APTD-1567
MARCH 1973
CHARACTERIZATION
OF PARTICIPATES
AND OTHER NON-REGULATED
EMISSIONS
FROM MOBILE SOURCES
AND THE EFFECTS
OF EXHAUST EMISSIONS
CONTROL DEVICES
ON THESE EMISSIONS
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Air and Water Programs
Office of Mobile Source Air Pollution Control
Emission Control Technology Division
Ann Arbor, Michigan 48105
-------
-------
11
The APTD (Air Pollution Technical Data) series of reports
is issued by the Office of Air and Water Programs, U.S.
Environmental Protection Agency, to report technical data
of interest to a limited number of readers. Copies of APTD
reports are available free of charge to Federal employees,
current contractors and grantees, and non-profit organiza-
tions - as supplies permit - from the Air Pollution Technical
Information Center, U.S» Environmental Protection Agency,
Research Triangle Park, North Carolina 27711 or may be
obtained, for a nominal cost, from the National Technical
Information Service, U.S. Department of Commerce, 5285 Port
Royal Road, Springfield, Virginia 22151.
This report was furnished to the U.S. Environmental Protec-
tion Agency by The Dow Chemical Company, Midland, Michigan
48640 in fulfillment of Contract Number EHS-70-101. The
contents of this report are reproduced herein as received
from The Dow Chemical Company, The opinions, findings, and
conclusions expressed are those of the author and not
necessarily those of the Environmental Protection Agency.
Publication Number APTD-1567
-------
Ill
TABLE OF CONTENTS
FOREWORD v
ABSTRACT vi
I. INTRODUCTION 1
II. CONCLUSIONS 3
III. EXPERIMENTAL PROCEDURES 6
A. Particle Generation 6
1. Engine Dynamometer Studies 6
2. Chassis Dynamometer Procedures 9
B. Particle Collection 11
1. Dilution Tube 11
2. Sampling Devices 13
C. Condensate Collection 15
D. Analytical Methods 15
1. Fuels 16
2. Oils 17
3. Diluent Air 17
4. Exhaust Gases 17
5. Oxides of Nitrogen 19
6. Exhaust Particles 21
7. Condensate Analyses 34
IV. EXPERIMENTAL RESULTS 41
A. Task I 44
B. Task II 81
C. Task III 92
D. Task IV 135
E. Task V 189
APPENDIX A 192
-------
FOREWORD
This report was prepared by the Transportation Chemicals Research
Group, Ag-Organics Department, The Dow Chemical Company, Midland,
Michigan, under Contract EHS 70-101. The work reported herein
was administered under the direction of the Office of Air and
Water Programs, Environmental Protection Agency, with Dr. Robert E.
Sampson and Mr. Chas. L0 Gray, Jr. serving as Project Officers.
The report covers work performed from Sept. 1, 1971, to Dec. 31, 1972,
The authors of this report are James E. Gentel, Otto J. Manasry,
and Joseph C. Valenta.
The authors wish to acknowledge the significant contributions
of the following individuals:
S. M. Sharp R. P. Himes
W. B. Tower M. Y. Kelly
J. D. McLean R. E. Mansell
R. B. Nunemaker P. P. North
C. E. Van Hall S. M. Richter
H. H. Gill N. J. Smith
G. E. Stobby S. Stell
H. D. Woodcock M. Griggs
R. Matalon K. Schmeck
S. W. McLean S. Love
J. F. Bartel B. Coleman
Jc T. Dumler M. J. Baldwin
T. A. Killer
-------
VI
ABSTRACT
This report describes work carried out on a research program to
determine the effect of emission control devices on the particulate
emissions of an automotive power plant. The work was divided into
five tasks as follows:
TASK I was the characterization of a particulate trapping system,
and the determination of what effects, if any, were noted as
conditions within the system were controlably varied.
TASK II was the definition of a particulate baseline for a 1972
Pontiac 400 CID engine, using non-leaded and low lead fuel. No
emission control devices were used for the baseline runs.
TASK III was the evaluation of the particulate emissions from
a 1972 Pontiac 400 CID engine equipped with the following control
devices: three different oxidation catalysts, one NO catalyst,
H
and one exhaust gas recirculation system.
TASK IV involved testing automobiles equipped with control devices
for particulate emissions. These vehicles were supplied by both
the contractor and the Government.
TASK V was to define a preliminary collection system for diesel
engine particulate sampling.
In all tasks, particulate mass emission rates were measured, as
well as particle mass size distribution, carbon and hydrogen, trace
metal, and benzo-a-pyrene content of the particulate. Ammonia and
aldehydes were measured in the exhaust gas condensate, and gaseous
emissions were determined as a routine check on engine operating
conditions.
-------
-1-
I. INTRODUCTION
The work presented in this report covers the second half of contract
EHS 70-101. In the previous work, reported in a final report
dated July 1971, some preliminary studies were made on the effect
of emission control devices on unregulated emissions. The extension
to Contract EHS 70-101, which is reported here, involved a more
detailed look at specific control devices and the resulting effects
on particulate size and mass, particulate composition, and condensate
composition. The work was divided into five specific tasks in order
to smoothly carry out the technical goals of the contract.
The major objective of Task I was to study the effect of specific
engine and sampling variables on certain non-regulated emissions
under highly controlled conditions. Studies were made on non-leaded
and low lead fuels.
During the course of some preliminary studies undertaken in the
previous year's work on Contract EHS 70-101, alarming differences
were noted in the mass of particulate emitted when non-leaded fuel
was used, and when the filtering systems and dilution tube were
operated at different flow rates and temperatures. In order to
reach an understanding of the above effects, and to define a
meaningful set of sampling parameters, a study was made of a number
of sampling variables and their effect on the mass of particulate
matter collected at a filter.
Task II involved running a 400 CID Pontiac engine, using non-leaded
and low lead (0.5 cc/gal) fuels, to determine a baseline for
subsequent studies. Operating conditions were varied and included
rich, standard, and lean air/fuel ratios, as well as advanced,
standard, and retarded spark timing. The sampling techniques
settled upon in Task I were used to collect particulate.
-------
-2-
The objective in Task III was to evaluate various emission control
devices with respect to their effect on non-regulated emissions.
The devices tested under this portion of the contract include:
Three oxidation catalysts
One reduction catalyst
One exhaust gas recirculation system
Initially, work was done using both non-leaded and low lead (0.5
cc/gal) fuels. The leaded fuel itself caused increases in the
amount of particulate and, after testing one catalyst on leaded
fuel, the rest were run only on non-leaded fuel. The low lead
fuel poisoned the catalyst sufficiently so that it was felt that
future runs would be more meaningful if only non-leaded fuel was
used.
The devices were obtained from either the manufacturer or an auto-
mobile company under a secrecy agreement, to protect any proprietary
rights involved. Consequently, the data on the devices are reported
with the only reference to the device being a code letter.
Task IV was an evaluation of the particulate emission levels of
vehicles equipped with various control devices. Several of these
vehicles were made available to Dow by the manufacturer, while others
were supplied by the Government for testing. Ten different vehicles
were tested, with 19 runs made on the 10. Each vehicle is discussed
in detail in the Experimental Section of Task IV.
Task V was a limited diesel engine study to establish baseline data
for emissions present in the exhaust. Due to an increased emphasis
on the vehicle studies, this task was only partially completed.
-------
-3-
II. CONCLUSIONS
1. The use of three different oxidation catalysts on an engine
stand, with non-leaded fuel, increased the amounts of particu-
late collected at 60 mph by a factor of 2-5, compared to a
baseline run, except under rich air/fuel ratios. Two catalysts
did not generally increase the particulate collected at 30 mph,
or under cyclic conditions, while an increase was noted with
one catalyst under both conditions. The total particulate
collected from the control devices was less than normally found
when using 3 cc leaded fuel.
2. There was no evidence in the particulate of catalyst degrada-
tion being the cause of the increase in particulate mass.
3. The mass medium equivalent diameter was shifted significantly
toward smaller particles, when compared to the baseline, for
all of the catalysts tested on an engine stand.
4. In almost all cases, 30 mph at steady-state, on an engine with
no control devices, gave higher particulate levels than the
corresponding 60 mph run.
5. An increase in particulate comparing 30 mph to 60 mph was noted
during the baseline runs. This appeared to be reversed when
running at standard conditions with two of the three oxidation
catalysts.
6. The three oxidation catalysts significantly lowered the
emission of aldehydes, as collected in the condensate, as well
as lowering the total hydrocarbons.
-------
-4-
7. The vehicles on which mileage accumulation tests were made
exhibited a large degree of fluctuation with respect to
grams/mile as a function of mileage. No clear trends have
been established.
8. The precision of measuring particulate mass from a vehicle
exhaust is substantially lower than that of measuring an engine
stand run, probably due to differences in operating conditions
prior to the tests.
9. In general, the particulate matter which exhibited higher
percentages of carbon also exhibited higher parts per million
of benzo-a-pyrene.
10. The mass medium equivalent diameter became larger with
mileage for two of the three mileage accumulation cars,
while decreasing for the other.
11. The mass medium equivalent diameter for the device equipped
vehicles in general correlates well with the numbers obtained
during the engine stand runs, even though the overall mass of
particulate changed.
12. In general, the low lead fuel gave higher particulate levels,
in grams/mile, than the non-leaded fuel.
13. Overall, the aldehyde content of the exhaust condensate was
not significantly different between the non-leaded and low
lead fuel.
14. Under rich air to fuel ratio conditions, both the oxidation
catalysts and the reduction catalyst gave a significant
increase in NH~ emissions.
lb. The concentration of benzo-a-pyrene in the particulate
varied widely with engine conditions, but did not appear
to be significantly changed by use of leaded vs. unleaded
fuel.
-------
-5-
16. Air/Fuel ratio had an effect on particulate with the
standard setting generally being lower than either rich or
lean settings.
17. The effect of particulate due to changes in spark timing
was sporadic and, therefore, inconclusive based on this data.
18. Different temperatures at the filter and the sample probe
caused differences in amounts of particulate collected.
19. Changes in flow rate through the dilution tube caused slight
variations in the amounts of particulate collected.
20. Dilution air temperature proved to be important since a lower
temperature showed definite increases in the amount of
particulate collected.
21. Sample probe location appeared to have only very small effects
on particulate samples.
22. Face velocity of a sample stream through a given filter
was important in that more sample/ comparatively, tended to
be collected at lower face velocities.
23. A majority of the sample during a steady-state run was
collected within the first 25 percent of the time period over
which the run was made.
24. Absolute measurements of grams/mile are misleading when
measured during a steady-state run of long duration for the
reason given in 23 above. Comparisons can be made, however,
between runs of like time periods, and are valid as measure-
ment of a trend.
25. The modified Federal cycle cold start gave more particulate
than the 23 minute Federal cycle.
-------
-6-
III. EXPERIMENTAL PROCEDURES
A. PARTICLE GENERATION
1. Engine Dynamometer Studies
The engines used in this study were 1972 Pontiacs, 400 CID.
These engines were mounted on the dynamometer bed plate and
attached to a fully instrumented Eaton Dynamatic dynamometer.
Appropriate control and sensing devices were attached to the
engine. The following procedure (Table 1) was then employed
to run-in the new engines, using Indolene low lead (0.5 cc/gal)
fuel.
TABLE I
NEW ENGINE BREAK-IN PROCEDURE
(28 hours)
1) Warm up engine to 180°F coolant outlet temperature at
1000 rpm, no load. Set spark advance and best idle
according to manufacturer's specifications.
2) Run one hour at 1500 rpm, no load, automatic spark
advance and fuel flow. Shut down, retorque cylinder
heads, drain and change lubricating oil.
3) Run Cycle 1
RPM Man. Vac. (In. Hg) Time (Hr)
1500 15.0 1.0
2000 14.0 1.0
2400 14.0 1.0
2600 14.0 1.0
2000 11.0 1.0
5.0
-------
4) Run Cycle 2
-7-
RPM
1500
2000
2500
3000
2000
Man. Vac. (In. Hg)
7.0
7.0
7.0
7.0
7.0
Time (Hr)
0.2
0.6
1.0
1.0
0.2
3.0
5) Repeat Cycle 2.
6) Run Cycle 3
RPM
2000
2500
3000
3500
2800
Man. Vac. (In. Hg)
WOT*
WOT
WOT
WOT
WOT
Time (Hr)
1.0
1.0
1.0
0.5
0.5
4.0x4 cycles
=16 hours
*WOT - wide open throttle
7) While engine is hot, run motoring compression and conduct
leak-down check.
The engine was removed from the dynamometer, drained, partially
dismantled, cleaned, reassembled, and placed back on the
dynamometer stand. A manufacturers original standard vehicle
exhaust system for the specific test engine was attached to
one bank of cylinders. The other bank of cylinders was attached
j
to the dynamometer cell exhaust system. Suitable engine monitors
-------
R
were attached to the engine in order to provide continuous
monitoring of oil pressure and temperature, coolant temperature,
carburetor air flow rate (using a Meriam Laminar Flow Element
50MC-2-45F) and temperature, etc.
The engine was then run for 75 hours using the following
"conditioning" sequence (Table 2) employing the specific test
fuel designated for that run. This sequence of testing was
used for the initial break-in of the- engine, as well as for
certain emission tests. It was not run prior to evaluation
of each condidate emission control device. During the
conditioning sequence, total unburned hydrocarbons, oxygen,
nitrogen, carbon monoxide, carbon dioxide, and oxides of
nitrogen were measured at frequent intervals by FID, gas
chromatography, chemical absorption, and a Scott N0/N0_ analyzer,
respectively. Air/fuel ratios were also calculated based upon
exhaust gas composition.
TABLE 2
TEST ENGINE CONDITIONING SEQUENCE
Cycle RPM Time (Min.) Vac. (In. Hg) Decay
1 800 2 18.8
2 1070 13 16.4 1/2 min.
3 1615 20 17.2 1/2 min.
4 2125 13 14.3 1/2 min.
5 1070 12 16.4 1/2 min.
Sequence repeats after each five cycles.
Following the conditioning sequence, the engine exhaust system
was attached to the dilution tube inlet pipe and the system
was ready for experimental particulate sampling. All subsequent
runs were 60 mph or 30 mph 2-hour steady-state runs.
-------
Q
2. Chassis Dynamometer Procedures
A Clayton CT-200-0 chassis dynamometer with a variable inertia
flywheel assembly was used in all tests conducted under this
program. A Chelsa direct-drive Model PLDUP-300A fan was
located in front of the test vehicle, and operated at 1750 rpm
providing 5,000 scfm air flow. In these tests, the vehicle was
operated under approximately 60 mph road-load cruise conditions
(2250 rpm - 17" Hg manifold vacuum) and under cyclic conditions
of the Federal Test Procedure (old California cycle) and LA-4
procedure driven by a vehicle operator following the cycle on a
strip-chart recorder driver aid.
Table 3 indicates specific procedures employed to prepare
each vehicle for test run.
TABLE 3
VEHICLE TEST PROCEDURE - CHASSIS DYNAMOMETER
1) General Vehicle Inspection
Exhaust System;
a) Inspect for holes or cracks, dents, and collapse
b) Inspect for leaking joints
Engine, check
a) All fluid levels
b) All coolant hoses
c) Air pump fan, power steering, and belts
d) Chec1. heat riser (if applicable) for fullness of operation
e) Check automatic choke operation and adjustment, where
possible
2) Engine Analysis and Tune-up
Scope Check
a) Start engine and allow to warm up for at least 15 minutes
b) With engine running at fast idle, check
Spark plugs
Spark plug wires
Distributor cap and rotor
Coil output
Points
-------
-10-
TABLE 3 (continued)
c) With engine running at idle, check
Dwell
Timing
d) With engine running at 1500 and 2400 rpm, check
timing device
e) Carburetor Adjustment
Tighten intake manifold and carburetor
Install new air cleaner element
With engine running at specified idle
speed, adjust air to fuel ratio to
specifications
Make final adjustment on idle speed
f) Recheck all scope patterns for normal appearance
3) Instrumentation and Equipment Installation
Vacuum and RPM monitors
a~)attach tachometer to ignition coil
b) connect "U" tube monometer to intake manifold
c) install throttle cable (if running under cruise mode)
Wheels
a) remove rear wheels
b) Install test tires and wheel assemblies
4) Procedure for Cold, Hot Starts, and Engine Temperature
Stabilization
Cold Start
a) Place vehicle on the dynamometer rolls, set inertia
weights for specific vehicle, and go through the
preparation for test as well as the tune-up procedure.
b) Allow at least a 12-hour soak period, but not more than
16 hours.
c) Connect vehicle tailpipe to dilution tube.
d) Start the vehicle and proceed with the individual test.
Hot Start
Continuation of the cold start only after the engine
temperature has stabilized.
-------
-11-
TABLE 3 (continued)
Engine Temperature Stabilization
Two-hour steady state runs were made only after a cold
start and one or more hot starts. Thus, the engine
and particle collecting system were always at operation
temperature before the steady state sampling was begun.
When preparation has been completed, the vehicle is placed
in gear and the speed is increased to 2250 rpm with the
intake manifold vacuum is set at 17.0" Hg by controlling
the amount of load imposed on the drive wheels. At the
time when the load and the speed become stabilized, the
tailpipe is connected to the dilution tube inlet pipe
and particulate collection is started when dilution tube
has come to equilibrium.
The procedures outlined in Table 3 were used whenever possible.
On certain vehicle tests where the vehicle was equipped with
proprietary systems, only visual checks were made of the
components and engine hardware. In some cases, the vehicles
were adjusted by personnel from the organization submitting
the vehicle for testing.
B. PARTICLE COLLECTION
Exhaust particles were collected after air dilution of the exhaust
in the large dilution tube described below. During the engine
stand studies, one-half of the engine exhaust was fed into the tube
while the other half was exhausted through the dynamometer cell
exhaust system. With vehicles, the entire exhaust stream was diluted.
Using EGR, the tests were conducted using full exhaust.
1. Dilution Tube
Air dilution and cooling of the exhaust was accomplished by
a dilution tube 16 inches in diameter and 27 feet in length
constructed of extruded polyvinyl chloride (PVC) pipe in
several sections with butt joints which were taped during
assembly prior to each run (Figure 1). The diluent air coming
into the tube is filtered by means of a Dri-Pak Series 1100
-------
Figure 2
Flow Diogrom tor Engine Exhoust
Particulate Collection
Air
out
Instrument
Filter
I
Engine Room RJ
< ^
Mix i ng
Air
in.
Eng ine
Dynamometer
Eng ine*
Flow >
Control
Gravimetric Fallout
Sampling Slits
'Tail Pipe
Standard Muffler
'5*
t
Exhaust P ipe
Scott Research ins,
NO and N02
Analysis
Fisher Gas Partitioner
CO, C02, N2, 02
Beckman 109A
Total Hydrocarbon
Analyzer
--»!
Anderson
Separator
M i11ipore
F ilter
Flow Meter >
Vacuum
Pump
A ir
Pump
to
i
Manometer
-------
-13-
Class II PIN 114-110-020 untreated cotton filter assembly.
This filter assembly is 24" x 24" and has 36 filter socks
which extend to 36 inches in length. This filter will pass
particles 0.3p in size and smaller. Pressure drop at 600 cfm
flow rate is minimal.
Exhaust was delivered to the tube via a tailpipe extension
which was brought into the bottom of the tube downstream of
the filter assembly. The extension was bent 90 degrees inside
the tube, thus allowing the introduction of the exhaust stream
parallel to the tube axis. Within the dilution tube, along
the perpendicular plane of the end of the exhaust extension
was a mixing baffle which has an 8-inch center hole and was
attached to the inside diameter of the tube. The baffle presented
a restriction to the incoming dilution air in the same plane
as the end of the exhaust extension and performed three essential
functions.
a) Provided a turbulent mixing zone of exhaust gas and
dilution air.
b) Eliminated engine exhaust pulsations in the tube.
c) Caused the tube to perform as a constant volume device
over a wide range of engine exhaust output volumes.
2. Sampling Devices
The particulate sampling zone for particles smaller than
15y is located at the exhaust end of the dilution tube. Four
isokinetic sample probe elbows are located in the exhaust-air
i
stream. One probe is connected}to an Andersen Impact Sampler
Model 0203, a filter assembly, and a vacuum pump, in that
sequence. The probes are 0.754
inch ID stainless steel tubes
which are located as shown in Fi.gure 1. A mercury manometer
is connected between the dilutipn tube probe and the exhaust
-------
-14-
side of the filter assembly, to measure the pressure drop
across the filter. A flow meter was used to monitor and
regulate the flow through the Andersen Sampler during the
course of each run. Two sample probes were both connected
to 1 cfm Millipore filter holders (142 mm) fitted with Gelman
Type A glass fiber filter pads and vacuum pumps. The fourth
filter was either a 293 mm, 4 cfm glass fiber; or a 142 mm,
4 cfm glass fiber, the former being used on engine stand runs
and the latter on vehicles.
Prior to use, all the filters were stored in the instrument
room which is temperature- and humidity-controlled. The
filters were placed on the tray of the Mettler Analytical
Balance, allowed to reach equilibrium, and then weighed
out to 0.1 milligram (mg).
After the test, the filters were removed from the holders and
again allowed to reach equilibrium, noted by no further change
in weight, and then weighed to 0.1 mg. This was done in the
same room in which the papers were stored. The Millipore filter
pads used were 142 mm Type AAWP 0.8y. The glass fiber filter
pads used were Gelman 0.3y Type A.
Andersen Sampler Model 0203 with a back-up 142 mm Millipore
filter was used as the basic particle collection device for
determining mass size distribution. Sample probes sized to
deliver an isokinetic sample from the dilution tube were
connected to the Andersen Sampler through which a proportional
sample was drawn at 1 cfm. The D_0 cut-off values for the
Andersen stages are listed in Table 4. The D5Q value is the
size at which 50% of those particles are collected, while the
remaining 50% pass on through to be collected on the next
stage.
-------
-15-
TABLE 4
D5Q VALUE - ANDERSEN MODEL 0203
Stage 1 D5Q 9y
Stage 2 DSQ 5.45y
Stage 3 DS() 2.95y
Stage 4 DSQ 1.55y
Stage 5 DS() 0.95y
Stage 6 D5Q 0.54y
Preweighed glass collection plates were used in this study.
Back-up filters were either Millipore type AAWP 0.8y or
Gelman 0.3y Type A 142 ram diameter. Gelman glass fiber
filters were routinely used while the Millipore filters
were used for special analytical applications.
C. CONDENSATE COLLECTION
Exhaust gas condensate was collected for aldehyde and NH_ analyses.
A tap was placed into the raw exhaust gas stream, as close to the
tailpipe of a vehicle as practical (about 12 inches in most cases)
and 8 feet from the muffler in an engine run. Raw exhaust was
drawn through a three-stage cold trap at the rate of 1 cfm. The
cold trap consisted of three flasks connected in series containing
40 grams each of DI water, immersed in an ice water bath. The
exhaust gas flow bubbles through the water in the flasks. Condensate
was collected for 41 minutes during a modified Federal cycle cold
start, and for 23 minutes during a Federal cycle hot start. Sampling
was terminated at 25 minutes during a steady state run.
The condensate from the exhaust gas was analyzed for ppm of HCHO and
NH.J. It was felt desirable to express this analysis in volume
percent to compare to the other components analyzed in the exhaust
gas. The procedure for this calculation is as follows:
-------
-16-
The "Ideal Gas Law" was used
PV = nRT
PV=
The total liters of exhaust that was put through the condenser is
known. The liters of the aldehyde can be calculated from the
formula above , so the volume percent can be calculated. This
volume percent is reported as volume parts per million in the
exhaust.
D. ANALYTICAL METHODS
Collected exhaust particles have been analyzed for both physical
and chemical character. Many analytical techniques have been
employed in the past, some of which provide very similar data in
the interest of correlating trends observed. This section reviews
the basic analytical concepts applied to each of the many test
components from fuels to exhaust particles. Detailed descriptions
of the specific analytical procedures employed are then presented.
Table 5 is a summary of the techniques used on the exhaust emissions.
TABLE 5
ANALYTICAL TECHNIQUES FOR EXHAUST SPECIES
02, N_, CO, CO_ Fisher Gas Partitioner
Total Hydrocarbons Beckman Model 109A Flame lonization
Detector.
Oxides of Nitrogen Beckman UV and IR Analyzer
C, H Pyrolysis
Benzo-ct-pyrene Chroma tograph , Fluorescence
Trace Metals Emission Spectroscopy, Atomic
Absorption
Aldehydes Polarography
NH~ Steam Distillation, Titration
-------
-17-
1. Fuels
Each test fuel was analyzed to verify concentrations of
additives under study. Additionally/ complete physical
analysis were determined on the base stock test fuel.
These analysis include (distillation, octane numbers,
fluorescence indicating analysis FIA composition, and
Reid vapor pressure [RVP] and trace metal).
2. Oils
Engine oils were examined for trace metals both before and
after test runs. Compliance of physical properties with
specifications was verified. The oils were only checked on
the engine stand runs, not on vehicles.
3. Diluent Air
Mass and composition of the filtered diluent air particulate
was determined with the engine or vehicle operating in the air
pick-up zone as during a test run. This data was necessary to
provide a correction factor applicable to the mass emission
rates determined during a test run.
4. Exhaust Gases
Engine exhaust gases were analyzed routinely several times
during the conditioning sequence and during sampling runs.
Schematically, exhaust gas sample points are as shown earlier
in Figure 1. The engine exhaust gas was analyzed for oxygen,
nitrogen, carbon monoxide, carbon dioxide, and total unburned
hydrocarbons. The hydrocarbons were broken down into saturates
and unsaturates. These analyses were done by gas chromatography,
chemical absorption, and a total hydrocarbon analyzer. Data
reduction was via an IBM 1800 computer through a Bell Telephone
ASR 33 Teletype interface.
a) Analytical Equipment
A Fisher Gas Partitioner was used for the analysis of
oxygen, nitrogen, carbon monoxide, and carbon dioxide.
The partition column consisted of a 6-foot section
containing hexamethyl phosphoramide and a 6-1/2 foot
section containing 13x molecular sieves in series.
-------
-18-
Total hydrocarbons were obtained from a Beckman Model
109A Total Hydrocarbon Analyzer. The concentration
of unsaturated hydrocarbons was determined by passing
the sample through an absorption tube (1/2" x 8") filled
with 30-60 mesh pink Chromosorb impregnated with 50 percent
mercuric perchlorate.
The output of the gas chromatograph was coupled with
a Hewlett-Packard Model 3370A Digital Intergrator which
has an ASCII coded output to drive an ASR 33 Teletype
and punch paper tape.
b) Sampling
A Neptune Dyna-Pump was used to pull the sample from
the exhaust pipe sampling point through 1/4" OD stainless
steel tubing and transfer it to the total hydrocarbon
analyzer and the gas sampling valve of the gas chromatograpl'
through 1/8" OD stainless steel tubing. A manifold system
was provided to allow the operator to calibrate the
equipment with the appropriate standards.
c) Standardization
A gas mixture containing known concentrations of oxygen,
nitrogen, argon, carbon monoxide, carbon dioxide, and
n-hexane was used as a reference standard for the total
hydrocarbon analyzer and the Fisher Gas Partitioner.
d) Operation
The operator typed the proper computer code and program
number on the teletypewriter, injected the reference
standard, and pressed the integrator start button.
As the peaks emerged, the time and area information
was encoded and stored on punched paper tape. Each
succeeding exhaust gas was identified along with the
total hydrocarbon level, and run in the same manner
as the standard. When the series was finished, the
punched tape was sent to the computer by teletype over
regular telephone lines.
-------
-19-
e) Data Reduction
A typical output format for the gas analysis is shown
in Figure 2. Identification of the components in the
standard was based upon each peak size in descending
order. Estimated retention time was the updated time
of each peak in the standard. Retention time windows
are 4 seconds plus 2 percent of the retention time.
Actual percent is a direct ratio of the area counts
in the unknown sample to the area counts in the standard
times the volume percent in the standard. The total
percent actual will normally be 97-98 percent since
water is removed from the saturated sample after the
sampling valve.
A correction for the unresolved argon in oxygen was
made based upon response factors and the amount of
argon found in a number of exhaust gas samples by mass
spectroscopy.
The actual percent was normalized to 100 percent in the
next column on a moisture free basis, and an Exhaust
Gas Analysis report was issued (Figure 2). The air-
to-fuel ratio was calculated from this analysis, the
total hydrocarbon content, and the percent carbon in
the fuel.
5. Oxides of Nitrogen
a) Equipment
Beckman Ultraviolet Analyzer
Beckman Infrared Analyzer
Recorder - Texas Instrument Company
The above pieces of equipment were in a single, self-
contained unit built by Scott Research Labs Inc.,
San Bernardino, California.
-------
-20-
Figure 2
G. C. A:,'. .LYSIS - TEC!:;«'IC;.L DATA -
CCV RUN £3 CCT 16
'CYCLE 2 72.9 K3UHS
KC 620.
PEAK TIME
NS. ACT. EST.
PCT. V3L.
ACTUAL N3RM.
10-16-70
Cf-r:r>ri:::D
IDENTIFICATION
1 22.
2 59.
3 83.
4 104.
5 107.
21.
59.
83.
107.
IRQ.
0*000
12.003
1 3
0.900
81.0CG
1.626
97.060
2.940
8.50
0«000
12.366
1.53C
0.9S7
03.452
1.675
1CO.OOO
CCMFC3ITZ
CAnr:.;j DIOXIDE
CXYGI;J
(\~GZti
niTnoGc:.'
CARG^iJ i^NOXIDE
T3TALS
DALA^CE DY DIFFEHZI-'C
TOTAL C':;rr/. :i NATI :;i
* r
LEVZL
EXKAUST GAS ANALYSIS
GGV RUi! 23 CCT 15
CYCLE 2 72c9 H3Uf?S
KC 620.
TIHE PERCi:;JT IDir.'TiriCATICM
83.
107.
C3.
1GC5.
59.
0.9 AHCVi!
C3.5 NIT/. . :~ri
i e 5 c .'\ V ;. -
12
DIOXIDE
1C 0.0 TOTAL
FRACTICN C/.r.-DN IN F
10-16-70
TOTAL
C::;:TC::T 620. PPH.
H.IYIO
-------
-21-
b) Calibrating Gases
Nitric oxide (3545 ppm in nitrogen)
Nitrogen dioxide (862 ppm in nitrogen)
These standard gases were purchased from Scott
Research Labs, Inc.
Nitrogen was used as zero calibrating gas.
c) Procedure
Before making NO, NO^ measurements, the paper filters
(Whatman #3) to each analyzer were changed and the
Drierite dryer in the exhaust sample line was replaced.
Both analyzers were standardized using the appropriate
calibrating gas at a constant flow. The zero standardizing
was done using nitrogen as the calibrating gas and using
the same flow rate.
After the instrument was standardized, the exhaust gas was
passed through the analyzer using the same flow rate
as in the standardization step. The NO, N02 values were
recorded by the dual pen Servo-riter recorder. Figure 1
indicates the source of the exhaust gas sample.
6. Exhaust Particles
The collection and classification techniques employed allow
the calculation of mass emission rates in grams/mile of
exhaust particulate. Additionally, cumulative mass distribution
data can be calculated. Several collection methods were used,
and have been discussed previously in Section III-B. The
specific techniques for chemical analysis of this particulate
matter are discussed in this section.
-------
-22-
a) Carbon and Hydrogen
The percentage of carbon and hydrogen in the particulate
was determined by pyrolysis and collection of the combustion
products. An entire 142 mm glass fiber filter containing
the particulate was placed in a large platinum boat. The
boat was then transferred to a combustion tube, and the
sample was combusted at 1100°C for 3/4 hour. Carbon dioxide
and water were absorbed in micro absorption tubes and weighed
in the conventional manner. The C and H values were then
calculated from the increase in weight using the given
weight of the particulate.
In general, this technique is quite accurate for carbon and
hydrogen analysis. However, the small sample sizes generated
in a 23 minute cycle or from vehicles or engines operating
on unleaded fuel make it difficult to obtain precise results.
For example, the 142 MM Gelman glass fiber filters have a
blank of approximately 7 mg for hydrogen and a spread of
nearly 1 mg. For carbon the blanks are over 2 mg with a
spread of 0.5 mg. It is not uncommon to have sample sizes
of less than 2 mg; therefore, the inherent inaccuracy of
weighings (even using a 5 place balance) plus the large
blank size make the results of a small sample only meaningful
in a gross comparative sense.
This technique can be used on samples collected on the
Andersen Sampler plates by careful transfer of the particulate
to the combustion chamber. However, even with the best
handling techniques the transfer of particulate is only
about 30 percent. In general, engine runs in which very
little sample was collected on the filter pads also gave
very little on the Andersen Sampler Plates.
-------
-23-
Nitrogen can also be determined by pyrolysis, but due to
the small sample size no meaningful results have been
obtained on nitrogen content.
b) Benzo(a)pyrene
Samples of exhaust particulate were collected on Gelman
142 mm glass fiber filter pads in a Millipore filter holder
operating at 1 cfm. Particulate weights gathered in this
fashion ranged from 0.2 to 35 mg. The samples on the
glass filter pads were analyzed for benzo(a)pyrene in the
following manner.
When available a sample of at least 10 mg (on either one
or two filter papers) was used for analysis. The filters
were folded and rolled with the particulates toward the
inside of the roll and tied with copper wire. The rolls
were Soxhlet extracted for at least 6 hours (with siphoning
four to six times per hour) with 75 ml of benzene. The
extracts were evaporated under a stream of filtered air at
room temperature to approximately 3 ml. This concentrate was
filtered through a M-fritted glass filter into a tared vial.
The flask and filter were washed three times with approx-
imately 2 ml of benzene for each wash. The combined
filtrates were evaporated to dryness at room temperature
with a stream of filtered air.
The residues obtained from both sample and blank filters
were weighed and the difference between them designated
"benzene soluble weight" for each sample. The residue was
dissolved in 0.2 ml of methylene chloride and a 10-40 yl
aliquot was spotted in 2 yl increments on a pre-conditioned
Alumina TLC plate along with a known standard of benzo(a)-
pyrene in methylene chloride. The TLC plates were conditioned
-------
-24-
by heating at 120°C for 1.5 hours and desiccating overnight
in a 45 percent relative humidity chamber (saturated
aqueous zinc nitrate). The TLC plate was developed in an
unsaturated tank containing 20 ml of ethyl ether in 200 ml
of n-pentane to a height of 15 cm (approximately 45 minutes).
The benzo(a)pyrene spots were identified by comparison of
Rf's with that of the standard spot under an ultraviolet
lamp. The spots, marked with a pencil, were circumscribed
with a #15 cork borer and scraped from the plate into vials.
All TLC work was performed as much as possible in a dimly
lighted area to avoid decomposition of the benzo(a)pyrene.
Five ml of 5 percent acetone in n-pentane was added to the
alumina in the vial and it was agitated for 15 minutes
on a mechanical shaker. The slurry was filtered through
a F sintered glass filter into a vial, washing the alumina
four times with approximately 2 ml of 5 percent acetone in
n-pentane with a 45-second soak period between each wash.
The combined filtrates were evaporated to dryness at room
temperature using a stream of filtered air. The benzo(a)-
pyrene residue was taken up in 2.0 ml of concentrated
sulfuric acid. This solution was evacuated for five
minutes to remove trapped air bubbles and its fluorescence
was measured in a one-cm cell at 540 nm while exciting at
470 nm on an Amino-Bowman Spectrophotofluorometer using a
#4 slit arrangement and a sensitivity of 30.
Standard and blanks were carried through the entire TLC
procedure. The blanks were subtracted from all fluorescence
readings and the net fluorescence values for each sample
were used to calculate the amount of benzo(a)pyrene present.
Throughout all steps in the procedure the samples were
refrigerated when not actually being processed and exposure
of the samples to light was kept at a minimum.
-------
-25-
c) Trace Metals
Both atomic absorption and emission spectrometry were
used for determination of metals in the particulate. Atomic
absorption was primarily used for lead determination. Trace
metals were determined by ES on millipore filters while lead
was determined as a percent of the particulate collected on
the 142 mm, 1 cfm fiberglass filter.
1) Emission Spectrometry
a. Principle
Organic matter in the sample is destroyed by wet ashing
in sulfuric, nitric and perchloric acids. The resulting
solution is taken to dryness and the residue is taken
up in a spectroscopic buffer solution containing the
internal reference element/ palladium. A portion of
the solution is dried on pure graphite electrodes.
The electrodes thus prepared are excited in an a.c.
arc discharge and the spectrum is photographed. The
intensity ratios of selected lines are determined photo-
metrically and the concentration of each element is
read from an analytical curve relating intensity ratio
to concentration.
b. Apparatus
1) Excitation. Excitation is obtained by the use
of a 2400 volt a.c. arc discharge - Jarrel-Ash Custom
Varisource, or equivalent.
2) Spectrograph - Baird 3 meter grating spectrograph.
Reciprocal dispersion is 5.55 A/mm in the first order.
3) Developing equipment - Jarrel-Ash Company. Plates
are developed in a thermostatically controlled
developing machine, washed and dried over heat
in a stream of air.
4) Densitometer. Spectral lines are measured with
a non-recording projection type densitometer. Densi-
tometer Comparator, Baird Associates Inc., or
equivalent.
-------
-26-
5) Calculating equipment. A calculating board is
employed to covert densitometer readings to log
intensity ratios. Jarrel-Ash Co.
6) Wet ashing equipment. A micro Kjeldahl digestion
rack is used for wet ashing the organic solvents.
Reagents and Materials
1) Distilled nitric and perchloric acids. Perchloric
acid is an intense oxidizing agent. Organic matter
should not be heated in perchloric acid unless in the
presence of sulfuric or nitric acid.
2) Sodium nitrate, reagent grade (NaNO_).
3) Palladium diamine nitrite, Pd(NH-)(N02)2.
4) Water soluble salts of the elements Al, Ca, Cu,
Fe, Mg, Mn, Ni, Pb, Sn, and Zn.
5) Electrodes, high purity graphite, 1/4" diameter
by 3/4" length. Ultra Carbon Corporation.
6) Photographic plates - Eastman Spectrum Analysis
No. 3.
7) Kjeldahl flasks, 10 ml.
Calibration
1) 0.2182 gm of palladium diamine nitrite
Pd(NH3)2N02)2 were dissolved in water. 10 ml of
concentrated reagent grade nitric acid were added
and the mixture diluted to volume with water in
a ICu ml volumetric flask. This solution contains
1 mg Pd per ml.
2) A buffer solution was prepared by dissolving 20
gm of sodium nitrate in water. 5.0 ml of the palladium
solution above and 7.5 ml of concentrated reagent grade
nitric acid were added and the whole diluted to 100
ml.
-------
27
3) A stock solution containing 0.01% (0.1 mg/ml)
each of the elements Al, Ca, Cu, Fe, Mg, Mn, Ni, Pb, Sn,
and Zn was prepared. Two aliquots of this solution were
diluted ten-fold and one hundred-fold to provide 0.001%
and 0.0001% solutions.
4) Standard additions of the impurity elements were
made to Kjeldahl flasks as shown in Table 6.
5) 0.5 ml of concentrated reagent grade sulfuric acid
was added to the Kjeldahl flasks and the solution
evaporated to dryness. After cooling, 1 ml of concen-
trated nitric acid was added and the mixture was
evaporated to dryness again. The residue was taken up
in 5 ml of buffer solution, warming, if necessary,
to put the salts into solution.
6) The end of the 3/4" graphite electrodes was polished
on filter paper and placed in a stainless steel drying
tray. A drop of kerosene was placed on the top of
each electrode to seal the porosity and the electrode
allowed to dry. One pair of electrodes was prepared
for each of the standard addition solutions by pipetting
0.03 ml of the solution onto the end of each electrode.
The electrodes were dried slowly over micro burners
in a gas drying oven and stored in a desiccator until
run.
7) The samples were excited in water cooled electrode
holders using the following conditions:
(1) Current, 4.0 amps, a.c. arc.
(2) Spectral region, 2150-3550 A.
(3) Slit width, 50y.
(4) Electrode gap, 2 mm.
(5) Pre-burn period, 10 seconds.
(6) Exposure period, 90 seconds.
-------
-28-
Conccntration
Table 6
nl. of standard addition impurity solution
Blank
0.00001%
0.000025%
O.CCOG5%
0.0001%
o.oco:,'5%
O.OGO.^
0.00075%
0.001%
O.C02S%
0.005%
0.01 %
0.5 ml.
1.25 nil.
V o «mJ &.*& o
0.5 nl.
1.25 nl.
2.5 til.
0.3715 nl.
O.5 Lil.
.1.23 nl.
2.5 ral.
5.0
0.0001% 8
it
o.coisi
II
II
II
0.01C^
It
II
II
II
10!
ti
it
it
it
it
ti
it
ti
it
ii
Element
Ni
Ni
Pb
Pb
fin
Sn
7n
Table 7
Analyticr.1 Liao
Analytical
Lino A
Al
Ca
Cu
Fe
Fe
£13-
US
l!n
Un
3032.71
3179.33
V
3273. CO
3021.O7
3020.84
XG02.G0
2779. £3
2&33.G3
27P-S.C2
Interns! Standard
_ Llc.0 A
3027.91 Pd
ii
it
it
?rVJ.7 <"-
wk/t> 4 . *- V
2873.33
2033.07
31 VS.02
2SG3.33
33':.'i.O,T
it
" cJoround
Concentration
Range %
0.000025-0.OO10
O.OOO25-0.010
O.00001-O.00025
O.OO01-O.O10
O.O00025-0.0050
0.000025-O.C010
O.O005-O.O10
O.OO05-O.O1O
O.OOC01-O.CD10
0.000025-O.0010
O.OC05-O.OIO
O.OO10-0.010
O.OO005-0.005O
O.00005-O.C05O
O.O0075-O.O1O
O.OOO1-O.O10
-------
-29-
8) The emulsion was calibrated by use of a stepped filter
or by other recommended methods described in the
"Recommended Practice of Photographic Photometry in
Spectrochemical Analysis" A.S.T.M. Designation:
E116, Methods for Emission Spectrochemical Analysis,
(1964) .
9) The emulsion was processed according to the following
conditions:
(1) Developer (D19, 20.5°C), 3 1/2 minutes.
(2) Stop bath (SB-4), 1 minute.
(3) Fixing bath (Kodak Rapid Fixer) , 2 minutes.
(4) Washing, 3 minutes.
(5) Drying, in a stream of warm air.
10) The relevant analytical line pairs were selected
from Table 7. The relative transmittances of the
internal standard line and each analytical line were
measured with a densitometer. The transmittance
measurements of the analytical line pairs were converted
to intensity ratios by the use of an emulsion calibration
curve and a calculating board.
11) Analytical curves were constructed by plotting
concentration as a function of intensity ratio on log-
log graph paper. For best results, the average of
at least four determinations recorded on two plates were
plotted.
e. Procedure
1) The available sample was weighed directly into
a Kjeldahl flask. Sulfuric acid was not used in the
wet ash procedure because test samples usually contained
-------
-30-
a large amount of lead which would form the insoluble
sulfate. Wet oxidation was carried out with nitric
and perchloric acid only. Extreme caution was exercised
in the use of this technique. Concentrated nitric
acid was added dropwise, a few tenths ml at a time,
to the hot mixture to aid in oxidation. A few drops
of concentrated perchloric acid may be added to the
hot solution after most of the free carbon has been
destroyed, to hasten complete oxidation. When the
solution became water clear, it was evaporated to dryness.
After cooling, 0.5 ml of nitric acid.was added and the
mixture evaporated to dryness. The addition of 0.5 ml
of nitric acid was repeated and the solution evaporated
to dryness again. The inorganic residue was dissolved
in dilute nitric acid and the volume adjusted to a
known concentration, usually 10 mg/ml. If the original
sample size was below 30 mg, a less concentrated solution
was usually made up. Aliquots of this solution were taken
to dryness and then the buffer solution (d2) added
in an amount to give a dilution factor of lOOx. One
sample was analyzed by the direct reader while a
second was examined photographically. Some samples
had to be run at factors larger than lOOx in order
to get the concentration for some elements to fall
within the range of the analytical curves. By varying
the sample to buffer ratio any number of concentration
or dilution factors could be achieved. A blank of
the acids used was carried through in the same manner
as the sample.
2) Proceed as in d(6), (7), (8), (9), and (10) of the
calibration procedure. Duplicate spectra were recorded
for each sample.
-------
-31-
f. Calculations
The intensity ratios were converted to concentration by
use of the analytical curves.
g. Precision and Accuracy
Representative precision and accuracy of the method are
given in Table 80 Each of the twelve samples A.. , A«, A_ ,
B, , B2, B.,, C, , , C.,, D, , D_, D_, was analyzed by means
of duplicate excitation.
2) Atomic Absorption
a. Method for Lead Determination
Following nitric acid digestion, particulate samples were
washed into 50-ml volumetric flasks and diluted to mark.
This normally put the concentration of lead in the flasks
between 20 and 200 yg Pb/ml. If the concentration was
higher than 200 yg Pb/ml, the sample required redilution.
The samples were analyzed on an atomic absorption spectro-
photometer (Perkin-Elmer Model 303) using a hollow cathode
lamp with a lead cathode filament. Operating conditions
were as follows: 10 milliamps tube current, light path
slit opening - 4, ultraviolet light range, acetylene-
air oxidizing flame, one-slot burner head, wavelength -
2170 angstroms. The sample solution is aspirated into
the flame where lead atoms present absorb the light from
the lead cathode filament. The amount of absorbed light
is proportional to the concentration of lead. The samples
were analyzed in conjunction with the following series
of lead standards: 10, 20, 40 , 60, 80, 100, 150 , and 200 yg
Pb/ml. The concentration of the standards was plotted
versus their absorbance values giving a standard curve.
With the absorbance values for the samples and the standard
curve, it was possible to determine the concentration of
-------
TABLE 8
0>
1-1
D.
E
a
w
Al
A.
A2
A3
El
JL
B2
E3
Cl
«L
C2
C3
Dl
JL
»2
D3
% Al
O.OOO044
O.OOOO52
0.000045
O.OOOO52
O . OCGO-1
O.OOOO52
O.OOO12
O.CC0097
O.OC0097
O.OCCO94
O.CCCG32
O.CC011
O.O0023
O.COD2O
O.O032O
o.cco::;3
O CC.'J24
0.00028
O.OCO74
O.OOOS4
O . CC053
O.COOGS
O.OO059
O.O0053
A. , A3, ted A3
070001% of Al
and O.O025% of
_ ^
KtPKtb
% Ca
0.00043
0.00050
0.00043
O.CC037
O.OCXM3
O.OO050
O.OO1O5
O.OOC33
O.OOOC5
O.OQOG3
O.COOS2
O.CC074
O.O023
O.0018
0. 00223
O.C02C3
O.O023-
O.OO275
O.C07O
O.CC'5'1
O.C019
O.G057
0.00-13
O.OO30
contain.
tN 1 A 1 1 Vt V
% Cu
O.OOOO48
O.OOO054
O.OOOO46
0.000047
O.OOOO50
O.OOOO48
O.OO012
O.OO010
O.OO0099
O.O30O95
0.000025
O.OOOOS6
O.O0023
O.O0020
O.OC023
0.00025
O.COO26
O.OOO28
__
__
__
__
0. 00005% of
KtLlblUIN
% Fe
O.OOO43
O.OO055
O. 00044
0.00043
O.OOO46
O.OOO46
O.OO1O
0.00094
0.00030
O.CO1O5
O.OO10
0.0010
O.O025
O.CO3O
O.C023
O.OO235
O.C0275
O. 00235
O.CDG5
O.G023
O.CO57
O.OO59
O.OO5O
O.OO55
Al and Cu
AINU AL
% Kg
0.00049
O.OOO52
O . OO04 7
O.OOO50
O.OOO53
O.OOO49
O.001O5
O.O0095
0. 00092
O.QOQ91
O.OO10
O.COO90
O.0023
O.OO23
O.C023
O.0024
0.0023
0.0024
O.0057
O.O051
O.0048
O.OO-17
O.O045
O.O055
, and 0
LUKALY (J
% Mn
O.OOO46
O.O0057
O.OO051
O.OO050
O.OOO49
O.OOO46
0.0010
O.OO12
O.OO11
O.OOO66
O.OOO86
O.OO092
O.002G5
O. 00195
O.CO2S5
O.OO275
0.00245
O.OO25
O.OO59
O.O058
0.0045
O.OO48
0.0047
O.0054
.OOO5% of
h bniSblU
r.- NI
O.OOO47
0.00055
' 0.000-15
0.00051
O . OOO4 7
O.OOO48
O.OO10
0.00096
O.0010
O.OO105'
O.OO10
O.OO105
O. 0024 5
O.C02S5
O.OO23
O.OO245
O.0026
0.00255
0.0035
O.OO58
O.OO56
O.O057
O.OO50
O.O055
each other
and Cu, and O.COIOJ of each other element. C-, , Cfi, and Ca
each other element.
Dlf DB aad D3 contain 6.OO65% of Al
N bPhU 1 H
% Pb
O.OOO56
O. OOO 59
O.OO050
0.00051
O.OOO52
0.00053
O.OO1O5
O.OO098
0.0010
O.OO105
O.OO10
O.OO1O
0.00235
0.00255
0.00245
O.CO26
O.0025
O.OO245
O.OO53
O.OO45
O.OO45
0.0043
0.0043
O.OO49
element
contain*
tUbUUPY
% Sn
O.OOO52
O. 00059
O.OOO53
O.OOO50
O.OOO5O
O.OOO46
0.0011
O.OOO94
O. 00105
0.00105
O.O0099
O.OO10
O. 00255
0.0027
O. 00215
O.OO23
0.0025
0.00265
O.OO54
O.OO59
O.OO53
O.O057
O.OO54
O.OO49
BI f B g »
o. 60025%
and Cu and O.OO5O%
%Zn
O.COO4O
0.00045
O.OOO54
O.OOO4O
O.OOO52
O.O0042
O.OOO94
0.0012
O.OO125
O.OO1O
O.OOO96
O.OO115
O.OO14
O.OO215
O.O0225
O.OO3O
O.OO3O
O.OO2O
O.OO58
O.005O
O.O050
O.OO3O
O.OO37
O.CXM1
and B3 contain
of Al and
of each "
Cu
»her
ro
i
elenent.
-------
-33-
lead in the samples. The sensitivity for the lead deter-
mination in an air-acetylene flame is about 0.25 yg Pb/ml
at 1 percent absorption. The detection limit is 0.1 yg
Pb/ml.
b. Determination of Lead and Iron in Engine Combustion
Chamber Deposits
These samples were thoroughly ground in a mortar prior to
analysis to obtain uniform samples. The ground sample was
dissolved in nitric acid and lead determined by atomic
absorption. A portion of the sample solution was also used
in the determination of iron. Iron is reduced with
hydroxylamine to the ferrous state, and reacted with
1,10-phenanthroline in an acetate buffered solution (pll 5)
to form an orange-red complex. Photometric measurements
were made using a Beckman DU-2 spectrophotometer. Operating
conditions were as follows: sensitivity setting - 2,
slit opening - 0.10 mm, wavelength - 510 nm, 40 mm optical
cells. The concentration of iron was determined from a
standard curve. For a one gram sample diluted to 100 ml,
the detection limit is about 1 ppm and the sensitivity
± 1 ppm.
c. Gravimetric Method for Lead Determination in Millipore
Filters
Following nitric acid digestion, concentrated sulfuric
acid was added to the sample to precipitate lead sulfate.
The solution was filtered, and the precipitate dried and
weighed to determine the amount of lead percent. In
addition, the filtrate was analyzed by atomic absorption
for trace amounts of lead. This analysis is included in
the total amount of lead reported for the sample.
-------
-34-
d. Determination of Lead and Other Metals in Fiberglass
Filters
The fiberglass filters cannot be digested completely with
nitric acid. They were cooked with concentrated nitric
acid for two hours to leach out the metals. The pulp was
filtered and washed and the filtrate analyzed by atomic
absorption for lead, and by emission spectroscopy for other
metals.
7. Coriden sate Analyses
Condensate was collected from the raw exhaust as described in
Section III-C. The condensate was analyzed for aldehydes and
NH3 using the procedures outlined below.
a) Aldehydes
The analytical method for the determination of carbonyl
compounds in automotive exhaust emissions employed
polarographic techniques. Samples for analysis were collected
from undiluted exhaust effluent using ice-water cooled
cold traps and via a sample probe welded into the engine
or vehicle exhaust system. A Princeton Applied Research
Model 170 Electrochemistry System was used as the monitoring
device. The derivative pulse polarographic mode yielded
the best combination of resolution and sensitivity for
the classification of carbonyl compounds. A dropping mercury
electrode with a Princeton Model 172 Drop Timer was employed
as the working electrode.
Hydrazine derivatives (hydrazones) were employed for the
determination of the carbonyl compounds, since hydrazones
are easier to reduce than the free compounds, thus eliminating
many possible interferences.
-------
-35-
An acetate buffer of approximately pH 4 (an equimolar mixture
of acetic acid and sodium acetate, 0.1M in water) was used
to control pH for hydrazone formation and also acted as
supporting electrolyte. Hydrazine was added as a 2 percent
aqueous solution. In this system formaldehyde gave a peak
potential (half-wave potential) of -0.92v vs. a saturated
calomel reference electrode. A platinum wire was employed
as the auxiliary electrode.
With the above system, it is possible to distinguish between
and simultaneously determine aromatic aldehydes, formaldehyde,
higher aliphatic aldehydes, and aliphatic ketones as shown in
Figure 3.
Since aromatic ketones, e.g. benzophenone, give polarographic
response in pH 4 buffer without hydrazine, it is also possible
to detect aromatic ketones. Lead and zinc could also be
determined from the samples under these conditions.
Since formaldehyde was the main carbonyl component of the
condensate samples, all results were calibrated against and
reported as formaldehyde. The upper curve in Figure 4 shows
an actual sample without hydrazine present and demonstrates
the lack of interference in the carbonyl region. The lower
curve shows the same sample after the addition of hydrazine.
Figure 5 shows the same solution after the addition of
a formaldehyde standard. These two figures clearly establish
the presence of formaldehyde in the exhaust samples.
Procedure
Pipet 2 ml of methanol sample into a 25-ml volumetric flask.
Add 10 ml of pH 4 acetate buffer and dilute to volume with
water. Transfer this solution to a polarographic cell and
deaerate with oxygen-free nitrogen for ten minutes. Record a
-------
-36-
Figure
H ; H H-Polarographic Determination of Aldehydes
-----i-
-
6v ivs ;SCF
i I . .; ....: I ,
. . , j. i
i i" i'LPT1
.1 ,.,. | .......
-------
-:''"' Figure
; ; _-_;_p<>larograph1c Determination of Aldehydes:}^"
Polarographlc Determination of Aldehydes
_asabove;withnydrazine
.
I
U>
-------
-38-
derivative pulse polarogram from 0 to -1.6 V vs. SCE. Add
2 ml of hydrazine reagent to the polarographic cell and
deaerate for 5 minutes Again, record the polarogram from
0 to 1.6 V vs. SCE.
Lead and aromatic ketones are determined from the waves
obtained without hydrazine at the peak potentials listed
above. Formaldehyde/ higher aliphatic aldehydes, aromatic
aldehydes, and aliphatic ketones can be determined from
the second polarogram with hydrazine present.
All responses should be calibrated by addition of known
amounts of standard compounds to actual runs. Peak heights
are linear with concentration.
In this system, zinc has a peak potential of -1.00 V vs.
SCE, but it can be differentiated from benzophenone by
the fact that it possesses only one polarographic wave.
A blind comparison of the polarographic technique vs.
the MBTH technique was made, and the results were as
follows, expressed as formaldehyde:
MBTH Polarographic
340 ppm 300 ppm
1500 ppm 1530 ppm
430 ppm 480 ppm
105 ppm 110 ppm
150 ppm 110 ppm
b) Ammonia
Ammonia is present in the exhaust gas condensate and is
analyzed in the following manner.
-------
5/1.
I '
*J
Distilling
Flask
ivacunced
Jacket \
Ilot Over
'Frcrr. Bottom
; I
; i
{
:
i;
: -\
MICRO-KJELDAHL APPARATUS {
(Mount on Rins Stand)
r
;
c
j
i
f
? .
fi
1 .
}
1 I
c
tf
J <.
5
1
i
. a
JS |
'
I
i
J
^
if
c
f
« ^ f. -,-.
^f.^ '~~'
\ /s-.^ir
\ ^ i
\
Trao i
3 - *
^/ i
' "^^ k
1*" Stcan /
Disti3.1?.tic^r^
FI^slc/"
iTucs to
:Fi12
; xTi'Fl3sK
r*
^ "~ ;
/ ~ ~ ~»^ ---__
* - iOv ^om ;
>~ - - i
ii
50 ml. Erlenneye:
Receiving Flask
. £. r-r:
!i
i
U)
FIGURE 6
.APPARATUS FOR DETERMINATIONOF NH,
-------
-40-
A 5-10 cc aliquot of condensate is added to a 50 percent
potassium hydroxide solution. This mixture is then steam
distilled into an excess of 0.010 N hydrochloric acid.
The excess acid is determined by adding potassium iodide
and iodate and titrating the liberated iodine with 0.010 N
sodium thiosulfate.
This technique is capable of determining ammonia as low
as 0.3 ppm. Figure 6 is a sketch of the apparatus used
for the determination.
The analytical procedures given herein have been adapted from
literature sources or developed upon the basis of experimental
data which are believed to be reliable. In the hands of a qualified
analyst they are expected to yield results of sufficient accuracy
for their intended purposes. However, The Dow Chemical Company
makes no representation or warranty whatsoever concerning the
procedures or results to be obtained and assumes no liability in
connection with their use. Users are cautioned to confirm the
suitability of the methods by appropriate tests.
-------
-41-
IV. EXPERIMENTAL RESULTS
The current extention to contract EHS 70-101 was outlined into
five basic tasks. These tasks are presented in Table 9 with a
brief description of the objective of each task and the approach
used to accomplish the objective.
TABLE 9
TASKS AND OBJECTIVES
TASK I - Evaluation of Particulate Sampling Parameters
Objective: To evaluate the relative significance of various sampling
parameters on the collection of particulate matter and to identify
therefrom a set of sampling conditions which will afford reliable
and meaningful data for the remainder of the work program.
Approach; Particulate matter from air diluted automotive exhaust
was sampled and the effect of changes in the following sampling
variables was monitored: sample line temperature, dilution ratio,
sample probe location, filter flow rate, and sampling time. This
work was performed using non-leaded fuel. Grams/mile of particulate
was measured.
TASK II - Determination of Baseline Data from Engine Dynamometer
Tests
Objective: To establish baseline data for particulate emissions ,
aldehydes, ammonia and benzo-a-pyrene present in the exhaust
effluent of an internal combustion engine operating on non-leaded
and low-lead gasoline under controlled conditions on an engine
dynamometer.
-------
-42-
Approach; A 1972 Pontiac 400 CID V-8 engine was procured and a
reference set of tests were run on this engine using non-lead
and low-lead fuels. The above emission data was generated under
the following test conditions:
Road load 60 mph
Road load 30 mph
Mild cycling (Dow cycle) to include cold start
For each of these conditions, the effect of the following variables
on such emissions were evaluated.
Air/fuel ratio (three different values at road
load 30 and 60, 1 value under
cycling)
Ignition timing (two different settings)
TASK III - Evaluation of Emission Control Devices by Engine
Dynamometer Tests
Objective; To evaluate the effect of various exhaust emission
control devices on the non-regulated emissions identified in
Task II.
Approach; For each emission control device to be considered the
tests outlined in Task II were repeated as applicable. The devices
tested included:
Sub-task IIIA Three HC-Co oxidation catalysts
Sub-task IIIB One NO reduction catalyst
X
Sub-task IIIC Exhaust gas recirculation
-------
-43-
TASK IV - Evaluation of the Effect of Emission Control Devices
on Non-Regulated Emissions by Vehicle Testing
Objective; To characterize the non-regulated emissions outlined
in Task II for vehicles equipped with emission control devices.
Approach: Vehicles equipped with emission control devices were
tested periodically (as a function of mileage accumulation) and
the above non-regulated emissions measured using the 1975-76
Federal Test Procedure on a chassis dynamometer. Four of these
vehicles were made available from automotive company durability
programs. Where possible, each vehicle was tested on at least three
separate occasions. Additional vehicles equipped by the Office
of Air Programs, Environmental Protection Agency (EPA), were tested
as scheduled by the Contract Officer.
TASK V - Diesel Engine Characterization
Objective: To initiate a limited diesel engine study to establish
baseline data foa^-emissions present in the exhaust stream.
Approach; A single cylinder Labeco diesel engine was used to study
sampling parameters and the effect of a dilution tube on the diesel
exhaust.
The data and conclusions for each task are presented separately, as
well as any discussion of operating parameters, analyses, or effects
of devices.
-------
-44-
A. TASK I
1. Introduction
The major objective of the first task outlined in the current
extension to the subject contract was to study the effect of
specific engine variables on certain non-regulated exhaust
emissions under highly controlled conditions. The emissions
defined for study included particulate matter (organic and
inorganic), aldehydes, ammonia and benzo-a-pyrene. Studies
were made on engines operating on non-leaded and low-lead fuels.
During the course of some preliminary studies undertaken in our
previous years work, alarming differences were noted when
particulate matter emitted in the exhaust stream of an engine
operating on non-leaded gasoline was collected on filters
operating at different flow rates and temperatures.
In order to reach an understanding of the above effects and
to allow the definition of a meaningful set of particulate
sampling parameters which could be used throughout the current
contract efforts, a study had been made of a number of sampling
variables and their effect on the mass of particulate matter
collected at a filter.
2. Experimental Procedure
All of the studies described herein were conducted using either
a 1971 Chevrolet Impala fitted with a 350 CID V-8 engine and
operated on a Clayton Model chassis dynamometer, or a 1972
Pontiac 400 CID V-8 engine operated on a General Electric Model
dynamometer. Both the vehicle and the engine were run on
Indolene 0 non-leaded fuel, except where noted. The exhaust
effluents from both the vehicle and engine were fed to similar
polyvinyl chloride (PVC) dilution tubes which have been described
in Section III-B. Separate dilution tubes were used for the
-------
-45-
vehicle and engine studies. Unless otherwise specified,
stainless steel sampling probes were located in the dilution
tube at the end remote from the air and exhaust inlets. All
tests were made with the vehicle or engine operating under 60
mph road load, steady-state conditions.
Appendix A is a report on work carried out at Dow's expense
prior to the current contract extension. This work led to
the identification of the parameters being studied in Task I.
Experiments for Task I were run to evaluate the effects of
various operating parameters as outlined below:
1. Effect of Dilution Tube Velocity
2. Effect of Filter Temperature
3. Effect of Dilution Air Temperature
\
4. Effect of Sample Probe Temperature
5. Effect of Sample Probe Location
6. Effect of Face Velocity Through the Filter Media
7. Effect of Sample Collection Time
8. Effect of Test Mode
It must be noted that in determining the effect of any one
variable, it was extremely difficult to hold all other
variables constant. Therefore, a complete analysis of the
effect of each variable by itself can only be made by inference.
In many cases, the same runs were used to try to evaluate several
parameters. For example, the effect of the sample probe location
was done on the same runs that were used to determine the effect
of filter temperature.
-------
-46-
3. Conclusions for Non-Leaded Fuel Particulate Sampling
a. A change in the temperature differential between the
filter and the sample probe caused differences in
amounts of collected particulate.
b. Changes in rate through the dilution tube caused slight
variations in the amounts of particulate collected.
c. Dilution air temperature proved to be important since
a lower temperature showed definite increases in the
amount of particulate collected.
d. Sample probe location appeared to have only very small
effects on particulate samples.
e. Face velocity of a sample stream through a given filter
was important in that more sample, comparatively, tended
to be collected at lower face velocities.
f. A majority of the sample, during a steady-state run, was
collected within the first 25 percent of the time period
over which the run was made.
g. Absolute measurements of grams/mile are misleading
when measured during a steady-state run of long duration
for the reason given in f above. Comparisons can be
made, however, between runs of like time periods, and
are valid as measurements of a trend.
h. The modified Federal cycle cold start gave more
particulate than the 23 minute Federal cycle.
-------
-47-
i. Most future work will be done at the following
conditions?
Filter temperature controlled to 100°F.
Dilution tube velocity controlled to 400 ft/min.
Inside dilution air used at all times.
Sample will be collected on 142 mm filters.
Steady-state runs will be 2 hours in duration.
Filter rate will be 1 cfm.
Sample probes will be used only at the end of
the dilution tube.
4. Effects of Various Operating Parameters
Dilution Tube Velocity - The effect of velocity of the diluted
exhaust in the dilution tube was studied in relation to the
effect it might have on the amount of particulate collected.
The raw data for this study is presented in Tables 10, 11, 12,
and 13.
The dilution tube flow rate was varied from 300, 400, and 500
ft/min by using increased amounts of dilution air. Table 10
is a study of the true velocity effect in the tube at the
three rates just mentioned. The flow rate would be expected
to show some side wall effect and, in fact, does. Gas flow
through the dilution tube was measured with an Anemotherm air
meter manufactured by Anemostat Corporation of America. The
general increase in measured flow as the velocity measuring
device was inserted can be attributed to the turbulence created
by the sampling probe itself, as well as some leakage at the
lower end,, The three sampling zones are described in Figure 7.
The first sample zone was 9 feet from the point of entry of the
exhaust; sample zone 2 was 16 feet; and, sample zone 3 was
23 feet.
-------
-48-
Table 10
DILUTION TUBE FLOW RATE PROFILE
Inches from
Bottom of Tube
2
4
6
8
10
12
14
15
400 feet/min
2
4
6
8
10
12
14
15
2
4
6
8
10
12
14
15
Sample Location
1
360 ft/min
360 ft/min
380 ft/min
380 ft/min
400 ft/min
400 ft/min
420 ft/min
420 ft/min
500 feet/mi
1
450 ft/min
470 ft/min
490 ft/min
500 ft/min
500 ft/min
500 ft/min
520 ft/min
500 ft/min
300 feet/mi
1
270 ft/min
280 ft/min
290 ft/min
280 ft/min
300 ft/min
300 ft/min
300 ft/min
330 ft/min
2
390 ft/min
410 ft/min
410 ft/min
400 ft/min
400 ft/min
400 ft/min
400 ft/min
410 ft/min
n
Sample Location
2
490 ft/min
500 ft/min
500 ft/min
500 ft/min
500 ft/min
510 ft/min
510 ft/min
530 ft/min
n
Sample Location
2
280 ft/min
300 ft/min
300 ft/min
300 ft/min
300 ft/min
300 ft/min
320 ft/min
320 ft/min '
3
400 ft/min
410 ft/min
400 ft/min
400 ft/min
400 ft/min
410 ft/min
410 ft/min
430 ft/min
3
460 ft/min
500 ft/min
510 ft/min
510 ft/min
500 ft/min
500 ft/min
510 ft/min
510 ft/min
3
300 ft/min
300 ft/min
295 ft/min
290 ft/min
300 ft/min
305 ft/min
310 ft/min
350 ft/min
-------
FIGURE 7
DILUTION TUBE SAMPLE POINTS
16" diameter Dilution Tube
A
A
Exhaust from Engine
Sample #1
-t
t
Sample #2 Sample #3
J -i
" -- j "7«o"
i
*.
vo
-------
-50-
All the sampling of particulate to be analyzed was taken from
the center of the tube (between 6 and 8 inches from the
walls). The flow rates in this zone were quite constant at
all three sampling points, indicating that complete mixing
of the air and exhaust was taking place as close as 9 feet
from the entry of the exhaust. Since the amount of exhaust
remained constant while the amount of dilution air was varied,
the temperature at various dilution rates also varied. Table 11
shows the effect of increased flow rate as well as the effect
of temperature differences on the amounts of particulate
collected. In all cases, except in the study of sample probe
location, all samples were collected 23 feet from the exhaust
inlet.
The following data, extracted from Table 11, shows that the
rate of flow through the tube has a small effect on the
grams/mile of particulate mass, independent of the temperature.
Flow Rate Temperature (°F) Grams Grams/Mile
300
300
400
400
500
500
The grams/mile of particulate collected varied from a high
of .0060 to a low of .0043, with the high point being
400 ft/min; 500 ft/min shows only a small decrease, which
is within experimental deviation.
Table 12 shows the particulate collected using .5 cc/gal
lead fuel at the same three flow rates. The low lead fuel
was used to generate much higher amounts of particulate in
126
106
117
99
108
96
.0044
.0046
.0045
.0045
.0031
.0033
.0043
.0045
.0060
.0060
.0054
.0057
-------
TAFLE 11
DILUTION TUBE FLOW RATE COMPARISON, UNLEADED FUEL
Air-Fuel Ratio (15.5)
Chassis Dynamometer
1971 Chevrolet 350 CID
No TEL Fuel
50 MPH - SS - Run 90
Air-Fuel Ratio (15.3)
Chassis Dynamometer
1971 Chevrolet 350 CID
No TEL Fuel
60 "MPH - SS - Run 91
Air-Fuel Ratio (15.4)
Chassis Dynamometer
1971 Chevrolet 350 CID
Ko TEL Fuel
60 MPH - SS - Run 92
01
.c
f\
C 3
r- H-
» C
0 0
r *'~
U. (-
3
U_ 'r-
<_> 0
467
467
467
467
342
342
342
342
601
601
601
601
c
o
w
3
r-
Q
+J
(/I
3 O
IO "-
J= <->
X IO
uj ce
. 4.49:1
4.49:1
4.49:1
4.49:1
3.24:1
3.24:1
3.24:1
3.24:1
6.59:1
6.59:1 '
6.59:1
6.59:1
en
«*
Q.
E i
Ol
C
Ol T-
J3 O
3 Q-
1
OJ
C r
0 CL
r E
+> IO
3 00
r- -M
O (O
131.5
131 .5
131.5
131.5
141.0
141.0
141.0
141.0
118.0
118.0
118.0
118.0
0
*J o
I- O Uu
o O£ o
E I- "
3 -r- Ol
3
O) C 4->
> O IO
+J 4J Ol
<0 = Q.
« > E
Ol *- Ol
0:0 H-
27/70
27/70
27/70
27/70
30/6?
30/69
30/69
30/69
25.5/73
25.5/73
25.5/73
25.5/73
01 en
.c >
o
Ol »
3
*> C
10 o
S--r-
01 -M
0.3
E
Q) *r~
l-Q
84
- 84
84
84
75
75
75
7.5
83.4
83.4
83. «
83.4
»
E f-^
3 T3
* fl>
L. *->
V r
*»~ »^ 3
+J *r- O
0* U.r
r «j
O
Qi (0"*-^"
400-Feet
72
32
49
.58
300 Feet
99
30
52
80
500 Feet
60
30
44
52
c u.
. 00
3 +J
i C
r- «^-
O 0
O.
10 OIT3
a>
' c E 10
0 00*3
CL. I)
01^-
S-Q m
Ol 3o
Per Minute
76
76
76
76
Per Minute
85
85
85
85
Per Minute
70
70
70
70
s
' E
CM
C
"r"
^
O
J-
u_ Ol
Sr-
OLL.
1 .0
1 .0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
i-
Ol
Lu
IO
Ol Lu
i- 0
3
10 01
L. U
Ol IO
Q.1*-
E S-
01 3
1 to
87.3
117.5
99.9
94.3
86.0
126.6
106.0
95.2
83.9
108.6
96.2
88.5
Ol
+j
10
3
U
f-
+*
u u
03 E 0
r- a 1-
<- to «-
a> 3
r o
20.0
20.0
20.0
20.0
20.0
20.0
20.0
20.0
22.5
22.5
22.5
22.5
01
f i.
i- 01
z: oi J3
t- (0 3
a. 3
U i_
l/> *^ O)
E*» ->
10 t- r-
U <0 -f
C3 a. LL.
1
.0060 2
.0060 3 i
u1
.0054 4 i-.
.OlfBT Avg. '
1
.0043 2
.0045 3
4
.OTJ4T Avg.
1
.0054 2
.0057 3
.0062 4
.0~OTT Avg.
*NG = no good, moisture on filter paper
Key: Filter #1 = Sample line water-cooled
#2 = Sample line insulated
#3 = Safnple'line water jacketed only
#4 = Sample line air-cooled
-------
TABLE 12
DILUTION TUBE FLOW RATE COMPARISON
Low Lead Fuel
Grams
Flow
ft/mi n
400
300
500
Flow in Tube
cfm
445
330
580
Exhaust Dilution
Ratio
4.49:1
3.24:1
6.59:1
Temperature at
Filter Surface
114°F
116°F
104°F
Parti cul ate
on Filter
.0087
.0095
.0072
Grams/Mi
.0322
.0261
.0348
le
Ul
to
Runs 100A, 100B, 100C
All runs on chassis dynamometer, 1971 Chevrolet 350 CID
0.5 cc TEL fuel, 60 mph Steady-state, 6 hour sampling time,
142 mm, 1 cfm fi1ter
-------
-53-
order to help minimize any deviations due to experimental
error. The temperature effects were minimized also. The
grams/mile of particulate mass increased from 0.0261 g to
0.0348 g as the flow went from 300 to 500 ft/min. This
data is shown graphically in Figure 8.
Comparing Runs 90, 91, and 92 (Table 11) to Runs 100A, 10OB,
and 100C (Table 12) shows an increase of 24 percent in grams/mile
of particulate mass in the first case, and an increase of 25
percent in the second, going from 300 ft/min to 500 ft/min. All
of the runs in Table 12 were done on the same day, thus
minimizing any ambient temperature or humidity differences.
Table 13 shows again the effect of flow rate versus particulate
collected. In this experiment, temperature differences between
flow rates, at the same sample points, were small. In each
case, the grams/mile of particulate mass appeared somewhat
higher at 500 ft/min, but some of the effect was probably due
to the small difference in temperature.
Conclusions
The differences in the amount of particulate collected
at the three flow rates studies were small enough so
that a flow rate of 400 ft/min was settled on as the rate
for future work. The differences between the flow rates
were attributed more to the temperature differences than
to any fundamental change due to the exhaust dilution.
Effect of Filter Temperature - Earlier attempts at defining
some of the variables in the dilution tube method of
collecting particulate samples indicated that the temperature
of the gas stream being sampled, at the filter, had an effect
on the amount of particulate collected. The data shown in
Tables 14 and 15 were generated as an attempt to further
characterize the filter temperature effect.
-------
.04
JE^
14
n
_4_
,03
to
3
0
s_
D_
O)
E
s_
tl
frS
02
01
300 400 500
Flow Through Tube, ft/min
Dilution Tube Velocity vs. Grams/Mile Particulate
4, Figure 8
-------
TABLE 13
DILUTION TUBE FLOW RATE COMPARISON, NON-LEADED FUEL
SAMPLE LINE
Ft/Min
Flow in
Tube
300
500
Sampling
Point
Temperature
132.8°F
109. 4°F
Heated
Temp . @
Filter
125°F
111°F
g/Mile
Collected
.0049
.0055
Ambient
Temp . @
Filter
108°F
97°F
g/Mile
Collected
.0060
.0070
Cooled
Temp . @
Filter
82°F
81°F
g/Mile
Collected
= 0064
.0070
Runs 113B and 113C
All runs made on 1972 Pontiac 400 CID engine dynamometer.
No lead fuel, 60 mph steady-state, 2 hour sample time.
142 mm, 1 cfm filter.
-------
-56-
TABLE 14
FILTER TEMPERATURE EFFECT ON PARTICULATE
Temperature
at Filter
Flow Rate in Tube
Feet/Minute
Grams/Particulate
on Filter Grams/Mile
106
126
94
99
117
88
96
109
300
300
400
400
400
500
500
500
.0046
.0044
.0040
.0045
.0045
.0036
.0033
.0031
.0043
.0045
.0054
.0060
.0060
.0062
.0057
.0054
Run on 1971 Chevrolet, 350 CID
Non-leaded fuel, 5.75 hours, 60 mph
Steady-state
-------
TABLE 15
FILTER TEMPERATURE EFFECT ON PARTICULATE
Sample Zone I
Run No .
107
105
102
106
Flow
Rate
300
400
400
500
Temp . Before
Filter
122°F
106°F
86°F
100°F
Grams
Collected
.0087
*
.0094
.0075
.0065
Sample Zone 2
Temp. Before Grams
Filter Collected
.0085
.0095
.0073
.0065
102°F
100°F
100°F
Sample Zone 3
Temp. Before Grams
Filter Collected
113°F
100°F
104°F
100°F
.0092
.0098
.0086
.0061
i
Runs 102, 105, 106, 107
All runs made on 1972 Pontiac 400 CID engine dynamometer
Non-leaded fuel, 60 mph steady-state, sample time 4 hours
142 mm, 1 cfm filter
-------
-58-
At 400 ft/min flow, a change in temperature from 117°F in an
insulated sample line to 99 °F in a water-jacketed line showed
no change in grams/mile of particulate mass collected. A change
from 126°F to 106°F showed a change of only 0.0002 grams out of
0.0045 grams collected. Changes from 109°F to 96°F to 88.5°F
showed weights of 0.0054 g, 0.0057 g, and 0.0062 g. These
changes are quite small, although it is felt that they are
meaningful. In all cases, the higher temperatures tended to
give the least amount of sample collected. However, grams/mile
of particulate mass tended to stay close to being constant with
temperature change. This effect is shown in Figure 9.
Table 15 shows a reversal of the above observation, however.
Note that in sample Zone 1 (not normally used in particulate
samplings), at 400 ft/min an increase in temperature from
86°F to 106°F gave an increase of 0.0019 g particulate. It
is felt that the results of Run 105 (ambient conditions,
etc.) were, for some reason, not comparable to Run 102. In
both cases, the runs were consistent with themselves with
respect to sample probe location. The data is plotted as a
function of temperature versus particulate mass in Figure 10.
There does not appear to be a clear trend in particulate
collected based on temperature effects.
Past experience on particulate sampling has shown that
approaching the dew point of the diluted exhaust stream has
a definite effect on the particulate picked up by the
filters. These runs were all sampled above the dew point.
Conclusions
Available data indicate that the higher filter
temperatures are likely to lead to lower amounts
of particulate collected. For that reason, the
filter probe temperature was maintained at 100°F
-------
£ .006
(O
S-
es
(O
3
O
S-
(O
Q.
.005
,004
,003
85
95
TOO 105
Temperature at Filter
110
115
120
125
EFFECT OF FILTER TEMPERATURE CHANGE ON GRAMS/MILE
Figure. 9
-------
,01
to
E
-------
-61-
by water jacketing the probes. This temperature was at
least 20°F above the dew point of a run at 400 ft/min.
The temperature of the diluted exhaust at 400 ft/min,
using indoor diluent air, is generally above 100°F with
no external controls. Cooling was almost always necessary
to give an average of 100°F. Under conditions of the
Federal cycle cold start, the sample probes were warmed
to keep them at 100°F during the first part of the run.
Effect of Dilution Air Temperature - Table 16 contains a
tabulation of data generated to assess the effect of using
cooler outside air as a dilution for the exhaust rather than
warmer inside air. During the course of these runs, the ambient
air being used for dilution was about 40°F, or roughly 35-40°F
cooler than internal air.
Under comparable conditions of flow rate, the grams/mile of par-
ticulate mass collected showed a 23 percent and a 17 percent
increase using cool air compared to warm air. This effect was
expected since previous experience had shown that some percentage
of the particulate component was due to condensed material, which
increased in amount as temperatures were lowered.
Since the ambient air was uncontrollable and since the tempera-
ture of the total diluted exhaust stream was difficult to
change except by changing the temperature of the diluent, all
additional work was done using room air for a diluent. However,
since the temperature effect is real, any condition of cold
starting can be expected to give larger amounts of particulate
than would the corresponding hot start, since the total
diluted exhaust stream is lower in temperature until the engine
reaches operating temperature.
-------
TABLE 16
EFFECT OF DILUTION AIR TEMPERATURE ON GRAMS/MILE
Ft/Min
Flow in
Run Tube
109A
109B
110A
HOB
400
400
400
400
Corrected
Flow, cfm
479
509
506
477
Dilution Tube
Temperature
120°F
92 °F
95 °F
122°F
Grams
Collected
.0066
.0077
.0085
.0077
g/Mile
Collected
.0351
.0432
.0478
.0408
Dilution Air
Inside
Outside
Outside
Inside
to
I
All runs made on 1972 Pontiac 400 CID engine dynamometer
Non-leaded fuel, 60 mph steady-state, 3 hour runs
Average of three filters
-------
-63-
Conclusions
A lower diluted exhaust stream temperature results in
higher amounts of collected particulate. Attempts to
control this variable will be made by using only inside
air for a diluent,
Effect of Sample Probe Temperature - Table 17 contains the
results of several tests to determine the effect of varying
the temperature of the sample probe. Water jackets were
installed on the tubes used as sample probes allowing either
cooling or heating to determine any differences. Runs were
made at 300, 400, and 500 ft/min. The 400 ft/min run was not
good since the sample collection temperature dropped below the
dew point, giving meaningless results.
In comparing the effect of a drop in sample probe temperature
at constant flow through the dilution chamber, it was shown
that a cooler sample probe, resulting in cooler temperatures
at the filter, does cause an increase in collected particulate.
This experiment is very closely related to the determination
of the effect of filter temperature previously discussed, and
further verifies the conclusions drawn from those runs. The
data is plotted in Figure 11.
Although the increase in collected particulate continues to
temperatures below 100°F, this temperature was felt to be the
lower limit of confidence with respect to maintaining an
adequate spread between dew point and filter temperature.
Most future work was done with the sample lines jacketed and
controlled to about 100°F filter temperature.
Effect of Sample Probe Location - The Dow dilution tube is
23 feet long from where the exhaust enters the tube to where
the majority of the sampling has been done. It was felt that
there might be an effect on the amount of particulate collected
due to sample probe location.
-------
TABLE 17
EFFECT OF SAMPLE PROBE TEMPERATURE
SAMPLE ZONE
Ft/Min
Flow in
Tube
300
500
Sampl ing
Point
Temperature
132. 8°F
109. 4°F
Heated
Temp . @
Filter
125°F
111°F
g/Mile
Collected
.0049
.0056
Ambient
Temp . @
Filter
108°F
97°F
g/Mile
Collected
.0060
.0070
Cooled
Temp . @
Filter
82°F
81°F
g/Mile
Collected
.0064
.0070
Runs 113B and 113C
All runs made on 1972 Pontiac 400 CID engine dynamometer
Non-leaded fuel, 60 mph steady-state, 2 hour sample time
142 mm, 1 cfm filter
-------
0070
0065
tQ
i.
CD
o
s_
(O
a.
,0060
0055
,005
85
90
5CO
ft"
:**
i
1
&
TT
I
-------
-66-
Figure 7 is a schematic diagram of the dilution tube and
the three sample zones tested. In Table 18, the data for
each of the zones at a specific flow rate is tabulated.
In general, the effect of sample probe location is shown to
be slight; for example, Run 105 at 400 ft/min shows a range
of .0094 g to .0098 g of particulate collected across the
three sample zones. There was a slight temperature decrease
across the three zones which was felt to be more likely the
cause of the slight sample size increase than was any effect
due to location. Run 106 shows again a very slight decrease
in collected sample at essentially constant temperature. The
data is graphed in Figure 12.
Conclusions
There appeared to be no large effect in particulate
sample size due to sample probe location. In all
further work, samples will be taken only at Zone 3,
which is 23 feet from the point of entry of the
exhaust.
Effect of Face Velocity Through the Filter Media - Table 19
is a tabulation of the data collected using a 142 mm glass
fiber filter at varying flow rates. As is clearly shown,
the gross amount of raw sample collected increases as the
flow increases. However, when the data was calculated on a
grams/mile basis, the slower rate of collection gives much
higher numbers. Figures 13 and 14 show this graphically.
Although no attempt was made to keep the sample probe or
filter temperature constant, and although we have noted a
temperature effect on sample size in previous work, it was
felt that the effect noticed in this instance was much
greater than would be expected from the temperature differences
noted.
-------
TABLE 18
EFFECT OF SAMPLE PROBE LOCATION
Sample Zone 1
Sample Zone 2
Sample Zone 3
Run No.
107
105
102
106
Flow
Rate
300
400
400
500
Temp. Before
Filter
122°F
106 °F
86°F
100°F
Grams
Collected
.0087
.0094
.0075
.0065
Temp . Before
Filter
lll'F
102°F
100°F
100°F
Grams
Collected
.0085
.0095
.0073
.0065
Temp. Before
Filter
113°F
100°F
104°F
100°F
Grams
Collected
.0092
.0098
.0086
.0061
i
en
All runs made on 1972 Pontiac 400 CID engine dynamometer
Non-leaded fuel, 60 mph steady-state, sample time 4 hours
142 mm, 1 cfm filter
-------
TABLE 19
EFFECT OF FLOW RATES THROUGH FILTER MEDIA
Run
96
96
96
96
98
98
98
98
Grams
Collected
.0045
.0053
.0057
.0066
.0051
.0051
.0054
.0072
Flow at
Filter
0.5
1
2
4
0.5
1
2
4
cfm
cfm
cfm
cfm
cfm
cfm
cfm
cfm
Temperature
at Filter
81°F
104°F
115°F
125°F
91°F
100°F
117°F
126°F
Grams Equilibrated
to 1 cfm
.0090
.0053
.0028
.0016
.0102
.0051
.0027
.0018
Grams/Mile
.0394
.0232
.0125
.0072
.0446
.0223
.0118
.0079
00
I
All runs on 1971 Chevrolet 350 CID chassis dynamometer
Non-leaded fuel, 60 mph steady-state, 2 hours
142 mm glass filter
-------
O)
O
(1)
O
C_5
O)
3
O
s-
(O
o.
,006
I
CTi
vo
I
Sample Location
EFFECT OF SAMPLE PROBE LOCATION
Figure 12
-------
-70-
.0400
O)
ID
S-
O)
+J
IO
3
O
.0300
S-
(C
0.
.0200
0100
2 3
Flow Past Filter, cfm
-------
-71-
040
to
i.
CD
OJ
to
o
S-
ta
Q.
,030
,020
,010
234
Flow Past Filter, cfm
-------
-72-
Table 20 is a measure of the effect of face velocity through
the filter obtained by varying filter diameter.
The collection of particulate sample at 1 cfm through the 142
mm filter was felt to be the best operating rate since the
additional amount of sample collected at the higher rates was
offset by the difficulty of maintaining these flow rates at a
temperature near 100°F. In addition, when calculated on a
grams/mile basis, the higher flow rates show a much lower number,
Conclusions
High face velocity through the filter media leads
to comparatively less sample collected. A rate
of 1 cfm through the 142 mm filter will be used
in future work.
Effect of Sample Collection Time - Table 21 is a tabulation
of several runs made to determine the effect of sample
collection time on the amount of particulate collected. As
would be expected, the longer collection times did result in
more sample collected. However, the rate of sample collection
was much higher in the initial few minutes of the collection
period than in the final few minutes. The data from Runs 99,
101, 104, and 108 (calculated in grams/mile) are presented
graphically in Figures 15, 16, 17, and 18.
The raw data from Run 108 shows about 22 percent of a 2-hour
sampling period being collected in the first 5 minutes. It
was obvious, therefore, that any attempt to attach quantitative
significance to the particulate mass grams/mile figure must be
done with extreme caution. If all sampling parameters are held
constant except one, comparative significance can be inferred
from a grams/mile calculation.
-------
TABLE 20
EFFECT OF FACE VELOCITY THROUGH FILTER
Filter Diameter
Flow Through
Filter
142 mm
47 mm
293 mm
1 cfm
1 cfm
1 cfn
Grams
Particulate
.0022
.0007
.0045
Face Velocity
7.346 ft/min
96.8 ft/min
0.5208 ft/min
Run made on 1971 Chevrolet, 350 CID chassis dynamometer
Non-leaded fuel, 60 mph steady-state, 3 hours
co
I
-------
-74-
TABLE 21
EFFECT OF SAMPLE COLLECTION TIME ON GRAMS/MILE
Run
99
99
99
99
99
99
101
101
101
101
104
104
104
104
104
104
104
104
108
108
108
108
108
108
108
Filter Time
0.5 hr
1.0 hr
1.5 hr
2.0 hr
1.0 hr
0.5 hr
0.5 hr
1.0 hr
1.5 hr
2.0 hr
5 min
10 min
20 min
30 min
5 min
10 min
20 min
30 min
5 min
10 min
20 min
30 min
1.0 hr
1.5 hr
2.0 hr
Grams
Collected
.0039
.0056
.0070
.0081
.0055
.0036
.0048
.0070
.0072
.0089
.0013
.0023
.0046
.0058
.0015
.0026
.0044
.0061
.0013
.0022
.0034
.0037
.0047
.0054
.0060
Grams/Mile
.0682
.0470
.0408
.0354
.0481
.0630
.0840
.0612
.0420
.0389
.1356
.1206
.1206
.1014
.1572
.1362
.1155
.1066
.1365
.1152
.0891
.0646
.0411
.0314
.0263
All runs on 1971 Chevrolet 350 CID chassis dynamometer
0.5 cc lead fuel except Run 108 which was non-leaded fuel
60 mph steady-state
142 mm, 1 cfm filter
-------
-75-
,0700
-------
-76-
0900
0800
Ol
1/1
E
a)
s_
03
O)
4->
(T3
0700
1 .0600
s-.
(O
Q-
0500
,0400
1 2
Collection Time, Hours
-------
-77-
(O
3
O
s_
(O
Q.
1600
1500
. L
^ . 1 400
1300
1200
1100
1000
r
N<>-
\
\ ; ;
J.MJH1
.11.5.
S/
f i
£1
LE FlOR
.4
; T
01
M
H-
1104 i
-t-t-t
10 15 20
Collection Time, Minutes
25
30
-------
-78-
.1400
,1200
O)
Z .1000
-------
-79-
In order to obtain sufficient raw sample for analysis,
future work will generally be done using a 2-hour steady-
state collection period, except where a Federal or other
cycle is noted.
Conclusions
The sample collection period has a definite effect
on sample size, with a large amount of the sample
being collected in the first few minutes. Grams/
mile figures of particulate mass are, therefore,
misleading unless used only in a comparative sense.
Effect of Test Mode - A series of tests were made to deter-
mine the amount of particulate collected during the 23 minute
Federal cycle versus the 41.4 minute modified Federal cycle.
In previous runs it was noted that similar weights were
obtained in both the 23 minute and the modified cycle.
Table 22 is a tabulation of the results and outline of the
procedure used to verify any differences. The tests were
run on the same day and with all variables essentially constant.
After a 23 minute cold start, 0.0014 g of particulate mass were
collected and after the additional run of 505 seconds, a total
of 0.0020 g were collected. This verified that somewhat more
sample was collected during the latter part of the modified run.
Tube air (filtered) was drawn through the dilution chamber
and the filter during part of the test to determine the
effect, if any, of the additional time of flow past the
filter surface. It appears that this effect was negligible.
-------
TABLE 22
EFFECT OF TEST MODE ON PARTICULATE COLLECTED
Procedure for Run 111
Federal Cycle Cold Start Modified
Filter #1
Filter #2
Filter #3
Filter #4
Cold start 23 min
4-
10 min tube air
4-
505 sec hot start
4-
Stop - weigh paper
4-
END
.0020 grams
Cold start 23 min
Filter OFF during
10 min shutdown
Filter ON again.
505 sec hot start
Stop - weigh paper
END
.0020 grams
Cold start 23 min
4-
10 min tube air
4-
Stop - weigh paper
4-
END
.0014 grams
Run made on 1971 Chevrolet 350 CID
chassis dynamometer
Non-leaded fuel
142 mm filter, 1 cfm
Cold start 23 min
4-
Stop - weigh paper
.0014 grams
Replace used paper into
filter. Using pre-
filtered air, draw 1 cfm
room air through filter
for 10 min
.0014 grams
Stop - re-weigh paper
If paper is not damaged
replace in filter.
Continue room air for
30 min
.0012 grams
Stop - re-weigh paper
i
oo
o
I
END
-------
-81-
B. TASK II
1. Introduction
The major objective in Task II was to establish the baseline
data for the Pontiac 400 CID engine using low lead (0.5 cc/gal)
and unleaded fuel.
The data for the non-leaded fuel runs are presented in
Table 23 and for leaded fuel in Table 24.
2. Conclusions
a. In general, the low lead fuel gave higher particulate
levels, in grams/mile of particulate mass, than the
non-leaded fuel.
b. Overall, the aldehyde content of the exhaust condensate
was not significantly different between the non-leaded
and low lead fuel.
c. The concentration of benzo-a-pyrene in the particulate
varied widely with engine conditions, but did not
appear to be significantly changed by use of leaded
versus unleaded fuel.
d. Air/fuel ratio had an effect on particulate with the
standard setting generally being lower than either rich
or lean settings.
e. The effect on particulate due to changes in spark
timing is sporadic and, therefore, inconclusive
based on this data.
f. In almost all cases, 30 mph at steady-state gave
50-100 percent higher particulate mass levels than
the corresponding 60 mph run.
-------
TABLE 23
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type: 1972 Pontiac 400 CID
Fuel Used: Indolene #15214 - No lead 91 octane
Converter Type: None
Run
No.
155A
128A
128B
158C
126A
158A
156A
130C
130A
130B
126B
126B
JL2 6B
126B
129A
129B
Air to Fuel
Range
L
L
L
S
S
S
S
S
S
S
S
S
S
S
R
R
Actual
17.1
15.0
16.7
14.7
14.7
14.5
15.5
15.1
15.1
15.1
-
""
-
13.7
12.7
Test Mode
30 mph CS
30 mph HS
60 mph HS
30 mph HS
30 mph HS
30 mph HS
30 mph CS
60 mph HS
60 mph HS
60 mph HS
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph HS
60 mph HS
Spark
Timing
Std
Std
Std
Adv
Std
Ret
Std
Adv
Std
Ret
Std
Std
Std
Std
Std
Std
Grams /Mile
Particulate Converter
T cfm Filter Temp. (°F)
.0757
.0338
.0218
.0284
.0238
.0260
.0445
.0420
.0167
.0117
.0209
.0209
.0209
.0209
.0387
.0255
Dilution
Tube
Temp. (°F)
93.0
9 3'. 2
147
91.4
95.0
91.0
87.8
140
143
149
91.4
91.4
136.4
Filter
Temp. (°F)
100-102
100-102
93-96
102-105
100-105
10'0-107
97-100
88-96
86-100
90-100
96-100.
99-100
102-104
ppm in
Exhaust Condensate
HCHO
613
300
400
660
210
410
591
1530
360
150
230
250
150
NH0
j
5
16
-
6
1
CO
_ NJ
13
10
10
12
6
14
Spark setting:
Adv = Std -10°
Ret = Std +10°
-------
-83-
Continuation of Table 23
ANALYSIS OF EXHAUST GAS
% by Volume
Parts per Million
Run
No.
155A
128A
128B
158C
126A
158A
156A
130C
130A
130B
126B
126B
126B
126B
129A
129B
C02
9.0
11.1
12.0
13.8
11.7
13.7
12.2
11.3
12.9
12.7
10.3
11.2
9.4
11.1
13.4
11.9
°2
10.35
7.0
5.7
3.2
4.5
3.4
5.2
6.5
3.5
4.3
7.6
6.1
9.3
6.0
2.9
3.5
N2
79.7
80.9
81.4
82.3
82.8
82.2
81.5
81.1
82.3
82.0
81.2
81.7
80.4
82.0
82.4
81.7
CO
.03
.03
.03
.03
.03
.03
.05
.17
.43
.03
.03
.03
.03
.03
.31
1.95
Total
H.C.
115
189
78
210
275
150
316
830
145
43
265
145
218
300
268
305
NO_
200
850
850
56
850
32
65
1300
850
900
-
-
-
-
400
360
NO
425
270
330
2100
260
650
960
650
575
340
-
-
-
-
1100
1500
NO
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-------
Continuation of Table 23
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
155A
128A
128B
158C
126A
158A
156A
130C
130A
130B
126B
126B
126B
126B
129A
129B
Trace Metals on Millipore Filter (% of particulate)
Fe Ni Cu Al
- - - -
_
.277 <.055 .250 .250
.538 <.153 .230 <.153
.416 <.166 .416 <.166
_
-
_
- - - -
_
_ _
Ca Mg Mn Cr Sn Zn Ti
_______
_______
1.75 .388 <.027 <.055 <.055 <.166 X.055
3.0 1.0 <.07 <.153 <.153 <.46 <.153
6.0 1.5 <.08 <.168 <.166 <.50 <.166
_______
_______
_"_____-
_______
_______
% Pb
Atomic
Absorp
0.7
0.7
1.2
0.9
0.3
1.6
1.9
.009
1.9
1.2
.05
-
-
-
1.2
1.1
% C on
Glass
Filter
47.6
104.5
58.8
72
76.9
65.7
78
59.4
53.4
151.0
-
-
-
50.2
65.4
ppm
BaP
22
146
14
33
112
69
148
192
12
94
71
-
-
-
113
161
1
00
1
-------
TABLE 24
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type: 1972 Pontiac 400 CID
Fuel Used: Indolene #15473 - 0.5 cc lead 91 octane
Converter Type: None
Run
No.
168A
135A
168B
170B
132A
170C
170A
176C
176A
176B
132B
132B
132B
132B
171A
171B
136B
Air
Range
L
L
L
S
S
S
S
S
S
S
S
S
S
S
R
R
R
to Fuel
Actual
16.0
16.0
16.0
15.8
14.6
15.9
15.8
14.6
15.0
14.8
-
-
- -
-
12.4
12.3
12.9
Test Mode
30 raph CS
30 mph CS
60 mph CS
30 mph HS
30 mph HS
30 mph HS
30 mph CS
60 mph HS
60 mph HS
60 mph HS
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph CS
30 mph HS
60 mph HS
Spark
Timing
Std
Std
Std
Adv
Std
Ret
Std
Adv
-Std
Ret
Std
Std
Std
Std
Std
Std
Std
Grams/Mile
Particulate Converter
1 cfm Filter Temp. (°F)
.0650
.0378
.0275
.0363
.0423
.0404
.0433
.0268
.0239
.0255
.0390
.0390
.0390
.0390
.0602
.0431
.0299
Dilution
Tube
Temp. (°F)
89.6
98.0
114.8
91.4
93.0
91.4
89.6
119
120
123
93-134
-
-
-
89.6
89.6
136
Filter
Temp. (°F)
97-100
100-
100-102
100-102
93-97
99-102
97-100
99-102
103-105
100-103
93-96
-
-
-
97-100
95-102
98-100
ppm in
Exhaust Condensate
HCHO
600
550
440
340
300
570
510
440
330
250
370
-
-
-
190
300
250
NH3
-
31
-
-
8
-
~ o>
en
-
.
13
.
-
-
-
-
-
Spark setting:
Adv = Std -10°
Ret = Std +10°
-------
-86-
Continuation of Table 24
ANALYSIS OF EXHAUST GAS
% by Volume
Parts per Million
Run
No.
168A
135A
168B
170B
132A
170C
170A
176C
176A
176B
132B
132B
132B
132B
171A
171B
136B
co2
12.2
11.8
12.8
13.0
12.7
12.9
12.7
13.1
13.0
13.4
12.2
11.8
12.5
11.8
9.8
10.1
13.0
°2
3.4
6.1
4.1
3.4
4.3
3.6
3.8
2.8
3.4
3.0
5.6
5.7
4.7
5.7
1.55
1.6
2.2
*2
82.6
81.2
81.8
82.0
81.7
81.8
81.7
82.4
82.0
82.1
81.4
81.2
81.6
81.2
79.8
80.4
82.2
CO
.64
.02
.31
.75
.17
.83
.95
.75
.61
.64
.03
.37
.25
.14
9.8
6.9
1.57
Total
H.C.
385
170
167
272
175
397
360
230
205
180
150
137
125
145
775
710
280
N02
40
500
40
33
550
33
33
23
2.3
13
-
-
-
-
7
15
1150
NO NO
1050
450
1400
505
380
1100
810
1800
1150
850
- -
- -
- -
- -
1050
1100
580
-------
Continuation of Table 24
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Trace Metals on Millipore Filter (%)
Run
No. Fe Ni Cu Al Ca Mg Mn Cr Sn Zn
168A - - - - --
135A __________
168B __________
170B - - --
132A - - - - - - - - --
170C __________
170A - - - - ' - - --
176C - - --
176A __________
176B - - - - - - . -
132B - - - -.-
132B - - - . -
132B - - - - - - - - - ' -
132B _--_-_____
171A - - - - -- - --
171B _ -_
136B
% Pb
Atomic
Ti Absorp
6.4
4.7
10.4
4.2
7.2
5.6
6.7
-
^ ^
14.4
-
'
-
8.4
12.1
12.8
% C on
Glass
Filter
68.4
-
41.0
66.1
-
57.4
57.0
m
-
-
-
-
58.0
51.0
32.1
ppm
BaP
424
63
<8
<13
<11
19
30
127
98 I
00
88 ^J
21
-
-
-
615
225
62
-------
-88-
g. Standard spark setting gave a higher percentage
of larger particles than either advanced or retarded
spark.
3. Discussion
This baseline data will have absolute significance only
when used as a comparison to the same engine equipped with
various converters. However, several interesting points
can be noted when looking at the differences in particulate
as a function of engine operating conditions.
As noted in the conclusions, low lead fuel seems to give
higher particulates than non-leaded. This was also the
conclusion of the work done under Contract CPA-22-69-145,
and reported by Moran et al. This data is presented
graphically in Figures 19, 20, and 21. The amounts of aldehyde
and benzo-a-pyrene did not appear to be significantly changed
by the use of 0.5 cc leaded fuel. This was not unexpected.
A more meaningful comparison will be the levels of these
compounds after the use of a converter. This data will be
discussed in a later section.
The air/fuel ratio of the engine had an effect on particulate
as shown in Figure 19. The important point to note here is
that the standard air/fuel ratio did seem to give the lowest
particulate. Figures 19, 20, and 21 show the effect of 60 mph
versus 30 mph. In almost every instance, regardless of the
air/fuel ratio or spark setting, the equivalent 30 mph run was
higher in particulate. This may in part be explained by the
difference in dilution tube temperature. Although the filter
temperature as shown in Tables 23 and 24 was held within 10°F
of 100°F, the dilution tube itself was 40-50°F higher at 60 mph
than at 30 mph.
-------
-89-
,04
o>
ISI
E
03
3
CJ
ra
a.
Lean
Standard
Rich
-------
-90-
E
10
i.
(O
s-
-------
-91-
,05
.04
i
1
-
-H
MM
-
j
^M
-
,
t
L
|
mtm
.
^^tf
!
!
1«HI
.__.
PA
S_P
BA
RT
EE
S£
~
1C
D
LI
j
!
.
i
!
MM
i
i
i
bi
AN
fop
AT
D
1
!
.....
:
ni
Fi
AS
R/
qu
A
:U
._!_
]^
1
re
F
IL
2
JN
_R
I
!
~r-
i^
i
i
1
:T|IO
msL
\
[
N
_EJ
i .
i
^H
)F
IR
;
;
j
*^m
.....
__.
i
'
i
!
i
i
i
j
i ! :
! :
| '
i
;
j
1 i :
!
1 ;
i i
i ;
'
;
(O
i.
CD
to
3
,03
." .02
.01
O O
CO to
Std
O O
CO tO
Lean
O O
co to
Rich
O O
co to
Std
O O
CO to
Lean
O O
co to
Rich
Unleaded Fuel
Leaded Fuel
-------
-92-
The mass medium equivalent diameter (MMED) can be determined from
the mass distribution charts in Figures 22-27. Mass medium
equivalent diameter is the diameter of the particle, in microns,
of which 50 percent are smaller and 50 percent are larger.
The mass distribution plots show the percent of particles
less than a given diameter.
In comparing the baseline data with MMED, it is apparent that
the standard spark setting gives larger particles than either
advanced or retarded spark setting. The MMED for standard
spark was approximately 0.6 microns, approximately 0.2 microns
for retarded spark, and less than 0.1 for advanced spark.
For leaded fuel, at standard spark and standard A/F, the MMED
was less than 0.5 microns, which was smaller than that seen
with non-leaded fuel.
C. TASK III
1. Introduction
The objective in Task III was to evaluate various emission
control devices with respect to their effect on non-regulated
emissions. The devices tested under this portion of the
contract include:
Three oxidation catalysts
One reduction catalyst
One exhaust gas recirculation system
Initially, work was done using both non-leaded and low lead
(0.5 cc/gal) fuel. The leaded fuel itself caused increases in
the amount of particulate and, after testing one catalyst on
leaded fuel, the rest were run only on non-leaded fuel.
The devices were obtained from either the manufacturer or an
automobile company under a secrecy agreement, to protect any
proprietary rights involved. Consequently, the data on the
-------
PROBABILITY 46 SO43
X 2 LOG CYCLES HADE IN U.S.A. .
KEUFFELa ESSCR CO.
99.99
0.2 0.1 0.05
0.01
Figure 22
MASS DISTRIBUTION
Baseline, Std Spark, Std A/F
Run No. 130A
Total in Particles of Diameter
-------
46 8043
MADf IN U.S.».
KEUFFEL & CSSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
0.2 O.J 0.05
0.01
O
s_
U
-------
46 8043
NADF IN U.S.A.
KEUFFEL a ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
0.01
c
o
s-
o
0)
!->
0)
E
10
T-
o
01
o
Figure 24
MASS DISTRIBUTION
Baseline Adv Spark, Std A/F
Run No. 130C
Total in Particles of Diameter
-------
46 SO43
NAOF IN U.S.A.
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
0.01
V)
E
O
s_
O
(V
4->
CD
E
(O
r
O
V
O
S-
; i -; «.- ri
Total in Particles of Diameter
-------
PROBABILITY
X 2 LOG CYCLES
46 8O43
MADE IN U. i.A.
KEUFPEL » ESSER CO.
99.99
0.01
to
C
o
J-
o
o>
E
(0
o
r"
4->
S-
Q.
Figure 26
MASS DISTRIBUTION
Baseline Adv Spark, Std A/F ^
0.5 cc Lead
Run No. 134C (176C)
Total in Particles of Diameter
-------
46 8043
MAOf IN U.S. A.
KEUPFEL a ESSER CO.
PROBABIUITY
X 2 LOG CYCLES
10-
9.
8.
7.
6..
5.
99.99
99.9 99.8
99 98
95 90
80 70
0.01
Total in Particles of Diameter
-------
-99-
devices are reported with the only reference to the device being
a code letter. The code letters and general description of the
devices are as follows:
A. Monolith, noble metal oxidation catalyst
B. Beaded, base metal oxidation catalyst
C. Beaded, base metal oxidation catalyst
D. Monolith reduction catalyst
E. Exhaust gas recirculation system
The data from these runs is shown in Tables 25 -31.
2. Conclusions
a. The use of three different oxidation catalysts on an
engine stand, with non-leaded fuel, increased the amounts
of particulate collected at 60 mph by a factor of 2-5,
compared to a baseline run, except under rich air/fuel
ratios. Two catalysts did not generally increase the
particulate collected at 30 mph, or under cyclic condi-
tions. The total particulate collected from the control
devices was less than normally found when using 3 cc
leaded fuel.
b. The three oxidation catalysts significantly lowered
the emission of aldehydes, as collected in the
condensate.
c. There was no evidence in the particulate of catalyst
degradation being the cause of the increase in
particulate mass.
d. An increase in particulate comparing 30 mph to 60 mph
was noted during the baseline runs. This appeared to
be reversed when running at standard conditions with
two of the three oxidation catalysts.
-------
TABLE 25
ENGINE DYNAMOMETER TEST OP CONVERTER EQUIPPED ENGINES
Engine Types 1972 Pontiac 400 CID
Fuel Used: Indolene #1.5214
Converter Type: "A" with 1975 Hardware, Monolith, Noble Metal
Run
No.
142B
142A
143B
140C
140A
140B
143A
143A
143A
143A
141B
141A
Air
Range
L
L
S
S
S
S
S
S
S
S
R
R
to Fuel
Actual
15.3
16.8
14.1
14.2
14.8
14.9
-
-
-
-
13.1
12.4
Test Mode
30 mph
60 mph
30 mph
60 mph
60 mph
60 mph
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph
60 mph
Spark
Timing
Std
Std
Std
Adv
Std
Ret
Std
Std
Std
Std
Std
Std
Grams/Mile
Particulate
1 cfm Filter
.1078
.1193
.0968
.0866
.0905
.0733
.0913
.0913
.0913
.0913
.0092
.0207
Converter
Temp. (°F)
825
1160
865
1210
1205
1240
820
940
1220
825
910
1180
Dilution
Tube
Temp. (°F)
87.7
127.4
91.4
131.0
131.0
132.8
_
91.4
131
ppm in
Filter
Temp. (°F)
96-100
100-102
98-100
100-102
100-102
102-103
_
100-102
100-102
Exhaust
HCHO
26
65
14
35
12
7.5
29
9.3
71
Condensate
NH,
-
'
7
4
2
M
.
-
M
O
O
1
Spark setting:
Adv = Std -10°
Ret = Std +10°
-------
Continuation of Table 25
ANALYSIS OF EXHAUST GAS
Before Converter
After Converter
% by Volume
Run
No.
142B
14 2 A
143B
140C
140A
140B
143 A
143A
14 3 A
143A
141B
141A
co2
11.7
9.9
6.9
12.0
12.3
12.6
8.4
11.1
11.8
10.8
12.3
12.3
°2
6.2
9.0
13.9
4.6
4.0
3.7
11.5
7.1
5.6
7.7
4.1
3.5
N2
81.2
80.1
78.3
81.9
82.4
82.2
. 79.1
80.8
81.3
80.6
81.7
81.4
CO
.03
.07
.07
.51
.37
.39
.08
.04
.34
.11
.82
2.2
Parts per
Total
H.C. N00
165
255
215
215
100
61
200
140
130
230
303
310
Million
NO NO^ CO.,
11.8
11.0
10.5
13.7
13.4
12.1
10.1
10.0
12.1
9.7
- . 13.7
12.3
% by Volume
°2
6.1
7.3
8.1
2.9
3.3
5.0
8.9
9.1
5.7
9.2
2.3
2.3
N2
81.2
80.8
80.5
82.5
82.4
82.0
80.1
80.0
81.4
80.2
81.6
81.9
CO
.03
.03
.03
.03
.03
.03
.03
.03
.03
.03
.21
2.3
Parts per Million
Total
H«C.
17
35
27
23
14
7
30
14
25
39
140
260
NO2
275
430
450
570
330
285
m
-
-
200
390
NO
590
600
570
850
1250
850
-
' -
320
1650
NO
-
-
-
-
-
1
(-
~ o
M
~ 1
-
-
-
-------
Continuation of Table 25
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
142B
14 2A
143B
14 OC
140A
140B
143A
143A
143A
14 3 A
141B
14 1A
Trace Metals on Millipore Filter (%)
Fe Ni Cu
_ _ _
_
_
_
.06 .02 .09
.09 .02 0.1
.08 .02 0.1
_
_
: : :
_
Al
-
-
-
-
.05
.04
.04
-
-
_
Ca
-
-
-
-
.60
.77
.83
-
-
_
Mg
-
-
-
.10
.16
.14
-
-
_
Mn Cr Sn
_
_
_
_
<.01 <.02 <.02
<.01 <.02 <.02
<.01 <.02 <.02
_
_
m. _
_ _ _
Zn
-
-
-
-
<.05
<.05
<.05
-
-
^
Ti
-
-
-
-
<.02
<.02
<.02
-
-
^
-
% Pb
Atomic
Absorp
0.2
0.2
0.1
0.1
0.2
<0.1
0.2
-
-
0.2
0.9
% C on
Glass
Filter
<1
3.1
<1
3.2
0.6
0.4
1.5
-
-
43.2
60.9
ppm
BaP
<5
21
35
6
17
<6
8
-
-
<53
17
1
o
NJ
-------
TABLE 26
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type: 1972 Pontiac 400 CID
Fuel Used: Indolene #15473, 0.5 cc lead 91 octane
Converter Type: "A" with 1975 Hardware, Monolith, Noble Metal
Run
No.
167A
15 IB
151A
166C
146A
166B
166A
145C
14 5A
145B
14 6B
146B
146B
146B
165A
150A
150B
Air to Fuel
Range
L
L
L
S
S
S
S
S
S
S
S
S
S
S
R
R
R
Actual
16.0
15.4
15.6
15.1
14.5
15.1
15.2
15.5
15.6
15.6
-.
-
12.0
14.2
13.3
Test Mode
30 mph CS
30 mph HS
60 mph HS
30 mph HS
30 mph HS
30 mph HS
30 mph CS
60 mph HS
60 mph HS
60 mph HS
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph CS
30 mph HS
60 mph HS
Spark
Timing
Std
Std
Std
Adv
Std
Ret
Std
Adv
Std
Ret
Std
Std
Std
Std
Std
Std
Std
Grams/Mile
Particulate
. 1 cfm Filter
.0388
.0364
.0742
.0256
.0245
.0222
.0257
.1268
.1130
.1001
.0580
.0580
.0580
.0580
.0592
.0354
.0305
Converter
Temp. (°F)
910
960
1250
950
920
1000
1000
1200
1210
1260
800
1075
1240
870
1150
1070
1250
Dilution
Tube
Temp. (°F)
8.9.6
93.2
127.4
93.2
93.2
89.6
87.0
127
140
127
-
98.6
95.0
129
Filter
Temp. (°F)
99-102
100-104
100-102
98-102
98-102
99-102
99-100
100-102
100-104
95-97
-
102-104
100-105
100-102
ppm in
Exhaust Condensate
HCHO NH_3
180
100
118
110
100
66
170 -
97 6
61 2
30 5
-
100
26
72
<10
Spark setting:
Adv = Std -10°
Ret = Std +10°
-------
Continuation of Table 26
ANALYSIS OF EXHAUST GAS
Before Converter
After Converter
% by Volume
Run
No.
167A
151B
15 LA
165C
14 6A
166B
166A
145C
14 3 A
14 3B
14 ',B
14 -SB
14 oB
14 5B
165A
150A
150B
co2
12.7
9.6
10.9
12.7
11.5
13.1
12.5
12.2
12.0
11.1
10.8
11.9
11.4
11.0
11.4
8.0
10.8
°2
4.3
9.2
7.1
4.1
6.5
3.4
4.0
5.1
5.8
6.8
7.4
4.9
6.3
7.3
2.6
11.5
4.8
81.8
80.0
80.8
81.6
81.3
82.0
81.7
81.6
81.2
80.4
80.8
81.3
80.9
80.8
81.0
79.0
80.7
CO
.32
.34
.38
.52
.06
.64
.70
.20
.11
.16
.04
.95
.38
.06
3.9
.61
2.79
Parts per
Total
H.C. NO0
298
260
178
375
200
225
315
170
140
175
200
288
190
165
496
280
425
Million
NO NO.. CO.,
13.2
9.8
8.9
13.6
11.6
13.9
14.0
11.5
10.6
11.2
_
_
_
- -
13.2
9.9
10.9
% by
°2
4.15
9.2
10.4 -
3.3
6.3
2.8
2.7
6.5
5.9
6.9
-
-
-
-
0.6
9.2
6.4
Volume
81.7
80.1
79.5
82.0
81.1
82.3
82.3
81.0
81.2
80.9
-
-
-
-
82.6
80.0
80.9
Parts per Million
CO
.03
.03
.03
.03
.03
.03
.03
.03
.29
.03
-
-
-
-
2.78
.06
.86
Total
H~C.
75
84
40
100
75
65
120
37
20
8
-
-
-
_
480
145
220
NO
40
' 200
650
33
240
33
10
625
555
250
-
-
-
-
40
240
460
NO
1100
500
520
1100
575
600
850
975
850
690
-
-
-
-
900
450
930
N0x
-
-
-
-
-
I
~ o
£».
1
-
-
-
-
-
-
-
-
-------
Continuation of Table 26
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
No. Fe Ni Cu
167A -
151B -
151A -
166C -
146A -
166B
166A -
145C .036 <.007 .056
145A .075 <.008 .058
145B 2.9 <.05 3.2
146B _ _ -
146B
146B - - - .
146B -
165A -
150A -
150B -
Trace Metals on Millipore Filter (%) A* .
Al Ca Mg Mn Cr Sn Zn Ti Absorp
___----- 1.4
-0.9
-------- 0.9
-- -2.5
--------1.8
- .- _ _ _ 2.3
----- -- -2.5
.020 .40 .040 X.004 <.007 <.007 <.020 <.007 0.7
.026 .41 .052 <.004 .015 <.008 <.021 <.008 0.7
.90 .21 2.2 <.25 .53 .50 <1.5 <.50 0.7
_'_ - - - - - -1.5
_-
__ _ _ _ _ _._
_-_ _ _._
------- - 11.9
-.-
'
% C on
Glass
Filter
7.3
12.1
12.3
12.0
5.3
8.3
20
3.9
3.1
3.9
3.4
-
-
41.2
20.4
20.2
ppm
BaP
90
19
22
<19
-
<21
62
«4
<4
-
<6
-
-
-
374
<13
24
cn
-------
TABLE 27
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
1972 Pontiac 400 CID
Indolene f15214, No lead.91 octane
Converter Type: "B" with 1975 Hardware, Beaded, Base Metal
Engine Type:
Fuel Used:
Run
No.
183A
182B
181A
181C
182A
182A
182A
182A
18 QA
179A
179B
Air to Fuel
Range
L
S
S
S
S
S
S
S
R
R
R
Actual
16.6
15.4
14.9
15.2
-
"
13.7
13.8
Test Mode
60 mph
60 mph
30 mph
60 mph
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph
60 mph
60 mph
Spark
Timing
Std
Ret
Std
Std
Std
Std
Std
Std
Std
Std
Std
Grams/Mile
Particulate
1 cfm Filter
.1165
.1134
.0048
.1052
.0256
.0256
.0256
.0256
Nil
.0522
.0183
Converter
Temp. (°F)
1140
1200
795
1235
825
1100
1190
975
950
1290
1235
Dilution
Tube
Temp. (°F)
129
129
92
131
112
89
130
122
Filter-
Temp . ( °F )
102-105
104-107
100-104
103-106
99-100
99-103
104-107
98-100
ppm in
Exhaust Condensate
HCHO
3.8
2.1
19
5.9
6.5
6.0
3.5
0.7
NH3
1.0
3.0
36
7
-
10.0
87
35
1700
1
h->
o
-------
Continuation of Table 27
ANALYSIS OF EXHAUST GAS
Before Converter
After Converter
Run
No.
183A
182B
18 1A
18 1C
182A
182A
182A
182A
18 OA
179A
179B
CO
11.2
12.5
12.0 .
12.7
12.5
12.7
12.5
12.3
14.5
12.9
14.4
% by
°2
7.0
4.7
5.7
3.8
4.8
4.7
5.8 '
5.1
0.9
3.4
1.1
Volume
N2
80.9
81.7
81.4
82.0
81.7
81.5
81.6
81.6
82.9
81.9
82.7
CO
.03
.03
.03
.27
.08
.15
.03
.06
.79
.03
.03
Parts per
Total
H.C. NO0
40
25
150
40
180
125
65
160
350
50
220
Million
NO NO,, CO,
11.4
12.8
12.4
13.1
13.0
13.1
13.0
13.0
14.9
14.8
15.1
% by
°2
6.7
4.3
5.2
3.3
4.3
4.0
4.2
4.1
0.4
2.5
0.5
Volume
N2
80.9
81.9
81.5
82.2
81.8
81.4
81.8
81.9
83.2
81.6
82.8
Parts per Million
CO
.03
.03
.03
.03
.03
.03
.03
.03
.55
.03
.40
Tptal
. H.C.
5
5
40
5
35
25
10
10
150
8
90
NO
. 85
55
-
65
-
65
75
72
7
37
17
NO
500
975
-
1600
-
920
1800
1000
930
1500
1600
X
-
-
-
-
-
I
I i
_ o
I
-
" -
-------
Continuation of Table 27
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
No . Fe Ni
183A
182B
181A
181C .21 .016
182A
182A
182A
18 2A
180A
179A
179B
Trace Metals
Cu Al Ca
_ _ _
_
_
.050 .240 .240
_
_
_
_
_
on Millipore Filter (%)
Mcj Mn Cr Sn Zn
_
- - - - -
_
.055 <.008 <.008 <.008 .058
_
_
_
-----
_
% Pb
Ti Absorp
-
-
-
<.008
-
-
-
-
-
% C on
Glass
Filter
5.0
5.1
7.9
2.9
48.5
-
-
-
;::
ppm
BaP
4
24
145
10
-
-
-
-
11 t-
o
39 «,
-------
TABLE 28
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type: 1972 Pontiac 400 CID
Fuel Used: Indolene 115214, No lead 91 octane
Converter
Run
No.
184A
185A
186D
18CA
185D
185B
185B
185B
187C
187B
187A
Air
Rang
L
S
S
S
S
S
S
S
R
R
R
Type: "C"
to Fuel
e Actual
16.6
14.9
15.0
14.7
14.5
14.7
14.8
14.8
13.4
13.4
13.3
with 1975 Hardware, Beaded, Base Metal
Test Mode
60 mph
60 mph
30 mph
60 mph
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 raph
60 mph
60 mph
Spark
Timing
Std
Ret
Std
Std
Std
Std
Std
Std
Std
Std
Std
Grams /Mile
Particulate
1 cfm Filter
.0052
.0533
Nil
.0554
.0383
.0383
.0383
.0383
.0117
.0360
.0171
Converter
Temp. (°F)
1140
1190
810
1150
750
975
1175
840
875
1440
1175
Dilution
Tube
Temp. («F)
134
126
92.0
125
115
95.0
142
115
Filter
Temp. (°F)
107-110
105-107
104-105
104-105
106-107
103-105
109-112
100-101
ppm in
Exhaust Condensate
HCIIO
0.8
0.3
28
5.3
4.8
1.5
0.2
0.2
NH
3.4
1.0
3.9
1.2
3.2
740.0
15.3
2180.0
ppm in
Exhaust Gas
HCHO
0.023
0.009
0.95
0.17
-
0.068
0.006
0.008
NH-
j
0.18
0.06
0.23
0.068
-
38.8
0.88
154.66
o
vo
I
-------
Continuation of Table 28
ANAL1
Before Converter
Run
No.
184A
185A
1EI6B
1H6A
1N5B
18 5B
1K5B
105B
187C
1H7B
1H7A
co2
12.0
12.8
12.2
12.6
12.4
11.8
12.9
12.2
14.3
13.1
14.2
% by
°2
5.5
4.4
5.2
4.8
4.8
5.9
4.1 '
5.2
1.0
2.3
0.7
Volume
N2
81.6
81.8
81.6
81.7
81.8
81.3
82.0
81.6
82.7
82.5
82.5
fSIS OF EXHA
Parts per Million
CO
.03
.03
.10
.03
.32
.11
.09
.03
.06
.95
.53
Total
H.C. NO.,
45
20
142
45
160
90
50
170
250
65
226
NO N0v
-
-
-
-
-
-
-
- -
-
- -
co2
12.5
13.2
12.5
13.1
12.7
12.3
13.1
12.4
15.0
14.9
15.1
After Converter
% by Volume
Parts per Million
°2
4.6
4.1
4.8
4.1
4.7
5.3
3.9
5.0
0.35
1.2
0.27
"2
81.6
81.8
81.8
81.9
81.7
81.5
82.1
81.5
83.1
83.0
83.0
CO
.03
.03
.03
.03
.03
.03
.03
.03
.41
.03
.20
93
33
7
65
8
58
65
33
15
10
8
NO
1500
1000
1100
1500
900
900
1850
1000
560
425
380
I
o
I
-------
Continuation of Table 28
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
No. Fe Ni
184A
185A
18 6B
186A .21 <.024
185B
185B
185B
185B
187C
187B
187A
Trace Metals on Millipore Filter (%)
Cu Al Ca Mg Mn Cr Sn
_______
_
_
.16 .093 .98 .25 <.024 .057 .045.
_
_ _
_______
_______
_______
_______
_ _ ' _ _ _ _ _
% Pb
Zn Ti Absorg.
_
-
_ _
.086 .024
_
_
_
_
_
_
_ _ _
% C on
Glass
Filter
57.9
7.6
_
11.1
31.6
-
-
-
91.0
20.0
43.4
ppm
BaP
36
13
-
16
<9
-
-
-
145
36
<13
-------
TABLE 29
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type: 1972 Pontiac 400 CID
Fuel Used: Indolene 115214, Ho lead 91 octane
Converter Type: NOx - "D", Monolith
Run
No.
201A
202A
202B
202C
202D
Spark
Air to
Fuel
Range Actual
R
R
R
R
R
setting:
13.9
13.9
13.8
13.8
13.8
Adv
Ret
Test Mode
60 mph
60 mph
60 mph
60 mph
30 mph
=> Std -10°
= Std +10°
Spark
Timing
Std
Std
Adv
Rtd
Std
Grams/Mile
Particulate
1 cfm Filter
.0052
.0126
.0129
.0036
.0031
Converter
Temp. (°F)
1090
1100
1105
-
"
Dilution
Tube
Temp. (°F)
125
111
113
-
"
Filter
Temp. (°F)
104-107
98-100
97-99
-
"
ppm in
Exhaust Condensate
HCHO NH3
33 310
32 320
13
13
"
ppm in
Exhaust Gas
HCHO NH
1.27 23.
1.33 23.
0.600
0.62
3
3
5
N)
I
-------
Continuation of Table 29
ANALYSIS OF EXHAUST GAS
Before Converter
After Converter
% by Volume
Run
No.
201A
202A
202B
13.7
14.0
13.3 ,
°2
0.9
0.73
0.75
82.8
83.2
83.8
CO
1.63
1.16
1.11
Parts per Million
Total
H . C .
250
240
250
7
16
10
NO NOX
1400
1600
1750
co2
13.8
14.1
13.7
% by Volume
5-2
0.7
0.56
0.6
83.0
83.6
84.1
CO .
1.5
0.72
0.67
Parts per Million
Total
H.C.
220
200
170
. 7
5
10
NO
1050
1150
1100
NO
-
-
u>
I
-------
TABLE 30
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type:
Fuel Used:
Converter Type
Run Air to
1972
Pontiac 400
Indolene #15214,
: EGR
Fuel
No. Range Actual
228G L .
228F S
228A S
228B S
228C S
22SC S -
228C S
228C S
228E R
228D R
16.7
15.0
15.7
15.6
15.3
15.3
15.3
15.3
13.5
13.5
- ON-
Test Mode
60 mph
60 mph
30 mph
60 mph
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph
60 mph
CID
No lead 91
Spark
Timing
Std
Ret
Std
Std
Std
Std
Std
Std
Std
Std
octane
Grams /Mile
Particulate
1 cfm Filter
.0124
.0071
.0314
.0097
.0183
-
-
-
.0227
.0079
Converter
Temp.(°F)
Dilution
Tube
Temp.(°F)
185
210
109
185
95-215
Filter
Temp.(°F)
100-120
125-130
92-98
110-118
100-125
85-90
225-245
ppm in
Exhaust Condensate
HCHO NH3
460.46 41.25
108.9 4.9
569.2 35.9
342.8 29.9
889.6
461.9
282.44
12.7
15.4
20.4
ppm in
Exhaust Gas
HCHO
21.0
5.2
31.4
7.5
41.9
27.1
3.3
0.4
3.5
1.2
2.4
3.4
Spark setting:
Adv = Std -10°
Ret = Std +10°
-------
-115-
Continuation of Table 30
EXHAUST GAS ANALYSIS
% by Volume Parts per Million
2
14.4
12.4-
10.9
11.4
11.3
11.7
11.9
11.4
11.6
12.4
0 2
7.3
4.8
6.8
6.0
6.6
5.6
5.4
6.5
4.9
2.7
N2
80.3
81.7
81.2
81.6
81.3
81.3
81.6
81.2
81.2
82.2
CO
.03
.03
.03
.03
.03
.03
.03
.03
1.20
1.67
Total
H.C.
80
40
206
110
160
135
75
135
175
75
NO,
65
45
50
65
48
48
72
48
30
32
NO NOX
600
650
100
1000
200
600
1220
275
150
940
-------
Continuation of Table 30
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
No. Fe Ni
228G
228G
228A
228B 14 <2
228C
228C
228C
228C
228E
228D
Trace Metals on Millipore Filter (%)
Cu Al Ca Mg Mn Cr Sn Zn
_._
________
_______
8 10 54 14 <2 <2 <2 14
_______
---
_______
_____ ___
_____ ___
________
% Pb
Ti Absorp
-
-
-
<2
-
-
-
-
T
-
% C on
Glass
Filter
36.7
49.8
55.6
47.3
50.8
-
-
-
46.0
46.4
ppm
BaP
21
33
120
28
68
-
-
230 £
120 °
-------
TABLE 31
ENGINE DYNAMOMETER TEST OF CONVERTER EQUIPPED ENGINES
Engine Type:
Fuel Used:
Converter Type:
1972 Pontiac 400 CID
Indolene #15214, Ho lead 91 octane
EGR - Off
Run
No.
229A
229D
229F
229C
229E
229E
229E
229E
229G
229B
Spark
Air to
Fuel
Range Actual
L
S
S
S
S
_
-
R
R
setting:
16.7
15.0
15.7
15.6
15.3
-
13.5
13.5
Adv
Ret
Test Mode
60 mph
60 mph
30 mph
60 mph
Dow Cycle 2
Dow Cycle 3
Dow Cycle 4
Dow Cycle 5
30 mph
60 mph
= Std -10°
= Std +10°
Spark
Timing
Std
Ret
Std
Std
Std
Std
Std
Std
Std
Std
Grams /Mile
Particulate Converter
T cfm Filter Temp. (°F)
.0019
.0087
.0021
.0105
-
.0208
-
.0231
.0056
Dilution
Tube
Temp. (°F)
185
200
105
195
100-230
110
225
Filter
Temp. (°F)
110-120
110-125
85-90
120-130
110-130
90-95
120-140
ppm in
Exhaust Condensate
HCHO
398'. 85
146.5
518.6
40.6
529.2
406.2
89.2
NH3
32.6
38.2
41.1
60.3
23.2
16.0
62,5
ppm 4-n
Exhaust Gas
HCHO NH3
12.8 1.8
6.4 2.9
20.8 2.9
18.9 4.9
-
-
5.1 6.3
-------
-118-
Continuation of Table 31
ANALYSIS OF EXHAUST GAS
% by Volume
Parts per Million
Run
No.
229A
229D
229F
229C
229E
229E
229E
229E
229G
229B
co2
11.3
17.2
11.5
11.9
11.7
11.7
12.2
11.6
11.0
12.9
°2
6.3
4.8
6.3
5.0
5.8
5.1
4.5
5.8
5.2
2.5
*2
81.6
81.6
81.3
81.3
81.4
82.2
82.3
81.6
81.9
81.8
CO
.03
.53
.03
.83
.03
.03
.03
.03
1.05
1.86
Total
H.C.
75
55
120
75
130
100
95
120
200
45
N09
^
100
72
50
80
55
55
120
55
10
30
NO NO
1400
1400
950
2000
850
1100
2500
1000
650
1100
-------
Continuation of Table 31
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
_ Trace Metals on Millipore Filter (%)
No. Fe Ni Cu Al Ca Mg Mn Cr Sn
229A - - - - - - ' -
229D ___-_--_-
229F _________
229C 12 <2 14 4 54 12 <2 <2 <2
229E _--__----
229E _--------
229E - - - --
229E - - --
229G ___-_--_-
229B _____----
% Pb
Zn Ti Absorp
_
- - -
_
10 <2
_ _ _
_
_
_
_
_ _ _
% C on
Glass
Filter
0.9
0.5
0.8
2.7
1.2
-
-
-
1.3
3.0
ppm
BaP
<13
<17
<17
21
39
-
-
-
48 !
<24
-------
-120-
e. The mass medium equivalent diameter was shifted
significantly toward smaller particles, when
compared to the baseline, for all of the devices
tested.
3. Discussion
The converters tested seemed to have a definite effect on
particulate in several ways. First, as mentioned in the
conclusions, the three oxidation catalysts all showed higher
grams/mile of particulate mass at 60 mph than did the baseline,
This is shown graphically in Figure 28. At standard air/fuel
ratio, the increase of particulate mass was significant. As
the engine was operated richer, however, the difference became
small enough to be considered less significant, although real.
The effect at 30 mph showed a reversal of the 60 mph effect
for two of the three oxidation catalysts. The two which show
a reduced particulate emission at 30 mph (Figure 29) are both
base metal beaded catalysts. The EGR system, (Converter E),
showed an increase at 30 mph, but was unchanged at 60 mph
compared to its own baseline. It is perhaps significant that
the same engine, when modified for EGR, showed a decrease in
particulate mass compared to the previous baseline.
The particulate mass collected during the Dow cycle (Table 2)
for the various converters showed little significant change,
although Converter A was quite high. This data is shown
graphically in Figure 30.
The increase noted in the particulate mass when using the
oxidation catalysts was not accounted for by anything which
was routinely measured as part of this contract.
-------
-121-
1200
100.0
0800
to
i-
CD
0)
-(->
(O
^
£
4->
s_
(O-
Q.
,0600
0400
,0200
Lean
Standard
Air/Fuel Ratio
Rich
-------
-122-
1200
"T
4-
1000
j I i !
NTT
0)
to
O)
0800
- .0600
-------
-123-
1200
1000
cu
ri .0800
(O
i.
- .0600
(O
(J
S-
ra
Q.
0400
,0200
i
1
I
_ , 1 '
~ 1
I [
i i
I j
I
__
-
i
!
i i
! |
I !
-
,
i j
^^
_.._
p;
C(
si
s\
-
JU
IN\
rfH
ar
m
fEfl
idf
idc
.111
J[
r(
re
A;
E.
/
<
j
!
!
]
i
^
1""" "
1
.._.
-+-
1
p I
p""*
--
I
--
...._.
.__..
E
_
i r
PC
AS
D(|
vn
rk
L
-
P
w
-UJ
F
cv
r
j_
Ul\
CL
He
i
i 1
! 1 i
i
;
! ;
;
«m
....
j. ^
^_
'
i
CT
E
1
10
N
" 1 ~
ii
0
'
OF
_..
i
j
I
: t
j
'
-4-
,
; , .
!
1 :
: i
1 . :
[ 1
r- 1
-
Baseline ,Con. " Con. Con. Con.
ABC E
'
~~
i
!
i
r'
,
i
I
i
1
|
! !
i
i i
i '
__j | ;_
!
__..
i
i ,
: !
i
Con.
E
Baseline
-------
X 2 LOG CYCLES BADE IN U.S.A.
KEUFFEL » ESSER CO.
99.99
99.9 99.8
0.01
CO
C
o
i.
EL.
Figure 31
MASS DISTRIBUTION
Converter A
Std Spark, Std A/F
Run No, 140A
Total in Particles of Diameter
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES K.OE IN u.i.A. .
KEUPFEL & ESSER CO.
99.99
0.2 0.1 0.05 0.01
to
c
o
j_
o
r
£
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES MADE IH u.>.>. .
KEUFFEL & ESSER CO.
99.99
99.9 99.8
99 98
95 90
80 70 60
0.01
Total in Particles of Diameter
-------
PROBABILITY
X 2 LOG CYCLES
46 8O43
HADE IN U.S.A.
KEUPFEL ft ESSER CO.
99.99
0.2 0.1 0.05 0.01
to
c
o
j-
o
O)
E
o
0)
o
S-
03
D.
Figure 34
MASS DISTRIBUTION
Converter A, 0.5 cc Lead Fue"E
Ret Spark, Std A/F
Run No. 145B
Total in Particles of Diameter
-------
46 8O43
MADE IN U. S. A.
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
99 98
0.2 0.1 0.05 0.01
to
E
O
j_
o
i
(O
r
O
(U
O
r
4J
S-
Figure 35
MASS DISTRIBUTION
Converter A, 0.5 cc Lead Fuel iT^l^i
Adv Spark, Std A/F
Run No. 145C
Total in Particles of Diameter
-------
PROBABILITY 46 6O43
X 2 LOG CYCLES HADE IH U.S.A. .
KEUFFEL ft ESSER CO.
99.99
0.2 0.1 0.05 0.01
J_
U
(U
E
(O
T-
o
0)
U
i.
to
CL.
Figure 36
MASS DISTRIBUTION
Converter B
Std Spark, Std A/F
Run No. 181C
Total in Particles of Diameter
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES M.DI IN U.S.A. .
KEUPFEL a ERSER CO,
99.99
99.9 99.8
0.2 O.I 0.05
0.01
Figure 37
MASS DISTRIBUTION
Converter C
Std Spark, Std A/F
186A
Total in Particles of Diameter
-------
46 8O43
HIDE IN U.S.A.
KEUFFEL. A ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
0.2 0.1 0.05 0.01
1/1
c
o
o
i"
E
s_
a>
-!->
?
o
Figure 38
MASS DISTRIBUTION
Baseline Converter E
Std Spark, Std A/F
Run No. 229C
Total in Particles of Diameter
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES »DE IN u.i... .
KEUFFEL & ESSER CO.
99.99
0.2 0.1 0.05 0.01
O
J-
u
O
0)
4J
(V
E
-------
-133-
The data for Converter D, a reduction catalyst (Table 29),
was not plotted or included in the discussions about the other
converters, since the conditions under which this converter
could be run were so limited. Several runs were aborted because
the air/fuel ratios were not in a range which would allow any
reduction in NO . The runs reported show only a minor reduction
X
in NO , but a noticeable increase in particulate mass. This
X
increase was attributed in part to the rich air/fuel ratio
necessary for operation of the converter. This device was not
run in conjunction with an oxidation catalyst.
A significant decrease in the MMED of the particulate mass col-
ted WTS noted in the r:ase of all t' '-f-o ox:i Lit don ta 1 y-7t :
and EGR. The mass size distributions for various runs are
plotted in Figure 31-39. Table 32 is a comparison of the dis-
tribution to the baseline runs made in Task II (Figures 22-27).
As is shown, the percent of the particulate less than 0.55
microns and 0.97 microns increased substantially with the addition
of a catalytic device. There was also a significant increase with
the addition of 0.5 cc lead. It is interesting to note that Con-
verter A plus leaded fuel was higher than 0.5 lead fuel alone.
The decrease in the particulate size noted when comparing the
Converter E baseline to the original baseline is partially
explained by the fact that the intake system and carburetor
of the engine was modified to take the EGR (Converter E).
Thus, the two baselines are not directly comparable. The
addition of EGR to the system further decreased the overall
particulate size.
-------
TABLE 32
MASS MEDIUM EQUIVALENT DIAMETER
FOR BASELINE AND CONVERTERS A, B, C, E
Conditions: 60 mph
Standard Spark
Standard Air/Fuel Ratio
Baseline
Converter
% Particle
<0055 Micron
% Particle
<0.97 Micron
No
Lead 0.5 cc Lead
45 73
A A w/0 . 5 cc Lead
88 89
B
94
C
80
E
85
E Baseline
73
i
i '
U)
i
60
87
89
92
97 88 88
82
-------
-135-
D. TASK IV
1. Introduction
The objective in Task IV was to evaluate the particulate
emission levels of various vehicles equipped with various
control devices. Table 33 is a description of the vehicles
tested and the number of runs on each one.
2. Conclusions
a. The vehicles on which mileage accumulation tests
were made exhibited a large degree of fluctuation
with respect to grams/mile of particulate mass as
a function of mileage. No clear trends have been
established.
b. The precision of measuring particulate mass from
a vehicle exhaust was substantially lower than
that of measuring an engine stand run, due to the
variations in driving conditions prior to testing.
c. In general, the particulate matter which exhibited
higher percentages of carbon also exhibited higher
parts per million of benzo-a-pyrene.
d. The mass medium equivalent diameter became larger
with mileage for two of the three mileage
accumulation cars, while decreasing for the other.
e. The mass medium equivalent diameter for the device
equipped vehicles in general correlates well with
the numbers obtained during the engine stand runs
even though the overall mass of the particulate
changed.
3. Discussion
The raw data for the vehicles described in Table 33 are presented
in Tables 34-43. The mass distribution plots corresponding to
-------
TABLE 33
VEHICLES TESTED AND NUMBER OF RUNS
Vehicle
Series
of Runs
1970 Chevrolet 350 CID 1
1971 Chevrolet 350 CID 3
1971 Chevrolet 350 CID 3
1972 Pontiac 400 CID 4
1972 Mercedes Benz Diesel 2
1971 Oldsmobile 350 CID 1
1972 Chevrolet 400 CID 2
Mail Jeep, Ford 1
1971 Pontiac 1
1970 Chevrolet 1
Control Device
Vehicle ID
Base metal, beaded ES 60311
Noble metal, monolith 61314
Base metal, beaded,
EGR
Base metal, beaded
None
Not known
Particulate trap
Stratified charge
Questor converter
61329
2477
EPA supplied
EPA supplied
PPG
EPA 801692
EPA supplied
Thermal reactor, EGR, Dupont
cyclone collector
Controlled Emissions
HC, CO
HC, CO
HC, CO, N0x
HC, CO
Particulate
HC, CO, N0x
HC, CO, NC)
HC, CO, NO , Particulate
X
-------
-137-
the 60 mph steady-state runs in Tables 34-43 are presented in
Figures 40-57, and follow the appropriate table. Data for the
vehicles as a function of mileage is plotted in Figures 58-62.
Many possible conclusions can be drawn from the large volume
of data generated on the various vehicles tested. Several
which are thought to be significant will be discussed.
First, it does not appear that any of the mileage accumulation
vehicles showed any marked trend toward higher particulate mass
levels with time. Car number 61329, a 1972 Chevrolet, increased
particulate mass somewhat as measured during a federal cycle hot
start, but showed sporadic mass collection rates at the 60 mph
steady-state run. Two other cars, the 400 CID Pontiac #2477
and another 1972 Chevrolet #61314, actually decreased slightly
with time. This data is plotted in Figures 58-62.
The Pontiac #2477 showed an actual decrease in particulate mass
collected when the control device was installed (Figures 60 and
61). This observation was in contrast to the reported increase
in particulate mass when oxidation catalysts were installed on
an engine stand. An explanation of part of the effect might be
that the exhaust system of the vehicle acts as a particle trap
in the early stages. However, the 60 mph steady-state results
for the Pontiac #2477 showed a general reduction in particulate
mass with mileage, out to about 12,000 miles, and the particulate
mass collected on the 4 cfm filter stayed essentially constant
after installation of the catalyst.
The effect of mileage accumulation on the mass medium equivalent
diameter was somewhat inconclusive, since two of the vehicles
(Car 61314 and the Pontiac) exhibited increased particulate size
with mileage, while the other (Car 61329) showed a particulate
size decrease.
-------
TABLE 34
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
ES 60311
1970 Chevrolet 350 CID
Non-noble Metal - Pelleted
Grams/Mile Particulate
Run
No.
68A
68B
68C
Car
Miles
80,000
80,000
80,000
Converter
Miles
50,000
50,000
50,000
Test Mode
FC CS
FC HS
60 mph
Andersen
Sampler
-
-
Millipore
Filter
-
-
-
Andersen +
Millipore
.
-
-
Glass 'Filter
1 cfm 4 cfm
.0242
.0016
.0174
u>
00
I
-------
Continuation of Table 34
EXHAUST GAS ANALYSIS
% by Volume
Parts per Million
Run
No.
68A
68 B
68C
CO
-
12.6
°2
-
3.9
N,
-
82.4
CO
-
-
.17
Total
H.C.
-
'-
45
N02
-
-
32
NO
-
-
1450
NO
-
_
Exhaust Condensate (ppm)
HCHO * NH,
VD
I
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES HADE IN U.S.A. .
KEUFFEL a ESSER CO.
99.99
99.9 99.8
99
0.2 0.1 0.05 0.01
10
C
o
s_
o
i.
-------
TABLE 35
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type;
61314
1971 Chevrolet 350 CID
Noble Metal - Monolithic
Grams/Mile Particulate
Run
No.
69A
69B
69C
94A
9
-------
Continuation of Table 35
EXHAUST GAS ANALYSIS
Run
No.
69A
69B
69C
94A
94B
189A
189B
18 9C
18 9D
* by Volume Parts per Million
Total
COj 0, N0 CO B.C. N02 NO NO^
____ ____
____ ____
11.2 5.0 82.3 .03 4 50 1200
____ ____
____ __-_
____ ____
13.5 3.4 82.2 .03 10 8 1070
_.___ _ - __
____ - - - -
to
1
ppm in
Exhaust Condensate (ppm) Exhaust Gas
HCHO NH3 HCHO NHj
-
-
-
-
-
100 6.6
51 3.6 2.77 0.34
-
-
-------
Continuation of Table 35
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
69A
69B
69C
94A
94B
189A
189B
189C
189D
Fe Ni
-
-
.011 .008
-
6.3 <1.0
2.9 <.34
-
- -
Trace Metals on Millipore Filter (%)
Cu Al Ca Mg Mn Cr Sn Zn
_-_
_________
.006 .002 .040 .008 <.00.1 <.001 <.001. .003
_"_
_
8.8 2.3 41.0 8.5 <1.0 <1.0 <1.0 4.4
2.1 0.6 16.0 3.7 <.34 40 .34 1.6
-._
--------
% Pb
Ti Absorp
-
-
<.001
-
-
<1.0
<.34
-
- '-
% C on
Glass
Filter
-
-
-
-
-
61.4
13.2
-
ppm
BaP
-
-
<19
-
-
100
<15
- - 1
- it*
U)
1
-------
99.99
CO
£=
o
s_
(J
o»
E
(O
r
Q
0)
s_
(O
o.
PROBABILITY 46 8O43
X 2 LOG CYCLES -ADE l« U.I... .
KEUFFEL ft ESSER CO.
0.01
Figure 42
MASS DISTRIBUTION
Car 61314
Run No. 189B
11 ;: i MI HI ii! i
H' TiTT "i TtTTiitr it I'ltn
Particles of Diameter
-------
Car Number:
Vehicle Type:
Converter Type:
TABLE 36
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
61329
1971 Chevrolet 350 CID
Base Metal Beaded + EGR
Grams/Mile Particulate
Run
No.
138A
138B
138C
139A
139B
139C
204A
204B
204C
204D
204E
205A
205B
205C
205D
205E
205F
205G
205H
231A
231B
231C
231D
231E
231D
Car
Miles
6,000
II
II
6,700
It
II
11,300
II
II
It
II
II
It
II
11
II
II
"
II
16,659
II
II
"
n
n
Converter
Miles
2,500
n
it
3,200
n
n
7,800
II
II
II
II
II
II
II
II
II
II
II
II
13,159
"
n
"
ii
n
Test Mode
FC HS
FC HS
FC HS
FC Mod.
60 mph
FC CS
FC Mod.
60 mph
FC HS
FC HS
FC HS
FC Mod.
60 mph.
FC HS
FC HS
FC HS
FC HS
FC HS
60 mph
FC Mod.
60 mph
FC Mod.
FC HS
FC HS
FC HS
Andersen
Sampler
-
-
.1434
.0108
.1533
.0826
.0067
-
-
.1130
.0073
-
-
-
-
-
.0139
.0826
.0068
.0435
-
-
Millipore
Filter
-
-
-
.0130
.0165
.0133
.0087
.0202
-
-
.0174
.0368
-
'
-
-
-
.0032
.1130
.0547
.0522
-
-
-
Andersen +
Millipore
-
-
.1564
.0273
.1666
..0913
.0269
-
-
.1304 .
.0441
-
-
-
-
-
.0171
.1956
.0615
.0957
-
-
-
Glass
1 cfm
.0233
.0080
.0149
.0060
.0519
.0166
.0065
.0291
.0200
.0266
.0233
.0239
.0510
.0167
.0100
.0133
.0800
.0733
.0143
.0348
.0620
.0391
.0300
.0300
.0300
Filter
4 cfm
.0070
.0080
.0130
.0087
-.0528
.0183
.0054
.0291
.0133
.0147
.0125 ,
.0097 £
.0487 <*
.0067
.0091
.0108
.0358
.0358
.0073
.0152
.0629
.0163
.0125
.0142
.0117
-------
Continuation of Table 36
% by Volume
EXHAUST GAS ANALYSIS
Parts per Million
Run
No.
138A
138B
138C
139A
139B
139C
204A
204B
204C
204D
204E
205A
205B
205C
205D
205E
205F
205G
205H
231A
231B
231C
231D
231E
231F
C02 Oj
-
-
-
11.5 5.8
-
-
11.3 5.8
-
-
-
-
11.8 5.9
-
-
-
-
-
11.6 6.37
12.0 5.4
-
-
-
-
Total
N2 CO H.C. N02 NO NO^
_ _ _ _
._ _ _
_ _
_ _
81.8 .03 10 8 270
_ _
_
81.5 .03 5 5 150
_ _
- _
. _ _ _ _
- _
81.3 .03 3 60 1050
- _
- _
_ _
- _
_
81.2 .03 28 40 200
81.6 .03 5 10 240
_ _
_ _
_ _
_ _
Exhaust Condensate (ppm)
HCHO NH,
27.3
30.8
58.5
33.6
143.0
50.12
1.24
98.9
16.4
2.8
13.9
9.73
2.4
3.26
ppra in
Exhaust Gas
HCHO NTT
.U
-J
I
0.92
0.87
2.0
0.29
7.8 1.3
0.07 0.25
-------
Continuation of Table 36
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
"Run
No.
"138A
138B
138C
139A
139B
139C
204A
204B
204C
204D
204E
205A
205B
205C
205D
205E
205F
205G
205H
231A
231B
231C
231D
231E
231F
Trace Metals on Millipore Filter (%)
Fe Ni Cu Al Ca Mg Mn Cr Sn Zn
-.-
_^_
__________
2.3 <.6 2.66 1.0 24 4.0 <.3 . <.6 <.6 2.0
0.2 <.04 .2 .06 1.6 .26 <.2 <.04 <.01 <.01
3.0 <1.0 6.0 1.5 37 6.0 <.3 <.2 <.6 <.3
1.9 <.5 4.7 <.50 16 1.5 <.5 <.5 <.5 1.6
.056 <.016 .20 <.016 .46 .64 <.01 <.01 .027 .12
__
__________
__________
__________
__________
_._ _ __
_._
---------
__________
---------
1.2 .20 1.6 .62 9.0 1.8 <.2 .24 <.2 1.2
8 <2 10 3 64 14 <2 <2 <2 6
8 <2 8 3 94 14 <2 <2 <2 10
8 <2 11 4 64 12 <2 <2 <2 8
__________
__________
_______
% Pb % C on
Atomic Glass ppm
Ti Absorp Filter BaP
- - - -
- - - -
_ _ _ _
<.6 7.5 63.8 320
<.04 5.0 1.4 <4
<.6 <.l 80.7 <93
<.5 Nil 100 670
<.01 - <5 <5
- - - -
_
(
_
_
_
_
_
<.2 <1.3 35.8 <27
4.8 - - <120
.2 - - 57
11.1 - - <100
_
_ _ _ -
_ - - -
-------
46 8O43
HIDE m u.l.i.
KEUFFEL A ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
0.2 0.1 0.05
0.01
o
S-
o
a>
E
-------
46 8043
MADE IN U.S.A. «
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
0.01
O
J_
O
01
£
S-
(O
0.
Figure 44
MASS DISTRIBUTION
Car 61329
Run No. 204B
ji | ; I ;
Total in Particles of Diameter 98 99.
99.8 99.9
99.99
-------
PROBABILITY
X 2 LOG CYCLES
46 8043
MADE IN U.S.A.
KEUFFEL a ESSEB CO.
99.99
99.9 99.8
0.2 0.1 0.05 0.01
to
C
o
i.
u
s_
a;
to
o
(J
r
4->
S-
Q.
Figure 45
MASS DISTRIBUTION
Car 61329
Run No. 205B
Total in Particles of Diameter
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES «.OE IK U.S.A. .
KEUFFEL & ESSER CO.
99.99
99.9 99.8
99 98
0.01
Figure 46
MASS DISTRIBUTION
Car 61329
Run No. 231B
Total in Particles of Diameter
-------
Car "Rumber:
Vehicle "Type:
Converter Type:
TABLE 37
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
2477
1972 Pontiac 400 CID
Base Metal Beaded "C"
Grams/Mile Particulate
Run
No.
162A
162B
162C
162D
177A
177B
177C
2 0*5 A
206B
206C
206D
205E
205F
206G
206H
226A
226B
226C
226D
226E
Car
Miles
4,325
It
"
II
6,000
n
n
10,841
n
n
it
ii
ii
it
it
15,851
II
tt
»
ii
Converter
Miles
455
n
it
n
2,130
n
n
6,971
n
n
ii
n
it
n
n
11,981
"
n
n
n
Test Mode
FC Mod.
60 mph
FC CS
FC HS
FC Mod.
60 mph
FC CS
FC Mod.
60 mph
FC HS
FC HS
FC HS
FC CS
FC HS
FC HS
FC Mod.
60 mph
FC Mod.
FC HS
FC HS
Andersen
Sampler
.1087
.0197
.1267
-
.0478
.0053
.0866
.0869
.0071
-
-
-
.1333
-
-
.0826
.0062
.0478
-
Millipore
Filter
.0826
.0636
.1800
-
.0652
.0373
.1266
.0869
.0515
-
-
-
.0133
-
-
.0651
.0329
.0130
-
-
Andersen +
Millipore
.1913
.0833
.3067
-
.1130
.0426
.2132
.1738
.0586
-
-
-
.1466
-
-
.1477
.0391
.0608
-
-
Glass
1 cfm
.0196
.0504
.0200
.0040
.2282
.0360
.0200
.0239
.0385
.0233
.0266
.0333
.0500
.0333
.0300
.0413
.0257
.0456
.0499
.0466
Filter
4 cfm
.0163
.0440
'.0067
.0040
.0163
.0337
.0100
.0163
.0473 ^
.0133 £
.0133 '
.0166
.0200
.0142
.0117
.0174
.0229
.0163
.0566
.0599
121A
2,000
Car before converter or 1975
FC Mod. .1391
hardware was installed
.1087 .2478
.0652
.0402
-------
Continuation of Table 37
EXHAUST GAS ANALYSIS
Run
No.
162A
162B
162C
1620
177A
177B
177C
206A
206B
206C
206D
206E
206F
206G
206H
226A
226B
226C
226D
226E
% by Volume
C02 0^ N2 CO
_
11.6 6.5 81.0 .03
_
_
-
11.9 5.4 81.8 .03
_
-
13.0 4.27 81.8 .03
'
_
_
_
_
_
'
11.3 6.4 81.4 .03
_
_
_ _ _
Total
H.C.
-
2
-
-
-
5
-
-
15
-
-
-
-
-
-
-
15
-
-
_
Parts per Million
N0_ KO NO,,
f.' "" A
100 1150
. -
_
_
40 1100
-
_
8 850
_
_
_
- _
- -
_
_
100 1600
_
_
_ - _
Exhaust Condensate (ppm)
HCHO NH,
-
-
-
7.0 19
3.2 8
20 3
177.8
11.5 4.41
-
-
-
26.14
-
-
36.38
115.4
77.4
-
_ _
ppm in
Exhaust Gas
HCHO NH3
-
-
-
-
-
-
-
-
0.72 0.49
-
-
-
-
-
-
-
5.8
-
-
- _
Car before converter or 1975 hardware was installed
121A
-------
Continuation of Table 37
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
162A
162B
162C
162D
177A
177B
177C
206A
206B
206C
206D
206E
206F
206G
206H
226A
226B
226t:
226D
Trace Metals on Millipore Filter (%)
Fe
.28
.041
.22
.54
.087
.41
.28
.063
-
_
-
-
-
10
-
8
_
Ni
<.05
<.006
<.037
.073
.016
<.05
<.05
<.006
-
_
-
-
-
<2
-
<2
_
Cu
.43
.042
.24
.43
.046
.37
.23
.069
-
_
-
-
- ' '
10
-
6
_
Al
.28
.020
.28
.17
.013
.084
.085
.020
-
_
-
-
-
4
-
4
_
Ca
4.3
.44
2.9
3.1
.35
2.0
1.5
.22
-
_
-
-
-
78
-
66
_
Mg
.48
.05
.35
.71
.08
.48
.30
.05
-
_
-
-
-
15
-
14
_
Mn
<.02
<.003
<.019
<.067
<.009
<.053
<.05
<.006
-
_
-
-
-
<2.
-
<2
_
Cr
<.05
.024
.05
<-.067
.022
<.053
<.05
.016
-
_
-
-
-
<2
-
<2
_
<
<
<
<
<
<
<
<2
<2
Sn
.05
.006
.041
.067
.009
.053
.05
.006
-
_
-
-
-
-
_
Zn
<.16
<.016
.19
<.2
<.03
<.16
<.15
<.95
-
_
-
-
-
<6
-
<6
_
% Pb % C on
Atomic Glass
Ti Absorp Filter
<.05 2.5 23.2
<.006 12.5 1.2
<.031 10.0 .5
<.06 <2.5
<.004 5.0
<.053 2.5
<.05 Nil 13.6
<.006 <.2 1.4
_ _ -
_ _ _
_ _
- ' - . ~
_
<2 - -
- - -
<2
_ _
ppra
BaP
100
<4
<93
<83
<5
<93
180
6
-
_
-
-
130
11
<43
_
1
cn
Ul
1
121A
.32
<.08
.24
Car before converter or 1975 hardware was installed
.08 2.76 .40 .04 .04 .08 .08
.24
.56 61.2
-------
46 8O43
M*0r IN U.3.A.
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
99
0.01
Figure 47
MASS DISTRIBUTION
Car 2477
Run No. 162B
Total in Particles of Diameter
-------
PROBABILITY
X 2 LOG CYCLES
46 8043
MADE IN U. S. A.
KEUFFEL ft tSSER CO.
99.99
0.2 0.1 0.05
0.01
to
£
O
J_
U
O
i-
(U
O)
a
O)
S-
rd
a.
Figure 48
MASS DISTRIBUTION
Car 2477
Run No. 177B
Total in Particles of Diameter
-------
i-r BO43
X 2 LOG CYCLES «AOE rn U.S.A. .
KEUFFEL A ESSER CO.
99.99
99.9 99.8
99
1 0.5 0.2 0.1 0.05 0.01
O
S_
o>
E
Q
0)
u
Figure 49
MASS DISTRIBUTION
Car 2477
Run No. 206B
irir.i-t
Total in Particles of Diameter
-------
PROBABILITY 46 SO43
X 2 LOG CYCLES MADE IN U.S.A. .
KEUFFEL ft ESSER CO.
99.99
99.9 99.8
0.01
C
o
i.
u
O)
E
O
S-
a.
Figure 50
MASS DISTRIBUTION
Car 2477
Run No. 226B
% Total in Particles of Diameter
-------
TABLE 38
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
Environmental Protection Agency
1972 Mercedes Benz Diesel (220)
None
. Grams/Mile Particulate
Run Car
No. Miles
K53A 3,171
2.17A 6,250
217B
217C
21 7D "
Converter
Miles Test Mode
FC
FC
FC
FC
60
Mod.
Mod.
IIS
HS
mph
Andersen
Sampler
.1735
.2739
.2333
.2333
.0390
Millipore
Filter
.4657
2.7696
2.3333
2.3066
.4784
Andersen +
Millipore
p
3.
2.
2.
0
6392
0435
5666
539
5174
Glass Filter
1 cfm
.7715
.6261
.5833
.5633
.2536
4 cfm
.6642
.5371
.4849 '
.4599
.20217
a\
o
I
-------
Continuation of Table 38
EXHAUST GAS ANALYSIS
Run Total
No. CO., 0., Nn CO B.C. NO0 NO NO,,
163A
217A
. _ No gaseous analyses
^X / a
217C
217D
Exhaust Condensate (ppm)
HCHO
66
71.5
50.9
-
19.9
NH-,
9.2
4.24
-
5.25
Exhaust Gas
HCHO NH3
-
- -
-
-
0.78 0.36
-------
Continuation of Table 38
ANALYSIS OF EXHAUST PARTICIPATE
Measured in Particulate
Run
No.
163A
217A
217B
217C
217D
Trace Metals on Millipore Filter (%)
Fe
.059
.03
.04
.007
_
Ni
.01
-
-
-
_
Cu
.067
.007
.003
.002
_
Al
.038
.05
.03
.005
_
Ca
.74
-
-
-
_
M£
.087
.02
.01
.002
_
Mn
.005
-
-
-
_
Cr
.012
-
-
-
_
Sn Zn
.01 .07
-
-
-
_ _
% Pb
Ti Absorp
.01 .1
.06
.05
.09
_
% C on
Glass
Filter
80.0
75.0
80.5
83.3
_
ppm
BaP
<3
24
5
-
4
to
I
-------
PROBABILITY 46 SO43
X 2 LOG CYCLES »DE IN U.S.A.
KEUFFEL & ESSER CO.
99.99
0.5 0.2 0.1 0.05 0.01
CO
C
o
s_
u
£-
0)
p
(U
E
(O
O)
o
r
+J
S-
«
Q.
Figure 51
MASS DISTRIBUTION
Merc Benz Diesel
Run No. 217D
:.ri'L i "tZTI
Total in Particles of Diameter
-------
TABLE 39
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
Environmental Protection Agency
1971 Oldsmobile DRX 401 - 350 CID
Not known
Run
No.
178A
4,285
Converter
Miles
2,200
Test Mode
FC Mod.
Andersen
Sampler
.0261
Grams/Mile Particulate
Millipore
Filter
.0087
Andersen +
Millipore
.0348
Glass Filter
1 cfm
.0087
4 cfm
.0098
!
M
*»
Additional Notes
1) This car did not start well, looking at the choke it did not appear to be closing completely. We
were told by Phillips that this is the way it was designed to work.
2) The filter papers with the particulate collected on it were very light in color. Not at all like
the 1971 Chevrolet, Federal Cycle cold start runs with choke on.
3) The dilution tube sweeping (particulate) did not resemble the usual type of material we have observed
in the past. There was a sparkle to the particulate and the density was apparently very low.
4) All samples of particulate were given to the E.P.A. for analysis at Ann Arbor.
-------
Continuation of Table 39
EXHAUST GAS ANALYSIS
Run
Mo.
178A
co2
% by Volume
°2
Total
H.C.
Parts per Million
NO, NO NO
""
-------
Continuation of Table 40
EXHAUST GAS ANALYSIS
Run
No.
203A
203B
203C
203D
203E
203F
203G
211A
211B
co2
11.7
-
-
-
-
-
-
12.0
°2
5.2
-
-
-
-
-
-
5.6
^2
81.8
-
-
-
-
-
-
81.4
CO
_
.03
-
-
_
_
-
-
.03
Total
H.C.
_
52
_
_
_
_
-
_
66
NO.,
40
-
_
_
-
-
_
40
NO NOX
-
1650
-
-
-
-
- .
-
1450
Exhaust Condensate (ppm)
HCHO NH3
474.8
266.3 19.6
-
-
328.0
-
-
363 9.26
600 21.24
ppm
Exhaust
HCHO
-
15.7
_
_
-
_
-
-
14.59
in
Gas
NH,
2.0
-
-
-
-
-
-
0.9
-------
' TABLE 40
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
Environmental Protection Agency Lease 4065
1972 Chevrolet 400 CID
PPG Trap Mufflers, etc0
Grams/Mile Particulate
Run
No.
203A
203B
203C
203D
203E
203F
203G
211A
211B
Car
Miles
14,566
II
II
II
II
II
II
16,000
n
Converter
Miles Test Mode
14,566 FC
60
i. FC
FC
ii FC
n pc
FC
16,000 FC
60
Mod.
mph
HS
HS
Mod.
HS
HS
Mod.
mph
Andersen
Sampler
.0520
.0150
-
-
.0782
.
-
.1043
.0134
Millipore
Filter
.0650
.1880
-
-
.2217
-
-
.0174
.0907
Andersen +
Millipore
.1170
.2030
-
-
.2999
-
-
.1217
.1041
Glass Filter
1 cfm
.0625
.2124
.0666
. .0733
.0978
.0800
.0667
.1054
.0933
4 cfm
.0550
.1791
.0517
.0500
.0586
.0516
.0483
.0619
.0826
I
a\
I
Additional Notes
1)
Water condensate which dripped out of the tail pipe connection during the test contained an orange-
yellow colored solid material. An analysis of this material was conducted.
Phosphorus = <.8 ppm
Sulfur = .016%
Zn = .02%
Fe = .23%
Pb = .25%
Br = .20%
2) Run 211B - 60 mph steady-state was interrupted by two blown tires during the run.
run was conducted with a 30-minute time interruption'to change tires.
A full two-hour
-------
Continuation of Table 40
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
203A
203B
203C
203D
203E
203F
203G
211A
211B
Trace Metals on Millipore Filter (%)
Fe Ni
1.2 <.067
.004 <.002
-
- -
o32 <.02
-
-
-
_ _
Cu
.43
.007
-
-
.29
-
-
-
_
Al Ca Mg
.13 2.7 .65
<.002 .016 <.006
_
_
<.02 160 .10
_
_
-
_ _ -,
Mn
<.067
<.002
-
-
<.02
-
-
-
-
Cr
.13
<.002
-
-
<.02
-
-
-
-
Sn
<.067
<.002
-
- '
<.02
-
-
-
-
Zn
.24
.027
-
-
.18
-
-
-
-
Ti
<.067
<.002
-
-
<.02
-
-
-
-
% Pb % C on
Atomic Glass ppm
Absorp Filter BaP
39 33.9 180
20.2 36.6 <1
_ _
_
29.8 20.2 <24
_ _ -
_ _ -
1
I
-------
46 SO43
MADE IN U.S.A.
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
1 0.5 0.2 0.1 0.05 0.01
c
o
i.
u
t-
O)
O)
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES NAD[ IN u i.».
KEUFFEL ft ESSER CO.
99.99
0.2 0.1 0.05 0.01
o
s_
u
i.
-------
TABLE 41
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
Environmental Protection Agency 801692
Mail Jeep Body
Stratified Charge Engine
Grams/Mile Particulate
Run
No.
219A
219B
219C
219D
Car
Miles
4,262
II
II
II
Converter
Miles
4,262
n
it
n
Test Mode
FC
FC
FC
60
Mod.
HS
HS
mph
Andersen
Sampler
.1043
.0999
.1066
.0091
Millipore
Filter
.0217
.0199
.0267
.0756
Andersen +
Millipore
.1260
.1198
.1332
.0847
Glass
1 cfm
.0283
.0500
.0666
.1046
Filter
4 cfm
.0358
.0533
.0516
.0942
Additional Notes
1) There was some question as to whether the engine was running right. It appeared to have a spark
plug miss.
2) Dilution tube sweepings at the end of this series of runs I219A, B, C & D was 45.8 grams which is
a gross amount compared to other vehicle tests. The sweepings were almost all magnetic indicating
iron from the exhaust system. We were told this vehicle had not been run for a prolonged period
which could account for the tube sweepings.
3) The test vehicle would not obtain 60 mph on the dynamometer so the test was conducted at 50 mph.
4) At 50 mph steady-state operation on the dynamometer there did not appear to be a miss in the engine.
-------
Continuation of Table 41
EXHAUST GAS ANALYSIS
% by Volume Parts per Million
Run
No.
219A
219B
219C
219D
Total
CO, 0, N, CO H.C.
- - - -
_
11.8 5.8 81.5 .02 10
Exhaust Condensate (ppm)
NO, NO NO
-
_
7.0 230
HCHO
33.7
10.0
-
10.3
NH.j
5.4
2.9
_
3.3
-J
to
1
ppm in
Exhaust Gas
HCHO NH,
-
-
0.6 0.35
-------
Continuation of Table 41
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
No.
219A
219B
219C
219D
Trace Metals on Millipore Filter (%)
Fe
2
5
4
.03
Tube 40
Sweepings
Ni
.09
.13
.006
.007
802
Cu
.9
.8
.06
.03
.03
Al
.4
1.5
.09
.009
.04
Ca Mg
1 10
16 16
12 .6
= 3 .2
o005 .004
Mn
.2
.3
.03
.006
.2
Cr
.3
.3
.03
.006
.001
Sn Zn
.10
.15
.006
.01
« w
Ti
.02
.10
.009
.003
.005
% Pb % C on
Atomic Glass ppm
Absorp Filter BaP
61.
109.
-
23.
» _
0 <31
0 <10
<31
2 3.2
_
00
1
-------
46 8O43
MADF IN U.S.A.
KEUFFEL ft ESSER CO.
PROBABILITY
X 2 LOG CYCLES
0.01
C
o
j-
o
(V
-IJ
(V
E
to
O)
4J
S-
rd
a.
Figure 54
MASS DISTRIBUTION
Stratified Charge
Run No. 219D
Total in Particles of Diameter i * i i i i i i i LJ : i i [ii L_J
20 30 40 50 60 70 SO
oni
98 99
99.8 99.9
99.99
-------
TABLE 42
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type;
Environmental Protection Agency
1971 Pontiac
Questor Converter System
Grams/Mile Particulate
Run
No.
221A
221B
221C
221D
Car
Miles
8,000
"
"
n
Converter
Miles
8,000
"
"
"
Test Mode
FC
FC
FC
60
Mod
HS
HS
mph
Andersen
Sampler
(Tail pipe
.1399
.1333
.0083
Millipore
Filter
disconnect
.0133
.0266
.0104
Andersen +
Millipore
failure)
.1532
.1599
.0187
Glass
1 cfm
.1600
.1533
.0292
Filter
4 cfm
.0883
.1049
.0175
01
Additional Notes
This car was driven on the dynamometer by the driver that delivered the vehicle. The choke on this
vehicle. The choke on this vehicle was so adjusted so that it would not close completely and fuel was
introduced at the carburetor to start it. No cold starts were conducted for this vehicle. Just hot
starts and 60 mph steady-state.
-------
Continuation of Table 42
EXHAUST GAS ANALYSIS
Run
No.
CO..
221A -
221B -
221C -
221D 10.7
% by Volume
0_2
7.7
80.6
.03
Parts per Million
Total
CO H.C.
10
135
NO
1760
Exhaust Condensate (ppm)
HCHO NH,
2.1
.13
97.7
4.7
ppm in
Exhaust Gas
HCHO NH
.007
0.4
-------
Continuation of Table 42
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
221A
221B
221C
221D
Fe
8
-
3
.6
Ni
-
.15
.09
Cu
3.3
-
1.4
.52
Trace Metals
Al Ca
1.3 <25
-
1.0 <12
.32 <2
on Millipore Filter
M£ Mn
>3 >.5
-
>12 >.2
>2 >.04
(%)
Cr
-
.3
.2
Sn
.10
-
.08
.06
Zn Ti
.13'
-
.05
.02
% Pb
Atomic
Absorp
-
-
-
% C on
Glass
Filter
-
-
5.1
4.5
ppm
BaP
<5
-
-
<7
Tube 25
Sweepings
15
»09 .2
o07
.02 .5
.5
,09
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES .HOE in B.I... .
KEUFFEL & ESSER CO.
99.99
99.9 99.8
99 98
95 90
80 70 60 50
9-
8-
7_.
6..
5-
4
in
c.
o
i.
u
i-
0>
+J
(U
e
(O
0>
u
r
4->
S-
-------
TABLE 43
CHASSIS DYNAMOMETER TEST OF CONVERTER EQUIPPED VEHICLES
Car Number:
Vehicle Type:
Converter Type:
DuPont
1970 Chevrolet
Thermal Reactor + EGR
Cyclone Collectors
Grams/Mile Particulate
Run
No.
222A
222B
222C
222D
222E
222F
222G
222H
2221
Car
Miles
11,376
II
II
II
II
It
II
II
II
Converter
Miles
2,000
it
it
n
it
ii
it
it
»
Test Mode
FC Mod.
FC HS
FC HS
FC Mod.
60 mph
60 mph
FC Mod.
FC HS
FC HS
Andersen
Sampler
.1565
.0733
.0533
.0695
.0147
.0077
.1043
.1666
.1399
Millipore
Filter
.0869
.1200
.1133
.0608
.0178
.0080
.2434
.1599
.1799
Andersen +
Millipore
.2434
.1932
.1666
.1303
.0325
.0157
.3477
.3265
.3198
Glass
1 cfm
.0804
.0466
.0499
.0826
.0243
.0191
.1674
.0799
.1399
Filter
4 cfm
.0619
.0283
.0333
.0652
.0199
.0155
.1283
.0449
.1183
1
M
>J
ID
1
Additional Notes: See attached.
-------
-ISO-
Additional Notes
Two different brands of fuel were used to make the series
of runs.
(1) Run #222 A, B, C, D, E were made on the fuel the vehicle
had in it when it was delivered. Supposed to be Sunoco #240,
(2) Run #222 F, G, H, I were made on Bay gasoline.
GASOLINE ANALYSIS
Bay Gas Sunoco #240
RVP
Gravity
I.B.P.
5% distillation
10%
20%
30%
40%
50%
60%
70%
80%
90%
95%
E.P.
RON
MON
Pb grams/gallon
Br grams/gallon
10.5
57.4
100.
121.
148.
175.
192.
221.
238.
260.
280.
320.
370.
390.
415.
95.4
84.8
3.30
1.27
7.3
59.1
96.
115.
125.
148.
171.
196.
219.
240.
264.
301.
351.
400.
410.
98.3
87.0
2.34
.91
-------
Continuation of Table 43
EXHAUST GAS ANALYSIS
% by Volume
Parts per Million
Run
No.
222A
222B
222C
222D
222E
222F
222G
222H
2221
C02 °2
_
. _
. -
13.1 3.7
13.5 3.1
-
-
_ ' _
Nj CO
-
-
-
-
82.3 .03
82.5 .03
-
-
_ _
Total
H.C.
-
-
-
-
1
0
-
-
_
NO,
£
-
-
-
24
7.5
-
-
_
NO
-
-
-
-
545
462
-
-
_
NO
-
-
-
-
-
-
-
-
_
LUSt
Condensate
HCHO NH
9.6
1.8
5.3
.5
1.1
1.2
15.
1.
"
.
,
8.
(ppm)
3
6
85
87
85
87
6
ppm in
Exhaust Gas
HCHO NH,
-
_
0.02 0.05
0.07 0.09
- -
1
I-1
CO
i
4.9
-------
Continuation of Table 43
ANALYSIS OF EXHAUST PARTICULATE
Measured in Particulate
Run
No.
222A
222B
222C
222D
222E
222F
222G
222H
2221
Trace Metals on
Fe
.7
.6
-
-
.5
.6
2
.4
-.
Ni
.03
.03
-
-
.03
.03
.15
.01
-
Cu
.4
.3
-
-
.7
.6
1.8
.1
-
Al
.5
.4
-
-
.3
.4
1.3
.2
-
Ca
3
3
-
-
1
2
9
2
-
Millipore Filter (%)
Mg
3
3
-
-
1
2
5
1
-
Mn
.05
.06
-
-
.02
.04
.2
.04
-
Cr
.1
.1
-
-
.1
.08
.3
.05
-
Sn
.02
.003
-
'
.01
.02
.19
.002
-
Zn Ti
.009
.008
-
-
.01
.008
.02
.004
-
% Pb
Atomic
Absorp
29.0
13
-
-
44
30
37
13
-
% C on
Glass
Filter
20.6
10.6
-
-
1.7
3.1
24.5
18.8
-
.ppm
BaP
53
120
-
-
<8
100
41 .
1
_ I-1
00
to
-------
PROBABILITY 46 8O43
X 2 LOG CYCLES «»OE
-------
46 8O43
«.ot IN u.j.». .
KEUFFEL & ESSER CO.
PROBABILITY
X 2 LOG CYCLES
99.99
99.9 99.8
0.2 0.1 0.05 0.01
O
J-
O
-------
-185-
,0800
0700
0600
0500
to
S-
O)
4J
IO
r
3
O
i.
0}
CL.
0400
0300
0200
,0100
Federal:
Cycle!-;
Tot
2500 5000 7500 10,000 12,500 15,000
Mileage, Converter
Car 61329, Basemetal, EGR - 142 mm filter, 4 cfm
-------
-186-
OJ
to
S-
C3
O)
(O
r
O
ro
0.
;0700
0600
,0500
0400
,0300
0200
0100
J.
T'
_..
-t
_LJ
T \
x
i i
2500
f.
]
! !
-fi-
i !
-jn mxh.
r
F rr
5000 7500. 10,000 12,500 15,000
Miles, Converter
Car 61329, Basemetal, EGR - 142 mm filter, 4 cfm
-------
-187-
0800
,0700
,0600
CD
.7 .050.0
in
e
-------
-188-
0700
,0600
Ol
- .0500
O)
)-)
tO
3
(J
0400
Q- .0300
0200
0100
2500
5000 7500 10,000
Miles , Converter
12,000 15,00(
Pontiac, Basemetal - 142 mm filter, 4 cfm
-------
-189-
008
0700
0600
Ol
(O
S-
C3
d)
4->
10
S-
(O
CL
0500
0400
0300
,0200
0100
5000 10,000 15,000 20,000 25,000 30,000 35,000
Car Miles
Car 61314, Noble Metal - 142 mm filter, 4 cfm
-------
-190-
E. TASK V
Task V involved a preliminary look at some of the factors
affecting the particulate mass sampling from a small, one-
cylinder diesel engine. Table 44 is a summary of the speci-
fications of the Labeco diesel used in the study. Figure 63
is a diagram of the dilution tube apparatus used for collecting
the particulate.
The tube was designed to give 400 fpm flow rate of air and
exhaust. This was roughly equivalent to what was used on
the gasoline engine studies.
The first runs were made with the exhaust from the diesel
entering the dilution tube at an orifice in the tube, hoping
that the turbulence set up by the orifice would allow complete
mixing. This proved to be unsatisfactory because of the large
degree of the exhaust pulsation occurring with the one cylinder
engine.
The pulsation effect was greatly reduced by introducing the
exhaust into the dilution air stream counter to the air flow.
The pulses were still strong enough, however, to necessitate
the placement of the exhaust inlet at least 5 feet from the
filter at the air inlet.
Only a few runs were made before the lack of time and funds
forced us to stop. Several preliminary conclusions can be
drawn:
1. The mass collected on the filter media, both
millipore and glass fiber, was high enough to
allow detailed analytical work on the particu-
late.
2. A single cylinder engine such as the one used in
this work can be a valuable tool for diesel
studies.
-------
-191-
3. The size of the particles collected appears to be
quite small, based on the figures in Table 45 showing
the amount collected on the Andersen plates vs. the
amount collected on the back-up Millipore filter.
TABLE 44
LABECO DI DIESEL, UNSUPERCHARGED
Bore: 3.80 in.
Stroke: 3.75 in.
Displacement: 42.53 cu. in.
Weight: 418 Ibs
Compression Ratio: 16.7:1
Brake Torque at 1600 rpm 26.7
Brake Horsepower at 2800 rpm 11.8
BMEP at 1600 rpm 94
FMEP at 2800 rpm 56
FMEP at 2000 rpm 44
IMEP at 1600 rpm 136
ISFC at 1200 rpm .365
BSFC at 1200 rpm . .511
TABLE 45
GRAMS/HOUR COLLECTED FROM A ONE CYLINDER
LABECO DIESEL ENGINE
(1500 rpm, 900 grams/hr fuel consumption)
Grams/Hour
Millipore Andersen + Fiberglass
Run No.
214A
214B
Time
20 min
60 min
Andersen
.0198
.0036
Filter
.0894
.1494
Millipore
,1092
.1530
Filter
.1221
.1378
Dilution tube flow rate was 400 ft/min, or 65 cubic feet
per minuteo
-------
FtOW -
CONTROL
FIGURE 63 - DIESEL DILUTION TUBE AND SAMPLING
SET-UP ' ;
-------
-193-
APPENDIX A
AN INVESTIGATION OF SOME VARIABLES OBSERVED
WHEN SAMPLING PARTICULATE MATTER FROM AIR DILUTED
AUTOMOTIVE EXHAUST
Otto J. Manary 0. C. Valenta
and
Michael J. Baldwin
The Dow Chemical Company
Midland, Michigan
October 1971
-------
-194-
SUMMARY
Both the quantity and quality of particulate matter* collected
from air diluted automotive exhaust are affected by such variables
as sampling temperature, dilution ratio, flow rate and by the
presence of gasoline additives. An investigation of the effect
of these variables is presented in this report.
*Particluate matter is defined to be that nongaseous matter
collected at filters under the sampling and operating conditions
specified for each separate run described herein.
-------
-195-
INTRODUCTION
In 1969 a study of the participate emissions present in the
exhaust effluent of automobile engines v/as initiated under a
government contract at Dow. The problem at hand was to
evaluate the effect of gasoline additives on the nature of
such emissions. This necessitated the development of a
reliable particulate sampling procedure. This procedure,
described in a previous report1 was found in duplicate tests
to afford samples of particulate matter repeatable to within
±10% on a weight basis.
During the first year of the above program particulate samples
were collected from the exhaust system of an internal combustion
engine operating under controlled conditions on a dynamometer.
The left bank of cylinders of a V8 engine was discharged via a
convectional automobile exhaust system into a 27' x 18" poly-
vinyl chloride dilution tube where the exhaust effluent was
diluted tenfold with filtered air. This condition permitted
isothermal, isokinetic sampling of particulate matter at the
end of the tube remote from the engine (See reference 1).
During the second year of the research program the results
obtained in the above engine studies were related to those
obtained for vehicles operating on a chassis dynamometer.
Physical restrictions necessitated that changes be made in
the particulate sampling procedure. A six hour sampling
period was found to be the most practical as compared to a
48 hour period in the previous engine dynamometer studies.
This reduction in sampling time would be expected to yield
an eightfold reduction in weight of particulate matter collected
However during the chassis dynamometer studies the total vehicle
exhaust was discharged to the dilution tube. An appropriate
change in the air dilution ratio was made to afford the same
overall flow rate within the dilution tube. Thus during the
six hour sampling period and under otherwise identical sampling
conditions it was expected that one fourth of the weight of
particulate matter collected in the engine dyno runs would be
collected during the chassis dynamometer tests.
This smaller weight of particulate matter proved inadequate
for comprehensive analysis. It was therefore decided to
supplement the Andersen sampler and filter combination
operated at 1 cfm and used exclusively to this point of the
program with an additional separate 142 mm filter operated
at 4 cfm. One might expect the 4 cfm filter to collect
four times the quantity of particulate matter collected at
the 1 cfm Andersen and filter combination. In fact the
Government Report # CPA-22-45-145)
-------
-196-
4 cfm filter collected from 20 to
matter than the ideal.
70% less participate
In viev/ of the trend to nonleaded gasolines and the necessity
for sampling participate matter in the relatively short cycling
sequences of the LA-4 and new Federal test procedures it was
felt to be .imperative that this anomaly be resolved and that
the particulate sampling procedure be refined to afford meaning
ful and reproducible results under the above conditions.
APPARATUS
1. Chassis Dynamometer
2. Engine Dynamometer
3. Andersen Sampler
4. Filter Holder
a. Glass fiber filter
b. Millipore membrane
5. Analytical balance
6. Electrical heated oven
7. Particulate collection
CLAYTON
DYNAMATIC
GELMAN
GELMAN
filter
system (See Figure I
PROCEDURE
Details of the operating and sampling procedures are presented
in Reference 1. All engines and vehicles were operated at the
equivalent of 60 mph road load conditions.
1. Handling and Weighing of Filter Papers
The normal procedure for handling and weighing of filter
paper was as fol1ows:
a. The paper was stored in an air conditioned room at
75°F and weighed prior to use.
After sample collection the paper was weighed within
minutes of removal from the filter holder and
when the weight had stabilized on the balance
which was in the same air conditioned room at
again
pan
75°F.
An experiment was conducted to evaluate the effect of
heating on the weight of glass fiber filter pads. The
results are presented in Figure §2. Since no weight
loss occurred between the temperatures of 75°F and
-------
-197-
100°F it was felt that the error in handling filter
papers in the above manner was negligible.
2. Milli pore ve r s u s Gel man F i1t e r Papers
In most of the work a glass fiber filter was used. However
in some incidents a membrane type filter was used to obtain
particulate on a soluble substrate which could be used in
various analytical procedures. It is important that both
filter papers be equally efficient in particulate collection
(Figure #3).
The data shows only a small difference in the efficiency of
the two filters opera-ting at 1 cfm flow rate. The difference
could be a function of the fuel additive. The data presented
includes no lead fuel and leaded fuel for comparison.
3. Exhaust Gas Velocity Through Filter Paper - 1 cfm vs. 4 cfm
If one compares a 1 cfm filter to a 4 cfm filter, the temperature
of the exhaust gas in the 4 cfm filter is higher due to the
shorter time it had in the sample line between the dilution
tube and the filter. Consequently in order to compare the
direct effect of flow rate on particulate collection the
sample line leading to the 1 cfm filter had to be heated to
maintain the same temperature as the 4 cfm filter.
As can be seen from Figure #4 there was a considerable loss
in filter efficiency as the face velocity of the gas stream
through the filter is increased. The exhaust particulate
from leaded gasoline were less sensitive to filter flow rate
than those from nonleaded fuel. This is probably due to a
much higher level of volatile organics associated with non-
leaded gasoline exhaust particulate.
4. Exhaust Gas Temperature at Filter
It is not surprising that the temperature of the filter would
have a pronounced effect on the amount of particulate matter
collected on the filter (See Figure #5). Again the use of
nonleaded gasoline has a greater effect on the percent of
particulate retained as the filter temperature is increased.
As was also expected the mild cycling conditions show the
greatest percent change due to the lower boiling materials
one would expect in the exhaust gas.
-------
-198-
5. Leaded Fuel Versus Nonleaded Fuel
The percent of lead in the gasoline had a definite effect
on the difference in the participate pick-up on the 1 cfm
filter plus Andersen sampler versus the 4 cfrn filter
only (See Figure #6).
A tabulation of the runs that were made at 60 mph road
load steady state operation shov/ that the particulate
matter collected from leaded gasoline exhaust is the
least sensitive to changes in sampling procedure.
It is felt that the increasing differences observed with
the reduction of lead in the fuel are due to the increased
level of volatile organics associated with the particulate
matter resulting from the combustion of nonleaded fuel.
The face velocity on the filter and the higher operating
temperature of the 4 cfm filter would have a greater
tendency to remove the more volatile compounds.
6. 1 cfm Filter Plus Andersen Sampler Versus 1 cfm Filter
Without Andersen Sampler
A comparison was also made between the efficiency of the
1 cfm filter alone and the 1 cfm filter used in conjunction
with an Andersen sampler. With leaded gasoline the difference
in efficiency varies from 2.8 to 13.9% (See Figure £7). Using
nonleaded gasoline the efficiency of the two systems varies
from 39.5 to 82.4%. Of particular significance is the fact
that the Andersen sampler collects 70 to 80% of the particulate
matter compared to the 1 cfm filter when using ncnleaded fuel.
However when leaded fuel is used the Andersen sampler accounts
for only 35 to 40% of the particulate matter collected. It
is felt that the difference in collection efficiency between
nonleaded fuel and leaded fuel is again due to the high ratio
of volatile organics present in the particulate matter from
the nonleaded gasoline. The magnitude of the thermal drop
across an Andersen sampler is similar to that of a filter so
one would expect to collect more particulate on the combined
.1 cfm filter/Andersen system.
7. Thermal Profile of Sampling System
The data tabulated in Figure #8 shows the average thermal
drop across a 1 cfm filter to be 36.4°F and that across a
4 cfm filter to be 12.7°F. The temperature drop across
the Andersen sampler was found to be 17.4°F and that across
the combined Andersen and 1 cfm filter to be 53.8°F.
-------
-199-
8. Unbound Mater In Two Types of Filter Media Before and
After Partlculate Col 1ection
In order to assess the possible effect of unbound v/ater
on the weights of particulate matter collected a glass
fiber and millipore filters, both types of filter pads
were desiccated both prior to and after sample collection
The data is tabulated in Figure #9.
It was found that the glass fiber filter has a very low
water retention this was the main filter used for the
particulate collection studies to date. The membrane
filter was very susceptible to water pick-up and must be
used with caution.
CONCLUSIONS
1. The temperature of the diluted exhaust gas has a significant
effect on the quantity of particulate matter collected in
the sampling system described in this report.
2. The flow rate of the exhaust gas being sampled through a
filter .paper was found to have a definite effect on the
amount of particulate collected on the filter paper.
3. Fuel additives such as TEL have an effect on the efficiency
of particulate collection system as a result of changes in
particulate composition.
4. The major filter media used for particulate collection was
glass fiber filter with no binders. This type of filter
media presented no gross problem if proper handling and
weighing procedures were followed.
RECOMMENDATIONS
1. The effects of exhaust gas dilution ratio, dilution air
moisture content, dilution air temperature, and residence
time in the dilution tube are other factors that should
be studied as to their effect on the collection of
particulate emissions.
2. Any definition of particulate matter must be referred to a
very well defined set of sampling parameters.
-------
Flow Diogrom for Engine Exhoust
Particulote Collection
Air
out
n
Filter
I
Instrument
_ . v7- and
^Engine Roo.n Rj Control Room _
Mix-ing -*-""
*. I fel _
Air
in.
Part iculate
Gravimetric Fallout
Flow
Control
Eng ine
Dynamometer
Eng ine
Sampling Slits
Tail Pipe
is
Standard Muffler
Scott Reseorch \r\s,
NO and N02
Analysis
Fisher Gas Part i'c i oner
CO, C02, M2, 02
Total Hydrocarbon
Analyzer
Exnaust Pipe
Anderson
Separator
Millipore
Filter
Flow Meter
Vacuum
Pump
*
/
Air
Pump
I
ro
o
I
Manome
Figures ]
-------
-201-
FIGURE-#2
TYPE "A" GLASS FIBER FILTER #61698 142 mm
All temperatures were held for one hour
before weighing. Six papers were used
to check reproducibility
Filter Paper No
Temperature °
Wt. grams at
l-.'t. grams at
Total grams 1
Wt. grams at
Total grams 1
Wt. grams at
Total grams 1
Wt. grams at
Total grams 1
Start wt. gra
48 hrs. wt. g
F
75°
100°
OSS
150°
OSS
200°
OSS
250°
OSS
The
ms at .
rams at
#1
.9341
.9341
.0000
.9338
.0003
.9335
.0006
.9333
.0008
six filter pa
room over
75° .9341
75° .9340
#2
.9572
.9572
.0000
.9567
.0005
.9563
.0009
.9563
.0009
pers were
weekend a
.9572
.9570
#3
.9404
.9404
.0000
.9400
.0004
.9398
.0006
.9396
.0008
returned
nd reweigh
.9404
.9403
//4
.9565
.9565
.0000
.9564
.0001
.9557
.0008
.9557
.0008
to storage
ed
.9565
.9564
#5
.9471
.9471
.0000
.9467
.0004
.9464
.0007
.9464
.0007
.9471
.9471
#6
.9453
.9453
.0000
.9450
.0003
.9444
.0009
.9444
.0009
.9453
.9449
Permanent wt. loss grams .0001
.0002
,0001
.0001
.0000
0004
-------
-202-
FIGURE #3
SAMPLING
Same Exhaust for Same Time Period
Flow Rates on All Filters 1 cfm
Mil 1i pore
0AAWP 14200
AA 0.8 p pore size
White Plain 142 mm
Membrane Type
Gel man
Glass Fiber Filter
Type A - 142 mm
99.7% Efficient for
Removal of .3 u particles^
Change
Run #44
Nonleaded Fuel
.0402 grams
.0377 grams
-6.2
Run #75
Leaded Fuel
.0652 grams
.0701 grams
+ 7.0
-------
«*- o
O -J
r t_
GJ
O +>
!-> i
o o
<_) C
a
E S_
<+- O
O 4-
4-
+20
+ 10
0
-10
-20
FIGURE #4
DILUTED EXHAUST PARTICULATE FILTRATION
- Nonleaded Fuel
Leaded Fuel
» ' ! i * ", ^» « «K ^^ ^^ ^^
- «
110°F - 79
-152°F - 79
133°F - 78
101°F -'77
93°F - 78
!' ""temp". '"'F'of"'
i Filters, Run #
- 77
!
! J. t ....
8 10 12 14 16 18 20 22 24
FOOT PER MINUTE GAS VELOCITY PAST FILTER
26 28
30
ro
O
CO
i
X =
1 cfm
= 4 cfm
-------
O J
S_
o +->
-(-> r
Lt.
CJ
i-
£ O
o o
o c:
o
E s-
4- CU
O >4-
FIGURE #5
DILUTED EXHAUST PARTICULATE FILTRATION
+20
+ 10
Nonleaded Fuel
- Leaded Fuel
\ Rn #79
cefm ;
i
#77 - 60 mph ss
; 4" cfm 60 moh ss
\\ #78 Mild Cycling
rv>
o
100
110
120 130 140 150
EXHAUST GAS TEMPERATURE °F AT FILTER
160
170
180
X =
1 cfm
= 4 cfm
-------
-205-
FIGURE f/6
The effect of leaded gasoline on the difference in
participate pick-up on the 1 cfm filter and Andersen
versus 4 cfm filter only.
Number Percent Less Particulate Collected
TEL in of Runs on 4 cfm Filter Compared to 1 cfm
Gasol ine Averaged Filter and Andersen
None 6 -72.23%
.5 cc/gal. 4 -40.60%
3. cc/gal. 9 . -19.29%
-------
FIGURE #7
Leaded Gasoline
Participate
Grams Per Mile '
% Change
1 cfni Filter
Run # 1
80
81
No n leaded
82
Mil 1 i pore
Glass
cfm Filter
.0117
(60.6%)
.0121 ' \
(65.0%) v-
.01228
(64,0%)
.01253
(66.5%)
Gasol ine
.0010
(14.4%)
.0019
(30.10%)
Andersen
.0075
(38.8%)
.0065
(39.6%)
.0069
(35.9%)
.0063
(33.4%)
\
\
.0057
(82.6%)
.0044
(69.8%)
Total
.0193
.0186
.01918
.01883
.0069
.0063
1 cfm
Only
.0166
.0164
.01826
.01883
.0012
.0038
Compared to
1 cfm & Andersen
-13.9%
-11 .82%
-4.7%
-2.8%
-82.4
d y . b /<>
I
l\5
O
0>
I
-------
THERMAL
-207-
FIGURE #8
PROFILE OF SAMPLING SYSTEM
Chassis dynamometer 60 mph steady
s t a t.e w i t h air to exhaust dilution
ratio between 4:1 and 6:1
Tempera
CC of TEL
3.0
0.0
0.5
3.0
0.5
0.0
3.0
0.0
0.0
3.0
3.0
3.0
0.0
i
Of Sample 1 cfm
167.2
144.2
147.0
147.2
141.4
139.0
156.6 108.
116.7 101.0
92.7
125.2 105.1
131.8 103
105
129.8 108
106
128.4 98.6
104.0
ture °F
Andersen
Plus 1 cfm
Filter
81 .5
92.0
90.0
87.5
84.3
82.4
86.0
86
86
88
86
82
82.4
i
4 cfm
Filter
143.5
142.0
137,0
136.1
132.0
126.0
139.
102.2
93.6
117.0
Run //
50
58
61
67
70
72
75
77
78
79
4
80
81
82
Avg. Temp, °F = 139.5 103.1 87.5 126.8
-------
FIGURE #9
DESICCATOR STUDY OF FILTER PAPER
New Tare
Tare
Mi
Fi
Gl
Fi
Mi
Fi
Gl
Fi
After 48 hrs. Moisture
in Desiccator Loss
IHpore t #1 3.2961 gm 3.2883 gm .0078 gm
Hers V. #2 3.2945 gm 3.2868 gm .0077 gm
ass / #3 1.1226 gm 1.1223 gm .0003 gm
ber x #4 1.1321 gm 1.1318 gm .0003 gm
Immediately After
Particulate Collection New Tare Particulate
11 ipore / #1
Her x #2
ass /
ber ^
#3
#4
3
3
1
1
.3019
.2984
.1295
.1394
gm
gm
gm
gm
After 48 hrs.
in Desiccator
Mi
Fi
Gl
Fi
11 ipo're / #1
Hers ^ #2
ass r
bers ^
#3
#4
3
3
1
1
.2934
.2919
.1293
.1390
gm
gm
gm
gm
3.2883
3.2868
1.1223
1 .1318
gm .0136 gm
gm . 01 1 6 gm
gm .0073 gm
gm .0076 gm
After 24 hrs. Moisture
in Desiccator Loss
3.2959 gm .0060 gm
3.2925 gm .0059 gm
1 . 1 294 gm .0002 gm - g
1 .1392 gm .0002 gm '
Moisture
Loss
.0085
.0065
.0003
.0004
gm
gm
gm
gm
-------
-209-
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
APTD-1567
2.
3. RECIPIENT'S ACCESSION'NO.
PB-224243/AS
4. TITLE AND SUBTITLE
Characterization of Particulates and Other Non-Regulatec
Emissions from Mobile Sources and the Effects of Exhaustje
Emissions Control Devices on These Emissions
5. REPORT DATE
March 1973
i. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
J. E. Gentel, 0. J. Manary, Joseph C. Valenta
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
The Dow Chemical Company
Midland, Michigan 48640
11. CONTRACT/GRANT NO.
EHS 70-101
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Mobile Source Air Pollution Control
Emission Control Technology Division
Ann Arbor, Michigan 48105 .
13. TYPE OF REPORT AND PERIOD COVERED
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
fhe effect of emission control devices on the particulate emissions of an automotive
power plant was investigated. The work was divided into five tasks as follows:
TASK I was the characterization of a particulate trapping system, and the determina-
tion of what effects, if any, were noted as conditions within the system were control-
lably varied; TASK II was the definition of a particulate baseline for a 1972 400 CID
engine, using non-leaded and low lead fuel - no emission control devices were used for
the baseline runs; TASK III was the evaluation of the particulate emissions from a
1972 400 CID engine equipped with the following control devices - three different
oxidation catalysts, one NOx catalyst, and one exhaust gas recirculation system; TASK
IV involved testing automobiles equipped with control devices for particulate emis-
sions - these vehicles were supplied by both the contractor and the Government; and
TASK V was to define a preliminary collection system for diesel engine particulate
sampling. In all tasks, r>articulate mass emission rates were measured, as well as
particle mass size distribution, carbon and hydrogen, trace matal? and benzo-a-pyrene
content of the particulate. Ammonia and aldehydes were measured in the exhaust gas
condensate, and gaseous emissions were determined as a routine check on engine
operating conditions.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
Air pollution
Particles
Exhaust emissions
Air pollution control equipment
Gas sampling
Catalytic converters
Spark ignition engines
Diesel engines
c. COSATI Field/Group
13B
13. DISTRIBUTION STATEMENT
19. SECURITY CLASS (ThisReport/
Unclassified
21. NO. OF PAGES
216
Unlimited
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
$13.00
EPA Form 2220-1 (9-73)
-------
-210-
INSTRUCTIONS
1. REPORT NUMBER
Insert the EPA report number as it appears on the cover of the publication.
2. LEAVE BLANK
3. RECIPIENTS ACCESSION NUMBER
Reserved for use by each report recipient.
4. TITLE AND SUBTITLE
Title should indicate clearly and briefly the subject coverage of the report, and be displayed prominently. Set subtitle, if used, in smaller
type or otherwise subordinate it to main title. When a report is prepared in more than one volume, repeat the primary title, add volume
number and include subtitle for the specific title.
5. REPORT DATE
Each report shall carry a date indicating at least month and year. Indicate the basis on which it was selected (e.g., date of issue, date of
approval, date of preparation, etc.).
6. PERFORMING ORGANIZATION CODE
Leave blank.
7. AUTHOR(S)
Give name(s) in conventional order (John R. Doe, J. Robert Doe, etc.). List author's affiliation if it differs from the performing organi-
zation.
8. PERFORMING ORGANIZATION REPORT NUMBER
Insert if performing organization wishes to assign this number.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Give name, street, city, state, and ZIP code: List no more than two levels of an organizational hirearchy.
10. PROGRAM ELEMENT NUMBER
Use the program element number under which the report was prepared. Subordinate numbers may be included in parentheses.
11. CONTRACT/GRANT NUMBER
Insert contract or grant number under which report was prepared.
12. SPONSORING AGENCY NAME AND ADDRESS
Include ZIP code.
13. TYPE OF REPORT AND PERIOD COVERED
Indicate interim final, etc., and if applicable, dates covered.
14. SPONSORING AGENCY CODE
Leave blank.
15. SUPPLEMENTARY NOTES
Enter information not included elsewhere but useful, such as: Prepared in cooperation with, Translation of, Presented at conference of,
To be published in, Supersedes, Supplements, etc.
16. ABSTRACT
Include a brief (200 words or less) factual summary of the most significant information contained in the report. If the report contains a
significant bibliography or literature survey, mention it here.
17. KEY WORDS AND DOCUMENT ANALYSIS
(a) DESCRIPTORS - Select from the Thesaurus of Engineering and Scientific Terms the proper authorized terms that identify the major
concept of the research and are sufficiently specific and precise to be used as index entries for cataloging.
(b) IDENTIFIERS AND OPEN-ENDED TERMS - Use identifiers for project names, code names, equipment designators, etc. Use open-
ended terms written in descriptor form for those subjects for which no descriptor exists.
(c) COSATI FIELD GROUP - Field and group assignments are to be taken from the 1965 COSATI Subject Category List. Since the ma-
jority of documents are multidisciplinary in nature, the Primary Field/Group assignment(s) will be specific discipline, area of human
endeavor, or type of physical object. The application(s) will be cross-referenced with secondary Field/Group assignments that will follow
the primary posting(s).
16. DISTRIBUTION STATEMENT
Denote releasability to the public or limitation for reasons other than security for example "Release Unlimited." Cite any availability to
the public, with address and price.
19. &20. SECURITY CLASSIFICATION
DO NOT submit classified reports to the National Technical Information service.
21. NUMBER OF PAGES
Insert the total number of pages, including this one and unnumbered pages, but exclude distribution list, if any.
22. PRICE
Insert the price set by the National Technical Information Service or the Government Printing Office, if known.
EPA Form 2220-1 (9-73) (Reverse)
------- |