner with estab-
lished maintenance facilities at central locations through the region of
operation. It has been recognized that present supplies of natural gas,
LNG, and even LPG may at times become limited which makes it desirable
to consider intermittent operation of such a supplementary system to
reduce emission of smoke and odor only in those areas where pollution is
most severe — in central city operation.
With this in mind, it was agreed to examine the feasibility utilizing
gaseous emission reduction systems on a bus fleet of 6, 000 vehicles
powered by 2-cycle Detroit Diesel \fodel 6V-71N or 8V-71N engines. These
supplementary fuel systems would only be operated within urban areas, to
reduce smoke and odor under city street traffic conditions (low speed, stop-
and-go operation with a high percentage of idle)and in enclosed bus terminals.
Two systems have been developed that utilize gaseous fuels in the diesel
engine and that are known to reduce smoke — the dual fuel and power boost
systems. It is also expected that these systems will reduce odor because
of their more homogeneous combustion, but such data are not available —
indeed direct measurement of odor is not feasible at present. These
systems, as applicable to the 71 series engine,are discussed subsequent to
8-1
-------
8929
the following discussion on current emission requirements and experience
relating to the 71 series diesel engine.
8.3. Emission Control Regulations
8.3.1. Current Status
There are currently no government standards in effect regulating gaseous
emissions from diesel engines. Adoption of regulations specifically for heavy
duty vehicles has lagged behind those for automobiles. There are several
reasons for the lag. Heavy duty trucks comprise less than 4
-------
8929
Q
o
t-
0
cr
0.
o
C5
&
»
O
%
IDLE
i o o 0 •) • •
I 23456 99 10 II 12 13
IDLE A ,,65
7
MODE
Figure 8.3.Z. -1. 13-MODE DIESEL EMISSIONS TEST CYCLE
Under this law, a given engine family must pass the dynanometer mode-
test for state approval. An engine family is established by models having
1) the same combustion system, 2) the same scavenge cycle (2 or 4), and
3) the same method of air aspiration. Models in a family may have different
numbers of cylinders, different configurations (in-line or Vee), and different
bore and stroke if they use the same block. The model tested for state
approval as representative of an engine family must be that unit with the
highest sales volume because the total gaseous emission quantity is the signi-
ficant pollution factor.
There are currently several existing or proposed state and Federal laws
regulating smoke from diesel vehicles. The Federal law establishing Federal
8-4
89Z9
Smoke Certification for engines was enacted in 1968 and sets smoke limits at
204. opacity, with 40<{ opacity allowed for 5 seconds. The testing is made
after a 5 minute idle, during full lug down and during an acceleration.
Measurement is made with a full flow light extinction smokemeter that
measures light obscured by the total plan. The model selected from an
engine family for certification must be the worst-case model, i.e. , have the
highest fuel rate in their family.' Similar regulations exist or are in some
stage of development in various states. Some such regulations call for "no
visible smoke, " which may be defined as 20
-------
8929
(that below the valve)was found to be the determining factor in hydrocarbon
emissions of the 2-stroke Detroit Diesel engines. The "E" series crown-valve
injector had an uncontrolled volume of 175 cu mm compared to the "N" ser.es
needle valve injector with a controlled "sac" volume of 3.5 mm. The
significance of the fuel trapped in the small uncontrolled volume is illustrated
by calculations showing that, for a Model 6V-71N engine operating ^00
rpm, one cu mm of unburned fuel per cylinder entering the exhaust would
result in an hydrocarbon emission level of 1100 ppm.
Experimental evidence indicated that a portion of the uncontrolled volume
of trapped fuel was expelled from the injector by heating and expansion late
in the combustion cycle and the early portion of the scavenge cycle, such that
it passed into the exhaust only partially burned. This type of emission was
greatly reduced through use of needle valve injectors with a minimized
uncontrolled "sac"volume.
8.4.2. Emissions Produced
As a resulUthe current 71N series engines are not smoke limited on power
output. While current Federal smoke standards are set at 20* opacity, the
TIN series smoke density is 2-4* opacity or below the visible threshold of
5-6« for properly maintained shift point rpms.""" Further there is no
concern about gaseous emission levels. It has already been noted that the
diesel is inherently somewhat superior in this respect. The current low
emission levels presented in.Table 8.4.2. -1 - which have resulted from
utilization of "low sac" injectors, and to some extent the catalytic muffler
for hydrocarbon emission control, the catalytic muffler for CO control, and
retarded ignition timing for NOX control - are well below existing standards.
Further, there is confidence on the part of General Motors that anticipated
59 '112
future requirements can be met.
These low level emissions were achieved through introduction in early 1970
of an emission reduction package for retrofit and new vehicle production
assembled by CMC Truck and Coach Division. This package contains the
8-6
8929
catalytic mufflers, further reduced "low-sac" volume injectors, vertical
exhaust stacks, with aspirator for odor dilution and also contains engine
sound proofing components. These modification allowed retarding the
injector timing to reduce NOx formation without adversely affecting other
emissions.
Table 8.4.2. - I.69 EXHAUST EMISSIONS FROMA6V-71 (2V.)
DETROIT DIESEL UIV. ENGINE
(including Exhaust Reactor)
Pollutant
Full Load
1/3 Load
CO g/bhp-hr
HC g/bhp-hr
NOX g/bhp-hr
Smoke £ Opacity
5-50*
0. 1-0.3
5-8
!8£
& 700 rpm
6-9
1.8-2. 7
9-11
H
3 2100 rpm
Range of values over engine speed range tested
Note: Data presented is for 1.490 injector timing; some variations in
relative emissions of each pollutant can be obtained by varying
injection timing.
8.4.3. Anticipated Effects of Converstion to Gaseous Fuels
No data could be found that would permit a direct comparison of the
emissions from a gas-fueled diesel engine using compression ignition with
that for the 6V-71N engine presented in Table 8.4.2. -1. However, a
general comparison of exhaust emissions from diesel and gas dual-fuel or
quasi-otto cycle operation has shown the diesel to have certain advantages.
It has somewhat lower exhaust hydrocarbons, comparable oxides of nitrogen,
and considerably lower carbon monoxide emissions.37It would not be surprising
then for a similar situation to exist for the specfic case of the 6V-71 engine
wit.i gaseous fuels inducted i .ito the intake air. That the uncontrolled injector
volume is the principal source of hydrocarbon emissions indicates that hydro-
carbon emission due to this source would be just as great for pilot injection
8-7
I N S T I I
TECHNOLOG
-------
8929
as for full load diesel operation. Thus dual or supplementary fuel operation
would result from the 30# excess intake air which is short circuited to the
exhaust during the scavenge cycle-. This air would also carry 30< of the
gaseous fuel into the exhaust unburned. A possible solution — the direct
injection of gaseous fuel into the cylinder after completion of the scavenge
cycle — was experimentally tested by Detroit Diesel a number of years ago."'
However, no gaseous emissions data was taken on this or any other Detroit
Diesel dual-fuel engine tests. In fact, experimental work at Detroit Diesel
with dual-fuel conversions ceased about 10 years ago because of the very
limited market prospects for such engines in the mobile vehicle field. Further,
direct injection conversion of the current 71 Series engines is not economically
feasible, as a new cylinder head design would be required.
The greatest potential for emission reduction from gaseous fuels is for
increasing smoke limited power output and reducing exhaust odor as a result
of increased preflame reactions and more homogenous combustion. ' However,
the TIN Series engine is not reduced with the "N" series modifications and
with the vertical exhaust stack-aspirator. Further, no current regulations for
odor are under enforcement, and are not likely for some time because of the
difficulty of odor measurement.
V'
.These factors have thus far deterred the General Motors Corporation, and
other major diesel engine manufacturers as well, from developing gas-fueled
versions of their automotive diesels. This does not however preclude the
conversion of such engines by others.
8.5. Engine Modifications Required
There are several ways in which a diesel engine can be modified to burn
a gaseous fuel such as natural gas or propane. All of them require very
extensive modification of the engine .however. Most of them actually convert
the engine to the otto cycle or some quosi-otto cycle. They are discussed in
turn below.
8-8
I N S T I T I' T I
8929
K.5.J. Conversion tc Spark-Ignition Engines
Virtually all high speed natural gas- or propane-fueled industrial engines
used in stationary engine applications are 4-stroke .spark-ignition engines.
Many of them are conversions of 4-stroke diesel engine designs in order to
take "advantage of the tooling and production facilities available. Although
most of the high speed automotive type diesels offered in this country are
•i-slroke engines, the GMC 71 series engine is a Z-stroke cycle engine.
While it can be converted to a spark-ignition engine, it would be a 2-stroke
spark-ignition engine. Such an engine can be expected to have higher emission
levels than the present engine. A 4-stroke spark ignition engine that
has been modified for use of natural gas or propane could be installed to
operate as a dual fuel engine using gaseous fuel in terminals and urban
areas, but a diesel engine cannot readily be switched back and forth from
a compression-ignition engine to a spark-ignition engine. However, it can
be approximated by modification to a dual-fuel cycle and this arrangement
can be applied to the 2-stroke cycle diesel engine.
8.5.2 Dual-Fuel Conversion
A dual-fuel conversion of the 71 series diesel engine is available on the
market, from Stuart S: Stevenson, a firm specializing in such conversion
work. It is applied to engines installed in stationary service for electric
generation, water pumping, compressor drive, etc.
In the dual-fuel engine, the principal fuel is natural gas or LPG which is
added to the intake air. As in diesel operation, power output is controlled
by varying the amount of primary fuel while the intake air remains unthrotUed.
ignition is provided by the pilot injection fuel, the amount of which is normally
a constant quantity fixed at 5-10=J of total full-load fuel. The dual-fuel engine
can operate at compression ratios and maximum power output levels
approaching the diesel, while operating an order of mapnitude leaner than
the spark-ignited engine/9' M> 8?
Advantages of the dual-fuel engine in addition to possible savings in fuel
costs, are""4"3-
8-9
-------
8929
1) less smoke than the dicsel
2) ICSK carbon monoxide
3) possible reduced engine wear bc'causn of cleaner ongine conditions
4) generally lower rates of pressure rise under nonknocking condition
5) less dilution of lube oil
Present drawbacks that may occur with dual-fuel operation are —
1) a knock-limited power output and, corresponding, a generally reduced
compression ratio and verified efficiency
2) a poor light load (lean) fuel utilization with corresponding poor light load
efficiency
3) extensive modification for diesel conversion to dual fuel may cost an
additional 30-40^! of the original diesel engine price
4) additional fuel supply system must be installed to supply amount of
gaseous fuel required
8.5.2.1. Dual-Fuel Engine Performance
Poor fuel utilization at light load results from appreciable quantities of
gaseous fuel surviving the combustion process when the fuel/air ratio falls
below a certain lean fuel limit that appears to generally correspond to the
flammability limits of gas-air mixtures as extrapolated to engine conditions at
ignition. Experimental evidence suggests that, for these lean mixtures,
flame fronts propagating from various ignition centers of injected fuel fail
to reach all regions of the cylinder, leaving some homogeneously dispersed
in the fringes of the combustion zone. At loads above which gaseous mixtures
exceed this limiting lean concentration, the flame sweeps throughout the mixture
•with corresponding faster heat release and pressure rise. As the load
increases, richer mixtures eventually lead to an upper limit condition of
knock and excessive pressure rise
54,86
As a. result of poor combustion, or even ignition failure at very light loads,
the specific fuel consumption of the dual-fuel engine is much inferior to the
diesel^ however, dual-fuel efficiency may surpass the diesel at full loads.
Associated with this poor fuel utilization, carbon monoxide production may
8-10
, ~ s
8929
increase to an order of magnitude above that of the diesel, but remains
considerably below that of spark ignition engines. Improved light load
efficiency with corresponding reduced emission of unburned hydrocarbons
(predominantly methane)and carbon monoxide, may be achieved by reducing
the lower mixture limits, or by enriching the fuel/air ratio by decreasing
intake air. 10JDownward e.vtenLion of the lower lean limit may be achieved
by-
1) injecting a larger proportion of diesel fuel at light loads, that
assists combustion of the gaseous fuel by increasing the overall
mixture temperature and providing more ignition centers. At
higher loads, diesel fuel injection is decreased to a minimum for
maximum efficiency.
2) injection pressures,as lo-w as will provide satisfactory diesel
operation, used -to prevent too rapid dispersion of the
pilot vapor before autoignition is achieved.
3) advanced injection timing increases fuel residence time in the23'86'87
cylinder and enhances the activity of preflame oxidation reaction.
Knock, produced at the upper (rich) mixture limit, is predominately
affected by the octane rating of the gaseous fuel, the degree of charge preheat,
and, to a lesser extent, by liquid fuel quality.9' ' Knock limited power output
for natural gas fuel may be as high as diesel power rating, 9fD"and high load
efficiency may surpass that of the diesel. ' Due to its lower octane rating,
knock limited power output for propane is about 70^ of that for natural gas.
Reduction of engine compression ratio to accommodate the gaseous fuel
sacrifices some efficiency.29 Studies have shown that the addition of reasonable
quantities of lead alkyl antiknock compounds to propane can increase its
knock limited power output to 30^ above diesel rating.2''54 Retarding injection
timing at full load will also increase knock limited output and improve emission
characteristics. Nevertheless, the addition of lead antiknock compounds is
not recommended if exhaust reactors are installed to reduce NOX emissions.
8.5.2.2 Emissions From £Hial-Fuel Engines
Very little emission data are available on dual-fuel engines, but smoke is
generally reported to be below the diesel.3'89 As mentioned, unburned
8-11
-------
hydrocarbon emissions from dual-fuel engines are largely methane. A
particular test has shown that maximum carbon monoxide production occurs
about mid-load and is of the order of 0.3^. NOX formation is not believed
higher than in the diesel in view of the more homogeneous mixtures and lower
cycle temperatures.
8.5.2.3. Engine Modifications Required for Dual-Fuel Operation
Rather extensive engine modifications may be required and the complete
installation may cost up to 30-40
-------
8929
common boost schemes have no real control over the resulting preQame
product concentrations, success is somewhat of a hit or miss process. As
expected from thermodynamic considerations, brake specific fuel consumr :ion
was slightly higher than the conventional diesel over the low and inte mediate
load range,but improved under heavy load, sometimes surpassing the liesel. '
Poor fuel utilization at light load is analogous to that observed for dual-fuel
operation.23 Benefits gained from fumigation vary widely with speed, generally
medium speed range,and the best performance is achieved at or near full load.
In summary, the extent of benefits achievable with power-boost schemes are
highly dependent on the individual situation.
60
No gaseous emissions data are currently available for power-boost systems.
Power-booster systems are generally frowned on by engine manufacturers
because of the potential to exceed the maximum design power rating which voids
the warranty/08'"2
8.5.4. Stratified Charge Gas Fueled Diesel Engine
Stratified charge engines have been known for many years and have been
used in some special applications where steady loads permit them to be used.
- For several years, the Southwest Research Institute has been conducting
research under the sponsorship of the American Gas Association aimed at
the development of a natural-gas-fueled engine operating on a diesel cycle.
At the present time technical details of the project are considered proprietary
pending completion of future development plans.
However, the method used has been demonstrated on both two- and four-
stroke diesel engines and should be applicable to automotive type engines as
well as stationary engines. No performance data and emissions data are
available nor can the suitability of a 71 series Detroit Diesel engine for
conversion to gas operation by this method be evaluated at the present time.
8.6. Alternatives Available
Of the various conversion possibilities discussed, the dual-fuel conversion
is the most appropriate alternative aside from complete replacement of the
8-14
INSTITl'TT OF ^ /• S r p £ M '.' ~ ' r c v
8929
71 series engines. Even this conversion does not promise great improve-
ment over the present engines and involves considerable expense and incon-
venience.
A dual-fuel conversion of the 8V-71N series diesel is available on the
market. Individual engine conversion cost (less fuel tank and supply lines
on the basis of fleet of 6000 bus = s is estimated to be approximately 26$ of
the cost of the unmodified 8V-71N. The dual-fuel version decreases the
compression ratio from 18.6:1 down to 17 :1. Estimated overall efficiency is
10< below that of the original engine, and has been operated with satisfactory fuel
consumption on the 71 series diesel. Conversion details are considerably
simpler than for a full dual-fuel conversion. Emission data for these
systems are not available from the manufacturer. However, smoke benefits
from these systems are known to be rather unpredictable, being highly
dependent on speed and load. A specific test of 71 series diesels in buses
with fumigation of diesel fuel (induction of atomized diesel fuel) resulted in
significantly decreased smoke only at speeds below the operating range. The
load and speed factors of city traffic are not generally considered to be
conditions favorable to power booster operation.
8.7. Conclusions and Recommendations
In summary, the methods currently available for converting the 6V-71N and
8V-71N diesel engines to the part-time use of natural gas cannot be recom-
mended in view of their ineffectiveness in improving exhaust emission control
and their high costs of conversion and inconvenience of operation. Except
for some reduction in odor, none of the conversions promise reductions in
regulated pollutants below those now reported for the engines when properly
equipped with available pollution control devices offered by the manufacturer.
While several schemes might be developed to accomplish the desired
objective, they will require extensive research and development and may
require replacement of the 71 engine. Such alternatives must be weighed
against the possible use of gas turbines or other new engines as existing
8-15
-------
8929
buses are replaced by new ones. The consideration of such longer range
alternatives is beyond the scope of this study.
8929
9. RECOMMENDATIONS
The findings and conclusions of this study support a recommendation to
further the development and use of natural gas and LPG as motor fuels
wherever logistical considerations permit their use. In line with that gen-
eral recommendation, a number of specific recommendations are made in
regard to problem areas that were uncovered during the course of the study.
These recommendations are presented below in relative order of importance.
9.1. Further information is needed on the reactivity of various types of
hydrocarbon emissions from engines in order to clarify the extent to
which various fias engine emiasions are harmful.
During the study it was discovered that opinions vary considerably
on the contribution to air pollution that has been made by various
species of hydrocarbons emitted from vehicle exhausts. Until suf-
ficient information is available to determine the relative contribu-
tions from various types of hydrocarbons, it is difficult to compare
various fuels on the basis of their pollution potential. In this study
no distinction wag made between various hydrocarbon species since
present testing procedures and emission standards do not provide
for them. Consequently, natural gas and LPG vehicles have corn-
parable emission characteristics. If, as has been reported by many
investigators, methane hydrocarbon emissions do not produce as
much smog as heavier hydrocarbons, this should be accounted for
in determining emission levels. This would make a significant
difference in the choice between LPG and natural gas and might
result in the elimination of exhaust reactor equipment on natural gas
vehicles as well as widening the usable operating ranges of fuel/air
ratio and ignition timing. This would make engine tuning less critical.
8-16
INSTITUTE
9-1
-------
8929
9. 2. It is recommended that an optimized design of an 1C engine for use
of gaseous fuels be built and tested to demonstrate the performance
attainable as well as the emissions produced by such an engine.
This design should take advantage of all the possible ways of -im-
proving engine performance as well as reducing emissions *'.at can
be used with gaseous fuels.-- The resulting engine should then be
tested both on a dynamometer and in actual road tests on a com-
parative basis with a gasoline-fueled engine and with a simple gas
engine conversion, where only the fuel supply has been changed.
The study revealed that additional ancillary equipment must be in-
stalled on gas-fueled engines in order to meet ultimate emission
requirements. The study also suggests that some of this equipment
might be eliminated if the engine were not required to operate on
gasoline and could be optimized for gas fuel use alone.
9. 3. Additional research and development is needed to better determine the
effectiveness of exhaust catalytic reactors and afterburners in handling
the lower concentration of pollutants from natural gas and LPG engines.
During the study it was noted that exhaust catalytic reactors operate
more effectively when the content of the pollutants in the exhaust
stream is high and that they become less efficient when operating on
exhaust streams that are relatively clean. This raises the question
of whether such devices manufactured for gasoline-fueled vehicles
will be well suited to use on gaseous fueled vehicles, or whether
special designs are needed.
INSTITUTE
9-2
GAS
TECHNOLOGY
8929
9. 4. Further research is needed on the automatic dual-fuel concept in
order to determine how to obtain the maximum number of clean vehicle
miles with a minimum amount of gaseous fuel.
It was impossible to make any precise estimate of the reduction in
gas fuel consumption obtainable with automatic dual-fuel use be-
cause of a lack of detail fue1 consumption and emission data for
each mode of a representative driving cycle - particularly for
gaseous fuels. These data should be obtained and a detailed analy-
sis made to estimate the practicality of such dual fuel operation.
This is potentially a very valuable technique for achieving the ex-
tremely low levels of emissions being projected for the future while
at the same time conserving the use of the cleaner fuels as much as
possible.
9. 5. A need exists for odorants that can be used in LNG in the same manner
as in natural gas.
The gas industry currently does not have a strong incentive to
develop odorants for LNG since in utility operations LNG is always
in the custody of a utility company and the revaporized gas is odor-
ized before distribution. However, in automotive use LNG would be
widely distributed and leakage from vehicles and service stations
would not be readily detectable unless a means is developed to add
some identifying odor to LNG. It is not a simple problem because
of the wide range of temperatures involved. A number of schemes
have been proposed but none has been carried far enough to demon-
strate feasibility.
9-3
-------
8929
g. 6. It is recommended that more experimental work be done in measuring
the pollution level from gas vehicles as a function of engine character-
istics and modifications.
During the study much of the data reported on gas vehicles emis-
sions proved to be of little value because they could not be correlated
with specific engine characteristics. Test conditions varied widely
and some significant parameters were not reported. The inability to
compare the data taken on earlier test methods with the results of
more recent measurements has made it difficult to compile com-
parable data that would indicate .the effect of specific engine char-
acteristics and modifications of gas engines on pollution levels.
Thus, there is a need for standardization of reporting practices
as well as the development of some method of correlating earlier
test results with data taken using the new method of measurement.
It is recommended that additional data be solicited on the operating
characteristics of gas and LPG engines.
The data reported in the literature on operating characteristics of
gas-fueled engines are less than anticipated. While there are
several reasons for this situation, the fact remains that informa-
tion of this kind is necessary in order to determine the engine oper-
ating conditions which will minimize harmful exhaust emissions
without penalizing the operating performance of the engine.
This work would be a necessary prerequisite to the project recom-
mended in 9. 2. '
9. 7. It is recommended that additional test data be taken on commercial
and industrial fleet vehicles to provide comparable emission data on
various types of power plants and vehicle operations.
Virtually no data are available on emissions from trucks, buses,
and other fleet vehicle operations that would provide a basis for
9-4
I N S T I T LJ
TECHNOLOGY
8929
comparison of one type of power plant with another. Such informa-
tion, if made generally available, would provide a basis for selec-
tion of fleet vehicle power plants on the basis of pollution control as
well as vehicle performance.'
9. 8. It is recommended that a survey be made of European experience in
gas-fueled vehicles, particularly in France and Italy.
Natural gas has been used in automotive vehicles in France and Italy
for many years. The relatively high cost of automotive fuels in
Europe causes greater concern for fuel efficiency and economy in
the operation of motor vehicles. Thus, there is not only background
of experience in this area, but this experience should be oriented
toward the economical operation of gaseous-fueled vehicles. There-
fore, a survey of vehicle users and equipment manufacturers in
Europe might be a rewarding investigation.
In addition to the technical and economic problems uncovered during the
study, a number of problems relating to logistics require further investiga-
tion. The more important questions are discussed below:
9. 9. It is recommended that additional urban areas besides New York City
be examined to determine the potential of gaseous fuels to reduce
overall pollution levels in other local urban areas,
It was noted in the study of the New York City area that of the three
gaseous fuels considered— propane, compressed natural gas, and
LNG — that LNG offers the easiest solution to the problem of logistic
supply in the New York City area. It is anticipated that similar
studies of other urban areas might show that LPG and, in some areas
compressed natural gas, may offer the best solutions depending upon
the relative availability of these fuels in individual geographic areas.
9-5
-------
8929
9. LO. It is recommended that further studies be made of the feasibility of
increasing the yield of propane from oil refineries at the expense of
gasoline production and the resultant effects upon cost of both fuels
and upon their distribution expense.
The possibility of convening gasoline production to propane pro-
duction at some increase in the cost of propane is the moot prom-
ising possibility for replacing gasoline as a motor fuel if that
were to become necessary to eliminate automotive pollution. Be-
cause of this, we believe that it is very desirable to undertake
further investigations of this potentially important question. The
comments that have been made in this report are based upon the
review of a relatively limited amount of information on this sub-
ject plus the results of interviews with knowledeable people in the
industry. There does not appear to be any publicly available de-
tailed examination of this question,
9.11. There is a need to examine in more detail the availability and potential
price of liquefied petroleum gas imported into the United States from
overseas.
This subject is discussed briefly in the report, but relatively little
information exists on this subject primarily because until the re-
cent energy shortage developed, the United States has been an ex-
porter of propane overseas and there has been little need to import
supplies except from Canada.
9-6
I N 5 T I T U T F
TECHNOLOGY
10.
2.
REFERENCES CITED
8929
Adams, W. E, and Boldt, K. , "What Engines Say About Propane
Fuel Mixtures, " S. A. E. Trans. 73, 718-42 (1965).
Agnew, W. G. , "Future Emission-Controlled Spark-Ignition Engines
and Their Fuels, " Proc. Amer. Petrol. Inat. 49_ Section 3, 242-80
(1969).
3. Alperstein, M. , Swim. W. 3. and Schweitzer, P. H., "Fumigation
Kills Smoke - Improves Diesel, " S. A. E. Trans. 66. 574-95 (1958).
4. Altshuller, A. P. et al. , "Photochemical Reactivity of n-Butane and
Other Paraffinic Hydrocarbons," J. Air Pollut. Contr. Ass. 19,
787-90 (1969) October.
5. Altshuller, A. P. et al., "Products and Biological Effects From
Irradiation of Nitrogen Oxides With Hydrocarbons or Aldehydes Under
Dynamic Conditions, " Air Water Pollut. Int. J. 10. 81-94 (1966).
6. "And Now— What of the Decade Ahead." Butane-Propane News
21-29 (1970) January.
7. Bachman, W. A., "Forecast for the Seventies. " Oil Gas J. 67, 160-
175 (1969) November 10.
8. Baker Sr. , R. A, andDoerr, R. C. , "Catalytic Reduction of Nitrogen
Oxides in Automobile Exhaust, " J. Air Pollut. Contr. Ass. 14, 409-14
(1964) October.
9. Balgord, W. D. , "Novel Catalysts and Catalyst Supports for Control
of NOX in Automobile Exhaust. " Paper presented at the 2nd Int.
Clean Air Congress, Washington, D. C., December 1970.
10, Bartholomew, E. , "Potentialities of Emission Reduction by Design
of Induction Systems. " S. A. E. Paper No. 660109(1966).
11. Bartok, W. e_t al., "Systems Study of Nitrogen Oxide Control Methods
for Stationary Sources, " Final Report GR-2-NOS-69 Vol. 2, 3-63.
Contract No. PH-22-68-55 for the Division of Process Control Engi-
neering, National Air Pollution Control Administration. Esso Re-
search and Engineering Company, Government Research Laboratory,
November 1969.
12. Bascom, R. C. and Hass, G. C., "A Status Report on the Develop-
ment of the 1973 California Diesel Emissions Standards, " S. A. E.
Paper No. 700671 (1970).
10-1
-------
..8929
13. Bascunana, J. L. and Conta, L. D. , "Further Research on Charpe
Stratification, " S. A. E. Paper No. 660095(1966).
14. Battigelli, M. C. , "Air Pollution From Diesel Exhaust," J. Occup.
Med. _5, 54-57 (1963) January.
15. Baudry, J. , "Some Recent Results Concerning the Reduction of Pollu-
tion by the Stratified Charge I. F. P. Process. " S. A. E. Trans. J4_,
662-72 (1966).
16. Baxter, M. C. , "LP-Gas - A Superior Motor Fuel. " S. A. E. Paper
No. 670054 (1966).
17. Baxter, M. C. , Leek III, G. W. and Mizelle, P. E. , "Total Emission
Control Possible With LP-Gas Vehicle, " S. A. E. Paper No. 680529
(1968).
18. Benson, J. D., "Reduction of Nitrogen Oxides in Automobile Exhaust, "
S.A.E. Paper No. 690019 (1969).
19. Blalock, W. S. and Little, R. N. , "Carburetion and Other Factors
Which Affect Propane-Fueled Engines," S.A.E. Paper No. 670058
(1966).
20. Boldt, K. , "Motor (LP) Knock Test Method Development, " Paper No.
670055 in S. A. E. Trans. 75, (1968)
21. Bolt, J. A. and Harrington, D. L. , "The Effect of Mixture Motion
Upon the Lean Limit and Combustion of Spark-Ignited Mixtures, "
S.A.E. Paper No. 670467 (1967).
22. Bonamassa, F. and Wong, H. W. , "Composition and Reactivity of
Exhaust Hydrocarbons From 1966 California Cars." Paper presented
at the 152nd National ACS Meeting, New York, September 1966.
23. Boyce, T. R. , Karim, G. A. and Moore, N. P. "An Experimental
Investigation Into the Effect of the Addition of Partially Oxidized Re-
action Products to the Intake Charge of a Compression Ignition Engine,1
J. Inat. Petrol. 52, 300-11 (1966) October.
24. B roe ring, L. C. , Rose Jr. , A. H. and Werner, W. J. , "Automotive
Mass Emission Analysis by A Variable Dilution Technique." Paper
presented at the APCA Annual Meeting, Cleveland, June 1967.
10-2
8929
25.
26.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
Brown, W. J. et al. , "Effect of Engine Intake-Air Moisture on Ex-
haust Emission. " S. A. E. Paper No. 700107 (1970).
Brownson, D. A. and Stebar, R. -F. , "Factors Influencing the
Effectiveness of Air Injection in Reducing Exhaust Emissions. " Ft. 2
"Vehicle Emissions, " S. A. E. Progr. Technol. \Z_ (1967).
Brubacher, M. L., "Reduction of Diesel Smoke in California, "
S.A.E. Paper No. 660548 (1966).
Bull, M. G. , "Development of an Ammonia-Burning Gas Turbine
Engine, " U.S. Army Eng. Res. Develop. Lab. Rep. AD-671-667.
Fort Belvoir, Va., April 1968.
Cabal, A. V., Felt, A. E. and Williams, Jr., H. A., "Performance
of Leaded LP-Gas in an EMD-567 Dual-Fuel Engine, " S. A. E. Paper
No. 763A (1963).
California Department of Public Health, Technical Report of Cali-
fornia Standards for Ambient Air Quality and Motor Vehicle Exhaust.
Sacramento, Calif., 1960.
California Motor Vehicle Pollution Control Board, California Test
Procedure and Criteria for Motor Vehicle Emission Control. Los
Angeles, 1964.
Campau, R. M. andNeerman, J. C. , "Continuous Mass Spectrometric
Determination of Nitric Oxide in Automotive Exhaust, " S. A, E. Pape r
No. 660116 (1966).
Cantwell, E. N. and Pahnke, A. J. , "Design Factors Affecting the
Performance of Exhaust Manifold Reactors, " Pt. 2 "Vehicfe Emis-
sions, " S._A._E._Prog_r._Te_ctaoli 12(1967).
Chandler, J. M. , Struck, J. H. and Voorhies, W. J., "The Ford
Approach to Exhaust Emission Control, "S.A.E. Paper No. 660163
(1966).
Cole, D. E. and Jones, C. , "Reduction of Emissions From the
Curtiss-Wright Rotating Combustion Engine With an Exhaust Reactor,"
S. A. E. Paper No. 700074(1970).
Colucci, J. M. and Barnes, G. J., "Evaluation of Vehicle Exhaust
Gas Odor Intensity Using Natural Dilution, " S. A. E. Paper No. 700105
(1970).
10-3
-------
8929
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
49.
50.
Cornelius, W., Huellmantel, L. W. and Mitchell., H. R., "Ammonia
as an Engine Fuel." S. A. E. Trans. 74. 29-44 (1966).
Daniel, W. A., "Engine Variable Effect on Exhaust Hydrocarbon
Composition (A Single-Cylinder Engine Study With Propane as the
Fuel)," S. A. E. Paper No. 670124(1967).
Daniel, W. A,, "Why Engine Variables Affect Exhaust HyJrc carbon
Emission," S. A. E. Paper No. 700108(1970).
Davis, S. J., "Instruction in Internal Combustion Engines, " F^roc^
Inst. Mech. Eng. 178. No. 26, 119-40(1963-64).
Deeter, W. F. , Daigh, H. D. and Wallin, Jr. , O. W. , "An Approach
for Controlling Vehicle Emissions." S. A. E. Paper No. 680400 (1968).
Dual Fuel Systems, Inc. , Fact Sheet, n. d.
Duckstein, L. , Tom, M. and Beard, L. L. , "Human and Traffic
Control Factors in Autombotive Exhaust Emission, " S. A. E. Paper
No. 680398 (1968).
Dunne, J. , "Ford's New Smog-Free Engine, " Pop. Sci. 55-57
(1970) May.
"Du Font's Exhaust-Manifold Thermal Reactor System" Oil Gas J. 68,
102 (1970) June.
Ebersole, G. D. , "Hydrocarbon Reactivities of Motor Fuel Evapora-
tion Losses, " S. A. E. Paper No. 690089 (1969).
Edwards, J. B. and league, D. M., "Unraveling the Chemical
Phenomena Occurring in Spark-Ignition Enginp.c, " S. A. E. Paper No.
700489 (1970).
El-Mawala, A. G. and Mirsky, W. , "Hydrocarbons in the Partial
Quench Zone of Flames: An Approach to the Study of the Flame
Quenching Process, " S. A. E. Paper No. 660112(1966).
EJston, J. C. , Andreath, A. J. and Milask, L. J. , "Reduction of
Exhaust Pollutant Through Automotive Inspection Requirement— The
New Jersey Repair Project." Paper presented at the 2nd Int. Clean
Air Congress, Washington, D. C. , December 1970.
Eltinge, L. et al. , "Emission Control Requires Good Carburetion
and Other Engine Modifications, " S. A. E. J. 76. 28-42 (1968)
September.
10-4
T U T F
8929
51. Eltinge, L. , Marsee, F. J. and Warren, A. J. , "Potentialities of
Further Emission Reduction by Engine Modifications, " S. A. E. Pape r
No. 680123 (1968).
52. Eyzat, P. and Guibet, J. C. , "A New Look at Nitrogen Oxides Forma-
tion in Internal Combustion Engines, " S. A. E. Paper No. 660124
(1968).
53. Farrar, G. L. , "Gas-Liqu:ds Production Up Sharply, " Oil Gas J.
(1970) July 13.
54. Felt, A. E. and Steele, Jr., W. A., "Combustion Control in Dual-
Fuel Engines," S. A. E. Paper No. 437B(196l).
55. Fitch, R. E. and Kilgroe, J. D. , "Investigation of a Substitute Fuel
to Control Automotive Air Pollution, " Final Report of Consolidated
Engineering Technology Corporation to National Air Pollution Control
Administration. NAPCA Contract CPA-22-69-70, Washington, D. C. ,
February 1970.
56. Fleming, R. D. , All sup, J. R. and Hum, R. W. , "Propane as an
Engine Fuel for Clean Air Requirement. Part 1 —A Progress Report."
Paper presented at the LP-Gas Engine Fuel Symposium, Detroit,
October 1970.
57. Floyd, C. M. , "A Comparison of the Performance of Highly Detergent
Oils and Highly Dispersant Oils in Natural Gas Engines. " S. A. E. Pape r
No. 680761
58. Ford, H. S. , Merrion, D. F. and Hames, R. J. , "Reducing Hydro-
carbons and Odor in Diesel Exhaust by Fuel Injector Design, " S. A. E.
Paper No. 700734 (1970).
59. Ford, S. , Detroit Diesel Div. , General Motors Corp., private com-
munication* September 1, 1970.
60. Fox, C. and Peyton, J., Marvel-Schebler Div. , Borg-Warner Corp. ,
private communication, December 16, 1970.
61. Fox, C. D. , "LPG- 'Torque-Topper' Development, " Report No.
4022023 Marvel-Schebler Div. , Borg-Warner Corp. (1966) December.
62. Galster, G. M. ?.nd Pigott, J. S. , "Some Effects of Altitude on LP-
Gas Tractor Operation, " Paper No. 670057 in S. A. E. Trans. 75,
(1968).
10-5
-------
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.
8929
Garwin, L. , "LP-Gas Supplies for the Engine Fuel Market." Paper
presented at LP-Gas Engine Fuel Symposium, October 21-22, 1970.
Gerstein, M. et al. . "Flame Propagation: II. The Determination of
Fundamental Burning Velocities of Hydrocarbons by a Revised Tube
Method. "J. ACS 73, 418-Z2 (1851) January.
Gilbert, L. F. et al. . "Oxide of Nitrogen in Engine Exhaust Gas. "
Paper presented at 132nd A. S. E. Meeting, September 1957.
Glass, W., Kim, D. S. and Kraus, B. J. , "Synchrothermal Reactor
System for Control of Automotive Exhaust Emissions, " S. A. E. Paper
No. 700147 (1970).
Glass, W. et al. , "Evaluation of Exhaust Recirculation for Control
of Nitrogen Oxide Emissions, " S. A. E. Paper No. 700146 (1970).
Glasson, W. A. and Tuesday, C. S. , "Hydrocarbon Reactivity and
.the Kinetics of the Atmospheric Photooxidation of Nitric Oxide. "
Paper presented at the 59th Annual Air Pollution Control Association
Meeting, San Francisco, June 1966.
CMC Truck & Coach Div. , General Motors Corp. , A Proposal to
Reduce Exhaust Emission and Decrease Vehicle Noise From Existing
Diesel Coaches. Detroit, 1970.
Golothan, D. W. , "Diesel Engine Exhaust Smoke: The Influence of
Fuel Properties and the Effect of Using Barium-Contining Fuel
Additive, " S. A. E. Paper No. 670092 (1967).
Graiff, L. B. , Legale, C. E. and Robinson, I. C. H. , "A Fast-
Response Flame lonizatiori Detector for Exhaust Hydrocarbons, "
S.A. E. Paper No. 660117 (1966).
Gray, Jr., J. T. et al. , "Ammonia Fuel — Engine Compatibility and
Combustion," S. A. E. Paper No. 660156(1966).
Groth, R. H. and Calabro, D. S. , "Evaluation of Saltzman and
Phenoldisulfonic Acid Methods for Determining NOX in Engine Ex-
haust Gas, " J. Air Pollut Contr. Ass. _!£, 884-87 (1969) November.
74. Hahn. A. V., Petrochemical Industry Market and Economics. 33-39.
New York: McGraw-Hill 1970.
10-6
I N S T I T I 7
8929
75.
76.
77.
78.
79.
80.
81.
82.
83.
84.
85.
86.
Hardison, L. C., "Techniques for Controlling the Oxides of Nitrogen, "
J. Air Pollut. Contr. Ass. 20, 377-82 (1970) June.
Harkins, J. and Goodwine, J. K. , "Oxides of Nitrogen in Diesel
Exhaust. "J. Air Pollut. Contr. Ass. 14, No. 1, 34-37 (1964) January.
Harris, W. L. and Hummel, J. W. , "Effects of Inlet-Air Tempera-
ture and Humidity on Performance of an LP-Gas Engine. " Paper
No. 670056 in S. A. E. Trans. 75. 1968.
Hittler, D. L. and Hamkins, L. R. , "Emission Control by Engine
Design and Development. " S. A. E. Paper No. 680110(1968).
Huls, T. A. and Nickois, H. A., "Influence of Engine Variables on
Exhaust Oxides of Nitrogen Concentrations From a Multi-Cylinder
Engine, " S. A. E. Paper No. 670482 (1967).
Huls, T. A., Myers, P. S. and Uyehara, O. A. "Spark-Ignition
Engine Operation and Design for Minimum Exhaust Emission, "
S.A.E. Paper No. 660405(1966).
Institute of Gas Technology, Seminar on Natural-Gas-Fueled Vehicles
(December 4, 1969) Chicago, 1970.
Jackson, M. W. , "Effects of Some Engine Variables and Control
Systems on Composition and Reactivity of Exhaust Hydrocarbons, "
S. A. E. Paper No. 660404 (1966).
Jackson, M. W. and Nebel, G. J. , "Some Factors Affecting the
Concentration of Oxides of Nitrogen in Automobile Exhaust Gases, "
J. Air Pollut. Contr. Ass. J5, 213-19 (1958) November.
Johnson, D. R. , "Free World Supply/Demand of LP-Gas and
Natural Gas." NLPGA Times. 18-21 (1969) Summer.
Kaleskas, E. W., "Performance of a Low-Emission Vehicle Oper-
ating on Propane. " Paper presented at the LP-Gas Engine Fuel
Symposium, Detroit, October 1970.
Karim, G. A. , "The Ignition of a Premixed Fuel and Air Charge
by Pilot Fuel Spray Injection With Reference to Dual-Fuel Combus-
tion, " S. A. E. Paper No. 680768 (1968).
10-7
-------
8929
87.
88.
89.
90.
91.
92.
93.
94.
95.
96.
97.
98.
99.
100.
Karim G A . "On the Emission of Carbon Monoxide and Smoke From
Compression Ignition Engines: Including Natural Gas-Fueled Engines. •
Paper presented at the 2nd Int. Clean Air Congress, Washington, D. C. ,
December 1970.
Kaufman, F. and Kelso, J. K. , "Thermal Decomposition of Ni' ric
Oxide, " J. Chem. Phys. 23, 170Z-07 (1955) September.
Khan, J. M. , "Formation and Combustion of Carbon in a Diesel
Engine, " S. A. E. Paper No. 70073S (1970).
Klosterman. D. L. and Sigsby, Jr., J. E. , "Application of Subtractive
Techniques to the Analysis of Automotive Exhaust. Presented at the
National ACS Meeting, Atlantic City, September 1965.
Kopa, R. D., "Control of Automotive Exhaust Emission by Modifica-
tion of the Carburetion System, " S. A. E. Paper No. 660114(1966).
Kiuvanen, G. J., West.a, R. O. ^d Lien, D. G. , '-Engine-Fuel
Relationships With LPG and Natural Gas, " S. A. E. Paper No. 670054
(1966).
Landen, E W. , "Nitrogen Oxides and Variables in Precombustion
Chamber Type Diesel Engines," j. A. E. Paper No. 714B (1963).
Lee. J. J. , "Ammonia Line Has Unusual Features, " Oil Gas J. .66.
103-108 (1968) November 4.
Lee. R. C. and Wimmer, D. B. , "Exhaust Emission Abatement by
Fuel Variations to Produce Lean Combustion, " S. A. E. Paper No.
680769 (1968).
Levedahl, W J , "Multistage Autoignition of Engine Fuels, "
Fifth Symposium (Int. ) on Combustion, 372-85, Reinhold 455.
Linden. H. R. . "U.S. Natural Gas Supply and Demand, " Public_
UU1. Fortn. 81, 13-18 (1968) February 1.
Linnell, R. H. and Scott, W. E., ''Diesel Exhaust Composition and
Odor Studies."^. Air Pollut. Contr. Ass. 12, 510-14, 545 (1962)
November.
Littler. D. L. and Hamkins, L. R. , "Emission Control by Engine
Design and Development, " S. A. E. Paper No. 680110(1968).
"LP-Gas Industry Future Up to Refiners. " Oil Gas J^ 68, 40-41
(1970) May 18.
10-8
INSTITUTE
TECHNOLOGY
8929
101. Maga, J. A. and Hass, G. C. , "Present and Future Emission
Standards for Heavy-Duty Vehicles. " S. A. E. Paper No. 690765(1969).
102. Maga. J. A. and Kinosian. J. R. , "Motor Vehicle Emission Standards
Present and Future, " S.A. E. Paper No. 660104(1966).
103. Manning, J. , Stewart & Stevenson Services, Inc., private communica-
tion and letter of September 17 and December 16, 1970.
104. Marshall, W. F. and Hu-n, R. W. , "Factors Influencing Diesel
Emissions." S. A. E. Paper No. 680528(1968).
105. Matsumoto, K. , Toda, T. and Nohira, H. , "Oxides of Nitrogen From
Smaller Gasoline Engine," S. A. E. Paper No. 700145(1970).
106. McConnell, G. , "Oxide of Nitrogen in Diesel Engine Exhaust Gas:
Their Formation and Control," Proc. Inst. of Mech. Engrs. 178,
Pt. I. No. 38, 1001-13 (1963-64).
107. McJones, R. W. ind Corbeil , R. J., "Natural Gas-Fueled Vehicles
Exhaust Emission and Operational Characteristics, " S. A. E. Paper
No. 700078 (1970).
108. McLaughlin, E. J. , Pinotti, P. L. and Sigworth, H. W. , "Power
Booster Fuels for Diesel Engines. " Paper presented at the S. A. E.
Natural Fuels and Lubricants Meeting, Chicago, October 1951.
109. Mehne. J. T. and Carufel, L. D. , "LP-Gas Engines for the 1970's. "
Paper presented at the LP-Gas Engine Fuel Symposium, Detroit,
October 1970.
110. Mengelkamp. R. A. and Hudson, A. C. , "LP-Gas Fuel System
Affects Engine Performance, " Paper No. 670052 in S.A. E. Trans.
15, (1968).
111. Merrion. D. F. . "Effect of Design Revisions on Two-Stroke Cycle
Diesel Engine Exhaust, " S. A. E. Paper No. 680422 (1968).
112. Merrion, D. F. , Detroit Diesel Div. , General Motors Corp. ,
private communication, December 16, 1970.
113. Meurer, J. S.. and Urlaub, A., "Man ModiBed M Combustion System
Burns Any Grade of Gasoline Efficiently and Without Knock, " S. A. E.
J. 77, 36-41 (1969) June.
10-9
INSTITUTE
T F C H N 0
-------
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
124.
125.
8929
Mick, S. H. and Clark, Jr., J. B., "Weighting Automotive Exhaust
Emissions, " S.A.E. Paper No. 690523 (1969).
Minkoff, G. J. and Tipper, C. F. H. Chemistry of Combustion Re-
actions, 3, 153. London: Butterworths, 1962.
Milks, D. et al. , "Studies and Analysis of Diesel Engine Odor
Production." Paper presented at the 2nd Int. Clean Air Congress,
Washington, D. C., December 1970.
Miller, C. O. , "Diesel Smoke Suppression by Fuel Additive Treat-
ment, " S. A. E. Paper No. 670093 (1967).
Moody's Investors Service, Inc., Moody's Public Utility Manual.
New York. 1970.
Mrochek, J. E. , "The Economics of Hydrogen and Oxygen Produc-
tion by Water Electrolysis and Competitive Processes," in
Abundant Nuclear Energy. U. S. Atomic Energy Commission Rep.
No. CONS-680810. Oak Ridge, Tenn. ; Oak Ridge National Labora-
tory, 1969.
National Energy Board of Canada, Annual Report: Energy Supply
and Demand in Canada and Export Demand for Canadian Survey.
Ottawa, 1969.
"New Engines are Designed for Use of Natural Gas, " Prod. Eng..
62-4 (1970) March 2.
Newhall, H. K. , "Control of Nitrogen Oxides by Exhaust Recircula-
tion —A Preliminary Theoretical Study, " S. A. E. Paper No. 670495
(1967).
Newhall, H. K. , "Kinetics of Engine Generated Nitrogen Oxide and
Carbon Monoxide, " in Proceedings of the 12th Int. Symposium on
Combustion, 603, Pittsburgh: Combustion Institute (1969).
NewhaU, H. K. and El Messiri, I. A., "A Combustion Chamber
Designed for Minimum Engine Exhaust Emissions, " S. A. E, Paper
No. 700491 (1970).
Newhall, H. K. and Starkman, E. S., "Direct Spectroscopic Deter-
mination of Nitric Oxide in Reciprocating Engine Cylinders, " S. A. E.
Paper No. 670122 (1967).
INSTITUTE
10-10
CAS
TECHNOLOGY
8929
126.
127.
128.
129.
New York City Department of Air Resources, information supplied
to the Institute of Gas Technology, August 1970.
"New York State Motor Vehicle Emission Program. " Final Report.
Division of Motor Vehicles, Department of Air Resources, City of
New York, August 1968.
Nichols. J. E. , El Messiri, i. A. and NewhaU. H. K. , "Inlet
Manifold Water Injection for Control of Nitrogen Oxides — Theory
and Experiment, " S. A. E. Paper No. 690018 (1969).
1968 LP-Gas Market Facts, National LP-Gas Association. Chicago,
1969.
130. 1970 National Petroleum News Factbook Issue. New York: McGraw-
Hill, May 1970.
131. Pahnke, A. J. and Conte, J. F. , "Effect of Combustion Chamber
Deposits and Driving Conditions on Vehicle Exhaust Emissions, "
S.A.E. Paper No. 690017 (1969).
132. Papa, L. J. , "Gas Chromatography — Measuring Exhaust Hydrcarbons
Down to Parts Per BUlion, " S. A, E. Paper No. 670494 (1967).
133. Patterson. D. J. , "Cylinder Pressure Variations, A Fundamental
Combustion Problem, " S.A.E. Paper No. 660129 (1966).
134. Pattison, J. N. et al. , "New Jersey 's Rapid Inspection Procedures
for Vehicular Emissions, " S.A.E. Paper No. 680111 (1968).
135. Pearsall, T. J. and Gar.abedian, C. G. , "Combustion of Anhydrous
Ammonia in Diesel Engine, " S. A. E. Paper No. 670947 (1967).
136. Pecsok, D. A., "News Release - Pollution Emission Results. "
St. Louis County Health Department, Clayton, Mo. , June 1970.
137. Perez, J. M. and Landen, E. W. , "Exhaust Emission Character-
istics of Precombustion Chamber Engines, " S. A. E. Paper No.
680421 (1968). - ~
138. Pratt, D. T. and Starkman, E. S. , "Gas Turbine Combustion of
Ammonia, " S.A.E. Paper No. 670938(1967).
139. "Propane Demand Will Outrun Supply, " Oil Gas J. 68, 44(1970)
March 2.
10-11
INSTITUTE
-------
8929
140. Quick, S. L. and Kittredge, C. K. , "Control of Vehicular Air Pol-
lution Through Modifications to Conventional Power Plants and
Their Fuels. " Paper presented at the 2nd Int. Clean Air Congress,
Washington, D. C. , December 1970.
141. Reid, R. S.. Mingle, J. G. and Paul, W. H. , "Oxides of Niti ogen
From Air Added in Exhaust Ports, " S. A. E. Paper No. 6601 V4_
(1966).
142. Robinson, E. and Robbins, R. C. , "Sources, Abundance, and Fate
of Gaseous Atmospheric Pollutants, " American Petroleum Insti-
tute Final Report SRI Project PR-6755. 69, 99. Huntsville, Ala. :
Stanford Research Institute, February 1968.
143. Robison, J. A., "Humidity Effect on Engine Nitric Oxide Emissions
at Steady-State Conditions, " S. A. E. Paper No. 700467(1970).
144. Santo, J. O. , Fagley, W. S. and Hunter, W. A., "Chrysler Evapora
tion Control System; The Vapor Saver for 1970, "- S. A. E. Paper No.
700150 (1970).
"Oxides of Nitrogen in the Combustion Pro-
ducts of an Ammonia Fueled Reciprocating Engine, " S. A. E. Paper
No. 680401 (1968).
Scheffler, C. E. , "Combustion Chamber Surface Area, A Key to
Exhaust Hydrocarbons, " S. A. E. Paper 660111 (1966).
Schneider, H. T. and Tooke r, R. S. , "The 19 "0 General Motors
Emission Control Systems, " S. A. E. Paper No. 700149 (1970).
Schwing, R. C. , "An Analytic Framework for the Study of Exhaust
Manifold Reactor Oxidation, " S. A. E. Paper 700109 (1970).
Schwochert, H. W. , "Performance of a Catalytic Converter on
Unleaded Fuel, " S. A. E. Paper 690503 (1969).
Shelef, M. , Otto, K. and Gandhi, H. , "The Heterogeneous Decom-
position of Nitric Oxide on Supported Catalysts, " Atmos. Environ.
_3, Pergamon Press, 107-22 (1969).
Sigsby, Jr., J. E. and Rose, Jr. , A. H. , "Evaluation of a Sub-
tractive Method in Auto Exhaust Analysis. " Presented at the
National ACS Meeting, New York, September 1966.
145. Sawyer, R. F, et al. ,
146.
147.
148.
149.
150.
151.
10-12
N S T I T U T
TECHNOLOGY
8929
152. Sneed, M. C. and Brasted, R. C. , Comprehensive Inorganic
Chemistry. Princeton, N. J. : D. Van Nostrand, 1956.
153. Spindt, R. S. et al., "Nitrogen Oxides, Combustion and Engine
Deposits," S. A. E. Trans. 64, 797(1956).
154. Squibb, C. E. , "An Observation by a Carburetion Equipment Manu-
facturer and Distribute •. " Paper presented at the LP-Gas Engine
Fuel Symposium, Detroit; October 1970.
155. Stahman, R. C., Kittredge, G. D. and Springer, K. L. , "Smoke
and Odor Control for Diesel-Powered Trucks and Buses, " S. A. E.
Paper No. 680443 (1968).
156. Starkman, E. S. and Newhail, H. K. , "Characteristics of the
Expansion of Reactive Gas Mixtures as Occurring in Internal Com-
bustion Engine Cycles, " Paper No. 650509. S. A. E. Trans. 74,
(1966).
157. Starkman, E. S. and Newhail, H. K. , "Thermodynamic Properties
of Methane and Air, and Propane and Air for Engine Performance
Calculation," S. A. E. Paper No. 670466 (1967).
158. Starkman, E. S. et al.. "Ammonia as a Spark-Ignition Engine Fuel:
Theory and Application, " S. A. E. Paper No. 660155(1966).
159. Starkman, E. S. et al. , "Alternative Fuels for Control of Engine
Emission. "J. Air Pollut. Contr. Ass. 20. 87-92 (1970) February.
160. Steffensen, R. J. , Agnew, J. L. and Olsen, R. H. , "Combustion
of Hydrocarbons Property Table," Engineering Bulletin of Purdue
University, Eng. Extension Series 122. Lafayette, Ind. , May 1966.
161. Swartz, D. N. , Wilson, K. W. and King, W. J. , "Merits of Lique-
fied Petroleum Gas Fuel for Automotive Air Pollution Abatement, "
Report Nos. 61-55, Los Angeles: University of California, August
1961.
162. Sweeney, N. , "Here's What Users Pay for Ammonia, " Hydrocarbon
Process. 4_7, 265-268 (1968) September.
163. Taylor, R. E. and Campau, R. M. , "The IIEC — A Cooperative Re-
search Program for Automotive Emission Control, " Proc. Amer.
Petrol. Inst. J% 309-27 (1969).
•- S T I T U T E
10-13
GAS
T E C H N n
-------
8929
164. Teague, D. M. , Lesniak, Jr., E. J. and Loeser, E. H. , "A Recom-
mended Flame lonization Detector Procedure for Automotive Ex-
haust Hydrocarbons, " S. A. E. Paper No. 700468(1970).
165. Thomas, H. R. and Grothusen, H. D. , "LP-Gas Sales far Outstrip
Production in Setting New Records in 1969, " Butane-Propane News
(1970) January.
166. Thomas, H. R. and Wetzel, O. K., "Propane Supply and Demand."
Panel presentation presented at National LP-Gas Association Con-
vention, May 3-6, 1970.
167. Thomson, J. C. , "NAPCA Findings on Gaseous Fuels. " Paper
presented at the LP-Gas Engine Fuel Symposium, Detroit, October
1970.
168. U.S. Bureau of Census, Inorganic Chemicals and Gases Summary
for 1968. Washington, D. C. : U. S. Department of Commerce, 1969.
169. U. S. Department of Health, Education, and Welfare, "Control of
Air Pollution From New Motor Vehicles and New Motor Vehicle
Engines," Federal Register 31, No. 61 (1966) March.
170. U.S. Department of Health, Education, and Welfare, "Control of
Air Pollution From New Motor Vehicles and New Motor Vehicle
Engines, " Federal Register 33. No. 2 (1968) January.
171. U.S. Department of Health, Education, and Welfare, "Control of
Air Pollution From New Motor Vehicles and New Motor Vehicles
Engines," Federal Register- Part III 35. No. 219, 17.288-313
(1970) November.
172. U.S. Office of Business Economics, Surv. Curr. Bus. 50, No. 11.
Washington, D. C. : U. S. Department of Commerce, November 1970.
173. U.S. Public Health Service, Environmental Health Se rvice, National
Air Pollution Control Administration, Air Quality Criteria for
Carbon Monoxide. Washington, D. C. : U.S. Department of Health,
Education, and Welfare, 1970.
174. U.S. Public Health Service, Environmental Health Service, National
Air Pollution Control Administration, Control Techniques for Carbon
Monoxide. Nitrogen, and Hydrocarbon Emissions From Mobile
Sources. Washington, D. C. : U.S. Department of Health, Education,
and Welfare, 1970.
10-14
t N S T I T U T r
TECHNOLOGY
8929
175.
176.
177.
178.
179.
180.
181.
182.
183.
184.
185.
186.
U.S. Public Health Service, Environmental Health Service, National
Air Pollution Control Administration, Control Techniques for
Particulatc Air Pollutants, AP-51. Washington, D. C. : U.S.
Department of Health, Education, and Welfare, 1970.
Walsh, A. D. , "The Knock Rating of Fuels. " Ninth Int. Symposium
on Combustion, 1046-55. Academic Press, 1963.
Wentworth, J. T. , "Piston and Ring Variables Affect Exhaust Hydro-
carbon Emissions, " S. A. E. Pape r No. 680109(1968).
Wimmer, D. B. and McReynold, L. A.. "Nitrogen Oxides and
Engine Combustion," S. A. E. Paper No. 380E (1951).
Wise, H. and Freeh, M, F. , "Kinetics of Decomposition of Nitric
Oxide at Elevated Temperatures, I. Rate Measurement in a Quartz
Vessel, " J. Chem. Phys. 20, 1724 (1952).
Witzky, J. E. and Clark, Jr., J. M. , "A Study of the Swirl Strati-
fied Combustion Principle," S. A. E. Paper No. 660092(1966).
Yagi, S. et al.. " A Study in the Development of Exhaust Emission
Controls for Small Displacement Honda Engines, " S. A. E Pape r
No. 680527(1968). —
Yumlu, V. S. and Carey, Jr., A. W., "Exhaust Emission Character-
istics of Four-Stroke, Direct Injection, Compression Ignition En-
gines, " S. A. E. Paper No. 680420(1968).
Agnew, J. T., "Unburned Hydrocarbons in Closed Vessel Explosions,
Theory Vs. Experiment Applications to Spark-Ignition Engine Ex-
haust, " S. A. E. Paper No. 670125 (1967).
Brinkman, W., "The Use of Natural Gas in Spark-Ignition Engines. "
Paper presented at the Institute of Marine Engineers, London,
February 15, 1968.
Svoboda, E. J., "Exhaust Emissions on Fork Lift Applications —
The Motivations, the Controls, the Benefits. " Paper presented
at the LP-Gas Engine Fuel Symposium, Detroit, October 1970.
Weil, N. A., Jenney, A. F. andGarrett, J. H. , "Development of
the New Family of Cummins Natural Gas Engines, " ASME Publica-
tion 65-OGP-13 (1965).
INSTITUTE
10-15
C- A S
T r r H N O I. O<~
-------
. 89Z9
187. Hopkins, H. IT., "Feasibility of IMilizing Gaseous Fuels for
Reducing Emissions Fi-oni Motor Vehicles,'1 Envi ronmenlal
Protection Agency, Ypsilanti, Michigan.
188. "Profile of Air Pollution Control in Los Angeles County, " Air
Pollution Control District , January 1969.
189. "1969 LP - Gas Market Facts, " National LP - Gas Association,
1970.
190. Murray, R. G. and Schocppel, R. J. , "Emission and Performance
Characteristics of an Air-Breathing Hydrogen-Fueled-Internal
Combustion Engine, " unpublished information, 1971.
10-16
CAS
ST.WD.ttO TITLE P*E ».• "**•*• £*W^>>
FORTECWHCALREPWm APTD-0698 '///////,
4. Title and Sttrtltlt
-. Emission Reduction Using Gaseous Fuels for Vehicula
' Propulsion
7. AuthOfU)
^TtWt^rSa^e^n'ology
3424 South Scate Street , .
IIT Center
Chicago, Illinois 60616
12. Sponsoring Aimcy Name ml Adfteu ' ' '
Air Pollution Control Office •
Environmental Protection Agency
Cincinnati, Ohio 45227 • '
13. Surplamenttry Hotel
?!/ 3. RKtfltm't CHalOf No.
S. Report Data
r June 1971
6. Pnformlni Orjsnljitlon Code
8. Performing Organliatlon Rept. No.
10. ProJect/Taik/Woik Unit No.
IT. eontraet?ffriB«"NoT '
fp^-VlPco -
70-69
13. Type ol Report & Period Covered
Final
14. Sponurlnt Agency C6\K
16. Abslracti- The. findings' of a STUfly art! preaenLefl as Cney relate to each ot the following
objectives: 1. To provide a comprehensive vehicular .pollution emissions by using
gaseous Fuels in commercially available engine systems. 2. To examine the logistics of
;aseous fuels to assess the availability, cost, storage and handling methods and safety
requirements of gaseous fuels. 3. To determine the feasibility of using natural gas as
a supplementary fuel in a two stroke cycle, diesel powered intercity bus as a means of
reducing exhaust pollution while operating in urban areas and bus terminals. 4. To
recommend specific research and development programs necessary to confirm or establish
the low emission characteristics and economic feasibility of selected gaseous fueled
vehicular engine systems.
17. Kay Vonb in) Doom* Anelyili. (•). Descriptors
Air pollution Fuel consumption
Exhaust emissions Fuel storage
Ground vehicles Costs
nternal combustion engines Feasibility
Natural gas Bus (vehicles)
.iquifled natural gas. Diesel engines
Availability Urban areas
Natural resources
in. Identlflm/OpMtCnted Tana
Air pollution control
17c. COSATI FtoU/Grav 13B
18. OUtriHuMon Sutvmt „ ., ., . 19. Security
Unlimited U||C
JEiKKlty
UNC
ClaulThlt Report) 21. No. of Pages
LASSIFIEO 268
Clasi. (Tnla Page) 22. Price
LASSIFIED
T E C H M 0
-------
_P.!SC_L_AIMER..
This report was furnished Co Che Office of Air
Programs by
Institute of Gas Technology
3424 South State Street
IIT Center
Chicago, Illinois 60616
in fulfillment of Contract No. 70-69
------- |