STUDY OF DECOMPOSED METHANOL
AS A LOW EMISSION FUEL
FINAL REPORT
OFFICE OF AIR PROGRAMS
ENVIRONMENTAL PROTECTION AGENCY
CONTRACT NO. EHS 70-118
APRIL 30, 1971
By
R. K. PEFLEY
M. A. SAAD
M. A. SWEENEY
J. D. KILGROE
R. E. FITCH
€/
-------
s
©
STUDY OF DECOMPOSED METHANOL
AS A LOW EMISSION FUEL
FINAL REPORT
CONTRACT NO. EHS 70-118
APRIL 30, 1971
By
R. K. Pefley
M. A. Saad
M. A. Sweeney
J. D. Kilgroe
R. E. Fitch
Prepared For
Environmental Protection Agency
Office of Air Programs
Division of Emission Control Technology
Ann Arbor, Michigan 48103
Attention:
Mr. Stephen Quick
Project Officer
SCHOOL OF ENGINEERING
ENGINEERING AND APPLIED SCIENCE RESEARCH
UNIVERSITY OF SANTA CLARA
SANTA CLARA, CALIFORNIA AREA CODE 4O8 - 246-32OO
-------
CONTENTS
I. SUMMARY (Results 6 Conclusions) 1
SUMMARY SUPPLEMENT 5
II. INTRODUCTION (Background 8 Program Objectives) 7
III. DESCRIPTION OF EXPERIMENTAL APPARATUS 9
A. General 9
B. Gas Analysis Meters 9
C. Gas Chromatograph 13
IV. EXPERIMENTAL PROCEDURES 16
A. General 16
B. Instrumentation 17
V. DATA REDUCTION 23
A. Engine Performance Data 23
B. Gas Analysis Meter Data 23
C. Gas Chromatograph 25
D. Wet Chemistry 27
VI. EXPERIMENTAL PROBLEMS 30
VII. DISCUSSION OF RESULTS 34
A. General 34
B. Performance 36
C. Emission Data 44
VIII. DECOMPOSITION FUELED ENGINE DESIGN 66
A. Engine Energy Analysis 66
B. Decomposition Chamber Design Analysis 73
C. Engine Performance Control 81
REFERENCES 88
APPENDICES
A. Data Calculation Method 90
B. Tabulated Data 93
C. Decomposition Cycle Energy Balance Analysis 99
-------
I. SUMMARY
Studies were conducted to evaluate blends of pure and
decomposed methanol (2H9 + CO) as fuels for reducing automotive
1C engine air pollution. These investigations included
laboratory 1C engine tests and analysis, and a preliminary
design study of possible methanol decomposition chambers with
associated engine air-fuel (A/F) ratio controls.
Steady-state performance and emission tests were made on
a variable compression ratio CFR engine operating at 900 RPM.
A total of 191 tests were conducted. They included 184 tests
with methanol blends and seven comparative gasoline fueled
tests. Engine test variables were A/F ratio, per cent
methanol dissociation*, compression ratio (CR), spark advance,
and intake manifold temperature. Instrumentation consisted of
apparatus for measuring air and fuel flow rates, engine load,
engine emissions and various engine temperatures. Emissions
instrumentation included CO, CO^, 0-, and NO., gas analyzers and
a gas chromatograph (GO using a flame ionization detector.
The engine operated easily on fuel blends from pure meth-
anol to completely dissociated methanol on tests conducted over
an A/F range of 5:1 to 9:1 and a CR range from 8.5:1 to 10.9:1.
Little power difference was noted between the gasoline and
pure methanol. Methanol blends of high dissociation percentage
(70 and 100) indicated a slight to moderate reduction in power
from the decrease in intake charge density. This power loss
was partially offset by increases in the combustion efficiency
and some recovery of fuel decomposition energy. The indicated
thermal efficiency using pure methanol was essentially the
same as that for gasoline.
For engine operation on methanol the number of different
exhaust species were significantly fewer than for gasoline.
Major unreacted emissions for the methanol fueled tests were
carbon monoxide (CO), oxides of nitrogen (NOV), methanol (CH,OH),
/\ o
s ;
Dissociation and decomposition of methanol are used
synonymously in this report.
-------
methane (CH^), formaldehyde (HCHO), and acetaldehyde (CHgCHO).
Low concentrations of ethane (C0HC), propane (C~H0), and an
L b - Jo
unknown hydrocarbon were also recorded.
Engine variables, particularly A/F ratio and spark advance,
affected exhaust emissions in a manner similar to that noted
for gasoline fueled engines. Near stoichiometric conditions
(A/F = 6.48) all emission concentrations except for NO., were
low. Emissions for a typical test for 30% dissociated methanol
at A/F = 6.5, CR = 8.5 and 10 degrees spark advance were:
methanol, 51 ppm; hydrocarbons, 59 ppm; aldehydes, 169 ppm;
CO, .08%; and NOV, .322%. At low A/F ratio test points (5.0 and
A
5.5) CO, CH_OH, and hydrocarbon emissions were higher and
aldehyde and NO., emissions were lower. As with gasoline, the
NOV concentrations for pure methanol peaked near the stoichio-
X
metric mixture. This peak was shifted to A/F = 7.5 for 100 per
cent dissociation. Hydrocarbon and methanol exhaust emissions
at A/F =7.5 were reduced below stoichiometric values but
aldehydes were at peak concentrations. At A/F = 9.0, all
emissions except methanol were quantitatively low.
The use of dissociated methanol tended to reduce overall
emissions (NO., excepted) and resulted in shifts in the relative
concentrations of the major species.
Decreasing the spark advance to between 0 and 10 degrees
significantly reduced emissions at stoichiometric and lean A/F
ratio without significantly affecting engine power. The effect
of changing the spark advance at rich A/F ratios was not
appreciable.
The effect of CR and intake manifold temperature on
emissions was, except for special conditions, not significant.
Comparison of methanol and gasoline emission data was
made for a limited number of tests near stoichiometric
conditions. The total grams of carbon and CO per cent in the
exhaust were in general an order of magnitude lower for
methanol than for gasoline. NOV emissions were equivalent for
A
both fuels.
-------
The preliminary study of the decomposition chamber and
engine control methods defined several design constraints and
indicated possible methods of solution. Major design
constraints include requirements for methanol or catalyst bed
heating to provide the endothermic heat of decomposition and
catalyst bed condition-cycling with engine start up and shut-
down. A constant decomposition flow rate with methanol
augmentation for power demands was indicated as the easiest
method of A/F ratio control.
Based upon the above results it is concluded that the use
of pure or blended methanol can reduce 1C engine hydrocarbon
and CO emission levels significantly below that obtainable with
gasoline. Nitrogen oxide levels will be similar to gasoline.
Aldehyde emissions can be reasonably controlled through A/F
and spark advance settings. These improvements are attainable
without significant changes in engine power and thermal
efficiency.
Use of a decomposition chamber with a methanol fueled
engine operating under a steady speed and load condition is
feasible. All of the fuel could be dissociated by the engine
exhaust waste energy. However, reformer experiments would have
to be made to ascertain the quality of the fuel decomposition.
Assuming that steady-state experiments produced high quality
decomposed products, it would then be possible to consider
whether a reformer would be feasible for handling the intermit-
tent performance requirements of an automotive engine.
Based upon these observations, it is recommended:
(1) That an automotive engine with carburetion modified
for use of gasoline and methanol be tested on a dynamometer.
That tests be conducted at constant speed and at speed-load
variations equivalent to those specified in federal exhaust
emission determination standards. That exhaust analyses be
made using instrumentation equivalent to that used in the
current test work.
-------
(2) That a bench model reformer designed for steady flow
operation be constructed and that the quality of dissociated
products be evaluated.
-------
SUMMARY SUPPLEMENT
A. Emission Evaluation Parameters:
It was found useful to rate the engine exhaust emissions
in terms of gm/ihp-hr. This or its comparison-rating, gm/bhp-hr,
are meaningful when comparing the pollution effects for different
fuels or for different engine speeds and loads under laboratory
conditions. With the rating base of ppm of engine exhaust for
different speed, load, and/or fuels, it is possible to decrease
the value of ppm and yet have an increase in pollution per unit
energy delivered to the pistons (ihp-hr) or drive shaft (bhp-hr).
This is avoided by using gm/ihp-hr or gm/bhp-hr. An alternative
rating method using gm/vehicle-mile is not satisfactory for
research comparisons, as an arbitrary conversion from engine
power to mileage must be made. It is therefore recommended that
the gm/ihp-hr or gm/bhp-hr ratings be standardized for laboratory-
engine comparisons. It is noted that the 1973 heavy-duty vehicle
standards for California utilize this form of evaluation.
Presuming that the recommended rating indices are attractive,
there should be some work done in relating this value to the
current standards in terms of gm/vehicle-mile. The work rec-.
ommended in Section I would provide an opportunity for an equiv-
alency investigation of the two rating parameters. For example,
it may be found that gm/ihp-hr may be transposed to gm/vehicle-
mile by assuming the vehicle is operating at 50 mph and devel-
oping 50 ihp. For such, the numerical value in gm/ihp-hr would
be the same in gm/vehicle-mile.
B. GC Emissions Analysis:
Gas chromatographic analysis and sampling techniques for
methanol fuel should be improved to minimize errors in data quanti-
fication. The possibility of sample degradation in the exhaust
line should be tested and, if necessary, eliminated. The
identify of the unknown peaks in the exhaust samples should be
established—particularly, the peak eluted just before ethane.
If it is an unsaturated C« hydrocarbon, it would contribute more
to air pollution than a saturated hydrocarbon.
-------
The feasibility of using the Porapak column for analysis
of gasoline aldehyde emissions should be evaluated. Techniques
for the gas chromatographic analysis of gasoline aldehyde
exhaust emissions with Porapak columns have been developed.*
However, these techniques do not determine formaldehyde, and they
involve a complex backflushing procedure. In the program
reported herein exploratory analyses of gasoline-fueled engine
emissions with Porapak T columns were conducted. These
analyses indicated that gasoline engine emissions could be
resolved into many separate peaks without backflushing. While
the laborious undertaking of identification of these peaks was
not made, the columns were found to be capable of resolving
formaldehyde. Based upon these results it is believed that a
column that would identify individual peaks, including
formaldehyde,with a less complex procedure can be developed.
T. A. Bellar and J. E. Sigsby, Jr., Env. Sci. Tech. 4_, 150
(1970).
-------
II. INTRODUCTION
Atmospheric air pollution from gasoline powered 1C engines
poses an environmental crisis in many urban areas. Some net
improvement in regional air quality has been experienced as a
result of automotive engine design changes and incorporation of
emission control devices, but growing car populations threaten
to negate these advances unless stricter standards are met [1].*
Since further reductions in emission controlled gasoline powered
1C engines are difficult, other methods of reducing passenger
car exhaust emissions have been sought. These methods have ranged
from the use of alternative 1C engine fuels to replacement of
the 1C engine with a different power plant. To date, no
completely acceptable power plant replacement has been developed.
An alternate fuel which has shown promise of reducing emissions
below that obtainable with gasoline is methanol.
In previous work [2] a standard automobile modified to run
on methanol fuel gave some evidence of reduced pollution in
comparison with gasoline fuel. This evidence was not conclusive
because of improper carburetion. Also included in this previous
work was an exploratory investigation into the use of fuel blends
of liquid methanol and dissociated methanol. A CFR engine was
used for this phase of the work. The ultimate goal of this
latter effort was to use waste thermal energy in the engine
exhaust in conjunction with a catalytic surface to vaporize and
dissociate the liquid methanol. The expected results were an
improvement in methanol performance from exhaust energy recovery
of the heat of decomposition, and a decrease in particulate and
exhaust gas pollution resulting from combustion of a chemically
simple fuel. These were to be achieved without major sacrifices
in either vehicle performance, equipment or fuel costs.
It was found that within a broad band of A/F values and
compression ratios (CR), the CFR engine could be operated on
methanol fuel blends ranging from pure methanol to 100% dis-
sociated methanol. It should be noted that the dissociated
[1] refers to references.
-------
methanol was simulated by a chemically correct mixture of CO and
H2 and intermediate fuel blends were achieved by mixing of pure
methanol and dissociated methanol. The exploratory evidence
from the CFR engine tests supported the thesis that methanol
fuels might reduce exhaust pollution relative to gasoline fuels.
The principal inadequacies in providing complete understanding
from these tests were:
(1) A very limited test matrix in terms of engine
variables.
(2) Limited emission instrumentation.
This exploratory evidence became the basis for the work
herein reported. The objectives of this project were:
(1) Using the CFR engine, to assess engine operational
control and steady-state, constant speed, power
performance as functions of:
(a) percentages of methanol and dissociated
methanol,
(b) air-fuel ratio,
(c) compression ratio,
(d) intake manifold temperature.
(2) To examine the exhaust emissions and relate the
results to engine performance in meaningful ways.
(3) To carry out a preliminary design study which
examines the feasibility of using engine waste
exhaust energy as an energy source for dissociating
methanol fuel.
-------
III. DESCRIPTION OF EXPERIMENTAL APPARATUS
A. General:
Tests were conducted on a single cylinder, variable
compression CFR engine manufactured by the Waukesha Company.
Figure III-l is a photograph of the experimental equipment and
Fig. III-2 is a schematic of the engine test set-up. The engine
carburetion and intake system were modified to allow injection
and mixing of methanol and air, including provision for
controlled heating of the fuel and/or fuel-air mixture. In
order to hold the speed constant and provide the desired load,
the engine was coupled to a synchronous motor running in
parallel with the standard cradled DC dynamometer which was
used primarily to start the engine. All tests were run at a
speed of 900 RPM regardless of changes in the A/F ratio, spark
advance, and CR.
The air flow to the engine was measured by a calibrated
nozzle mounted on a plenum chamber. The flow of the liquid fuel
and gaseous fuel was measured by rotameters. Both the air flow
meter and the gaseous fuel meter were calibrated with positive
displacement meters and readings are estimated to have a maximum
uncertainty of +_ 2%. During all tests the engine crankcase
temperature was maintained at a constant value by a control-
lable electric heater.
A sampling line connected at the exhaust port of the
engine fed exhaust gases to 02, N0y» CO and C02 meters and a
gas chromatograph (GO—see Fig. III-3. A second sampling line
was provided for collection of aqueous samples. These samples
were later subjected to post test GC analysis. A more detailed
description of the emission instrumentation is presented in the
following sections.
B. Gas Analysis Meters;
Continuous exhaust emission measurements were made using
individual instruments for determination of 0^, NO,,, CO, and
-------
FIG.III-1: CFR ENGINE AND INSTRUMENTATION
-------
NOZZLE
AIR SURGE
TANK
SPRAGUE
DYNOMOMETER
V-BELTS
AIR
PREHEAT
WATT-
METER
SYNCHRONOUS
MOTOR
CRANKCASE
HEATER
FORCE
EXHAUST
BUBBLER
TO ATM.
EXHAUST
SAMPLE
(7)TT
CFR
ENGINE
WET TEST
METER
iHEATER TAPE
VALVE TO ATM.
> HEATER TAPE
COMPRESSION
RAT'°
FLOWMETER
SOLENOID
VALVE
EXHAUST TO
GAS ANALYSIS
LIQUID
METHANOL
TANK
DISSOCIATED METHANOL
TANKS
MEASURED TEMPRATURES•
0 INLET AIR HEATER
® OUTLET AIR HEATER
S COOLING WATER IN
COOLING WATER OUT
(D ENGINE INLET MIXTURE
© CRANK CASE OIL
(?) ENGINE EXHAUST
FIG.HI-2: EQUIPMENT -SCHEMATIC METHANOL FUEL STUDY
-------
TO AMPLIFIERS
A,
ENGINE
EXHAUST*
EXHAUST
SAMPLING
VALVE
MANOMETER
(H20) ICE BATH
8 TRAP
HUMIDIFIER
CALIBRATION
GAS
MANOMETER
(H20)
GAS
CHROMATOGRAPH
PROGRAMMER
SENSOR
_
(T\
VACUUM
PUMP
TRAP
MANOMETER
(Hg)
RECORDER
AMPLIFIER
FIG. Iff-3: SCHEMATIC DIAGRAM OF GAS ANALYSIS INSTRUMENTATION
-------
13
C07 exhaust concentrations. Carbon dioxide concentrations were
measured with a Beckman LB-1 Medical Gas Analyzer and carbon
monoxide with a Model 315a Beckman Infrared Analyzer. The
oxygen content was measured with a Beckman Model 778 Process
Oxygen Analyzer and the oxides of nitrogen were monitored by a
Dynasciences Corporation Model NX 130 Air Pollution Monitor.
Pressure in the exhaust manifold was maintained at 10"H20
by a valve at the exhaust pipe outlet. Pressures and flow rates
in the sample line were adjusted and monitored by a series of
valves, manometers and flow meters. A cold trap was provided
upstream of the first instrument (see Fig. III-3) to prevent
water condensation on the sensor elements. Heating tape was
wrapped around the exhaust line between the exhaust manifold and
first cold trap to prevent condensation in the sensor line. A
pump was placed downstream of the last sensor to maintain a
sufficient sample flow rate for optimum instrument response.
It should be noted that all sensor pressures were slightly above
atmospheric pressure to avoid exhaust dilution should there be
any undetected leaks.
Premixed calibration gases from high pressure storage
bottles could be fed into the sample line just upstream of the
gas analyzers for instrumentation calibration. Valves in the
exhaust sample and calibration gas lines permitted selection
of gases from either of these sources. Calibration gas flow
rates were adjustable by use of the bottle gas regulator and
flow throttling valves by reference to manometers and flow
meters in the sensor line. Calibrating gas flow rates and
pressures were set to match the test pressures and flow rates.
C. Gas Chromatograph:
The GC (Hewlett-Packard Model 700, dual column, dual
flame ionization detector) was fitted with two columns: one
Porapak Q, the other Porapak T (manufactured by Waters
Associates). Each column, made of 1/8-inch O.D. aluminum
tubing, was six feet long and used 80/100 mesh packing.
-------
Two different systems were provided for taking exhaust
samples for the GC. In the first, a silicone septum was
compressed in a port just downstream of the exhaust valve. A
hypodermic needle of a heated gas syringe could be inserted
through the septum and could sample the exhaust by filling the
syringe. The second, more convenient system, consisted of a
two-position valve (see Fig. III-4) connected to the exhaust
gas sampling line. When the valve was closed, the exhaust
sample passed continuously into the valve through a loop of
known volume back into the valve, and then was vented to the
atmosphere. Simultaneously, the carrier gas for the GC passed
from the nitrogen cylinder through the valve, and into the
columns. When the valve was opened, both flows were diverted.
The exhaust stream passed directly through the valve to the
vent, and the carrier gas passed through the loop, sweeping the
exhaust sample in the loop into the column.
-------
•15
EXHAUST
TO LOO
N,
EXHAUST
EXHAUST VENT
FROM LOOP
TO GC
VALVE CLOSED
y EXHAUST VENT
FROM LOOP
TO GC
VALVE OPENED
FIG. IIT-4: SAMPLING VALVE
-------
16
IV. EXPERIMENTAL PROCEDURES
A. General:
Start up of the engine was accomplished by motoring with
the dynamometer until combustion was sustained. The cradled
dynamometer was then unloaded and allowed to rotate freely.
The engine was set at a mid-range A/F ratio and allowed to warm
up until the desired operating temperatures of the intake
manifold and crankcase were stabilized.
Setting up a test run began by first stabilizing the
crankcase temperature at 61-64°C. The desired compression
ratio was then set at the required value with a hand crank and
the cylinder head was clamped to prevent further movement. The
barometric pressure, the ambient air temperature, and the
pressure drop across the air intake nozzle were recorded and
the air mass flow rate was calculated. The desired A/F ratio
and per cent of fuel dissociation were then set using the
calibration curves for the rotameters. If the air flow changed
because of the fuel flow adjustments, the procedure was repeated
until the proper setting was established.
The intake manifold temperature was set at the desired
temperature by a variable resistance heater and the fuel flow
rates were again checked and corrected, if necessary. The
spark advance was either set at a desired point or optimized
at the setting which produced the maximum power reading.
All of the above readings were then checked to insure
that they were both correct and steady, and the test was begun.
During the test run the readings were checked periodically.
Some modes of operation required nearly continuous adjustments
to offset drifting of the fuel flow rate or intake manifold
temperature.
During the time of the test run, a portion of the exhaust
was routed through the exhaust bubbler with the exhaust sample
volume being measured by the wet test meter corrected for the
temperature measured at the meter. Between tests the bubbled
-------
17
sample was washed into another bottle and cataloged. The
bubbler was then refilled with distilled water for the next test
Another portion of the exhaust was routed to the gas
analysis meters, where the CO, CO-, 02 , and NO^. readings were
regularly monitored. A gas chromatograph sample was taken
once during each test run at a period of steady operation.
During each test, the following engine data were
recorded:
Dynamometer Force
Spark Advance
Air Nozzle Pressure Difference
Barometric Pressure
Ambient Air Temperature
Fuel Rotameter (methanol and/or dissociated methanol)
Dissociated Fuel Pressure
Wattmeter Power*
Compression Ratio
Temperature
(1) Intake Manifold
(a) Heater Inlet
(b) Heater Outlet
(c) Engine Intake Port
(2) Cooling Water In
(3) Cooling Water Out
(4) Crankcase Oil
(5) Engine Exhaust
Test runs averaged about 22 minutes of steady-state
operation, the approximate time necessary to cycle the GC. The
time between runs was a function of the parameter being changed.
Test runs were only begun after steady-state operation was
established. All engine data were recorded at the beginning of
the run and readings were checked at the middle and at the end
of the runs. The average was recorded if a variation occurred.
The gas analysis meters were monitored during the run and
frequent data was recorded.
B. Instrumentation:
Calibration - Calibrations of the 02, NO^, CO, and CO
meters were made at the beginning and end of each test day
e.
The wattmeter power combined with the calibrated
efficiency curve of the synchronous motor provided measurement
of engine output to the synchronous motor.
-------
18
and at intervals of approximately three hours throughout the
day. The calibration duration was in each case long enough to
insure that steady-state conditions had been established in the
sensor line. These calibrations were accomplished by introducing
premixed calibration gases into the instrumentation sampling
line just upstream of the 0* meter. The 02, CO, and NO., meters
were calibrated using two reference gas concentrations, one of
the references being zero. The C02 meter, which was not as
linear, required three reference points. Zero concentration
levels for the NO.., CO, and C02 meters were obtained using air.
A mixture of N~ and C02 was used to zero the 0~ meter. The
second calibration point was obtained using a reference gas
mixture with concentrations of: C02, 12.0%; 02, 4.0%; CO, 3.0%,
N02, 0.15%; and N2, 80.85%. The third calibration point for the
C0? meter was established using a mixture of 5.0% C02 and
95.0% 02.
Instrument gain settings were selected which gave good
accuracy for all but the most extreme test conditions. The
C02 meter was calibrated on a range from 0 to 16%, the CO meter
on a range from 0 to 5%, the 02 meter on either a range of 0 to
5% or a range from 0 to 25%, and the NO,, meter on one of the
three ranges: 0 to 0.1%, 0 to 0.3%, 0 to 1.0%.
Three columns were used on the GC over the course of the
test program. During initial check-out tests with the first
column, the various methanol exhaust species, which were to be
distinguished by the GC, were defined by their elution times
as determined by authentic compounds. Subsequently, through the
test program all columns were calibrated for elution time and
peak sensitivity with pure reference samples of the primary
exhaust emission contaminants. Periodic calibrations on each
column were made with formaldehyde and methanol. For each
column the elution time for a given constituent was constant
until the effective lifetime of the column had been reached.
At that time the column was replaced. Elution times and peak
sensitivities were determined for methane on each of the three
-------
19
columns and for ethane and propane on the first two columns.
Complete calibration of all constituents on all three columns
was not necessary since the elution times and peak sensitivities
were constant for a given column and were in direct proportion
to similar events on the other columns.
Testing - Exhaust sampling for the gas analysis meters
was accomplished by shutting the valve to the calibration gas
bottles and by opening the valve to the exhaust manifold. The
valve downstream of the cold trap was then adjusted to give a
positive pressure matching the calibrating gas pressure and
guaranteeing a match of the exhaust and calibration flow rates
through the sensors. Steady-state conditions were usually
obtained in 5 to 8 minutes. An additional 10 to 12 minutes
were then allowed to obtain average steady-state readings for
all test instrumentation.
At the end of experimentation the instrumentation
calibration was checked. If the calibration difference due to
drift was greater than the reading error, then linear extrapola-
tion in time was performed. If the drift was excessive, the
experiments were repeated.
One or more gas chromatograms were taken during each test
cycle. During early tests two or three were recorded. Later,
after the test procedures became well established and it was
determined that all the chromatograms obtained during a given
run gave consistent data, only one was recorded per test.
Exhaust samples were introduced into the GC column by cycling
the gas sample valve, or for special tests by use of a heated
syringe. Elution of the exhaust constituents from the column
was accomplished by either one of two compound temperature
programs. The primary program consisted of holding the column
at 110°C for two minutes followed by a linear increasing tem-
perature of U°C per minute to a maximum temperature limit
of 165°C. The second program, used to obtain more distinct
separation of some peaks, consisted of holding the column at
60°C for four minutes with a subsequent linear temperature
-------
20
increase of 4°C per minute to a limit of 165°C. Figures IV-1
and IV-2 show representative GC data for the two different
temperature programs.
-------
ATT.
32
4
'«
CH.OH
, t-
60'
TEMPERATURE INCREASE AT 4VMIN
TIME MINUTES
FIG EZ-h CHROMATOGRAM FOR 60 °C TEMPERATURE PROGRAM
(TEST 46-2)
-------
ATT.
32
16
22
CH3OH
-^TEMPRATURE
INCREASE AT4°/MIN
TIME MINUTES
FIG.Iff-Z'CHROMATOGRAM FOR 110°C TEMPERATURE
PROGRAM (TEST 65)
-------
23
V. DATA REDUCTION
A. Engine Performance Data:
The determination of engine performance parameters of
interest, apart from the exhaust gas constituent analysis, was
by standard methods. (Sample engine performance data calcula-
tions are presented in Appendix A.)
The A/F values were determined from experimental air and
fuel flow rate data.
The bhp and ihp determinations were complicated by
simultaneous use of a synchronous motor as a means of loading
the engine while holding its speed constant, and a cradle
dynamometer for engine start. Engine brake power was used to
drive the unloaded cradle dynamometer and the synchronous motor
which was V-belt driven from the engine flywheel. The power to
the cradled dynamometer could be directly determined. The
power to the synchronous motor was determined by measuring its
power output and dividing by the motor and belt efficiency.
The synchronous motor efficiency was obtained by using the
cradled dynamometer as a known power source and motoring the
engine and synchronous motor at 900 RPM and at different load
conditions. This same technique with the synchronous motor
disconnected provided the engine friction horsepower at a
nominal CR, which could be subtracted from the total motoring
power to obtain the power supplied to the synchronous motor and
V-belt drive.
Indicated horsepower was determined from the sum of
brake horsepower and engine friction horsepower.
Indicated specific fuel consumption was determined by
ratioing of the fuel flow rate and indicated horsepower.
B. Gas Analysis Meter Data;
The gas analysis meter data was reduced to determine the
molar fractions of C^, CO, 02, and NO., in the exhaust sample
of each test. Data reduction calculations were aided by use
of a data reduction computer program. Input into the program
-------
24
included steady-state or average test values and calibration
data for each gas. Correction factors and assumptions used in
reducing the data included:
(1) Oxygen meter electrode residual correction -
instrument usually read 0.2% 02 when nitrogen or
carbon dioxide was purged under a slight positive
pressure (1" H20).
(2) NO,, meter response to concentrations of CO
according to the following expression:
Indicated ppm NOX = 40 (% CO)
(3) Assumptions that the oxides of nitrogen from the
engine exhaust were primarily NO. The response
of the NO,, meter to NO was approximately 10%
less than to N02. The NO,, meter was calibrated
with NO2 and thus an additional correction factor
was introduced to take this effect into account.
(4) Correction factor in the molar balance due to
condensation of water in the cold trap. This
correction factor assumed that the effect of
ignoring the concentrations of NO,, and hydro-
carbons in the exhaust was negligible.
The measurement accuracy of the 02 content on the rich
side of stoichiometric A/F was poor due to the almost complete
lack of 02. On the lean side of stoichiometric the CO measure-
ment accuracy was likewise poor due to the almost negligible
amount of the gas. At A/F values of 5, the CO content was
greater than 5% and, therefore, could not be measured. For
these cases it was necessary to assume a value in order to esti-
mate a chemical balance and to express the emissions in terms of
grams/ihp-hr. The CO and NO,, meters tended to be the most
stable, while the C09 meter tended to baseline and sensitivity
shifts. This was primarily a temperature phenomena due to lack
of internal temperature compensation within the instruments.
Maximum uncertainty in measurement, including errors in reading
the meter values, extrapolating and reading from graphs (C02
-------
25
and CO only), instrument sensitivity shift, baseline shift,
and accuracy of calibration gas values are indicated below for
nominal A/F values for tests using liquid methanol:
Maximum Uncertainty
Rich
A/F 5.5
6%
10%
11%
100%-
Stoic
A/F 6.5
6.5%
50%*
11%
9%
Lean
A/F 9,0
7%
150%*
18%
13%
co2
CO
NOX
2
High due to low-end scale reading on
the instrument
C. Gas Chromatograph:
Quantitative analysis of emissions with the GC required
identification of the peaks in the chromatograms and determina-
tion of constituent concentration corresponding to the area under
the peak. Identification of the peaks was made by comparison
of their elution times with those of authentic compounds. Since
the elution time of many compounds displayed a small dependence
on sample size, an extrapolation of the elution times with
reference to standardization chromatograms made with different
sample concentrations was required. Only one significant peak
could not be unambiguously identified. In chromatograms with
temperature programming starting at 110°C, this compound was
eluted slightly before or during the peak for ethane (see
Fig. IV-2). In chromatograms beginning at 60°C, this unknown
component was clearly distinguished from ethane (see Fig. IV-1).
When the two peaks could be resolved they were reported
separately; otherwise, they were reported as a single number.
Several other smaller peaks could not be identified but were
sufficiently small as to represent an insignificant proportion
of the exhaust concentrations.
-------
26
Quantitative calibration and reduction of the chromatograms
was done in several different ways. For gaseous compounds, such
as methane, the pure gas was passed through the sampling valve,
and the instrument response to the pure gas was measured. (The
. instrument response was defined as the peak area—height times
the width at half maximum times the attentuation.) The volume
fraction of that gas in the exhaust was then the ratio of the
peak area in the chromatogram of the exhaust sample to the peak
area for the pure gas. For liquids the sensitivity of the
instrument was measured for small samples (usually one to two
microliters). Knowing the peak area corresponding to a given
volume, the peak area corresponding to a given mass was
calculated from the density of the liquid. The volume of this
mass in the vapor phase at 110°C, the usual temperature of the
sampling valve, was calculated from the molecular weight of the
compound and the ideal gas law. The volume fraction of the
compound in the exhaust was then the ratio of this volume to
2.9 ml, the effective volume of the sampling valve. Calculations
for the different compounds are described in Appendix A.
For some compounds it was inconvenient to measure the
sensitivity directly (for example pentane, for the gasoline-
fueled runs). For these cases the sensitivities of the GC to
several Cg , C-, and Cg hydrocarbons were measured. Interpolation
of these data with the information on CH., C2Hg, and C-Hg was
used to determine the GC sensitivity to other hydrocarbons.
Two runs were made with gasoline as fuel. A 2.9 ml sample
of exhaust was injected into the gas chromatograph and over
thirty separate components were resolved using the 60° program.
The chromatograms from these two runs were reduced by measuring
the total peak areas and by applying a standard hydrocarbon
sensitivity factor to obtain the total hydrocarbon concentration
in the exhaust.
The gas chromatograph appears to give data which at
moderate to high concentrations is accurate to within about 10%
of the actual constituent concentrations. This accuracy was
-------
27
noted in the calibration tests with prepared samples. Repeat-
ability was demonstrated on replicate chromatograms during
continuous steady-state operation of the engine at specific test
conditions (see Table V-l). Occasional wide variations of a
constituent during a given run (for example methane during
run 17) were noted but are attributed in most cases to
variations in the instantaneous exhaust emissions. In analyzing
the data, considerable scatter was noted for some species on
tests at similar test conditions but conducted during different
days. The data scatter is believed due to the inability to
exactly reproduce given engine test conditions rather than
significant data inaccuracies. This would seem to indicate that
the reaction kinetics of some of the species are sensitive to
small changes in engine operating variables. An example of this
data scatter is seen by comparing runs 6 , 12 , and 83 (see Table 1
in Appendix B). Although engine operating variables were
essentially the same the C02> CO, and CH^ exhaust emissions
varied systematically. In this case, the disagreement in GC
methane emission data cannot be considered entirely spurious
since the C02, CO, and CH^ exhaust concentrations were measured
independently.
D. Wet Chemistry:
Wet chemical techniques were used for formaldehyde deter-
mination as a comparison for the GC data. For each run the
exhaust was passed through a heated line and through a glass
bubbler filled with water. Formaldehyde and other water-
soluble components were retained in the scrubber, while the rest
of the exhaust was vented to the air. Formaldehyde absorption
is nearly quantitative under these conditions [3].
Several techniques were tried but did not produce
satisfactory data [4, 5, 6]. The most consistent data were data
obtained with a colorimetric method using 2,4-dinitrophenyl
hydrazine [7]. In this method, carbonyl-free methanol was made
-------
28
TABLE V-l
RESULTS OF REPLICATE ANALYSES
Run No.
5
11
12
13
15
17
160
CH^
.108
.108
.083
.083
.062
.072
.008
.008
.002
.002
.074
.110
.172
.406
.427
mg C/g Exhaust
C0HC+ unknown
i b
.008
.009
.008
.006
.008
.008
.003
.003
.002
.002
.007
.007
.006
.018
.020
ppm
C3H8
.001
.001
.001
.001
.001
.001
.0002
.0003
.0001
.0001
.0005
--
.0004
tr
tr
HCHO
51
51
32
34
54
60
22
27
13
14
46
50
48
25
69
CH3OH
560
570
783
689
641
469
192
142
394
394
324
—
251
457
391
by refluxing spectrograde methanol with a 2,4-dinitrophenyl
hydrazine and by distilling. A 1 ml aliquot of scrubber liquid
was mixed with 1 ml of saturated 2,4-dinitrophenyl hydrazine in
carbonyl-free methanol. One drop of concentrated HC1 was added
and the mixture was heated at 60°C for about 20 minutes. After
cooling, the sample was mixed with 5 ml of 10% KOH in 80%
methanol. After the color developed for a few minutes, the
o
optical density of the sample was measured at 4800 A on a
Beckman Model B spectrophotometer. Reproducibility and
accuracy were good on standard samples, and the molar
-------
29
extinction coefficient measured on standards agreed with
published values [7].
This method does not measure formaldehyde, nor total
aldehydes, but rather total carbonyl content of the sample.
Since the exhaust should contain little ketone, the total
carbonyl and total aldehyde concentration should not be too
different. The method is reputed to be insensitive to inter-
fering substances, but results on the scrubber samples gave
poor reproducibility as compared to the GC data. It is
believed that the gas chromatographic data are more reliable
and that the wet chemistry results serve as a poor check on
the total aldehydes in the exhaust samples. For this reason,
only that data obtained from the GC are reported.
-------
30
VI. EXPERIMENTAL PROBLEMS
The principal experimental problems encountered were in
the calibration and reduction of exhaust emission data from the
gas analysis meters and gas chromatograph. Reduction of data
from the gas analysis meters was hampered for a considerable
length of time because of difficulty in establishing the gas
concentrations in the pre-mixed calibration gases. Analysis
of the NO2 content in the purchased gas mixture was originally
given by the manufacturer as 1400 ppm and later amended to
1050 ppm. Final analysis of the original gas mixture by
comparison with two properly analyzed bottles from separate
sources showed that the real content was 780 ppm N0».
Originally, it had been speculated by the manufacturer that the
plain carbon steel bottle was absorbing some of the N0»;
however, identical runs over the course of experimentation could
not produce any such evidence. In the data reduction program,
as a result, correction factors were inserted depending upon
which calibration gas bottle had been used.
The manufacturer of the NO., meter stated in their manual
that the instrument did not respond to CO. However, during
experimentation the NO meter was found to respond to a bottled
A
mixture of the H2 and CO. After notifying the manufacturer on
this point it was indicated that the instrument would respond to
concentrations of CO as discussed previously—see Item 2, Page 24.
Response of the sensor to hydrogen was found to be minimal.
Some experimentation was required in selection of the gas
chromatograph columns and in determination of chromatographic
techniques to be used with the selected columns. Additional
problems were encountered in calibrating the volume of exhaust
gas in the GC sample loop.
Extreme polar compounds, such as water and formaldehyde,
are difficult to analyze on conventional packed columns, as
they are irreversibly absorbed. This irreversible absorption
causes broad, low, tailing, and non-reproducible peaks. In the
-------
31
past this problem has usually been solved for aldehydes by
forming volatile non-polar chemical derivatives of the aldehydes
This approach could not be used for the current analysis because
of the low aldehyde concentrations. Consequently, a special
column packing, unlike the usual solid support coated with a
thin layer of absorbant was tried. Porapak (Waters Associates)
is a porous polymer, principally polystyrene, fabricated into
beads. When used as a column packing it does not irreversibly
absorb polar compounds, and their peaks are sharp and
reproducible. Several varieties of Porapak are available,
modified to show different degrees of polarity. Evaluation of
several of these indicated that one column, Porapak T, success-
fully resolved a mixture of formaldehyde, acetaldehyde, water,
and methanol. An even more attractive feature of this column
packing was that operating above room temperature it could
resolve a mixture of methane, ethane, propane, and butane not
only from one another but also from methanol and aldehydes.
Thus, one analysis could furnish the concentrations of both the
hydrocarbons and the aldehydes in an exhaust sample.
After selection of Porapak T as the best column packing
material, the effect of column size, flow rates, operating
temperatures and temperature programming on constituent elution
times were investigated. It was found that the most effective
Porapak T columns were made from 1/8" O.D. aluminum tubing, 61
long. The optimum flow rates were 18 ml/min for nitrogen
carrier gas, 20 ml/min for hydrogen, and 120 ml/min for air.
Column temperatures hot enough to elute methanol quickly were
too hot to separate methane from other exhaust components. Two
compound temperature programs were found to provide adequate
resolution of data. In the primary program the columns were
held at 110°C for two minutes, and then the temperature was
automatically increased at 4°C/min to a maximum temperature of
about 165°C. A second program was used to verify the elution
times of some constituents. This program consisted of holding
the column temperature constant at 60°C for four minutes
-------
32
followed by an increase of U°C/min to about 165°C. Figures IV-1
and IV-2 show typical runs under these conditions.
The GC exhaust samples used during the test program were
collected using a Hewlett-Packard gas sampling valve. Special
tests using a secondary sample collecting system were necessary
to calibrate the volume within the gas sample valve loop. A
port in the exhaust line about six inches downstream from the
exhaust valves was fitted with an elastomer septum. A heating
jacket was made for a gas syringe to keep it above 100°C. The
needle of the heated syringe was inserted into the exhaust
stream through the septum and the syringe was filled with
exhaust gas. The needle was then withdrawn from the septum in
the exhasut line, inserted into a similar septum in the gas
chromatograph, and the contents of the syringe injected into
the column. The volume of the syringe was verified to be 2 ml
by water displacement (the nominal volume was correct within
2%). Comparative analysis of the calibrated syringe and sampling
value data showed a sample value volume of 2.9 ml. This value
was used for the value sample volume in all calculations
involving this parameter (see Appendix A).
Another problem of concern which required some analysis
was a shift in the chromatogram base line during temperature
programming as shown in Figs. IV-1 and IV-2. This shift did
not occur with a new column, and was not due to column bleeding
because dual-column operation did not help. It is believed
that unstable compounds, perhaps from the crankcase, were
included in the exhaust samples and, hence, were loaded and
retained on the columns during sampling of the exhaust. At
low temperatures they were stable, but as column temperature
rose they slowly decomposed, causing a flow of small amounts of
decomposition products down the column and to the detector.
This would cause the base line to rise, as observed. The base
line shift did not interfere to a great extent with data
analysis. At low temperatures the drift was not significant.
At high temperatures the methanol and aldehyde peak areas
-------
33
could be measured by interpolating the base line between the
beginning and end of the peak. Although the area could not be
measured so accurately as with a flat base line , inaccuracies ,
which are believed to be small, are restricted to the
determination of acetaldehyde and methanol. Errors in
distinguishing between these two species may account for some
of the significant data scatter in methanol and acetaldehyde
yield which were noted for similar test conditions.
-------
VII. DISCUSSION OF RESULTS
A. General:
A total of 191 performance and emission tests were
conducted on the CFR engine. Seven of these tests used gasoline
as a fuel. The remainder used fuel blends ranging from pure
methanol to 100% dissociated methanol. Table VII-1 represents
a summary of engine test conditions used during these experi-
ments. Complete engine test-variable, performance and exhaust
emission data for each of the tests are presented in Appendix B.
All tests were carried out at steady-state conditions.
No significant problems were encountered in starting or stopping
nor while operating over a wide range of CR and A/F values. The
ease of engine operation and the power performance using blends
of methanol and dissociated methanol compared very well with
gasoline performance. The peak power location and power
variation of methanol and gasoline at identical fuel equivalence
ratios (actual A/F/stoichiometric A/F) showed striking similari-
ties. Indicated thermal efficiencies for equivalent conditions
also produced nearly the same results. The differences observed
in power between methanol and dissociated methanol were primarily
due to differences in engine intake conditions. Under duplicate
intake conditions, it is expected that the contrast would be
slight.
The principal exhaust emissions analyzed were carbon
monoxide, hydrocarbons, methanol, oxides of nitrogen, and
aldehydes. In aggregate, the data indicated that the dissociated
methanol is approximately equal to or superior to liquid methanol
in reducing exhaust hydrocarbons and carbon monoxide for all
A/F and CR values. The principal dissociated fuel disadvantage
was higher NO,, emissions in the lean A/F range. Although a
fuel comparison was not part of the study, limited data
indicated that methanol is significantly superior to gasoline
in reducing hydrocarbons and CO emissions.
-------
35
TABLE VII-1
TEST MATRIX
Parametric Base Line Performance (maximum power spark advance--
full
throttle)
CR
8.4
9.2
11.0
Variable Intake
full
Spark
throttle)
CR
8.4
11.0
A/F
5.0
5.5
6.5
7.5
9.0
Temperature
A/F
6.5
7.5
% Diss.
0
30
70
100
(maximum power
% Diss.
0
30
70
100
T
intake
nominal
(70°F)
spark advance
T
intake
40°F
110°F
Timing (full throttle)
CR
8.4
11.0
A/F
5.5
6.5
7.5
9.5
% Diss.
30
Timing
(°BTDC)
0
10
20
25
-------
36
The following sections present an analysis of the CFR
engine performance and emission data.
B. Engine Performance:
Before examining the details of the exhaust emissions it
is of importance to obtain a general view of the engine's
performance in terms of power and thermal efficiency using the
methanol fuel blends. Thermal efficiency is adopted as a basis
of comparison rather than specific fuel consumption because of
the significant difference in heating values of the fuels to be
compared.
(1) Dissociated Fuel Power Effects. The experimentally
observed values of indicated horsepower (ihp) developed by the
engine for pure methanol and 100% dissociated methanol as a
function of A/F ratio are shown in Fig. VII-1 for various
compression ratios. The increase in ihp with compression ratio
approximately reflects that predicted by Otto cycle efficiency.
However, all of the experimentally observed ihp values for pure
methanol fuel as a function of A/F ratio are seen to be
significantly higher than those for 100% dissociated methanol.
This appears to be inconsistent with the fact that the dis-
sociated methanol has 20% more energy per unit mass than the
pure methanol. by virtue of its recovery of decomposition energy,
This apparent discrepancy is resolved by noting that the energy
per unit volume of cylinder intake and not the specific fuel
energy is the parameter which governs the engine power.
To consider changes in the intake charge energy with
dissociated methanol, a volumetric energy parameter depending
upon the degree of fuel dissociation was developed. This
parameter, which is based upon a dimensionless energy per unit
volume, is given by the equation:
0.2x
p K H
^=^) -4_ + A/F
o Mf,x
-------
37
(T
UJ
O
Q.
UJ
CO
O
UJ
I-
<
O
o
z
6.0
5.0
4.0
3-0
2.0
1.0
CR
8.55
9.20
10.90
CH3OH
COtH,
o
A
D
*
5-0
6.0
70
8.0 9.0
AIR/FUEL lba/lbf
FIG W-I: FUEL DISSOCIATION SCR EFFECT ON IHP
-------
38
E - energy per unit volume of change
(LHV) - lower heating value of pure methanol
p - total pressure in the intake manifold
M - molecular weight of air
a
MP - molecular weight of fuel blend
1 ,X
R - universal gas constant
T - absolute temperature
x - mass fraction of dissociated methanol
It was further assumed that the available fuel energy is
restricted at a fuel rich mixture of 5.5:1, according to the
reaction:
CH,OH + 1.17 [09] + 4.40 N9 + 0.67 CO + .33 C00 + 2 H90 + 4.40 N
O £. £ £. L.
Figure VII-2 shows the volumetric energy as determined from these
equations as a function of the methanol dissociation for various
A/F ratios.
Analysis of Fig. VII-2 indicates that if-the intake
pressure and temperature are held constant, the volumetric intake
energy for a given A/F ratio decreases slightly with increased
dissociation. This power reduction with dissociated fuel must
have resulted primarily from changes in the intake charge density
associated with charge temperature variations. This latter
postulation was confirmed by showing that the experimental engine
ihp performance with liquid and 100% dissociated methanol would
be almost identical under equivalent intake energy conditions.
Figure VII-3 shows the results of correcting for intake
charge effects. Factors accounted for included:
(a) Inlet manifold temperature differences between
the liquid and dissociated fuel.
(b) Charge density increases through endothermic
cooling from that portion of the fuel not
previously vaporized at the saturated vapor
conditions corresponding to the inlet manifold
pressure and temperature (see Fig. VII-4).
-------
LU
39
o
24
(T
LU
0.
>-LU
O
or
LU
z
LU
CO
CO
LU
O
CO
LU
0:° 20
16
12
8
A/F=6.43
A/F = 5.0
A/F = 7. 5
0
0
20
40
60
80
100
%DISSOCIATED FUEL
(MASS BASE)
FIG.W-2
-------
(T
UJ
O
0-
UJ
V)
tr
o
a
UJ
o
o
6.0
5.0
4.0
3.0 -
2.0
1.0
CR CH3OH CO+H2
5.0
6.0
70 8.0 9.0
AIR/FUEL lba/lbf
FIG.YD-3= FUEL DISSOCIATION &CR EFFECT ON IMP
-------
UJ
m
100
80
i
_J
UJ
3
U_
\
or
2 60
CE
? 40
30
20
10
8
4
3
10
- METHANOL CH^OH
AIR / FUEL VAP ( SAT.)
VS
TEMPERATURE FOR A
TOTAL PRESS = I ATM
VAPOR LIQUID ^
DROP MIXTURE
^VAPOR
ONLY
STOICHIOMETRIC
20
30 40
60 80 100
TEMP (°F)
FIG.M-4 : METHANOL AXF(VAPOR) VS TEMP
-------
In Fig. VII-3 curves A-A and B-B represent, respectively
the experimental results for the liquid and dissociated methanol.
Curve A1 - A' is a correction for the dissociated methanol,
curve A-A, taking charge density effects into account. Good
agreement is indicated. It is thus seen that fuel vaporization
effects and intake temperature differences existing in the
test work exaggerated the differences in power performance of
the pure methanol and the 100% dissociated methanol. Under
identical intake manifold conditions the ihp contrast would
be slight.
(2) Methanol and Gasoline Performance Comparison. The
difference in power performance of the CFR engine operated on
methanol and gasoline at equal equivalence ratios was found to
be slight in terms of power and thermal efficiency. The
maximum power the engine developed using methanol fuel at 900
RPM and a CR of 9.2 was 3.8 ihp. For gasoline at the same
conditions it developed 3.9 ihp. These values agree quite well
with what is expected when the fuels are compared on the
dimensionless energy per unit volume base.
0.136 for methanol
PTM.
LHV (-i-*)
o stoich.
= 0.137 for octane
Figure VII-5 shows experimental comparisons using the equiv-
alence ratio as a basis. The diminution of power from its peak
shows very similar results for the gasoline and methanol. The
indicated thermal efficiency of the two fuels are also quite
similar. The dissociated methanol produced a somewhat lower
thermal efficiency curve. An explanation for this reduced
performance warrants further investigation. As indicated by
the emission data, it is not due to incomplete combustion.
The general evidence thus indicates that there are no
major disadvantages in engine power and efficiency performance
when operated on methanol as compared to gasoline. These
-------
100
o
UJ
5
o
o
90
cc
UJ
I
70
60
RICH
METHANOIIX \
GASOLINE
LEAN
0.6
0.8
1.0
1.2
1.4
1.6
STOICHIOMETRIC A/F1
-I
ACTUAL A/F
FIG.YII-5: GASOLINE 6 METHANOL PERFORMANCE COMPARISON
-------
44
general observations are in agreement with others [8, 8a], The
relative performance merits of the two fuels must be established
by the amounts and types of pollutants that each fuel produces.
C. Emission Data:
Engine emission data are normally presented in volume per
cent of exhaust concentration or as parts per million (ppm) of
exhaust concentration. In analysis of the data it was found to
be desirable to use an additional parameter, grams of emission
per indicated horsepower hour (gm/ihp-hr), as an evaluating
criterion. This rating parameter indicates the emissions
produced per unit of work delivered to the piston face. It is
analogous to the "grams of pollutant per mile" used as a
vehicle rating. However, the latter rating is not particularly
meaningful for a laboratory engine operation at constant speed
and load. In contrast the gm/ihp-hr rating was found to be
particularly valuable in this investigation since the engine
power output varied with percentage of fuel decomposition. Its
use is encouraged as a generally useful parameter in comparing
performance of different fuels, engine sizes, and loadings.
A selected amount of the total data is graphically
presented for analysis. The three principal variables used for
evaluation of the test results were the A/F ratio, the compres-
sion ratio (CR) and per cent of fuel in dissociated form. In
a majority of cases the A/F ratio was used as the independent
variable and CR or per cent of dissociation were used as
parameters. Cross-plotting with per cent dissociation or CR
as the independent variable was made as necessary to illustrate
the effects of these variables. In some tests the effect of
the per cent dissociation on the independent variable was not
large and the data for intermediate dissociation percentages
were bounded by the zero and 100 per cent extremes. In these
cases, only the data for the extreme blends were plotted.
Analysis of portions of the emission data was hindered
because of significant data scatter and the large number of
interacting engine variables involved. For some emissions,
particularly the aldehydes, the data scatter made graphical
-------
interpretation of the results difficult. In these cases tabular
data were analyzed for identifiable trends.
The effect of engine test variables upon emissions was
qualitatively similar to that reported for gasoline. Hydro-
carbon and unburned methanol emissions were high at low and very
high A/F ratios and were reduced near stoichiometric conditions
(see Fig. VII-6). Total aldehyde emissions were low at low A/F
and increased for lean A/F ratios.
Increased spark advance tended to increase hydrocarbon,
aldehyde and NOy emissions at stoichiometric and lean A/F
ratios (see Table VII-2). For these cases, maximum power spark
advance occurred between 10° and 15° before top dead center
(BTDC). Emissions improved appreciably below 10° with small
loss in power. At rich A/F ratios the effect of spark advance
was not as pronounced.
The percentage of fuel dissociation had varying effects
upon the different emission species (see Fig. VII-7). As one
would expect, unburned methanol concentrations were decreased
for increasing fuel dissociation at all A/F ratios. At high
dissociation percentages the decrease in methanol concentrations
appeared to be at the expense of increased hydrocarbon and
aldehyde emissions. For these cases the reaction kinetics
appeared to favor methane production at low A/F ratios and
acetaldehyde production at high A/F ratios. NO,, emissions were
increased with increasing dissociation at both high and low A/F
ratios.
Comparison of methanol and gasoline emission data was
made for a limited number of test conditions. Near stoichio-
metric the total grams of carbon and per. cent of CO in the
exhaust were in general an order of magnitude lower for methanol
than for gasoline. NO., emissions were equivalent for both fuels.
A further discussion of the effect of engine operating
variables on major emission species and comparison of gasoline
and methanol emission data are presented in the following
paragraphs.
-------
1800
1600
or
x
i
Q. 1400
O
QD
o:
<
o
L_
O
1200
1000
i
-I
- 800
600
400
200
CR=920
CH.OH CO*H2
HYDROCARBON
TOTAL
METHANOL
METHANE
OTHER
CH,OH FUEL
+CO FUEL
56789
AIR/FUEL lba/lbf
FIG2n-6-- EXHAUST HYDROCARBON DISTRIBUTION
-------
TABLE VII-2
SPARK ADVANCE EFFECT ON EXHAUST EMISSIONS
Test
142
141
138
137
161
162
163
164
165
61
62
63
64
65
71
72
73
74
75
A/F
6.5
6.5
6.5
6.5
7.5
7.5
7.5
7.5
7.5
5.4
5.5
5.5
5.5
5.5
7.5
7.5
7.5
7.4
7.4
CR Oils SA Tim ihp isfc C02 °2
10.9
10.9
10.9
10.9
10.9
10.9
10.9
10.9
10.9
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
A/F
CR
% Diss
SA
T.
im
ihp
isfc
--
Tr
(%) (T)
30 2 70 3.52 .985 10.82 1.24
30 10 70 3.58 .97 10.93 1.04
30 20 70 3.55 .98 10.93 1.03
30 23 70 3.49 .98 10.97 0.95
30 1 70 3.10 1.005 9.54 3.55
30 10 70 3.27 .455 9.55 3.53
30 15 70 3.31 .94 9.53 3.56
30 20 70 3.26 .955 9.53 3.57
30 23 70 3.22 .97 9.61 3.42
30 2 73 3.80 1.07 9.87 0.07
30 10 72 3.73 1.04 9.97 0.09
30 12 71 3.77 1.07 10.06 0.09
30 20 71 3.61 1.12 9.79 0.11
30 23 72 3.57 1.14 9.74 0.11
30 2 70 3.01 1.01 9.02 4.49
30 10 70 3.14 0.96 9.02 4.49
30 14 70 3.25 0.93 9.18 4.20
30 20 70 3.22 0.94 9.22 4.14
30 23 70 3.11 0.97 9.19 4.19
air- fuel ratio
compression ratio
per cent of methanol in dissociated
spark advance
intake manifold temperature
indicated horsepower
indicated specific fuel consumption
no data
trace of specie less than . 5 ppm
CO
(%)
0.08
0.08
0.09
0.09
0.08
0.08
0.09
0.09
0.09
2.12
1.97
1.85
2.18
2.26
0.08
0.08
0.08
0.08
0.08
form
NOX
(%)
0.299
0.347
0.404
0.406
0.142
0.252
0.295
0. 357
0.407
0.142
0.173
0.187
0.167
0.166
0.101
0.206
ND
ND
ND
CHgOH
(ppm)
40
83
115
114
59
114
159
134
133
237
227
207
249
223
101
130
162
106
— _
OTJ f»
^"14 L
(ppm)
47
47
50
48
12
16
24
21
21
483
476
431
654
548
9
18
35
40
40
2H6+U
(ppm)
16
17
17
17
9
10
12
11
11
11
16
16
17
16
4
8
10
12
11
C3H6
HCOH
(ppm) (ppm)
1
1
Tr
Tr
Tr
_
Tr
Tr
Tr
1
Tr
Tr
Tr
1
Tr
Tr
Tr
Tr
Tr
37
44
52
60
26
47
53
46
36
62
71
77
64
61
68
47
68
62
61
CH3CHO
(ppm)
127
138
139
139
4
55
221
374
237
12
8
17
12
15
10
36
128
384
487
, Ib/hp-hr
-P
^
-------
20 40
60 80 100
% DISSOCIATION
O5.40
20 40 60 80 100
% DISSOCIATION
40 60 80 100
% DISSOCIATION
X
QL
O
U.
O
CO -
U> 8
40 60 80 100
% DISSOCIATION
FIG.VII-7' EXHAUST EMISSIONS VS METHANOL DISSOCIATION
-------
(1) Carbon Monoxide. Consistent with the behavior of
hydrocarbon fuels the methanol and dissociated methanol exhibit
a rapid increase in CO as the A/F ratio moved into the fuel
rich region. Similarly, they were at trace levels at A/F
ratios in excess of stoichiometric (see Fig. VII-8a,b). The
most significant contrast between the pure methanol and the
dissociated methanol was found at the stoichiometric condition.
At this condition the dissociated methanol shows a threefold
higher percentage of CO in the exhaust than does the pure
methanol. It is postulated that this results from the
probability that any unburned fuel in the 100 per cent dis-
sociated methanol would preferentially show up as unreacted CO,
while in the case of liquid methanol the unburned fuel would
more probably be unburned methanol or hydrocarbons. If this is
true, then the CO exhaust concentrations at all fuel rich
mixtures should be higher for dissociated than undissociated
methanol. This is seen to be confirmed by the data in Fig. VII-8a,
b.
(2) Hydrocarbons. The major unburned fuel species in the
exhaust were methanol and methane.* The amount of methanol and
methane were highly dependent upon A/F and per cent dissociation
and to a lesser extent on CR. Other hydrocarbons, including
C2Hg and C^Hg were also present but were in small concentrations
relative to the combined amounts of CH-OH and CHU components.
Large methanol exhaust concentrations were noted at low
and high A/F ratios for engine operation on liquid methanol
(see Fig. VII-9). These high concentrations are attributed to
insufficient oxidizer for rich mixtures and poor combustion
efficiency for lean mixtures. The advantage of combusting
dissociated methanol was most pronounced for very rich mixtures
and lean A/F ratios. As expected, methanol exhaust concentra-
tions decreased proportionally with increased per cent dis-
sociation until only trace concentrations were present for
completely dissociated fuel.
Although methanol is an alcohol rather than a hydrocarbon
it is grouped with the latter for ease of reporting.
-------
CH3OH
C0+H0
I
X
LJ
O
O
LJ
cr
LJ
Q_
AVE FOR CH.OH
3
6789
AIR/FUEL lba/lbf
FIG. 301-80: CARBON MONOXIDE EMISSIONS,%
tr
x
i
0_
X
\
O
O
O
CO
160
140
120
100
AVE FOR CO*H2
AVE FOR CH^DH
50
FIG
5 6
YO-Sb. CARBON MONOXIDE EMISSIONS.GRAM/IHP-HR
789
AIR/FUEL lba/lbf
-------
51
800
<
»-
LJ
Q.
a.
600
400
200
7 8
AIR/FUEL Ib./lb
FIG.W-9a= METHANOL
a' '"f
EMISSIONS , PPM
0
6789
AIR/FUEL lba/lbf
FIG.W-9b= METHANOL EMISSIONS, GRAMS/IHP-HR
-------
52
Significant concentrations of methane were found in the
engine exhaust when it was operating at rich A/F ratios (see
Fig. VII-10a,b). It was also noted that dissociated methanol
produced a higher yield of CH. than the pure methanol in this
region. The source of this methane is probably in part from
recombination of the CO and H2 according to the reaction:
CO + 3H2 •> CH4 + H20
This process was shown by Sabatier to proceed readily at tem-
peratures above 200°C in the presence of iron [9]. A related
reaction may also be the source of the ethane found in the
exhaust. The relative absence of methane at higher A/F ratios
probably results from the higher oxygen concentrations which
enable oxidation of methane.
The total carbon exclusive of methanol in the hydrocarbon
exhaust components given as microgram carbon/gram exhaust and as
gram carbon/ihp-hr are shown in Fig. Vll-lla, b. Hydrocarbons
exclusive of methanol and methane were found only in small
concentrations (compare Fig. VII-6 with Fig. Vll-lla). With one
exception, the maximum concentration for all these species was
less than 40 ppm for all conditions (see Table I, Appendix B).
As expected, these hydrocarbon concentrations were highest at
low A/F ratios. As in the case of methane, the concentrations
for dissociated fuel were generally higher than for liquid
methanol. This would indicate that the H~ and 0,, react more
readily to form hydrocarbons than does methanol.
It was difficult to assess the exact effect of CR on the
methanol, methane and other hydrocarbon emissions. Sometimes
they appeared to increase then decrease with increasing
compression ratio. For other conditions no appreciable changes
were noted.
(3) Oxides of Nitrogen. Changes in A/F ratio and the
per cent fuel dissociation were shown to have a pronounced effect
on the level of NOV emissions (see Fig. VII-12a). As with
A
gasoline, liquid methanol was shown to exhibit a large NO,, peak
near stoichiometric conditions. At fuel rich conditions there
-------
1600
o
1200
800
400
i'
2350
a
o
0 „
5678
AIR/ FUEL lba/lbf
FIG. Sff-lOa METHANE EMISSIONS, PPM
DC
I
I
a_
o
u_
o
C/)
oc
CD
8
53
5678
AIR/ FUEL lba/lbf
FIG-YE-lOb: METHANE EMISSIONS, GRAMS/ IHP-HR
-------
en
ID
800
2
O
CD
O
8
AIR/ FUEL lba/lbf
TOTAL EXHAUST HYDROCARBONS,
MICROGRAMS/GRAM EXHAUST
QL
X
O
CD
or
<
O
• A
56789
AIR /FUEL lba/lbf
FIGTfl[-llb= TOTAL EXHAUST HYDROCARBONS,
GRAMS/IHP-HR
-------
or
i
i
CL-
OT
O
56789
AIR/FUEL
FIG.W-l2a: OXIDES OF NITROGEN EMISSIONS,
GRAMS /IHP-HR
55
AVE.FOR CO*H2
5678
AIR/FUEL
FIGM-12b .OXIDES OF NITROGEN EMISSIONS,
GRAMS/IHP-HR
-------
56
was little observable difference between the blends and liquid
fuel. However, on the lean side there is a pronounced effect--
100% dissociated methanol produced far more NO,, per unit of
energy delivered than did pure methanol. Since the dissociated
methanol has 20% more energy per pound of fuel, the rise in
NOV with dissociation in the lean A/F range is consistent with
X
the expectation of higher cylinder temperature. Why this is
not matched by a like gain in isfc requires further inquiry.
Compression ratio effects when contrasted with A/F and
fuel dissociation effects on the N0y emissions were not very
significant, as shown in Fig. VII-12b.
(4) Aldehydes. The aldehyde emissions consisted almost
exclusively of formaldehyde (HCHO) and acetaldehyde (CH3CHO).
Both were found to be appreciably affected by A/F ratio, per
cent dissociation, CR and spark advance. Formaldehyde concentra-
tions varied from a low of 3-4 ppm to a maximum of 170 ppm with
the normal range of values being between 20 and 80 ppm.
Acetaldehyde exhibited a much greater range of variation--from
trace values to almost 500 ppm. The tabular data did not show
any evident interdependency of the formaldehyde and acetaldehyde
emissions. However, a clear relationship between the acetaldehyde
and methanol emissions was sometimes seen. For some conditions
when acetaldehyde emissions were high, methanol emissions were
low, indicating that methanol emissions are reduced at the
expense of increased aldehyde emissions. It should be emphasized
that this relationship did not always hold and that under some
conditions methanol and aldehyde emissions were both low.
The A/F ratio, compression ratio and spark advance were
the principal variables which determined the level of aldehyde
emissions. At low A/F (from 5.0 to 5.5) the total aldehyde
emissions were low regardless of compression ratio or spark
advance, the worst case being slightly greater than 100 ppm.
At stoichiometric or higher A/F ratios, the compression ratio
and spark advance significantly affected aldehyde emissions.
Moderate to high spark advances (14-25°BTDC) were in general
associated with increased aldehyde emissions, particularly at
-------
57
high A/F and CR ratios. Although some large aldehyde emissions
were noted at low spark advances (0-10°BTDC), they were on the
average lower than, for large spark advances and generally
appeared in conjunction with significant decreases in the
methanol and hydrocarbon emission levels. The increased emis-
sions at high A/F ratios and spark advances are seen to be in
agreement with aldehyde emission trends in gasoline [10]. The
effect of CR on aldehyde emissions was not clear.
Fuel dissociation was found to promote aldehyde formation.
The worst aldehyde emissions for pure methanol fuel was 205 ppm
for test 57 (A/F = 7.5, CR = 10.5, spark advance = 20.0).
Worst case emissions for 30, 70, and 100 per cent dissociated
fuel were 455 ppm, 425 and 428, respectively. These emissions
were also obtained at A/F = 7.5 and CR = 10.5.
Because of data scatter and lack of emission data at
some test conditions it was difficult by graphical methods to
assess the exact effect of various engine variables on formalde-
hyde and acetaldehyde emissions. Figure VII-13 presents
formaldehyde emission data for maximum power spark advance for
various CR and per cent dissociation as a function of A/F ratio.
The only clear trends are an apparent reduction of emissions
for the 'dissociated fuel and a tendency of increased data
scatter for liquid methanol emissions at increasing A/F ratios.
Also, the average formaldehyde emissions for dissociated fuel
appear to be higher than for the liquid fuel at low A/F, while
just the opposite is seen to be the case at high A/F ratios.
Selected acetaldehyde emission data are shown in Fig. VII-14, as
a function of spark advance and compression ratio. For 30 per
cent dissociation at compression ratios of 8.4 and 10.9, the
acetaldehyde emissions are seen to be significantly affected
only at an A/F ratio of 7.5. For A/F = 7.5, the aldehydes
increase rapidly with increased spark advance. At other A/F
ratios, the effects of spark advance appears minimal. In
Fig. VII-14 no clear trend is seen with compression ratio
except that the variation with compression rates is highly
dependent upon A/F ratio.
-------
180
Q.
CL
UJ
O
UJ
O
cr
o
u.
160
140
120
100
80
60
40
20
D
>
A
D
A
< X
O
4
I
I
/>
•
A
A
i
L
' 6
•
i
V
1
A
AVE. FOR /
X
D
A
A
^o±
. A
CR CHjOH CO+H2
8.55 • 0
9.!
• I0.1
X
•
A
kLL CH3OH
A
D
1
A
1
20 A A
30 • D
AVERAGE VALUES
FOR DISSOCIATED
METHANOL
<
1
1
RUNS
4
>
1
1
»
-i
X
A
1
89
AIR/ FUEL Ib/lb
f
58
FIG. 2DI-I3 : FORMALDEHYDE EMISSIONS
-------
59
0487
400
I I
CR = 8.40
30% DISS
A/F
O 3-50
A 6.50
Q 7.50
t 8.90
CR=I0.90
30% DISS
A/F
A 6.50
7.50
15 20 25 30
SPARK ADVANCE, "BTDC
10 15 20 25 30
SPARK ADVANCE,°BTDC
160
Q.
Q.
Ill
O
X
o
_l
<
»-
LU
O
I 20
A/F
O 5-50
A 6.50
+ 6.90
160
o.
Q.
Ul
O
>•
X
UJ
o
Ul
o
120
ao
9.0 as 10.0 10.5 11.0
COMPRESSION RATIO
8,0
9O
9.5 IOO 105 II,O
COMPRESSION RATIO
FIGffl-14' AC ET ALDEHYDE DATA
-------
60
(5) Oxygen and Carbon Dioxide. The values of 0^ and C02
in the exhaust vary with A/F ratio, as expected, and are similar
to those observed for gasoline when stoichiometric mixture is
used as a reference point (see Figs. VII-15 and VII-16). It is
of some interest to note that in the lean mixture region the
dissociated methanol shows a more complete combustion capability
in that there is more CO- and less 02 present. This is probably
a partial explanation for the relatively higher NO^ values
observed previously to be produced by the dissociated methanol
in this region.
D. Comparative Gasoline and Methanol Emissions:
Only tentative comparison of gasoline and methanol
emissions can be made because of the limited gasoline emission
data acquired during the test program. Also, the GC emission
data for gasoline were not broken down into individual species
but were reported as grams carbon per grams of exhaust. Thus,
the relative portions of aldehydes and hydrocarbons in the
methanol and gasoline CFR engine tests cannot be compared.
Gasoline and methanol emission data at stoichiometric A/F
ratios and at otherwise comparable test conditions are presented
in Table VII-3. The CO and total grams carbon per ihp-hr are
seen to be in general almost an order of magnitude higher for
gasoline than for methanol. The NOV data are seen to be
/\
approximately equivalent. Comparative CO and NO., data for
gasoline and methanol at other than stoichiometric conditions
are shown in Fig. VII-l7s plotted against the inverse of the fuel
equivalence ratio (actual A/F per stoichiometric A/F). This data
indicates that the methanol NO,, emissions for liquid methanol are
less tiian those for gasoline for all 4>,while the fully dissociated
methanol N0y are higher than gasoline NO., for lean A/F ratios. The
methanol CO emissions for both liquid and dissociated fuel are
reduced to trace values at much lower equivalence ratios than
the gasoline CO emissions. This latter factor seems to indicate
a higher combustion efficiency at equal equivalence ratios than
gasoline.
-------
61
o
CO
O 7
• ,
LL.
o
H 6
LJ
O
or
uj 5
Q_
4
3
2
1
O
—
—
^10
1
CR CH.OH CO+H,
<3 C
8.55 • o
9.20 A A
-10.90 • D
^^§9*3
*
§
I
1
%
A
8
A
1
1
•
1
1
1
A
A
1
5 6
FIG.M-15;
789
AIR / FUEL lba/lbf
OXYGEN EMISSIONS
-------
62
\
-------
50
I
Q.
-------
TABLE VII-3
Run
%
Diss
A/F
«,*
CR
SA
In
Tern
-OY
ihp
isfc
(Ib/hp-hr)
CO
rw
TkT f\
NOX
(%)
Total
Exhaust
Carbon
(gm/ihp-hr)
Gasoline :
185
188
191
NA
NA
NA
14.9
14.0
14.9
1.01
1.07
1.01
9.2
9.2
9.2
18
18
18
86
56
88
3.90
3.90
3.90
.461
.492
.461
2.51
—
2.76
.306
.527
.427
3. 32
3.74
--
Methanol:
7
13
22
24
120
32
123
158
0
0
30
30
30
70
70
100
6.5
6.5
6.5
6.5
6.5
6.4
6.5
6.5
1.00
1.00
1.00
1.00
1.00
0.99
1.00
1.00
9.2
9.2
9.2
9.2
9.2
9.2
9.2
9.2
14
14
14
14
10
9
7
6
70
73
70
70
110
70
110
110
3.65
3.79
3.58
3.58
3. 36
3.45
3.22
3.15
.98
.97
.99
.99
0.99
1.02
1.00
1.01
0.08
0.17
0.08
0.12
0.08
0.53
0.17
0.48
0.391
0.372
0.471
.390
0.302
ND
0.395
0.320
.213
. 316
—
.198
.500
.356
.460
.617
- equivalence ratio
A rough comparison of these emissions with federal and
California state standards is possible if the assumption is made
that the basis for generating these standards is equivalent to
steady-state operation of a vehicle at 50 mph while developing
50 ihp. With this assumption, the emissions in gm/ihp-hr can be
converted to gm/mile on a one to one basis. Then, from
Fig. VII-17, it is seen that at an equivalence ratio of 1.0
the CFR engine CO emissions for gasoline (82 gm/mile) are about
twice the 1972 federal standard of 39 gm/mile. At the same
equivalence ratio the methanol CO emissions are less than half
the 1972 federal standard. For NO.,, both the gasoline and
methanol are seen to exceed by a considerable amount the
tentative range for the 1975 federal standards of 0.3 to 0.4
gm/mile. The total carbon in the gasoline exhaust at stoichio-
metric A/F ratio, as given in Table VII-3, includes hydrocarbons
and aldehydes. Hence, the CFR engine hydrocarbon emissions are
seen to be less than the 1972 federal standard of 3.4 gm/mile.
-------
65
This is as expected considering that the federal standard is
based upon a cycle average while the CFR engine data are for
steady-state operation at stoichiometric A/F ratio. In
comparison, the total methanol carbon emissions including
hydrocarbon, methanol and aldehydes are significantly less than
the 1972 federal standards. These limited data indicate that
the methanol fuel at stoichiometric and leaner A/F ratios has
inherently less CO and hydrocarbons in the exhaust than
gasoline.
-------
66
VIII. DECOMPOSITION FUELED ENGINE DESIGN
The following section presents preliminary studies of
possible methanol decomposition methods, the related engine
control, and exhaust energy requirements needed for automotive
engine operation on partially or fully decomposed methanol
fuel.
A. Engine Energy Analysis:
It is proposed that methanol decomposition for engine
consumption utilizes engine exhaust energy for fuel preheating,
vaporizing, superheating and decomposing. At atmospheric
pressure approximately 2438 BTU are required to heat one pound
of liquid methanol from 68°F to 554°F (405° superheat) and
decompose it to CO and H~. The decomposition energy
(1720 BTU/lb) is 70.7% of this heat requirement (see Table VIII.1),
Total exhaust energy recycle requirements are defined by
the fuel consumption and desired percentage of methanol decomposi-
tion at given engine operating conditions. An estimate of total
recycle heat requirements for various operating conditions may
be obtained by examining methanol fuel consumption over a range
of engine loads and speeds. Comparative engin'e performance
characteristics at one-third full load for gasoline and methanol
fueled engines were calculated using gasoline engine performance
mappings [11] and assuming that the methanol isfc was twice that
of gasoline (see Fig. VIII-1). The relative gasoline to
methanol isfc value was obtained from tests on the CFR engine.
Methanol consumption for these speed and load conditions is
presented in Fig. VIII-2, along with energy rate requirements
for 30 per cent methanol decomposition. Energy rate requirements
for other decomposition percentages would be proportional.
The significance of these heat requirements is clarified
by comparing them with the available exhaust energy. The
available exhaust energy is a function of engine load, fuel •
consumption, speed and other variables and cannot be rigorously
predicted by analytical methods alone. However, an estimate of
-------
67
120
2 100
UJ
2.5 o
o
2.0
UJ
:D
o
o
UJ
Q.
W
1.5
1.0
0.5
8
16 24 32 40 48
ENGINE SPEED RPM/IOO
FIG.W-P ENGINE PERFORMANCE CHARACTERISTICS
1/3 FULL LOAD
-------
68
140 _
1000
2000
3000
4000
5000
ENGINE SPEED -RPM
FIG.Snr-2' METHANOL FUEL CONSUMPTION AND DECOMPOSITION
ENERGY RATE AS A FUNCTION OF ENGINE SPEED
-------
69
TABLE VIII-1
Energy Requirement Per cent
(BTU/lb) of total
Preheating
68° to 148°F
Vaporization
1480 @ 1 atm
Superheating
148° to 554°F
Decomposition
Total:
48
482
188
1720
2438
1.9
19.7
7.7
70.7
the required percentage of exhaust energy as a function of
decomposition can be made by using published engine data to
establish base line conditions and by assuming that the exhaust
energy will vary with engine performance in accordance with
simple analytical expressions. Using this approach a computer
program was written and results were obtained for a limited
range of engine operating conditions. Assumptions and method of
calculation used in this computer program are discussed in
Appendix C. Basic variables were RPM, per cent dissociation,
inlet temperature and isfc. The A/F ratio was held constant at
6.5:1. An energy and mass flow diagram showing the counterflow
heat exchange cycle used in the program is shown in Fig. VIII-3.
For this cycle, the decomposed and undecomposed energy circuits
are separated. Calculations indicate that at normal conditions
sufficient exhaust energy is available to decompose 30 and 100%
of the methanol (see Figs. VIII-4 and VIII-5). This data
indicates that the exhaust contains sufficient energy to
perform its decomposition and vaporization fuctions; however,
-------
DECOMPOSED FUEL CIRCUIT
HEATER EVAP HEATER REFORM'
QPR
OR)
T
QER
(2R)
T
QSR
(3R)
14R)
XW,
m
QRR
(OM) (IM)| |(2M)| -|(3M) (l-x)Wm
QPM OEM
QSM
UNDECOMPOSED FUEL CIRCUIT
(14)
QPM OEM
t f
KI3)|
(12)
QSM QPR
^ r-*-
(II)
(10)
QER QSR
(9)
(5)
COOL
(8)
QRR
f
Wm(A/F)
(a)
(6)
ENGINE
(7)
FIG.3fflI-3 : ENGINE ENERGY SCHEMATIC
o
-------
Wm = 90.8 Ib/Hr
RPM = 2800
ISFCN= 0.938
DISSOCIATION = x=0.30
low
1600
or
° 1400
£ 1200
13
< 1000
or
UJ
^ 800
Ul
•" 600
400
200
0
EXHAUST
FINAL
il=x}^
IN
m7Q%
/
4
0
10
10
try
o
\
^
(l-x)W
OUT
IM
xWm
EXHAUST
CO
O
I
^^
.
-"
in
(D
rf>
\
f
. — • —
f
3
(D
0)
1
f
rw ^
(Jlfc^
/
/
cn
CO
00
ro
1,
/
LXMAU5I
AVAILABLE
x Wm
OUT\
•X
(l-x)Wm \
OUT /
AIR
PRE EVAP SUP PRE EVAP SUP REFORM
HTR HTR HTR HTR
LOCATION INv SYSTEM
0
f
1
1
ENGINE
1
P
* BTU/HR
UNDECOMPOSED
FUEL
DECOMPOSED FUEL
FIG-VDT-4 TYPICAL TEMPERATURE PROFILE 30 % DISSOCIATION
-------
= 90.80 Lb /Hr
RPM = 2800
IFSCN=0.938
DISSOCIATION = x =1.0
C. \J\J\J
1600
(T
°. 1600
a:
z
uj 1400
1200
1000
800
600
400
200
0
—
PRE
HTR
EVAP
EXHAUST
FINAL
Hc_, — J
wm
IN"*"
SUP
HTR
,> *_,
o>
10 1
tf\
n3 1
CM
11 I
0 f
^-— -^
PRE
HTR
/
CD
in
10
r,
N-
00
fO
1 !
EVAP SUP
HTR
1
f
L
(VI
10
P
/
/
EXHAUST
AVAILABLE
Wm
OUT \
AIR /
Q
t
ENGINE
\
I
P
REFORM
j
UNDECOMPOSED FUEL
DECOMPOSD FUEL
FIG.W-5' TYPICAL TEMPERATURE PROFILE 100% DISSOCIATION
-------
73
at 100 per cent decomposition, the exit temperature difference
between the exhaust and decomposition streams drops to 400°F.
This low temperature difference would require large heat
exchange elements.
B. Decomposition Chamber Design Analysis:
(1) Design Considerations. Catalytic decomposition is
essentially an empirical art. However, a decomposition
chamber design analysis with data from the literature will
provide information on the probable size, performance,
operating characteristics and problem areas.
A survey was conducted of catalytic decomposition and
synthesis. Performance data was found for synthesis of methanol
with currently used catalysts. In addition, decomposition data
was acquired in the open literature from basic research for a
number of catalysts tested over a limited range of variables
[12, 13].
Methanol is synthesized or decomposed according to the
reaction:
decomposition
CH3OH + CO + 2H2
synthesis
Decomposition is endothermic with an energy requirement
of 1720 BTU/lb. Other reactions involving decomposition to
formaldehyde, methylformate, methane, H2, and 0^ are also
possible but low in probability and are not of major con-
sequence. Catalysts for methanol decomposition and synthesis
contain ZnO, CuO, A1203, or Cr^O,,, as the primary constituents
[12,13,14,15,16]. The exact composition for commercial cat-
alysts is proprietary. A catalyst used for synthesis may also be
used for decomposition, although one optimized for synthesis
will not necessarily be optimized for decomposition. With a
given catalyst the optimum reaction (decomposition or synthesis)
is determined primarily by the catalytic bed pressure. High
-------
pressures (1000 to 5000 psia) promote synthesis while low
pressures favor decomposition.
Methanol catalysts are generally made up into small
porous pellets and are packed into a catalytic bed into which
the gaseous feed stock is introduced. Because of their
porosity the catalytic pellets have low thermal conductivity
and compressive strength. Their effective surface to volume
ratio is high and their heat capacity is low.
Prior to initial use, methanol catalysts are generally in
an oxidized state for ease of handling. After installation in
the catalyst bed they are reduced by heating in a hydrogen
atmosphere. In the reduced state, although not pyrophoric,
they react readily with oxygen. Since oxygen reduces the
catalytic effectiveness, air must be excluded from the decom-
position chamber. A partially oxidized catalyst can be regen-
erated by reduction. Sulfur and chlorides are permanent poisons
and should be excluded from the methanol feed stock.
Severe pressure and temperature cycling during continuous
methanol synthesis can reduce catalyst effectiveness after
30 days by about 10 per cent [I1*]. A similar degradation might
be expected for continuous decomposition cycling. The effects
of cyclic on-off operation of the type expected in an automotive
engine decomposition chamber have not been assessed. It has been
reported that condensation of methanol in the catalyst bed
should be avoided since reheating and vaporization of saturated
pellets will destroy some types of pellets. However, this can
probably be avoided by reducing thermal shock through gradual
heating of the bed and by use of a low initial feed of super-
heated methanol during start up.
The rate of decomposition or synthesis in a given catalyst
bed is determined by the space velocity, the pressure and the
temperature. Optimum temperatures, which are pressure dependent,
range from approximately 450 to 750°F [12,14]. The rate of
synthesis increases with pressure with a higher yield at
-------
75
pressures of 2000 psig and above. Figure VIII-6a illustrates
the effect of pressure and temperature on methanol synthesis
and Fig. VIII-6b shows the variation of decomposition with
temperature. Although no data was found on the effect of
pressure on decomposition, it is believed that pressures of
one atmosphere or less are best for decomposition. This
is concluded from the facts that all reported data on
decomposition were at one atmosphere and the synthesis rates
show an increase with pressure.
The catalyst bed space velocity is defined as the ratio
of the gaseous feed stock volumetric flow rate at standard
conditions to the effective catalyst volume. The space velocity
is an empirical sizing parameter which defines the relative
volume of catalyst to react a given volumetric flow rate.
Data for the Catalysts and Chemicals, Inc. (CCI) C-79 catalyst
(CuO, ZnO, A^O-) indicates that methanol yields at a given
temperature and pressure increases linearly for space velocities
between 10,000 and 40,000 hr"1 (see Fig. VIII-7). No decom-
position data showing the effects of space velocity on per-
centage decomposition was found. From data for a zinc-copper-
chromium oxide catalyst at a space velocity of 25 hr~ , it is
estimated that decomposition percentages above 90 per cent can
be expected (see Fig. VIII-6b). As will become evident later,
3
the space velocities of interest here are of the order of 10
hr~ . While it is not rigorous to assume that decomposition
percentages will increase linearly in an inverse manner when
compared with the synthesis yield, it seems reasonable to assume,
based on Fig. VIII-7, that the decomposition space velocity can
be increased over a considerable range without a drastic
reduction in decomposition percentages. However, it is not
possible to predict with confidence from the data available if
the dissociation yield for the desired space velocity will be
satisfactory. This question can only be determined by further
exploration.
-------
o 4
o
QC
x
\
o
o
o
UJ
X
UJ
o
-1
uj 3
0
SPACE VELOCITY = 10,000 Hr
( FTEF 14 )
2000 psig
FEED GAS COMPOSITION
18.0% CO
5. 0% C02
77. 0 % H,
1000 psig
420
440
460
480
500 520 540
TEMPERATURE ( °F)
FIG.2nr-6o : EFFECT OF TEMPERATURE AND PRESSURE
ON SYNTHESIS YIELD
100
SPACE VELOCITY= 25 Hr'1
(REF 12)
CARBON MONOXIDE
200 210 220 230 240 250 260 270 280 290
TEMPERATURE (°C )
FIG5Zm-6b' EFFECT OF TEMPERATURE ON DECOMPOSITION
( P= I ATM )
-------
77
U
O
o:
i
V.
O
O
UJ
UJ
Z
o
o
UJ
o
>
FEED GAS COMPOSITION
8.0% CO
5.0% C02
77-0% H2
10,000 20,000 30,000 40000 50,000
SPACE VELOCITY (VOL/VOL/HR)
FIG.Effl-7 : EFFECT OF SPACE VELOCITY ON METHANOL
SYNTHESIS (T=475°F)
-------
78
A primary design objective for a decomposition chamber and
associated equipment is to supply the high endothermic heat of
decomposition (1720 BTU/lb). Also, acceptable decomposition
rates are achievable only at temperatures in excess of 400°F.
These temperature and energy requirements can be met by super-
heating the methanol feed stock above 400°F prior to contact
with the catalyst, direct heat transfer to the catalytic bed, or
by both.
If pure methanol is superheated then the amount of energy
available for decomposition (assuming no methanol condensation)
is limited to superheat energy. To completely decompose one
pound of methanol requires energy approximately equivalent to
2300°F of superheat. Thirty per cent decomposition requires
almost 1000°F of superheat (see Fig. VIII-8). Since most
catalysts are not capable of withstanding temperatures above
600°F, and since the rate of decomposition falls off drastically
below approximately U60°F, the available range of superheat is
limited to approximately 140°F. With this constraint, only
3.69 per cent of the methanol can be decomposed during a single
pass.
Possible alternate designs which can be used to provide
the necessary decomposition energy are: a recycle separator
system, a series heating and decomposition system or a system
combining catalyst bed heating with recycling or series heating
(see Fig. VIII-9). In the recycle separator system, undecom-
posed methanol exiting from the catalyst bed would be separated
from the CO and H2 and would be recycled (see Fig. VIII-9a).
The separator process would be involved in terms of energy
exchange and efficient separator design. Heating of the
catalyst bed by itself does not appear to present a reasonable
method of providing the required decomposition energy.
Combined methanol and catalyst bed heating is more feasible
(see Fig. VIII-9b). However, the percentage of total energy
supplied via catalyst bed heating would be limited by. the low
catalyst thermal conductivity and heat capacity, especially
-------
79
I-
X
0 £
uj o
h- Q.
2500
2000
-*
ui z
1500
1000
o cc
UJ O
o u.
500
DECOMPOSITION
CHAMBER
20
40 60 80 100
PERCENT METHANOL DECOMPOSITION
FIG.SDI-8 • REQUIRED SUPERHEAT TEMPERATURE AS A
FUNCTION OF METHANOL DECOMPOSITION
-------
CH5OH
FEED
CH_OH
FEED
VAPORIZER -
SUPERHEATER
DECOMPOSITION
CHAMBER
PUMP
START-UP
BY PASS
H,,CO
SEPARATER
(CONDENSER)
RECYCLE CH3OH
o. RECYCLE SEPARATOR SYSTEM
VAPORIZER -
SUPERHEATER
11 11 i,
m
(CH,OH)
3 ^ VAPOR
CO, H2
HEATED BED DECOMPOSITION
CHAMBER
b. HEATED BED DECOMPOSITION
CH3OH
FEED
VAPORIZER-
SUPERHEATER
SUPERHEATER
V^n_ wn
CO. H«
VAPOR
DECOMPOSITION
CHAMBER
DECOMPOSITION
CHAMBER
c. SERIES HEATING AND DECOMPOSITION
FIGWI-9: DECOMPOSITION CHAMBER SCHEMES
-------
81
during transient start up periods. The best system for the
current application is believed to be an alternate series of
heating and decomposition chambers, as shown in Fig. VIII-9c.
In this system each of the decomposition and heating stages
would be designed to operate over the optimum superheat decom-
position temperatures ranging between 460° and 600°F. Some
heating of the catalyst bed could also be expected because of
the anticipated sandwich type structure of the alternating
decomposition and heating sections.
(2) Decomposition-Chamber Design. Based upon the design
considerations in the previous section, a series heating
decomposition bed with limited catalyst heating is believed to
be the most feasible design configuration. Figure VIII-10
shows a schematic of a compact series decomposition-heating
chamber. Calculations based upon 620°F superheated methanol
inlet into the decomposition stages and 460°F inlet into the
heating stages indicate that 10 states will be required for
30 per cent methanol dissociation (any heat transfer to the
catalyst is neglected).
Sizing of the decomposition chamber is not possible in a
rigorous manner because of the lack of catalyst decomposition
performance data. However, verbal information from a catalyst
manufacturer indicates that a catalyst volume of 0.25 cu ft
should be adequate for decomposition of 42 Ib/hr. The heat
transfer stages could reasonably add an additional .12 cu ft.
This total is about that required by a muffler on a current
automotive engine. In fact, the baffled heat transfer surfaces
in the exhaust side of the decomposition chamber would
obviate the need for a muffler.
C. Engine Performance Control:
Decomposition chamber design studies and engine energy
exhaust analyses indicate a methanol decomposition limit of
approximately 30 per cent at a total fuel flow rate of 120 Ib/hr.
Assuming that a practical decomposition device can be provided,
-------
(g)
CH3OH(g)
IN
SUPERHEATER
TO ENGINE
CATALYST
BEDS
EXHAUST
GAS OUT
HOT
EXHAUST
GAS IN
FIG.2III-IO= DECOMPOSITION CHAMBER DESIGN CONCEPT
en
ro
-------
83
it is important to consider methods of A/F ratio control during
engine warm up and during transient load and speed conditions.
Although definition of exact control components is beyond the
scope of this study, control problems and general methods of
solution will be reviewed.
In order to gain enough energy for methanol decomposition,
the total methanol flow must be superheated and routed through
the decomposition chamber. The characteristics of the decom-
position chamber will probably be such that for quasi steady-
state conditions (slowly varying speed and load) the per cent
decomposition will increase with engine speed, i.e., available
exhaust energy. A/F ratio control under these conditions would
be relatively simple with perhaps some fuel temperature
compensation to account for density effects in the superheated
methanol.
Fuel demands resulting from transient increases in the
load or RPM can be met by by-passing superheated or liquid
methanol around the decomposition chamber. Transient demands
probably could not be met by the decomposition chamber since a
probable time lag of 5-10 seconds through the decomposition
chamber would not be acceptable. If superheated methanol is
used in conjunction with a base flow of 30 per cent dissociated
methanol then changes in A/F ratio resulting from variations in
fuel mixture gas properties should not be significant. Use of
liquid methanol would require gaseous and liquid fuel metering
elements with some type of sensing and control elements. These
factors tend to favor a single phase system; however, during
warm up vaporized methanol would not be available unless an
auxiliary vaporization system is provided. Consequently, design
conditions necessary to satisfy both transient and warm up
operation may favor liquid methanol augmentation.
It is probable that the decomposition chamber start up
transient will last at least five minutes. However, it is
estimated that exhaust heated,vaporized methanol would be
-------
available after one minute. During this first minute of
operation either a liquid metering system must be used or the
methanol must be vaporized by an electric heater, a small
combustor, by chemical reaction or by other suitable means.
The highest start-up fuel consumption should not exceed 2 Ib/min
with a required heat energy input of 600 BTU/lb or 1200 BTU/min.
This rate of energy expenditure would require a 21 kw electrical
heater (100 per cent conversion efficiency) or combustion of
0.25 lb (142 cc) of methanol in a combustion heater (50 per cent
energy conversion efficiency). Chemical reaction of methanol
with a reactive chemical would probably require at least an
order of magnitude more total methanol than a combustor and
would present greater design, maintenance and servicing problems
than a combustor. A final approach would be compression and
atomization (pseudo-vaporization) of the methanol to the extent
that it could be metered in the same fashion as fully vaporized
methanol.
Major components for a dual liquid and gas fuel metering
system are shown in Fig. VIII-11. Fuel is pumped to a flow
divider which meters fuel to decomposition and liquid fuel flow
paths. Metering in the flow divider is controlled by a small
A/F ratio control computer which determines appropriate flow
divider valve settings by temperature and pressure signals from
sensor locations in the air intake and fuel flow lines, and by
throttle position and rate of movement. During start up the
primary fuel flow is through the liquid line. A start up
heater is used to prevent cold start problems (the methanol
need not be completely vaporized). While the decomposition
chamber is warming up, superheated methanol is by-passed into
the gaseous fuel metering leg. As the decomposition chamber is
heated and residual methanol from the previous operation cycle
is vaporized, superheated methanol is gradually emitted into
the chamber until steady-state conditions are reached and the
by-pass vale is closed. Transient acceleration fuel demands
-------
AIR
A/F RATIO
CONTROLLER
CHECK
VALVE
BY-PASS
VALVE
DECOMPOSITION
CHAMBER
i
AIR
INTAKE
THROTLLE
FLOW
I SENSOR
1
FLOW
DIVIDER
t
1AAJUUU
FLOW
SENSOR
fQQ001
GASEOUS FUEL
METERING
LIQUID FUEL
METERING
COLD WEATHER
START-UP
HEATER
ENGINE
EXHAUST
EVAPORATOR 8
SUPERHEATER
FIG. ME-II : LIQUID-GAS FUEL METERING SYSTEM
oo
en
-------
86
are met during warm up and steady-state operation by proper
increases in the liquid fuel flow rate. A schematic showing
the arrangement of components in a gaseous fuel system is
shown in Fig. VIII-12. Operation of the system is similar to
that of the dual system except that an auxiliary fuel
vaporizer is used during the initial engine warm up. Transient
acceleration fuel demands are met with superheated methanol
which by-passes the decomposition chamber.
-------
FLOW
DIVIDER
DECOMPOSITION
CHAMBER
FLOW
SENSOR
START-UP
VAPORIZER
BY-PASS
VAPORIZER
LJ
r
FLOW
CONTROLLER
J
uwuu
EVAPORATOR AND
SUPERHEATER
I AIR
NTAKE
GASEOUS FUEL
METERING
ENGINE
FIG. 3Zm-l2: GASEOUS FUEL METERING SYSTEM
CO
-J
-------
88
REFERENCES
[1] "Air Pollution Control-in California—1970 Annual Report",
State of California Resources Agency, Air Resources Board.
[2] Fitch, R. E., Kilgroe, J. D., "Investigation of a
Substitute Fuel to Control Automotive Air Pollution—
Final Report", prepared for the Division of Motor Vehicle
Research and Development, National Air Pollution Control
Administration, Contract No. CPA 22-69-70, CETEC Corp.,
Mountain View, Calif., February 1970.
[3] American Standards Association, "Allowable Concentrations
of Formaldehyde", New York, American Standards Association,
1944.
[4] Malmberg, E. W., J. Am. Chem., Sec. 76, 980 (1954).
[5] Walker, J. F., Formaldehyde, 2nd Ed., ACS Monograph 120,
Reinhold, New York, 1953.
[6] Satterfield, C. W., et al., Anal. Chem., 26, 1792 (1954).
[7] Lappin, G. L., and Clark, L. D., Anal. Chem., 23, 541 (1951).
[8] Bolt, J. A., "A Survey of Alcohol as a Motor Fuel", ASME
Conference Paper, SP 254.
[8a] Starkman, E. S., Newhall, H. K., and Sutton, R. D. ,
"Comparative Performance of Alcohol and Hydrocarbon Fuels",
ASME Conference Paper, SP 254.
[9] Fisher, F. , and Tropsch, Chm. Ber. , 68A, 169 (1935).
[10] Oberdorfer, P. E., "Determination of Aldehydes in Automobile
Exhaust Gas", SAE Paper 670123, 1967.
[11] Obert, E. F., Internal Combustion Engines, 3rd Ed.,
International Textbook Co., Scranton, Pa.
[12] Fenske, M. R., and Frolich, P. K., Ind. Eng. Chem., 21,
1052 (1929).
[13] Huffman, J. R., and Dodge, B. F., Ind. Eng. Chem., 21, 1056
(1929)
[14] "C79 Methanol Synthesis Catalyst Summary of Laboratory
Studies and Commercial Experience", Catalysts and Chemicals,
Inc., Louisville, Ky., June 1970.
[15] Emmett, P. H., Catalysis, Vol. Ill - Hydrpgeneratiori and
Denydrogeneration, Reinhold, New York, 1955.
-------
89
[16] "Catalytic Production of Mixtures of Carbon Dioxide and
Hydrogen from Aqueous Methanol", U. S. Patent No. 3,393,979,
July 1968.
[17] "Low Pressure Catalyst Operating Instructions", Catalysts
and Chemicals, Inc., Louisville, Ky.
-------
APPENDICES
-------
90
APPENDIX A
CALCULATION METHODS
The mass flow rate of air to the engine is calculated
from:
where
air
m.
d
Cd
Ap
= d C, / 2 gc Ap p
mass flow rate of air
diameter of nozzle
coefficient of flow
pressure difference across nozzle
density of air
A Fisher Porter Tri Flat Meter (FP-±-12-G-5) was used to
o
measure the fuel flow rate.
m,- = CB
Formulas used are
1 Ibm
and
where
» '£
- Pmeth^Pmeth
N
A
y :
pmeth:
B =
C
m,. :
viscous influence number
size factor (Table 7)*
methanol viscosity, centipoise
float density
methanol density, g/cc
size factor #2(table D*
flow coefficient (table 104 - function of N)*
flow rate, Ibm/min
The A/F was obtained by dividing m . by ITU.
The brake horsepower is:
nun w n 11,1
BHP = -x 1.341
Watt
F X N
Fisher g Porter Co., Tri-Flat Variable Area of Flow Meter
Handbook (#10A9010).
-------
91
where
w = synchronous motor output to wattmeter, watts
e = synchronous motor efficiency
F = dynamometer force, Ibf
N = speed of engine, rpm
The indicated horsepower is:
IHP = BHP + FHP
The indicated specific fuel consumption is:
m
ISFC =
The sensitivity of the chromatograph to different compounds
was established by injection of standards and measurement of peak
areas as a function of sample size. The calculations for
individual compounds are described here.
Methane: Pure CHU, in the loop at 105°C, gives an average
7
peak of 1.62 x 10 units so that,
. .units, in .CH^peak
fraction of CHU in exhaust = =—
H 1.62 x 10'
, g units in CH^peak
311 ppm CHU in exhaust = (10 ) B—
1.62 x 10'
To convert to g carbon/g exhaust, using the ideal gas equation,
gCH g
PV = ,c RT = T-£ RT for methane and carbon (V = Vpu or
±b Lt. ^"n
Vcarbon • respectively).
The molecular weight of the exhaust averaged 27 during these runs,
and
PV = gexhaust RT f
1» O*7 *\ J. iWl.
Dividing,
Carbon 12(vo1 CH4)
Exhaust 2rCVorexhaustT
= 0.444 (fraction of CH^ in exhaust)
= 0.444 (ppm x 10"6)
mgcarbon = 0.444 x lO'3 (ppm)
^exhaust
-------
92
Methanol: Liquid samples of methanol ranging from microliters
down to 0.2 microliters (injected as 2 microliters of a 10%
7
solution of methanol in acetone) averaged 1.04 x 10 units per
microliter. As the density of methanol is 0.79 g/ml, this represents
1.3 x 10 units per microgram. One microgram is 1/32 x 10~ moles,
or 31 nanomoles. Assuming methanol vapor to follow ideal gas
behavior and occupy 22.4 liters at 1 atm and 0°C, 1 mole will
h
occupy 3.18 x 10 ml at 115°C, and 31 nanomoles will occupy 31
(3.2 x 10 ) ml. The total sample volume is 2.9 ml, so
ppm = (31) (3.2 x 1Q-5)(106)( units N
2'9 1.3 x 104
_9
ppm = (2.6 x 10 x units)
Hydrocarbons: For methane, a 1.0 ml gaseous sample at 1 atm
and 20°C gives a peak of area 1.58 x 10 units. Assuming an ideal
-4 -4
gas, this 1 ml sample contains 6.68 x 10 g CH^, or 5.0 x 10 g
carbon.
Sensitivity = ^ x ^ units = 3.4 x 1010 units/g C
5.0 x 10 g carbon
For 3-methyl hexane, 1 microliter = 0.686 mg C-H,g, and the
peak area from 1 microliter is 2.6 x 10 units. 0.686 mg C-H, K
84 1
contains (0.686) .. ' 2 mg carbon = 0.573 mg carbon.
Sensitivity = 2'B x 10? ^its = 4.5 x 1010 units/g C
0.573 x 10 g carbon
Similar calculations for other hydrocarbons give an average of
4 x 10 units/g C. This figure was used for all the hydrocarbons
in runs 185 and 188, and for ethane, propane, and the unknown peak
in the methanol-fueled runs. This gives grams of carbon, and the
calculation of grams of exhaust is given above, under methanol.
Formaldehyde; Sensitivity was 1.02 x 10 units per microliter.
A calculation exactly the same as given above for methanol except
for the figure 1.02 x 10 gives the concentration in ppm.
Acetaldehyde: Sensitivity was 8 x 10 units per microliter.
Again, the usual calculation gives ppm. Note that the sensitivity
of the instrument for aldehydes, and especially formaldehyde, is
much less than for hydrocarbons.
-------
APPENDIX B
TABLE 1
'est
4
5
6
7
8
10
11
12
13
14
15
21
22
23
24
25
26
27
29
30
31
32
33
A/F
13. 7
4.9
5.5
6.5
7.5
8.9
5.0
5.4
6.5
7.5
9.0
5.5
6.5
5.5
6.5
7.5
9.0
5.0
6.4
7.6
9.0
6.4
5.5
A/F
CR
% Diss
SA
T.
im
ihp
isfc
Tr
CR
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
9
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
.2
-
-
T . rn
Diss SA im ihp isfc 2
(%)
0 12 100 3.91 .474 12.07 0
0 14 70 3.72 1.24 6.60 0
0 14 70 3.73 1.11 9.26 0
0 14 70 3.65 .98 11.17 0
0 14 70 3.22 .965 9.45 3
0 20 70 2.81 .965 8.05 6
0 14 70 3.81 1.21 6.56 0
0 14 71 3.78 1.14 8.79 0
0 14 73 3.79 .97 11.20 0
0 14 68 3.37 .94 9.44 3
0 20 71 2.75 .98 7.82 6
30 14 70 3.64 1.10 9.50 0
30 14 70 3.58 .99 11.22 0
30 14 69 3.64 1.15 8.87 0
30 14 70 3.58 .99 11.20 0
30 14 71 3.31 .94 9.96 2
30 20 69 2.99 .91 8.57 5
30 14 70 3.57 1.44 6.66 0
70 9 70 3.38 .98 11.00 0
70 9 70 3.11 .94 9.62 3
70 12 78 2.81 .90 7.89 6
70 9 70 3.45 1.02 11.02 0
70 6 78 3.49 1.13 9.13 0
air- fuel ratio
compression ratio
per cent of methanol in dissociated
spark advance
intake manifold temperature
indicated horsepower
indicated specific fuel consumption,
indicates no data were recorded for
indicates only trace elements of the
°2
TTT
.14
.03
.05
.60
.71
. 31
.03
.03
.41
. 74
.67
.06
.50
.21
.49
.78
. 31
.14
.90
.42
.55
.23
.07
form
CO
0.81
—
2.90
0.08
0.08
0.04
—
3.52
0.17
0.09
0.09
2.59
0.08
3. 30
0.12
0.08
0.08
—
0.08
0.07
0.08
0.53
3.06
Ib/hp-hr
this point
compound
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
NOX
(%)
.410
.060
.113
.391
.117
.007
.036
.078
. 327
.142
.012
.107
.471
.072
. 390
.357
.111
.030
—
—
.207
—
.128
CHgOH
(ppm)
—
565
441
107
138
424
736
655
167
140
394
222
—
221
72
62
155
131
23
43
77
71
201
less than .5
(ppm)
—
292
174
20
13
13
184
150
17
8
94
525
—
435
32
32
13
236
44
53
51
88
586
ppm
C0H_+U
Z b
(ppm)
— —
10
12
5
3
2
8
10
4
4
2
10
—
8
8
5
4
2
4
2
1
8
6
C3H6 HCOH
.ppm)(ppm) (ppm)
1
2
Tr
1
1
Tr
Tr
Tr
1
Tr
Tr
1
Tr
Tr
Tr
51
79
Tr
65
18
33
57
25
21
13
75
44
13
5
3
12
19
9
6
57
55
to
CO
-------
Test
34
35
36
37
38
39
40
41
42
43
44
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
A/F
5.1
9.0
7.4
6.5
6.0
4.9
5.0
5.5
6.5
7.5
9.0
6.5
6.5
7.5
7.5
5.0
5.5
6.5
6.5
6.2
7.5
7.5
7.5
9.0
5.0
5.4
5.5
5.5
5.5
5.5
8.9
8.9
8.9
8.9
8.9
CR
9.2
9.2
9.2
9.2
9.2
9.2
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
10.9
10.9
10.9
10.9
10.9
10.9
10.9
10.9
10.9
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
8.5
T?
Diss
70
100
100
100
100
100
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
30
30
30
30
30
30
30
30
30
30
30
SA
6
12
12.
12
12
12
15
15
14
14
15
12
12
20
20
12
12
12
12
12
20
20
20
20
8
2
10
12
20
23
23
20
14
10
2
T.
im
65
105
108
107
108
105
69
68
58
69
67
109
44
44
109
70
69
70
110
43
42
110
71
76
70
73
72
71
71
72
80
77
70
67
68
ihp
3.43
2.68
2.93
3.06
3.03
2.66
3.96
3.97
3. 80
3. 36
2. 80
3.58
3.93
3.58
3. 31
4.08
4.13
4.05
3. 87
4.15
3.69
3.48
3. 34
2.99
3. 73
3. 80
3. 73
3. 77
3.61
3.57
2. 72
2.69
2.64
2.67
2.47
isfc
1.22
.91
.97
1.04
1.21
1.43
1.18
1.06
.95
.95
.98
.98
.95
.93
.94
1.17
1.03
.91
.93
.95
.90
.90
.95
.91
1.16
1.07
1.09
1.07
1.12
1.14
.965
.975
.995
.98
1.06
co2
(%)
6.67
8.53
10.38
10.67
8.51
5.00
6.60
9.65
10. 75
9.25
7. 71
10.88
10. 80
9.55
9. 73
6.50
9.11
11.11
11.20
10.90
9.03
9.43
8. 86
7.52
9.00
9. 87
9.97
10.06
9.79
9. 74
7. 86
7. 79
7. 73
7. 85
7. 79
°2
Try
0.05
.5.38
2.06
0.46
0.05
0.06
0.03
0.06
1. 35
4.09
6. 88
1.12
1.22
3.54
3.21
0.06
0.07
0.64
0.48
1.02
4.41
3.69
4. 78
7.22
0.07
0.07
0.09
0.09
0.11
0.11
6.59
6. 73
6. 84
6.63
6. 73
CO
~T%7
—
0.08
0.06
0. 81
—
—
—
2.40
0.08
0.08
0.09
0.08
0.12
0.08
0.08
—
3.09
0.12
0.12
0.12
0.13
0.13
0.08
0.08
3.22
2.12
1.97
1.85
2.18
2.26
0.09
0.08
0.08
0.08
0.08
NOX
(%)
0.045
—
—
—
0.116
0.011
0.015
0.097
—
0.096
0.009
—
—
0.178
0.237
0.009
0.066
— —
— —
—
0.185
—
0.165
0.009
0.077
0.142
0.173
0.187
0.167
0.166
0.089
0.048
0.028
0.023
0.014
CH3OH
(ppm)
176
39
77
2
5
1
646
319
107
158
440
102
199
246
195
810
500
162
228
425
413
376
—
—
252
219
227
207
249
223
370
334
334
356
286
CHU (
(ppm)
660
45
55
78
808
950
746
498
2
9
8
36
46
38
33
1420
675
42
51
47
33
50
28
26
673
483
476
431
654
548
50
31
25
31
23
:2H6+u
(ppm)
12
Tr
6
5
22
4
4
3
7
1
3
10
10
6
6
Tr
5
10
10
10
5
6
7
4
11
11
16
16
17
16
6
6
5
7
8
C3H6
HCOH CH3CHO
(ppm) (ppm)
Tr
—
— —
Tr
1
—
—
Tr
Tr
—
—
1
1
Tr
Tr
—
—
Tr
1
1
Tr
Tr
—
—
1
Tr
Tr
Tr
1
1
Tr
Tr
Tr
Tr
Tr
74
7
8
89
133
20
33
57
22
90
103
55
87
101
109
47
38
46
54
114
122
164
135
89
77
62
71
77
64
61
72
103
218
74
79
(ppm)
—
__
1
__
_ _
—
Tr
8
12
29
Tr
3
6
8
42
6
21
Tr
21
21
10
41
41
10
15
12
8
17
12
15
6
17
4
15
13
CO
-p-
-------
Test
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
A/F
7.5
7.5
7.5
7.4
7.4
7.4
7.5
9.1
9.5
9.5
9.5
5.5
5.5
5.5
5.5
9.0
9.0
9.0
9.0
5.5
5.5
5.5
5.5
9.0
9.0
9.0
9.0
9.0
5.5
5.5
5.5
5.5
9.0
9.0
9.0
CR
8.5
8.5
8.5
8. 5
8.5
8.5
8.5
10.9
10.9
10.9
10.9
8.5
9.1
10.0
10.9
10.9
10.0
9.2
8.5
8.5
9.2
10.0
10.9
10.9
10.0
9.2
8.5
8.5
8.5
9.2
10.0
10.9
10.9
10.0
9.2
o
Diss
30
30
30
30
30
30
30
30
30
30
30
0
0
0
0
0
0
0
0
30
30
30
30
30
30
30
30
30
70
70
70
70
70
70
70
SA
2
10
14
20
23
14
14
18
10
20
23
15
15
12
16
20
20
20
20
12
12
12
12
16
18
16
16
16
5
6
6
4
12
12
12
T.
im
70
70
70
70
70
110
44
71
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
70
72
70
70
70
76
75
76
ihp
3.01
3.14
3.25
3.22
3.11
3.22
3. 39
2. 86
2.69
2. 84
2. 80
3.66
3. 70
3. 77
3. 81
2. 86
2. 77
2. 73
2.69
3.49
3.56
3.60
3.62
2.94
2.90
2. 80
2.69
2.60
3. 35
3.39
3.43
3.49
2.80
2. 77
2.69
isfc
1.01
.96
.93
.94
.97
.94
. 89
.895
.95
.90
.91
1.12
1.11
1.09
1.08
.945
.975
.99
1.01
1.15
1.13
1.12
1.11
.895
.905
.94
.98
.96
1.09
1.09
1.07
1.06
.87
. 88
.905
CO 2
(%)
9.02
9.02
9.18
9.22
9.19
9.60
9.19
7.61
7.28
7.41
7.43
9. 71
9.74
9.72
9.60
7. 86
7.79
7. 76
7. 88
9. 73
9. 84
9.57
9. 70
7.98
8.08
7.98
7. 78
7. 37
10.73
10.71
10.62
10.63
7. 79
7. 70
7.75
°2
~T¥7
4.49
4.49
4.20
4.14
4.19
3.44
4.19
7.06
7.65
7.41
7. 39
0.03
0.03
0.07
0.11
6.60
6. 72
6. 78
6.55
0.13
0.15
0.17
0.19
6. 37
6.21
6. 37
6. 75
7.49
0.23
0.25
0.25
0.29
6.66
6.83
6. 80
CO
(%)
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
2. 34
2. 30
2. 30
2.43
0.09
0.09
0.08
0.09
2.25
2.10
2.43
2.24
0.08
0.08
0.08
0.09
0.08
0.90
0.91
1.03
0.99
0.13
0.13
0.08
NOX
(%)
0.101
0.206
—
—
—
—
—
0.044
0.012
0.030
0.035
0.104
0.109
0.106
0.106
0.013
0.009
0.007
0.007
0.142
0.152
0.127
0.146
0.057
0.057
0.043
0.024
0.010
0. 336
0. 352
0. 342
0.346
0.122
0.104
0.094
CHgOH
(ppm)
101
130
162
106
—
114
118
365
456
398
416
410
439
467
690
713
780
785
585
321
240
206
311
287
311
304
510
479
856
88
134
102
117
121
86
CH4 (
(ppm)
9
18
35
40
40
30
40
35
38
50
41
496
629
520
809
45
37
34
30
563
480
685
613
26
31
31
32
69
241
280
333
299
50
44
44
:2H6+u
(ppm)
4
8
10
12
11
11
13
5
6
6
6
7
20
20
21
7
5
7
6
12
18
17
21
6
7
7
6
6
17
20
25
25
10
10
8
C3H6
(ppm)
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
Tr
1
1
1
2
Tr
—
Tr
—
Tr
1
Tr
1
—
—
Tr
Tr
—
Tr
Tr
1
1
—
Tr
Tr
HCOH
(ppm)
68
47
60
62
61
41
67
87
61
39
61
55
65
78
73
55
45
30
33
63
53
68
71
75
86
76
88
35
49
48
83
73
21
26
15
CH3CHO
(ppm)
10
36
128
382
487
335
324
6
—
—
—
3
41
30
20
20
41
10
—
6
21
13
21
17
6
6
2
Tr
4
2
14
8
2
2
2
to
en
-------
Test
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
A/F
9.0
9.0
9.0
9.0
9.0
9.0
9.6
5.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
6.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
5.5
6.5
6.5
6.5
6.5
6.5
CR
8.5
8.5
9.2
10.0
10.9
10.0
10.0
9.2
8.5
9.2
10.0
10.9
10.9
10.0
9.2
8.5
8.5
9.2
10.0
10.9
10.9
8.5
8.5
10.9
10.9
8.5
8.5
10.9
10.9
8.5
8.5
10.9
10.9
8.5
8.5
o
Diss
70
100
100
100
100
70
70
0
0
0
0
0
30
30
30
30
70
70
70
70
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
SA
12
10
10
10
10
12
6
15
16
16
16
16
10
10
10
10
7
7
7
7
2
2
5
5
10
10
20
20
23
23
23
23
23
20
10
T.
im
76
102
105
107
106
75
75
70
108
110
108
110
110
110
110
109
110
110
110
109
70
70
70
70
68
73
70
70
70
70
70
70
70
70
70
ihp
2.64
2.64
2.68
2. 71
2. 72
2.81
2. 77
3.68
3.41
3.45
3.50
3.56
3.45
3.41
3. 36
3. 31
3.17
3.22
3.29
3.31
3.63
3.50
3.58
3.73
3. 74
3.59
3.52
3.66
3.55
3.44
3.33
3.49
3.55
3.42
3.47
isf c
.92
.93
.915
. 895
.885
.905
.86
1.11
1.015
1.005
.99
.975
.96
.97
.99
1.00
1.015
1.00
.98
.97
1.09
1.13
1.10
1.06
1.08
1.12
1.145
1.10
1.14
1.17
1.04
.995
.98
1.01
1.00
co2
(% )
7. 70
8.26
8.21
8.27
8.21
8.03
7.49
9.76
10.95
10.86
10.95
11.02
11.03
11.03
10. 84
10.98
10.98
10.96
10.92
10.92
10. 38
10.53
10.49
10. 37
9. 78
9.69
9.65
9.64
9. 70
9.64
11.00
10.97
10.93
10.99
10.93
°2
(%)
6. 89
6.01
5.97
5. 85
5. 89
6. 30
6. 79
0.14
1.00
1.15
1.00
0. 88
0. 84
0. 84
1.19
0.94
0.77
0. 86
0.92
0. 73
0.04
0.03
0.03
0.05
0.05
0.03
0.03
0.05
0.07
0.06
0.91
0.95
1.03
0.92
1.03
CO
( %)
0.08
0.00
0.08
0.09
0.13
0.08
0.43
2.21
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.21
0.17
0.17
0. 31
1.48
1. 30
1. 36
1.49
2.24
2. 37
2.42
2.42
2. 33
2.42
0.09
0.09
0.09
0.08
0.08
NOX
( %)
0.082
0. 315
0. 337
0. 369
0. 386
0. 351
0.130
0.074
0.254
0.257
0.273
0.284
0. 329
0. 325
0.302
0.294
0. 385
0. 395
0.405
0.405
0.152
0.146
0.148
0.149
0.116
0.100
0.105
0.111
0.120
0.111
0. 358
0.406
0.404
0. 364
0. 322
CH3OH
(ppm)
96
13
2
1
1
387
—
508
—
—
—
—
—
85
77
65
23
30
34
—
—
—
—
—
—
—
—
284
288
318
86
114
115
62
51
CH^ (
(ppm)
36
72
46
38
43
43
—
672
— —
—
—
—
—
57
54
—
84
73
55
116
—
—
—
—
—
—
—
741
716
1210
44
48
50
48
44
:2H6+u
(ppm)
8
7
10
9
10
10
—
14
—
—
—
—
—
37
38
31
27
30
29
26
—
—
—
—
—
—
—
14
17
75
20
17
17
21
15
C3H6
(ppm)
Tr
Tr
Tr
Tr
Tr
Tr
—
1
—
—
—
—
—
1
Tr
1
1
1
1
1
—
—
—
—
—
—
—
Tr
Tr
—
Tr
Tr
Tr
Tr
Tr
HCOH
(ppm)
19
13
16
20
20
31
. —
55
—
—
—
—
—
29
33
29
26
21
29
29
—
—
—
—
—
—
—
35
40
25
47
60
52
52
44
CH3CHO
(ppm)
—
82
303
397
405
74
—
4
—
—
—
—
—
10
52
63
13
72
88
68
—
—
—
—
—
—
—
5
4
6
21
139
139
118
125
CO
CD
-------
Test
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
A/F
6.5
6.5
6.5
6.5
6.5
7.5
7.5
7.5
7.5
7.5
5.7
5.6
5.6
5.7
5.0
6.5
6.5
6.5
6.5
5.5
7.5
7.5
7.5
7.5
7.5
7.5
7.5
5.1
5.0
5.0
6.5
6.5
7.5
7.5
7.5
CR
10.9
10.9
8.5
8.5
10.9
10.9
10.9
10.9
10.9
8.5
8.5
9.2
10.0
10.9
10.9
10.9
10.0
9.2
8.5
9.2
10.9
10.9
10.9
10.9
10.9
10.9
8.5
8.5
8.5
10.9
10.9
8.5
10.9
8.5
10.9
•o
Diss
30
30
30
30
30
30
70
70
70
70
100
100
100
100
100
100
100
100
100
0
30
30
30
30
30
100
100
100
70
70
70
70
70
70
30
SA
10
2
2
10
10
13
2
10
6
6
5
5
5
5
5
6
6
6
6
15
1
10
15
20
23
10
10
5
6
6
6
6
10
10
12
T.
im
70
70
70
46
45
45
70
70
70
70
110
110
110
110
110
110
110
110
110
70
70
70
70
70
70
113
113
111
70
71
71
70
110
110
110
ihp
3.58
3.52
3.38
3.49
3.65
3. 38
3.12
3.17
3.20
3.05
3.10
3.12
3.17
3.18
3.08
3.24
3.18
3.15
3.10
3.70
3.10
3.27
3.31
3.26
3.22
3.03
2.94
3.00
3. 32
3.45
3.41
3.29
3.16
3.01
3.28
isf c
.97
.985
1.03
.995
.95
.915
.955
.94
.93
.975
1.185
1.18
1.16
1.155
1.285
.98
1.00
1.01
1.025
1.14
1.005
.955
.94
.955
.97
.95
.98
1. 345
1.24
1.195
.90
.935
.93
.975
.93
co2
( %)
10.93
10. 82
10. 89
10.69
10.66
9.46
9.40
9.51
9.44
9. 39
8. 83
8. 82
8. 89
8.62
6.55
11.03
10.98
10.98
10.98
8.96
9.54
9.55
9.53
9.53
9.61
10.29
10.24
6.57
6.91
6.93
10. 86
10. 83
10.15
10.07
9.69
°2
(%)
1.04
1.24
1.10
1.47
1.53
3.69
3.80
3.61
3. 73
3.82
0.07
0.07
0.08
0.07
0.05
0.39
0.38
0. 39
0.38
0.03
3.55
3.53
3.56
3.57
3.42
2.19
2.28
0.06
0.02
0.03
1.17
1.22
2.46
2.60
3.28
CO
(%)
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
0.08
3.44
3.46
3.36
—
—
0.41
0.48
0.48
0.47
3.31
0.08
0.08
0.09
0.09
0.09
0.08
0.08
—
—
—
0.08
0.08
0.08
0.08
0.08
NOX
(
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
% )
347
299
277
305
342
244
302
398
369
331
136
128
128
122
074
335
329
320
307
058
142
252
295
357
407
489
471
064
047
051
440
406
442
408
300
CH3OH
(ppm)
83
40
23
39
67
109
—
36
28
23
7
5
8
46
20
8
14
3
2
411
59
114
159
134
133
1
Tr
2
70
143
25
12
50
18
228
CH4 C2H6+U
(ppm)
47
42
36
41
38
38
—
25
28
30
1675
1500
1370
1713
2350
396
304
349
386
754
12
16
24
21
21
28
28
1030
490
957
46
37
45
55
38
(ppm)
17
16
16
17
16
13
—
14
14
16
15
17
36
30
38
23
28
26
24
19
9
10
12
11
11
14
12
25
15
30
20
17
21
17
16
C3H6
HCOH
CH3CHO
(ppm) (ppm) (ppm)
1
1
Tr
Tr
Tr
Tr
—
Tr
Tr
Tr
Tr
1
1
5
6
—
Tr
Tr
—
Tr
Tr
—
Tr
Tr
Tr
Tr
—
2
1
3
1
Tr
Tr
Tr
Tr
44
37
26
25
50
91
—
45
45
46
45
47
65
124
170
33
44
30
22
74
26
47
53
46
36
24
12
81
59
66
17
5
18
17
--
138
127
92
151
221
364
—
395
288
277
20
Tr
6
58
180
55
21
6
2
7
4
55
221
374
232
227
134
18
Tr
Tr
48
013
410
2154
2108
to
-------
Test A/F
CR
"O
Diss
SA
T-
im
ihp
sfc
co2
°2
CO
NOX CH3OH CH4
C2H6+U C3H6 HCOH CH3CHO
(%) (%) (%) (%) (ppm) (ppm) (ppm) (ppm)
176 5.0
177 5.5
178 5.5
179 5.5
180 9.0
181 9.0
182 5.0
183 6.5
184 6.5
185 14.9
Tests 186
186 10.0
187 13.1
188 14.0
189 16.1
190 17.9
191 14.9
10.9
9.2
9.2
8.5
8.5
9.2
10.9
8.5
10.9
9.2
through
9.2
9.2
9.2
9.2
9.2
9.2
30
30
30
30
30
0
0
100
0
0
191
10
16
10
10
16
20
5
14
14
18
were
18
20
18
18
18
18
70
70
70
70
70
70
110
69
71
86
3.65
3. 78
3.61
3.59
2.61
2. 70
2.83
3.65
3. 86
3.90
conducted
80
84
87
90
91
88
3.67
3. 84
3.90
3. 87
3. 72
3.90
1. 205
1.13
1.11
1.115
.955
1.00
1.42
1.01
.955
.461
6. 80
9.29
9. 86
9.91
7. 32
7.65
6.56
11.16
11.16
10. 82
with gasoline ;
.735
.535
.492
.431
.404
.461
5. 71
7.21
9.66
11. 39
12.06
10.63
0.08
0.04
0.04
0.04
7.69
7.09
0.03
0.62
0.62
0.08
all
0.08
0.06
0.08
0. 84
0. 88
0.08
— -
2.88
2.15
2.08
0.00
0.00
—
0.08
0.08
2.51
previous
—
—
—
1.15
0.23
2.76
0.066
0.054
0.106
0.107
0.005
0.004
0.027
0.278
0. 314
0.306
tests
1.080
0.566
0.527
0.350
0.358
0.427
441
341
216
189
299
508
56
were
1191
679
444
401
29
27
--
with
23
13 Tr
12
10
9 Tr
9 Tr
13 Tr
methanol .
(ppm)
75
28
30
18
38
44
41
(ppm)
2
Tr
52
—
53
65
62
to
00
-------
99
APPENDIX C
REFORMING CYCLE
ENERGY BALANCE ANALYSIS
A. Reforming Cycle:
A reforming cycle is one in which exhaust energy from a
heat engine is used to dissociate some or all of the fuel
before injection into the engine. Associated with reforming
are preheating, evaporation and superheating of the fuel.
There are several sequences in which energy can be withdrawn
from the exhaust to accomplish these processes, especially if
the reformed and non-reformed portions of the fuel are treated
separately.
For the reasons described in the body of the report, the
cycle was set up, as shown in Fig. VIII- . This arrangement is
one of counterflow in which the reforming of the fuel is
accomplished by energy extracted from the exhaust first, at
the most elevated temperature. The schedule of heat transfer
events is as shown.
The significant assumptions which were made in setting up
the energy balance are as follows:
(1) Fuel in pure methanol (CHgOH)
(2) A/F = 6.43
(3) The fuel dissociates into the components CO + 2 H?
(4) All mixing processes are perfect
(5) No energy is lost from the system except for
engine cooling
(6) The fuel is completely reacted with the oxygen in
the air so that the exhaust composition (mole
. fraction) is:
C02 - 0.115
H20 - 0.231
N2 ~ °-654
1.000
-------
100
B. Analysis:
The fuel flow rate is specified along with the fractions
of the fuel which pass through the reforming and non-reforming
branches; the temperatures at the steps in both branches; the
nominal ISFC; the engine RPM; and intake (evaporative) pressure.
The calculations pass through the cycle with no iteration or
loops (except for iterations to calculate temperature as a
function of enthalpy). Engine performance, enthalpy levels,
exhaust temperatures are calculated.
(1) Method of Calculating Enthalpy. The enthalpy of a
chemical species is taken to be:
H
i
= AH°
+
i
T
C
o
P
i
dT
where AH?, is heat of formation of the species at the reference
temperature T (298.15°K) and the integral expression is the
sensible enthalpy of the species of the temperature T, above
or below the reference temperature. The enthalpies of all
species except methanol were obtained from the JANAF Thermo-
chemical Tables; the heat of formation of methanol was calculated
its heat of combustion and its specific heat (source unknown)
curve fitted and integrated to produce the sensible enthalpy as
a function of temperature.
Under adiabatic conditions the enthalpy of the reactants
in a process is equal to the enthalpy of the products. The
enthalpies of the products can then be calculated according to
the following relation provided that the product composition is
known:
z
react
T
react
(AH° +
Cp dT)
To
T
prod
= I (AH° +
prod
Cp dT)
To
-------
101
(2) Effects of Intake Variables and Engine Performance
on Exhaust Properties. Little data exists on the performance
and operation of a methanol engine; consequently, the treatment
of the effects of the intake and engine operating variables on
performance and exhaust conditions is highly simplified. This
can be easily modified as more data becomes available. At
present, the engine ISFC variation with inlet mixture enthalpy
is calculated from the expression:
ISFC = ISFC - .00055 (Hc + 461.51) ,
nom 6 '
derived from the experimental data of the previous program,
where ISFCnQm is input, and - 461.51 BTU/lb is the enthalpy of
the nominal intake mixture. The exhaust enthalpy is assumed to
vary by 40% as much as the intake enthalpy. The effect of fuel
flow rate is calculated by an empirical expression which alters
the exhaust temperature from its input nominal value - 1460°R.
C. Computer Program:
The program (Fortran IV) is set up to calculate a broad
matrix of data points of the particular engine and nominal
operating point selected for this study. There is presently
some internal override of the input variables.
(1) Input Variables. The input is contained in two
cards:
Card 1 - 5-10 column fields
RPM Engine speed - revs, per minute
PIN Intake pressure - psia
ISFCN ISFC at nominal operating condition
XR1 Fraction of fuel through reforming branch
XR2 Fraction of XR1 fuel which is dissociated
Card 2 8-10 column fields
TA Air intake temperature
TO Fuel intake temperature
T1R Reform branch temperature after preheating
T3R Reform branch temperature after superheating
T4R Reform branch temperature after reforming
TIM Non-reform temperature after preheating
T2M Non-reform temperature after evaporation
T3M Non-reform temperature after superheating
-------
102
(2) Program Structure. The program is now set up to
function through a succession of DO loops in the following
order:
ISFCN = .888, .915, .938, .963 (Ib/hp-hr)
RPM = 2400, 2800, 3200
T4R = 860, 960, 1060 (°R)
XRL = 0, .3, .7, 1.0
wfuel = 80.8, 85.8, 90.8, 95.8, 100.8 (Ib/hr)
(3) Output: The output consists of 57 variables for
each case: temperatures, flow rates, all input variables except
ISFCN, enthalpies, and heat exchange values. Temperatures are
in °R, enthalpies in BTU/lb, and heat transfer values (according
to the flow diagram) in BTU/hr.
The output format is printed at the top of each page.
------- |