EPA-600/2-77-237
December 1977
Environmental Protection Technology Series
   GASEOUS,  PARTICULATE,  AND  SULFUR-RELATED
 EMISSIONS  FROM NON-CATALYST AND CATALYST
                                   EQUIPPED  VEHICLES
                                 Environmental Sciences Research Laboratory
                                     Office of Research and Development
                                     U.S. Environmental Protection Agency
                                Research Triangle Park, North Carolina 27711

-------
                RESEARCH REPORTING SERIES

Research reports of the Office of Research and Development, U.S. Environmental
Ptolection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:

      1.  Environmental Health  Effects Research
     ?..  Environmental Protection Technology
     3.  Ecological Research
     4.  Environmental Monitoring
     5.  Socioeconomic Environmental Studies
     6.  Scientific and Technical Assessment Reports (STAR)
     7.  Interagency Energy-Environment Research and Development
     8.  "Special" Reports
     9.  Miscellaneous Reports

This  report has  been assigned  to the ENVIRONMENTAL PROTECTION TECH-
NOLOGY series. This series describes research performed to develop and dem-
onstrate instrumentation, equipment, and methodology to repair or prevent en-
vironmental degradation from point and non-point sources of pollution. This work
provides the new or improved technology required for the control and treatment
of pollution sources to meet environmental quality standards.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.

-------
                                            EPA-600/2-77-237
                                            December 1977
GASEOUS, PARTICULATE, AND SULFUR-RELATED EMISSIONS
 FROM NON-CATALYST AND CATALYST EQUIPPED VEHICLES
                        by
                 James N. Braddock
Emissions Measurement and Characterization Division
    Environmental Sciences Research Laboratory
        Research Triangle Park, N.C. 27711
    ENVIRONMENTAL SCIENCES RESEARCH LABORATORY
        OFFICE OF RESEARCH AND DEVELOPMENT
       U.S. ENVIRONMENTAL PROTECTION AGENCY
   RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711

-------
                                 DISCLAIMER  .

This report has; been reviewed by the Environmental Science Research
Laboratory, U.SJ. Environmental Protection Agency, and approved for
publication.  Mention of trade names or commercial products does not
constitute endorsement or recommendation for use.
                                   ii

-------
                                  ABSTRACT

     The gaseous, participate, and sulfur-related emissions from six 49-
state 1975 domestic production catalyst equipped vehicles (two Chevro-
lets, two Fords, and two Plymouths) and one 1975 imported prototype non-
catalyst vehicle (Honda CVCC) were measured by a variety of chemical
analysis techniques as a function of fuel sulfur level.  Four separate
and different sulfur level fuels were tested for each vehicle.  The
seven test vehicles were driven alternatingly, in accordance with the
1975 Federal Test Procedure (75 FTP) and Highway Fuel Economy Test
(HWFET) procedures.  All vehicles tested met the level of the 1975
interim emission standards of 1.5 g/mile hydrocarbon (HC), 15 g/mile
carbon monoxide (CO), and 3.1 g/mile oxides of nitrogen (NO ).  The
                                                           n
Honda CVCC gave the lowest gaseous emissions and best fuel economy.
Retarding the stock spark timing 5° on those vehicles that "pinged" on
the regular no lead test fuel eliminated the pinging, but decreased fuel
economy and increased regulated emissions. Sulfur dioxide (S02), parti-
culate matter, and sulfate emissions increased with increasing fuel
sulfur and were greater during the HWFET cycle than during the 75 FTP
cycle.  The greatest particulate and sulfate emitters were the air pump
equipped Ford vehicles, while the lowest particulate and sulfate emitter
was the Honda CVCC.  Particulate matter was composed mostly of sulfuric
acid; the non-sulfate portion of the particulate matter was composed of
carbonaceous material and aluminum and iron oxidation products.  No
platinum was found in the particulate matter.
                                     iii

-------
                                  CONTENTS
Abstract	iii
Figures	vi
Tables   	vii
Acknowledgment	viii
     1.   Introduction 	   1
     2.   Conclusions  	   3
     3.   Mechanism of Sulfate Formation 	   5
     4.   Experimental	12
     5.   Results and Discussion	17
               Gaseous emissions results 	  17
               Sulfur-related emissions results  	  20
               Non-sulfate particulate matter results  	  23
References	44
Appendices	47
     A.   Unleaded gasoline analysis 	  47
     B.   Fuel sulfur analysis	48

-------
                                 FIGURES
Number                                                            Page
   1     Schematic of sulfur oxidation reactions
          throughout the exhaust system 		  .    6
   2.    Combustion reactions of sulfur related
          compounds in the exhaust	    9
   3.    Sampling system	15
   4.    Vehicular particulate material balance  	   43
                                   VI

-------
                                TABLES
Number                                                            Page
   1.    Description of 1975 model  year vehicles  	   12
   2a.  1975 FTP gaseous emissions (grams/mile)  .	17
   2b.  1975 HWFET gaseous emissions (grams/mile)  	   18
   3.    Comparison of 1975 FTP gaseous emissions from
        vehicle timed at stock spark ignition with
        vehicle timed at 5° retarded spark from
        manufacturer specifications 	   19
   4.    Chevrolet Impala S02, S04, and particulate emissions   .  .   24
   5.    Chevrolet Nova S02, SO., and particulate emissions   ...   25
   6.    Plymouth Fury S02, SO*, and particulate  emissions  ....   26
   7.    Plymouth Scamp S02, SO., and particulate emissions   ...   27
   8.    Ford Granada SCL, SO., and particulate emissions	28
   9.    Ford Torino S02, SO., and particulate emissions	28
   10.  Honda CVCC S02> SO^, and particulate  emissions	30
   11.  S02 linear regression analysis	   31
   12.  SO. linear regression analysis  	   32
   13.  Particulate matter linear regression  analysis 	   33
   14.  Overall ranking by S02, particulate,
          and SO. emission rates	34
   15.  Elemental analysis of particulate  matter by x-ray
          fluorescence:  1975 FTP results	39
   16.  Elemental analysis of particulate  matter by x-ray
          fluorescence:  HWFET results  	   40
   17.  Elemental analysis of particulate  matter by x-ray
          fluorescence:  1975 FTP results	41
   18.  Elemental analysis of particulate  matter by x-ray
          fluorescence:  HWFET results  	   42
                                   vn

-------
                              ACKNOWLEDGMENTS
The author gratefully acknowledges the assistance of Dr. Silvestre Tejada
for performing 1:he sulfate analyses and Mr. Robert Kelloqq for performing
the x-ray fluorescence analyses.
                                    vm

-------
                                 SECTION 1

                               INTRODUCTION

     It has been demonstrated that oxidation catalyst equipped vehicles
emit more particulate matter, consisting mostly of aqueous sulfuric acid
(H2S04) droplets, than do their non-catalyst equipped counterparts.
Because concern has been expressed about the environmental impact of
                       2 3
these sulfate emissions '  we have undertaken a study of six 49-state
1975 production catalyst cars and one 1975 prototype non-catalyst vehicle.
This study examines the gaseous and particulate emissions of these
vehicles as a function of fuel sulfur, type of catalyst, and driving
cycle with particular emphasis upon sulfate formation.

BACKGROUND

     With the advent of the 1975 automobiles, we were introduced to a
                                                                    A
new auto exhaust emission control device — the catalytic converter.
This device, which oxidizes excess hydrocarbons (HC) and carbon monoxide (CO)
in the exhaust, replaces former fuel  inefficient controls such as retarded
spark timing.  Consequently, auto manufacturers in 1975 were able to
increase fuel economy substantially,  10-20% over 1974 model  automobiles,
while meeting the tighter 1975-1977 interim exhaust emission standards.
In addition to an increase in fuel economy, there was also an improvement
in the driveability of 1975 automobiles compared to previous years.
Although the catalytic converter emission control device decreases
regulated gaseous emissions, it unfortunately increases non-regulated
particulate emissions under certain operating conditions, specifically
that of sulfuric acid (HUSO.).  This  emission of sulfuric acid is a
product of the combustion of fuel sulfur and the mechanism of catalytic
oxidation. In the following pages we shall look in some detail at

                                      1

-------
how the catalytic converter functions.

     Most catalysts, including those studied in this work, are composed
of noble metal Pt/Pd (platinum/palladium) blend.  The Pt/Pd blend was
chosen because of its relative chemical inertness and its resistance to
chemical poisoning, especially by sulfur compounds.  Approximately 1.5
grams of the noble metal blend are used per;catalyst.  Currently there
are two geometric catalyst configurations:  pellet (beaded) and monolith.
The container s.hape of the pelleted catalyst is a >bedpan :and,, internally,
the pelleted catalysts contain.pel lets approximately 1/6 - 3/8 inch in
diameter composed mostly of aluminas  .The pelleted.and exhibited a
large surface area'per. unit massv,;'The containerjshapeVof the monolithic
catalyst is an elliptical, cylinder approximately3-6 inches, in ,d1ameter>
Internally, the monoliths;-contain;a.;honeycomb type>^truetureiextending,
the length and diameter pf the tyl-inder aridi are .surrounded:tiy, an,integral
outer skin of alumina orv.corderitev^ Compared to pellet cataly.st5,
monoliths offer greater mechanical strength, lower pressure drop, smaller
surface area por unit mass, and quicker warm-up time.  Both
types of catalysts are mounted before the automobile muffler, as close
as possible to the' engine's, exhaust manifo.Td on the!;ex^ausfc,pjpe.

-------
                                SECTION 2

                               CONCLUSIONS

     On basis of the gaseous, particulate, and sulfur-related emission
results obtained from the seven vehicles in the experimental  test program,
the following conclusions apply:

     1.   All test vehicles met the level of the 1975 interim gaseous
     emission standard of 1.5 g/mile HC, 15 g/mile CO, and 3.1  g/mile NO .
                                                                        /\

     2.   Retarding the stock spark timing 5° from manufacturer specifica-
     tions on those vehicles that "pinged" on the unleaded test fuel
     eliminated the "pinging" at the expense of decreased fuel  economy
     and increased regulated gaseous emissions.

     3.   Sulfur dioxide, particulate matter, and sulfate emission rates
     increased with increasing fuel sulfur.

     4.   Total sulfur balances were related to driving cycle characteris-
     tics.  1975 FTP sulfur balances were consistently less than HWFET
     sulfur balances.

     5.   The sulfur oxide storage/release phenomenon depended upon the
     type and operating conditions of the catalyst.  Pelleted catalysts
     were among the lowest 1975 FTP sulfate emitters and among the highest
     HWFET emitters.  The air pump equipped monolithic catalysts emitted
     the greatest sulfate and particulate matter in both the  1975 FTP and
     HWFET.

-------
6.   Trie catalyst vehicles' particulate matter was composed mostly
of sulfate; material and its associated bound water molecules.  The
non-sulfate portion of the particulate was composed of carbonaceous
material and inorganic oxidation products.  Iron and aluminum were
the most prominent inorganic oxidation products emitted.  No
platinum v/as detected in the particulate matter.

7.   The non-catalyst vehicle's particulate matter was composed
mostly of carbonaceous material.

-------
                                SECTION 3

                     MECHANISM OF SULFATE FORMATION

     In the combustion process detailed in Figure 1, the fuel sulfur is
oxidized to sulfur dioxide (S02).  The exhaust gases, including hydrocarbons
(HC), carbon monoxide (CO), oxides of nitrogen (NO ), and sulfur dioxide
                                                  J\
(S02) then proceed via the exhaust manifold to the oxidation catalyst.
In the catalyst, excess hydrocarbons and carbon monoxide are oxidized to
water vapor [HpO(g)] and carbon dioxide (C02), while some of the S02 is
oxidized to sulfur trioxide (S03).  Proceeding through the catalyst and
into the muffler, the S03 combines with H20 (g) to form gaseous H2S04.
This gaseous H2$04 condenses to a finely dispersed aerosol after exiting
from the tailpipe.

     In dealing with formation of sulfuric acid in the exhaust, one must
consider the following reactions:

     (1)  S + 02  	»-  S02  (g)      Assume complete combustion
                                        of fuel sulfur to S02

     (2)  S02 + 1/2 02—»- S03 (g)     Oxidation of S02 to S03 via
                                        catalyst

     (3) !S03 + H20 —f  H2S04 (g)    Hydration of S03 to gaseous
                                        H2S04

     (4) H2S04(g) + n H20  	*-        Condensation of gaseous H2S04
                                        to finely dispersed aerosol outside
         H2S04 * nH20                   tailpipe

-------
 AIR
FUEL
COMBUSTION PROCESS:
OXIDATION OF FUEL
SULfURTOS02

TEMPERATURE:'650°C

REACTION: Sf Oj —•+S
                             ENGINE
                            EXHAUST
OXIDATION CATALYST:
OXIDATION OF S02 TO
SO* VIA CATALYST

TEMPERATURE: 650-500°C

REACTION: S02 +%02 —»S03
CATALYST
EXHAUST
MUFFLER:
HYDRATIONOFS03
TO GASEOUS
SULFURIC ACID

TEMPERATURE: SOO-250°C

REACTION: SOa+HaO-*
                                                 ATMOSPHERE:
                                               CONDENSATION OF
                                                 SULFURIC ACID
                                               OUTSIDE TAILPIPE
                                              TEMPERATURE: 250°C
                                                   AMBIENT
                   Figure 1. Schematic of sulfur oxidation reactions throughout the.,exhaust system.

-------
It is apparent that the sulfate determining reaction is reaction (2),
which is the heterogeneous catalytic oxidation of S02 to SOo.

     A probable mechanism with the following sequence of steps is proposed
for the catalytic oxidation of SO,,:

     (a)  diffusion of the reactants S02 and 02 to the catalyst surface.
     (b)  adsorption of the reactants on the surface.
     (c)  chemical reaction (oxidation) on the surface.
     (d)  desorption of the product S03 from the surface.
     (e)  diffusion of the product SO., from the surface.

     Because steps (a) and (e) are usually very fast, step (b), step
(c), or step (d) may be rate determining for this mechanism.   The
literature covering the mechanism of S09 catalysis by platinum in
                                         5-12
automotive applications is quite limited.      Consequently,  we must
turn to industrial heterogeneous catalysis experience in order to
formulate a general rate expression for the catalytic oxidation of
S02. The rate equation of Uyehara - Watson is perhaps the best industrial
formulation with respect to generality, mathematical depiction, and
                                  13
experimental data representation.     Their rate equation is  first order
with respect to S02 partial pressure and is proportional to the square
root of oxygen partial pressure.  In addition, they define the rate
determining step as the surface reaction between adsorbed 0,  and adsorbed
                         10
S02.  Hammerle and Truex    in their recent flow reactor investigations
of the kinetics of S02 oxidation for various automotive catalyst compositions
have used a similar expression, approximating that the rate of sulfur
dioxide oxidation is first order in S02.  This approximation, equating
the automotive catalytic S02 oxidation rate expression with the industrial
S02 oxidation rate expression, appears valid.

     The quantity of S03 formed is influenced by many factors including
the concentration of 02 in the exhaust, the mass flow of sulfur through
the engine, the type of catalyst, the temperature of the catalyst, and
possible reactions of S02 and SOg with other exhaust constituents.  In
                                      7

-------
addition, the quantity of SCL formed is complicated by the problem  of
sulfur oxide storage/release within the catalyst. That is, investigators
have not been ab'Je to obtain a chemical mass balance between the  sulfur
burned in the fuel and the sulfur emitted from the tailpipe in  catalyst
equipped vehicle:;.  Inherent in these investigations is the assumption
that the analytical methodology for S02 and sulfate detection   is
sufficient ard accurate.  Therefore, the problem of obtaining a sulfur
mass balance directly relates to the intra-catalyst storage/ release
phancrcena.  Investigators of non-catalyst vehicles have usually been
                                             o
able to obtain credible sulfur mass balances.   Consequently * owe; ; need to
focus our attention upon the intraKcatalyst stbrage-/fe:>6aseophehomenabin
order to better jnderstand the complicated mechanism(s) of sulfate
formation ant;
     In the exhaust . system ,we have: thus far jioted^that SOo may.. fee  ca.talytic.ally
oxidized to ,SOx,  ~Jhe.se two aasBous 5u9fur oxides., SO/, and S0,0  may
react with other, exhaust gas cpnstj.tjuents and  in addition, react  with
the aluminum oxide catalyst. substrate material and with the iron contained
in the engine and exhaust system.  The s.uJfur oxide reactions with the
aluminum oxide catalysi substrate material are referred to as the  intra-
catalyst storane/releas£ phenomena and are iHtlustrated in Fiqure. 2L
Reactions (5") and. (5) with, iron. oxide are chemically sim.ilar and are
                                                         1 ^*-T7
thernodynamica'lly favorable at temperatures below^450°C.        Below
450°CD sulfur oxide storage .occurs with iron sulfite/suJfate as the
product.  As the temperature increases above 450°C, s¥lfur oxide is
released according to reaction (7). '°»19

     (5)  3St)2 -»• Fe203  — 9~  Fe2(S03)3                  Sulfition process
     (6)  3S03 -i- Fe203  — *-  Fe2(S04)3                  Sulfation process
     (7)  2F(i2{5,04)3 — ^—*-  2Fe203 + 6S02 + 302        Sulfur oxide release

Consequently;, depending upon the temperature and sulfur oxide concentration,
sulfation is possible over a large portion of the exhaust system.  Those
zones in the exhaust system in which the temperature varies may therefore
either store or release sulfur oxides.

                                      8

-------
                                                              C02
                  S+02 C
  INTRA-CATALYST
  STORAGE • RELEASE PHENOMENON
S02
                                        02
                      /
     Pt/Pd
   CATALYST
V
                    Al2
 S03
         AI203
  AI203 + S02 + 02 «    A\2 (804)3 /'
      AI203 + S02 + CO?
1                 x
                          /  /
                       //         \
H2S04
                  S02 +C02
           Figure 2. Combustion reactions of sulfulr related compounds in the exhaust.
             OUTSIDE TAILPIPE

-------
     A similar reaction mechanism is proposed for sulfur oxide reaction
with the aluninum oxide catalyst substrate.  Thermodynamic data indicate
                                                            20
that reactior: (8) is favorable at temperatures below 425°C.     With
increasing temperatures, the decomposition reaction, (9), favors the
formation of S02 and 0,,.  Consequently, S03 formed on the catalyst may
react with the substrate, A1203> only to be released later as S02.
Another routo fcr aluminum sulfate decomposition is the reduction
                      on
reaction (10) with CO.

     (8)  3S03 •»• A1203   	*-  A12(S04)3               Sulfation process

     (9)  2Ai2(S04)3      A  »•   2A12°3 * 6S02 + 302     Thermal  decomposi-
                                                         tion process

     (10) 3C3 + A12(S04)3 —»~  £1203 + 3S02 + 3C02     CO reduction process

     In the preceding reactions (4 -10) of the sutfur oxides with the
aluminum oxide catalyst substrate material and with the iron contained
in the engina and exhaust system we observed that sulfur oxide storage
occurred at temperatures below 450°C  a'nd sulfur oxide release occurred
at temperatures above 450°C.  Becatise temperature is a function of driving
cycle, it fellows that prolonged high speed driving cycles as the highway
fuel economy test (48 miles per hour average speed for 12.25 minutes)
should release Dreviously stored sulfur oxides.  Conversely, transient
driving cycles such as the 75 FTP (22 miles per hour average speed for
31 minutes) shojld store sulfur oxides.  The storage/release phenomena
are discussed in more detail in the sulfur related emissions section.

     The most important exhaust gas constituents with which S02 and S03
can react are ammonia (NH3) and CO.  Reaction (11) of SCL with ammonia
is thermodys'.ar.ikally unfavorable at temperatures above 225°C.  Because
the average catalyst temperature is greater than 225°C for all driving
cycles studied, the formation of ammonium sulfate in the exhaust system
                                                          20 21
could occur only during warm-up and some idle conditions.   '    In
additions U is more probable that ammonia would be catalytically
                                     10

-------
oxidized over platinum according to reaction (12) before reaction with
                 22
SO, could occur.     The carbon monoxide reductions of SO, to SOp (13) and
S02 to elemental sulfur (14) are thermodynamically favorable but kineti-
cally limited.

     (11) 2NH3 + S03 + H20  	»-  (NH4)2S04         Sulfation process

     (12) 4NH3 + 502   	»-  4NO + 6H20        Catalytic oxidation
                                                     process
     (13) S03 + CO   	*~  S02 + C02         CO reduction process
     (14) 2S02 + 4CO  	*-  S2 + 4C02         CO reduction process

     In summary, the possibility exists for reaction of sulfur oxides
with both exhaust gas constituents on the one hand and, on the other,
with the catalyst substrate material and iron contained in the engine
and exhaust system.  Reactions (11-14) with exhaust gas constituents are
generally thermodynamically favorable but kinetically limited.  Reactions
(5-10) with Alp03 and Fe203 are thermodynamically and kinetically more
favorable; hence their occurrence is more probable.
                                      11

-------
                                  SECTION 4
                                EXPERIMENTAL
     The experimental  apparatus  used consisted of (1) seven separate
automobiles tested on  four separate sulfur level fuels, (2) chassis
dynairometer test equipment, (3)  a dilution tube constant volume sampling
system (CVS), and (S4) the analytical  instrumentation necessary to
perform the chemical analyses.

     Table 1 describes the characteristics of the seven test vehicles
WUh the eXcaptTorV 6f  the Honda  CVCC,  all were 1975 production vehicles
             Table I. DESCRIPTION OF 1975 MODEL YEAR VEHICLES
Make, mo; lei
Chevrolet Ir.tpuhi
Chevrolet Nov.!
Plymouth Fjry
Plymouth S<;arnp
Ford Granada
Ford Torino
Honda CVCC
Inertia
weight, Ib
5000
4000
5000
4000
4000
4500
2000
Engine
350CID V-8, 2bbl
250CID6, 1bbl
318CIDV-8, 2bbl
225 CID 6, 1 bbl
302CID V-8, 2bbl
351 CID V-8, 2 bbl
91 CID 4, 3 bbl
Exhaust emission control system
GM/AC Catalyst (Pellet), EGR.
GM/AC Catalyst (Pellet), EGR.
Universal Oil Products Catalyst
(Monolith), EGR.
Universal Oil Products Catalyst
(Monolith), EGR.
Matthey - Bishop Catalyst
(Monolith), EGR, Air Pump.
Engelhard Catalyst (Monoiith),
EGR, Air Pump.
Stratified Charge Engine
                                    12

-------
and were leased locally.  The lease vehicles had  acquired a maximum of
800 road miles prior to our lease and were driven on unleaded gasoline
only.  The Honda CVCC was a 1975 prototype designed to meet the 1975
statutory emissions standards of 0.41 g/mile HC, 3.4 g/mile CO, and 3.1
g/mile NO .  All were tuned to manufacturers' specifications and checked
         A
for exhaust leaks prior to testing.  Each vehicle was conditioned via
highway driving on its particular test fuel for a minimum road distance
of 250 miles prior to chassis dynamometer testing.

     The vehicles were tested using both the 1975 Federal Test Procedure
(75 FTP) and the Highway Fuel Economy Test (HWFET) procedure as described
                        23
in the Federal Register.    Cycles were manually driven on a Clayton
CT-50 water brake dynamometer with direct drive flywheels.  The dynamometer
had been calibrated for load via the Federal Register coast down procedure
prior to the vehicular test sequences.  A standard 5000 CFM fan, as
specified in the Federal Register, was used to cool the vehicles.  No
supplemental fan was used to cool the vehicle catalyst during testing.
Vehicle exhaust samples were accumulated using a conventional Scott
constant volume sampling system arranged for 6 bag sampling, with a
total exhaust plus dilution air flow rate of 406.9 ACFM. This system
was flow calibrated by the propane injection method described in the
Federal Register.

     Gaseous exhaust emission analyses were made using Federal Register
specified instrumentation for hydrocarbons, CO, NO , and C0?.  The
                                                  A        £
total hydrocarbons in bag samples were determined using a Beckman 400
flame ionization detector.  Individual C-, - C-JQ detailed hydrocarbons
were determined by computer operated gas chromatographic analyses of
                                                             24
the. bag samples.  The results have previously been reported.
Carbon monoxide and C09 were determined in bag samples using a Philco-
                                                   25
Ford prototype gas filter correlation spectrometer.    Oxides of nitrogen
(NO and N02) were determined in bag samples using a Thermo-Electron
model 10A Chemiluminescence Analyzer. These routine gas analyzers were
mounted in an instrument train conforming to Federal Register specifications
and were calibrated using span gases that had been checked against NBS
                                     13

-------
certified standard reference material gases.  All calculations  for
gaseous emission rates (HC, CO, NO  , CO,,) and fuel economy  (miles-per-
                                  A    t
gallon or mpg) were made according to Federal Register specifications.
Sulfur dioxide ($02) was measured in the dilute exhaust by  pulling  the
gaseous sample tnrough heated (100°C) stainless steel lines into a  50 ml
bubble-breaking type fritted glass absorber containing 25 ml of
tetrachloromercurate solution. Sulfur dioxide  was then determined  by
                      t
the West-Gaeke method '
Monitor IV instrument.
                      26
the West-Gaeke method    using; a modified Techhicon Automated Air
     Particulate emissions were measured using a cylindrical  stainless
steel dilution tube sampling system, 18 inches in diameter and 15 feet
long; it-was equipped with four 0.900 inch internal diameter  stainless
steel sampling probes operating teokinetically at 1.06 CFM.   The dilution
tube sampling-system is illustrated in Figure 3.  Tunnel and  probe
losses were less tnarv:\l percent of the collected particu3ater! Routine
particulate' matter samples wemcollected oh ei'.ther. .47-mrn.. fluorocarbon
type filters, (Fluoropore type FA from. Mil ,li pore Corporation),".having a 1
micron nominal porosity, or on •Gelmari: type "A" gilass fiber fidters   In
usual practice, three: probes (employed the-FTuoropore filters  and one
probe employed the glass fiber filter   Particuflate samples were (Obtained
over the 31 minute active portion of the 75 FTP and over the  entire
Highway Fuel  Economy Test.  Filter weighings were performed on a microbalance
to the nearest 2 microgramsrafter 48 hours, conditioni.n.g at 50% relative
humidity and 2f>°C.  Sulfate analyses were performed upon the  filters
                                               27
using the automated.barium chloranilate method, '  X-ray fluorescence
analysis for trace metal-s was performed, us ing a computer operated -17
                         (28)
channel Siemens analyzer.v  ;  Elemental, analyses for percent carbon and
hydrogen were performed.upon the glass fiber filters using a  Perkin-
Elmer automated combustion analyzer.

     The gasol ire used in ."all experiments was an unleaded, reference
test fuel.  Operation .with this fuel was satisfactory for all test
vehicles.   The sulfur level of the reference or base fuel was 0.020. weight
percent sulfur.  Sulfur levels were increased by doping the :base fuel with

                                     14

-------
AIR
 IN"
     ABSOLUTE
      FILTER
                                              PARTICULATE
                                                ANALYSIS:
                                              FOUR 1-in. DIAM.
                                                 PROBES
•*—£. Tt. — ••


/
<
^
/f
TO fcT
VACUUMF^
SOURCE ^
'*>r t |
•» ^O 18 -in. DIA J£
if


^ORIFICE MIXING PLATE
GAS ANALYSIS
   HC, NOx       S02
   CO, CO2     ANALYSIS
                            EXHAUST
                           '   OUT
                                                          4-in. DIA
                 EXHAUST PIPE
                                       FigureS. Sampling system.

-------
reagent grade Thiophene, C.H.S,  from Eastman  Kodak  Company,  Eastman
Organic Chemicals, Rochester, New York.   Appendix A lists  the  properties
of the test fuel; appendix B lists the amounts  of thiophene  added and
the analysis for the sulfur content of each fuel.
                                   16

-------
                                 SECTION 5
                          RESULTS AND DISCUSSION
GASEOUS EMISSIONS
     Table 2 lists the gaseous emissions data for the test vehicles for
the 1975 FTP and the 1975 HWFET.  All vehicles tested met the level of
the 1975 interim emission standards of 1.5 g/mile HC, 15 g/mile CO, and
3.1 g/mile NO .   The Honda CVCC, the only non-catalyst vehicle tested in
             A
these experiments, gave the lowest overall emissions for HC, CO, and
NO .   In fact, the Honda met the level of the proposed 1977 standard of
  A
0.41  g/mile HC,  3.4 g/mile CO, and 2.0 g/mile NO .   Of the six domestic
                                                ^
production vehicles, the Chevrolets gave the lowest overall emissions,
followed by the  Ford vehicles and the Plymouth vehicles, respectively.
In terms of fuel economy, the Honda CVCC, as expected, achieved the
highest mileage  in both the 75 FTP and HWFET. Of the domestic production
vehicles, the Plymouths achieved the best overall mileage, followed
closely by the Chevrolets and the Ford vehicles, respectively.

              Table 2-a. 1975 FTP GASEOUS EMISSIONS (grams/mile)
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
N
9
10
9
9
25
30
29
HC
0.281
0.284
0.558
0.716
0.581
1.110
0.245
CO
4.213
3.189
5.393
13.042
4.665
4.123
2.890
NOX
1.933
1.786
2.624
1.364
1.060
1.492
1.270
C02
635
522
675
488
566
605
366
Fuel economy,
mpg
13.63
16.55
12.80
17.02
15.24
14.29
23.56
                                   17

-------
              "able 2-b. 1975 HWFET GASEOUS EMISSIONS (grams/mile)
Vehicle
Impala
Nova
Fury
Scamp
Grartaca
Torino
Honda
N
9
9
8
7
25
30
25
HC
0.119
0.126
0.155
0.420
0.300
0.543 .
0.058
CO
1.815
0.399
0.685
1.284
0.873
2,436
0.853
NOX
2,496
1.392
3.198
1.793
1.342
2.077
1,494
C02
435
387
451
341
457
478
300
Fuel economy,
mpg
19.89
22.44
19.29
25,41
19.17
18.06
28.85
     With the stock spark timing it was noticed that the Chevrolet and
Plymouth vehicles would ping under mild acceleration. Therefore, the
spark timing of the ChevroTet Impala and Nova and the Plymouth Fury and
Scamp was retarced 5° for some 75 FTP runs.  This experiment was performed
for two reasons:  first, to check upon the emission levels and fuel
economy with the' retarded spark, and second, to check upon vehicle
driveability and the"level of "pinging  With the retarded spark. The
results are listed in Table 3 and in all cases there was an overall 6$
decrease in fuel economy associated with the retarded spark.  Both
Chevrolets showed an increase in HC and CO emission levels and a decrease
in the NO  emissiipn level.  Both PLymouths exhibited a decrease in
         A
observed HC emissions-and an increase in CO emissions.  The Plymouth
Fury emitted 8% .less NO  under the retarded spark condition while the
                       X
Plymouth Scamp omitted 9% more NO  under.identical conditions.  The
                                 A
driveability of all vehicles with the retarded spark remained essentially
the same as whei^tested with the stock spark,  The. "pinging" associated
with the stock spark under mild acceleration did disappear with the 5°
spark retard. Thus it appears that one could retard- the spark timina
approximately 5(> in order to decrease the "pinging" arising from use of
currently available unleaded-gasolines.  However, the penalty associated
with this operation is decreased fuel economy and an increase in regulated
emissions.
                                     18

-------
Table 3. COMPARISON OF 1975 FTP GASEOUS EMISSIONS FROM VEHICLE TIMED
  AT STOCK SPARK IGNITION WITH VEHICLE TIMED AT 5° RETARDED SPARK
               FROM MANUFACTURER SPECIFICATIONS
1
Vehicle
Impala
Impala
Nova
Nova
Fury
Fury
Scamp
Scamp
Spark
Stock
5° Retard
Stock
5° Retard
Stock
5° Retard
Stock
5° Retard
, 1975 FTP Gaseous Emissions, grams/mile ,
N
9
4
10
4
9
5
9
5
Chevrolets:
Average Difference
(5° Retard/Stock)
Plymouths:
HC
0.281
0.344
0.284
0.313
0.558
0.445
0.716
0.641
+16%
- 15%
CO
4.213
9.127
3.189
5.718
5.393
6.841
13.042
13.780
+ 98%
+16%
NOX
1.993
1.604
1.786
1.587
2.624
2.421
1.364
1.493
-15%
+ 1%
C02
635
700
522
545
675
702
488
548
+7%
+8%
Fuel economy,
mpg
13.63
12.31
16.55
15.80
12.80
12.27
17.02
15.75
-7%
-6%
                              19

-------
SULFUR-RELATED EMISSIONS

     The sulfur-related emissions listed in Tables 4-10 detail the fuel
sulfur levels, the SOo emission rates in grams per mile, the S02 sulfur
balances, the particulate emission rates in milligrams per mile, the
sulfate emission rates in milligrams per mile, the sulfate percent of
the particulate matter, the SO^ sulfur balances, and the total sulfur
balances for the 75 FTP and the HWFET for the seven test vehicles studied.
Table 11 lists the S02 linear regression analysis, Table 12 lists the
SO. linear regression .analysis-,, and Table 13 lists the particulate
matter linear regressioii'analysis, aJT a5 a function.of fuel sulfur
level.  After 'initial highway conditioning on a particular test fuel,
the driving schedule-involved alternating 75 FTP'S and HWFET '•$' for each
vehicle.  The Chevrolet and Plymouth vehicles  sequential order of fuel
sulfur level during testing^as 0-087, 0.020, 0.036, and 0.053 weight
percent sulfur. The Ford and. Honda vehicles' seguential order of fuel
sulfur level during testing was 0.020, 0,041, 0,059. and Q.084 weight
percent sulfur. Consequently, there is a substantial amount of data and
analyses presented for^the seven test vehicles in the two driving cycles
using four different sulfur level fue^s rangwg in vjalue from O.^EO to
0.087 weight percent sulfilir.

     The S02 emission rates a-nd SOo sulfur balances vary considerably.
They are dependent upon the type of vehicle/catalyst, the fuel sulfur
level, and the driving cycle.  For example, the pellet catalyst equipped
Chevrolets (Nov«i and Impala) are the lowest S02 emitters in the 75 FTP,
and, conversely, are the highest SOp emitters in the HWFET.  This is
expected due to the nature of the pellet catalyst which, because of its
large surface and alumina contact area, has a greater capacity to store
sulfur oxides than do monolithic catalysts.  As mentioned previously,
sulfur oxide storage tends to occur during low speed/temperature driving
modes as the 75 FTP, while sulfur oxide release tends to occur during
high speed/teirporature driving modes as in the HWFET.  The pellet catalyst's
S02 emission rate supports this mechanism.  The correlation coefficients
                                     20

-------
from the linear regression analysis in Table 11 for the 75 FTP ranged in
value from 0.962 to 0.997 (average 0.984 ± .012) indicating excellent
linear association of S02 emission rate with fuel sulfur level.  The
regression slopes ranged in value from a high of 3.918 (Ford Granada) to
a low of 0.575 (Chevrolet Nova).  The order of decreasing FTP S02
emission rate was Fury > Granada > Torino > Honda > Scamp > Impala > Nova.
The correlation coefficients for the HWFET ranged in value from 0.899 to
0.996 (average 0.965 ± .033) again indicating very good linear association
of SOo emission with fuel sulfur level.  As expected, the regression
slopes were greater in the HWFET than in the 75 FTP and ranged in value
from a high of 5.584 (Chevrolet Nova) to a low of 1.776 (Honda).  The
order of decreasing HWFET SO^ emission rate was Impala > Nova > Torino >
Granada > Sfiafep > Honda > Fury.

     The SO^ Emission rates and S04 sulfur balances vary considerably
also, depending on the type of vehicle/catalyst, on the fuel sulfur
level, and on the driving cycle.  The correlation coefficients from
the linear regression analysis in Table 12 for the 75 FTP ranged in value
from 0.674 to 0.990 (average 0.909 ± .116) indicating good linear association
of S0« emission with fuel sulfur level.  The regression slopes ranged in
value from a high of 463.369 (Ford Torino) to a low of 15.086 (Honda).
The order of decreasing 75 FTP SO^ emission rate was Torino > Granada >
Fury > Nova > Scamp > Impala > Honda. The correlation coefficients for
the HWFET ranged in value from 0..959 to 0.996 (average 0.976 ± .017)
indicating very good linear association of SO^ emission with fuel sulfur
level.  As usual, the regression slopes were greater in the HWFET and
ranged in value from a high of 1336.073 (Chevrolet Nova) to a low of
81.346 (Honda).  The order of decreasing HWFET SO^ emission rate was
Torino > Granada > Nova > Fury > Impala > Scamp > Honda.

     The particulate emission rate data was likewise variable depending
on the type of vehicle/catalyst, on the fuel sulfur level, and on
the driving cycle.  The correlation coefficients from the linear regression
analysis in Table 13 for the 75 FTP ranged in value from 0.164 to 0.994
(average 0.788 ±  .285) indicating fair linear association of particulate

                                   21

-------
emission with fuel sulfur level.  The regression slopes ranged in absolute
value from a high of 909.846 (Ford Torino) to a low of 4.607 (Honda).  The
Chevrolet Nov Fury >
Granada > Nova > Impala > Scamp > Honda.  The correction coefficients
for the HWFET ranged in value from'O.'SSl to 0.999 (average Q.96&'±-  .040)
indicating very good linear-association of ^articulate'emission with
fuel sulfur level.  As expected,- the' regress ion* slopes We re* .greater 'in
the HWFET and rangefd in Va'Tue from a high of 2303V632; (Chevrolet Nova)
to a low^of 148.419 (Honda).  The order of •'decreasing HWFET-particulate
emission rate was Torino > 'Nova > Granada > Fury > F'mpala > Scamp > Honda.
The sill fate percent) of thef^particulate matter varied from* 4.2 to 55.1%
of the particulate matter (average 23.0 ± 14.5%?) for all seven
vehicles in tire -75 FTP and Varied from 25.0 to 69:0% (average 50.3 ±"8.9$)
in the HWFET'.  Thus-approximately one half of the- HWFET partfrculate matter
was sulfate, while •approximately one quarter of the 75 FTP particulate matter
was sulfate.  The total sulfur balances presented in
-------
SO, emitter in the 75 FTP and HWFET was the air pump equipped Ford Granada
with the Matthey-Bishop monolith.  The addition of excess air resulting
in a 4% to 7% tailpipe oxygen level apparently enhances the kinetic
rate of oxidation of SOp to S(L;  hence, the greater sulfate emission
rates for air pump equipped vehicles.  The Chevrolet vehicles with the
GM/AC pellet catalysts were the lowest S02 emitters and, with the exception
of the Honda and Plymouth Scamp were also the lowest emitters of SO^ and
particulate matter during the 75 FTP.  This result is not surprising due to
the large storage capacity of the pellet catalyst.  During the HWFET,
however, the Chevrolets were the greatest S02 emitters, the Nova the
third greatest S04 emitter and the Impala the fifth greatest S04 emitter.
The Plymouth vehicles were equipped with the Universal Oil Products
monolithic catalysts.  In the 75 FTP, the Fury ranked first, second, and
third in SOp, particulate matter, and SO, emissions respectively, while the
Scamp ranked fifth, sixth, and fifth respectively.  In the HWFET, the
Fury ranked seventh, fourth, and fourth while the Scamp ranked fifth,
sixth, and sixth in S02, particulate matter, and S04 emission respectively.
The non-catalyst Honda emitted the least particulate matter and the
least S04 in both the 75 FTP and HWFET.  Overall, the Honda had the
lowest regulated gaseous emissions and the lowest sulfur-related emissions.

NON-SULFATE PARTICULATE MATTER

     The fuel parameter exerting the greatest influence over total
particulate emissions was the fuel sulfur level.  Tables 4-10 describe
the particulate emission rates in milligrams per mile as a function of
fuel sulfur for both the 75 FTP and the HWFET.  Table 13 describes
the linear regression analysis for the particulate emission rate as a
function of fuel sulfur level.  The correlation or measurement of linear
association between the particulate emission rate and the fuel sulfur is
especially good for the six catalyst vehicles.  Excepting the Chevrolet
Nova, which had an unusually high particulate emission rate using 0.020%
sulfur fuel, the correlation coefficients for the five other catalyst
vehicles in the 75 FTP ranged in value from 0.904 to 0.994 (average
0.958 ± .038).  In the HWFET the.correlation coefficients ranged in
                                     23

-------
ro
                        Table 4. CHEVROLET !MPALJA §Q2 SO4' AND-PAHTlBUUVre EMISSIONS



                                                75rF.TC« Results
                                               HWFiET Results
No.
of
tests
3
3
o
6
Fuel
% sulfur
level
0.020
0.036
Ontrc
0.087
S02,
g/miie
0.017
0.043
0.168
FuelS
asS02
18.7%
29.3%
49.7%
Partfcutetei
m'd/mfle
4.S
5.4
*j a
/.o
9.4
S04l
m(J7rn7fe~
is
£6
jho
f.6
<96sg>4
partictilate
6.7
IV!
1 R'^.
ir^o
FuelS
asSO4
0.3%
0.3%
0-AOi
,*r/0
0.3%
Total
sulfur
balance
0.190
0.296
0.500
3
3
3
6
0.020
0.036
0.053
0.087
0.070
O..WO
0.378
0.4^14
1 2KO%
119?0%
252,1%
17U8%
3.1
w.q
M. S
32.4
1.4
TiO
11.8
17i4
45i2
^74
54.J6
53.7
1.7%
5.0%
5.5%
3.3%
1.267 ,
1.150
2.626
1.751

-------
ro
in
                        Table 5. CHEVROLET NOVA SO2, SO4, AND PARTICULATE EMISSIONS


                                               75 FTP Results
No.
of
tests
3
3
3
7
Fuel
% sulfur
level
0.020
0.036
0.053
0.087
S02,
g/mile
0.002
0.022
0.023
0.044
FuelS
asS02
2.8%
18.2%
12.9%;
5.5%
Particulate,
mg/mile
73.9
16.4
9.2
21.3
S04,
mg/mile
3.1
0.8
1.1
6.6
%S04
Particulate
4.2
4.9
12.0
31.0
FuelS
asS04
3.2%
0.4%
0.4%
1.6%
Total
sulfur
balance
0.060
0.186
0.133
0.171
                                              HWFET Results
3
3
3

0.020
0.030
0.053
0 087

0.015
0.090
0.199

33.3%
101.1%
153.1%

7.2
25.3
42.0
15Q A

1.8
11.5
21.4
RQ Q

25.0
45.5
51.0
Rfi A
OO.*T
4.3%
9.8%
10.5%
09 oq£,
O£.4£70
0.376
1.109
1.636


-------
ro
                         Table 6. PLYMOUTH FURY SQ2, SO4, AND PARTICU LATE EMISSIONS
                                                75'FTP Results
No.
of :
tests
3
3
3
6
Fuel
% sulfur
level
0.020
0.036
ap53
0.087
S02,
fi/milej
0.060
0.139
0.209 ;
0.290
FuelS
asSO2
€3.2%
88.5%
90.1'% i
79.2%
Participate, '
mg/mile
5:4
8.5
.15.6
35.0
S04,
mg/mile
^.t
1.9
4.T
19.3
%S04
Part icu late
20.4
22.4
26.3
55.1
FuelJS
asSO4
0.9%
0.8%
1.2%
3. .5%
Total
sulfur
balance
0.641
0.893
0.913
0.827
                                                HWFET Results
3
3
3
5
o!b2b
,0.036
0.053
0.087
0.047
0,086
0.112
0.200
75.8%
8t9%
72.7% :
82.0%
id.2
T6.7
38.1
126. 1
5.5
9-7
20.6
71.3
53.9
58,-T
54.1
56.5
6.7%
6.5%
9.5%
14.1%
0.825
0.884
0.822
0.961

-------
ro
                         Table 7. PLYMOUTH SCAMP SO2, SO4, AND PARTICULATE EMISSIONS



                                               75 FTP Results
No.
of
tests
3
3
3
6
Fuel
% sulfur
level
0.020
0.036
0.053
0.087
S02,
g/mile \
0.017
0.073
0.131
0.186
FuelS
asS02
25.4%
64.0%
78.4%
69.9%
Particulate,
mg/mile
2.5
4.1
3.2
8.8
S04,
mg/mile
0.2
1.3
0.7
2.0
%SO4
Particulate
8.0
31.7
21.9
22.7
FuelS
asSO4
0.2%
0.8%
0.3%
0.5%
Total
sulfur
balance
0.256
0.648
0.787
0.704
                                               HWFET Results
3
3
3
4
0.020
0.036
0.053
0.087
0.042
0.117
0.168
0.209
87.5%
140.0%
118.0%
11376%
2.7
8.9
14.2
22.5
i 1.8
4.4
9.8
14.3
66.7
49.4
69.0
63,6
2.1%
3.5%
6.2%
5.0%
0.896
1.435
1.242
1.186

-------
ro
oo
                          Tabie 8.  FORD GRANADA SO2, SG4, AND PARTiCULATE EMISSIONS



                                                75 FTP Results
No.
of
tests
7
8
5
5
Fuel
% sulfur
level
0.020
0.041
0.059
0.084
S02,
g/mile
0.048
0.111
0.188
0.296
FuelS
asSi02
*65,B%
76.0%
87.0%
96.7%
Paniculate,
mg/rflile
JB3
16.9
10.-9
20.6
(
S04,
ml/mile
l.-l
3.9
5.0
8.7
%SO4
Paniculate
15.9
35.8
45.9
42.2
FuelS
asS04
1.0%
1.8%
1.5%
1.9%
Total
sulfur
balance
0.668
0.778
0.885
0.986
                                                      Results
6
9
5
5
0.020
0.041
0.057
0.084
0.089
0:«2
0.193
0.272
rs3.-4%
94v9%
1 10.9%
11076%
1-9.3
3SV6
39.2
56.1
8.6
1-8;4'
18.5
26.4
44.1
51.7
47.-2
47.1
11.2%
10.5%
7.1%
7.2%
1.646
1.054
1.180
1.178

-------
ro
                           Table 9. FORD TORINO SO2, SO4, AND PARTICULATE EMISSIONS



                                                75 FTP Results
No.
of
tests
8
10
6
6
Fuel
% sulfur
level
0.020
0.041
0.059
0.084
Sp2,
g/mile
0.067
0.101
0.160
0.226
FuelS
asS02
83.8%
64.3%
70.2%
71.1%
Paniculate,
mg/mile
19.2
33.8
50.0
77.2
S04,
mg/mile
3.8
13.8
21.2
33.7
%SO4
Paniculate
19.8
40.8
42.4
43.7
FuelS
asSO4
3.2%
5.8%
6.2%
7.0%
Total
sulfur
balance
0.870
0.701
0.764
0.781
                                                HWFET Results
6
9
6
5
0.020
0.041
0.059
0.084
0.084
0.109
0.195
0.249
133.3%
89.3%
108.9%
95.4%
33.5
66.2
97.4
135.1
15.4
34.1
43.2
62.7
46.0
51.5
44.4
46.4
16.3%
18.7%
16.1%
16.1%
1.496
1.080
1.250
1.115

-------
CO
o
                           Table 10. HONDA CVCC SO2, 864, A1\!D PARTICULATC EMISSIONS


                                                 75 FTP Results
No. ;
:of
tests
8-
9
6
6
Fuel
% sulfur
level
0:020
0.041
0.059
0.084
S02,
,,g/rnjle
0.071
0.093
0.134
0.226
FueTS
asS02
127.7%
98.9%
96.4%
112.4%
Particulate,
ma/mile
5.2
5.Z
4.0
-5.8
S04,
mq/,mile
0.2
0.4.
0.5
1*2
%SO4
Particulate
4.6
7.7
12.5
20.7
FuelS
as 6Q4
0.3%
0.3%
0.2%
0.4%
Total
sulfur
balance
1.280
0.992
0.996
1.128
                                                HWFET Results
6
7
6
6
.0.020
.Q.Q41
: 0,059
0.084
0.072
0:087
0.1-16
0.185
189.5%
108.8%
1.03v6%
114.2%
13,3,.
13.4
14.£
22.9
*.6
6.3
8.2
9.7
34.6
47,6
55;0
42.4
8.2%
5.3%
4^%
4.0%
1.977
1,141
1.085
1.182

-------
Table 11. SO2 LINEAR REGRESSION ANALYSIS
SO2 (g/mile) = a + b (FUEL WEIGHT % SULFUR)
                 75 FTP
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
a = Intercept
-0.034
-0.005
0.010
-0.020
-0.039
0.008
-0.004
b = Slope
2.302
0.575
3.364
2.490
3.918
2.552
2.593
P _ Correlation
coefficient
0.997
0.962
0.983
0.979
0.996
0.992
0.980
               HWFET
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
a = Intercept
-0.030
-a 102
0.001
0.015
0.014
0.019
0.024
b = Slope
5.581
5.584
2.254
2.405
3.000
2.748
1.776
R _ Correlation
~ coefficient
0.899
0.992
0.996
0.953
0.978
0.977
0.962
                  31

-------
Table 12. SO4 LINEAR REGRESSION ANALYSIS
SO4 (mg/mile) = a +.b f FUEL WEIGHT % SULFUR)
                75 FTP
Vehicle
Impala
Nova
Fury
Scannp
Granada
Torino
Honda
as* intercept
0.049
0.173
7.152
0.063
, 1.176
5.507
0.194
b~Stop"e
T9.!879
62.713
! 280.648
22.713
114.724
483.369
15.086
R _ Correlation
coefficient
0.975
0.674
0.941
0.839
0.991
0.999
0.942
               HWFET
Vehtete
Impala
Nova
Furv
Scamp
Granada
Torino
Honda
a — Intercept
2.043
34,318
22.943
1.812
4.877
2.079
3.051
fa - Slope
233:522
J336.073
i$t4.656
191.579
256.820
721.003
81-346
R Correlation
coefficient
0.9B2
0.959
0.959
0.984
0.958
0.996
0.994
                    32

-------
                  Table 13.
PARTICULATE MATTER LINEAR REGRESSION ANALYSIS
PARTICULATE (mg/mile) = a + b (FUEL WEIGHT % SULFUR)
                    75 FTP
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
a = Intercept
3.051
60.157
6.199
0.258
2.095
1.352
4.815
b = Slope
75.992
611.377
455.587
89.636
200.587
909.846
4.607
R _ Correlation
coefficient
0.975
0.593
0.984
0.904
0.904
0.994
0.164
                    HWFET
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
a = Intercept
2.563
54.354
40.133
2.165
9.856
1.574
8.556
b = Slope
419.150
2303.632
1794.049
290.607
543.993
1597.561
148.419
Correlation
coefficient
0.979
0.961
0.961
0.994
0.984
0.999
0.881
                      33

-------
Table 14. OVERALL RANKING BY S02, PARTICIPATE,
           AND SO4 EMISSION RATES

                   75 FTP
Vehicle
Impala
Nova
Fury
Scamp
Granada
Torino
Honda
S02 emission .
6
7
1
5
2
3
4
Paniculate emission
5
4
2
6
3
1
7
S04 emission
6
4
3
5
2
1
7,
                  HWFET
Vehicle
Impala
Novel
Fury
Scamp
Granada
Torino
Honda
SO2. emission
1
2
7
5
4
3
6
ParticuHafe emission
5
2
4
6
3
1
7
S04 -emission
5
3
4
6
2
1
7
                    34

-------
value from 0.961 to 0.999 (average 0.980 ± .016) for the six catalyst
cars.  The Honda CVCC particulate emission rate in the 75 FTP was
constant as 5.1 ± 0.8 mg/mile, while its participate emission behavior
in the HWFET was similar to that of the catalyst vehicles with a
correlation coefficient of 0.881.  The Honda was generally the lowest
particulate emitter in both the 75 FTP and the HWFET for most fuels
studied.

     As expected, a great portion of the particulate matter was sulfate and
its associated bound water.  The sulfate percent of the particulate
matter averaged 50.3 ± 8.9% for all seven test vehicles in the HWFET
and 23.0 ± 14.5% in the 75 FTP.  In other words, at least one half of
the particulate matter was sulfate in the HWFET runs decreasing to a
quarter of the particulate matter being sulfate in the FTP runs.

     The non-sulfate portion of the particulate matter may be divided
into two general categories:  organic matter and inorganic matter.
Organic matter was determined by elemental combustion analyses performed
upon Gelman type A glass fiber filters.  These analyses, reporting percent
carbon and hydrogen, indicated that organic material comprised from a
minimum of ^2% to a maximum of ^35% of the total particulate mass for
the six catalyst vehicles.  In the case of the Honda CVCC, the organic
material material composed ^23% of the total  particulate mass in the 75
FTP and -x-51% of the total particulate mass in the HWFET.  No organic
extractions have been performed upon the particulate samples in order to
clarify the composition of the organic material.  Further investigation
into this area is warranted.

     The remaining non-sulfate, non-organic particulate matter is of
inorganic origin.  Accordingly, the following metals were measured by
x-ray fluorescence for the seven test vehicles:  silicon (Si), cadmium (Cd),
aluminum (Al), sulfur (S), magnesium (Mg), potassium (K), zinc (Zn),
nickel (Ni), iron (Fe), titanium (Ti), barium (Ba), calcium (Ca),
platinum (Pt), and vanadium (V).  The elemental carbon, hydrogen and
metal emission rates in milligrams per mile are entered in Tables 15-18.
                                     35

-------
Because x-ray fluorescence does not provide information concerning  the
combination state of the analyzed metal, it is a reasonable assumption
that the metals are combined with other elements in the form of oxidation
products.

     Elements oF prominent concentration in the particulate matter  were
sulfur, iron, aluminum, zinc and calcium.  The other elements detected
were present in trace quantities-only and appear to follow no pattern
with regard to fuel sulfur level or with regard to the driving cycle
selected.  No platinum was detected-in the analyses.  With the exception
of sulfur, thenost prominent element detected--iji the ^articulate matter
was iron.  The iron emission rajte for all test.vehicles, wa'th ,-the exception
of the Honda and the Chevrplets, wa:s greater in therJS FTP than in  the
HWFET.  Also, the, :ca,talyst vehicle ^i,r on ^mission, .rate was, greater than
that of the non-catalyst Honda CVCC.  The iron presumably arises from
engine and exhaust system wear in 
-------
were the greatest aluminum emitters, followed by the Engelhard monolith
(Ford Torino), the Universal Oil Products Monoliths (Plymouths), and the
Matthey-Bishop monolith (Ford Granada).

     The fourth and fifth most prominent elements detected in the
particulate matter were calcium and zinc respectively.  The calcium and
zinc emission rates were greater for the non-catalyst Honda than for the
other six vehicles, with the exception of the Chevrolets in the case of
zinc emissions. The reason for the GM/AC pelleted catalyst-equipped
vehicles generating the greatest zinc emission is uncertain.  Perhaps
the zinc emitted is a corrosion or reaction product of H2S04 with the
Chevrolet exhaust system.  Similar observations have been reported by
         29
Beltzer.    In the case of the Honda, the increased calcium and zinc
emissions appear derived from lubricating oil consumption.  We may
determine a mass balance for the particulate matter by assuming that the
prominent elements listed in Tables 15-18 are in reality emitted as
oxidation products.  That is, Al is emitted as A1203, Fe is emitted as
FepO-, and S is emitted as H2SO. aerosol.  Carbon, however, is assumed
to be emitted in its elemental form as a carbonaceous product only.
Consequently, we may sum the masses of the individual products and
compare this resultant sum to the experimental particulate emission
rates listed in Tables 4-10.  For the six catalyst cars, the sum of the
individual products gives 85 + 19% of the experimental particulate
emission rate in the 75 FTP and 66 ± 16% in the HWFET.  In the case of
the Honda CVCC, the sum of the individual products gives approximately
55% of the experimental particulate emission rate in the 75 FTP and
approximately 103% in the HWFET.  The overall mass balance for the seven
test vehicles was 81 ± 21% in the FTP and 72 ± 20% in the HWFET, and
considering the rather simple assumptions used, this gives a fair
indication of the chemical composition of the particulate.  The particulate
material balances are illustrated in Figure 4.

     In summary, S02, particulate matter, and H2S04 emission rates
increased with increasing fuel sulfur for all vehicles.  The reproducibility
and magnitude of the S02, particulate matter, and SO. emissions were
                                     37

-------
greater in the. HWFET than in the 75 FTP.  The accuracy and magnitude  of
the emission rates are directly attributable to the characteristics of
the driving cycles.  The HWFET is essentially a steady state operation
at 48 rjph covering 10.24 miles in 12.75 minutes while the 75 FTP  represents
city driving with frequent accelerations and decelerations for a  transient
type of operation with an average speed of 22 mph covering 11.09  miles
in 31 m'nutes   In addition, the average catalyst temperature is  approximate-
ly 25% greater during the HWFET than during the 75 FTP.  Sulfur oxide
storage was ol»served to occur, during the low speed/temperature transient
modes (75 FTP) only to be released during the higher speed/temperature.
modes (HWFET).  fhat i.s,\- sulfur,, oxide storage accounts for total  sulfur
balances bein« consistently less than, theoretical in the 75 FTP and
consistently greater than theoretical. In the HWFET.  Ttie sulfur joxide
storage/release phenomena is,also dependent,upotn the type-of catalyst in
add! it Ion, to the type of driving cycle.  Pelleted catalysts'with their
large surface and alumina contact; areas, had the greatest.sulfur oxide
storage capacity.,,.;As.a .result, we observed that the. pelleted catalysts
           ......     •.   ,-. v's   ~
were the lowest sulfur emitters-,in the 75 FTP,.and-among the:highest
emitters in the hWFET.  The^a.ir,pump.equipped monolithic,catalysts
operating at 4-7X tailpipe oxygen emitted the greatest, swl fate ^and
particulate matter. ^Tq decrease, the rate of catalytic, oxidation-of SOx
to SOo and, hence tp HgSO. would-be both %a; kinet-iq ^andOthenriodynamic
problem and wou'lc affect catalytic oxidation of excess HC-and CO, since
the catalytic properties that affect HC and CO. pxi.dation likewise affect
SO^ oxidations  Catalyst vehicle particulate matter was.composed  mostly
of sulfate; the ron-sulfate portion was composed,of carbonaceous  material
and aluminum and iron oxidation -products. .No...platinum was,detected in
the particulat.e matter.
                                     38

-------
        Table 15. ELEMENTAL ANALYSIS OF PARTICULATE MATTER
             BY X-RAY FLUORESCENCE:  1975 FTP RESULTS?
                           (milligrams/mile)
"^*\^ Vehicle
.Element ^^\^
SI
Cd
Al
S
Mg
K
Zn
Ni
Fe
Ti
Ba
Ca >
Pt
V
Cb
Hb
Chevrolet
Impala
0.052
0.001
0.133
0.757
0.004
0.008
0.065
0.008
1.085
0.002
0.005
0.007
ND
0.003
1.97
1.13
Chevrolet
Nova
0.022
0.002
0.435
2.740
0.004
0.012
0.080
0.008
3.224
0.002
0.021
0.021
ND
ND
7.46
3.37
Plymouth
Fury
0.012
ND
i 0.045
6.212
0.003
'0.003
ND
ND
0.575
ND .
0.003
0.005
ND
ND
2.64
2.35
Plymouth
Scamp
0.013
ND
0.054
0.853
0.001
0.004
0.019
0.006
1.035
ND
0.003
0.014
ND
ND
2.29
1.56
Detection
<\ limit •' \
0.001
0.001
0.005
0.005
0.001
0.001
0.010
0.001
0.022
0.001
0.002
0.002
0.001
0.001
—
—
a Vehicles run on 0.087% sulfur fuel.

"Combustion analysis results.

ND = Not detected.
                                 39

-------
Table 16. ELEMENTAL ANALYSIS OF PARTICULATE MATTEH
BY X-RAY FLUORESCENCE:; HWFET (RESULTS8
(milligrams/mileT
\^ Vehicle
Element ^^^^^^
/ » • ' *' ' *
=• t j
"» 0 Cd 'EN
i

!„ ,
;-

: •<£• .0 Mgi --1
rc"'.c K 1 *3
i
S^.OZnj  !
Hb
Chevrolet
Impala .
~- 0.026
•J 0.001
t
' O.OJ54
'- 5.721
> •' ' *
JA> 0.003
k n i f
r 0.007
ij 0.091
•10,0.665
*s n
J°2.309
|

1 0.005
I
°%008
3." ND|
i
1/1 NDi

12.82
2.27
Chevrolet
Nova
O.O20
' 0.003
• 1
.7- i '
0.120,
"21885!

! 'oloos!
;r.r, ..."; I

iU O.120!
^.Oni

5.456 '
•* ^ •* r*
ISI D
*T/|
"0.004

"i;6io8
at'M^D
i . /
ND j

"3.03 !,
4.72
Plymouth .
* •- Fury
f 77 0.01(7" ;g
..'^aooii !.

..^^osj7 r
Ol2t694 ..
j-.|, ,,.(S 	 „ Vr-
* 0,005 ..
Kjt .0 ; **
,.,0,007 a(
*-• " '>" ""^ 1
IM ^J I
w Vo08 '
80C.O i H
0.316 .
KC"*?* 4-* j ." p
1 0-003, c
Aw'o!bo4l "
J'SC '} ' *'
' 0-087I ,
T'ND j •

"',ND \ t
•E.i 1 f
'4.49 1
V()4 j

.iPlymouthj i .
.^rScamp;" j
j'" a013 "[;;*
;-• ND >'

r_n 0.016 ;/t
v- , 4.686
i ' , i
" " " j
•0 0 ND ' ! ^
^D0-^6 !'^
r,n 0.026 ;nc
'-• ' ' ^
,'j.O ' i l;/"
,.. . NP
3 v - . f ' " '
r ND ! •. r
*. ,,0.007 • i B:
f-sj.i. ' Sv
.-,0.0,15 : -
^ ... ND
Qv! {
,„„ ,. ND

', Jj3-49 • ;;
2.79
                                                                        •J
a Vehicle:; run on 0.087% sulfur fuel.
b Combustion analysis results.
ND = Not detected.
                                 40

-------
                Table 17.
 ELEMENTAL ANALYSIS OF PARTICULATE
   MATTERJY X-RAY FLUORESCENCE:
           1975 FTPTRESULTS8
             (milliyams/mile)
^-^^ Vehicle
; Element ^""^^^^
Si
Cd
Al
S
Mg
K
Zn
Ni
Fe
Ti
Ba
Ca
Pt
V
Cb
Hb
Ford
Granada
0.007
ND
0.029
1.397
0.002
0.002
0.017
ND
0.729
ND
ND
0.005
ND
ND
2.39
0.06
Ford
Torino
0.011
ND
0.190
4.729
0.002
0.002
0.017
0.005
1.614
0.002
0.003
0.011
ND
ND
6.41
0.07
Honda
CVCC
0.009
ND
0.008
0.125
0.005
0.002
0.030
0.005
0.125
0.001
0.003
0.041
ND
ND
2.13
0.05
a Vehicles run on 0.041% sulfur fuel.

b Combustion analysis results.

ND = Not detected.
                   41

-------
               Table 18.
 ELEMENTAL ANALYSIS OF PARTICULATE
   MATTER BY X-RAY FLUORESCENCE:
           HWFET RESULTS*
            (milligrams/mile)
^""^\^ Vehicle
Element ^^\^
Si
Cd
Al
S
Mg
K
Zn i
Ni i
Fe
Ti
Ba
Ca
Pt
V
Cb
Hb
Ford
Granada
0.006
ND
0.008
5.017
0.019
0.002
Oi016
ND
0.153
Oi006
ND
0.002
ND
ND
5.27
! 1.10
Ford
Torino
0.005
0.001
0.047
10.105
0.001
0.002
0.015
0.005
0.564
0.001
0.005
0.008
ND
ND
2.14
0.20
Honda
CVCC
0;062
ND
0.060
2.019
0.007
o.on
0.035
0.007
0.224
0.005
0.003
0.051
ND
ND
6:75
0.26
 Vehicles run on 0.041% sulfur fuel.

b Combustion analysis results.

ND = Not detected.
                  42

-------
                 1.20
             S   1.00
             CO
CO
                 0.80
                 0.60
             D
             O



             5   0.40
             Q,
                 0.20
                                                   I* • ' • I

                                                   llX'v  ORGANIC MATERIAL
                                                   |.«« . I
                                                          INORGANIC MATERIAL
                                                          SULFATE MATERIAL
                                      till
                          IMPALA
NOVA
FURY        SCAMP     GRANADA


       FTP/HWFET RESULTS
TORINO
HONDA
                                              Figure 4. Vehicular particulate material balance.

-------
                                 REFERENCES

1.   R.L.  Bradow and J.B.  Moran,  "Sulfate Emissions  from Catalyst  Cars,
     A Review"  SAE Paper 750090, February 1975.

2.   L. Niemeytjr, "Estimated Changes in Human  Exposure  to Sulfates
     Attributing to^fquipping Light Duty.Motpr, Vehicles with  Oxidation
     Catalysts^ . (Paj^rjpKesen'tedral%^ational''lristitute:  of Environmental
     Heal th Sclehe^s^Syfliposium^'We^h;Consequence^ of  Envi ronmental
     Controls", [Durham, N.C., April  1974.
      *.

3.   "issue Pa^>eV;:*l£t5n§^^                        a,  Result of Equipping
     Ltght-Dut^ifet0f^e^Si«wSrOxidatibn'Catalysts", EPA,  January
     1|75.

4.   Us Gibtiey  ^Environ. Sci. Technol., 8, 793 (1974).
     W,.iR.  Lepp»r^f ^ter^^e Seartb >, inj^rinLBepjprt  on  Task  1  of
     Contract ^of>||^^4^ ^»ate:Cohfr61 .Technology  Assessment. "
     Egvi ronmerit^OrotejEtii&jai^gjfi^  JEral Ss'iJoja-* Control  Technol ogy
     Division,   pihn  Arbor, Michigan, November 1974.
6.   Ws-R. Pieriioh^d1>f|i^rl€^>ap|f-4.T.  Kummer,  "Sulfuric Acid Aerosol
     Eii ssi on£ f fe^^t<(^t4«I^PMl iM+nes "   SAE Paper 748287 .
     February %w*'.

7.   Me Beltzer,|R.J.  Campion, J.  Harlan,  and A.M.  Hochhauser,  "The
     CBijversion ^Sefe'^vejp^AB^Hr^^ie Oxidation Catalysts"  SAE Paper
     750095,
R.   D.A. Tray&efi t^;^^f^C|fSif3iPi^^^               Piehson,  "Sulfuric
     Acid and Nitrate Imissionlr from Oxidatibh Catalysts" SAE Taper
     750091, February 1975.

9.   R.H. Hammerle and M.  Mikkor, "Some Phenomena  which  Control  Sulfuric
     Acid EmisM'ori from automotive Catalysts"   SAE Paper 750097, February
     1975.

10.  R.H. Hammerle and T.J.  Truex, "The Kinetics of S02  Oxidation  for
     Various Catalyst Compositions"  SAE Paper 760090, February  1976.

11.  B. Krause, R.A.  Bouffard, T. Karmllovich, and E.L.  Kayle, "Critical
     Factors Affecting Automotive Sulfate Emissions" SAE Paper  760091
     February 1976.
                                     44

-------
12.  M. Beltzer, "Particulate Emissions from Prototype Catalyst Cars"
     EPA Report No. EPA-650/2-75-054, Chemistry and Physics Laboratory,
     National Environmental Research Center, U.S. Environmental Protection
     Agency, Research Triangle Park, N.C., May 1975.

13.  O.A. Uyehara and K.M. Watson, Ind. Eng. Chem., 35_, 541 (1943).

14.  J.H. Somers, R. Lawrence, E.E. Fett, T.M. Baines, and R.J. Garbe,
     "Sulfuric Acid Emissions from Light Duty Vehicles"  SAE Paper
     760034, February 1976.

15.  R. Knietsch, Chem. Ber., 34_, 4069 (1901).

16.  H.H. Krause, A. Levy, and W.T. Reid, J. Eng. Power 90_, 38 (1968).

17.  A. Levy, E.L. Merryman, and W.T. Reid, Environ. Sci. Technol., 4,
     653 (1970).

18.  N.A. Warner and T.R. Ingraham, Can. J. Chem., 38, 2196 (1960).

19.  N.A. Warner and T.R. Ingraham, Can. J. Chem. Eng., 40, 263 (1962).

20.  K.K. Kelly, C.H. Shomate, F.E. Young, B.F. Naybor, A.E. Salo,
     and E.H. Hoffman, U.S. Bur. Mines Tech. Paper, 685, (1949).

21.  Ford Motor Company, Ford Response to EPA Request for Data on
     Automotive Sulfate Emissions, May 1974.

22.  F.A. Cotton and G. Wilkinson, Advanced Inorganic Chemistry, Interscience
     Publishers, John Wiley and Sons, New York, New York, 1967, p.333.

23.  Federal Register, 37^, No. 221, p. 2429, November 1972.

24.  F.M. Black and R.L. Bradow, "Patterns of Hydrocarbon Emissions from
     1975 Production Cars"  SAE Paper 750681, June 1975.

25.  D.E. Burch and J.D. Pembrook, "Instrument to Monitor CH., CO, and
     C02 in Auto Exhaust." Environmental Protection Technology Series
     Report  No. EPA-650/2-73-030,  Chemistry and Physics Laboratory,
     National Environmental Research Center, U.S. Environmental
     Protection Agency, Research Triangle Park, N.C., October 1973.

26.  Federal Register. 36_, No. 84, p. 8187, April 1971.	

27.  R.L. Bradow, D.A. Carpenter, D. Klosterman, F.M. Black, and S. Tejada,
     "Sulfate Emissions from Catalyst and Non-Catalyst Cars" SAE Paper
     740528, October 1974.
                                     45

-------
28.  J. Wagman, R.L. Bennett, and K.T. Knapp, "X-Ray Fluorescence
     Multispectrometer for Rapid Elemental Analysis of Particulate
     Pollutants.."  Environmental Protection Technology Series Report
     No. EPA-600/2-76-033, Environmental Sciences.Research Laboratory,
     Office of Research and Development, U.S. Environmental
     Protection Agencyr, Research Triangle Park, N.C., March 1976.

29.  M. Beltzer, "Non-Sulfate Particulate Emissions From Catalyst Cars"
     SAE Paper 760038, February 1976.
                                     46

-------
:            APPENDIX A

'    UNLEADED GASOLINE ANALYSIS
•;	I
; Research Octane Number
; Motor Octane Number
i Reid Vapor Pressure, psia
I
' Distillation, ASTM D-86, °F
! IBP
10%
50%
90%
EP
Fia Analysis
Aromatics %
Olefins %
Paraffins %
API Gravity @60°F
Weight % C
Weight % H
Weight % S
Lead, g/gallon
93.2
84.7
10.2

90
124
203
290
372

24.0
8.3
67.7
61.6
85.26
14.01
0.020
0.00004
                47

-------
                  APPENDIX B

            FUEL SULFUR ANALYSIS
Amount of Thiophene in ml
added to 55 gal drum of fuel
0.0
37.1
69.0
74.1
139.0
148.2
222.4
278.7
Theoretical
% sulfur
0.020
0.030
0.038
0.040
0.058
0.060
0.080
0.095
Measured
% sulfur
0.020a
0.03db
0.036'9
0.041 b
0.053a
0.059lb
0.084b
0.0873
3 Exxon Research and Engineering analysis via X-ray
  Method.

'" Environmental Protection, Agency analysis by ASTM
  Lamp Method. Estimated measurement uncertainty
  is+10%.
                       48

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
i. REPORT NO.
  EPA 600/2-77-237
                                                           S.flECIPIENT'S ACCESSIOI»NO.
4. TITLE AND SUBTITLE
  GASEOUS,  PARTICULATE, AND SULFUR-RELATED  EMISSIONS
  FROM NON-CATALYST AND CATALYST EQUIPPED VEHICLES
5. REPORT DATE
 December 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                           8. PERFORMING ORGANIZATION REPORT NO.
         James  N.  Braddock
9. PERFORMING ORGANIZATION NAME AND ADDRESS
  Environmental  Sciences Research Laboratory
  Office of  Research and Development
  U.S. Environmental Protection Agency
  Research Triangle Park, North Carolina  27711
                                                           10. PROGRAM ELEMENT NO.
   1AA601
CA-18    (FY-76)
11. CONTRACT/GRANT NO.
12. SPONSORING AGENCY NAME AND ADDRESS
  Environmental  Sciences Research Laboratory-RTP,  NC
  Office of  Research and Development
  U.S. Environmental Protection Agency
  Research Triangle Park, North Carolina  27711
13. TYPE OF REPORT AND PERIOD COVERED

 In-House   11/74-11/75
14. SPONSORING AGENCY CODE
     EPA/600/09
15. SUPPLEMENTARY NOTES
16. ABSTRACT
       Gaseous,  participate, and sulfur-related emissions from six 49-state  1975
  domestic  production catalyst equipped  vehicles (two Chevrolets, two  Fords, and two
  Plymouths)  and one 1975 imported prototype non-catalyst vehicle (Honda  CVCC)  were
  measured  by a  variety of chemical analysis techniques as a function  of  fuel  sulfur
  level.  Four separate and different sulfur level  fuels were tested for  each vehicle.
  The seven test vehicles were driven in accordance with the 1975 Federal  Test Pro-
  cedure  (75  FTP) and the Highway Fuel Economy Test (HWFET) procedure.  Sulfur dioxide,
  particulate matter, and sulfate emissions  increased with increasing  fuel sulfur and
  were greater during the HWFET cycle than during the 75 FTP cycle.  The  greatest
  particulate and sulfate emitters were  the  air pump equipped Ford vehicles, while the
  lowest  particulate and sulfate emitter was the Honda CVCC.  Particulate matter was
  composed  mostly of sulfuric acid; the  non-sulfate portion of the particulate matter
  was composed of carbonaceous material  and  aluminum and iron oxidation products.  No
  platinum  was found in the particulate  matter.
17.
                               KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                             b.lDENTIFIERS/OPEN ENDED TERMS
              c.  COSATI Field/Group
  * Air  Pollution
    Automobiles
  * Exhaust emissions
    Catalytic converters
  * Sulfur dioxide
  * Sulfates   .
                  13B
                  13F
                  21B
                  07A
                  131
                  07B
13. DISTRIBUTION STATEMENT
       RELEASE TO PUBLIC
                                              19. SECURITY CLASS (ThisReport)

                                                 UNCLASSIFIED	
                                                                        21. NO. OF PAGES
                    57
                                              20. SECURITY CLASS (Thispage)

                                                 UNCI ASSTFTTD  	
                                                                        22. PRICE
EPA Form 2220-1 (9-73)
                                            49

-------