EPA-600/2-76-026
February 1976
Environmental Protection Technology Series
EFFECT OF
LIBEABY
ff.
I
Environmental Sciences Research
Office of Research and
U.S.
Research Triangle Park,
lllll
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2 Environmental Protection Technology
3 Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ECOLOGICAL RESEARCH series. This series
describes research on the effects of pollution on humans, plant and animal
species, and materials Problems are assessed for their long- and short-term
influences Investigations include formation, transport, and pathway studies to
determine the fate of pollutants and their effects This work provides the technical
basis for setting standards to minimize undesirable changes in living organisms
in the aquatic, terrestrial, and atmospheric environments.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/2-76-026
EFFECT
OF GASOLINE ADDITIVES
ON GASEOUS EMISSIONS
(PARTH)
by
R. W. Hurn, F. W. Cox, and J. R. Allsup
Fuel/Engine Systems Research Group
Bartlesville Energy Research Center
Energy Research and Development Administration
Bartlesville, Oklahoma 74003
Interagency Agreement No. EPA-IAG-D4-0040
and
Interagency Agreement No. EPA-IAG-D4-0453
EPA Project Officer: J. E. Sigsby, Jr.
Environmental Sciences Research Laboratory
Office of Research and Development
U. S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
J ^vlfw.'^vL^T.-L >ot<- "
OFFICE OF. RESEARCH AM^BEVELOPMENT
U.S. ENVrfeNMEOTSL PR'OTECTION AGENC
WASHINGTON, D.C. 20460
February 1976
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publication.
Approval does not signify that the contents necessarily reflect the views and
policies of the U.S. Environmental Protection Agency, nor does mention of
trade names or commercial products constitute endorsement or recommendation for
use.
-11-
-------
CONTENTS
List of Figures iv
List of Tables v
I Introduction 1
II Conclusions 3
III Experimental Apparatus 5
IV Experimental Design 9
V Vehicle Malfunctions .13
VI Results and Discussion 14
VII References 26
Appendix A. - Raw Emissions Data , 39
-iii-
-------
LIST OF FIGURES
Number
1 City Route Driven for Mileage Accumulation. ............ ....... . .... 27
2 Chromatographic System for Analysis of Nitrogen Compounds .,, ....... 27
3 Exhaust Analysis for Nitrogen Compounds --Carbopack Column..., ..... . 28
4 Exhaust Analysis for Nitrogen Compounds --Chromos orb Column. ........ 28
5 Nitrogen Compound Emissions --Mazda Engine .......................... 29
6 Nitrome thane Emissions ...................... . . . .................... 30
7 Nitroethane Emissions. ....................... , .......... . ...... ,... 31
8 Hydrogen Cyanide Emissions .................... . .................... 32
9 Effect of Mileage Accumulation on CO, HC, and NO Emissions for
the Volkswagen. .... .......................................... .... 33
10 Effect of Mileage Accumulation on CO, HC, and NO Emissions for
the Ford ................ , ..................... * .................. 33
11 Effect of Mileage Accumulation on CO, HC, and NO Emissions for
the Chevrolet [[[ 34
12 Effect of Mileage Accumulation on CO, HC, and NO Emissions for
the Mazda Vehicle.... ..................... „ ................... ... 34
13 Effect of Mileage Accumulation on CO, HC, and NO Emissions for
the Stationary Mazda ............................................. 35
14 Effect of Mileage Accumulation on Nitrogen Compound Emissions for
the Volkswagen ............................. . ..................... 35
15 Effect of Mileage Accumulation on Nitrogen Compound Emissions for
the Ford ........... . .............. . .............................. 36
16 Effect of Mileage Accumulation on Nitrogen Compound Emissions for
the Chevrolet ...... . .......... . ..... . ...... „ „ ...... . ....... ...... 36
17 Effect of Mileage Accumulation on Nitrogen Compound Emissions for
the Mazda Vehicle. . . . . . ....... . ...... „....,..,..,,,, ...... . ......... 37
18 Effect of Mileage Accumulation on Nitrogen Compound Emissions for
the S tationary Mazda „ ........... „....<> ..... „ ........... » » . . . ..... 37
-------
I
LIST OF TABLES
• Number Page
_ 1 Inspection Data for Test Fuel. ., 7
1
2 Inspection Data for High Aromatic Test Fuel 8
I 3 Independence of Emissions With and Without Additive................. 20
A-l Raw Emission Data (Vehicle No. 64, Volkswagen) 40
I A-2 Raw Emission Data (Vehicle No. 68, Ford Torino) 43
A-3 Raw Emission Data (Vehicle No. 67, Chevrolet). 46
I
I
I
I
I
I
I
I
I
I
I
I
A-4 Raw Emission Data (Vehicle No. 66, Mazda). 49
A-5 Raw Emission Data (Mazda Stationary Engine) 52
-v-
-------
SECTION I
INTRODUCTION
The continuing desire for lower and lower exhaust emissions demands that engines
be operated in a narrow range of engine adjustment; and as lower emissions are
required, the available range of engine adjustment becomes increasingly narrow.
This demand for invariant engine alignment creates a great need to keep deposits
from forming in sensitive areas such as the carburetor and intake manifold. The
obvious way to keep these sensitive areas clean is to use an effective cleaning
agent or additive in the fuels. Various suppliers have formulated many fuel
additives that effectively keep these sensitive areas clean, but have given little
attention to what happens to the additives during combustion and subsequent
emission as exhaust.
No standard procedure has yet been specified for testing the effect of fuel
additives in keeping engines clean, and no chemical procedures are available
for determining the amount or character of any additive-related materials that
may be emitted in the exhaust.
OBJECTIVE
The Environmental Protection Agency (EPA) has contracted with the Bartlesville
(Okla.) Energy Research Center (BERC) of the Energy Research and Development
Administration (ERDA) to develop a basic methodology for the standardization of
test procedures involving fuel additives, and to supply information relating
the effects of fuel additives upon pollutants emitted by late-model, spark-
ignition, reciprocating and rotary engines. The BERC has performed investigations
under three separate interagency agreements. The results of work performed under
the first agreement, No. EPA-IAG-097(D), have been reported to EPA under the
title,"Effect of Gasoline Additives on Gaseous Emissions"1. Briefly, the objec-
tive of that study was to establish a methodology for testing the effects of
fuel additive. The program included tests with a nonmetallic, multifunctional
cleaning additive and a metallic, octane-improving additive used in conjunction
with five reciprocating engines (two stationary engines and three vehicles).
The results of work performed under the second and third agreements are presented
in this report. The effect of gasoline additives upon the emission of pollutants
from reciprocating engines and rotary engines was investigated. Although the
engine types tested under the two agreements differ considerably, the program
goals were similar, and separate reports would be redundant. Consequently, no
differentiation is made between elements of the two agreements.
The purpose of the study was to determine whether additive fragments or additive-
related derivatives appear in the exhaust as a result of the presence of various
additives in the fuel and whether the use of these gasoline additives directly
effects the character and/or composition of normally emitted exhaust components.
Toward this objective, the number of fuel additives tested was increased to six,
and each additive was tested with four 1974 model vehicles. Vehicle and engine
classes represented were: (1) one economy vehicle with air-cooled engine, (2) two
vehicles with medium-sized engines representing a high, nationwide population
Effect of Gasoline Additives on Gaseous Emissions. Environmental Protection
Technology Series, Report No. EPA-560/2-75-014, 1974, 64 pp.
-1-
-------
percentage, and (3) one medium-to-light vehicle with rotary engine. In addition,
extended mileage tests were made using three of the six additives with a 1973
rotary engine mounted on a test stand.
As a result of the study, analytical methods developed specifically for nitrogen
compounds have been improved, and some insight has been gained concerning the
stability of several nitrogen compounds in the presence of auto exhaust,,
-2-
-------
SECTION II
CONCLUSIONS
This study was designed to produce test data from which conclusions could be
drawn concerning the effect of gasoline additives upon the emissions of pollu-
tants from spark ignition automotive engines. The scope of the study was as
broad as practical and included tests of several additives with both recipro-
cating and rotary engines. Periodic engine adjustment checks were made to
ensure their stability throughout the program.
Analytically, both "routine" and specific exhaust measurements were made in an
effort to determine the fate of the additive material. Methods for the routine
measurements (CO, C02, HC, NOX, and aldehydes) are well established but prelim-
inary exhaust spiking experiments with the compounds included in the specific
measurements (nitrogen containing compounds other than NOx) were necessary to
eliminate repeated, nonproductive analyses.
The following are conclusions derived from the information generated by this
study.
In the presence of auto exhaust, the nitrogen compounds proposed for analysis
fall into one of four categories:
1. Unstable and not detectable—ammonia, alkyl and aryl amines, pyridines,
alkyl (and probably aryl) nitriles, and dialkyl N-nitrosoamines.
2. Stable but not produced in detectable, quantities--aryl nitro compounds.
3. Unstable but usually detectable--hydrogen cyanide and cyanogen.
4. Stable and produced in detectable quantities--alkyl nitro compounds„
Aryl nitriles were not detectable in the exhaust samples, but since their insta-
bility was not determined, nondetectability may have been the result of either
decomposition or extremely low emission levels. The apparent instability of
hydrogen cyanide and cyanogen may have been caused by analytical inadequacies.
At the commercially recommended dosages, the presence of the tested additives
in gasoline would not have an immediate, measurable effect on the emission levels
of carbon monoxide (CO), hydrocarbon (HC), oxides of nitrogen (NOX), and aldehydes
unless several active fragments were produced by each additive molecule. The
recommended dosages are too small to produce (directly or indirectly) these
exhaust components in quantities large enough to measurably change the normal
exhaust levels unless some synergistic mechanisms were involved. Conversely,
additive fragmentation or action within the exhaust could conceivably effect
the much lower emission levels of CH3N02, CH3CH2N02, HCN, and NCCN.
The data, however, presented no evidence that the presence of any of the additives
in the fuel had an immediate effect upon the emission level of any of the measured
exhaust components. Though the fate of additive nitrogen was not determined
directly, it could, as stated previously, all appear as NOX without measurably
changing the exhaust concentration.
-3-
-------
An inverse relationship was found to exist between NO and nitroraethane exhaust
levels of the vehicles involved in the program„ The relationship of these
emissions from the stationary Mazda engine did not conform to that established
for the vehicles. Additional data points with NOX emission levels in the ranges
0 to 5 and 9 to 14 grams/test are needed to completely establish the
relationship.
Developmental experimentation specifically with cyanogen and hydrogen cyanide is
needed to establish the source of, and correct the analytical nonreproducibility
associated with these compounds0
-4-
1
I
1
I
I
-------
SECTION III
EXPERIMENTAL APPARATUS
EXHAUST SOURCES
The vehicles, engines, and transmissions used in the program were as follows:
Vehicle Engine Transmission
1974 Ford sedan.. „ 351-CID. ....... automatic
1974 Chevelle sedan „... 350-CID automatic
1974 Volkswagen sedan .1,500 CC standard
1974 Mazda sedan rotary 2X35-CID automatic
1973 Mazda engine „ .rotary 2X35-CID .automatic
The 1973 Mazda engine was mounted on a test stand and coupled to an eddy-current
dynamometer with an inertial system to simulate actual vehicle driving.
The prototype staged-combustion or stratified-charge engine and the prototype
1975 catalyst-equipped medium sedan specified in the original work plans were
omitted from the program by mutual consent of EPA and BERC.
ADDITIVES
The fuel additives specified in the program were all nonmetallic additives and
are described below:
1. Chevron F310 (polybutene amine in a carrier oil) is a multifunctional
cleaning additive and deposit modifier. F310 dosage was 1,232 lb/1,000 bbl of
gasoline.
2. DuPont DMA4 (amine neutralized alkyl phosphate) is a multifunctional
cleaning additive and controls carburetor deposit formation,, DMA4 dosage was
15 lb/1,000 bbl of gasoline.
3. Lubrizol 8101 (succinamid) is a multifunctional dispersant-type additive,
8101 dosage was 140 lb/1,000 bbl of gasoline.
4. Texaco TFA 318 (polyisopropylene carrier oil) primarily controls induc-
tion system deposit buildup and especially prevents the adherence of deposits
to the intake valve tulip. TFA 318 dosage was 220 lb/1,000 bbl of gasoline.
5. DuPont DMA 51 (carboxylate) is a multifunctional cleaning additive and
deposit modifier. DMA 51 dosage was 15 lb/1,000 bbl of gasoline.
6. Lubrizol 8101 and Texaco TFA 318 described above were used in combina-
tion with a dosage of 140 Ib Lubrizol 8101 plus 220 Ib of Texaco TFA 318 per
1,000 bbl of fuel.
-5-
-------
FUELS
The primary, unleaded fuel for the program was supplied by EPA. The high aroma-
tic fuel used in the program was a blend of the EPA fuel and a heavy platformate
stock. Inspection data for the fuels are presented in tables 1 and 2.
-6-
I
I
1
1
1
I
-------
I
I
I
I
I
I
I
I
I
I
1
I
I
I
I
I
t
I
I
TABLE 1. - Inspection data for test fuel
Distillation ASTM D-86
Evaporated, pet
5
10
20
30
40
50
60
70
80
90
95
0 F
112
122
141
160
172
194
210
227
240
285
341
Reid vapor pressure 9.1psia
Specific gravity 0.7334
FIA analysis, pet
Aromatic 23
Olefin 7
Saturate 70
Mole fraction summation
Carbon No.
1
2
3
4
5
6
7
8
9
10
11
Total
Paraffins
0
0
0
0.0716
.3795
.0532
.0642
.1244
.0219
.0056
.0023
0.7227
Olef ins
0
0
0
0
0.0096
.0128
.0270
.0035
.0024
0
0
0.0553
Aromatics
0
0
0
0
0
0.0012
.1436
.0469
.0208
.0096
0
0.2221
-7-
-------
TABLE ?. - Inspection data for high aromatic test fuel
Distillation ASTM D-86
Evaporar-ed, pet
5
10
20
30
40
50
60
70
80
90
95
n F
128
140
162
132
200
216
222
240
252
290
338
Specific gravity 0.7519
FLA analysis, pet
Aromatic 30
Olefin 7
Saturate 63
Mole fraction summation
Carbon No.
Paraffins
1 j 0
2 I 0
3
4
~t
6
1
F
q
10
11
0
0.0319
.1195
.1467
.0754
.1581
.0212
.0074
.0019
Total i 0,5621
Clef ins
0
0
0
0
0.0096
.00;)Q
.0327
.0047
.0052
. 0007
0
C.0619
Aromatics
0
0
0
0
0
0.0015
.2506
.0736
.0331
.0173
0
0.3761
-8-
-------
SECTION IV
EXPERIMENTAL DESIGN
The vehicles were acquired new and driven in moderately severe highway driving
for 2,000 miles using clear test fuel to "break in" the engine and stabilize
emission levels. The engines were adjusted to factory specifications and checked
regularly to ensure adjustments stability. Other than the specific items referred
to later in this report, no engine adjustments were changed or any items replaced
(other than regular oil changes) during the test program,,
All vehicles were "soaked" before testing in an appropriate 75° F .^rea: hovever,
the Ford and Chevrolet vehicles were tested on a large roll chassis dj'namoraeter
at ambient temperature, Efforts were made to run the test with a minimum elapsed
time between the 75° F soak area and the uncontrolled ambient temperature area.
The Volkswagen and Mazda vehicles were both "soaked" and tested in a controlled
environment of 75° F. The stationary engine while in an uncontrolled ambient
area used controlled temperature intake air.
Three separate routes or duty cycles were chosen for the different segments of
the program. Each vehicle was fitted with a recording tachograph to ensure
proper route profiles. The three routes were:
10 The city route, shown in figure 1, was chosen to simulate the driving
cycle of the Federal test procedure. The city route contains the same number
of stops, same average speed, and a similar 55 to 57 mph portion a? tbo rederal
driving cycle and requires 20 minutes to complete.
2. The combined city and highway route consisted of 1 hour spent on the
city route described above followed by 1 hour of highway driving at an average
speed of 55 mph, resulting in an overall average speed of about 35 mph. The
highway portion involved a round trip from Bartlesville, OK to Pawhuska, OK,
some 50 miles per trip.
3. The highway route (duty cycle) was used for the test stand ermine only
and consisted of 50 mph constant speed at road load.
Data points for each additive for the reciprocating-engine vehicles f'Ford, Chev-
rolet, and Volkswagen) consisted of a single test point with the clear fuel
followed by duplicate test points with the additive-treated fuel. Additional
single tests were conducted at 500 miles and 1,500 elapsed miles with the additive-
treated fuel followed by a single test with the high aromatic fuel treated with
the additive. The next series of tests for the next additive was t-oo.n begun
without further engine conditioning. The driving cycle for the rec i^roopt-ing-
engine vehicles consisted entirely of the repetitive city routes previously
described. All six fuel additives were used in each reciprocating-^ngire vehicle.
The rotary-engine vehicle tests consisted of a single data point with the clear
fuel followed by duplicate data points with the additive-treated fuel. Additional
single data points were collected at 1,000; 2,000; and 3,000 elapsed miles with
the additive-treated fuel. The final data point consisted of a sirglo fest with
the clear fuel. The vehicle was then conditioned with clear fuel for 1,000 miles
of moderately severe highway driving before the series of tests with the next
-9-
-------
additive was begun. The combined city and highway driving route was used with
the rotary-engine vehicle to accumulate mileage while using the fuel additives.
All six fuel additives were used in the rotary-engine vehicle.
Extended mileage-accumulation tests were conducted using a rotary engine mounted
on a test stand -md coupled to a dynamometer and an inertia system. A series
of tests for a single additive consisted of a single test with clear fuel at the
start, immediately followed by duplicate tests with the additive-treated fuel.
Comparison '-e^t-, one test with clear fuel, and one test with additive-treated
fuel, were conducted at 1,000; 3,000; 9,000; and 15,000 elapsed miles. Single
tests with the additive-treated fuel were conducted at 6,000 and 12,000 elapsed
miles. The engine vas then conditioned for 1,000 miles with clear fuel before
tests with the next additive were begun. The highway duty cycle was used for all
accumulation t.;ork with the stationary rotary engine. Three of the six fuel addi-
tives were used for the extended mileage tests with the stationary rotary engine.
ROUTINE ANALYSES
The 1975 Federal test procedure8 was used on all vehicular and engine testing.
Analytical methods for determining exhaust components included in the program
and considered to be routine are:
1o Total hydrocarbon by flame ionization detection (FID)--Beckman 400.
2. Nitro?c?ti dioxide (N02) and oxides of nitrogen by chemiluminescence—
Thermo Electron 10A.
3. Carter monoxide and carbon dioxide (C02) by nondispersive infrared
(NDIR) absorption--3ecknan 315.
4. Total aldehydes by 3-methyl-2-benzot.hiazolone hydrozone (MBTH) colori-
metry3—Spectrcnic 20.
The sample? for tn'inl aldehyde analysis were metered directly from the constant
volume sampling (TVS) system into the MBTP reagent solution. With this exception,
samples for all routine analyses were collected from the CVS system in light-
proof Tedlar hags.
ANALYSIS FOR *'ITR->^T COMPOUNDS
The basic mathrdo! og-7 for nitrogen compound analysis was developed as a part of
a previous strud-r , an;! initially, these analytical procedures essentially were
2U.S. Code of rod;-ra1 Regulations. Title 40--Protection of Environment; Chapter I —
Environmental ?r.->ruction Agency; Part S5--Control of Air Pollution from New Motor
Vehicles and Jtov ibtor Vehicle Engines. Federal Register, v. 39, No. 101,
May 23, 1974, pp. Ih07ft-18084.
Coordinating Research Council, Inc. Oxygenates in Automotive Exhaust Gas:
Part I. Techniques for Determining Aldehydes by the MBTH Method. Report No.
415, June 1968, 21 pp.
4 Effect of Gasoline Additives on Gaseous Emissions. Environmental Protection
Technology Series. Reoort No0 EPA-650/2-75-01&, 1974, 64 pp.
-10-
-------
duplicated. A PE-900 gas chromatograph was fitted with a Coulson electrolytic
conductivity detector and the appropriate column. Vapor samples were taken
directly from bags containing exhaust collected according to the 1975 Federal
test procedure and injected into the chromatograph via a 25 cm^ gas-sample loop.
Figure 2 shows a schematic of the analytical system. Separate injection systems
were installed on the chromatograph for basic or acidic compound analysis, and
each system was preconditioned with ammonia or hydrogen cyanide.
Three separate chromatographic columns provided the capability to separate and
distinguish the various nitrogen-containing compounds, and a fourth column was
ultimately used to obtain most of the nitrogen compound data.
Chromatographic conditions for the analysis of ammonia, light aliphatic amines,
and pyridine were:
1. Column: 10 feet by 1/8 in O.D. stainless steel tubing packed with 15 pet
Carbowax 600 plus 10 pet KOH on 80/100 mesh Gas-Chrom R.
2. Carrier: Helium flowing at 48 cc/min.
3. Temperature program: Hold at 25° C for 2 minutes, then program at 5° C/
min to 120° C.
Substances such as acetonitrile, pyrrolidine, and cyclohexylamine also can be
analyzed on this column.
Chromatographic conditions for the analysis of all of the preceding nitrogen
compounds (but with less resolution), N-nitrosoamines, nitrosoaromatics, nitro-
aromatics, aromatic nitriles, and aromatic amines were:
1. Column: 3 feet by 1/8 in O.D. stainless steel tubing packed with 15 pet
Carbowax 1540 plus 10 pet KOH on 80/100 mesh GC-22.
2. Carrier: Helium flowing at 52 cc/min.
3. Temperature program: Hold at 35° C for 2 minutes, then program at 6.5° C/
min to 180° C.
Molecular size for this column is limited to about Cg.
Initial chromatographic conditions for the analysis of cyanogen, hydrogen cyanide,
nitromethane, nitroethane, and acetonitrile were:
1. Column: 2-1/2 feet by 1/8 in O.D. stainless steel tubing packed with
Carbopack B treated with three to four drops of ^PO^.
2. Carrier: Helium flowing at 42-1/2 cc/min.
3. Temperature program: Hold at -70° C for 6 minutes, then program at
13° C/min to 180° C.
-11-
-------
Over 80 pet of the cyanogen, hydrogen cyanide, nitromethane, and nitroethane
data presented in this report was obtained using the following chromatographic
setup:
1„ Column: 8 feet by 1/8 in O.D. stainless steel tubing packed with
80/100 mesh Chromosorb 101.
2_ Carrier: Hydrogen flowing at 165 cc/min.
30 Temperature program: 0° to 180° C at 13° C/min, then purge isothermally
at 280° C for 3 to 5 minutes.
Using a Soxhlet apparatus, the Chromosorb 101 was extracted for 4 to 5 hours
with methanol then 1 to 2 hours with constant-boiling hydrochloric acid prior
to column packing.
Nitrogen compound detection was provided by a Coulson electrolytic conductivity
cell. Chromatographic effluent was fed into a quartz catalyst tube at 700° C
where the nitrogen compounds were reduced to ammonia. A nickel wire bundle,
about 4-1/2 inches in length, acted as the reduction catalyst.
-12-
-------
SECTION V
VEHICLE MALFUNCTIONS
Each reciprocating-engine vehicle had an incident worth noting. Approximately
800 miles into the Chevron F310 additive test, the Ford began making a tappet-
like noise; the source of the noise was found to be an untrue valve guide in the
engine headc The F310 tests were completed, the head with the faulty valve guide
was replaced, and tests were begun with the next additive.
A problem with the Volkswagen was encountered at about 500 miles into the test
with the Lubrizol 8101 fuel additive when a cylinder misfire was noted. The
misfire was caused by a loose tappet adjusting nut, and the result was a valve
that was not seating and a bent push rod. The push rod was replaced and the
valve readjusted^ The test was continued rather than repeated from the beginning
after an emission check showed the emissions to be normal.
The Chevrolet vehicle was involved in a minor accident at about 200 miles into
the test using Texaco TFA 318. The accident resulted in damage to the front
bumper and front fender. Exhaust emissions were not measurably affected;
therefore, the test was continued.
-13-
-------
SECTION VI
RESULTS AND DISCUSSION
The nitrogen compound classes to be analyzed were amines, pyridines, N-nitro-
soamines, nitriles, and nitro compounds. Individual nitrogen compounds included
were hydrogen cyanide and cyanogen. Of these compounds, only a few were found
to be present in exhaust at a detectable level. Analysis of exhaust samples
taken from the autos discussed in the Experimental Apparatus section of this
report gave peaks corresponding to (1) cyanogen, (2) hydrogen cyanide, (3) nitro-
methane, and (4) nitroethane.
Using the chromatographic conditions previously described for basic and neutral
compounds up to Cg, experiments were conducted in which light-proof bag samples
of CVS auto exhaust were spiked with compounds representative of the remaining
classes. The discussion of the results of these experiments is not offered as
proof that any particular nitrogen compound is not generated in the combustion
process, but as an indication of which compounds are likely to produce reliable
analytical results if generated in sufficient quantity.
The spiking experiments showed that most of the proposed nitrogen compounds are
unstable in auto exhaust with 30 pet or less ammonia, alkyl amines, aryl amines,
pyridine, and N-nitrosoamines remaining after 30 minutes. Acetonitrile seemed
to be somewhat more stable, losing about 60 pet in 60 minutes. The concentration
of nitrobenzene in exhaust remained stable for more than 60 minutes.
When mixed with exhaust, several of the nitrogen compounds produce reaction
products which are resolved by Carbowax 1540-KOH. The reaction products of
others appear as chromatographic smears, and in some cases, both peaks and smears
appear on the chromatogram. Some of the reaction products are relatively stable,
but generally they too decrease in concentration upon aging.
Since the objective of the spiking experiments was to determine the stability of
the nitrogen compounds when exposed to auto exhaust, only qualitative measurements
were made. The quantity of nitrogen compound injected into the exhaust sample
was several times that needed to produce a detectable level. This creates an
unrealistic situation with respect to the expected nitrogen compound levels in
exhaust, but destruction of large quantities of the nitrogen compounds indicates
the capacity of exhaust to reduce small quantities to levels below the detection
limit within relatively short periods.
Little effort was directed toward identifying the reaction products resulting
from the spiking experiments or determining the reaction mechanisms involved.
In some instances, more than one well-defined chromatographic peak appeared; in
others, only the destruction of the introduced compound could be followed. The
formation of N-nitrosoamines from the action of auto exhaust upon secondary
alkylamines is noteworthy. Dimethylamine and diethylamine were injected into
separate exhaust samples. The retention time of the major peak to appear in
each of the exhaust chromatograms agreed exactly with that of the corresponding
N-nitrosoamine. As previously stated, however, the N-nitrosoamine peaks were
transient.
-14-
-------
CVS exhaust samples from the vehicles involved in the program failed to show
detectable levels C above 0.025 ppm nitrogen atom) of the basic compounds, nitriles,
or aromatic nitre compounds „ Analysis with the Carbowax 1540-KOH column was,
therefore, discontinued.
The nitrogpn-crntr in'r,,t compounds proposed for analysis then fall into one of
these four cr^^'~-.'-i e~ :
1. Unstable ind -3*: detectable—ammonia, alkyl and aryl amines, pyridines,
alkyl (and probn-. LY a^yl) nitriles, and dialkyl N-nitrosoamines .
2. Stable but not produced in detectable quantities--aryl nitro compounds.
3. Unstable hut usually detectable--hydrogen cyanide and cyanogen.
&, Stable and produced in detectable quantities--alkyl nitro compounds.
The stability of a'ky1 nltri compounds, cyanogen, and hydrogen cyanide has not
yet been discussed, and >t is sufficient to say that nitromethane and nitroethane
analytical result wci relatively reproducible over a period of at least 1 to 2
hours. On the other hand, cyanogen and hydrogen cyanide are rather elusive, and
successive analyses seldom produce like results. Unlike those compounds which
are unstable and not detectable, the concentrations of cyanogen and hydrogen
cyanide do not clearly diminish with time hut may actually appear to increase.
No reasonable explanation can be offered by the investigators for an increase
of these materials in exlrmst standing at room temperature, and the nonrepeata-
bility has ^een attributed (at least in part) to inadequacies of the analytical
method. A? a nattar or" routine, the analytical cycle was kept as constant as
practicable for all samples with respect to instrumentation and sample age.
Precise procedure renlicrtion was not always possible or altogether successful,
as will bee -Tie inocrent 1'tter in the discussion.
One recc?.ni od analytical deficiency, which defied all attempts to recify, was
an intorf ornr.r e CT:,-I<' by water vapor in samples analyzed for hydrogen cyanide.
Fvsn molo.-.u .';• n ! VIT-" .-n containing water vapor r.avc a peak with a retention time
equivaie-.t ' ' that of hydrogen cyanida. This interference persisted regardless
of the column tT'r~ or "hrrnatographic parameters (carrier, flow rate, temperature
program) . The detection F* stem was essentially nitrogen compound specific, and
efforts to es^ah^i-h the source of this unorthodox behavior were unproductive.
The erratic bentr'l ^r of cyanogen and hydrogen cyanide, however, cannot be explained
in term? of ir.*-
-------
For about 1 month after exhaust testing was resumed with the Chromosorb 101
column, hydrogen cyanide and cyanogen values were exceptionally high and tended
to drop as the column aged. Overnight conditioning of the column produced much
higher hydrogen cyanide and cyanogen values on one occasion; and on another,
inadvertent injection of air into the hot column produced the same results. The
additives being tested with the vehicles during this period were primarily DMA4
and Lubrizol 8101 with high test values for the column conditioning and air
injection into the hot column occurring during the DMA51 and TFA 318 tests,
respectively. F310 was being tested with the stationary Mazda engine during
this period. The variability for hydrogen cyanide and cyanogen during the
latter part of the program was greater than would normally be expected, but the
day-to-day fluctuations were not nearly so great as they were when the Chromo-
sorb 101 column was new. The evidence then points to some analytical deficiency
being responsible for the highly anomolous hydrogen cyanide and cyanogen values
with sample stability possibly entering into the less radical value fluctuations.
The analytical difficulties encountered during the program have largely decreased
the value of the hydrogen cyanide and cyanogen information given in this report.
The levels of these compounds in exhaust can, at best, only be considered as
estimates, and any particular conclusions drawn must be viewed with a certain
amount of reserve. The evidence, however, does strongly suggest that these
compounds are commonly emitted auto exhaust components. The most likely route
in establishing the source (or sources) of the analytical variability would be
development of a method for direct hydrogen cyanide calibration and experimentation
with known quantities of the compounds exhibiting unusually high analytical
variability.
The chromatographic system for nitrogen compound analysis is illustrated in
figure 2. Examples of exhaust chromatography are presented in figures 3 and
4. The original retention times on the Carbopack B-HoPO/ column were up to
2 minutes longer, but were shortened over a period of several days by periodic
calibration with a methanol solution of nitromethane and nitroethane. Cyanogen
is below the detection limit in figure 3, but the retention time is indicated.
Oxides of nitrogen injected into the columns give backgrounds similar to those
in the figures.
A water solution of known quantities of nitromethane and nitroethane was prepared
for calibration when it became evident that the retention characteristics of the
Carbopack B-H^POA column were changing and methanol was suspected as the cause
of the change. There was no measurable difference in detector response to these
nitro compounds eluted from either the Carbopack B-^PO/^ (before peak broadening
occurred) or the Chromosorb 101 column. The mean and standard deviation for
detector response was calculated from all of the daily calibrations made during
the program and was found to be 4.05 x 10~10 ± 0,99 x 10 ~ ° nitrogen atom per milli-
volt. The noise level was 0.02mV-0.04mV and the detection limit (twice the noise
level) was about 2.5 x 10-11 nitrogen atom. Considering day-to-day fluctuations
of sensitivity and noise gives a limiting range of 1.2 x 10" - 4.0 x 10"
nitrogen atom.
The raw data for both routine and nitrogen compound measurements are given in
appendix A. All tests were made according to the 1975 Federal test procedure and
values reported for the individual bags and for the weighted composite. Units
-16-
-------
are grams/test for the individual bag samples and grams/mile for the composites.
Individual bag concentrations were calculated from the experimental data according
to:
RxFRxVx°RxMx 1630.55
bag concentration, g/test = (1)
where,
R = detector response (divisions)
•F*
R = response factor (moles/division)
V = standard volume of exhaust plus dilution air (cu ft/test)
°R = test temperature (degrees Rankine)
M = molecular weight (grams/mole)
P = barometric pressure (mm Hg)
Composites were calculated using the formula:
0.43(Bag;L, g/test) + Ba§2, g/test + 0.57(Bag3, g/test)
composite, g/mi = - - — - - (2)
/ • 3
When one, or more, of the three bag samples from a test contains an immeasurable
level of an exhaust component, a choice must be made concerning the calculation
of the composite (formula 2). An immeasurable level can be considered to be
zero or an estimate of the probable level can be made. Considering a component
that is normally found in one or more of the three bag samples, it is unlikely
that the level x^ill be absolutely zero in the remaining bag or bags. Also, it
is unlikely that one spark-ignition, internal-combustion engine will produce a
measurable quantity of a substance and another produce absolutely zero. There-
fore, maximum probable levels were estimated for bag samples falling in this
category before the composite sample values were calculated. A very small (less
than 0.04mV) but definite recorder deflection (cf - cyanogen, figure 4) has been
designated as trace (T), and no discernible recorder deflection at the retention
time of a component has been designated as below the detection limit (BDL). For
composite calculations, a trace level was estimated to be no more than 2.9 x 10"
nitrogen atom per test in sample bag No. 1 or 3 and no more than 4.9 x 10~ nitro-
gen atom per test in sample bag No. 2. These values are simply those giving 0.5
to 0.75 divisions of recorder deflection at 4.0mV full scale. When a sample
produced no definite recorder deflection for a component, these values were halved
for the composite calculation. The estimated T and BDL levels all fall below
the reported detection limit and are considered only as the maximum levels that
could have been present in the samples. Calculating composite values in this
manner allowed assignment of real numbers for the statistical analysis given in
table 3 and discussed later in this report.
-17-
-------
I
I
I
I
I
I
I
I
I
1
I
I
I
I
t
I
I
I
I
To determine the deleterious effect of a gasoline additive upon exhaust emissions,
consideration must be given to both immediate and long-range emission level and
composition. Immediate changes (from clear fuel to fuel with additive) in level
or composition may indicate a direct effect with the additive or additive pro-
ducts appearing in the exhaust, or indirect effects by altering the combustion
characteristics of the fuel or acting upon normally emitted products in the
exhaust. Long-range changes (extended use of fuel with additive) attributable
to an additive are indirect when viewed as the result of the additive's ability
to alter, form, prevent formation of, or remove engine deposits. However, the
immediate effect of the additive may change as deposits change.
All of the additives tested in this program are nonmetallic, engine-cleaning
agents, and the engines used were initially clean. Therefore, changes in
emissions should have resulted from the appearance of additive or additive pro-
ducts in the exhaust, additive action upon normally emitted products in the
exhaust, and/or deposit formation. Thus, with initially clean engines and no
control (a second engine operated exclusively on clear fuel) run in parallel
with the extended mileage test engine, program design places emphasis upon the
immediate effects of the additives even though these effects might be the result
of long-term additive use.
The highest level of nitrogen in the exhaust which could be derived solely from
additive was about 0.015 gram/test in the first bag. The nitrogen contents
of the various additives were obtained from the manufacturers. Using these
values and the dosages reported in the Experimental Apparatus section of this
report, the additive-nitrogen levels in the fuel were calculated as grams nitrogen
per liter fuel:
Additive N content of fuel
F310 0.01134
Lubrizol 8101 00513
Lubrizol 8101 + TFA 318 00513
DMM 00189
DMA51... 00127
Fuel consumption and CVS system output from a Mazda vehicle test were used to
calculate the maximum additive-derived nitrogen exhaust level reported above.
Assuming 100 pet corversion of the additive-nitrogen to a single nitrogen com-
pound in the exhaust, this level would give a deflection of about 25 divisions
on a 4mV full-scale recorder.
By comparing the lowest maximum possible emission level for additive-nitrogen to
the emission levels of the exhaust components (appendix A), it is obvious that addi-
tive-nitrogen apoearance as nitric oxide or NC>2, or additive action upon any of
the routinely measured exhaust components would not measurably affect their emis-
sion levels (except in the event that several active fragments were produced by
each additive molecule). On the other hand, emission levels of nitromethane,
nitroethane, hydrogen cyanide, and cyanogen could be affected to a large extent
if one or more of these compounds were additive-related reactants or end products
of the additive combustion process.
-18-
-------
There was no immediate additive-related effect upon emissions. This is shown
by comparison of emission levels obtained using fuel with additive to those
obtained using high aromatic fuel with additive and clear fuel. This information
is given in table 3. With the exception of those indicated as not being included
in the reduced data, all values in appendix A were used for the table 3 computa-
tions. For each group of values, the mean (x) is given by:
...
x = - (3)
where n is the number of values within the group. The standard deviation (a) of
each group is expressed as:
n
(4)
n - 1
and the test for the independence of two groups of values (t.:) was calculated
according to the formula:
A y (5)
'(n - 1) a 2 -+- (n - 1) c ?
x x y y
n + n - 2
X y
x and y representing the two groups of values being compared. In all cases, *
x is used to represent the values for the standard fuel with additive. Because
there were only two data points for clear fuel with each additive, two t^ values |
were calculated for the Mazda vehicle. All t^ values enclosed by parentheses {
compare all values from a particular engine using a specific additive to the
values from all tests of that engine using either clear or high aromatic fuel. •
Those t-[ values not enclosed by parentheses compare the additive values only to 1
the clear fuel values obtained during tests made using that additive. The
significance of t± increases as n for both groups being compared increases.
Therefore, the enclosed t^ values are the more significant. In all cases, n 1
for at least one of the two compared groups is as small as 4 to 7. For n values I
in this range, t^ shows some degree of independence between the two groups of
values when its absolute value is greater than approximately 2.5. Sign denotes I
the direction of deviation of the y group from the x group. |
-19-
-------
TABLE 3. - Independence of emission^ with and without additive
Additive
Carbon Monoxide
Emissions
1
Fuel with
additive
g/mi (mean 6. S.D.)
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(16.2) | ti(l&3)
Hydrocarbon
Emissions, g/mi (mean & S.D.)
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(l&2) ( ti(l&3)
VOLKSWAGEN
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
20.5+2.6
24.0+3.2
27.0+4.6
30.6+3.2
29.1+2.6
29.8+1.4
26.8+4.5
19.5
24.5
25.9
30.6
30.7
30.5
27.0+4.5
18.6
26.4
32.7
27.8
28.5
30.3
27.4+4.8
(-2.56)
(-1.14)
( .02)
( 1.38)
( .84)
( 1.18)
-0.07
(-2.59)
(-1.24)
(- .13)
( 1.15)
( .64)
( .94)
-0.28
2. 39+. 29
2. 51+. 14
2. 63+. 16
2. 63+. 15
2. 47+. 10
2. 58+. 09
2.53+.18
2.39
2.35
2.73
2.51
2.65
2.61
2. 54+. 15
2.38
2.78
2.76
2.49
2.46
2.76
2. 61+. 18
(-1.11)
(- .37)
( .91)
( .88)
(- .81)
( .50)
-0.09
(-1.48)
(- .93)
( -22)
( .18)
(-1.34)
(- .23)
-0.90
FORD
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
31 . 0+3 . 6
29.6+3.9
36.6+3.5
25.8+1.2
26.7+1.9
32.3+2.0
30.3+4.5
39.6
24.3
38.4
37.0
30.2
35. A
34 . 2+ 5 . 8
39.7
37.6
32.6
30.3
34.6
32.4
34.5+3.5
(-0.95)
(-1.35)
( .75)
(-2.78)*
(-2.42)
(- .59)
-1.75
(-1.53)
(-2.06)
( -92)
(-4.70)*
(-4.01)*
(-1.12)
-1.69
2. 61+. 12
2. 56+. 22
2. 87+. 34
2. 44+. 24
2. 44+. 27
2. 35+. 13
2. 54+. 27
2.74
2.19
2.95
2.69
2.51
2.64
2.62+. 26
3.17
2.97
2.14
2.63
2.41
2.30
2. 60+. 40
(-0.05)
(- .37)
( 1.31)
(-1.13)
(-1.09)
(-1.92)
-0.63
( 0.04)
(- .18)
( 1.07)
(- .74)
(- .73)
(-1.20)
-0.44
CHEVROLET
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
37.7+5.1
40.0+7.5
46.0+7.3
36.5+4.3
35.3+4.1
42.0+6.3
39.6+6.4
34.0
59.7
50.3
47.7
47.5
38.1
46.2+9.1
33.4
51.0
56.4
40.5
35.6
43.7
43.4+8.9
(-1.69)
(-1.13)
(- .04)
(-1.97)
(-2.22)
(- .80)
-2.09
(-1.16)
(- -63)
( .48)
(-1.44)
(-1.70)
(- .28)
-1.23
1.21+.19
1.24+.35
1.30+.19
1.13+.12
1.36+.27
1.22+.22
1.24+.22
1.01
1.30
1.49
1.27
2.08
1.09
1.37+.39
0.93
1.51
1.50
1.49
1.02
1.24
1.28+.26
(-0.80)
(- .58)
(- .36)
(-1.21)
(- .08)
(- .71)
-1.13
(-0.50)
(- .24)
( .10)
(-1.08)
( .43)
(- .39)
-0.40
MAZI
F310
DMA 4
LUB 8101
DMA51
TFA318
8101+318
Total
22.5+1.8
18.1+3.4
20.0+1.7
21.3+6.1
22.3+1.8
22.4+3.0
21.1+3.4
21.6+0.9
18.2+0.1
20.0+0.1
28.9+11.8
20.7+3.8
19.4+2.1
21 .4+5.3
0.71(0.44)
- .03(-1.29)
.06(-0.58)
-1.19(-0.04)
.84(0.37)
1.30(0.38)
-0.23
)A VEHICLE
2. 40+, 20
1.81+.37
1.78+.10
1.59+.26
2. 08+. 23
1.90+.17
1.93+. 34
2. 15+. 15
1.76+.16
1.74+.21
2. 10+. 58
2. 00+. 07
1.69+.11
1.91+. 28
1.60(3-60)*
.20(-0.57)
.33(-1.00)
1.75(-2.18)
.44(1.21)
1.55(-0.05)
0.18
STATIONARY MAZDA
F310
DMA4
LUB 8101
Total
24.2+4.3
19.0+4.0
24.7+3.2
22.7+4.5
20.6+4.1
18.9+4.8
22.4+1.6
20.6+3.8
1.58
.01
1.44
1.47
3. 03+. 35
2. 41+. 43
2. 99+. 80
2. 82+. 59
2. 82+. 53
2. 32+. 29
2.55+. 37
2. 58+. 44
0.89
.42
1.16
1.37
-'•No overlap of standard deviations.
( ) Each additive compared to total value for columns 2 and 3.
-20-
-------
TABLE 3. - Independence of emissions with and without additive
Continued
Additive
Nitrogen Oxides
Emissions, g/mi (mean & S.D.;
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(l&2) I ti(l&3)
Aldehydes
Emissions, g/mi (mean & S.D.)
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(l&2) |ti(l&3)
VOLKSWAGEN
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
3. 17+. 30
3. 70+. 52
4. 09+. 11
4. 93+. 22
4. 18+. 27
4. 35+. 31
4. 07+. 62
3.19
3.88
3.88
4.19
4.45
3.95
3. 92+. 42
-__ _ _
4.13
4.18
3.86
4.47
3.96
4.71
4. 22+. 32
(-3.07)*
(- .74)
( .77)
( 4.34)*
( 1.06)
( 1.73)
0.54
(-5.20)*
(-1.97)
(- .75)
( 3.86)*
(- .21)
( -65)
-0.56
0.071+.012
.079+. 013
.086+. 009
.104+. 012
.081+. 004
.072+. 010
.082+. 015
0.077
.070
.084
.093
.088
.067
.080+. 010
0.082
.076
.082
.099
.065
.048
.075+. 017
(-1.19)
(- .15)
( -92)
( 3.45)*
( .21)
(-1.20)
0.36
(-0.40)
( .33)
( 1.09)
( 2.88)*
( .631
(- .35)
0.98
FJJ.O
DMA 4
LUB 8101
DMA51
TFA318
8101+318
Total
3. 13+. 22
3. 63+. 39
3. 70+. 27
3. 87+. 54
3. 57+. 26
4. 00+. 32
3.65+. 41
3.09
2.60
4.56
3.76
3.22
3.84
3. 51+. 69
3.47
3.94
2.83
3.75
4.00
4.18
3. 70+. 49
(-1.05)
( .32)
( .51)
( -86)
( .17)
( 1.31)
0.64
(-2.13)
(- .21)
( .01)
( -52)
(- .45)
( 1.09)
-0.23
0. 106+.034
.140+. 018
.122+. 003
.152+. 007
.145+. 01 7
.115+. 017
.130+. 024
0.109
.123
.125
.115
.151
.117
.123+. 015
0.131
.093
.113
.150
.113
.118
.120+. 019
(-1.14)
( 1.59)
(- .15)
( 3.62)*
( 2.08)
(- .85)
0.65
(-0.83)
( 1.66)
( .20)
( 3.20)*
( 2.07)
(- .41)
0.98
CHEVROLET
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
1.84+.16
1.95+.12
1.85+.29
2. 03+. 17
1.88+.31
1.94+.42
1.91+.24
2.20
2.24
2.00
1.80
1.62
1.94
1.97+.24
2.02
1.71
1.86
1.92
2.14
2.01
1.94+.15
(- .95)
(- .13)
(- -70)
( .44)
(- .49)
(- -14)
-0.47
(-1.08)
( -07)
(- .68)
( .84)
(- .42)
(- -03)
-0.28
0. 120+.012
.117+. 014
.122+. 006
.129+. 020
.121+. 008
.103+. 012
.119+. 01 4
0.110
.086
.116
.128
.125
.119
.114+. C
15
0.112
.095
.087
.114
.105
.112
.104+. Oil
( 0.69)
( .31)
( .92)
( 1.34)
( .80)
(-1.24)
0.70
( 2.20)
( 1.62)
( 2.84)*
( 2.53)
( 2.55)
(- .19)
2.32
MAZDA VEHICLE
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
1.33+.16
1.30+.07
1.29+.12
1.18+.07
1.21+.03
1.25+. 04
1.26+.10
1.20+.11
1.27+.04
1.31+.06
1.18+.07
1.16+.02
1.23+. 03
1.22+.07
1 .08(1.15)
.52(1.99)
- .26(1.34)
- .07(-1.23)
2.72M-0.28)
.76(0.89)
1.14
STATIONARY MAZDA
*No overlap of standard deviations.
( ) Each additive compared to total value for columns 2 and 3.
0. 162+.016
.120+. 025
.141+. 018
.122+. 027
.162+. 025
.143+. 012
.142+. 026
U31+.003
.100+. 013
.120+. 013
.169+ .098
.151+. 022
.127+. 004
.133+. 038
2.53*(1.61)
1.07(-0.66)
1.53( 0.48)
-1.1H~0.56)
.58 (1.58)
1.74*(0.55)
0.89
F310
DMA4
Lub 8101
Total
0.98+.22
.72+. 06
.72+. 15
0.^,2 K20
0.96+.27
.69+. 06
.67+. 12
0.78+. 22
0.18
.81
.68
0.47
0. 202+.043
.161+. 047
.218+. 090
.192+. 063
Q.160+.029
.163+. 050
.223+. 086
.182+. 062
2.05
- .09
- .11
0.48
-21-
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
TABLE 3, - Independence of emissions with and without additiyg
Continued
Additive
HYDROGEN CYANIDE
Emissions g/mi (mean & S.D.)
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(16,2)
ti(l&3)
CYANOGEN
Emissions, g/mi (mean & S.D.)
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(lH)
ti(l&3)
VOLKSWAGEN
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0115±.0028
--
.0208
.0234±.0048
.0181±.0046
.0204±.0028
0.0185±.0056
0.0151
--
--
.0328
.0178
.0297
0.0239±.0087
--
--
0.0208
.0139
.0260
.0187
0.0199±.0050
(-2.69)*
--
--
(- .09)
(-1-03)
(- .77)
-1.54
(-2. 89)*
--
--
( 1.02)
(- .47)
( .17)
-0.44
O.OOOli.OOOO
.0022
--
.0025±.0014
.0006±.0002
,0009±.0008
0.0011±.0013
0.0001
--
-.
.0037
.0003
.0014
0.0014±.0017
--
._
0 .0028
.0010
.0029
.0001
0.0017±.0014
(-1.54)
__
--
( 1.06)
(- .66)
(- .57)
0.30
(-2.32)*
__
__
( .84)
(-1.11)
(-1.07)
-0.77
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0125±.0022
.0071
.0343±.0062
.0160±.0094
.0140±.0039
.0144+. 007 5
0.0163±.0090
0.0048
--
-_
.0065
--
.0087
0.0067±.0020
--
—
0.0114
.0107
.0257
.0100
0.0145+.0075
( 3.43)*
--
( 7.76)*
( 1.68)
( 2.92)*
( 1.71)
1.79
(-0.43)
--
( 3.17)'
( .24)
(- .09)
(- .01)
0.37
O.OOOli.OOOO
.0001
.0019
.0011±.0008
.0002±.0002
,0002±.0001
0. 000 5±. 0006
0.0001
—
—
.0009
.0025
.0001
0.0004±.0005
—
.-
0.0014
.0006
.0001
.0004
0.0006±.0006
(-1.00)
-.
—
( 1.38)
(- .59)
(- .73)
0.26
(-1.60)
--
--
( .94)
(-1.25)
(-1.51)
-0.45
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0114+.0019
.0028
.0196±.0068
.0087*. 0057
.0072±.0033
.0070±.0058
0.0094±.0060
0.0046
-_
--
.0033
.0104
.0120
0.0076±.0043
--
--
0.0076
.0049
.0073
.0032
0.0058±.0021
( 1.42)
—
( 2.77)*
( .30)
(- .14)
(- -18)
0.57
( 3.66)'1
--
( 4.16)<
( .96)
( .71)
( .39)
1.18
O.OOOli.OOOO
.0001
.0004
.0003±.0001
.0003±.0001
.0002±.0002
0.0002±.0001
0.0001
.-
--
.0004
.0002
.0001
0.0002±.0001
--
--
0.0006
.0002
.0001
.0001
0.0003±.0002
(-1.20)
--
--
( .79)
( .52)
( .21)
-0.30
(-1.07)
-.
-.
( .19)
( .00)
(- .16)
-0.30
MAZDA VEHICLE
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0014±.0001
.0040±.0007
.0093±.0021
.0084±.0040
.0083±.0030
.0039±.0013
0.0066±.0036
0.0176
.0025
.0055
.0100±.0033
.0050±.0010
.OOSOi.OOOO
0.0060±.0029
(-2.12)*
(- .92)
( 2.16)
- .48(1.19)
1.45(1.37)
-1.16(-1.50)
0.43
0.0001±.0000
.0003±.0003
.0003±.0000
.0002±.0000
.OOOli.OOOO
.0001±.0000
0.0002±.0001
0.0001
.0002
.0002
.0003±.0001
.0002±.0001
.OOOli.OOOO
0.0002±.0001
(-1.09)
( 1.03)
( 3.06)*
-1.92( .05)
-1.89(-1.76)
.00(-1.24)
0.20
STATIONARY MAZDA
F310
DMA4
LUB 8101
Total
0.0182±.0220
.0030±.0018
.0027±.0013
0.0029±.0015
0.0079±.0081 I/
.0014±.0008
.0025±.0014
0.0019±.0012
0.98
1.93
.28
1.62
0.0003±.0002
.0001±.0000
.0001±.0000
O.OOOli.OOOl
0. 000 2±. 0002
.0001±.0000
.0001±.0000
O.OOOli.OOOl
0.73
.00
.00
0.47
;'-No overlap of standard deviations.
( ) Each additive compared to total valu
I/Not included in total.
for columns 2 and 3.
-22-
-------
TABLE 3. - Independence of emissions with and without additive
Continued
Additive
NITROMETHANE
Emissions, e/mi (mean & S.D.)
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(l&2) |ti(l&3)
NITROETHANE
Emissions, e/mi (mean & S.D.I
1
Fuel with
additive
2
Clear
fuel
3
High
aromatic
fuel with
additive
Test for
independence
ti(l&2) | ti(l&3)
VOLKSWAGEN
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0026±.0012
.0036
.0040
.0042±.0009
.0030±.0007
.0030±.0003
0.0033±.0009
0.0041
--
.0035
.0038
.0027
.0039
0.0036±.0005
--
0 . 0040
.0020
.0028
.0020
.0029
0.0027±.0008
(-1.65)
—
--
( 1.23)
(-1.44)
(-2.06)
-0.71
(-0.15)
--
—
( 2.56)
( .56)
( .54)
1.29
0.0007+.0002
.0008
.0004
.0004±.OOQ1
.0004±.0001
.0004±.0000
0.0005±.0002
0.0010
—
.0005
.0004
.0004
.0004
0.0005±.0003
__
0.0005
.0004
.0004
.0004
.0004
0.0004±.0000
( 0.93)
__
(- .86)
(- .86)
(-1.06)
-0.56
( 3.10)*
__
( .16)
( .16)
(-1.00)
9.97
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0051±.0012
.0058
.0061±.0013
. 004 5±. 0009
. 004 3±. 0008
.0047±. 0004
0.0049±.0010
0.0037
-_
.0080
.0036
.0048
.0038
0.0048+.0019
--
—
0.0027
.0033
.0051
.0040
0.0038±.0010
( 9.18)
-_
( 1.04)
(- .27)
(- .48)
(- .08)
0.19
( 1.37)
__
( 2.65)*
( 1.04)
( .81)
( 1.66)
2.02
0.0010±.0002
.0006
.0006±.0002
.0004±.0000
.0004±.0000
.0004±.0001
O.C005±.0002
0.0008
__
.0014
.0004
.0004
.0004
0. 0007±.0004
--
__
0.0004
.0004
.0007
.0004
0.0005±.0002
( 0.80)
__
(- .33)
(-1.26)
(-1.26)
(-1.14)
-1.24
( 3.28)*
__
( .91)
(-1.00)
(-1.00)
(- .63)
0.37
CHEVROLET
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0.0081±.0008
.0132
.0136±.0032
.0118±.0008
.0113±.0017
.0109±.0020
0 .0113±.0024
0.0101
-_
.0208
.0094
.0121
.0138
0.0132±.0046
--
--
0.0123
.0093
.0124
.0112
0.0113±.0014
(-1.86)
--
( .13)
(- .54)
(- .79)
(- .95)
-1.35
(-3.40)-."
--
( 1.31)
( .50)
( .02)
(- .33)
-0.03
0. 0012±.0002
.0025
.0024±.0006
.0018-^.0007
.001S±.0003
.OOlHi.0005
0. 001 Q±. 0006
0.0012
--
.0041
.0014
.0017
.0027
D.0022±.0012
--
--
0.0025
.0014
.0022
.0017
0.0020±.0005
(-1.37)
--
( .28)
(- .59)
(- .63)
(- .60)
-0.98
(-2.38)*
--
( 1.19)
(- .30)
(- .42)
(- .34)
-0.28
MAZDA VEHICLE
F310
DMA4
LUB 8101
DMA51
TFA318
8101+318
Total
0. 009 9±. 0008
.0121±.0018
.0117±.0015
.0107±.0037
.0176±.0011
.0145±.0021
0.0131±.0033
0.0131
.0117
.0099
.0179±.0013
.0141±.0001
.0128±.0006
0.0138±.0027
(-1.98)*
(-1.02)
(-1.56)
-2.56*(-l .83)
4.03*(2.93)"
1.11 ( .53)
-0.53
3.0016±.0000
.00221.0002
.0019±.0003
.0019*. 0004
.0028±.0003
.0022+.0005
ja0021±.0005
i
0.0019
.0024
.0012
.0028±.0002
.0023±.0001
.0018±.0000
3.002 1±. 0005
(-1.41)*
( .11)
(- .99)
-2.83*(- .95
1.92*(2.53)'
1.04 ( .23)
0.01
STATIONARY MAZDA
F310
DMA4
LUB 8101
Total
0.0126±.0032
.OOS3±.0024
.0098±.0038
0.0093±.0040
X009^±.0039
.0059±.0009
. 008 0±. 0042
0.0079±.0035
1.23
.39
.77
1.04
0.0fi31±.f)009
.0013-t.0005
.0023 --.0009
O.OOrZi.P'HO
0.0023±.0008
.0012±.0003
.0022±.0010
0.0019±.0009
1.34
.43
.33
0.76
-^No overlap of standard deviations.
( ) Each additive compared to total value for columns 2 and 3.
-23-
-------
For an additive to show an influence upon the production of an exhaust component,
all (or at least the majority) of the engines tested should give significant t^
values (with the same sign) for that additive and that component when values
obtained using fuel with additive are compared to values obtained from using
clear fuel [t^l & 2)]. This is not the case. The F310 influence upon hydro-
gen cyanide production may at first glance appear to be real with three of five
engines giving t^ values with some degree of significance. However, several
values for F310 are missing from the raw data, and two of the three significant
ti values are negative while the third is positive. The same arguments essen-
tially negate a certain amount of independence shown for nitroethane and hydro-
gen cyanide production between the fuel with F310 and high aromatic fuel with
F310 [ti(l & 3)].
Bar graphs have been constructed for quick, unambiguous comparison of nitrogen
compound emission levels with and without the various additives tested. These
data are presented in figures 5 through 8. Values for the routine emission
measurements have been omitted from these graphs because of their relatively
high emission levels, and cyanogen, with exceptionally high standard deviations,
also has been omitted.
The duty and test cycles for the Mazda stationary engine were considerably
different from those for the vehicles. The data for the Mazda engine have,
therefore, been presented separately and in slightly different form from those
of the vehicles. In figure 5, the mean exhaust level and standard deviation
for each component from all tests with an additive are compared with those from
tests made with clear fuel when that particular additive was being tested0 The
number of samples is shown at the lower end of each bar. Data for the four
vehicles have been grouped for comparison of exhaust levels from the various
vehicles as well as exhaust levels from a particular vehicle using fuel with
additive or clear fuel. Each additive-labeled bar gives the mean value and
standard deviation for all tests made with that vehicle using that additive in
the fuel. (The tests with high aromatic fuel and additive are also included.)
Each bar designated by the word "clear" gives the mean value and standard devia-
tion for all tests made with that vehicle using clear fuel regardless of the
additive being tested. The number of test values involved in calculating the
mean and standard deviation is again given in each bar. Figure 6 through 8
show rather distinctive source differences for nitromethane and nitroethane
emission levels and, to a lesser degree, for hydrogen cyanide emission levels.
Also, including figure 5, it is obvious that the few additive test values that
differ significantly from the clear fuel test values offer no evidence of an
additive-related influence upon emissions.
The program was aimed primarily at determining immediate effects of several
commercially available gasoline additives upon emission levels or composition.
Changes in engine parameters may alter emissions serving either to create effects
that are independent of additive use or to mask effects of the additive. Except
in the special cases where malfunctions occurred, engine parameters were not
adjusted during the program. They were, however, checked periodically to assure
parametric consistency and to minimize the chance that anomolous exhaust levels
or composition might occur from parameter changes and obscure possible additive
effects. No radical changes in CO, HC, and NOX emission levels occurred during
the program. This is shown in figures 9 through 13 where emission levels have
been plotted as a function of mileage accumulation. Because no measurable
-24-
-------
immediate effect was found for any of the additives tested, the average emission
level was plotted when more than one test was made at a particular mileage.
Figures 11 through 13 show good overall constancy for CO, HC, and NOX emissions.
There were gradual increases in CO and NOX emitted by the Volkswagen (figure 9)
and in NOX emitted by the Ford (figure 10)„
Nitrogen compound emission levels, as a function of accumulated mileage, are
presented in figures 14 through 18. Here again, multiple test points were
averaged. Considering that emission levels were quite low and that the emission
scale was expanded, overall emission constancy for nitrogen compounds was good.
The relatively large fluctuations from 1,500 to 4,500 miles in figures 14 through
16 and from 0 to 15,000 miles in figure 18 reflect the analytical difficulties
previously discussed.
Aside from the question of additive influence upon emissions, the program data
show an interesting relationship between the exhaust levels of nitrogen oxides
and nitromethane. In all probability, this relationship could be extended to
include nitroalkyls as a class. Oxides of nitrogen vs CH3N02 was plotted for bag
No. 1 of each test. All such points from the reported tests are included in
figure 19. The relationship is apparent from the plotted data points; however,
it seems to be more clearly defined by reduced data scatter when the stationary
Mazda (points enclosed by the rectangle) is omitted. The best linear fit for the
data from the four vehicles is expressed by the least squares regression equation:
NOX = -202.13(CH3N02) + 17.919 (6)
With all 118 data points, the correlation coefficient, -0.855, shows a high
degree of significance for the relationship. The 95 pet confidence interval for
the prediction of nitromethane was found to be CH3N02 ± 0.002 g/test.
With the vehicles tested, very few NOX levels fell in the range, 9 to 14 g/test.
Also, the only emission source which gave NOj,- levels below 5 g/test was the
stationary Mazda, and these points do not conform to the relationship discussed
above. Additional data points obtained from sources emitting NOX in these ranges
are needed to determine whether the relationship is truly linear (the stationary
Mazda data suggest that it may not be).
-25-
-------
SECTION VII
REFERENCES
1. Environmental Protection Agency. Effect of Gasoline Additives on Gaseous
Emissions. Environmental Protection Technology Series, Report No. EPA-
560/2-75-014, 1974, 64 pp.
2. U.S. Code of Federal Regulations. Title 40--Protection of Environment;
Chapter I—Environmental Protection Agency; Part 85—Control of Air
Pollution from New Motor Vehicles and New Motor Vehicle Engines. Federal
Register, v. 39, No. 101, May 23, 1974, pp. 18076-18084,,
3. Coordinating Research Council, Incc Oxygenates in Automotive Exhaust Gas
Part I. Techniques for Determining Aldehydes by the MBTH Method. Report
No. 415, June 1968, 21 pp.
4. Environmental Protection Agency. Effect of Gasoline Additives on Gaseous
Emissions. Environmental Protection Technology Series. Report No. EPA-
650/2-75-014, 1974, 64 pp.
-26-
-------
KEY
• Stop sign AYield sign
B Signal light
Sunset Blvd
35 to 45 mph
55 mph
Hyd roge n
FIGURE 1. - City Route Driven for Mileage Accumulation
Hydrogen
Manga nous
oxide
T
i.
o
I n j ector
r
Sample ~\
mV lout
25ml L-
sample /|
loop [_ Sample valve j
i
Column oven
Injection
COULSON ELECTROLYTIC CONDUCTIVITY DETECTOR
FIGURE 2. - Chromatographic System for Analysis
of Nitrogen Compounds
-27-
-------
2 5 feetCorbopack B-H3P04
Carrier He, 42 cc/rninutes
Initial temp -7Q° for 6 minutes
Final temp I 80°C
Temp program l3°C/minute
10
TIME, minutes
FIGURE 3. - Exhaust Analysis for Nitrogen
Compounds—Carbopack Column
8 fee t Chromosorb IO I
Carrier H2 , 165 c c / m inute
Initial temp 0°C
Final temp 180° C
Temp program I3°C /minir
[-»• 4 mv F S
10 12
TIME, m i n u tes
FIGURE 4. - Exhaust Analysis for Nitrogen
Compounds--Chromosorb Column
-28-
-------
0.020
.0 I 8
.0 16
.0 14
0 12
.010
.008
.006
.004
.002
Nitromtthont Nitroethone Hydrogen Cyanide
r
—
—
fi
/
•s
V
^ f *\ f ^
I
R
1
7
/,
1
5
I
7
'/
I
S
T T DO
71 I | ~ rn
/ / / >; ~ ~ / 'X /i
'/ '/ 1 / ' \ \ 7 // /
55 71 Ul 75 2. 2. 7 [5| 75
/y /, ^1 VA //
-------
b^^^N^^^^
I^SS^^^^^
*<
o
6
o
o
Clear
8101 +318
TFA 318
DMA5I
LUB 8101
DMA4
F3IO
Clear
8101 + 318
TFA 318
DMA 51
LUB 8101
DMA4
F 310
Clear
8101 + 318
TFA 318
DMA 51
LUB 8101
DMA4
F3IO
Clear
8101 + 318
TFA 318
DMA 51
LUB 8101
DMA 4
F3IO
FIGURE 6. - Nitromethane Emissions
-30-
-------
KSNXSXSX->
Clear
8101+318
TFA3I8
DMA 51
LUB 8101
DMA 4
F3IO
C\»ar
8101 4318
TFA 318
DMA 51
LUB 810\
DMA 4
F 310
Clear
8101 + 318
TFA3I8
DMA 51
LUB 8101
DMA 4
F 310
Clear
8101 + 318
TFA 318
DMA 51
LUB 8101
DMA 4
F 310
w
8
H
M
53
o
o
o
o
o
aiiw/swvao
FIGURE 7. - Nitroethane Emissions
-31-
-------
!<
*<
u. \
No voluts
Cltor
8101 4318
TFA 318
DMA 51
LUB 810!
DMA4
F3IO
Clear
8101 4 318
TFA 318
DMASI
LUB 8101
DM A 4
F3IO
Clear
8101 4 318
TFA 318
DMASI
LUB 8101
DMA 4
F3IO
Clear
8101 4 318
TFA3I8
DMA 51
LUB 8101
DMA 4
F 310
8
q
t
O
31IW/SWVHS
FIGURE 8. - Hydrogen Cyanide Emissions
-32-
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
7 8
I
F 310
DMA 4
LUB8IOI DMA5I TFA3I8 8101+318
MILEAGE , thousands
FIGURE 10. - Effect of Mileage Accumulation on CO, HC,
and NO Emissions for the Ford
X
-33-
10
F 310 DMA4 LUB8IOI DMA5I TFA 318 8101+318
M IL EAGE , thousands
FIGURE 9. - Effect of Mileage Accumulation on CO, HC, and NO
Emissions for the Volkswagen x
-------
DS'A M
MILEAGE , 'h- u- inds
7
TFA~3l8~
10
8101 + 318
FIGURE 11. - Effect of Mileage Accumulation on CO, EC,
and NOV Emissions for the Chevrolet
4 t* f- 10
L _ i._ .J - - - ._ i . _l_
^IC "'.'A * LU8 8101
FIGURE 12. - Effect of Mileage Accumulation on CO, HC,
and NOV Emissions for the Mazda vehicle
-------
F 3IO
DMA 4
MILEAGE , thousands
LUB 8101
0.030
.025
1.020
015
FIGURE 13. - Effect of Mileage Accumulation on CO, HC, and
NOX Emissions for the Stationary Mazda
KEY
o Nitromethone
• Nitroethane
A Hydrogen cyanide
.010
.005 •
F 310
DMA 4
LUB 8101 DMA 51
MILE AGE , thousands
TFA 318
8101 +318
FIGURE 14. - Effect of Mileage Accumulation on Nitrogen
Compound Emissions for the Volkswagen
-35-
-------
0 035
Ford
KEY
o NHromethone
• Nitroethane
A Hydrogen cyanide
F 310
DMA 4
LUB 8101 DMA 51
MILEAGE , thousands
TFA 318
8101 +318
FIGURE 15. - Effect of Mileage Accumulation on
Nitrogen Compound Emissions for the Ford
0.025p
KEY
O Nitromethone
• Nitroethone
i Hydrogen cyanide
F 310 DMA 4 LUB 8101 DMA 51 TFA 318
MILEAGE , thousands
8101 + 318
FIGURE 16. - Effect of Mileage Accumulation on Nitrogen
Compound Emissions for the Chevrolet
-36-
-------
Mazdo vehicle
° Nitromethane
• Nitrotthane
& Hydrogen cyanide
005
O I 2 3 4 5 6 7 8 9 10 I I 12 13 14 15 16 17 18 19 20 21 22 23 24 25
MILEAGE, thousands
FIGURE 17. - Effect of Mileage Accumulation on Nitrogen
Compound Emissions for the Mazda Vehicle
DMA 4
MILEAGE , thousonds
FIGURE 18. - Effect of Mileage Accumulation on Nitrogen
Compound Emissions for the Stationary Mazda
-37-
-------
Bag
\
20
KEY
Individual test points
• — Area of stationary Mazda test points
17919 -NOX ,
CHJN02- + 022 g/test
202.13
(least squares regression equation)
correlation coefficient = -0 855
™^— 95 Percent confidence interval
• N •
.•••••v
15
I 0
02 .03 04 .05 06 .07 .08 .09 .10
\
.01
FIGURE 19. - Relationship Between NOX and CH3N02 Emission Levels
-38-
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
APPENDIX A. - RAW EMISSIONS DATA
-39-
-------
o o o o
en CM r*- oo
O so ui .-I
•-> O O O
oa to PQ o
pa po CD o
oo co en o
-- o o c
f*-. CM r~~
m co i-i £N
CM in O CM
1-1 O .-i O
CO 0) O I-H
O O i-i O
O co O O
O O C O
i-l CO U-l UO
\o o
co eo in O
•-H vo r- CM
^-t O O O
r- in O r^
CM co m •-(
O O --» O
o o o o
O O O O
! CD O .-HO*
ass
rt O O
J *J hJ O
Q Q Q O
03 CD 03 O
•-( *S" tn
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
£2
•-< ON co CM in *
OOONON OS. i-t Q Ol
OlACOON O •-< •-! O
en -3- CM O OOOO
ggg§
eo po co o
hJ J J O
Q O Q O
CO CO CQ O
•-i O CM en
§m en 1-1
O O o
CM CO ON ON
ON vo vO i-H
oooo
CO nJ in r-i
^O _J J O
O O Q O
o ca ea o
ON -<
co r- in oo
1-1 CM cor-
in i— v to r^ -j-
O CN »-l i— < O
r-i OOOO
01 incof>
CSIOi-^O
OOOO
\ csi in u~i ,— i
-
O
4J
T
T3
T
E
v :
H a:
• ' '• B • : : 0 - : : 0 : : : B : : : 'e
SSS^0 SSSM SSS60 SSS00 3££°°
•H -H -H -H -H
o. P- ft a, Q,
i 00 00 bC 0 00 W) W) B 00 QC 00 6 00 60 00 B 00 GO 00 0
CQcQCQCJ CQ CQ pa U CQBQcQO CO CQ pa CJ M M 03 O
§ B i
• • - -H . . . TH ! I ! -H
^Ju-5 uJJ-? JJ^-S
•a-a-aa -a-a-aa -a-a-aa
a o. ""* ^ a.
OO 00 00 0 00 00 OC B OC 00 OC B
01
c
41
^
•sa
-------
CN in Ol OO
Cs| vD cTi O
CM csi in \o m-^-m-d'
r~- r~* *& r*- i—< ,—i i—i o
CM en CM o oooo
y •-! a <
CO Q CO <;
m m -j- v£>
i ON \o \o m m o m
) ON CO ON i-H r-J i-l O
) cn CM O OOOO
i-l CNJ i-« O
o co r-. CM
X) i—I -* -* i-"
> o o o
co r~- *—i in
OOOO
* ^0 o -j- CN in r-j
J CM <—i \o ^3- •—' in
J O O OOOO
-U M 00 00 W
X M 60 00 B
4J (fl CO Cfl O
•H 03 CO CQ U
CO CQ CQ CJ
00 00 M) 0
Sfl) CO O
CQ CQ U
i^ a tH
rt ai cu
CQ H Pi
I
42
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
a
CO
O Q\ ON CO
O O
CO O
O O O O
tn i-i CO O
OJ CO CM •-!
O O O O
O en •-i ^
LT, ul CM
tr, co r-
CO OO
O !—I
o o
(N c*i CO O
CN c~i m i—i
O O O O
CT, fsl csi >£>
• --< o o o o
m
-------
O O O O
cq oa pa O
) i-H O
) O O i
• •-* o o o o
O O O O
uo ro ro ro
--• CPl -J- -sf
-I -H r-l O
O O O O
O O ro 1-1
oo f^» m ,3-
CM m ON »*
ON
O O O O
ON °O 00 CM
CM CM i
O A
O CO \O i—I
•-I O O O
o o o o
o o o o
pa CQ PQ O
O co -d- O
ro m Vt
CO .-J r-4 O
O Q O O
O CQ O O
oo co CO r-
m co co -
co cn M oo en w w
0) HJ Ol QJ Q> HJ
M 00 00 -M 00 OO 00 .
-H
>-t CM CO O i-l CS CO
CX
00 00 00 0 00 00 M
ro w cij o n) rt n)
CQ CQ tQ O PQ pa PQ I
OOOOOOW
„ „ „ "5
r-tcsiroo
GOM600
flcdTOO
CQCQCQU
rH CM CO I
00 SO 00 I
to nj n) <
pa pa pa c
•H C
s
[
\
44
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
ggg8
*a PQ eo o
O* f*. CM O
c-J m m ,-1
o o o o
i O >-O oo (si oo CM OO
00 A A
hJ iJ O
o o o
OQ pa o
OOOO
1 --^ OOOO
LA i—i i-l O
oooo
CM O -* i-l
CO (*! CO --1
OOOO
-i-i OOOO
O CO >n ^
CM CM CM
OOO
U U CO
<8 ffl O
O oo fi CM
QJ
00 OO 00
00 OO 00
CQ CO CO
45
-------
>O in CM rH
oooo
CTl rH \O (N
rH p- ro cn
CM O rH O
OOOO
CO rH O rH
O"i O"i CO CTi
in CM CM
c*i ^r en rH
^-ITimrH
OOOO
r-* r*. i-t >a-
ro in ^o rH
oooo
OOOO
rH \O CO CM
CM r-t TH 01
00 \O in ^3-
CM rH rH O
OOOO
u u o
tj (3 (j O
E- PQ E-» •
r-~(MOr-
CO O rH m CM m rH
• cr>
CO CO CM O
OOOO
^O CO f- CM
O rH O O
OOOO
u-ico-r-,-i
mio-jrH
OOOO
rnooo
OOOO
m-^-ro^H
OOOO
rJ rJ rJ <
P Q Q <
PQ eo PQ
r- r- O i— 1 iH ro
oo m o **
in o o CM
-d- m - CM
OOOO
OOOO
o CM -^- rH
OOOO
o ro r- m
CO <3- o" 00
co r- co in
r- ro co rH
r—i r~~- r— o
-3" ro ro t—(
OOOO
O 0"> rH
cd o o
MOO
ro r- r- o
OOOO
u o
g£ se
ea CQ H •
r--£> -J- rH
^TCMP^O
OOOO OUOO OOOO
wwcnoo uimtnoo 01 w w oo oiuiaioo WIWUJ
QJQJOt QJOJOt (UQJQJ QJQJO> QJOJQJ
. , , .
bOOOOO*J OOoOOOJ-i
- *^. -^ — -- *~ *- * - - -
OOM604-1 0000004J 0000604-1 OOM004J
OMOo*
fflfflo
pqcQU
Moot
tflctft
FQC30
ftoaowi' ooootjoB*
toraojo tfltocco
»(OpqU CQPQPQCJ
r-icNcno
MMOOB MOOOO
njnjnjo tnnjro
pQpQcOCJ pQpQCQ
°?5P2pB
555..^
CQCQPQU
I
I
46
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
•-H !-* -3- m
o ^H o c*
CO O m
vO OO (M CM
ON O
-------
CO
+ s
3 ^
o
m o
8-
o
CM
O
fl
CD o
CO
m
' O
£ ^
s
u-
rH O
O m
CO
J
O
CM
O
r
tt) 0
0
i-H QJ
(U rH
3 T-
ON r-- O> in
-J O CO ON CM ^£) O
CM 00 CT\ v£) 00
.* GO S co rS £ £
ooom-n
5! £ S CM £ £ 3
T* CM CN ^D CM,
CM in m r- co
CM o ro i-t co ^o i-H
<± QO
-J- co r- CM co r- -J-
1-^ CM I-H rH
oo in in o"^ o
oo o I-H -, 00 60 00 B
*_> w co re re o
•H -H CQ CO ffl CJ
•a T)
<= -d
O 0
CJ QJ 3
t-> X u
w vj a T3
tn tu u a> -H
at u nj p» x
H QJ VJ T-I CO
G at 4J o c
o p, n) j= o
tj 0 rH MS
« >u a) «
CO H Pi tJ
CM m CM ON
oo * r-~,
I-H oo r*» i— i
CM
O O -O CM CO
i— i i£) i— i r--
5SSS
co vo r— CM
to co en 00
OJ 4) OJ
00 00 00 4J
•H
rH CM CO O
ex
00 00 00 G
cd n) cd O
C
M-l QJ
O 00
o
en i-i
"C3 iH
X
O
\O O CO ^H
in -£" co i— i
O
CO
r-- ^o ** I-H
O
CO
3 5 5 £
o
o- m -a- I-H
o
55^^
o
CO
5 S ^ 2
o
in co vo i^-
CM CO O O
o
in in co t— i
o*
co m
o i i~-
I-H m co >*
o o o O
o
co
O O O O
0
0
O O O O
O
CM CO CO CM
O O O O
o
CM >X3 in O*i
o o o o
o
o o o o
o
(--. r-- CM ^>
o o o o
o
•
•-H O O O
o o o o
o
00 O ON r^
o o o o
o o o o
o
O Q) ro i-H
o n o o
o
co o OJ CM O
O \O t** CM
o o o o
o
r-.
o o o o
o
m CN
as r- r-- CM
o o o o
o
m r- £) ON co
CO i-H \O CO
O O O O
o
O CM vO vD
CM r- in r*»
o o o o
o
m CN cri co
in I-H ON
CO O CM CM
O CO CO O
o
in o
I-H | ^ |
CO 1 \C 1
o
. . . B
W CO CO 00
OJ QJ Ol
--. \ -- OJ
00 00 00 -U
,. . * 0)
rH CN CO O
oo oo oo B
eq CQ ea U
0)
C T3
HI -H
00 C
l-t >1
-fl CJ
OS
CM
CO P efl O
K CO li O
H H Q-
^
CQ CQ CO O
O
in QJ cu co
rH CJ CJ O
O 03 co O
O l-t l-t O
O
o o o o
O CQ O O
o
rH CN CO
§Q O O
CO O O
o
CJ O CM O
0
CM U
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
cs) m o\ ^-i co O c
vC Oi ^O O * OO -*
00 OO *£> CN
I-H OJ •-< O
OhOO
,-4 O fi O O O i
OOOO O CO I
ii-l OOCM-JCO
CM O c^ oo
O ^ OOOO
oQ^-iO
OWOO
co O co c
OO CO
O O ^-i O
-H
•o
» 1
l-i 33
m to en
00 00 00
333
01 0)
-~. --, ^- 0)
00 00 00 4-1
(fl (0 (0 O
pa PQ CQ u
g
en ui tn oo
~~- -^ ^^ QJ
00 00 00 W
oo oo oo B
« CO CO O
C
0)
cn in oo
0) Q;
00 00 00 *J
•r4
00
(0
ffl
oo oo 0
ffl CO O
CO CO O
05
00
00
tO
OJ
C
(0
2 s M
-^ -^ a>
00 00 4J
•H
00 00 S
SCO O
03 O
m « m oo
ao ao H> 4-i
00 0* 60 S
td to to o
BQ CO PQ CJ
01
c
tn w 05 t*
00 M 00 4J
00 00 00 0
CO (Q (O U
S S S
00 00 00
00 00 00
CO CO CO
i (0 >
Li -H
CO (- 06 O
60 C
O (0
ki ?-,
49
-------
-1
o
o
o
o
5
a
o
o
CM
2
0
3
j
g
o
r-i 0
CO
o
o
CM
0
rl
41 O
a
r-l OJ
41 r-
-.--.-.
CO -d"
3££ ££££
en >£> CM •-" o m cr. o
m m cn r^ ^- m CM
0 ^=m^0
3 £ 5 S£ 2 S
mocnco
O f* O CM CM
•* r*. vt O CM CO en
0 0-cftMc*
•d" r^ m o co CM r-i
0 ^x,^
~* r- m en m m CM
a, «,stm~
-J r-. -d- o m CM CM
« « •". ~ ^
eO > M
H 41 ri -H 00
S 4> 4-t o C
O CX CO J3 Q
M S iH MS
co o) ot n
PQ t-i PS t_)
en
r-. CM r~ r-i
O
en in cp en
O
en r^ CM en
M5 CM CM CM
r- CM r- <-<
O
en to en oo
01 41 Ot
00 00 00 4J
rH CM en O
00 00 00 S
m r-^
*4D CM <-< i— i
O O 1-1 O
o
r-i en CM CM
O O O O
o
m m
O O O O
o
-d- co >J
0 0 r-l 0
O
rH O CN CO
co cn CM co
o o o o
o
co en o
en en 41
00 00 DO 4J
•H
rH CM CO O
00 00 00 S
rt co n o
oa pq oa a
at
00 C
o rt
03
3 S .*
CM J CM 0
O D O O
o ca o o
o
41 41 M
« ca 03 o
Ho
4) CM
Q Q «J O
P3 EQ h O
°
V 01 CM
Js -3 eo o
0
4) 1-4
0 rJ rJ 0
t-i eo pa o
H o
41 4) g
rl P3 rl O
°
C|) a) CM
rl P3 M 0
o
rl P3 Jj O
0
rH 0 12 0
o CQ o o
0
co cn
o pa o o
o
r-t 41 m
CO rJ O O
o « v o
o" H
0 rJ CM O
« Q O O
O
a 5 S o
pa pa o o
0
01 01 41
--.-— *— 41
00 00 00 4J
00 00 00 B"
rt n fa o
« PS. PQ u
0)
00
o
n
(0
u
(0
4->
cO '
H
1 c
01 O
£ OJ
*Not included
BDL-Below de
50
I
I
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
i—
i~i r-. .-i OOOO
i c* O i-t \o u en in
• a «* o cv to OOi-tO OOOO O O O (
i fn cj-i m oo co i—' -J*
J co -d- m o c
1 -J O
I Q o
> ca o
) 1-1 OOOO O
too m o CM o
i O O OOOO
r oo .-H & o i
> 0*1 m -j- o> •
m-JooO OflJO^t ^-lOcor
cofni-i-* cnumcM CMOO--IL
in, tn co i—i I-H qj i—' O mocMC
oooo OMOO oooc
\o ro CM iH
§Q\ i—I r-- r-l
OOOO
^o>-d--j in-joor-
OO^CM r-u*ior—
OMOO oooo
M CQ C
\ CM m i-* CM (U
•--•— —.qj -~-.~--.-— 0> --- '—."--.(
00 00 00 U 00 00 00 W QQ 00 OB 4.
« Q « O
M CQ M O
0 00 Q 00 00 00 I
to o cd tc (0 i
QCQD CO 00 03 t
00 60 00 B
oj n rt o
CQ pa eo u
00 00 60 B
CQ CQ CQ U
51
-------
—* r-- 03 CO
» •
I** -3 —« !
> •—I O O
.-* O r- <
CT\ -d" O c
1 r-i r-- oo (•*•» c
1 in 1/1 CM en c
) CM O O O C
• cor^oo ,_i o o oooo
IT) U->l f»^ C-) U"> V.
OOOO Of-iOO
CO OD CO O O
1 -J 1-1 O
1 Q O O
I 03 O O
i-i OOOO
O H O O
OOOO O O O C
P3 CQ O O
cOcr>fsJCO cof^^iri ^-JCOMU-l
io\£)LOro CTNf-ii—i^-i oooo oooo oooo eacaeac
"icom ooi~^--'i^ oc^i-J-f
1 CTv cu QJ
00 CC 60
I
r-l CO
•a -H
H c
I
52
I
-------
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
i-i CJ CM O
r-f 00 CM i
ui to cn '
m CM r^ '
m co o m
u -J
.-J J hJ O
Q Q Cl O
03 P3 CD O
vD vD Cvl CTi
-d- r~^ r-» cyi
-* co oo en
CM o t-t O
O O O O
r-. co o cy
O CM' oo CM
m O o CM
0*1 m m CM
co CM m 1-1
88§8
OO ^ vj •—'
-1000
o o o o
O O O O
•J- 0*1 O c-l
i-« O i-H O
o o o o
CM m <{• oo
Sf O CM CO
rH m r-.
> O r-
o in m o
CM o O CM
O O I-H O
o o o o
O OO CO O
o\ m o oo
fi m CM o
o o o o
O CM r^. CM
r-t r-l O O
o o o o
CM CM in CM
m o o O
O O O O
>
J-
T
si
« 0) tO
00 00 00
i-f CN ro
•> 00 00 OD
cct ca to
CO CQ 03
CU
00
CU
4_)
•H
o
o
OJ to 07
CU CU CU
^va
rH CM ro
00 00 00
« cq CQ
c
o
oo
4-1
•H
O
1
U
07 CO VI OO
d) o> cu
*^. --~. -- CU
00 00 60 -t-i
•H
fH CM m O
ex
00 00 00 B
c
to w ca oo
CU CU CU
•*-. ~-- -^. cu
00 OO 00
-------
in r— co CM
CD \O
a o i
« .-I « a
Li O I- O
H CO H O
•J J iJ O
Q Q Q O
CO P3 P2 O
U~l CO OJ eg
O in CTi CM
^4 J ,-J O
Q P O O
-* cj oo
r- ^ 09 O
O CO H O
eg cj -J- ^-i
r^ BJ in r-i
eg t-i O O
O H O O
, ,-1 O —I
^-l O OO O>
o a o eg
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1 REPORTNO. 2.
EPA-60CV2-76-026
4 Til LE AND SUBTITLE
EFFECT OF GASOLINE ADDITIVES ON GASEOUS
MISSIONS (PART II)
7 AUTHOR(S)
R. W. Hum, F. W. Cox, and J. R. Allsup
9 PERFORMING ORGANIZATION NAME AND ADDRESS
Fuel/Engine Systems Research Group
Hart lesvil le Energy Research Center
Energy Research and Development Administration
Bart lesville , Oklahoma
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
3. RECIPIENT'S ACCESSION- NO.
5. REPORT DATE
February 1976
6. PERFORMING ORGANIZATION CODE
8. PERFORMING ORGANIZATION REPORT NO
10. PROGRAM ELEMENT NO.
1AA002
11. CONTRACT/GRANT NO.
EPA-IAG-D4-0040
EPA-IAG-D4-0453
13. TYPE OF REPORT AND PERIOD COVERED
Final Report 1974 - 1975
14. SPONSORING AGENCY CODE
EPA - ORD
15. SUPPLEMENTARY NOTES
Supplements and Extends, Part I
16. ABSTRACT
A study has been conducted to determine the effects of nitrogen-containing fuel
additives in gasoline on regulated and nonregulated automotive emissions. Method-
ology was developed to measure possible nitrogen-containing compounds and was
used to analyze the emissions from a variety of cars without catalysts. No effects
due to the additives could be discerned. Of the nonregulated nitrogen compounds
analyzed, ammonia, amines, nitriles, nitrosoamines, and aryl nitro compounds were
not detected; HCN, cyanogen, and alkyl nitro compounds were measured. Emission
data are included from a rotary engine (Mazda), an air-cooled engine (Volkswagen),
and two standard V-8 engines (Chevrolet and Ford). Six nitrogen-containing addi-
tives chosen for their common usage were tested.
17. KEY WORDS AND DOCUMENT ANALYSIS
.1 DESCRIPTORS
Evaluation '''Nitrogen organic <
Gasoline "'Nitrogen inorganii
'--Fuel additives Chemical analysis
Automotive engines
-'Exhaust emissions
Air pollution
13 DISTRIBUTION STATEMENT
Release to Public
b. IDENTIFIERS/OPEN ENDED TERMS
:pds .
: cpds.
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20 SECURITY CLASS (This page)
UNCLASSIFIED
c. COSATI Held/Group
14G
21D
21K
21J
13B
07C
07 R
07D
21 NO OF PAGES
61
22 PRICE
EPA Form 2220-1 (9-73)
-55-
------- |