-------
TABLE OF CONTENTS (Cont'd)
F. Constituents of Organic Soluble Fraction of
Particulate 52
REFERENCES 61
APPENDIXES
A. Details of Analytical Procedures
B. Calibration and Calculations on Tunnel and
Sampling Devices
C. Data Reduction
D. .Smoke and Gaseous Emissions Data
E. Average Mass Rate and Concentration Results
F. Trace Metals in Particulate
G. Summary of BaP Results
VI
-------
LIST OF FIGURES
Figure
1 Detroit Diesel 6L-71T Test Engine 4
2 Cummins NTC-290 Test Engine 4
3 Boiling Ranges of Fuels Used in Methodology
Demonstration 7
4 Overall View of Diesel Particulate Dilution Tunnel 12
5 Detail of Exhaust Muffler Used for Testing 12
6 Detail of Perforated Exhaust Restrictors Used to
Limit Exhaust Flow Into Tunnel 12
.?
7 Tunnel Blower Counter and Timer 15
8 Schematic Section of Dilution Tunnel for
Diesel Particulate Sampling 13
9 Probes Used With 47mm Sampling Systems 15
10 47mm Filter Holders 15
11 Flowmeters and Totalizing Meters Used -with
47mm Filters 15
12 Sampling Nozzle Used with Hi-Vol System 16
13 Blower and Flow Measurement Apparatus Used
with Hi-Vol System 16
14 Modified Hi-Vol System Used with 293mm
Fluoropore Filters 16
15 Gas Pump Used with 293mm Fluoropore Filters 16
16 Flow Measurement System Used with 293mm
Fluoropore Filters 17
17 ERG Sampler-Diluter 17
18 PHS Smokemeter and Bosch Sampling Unit Being
Employed for Diesel Smoke Measurement 17
VII
-------
LIST OF FIGURES (Cont'd.)
Figure Page
19 Humidity- and Temperature-Controlled Chamber
Housing Micro-Balance for Filter Weighing 17
20 Fuel Specific Particulate Emissions from a Detroit
Diesel- Allison 6L-71T Engine, Fuel as Parameter 41
21 Fuel Specific Particulate Emissions from a
Cummins NTC-290 Engine, Fuel as Parameter 42
22 Boiling Ranges of Fuels, Lubricating Oil, and Or-
ganic Soluble Fractions of Particulate for Both En-
gines Used in Methodology Demonstration 55
Vlll
-------
LIST OF TABLES
Table Page
1 Engine Design and Operating Data 3
2 Fuels Used for Diesel Particulate Studies 6
3 Comparison of Test Fuels With "Typical"
Diesel Fuels 8
4 Typical Properties of Additives Used for Methodology
Evaluation 8
5 Coding and Composition of Fuels and Fuel-Additive
Mixtures 9
6 Summary of Analyses for Particulate Composition 19
7 Minimum Test Plan and Data Matrix for Each
Engine and Each Fuel 28
8 Planned and Actual Numbers of Analyses Carried
Out on Each Engine 28
9 Data and Calculations Used to Determine Mode
Times for the Detroit Diesel 6L-71T Engine 30
10 Data and Calculations Used to Determine Mode
Times for the Cummins NTC-290 Engine 31
11 Weighting Schedule for 11-Mode "Composite" Runs 31
12 Average Federal Smoke Test Results, Two Engines
and Six Fuels, Compared to Federal Certification Results 35
13 Summary of Steady-State Smoke Data for Detroit
Diesel 6L-71T Engine 36
14 Summary of Steady-State Smoke Data for
Cummins NTC-290 Engine 37
15 Average 13-Mode Gaseous Emissions Results 38
16 Fuel Specific Partieulate Results from Two Engines
Operated on Six Fuels, 47mm Glass Fiber Filters 40
-------
LIST OF TABLES (Cont'd. )
Table Page
17 Composite Brake Specific Particulate Emissions from
Two Engines Operated on Six Fuels, 47mm Glass
Fiber Filters 43
18 Composite Particulate Concentration Data From Two
Engines Operated on Six Fuels, 47mm Glass Fiber Filters 44
19 Analysis for Carbon, Hydrogen, Nitrogen, and Sulfur
in Particulate Samples, Detroit Diesel 6L-71T Engine 46
20 Analysis for Carbon, Hydrogen, Nitrogen, and Sulfur
in Particulate Samples, Cummins NTC-290 Engine 47
21 Average Percentage of C, H, N, and S in Particulate
by Engine and Fuel 48
22 Organic Solubles as Percent of Total Particulate 49
23 Material Balances for Metals in Organo-Metallic
Smoke-Suppressant Additive, Fuels B+ and C+ 50
24 Carbon, Hydrogen, Nitrogen, Sulfur, and Oxygen in
the Organic Solubles of Diesel Particulate 53
25 Boiling Point Distributions for Fuels, Lubricating Oil,
and Organic Soluble Fractions of Particulate Samples 56
26 Average Fuel Specific BaP Results 58
-------
SUMMARY AND CONCLUSIONS
The objectives of this project, which were to develop and demon-
strate a methodology for evaluating fuel and additive effects on diesel par tic -
ulate, have been achieved. In addition, some of the data acquired during
the demonstration phase can serve as characterization of the engines tested,
although the data on the NTC-290 will serve this latter purpose much better
than those on the 6L-71T.
Development of the methodology included calculation procedures
for relating particulate acquired to total particulate emitted by the engine,
sizing and calibration of sampling devices, and calculation procedures for
proper weighting of modes to simulate the Federal 13-mode gaseous
emissions test. This development also included quality check procedures
and programs for data reduction. On the analytical side, the methodology
development included refinement of BaP analysis procedures, development
of procedures for determination of phenols and nitrosamines in particulate,
and construction of a precision weighing chamber with controlled temperature
and humidity.
The methodology demonstration included a number of test runs on
each engine, with a total of over 1100 filter samples being collected.
Statistical analysis on the data was limited to basics, since the contract did
not allow for a large analysis effort.
A number of conclusions have been reached as a result of this
research project, and the most important ones are:
1. Due to the value of results obtained in the methodology dem-
onstration part of this project, it is recommended that the
following types of analysis be performed in future research on
a more or less routine basis: S, C, H, and N in particulate;
total particulate mass; total organic solubles; S, C, H, N, and
O in organic solubles; BaP; metals; boiling point distribution
of solubles and paraffin fractions; NMR and IR analysis (limited);
phenols (if a more acceptable technique can be worked out); and
smoke by the PHS smokemeter.
2. Due to the lack of value of results obtained thus far, it is recom-
mended that the following types of analysis not be performed in
future research unless a special need exists: nitrosamines (not
found at the limits of detection used in this project); Bosch
smoke number (superfluous when PHS numbers obtained); and
total particulate by ERG sampler (less accurate and more com-
plicated than tunnel measurements).
-------
3. The dilution tunnel technique, with proper attention to cali-
bration and sizing of all flow systems, seems to be accurate
and repeatable for determining particulate mass rates (and
consequently mass rates of particulate constituents). This
technique is recommended as the reference method for future
studies of dies el particulate, including both large engines (where
a fraction of the exhaust is diluted) and automotive-size engines
(where all the exhaust can be diluted).
4. Specific particulate emission rates (g particulate per kilogram
fuel consumed) were strongly influenced by fuel type and by
the presence of the organo-metallic smoke-suppress ant ad-
ditive. Using data from the NTC-290 engine as an example
(fuel-to-fuel comparisons are held to be more valid for this
engine), the following observations have been made:
(a) Average fuel specific particulate emissions using fuel
"B+ additive" were higher than those using fuel "B" for
all 11 steady-state conditions and the composite run.
(b) Average fuel specific particulate emissions using fuel
"C + additive" were higher than those using fuel "C" for
all 11 steady-state conditions and the composite run.
(c) Average fuel specific particulate emissions using fuel "B"
were higher than those using fuel "C" for all 11 steady-
state conditions and the composite run.
(d) Average fuel specific particulate emissions using fuel "C"
were higher than those using fuel "A" for 9 of 11 steady-
state conditions (plus one tie) and the composite run.
The same general trends were in evidence for the 6L-71T en-
gine, although not as clearly. These results are shown below
(same comparisons as parts a through d):
(e) B+ higher than B: 10 of 11 + composite
(f) C+ higher than C: 10 of 11 + composite
(g) B higher than C: 10 of 11 (+ one tie) + composite
(h) C higher than A: 7 of 11
5. As a general trend, particulate mass emission rate from both
engines increased with increasing power output at steady speeds.
Average ranges at steady-state conditions were approximately
30 g/hr to 150 g/hr for the Detroit Diesel-Allison 6L-71T engine
and 10 g/hr to 70 g/hr for the Cummins NTC-290 engine.
XII
-------
6. Particulates from the 6L-71T engine were indicative of a
hydrocarbon-type material with a small amount of excess
carbon or soot (17 percent), while those from the NTC-290
were indicative of a hydrocarbon material with a great deal
of excess carbon (160 percent). These figures were determined
from elemental analyses conducted by a commercial laboratory.
7. As averages for fuels not containing additives, 50 percent of
particulate weight could be extracted as solubles from samples
run on the 6L.-71T as compared to 9 percent of particulate
weight from samples run on the NTC-290. This result lends
confirmation to the observation that the Detroit Diesel's part-
iculate seemed more oily than that from the Cummins engine.
8. Elemental composition of the organic solubles obtained by
commercial laboratory combustion analysis indicates a pri-
marily hydrocarbon material with relatively small fractions
of nitrogen, sulfur, and oxygen.
9. Boiling point distributions of the organic soluble fractions of
particulate from the 6L-71T engine confirm that a great deal
of the material boils in the same range as lubricating oil.
Similar data from tests on the NTC-290 confirm that most of
the solubles from that engine boil in a higher range.
10. Metals in the fuel (as in the form of an organo-metallic addi-
tive) can be reconstituted in the particulate with reasonable
accuracy by X-ray analysis of particulate samples.
11. Fuel specific BaP emissions during composite runs averaged
about 0.68 mg BaP/kg fuel for the 6L-71T engine and 0.21 mg
BaP/kg fuel for the NTC-290. In brake specific terms, these
values amount to 180jUg/kW hr (130jL/g/hp hr) for the 6L-71T
and 59/^g/kW hr (44JLJg/hp hr) for the NTC-290. These results
have been checked carefully, but they are much higher than those
reported earlier by others and thus must be treated with caution.
12. As a general trend, but with a great deal of variability, smoke
density from both engines increased with increasing power out-
put at steady speeds (although baseline levels were very low
for both engines). Average ranges at steady-state conditions
were approximately 0. 2 to 2.4 percent opacity for the Detroit
Dies el-Alii son 6L-71T engine and 0.5 to 2.2 percent opacity
•for the Cummins NTC-290 engine.
13. Major gaseous emissions as measured on the Federal 13-mode
test (HC, CO, and NOX) did not appear to be strongly influenced
by fuel type or the presence of additives.
• • *
X11X
-------
14. Compared to runs on untreated fuel, runs with the organo-
metallic additive (used in fuels B+ and C+) had significantly
lower visible smoke. This trend was reversed for total part-
iculate mass emissions, however, with values for fuels B+
and C+ being consistently higher than for fuels B and C, res-
pectively.
xiv
-------
I. INTRODUCTION
Section 211 of the Clean Air Act' ' requires the Administrator
of EPA to register all fuels and fuel additives used in interstate commerce,
and to require that the additive manufacturers document additive effects
on gaseous and particulate emissions, control system performance, and
visibility. This project was initiated to develop methods for the sampling
and analysis of diesel particulate, so that the effects of fuel and additive
composition could be identified. Research and development efforts were
centered on three major areas, namely:
development of sampling and measurement procedures
for the reproducible determination of diesel particu-
late mass emissions;
development of analytical procedures for identification
of potentially harmful substances in particulate matter;
and
demonstration of the sampling and analysis methodology
using six fuels (or fuel-additive combinations) in both
a 4-stroke cycle diesel engine and a 2-stroke cycle diesel
engine.
The approaches used in this research project resulted from the
collective experience of the contractor, EPA, and others involved in air
pollution studies with diesels. Particulate sampling via dilution tunnel
as the reference technique came from both the literature'2-4) and the
direct experience of EPA personnel. The systems used to extract iso-
kinetic samples from the tunnel were suggested by EPA, but a great deal
of effort went into sizing them, calibrating them, and refining their phy-
sical construction to the extent that they became accurate and dependable.
Visible smoke measurements were conducted with the PHS opacity meter
and the Bosch sampling spotmeter due to long experience with both tech-
niques and their widespread use. Yet another measure of particulate
mass was provided by using a portable sampler-diluter built by Environ-
mental Research Corporation for EPA under contract. An evaluation of
this unit was conducted concurrently with the data-taking operation.
Some of the analytical techniques applied to diesel particulate
and/or the organic soluble fraction of diesel particulate had been quite
well developed prior to this research project. These well-developed
techniques include those for analysis of sulfur, carbon, hydrogen, nitro-
gen, oxygen., trace metals, boiling point distribution of organic solubles,
and NMR and IR spectra. A procedure for analysis of BaP was available
from the ambient air literature, but it had not been applied to the solubles
'^Superscript numbers in parentheses designate references at end of report.
-------
in diesel particulate. Consequently, a considerable amount of develop-
ment work had to be done for this application. New techniques were
developed essentially from scratch for analysis of phenols and nitro-
samines in particulate. No success was achieved in developing a pro-
cedure for organic peroxide analysis. Gaseous emissions and smoke
were measured by well-accepted instrumentation and procedures.
The engines used in the methodology development and in its dem-
onstration were a Detroit Diesel 6L71-T turbocharged 2-stroke cycle
engine and a Cummins NTC-290 turbocharged 4-stroke cycle engine. Both
engines had comparatively low smoke characteristics and were somewhat
more representative of engines used for intercity hauling than urban or
suburban service. All the fuels used were special blends, not very sim-
ilar to pump fuels which would be available for commercial use. They
included a No. 1 kerosene-type fuel (with and without a primary hexyl
nitrate "cetane improver"); a midrange fuel, considering aromatic content,
sulfur, and density as criteria (with and without an organo-metallic smoke
suppressant); and a No. 2 diesel smoke test fuel (also used with and without
the smoke suppressant).
-------
II. ENGINES, FUELS, AND ADDITIVES
The engines used to generate exhaust samples for this study
were a Detroit Diesel 6L-71T and a Cummins NTC-290, both chosen
by the Project Officer. A summary of the pertinent design and operating
data for each engine is given in Table 1, and while a number of differences
exist, the most important difference for the particulate work is probably
in the operating cycle (2-stroke vs 4-stroke). This characteristic appears
TABLE 1. ENGINE DESIGN AND OPERATING DATA
Data item
Engine
Detroit Diesel
Cummins
NTC-290
Cylinders and arrangement
Displacement, liters (in )
Operating cycle
Induction system
Maximum boost pressure, kPa (in
Combustion chamber design
Fuel system
Maximum power, kw (hp)
Rated rpm
Maximum torque, N-m (ft Ibf)
Peak torque speed, rpm
Manufacturer's "intermediate"
speed for certification, rpm
1-6
6. 98 (426)
2-stroke
Turbocharger feeding
scavenging blower
Hg) 67.7(20)
Open
Unit injectors
198 (265)
2100
1044 (770)
1600
1600
1-6
14.0 (855)
4-stroke
Turbocharger
47.4 (14)
Open
High-pressure pump
216 (290)
2100
1135 (837)
1500
1500
to have had a strong influence on the character of the particulate emitted,
and it probably has an influence on the mass rate of particulate emitted as
well. Photographs of the test engines are shown in Figures 1 and 2 (De-
troit Diesel and Cummins, respectively). Both engines were operated in
a test cell adjacent to the particulate tunnel, and the Detroit Diesel was
used for all the initial setup and calibration of the sampling systems as
well as for the first test (methodology demonstration) engine. The Cummins
engine was used as a test engine only, once the necessary recalibrations
had been performed, so it was not on the stand very long as compared to the
Detroit Diesel engine.
These two engines were both "low smoke" types, perhaps somewhat
representative of newer engines used in intercity truck service. Engines
used in the urban/suburban situation tend to be smaller than the test engines,
and are less likely to be turbocharged. Engines like the two test units may
not contribute greatly to urban particulate; and it is difficult to say at this
point whether smaller, non-turbocharged diesels emit similar particulate
at similar rates.
-------
FIGURE 1. DETROIT DIESEL 6L-71T TEST ENGINE
FIGURE 2. CUMMINS NTC-290 TEST ENGINE
-------
Quite a bit of time was spent during the initial project stages
in choosing the fuels and additives to be used. The major criteria
were to have ranges of sulfur levels, densities, and percentages
of aromatic hydrocarbons, among the fuels. Fuel B was selected to
represent No. 2 diesel smoke emissions test fuel as specified in EPA
regulations'^', and is considered a "worst case" fuel (high sulfur,
high aromatics, high density). Fuel A is considered to be on the other
end of the spectrum, with low density, sulfur, and aromatics. It is a
No. 1 kerosene or DF-1 type fuel, and was expected to yield the lowest
smoke.
Fuel C was selected to be between fuels A and B in density,
sulfur, and aromatics. It has been referred to as a "No. 1 1/2" fuel,
and due to blending it turned out to have a peculiarly truncated boiling
range. Specifications for all three test fuels are presented in Table 2,
and the boiling ranges are illustrated by Figure 3. Note that the simu-
lated distillation by gas chromatograph (D2887) presents a more realistic
and more accurate picture of fuel boiling range than straight distillation
(D86). Figure 3 also shows the near-total absence of light ends in fuel
C, and the general trend toward heavier constituents from fuel A to
fuel C to fuel B.
Using the data generated by ASTM D2887, it appears that fuel A
spans hydrocarbons from about €9 to C\f (median around Cj^). Fuel C
appears to include materials from about Cll to C21 (median around Ci3),
and fuel B contains hydrocarbons from about Cg to C26 (median around
0^5). Distillation by D86 and simulated distillation by D2887 were con-
ducted on all three of the "fuel + additive" combinations also, but the
results were so nearly identical to those for the fuels alone that it is not
worthwhile to discuss them here.
The test fuels do not necessarily represent fuels used in normal
vehicle operation, and they were not intended to do so. It is useful, how-
ever, to compare them to some more typical diesel fuels, because fuel
composition does appear to have a significant effect on particulate emis-
sions. A brief comparison is given in Table 3, with data on the "typical"
fuels coming from the U.S. Bureau of Mines'"). Fuel A is shown to be
somewhat lighter than a "typical" No. 1 fuel, and to have a much lower
sulfur content. Fuels B and C bracket the "typical" No. 2 fuel in terms
of boiling range, density, and sulfur content.
The additives used for methodology evaluation during this project
were Ethyl DII-2 and Lubrizol 8005. The Ethyl material is a primary
hexyl nitrate' (organic) material intended for use as an ignition accelerator
or "cetane improver". Lubrizol 8005 is an organo-metallic used as a
smoke suppressant, containing calcium and a small amount of barium.
Typical properties of these two materials are given in Table 4, verifying
that they are physically as well as chemically dissimilar.
-------
TABLE 2. FUELS USED FOR DIESEL PARTICULATE STUDIES
Project fuel code
SwRI fuel code
Fuel type
Distillation, °C, (°F)
IBP
10%
20%
30%
40%
50%
60%
70%
80%
90%
EP
% Recovery
% Residue
% Loss
Aromatics, %
Olefins, %
Saturates, %
Gravity, g/ml (°AI
Cetane (calculated)
Total sulfur, wt. %
Weight %C
Weight %H
Weight %N
Viscosity, cs
A
EM-197-F
No. 1 Kerosene
B
EM-195 & 204-F
2D Emissions
C
EM-198-F
"No. 1 1/2"
Boiling range by ASTM method No.
D86a
166 (330)
179 (354)
182 (360)
186 (366)
191 (376)
197 (387)
206 (420)
214 (418)
225 (437)
238 (460)
274 (525)
99
1
0
D2887b
148 (298)
163 (326)
173 (344)
181 (357)
195 (383)
203 (397)
217 (422)
229 (445)
238 (460)
253 (488)
301 (574)
— — _ —
9.2
0.8
90.0
•I) 0.794 (46.8
51.0
0.003
85.0
13.9
0.08
1.62
Flashpoint, °C (°F) 54 (130)
D86
196 (384)
223 (434)
238 (460)
251 (483)
260 (500)
270 (518)
277 (531)
287 (548)
298 (569)
316 (601)
356 (673)
98
2
0
D2887
161 (321)
198 (388)
217 (422)
240 (464)
257 (494)
271 (519)
284 (544)
301 (573)
316 (601)
338 (640)
417 (782)
_ — — _
35. 1
0.0
64.9
0.850 (34.9)
49.5
0.319
86.5
12.8
0. 10
2.7
80 (176)
D86
210 (410)
219 (426)
227 (441)
230 (446)
232 (450)
235 (455)
238 (460)
241 (465)
248 (479)
261 (502)
312 (594)
99
1
0
D2887
191 (376)
218 (425)
228 (442)
233 (452)
237 (458)
239 (463)
248 (479)
254 (490)
261 (502)
274 (525)
356 (672)
_
23.0
1.0
76.0
0.828 (39.4)
49.5
0.010
85. 3
13.9
0.04
2. 12
85 + (185+)
1 thermal distillation, % by volume
gas chromatograph simulated
distillation, % by weight
-------
100
90
80
70
60
50
40
30
10
0 L
Normal paraffin boiling points by carbon number
9 10 11 12 13 14 15 16 18 20 22 24 26
Fuel A
o ASTM D86 (distillation) volume %
A ASTM D2887 (chromatograph)
weight %
150
i
300
200
400
250
300
I I
350
400° C
500 600
Temperature
700°F
FIGURE 3. BOILING RANGES OF FUELS USED IN METHODOLOGY DEMONSTRATION
7
-------
TABLE 3. COMPARISON OF TEST FUELS WITH "TYPICAL"a DIESEL FUELS
Fuel property
Distillation, °C (°F)
IBP
10%
50%
90%
EP
Gravity, g/ml
Gravity, °API
Total sulfur, wt. %
Method
D86
D287
D129
Data by fuel code or type
A
166 (330)
179 (354)
197 (387)
238 (460)
274 (525)
0.794
46.8
0.003
B
196 (384)
223 (434)
270 (518)
316 (601)
356 (673)
0.850
34.9
0.319
C
210 (410)
219 (426)
235 (455)
261 (502)
312 (594)
0.828
39.4
0.010
"Typical"
No. 1
177 (351)
199 (391)
228 (443)
263 (505)
284 (544)
0.818
41.4
0.096
"Typical"
No. 2
189 (373)
219 (426)
257 (495)
302 (575)
327 (620)
0.843
36.4
0. 228
aThese values are averages of regional data in the 1973 Bureau of Mines Diesel
Fuel Survey and are not sales-weighted
TABLE 4. TYPICAL PROPERTIES OF ADDITIVES
USED FOR METHODOLOGY EVALUATION
Lubrizol 8005
Property
Density, g/ml @ 16°C
Viscosity, cP @ 38° C
Flashpoint, °C (°F)
Wt. percent Ca
Wt. percent Ba
Wt. percent S
Wt. percent sulfated ash
Value
1. 14
71
>93(>200)
11.4
2.22
1.0
42.2
Ethyl DII-2
Property
Density, g/ml @ 20°C
Viscosity, cP @ 38°C
Flashpoint, °C (°F)
Value
0.97
0.95
69 (156)
-------
Additive treatment levels used were those indicated to be re-
presentative of field usage hy the additive manufacturers, namely 0. 1%
by volume for DII-2 and 0. 25% by volume for 8005. Composition and
coding of the fuels is summarized in Table 5 for reference.
TABLE 5. CODING AND COMPOSITION OF FUELS
AND FUEL-ADDITIVE MIXTURES
Fuel Name
A
B
C
A + additive
B + additive
C + Additive
Fuel Type
No. 1 kerosene
2D Emissions test
"No. 1 1/2"
}As above
but with
additives
SwRI code
EM-197-F
EM-195-Fa
EM-198-F
EM-209-F
EM-210-Fb
EM-211-F
Additive & Vol. %
None
None
None
DII-2 at 0. 1
8005 at 0. 25
8005 at 0. 25
aEM-204-F after 8/22/74 (new batch)
bEM-224-F after 8/22/74 (new batch)
Lubrizol 8005 was added to the two heavier fuels because it was assumed
that they were most likely to produce high smoke levels. Although the
cetane number of fuel A was adequate for satisfactory performance with
either test engine, the DII-2 ignition accelerator was added to ascertain
its effects, if any, on particulate mass and composition.
-------
IE. INSTRUMENTATION AND TEST EQUIPMENT
This section is included primarily as descriptive material on
the hardware used to accomplish the necessary measurements, and
it should provide adequate background for the technical discussions
to follow. Both engines were operated on a conventional test stand
which employed a 373kw (500 hp) capacity eddy-current dynamometer
and a large inertia wheel. This stand did not have motoring capability
(of little use with diesels), but the inertia wheel could be selectively
coupled or uncoupled depending on procedural requirements.
Engine fuel flow was measured using a bridge-type mass flow-
meter manufactured by Flo-Tron, Inc. , which is a -widely-used indus-
trial instrument. Engine air flow was measured using a sharp-edged
orifice installed downstream of an MSA Ultra-Aire filter system, and
the flanges holding this orifice are visible on the vertical black duct
at the left in Figure 4. Figure 4 also shows the whole dilution tunnel,
which used another MSA filter for dilution air cleanup. The engine
being operated for particulate collection is on the other side of the wall
at left in Figure 4, and the instrument under the center of the tunnel is
the chemiluminescent analyzer used for computing actual exhaust dilution.
The muffler used on these engines is shown in detail in Figure 5,
with special modifications to the upper end to provide three separate
exhaust outlets. The two larger outlets (with valves) permitted most of
the engine's exhaust to escape to the atmosphere, but the portion in the
small pipe went directly into the tunnel. To prevent an overly large
amount of exhaust from going into the tunnel, restrictor tubes were con-
structed as shown in Figure 6. These tubes were inserted into the exist-
ing 3-inch muffler outlet on a trial-and-error basis, and they were built
with the same inside length and perforation pattern as the existing tube.
This technique provided more restriction between muffler and tunnel than
the existing tube, but presented a passage for the exhaust which was as
normal in configuration as possible.
The dilution tunnel itself was flow-calibrated in much the same
way as one would calibrate a constant-volume sampler (CVS). It employs
a positive-displacement rotary vane gas pump, and the calibration per-
mitted flowrate (and thus isokinetic velocity) calculations to be made utili-
zing blower rotations and elapsed time. The counter and timer are shown
in Figure 7 along with gauges used to measure pressure at the sampling
section and across the blower. Figure 8 is a schematic section of the
dilution tunnel with its important dimensions specified. This tunnel (the
first one built by the contractor) was made of 3. 2mm (1/8 inch) stainless
steel sheet rolled into duct form and welded along a seam at the bottom.
All tunnel connections were made with flanges and gaskets to insure against
leaks, even though the vacuum in the tunnel was very slight (approximately
11
-------
FIGURE 4. OVERALL VIEW OF DIESEL
PARTICULATE DILUTION TUNNEL
FIGURE 6. DETAIL OF PERFORATED
EXHAUST RESTRICTORS USED TO LIMIT
EXHAUST FLOW INTO TUNNEL
FIGURE 5. DETAIL OF EXHAUST
MUFFLER USED FOR TESTING
12
-------
6lOmm( 24in)
t
61 Omm
(Z4in) 450mm
I
76mm (Sin) raw
exhaust transfer tube
dilution air
filter enclosure
hi-vol —
sample probe
230mm ( 9in)
mixing orifice
•700mm ( 27. Sin)
127mm
( Sin) dia
FIGURE 8. SCHEMATIC SECTION OF DILUTION TUNNEL FOR DIESEL PARTICULATE SAMPLING
-------
0. 5 kPa or 2 inches
Three sampling systems were used (one at a time) with the
dilution tunnel for different types of filters. The "47mm" system is
shown in Figures 9 and 10, and it was installed (as were all the others)
with probe pointing upstream. Each of the four separate probes led
to a 47mm filter case, three holding glass fiber filters and one holding
a Fluoropore* filter. Flows through these systems were set and main-
tained using the flowmeters (with metering valves) shown in Figure 11,
and flows were totalized using the dry gas meters also shown in Figure
11. Each sampling system was connected to a separate pump to prevent
unwanted interactions and permit stopping sample acquisition at will to
replace filters. The reducers inserted in the 25mm (1 inch) tubes shown
in Figure 9 were necessary for isokinetic sampling through the small
filters.
A modified "hi-vol" sampling system was used to collect samples
on 203mm by 254mm (8 by 10 inch) rectangular glass fiber filters as
shown in Figures 12 and 13. The sampling nozzle (Figure 12) was inside
the tunnel down to the flange when operating, and the upper half of the
filter holder plus the whole unit shown in Figure 13 extended downward
from the tunnel. The "tailpipe" added to the hi-vol blower contained a
straightening section and a sharp-edged orifice to provide for computation
and monitoring of sample flow.
It became necessary during the test sequence to take some samples
on large Fluoropore filters (293mm diameter), requiring a modification
of the hi-vol system due to the high pressure drop of these filters. Although
it is somewhat difficult to see in Figure 14, the filter holder was modified
to include another gasket and a spacer which restricted the effective fil-
tration area slightly. The "tailpipe" was sealed to prevent in-leakage, and
the Roots gas pump shown in Figure 15 was connected to its lower end by
tubing and mounted outside the building. Other items added were the by-
pass system and vacuum guage shown in Figure 16, permitting accurate
orifice flow calculations and vacuum relief before the filter plugged. This
system could have been used with glass fiber filters as well, but the regu-
lar hi-vol system was less complicated to operate and thus was used where-
ever possible.
The final particulate sampling system to be discussed withdrew
samples not from the dilution tunnel, but from the exhaust pipe upstream
of the muffler. This system was a portable sampler-diluter developed for
the Environmental Protection Agency by Environmental Research Corporation
under Contract No. 68-02-0589. It will be referred to from this point on as
the "ERC sampler" for brevity. The unit is shown in Figure 17 along with
the sample pump and flow measuring devices necessary for its operation,
^Trademark of the Millipore Corporation
14
-------
FIGURE 7. TUNNEL, BLOWER
COUNTER AND TIMER
FIGURE 9. PROBES USED WITH
47mm SAMPLING SYSTEMS
FIGURE 10. 47mm FILTER HOLDERS
FIGURE 11. FLOWMETERS AND
TOTALIZING METERS USED WITH
47mm FILTERS
15
-------
FIGURE 12. SAMPLING NOZZLE
USED WITH HI-VOL SYSTEM
FIGURE 13. BLOWER AND FLOW
MEASUREMENT APPARATUS USED
WITH HI-VOL SYSTEM
FIGURE 14. MODIFIED HI-VOL
SYSTEM USED WITH 293mm
FLUOROPORE FILTERS
FIGURE 15. GAS PUMP USED WITH
293mm FLUOROPORE FILTERS
16
-------
FIGURE 16. FLOW MEASURE-
MENT SYSTEM USED WITH
293mm FLUOROPORE FILTERS
FIGURE 17. ERG SAMPLER-DILUTER
FIGURE 18. PHS SMOKEMETER
AND BOSCH SAMPLING UNIT
BEING EMPLOYED FOR
DIESEL SMOKE MEASUREMENT
FIGURE 19. HUMIDITY - AND
TEMPERATURE-CONTROLLED
CHAMBER HOUSING MICRO-
BALANCE FOR FILTER WEIGHING
17
-------
but it is shown not connected to the exhaust pipe at its normal sample
point. The fitting mounted at the elbow leading into the muffler held
the probe for the ERG sampler when in operation, and the sample line
was heated from the sampling point back to the sampler.
In order to obtain some smoke density measurements against
which particulate emissions data could be compared, both the Federal
(PHS) smokemeter and the Bosch sampling smokemeter were employed.
These systems are shown being used in Figure 18, with the optical unit
of the PHS meter mounted on the end of the stack and the Bosch probe
inserted into the stack. Thg control and readout unit of the PHS meter
was located at the engine operator's position, and the filters through
which exhaust gas was sampled for the Bosch unit were later measured
for reflectance.
The final major item of equipment necessary to this project was
the humidity— and temperature — controlled weighing chamber and
microbalance shown in Figure 19. It was made of plexiglass and equip-
ped -with a flow-through ventilation system providing filtered air con-
trolled to ±0. 6°C and ±2% relative humidity. The air system includes
a chilled water spray chamber for absolute humidity control, followed
by controlled electrical reheat and an MSA Ultra-Aire filter. The box
is maintained at a positive pressure of 37 Pa (0. 15 inch H2O) by a blower
and an outlet orifice, and total flow through the box is about 0. 3 m^/min
(10 ft min). The nominal conditions to which air in the box is controlled
are 72°F dry bulb and 39 percent relative humidity (46 grains H2O/lbm
dry air).
18
-------
IV. PROCEDURES FOR ANALYSIS OF PARTICULATE COMPOSITION
Once the particulate matter had been collected and weighed, it
was sent to one of several other laboratories for further study. Table
6 summarizes the analysis activities, with three of the five other
TABLE 6. SUMMARY OF ANALYSES FOR PARTICULATE COMPOSITION
Sample form
Particulate on
47mm glass
fiber filters
Particulate on
47mm Fluoro-
pore filters
Particulate on
293mm Fluoro-
pore filters
Particulate on
8 x 10 glass
fiber filters
Organic solubles
Laboratory
Galbraitha
S. E. A. L.
SwRI Dept. lc
SFRE<*
Galbraitha
SwRI Dept.
10e
Analysis for
Carbon, hydro-
gen, nitrogen,
sulfur
Metals (Ca, V,
Mn, Ni, Cu, Zn,
Pb, Sr, Sn, Ba)
Phenols, nitro-
samines
Total organic
solubles; NRM,
IR, and BaP in
organic solubles
Carbon, hydro-
gen, nitrogen,
fulfur and oxygen
Boiling point dis-
tribution of paraf-
fins in organic
solubles
Method
Combustion
X-ray
fluorescence
Extraction
and G. C.
Soxhlet
extraction
plus spec-
tral analysis
or TLC and
fluorescence
spectro-
photometry
Combustion
Fraction-
action and
GC analysis
Output
Data only
Data only
Data only
Data and
remaining
organic
solubles
Data only
Data only
aGalbraith Laboratories, Inc., Knoxville
Scanning Electron Analysis Laboratories, Inc. , Los Angeles
GDepartment of Chemistry and Chemical Engineering
"Southwest Foundation for Research and Education
eU. S. Army Fuels and Lubricants Research Laboratory
19
-------
laboratories being located within the Southwest Research Center complex.
Most of the analyses desired under the subject contract were performed
with good results, but no success was achieved in developing a method for
analysis of organic peroxides. It was felt that effort needed on procedural
development of the phenol and nitrosamine analyses would be compromised
if the required effort were to be expended on peroxides, so attempts at
developing a peroxide technique were dropped early in the program. In
addition, the time required for Soxhlet extractions, BaP analysis, and
subsequent NMR and IR spectra and sample drying created quite a delay
in making samples of organic solubles available for CHNSO and paraffin
boiling point analyses.
A. Elemental Analysis by Commercial Laboratory
\
Determination of carbon, hydrogen, nitrogen, and sulfur weight
percentages in diesel particulate and organic solubles, and of oxygen weight
percentages in organic solubles only was performed by Galbraith Labora-
tories. Their results were determined by gas analysis following combus-
tion of whole 47mm glass fiber filters (or samples of organic solubles),
so the percentages of each compound which constituted final data depended
on the total particulate weights supplied to them with the samples. The
results were corrected for blank filter content, which was reported to be
very low. Galbraith had to split the samples of organic solubles themselves,
and sometimes there was not a sufficient amount for oxygen analysis. They
were usually sent 3 filters for each particulate sample; one for CHN analysis
one for S analysis, and a third to keep should a repeat run be necessary.
B. Analysis for Metals by Commercial i Lab oratory
Analysis for ten metals (listed in Table 6) was conducted using
X-ray fluorescence analysis by Scanning Electron Analysis Laboratories,
Inc. They reported their results injL/g/cm on the filter using samples
on 47mm Fluoropore filters. This particular filter type was chosen due
to its relatively low background metal content and retention properties
similar to glass fiber filters (these Fluoropores had 0. 5jUm mean flow pore
size). Knowing the effective filtration area of the filters and the total parti-
culate weight, the data from S. E.A. L. were adequate for computation ofjllg
metal per filter and fraction of total particulate appearing as each metal.
Filter background levels and minimum detection limits were determined
separately for each batch of filters analyzed. The portion of each filter
analyzed was its center, so concentrations of some materials reported may
be somewhat high due to particle dynamic effects.
To illustrate the precision of analysis for metals, Table A-l of
Appendix A lists the minimum detectable limit for each element in each
batch of filters submitted. These limits were calculated by the formula
20
-------
M.D.L. (element "Z") = sVbackground counts under peak/(peak area counts)
x actual concentration of "Z" .
where the actual concentration is obtained from a standard (known) sample.
Variation in M.D. L. occurred due to detector aging, instrument anomalies,
and so forth. A typical graphical representation of the analyzer output
for a sample standard (lOjL/g each Mn, Ni, Cu, Zn, Pb, Sr) is presented
in Figure A-l, and Figure A-2 shows a similar output for a typical parti-
culate sample.
C. Analysis for Phenols and Nitrosamiijes
The SwRI Department of Chemistry and Chemical Engineering worked
under this contract to develop analysis methods for phenols, nitrosamines,
and organic peroxides in diesel particulate. The nitrosamine technique is
capable of detecting 0. IjL/g or less of dimethylnitrosamine in a sample col-
lected on an 8 x 10 inch glass fiber filter, and approximately 1/^g of any
of 9 phenols in a sample collected on a 293mm Fluoropore filter. Inter-
ference of phenols present on the filter media has been a continuing problem
with that analysis. In brief, the phenol and nitrosamine procedures involve a
a series of extractions followed by gas chromatograph analysis, and both pro-
cedures are given in full on Appendix pages A-5 through A-7. A calibration
of the nitrosamine system is shown in Figure A-3, and a typical sample out-
put is shown in Figure A-4. An early calibration of the phenol system is
given in Figure A-5, followed by a reagent blank (Fig. A-6), an unused glass
fiber filter (Fig. A-7), and a particulate sample on a glass fiber filter (Fig.
A-8). A later standard run is shown in Figure A-9, and two blank Fluoro-
pore filter runs in Figures A-10 and A-11. A particulate sample on a
Fluoropore filter is given in Figure A-12. Fluoropore filters were used for
all record runs.
D. NMR and IR Spectra, and Benzo (00 Pyrene Analysis of
Organic Solubles
Southwest Foundation for Research and Education (SFRE) was engaged
to assist in analysis by extracting organic solubles from particulate samples
on 8 x 10 glass fiber filters, running NMR and IR spectra on the solubles,
and finally using the method of Sawicki, et aP ', to determine Benzo (aQ
pyrene content of the solubles. Soxhlet extraction was used on most of the
samples, with benzene, hexane, and methylene chloride all being used as
solvents with approximately equivalent results. Ultrasonic extraction
was used on some samples for comparison, but its accuracy was not
considered superior to Soxhlet, so it did not become the standard
technique. A report from SFRE on the BaP analysis, its accuracy and
its problems is included in Appendix A, pages A-18 through A-36 .
21
-------
To perform the NMR and IR absorptions, a portion of the solubles
was re-dissolved in carbon tetrachloride. Analysis by UV absorption
was also attempted early in the program, but the results showed an
undefinable hydrocarbon "lump" which made it impractical to pursue UV
any further. The IR scan covered wavelengths from 2. 5 x 10 m to
16 x 10~°m. The NMR spectrum ranged from about 0 to 8 ppm (or 0 to 8 Hz
per MHz), with an oxcillator frequency of 60 MHz. A typical IR analysis
output is given in Appendix A, Figure A-13, and a similar example of an
NMR output is given in Figure A-14.
E. Boiling Point Distribution of Paraffins in Organic Solubles
The U. S. Army Fuels and Lubricants Research Laboratory at
SwRI agreed to perform analysis required for boiling point distribution
of paraffins in the organic soluble portion of diesel exhaust particulate.
Saturate fraction isolation on an activated silica gel column was per-
formed according to ASTM D1319-70, and gas chromatograph analysis
of boiling point distribution was performed on organic solubles (both whole
sample of solubles and isolated paraffins) according to a modified version
of ASTM D2887-73. Fuel samples were subjected to the same analysis
procedure. A description of the "Procedure for Saturates in Fuels" is
included in Appendix A, pp. A-39 through A-44.
The fuel and all samples and part of the particulate (solubles)
samples were chroma tog raphed on a Dexsil 300 column,which is one of the
allowable columns for ASTM D2887-73. The remainder of the solubles
were chromatographed on SE-30 to improve baseline stability and thus
enhance the accuracy of the derived boiling point distributions. A cali-
bration of the Dexsil column is shown in Appendix A, Figure A-15. A
calibration of the SE-30 column is shown in Figure A-16, a fuel analysis in
in Figure A-17, an oil analysis in Figure A-18, and a particulate (solubles)
analysis in Figure A-19.
-------
V. CONSTRUCTION AND CALIBRATION
OF DILUTION TUNNEL AND SAMPLING DEVICES
In constructing a particulate dilution tunnel for use with large
diesel engines, several design conflicts became apparent in transferring
the technology and hardware formerly used with light-duty gasoline ve-
hicles. The tunnel as specified by EPA was originally designed and
sized for vehicles which produce a maximum of perhaps 3m (106 ft )
of exhaust gas per minute on the LA-4 cycle. The large diesels used
for this project produced much higher exhaust flows, the maxima being
some 18m3/min (650 ft3/min) for the Cummins NTC-290 and about 27m3/
min (950 ft3/min) for the Detroit Diesel 6L-71T. These maximum exhaust
flows were far in excess of total tunnel capacity with no dilution at all, so
it became necessary to "split" the exhaust flow and admit only a fraction
into the tunnel.
The separation of a fraction of the exhaust to enter the tunnel was
accomplished by sizing and restricting exhaust outlets from the muffler,
as already described in Section III. Using a fraction of the exhaust in the
tunnel introduces another complication, however, because that fraction
must be measured without "handling" it (such as with an orifice or laminar
flow element). The problem can be resolved by calculating exhaust flow
into the tunnel subtractively as follows:
flows into tunnel: dilution air (A) and exhaust (B)
flows out of tunnel: sample (C) and (discarded) dilute exhaust (D)
material balance: A + B = C + D
therefore: B = C + D - A
Flow C is known by calibration of sampling systems, flow D is known by
calibration of the tunnel's primary blower, and Flow A can be measured by
an orifice or some other device. Thus flow B (exhaust) can be determined
subtractively, and the engine's total particulate emission rate can be com-
puted by the relationship
total particulate rate -(sampled particulate rate) (^^) (^^ ffi^A ****
Another method for determining total particulate rate (which is the
desired end result rather than flow of exhaust into the tunnel) is measure-
ment of both raw and dilute concentrations of a "tracer" gas along with
quantities of sample and (discarded) dilute exhaust. For the tunnel used in
this project, the "tracer" gas used was NOX» since it is present in sub-
stantial concentrations and remains stable over at least the interval neces-
sary for measurements (approximately 5 seconds). The derivation of this
technique is as follows:
23
-------
RNOX = ppm NOX in raw exhaust = moles NOX/10° moles raw exhaust
DNOX = ppm NOX in dilute exhaust = moles NOX/10^ moles (raw exh. + dil. air)
material balance: moles NOX = constant
therefore: moles raw exhaust = (DNOX/RNOX) (moles raw exh. + dil. air)
= (DNOX/RNOX) (C + D)
and C and D are known quantities.
Total particulate emissions from the engine are computed by
total part, rate = (sampled part, rate) (^P) (gggX) ^engine e^rate ^ _
The latter technique was used during this project, but the former is theo-
retically equivalent and would have the advantage of not requiring a dedi-
cated chemiluminescent NOX analyzer.
The positive displacement blower used to pull dilution air and exhaust
through the dilution tunnel system was calibrated using a large laminar flow
element and a bank of electric heaters. The result is shown on page B-2
of Appendix B, with supporting data and calculations on pages B-3 and B-4.
It was necessary to extrapolate the line back to a blower Ap of 0. 87kPa (3. 5
inches H2O). The higher blower speed (36:26 drive ratio) was used to permit
higher tunnel velocities and consequent higher sample acquisition rates.
Calibration with the 36:26 drive ratio and 0. 87kPa (3. 5 inches E^O) blower
Ap yielded a tunnel bulk velocity at the sampling station
v _ ,yj blower revolutions (counts) I
vsb ~ 1N time, sec
X.T8 v PB
where: N = 11. 9 for the 4-probe system using 47mm filters
N = 12.4 for the hi-vol system
subscript b indicated "bulk"
subscript s indicated sampling station
subscript B indicates blower inlet.
In order to withdraw an isokinetic sample, however, the tunnel
centerline velocity must be used rather than the bulk velocity. The tunnel's
velocity profile at the sampling station provides the information required to
calculate the centerline velocity from the bulk velocity, in addition to pro-
viding data on velocity variation in the sampling zone. Velocity profiles
(horizontal and vertical) at the sampling station were acquired with a Thermo
Systems hot-film anemometer. The results were less precise than antici-
pated, due primarily to the influence of large-scale turbulence in the tunnel.
Averages over a number of runs gave usable values, however, and the plots
24
-------
shown on page B-5 were constructed from these data. Data and cal-
culations as well as the velocities in the profiles are referred to the
duct centerline, and although the "flat" sections of the profiles are
not quite normal to the tunnel axis, the deviation from the centerline
velocity (V> ) is only about ±2% in the sampling zone. The reason for
the higher velocities in the upper right section of the tunnel is probably
the overhead lighting which strikes the tunnel from that direction. The
±2% velocity gradient in the sampling zone is probably much less a cause
of anisokinetic sampling than the large scale turbulence mentioned
earlier. Temperature and gas concentration profiles were also taken
(vertical only), and the results are shown on pages B-9 and B-10.
Measurement of flow through the hi-vol sampling system using
8 inch by 10 inch glass fiber filters was performed using an orifice mounted
about 0. 79m (31 in. ) downstream of the sampling blower outlet in a 76mm
(3 in) O. D. "tailpipe". The orifice equation was determined by the ASME
flowmeter handbook* ' procedure as
mass flow = 4. 65 (APp)°- 5 Ibm/min = 2. 11 (App)0' 5 kg/min,
where Ap is in inches of H2O and p is in Ibj-n/ft. Flow through the hi-vol
unit was set according to the relationship
Apor (isokinetic sample) =
where the subscript "or" denotes the orifice station downstream of the hi-
vol blower.
Flow through the 4-probe system -was set by calibrated rotameters
and measured (totalized) by calibrated dry gas meters. Using the cali-
brations, flow through this system was set according to the relationship
m = (x) Vsb ,
where x was 0. 00151 for two of the systems, and 0. 00152 for the others.
(The probe inlet area of both the hi-vol system and the 4-probe system
had to be reduced from their original sizes (as specified for testing of
gasoline engines) to prevent inordinately high filter face velocities and
pressure drops. Even with these area reductions (and their consequent
flow reductions), filters often plugged before the desired sampling time had
elapsed. Calibration curves and calculations on these two sampling systems
(hi-vol and 4-probe) are given on pages B-ll through B-13.
When the filter samples for phenol analysis were taken, Fluoropore
filters had to be used to minimize background problems. These filters (293 mm
diameter) had a much higher pressure drop per unit volume flowrate than
glass fiber filters, so the standard hi-vol blower could no longer be used.
25
-------
In its place, an.other system was constructed using an orifice and a
small Roots pump to measure and withdraw sample. This system used
the same orifice equation as the hi-vol system, but the operating point
had to be changed as the filter plugged because the orifice was under
an increasing vacuum as the run progressed. To maintain the isokineticity
of the sampling as long as possible, the curve shown on page B-14 of Appen-
dix B was used by the sampler operator to regulate a bypass (inlet) down-
stream of the orifice. Since the Fluoropore filter did plug rather rapidly,
sampling was usually conducted until the sample rate fell to about 0. 6 times
the isokinetic rate. Simplified operating guides for the tunnel and samples
are given on page B-15.
The orifice used to measure engine air flow was calibrated against
a laminar flow element having calibration traceable to NBS standards. The
final air flow equation (derived by application of the least squares method
to logs of Ap and mass flow) is
Ma = 40.64 (APp)°-4842 kg/min = 89.60 (App)0'4842 lbm/min,
and the applicable calibration data and curve are given on pages B-29 and
B-30. Exhaust mass flow is air flow plus fuel flow (which was measured
by a flow bridge-type instrument).
The final particulate sampling system to be discussed was called
the "ERG sampler, " and it was built by the Environmental Research
Corporation for EPA under an earlier contract. This sampler was inde-
pendent of the dilution tunnel, and withdrew its sample from the engine's
exhaust pipe upstream of the muffler. After the unit was made operable
by a number of minor modifications and repairs, it was decided that the
instructions supplied with it were indecipherable. The sampler's princi-
ples of operation were reviewed, and a more usable set of instructions
was devised with considerable effort. These instructions appear as pages
B-16 through B-23 of Appendix B, and the calculations and considerations
leading to the instructions are given as pages B-24 through B-28.
The ERG sampler drew both exhaust and dilution air into a central
chamber, using a vacuum blower as the gas-handling device. Exhaust
was drawn through a heated sample line containing a venturi for flow
measurement. This exhaust was allowed to impinge upon a (theoretically)
equal flow of air in the center of the dilution chamber, and additional dilu-
tion air was added through the wall of the chamber (it was made of fine
mesh). Provision was made for sampling from the total dilute flow at
either 1 CFM (0.028 m3/min) or 5 CFM (0.142 m3/min), using a separate
filter holder, sample pump, flowmeter, and dry gas meter. The 1 CFM
flowrate was used for this project, and the remainder of the (nominal) 50
CFM (1.42 m3/min) dilute exhaust flow was discarded.
26
-------
VI. PLANS FOR ACQUISITION AND REDUCTION OF DATA
Since the total number of samples taken in the methodology dem-
onstration portion of this project was extremely large, a test plan was
devised to meet project objectives while avoiding unnecessary duplication.
This plan provided ample information on which types of analysis are
most useful for fuel and additive qualification. The minimum plan is
summarized in Table 7, and it was followed more closely for the Cum-
mins NTC-290 engine than for the Detroit Diesel-Allison 6L-71T engine
(the test plan was still being finalized while the 6L-71T engine was under
test). A number of duplicate and supplementary runs was made on each
engine as required to fully document the test results. The actual number
of independent analytical determinations made on each engine is summarized
in Table 8 along with the number which would have resulted had the test
plan followed strictly.
Procedures for data reduction were really the final technical de-
velopments necessary to calculate engine total particulate output from
data obtained during the course of a test, based on calibrations and com-
putations already presented. Data acquired during a test are perhaps
best illustrated by the data forms actually filled out during a test, so
examples of the three types of forms are given as pages C-2 through C-4
of Appendix C. For a given test only half of each data form would be
completed (top or bottom).
Mathematical development of data reduction procedures is given
as pages C-5 and C-6, based largely on the results of calibrations and
calculations discussed in Section V. Although the calculations are com-
pact enough to be performed by hand for a few cases, the large number
of samples taken for this project made computer processing more eco-
nomical in the long term. Examples of the encoding sheets from which
data were keypunched (12 data cards per test) are given as pages C-7 and
C-8. The computer program used for processing is included as pages C-9
through C-12, and sample results are given on pages C-13 and C-14.
Developing procedures for steady-state runs (all those other than
the composites) was relatively simple, since most of the major data items
remained rather stable. Obtaining a sample on one filter which was a
true composite for the 13-mode test, however, required that the total
amount of raw exhaust gas filtered in each mode be proportional to the
product of engine exhaust mass flowrate and the time-based weighting
factor for that mode. In mathematical terms
27
-------
TABLE 7. MINIMUM TEST PLAN AND DATA MATRIX
' FOR EACH ENGINE AND EACH FUEL
Operating condition
Speed
Idle
Peak torque
Peak torque
Peak torque
Peak torque
Peak torque
Rated
Rated
Rated
Rated
Rated
Load, %
0
0
25
50
75
100
0
25
50
75
100
Composite
Analysis codesa by sampling system
Four 47mm filters
First run
1,2,5,8
1,2,5,8
l,2b, 5
1,2,5,8
l,2b, 5
1,2,5,8
1,2 5,8
i.
l,2b, 5
1,2,5,8
1,2^,5
1,2,5,8
1,2,5
Repeat
1,5
Hi-vol system
First run
1,3,4,6
1,3,4,6
1
1,3,4,6
1
1,3
1,3
1
1,3
1
1,3,4,6
1,3,4,6
Repeat
1,7
1,7
1,7
1,7
Analysis codes
1. gravimetric
2. metals
3. organic solubles, BaP, IR, NMR
, , ,
paraffins in organic solubles
Fuel "B" only
4.
5. SCHN in particulate
6. SCHNO in organic solubles
7. phenols and nitrosamines
8. ERG gravimetric
TABLE 8. PLANNED AND ACTUAL NUMBERS OF ANALYSES
CARRIED OUT ON EACH ENGINE
Number of analyses by engine
Type of analysis
gravimetric
metals
SCHN (particulate)
organic solubles,
(BaP, IR, NMR
paraffin boiling
range (solubles)
SCHNO (solubles)
phenols, nitrosamines
Planned (both)
408
54
78
48
30
30
24
D.D. 6L-71T
592
54
65
41
21a
13
20
Cum. NTC-290
524
58
72
54
8a
8a
21
Insufficient sample to perform the planned number of tests
28
-------
where: i = individual mode, i = 1, 2, 3, , 13;
mj = (m^) (time)i = total dilute exhaust filtered in mode i, 4 x 47
system, lbm;
(Mor)i = (Mor)j, (time^ = total dilute exhaust filtered in mode i,
hi-vol system, lbm;
i_, = time in mode i, sec;
)^ = engine exhaust flowrate in mode i, lbm/min;
W^ - time-based weighting factor;
E^ = exhaust flow through tunnel, lbm/min; and
D^ = diluent flow through tunnel, lbm/min.
Note that
so ^ _ ^ "M
Therefore, since both m^ and (MQr)^ are essentially fixed by isokinetic
considerations, it is sufficient to use only one of the (time)^ equations
above for computation purposes. The quantities ( — ^~) » ^ij ^ek'
• \ E / i
and (Mor). are known or can be calculated from experimental data. If
a value for any (time)^ is assumed, the constant C% can be calculated
and then the other (time)^ can also be calculated. To determine whether
or not our choice of G£ is reasonable, we can compute
(time) =
and choose a higher or lower value of G£ to make (time) more reasonable.
To minimize the complexity of the 13-mode test, it was decided
to determine the (E+D/E)^ with both the dump valves in a constant position
such that maximum exhaust back pressure existed only for the rated speed,
100 percent load condition. This decision means that we did not have the
absolute maximum particulate collection per unit time, but experimental
data showed that an adequate amount should be collected in a test of about
40 minutes' duration. It was also possible, of course, to determine (Me)j
and a good average value for MQr while measuring the dilution ratios.
Data and some calculations are given in Table 9 and on the lines just
below Table 9 for the Detroit Diesel 6L-71T engine. Note that the technique
29
-------
TABLE 9. .DATA AND CALCULATIONS USED TO DETERMINE
MODE TIMES FOR THE DETROIT DIESEL 6L-71T ENGINE
i =
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
wi
0.20-i- 3
0.08
0.08
0.08
0.08
0.08
0.20-r3
0.08
0.08
0.08
0.08
0.08
0.20-r 3
(E+D\
VTT^ m
/ ]_
32.2
22.6
19.8
20.4
19.4
18.1
31.5
12.9
14.4
15.8
16.3
17.6
27.1
•
(M )•
9.85
35.78
37.63
41.64
46.73
54.14
9.85
74.43
67.45
60.36
54.16
50.79
9.85
TOTAL = (time) =
Time in mode (min) by assumption
1
2.00
6.12
5.64
6.43
6.86
7.42
1.96
7.27
7.35
7.22
6.68
6.76
1.68
73.39
2
1.08
3.31
3.05
3.48
3.71
4.01
1.06
3.93
3.97
3.90
3.61
3.66
0.91
39.68
ASSUMPTION 1: (time)1 = 2.00 min .'. C2 = 0.1665
conclusion: (time) too long ,\ assume smaller
ASSUMPTION 2: C2 = 0.09
conclusion: (time) is OK, but combine 1, 7, and 13 to make one
(longer) idle mode so technicians will have adequate
time to gather data
converges rapidly on the desired value of (time). Similar information
for the Cummins NTC-290 engine is shown in Table 10, although the
schedule there has already been reduced to 11 modes as suggested fol-
lowing Table 9. The final mode times in seconds and cumulative seconds
are shown for both engines in Table 11. Although the constants are dif-
ferent for any two engines and dilution tunnels, a similar approach to mode
weighting by controlling sampling times should be valid for any system.
This schedule yields the desired result, that is, weighting of modes so
that a single filter is representative of a 13-mode test as that test is de-
fined. Runs were sometimes repeated with the same filter in place when
too small an amount of particulate was collected.
30
-------
TABLE 10. DATA AND CALCULATIONS USED TO DETERMINE
MODE TIMES FOR THE CUMMINS NTC-290 ENGINE
i =
Mode
1
2
3
4
5
6
7
8
9
10
11
Wi
0.08
0.08
0.08
0.08
0.08
0.20
0.08
0.08
0.08
0.08
0.08
/E+D\ a
22.2
22.5
21.9
20.0
19.9
25.3
12.9
14.1
15.4
16.7
16.7
f
23.12
24.56
26.59
29.38
33.02
9.53
48.63
43.07
38.42
34.97
31.75
TOTAL = (time) =
Time in mode (min) by assumption
1
1.41
1.52
1.60
1.62
1.81
1.66
1.73
1.67
1.63
1.61
1.46
17.73
2
3.18
3.43
3.61
3.64
4.07
3.74
3.89
3.77
3.67
3.62
3.29
39.91
Valve positions: large—open; small--5.5 turns closed
ASSUMPTION 1: C2 = 0.06 .'. (timeh = 0.0344
\
conclusion: (time) too short /. assume smaller C2
ASSUMPTION 2: C2 = 0.135 /. (time). = 0.0775
conclusion: (time) is OK
TABLE 11. WEIGHTING SCHEDULE FOR 11-MODE "COMPOSITE" RUNS
i =
Mode
1
2
3
4
5
6
7
8
9
10
11
Condition
rpm
peak torque
peak torque
peak torque
peak torque
peak torque
Idle
rated
rated
rated
rated
rated
Load, %
0
25
50
75
100
0
100
75
50 -
25
0
D. D. 6L-71T
Mode
time,
sec.
198
183
208
222
240
182
235
238
234
216
219
Curnul.
time,
sec.
198
381
589
811
1051
1233
1468
1706
1940
2156
2375
Cum. NTC-290
Mode
time,
sec.
191
206
217
219
244
224
233
226
220
217
197
Cumul .
time,
sec.
191
397
614
833
1077
1301
1534
1760
1980
2197
2394
31
-------
VII. RESULTS OF METHODOLOGY DEMONSTRATION
The project being reported on here was a very ambitious one,
including development or perfection of several particulate sampling
techniques, acquisition of gaseous emissions and smoke data as a
function of fuel composition, and development and demonstration of
a number of techniques for analyzing particulate content. Accordingly,
since no data base was available which applied to many of the consti-
tuents of interest, most of the results presented here cannot be considered
entirely authoritative until they can be tested against further research.
Time and cost limitations also served to restrict the number and range
of particulate analyses conducted, resulting in a less-than-complete
picture of fuel, additive, and engine effects on emission of a number of
constituents in particulate form. The demonstration did, however,
provide useful information regarding several classes of compounds and
elements. It also gave strong indications of the priorities which should
be attached to further research on many constituents by finding them
to be present in either significant or insignificant (or unmeasureable to
the limits of present techniques) quantities.
Due to the variety of analyses conducted, this section will be
subdivided by type of analysis for clarity. These subdivisions are
ordered as follows:
Federal smoke (opacity) evaluations
steady-state smoke measurement by Bosch spotmeter
and PHS opacity meter
gravimetric (emission mass rate and concentration)
analysis
constituents of particulate
sulfur, carbon, hydrogen, and nitrogen by combustion
total organic solubles by soxhlet extraction
metals by X-ray fluorescence
phenols and nitrosamines by extraction and chroma tog raphy
analyses of the organic soluble fraction of particulate
sulfur, carbon, hydrogen, nitrogen, and oxygen
by combustion
paraffin boiling point distribution
BaP (benz-oc-pyrene)
-- NMR and IR spectra
Care must be taken to note the qualifications attached to all the results,
especially those for which no comparative data are available in the litera-
ture.
It should be noted here that some of the particulate data acquired for
the Detroit Diesel 6L-71T engine are probably representative of an engine
with one or more malfunctioning injectors, limiting the usefulness of these
33
-------
data for fuel-to-fuel and engine-to-engine comparisons. As the data
are presented, those suspect from this standpoint will be identified and
qualified. No such problem occurred during operation of the Cummins
NTC-290 engine.
A. Federal Smoke (Opacity) Evaluations
One intent of these measurements was to document the extent
to which the two engines tested appeared to be typical of their respective
makes and models as compared to Federal certification data. They also
served as a comparison between fuels and could perhaps be used in a
correlation format as a rough predictor of total particulate emission rate
(although the establishment of such a correlation is beyond the scope of
this project). It should be noted that the "Certification data" do not agree
in all cases with smoke from the test engines but that disagreement such
as that present can exist from engine to engine of a given type. Further,
the "B" factor data for the Cummins NTC-290 appear to be quite a bit
higher than "Certification" results, but data from various issues of the
Federal Register strongly support a value of 4. 0 to 5.0 percent rather
than the 1. 7 percent figure supplied by Cummins personnel.
Both the Federal Smoke Tests and the steady-state tests (described
in the next section) are measures of the density or "opacity" of the smoke
plume generated by dies el engines. The measurement device employed
for these tests was the PHS smokemeter, consisting of a light source and
a. detector on opposite sides of the plume (plus associated control and read-
out equipment). Most of the smoke measurements taken on the two test
engines were quite low given the operating conditions, especially the steady-
state values. Smoke opacity of about 2 percent is considered near the visi-
bility limit, while 20 percent opacity is considered quite heavy. Federal
regulations on smoke applicable to the test engines are an "A" factor of
40 percent and a "B" factor of 20 percent. The regulation including the
11C" factor (50 percent maximum) did not go into effect until the 1974 model
year.
Federal smoke data on both engines are summarized in Table 12,
indicating mixed and rather minor variations among the fuels not containing
additives. The additive in fuel "A+" seemed to have little effect on smoke
from the Detroit Diesel engine, but a measureable suppressant effect for
the Cummins engine. The organo-metallic additive in fuels "B+" and "C+",
however, had a distinct effect on visible smoke from both engines in this
transient procedure (achieving nominal 50% reductions). These reductions
are considered highly significant. Later in this section the degree to which
this trend does or does not hold for particulate mass emissions will be ex-
amined. Complete data on Federal smoke tests are presented on pages
D-2 and D-3 of Appendix B.
34
-------
TABLE 12. AVERAGE FEDERAL SMOKE TEST RESULTS, TWO
ENGINES AND SIX FUELS, COMPARED TO FEDERAL
CERTIFICATION RESULTS
Engine
D.D. 6L-71T
Cum. NTC-290
Result
"A" factor
"B" factor
"C" factor
"A" factor
"B" factor
"C" factor
Certifica-
tion data
12. 8d
2.4d
27. Od
11. 2e
1.7e
14. 8 e
Average PHS smokemeter % opacity by^uel
A
14.3
3.6
20.4
7.0
4.3
10.8
B
12.3
1.9
19.1
7.8
4.2
11. 1
C
15.3
2.3
22. 2
7.9
4.4
10. 3
A+*
15.5
3.5
22.5
4.8
3.6
7.0
B+D.C
3.2
1.0
5.2
2.5
1.4
4. 1
C+°'c
8.0
1.4
11.8
3.8
1.6
6. 1
acontains ignition accelerator additive
contains smoke suppressant additive
cinjector repair prior to running this fuel for 6L-71T engine
daverage of 1973 and 1974 data
edata from factory, reportedly a composite from several similar engines
B.
Steady-State Smoke Tests (PHS opacity and Bosch spot)
These tests were conducted to document via accepted methods the
appearance or density of the exhaust plumes on which particulate mass
and concentration data were acquired. The operating conditions for
these tests were the same steady-state modes used in the particulate
evaluations, and Tables 13 and 14 present the results as averages for
the two engines. It is apparent that for steady-state conditions, both
engines have low smoke levels for all fuels and operating points. Both
engines employ turbochargers, meaning that all steady-state operation
occurs with comparatively lean mixtures (as compared to some naturally-
aspirated engines).
Even though smoke levels were low, limiting results to one or two
significant figures, there were mixed trends toward higher opacity and
higher Bosch numbers as power output increased at a given speed. It
is assumed that these trends were somewhat dependent on turbocharger
performance and the resulting fuel/air ratios, although a direct correlation
is not apparent. Reductions in smoke opacity and Bosch numbers due to
use of the smoke-suppressant additive (Lubrizol 8005 in B+ and C+ ) were
not so dramatic as for Federal smoke tests, but they still ran around 50
percent for conditions other than idle and zero load. These reductions
are considered quite significant.'
35
-------
TABLE 13. SUMMARY OF STEADY-STATE SMOKE DATA
FOR DETROIT DIESEL 6L-71T ENGINE
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine rpm
and load
Idle
1600 - 0%
1600 - 25%
1600 - 50%
1600 - 75%
1600 - 100%
Idle
2100 - 100%
2100 - 75%
2100 - 50%
2100 - 25%
2100 - 0%
Idle
PHS smokemeter % opacity by fuel code
A
0.3
0.7
0.8
1.0
1.7
3.4
0.2
1.9
1.4
1.2
1.0
1.1
0,2
B
0.3
0.4
0.6
0.8
1. 1
1.6
0.2
1.0
0.8
0.7
0.8
0.9
0.3
C
0.3
0.3
0.5
0.8
1. 1
1.9
0.2
1.1
0.9
0.9
0.8
0.9
0.2
A + add.
0.4
0.6
0.8
1.0
1.6
3.3
0.2
1.6
1.0
1.0
0.9
1.2
0.2
B + add. a
0.2
0.6
0.6
0.6
1.2
2.6
0.2
1.6
1.2
0.9
0.9
1.4
0.2
C + add. a
0.3
0.4
0.6
0.7
0. 8
1.3
0.3
1.0
0.9
0.7
0. 8
0. 8
0.3
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine rprn
and load
Idle
1600 - 0%
1600 - 25%
1600 - 50%
1600 - 75%
1600 - 100%
Idle
2100 - 100%
2100 - 75%
2100 - 50%
2100 - 25%
2100 - 0%
Idle
Bosch smoke number by fuel
A
0. 1
0.2
0.5
0.7
1. 1
1.9
0.1
0.9
0.7
0.5
0.5
0.7
0. 1
B
0. 1
0.1
0.2
0.4
0.6
1.0
0.2
0.3
0.3
0.3
0.2
0.4
0.2
C
0. 1
0. 1
0.2
0.4
0.6
1.1
0. 1
0.3
0.2
0.2
0.2
0.4
0. 1
A + add.
0.1
0.2
0.4
0.6
1.1
1.8
0. 1
0.8
0.4
0.4
0.3
0.7
0.1
B + add. a
0.2
0.3
0.3
0.7
0.7
1.2
0.2
0.4
0.4
0.4
0.4
0.6
0.2
C + add. a
0. 1
0. 1
0. 1
0.2
0.3
0.6
0. 1
0.2
0.2
0. 1
0.1
0.3
0. 1
asome of these results reflect minor injector problems - use care in comparing
them to those from other fuels
36
-------
TABLE 14. SUMMARY OF STEADY-STATE SMOKE DATA
FOR CUMMINS NTC-290 ENGINE
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine rpm
and load
Idle
1500 - 0%
1500 - 25%
1500 - 50%
1500 - 75%
1500 - 100%
Idle
2100 - 100%
2100 - 75%
2100 - 50%
2100 - 25%
2100 - 0%
Idle
PHS smokemeter % opacity by fuel code
A
0.6
0.7
1.5
1.6
1.4
1.9
0.3
2.6
1.8
2.0
2.4
1.2
0.4
B
0.8
1.1
1.8
2.5
2.1
2.1
0.5
3.3
2.1
2.5
2.2
1.1
0.6
C
1.0
1.0
2.0
2.2
1.7
2.0
0.4
2.8
2.2
2.7
2.8
1.8
0.4
A + add.
0.8
0.5
1.0
1.4
0.8
1. 1
0.3
2.2
1.6
1.8
2.0
1.0
0.4
B + add.
0.8
0. 6
0.8
0.6
0.5
0.7
0.2
0.8
0.4
0. 6
0.8
0.8
0.4
C + add.
0.6
1.0
1.0
1.0
1.0
1. 1
0.7
1.3
1.0
1.0
1.2
1.3
0.8
Mode
1
2
3
4
5
6
7
8
9
10
11
12
13
Engine rpm
and load
Idle
1500 - 0%
1500 - 25%
1500 - 50%
1500 - 75%
1500 - 100%
Idle
2100 - 100%
2100 - 75%
2100 - 50%
2100 - 25%
2100 - 0%
Idle
Bosch smoke number by fuel
A
0.1
0.4
0.9
1.1
0.9
1.2
0.2
1.6
1.2
1.3
1.4
0.7
0.2
Ba
0.2
0.6
1. 1
1.4
1.1
1.2
0.2
1.8
1.4
1.4
1.5
0.8
0. 1
C
0.2
0.6
1.1
1.2
1.0
1. 1
0.2
1.7
1.2
1.5
1.5
0.9
0.2
A + add.
0.2
0.4
0.9
1.0
0.9
1.1
0.2
1.6
1.2
1.4
1.5
0.8
0.2
B + add.
0.2
0.5
0.6
0.6
0.4
0.6
0.2
0.6
0.4
0.5
0. 6
0.6
0.2
C + add.
0.2
0.4
0.4
0.4
0.2
0.4
0.2
0.6
0.3
0.4
0.6
0.4
0.2
acheck after particulate work on fuel "B" gave 0. 2 at idle, 1. 2 at 1500 rpm
and J00% load, and 1.8 at 2100 rpm and 100% load
37
-------
It is interesting to note that different mixed trends toward higher
values were also the case for particulate mass emissions and particulate
concentrations. It is possible, therefore, that while smoke opacity is
due to some sort of exhaust particulate, this light-bio eking material is
not the major factor in particulate mass. This situation is obviously the
case for tests involving use of the organo-metallic smoke suppressant
additive, since opacity decreased significantly and particulate mass gen-
erally increased as compared to tests conducted without the additive.
C.
13-Mode Gaseous Emissions Tests
Gaseous emissions from the two engines were measured on all
six test fuels using the well-accepted 13-mode procedure for diesels.
These data are intended to document any variation in emissions caused
by fuel composition, and the average results are given in Table 15.
TABLE 15. AVERAGE 13-MODE GASEOUS EMISSIONS RESULTS
Engine
Fuel
code
Number
of runs
Statutory limits (1974)
Detroit Diesel
6L-71T
Cummins
NTC-290
A
B
C
A+
B+
C +
A
B
C
A+
B+
C +
2
6
3
2
2
2
2
8
2
2
2
3
Cycle composite emissions, g/hp hr
HCa
0.96
0.68
1.00
1.12
1.18
1.32
0.30
0.33
0.28
0.26
0.40
0.36
CO
40.
2.20
2.46
2.33
2.22
3.14
3.02
2.50
2.46
2.38
2.46
2.79
2.93
NOX
11.8
12.5
11.4
10.4
11.2
11.5
12.0
12.5
12.4
12.1
12.5
12.6
HCa + NOV
Ji
16.
12.8
13.2
12.4
11.5
12.4
12.8
12.3
12.8
12.7
12.4
12.9
13.0
asome data for 6L-71T engine may be representative of slight malfunctions
in injection system
The data in Table 15 show that fuel variation had little influence on regu-
lated emissions. There is perhaps a weak indication that HC and CO
emissions from these engines increased slightly when fuels containing
the organo-metallic additive were used (B+ and C+), but a great deal
more testing would be necessary to substantiate it. Gaseous emissions
data are given by individual runs on pages D-4 and D-5 of Appendix D.
38
-------
D. Gravimetric Analysis - Particulate Mass Emissions and
Concentrations
The data in this subsection are the cornerstone of all the chemical
and physical analyses performed on particulate samples from the engines
tested. Computation of particulate mass rates and concentrations has
already been discussed, but it is important to recall that all rates and
concentrations of particulate constituents are calculated by assuming that
they are known fractions of the total collected sample. For these reasons
the multiplicity of data presented here is not superfluous, but rather only
necessary documentation.
Basic processing of data by computer yielded particulate data in
grams per hour and in micrograms per standard cubic meter of exhaust
gas. The mass rates turned out to be in a useable range numerically
(mostly between 1 and lOOg/hr), but the concentrations were unwieldy
(mostly in the range of 1000 to 100,000). Accordingly, particulate con-
centration data will be summarized in milligrams per standard cubic
meter of exhaust gas for ease of presentation. Summaries of average
particulate mass rates (g/hr) and concentrations (mg/m ) are given by
engine, fuel, and operating condition in Appendix E, pages E-2 through
E-ll.
For presentation in the text, particulate emissions have been
calculated on a specific basis. The average data from runs using 47mm
filters have been divided by the average fuel rates (different for each
fuel), yielding fuel specific data for each mode (and for composites) in
grams of particulate per kilogram of fuel consumed. This technique
circumvents the problems inherent in presentation of brake specific data
by mode where the power output can be zero (or near zero). The fuel
specific data are presented in Table 16 for both engines, and they have
been graphed in Figures 20 and 21.
In discussing these fuel specific results, it should first be noted
that some of the data for the 6L-71T engine are probably representative
of an engine with one or more malfunctioning injectors. These problems
occurred during testing of fuels B+ and C+; and efforts were made to
correct them, but they obviously had some influence on the results. If
fuels B+ and C+ are eliminated from the comparison, the 6L.-71T engine
still emitted considerably more particulate per unit of fuel consumed
than the NTC-290. As an average over the composite runs, fuel spe-
cific particulate emissions from the 6L-71T engine were about 3. 2 times
those from the NTC-290 (factors varied from 2. 3 to 3. 9 by fuel).
Other trends became obvious by examining Figures ZO and 21,
beginning with distinct variation patterns related to engine power level
and operating speed. Another trend visible in the Cummins NTC-290 data
is consistently lower fuel specific particulate for fuels A and A+ than for
39
-------
TABLE 16. FUEL SPECIFIC PARTICULATE RESULTS FROM
TWO ENGINES OPERATED ON SIX FUELS
47mm GLASS FIBER FILTERS
Engine
6L-71T
NTC-290
rpm
idle
1600
1600
1600
1600
1600 .
2100
2100
2100
2100
2100
Load, %
0
25
50
75
100
0
25
50
75
100
Composite
idle
1500
1500
1500
1500
1500
2100
2100
2100
2100
2100
0
25
50
75
100
0
25
50
75
100
Composite
Grams particulate per kilogram fuel
by fuel code
A
5.9
5.0
2.9
2.6
2.4
1.8
5.8
3.8
3.9
2.5
2.6
3.8
0.49
1.2
1.1
0.87
0.81
0.54
1.6
2.0
1.4
1.0
0.93
1.0
B
15.
6.5
4.0
3.5
3.7
1.9
6.3
3.6
3.5
3.1
2.9
3.7
2.8
2.1
1.9
1.7
1.2
0.98
2.6
3.3
2.1
1.7
1.3
1.6
C
6.6
5.3
3.5
3.0
2.5
1.7
5.2
3.1
3.5
2.6
1.8
3.4
1.2
1.7
1.4
1.3
0.78
0.62
2.3
2.4
1.6
1.0
1.0
1.2
A+
3.0
5.1
3.6
2.8
2.3
1.7
4.8
3.0
3.4
2.9
2.7
3.7
0.35
1.4
1.2
0.95
0.73
0.54
1.9
1.9
1.1
0.96
1.1
0.95
B +
1.7
9.4
13.
15.
4.7
3.8
15.
24.
11.
6.1
4.1
7.2
2.9
3.4
2.3
2.1
1.8
1.7
3.4
3.9
2.2
1.9
1.8
2.0
C +
6.5
12.
13.
6.4
3.4
2.6
7.7
6.5
5.0
4.7
3.0
3.7
2.2
2.6
2.1
1.8
1.6
1.3
2.8
3.0
1.8
1.6
1.5
1.3
40
-------
24
v
20
16
u 12
4J
M
rt
a
bO
8
Note: Some data for fuels B+
and C+ may be representative
of injector malfunction
B+
B+
0 25 50 75 100
Percent of full load at 1600 rpm
0 25 50 75 100
Percent of full load at 2100 rpm
FIGURE 20. FUEL SPECIFIC PARTICULATE EMISSIONS FROM A DETROIT DIESEL-
ALLISON 6L-71T ENGINE, FUEL AS PARAMETER
-------
CO
d>
,5 2
o
t«
«J
A+
0 •-
0 25 50 75 100
Percent of full load at 1500 rpm
0 25 50 75 100
Percent of full load at 2100 rpm
FIGURE 21. FUEL, SPECIFIC PARTICULATE EMISSIONS FROM A CUMMINS
NTC-290 ENGINE, FUEL AS PARAMETER
-------
the others. Fuel A was the lightest fuel tested, while C and B were pro-
gressively heavier. Fuel specific particulate was consistently higher for
fuel B+ than for fuel B and consistently higher for fuel C+ than for fuel C.
These results indicate that particulate emissions were greater when the
organo-metallic smoke-suppressant additive was used for the engines and
fuels tested in this study.
Although it is not practical to attempt mode-by-mode brake
specific particulate computation, as mentioned earlier, it is relatively
straightforward to compute brake specific values for composite runs.
The results of such computations are shown in Table 17t confirming
the previously-mentioned difference between the two engines tested.
TABLE 17. COMPOSITE BRAKE SPECIFIC PARTICULATE
EMISSIONS FROM TWO ENGINES OPERATED ON
SIX FUELS, 47mm GLASS FIBER FILTERS
Engine
6L-71T
NTC-290
grams
A
0.90
0.27
^articulate per k.W hr by fuel code
B
0.94
0.46
C
0.87
0.34
A+
0.94
0. 27
B+
2.0a
0.58
C+
1.0a
0. 38
athese data may be representative of an engine with one
or more malfunctioning injectors
Although the sampling and calculation methods were quite different,
the data in Table 17 agree reasonably well with earlier work on diesel
particulate^' which showed a range of 0. 28 to 3. 0 g/kW hr for eight
engines operating on a fuel similar to fuel B.
For this presentation of specific particulate data, values obtained
using 47mm glass fiber filters have been used. The reasons for this
choice are primarily the sheer weight of data available and the good
consistency observed in the data. Values from tests using the hi-vol
system averaged somewhat higher than those from tests using the 47mm
system, probably due to the smaller surface-to-volume ratio of the hi-
vol system. Hi-vol data were not quite as comprehensive as 47mm
glass fiber data, however, thus the latter was chosen. Data generated
using 47mm Fluoropore filters could have been used, but three glass
fiber filters were acquired for each Fluoropore, providing a better data
base.
Only a few runs were made with 293mm Fluoropore filters, so
they were not considered for further analysis. A number of runs were
made with the ERC sampler-diluter, but it was incapable of sampling
during composites. In addition, the repeatability of ERC data was
-------
generally poor,' and nominal values deviated widely from those obtained
via dilution tunnel measurements. For mass particulate rates averaged
over four fuels (6L-71T) or six fuels (NTC-290) at single operating con-
ditions, coefficients of variation for ERG samples were consistently higher
than tunnel-collected samples (coefficient of variation is standard deviation
divided by mean = s/x). For the NTC-290, the average coefficient of var-
iation for ERG samples was 0. 46, while that for similar filters (47mm
Fluoropore) collected via tunnel was 0.39. For the 6L-71T, the average
coefficient of variation for ERG samples was 0. 47 and that for tunnel-
collected samples was 0.25.
During most sampling conditions with the NTC-290 engine, ERG
particulate mass rate results were from 40 percent to 60 percent of those
obtained by tunnel measurements. Data from runs on the 6L-71T engine
at 1600 rpm agreed quite well between ERG and tunnel methods, but at
2100 rpm the ERG data ran from 45 percent to 80 percent of the tunnel
data. Problems with the ERG sampler which are likely causes of the dis-
agreements noted are deposition of particles in the sample line and venturi,
inability to sample isokinetically due to lack of control range, and inability
to balance sample and primary dilution flows due to lack of control range.
For these reasons, further analysis of ERG data in fuel specific or brake
specific terms was not considered.
Although all the average particulate concentration data are given
in Tables E-2 and E-4, the results for composite runs are restated as
Table 18 for convenience. Here the values for the two engines appear
TABLE 18. COMPOSITE PARTICULATE CONCENTRATION
DATA FROM TWO ENGINES OPERATED
ON SIX FUELS, 47mm GLASS FIBER FILTERS
Engine
6L-71T
NTC-290
mg particulate per standard m exhaust gas by fuel
A
59.
28.
B
65.
48.
C
59.
36.
A+
61.
27.
B+
129. a
61.
C+
67.a
39.
athese data may be representative of an engine with one
or more malfunctioning injectors
to agree more closely than in the specific data, but it should be noted
that the exhaust volume per unit work from the 6L-71T is much larger
than that from the NTC-290 due to the blower-scavenged 2-stroke design
of the former.
44
-------
E. Constituents of Particulate
Several analyses were conducted to determine particulate com-
position, and they will be outlined in this subsection. Analyses run spe-
cifically on the organic solubles found in particulate samples will be
presented in the next subsection.
1. Carbon, Hydrogen, Nitrogen, and Sulfur by Combustion
In order to determine the gross composition of diesel parti-
culate, samples were analyzed for C, H, N, and S by commercial
laboratory. The data resulting from these analyses are presented as
Tables 19 and 20, and a great deal of useful information can be obtained
by examining them. Perhaps the most interesting and most obvious
trends in the data are the large differences in carbon/hydrogen ratio
between the two engines and the distinct drop in total measureable con-
stituents from fuels B and C to fuels B+ and C+. It is assumed that
the percentage by which the sums fall short of 100. 0 represent the
fraction of particulate which exists as ash, metals, and possibly other
substances.
To quantify the above observations to some extent, Table 21
has been prepared to show the average percentage of C, H, N, and S in
particulates for each engine and each fuel. The average carbon/hydrogen
ratio by mass for untreated fuels (no extra additives) was about 6. 8 for
the Detroit Diesel 6L.-71T, and about 15 for the Cummins NTC-290. This
difference means that the particulate from the 6L.-71T was mostly a hydro-
carbon-like material (n-Cj.6 has a carbon/hydrogen ratio of about 5. 6),
and that particulate from the NTC-290 was more a carbon- (or soot-) like
material. The difference was also confirmed by the appearance of filters
collected. To document the effect of the metal-containing additive (used
in B+ and C+), the average total particulate fractions made up of C, H,
N, and S were reduced 20% for the 6L.-71T and 39% for the NTC-290 when
additive-containing fuels B+ and C+ were used in place of fuels B and C.
The difference in reductions is mostly traceable to the difference in basic
particulate rates from the two engines, and the reductions themselves to
the presence of metals and (possibly) ash on the filters.
2. Analysis for Total Organic Solubles in Particulate
Organic solubles were extracted from particulate samples
collected on rectangular (8 inch by 10 inch) glass fiber filters by the
Soxhlet technique. The extractions were performed in methylene chloride,
benzene, or hexane for a period of four hours with refluxing at 20 cycles
per hour. Following extraction, the solvent was evaporated to permit
determination of the weight of soluble material present. Ultrasonic ex-
45
-------
TABLE 19. ANALYSIS FOR CARBON, HYDROGEN, NITROGEN, AND
SULFUR IN PARTICULATE SAMPLES, DETROIT DIESEL 6L-71T ENGINE
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
21 00 rpm - 1 00% load
Composite
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm • 0% load
21 00 rpm - 25% load
2100 rpm - 50% load
21 00 rpm- 75% load
2100 rpm - 100% load
Composite
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
21 00 rpm - 75% load
2100 rprn - 100% load
Composite
Weight % - fuel "A"
C
53. 7a
70.4
81.1
72.9
71.2
69.6
71.2
64.8
75.1
70.8
70.8
64.9
H
7.3
11.3
12.6
11.1
9.5
6.3
10.9
10.4
11.8
11.4
9.7
10.5
N
b
b
b
b
b
0.5
b
b
b
b
b
b
S
b
b
b
b
0.2
b
b
b
0.4
3.9
1.1
1.0
£%
61. Oa
81.7
93.7
84.0
80.9
76.4
82.1
75.2
87.3
86.1
81.6
76.4
Weight % - fuel "C"
C
68.4
73.3
77.2
75.0
72.7
77.6
74.5
75.0
H
9.1
11.1
10.7
a
9.7
11.7
12.0
11.3
N
b
0.1
0.3
0.1
0.9
0.6
b
b
S
2.1
0.7
0.6
0.5
1.0
0.6
0.9
1.7
s>
79.6
85.2
88.8
a
84.3
90.5
87.4
88.0
Weight % - fuel "B + additive"
C
67.4
64. 6C
56.8
61.5
39.9
48. 9C
73.3
63.9
66.2
55.7
51.6
52. 8C
H
10.8
7.7C
6.2
3.1
4.4
3.2C
5.6
2.5
3.6
3.6
4.2
5.5C
N
2.0
0.2C
0.7
°'5b
1.8C
0.8
°'7b
b
b
0.6C
S
5.6
4.3C
2.0
1.5
5.1
7.4C
2.9
1.2
2.7
4.3
4.9
4.7C
E%
85.8
76.8
65.7
66.6
49.4
61. 3C
82.6
68.3
72.5
63.6
60.7
63. 6C
Weight % - fuel "B"
C
60.9
69.1
76.8
63.5
71.1
66.7
66.6
65.9
65.8
69.9
70.7
71.8
H
7.7
10.3
12.8
10.9
10.5
8.8
9.8
10.4
9.8
a
10.9
10.0
N
b
1.0
1.0
b
0.3
0.8
0.8
b
b
b
0.4
0.3
S
3.0
2.1
2.6
2.9
2.1
2.3
1.8
2.8
2.4
1.6
2.2
2.0
£*
71.6
82.5
93.2
77.3
84.0
78.6
79.0
79.1
78.0
a
84.0
84.1
Weight % - fuel "A + additive"
C
71.4
65.3
69.3
62.0
63.2
73.0
66.8
67.1
H
11.8
10.4
11.6
6.2
10.4
11.9
10.7
10.3
N
0.1
b
0.4
b
b
b
b
b
b
b
S
3.9
0.6
0.8
1.3
0.4
0.7
0.8
0.8
£%
87.0
76.3
82.1
69.5
74.0
85.6
78.3
78.2
Weight % - fuel "C + additive"
C
68.6
58.3
57.4
42. 3C
58.7
52.5
45.6
50. lc
H
11.5
8.8
5.3
5.2C
7.4
7.8
7.1
6.6C
N
b
b
0.3
b, c
1.6
b
b
c
S
6.9
3.8
1.8
2.7C
3.6
2.4
2.1
2.7°
E%
87.0
70.9
64.8
50. 2C
71.3
62.7
54.8
59. 4C
Questionable data "r
Below detectable limit
c Average of two rune
46
-------
TABLE 20. ANALYSIS FOR CARBON, HYDROGEN, NITROGEN, AND
SULFUR IN PARTICULATE SAMPLES, CUMMINS NTC-290 ENGINE
Operating condition
Idle
1500 rpm - 0% load
1500 rpm 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Weight % - fuel "A"
C
82.2
67.9
82.1
90.3
84.3
61.2
70.1
88.6
87.3
72.0
82.2
69.8
H
10.0
6.0
6.8
3.4
5.0
2.2
5.5
5.3
3.6
5.4
2.5
3.2
N
4.1
4.0
1.4
2.9
0.5
1.6
0.6
0.9
0.5
0.6
a
2.9
S
2.0
1.2
1.9
1.6
5.5
3.4
4.5
2.2
1.4
1.3
2.7
5.4
"£%
98.1
79.1
92.2
98.2
95.3
68.4
80.7
97.0
92.8
79.3
87.4
81.3
Weight % - fuel "C"
C
60.3
84.4
77.1
86.0
83.0
75.2
78.3
79.3
83.9
84.9
88.2
85.4
H
9.5
9.5
6.8
12.8
6.5
8.6
9.2
5.5
6.1
6.0
7.5
9.0
Weight % -
C
48.0
37.8
34.1
34.2
32.9
17.3
47.8
38.1
31.3
20.6
17.1
34.6
H
4.5
4.8
2.7
2.5
2.0
2.1
3.7
2.5
2.4
1.7
1.1
3.6
N
3.8
0.5
a
a
1.6
0.7
a
a
0.6
0.3
0.5
2.3
fuel "B
N
0.3
5.4
1.2
1.6
1.7
1.2
1.4
1.0
1.2
0.8
1.1
1.7
S
5.7
4.4
4.3
1.1
2.0
4.6
5.4
0.9
3.0
2.2
3.2
3.1
£%
79.3
98.8
88.2
99.9
93.1
89.1
92.9
85.7
93.6
93.4
99.4
99.8
+ additive"
S
7.5
8.5
8.1
7.1
8.0
9.0
6.1
4.0
6.8
8.9
10.6
8.5
£%
60.3
56.5
46.1
45.4
44.6
29.6
59.0
45.6
41.7
32.0
29.9
48.4
Weight % - fuel "B"
C
51.4
54.4
70.2
51.8
58.3
54.6
63.3
70.5
70.2
62.7
61.2
61.9
H
4.7
6.1
4.5
3.1
2.4
2.4
3.6
2.8
3.0
2.3
2.1
2.3
N
a
a
a
a
a
a
a
0.3
a
0.5
1.6
a
S
4.8
4.2
4.3
5.0
6.7
5.9
3.3
3.0
3.3
2.2
5.6
3.3
Z%
60.9
64.7
79.0
59.9
67.4
62.9
70.2
76.6
76.5
67.7
70.5
67.5
Weight % - fuel "A + additive"
C
71.1
53.8
87.4
45.1
82.6
71.8
81.3
88.0
83.6
85.0
84.0
70.2
H
3.1
5.7
3.4
a
2.9
1.6
4.6
3.3
2.9
2.3
2.2
2.7
N
a
a
0.9
a
0.4
a
0.5
0.3
1.8
1.4
2.3
a
Weight % - fuel "C
C
42.9
59.1
40.8
46.9
36.4
30.2
52.4
57.6
39.7
35.4
27.2
73.2
H
4.2
4.9
2.9
2.8
1.4
1.4
3.9
2.2
2.7
1.5
2.4
5.1
N
0.8
6.5
2.2
2.4
1.5
2.8
3.0
1.9
2.1
0.9
2.0
6.3
S
9.4
4.5
1.8
1.9
1.3
3.2
2.1
1.3
2.1
1.3
1.7
2.7
£*
83.6
64.0
93.5
47.0
87.2
76.6
88.5
92.9
90.4
90.0
90.2
75.6
+ additive"
S
5.2
3.9
1.6
1.6
1.4
1.6
2.1
1.5
1.6
1.6
2.4
3.0
£%
53.1
74.4
47.5
53.7
40.7
36.0
61.4
63.2
46.1
39.4
34.0
87.6
Below detectable limit
47
-------
TABLE 21. AVERAGE PERCENTAGE OF C, H, N, AND S
IN PARTICULATE BY ENGINE AND FUEL
Engine
Detroit Diesel 6L-71T
Cummins NTC-290
Fuel
A
B
C
A+
B+
C+
A
B
C
A+
B+
C+
Average weight % by constituent
C
69.7
68.2
74.2
67.3
58.6
54.2
78.2
60.9
80.5
75.3
32.8
45.2
H
10.2
10. 2
10.8
10.4
5.0
7.5
4.7
3.3
8.1
2.9
2.8
3.0
N
0.1
0.4
0.2
0. 1
0.6
0.2
1.7
0.2
0.9
0.6
1.6
2.7
S
0.6
2.3
1.0
1.2
3.9
3.2
2.8
4.3
3.3
2.8
7.8
2.3
£
80.5
81. 1
85.3
79.0
68. 1
65. 1
87.4
68.7
92.8
81.6
45.0
53.2
traction -was evaluated also, but it did not show any improvement over the
Soxhlet technique. Data on organic solubles as percent of total particulate
is presented in Table 22 by engine, fuel, and operating condition.
The data in Table 22 show that the percentage of organic
solubles in particulate is considerably higher for the 6L-71T than
for the NTC-290. This is an expected result, given the carbon and
hydrogen percentages in the particulate samples discussed earlier. No
trend of percentage organic solubles as a function of operating condition
is readily obtainable from the data on the 6L-71T, but it appears that
this percentage tended to decrease with increasing load for the NTC-290.
No distinct trend of percentage organic solubles as a function of fuel is
apparent for either engine.
3. Metals in Particulate by X-ray Fluorescence
As described in an earlier section, X-ray fluorescence was
used to analyze samples on Fluoropore filters for 10 metals. The re-
sults of all 112 determinations are presented in full in Appendix F, pages
F-2 through F-5. The only metal found consistently in nearly all the
samples was zinc, normally at levels of about 1 microgram per cm^ on
the filters, which translates to an emission rate of perhaps 1 g/hr or less.
The most interesting results were for those samples taken while the en-
gines were operated on fuels containing the organo-metallic additive
(fuels B+ and C+). For these tests we have both the metal input to the
engine (in the fuel) and the metal output in the particulate, so a material
balance can be attempted. A summary of these results is given in Table 23,
and although the agreement between metals collected and metals in the fuels
48
-------
TABLE 22. ORGANIC SOLUBLES AS PERCENT OF TOTAL PARTICULATE
Com
rpm
idle
PTa
PT
PT
rated
rated
rated
iition
Load, %
— V —
0
5C
100
0
50
100
Composite
idle
PT
PT
PT
rated
rated
rated
_.•.
0
50
100
0
50
100
Composite
idle
PT
PT
PT
PT
rated
rated
rated
rated
rated
...
0
50
75
100
0
25
50
75
100
Compos ito
Engine
6L-71T
Fuel
A
B
C
Percent
solubles
36.
72.
31.
55. b
37. b
48.
50.
32.b
47. b
47. b
64.
52. b
48.
32.
57.
55.
76.
73.
67.
64. b
Fuel
A+
B+
C+
Percent
solubles
58.
20.
54. b
24. b
14. b
Engine
NTC-290
Fuel
A
B
C
Percent
solubles
10.
8.1
7.0
22.
16.
3.9
3.3b
15.
34.
4.5
0.91b
12.
7.3
1.8
6.0
9.9
16.
7.4
2.8
14.
16.
c
2.4
Fuel
A+
B+
C+
Percent
solubles
9.0
18.
10.
c
13.
16.
12.
8.1b
29.
25.
15.
8.7
13.
9.1
7.2b
c
9.2
1.6
3.5
16.
12.
1.3
7.0
* Manufacturer's peak torque speed; 1600 rpm for 6L-71T and 1500 rpm for NTC-290
b Average of two or more determinations
c Too small to measure
49
-------
TABLE 23. MATERIAL BALANCES FOR METALS IN ORGANO-METALLIC
SMOKE - SUPPRESSANT ADDITIVE, FUELS B+ AND C+
Engine
6L-71T
NTC-290
Fuel
B+
C+
B+
C+
Condition
rpm
idle
1600
1600
1600
1600
1600
2100
2100
2100
%load
0
.0
50
100
100
0
50
100
composite
composite
idle
1600
2100
50
100
composite
composite
idle
1500
1500
1500
2100
2100
2100
0
50
100
0
50
100
composite
composite
idle
1500
1500
1500
2100
2100
2100
0
50
100
0
50
100
composite
composite
Calcium, g/hr
Metal in fuel
0.34
2.0
2.1
8.8
16.
15.
3.7
11.
19.
9.0
9.2
0.35
8.4
18.
9.0
9.1
0.70
2.4
8.5
16.
3.9
12.
20.
8.4
8.3
0.59
2.1
8.4
16.
4.0
12.
20.
8.2
8.2
Metal collected
0.07
1.0
1.8
11.
19.
18.
4.9
14.
23.
14.
11.
0.57
8.3
25.
12.
7.4
0.45
2.3
7.7
16.
4.2
12.
20.
3.6
9.4
0.33
2.1
7.9
13.
4.3
a
19.
11.
12.
Barium, e/hr
Metal in fuel
0.07
0.38
0.40
1.7
3.0
3.0
0.71
2. 1
3.6
1.7
1.8
0.07
1.6
3.5
1.7
1.8
0.13
.0.45
1.6
3.1
0.75
2.2
3.8
1.6
1.6
0.11
0.40
1.6
3.1
0.78
2.2
3.8
• 1.6
1.6
Metal collected
a
a
a
1.5
3.0
2.1
a
1.6
2.4
3.4
2.7
a
1.7
3.6
2.2
1. 1
0.07
.0. 44
1.3 .
2.8
0.73
1.9
3.0
a •
1.7
0.05
0.39
1.3
1.8
0.83
1.9
2.9
2.0
2.1
no data
50
-------
is not perfect, a good correlation certainly exists. It should be noted that
barium and calcium have the highest minimum detectable limits of the ten
metals analyzed for, due to optimization of the detection parameters for
heavier metals (see page A-2). If the X-ray system were optimized for
calcium and barium, the results could be expected to have smaller vari-
ability.
To document the relationships between metals in fuel and
metals collected, regression equations have been computed for both engines
and both fuels as follows:
metal collected = a0 +
(metal in fuel)
Engine
6L-71T
NTC-290
Fuel
B +
C+
B+
C+
Metal
Ca
Ba
Ca
Ba
Ca
Ba
Ca
Ba
ao
-0. 19
0.17
-1.83
-0.09
-0.60
0. 10
1.21
0.35
al
1.25
0.83
1.39
1.04
1.01
0.82
0.89
0.67
r2
0. 980
0.575
0. 930
0.778
0.935
0. 976
0.897
0.789
Perfect correlations would have r2 of 1.0, a0 = 0, and a^ = 1. 0.
Expressed in terms of percentage of total particulates for runs on
fuels B+ and C+, metals increased with engine power output, which in turn
increased with engine fuel rate. Calcium ranged from about 3 percent to
17 percent of total particulate for the 6L-71T engine and constituted about
8 percent to 29 percent of total particulate for the NTC-290 engine (slightly
higher percentages for fuel C + than for fuel B+). Barium constituted about
1 percent to 2 percent of particulate for the 6L-71T engine and some 1 per-
cent to 4 percent of particulate for the NTC-290 (again slightly higher for
fuel C+ than for fuel B+).
4. Phenols and Nitrosamines
In summary, no nitrosamines were found in the samples
submitted for analysis to a detection limit of 0. IjL/g on a hi-vol filter.
In terms of particulate quantity, this limit means that if nitrosamines
were present, they constituted less than 1 x 10~^% of particulate in
all cases and less than 1 x 10"4% of particulate for typical cases. Due
to filter backgrounds, phenols found were in such quantity that no real
credibility can be attached to their existence in particulate samples.
Of the 9 phenols analyzed, no trace of 2, 6 xylenol or 3,4 xylenol was
51
-------
found in any sample. Quantities of other phenols were essentially traces,
in no case more than IjUg on a filter (or 1 x 10~3% of particulate collected).
Twenty samples from runs on the 6L-71T were analyzed, and 21 samples
from the NTC-290 were analyzed, in addition to a number of blanks.
F. Constituents of Organic Soluble Fraction of Particulate
A variety of analyses were conducted on organic solubles, and
they will be presented separately for clarity.
1. Carbon, Hydrogen, Nitrogen, Sulfur, and Oxygen
by Combustion
The gross composition of the organic soluble fraction of a
number of diesel particulate samples was determined by a commercial
laboratory using combustion analysis. The samples submitted were a
part of those used in the BaP analysis and for NMR and IR spectral analysis,
and most of them were quite small (10 to 100 rng). In a number of cases,
sample size was not sufficient for oxygen determination, and sufficient for
only a rudimentary sulfur determination. The number of samples submitted
for analysis was limited because relatively few of them contained enough
sample for a proper analysis (15 mg was the nominal lower limit for a
complete analysis).
Data on composition of organic solubles are presented in
Table 24 for all the samples submitted. No samples from runs on fuels
B+ or C+ were submitted because metals supposedly create an interference
with the oxygen measurements.
The most obvious conclusion to be reached about the data in
Table 24 is that they are highly indicative of a hydrocarbon-type material
with relatively small amounts of nitrogen, sulfur, and oxygen. The average
carbon/hydrogen ratio by weight for the solubles from the 6L.-71T engine
was 6. 5, and that for the solubles from the NTC-290 engine was 7. 1. For
comparison, the carbon/hydrogen ratio by weight of n-Cj^H^ is 5.65.
Although lack of accuracy in oxygen measurement restricts the strength
of conclusions to be drawn, it appears that surprisingly little oxygen was
present in the organic solubles. The presence of oxygenated compounds
was detected in a few of the samples by spectral analysis, as will be dis-
cussed in a later subsection.
2. Paraffin Boiling Point Distribution
Boiling point distributions have been obtained for a number of sam-
ples of organic solubles in the range of 230 to 540° C (450 to 1000°F) by
gas chromatograph. Most of the samples have been run both "as extracted"
and as paraffins only (following fractionation). In general, the boiling point
distribution of the paraffins recovered followed that of the total solubles
52
-------
TABLE 24. CARBON, HYDROGEN, NITROGEN, SULFUR,
AND OXYGEN IN THE ORGANIC SOLUBLES OF DIESEL PARTICULATE
Engine
D.D. 6L-71T
NTC-Z90
Fuel
A+
B
C
A
A+
B
C
Condition
rpm
1600
load, %
50
composite
idle
1600
2100
50
100
composite
idle
idle
1600
1600
2100
2100
50
75
0
75
composite
2100
1500
2100
idle
1500
2100
2100
2100
50
50
0
0
0
50
50
Percent by weight in organic solubles
C
86.
85.
86.
83.
79.
82.
85.
85.
86.
85.
85.
86.
80.
70.
71.
83.
72.
83.
84.
83.
78.
H
13.
13.
13.
13.
12.
12.
13.
13.
13.
13.
13.
14.
12.
11.
11.
12.
8.7
11.
12.
12.
11.
N
0. 1
0. 1
0.2
0.4
0.7
0.2
0.1
0. 1
0. 1
0. 1
0.2
0.0
0.4
3.6
0. 1
1. 1
0.8
0. 1
0.6
0.5
0.0
S
0. 1
0. 1
0.3
0.2
0.2
0. 8
0. 1
0. 1
0.2
0.2
0.4
0. 1
0.1
b
b
___b
b
-lib
_"b
___b
O
0.2
0.7
3.4
8.4
4. 5
a
a
0.6
a
a
a
a
a
a
a
a
a
a
a
a
Z*
99.+
98.+
100.
100.
99.+
99.+
98.
98.
100.
98.
99.+
92.
85.
82.
96.
82.
94.
97.
97.
96.
89.
M
insufficient sample for oxygen determination
"sulfur under 0. 5%, insufficient sample for more precise determination
recovered; although in most cases, each percentile of the paraffins oc-
curred a few degrees above the total solubles. Differences occurred,
however, in total boiling range relative to the original amount of solubles
present due to differences in recovery (solubility and boiling range) be-
tween samples.
The appearances of the GC output from several calibration and
sample runs are shown in Appendix A, Figures A-15 through A-21.
Figures A-15 and A-16 show calibrations of the system with pure normal
paraffins, producing strong peaks with baseline separation. Figure A-17
shows one of the fuels used for testing (fuel A), indicating strong presence
of normal paraffins but also a "hump" consisting primarily of overlapping
aromatics, isoparaffins, and cycloparaffins. Figure A-18 is the output
from an analysis of the lubricating oil used in the Detroit Diesel 6L-71T
engine (same as that used for the NTC-290), indicating few normal paraffins
and a much higher boiling range than the fuels. The single small peak
53
-------
between the calibration standards and the oil itself is n-cetane, which was
added to the oil at a concentration of about one percent for possible use as
a tracer. This peak was not readily identifiable in any of the samples of
solubles.
As an aid in the interpretation of the boiling range data, Figure
22 has been prepared with percent by weight (distilled) plotted versus
temperature and carbon number of n-paraffins. Solubles from samples
taken during operation of the 6L-71T engine, as noted in Figure 22, match
the boiling range of the lubricating oil quite closely. Solubles from the
NTC-290 engine, however, seem to be composed mainly of material which
boils above 500° C. The existing GC analysis simply cannot be extended
far enough to reach the higher-boiling materials, so it was possible to
characterize only the lighter ends of these materials. The appearance
of a typical gas chromatograph output for solubles from the 6L-71T engine
is shown as Figure A-19, and a typical output for the NTC-290 engine is
shown as Figure A-20.
Figure A-21 shows a particular sample from the NTC-290 engine
which has an anomalous peak at about 415° C. The fuel used for this test
was "B+ additive", and the engine was running at 1500 rpm and 50 percent
load. One other sample was analyzed which had been taken during a run
on the same fuel (2100 rpm and 50 percent load), and a peak occurred at
the same temperature but had only about 6 percent of the height of the
previous peak. It is planned to run a portion of these samples on a GCMS
(gas chromatograph-mass spectrometer) for further identification, but
that information is not available for this report. The first peak mentioned
also resulted in the near-vertical portion of the upper curve for solubles
from the NTC-290 engine at about 415° C in Figure 22.
To complete the presentation of data on boiling point distribution
of solubles, Table 25 presents numerical information in a fairly compre-
hensive form. These data, with additional detail as necessary to form
continuous plots, are the basis for Figure 22. Distribution of normal
paraffin boiling points is given in Figure 22, so it is not considered neces-
sary to duplicate it in Table 25.
3. Analysis for Benzo (oc) Pyrene in Organic Solubles
A number of BaP determinations were conducted on samples
from both engines and all fuels, and the results showed a considerable
amount of unexplained variability. Although various techniques have
been used for BaP measurement for a number of years, determination
of BaP in diesel particulate is still in its infancy. It is recommended,
due to unexplained variability in results and the lack of similar data in
the literature, that the BaP results be treated as preliminary and tenta-
tive. Projections of emission contributions should not be made using
54
-------
in
100 r-
90 -
80 -
70
tj 60
30
20
10
fuels (range)
4
0
u
w
Q<
so
•rt
50
40
10
• « 1
.
solubles from
particulate, 6L-71T
engine (range)
20 25
I I I I | I I I I | I
n - paraffin boiling
points by carbon
number
solubles from
'/ parti culate, NTC-290
engine (range)
lubricating oil
(both engines)
150
300
40 45
fit l | I I I I | -
200
250
300 350
Temperature, °C
400
450
500
550
400
500 600 700
Temperature, °F
800
900
1000
FIGURE 22. BOILING RANGES OF FUELS, LUBRICATING OIL, AND ORGANIC SOLUBLE
FRACTIONS OF PARTICULATE FOR BOTH ENGINES USED IN METHODOLOGY DEMONSTRATION
-------
TABLE 25. BOILING POINT DISTRIBUTIONS FOR FUELS, LUBRICATING OIL, AND
ORGANIC SOLUBLE FRACTIONS OF PARTICULATE SAMPLES
Sample type
Engine
Fuel
Condition
rpm
% load
Fuel A (straight)
Fuel B (straight)
Fuel C (straight)
Lubricating oil (straight)
Total
solubles
Paraffins
only
6L-71T
NTC-290
6L-71T
NTC-290
A
B
C
A +
B+
A
C
B+
C+
A
B
C
A+
B+
A
C
B +
C +
1600
comp.
idle
1600 .
idle
1600
1600
1600
1600
2100
2100
idle
comp.
comp.
1500
1500
2100
mixb
1500
2100
1500
mixb
1600
comp.
idle
1600
idle
1600
1600
1600
1600
2100
2100
idle
comp.
comp.
1500
1500
2100
mix
1500
1500
mi-u-b
50
var.
0
0
50
50
100
25
100
var.
var.
0
0
0
50
50
0
50
var.
0
0
50
50
100
25
100
var.
var.
0
0
0
50
0
Recovery,
%
34
64
100
99
89
100a
I00a
100a
88
100a
89
100a
100a
100a
100a
100a
100a
89
14.1
5.4
7.4
2.9
24.8
12.7
9.6
9.6
68
93
94
100a
72
100a
78
100a
100a
I00a
100a
88
86
65
4.7
4.5
6.6
1.4
5.7
7.2
c, i
Tern
0.5
148
161
191
215
322
284
278
279
319
255
282
259
276
259
281
289
287
258
364
375
397
405
343
382
341
393
316
291
298
295
309
249
292
261
277
261
297
280
285
313
374
372
409
429
381
400
4CIR
10
163
198
218
364
391
384
378
379
385
364
386
369
375
378
363
385
383
373
462
c
c
c
418
473
c
c
397
391
381
377
392
362
388
371
381
389
367
388
386
394
c
c
c
c
c
c;
c
perature in ° C at weight % off
20
173
217
228
413
409
397
391
392
403
396
399
391
393
404
381
407
401
395
c
c
c
c
465
c
c
c
414
408
398
395
410
389
409
388
394
411
389
411
403
417
c
c
c
c
c
c
c
40
195
257
237
457
436
419
413
414
429
422
425
418
414
436
407
443
428
431
c
c
c
c
__c
c
c
c
454
434
424
421
441
420
436
415
417
451
411
446
439
461
c
c
c
c
c
i;
c
60
217
284
248
490
466
441
434
436
451
453
447
449
432
470
429
484
458
465
c
c
c
c
__c
c
c
c
516
469
446
443
487
446
478
446
434
506
437
499
477
533
c
i;
c
c
c
c
c
80
238
316
261
518
511
466
456
462
497
490
489
501
458
518
451
523
502
529
c
c
c
c
__c
c
c
c
c
514
492
477
488
__c
503
460
533
467
537
536
c
c
c
c
c
c
(~
c
90
253
338
274
532
__c
491
473
487
_ _c
518
__c
527
475
536
472
537
524
--c
c
c
c
c
__c
c
c
c
(.
533
524
506
__c
516
__c
532
505
541
500
c
c
c
c
c
c
c
c
c
c
100
301
417
356
545
__c
534
522
538
_ _c
545
c
546
523
545
529
546
545
^
c
c
c
c
_ c
c
c
c
__r
__c
c
537
_ _c
545
c
546
533
545
532
c
c
c
c
c
c
c
<-"
c
value as calculated exceeded 100 percent due to baseline upset
several wamplee mixed together
out of range of analysis
56
-------
these data as basis. Thus qualified, average data are presented in Table
26 in terms of milligrams BaP per kilogram fuel consumed. Complete
data are given in Appendix G, Tables G-l and G-2, in terms of milligrams
per hour and micrograms per standard cubic meter of exhaust gas.
Several trends are apparent in the data from Table 26, perhaps
the most important being the decrease in fuel specific BaP with increasing
power (at a given speed) for the Cummins NTC-290 engine. This was a
very strong trend, but no confirmation exists for other engines at this point.
Analysis of the Cummins NTC-290 data is much more useful than analysis
of the 6L-71T data presented, because quite a bit of development in the BaP
analysis procedure was still underway when the latter engine was run. The
data also indicated higher BaP, in most cases, for samples taken during
runs on fuels B+ and C+ than for comparable samples taken during runs on
fuels B and C.
Although the comparison must be qualified heavily due to the
holes in the 6L-71T data, it appears at this point that BaP emissions
from the 6JL-71T engine were considerably higher than from the NTC-290.
The average fuel specific BaP result from composites run on the 6L-71T
was 0.68 mg/kg fuel, while that from the NTC-290 was 0.21 mg/kg fuel.
In brake specific terms, these averages of composite data are 180jWg/kW hr
(130jUg/hp hr) for the 6L-71T and 59/^g/kW hr (44jLfg/hp hr) for the NTC-
290. If the lubricating oil acts as a "sink" for BaP, both the fact that the
6L-71T's particulate seemed more oily than that from the NTC-290 and
the longer overall operating time for the 6L-71T are possible (partial)
explanations for the apparent difference between engines.
4. NMR and IR Spectra of Organic Solubles
Peaks of significant proportion, other than those denoting hydro-
carbon chains, were noted on 14 of the 93 NMR spectra. These peaks may
also have been present in other spectra, but some of the samples were
unavoidably too dilute (due to small amount of solubles) to obtain a proper
signal. The peaks noted fall generally into three categories by chemical
shift as follows:
0, ppm Proton location Grouping
H O
I II
1.8 C — C carbonyl
H
I
4.2,4.8 C—-O—C ester
7.5 R—(p—H benzene ring
57
-------
TABLE 26. AVERAGE FUEL SPECIFIC BaP RESULTS
Condition
rpm
idle
PTa
PT
PT
rated
raised
rated
com
idle
PT
PT
PT
rated
rated
rated
com
idle
PT
PT
PT
PT
rated
rated
rated
rated
rated
load, %
-
0
50
100
0
50
100
DO site
-
0
50
100
0
50
100
so site
-
0
50
75
100
0
25
50
75
100
composite
Engine
6L-71T
Fuel
A
B
C
BaP,
mg/kg fuel
0. 18
0.086
0.28
0. 17b
2.4&
5.1
0.22
0.47b
0.53b
4.9b
1.0
0.25b
0.49
0.21
4.5
0.99
1. 1
0. 62
0.018
1.2b
Fuel
A+
B+
C+
BaP,
mg/kg fuel
1.4
0.70
0. 54b
0.94b
0.69b
Engine
NTC-
290
Fuel
A
B
C
BaP,
mg/kg fuel
0.25
0.30
0.047
0.027
0.20
0.20
0.090
0.060b
0.40
0.47
0.014
0.022b
0.46
0.073
0.049
0. 19
0.30
0.52
0.014
0.018
0.087
0.082
0.034
0. 14b
Fuel
A +
B+
C+
BaP,
mg/kg fuel
0.06
0.38
__c
__c
__c
0. 13
__c
0.039b
0.83
0.52
0. 17
0. 17
1.9
0.21
0. 11
0.40b
0.78
1.2
0. 14
0. 15
0. 55
0.39
0.27
0.45b
(J\
00
amanufacturer's peak torque speed; 1600 rpm for 6L-71T and 1500 rpm for NTC-290
baverage of two or more determinations
cnot enough to measure
-------
The infrared spectra mostly contained peaks confirming the pre-
sence of saturated hydrocarbon chains. In 19 of the 94 IR spectra, a
band appeared at 1740 cm"*, indicative of an ester carbonyl. A positive
match between IR and NMR spectra occurred for eight samples, with one
other instance being probable. Indication of an ester by IR without an
NMR match creates no conflict, since NMR is not as sensitive as IR for
this grouping. No strong relationships between the occurrence of signi-
ficant peaks and operating condition, fuel, or engine were noted. Further
analysis of these data does not seem appropriate for this report due to the
small mass of usable data, but all the spectra and a copy of the analysis
summary is being supplied to EPA with this report should further work be
recommended.
59
-------
REFERENCES
1. Section 211, "Regulation of Fuels", Clean Air Act Amendments
of 1970 (P. L. 91-604) to Clean Air Act of 1969 (P. L. 88-206).
2. Habibi, K. , et al. , "Characterization and Control of Gaseous
and Particulate Exhaust Emission From Vehicles", presented
at the Air Pollution Control Association West Coast Section,
Fifth Technical Meeting, October 1970.
3. Wagman, Jack, "Recent Developments in Techniques for
Monitoring Airborne Particulate Emissions from Sources",
AIChE Symposium Series, No. 137, Volume 70, pp. 277-284.
4. Gentel, James E. et al. , "Characterization of Particulates
and Other Non-Regulated Emissions from Mobile Sources
and the Effects of Exhaust Emissions Control Devices on
These Emissions", APTD-1567, National Technical Information
Service, March 1973.
5. Federal Register, Vol. 37, No. 221 Part II, Subparts H and J,
November 15, 1972.
6. Petroleum Products Survey No. 73, U. S. Department of the
Interior, Bureau of Mines, January 1972.
7. "Tentative Method of Microanalysis for Benzo(a)Pyrene in
Airborne Particulates and Source Effluents", Health Laboratory
Sciences Supplement, Vol. 7, No. 1, January 1970, pp. 56-59.
8. Flowmeter Computation Handbook, American Society of
Mechanical Engineers, 1961.
9. Hare, C. T. and K. J. Springer, "Exhaust Emissions from
Uncontrolled Vehicles and Related Equipment Using Internal
Combustion Engines, " Final Report - Part 5, Heavy Duty
Farm, Construction, and Industrial Engines, Contract No.
EHS 70-108, Environmental Protection Agency, October 1973.
61
-------
APPENDIX A
DETAILS OF ANALYTICAL PROCEDURES
-------
TABLE A-l. MINIMUM DETECTION LIMITS FOR METALS
BY ELEMENT AND FILTER BATCH
Filter
batch(es)
1 and 2
3
4
5
6
Sample
numbers
1-38
39-54
55-76
77-94
95-112
Average
*y
Minimum detection limit, jL/g/cm
Ca
0.90
0.41
0.53
0.44
0.38
0.53
V
0.16
0.11
0.11
0.14
0.19
0.71
Mn
0.20
0.13
0.15
0.15
0.23
0.17
Ni
0.08
0.05
0.06
0.06
0.09
0.07
Cu
0.07
0.05
0.07
0.06
0.08
0.07
Filter
batch(es)
1 and 2
3
4
5
6
Sample
numbers
1-38
39-54
55-76
77-94
95-112
Average
Minimum detection limit, jL/g/cm^
Zn
0.06
0.04
0.05
0.04
0.06
0.05
Pb
0.22
0.17
0.21
0.20
0.21
0.20
Sr
0.07
0.05
0.06
0.06
0.05
0.06
Sn
0.64
0.39
0.54
0.46
0.48
0.50
Ba
1.22
1.80
1.63
1.31
1.28
1.45
A-2
-------
• 10 X to TOTIIE CENTIMETER AO IBIS
I 10 X 23 CM. MIDI !• v.f.a.
»CR CO.
ELECTRON MICROPROBE X-RAY SPECTRUM
Scanning E£ec£ton
Labo/lato/u.U, Inc..
90066.. .
X-ray Photon Energy (keV) Hu Hi plication Factor: D x 1, D x 2, H x
FIGURE A-l. X-RAY SPECTRUM OF SAMPLE STANDARD CONTAINING SEVEN METALS
-------
K-
'
; IO X 1C TO THE CENTIMETER 46 1013
i 10 X 25 CM • ••( H W.Vfc.
Kcurru. a ctyta ca.
ELECTRON MICROPROBE X-RAY SPECTRUM
IPfiJillH
.::.L-II ;^ . : , ..''.•••i i;.i..;.i
bovScan .Rateivi-
Jt j' . '!• n • •• I V I "!
X-ray Photon Energy (keV) Multiplication Factor: Dxl, D x 2, H
*
FIGURE A-2. X-RAY SPECTRUM OF A DIESEL PARTICULATE SAMPLE
-------
Method for Determination of Phenols and N-Dimethylnitrosarnine
in Particulate Matter Collected on Glass Fiber Filter
1. Cut filter in pieces approximately 5 x 40 mm and place in
200-ml round bottom distillation flask.
2. Add 70 ml of 1% H3?O4 in water.
3. Connect distillation flask to vertically mounted small diameter
(8 mm O. D.) water cooled condensing tube. This distillation set-
up is similar to a Kjeldahl distillation apparatus.
4. Place 5 ml of 50% KOH in 50 ml beaker and place beaker so that
outlet end of condenser tube is immersed in KOH solution.
5. Distill over 35 ml H2O and rinse condenser tube with 5 ml H^O.
Should now be approximately 45 ml in beaker.
6. Transfer, without rinsing, contents of beaker to 125-ml separa-
te ry funnel.
7. Add 13 gm NaCl to funnel and shake to dissolve.
8. Rinse condenser tube with 10 ml benzene and collect in 50 ml beaker.
9. Transfer benzene to separatory funnel containing distillate and shake
vigorously for 1 minute.
10. Drain aqueous phase into another 125-ml separatory funnel. Discard
benzene.
11. Add 10 ml dichloromethane (DCM) to separatory funnel containing
aqueous phase and shake vigorously for 1 minute.
»
12. Collect DCM in small vial and save.
A-5
-------
13. Add 10-ml hexane to separatory funnel and shake well.
14. Drain aqueous phase into 100-ml volumetric flask. Discard
hexane.
15. Add 1 drop Phenolphthalein Indicator Solution to aqueous phase.
16. Add concentrated P^PCXj to aqueous phase to indicator end-
point then add 2-3 drops excess t^PO^
17. Cool to room temperature and add 0.5 ml diisopropyl ether (DIE).
18. Shake vigorously for 1 minute and immediately pour into 50-ml
volumetric flask using appropriate funnel.
19. Swirl contents of stoppered flask and then allow DIE to collect on
aqueous surface in neck of flask.
20. Insert ground glass stopper, to which has been attached a short
length (60 mm) of 2-mm I. D. capillary tubing, into mating glass
joint on flask.
21. Using a syringe and needle, inject water into flask through pre--
viously inserted silicone plug in flask body, so as to force the
DIE up into the capillary tube.
22. Using a micro syringe, withdraw 5[ll of DIE and inject into gas
chromatograph for analysis of phenols.
23. The DCM previously saved is transfered to a micro concentrator
and evaporated down to 0. 5-0. 75 ml.
24. 20jUl of the concentrate DCM extract is injected into a gas
chromatograph equipped with an Electrolytic Conductivity Detector
used in the pyrolitic mode for selective detection of N-nitrosamines,
A-6
-------
CHROMA TOGRAPHIC CONDITIONS
Column:
Column Temp:
Detector:
Detector Lens:
Phenols
6' 10% OV-3 + 1% FFAP on 80-100 mesh
Gas-Chrom Q-AWDMS
1Z5°C
FID
16X
Dimethylnitrosamine
Column:
Column Temp:
Detector:
Detector Lens:
6' 10% Carbowax 1540 + 10% KOH
on 60-80 Gas-Chrom Q
1Z5°C
Electrolytic Conductivity (Pyrolytic mode)
IX
A-7
-------
•vs-n I
iT-
t
I
I
t
T- • •;-
..,...-.(_
-L . U.
-Q- --!-r-|- "T-
ftS «
.7.4jttLj
o..
••rr1
.QLH
i~i, f "t"
—^-^CD-
-r-H-
±r
—4-
SEE
T I '
^
~'O~ • I "'
. :.*_!_
i(0-..-
-|-
-r^-H-
_,_ ^ ,-*-.-- t—^ —
I • "J^C" : • • L 1
:vfclij^^p^z~ .jSa^fx-": pi- .H...
t
tt-h-
:«::
-I-U
H
..1:4.1
. . I . * CM
i L ; l . ~
_]..
— 1-—V--1 i jw
.- p -\ i- i—- .'-i —
_ET '\ I ... i • -i __
L-TF-A-I -]•---(-
k=4=
FIGURE A-3. SAMPLE STANDARD DETECTION
BY NITROSAMINE ANALYSIS SYSTEM
A-8
-------
1 ' 1 1 t 1 1
FIGURE A-4. DIESEL PARTICULATE SAMPLE ANALYZED
BY NITROSAMINE DETECTION SYSTEM
A-9
-------
i ( ( ( i i : ( l ( ( ( i f ; '
.._........j ...... ;.u.v:.,h......!. .:_v!; ._,..,!
__
«-• !-'-••• —
f-. ...... l ,
y.,_._Pkenej ?,..
r. Jii^.Q-.
•_ •!/... .ij 6. J^y /.c«a./. 1 •_ •*, V Xy/e.tio./ —ril-
T."•i :'L" ~
ii—-.kr-rhpzn
I ' . \J-t : l '
. 1^ • t . '_
.. - L
FIGURE A-5. SAMPLE STANDARD ANALYZED
BY PHENOL DETECTION SYSTEM
A-10
-------
FIGURE A-6. REAGENT BLANK ANALYZED
BY PHENOL DETECTION SYSTEM
A-ll
-------
I \
FIGURE A-7. UNUSED GLASS FIBER FILTER ANALYZED
BY PHENOL DETECTION SYSTEM
A-12
-------
1 ( (
--; ------ j.;.; | ::::^-|^:l: I -"ife-H
tf'jv.jjjn y:«^j".//ct/./*iy e. t t •—
FIGURE A-8. DIESEL PARTICULATE SAMPLE ON GLASS FIBER FILTER
ANALYZED BY PHENOL DETECTION SYSTEM
A-13
-------
* IN tf.rjb HONIVWtU. fOMT WMMIMBfOM. PA.
CHART NO. »M4N
FIGURE A-9. LATER SAMPLE STANDARD ANALYZED
BY PHENOL DETECTION SYSTEM
A-14
-------
!_£7ri"J£j" '.'l'1;"
CHAHT HO. UMN
FIGURE A-10. UNUSED FLUOROPORE FILTER ANALYZED BY PHENOL DETECTION SYSTEM
(FIRST OF TWO SAMPLES)
-------
FIGURE A-ll. UNUSED FLUOROPORE FILTER ANALYZED BY PHENOL DETECTION SYSTEM
(SECOND OF TWO SAMPLES)
-------
H| UAA. HpNtVWtU. WMT WUHlMuMH. M.
CHART NO. *M4N
FIGURE A-12. DIESEL PARTICULATE SAMPLE ON FLUOROPORE
FILTER ANALYZED BY PHENOL DETECTION SYSTEM
A-17
-------
Diesel Emissions Analysis - Benzo-a-Pyrene
Interim Report - Engine #1
Introduction
Prior to undertaking the analysis of glass fiber filters containing
deposits of diesel exhaust emissions for benzo-a-pyrene we reviewed four
1-4 1
published methods of analysis. Our choice was that of Sawicki et al.
as being the best suited for use in our laboratory. During the period of
analysis we found that Sawicki had since changed his method for extraction
of organic solubles from his original paper. A critical comparison of the
old and new methods was then made including' a comparison of precision and
accuracy of both methods. The nuclear magnetic resonance (proton) and
infrared spectra of the extracted materials were also obtained.
A-18
-------
Procedure
The method of Sawicki e£ al. was chosen as that most adaptable to our
laboratory methods. All or a known portion of the filters submitted for
analysis was carefully placed in a Soxhlet extraction thimble and inserted
in the extraction apparatus. One hundred ml of methylene chloride was then
placed in a 250 ml Erlenmeyer flask. Heat was applied via hotplate and
refluxing was allowed to occur for four hours at the rate of about 20
cycles/hour. At* the end of this time the extract was transferred to a pre-
weighed evaporating dish and the methylene chloride driven off at low heat.
When dry, the dish was re-weighed to determine the weight of extractable
organic material. This material was then redissolved in 5 ml of methylene
chloride. It was this stock solution that was used for the remainder of
the analyses.
Ten nl of solution were placed on an alumina thin layer plate
(subsequently silica gel was also used) which was developed in a 19:1
pentanc:ethyl ether solution. Markers of BaP in high concentration were
also spotted so that the position of the "unknown" spots could be readily
identified under ultraviolet light. After development, each BaP spot was
scraped and washed with 50-100 ml of ethyl ether to remove the BaP from
the substrate. The ether was then driven off under vacuum and the residue
taken up in 1 ml of sulfuric acid. Once the acid was added, the solution
was analyzed immediately by fluorescence spectrophotometry. The excitation
wavelength was 470 nm while emission was measured at 540 nm. The intensity
of the emitted light was then compared to a standard curve made from known
amounts of BaP.
A-19
-------
Several questions arose on various portions of the above procedure,
therefore, we decided to speak directly with Dr. Sawicki. Two points of
significance were determined during this conversation. The first was that
certain chlorinated compounds may suppress the fluorescence emission. The
second was that an ultrasonic extraction technique developed by his labora-
tory appeared to be much more efficient than the Soxhlet technique in
i
removing soluble organic material from the filters. We, therefore, began a
comparison program to evaluate the differences in the extraction methods.
At the same time we undertook a second program to determine the precision
and accuracy of each method. To do this, a single filter was divided
equally into quarters which were in turn divided in quarters. This allowed
duplicate analyses to be made of each quarter by each technique and each
quarter could also be spiked with a known amount of BaP for recovery studies.
The results of these analyses are shown in Table II.
After reviewing the data obtained by the above procedures it was
decided, for the sake of continuity, to continue the Soxhlet method of
extraction. Other procedural changes which will be continued through
Engine #2 are finalized as follows:
1. Extract 1/2 filter (divided into 2 equal sections for
duplicate analyses) using hexane for 4 hours in a Soxhlet
extractor.
2. Remove the hexane, except for 1-2 cc and quantitatively
transfer to a 20 ml pre-weighcd screw cap vial with repeated
hexane rinses. (This allows any glass fibers in the extract
to be removed prior to analysis of the extract.)
A-20
-------
3. Evaporate to dryness at low temperature (under N2 flow
if possible).
4. Re-weigh vials to determine the amount of organic material
extracted.
5. Add exactly 5 ml of hexane to the extract and shake
until all residue is dissolved.
6. Spot a thin layer plate (alumina or silica gel) with
20 nl of solution and develop with 19:1 hexanerethyl ether.
7. Scrape the spots due to BaP and extract with 1 ml of
hexane, quantitatively filter and evaporate to dryness.
/
8. Add 1 ml sulfuric acid to the dried residue and analyze
immediately using fluorescence measurements excitation: 470 nm
and emission: 540 nm.
9. Using the remainder of the 5 ml portion (§ 5.) again dry
and then add 1 ml CC1, and shake until dissolved.
4
10. Use appropriate portions of this solution to determine
IR and NMR absorptions.
11. Transfer solutions to tapered vials for return to sender
and further analyses.
A-21
-------
Results
The final data for Engine #1 is shown in Table I. Calculations were
based on a composite of as many external standard curves as were prepared
during the analysis of each batch of filters.
Table II and Ila show the results from our analysis of the precision and
"V
accuracy of the technique employed.
Figure I illustrates the composite standard curves obtained for each
batch of filters received while Figure II and Ila illustrate fluorescence
spectra of standards and filters.
A-22
-------
Discussion
While performing the initial analyses according to the method of Sawicki
we found several features of the protocol that were subject to question.
These were:
1. The solvent front would not travel more than about 8 cm on
the TLC plate.
2. The volume of washings and the relatively large surface
area required for completion of the extractions seemed to be a
possible source of contamination and/or loss of BaP.
3. The application of a vacuum for removal of the ether may
cause sublimation of the BaP. There is also occasional
contamination from ether extraction of the stopper in the flask.
4. The volatility of the methylene chloride is such that
pipetting is made difficult.
5. The completeness of the Soxhlet technique on the filters
is not fully described.
We, therefore attempted several changes in protocol. Initially, we
changed extraction solvent from methylene chloride to carbon tetrachloride,
a less volatile solvent, to make pipetting easier. However, we found that
the fluorescence emission was suppressed. We also attempted transfers in
A-23
-------
a cold room to decrease volatility. Prior to attempting further changes
in protocol we contacted Dr. Sawicki for his comments and advice.
At this time we learned that chlorinated compounds suppress the fluorescence
of BaP, that Dr. Sawicki is currently promoting ultrasonic extraction for
removal of organic materials from glass fiber filters, and that silica
gel could be used in place of alumina for TLC.
X
Changing the TLC substrate to silica gel and the eluting agent from
pentane to hexane permitted the solvent front to move the full length of
the plate thus allowing the BaP spot to fully develop rather than remain
in the solvent front as before. The use of hexane as a solvent allowed
the removal of all chlorinated compounds from the system and thereby any
possible suppression of the fluorescence. Also, continuous shaking in a
small volume of solvent and subsequent filtering of the substrate allows
the BaP to be isolated without the use of large surface areas or vacuum.
All of the above factors were considered with the purpose of reducing
surface areas and solvent volumes to minimize the chances of contamination
and/or loss of the BaP sample.
It has been decided to continue with the Soxhlet extraction technique
for two reasons. The ultrasonic technique is highly abrasive in that
ground glass is used to shred the filters and we feel that the apparent
increased recovery of solubles may be due to factors other than normal
chemical processes thereby changing the standard definition of "soluble".
The other reason is that we have not found the precision of recovery to be
any better for the ultrasonic than for the Soxhlet method. In fact the
most recent recovery (for an Engine S2 filter and not shown here) has a
A-24
-------
coefficient of variation of 4% on extractables from 4 separate quadrants
(quadruplicate analysis) using the Soxhlet method.
All filters have not been as uniform in their coatings as the one
just mentioned, however. Through handling and storage notable variation
in particulate density have been noted in several filters. This too may be
one of the possible causes of generally poor precision in the overall
analytical scheme.
High pressure liquid chromatography was briefly investigated as a
possible alternative to the above analytical sequence. Although this
technique may be promising we had difficulty in both separation of BaP
from co-elutors and a too high detection limit. Since our time was
limited we did not pursue this method any further.
The decreasing slopes of the standard curves over the several months
of these tests has caused us concern. The reason for this change became
obvious as tho fluorescence spectrophotomcter we were using had a complete
breakdown just prior to the writing of this report. The circuit balance
(dark current adjust) has been steadily drifting since the start of this
project. Upon completion of repairs we expect that the gain and thus
sample peak heights will return to those obtained initially.
It should be pointed out, however, that standard curves were run
with each butch of filters. Therefore, the data obtained, even though
the instrument, characteristics varied, is consistent between batches.
A-25
-------
Conclusions
The data reported is reproducible to better than ± 50%, and generally
better than 10% for the BaP determinations using the Soxhlet technique.
However, there are many factors, some of which have been corrected, that
contribute to overall variations in the analysis. The most important of
these are the amount of glassware used and the occasional irregularity in
the particulate material deposited on the glass fiber filter. Recovery
experiments at this time show poor (10-50%) accuracy. It is felt, however,
that a misinterpretation has been made and that further study is necessary.
A-26
-------
Future Work
During the week of February 24 the DuPont GC - mass spectrometer
system 21-490B being housed at the Foundation will go on line. As soon
as possible thereafter we plan to use the chemical ionization system to
monitor our organic extracts. If, as we hope good separation and sensi-
tivity is obtained we will quantitate the BaP in organic extracts using
this system. This will eliminate almost totally the need for multiple
transfers of extract solution and bring to a minimum the chance of BaP
loss or sample contamination.
A-27
-------
Bibliography
1. Sawicki, E., Corey, R. C., Dooley, A. E. et^ ajU: Tentative method of
microanalysis for benzo(a)pyrene in airborne particulates and source
effluents. Health Lab. Sci. £(Suppl. l):56-59, Jan. 1970.
2. Sawicki, E., Corey, R. C., Dooley, A. E. et aK: Tentative method
of analysis of polynuclear aromatic hydrocarbon content of atmospheric
particulate matter. Health Lab. Sci. 7_(Suppl. l):31-44, Jan. 1970.
3. Sawicki, E., Corey, R. C., Dooley, A. E. et_ a^.: Tentative method of
routine analysis for polynuclear aromatic hydrocarbon content of
atmospheric particulate matter. Health Lab. Sci. 7_(Suppl. l):45-55,
Jan. 1970.
4. Brown, R. A., Searl, T. D. et. al_.: Final Report CRC-APRAC Project
CAPE-12-68, Esso Research and Engineering Co., 52 pp., 1971.
5. Golden, C. and Sawicki, E.: Ultrasonic extraction of total particulate
aromatic hydrocarbons (TpAH) from airborne particles at room temperature.
Int. J. Environ. Anal. Chem. in press, 1975.
A-28
-------
DIESEL EMISSIONS STUDY
Table I
Benzo(a)Pyrene Anal/sis
vO
Fuel
B
B
B
B
C
C
C
A
A
A
B
B
B
B
B
C
C
C
C
C
C
C
C
C
C
A+
DII-2
DII-2
DII-2
Condition
2100-100%
1600-50%
Idle
Composite
Composite
Idle
1600-50%
Idle
1600-50%
Conposite
Composite
Conposite
Idle
1600-0%
1200-100%
Composite
Composite
1600-0%
1600-100%
2100-0%
2100-50%
2100-100%
Idle
1600-50%
1600-50%
Idle
1600-50%
Composite
Composite
Sample No.
X
X
GS-1
GS-2
GS-3
GS-4
GS-5
GS-6
GS-7
GS-8
GS-9
GS-10
GS-11
GS-12
GS-15
GS-14
GS-15
GS-16
GS-1 7
GS-18
GS-19
GS-20
GS-21
GS-22
GS-23
GS-24
GS-25
GS-26
GS-27
GS-28
GS-29
Filter No.
AR- 7
AR-30
AR-36
AR-43
AR-46
AR-54
AR-73
AR-77
AR-79
AR-92
AR-93
AR-94
AR-56
AR-57
AR-5S
AR-59
AR-60
AR-74
AR-75
AR-78
AR-SO
AR-81
AR-S2
AR-S3
AR-84
AR-S5
AR-36
AR-95
AR-96
AR-97
AR-9S
P articulate
Keight (g)
0.2605
0.2171
0.2037
0.2319
0.1425
0.1106
0.0979
0.0759
0.2043
0.0645
0.1629
0.1013
0.1073
0.1148
0.0501
0.0704
0.1797
0.0989
0.1025
0.1224
0.1208
0.0965
0.1525
0.1727
0.0515
0.1809
0.1627
0.0569
0.1718
0.1425
0.0921
Extract
Weight, Cg)
0.1508
0.1363
0.1051
0.1165
0.0413
0.0487
0.0587
0.0315
0.1459
0.0234
0.1170
0.0632
0.0621
0.0652
0.0230
0.0341
0.0210
0.0655
0.0670
0.0789
0.0385
0.0549
0.1162
0.1163
0.0236
0.0280
0.1149
0.0305
0.0323
0.0975
0.0654
Wt. % Extract
in Particulates
57.9
62.8
51.6
50.2
29.0
44.0
60.0
41.5
71.4
36.3
71.8
62.4
57.9
56.8
45. 9
48.4
11.7
66.2
65.4
64.5
31.9
56.9
76.2
67.3
45.8
15.5
70.6
53.6
18.8
68.4
71.0
Total
BaP Cfg)
1.8
0.5
15.5
14.2
2.4
13.9
4.7
2.5
S.4
2.7
4.5
2.8
19.8
20.2
13.0
51.2
42.4
0.7
84.6
36.5
13.0
77.7
40.4
1.2
43.4
30.6
4.2
15.0
37.5
29.6
4.0
Wt. % BaP
in Particulates
.0007
.0002
.0066
.0061
.0017
.0126
.0048
.0033
.0041
.0042
.0028
.0028
.0184
.0176
.0259
.0727
.0236
.0007
.0825
.0298
.0108
.0805
.0265
.0007
.0843
.0169
.0026
.0264
.0218
.0208
.0043
Wt. % BaP
in Extract
.0012
.0004
.0128
.0122
.0058
.0285
.0080
.0079
.0058
.0115
.0058
.0044
.0319
.0310
.0565
.1501
.2019
.0011
.1263
.0463
.0338
.1415
.0348
.0010
.1839
.1093
.0037
.0492
.1161
.0304
.0061
Sample
Batch Xo.
1
1
1
1
1
1
2
2
2
2
2
2
3
3
5
3
3
3
3
3
3
3
3
5
3
5
3
3
3
3
3
-------
Table I (Continued)
Fuel
C
C
C
C
C
C
C
B+
5+
C*
C+
B
B
A
A
Condition
2100-50%
1600-75%
2100-100%
1600-75%
2100-25%
2100-75%
Idle
Coaposite
Coaposite
Composite
Composite
Coaposite
Cocposite
1600-100%
Composite
Sample No. Filter No.
GS-30
GS-31
GS-32
GS-33
GS-34
GS-35
GS-36
GS-37
GS-38
GS-39
GS-40
GS-41
GS-42
GS-43
GS-44
AR-69
AR-S8
AR-72
AR-65
AR-68
AR-71
AR-76
AR-106
AR-107
AR-121
AR-122
AR-130
AR-133
AR-143
AR-147
Particulate
Weight (g)
0.1384
0.2156
0.1433
0.0967
0.0424
0.1493
0.0519
0.1758
0.2153
0.1379
0.1435
0.1257
0.1237
0.1577
0.0825
Extract Wt. % Extract Total
Weight (g) in Particulates BaP (;ig)
Used for high pressure liquid
0.04644*
0.02352*
0.10839*
0.01735*
0.05208
0.04060
0.01316
0.02791
. 0.03956
0.05392
0.04892
0.03924
47.5
54.6
72.6
54.3
29.6
18.9
9.5
19.4
31.5
43.6
31.0
47.6
17.5
20.8
28.0
58.5
19.1
30. 4
22.2
14. S
7.6
4.6
19.9
4.7
Wt. % BaP Wt. % BaP
in Particulates in Extract
chromatographic
.0181
.0491
.0188
.1207
.0109
.0141
.0161
.0103
.0060
.0037
.0126
.0057
analyses
.0377
.0898
.0258
.2219
; .0367
.0749
.1687
.0530
.0192
.0085
.0407
.0120
Sample
Batch No.
4
4
4
5
5
5
5
5
5
5
5
5
5
5
5
•Sum of extract weight from 3 quadrants x 4_ to obtain whole filter extract weight.
3
-------
Table II
Comparison of BaP Analysis between
Soxhlet and Ultrasonic Extraction Processes
Wt. Solubles Wt. BaP
Filter* BaP Extracted (jug) Wt. Extracted (g) "Wt. Extract
65 A-l 3.35 0.01190 281.5
A-2 1.75 0.01382 126.6
B-l 2.06 0.05409 38.1
B-2 3.84 0.02488 154.2
68 A-l 3.55 0.00743 417.8
A-2 3.25 0.00433 750.6
B-l 2.71 0.01809 149.6
B-2 1.33 0.00989 134.8
71 A-l 3.45 0.02913 118.4
A-2 3.20 0.03036 105.4
B-l 2.26 0.04698 48.1
B-2 0.93 0.03425 28.6
76 A-l 2.85 0.00499 571.1
A-2 2.50 0.00601 416.0
B-l 1.86 0.02178 85.3
B-2 3.11 0.01641 189.4
Results reflect analysis normalized to 1/4 filter.
*A-Soxhlet
B-Ultrasonic
A-31
-------
Table Ha
Precision of Duplicate Analyses
Filter No.
Soxhlet
65
68
71
76
Ultrasonic 65
68
71
76
BaP (1/4 filter)
Mean Precision (%)
Extractables (1/4 filter)
2.55
3.40
3.32
2.67
2.95
2.02
1.60
2.48
40.5
5.7
4.9
8.4
38.9
44.1
53.8
32.5
Mean
.01286
.00588
.02974
.00549
.03948
.01398
.04061
.01909
Precision
9.6
34.0
2.7
12.0
47.7
37.8
20.2
18.2
('")
Accuracy of the Techniques
At this time values appear to be between 10 to 50% recovery. We feel
that these estimates are in error, however, further review of the data is
necessary.
A-32
-------
Figure I
Linear Regression Data for Standard Curves
Number of Correlation Probability
Batch No. Data Points Coefficient Intercept* Slope** of Randomness
1 4
2 23
3 8
4 4
5 17
0.9987
0.9836
0.9978
0.9907
0.9244
-2.65
-4.39
+3.40
+4.00
-1.35
6860
5502
5546
3885
850
<1:1000
<1:1000
<1:1000
<1:1000
<1:1000
* Peak heights measured in ram,
** Variations due to change in instrumental parameters.
A-33
-------
-------
Jprr U-^ttTi
. .aTt.h-.jt ft*L js
N5\
J O-» • i i L
zti^zzr r "••"•
.... ...71-—j
\ I
—1««.
.
o
\ _l—.-
-------
M i ; < f f M
• .„ . ^ i : ! i i 1 ! 1
O . O i f i
: . • ' i i
. • .. _ i ; • ! i '
; ;- 7 • ; f\/Q'J 3j£^ ~" T i
^ : ^"\ ^^"/^ ^ ' ^? f- / '. f~~ /"*" ' '
! •»«'//C'^"* C^-if^-rJ • _? :
i • . * - ' ' '
>v f <-r» : m '. / _f f~ / \J- jf- / //. f** 1 JL 1
1 o ' o • 'W*'/ &j*rj i'v {~£~£if& . > . i t i „ « i' t
i ! . ; j J I l/\ .£> ^\ i
' L ' ' ' '• 1 ! t*3 r \ O *'
: J i : !. : , ' r " 1 ^>~
• i ! ! H* ! £ i •
.' . »,. ,. ' | ! I lA • k. " i
"i ' 1 1 1 *? K r P '
: i j i i j C3-^ \Al) M>
r-% j^. 1 ! i ' t
- - - ci o 1 .._• : . ; : - ; . 6<5 0
; • ' - • i ! • • i , . \
i - - ! - ! j ' ! ! "
ii> i I ' ! i i ' [T '.
ON ' i 1 ! -. i . ! ! ^ :
O O 1 ; . : i i Si
' i : : I i u
• . ' ! ! ' ' i I — — Ki '
- - -: - - - ..- ! - 1 i . ; • ! w^, \l
i ^^~~ : j ^-L_ i i i^^x" i ^v.
. - I j^,^1 • ; -f • ;- • j ! • !• —
•'QUO/1*? ' •• ! ! ' :
».— v, . i j ! i .
; o ^v^AXV^^TN :!!•':•
. •- . ^ fVv \ \ ; : ; ; j |
da^' ! . - '; i ! • i i.
C^* > J ! ^1 j
\* ^^ " :ll t ! r\r^^} ' ' ""^
*y^ TC"* 0^ \ 1 £"":i * •" i .C/^5^: -t —
S" **•' t*^"cn'i ' 1 ' ' t ""
* - 4 I \d> 1 L ! ' ' .*S" "~"N ' '
1 I ' V " "*" 1 i ' S*~*~*m\S^*^r*^ '^ ' I
V* f -* ' I ' sv * * ' -^^' : * • ! • '
""s-*^^ • C^i **'! 1 '/I * ' •
• O^i i "*"* cOi I ' ' ' '
t ^^^ i *O* • -^i lr^ • i I j i '
: T^ ' '«^ • ' ' ! SO 1
'-r ^0 * ** '• • £— * i — . • C- :
1 1^ I ^ i *y ^O* | ^* ' 'Z. • '
i ^ 1? 1 -Np!< " V5^ • "f !
i •* ! : l^( ?^ 1
1 *- • ~ i ! ! i J i
l|! 1 1
i ! i i !
. -'. 1. '. !. ,! ! • i1 i : ,! ,
!
«
J
i
.. .
L
t
~"i
i
«
j
i y» *
*vj^
7"^
;^J '
Lr
.. L .
1
(
;
i
. .. — .
!
i j
i ; ' i :'
• ! ; i
1 1 7 , j
! ' \ i
1 | . ___
| ! | t i
i : I • ! !
1 i
i ! *
: ! i i
i ! i !
- : • i ' ,
i
I , _ , . j . .... .
l j ! !
. ! „ . . . ,
1 > i
1 1 ! '
1 ' !
• 1 ! j
1 ! '
i .
: 1 .. '
!
i j
1 1
i
! i i
i ' ':
1
1 i
I
i i
i i
l (
_^_j , — ^ 1 j
n, i-i 1 ^'"- 1 t '
UOLCJ N ^-;
vl to V ! > ~j
\ s^s; ^ , '
! •> i t
! i
i ! \
• ' ! i
• ! :
l ' 1
r "~ir. ] i
• \ i i i
vri i ,
i - ! i
!-»
i JA
^ \ •• \ \
f '
... - . }.... j. . . -u-.-i —
T?
, r1, i
' i i
i * i
: '
' ' , - -
...... .
T ' I • i
;• -— i , - j r __ _ _ _
• i - .:
i ^ ... j. T .... .
1 • -•
1 . : . . _
! i "- j ;.:..!_.
----;• ,
j ; • i _
1 ' f '
' 1 !..._.._. ... 1 _
1
: i IT j— »
i ' ! .;.._• J _
! ' i •, l ^
| : ; r
!(:!'•
J „.; J .... . -.--
i • 1 1 - ' - _ . ^>>
' ....!_..; ._ .. : ... x>
----••-,-- , • . >
; i ' : ^^
! i ' ' ' ' • ~
{ i . ' r*
! 1 ' - ' J,
1 "• " ! vfr
! ' . : , ; V
\s^*"^ • » ^^~*^^ ^
; . ' . .. • >»^_ ~^^
! i : t ' —"~~^
j : 1
: . I . I -
• ! ' 1 -- ;
• I i
1 : ' '
1
"— i"" 1 !
1 : ; i , , : . :
, ' <-. i i
-T- \ \r. i : j
1 i — i
1 JL ' ' ;
! "^ i I ' '
1 --J -i- ! i :
i i j : : j
i ' 1
I * . t
i ' i
i , .
1 ; : j| i
J I '
^- "?'/•• !
C" ' .v -yf— •
.. i- :•? !..: j/4. ..• .. ..t .-: - -'
'! ' 1 Vi -
I f
-------
64 MCTONS 6O
1 ! '. ' i' '..' i'.'!.' i'.'.>-
...... i • • i • ,• - - ' ''I'!'
3000
WAVE-NUMBER (CM-1)
1400 1200
WAVENUMOER (CM ')
4 1? «o
6S /5-
CRI'j'1; . . _. ... . :•
SOLVENT &«!.., CW*M^
rrvjcajriJATiON
CELL PATH .. __..__.
REFERENCE. ... ...-._
REMARKS
•
SCAN SPEED M
SLIT . . 4
PEBKIN ELMER
PART NO. 473-5089
OPERATO1?
DATE ^00^./<}>y
REF. No.
-- -o
------ o- - -.. o- -
- o- <> o o- - - -o- -
FIGURE A-13. INFRARED SPECTRUM OF ORGANIC SOLUBLE FRACTION OF A DIESEL PARTICULATE SMAPLE
-------
K
I
|
00
L0
" " " | " " ' '
e : • ¥
9 .....
9 . .
J
. . •-...'.
• .
-
. .
-
.
...
.......
-—.-.._-.!—.
.. .
,
. . . 1
M
I
r . . . .
. .
. • 1 '
•
*" i
*
. . . : .
- i - ,
1 '
.
• •
'
4
!
7
A
1 * * "
«.
1
i i •*
: • J ..
'• t
\
• • } •
1 ~
i
:
,
•
i
i
A
• i •
9 •
lo ;
- :
p i
>. i .
„
i .
• ! -
i
'
\
i
i
i
.
',
i
,
i * •
'
..
•
i
4
•
'
i
.__
4J
•
»
,
-
"
1
^—\
...
— ^
•
r —
V
•
i
•_,
•
•
•
"*
1
•j
1
.
1
:
! 1
' ;
-
i
c 1
; j
• !
; j
• 'i'
ij ]
/
i
1
•1
j
3.
u
--
»
:"•
u
1 •
j
1
1
J_
•
:
1 .
.
}~
7*"
|
u
a
.
„
-f-
1
'
I
-1-
•
r
-A.
1
i
i
:
j
'I'
1
'
_•
;
«
4-
!
t
1
I
1
I
!
:
"*"
•
!
i
i
I
1
i
»•
T
i.
t
...
..;.
-4-
T
*
1
!'
i.
•i
!
•
._.
A.
.L
'
'
i
'
*
.-
«j
0
'
-j-
'
;
B
•
t
•_
i
,
--
t .
i
'
T
i
.1
l
i
.;.
4
1
t
1
..
(
i
•
!
.
D
0
a
-
-
i
u
a
'
:
j
(-
,
i
j.
~; "
:
!
•
i-
*
.__
i ,
B
t
•
!-
;
.1
i
•
f
:
•
i
1
,
l_
•
t
,
,
'
1
:
*
i
*
a
i
,
i
_•«
•
—
^-t
(
U
J>
-;
*ri
*
>
• «
•-
-
So
•
&
-•
i
1
N
-
1
.
.:.J
_i_
I
1
I
* )j
1
/ '
u
I
ifl
-
— —
1
.
.
!
1
(...
1"
1
I/
L •/
\J
- —
u
A
1
:
'".
/
>
,i
i
I
-
L
1
-
'•^
f
V
.
1-
1 -
1
1
J-
1
,
1
.
1
••
...
1
i
,
-
•
•
•
.
i
—
i f
i—
i
i
*. i
LI
•
•
••-•
-r
.
^_
i
,
~
-
.
•
.
—
a
1 PfM (I
J •
| " *"
r i ,
D J-
1
' '.-
. ]..
1
1
— •—
•-t-
• ••«-
. i
: l •
"1 •
-. 1-
... .
j
I .
•
rw
i
•OMMiNMt
IMCTIUM NO. ..._
tS-f
cD O J
.
•
(2.ff i$
mill lANOWlOTH.....^ Hi
t f. raft j.fi •«
IWWP WIDTH: ft??.. Hi
tWCt? OfKOi C1 Mi
MTKUi AMPl
UMJUUdi
*
^^^^ INftTAUHIMT OlVIVIOft
CHAIT I-4GA
FIGURE A-14. NUCLEAR MAGNETIC RESONANCE SPECTRUM OF THE ORGANIC
SOLUBLE FRACTION OF A DIESEL PARTICULATE SAMPLE
-------
U.S. ARMY FUELS AND LUBRICANTS RESEARCH LABORATORY
Procedure for Saturates in Fuels
Scope
This method covers the determination of hydrocarbon saturates
in samples having boiling points from 450°F to 1000°F.
Summary of Method
An accurate amount (0.03 to 0.1 g) of sample is diluted with
a solvent containing 5% wt internal standard and analyzed by
gas chromatography before and after saturate fraction iso-
lation on an activated silica gel column. The amount and
boiling point distribution of the hydrocarbons and the
saturates in the sample are calculated.
Apparatus
Gas chromatograph equipped to meet conditions in Table I and
apparatus for ASTM D1319-70.
Materials
1. Solvents: Pure n-hexane
Pure cylohexane
Pure 3-methylpentane
Pure n-nonane
Pure 4-methylnonane
Pure n-decane
Pure 2-methyldecane
Pure n-undecane
2. Column: Dexsil 300
Chromosorb P, AW, 45/60 mesh.
Procedure
In a weighted container containing a (0.03 to 0.1 g) amount
of sample (Ws), add 0.6 g (0.8 ml) of solvent (60% wt
cyclohexane, 17.5% wt n-hexane, 17.5% wt 3-methylpentane)
containing a known percentage (5% wt) of internal standard
(n-nonane, 4-methylnonane, n-decane, 2-methyldecane, n-
undecane). Shake sample and analyze under conditions given
in Table I for sample. Transfer 0.6-0.7 ml of diluted sample
to column prepared for ASTM D1319 analysis (Standard Method
of Test for Hydrocarbon Types in Liquid Petroleum Products
by Fluorescent'Indicator Adsorption). Column is prepared
with 3/4 inch of glass hair in end. Allow 90% of saturate
fraction to collect in 2 ml viel, cap, and analyze in same
A-39
-------
manner as for sample in Table I. Determine the boiling
point distribution over the range 450-1000°F using simulated
distillation software for the sample portion of the chroma-
togram. Calibration standard components are in Table 2.
Calculations
Using the chromatogram areas defined below, calculate the
amount of hydrocarbons and amount of saturates in the sample.
6
Chromatogram 1
Chromatogram 2
Ws -
Wis =
A
B
C
D
% Hyd =
% Sat =
S
Weight of sample
Weight of Internal Standard
Area of Internal Standard from Chromatogram 1
Area of sample from Chromatogram 1
Area of Internal Standard from Chromatogram 2
Area of sample from Chromatogram 2
Percent hydrocarbons in sample
Percent saturates in sample
Solvent
% Hyd = [(W±s £)/(Ws)](100)
nv
% Sat = [(% Hyd) 5^ ]
(1)
(2)
Example: Saturate sample obtained from a diesel fuel
Ws "
Wis =
A
rj __
C
D
0.0463 g
0.03122 g
27,232,804 mv-sec
39,351,823 mv-sec
31,678,205 mv-sec
45,332,543 mv-sec
A-40
-------
0 03122 (39341823)
*272328Q4J
070463
% Hyd =97.4
_ 27232804 _
K " 31678205 "
% Sat = 97.45 () 0.85967
% Sat =96.5
Note: This sample was obtained by HPLC and known to be approx-
imately 99% pure saturaten from a diesel fuel.
A-41
-------
TABLE 1
Gas Chromatograph Operating Conditions -
Procedure for Saturates in Fuels
Gas Chromatograph
Column: Dual 4 ft x 1/8" O.D. (0.086" I.D.), 10%
Dexsil 300 on chromasorb P, AW, 45/60 mesh
Detector: Dual Flame lonization Detector 240 ml air, 40
ml hydrogen, 60 ml helium (column plus auxiliary)
Temperatures:
Inlet: Water cooled septum inlet
Oven: 50°C to 390°C, 16°C/min, hold 4 min.
Detector: 400°C
Carrier Gas Flow Rate: 25 ml/min helium
Injection: ly on column
Integrator
Integrate
Output
Electrometer
Baseline Shift
Digital Baseline Corrector
Peak Corrector
Slope Sensitivity
Peak width/sec
Filtering
Area Reject
Teletype hard copy
Punched paper tape
Calibration
Standard*
Automatic
Automatic
peaks_Q
256x10 *AFS
ON,50yV/peak
Auto, max rate
Level
10
10
10
100
ON
ON
Samples
ON
12 sec intervals
256xlO-10AFS
OFF
manual
Level
10
10
10
100
ON
ON at 11 min.
A-42
-------
TABLE 2
Calibration Standard - Procedure for
Saturates in Fuels
n-saturate
Component Boiling Point
Carbon Number v% (or weight) °C °F
5 10.8 97 36
6 2.7 156 69
7 5.4 209 98
8 5.4 258 126
9 10.8 303 151
10 5.4 345 174
11 5.4 385 196
12 21.6 421 216
14 10.8 488 253
15 5.4 519 271
16 10.8 548 287
17 5.4 576 302
18 1.8g 602 317
20 1.8g 651 344
24 l.lg 736 391
28 0.7g 809 432
32 0.7g 874 468
36 0.7g 928 498
40 0.4g 977 525
A-43
-------
Report
Report the percent and boiling point distribution of saturates
and hydrocarbons in the sample.
Note: Precision and accuracy of this procedure are best in
the BP range 450-900°F using an adequate sample sizeg
(minimal dilution) and a detector sensitivity of 10
AFS.
A-44
-------
01
FIGURE A-15. SAMPLE STANDARD CALIBRATION OF DEXSIL COLUMN USED FOR PARAFFIN ANALYSIS
-------
FIGURE A-16. SAMPLE STANDARD CALIBRATION OF SE-30 COLUMN USED FOR PARAFFIN ANALYSIS
-------
FIGURE A-17. ANALYSIS OF FUEL A ON SE-30 COLUMN
-------
I
^
00
FIGURE A-18. ANALYSIS OF LUBRICATING OIL ON DEXSIL COLUMN
-------
I
rf>>
NO
FIGURE A-19. ANALYSIS OF THE ORGANIC SOLUBLE FRACTION OF A TYPICAL DIESEL
PARTICULATE SAMPLE FROM THE 6L-71T ENGINE
-------
Ul
o
FIGURE A-20. ANALYSIS OF THE ORGANIC SOLUBLE FRACTION
OF A TYPICAL DIESEL PARTICULATE SAMPLE FROM THE NTC-290 ENGINE
-------
>
Ul
jit; in ii!
•: $;•• I:!! :i!l: i |! i"i!i!i lit .):i!:.i
• I •.•,'*• '.It. •<>
i^'ii'i^^rT^ -4-M—rr
iiil
4i**l- 1- *»- W *..' *1»
•''.tnii'T^i 1TH i!F'i|ii^;<:P^tin:'!-'"ri;'i •'•"•: T""^'
r ''!!!i ::' Ml! .!!|!l'!'!'-i •••.:.'::|.: •!.':•!•• i •: • ::
^rvi-^-^'i-^fiiJrBni^^
juu3uJ93uuoou3oaoo43aoooi;jj jwJujvu
j j u J j auu , . .
FIGURE A-21. ANALYSIS OF THE ORGANIC SOLUBLE FRACTION
OF A DIESEL PARTICULATE SAMPLE FROM THE NTC-290 ENGINE
SHOWING AN UNIDENTIFIED PEAK AT ABOUT 415° C
-------
APPENDIX B
CALIBRATION AND CALCULATIONS ON TUNNEL
AND SAMPLING DEVICES
-------
: i j : i*j
,^ .
TO i eo
jp 100 i no ! Uo : »ac
^ _... | __, __|__.__.
J*.
_.i
B-2
-------
l\--2>7lb- ooi
. 1 / Ho
PUUIEV 3k
BU>tU£.R- PUUl&y ^^
I^
"""
BLOUJEK- WLE1
TAK.&ETT
-15
75"
7S
voo
too
10 0
IIJT
I2S
I2C
1ST)
is»
1ST)
7ST
7T
(00
Ift*
i?r
U7
' itrvvP. , • F
AC,TUA<_
75
74
74-
»o9
> 1 1
,1,
1*0
,v
1ST)
I5T>
1^2.
T)
7t
lOb
no
»*i
BMHOtR-
A^^^D
,7,
IS
-zo
17-
»S
70
IT
IS
-zo
tT-
,r
10
»Z
»c
|S
TO
»7-
TO
BLocoEC. LAM* wfl
*tv/Uv Q,o»ocoB.e..
lfci« 520.
11, £8 50>.
»U»% 1 4J>.
IU34- S4B.
1
ir *-^ i 5"i7
»OtO S13.
>l»34- ' S toitfc.
0.9741
0.974-U
0.9151
0-974^
0.9148
00753
0.9148
0-91SI
0.91SU
0.97S-I
0.975^
0.3758
0-974U
o.9i4g
0.97STI
0,9751 :
0.974ft
0,«)1S3
EUEAM
T CJOR.R-.
0-9634
o.<)834
0,9^34-
0.b7U
O.fcTfct
O.BtBt
o.&n?
o.&is?
0.0107
0.1U7S
0.7V3Z
0.75ft?
0 *9 pO"Z-
0* / wO^-
0 * v * w W
0 • D V 0 V
O.bll?
j
O* U 1 v 1
PAJT
Q, CJORJL.
498,
488-
413,
4C.&.
45J.
4-43. ,
449.
44-1.
4^-7.
437.
4^7.
413.
494,
467.
4C.O.
44-t.
450.
42>0.
BL5WEK.
V, CW
0.2^4-
o.%\7
0-315-
0.3^.2
0.32.0
0.3IS
0. 32.0
0. 3 1 Q
0. 3(4-
0. 32.Z
0. B\ 9
0. 3 IS
0.'2>2.2.
0.32Q
O.^iO
0,317
0-3Z\
..91*17 -
ATm.oSPHEe.UL-
B-3
-------
\\-'b~l\b-ooi
' /lfe/74-
6LOO)Eft- IWLE1
-TAK.&C-T
75
75"
7S
(00
\ 00
V 00
MS
I2S"
tXS"
ISO
ISO
ISO
7S
75-
«»0
\00
|2S
12.5
lE-rnP. , • p
AC,TUA(_
-?(.
78
7t
97
91
,(,
H9
lift
U9
_ J.13
I4Z
139
7S
7S"
98
91
113
tl8
fcuotut^
AK^rt«.o
»0
IS
\0
IS ^
zo
10
IS
zo
10
IS-
10
10
IS
IS
20
10
2*
BLOtOEK- LA.IVV* WA
REV/^v Q.OMtORd.
1334 42£.
\330 414.
I
I^L J ^C ^t ft ft
3 ^9 O / w •
.j
»332 42.8.
i
132-C i 415.
1337 458.
1332 447.
I-328 432.
H,3t 48\.
1-532 457.
132-7 440.
)33(. 4\9.
1332 ! 407.
• 33 1 i 42,7.
1
'i'Z.S 410.
»33S 1 4-57.
!
»"bX7 4-5,2..
fl FLO^)
l> CDR-fc.
0.91C.4
o.9i_y
fti^uT
0.97U9
°.9!7J_
0. 9 1 tt
0.9171
O.M7C
0.97fr9
0.9TTJ
0.970t
0.9U9
0.91fc9_
"
0.9171
o.nfc*
0,9114-
ECE»WI
T CJOR.R-.
0-9602,
0.972)7
0.9H9
0.9IS&
0.9»29
0,6477
0,b411
0(&4S2
0.7fc32
0,7ft9B
0,199.0
0-9BG.7
0.9&fc7
0,9129
0.91SB
0 . 8477
0.&42£
=/ur
Q , CO R.IL .
405. j
394,
_»Ak_
3B3.
370,
379.
370.
3S?,
3S9,
3S3.
344. J
404.
391,
161-
*fc7.
390.
3S(p.
-2-9.4Z.
B-4
-------
"T
5
.- fc.
...
_. .
.. .
- •
....
5
£ 6
^t
K
5 6
*
i
]
AT
— -r
_ . -
i
. •
i o
i
i
SAMP
w
1
J
'
,
ST/VJIOM
1
l
!
l
! . .:....
•
- — . .
i
i
1
1
i
-------
PROFILE-
C*C(\«S
-HV^+V 4^
t
&.G
&.»
7.t
(o.C
s.t
4.t
3.C.
2,C.
1.1
o.(,
-0.4
-1.4
-i-4
-^ A.
A«*T
-4.4
-S.4
-t.4
!
-7/V
-7.%
V, «*
Ru*l
s.4
s.ss
fc.1
&.2>
t.4
R«*l
o.b(»7
o.«»l
0.913
1. OVI
1.02-1
t.04-3
\.0f2-
I.OIJ
o-m
o.^^s
1,001
0.971
0-H3
n.iTJ
0.95*
0.9Ui
O.B99
O.ftJI
0.8SJ
'/v/t
R^i
0-915
0,9(03
1. 012.
i.oto
1.040
1.051
—
0.992,
—
1.001
—
o.ni
0,9iV
o.bJt
o.fl%
o.o48
^y^.
O.b9l
0-917
0.99U
I.03&
1-054-
1.041.
1.057
1.019
0-9H
0.955
1.003
0-971
0-973
0. 1~lH
•7 n
0.953
o.<)ti
0-b97
0.%93
0.8&3
CwcU«J
i.r+ nj>
i<-rT -OT
t
&,(.
8.1
7.1
(..(.
£.t,
4.C
^.U
^.k
M
O.L
-0.4
-1.4
-2 A
-5.4
-45
-S.4
-C.4
-7.^
-7.B
V,
fc»«2.
5.45"
7.05
7.4
5.9
a.9
9.0
95
>.^
9.L
9.4
ie.0
(O.I
10. J_
9-8
\t>.S
lo.t
<0.4
10.?
10,0
*V*u
Run. 4
t,l
b.S"
9-0
U
M.
—
(0.0
-1 —
9.9
\0.0
10. "S
10.-J,
10,4
9.8
10. b
,
RMM£
4.0
S.4
S.B
S,U
&,4
S^
. —
5,5
—
5.$
•
(..0
&,9_^
—
t.\
S-B
Sr.3.
Run r
O.SS&
O.TXl-
O.TS&
0-912.
0,
o.9(,^
0-984-
0-9tJ
1,02.5"
I.04S
1.056
i Art A.
1, OUT1
I,07t
\.045
I-OU
l/07(.
J.OZS
V/,
Euv> 4
O.OU
0.%S2
0,>»Z
0.9C1
0.912
i.OOZ
—
o.39i
—
l,oo^
—
1.02,1
,
i.oit
—
»-04^-
0-9&Z
1.001-
*/t
K«K£
O.L«7
o,9^»
0-9*£
0.9tl
0.9^8
O.S9S
0-9*5
• —
I.OI4
l.OJI
i,oi4-
— •
I.04&
o.<)97
0.9II
Av^.
0.&I9
0.&i4
o,ft(,6
0.94?
o.9i4
0.912.
0.9S&,
0.9U"i
0.974-
o.9fi
1-014-
I.04S-
l,03>9
1 . ftf\4-
t.04l
I.OO-&
l.o£2
\.otg
0.979
16.
9.96
-ft
"7 -
B-6
-------
- OO
U
». -
JU4t
PL *• Q -
(..I
3.1
J. 03^
_M>14
1.043
t.ov?
.SJ,_
4-.C
z.U
0,913
~\A
-2A
OA11
•**_ ______________ _ J.OJ) ___ _
-4.4
p.933
-4.4-
I ,
X = O.IO
O = 0.9911
x * o. 10
, (o 4-
o , oo 53 9 (.
_i __i_o? JLV zs..
Q - o.)9 4 — o.01ozs »
B-7
-------
•Z-/IZ./74-
_
( flfrv^v _j44*ar
N/b- "*>
TOP
8.85
e.t
7.L
(..(.
J'L
4.t
i.l
I.I
A't_
0
_0
*•..*".
0.92-7
0.9U
\.OS4
1.
1.000
TUfcZl
5.7994-
4.1)74
±±'0-
0
-i
•TO
-£
-+.V&S-
-\t*
•>*)¥)-+-*T«-
«,Z?ftS—l-Ott
-0.1^
0,S
4^*-
-M>
>-.--vv%-
^s-e-
hrO^*7
S8.1814-
6.6S
7-8
O
0.&S3
'
Ml 9
1.003
t.OOO
5.74*5
5 St
-G-t*0&
r^-19?--
-ZrrftM^-
0, ft 601
-o.wofe—
^rO
-IrO-
—Cr«—
—lr«—
--I.O
-OA
-3 .
J^»»4J
-*r*M5-
».) = O.Biy
Vr
6.85
ft. I
0
O.lslJ
8.BS
k'liSJ:
(p.5%8
0
r
7.5331
H
4-1 1
-VrO-
-1 (t
t.osz
-?.
-Vr^VJJ-
(,.1570
Irft-
5.4
1-04S
-4.05-
e.4-
-ir«-
-»T» 1
~»r&-
-4rO«90
-l.ASift
0.53/1-
0.922
0.9SL
-4-^7158-
4.SB04-
5*970
h^4-
l.o«4
-- -J.-T!
1.035.
—U«—
I.C.
l-SSftt
O.S7.7B.
0
4rO(r&f-
0.4-
1.04S
1.014-
0.40 St,
r-4 (M>&H
-4,4.
0,983
B-8
-------
. 4
rtr
.1-1
"t
~T
*r=
41)
RfT
:in±
lili
a'-H- ^+—r -i-f»-
^^^. ^~*~.-. ^-Y-> ^-
tt
BEUHO
T-r-tr:
4t T*M?
Dll
-
vmo*Jl
. AVEIJSO-fe
....--(••--..
-—I:
.v.
frffn
• : I
NS
! I I
.:_|.__;._.j__
!' f I I
I L_! i l
rrrfT
-------
I I I
iB'
ri<
•T-t
li
•-' I - — i
-.-±r -.m.
&
-D
t~"T~* T
4r
fE
7^
rrS
it:
-H-1--
il'
1o
^U
a »o r rnoc
rir
tow
B-10
-------
pLOlOMETB*. READING-.! * IpOU-
T.j_L -i
B-ll
-------
R-ATE.
PfLOBE.
LOUVHOAJ
V\l -VOL Cft**tO
4 * 47 uo*vA
EfcC <*~lvfc/l
N0.
I
Z.
3
4-
2
3
4-
5
fc
TV p
I.D., i-v
5,^(U^^
o.soo
0.5^00
o.s^i
o.&ol
O.ZS3S
0.2.00S
O.SOSS"
o.\sio
o. looS
AreeA,^^
9.40
o.i9U
o,\9t
o.t?i
o,\97
0.0505
o.o^tU
0.012JJ
0. OIT9
0.0019^
Alt 6 A , W
O.OC.S7
0. 00 I3>(.
o. ooisc.
O.OOIS7
0.00\37
O.OOOiSO
O.OOQ2.19
O.OOOS02
O. 000 124-
o.ooooss-i
M= U.3
'• V,,^ i.i\. V.
,io
M
•••Msb =
W.7SV?»P«» «i/ U P L
a /IWliA '•*
\s
"
,4ltf, j *:^
40 M tofc
)V«b *c
PROBE r4«.
1 '
2
5
4-
T\p r.D. ,,-n
o.soo
O.^>0
0. £0\
o.soi
oo
O.OO»S|
O.OOlS t
O.OOIS7.
O.OOIS2
B-12
-------
11 -37*6-00 I
7/18/74-
SAMPLER.
ISOM/OE.TIC,
>. I(V
.02.89
UBfc
2.
3
4-
0.
O.J03S
0-lSIO
O.IOOS
0.00291
0.00(6(0
6.0010k
Aid FLOW
= t\j 69.(bQ (^l)'
T =7OF,
Kio
PACTOR.
1.044
1.0
B-13
-------
-------
D\l_\JT»OAJ -nJM/VJEU OP Evfc. A-Tt /O G-
C.T1 OAJS
H\-VOl_ (^fcXIO FILTER.)
1. SET &|ao1l (OR.IFICE IN TAILPIPE.) = 2.,
OFF AMD DVUOTtOAj TOMNEL &L-OW ER. ON
— ^«-OlTCH OFF
AS POSSIBLE.
S.TAR.T SAIVV^Lt BLOUJEje.,COOWTErt,^ TIMER. >»MUU-TA/OEOO^L.y;. CORRECT
AS
4. STOP
OF-
Sirv\OLTAAJEOU^CV AT
B.
). SET
CAL-VB P-ATlOAJ .CO.l2.yfcS FOI2.
AS
i — 4
OK )
V
2., STOP
TAICE
AS
AS
3. STAW
AS
SI *VUUT AMEOOSL.y V
USIM&. VALVES.
. c.-Tt>P PU»*P^ COOiOTER-,
ftAULTAAJ EOUS LV AT EAJ b 0^ I?-U/0
B-15
-------
PUKPS ON - DUMMY FILTER.- S.E.T FL-OU>N1ETE. R. S AT "70
_QM_
.S AMD 2,0 v»
__
2-, (LALdULATE EXHAUST MASS. FLOCO _- Me = _M
- ----- ..... ------ ------- .......... -------- ------- - - --•- -- -- - / NO.
=. .89.60
OK. R£At> Kt. -Cvow EE-C 6-KAPt\ Q ............ ________________________________
' T \S_ .EAJ&IA) E.
(2-EADOUT".)
3, CALCULATE R.AVAJ. SAMPLE 4>._-_ A>v...r_l9_. ?Q_*_Vp7.t.) Mt_* _O_R._
RE.AD A|pv £»OIM. Eg-C, 6-B-APH 1
.4, S,E1__C>iLU.T_E_. Sft.>APL_t__J:L.OWJ_^.AT1E_. .AT ._.SP._ AC.Ft^ ..OS»V /OG-
_2^._ A"T•__ Tds _((3)_ON_TV*JWEL' ^ITAU
S. CALCULATE PRIMARY Dl LOT* OM A|f> » A.
fo, SET _A>V .A.NJD
7.
4-
6. TO.R-KJ OFF.. .LC.F|V\_ SAMPLE, _.PojW p AMJD__JW^TALU
\0. S.\ MULT A/0 EOU SLY START '_ \ .. C.FIV\ SAIW P uE. PUiV\F> AAJD
RtSET M«. S FCoUiWETaR. [t ACFH) ,. A>OD M». S CAUBeATIOFJ .
.._.„„....—._--.---..— — ..^_ -. , j -OS ""~ " * ~ " ~*
AAJD R^COfcD DATA. AT. I«S)DICATE£> V/VJT^EE-V^AUS ._OM.. SAmPJUE
j
\\. tf\UT DOlON) I CFlU POrWP AMD TUMEC. vJKtti IKJWCATEP OK .I.F FILTER
B-16
-------
U-ill 6-ool
_ __.. . _ ITS
t. 2.9.0 ^ |\. fe |fV - 2-9-1 ^~ A^,
^ . /»ou»r /ift-v^loAi. TJA-jU^vtdbu/JL SO F fc Tv ^ ^00 F
3. tKC. uiC«-»-iN- ow ju-»aiA»«AAjL oQF ^ "^ &OF
4. EK.C- A«L-~v.y^6*^K
S- tXHMjST .P^PE
B-17
-------
E RC
GRAPH O
-------
ERG
GRAPH I
-------
O)
•
5
o
:s
o
_j
u_
2 iu
CL
2
to
Q
LU
20
30 40 50 60
DILUTED SAMPLE FLOW RATE, acfm
Figure 15, Diagram of the Diluted Sample
Flow Rate
B-20
-------
B-21
-------
! i
I I
1
i
!
i
|
I
i
I
1
I
. j .
i
,
i
— j-
'
I
_ ._ __
1
j
1
— j — - —
i
J
.._ . .
.
-
I
,
.
j
/
/
j
i '
/
:./
/
'—- • •
!
. _..
•
. .
— -»
'
.. ._
'
,
!
• ••" • »-
j
(
•
B-22
-------
B-23
-------
\\-V7l 6-00 \
7/1 7 h 4
2-
fe5 ------
H ___ _
? .-.--
.9ilV^: ____ 0.- J tt A-J?5.Q ______ 0 . ?.J?.0 ___________ b = _o : s 4 3 B ____ d_ ^ 0
TV-
10
^' ..........
.TL !.'*J^ ____ C
- O.S3.B2. A
**- 0.9991 ___ ^lf_A.?
T*
Z. 4- 6 10
___ __ I. .
Co.o'Hs) "
O.Z4Q _________ __ __
^- 0.999t
- o.onc.70
. K_'._ vi = o.on0,7
— pet \^-
B-24
-------
M -3718-001
_ ,^._.____——.- Ad.^—._...._^. . ._..-—.__,.. -_--^-.A--_^ — ^ _._ •*_-_-"-- -
.T~_ 7 Via. ^ ~ " TbT ~ i+~
^^<^^^m^: ™<&
-------
USED
H-37IB-OOI
TO G-EAJ&R-ATt. A^d^_N»> J^Pv _ C-UEA/E^J fk,, Ts ., ~T,u. AS P_AfcArv\ ETER.^
b^-2-9-3 %. ^^^
550 R- (W
SIO li(o(«2. O.Z9& \'(tfi2-
S-bo ».ftB2 o.^3>4- !.»»«•
Q> V 0 Z. 1 I 0 0 . 3>7 5 i' I 1 °
(D G> 0 . _ 2.143 o .4l(o 2--M5
1 \ 0 2.583 0.4SJ ^.^^
^^ Z'82-8 °'502- *'B2B
S-41
fc.12.
fc.87
7.&1
8.41
9.2^
sio
S(» 0
Co I 0
G>(pb
HO
7U
i.bsi
l.fc-M ..._
2.076 _
2.329
^.S(b7
2.. 811
510
Gl 0 2.0
(o(j>0 Z.i
B-26
-------
T OSfcD
H-V71&-001
^ SZO R. (,fcOF)
v
n) T,.
<5\ D
Z. \-Z-3
2.3^8
= 0:94 x\q )
*«.
MO
_ Aj»v.« ?. 9.0
O.I
SO- USE FIG-UP-ES FOR. TDF
•2-9- *!> l< \V.
1% (oO F £= T^ 6, 80 F ) A,w D
B-27
-------
\\-2nt6-00 \
OF A fr
y
SH \ P
- \,0
, r >».&(> / Nl.&t
vy _*-AUL«.T.l""J5*__"« AP - V^.OT- >w J ~ v^3.04 * Q.D133J = 0.9
iv... a«JA_\_.Ak..Jj.iS.B ^
; _=. _fe.A<
, .s^.Bfe r ^ N»-6fe
=. .L»3j.O^.*L). ~ (>S.04- ^ 0.147 )
i
i
j' "
B-28
-------
11-3716-
479/74^
INLET
R.01J 7-
OKI net
, owcoefc.
0.1
_o.A.
0,0
-?.'*_
1.0
2-3 S,
0.9U6
9 .\ \
\\AS_
\ 9.4ft
2-32.
0.9*40.
.o.jfcsj
0.981?
J^O^BO
19.95
318.
0.9SJ1
2.5.02.
2.5.3*-
!•>-..
(.4
1-V
Z.O
Z.S
2.0
4.0
.353..
438.
4-9 J.
0.1955
0-H70
0.96^0
0.91SZ-
0-584?
51.45
SS.70
4S4.,
_47S-.
6"00,
0.9 BS4
o.^8o^
.578.
4-1.10
0.981^
.J>'1170.
.^ZM-
ST5.0*
JZo4._
5.0
C|?
C..6
JLJL
SJ..JL8
0.371?
6-75*
93S,
O.V70S-
0.1U4-
GC..&I
o-nos
0.9600
0.9795
0.^770
0.973?
34/_
it.04-
4-0.78
_4-4,0&
4-7.(.S"
5S.87
ffft. 42.
61. OU
(ct.04-
B-29
-------
Alfl-
\i -2>7\&-oo\
4/9/74-
li 2 2.5 3 4 5 6 7 8 9 10
1.5 2 2,5 3 4 56789
V
= 2.5,7
wo
M =
'*
B-30
-------
APPENDIX C
DATA REDUCTION
-------
SAMPLER
FOR MUL.TIMOOE RONS
OPERATOR
DATE
RUN
MODE
1
2
3
4
5
6
7
8
9
10
1 1
12
13
TIME l»«e) AT
START
END
ENGINE
RPM
kOAD,
Vfc
TEMP'S., *F
TV (D
Tv
-------
TUNNE.U
FOR PlOLTlMODE RUNS
OPERATOR
/ /
RUN
MODE
I
2
3
4
5
6
7
&
9
1 O
1 i
I 2
I 3
TIME (MO AT
START
END
EIM&-INE.
RPM
LOAD,
TEflP'S. , °F
T*(§)
T.(D
T;©
-
"&«dD
T.®
0UXOER
A. |f>
pf
VAiyETOHNSCUp
UAR&E
SMALL
MO, RAW
RAM&E
CHART
NO, DILUTE
RAM&E
CHART
O
FO
RP
I
1
3
4
S
ft
7
6
»
ID
II
u
R. S1N&UE - MODE ROMS t STEADY - STAT E)
K\- %UOAO« LOAOCl^) =
TIME,
sec
TIME,
miit
TEMP'S., %P
TA(§)
T/9)
Tf©
To* (LD
T*©
BLOWER
A|»
?,
VALVE TORMSCLO.
LAR&E
SMALL
TUNNEL BLOWER COUNTS | (
SAMP LI N& TIME., S«t ^
NO..RAW
RANfrE
CKAfcT
K0« DILUTE
RANftt
CHART
ENGINE
RUTERC*)
-------
*->-W» IX t_ \J (
DATE. / /
KUN
FOR. MOLTIMODE RONS
MODE
1
2
i
4
S
6
7
8
9
1 0
1 1
I 2
1 3
TIME.C»tt} AT
START
END
ENGINE
RPM
LOAD,
%
UOAO,
»k«
INTAKE AIR
i» tfio
^e. »p
RESTRICTIONS
INTAKE
l» «iO
EXMAMSX
in Ha
FOEL,
IVVn.
FOR SINGLE-MODE RONS CSTEAOY -STATED
1
2
3
4
5
6
7
8
)
10
11
It
TIME,
itc.
TIME,
mt H
INTAKE A:IR
At\,
i» «»O
Tt,ig
RESTRICTIONS
INTAKE,
«V M*O
CXHMIVT
VK H4
FUCL,
1W
VK*-
EMG-INE
BULB •>
BUL.0 -
FOE.L,
•p
C-4
-------
6/13/74-
DATA
DATA
ONE. vACOE PER
EPUAT10NJS
"
£ =
H\-vOU ^AKPL&S ONuV
FIL.-TERE.&J. ZvCM^Ak^M^J
\
C-5
-------
FOR.
.AT ss
.1 at*.
!=»
4-"7 ~»* SAKPLJ&S QAJW
./ v .
'*v* / ~ /. \
24 "
C-6
-------
PAGE
OF 2
1
2
3
4
5
RUN
NO.
d
d
hi
d
d
6
7J8
*EQ
H
1C
[
mm/
b
l|
|/
FUE
COI
d
°HEH
dd
0
3
0
**
0
s
0
6
0
'I
V
d
I|M
T
®
T
T
11
12
«
n
15
ATE
-
L
E
d
|/
d
H
d
G
.
11
. dd
d
j
d
E| | |
"II
E] f
"II
d
d
1
d
1
<
1
d
1
dd
16
iMa
RPM
20
LI
d
\R OR G47
d
d d d
d
V21
2
d
1
2
d
!
2
d
1
2
d
1
2^22(23^425
AD
NO.1
1 G4
d
I
V
I
3
I
3
I
3
I
3
I
— I — 1 — I — I —
!fl272829p(
- R
-
PX
n
\
DRY
2
1
4
1
4
I
4 _j
1
B
4
1
H4^
35
N(
1
RTICULA1
G473
GAS METE
V22
TIME
5
6
1
ORIFICE 1
5 6
1
NTAKE Ai
5 6
-OWER IN
5 6
1
3637
38
39J40
>. 2
—
E WEIGHT
G
1
R READIN
V13
I
NMODE
7
1
EMP. BY V
7
I
\ TEMP. B>
7
1
.ET TEMP.
7
1
11)42)43
44
FILTER
4E
C
NO
I
-
74
3S
V
8
ODE
8
MODE
8 J
3Y MODE
8
1
!
<4"H
f
5DES BY
50
Pi
3
]
GERC
23
9
9
1
1
9 V
9
1
5152J53
54
55
>SITION
NO
-
1
dd
I
V14
) 11
I
) 11
I
I 11
1
) 11
1
SCJBTJSijSS
60
4
A
.
dd
d d
V24
12
12
12
12
!
61 62J63J64656e
67*6£69707l[72
NO. 5
-
T
TOTAL
CB TIME
dddddddd.d
]
13
13
13
13
73 74 75 76
GINE CC D
77 78 ?9 90
E
N
dd
I
SOUTHWEST RESEARCH
INSTITUTE
DEPARTMENT OF
EMISSIONS RESEARCH
DIESEL PARTICULATE
GRAVIMETRIC ANALYSIS
PROJECT
ENCODED
DATE ENC
B
:o
11-
-
-------
PAGE —£ OF
O
oc
1
d
2 13 l*l6
RUN
NO.
-d d d
'III
6
7
8
9
1C
SEQ
0
a|o|p|o
•HP
1
i
1
U
E
o| |D|P
1JOJN
•1
R
N
9
9
1*12
13
d
"II
L|
d
h
-
11
1
,
1
d
d
I
.
1
d
E|
d
.
d
1
d
A
d
d
.
1
dd
*T
d
leUialifiizc
2
d
I
2
d
1
2
d
1
2
d
1
2
d
1
n
7-
7,
74
h
3
3
3
3
1
3
,h44-
4 5
I I
4 5
I I
4
I
D
4
I
4
I
»1
M*
J6
ORIFICE A
6
•
ENGI
6
ENGINE,
5
I
LUTE
NOX
5
I
AW
NOXC
5
I
1'
Mp7|383940
> FOR HI-V
7
I
JEFUELR
7
I
AIRFLOW (
6
I
CONCENT
6
_L _
(1
da
4445
OL SAMPL
8
I
VTE
INLB
8
I
RIFICEAP
7
I
NATION IN
7
I
3NCENTRATION IN
6
I
P
7
I
ttM^Sllw
• UY MODE
9 1
I
T/HR BY M
9 1
I I
BY MODE
8
I
PPMBYMC
8
I
M BY MOC
8
I
515JK
)
J,
11
)DE
)
11
1
9
!
DE
9
I
E
9
1
sddd&sleo
12
II
12
1
10
I
10
I
10
I
eiUeaUU
13
1
13
1
11
I
11
I
11
I
6367)56
6S
J70
12
I
12
I
12
I
nh
fc
m
75
13
I
13
I
13
I
/^yfy^id
J
I
I
-------
SER,PERCCl. DAN M.
PERATE,P2U,T20,PP100,PL20.CMSOOOO.
10 113718001 SRIlOb? 23RATECC301380001
RUN(S)
COPYCRUNPUT, OUTPUT)
LDSET(PR£sET=ZERO)
MAP(OFF)
REDUCE.
LGO,
COPYCRdNPUT, OUTPUT)
*
NOLIST
PROGRAM RATECCUNPUT, OUTPUT, TAPEbO=INPUT)
DIMENSION TOR(13),DPE(13),TE(13),FUELU3),RNOXC13),ANOXri3)
DIMENSION TB(13),DPOR(13),TIME(i3)
DIMENSION EXH(13),AIR(13),FLOWOR(13)
DIMENSION SAMP(t),ENGCODEC2),IFLTRS(S),GJCS)
EQUIVALENCE (GJ(1) , G«f 71) , (CJ(S) ,GH72) , (GJ(3) ,Gf 73) , CGJ(««) ,
EQUIVALENCE (GJ(5),GERC)
C INITIALIZE ARRAYS TO ZERO
301 DO 1 1=1,13
TOR(I)eDPE(I)aTE(I)sFUEL{I)sO
RNOX(I)=DNOX(I)=TB(I)sDPOR«)sTIME(I) = 0
EXH(I)=AIR(I)=FLOWOR(I)=0
1 CONTINUE
C READ HEADER CARD
REAOCbO, 100) JRUN,JSEQ,JDATE,JRPM, LOAD, 1FL1RS,ENGCODE
IF(EOF,bO) 80, Z
2 PRINT EDO
REAO(bO,101)KRUN,KSEQrFUELCrGH71,Gf?2,GH73fGlt7f ,GERC,P4,CB,TOT1ME
1 ,N
PRINT 201, JRUN,JDATE,ENGCODE,FU£LC, LOAD, JRPM
C CHECK FOR FILTERS
IS=1HS
IF(IFLTRS(2).NE,<4H ) GO 70 3
GO TO S
3 IFCT=a
IF(IFLTRS(3).NE,«IH )IFCT=3
IF(IFLTRS(H).NE.«1H )IFCTs«»
IFCIFLTRS(5).EQ,t
IF(KRUN.NE.JRUN.OR.KSEQ,NE,3 )GO TO 8
IFCIS.EQ.IH ) GO TO 30b
PRINT a05,VU,V21,TOTIME,PA
C-9
-------
PRINT 20b,V12,V22
PRINT S07,V13,V23
PRINT 208, Vlf ,V2"*,CB,N
GO TO 7
30fa PRINT 218,TOTIME,PA,CB,N
C . READ REST OF DATA FOR ALL MODES
7 REAO(bO,103)KRUN,KSEQ,(TIME(I),I=lrN)
IPURUN.NE.JRUN.OR.KSEQ.NE.t) GO TO 8
REAOCbO,m)KRUN,KSEQ,
IF,TEm,FuEL
-------
PRMsPER/GERC
SUMEXHsO.
00 IS 1=1, N
SUMEXH=SUMEXH+EXHm
15 CONTINUE
PC=4.2b3E+12*GERC*N/TOTIME/SUMEXH
PCMaPC/GERC
PRINT 21*
60 TO 19
•» X *7 SYSTEM
lb V=V21-V11
IF(VfLT,0)V=V+100,
SAMP(1)=,Q753*V
IF(V.LT.O)V=V+100,
SAMp(2)=,07fa7*V
VBVB3-V13
IF(V.LTtO)V=V+100,
SAMp(3)=.07S9*V
IF(V,LT.O)V=VilOO,
SMEXTI=0.
SMTINO=0,
00 17 1=1, N
SMEXTI=SMEXT1+EXH(I)*TIME(I)
SMTINO=SMTINO*T1ME(I)*DNOX(I)/RNOX(I)
17 CONTINUE
RATIO=SMEXTI/SMTINO * bO,
PRINT 215
00 18 J=l,f
PERa GJ(J)*RATIO/SAMP(J)
PRMs PER/CJ(J)
PCe 8tb43E+b*CJCJ)*TOTIME/SMTINO/SAMP(J)
PCMsPCXGJ(J)
PRINT 21b,PERfPRW»PC,PCM
18 CONTINUE
GO TO 20
19 PRINT 21b,PER,PRM,PC,PCM
1F(IFUTRS(2) .NE. 1H ) GO TO Ifa
IFCIFLTRSC3) ,NE, 9H ) GO TO lb -
IF(IFLTRS(H) ,NEt qH ) GO TO lb
BO GO TO 301
BO STOP 100
C " INPUT FORMATS
100 FORMAT(AS,I2,A8,AH,A3,5A«1,A10,A3)
101 FORMAT(A5,I2,Ab,SF8.b,F5,2,Fb,0,Fb.l,fiX,l2)
102 FORMAT(AS,I2,8F7.3)
103 FORMAT(AS«I2,SXil3F>»,0)
101* FORMAT(AS,I2,5X,13Ff.O)
105 FORMAT(45,I2,5X,13FH,0)
IQb FORMAT(A5,I2,5X,13FH.O)
107
108
109 FORMAT(A5,I2,'*X,13F5t2)
110 FORMAT(A5,12,HX,19F5.1)
111 FORMAT(AS,I2,4X,13F5.0)
C OUTPUT FORMATS
200 FOR«AT(*1 TABLE PARTICULATE EMISSION RATE AND CONCEN
ITRAylON CALCULATIONS* )
201 FORMAT(30X,*RUN *,A5,2X,A8 /15X,*ENGINE *,A10,A3,* FUEL *,Ah,
1 * LOAD *,A3,* RPM *,A>» )
C-ll
-------
202 FORMAT(*0 FILTER*, Al, 12,IX, AS,<»(I3,1X, A«O ) I--
203 FORMAT(*0 PARTICLES*,FlO,b,HF13,b)
20* FORMATC*0 DRY GAS METER READINGS INITIAL FINAL. TOTAL T
1IME ATM PRESSURE*)
205 FORMAT(* l*,F8.3,F%3,Flf .1,F13,2)
20fa FORMATt* 2*,F8.3,F9.3)
207 FORMAT(* 3*,F8,3,F1,3, * BLOWER
1COUNT NO, OF MODES*)
208 FORMAT(* **,F8,3,FS,3,F1S,0,1101
209 FORVAT(* CARD OUT OF SEQUENCE *,AS,I3)
210 FORMAT(*0
1 TB*)
211 FORMAT(*
IPb.O )
212 FORMAT(*0
1 TB DPOR*)
217 FORMAT(*
213 FORMATC*O
21H FORMAT(*0
215 FORMAT(*0
21b FORVIAT(*0
i *
2 *
3 *
218 FORMAT(*U
10DES*/
END
TIME
OPE
TE FUEL
*,F*,0,FB,2, Fb,0, F7,l,
TIME TOR
HI»VOL*)
E R C
•» X *7
OPE
TE FUEL
F7.1, FS
RNOX
RNOX
DNOX
DNOX
.O»F7.2)
SYSTEM*)
SYSTEM*)
PARTICULATE EMISSION RATE 5* ,F12,Z/
MULTIPLIER =* ,F10fOf
PARTICULATE CONCENTRATION s* ,FlO,0/
MULTIPLIER s* ,F10-,0 ^
TOTAL TIME ATM PRESSURE BLOWER COUNT NO, OF
F17.1 »F15.2 fFlS.O ,1123
PM OAYFILE NEXT
«•
PM USE REVERSE SIDE OF 1-PART PAPER
C-12
-------
TABLE PARTICULATE EMISSION RATE AND CONCENTRATION CALCULATIONS
RUN 8-311 02/21/75
ENGINE CUMS. NTC-210 FUEL EM«?2*F LOAD 100 RPM 2100
FILTERS 1 FP47-315
PARTICLES .00247*
DRY GAS METFR READINGS
TIME
210
240
240
240
240
2HO
£10
240
240
2HO
OPE
3,80
3,80
3.80
3,80
3,80
3,80
3,80
3*80
3,80
3,80
2
IS
1
2
.3
*
TE
7*
80
8b
75
71
71
73
72
7b
80
A*7-b21
.OD247b
INITIAL
11.174
35.118
7b.527
10.3b5
FUEL
114,0
114,0
114,0
11*. 0
113,0
113,0
113.0
113,0
113,0
113,0
3 A*7-b22 4 A47-G23 5 FP47-422
,OU2**2 .002*18 .00113
FINAL
17.b73
S3.83S
15.075
1.58*
RNOX
1717.0
1717.0
181*. 0
1832.0
1841.0
1101.0
188*. 0
110*. 0
188*. 0
1108.0
TOTAL TIME
2*00.1
BLOWER COUNT
b5b87
DNOX TB
12, fa lib
108,1 123
108,3 127
107,0 127
101,0 127
107,0 125
1Gb, 3 127
100,7 125
103.5 125
10*, 2 125
ATM PRESSURE
21.38
NO. OF MODES
10
E R C SYSTEM
PARTICULATE EMISSION RATE r
MULTIPLIER =
PARTICULATE CONCENTRATION =
MULTIPLIER -
45, If
«f0332
*USbO
35bl0552
«» X 17 SYSTEM
PARTICULATE EMISSION RATE
MULTIPLIER
PARTICULATE CONCENTRATION
MULTIPLIER
PARTItULATE EMISSION RATE
MULTIPLIER
PARTICULATE CONCENTRATION
MULTIPLIER
PARTICULATE EMISSION RATE
MULTIPLIER
PARTICULATE CONCENTRATION
MULTIPLIER
PARTICULATE EMISSION RATE
MULTIPLIER
PARTICULATE CONCENTRATION
MULTIPLIER
14.2b
38100
83241
33b*f135b
91.13
37127
81181
327105*3
12.Ob
37b11
81307
332151bl
11,3b
3bG75
80b13
32302154
C-13
-------
TABLE PARTICULATE EMISSION RATE AND CONCENTRATION CALCULATIONS
RUN b-377 02/11/75
ENGINE CUMS. NTC-290 FUEL EM22tF LOAD JLOO RPM 2100
FILTER 1 AR-223
PARTICLES .1321UO
TOTAL TIME
ATM PRESSURE
29,21
TIME
2HO
2MO
2'IO
2HO
2HO
2MO
2 '10
2 '10
2HO
2HO
TOR
105
113
118
12*
127
135
1HO
1*5
1*8
150
OPE
3,30
3.bO
3.bO
3.bO
3,bO
3,bO
3.60
3,bO
3,bO
S.faU
TE
80
7S
81
80
80
81
81
79
79
78
FUEL
llb,0
112,0
112.0
112,0
112,0
112.0
112,0
112,0
112,0
112,0
BLOWER COUNT
bS705
RNOX
18H2.0
1S80.0
NO, OF MODES
10
2022.0
2032. U
2015.0
2004.0
1180. U
2032.0
DNOX
l*b.G
125,3
125,9
117,7
118,*
115,2
113,9
109,7
111,1
11*, 3
TB
133
131
119
131
131
129
131
131
130
130
DPOR
3,30
3,30
3.30
3.30
3,30
3,30
3,30
3,30
2.90
2.55
HI-VOL
PARTICULATE EMISSION RATE =
MULTIPLIER «
PARTICULATE CONCENTRATION s
MULTIPLIER =
715
8b251
b52920 m-3 (106)
C-14
-------
APPENDIX D
SMOKE AND GASEOUS EMISSIONS DATA
-------
TABLE D-l. RESULTS OF FEDERAL SMOKE TESTS ON A DETROIT,
DIESEL 6L71-T ENGINE USING SEVERAL FUELS
Fuel codes
"A" or
EM-197-F
"B" or
EM-195-F
"C" or
EM-198-F
"A + additive"
or EM-206-F
"B + additive"
or EM-207-F
"C + additive"
or EM-208-F
Fuel type
No. 1 Kerosene
Averages
No. 2 Emissions
test fuel
Averages
"No. 1-1/2" Special
blend
Averages
No. 1 + Ethyl
DII-2
Averages
No. 2 + Lubrizol
8005
Averages
"No. 1-1/2" + Lubrizol
8005
Averages
% opacity by PHS smokemeter
"A" factor
11.2
12.5
16.7
16.7
14.3
11.9
12.8
13.0
13.2
13.6
11.2
10.1
12.3
19.2
17.9
18.5
10.5
10.2
15.3
14.8
15.0
16.6
15.7
15.5
3.5
3.0
3.2
9.9
10.8
6.2
5.1
8.0
"B" factor
2.1
2.1
4.8
5.4
3.6
1.4
2.1
1.9
2.4
1.9
1.8
1.5
1.9
2.3
2.8
2.8
1.9
1.7
2.3
2.7
2.5
4.7
4.1
3.5
1.0
0.9
1.0
1.8
1.8
1.0
1.0
1.4
"C" factor
18.0
19.5
22.6
21.4
20.4
20.6
20.8
21.8
19.9
20.8
15.6
14.3
19.1
28.7
26.5
26.9
14.6
14.3
22.2
23.6
24.6
21.2
20.7
22.5
5.7
4.6
5.2
14.8
16.3
8.9
7.2
11.8
D-2
-------
TABLE D-2. RESULTS OF FEDERAL SMOKE TESTS ON A CUMMINS
NTC-290 ENGINE USING SEVERAL FUELS
Fuel codes
"A" or
EM-197F
"A + DII-2"
or EM-209-F
"B" or
EM-204-F
"B + 8005"
or EM-224-F
"C" or
EM-198-F
"C + 8005"
or EM-211-F
Fuel type
No. 1 Kerosene
Averages
No. 2 Emissions
test fuel
Averages
"No. 1-1/2" Special
blend
Averages
No. 1 + Ethyl
DII-2
Averages
No. 2 + Lubrizol
8005
Averages
"No. 1-1/2" + Lubrizol
8005
Averages
% opacity by PHS smokemeter
"A" factor
7.3
6.8
7.1
5.2
4.5
4.9
7.5
7.8
7.0
8.7
8.8
7.9
7.5
7.5
7.3
7.9
7.8
2.5
2.5
2.5
8.2
7.6
7.9
4.5
3.0
3.8
"B" factor
4.4
4.2
4.3
3.6
3.7
3.7
4.1
4.0
3.8
4.4
3.8
4.3
4.6
4.7
4.2
3.8
4.2
1.5
1.2
1.4
4.4
4.5
4.5
2.0
1.1
1.6
"C" factor
11.4
10.1
10.8
7.2
6.7
7.0
10.3
10.4
9.9
11.8
12.8
11.9
11.2
10.4
10.1
12.3
11.1
3.8
4.4
4.1
10.6
10.0
10.3
7.1
5.1
6.1
D-3
-------
TABLE D-3. GASEOUS EMISSIONS FROM A DETROIT DIESEL 6L71-T
ENGINE OPERATED ON THE 13-MODE PROCEDURE
Fuel codes
"A" or
EM-197-F
"B" or
EM-195-F
"C" or
EM-198-F
"A + additive"
or EM-206-F
"B + additive"
or EM-207-F
"C + additive"
or EM-208-F
Fuel type
No. 1 Kerosene
Averages
No. 2 Emissions
test fuel
Averages
"No. 1-1/2" Special
blend
Averages
No. 1 Kerosene +
Ethyl DII-2
Averages
No. 2 + Lubrizol
8005
Averages
"No. 1-1/2" + Lubrizol
8005
Averages
Cycle comp. emis., g/hp hr
HC
0.97
0.94
0.96
0.49a
0.44a
0.58a
0.81
0.84
0.89
0.68
0.91
0.97
1.11
1.00
1.12
1.12
1.12
1.12
1.25
1.18
1.21
1.43
1.32
CO
2.19
2. 22
2.20
2.66
2.33
2.10
2.80
2.53
2.33
2.46
2.28
2.37
2.34
2.33
2.35
2.10
2. 22
3.07
3.22
3.14
2.85
3.18
3.02
NOX
11.6
12.0
11.8
11.6
11.7
12.6
13.0
13.1
12.9
12.5
11.3
11.4
11.4
11.4
10.3
10.6
10.4
11.3
11.2
11.2
11.8
11.2
11.5
HC+NOX
12.6
13.0
12.8
12.1
12.2
13.1
13.8
13.9
13.8
13.2
12.3
12.3
12.5
12.4
11.4
11.7
11.6
12.4
12.4
12.4
13.0
12.7
12.8
lRuns conducted January 1974, all others September and October 1974
D-4
-------
TABLE D-4. GASEOUS EMISSIONS FROM A CUMMINS NTC-290
ENGINE OPERATED ON THE 13-MODE PROCEDURE
Fuel codes
"A" or
EM-197-F
"A + DII-2"
or EM-209-F
"B" or
EM-204-F
"B + 8005"
or EM-224-F
"C" or
EM-198-F
"C + 8005"
or EM-211-F
Fuel type
No. 1 Kerosene
Averages
No. 2 Emissions
test fuel
Averages
"No. 1-1/2" Special
blend
Averages
No. 1 Kerosene +
Ethyl DII-2
Averages
No. 2 + Lubrizol
8005
Averages
"No. 1-1/2" + Lubrizol
8005
Averages
Cycle comp. emis., g/hp hr
HC
0.31
0.30
0.31
0.26
0.25
0.26
0.29
0.36
0.34
0.35
0.31
0.32
0.32
0.33
0.31
0.35
0.33
0.44
0.33
0.33
0.34
0.41
0.38
0.40
0.29
0.28
0.29
0.34
0.39
0.34
0.36
CO
2.43
2.57
2.50
2.49
2.44
2.47
2.46
2.53
2.31
2.24
2.49
2.32
2.68
2.63
2.47
2.49
2.66
2.81
2.65
2.57
2.52
2.96
2.62
2.79
2.29
2.41
2.35
2.72
2.89
3.17
2.93
NOX
12.1
11.9
12.0
11.7
12.5
12.1
12.6
12.1
12.6
11.6
12.7
12.8
12.6
12.5
12.4
11.9
12.0
12.4
12.4
13.7
12.5
12.1
12.9
12.5
11.3
13.4
12.4
12.8
12.3
12.6
12.6
HC+NOX
12.4
12.2
12.3
12.0
12.7
12.4
12.9
12.5
12.9
11.9
13.0
13.1
12.9
12.8
12.8
12.2
12.3
12.8
12.7
14.1
12.8
12.5
13.3
12.9
11.6
13.7
12.7
13.2
12.7
13.0
13.0
D-5
-------
APPENDIX E
AVERAGE MASS RATE AND CONCENTRATION RESULTS
-------
TABLE E-l.
MASS RATE RESULTS OF PARTICULATE SAMPLING
ON DETROIT DIESEL 6L71-T ENGINE
Dilution Tunnel, 47mm Glass Fiber Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
5.3
26.
33.
49.
67.
65.
47.
55.
78.
84.
107.
64.
B
9.2
36.
49.
72.
111.
77.
56.
62.
94.
114.
137.
71.
C
5.1
25.
41.
59.
72.
65.
43.
51.
87.
94.
84.
65.
A+add-
itivea
2.5
27.
36.
54.
64.
59.
45.
52.
90.
91.
105.
67.
B+add-
itiveb
1.4
48.
169.
333.
145.
155.
147.
465.
300.
247.
198.
144.
C+add-
itiveb
5.9
67.
183.
136.
104.
100.
67.
116.
138.
176.
142.
73.
Dilution Tunnel, 47mm Fluoropore Filters
Grams particulate per hour by fuel
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rr>m - 75% load
A \s \S \J JL LJ HI 1 •— ' /if J.V/ drWL
lf>00 mm - 100% load
J» \J \J \J J. Lj LL1 A v \J f\j XU CUvJi
? 1 00 -rom - 0% load
L* JL V/ \J A LJ III. VJ fO ±\J Ot\+
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
A
9.1
29.
36.
56.
68.
69
** / •
64
*s * •
56.
137.
136.
203.
60.
B
13.
42.
64.
54.
83.
150.
C
8.1
24.
55.
114.
100.
78
i ^ •
116
X A w •
64.
99.
169.
138.
86.
A+add-
itivea
9.3
30.
45.
72.
84.
\s A*
70
i v •
57
«* i •
71.
111.
123.
122.
79.
B+add-
itive
1.6
42.
103.
356.
146.
161.
144.
464.
307.
252.
203.
149.
C+add-
itive
6.0
67.
197.
155.
106
x \J \J •
105
X \J -J •
80
\J\J •
135
X *J 'J •
154.
179
A 1 7 •
146.
81.
""Primary hexyl nitrate (cetane improver)
3Organo-metallic (smoke suppressant)
E-2
-------
TABLE E-l (Cont'd). MASS RATE RESULTS OF PARTICULATE
SAMPLING ON DETROIT DIESEL 6L71-T ENGINE
Dilution Tunnel, 8x10 Glass Fiber Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
3.3
32.
63.
79.
51.
83.
104.
82.
B
9.8
42.
48.
81.
87.
92.
70.
69.
114.
112.
150.
92.
C
5.5
40.
56.
67.
74.
73.
59.
49.
107.
117.
102.
76.
A+add-
... a
itive
4.3
29.
62.
71.
50.
86.
87.
76.
B+add-
itiveb
21.
71.
73.
155.
161.
148.
C+add-
itiveb
5.6
56.
74.
72.
118.'
162.
191.
104.
ERG Sampler, 47mm Fluoropore Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Comnosite
Grams particulate per hour by fuel
A
21.
42.
28.
39.
108.
B
11.
42.
122.
162.
-
C
118.
38.
47.
56.
192.
83*
A+add-
itivea
9.8
15.
31.
36.
57.
73.
39.
76.
B+add-
itiveb
10.
42.
77.
193.
82.
85.
210.
112.
79.
C+add-
itiveb
12.
10.
37.
140.
19.
49.
102.
106.
77.
Primary hexyl nitrate (cetane improver)
Organo-metallic (smoke suppressant)
E-3
-------
TABLE E-l (Cont'd.) MASS RATE RESULTS OF
PARTICULATE SAMPLING ON DETROIT DIESEL 6L-71T ENGINE
Dilution Tunnel, 293mm Fluoropore Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - " 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
6.90
57. 1
89.8
51. 8
B
11.2
33.4
66.6
128.
53.3
102.
128.
60.4
C
A + add-
itivea
B + add-
itive13
C + add-
itive15
TABLE E-2.
CONCENTRATION RESULTS OF PARTICULATE SAMPLING
ON DETROIT DIESEL 6L-71T ENGINE
Dilution Tunnel, 47mm Glass Fiber Filters
Milligrams particulate per standard cubic meter
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
A
22.
33.
38.
54.
67.
58.
43.
46.
60.
58.
69.
59.
B
41.
46.
57.
77.
106.
65.
52.
52.
71.
77.
83.
65.
C
12.
31.
48.
63.
70.
56.
39.
43.
99.
65.
52.
59.
A + add-
itivea
11.
34.
43.
57.
63.
52.
41.
44.
69.
63.
68.
61.
B + add-
itive13
5.6
52.
197.
349.
135.
129.
131.
372.
216.
164.
122.
129.
C + add-
itive13
25.
81.
206.
139.
99.
85.
60.
96.
102.
117.
89.
67.
a-Primary hexyl nitrate (cetane improver)
-metallic (smoke suppressant)
E-4
-------
TABLE E-2. (Cont'd). CONCENTRATION RESULTS OF PARTICULATE
SAMPLING ON DETROIT DIESEL 6L-71T ENGINE
Dilution Tunnel, 47mm Fluoropore Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Milligrams particulate per standard cubic meter
A
38.
37.
42.
60.
67.
62
58.
46.
105.
94.
129.
68.
B
56.
53.
75.
58.
_ _ _ _ _
63.
93.
C
18.
31.
64.
124.
96.
67.
59.
53.
75.
116.
85.
90.
A + add-
itivea
40.
37.
53.
76.
82
62.
52.
59.
85.
85.
78.
82.
B + add-
itive15
7.1
53.
118.
573.
137.
134.
129.
371.
221.
165.
124.
134.
C + add-
itive0
25.
81.
221.
158.
100.
90.
71
111.
115.
119.
91.
73.
Dilution Tunnel, 8 x 10 Glass Fiber Filters
Milligrams particulate per standard cubic meter
Operating condition
Idle
1 AOO t*rvm — O0£> Irtor3
1 AOO T-ri-m 7 f^PjL lr\arl
1600 rpm - 50% load
1600 rpm - 100% load
9 1 nO i-rvm — nP//» l^ttrl
7100 rnm - ?5*& Inarl
2100 rpm - 50% load
21 00 rnm - 759k InaH
2100 rpm - 100% load
Composite
A
19.
tyJ .
69.
71.
47
63.
66.
67.
B
47.
55
62
91.
88
73.
64
/TI
90.
80
94.
83.
C
30.
52
66
74.
74
64.
55
41
82.
82
64.
72.
A + add-
itivea
20.
36
67.
65.
46
67.
56.
69.
B + add-
itive
101.
76.
64.
117.
99.
132.
C + add-
itive13
30.
7?
80.
63.
1 O7
122.
120.
94.
aPrimary hexyl nitrate (cetane improver)
kOrgano-metallic (smoke suppressant)
E-5
-------
TABLE E-2 (Cont'd). CONCENTRATION RESULTS OF PARTICULATE
SAMPLING ON DETROIT DIESEL 6L-71T ENGINE
ERG Sampler, 47mm Fluoropore Filters
Operating condition
Idle
1600 rpm - 0% load
1600 rpm - 25% load
1600 rpm - 50% load
1600 rpm - 75% load
1600 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Milligrams particulate per standard cubic meter
A
5.9
27.
46.
82.
25.
30.
70.
B
10.
53.
143.
174.
C
44.
47.
54.
60.
185.
58.
30.
27.
32.
57.
66.
A + add-
itive
31.
19.
36.
38.
56.
65.
18.
34.
30.
53.
43.
B +' add-
itive
41.
53.
89.
202.
69.
58.
40.
168.
96.
64.
37.
C + add-
... b
itive
33.
12.
21.
37.
40.
112.
16.
36.
66.
62.
63.
Dilution Tunnel, 293mm Fluoropore Filters
Milligrams particulate per standard cubic meter
Operating Condition
Idle
1600 rpm - 25% load
1600 rpm - 50% load
1 Ann v«rfm 7 ^Q7/\ "\r\Sif\
1 Ann *»rvrvi 1 nn0£« 1r\9/1
O 1 An *»i-k-ry-i nP/£» ls\arl
O 1 nn >-I-M-»-» 7 C^O^ Irt9r3
7 1 nn T"r\rT» Rn*7 loa.H
2100 rpm - 75% load
2100 rpm - 100% load
Composite "^
A
14.
63.
_____
59.
49.
B
22.
42
73.
m-
48
77
81.
64.
C
A + add-
itivea
'
B + add
itiveb
C + add-
itive15
aPrimary hexyl nitrate (cetane improver)
kOrgano-metallic (smoke suppressant)
E-6
-------
TABLE E-3. MASS RATE RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
Dilution Tunnel, 47mm Glass Fiber Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
0.78
6.6
15.
19.
24.
22.
15.
38.
40.
39.
46.
20.
B
3.8
12.
25.
37.
37.
42.
26.
65.
64.
68.
69.
35.
C
1.9
9.
18.
27.
24.
26.
23.
46.
50.
42.
52.
26.
A + add-
itivea
0.60
7.5
16.
20.
22.
22.
19.
38.
32.
37.
52.
20.
B + add-
itive)3
5.0
20.
31.
48.
56.
73.
35.
79.
65.
79.
92.
44.
C + add-
itive13
3.5
14.
28.
40.
49.
53.
29.
59.
54.
66.
74.
29.
Dilution Tunnel, 47mm Fluoropore Filters
Grams particulate per hour by fuel
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
A
0.82
7.7
17.
18.
21.
24.
24.
37.
43.
41.
50.
26.
B
4.1
12.
28.
40.
38.
45.
24.
6B.
66.
70.
70.
37.
C
2.6
9.1
18.
28.
24.
24.
23.
48.
48.
42.
46.
29.
A + add-
itivea
0.76
7.4
16.
21.
22.
20.
19.
37.
34.
38.
54.
22.
B + add-
itive13
5.3
21.
31.
46.
55.
75.
37.
79.
67.
80.
94.
50.
C + add-
itive13
2.9
15.
28.
40.
50.
44.
30.
60.
55.
53.
76.
40.
aPrimary hexyl nitrate (cetane improver)
^Organo-metallic (smoke suppressant)
E-7
-------
TABLE E-3 (Cont'd. ) MASS RATE RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
Dilution Tunnel, 8 x 10 Glass Fiber Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load'
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
1.7
10.
23.
29.
25.
.32.
25.
48.
50.
44.
52.
32.
B
5.3
17.
30.
46.
45.
55.
32.
43.
56.
74.
85.
47.
C
3.6
16.
27.
28.
29.
34.
34.
54.
59.
46.
56.
33.
A + add-
itive a
2.2
13.
20.
23.
22.
28.
26.
48.
47.
38.
48.
29.
B + add-
itive
5.7
19.
35.
48.
56.
69.
38.
55.
70.
83.
94.
51.
C + add-
itive13
3.3
17.
28.
38.
47.
52.
34.
51.
59.
65.
90.
40
ERC Sampler, 47mm Fluoropore Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
1.9
5.5
13.
9.8
7.7
18.
22.
B
3.1
4.8
8.9
24.
35.
12.
35.
54.
C
4.5
4.8
9.6
7.8
11.
25.
33.
A + add-
itivea
2.6
1.4
8.8
8.8
11.
16.
21.
B + add-
itive0 .
0.83
6.9
22.
25.
14.
33.
46.
C + add-
itive0
0.58
7.6
17.
19.
21.
41.
48.
Primary hexyl nitrate (cetane improver)
kOrgano-metallic (smoke suppressant)
E-8
-------
TABLE E-3 (Cont'd. ) MASS RATE RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
Dilution Tunnel, 293mm Fluoropore Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 178% load
2100 rpm - 100% load
Composite
Grams particulate per hour by fuel
A
B
3.9
15.
37.
44.
63.
64.
34.
C
1.7
27.
31.
51.
32.
A + add-
itivea
1.1
21.
27.
40.
24.
B + add-
... b
itive
C + add-
itive13
TABLE E-4. CONCENTRATION RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
Dilution Tunnel, 47mm Glass Fiber Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 7 5%' load
2100 rpm - 100% load
Composite
Milligrams particulate per standard cubic meter
A
3.5
12.
27.
30.
37.
29.
21.
47.
45.
41.
41.
28.
B
17.
23.
43.
60.
54.
55.
35.
74.
70.
68.
61.
48.
C
8.1
17.
33.
44.
35.
34.
32.
57.
54.
42.
47.
36.
A + add-
itivea
2.5
14.
28.
33.
34.
29.
27.
48.
36.
39.
48.
27.
B + add-
itive13
23.
37.
55.
77.
86.
95.
48.
98.
73.
79.
81.
61.
C + add-
itive13
15.
26.
48.
64.
72.
70.
40.
72.
59.
67.
67.
39.
aPrimary hexyl nitrate (cetane improver)
"Organo-metallic (smoke suppressant)
E-9
-------
TABLE E-4 (Cont'd.) CONCENTRATION RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
Dilution Tunnel, 47mm Fluoropore Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 1QX)% load
Composite
Milligrams particulate per standard cubic meter
A
3.7
14.
30.
30.
32.
32.
32.
46.
48.
43.
45.
36.
B
18.
22.
48.
64.
55.
59.
33.
77.
72.
70.
62.
51.
C
11.
17.
32.
46.
36.
31.
32.
60.
52.
42.
41.
39.
A + add-
itivea
3.2
13.
29.
34.
33.
27.
27.
46.
38.
40.
50.
31.
B + add-
itive
24.
39.
55.
74.
84.
98.
51.
97.
75.
80.
83.
69.
C + add-
itive
13.
28.
49.
65.
74.
58.
41.
75.
61.
54.
69.
55.
Dilution Tunnel, 8 x 10 Glass Fiber Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
Composite
Milligrams particulate per standard cubic meter
A
7.5
19.
41.
47.
37.
44.
35.
60.
57.
45.
47.
45.
B
24.
33.
54.
75.
67.
74.
45.
54.
64.
76.
77.
66.
C
15.
29.
48.
46.
42.
45.
46.
67.
64.
47.
49.
45.
A + add-
itivea
9.5
24.
35.
37.
33.
38.
35.
59.
54.
39.
44.
40.
B + add-
itive15
26.
36.
62.
78.
83.
92.
53.
68.
79.
84.
86.
71.
C + add-
itive15
15.
32.
49.
61.
67.
75.
47.
63.
66.
64.
66.
54.
aPrimary hexyl nitrate (cetane improver)
^Organo-metallic (smoke suppressant)
E-10
-------
TABLE E-4,(Cont'd). CONCENTRATION RESULTS OF PARTICULATE
SAMPLING ON CUMMINS NTC-290 ENGINE
ERC Sampler, 47mm Fluoropore Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% load
(~*. rmnnnsi f f>
Milligrams particulate per standard cubic meter
A
8.0
10.
21.
13.
11.
20.
20.
B
14.
9.
15.
39.
46.
17.
38.
48.
C
19.
8.7
16.
10.
15.
27.
29.
A + add-
itivea
11.
2.5
14.
12.
15.
19.
19.
B + add-
itive
3.7
13.
35.
32.
20.
36.
41.
C + add-
• j_- b
itive
2.5
14.
27.
25.
28.
45.
44.
Dilution Tunnel, 293mm Fluoropore Filters
Operating condition
Idle
1500 rpm - 0% load
1500 rpm - 25% load
1500 rpm - 50% load
1500 rpm - 75% load
1500 rpm - 100% load
2100 rpm - 0% load
2100 rpm - 25% load
2100 rpm - 50% load
2100 rpm - 75% load
2100 rpm - 100% -load
Composite
Milligrams particulate per standard cubic meter
A
B
18.
28.
59.
58.
-
71.
58.
47.
C
7.3
45.
42.
45.
43.
A + add-
itivea
4.7
34.
36.
37.
33.
B + add-
itiveb
C + add-
itive13
aPrimary hexyl nitrate (cetane improver)
bOrgano-metallic (smoke suppressant)
E-ll
-------
APPENDIX F
TRACE METALS IN PARTICULATE
-------
TABLE F-l. ANALYSIS FOR METALS IN PARTICULATE SAMPLES
FROM FOUR FUELS, DETROIT DIESEL 6L71-T ENGINE
Fuel
Operating
condition
"A" or
EM-197-F
"B" or
EM-195-F
"C" or
EM-198-F
"A+additive"
or EM-209-F
Idle
1600- 0%
1600- 25%
1600- 50%
1600- 75%
1600- 100%
2100- 0%
2100- 25%
2100- 50%
2100- 75%
2100- 100%
Composite
Idle
1600- 0%
1600- 50%
2100- 0%
2100- 50%
2100- 100%
Idle
1600 - 0%
1600- 50%
1600 - 100%
2100- 0%
2100- 50%
2100-100%
Composite
Idle
1600 - 0%
1600- 50%
1600-100%
2100- 0%
2100- 50%
2100- 100%
Composite
^
Metal analysis in/^g/cm on filter
Ca
1.35
tra
2.32
1.28
tra
2.41
2.09
1.16
tra
1.30
3.10
1.13
V
Mn
tra
Ni
_ m, _ _
Cu
tra
tra
tra
0.12
0.07
tra
0.09
tra
0.07
0.07
Zn
1.80
0.46
0.84
tra
2.07
1.04
1.60
0.66
1.06
2.50
0.50
1.05
0.70
0.75
0.23
1.13
0.46
0.84
0.53
1.46
0.58
0.78
0.33
4.02
2.40
3.58
0.74
1.70
0.50
0.23
0.30
1.04
0.80
0.75
0.15
0.35
tra
Pb
0.37
tra
0.38
tra
tra
tra
0.33
tra
0.28
0.28
0.48
0.22
Sr
0.07
tra
Sn
Ba
a"trace" means an amount less than the minimum detectable limit
but more than 2/3 of that limit
F-2
-------
TABLE F-l (Cont'd). ANALYSIS FOR METALS IN PARTICULATE SAMPLES
FROM TWO FUELS, DETROIT DIESEL 6L71-T ENGINE
Fuel
Operating
condition
"B+additive"
or EM-210-F
"C+additive"
or EM-211-F
Idle
1600- 0%
1600- 50%
1600-100%
2100- 0%
2100- 50%
2100-100%
Composite
Idle
1600- 50%
2100-100%
Composite
Metal analysis in/^g/cm^ on filter
Ca
2.25
1.85
1.66
33.28
40.84
30.55
11.41
25.66
41.16
23.49
18.17
3.70
16.08
52.57
23.79
71 77
V
0.21
0.27
0.14
0.41
0.42
Mn,
Ni
Cu
Zn
0.07
0.38
0.47
0.29
0.57
0.24
0.20
0.45
0.55
0.45
0.20
0.19
1.40
1.44
0.46
0.4.Q
Pb
tra
0.23
0.23
0.25
0.34
0.19
tra
0.33
tra
Sr
tra
0. 12
0.13
0.09
0.05
0.11
0.15
0. 10
0.07
tra
0.07
0.14
0.09
fK OR
Sn
_ _ _ _
Ba
tra
2.66
4.79
6.38
3.72
tra
2.82
4.26
4.26
4.63
3.30
7.66
4.49
^ 7A
"trace" means an amount less than the minimum detection limit
but more than 2/3 of that limit
F-3
-------
TABLE F-2. ANALYSIS FOR METALS IN PARTICULATE SAMPLES
FROM FOUR FUELS, CUMMINS NTC-290 ENGINE
Fuel
"A" or
EM-197-F
"B" or
EM-204-F
"C" or
EM-198-F
"A+additive"
or EM-209-F
Operating
condition
Idle
1500 - 0%
1500 - 50%
1500 - 100%
2100 0%
2100 - 50%
2100 - 100%
Composite
Idle
1500 - 0%
1500 - 25%
1500 - 50%
1500 - 75%
1500 - 100%
2100 - 0%
2100 - 25%
2100 - 50%
2100 - 75%
2100 - 100%
Composite
Idle
1500 - 0%
1500 - 50%
1500 - 100%
2100 - 0%
2100 - 50%
2100 - 100%
Composite
Idle
1500 - 0%
1500 - 50%
1500 - 100%
2100 - 0%
2100 - 50%
2100 - 100%
Composite
Metal analysis injUg/cm on filter
Ca
0.65
1.38
1.61
1.53
tra
tra
tra
1.19
tra
V
Mn
tra
Ni
Cu
tra
tra
tra
0.10
0.07
Zn
tra
1.89
1.29
tra
0.93
1.06
0.07
0.05
0.27
0.10
0. 25
tra
0.88
0.13
0.08
0.10
0.07
0. 12
0.07
0.06
0.20
0.13
0.34
0.08
tra
0.15
0.22
tra
0.12
Pb
tra
0.22
tra
0.21
tra
0.04
0.33
tra
0.52
0.47
0.44
0.44
0.31
0.36
0.33
0.72
0.61
tra
0.34
0.21
0.29
0.25
0. 52
tra
tra
0.24
tra
0.32
0.30
Sr
tra
tra
0.06
0.10
0.06
0.07
tra
tra
0.08
tra
0.06
tra
tra
tra
Sn
Ba
a "Trace" means an amount less than the minimum detection limit, but more than 2/3
of that limit
F-4
-------
TABLE F-2 (Cont'd). ANALYSIS FOR METALS IN PARTICULATE SAMPLES
FROM TWO FUELS, CUMMINS NTC-290 ENGINE
Fuel
"B+additive"
or EM-224-F
"C+additive"
or EM-211-F
Operating
condition
Idle
1500 - 0%
1500 - 50%
1500 - 100%
2100 - 0%
2100 - 50%
2100 - 100%
Composite
Idle
1500 - 0%
1500 - 50%
1500 - 100%
2100 - 0%
2100 - 50%
2100 - 100%
Composite
o
Metal analysis in/^g/cm on filter
Ca
8.13
12.13
40.56
45.19
14.92
30.51
38.07
9.06
22.49
8.89
11.29
44.16
35.79
18.94
3.86
58.58
30.58
31. QR
V
Mn
tra
0.26
tra
tra
0.28
tra
tra
Ni
Cu
0.12
0.11
0.12
tra
0.12
0.17
0.14
0.16
0.18
0.11
0.12
0.12
0.17
0.18
0.18
0.11
0.15
n 1 e;
Zn
0.28
0.17
0.37
0.20
0.46
0.53
0.10
0.43
0.47
tra
0.40
0.37
0.31
0.25
0.32
0.19
0.13
n i R
Pb
0.29
0.34
0.31
tra
0.27
tra
0.42
0.26
0.28
«-^a
Sr
0.07
0.07
0.13
0.15
0.07
0.14
0.14
0.07
0.11
0.05
tra
0.15
0.18
0.08
0.12
0.19
0.14
n i 7
Sn
Ba
1.35
2.31
6.74
7.58
2.61
4.81
5.49
tra
4.11
1.42
2.16
7.46
4.91
3.65
7.18
8.94
5.50
Ci Cl
"Trace" means an amount less than the minimum detection limit, but more than 2/3
of that limit
F-5
-------
APPENDIX G
SUMMARY OF BaP RESULTS
-------
TABLE G-l. SUMMARY OF BaP RESULTS, DETROIT DIESEL 6L-71T ENGINE
Fuel
B
C
A
B+
C+
A+
Engine rpm
and load
Idle
1600 - 0
1600 - 50
2100 - 100
Composite
Idle
1600 - 0
1600 - 50
1600 - 75
1600 - 100
2100 - 0
2100 - 25
2100 - 50
.2100 - 75
2100 - 100
Composite
Idle
1600 - 50
1600 - 100
Composite
Composite
Composite
Idle
1600 - 50
Composite
mg/hr
1
0.139
28.5
4.17
9.92
10.3
0.135
11.1
3.02
14.1
7.94
37.1
16.5
26.6
22.4
0.623
3.52
0.165
1.62
9.93
2.28
14.4
16.0
1.13
13.5
16.2
2
2.77
35.0
15.2
2.86
10.2
.
0.545
3.56
23.0
11.2
3.18
3
15.4
8.32
1.44
65.3
4
5.98
5
4.00
Avg.
1.45
28.5
4.17
22.5
10.2
3.77
11.1
4.89
14.1
7.94
37.1
16.5
26.6
22.4
0.823
23.1
0.165
1.62
9.93
2.92
18.7
13.6
1.13
13.5
9.69
tfK/m*
1
0.739
36.4-
4.81
6.23
9.66
0.755
14.7
3.37
14.0
7.09
33.6
14.4
20.1
16.0
0.512
3.30
0.898
1.79
8.96
2.11
12.9
14/5
5.29
14.7
14.8
2
12.8
21.9
14.0
16.2
11.2
0.510
3.29
20.7
10.0
2.91
3
14.2
38.3
\
1.56
61.8
4
5.40
5
3.50
Avg.
6.77
36.4
4.81
14.1
9.36
18.4'
14.7
5.38
14.0
7.09
33.6
14.4
20.1
16.0
0.512
21.9
0.898
1.79
8.96
2.70
16.8
12.2
5.29
14.7
8.86
9
-------
TABLE G-2. SUMMARY OF BaP RESULTS. CUMMINS NTC-290 ENGINE
Engine rpm
and load
Idle
1500 - 0-
1500 - 50
1500 - 100
2100 - 0
2100 - 50
2100 - 100
Composite
Idle
1500 - 0
1500 - 50
1500 - 100
2100 - 0
2100 - 50
2100 - 100
Composite
Idle
1500 - 0
1500 - 50
1500 - 100
2100 - 0
2100 - 50
2100 - 100
Composite
Fuel
B
A
C
mg/hr
1
0.55
2.75
*0.29
1.13
4.49
2.17
2.53
4.04
0.40
• 1.72
1.01
1.08
1.94
5.82
4.44
2.12
0.49
2.82
••0.26
0.76
0.87
2.47
1.77
5.94
2
0.73
«0.61
*
*
Avg.
0.55
2.75
•*0.3
0.93
4.49
2.17
2.53
4.04
0.40
1.72
1.01
'1.08
1.94
5.82
4.44
1 . 06 to
1.36
0.49
2.82
-eO.3
0.76
0.87
2.47
1.77
—3.0
Uf.
I
2.48
5.29
«a.48
1.54
6.37
2.47
2.32
5.66
1.73
3.24
1.65
1.46
2.71
6.68
4.01
2.87
2.1
5.24
«0.43
. 1.00
1.20
2.70
1.57
8.03
/m3
2
0.97
«0. 85
...
*
Avg.
2.48
5.29
*0.5
1.25
6.37
2.47
2.32
5.66
1.73
3.24
1.65
1.46
2.71
6.68
4.01
1.44 to
1.86
2.1
5.24
««0.4
'l.OO
1.20
2.70
1.57
-4.0^
Fuel
B+
A+
C+
mg/hr
1
1.43
2.98
3.86
7.33
19.4
6.36
5.51**
7.71
*0. 13
2.07
*
* .
*
3.62
*
*
1.21
6.53
3.08
6.31
5.75
11.6
13.5
7.07
2
9.66
1.54
12.1
Avg.
1.43
2.98
3.86
7.33
19.4
6.36
5.51
8.68
•*0.1
2.07
*
*
*
3.62
*
~0.08
1.21
6.53
3.08
6.31
5.75
11.6
13.5
9.58
tfg/m3
1
6.42
5.63
6.35
9.83
26.8
7.14
5.03**
10.7
•*0.55
3.83
*
*
*
4.15
*
*
5.60
12.2
4.93
8.15
7.87
12.9
9.99
9.59
2
13.3
2.11
15.9
Avg.
6.42
5.63
6.35
9.83
26.8
7.14
5.03
12.0
«0.5
3.83
*
*
*
4.15
*
—1.0
5.60
12.2
4.93
8.15
7.87
12.9
9.99
12.8
a
* Not enough to measure
** Small amount of solubles
-------
TECHNICAL REPORT DATA
(Pleaseread Instructions on the reverse before completing)
1. REPORT NO.
EPA-650/2-75-056
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Methodology for Determining Fuel Effects on Diesel
Particulate Emissions
5. REPORT DATE
March 19,75
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
Charles T. Hare
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Southwest Research Institute
8500 Culebra Road
San Antonio, Texas 78284
10. PROGRAM ELEMENT NO.
1AA002; ROAP 26AAE-19
11. CONTRACT/GRANT NO.
68-02-1230
12. SPONSORING AGENCY NAME AND ADDRESS
U.S. Environmental Protection Agency
Office of Research and Development
Washington, D. C. 20460
13. TYPE OF REPORT AND PERIOD COVERED
. Final Report 6/73-2/75
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
16. ABSTRACT
To develop a methodology for characterizing particulate emissions from
diesel engines, one 2-stroke cycle engine and one 4-stroke cycle engine were
operated in both individual steady-state modes and according to a variation of the
13-mode diesel emissions measurement procedure. Both engines were operated
on three fuel types, each of which was used with one of two available diesel fuel
additives as well as by itself.
The primary particulate sampling technique employed was a dilution tunnel;
and secondary evaluation techniques included a diluter-sampler developed under
contract to EPA by another organization, a light extinction smokemeter, and a
filter-type sampling smokemeter. Gaseous emissions were also measured, pro-
viding a running check on engine conditon.
Particulate mass rates were calculated from gravimetric data; and analysis
of particulate included determination of sulfur, carbon, hydrogen, nitrogen, phe-
nols, nitrosamines, trace metals, and organic solubles. Analysis of the organic
soluble fraction included NMR, IR, paraffin boiling point distribution, benz(a)-
pyrene, sulfur, carbon, hydrogen, nitrogen, and oxygen.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COS AT I Field/Group
Diesel Engine Emissions
Particulate Emissions
Fuel Effects
Dilution Tunnel
Gravimetric Analysis
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
200
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
-------